* 'rbd-sysfs' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
rbd: replace the rbd sysfs interface
</sect2>
</sect1>
</chapter>
- <chapter id="clk">
- <title>Clock Framework Extensions</title>
-!Iinclude/linux/sh_clk.h
- </chapter>
<chapter id="mach">
<title>Machine Specific Interfaces</title>
<sect1 id="dreamcast">
The representation of the above is reflected in the directory tree
in EDAC's sysfs interface. Starting in directory
/sys/devices/system/edac/mc each memory controller will be represented
-by its own 'mcX' directory, where 'X" is the index of the MC.
+by its own 'mcX' directory, where 'X' is the index of the MC.
..../edac/mc/
....
Under each 'mcX' directory each 'csrowX' is again represented by a
-'csrowX', where 'X" is the csrow index:
+'csrowX', where 'X' is the csrow index:
.../mc/mc0/
In 'mcX' directories are EDAC control and attribute files for
-this 'X" instance of the memory controllers:
+this 'X' instance of the memory controllers:
Counter reset control file:
'csrowX' DIRECTORIES
In the 'csrowX' directories are EDAC control and attribute files for
-this 'X" instance of csrow:
+this 'X' instance of csrow:
Total Uncorrectable Errors count attribute file:
Geert Uytterhoeven <geert@linux-m68k.org>
00-INDEX
- - this file
+ - this file.
arkfb.txt
- info on the fbdev driver for ARK Logic chips.
aty128fb.txt
- info on the ATI Rage128 frame buffer driver.
cirrusfb.txt
- info on the driver for Cirrus Logic chipsets.
+cmap_xfbdev.txt
+ - an introduction to fbdev's cmap structures.
deferred_io.txt
- an introduction to deferred IO.
+efifb.txt
+ - info on the EFI platform driver for Intel based Apple computers.
+ep93xx-fb.txt
+ - info on the driver for EP93xx LCD controller.
fbcon.txt
- intro to and usage guide for the framebuffer console (fbcon).
framebuffer.txt
- introduction to frame buffer devices.
-imacfb.txt
- - info on the generic EFI platform driver for Intel based Macs.
+gxfb.txt
+ - info on the framebuffer driver for AMD Geode GX2 based processors.
intel810.txt
- documentation for the Intel 810/815 framebuffer driver.
intelfb.txt
- docs for Intel 830M/845G/852GM/855GM/865G/915G/945G fb driver.
internals.txt
- quick overview of frame buffer device internals.
+lxfb.txt
+ - info on the framebuffer driver for AMD Geode LX based processors.
matroxfb.txt
- info on the Matrox framebuffer driver for Alpha, Intel and PPC.
+metronomefb.txt
+ - info on the driver for the Metronome display controller.
modedb.txt
- info on the video mode database.
-matroxfb.txt
- - info on the Matrox frame buffer driver.
pvr2fb.txt
- info on the PowerVR 2 frame buffer driver.
pxafb.txt
- info on the fbdev driver for S3 Trio/Virge chips.
sa1100fb.txt
- information about the driver for the SA-1100 LCD controller.
+sh7760fb.txt
+ - info on the SH7760/SH7763 integrated LCDC Framebuffer driver.
sisfb.txt
- info on the framebuffer device driver for various SiS chips.
sstfb.txt
- info on the frame buffer driver for 3dfx' Voodoo Graphics boards.
tgafb.txt
- - info on the TGA (DECChip 21030) frame buffer driver
+ - info on the TGA (DECChip 21030) frame buffer driver.
+tridentfb.txt
+ info on the framebuffer driver for some Trident chip based cards.
+uvesafb.txt
+ - info on the userspace VESA (VBE2+ compliant) frame buffer device.
vesafb.txt
- - info on the VESA frame buffer device
+ - info on the VESA frame buffer device.
+viafb.modes
+ - list of modes for VIA Integration Graphic Chip.
+viafb.txt
+ - info on the VIA Integration Graphic Chip console framebuffer driver.
vt8623fb.txt
- info on the fb driver for the graphics core in VIA VT8623 chipsets.
improve throughput, but will also increase the
amount of memory reserved for use by the client.
+ swapaccount[=0|1]
+ [KNL] Enable accounting of swap in memory resource
+ controller if no parameter or 1 is given or disable
+ it if 0 is given (See Documentation/cgroups/memory.txt)
+
swiotlb= [IA-64] Number of I/O TLB slabs
switches= [HW,M68k]
Count buffering overhead as bytes/2^tcp_adv_win_scale
(if tcp_adv_win_scale > 0) or bytes-bytes/2^(-tcp_adv_win_scale),
if it is <= 0.
+ Possible values are [-31, 31], inclusive.
Default: 2
tcp_allowed_congestion_control - STRING
+++ /dev/null
-Clock framework on SuperH architecture
-
-The framework on SH extends existing API by the function clk_set_rate_ex,
-which prototype is as follows:
-
- clk_set_rate_ex (struct clk *clk, unsigned long rate, int algo_id)
-
-The algo_id parameter is used to specify algorithm used to recalculate clocks,
-adjanced to clock, specified as first argument. It is assumed that algo_id==0
-means no changes to adjanced clock
-
-Internally, the clk_set_rate_ex forwards request to clk->ops->set_rate method,
-if it is present in ops structure. The method should set the clock rate and adjust
-all needed clocks according to the passed algo_id.
-Exact values for algo_id are machine-dependent. For the sh7722, the following
-values are defined:
-
- NO_CHANGE = 0,
- IUS_N1_N1, /* I:U = N:1, U:Sh = N:1 */
- IUS_322, /* I:U:Sh = 3:2:2 */
- IUS_522, /* I:U:Sh = 5:2:2 */
- IUS_N11, /* I:U:Sh = N:1:1 */
- SB_N1, /* Sh:B = N:1 */
- SB3_N1, /* Sh:B3 = N:1 */
- SB3_32, /* Sh:B3 = 3:2 */
- SB3_43, /* Sh:B3 = 4:3 */
- SB3_54, /* Sh:B3 = 5:4 */
- BP_N1, /* B:P = N:1 */
- IP_N1 /* I:P = N:1 */
-
-Each of these constants means relation between clocks that can be set via the FRQCR
-register
BONDING DRIVER
M: Jay Vosburgh <fubar@us.ibm.com>
-L: bonding-devel@lists.sourceforge.net
+L: netdev@vger.kernel.org
W: http://sourceforge.net/projects/bonding/
S: Supported
F: drivers/net/bonding/
INTEL DRM DRIVERS (excluding Poulsbo, Moorestown and derivative chipsets)
M: Chris Wilson <chris@chris-wilson.co.uk>
-L: intel-gfx@lists.freedesktop.org
+L: intel-gfx@lists.freedesktop.org (subscribers-only)
L: dri-devel@lists.freedesktop.org
T: git git://git.kernel.org/pub/scm/linux/kernel/git/ickle/drm-intel.git
S: Supported
FRAMEBUFFER LAYER
L: linux-fbdev@vger.kernel.org
W: http://linux-fbdev.sourceforge.net/
+Q: http://patchwork.kernel.org/project/linux-fbdev/list/
T: git git://git.kernel.org/pub/scm/linux/kernel/git/lethal/fbdev-2.6.git
S: Orphan
F: Documentation/fb/
-F: drivers/video/fb*
+F: drivers/video/
+F: include/video/
F: include/linux/fb.h
FREESCALE DMA DRIVER
W: http://www.tilera.com/scm/
S: Supported
F: arch/tile/
+F: drivers/char/hvc_tile.c
+F: drivers/net/tile/
TLAN NETWORK DRIVER
M: Samuel Chessman <chessman@tux.org>
VERSION = 2
PATCHLEVEL = 6
SUBLEVEL = 37
-EXTRAVERSION = -rc3
+EXTRAVERSION = -rc4
NAME = Flesh-Eating Bats with Fangs
# *DOCUMENTATION*
reloc_end:
.align
- .section ".stack", "w"
+ .section ".stack", "aw", %nobits
user_stack: .space 4096
user_stack_end:
.bss : { *(.bss) }
_end = .;
- .stack (NOLOAD) : { *(.stack) }
+ .stack : { *(.stack) }
.stab 0 : { *(.stab) }
.stabstr 0 : { *(.stabstr) }
--- /dev/null
+CONFIG_EXPERIMENTAL=y
+# CONFIG_LOCALVERSION_AUTO is not set
+# CONFIG_SWAP is not set
+CONFIG_SYSVIPC=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_MODULES=y
+CONFIG_MODULE_FORCE_LOAD=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODVERSIONS=y
+CONFIG_MODULE_SRCVERSION_ALL=y
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_CFQ is not set
+CONFIG_ARCH_AT91=y
+CONFIG_MACH_ONEARM=y
+CONFIG_ARCH_AT91RM9200DK=y
+CONFIG_MACH_AT91RM9200EK=y
+CONFIG_MACH_CSB337=y
+CONFIG_MACH_CSB637=y
+CONFIG_MACH_CARMEVA=y
+CONFIG_MACH_ATEB9200=y
+CONFIG_MACH_KB9200=y
+CONFIG_MACH_PICOTUX2XX=y
+CONFIG_MACH_KAFA=y
+CONFIG_MACH_ECBAT91=y
+CONFIG_MACH_YL9200=y
+CONFIG_MACH_CPUAT91=y
+CONFIG_MACH_ECO920=y
+CONFIG_MTD_AT91_DATAFLASH_CARD=y
+CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
+CONFIG_AT91_TIMER_HZ=100
+# CONFIG_ARM_THUMB is not set
+CONFIG_PCCARD=y
+CONFIG_AT91_CF=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_PREEMPT=y
+CONFIG_AEABI=y
+CONFIG_LEDS=y
+CONFIG_LEDS_CPU=y
+CONFIG_ZBOOT_ROM_TEXT=0x10000000
+CONFIG_ZBOOT_ROM_BSS=0x20040000
+CONFIG_KEXEC=y
+CONFIG_FPE_NWFPE=y
+CONFIG_BINFMT_MISC=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_XFRM_USER=m
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_NET_IPIP=m
+CONFIG_NET_IPGRE=m
+CONFIG_INET_AH=m
+CONFIG_INET_ESP=m
+CONFIG_INET_IPCOMP=m
+CONFIG_INET_XFRM_MODE_TRANSPORT=m
+CONFIG_INET_XFRM_MODE_TUNNEL=m
+CONFIG_INET_XFRM_MODE_BEET=m
+CONFIG_IPV6_PRIVACY=y
+CONFIG_IPV6_ROUTER_PREF=y
+CONFIG_IPV6_ROUTE_INFO=y
+CONFIG_INET6_AH=m
+CONFIG_INET6_ESP=m
+CONFIG_INET6_IPCOMP=m
+CONFIG_IPV6_MIP6=m
+CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m
+CONFIG_IPV6_TUNNEL=m
+CONFIG_BRIDGE=m
+CONFIG_VLAN_8021Q=m
+CONFIG_BT=m
+CONFIG_BT_L2CAP=m
+CONFIG_BT_SCO=m
+CONFIG_BT_RFCOMM=m
+CONFIG_BT_RFCOMM_TTY=y
+CONFIG_BT_BNEP=m
+CONFIG_BT_BNEP_MC_FILTER=y
+CONFIG_BT_BNEP_PROTO_FILTER=y
+CONFIG_BT_HIDP=m
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_MTD=y
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_AFS_PARTS=y
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+CONFIG_MTD_CFI=y
+CONFIG_MTD_JEDECPROBE=y
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_MTD_COMPLEX_MAPPINGS=y
+CONFIG_MTD_PHYSMAP=y
+CONFIG_MTD_PLATRAM=y
+CONFIG_MTD_DATAFLASH=y
+CONFIG_MTD_NAND=y
+CONFIG_MTD_NAND_ATMEL=y
+CONFIG_MTD_NAND_PLATFORM=y
+CONFIG_MTD_UBI=y
+CONFIG_MTD_UBI_GLUEBI=y
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_NBD=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_ATMEL_TCLIB=y
+CONFIG_EEPROM_LEGACY=m
+CONFIG_SCSI=y
+CONFIG_BLK_DEV_SD=y
+CONFIG_BLK_DEV_SR=m
+CONFIG_BLK_DEV_SR_VENDOR=y
+CONFIG_CHR_DEV_SG=m
+CONFIG_SCSI_MULTI_LUN=y
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_NETDEVICES=y
+CONFIG_TUN=m
+CONFIG_PHYLIB=y
+CONFIG_DAVICOM_PHY=y
+CONFIG_SMSC_PHY=y
+CONFIG_MICREL_PHY=y
+CONFIG_NET_ETHERNET=y
+CONFIG_ARM_AT91_ETHER=y
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+CONFIG_USB_CATC=m
+CONFIG_USB_KAWETH=m
+CONFIG_USB_PEGASUS=m
+CONFIG_USB_RTL8150=m
+CONFIG_USB_USBNET=m
+CONFIG_USB_NET_DM9601=m
+CONFIG_USB_NET_GL620A=m
+CONFIG_USB_NET_PLUSB=m
+CONFIG_USB_NET_RNDIS_HOST=m
+CONFIG_USB_ALI_M5632=y
+CONFIG_USB_AN2720=y
+CONFIG_USB_EPSON2888=y
+CONFIG_PPP=y
+CONFIG_PPP_MULTILINK=y
+CONFIG_PPP_FILTER=y
+CONFIG_PPP_ASYNC=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_MPPE=m
+CONFIG_PPPOE=m
+CONFIG_SLIP=m
+CONFIG_SLIP_COMPRESSED=y
+CONFIG_SLIP_SMART=y
+CONFIG_SLIP_MODE_SLIP6=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=640
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=480
+CONFIG_INPUT_EVDEV=y
+CONFIG_KEYBOARD_GPIO=y
+# CONFIG_INPUT_MOUSE is not set
+CONFIG_INPUT_TOUCHSCREEN=y
+CONFIG_SERIAL_ATMEL=y
+CONFIG_SERIAL_ATMEL_CONSOLE=y
+CONFIG_LEGACY_PTY_COUNT=32
+CONFIG_HW_RANDOM=y
+CONFIG_I2C=y
+CONFIG_I2C_CHARDEV=y
+CONFIG_I2C_GPIO=y
+CONFIG_SPI=y
+CONFIG_SPI_ATMEL=y
+CONFIG_SPI_BITBANG=y
+CONFIG_GPIO_SYSFS=y
+CONFIG_HWMON=m
+CONFIG_SENSORS_ADM1021=m
+CONFIG_SENSORS_ADM1025=m
+CONFIG_SENSORS_ADM1026=m
+CONFIG_SENSORS_ADM1029=m
+CONFIG_SENSORS_ADM1031=m
+CONFIG_SENSORS_ADM9240=m
+CONFIG_SENSORS_DS1621=m
+CONFIG_SENSORS_GL518SM=m
+CONFIG_SENSORS_GL520SM=m
+CONFIG_SENSORS_IT87=m
+CONFIG_SENSORS_LM63=m
+CONFIG_SENSORS_LM73=m
+CONFIG_SENSORS_LM75=m
+CONFIG_SENSORS_LM77=m
+CONFIG_SENSORS_LM78=m
+CONFIG_SENSORS_LM80=m
+CONFIG_SENSORS_LM83=m
+CONFIG_SENSORS_LM85=m
+CONFIG_SENSORS_LM87=m
+CONFIG_SENSORS_LM90=m
+CONFIG_SENSORS_LM92=m
+CONFIG_SENSORS_MAX1619=m
+CONFIG_SENSORS_PCF8591=m
+CONFIG_SENSORS_SMSC47B397=m
+CONFIG_SENSORS_W83781D=m
+CONFIG_SENSORS_W83791D=m
+CONFIG_SENSORS_W83792D=m
+CONFIG_SENSORS_W83793=m
+CONFIG_SENSORS_W83L785TS=m
+CONFIG_WATCHDOG=y
+CONFIG_WATCHDOG_NOWAYOUT=y
+CONFIG_AT91RM9200_WATCHDOG=y
+CONFIG_FB=y
+CONFIG_FB_MODE_HELPERS=y
+CONFIG_FB_TILEBLITTING=y
+CONFIG_FB_S1D13XXX=y
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=y
+CONFIG_BACKLIGHT_CLASS_DEVICE=y
+# CONFIG_BACKLIGHT_GENERIC is not set
+CONFIG_DISPLAY_SUPPORT=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+CONFIG_FONTS=y
+CONFIG_FONT_MINI_4x6=y
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+CONFIG_USB=y
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_DEVICE_CLASS is not set
+CONFIG_USB_MON=y
+CONFIG_USB_OHCI_HCD=y
+CONFIG_USB_ACM=m
+CONFIG_USB_PRINTER=m
+CONFIG_USB_STORAGE=y
+CONFIG_USB_SERIAL=y
+CONFIG_USB_SERIAL_CONSOLE=y
+CONFIG_USB_SERIAL_GENERIC=y
+CONFIG_USB_SERIAL_FTDI_SIO=y
+CONFIG_USB_SERIAL_KEYSPAN=y
+CONFIG_USB_SERIAL_KEYSPAN_MPR=y
+CONFIG_USB_SERIAL_KEYSPAN_USA28=y
+CONFIG_USB_SERIAL_KEYSPAN_USA28X=y
+CONFIG_USB_SERIAL_KEYSPAN_USA28XA=y
+CONFIG_USB_SERIAL_KEYSPAN_USA28XB=y
+CONFIG_USB_SERIAL_KEYSPAN_USA19=y
+CONFIG_USB_SERIAL_KEYSPAN_USA18X=y
+CONFIG_USB_SERIAL_KEYSPAN_USA19W=y
+CONFIG_USB_SERIAL_KEYSPAN_USA19QW=y
+CONFIG_USB_SERIAL_KEYSPAN_USA19QI=y
+CONFIG_USB_SERIAL_KEYSPAN_USA49W=y
+CONFIG_USB_SERIAL_KEYSPAN_USA49WLC=y
+CONFIG_USB_SERIAL_MCT_U232=y
+CONFIG_USB_SERIAL_PL2303=y
+CONFIG_USB_GADGET=y
+CONFIG_USB_ETH=m
+CONFIG_USB_MASS_STORAGE=m
+CONFIG_MMC=y
+CONFIG_MMC_AT91=y
+CONFIG_NEW_LEDS=y
+CONFIG_LEDS_CLASS=y
+CONFIG_LEDS_GPIO=y
+CONFIG_LEDS_TRIGGERS=y
+CONFIG_LEDS_TRIGGER_TIMER=y
+CONFIG_LEDS_TRIGGER_HEARTBEAT=y
+CONFIG_LEDS_TRIGGER_GPIO=y
+CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
+CONFIG_RTC_CLASS=y
+# CONFIG_RTC_HCTOSYS is not set
+CONFIG_RTC_DRV_DS1307=y
+CONFIG_RTC_DRV_PCF8563=y
+CONFIG_RTC_DRV_AT91RM9200=y
+CONFIG_EXT2_FS=y
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT3_FS=y
+# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_REISERFS_FS=y
+CONFIG_AUTOFS4_FS=y
+CONFIG_ISO9660_FS=y
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_UDF_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_NTFS_FS=m
+CONFIG_TMPFS=y
+CONFIG_CONFIGFS_FS=y
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_SUMMARY=y
+CONFIG_JFFS2_COMPRESSION_OPTIONS=y
+CONFIG_JFFS2_LZO=y
+CONFIG_JFFS2_RUBIN=y
+CONFIG_CRAMFS=y
+CONFIG_MINIX_FS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+CONFIG_NFS_V3_ACL=y
+CONFIG_NFS_V4=y
+CONFIG_ROOT_NFS=y
+CONFIG_NFSD=y
+CONFIG_SMB_FS=m
+CONFIG_CIFS=m
+CONFIG_PARTITION_ADVANCED=y
+CONFIG_MAC_PARTITION=y
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_CODEPAGE_737=m
+CONFIG_NLS_CODEPAGE_775=m
+CONFIG_NLS_CODEPAGE_850=m
+CONFIG_NLS_CODEPAGE_852=m
+CONFIG_NLS_CODEPAGE_855=m
+CONFIG_NLS_CODEPAGE_857=m
+CONFIG_NLS_CODEPAGE_860=m
+CONFIG_NLS_CODEPAGE_861=m
+CONFIG_NLS_CODEPAGE_862=m
+CONFIG_NLS_CODEPAGE_863=m
+CONFIG_NLS_CODEPAGE_864=m
+CONFIG_NLS_CODEPAGE_865=m
+CONFIG_NLS_CODEPAGE_866=m
+CONFIG_NLS_CODEPAGE_869=m
+CONFIG_NLS_CODEPAGE_936=m
+CONFIG_NLS_CODEPAGE_950=m
+CONFIG_NLS_CODEPAGE_932=m
+CONFIG_NLS_CODEPAGE_949=m
+CONFIG_NLS_CODEPAGE_874=m
+CONFIG_NLS_ISO8859_8=m
+CONFIG_NLS_CODEPAGE_1250=m
+CONFIG_NLS_CODEPAGE_1251=m
+CONFIG_NLS_ASCII=m
+CONFIG_NLS_ISO8859_1=y
+CONFIG_NLS_ISO8859_2=m
+CONFIG_NLS_ISO8859_3=m
+CONFIG_NLS_ISO8859_4=m
+CONFIG_NLS_ISO8859_5=m
+CONFIG_NLS_ISO8859_6=m
+CONFIG_NLS_ISO8859_7=m
+CONFIG_NLS_ISO8859_9=m
+CONFIG_NLS_ISO8859_13=m
+CONFIG_NLS_ISO8859_14=m
+CONFIG_NLS_ISO8859_15=m
+CONFIG_NLS_KOI8_R=m
+CONFIG_NLS_KOI8_U=m
+CONFIG_NLS_UTF8=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_FS=y
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_RCU_CPU_STALL_DETECTOR is not set
+# CONFIG_FTRACE is not set
+CONFIG_CRYPTO_PCBC=y
+CONFIG_CRYPTO_SHA1=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_ARCH_AT91RM9200DK=y
-CONFIG_MACH_ECO920=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_PCCARD=y
-CONFIG_AT91_CF=y
-CONFIG_LEDS=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,115200 initrd=0x20410000,3145728 root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_IPV6 is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_JEDECPROBE=y
-CONFIG_MTD_CFI_AMDSTD=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_I2C_GPIO=y
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_GADGET=y
-CONFIG_MMC=y
-CONFIG_RTC_CLASS=y
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_LL=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_LOCALVERSION_AUTO is not set
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_AT91RM9200EK=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_LEDS=y
-CONFIG_LEDS_CPU=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,115200 initrd=0x20410000,3145728 root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_IPV6 is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_JEDECPROBE=y
-CONFIG_MTD_CFI_AMDSTD=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_I2C_GPIO=y
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-CONFIG_FB=y
-CONFIG_FB_S1D13XXX=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_GADGET=y
-CONFIG_MMC=y
-CONFIG_RTC_CLASS=y
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_LL=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_EMBEDDED=y
-CONFIG_SLAB=y
-CONFIG_PROFILING=y
-CONFIG_OPROFILE=m
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_ATEB9200=y
-CONFIG_PCCARD=m
-CONFIG_AT91_CF=m
-CONFIG_PREEMPT=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_FPE_NWFPE=y
-CONFIG_PM=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_NET_KEY=y
-CONFIG_INET=y
-# CONFIG_IPV6 is not set
-CONFIG_BRIDGE=m
-CONFIG_VLAN_8021Q=m
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK_RO=y
-CONFIG_BLK_DEV_LOOP=m
-CONFIG_BLK_DEV_NBD=m
-CONFIG_SCSI=m
-CONFIG_BLK_DEV_SD=m
-CONFIG_BLK_DEV_SR=m
-CONFIG_BLK_DEV_SR_VENDOR=y
-CONFIG_CHR_DEV_SG=m
-CONFIG_SCSI_MULTI_LUN=y
-CONFIG_NETDEVICES=y
-CONFIG_DUMMY=m
-CONFIG_TUN=m
-CONFIG_PHYLIB=y
-CONFIG_DAVICOM_PHY=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-CONFIG_USB_USBNET=y
-CONFIG_USB_NET_GL620A=y
-CONFIG_USB_NET_PLUSB=y
-CONFIG_USB_NET_RNDIS_HOST=y
-CONFIG_USB_ALI_M5632=y
-CONFIG_USB_AN2720=y
-CONFIG_USB_EPSON2888=y
-CONFIG_PPP=m
-CONFIG_PPP_ASYNC=m
-CONFIG_PPP_SYNC_TTY=m
-CONFIG_PPP_DEFLATE=m
-CONFIG_PPP_BSDCOMP=m
-CONFIG_PPPOE=m
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_I2C=m
-CONFIG_I2C_CHARDEV=m
-CONFIG_I2C_GPIO=m
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_SOUND=y
-CONFIG_USB_HID=m
-CONFIG_HID_PID=y
-CONFIG_USB_HIDDEV=y
-CONFIG_USB=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_ACM=m
-CONFIG_USB_PRINTER=m
-CONFIG_USB_STORAGE=m
-CONFIG_USB_STORAGE_DATAFAB=m
-CONFIG_USB_STORAGE_FREECOM=m
-CONFIG_USB_STORAGE_USBAT=m
-CONFIG_USB_STORAGE_SDDR09=m
-CONFIG_USB_STORAGE_SDDR55=m
-CONFIG_USB_STORAGE_JUMPSHOT=m
-CONFIG_USB_SERIAL=m
-CONFIG_USB_SERIAL_GENERIC=y
-CONFIG_USB_SERIAL_FTDI_SIO=m
-CONFIG_USB_SERIAL_PL2303=m
-CONFIG_USB_GADGET=m
-CONFIG_USB_ETH=m
-CONFIG_USB_GADGETFS=m
-CONFIG_USB_FILE_STORAGE=m
-CONFIG_USB_G_SERIAL=m
-CONFIG_MMC=m
-CONFIG_MMC_DEBUG=y
-CONFIG_RTC_CLASS=y
-# CONFIG_RTC_HCTOSYS is not set
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=m
-CONFIG_EXT3_FS=m
-CONFIG_REISERFS_FS=m
-CONFIG_INOTIFY=y
-CONFIG_ISO9660_FS=m
-CONFIG_JOLIET=y
-CONFIG_ZISOFS=y
-CONFIG_UDF_FS=m
-CONFIG_MSDOS_FS=m
-CONFIG_VFAT_FS=m
-CONFIG_NTFS_FS=m
-CONFIG_NTFS_RW=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=m
-CONFIG_NFS_V3=y
-CONFIG_NFS_V3_ACL=y
-CONFIG_NFS_V4=y
-CONFIG_NFSD=m
-CONFIG_NFSD_V4=y
-CONFIG_PARTITION_ADVANCED=y
-CONFIG_MAC_PARTITION=y
-CONFIG_BSD_DISKLABEL=y
-CONFIG_MINIX_SUBPARTITION=y
-CONFIG_SOLARIS_X86_PARTITION=y
-CONFIG_UNIXWARE_DISKLABEL=y
-CONFIG_NLS_CODEPAGE_932=m
-CONFIG_NLS_ASCII=m
-CONFIG_NLS_ISO8859_15=m
-CONFIG_NLS_UTF8=m
-CONFIG_CRYPTO_MD5=y
-CONFIG_CRYPTO_MICHAEL_MIC=m
-CONFIG_CRYPTO_ARC4=m
-CONFIG_CRC16=m
-CONFIG_LIBCRC32C=m
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_EMBEDDED=y
-# CONFIG_HOTPLUG is not set
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_CARMEVA=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_MULTICAST=y
-CONFIG_IP_PNP=y
-# CONFIG_IPV6 is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-CONFIG_SERIO=m
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_MMC=m
-CONFIG_MMC_DEBUG=y
-CONFIG_EXT2_FS=y
-CONFIG_EXT2_FS_XATTR=y
-# CONFIG_DNOTIFY is not set
-CONFIG_JFFS2_FS=y
-CONFIG_JFFS2_COMPRESSION_OPTIONS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_NFS_V4=y
-CONFIG_ROOT_NFS=y
-CONFIG_NFSD=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_LOCALVERSION_AUTO is not set
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_BLK_DEV_BSG is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_CPUAT91=y
-CONFIG_AT91_TIMER_HZ=100
-# CONFIG_ARM_THUMB is not set
-CONFIG_PREEMPT=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
-# CONFIG_INET_XFRM_MODE_TUNNEL is not set
-# CONFIG_INET_XFRM_MODE_BEET is not set
-# CONFIG_IPV6 is not set
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_CFI_INTELEXT=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PLATRAM=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_BLK_DEV_NBD=y
-CONFIG_BLK_DEV_RAM=y
-# CONFIG_MISC_DEVICES is not set
-CONFIG_SCSI=y
-CONFIG_BLK_DEV_SD=y
-CONFIG_SCSI_MULTI_LUN=y
-# CONFIG_SCSI_LOWLEVEL is not set
-CONFIG_NETDEVICES=y
-CONFIG_PHYLIB=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
-CONFIG_PPP=y
-CONFIG_PPP_ASYNC=y
-CONFIG_PPP_DEFLATE=y
-CONFIG_PPP_BSDCOMP=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_LEGACY_PTY_COUNT=32
-# CONFIG_HW_RANDOM is not set
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_I2C_GPIO=y
-CONFIG_GPIO_SYSFS=y
-# CONFIG_HWMON is not set
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_HID_SUPPORT is not set
-CONFIG_USB=y
-# CONFIG_USB_DEVICE_CLASS is not set
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_GADGET=y
-CONFIG_USB_ETH=m
-CONFIG_MMC=y
-CONFIG_MMC_AT91=m
-CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=y
-CONFIG_LEDS_GPIO=y
-CONFIG_LEDS_TRIGGERS=y
-CONFIG_LEDS_TRIGGER_TIMER=y
-CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_LEDS_TRIGGER_GPIO=y
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
-CONFIG_RTC_CLASS=y
-# CONFIG_RTC_HCTOSYS is not set
-CONFIG_RTC_DRV_DS1307=y
-CONFIG_RTC_DRV_PCF8563=y
-CONFIG_EXT2_FS=y
-CONFIG_EXT3_FS=y
-# CONFIG_EXT3_FS_XATTR is not set
-CONFIG_INOTIFY=y
-CONFIG_AUTOFS4_FS=y
-CONFIG_MSDOS_FS=y
-CONFIG_VFAT_FS=y
-CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
-CONFIG_JFFS2_SUMMARY=y
-CONFIG_CRAMFS=y
-CONFIG_MINIX_FS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_ROOT_NFS=y
-CONFIG_PARTITION_ADVANCED=y
-CONFIG_NLS_CODEPAGE_437=y
-CONFIG_NLS_ISO8859_1=y
-CONFIG_NLS_UTF8=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_BLK_DEV_BSG is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_CSB337=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_PCCARD=y
-CONFIG_AT91_CF=y
-CONFIG_LEDS=y
-CONFIG_LEDS_CPU=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,38400 initrd=0x20410000,3145728 root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_DHCP=y
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_INET_LRO is not set
-# CONFIG_IPV6 is not set
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_CFI_INTELEXT=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_ATMEL_SSC=y
-CONFIG_SCSI=y
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_HW_RANDOM is not set
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_I2C_GPIO=y
-# CONFIG_HWMON is not set
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_SERIAL=y
-CONFIG_USB_SERIAL_CONSOLE=y
-CONFIG_USB_SERIAL_GENERIC=y
-CONFIG_USB_SERIAL_FTDI_SIO=y
-CONFIG_USB_SERIAL_KEYSPAN=y
-CONFIG_USB_SERIAL_KEYSPAN_MPR=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28X=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28XA=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28XB=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19=y
-CONFIG_USB_SERIAL_KEYSPAN_USA18X=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19W=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19QW=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19QI=y
-CONFIG_USB_SERIAL_KEYSPAN_USA49W=y
-CONFIG_USB_SERIAL_KEYSPAN_USA49WLC=y
-CONFIG_USB_SERIAL_MCT_U232=y
-CONFIG_USB_GADGET=y
-CONFIG_MMC=y
-CONFIG_RTC_CLASS=y
-CONFIG_RTC_HCTOSYS_DEVICE="rtc1"
-# CONFIG_RTC_INTF_SYSFS is not set
-CONFIG_RTC_DRV_DS1307=y
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_NFS_V4=y
-CONFIG_ROOT_NFS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_LL=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_BLK_DEV_BSG is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_CSB637=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_PCCARD=y
-CONFIG_AT91_CF=y
-CONFIG_LEDS=y
-CONFIG_LEDS_CPU=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,38400 initrd=0x20410000,3145728 root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_DHCP=y
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_INET_LRO is not set
-# CONFIG_IPV6 is not set
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_CFI_INTELEXT=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_SCSI=y
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_SERIAL=y
-CONFIG_USB_SERIAL_CONSOLE=y
-CONFIG_USB_SERIAL_GENERIC=y
-CONFIG_USB_SERIAL_FTDI_SIO=y
-CONFIG_USB_SERIAL_KEYSPAN=y
-CONFIG_USB_SERIAL_KEYSPAN_MPR=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28X=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28XA=y
-CONFIG_USB_SERIAL_KEYSPAN_USA28XB=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19=y
-CONFIG_USB_SERIAL_KEYSPAN_USA18X=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19W=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19QW=y
-CONFIG_USB_SERIAL_KEYSPAN_USA19QI=y
-CONFIG_USB_SERIAL_KEYSPAN_USA49W=y
-CONFIG_USB_SERIAL_KEYSPAN_USA49WLC=y
-CONFIG_USB_SERIAL_MCT_U232=y
-CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=y
-CONFIG_LEDS_GPIO=y
-CONFIG_LEDS_TRIGGERS=y
-CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_NFS_V4=y
-CONFIG_ROOT_NFS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_LL=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-CONFIG_SYSVIPC=y
-CONFIG_IKCONFIG=y
-CONFIG_IKCONFIG_PROC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SLAB=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_ECBAT91=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-CONFIG_PCCARD=y
-CONFIG_AT91_CF=y
-CONFIG_PREEMPT=y
-CONFIG_LEDS=y
-CONFIG_LEDS_CPU=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="rootfstype=reiserfs root=/dev/mmcblk0p1 console=ttyS0,115200n8 rootdelay=1"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_DHCP=y
-# CONFIG_IPV6 is not set
-CONFIG_CFG80211=y
-CONFIG_MAC80211=y
-# CONFIG_STANDALONE is not set
-# CONFIG_PREVENT_FIRMWARE_BUILD is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_AFS_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_DATAFLASH=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_SCSI=y
-CONFIG_BLK_DEV_SD=y
-CONFIG_CHR_DEV_SG=y
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
-CONFIG_PPP=y
-CONFIG_PPP_MULTILINK=y
-CONFIG_PPP_FILTER=y
-CONFIG_PPP_ASYNC=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_HW_RANDOM=y
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_SPI=y
-CONFIG_SPI_BITBANG=y
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-# CONFIG_VGA_CONSOLE is not set
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEVICEFS=y
-# CONFIG_USB_DEVICE_CLASS is not set
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_PRINTER=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_GADGET=y
-CONFIG_MMC=y
-CONFIG_MMC_DEBUG=y
-CONFIG_MMC_AT91=m
-CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=y
-CONFIG_RTC_CLASS=y
-# CONFIG_RTC_HCTOSYS is not set
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_EXT3_FS=y
-CONFIG_REISERFS_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CONFIGFS_FS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_NFS_V3_ACL=y
-CONFIG_NFS_V4=y
-CONFIG_ROOT_NFS=y
-CONFIG_PARTITION_ADVANCED=y
-CONFIG_DEBUG_USER=y
-CONFIG_CRYPTO_PCBC=y
-CONFIG_CRYPTO_SHA1=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_LOCALVERSION_AUTO is not set
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SLAB=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_KAFA=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_PREEMPT=y
-CONFIG_LEDS=y
-CONFIG_LEDS_CPU=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,115200 initrd=0x20800000,10M root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_BINFMT_MISC=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-# CONFIG_INET_DIAG is not set
-# CONFIG_IPV6 is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK_RO=y
-CONFIG_NETDEVICES=y
-CONFIG_PHYLIB=y
-CONFIG_DAVICOM_PHY=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_LEGACY_PTY_COUNT=32
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_I2C_GPIO=y
-# CONFIG_HWMON is not set
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_RTC_CLASS=y
-# CONFIG_RTC_HCTOSYS is not set
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT3_FS=y
-# CONFIG_EXT3_FS_XATTR is not set
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=m
-CONFIG_NFS_V3=y
-CONFIG_CRYPTO_MD5=y
-CONFIG_CRYPTO_DES=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_AUDIT=y
-CONFIG_IKCONFIG=y
-CONFIG_IKCONFIG_PROC=y
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_KALLSYMS_EXTRA_PASS=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_MODVERSIONS=y
-CONFIG_MODULE_SRCVERSION_ALL=y
-# CONFIG_BLK_DEV_BSG is not set
-# CONFIG_IOSCHED_DEADLINE is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_KB9200=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-CONFIG_NO_HZ=y
-CONFIG_HIGH_RES_TIMERS=y
-CONFIG_PREEMPT=y
-CONFIG_AEABI=y
-CONFIG_ZBOOT_ROM_TEXT=0x10000000
-CONFIG_ZBOOT_ROM_BSS=0x20040000
-CONFIG_CMDLINE="noinitrd root=/dev/mtdblock0 rootfstype=jffs2 mem=64M"
-CONFIG_KEXEC=y
-CONFIG_FPE_NWFPE=y
-CONFIG_BINFMT_MISC=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_DHCP=y
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
-# CONFIG_INET_XFRM_MODE_TUNNEL is not set
-# CONFIG_INET_XFRM_MODE_BEET is not set
-# CONFIG_INET_LRO is not set
-# CONFIG_INET_DIAG is not set
-# CONFIG_IPV6 is not set
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-# CONFIG_FIRMWARE_IN_KERNEL is not set
-CONFIG_MTD=y
-CONFIG_MTD_CONCAT=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_CFI_INTELEXT=y
-CONFIG_MTD_COMPLEX_MAPPINGS=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_NAND=y
-CONFIG_MTD_NAND_ATMEL=y
-CONFIG_MTD_UBI=y
-CONFIG_MTD_UBI_GLUEBI=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=16384
-CONFIG_ATMEL_TCLIB=y
-CONFIG_ATMEL_SSC=y
-CONFIG_SCSI=y
-CONFIG_BLK_DEV_SD=y
-CONFIG_CHR_DEV_SG=y
-CONFIG_SCSI_MULTI_LUN=y
-CONFIG_SCSI_CONSTANTS=y
-CONFIG_SCSI_LOGGING=y
-CONFIG_SCSI_SPI_ATTRS=m
-# CONFIG_SCSI_LOWLEVEL is not set
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_LEGACY_PTYS is not set
-# CONFIG_HW_RANDOM is not set
-# CONFIG_HWMON is not set
-CONFIG_WATCHDOG=y
-CONFIG_AT91RM9200_WATCHDOG=y
-CONFIG_FB=y
-CONFIG_FB_MODE_HELPERS=y
-CONFIG_FB_TILEBLITTING=y
-CONFIG_BACKLIGHT_LCD_SUPPORT=y
-# CONFIG_LCD_CLASS_DEVICE is not set
-CONFIG_BACKLIGHT_CLASS_DEVICE=y
-# CONFIG_BACKLIGHT_GENERIC is not set
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_FRAMEBUFFER_CONSOLE=y
-CONFIG_FONTS=y
-CONFIG_FONT_MINI_4x6=y
-# CONFIG_HID_SUPPORT is not set
-CONFIG_USB=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_LIBUSUAL=y
-CONFIG_MMC=y
-CONFIG_MMC_AT91=m
-CONFIG_RTC_CLASS=y
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_EXT3_FS=y
-# CONFIG_DNOTIFY is not set
-CONFIG_INOTIFY=y
-CONFIG_VFAT_FS=y
-CONFIG_TMPFS=y
-CONFIG_CONFIGFS_FS=y
-CONFIG_JFFS2_FS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_ROOT_NFS=y
-CONFIG_NLS_CODEPAGE_437=y
-CONFIG_NLS_UTF8=y
-CONFIG_MAGIC_SYSRQ=y
-CONFIG_DEBUG_FS=y
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_SCHED_DEBUG is not set
-# CONFIG_DEBUG_PREEMPT is not set
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-CONFIG_EMBEDDED=y
-CONFIG_SLAB=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_ONEARM=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_PCCARD=y
-CONFIG_AT91_CF=y
-CONFIG_LEDS=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="console=ttyS0,115200 root=/dev/nfs ip=bootp mem=64M"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_BOOTP=y
-CONFIG_IPV6=y
-# CONFIG_INET6_XFRM_MODE_TRANSPORT is not set
-# CONFIG_INET6_XFRM_MODE_TUNNEL is not set
-# CONFIG_INET6_XFRM_MODE_BEET is not set
-# CONFIG_IPV6_SIT is not set
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_JEDECPROBE=y
-CONFIG_MTD_CFI_AMDSTD=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_NBD=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-# CONFIG_VT is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_HW_RANDOM is not set
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=y
-# CONFIG_USB_HID is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_GADGET=y
-CONFIG_MMC=y
-CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
-CONFIG_TMPFS=y
-CONFIG_CRAMFS=y
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-CONFIG_NFS_V3_ACL=y
-CONFIG_ROOT_NFS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_LL=y
+++ /dev/null
-CONFIG_EXPERIMENTAL=y
-CONFIG_SYSVIPC=y
-CONFIG_IKCONFIG=m
-CONFIG_IKCONFIG_PROC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_EMBEDDED=y
-# CONFIG_KALLSYMS is not set
-CONFIG_SLAB=y
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_MACH_PICOTUX2XX=y
-CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
-CONFIG_AEABI=y
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_KEXEC=y
-CONFIG_FPE_NWFPE=y
-CONFIG_BINFMT_MISC=m
-CONFIG_NET=y
-CONFIG_PACKET=m
-CONFIG_UNIX=y
-CONFIG_XFRM_USER=m
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_BOOTP=y
-CONFIG_NET_IPIP=m
-CONFIG_NET_IPGRE=m
-CONFIG_INET_AH=m
-CONFIG_INET_ESP=m
-CONFIG_INET_IPCOMP=m
-CONFIG_INET_XFRM_MODE_TRANSPORT=m
-CONFIG_INET_XFRM_MODE_TUNNEL=m
-CONFIG_INET_XFRM_MODE_BEET=m
-CONFIG_INET_DIAG=m
-CONFIG_IPV6_PRIVACY=y
-CONFIG_IPV6_ROUTER_PREF=y
-CONFIG_IPV6_ROUTE_INFO=y
-CONFIG_INET6_AH=m
-CONFIG_INET6_ESP=m
-CONFIG_INET6_IPCOMP=m
-CONFIG_IPV6_MIP6=m
-CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m
-CONFIG_IPV6_TUNNEL=m
-CONFIG_BRIDGE=m
-CONFIG_VLAN_8021Q=m
-CONFIG_BT=m
-CONFIG_BT_L2CAP=m
-CONFIG_BT_SCO=m
-CONFIG_BT_RFCOMM=m
-CONFIG_BT_RFCOMM_TTY=y
-CONFIG_BT_BNEP=m
-CONFIG_BT_BNEP_MC_FILTER=y
-CONFIG_BT_BNEP_PROTO_FILTER=y
-CONFIG_BT_HIDP=m
-CONFIG_FW_LOADER=m
-CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_CFI_AMDSTD=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_BLK_DEV_LOOP=m
-CONFIG_EEPROM_LEGACY=m
-CONFIG_SCSI=m
-CONFIG_BLK_DEV_SD=m
-CONFIG_BLK_DEV_SR=m
-CONFIG_BLK_DEV_SR_VENDOR=y
-CONFIG_CHR_DEV_SG=m
-CONFIG_NETDEVICES=y
-CONFIG_TUN=m
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-CONFIG_USB_CATC=m
-CONFIG_USB_KAWETH=m
-CONFIG_USB_PEGASUS=m
-CONFIG_USB_RTL8150=m
-CONFIG_USB_USBNET=m
-CONFIG_USB_NET_DM9601=m
-CONFIG_USB_NET_GL620A=m
-CONFIG_USB_NET_PLUSB=m
-CONFIG_USB_NET_MCS7830=m
-CONFIG_USB_NET_RNDIS_HOST=m
-CONFIG_USB_ALI_M5632=y
-CONFIG_USB_AN2720=y
-CONFIG_USB_EPSON2888=y
-CONFIG_USB_KC2190=y
-CONFIG_PPP=m
-CONFIG_PPP_FILTER=y
-CONFIG_PPP_ASYNC=m
-CONFIG_PPP_DEFLATE=m
-CONFIG_PPP_BSDCOMP=m
-CONFIG_PPP_MPPE=m
-CONFIG_PPPOE=m
-CONFIG_SLIP=m
-CONFIG_SLIP_COMPRESSED=y
-CONFIG_SLIP_SMART=y
-CONFIG_SLIP_MODE_SLIP6=y
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_SERIO is not set
-# CONFIG_VT is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_LEGACY_PTYS is not set
-CONFIG_I2C=m
-CONFIG_I2C_CHARDEV=m
-CONFIG_I2C_GPIO=m
-CONFIG_HWMON=m
-CONFIG_SENSORS_ADM1021=m
-CONFIG_SENSORS_ADM1025=m
-CONFIG_SENSORS_ADM1026=m
-CONFIG_SENSORS_ADM1029=m
-CONFIG_SENSORS_ADM1031=m
-CONFIG_SENSORS_ADM9240=m
-CONFIG_SENSORS_DS1621=m
-CONFIG_SENSORS_GL518SM=m
-CONFIG_SENSORS_GL520SM=m
-CONFIG_SENSORS_IT87=m
-CONFIG_SENSORS_LM63=m
-CONFIG_SENSORS_LM75=m
-CONFIG_SENSORS_LM77=m
-CONFIG_SENSORS_LM78=m
-CONFIG_SENSORS_LM80=m
-CONFIG_SENSORS_LM83=m
-CONFIG_SENSORS_LM85=m
-CONFIG_SENSORS_LM87=m
-CONFIG_SENSORS_LM90=m
-CONFIG_SENSORS_LM92=m
-CONFIG_SENSORS_MAX1619=m
-CONFIG_SENSORS_PCF8591=m
-CONFIG_SENSORS_SMSC47B397=m
-CONFIG_SENSORS_W83781D=m
-CONFIG_SENSORS_W83791D=m
-CONFIG_SENSORS_W83792D=m
-CONFIG_SENSORS_W83793=m
-CONFIG_SENSORS_W83L785TS=m
-CONFIG_WATCHDOG=y
-CONFIG_WATCHDOG_NOWAYOUT=y
-CONFIG_AT91RM9200_WATCHDOG=m
-CONFIG_HID=m
-CONFIG_USB=m
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_OHCI_HCD=m
-CONFIG_USB_ACM=m
-CONFIG_USB_PRINTER=m
-CONFIG_USB_STORAGE=m
-CONFIG_USB_SERIAL=m
-CONFIG_USB_SERIAL_GENERIC=y
-CONFIG_USB_SERIAL_PL2303=m
-CONFIG_MMC=m
-CONFIG_MMC_AT91=m
-CONFIG_RTC_CLASS=m
-CONFIG_RTC_DRV_AT91RM9200=m
-CONFIG_EXT2_FS=m
-CONFIG_EXT3_FS=m
-# CONFIG_EXT3_FS_XATTR is not set
-CONFIG_INOTIFY=y
-CONFIG_ISO9660_FS=m
-CONFIG_JOLIET=y
-CONFIG_UDF_FS=m
-CONFIG_MSDOS_FS=m
-CONFIG_VFAT_FS=m
-CONFIG_NTFS_FS=m
-CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
-CONFIG_JFFS2_SUMMARY=y
-CONFIG_JFFS2_COMPRESSION_OPTIONS=y
-CONFIG_NFS_FS=m
-CONFIG_SMB_FS=m
-CONFIG_CIFS=m
-CONFIG_PARTITION_ADVANCED=y
-CONFIG_AMIGA_PARTITION=y
-CONFIG_NLS_DEFAULT="utf-8"
-CONFIG_NLS_CODEPAGE_437=m
-CONFIG_NLS_CODEPAGE_737=m
-CONFIG_NLS_CODEPAGE_775=m
-CONFIG_NLS_CODEPAGE_850=m
-CONFIG_NLS_CODEPAGE_852=m
-CONFIG_NLS_CODEPAGE_855=m
-CONFIG_NLS_CODEPAGE_857=m
-CONFIG_NLS_CODEPAGE_860=m
-CONFIG_NLS_CODEPAGE_861=m
-CONFIG_NLS_CODEPAGE_862=m
-CONFIG_NLS_CODEPAGE_863=m
-CONFIG_NLS_CODEPAGE_864=m
-CONFIG_NLS_CODEPAGE_865=m
-CONFIG_NLS_CODEPAGE_866=m
-CONFIG_NLS_CODEPAGE_869=m
-CONFIG_NLS_CODEPAGE_936=m
-CONFIG_NLS_CODEPAGE_950=m
-CONFIG_NLS_CODEPAGE_932=m
-CONFIG_NLS_CODEPAGE_949=m
-CONFIG_NLS_CODEPAGE_874=m
-CONFIG_NLS_ISO8859_8=m
-CONFIG_NLS_CODEPAGE_1250=m
-CONFIG_NLS_CODEPAGE_1251=m
-CONFIG_NLS_ASCII=m
-CONFIG_NLS_ISO8859_1=m
-CONFIG_NLS_ISO8859_2=m
-CONFIG_NLS_ISO8859_3=m
-CONFIG_NLS_ISO8859_4=m
-CONFIG_NLS_ISO8859_5=m
-CONFIG_NLS_ISO8859_6=m
-CONFIG_NLS_ISO8859_7=m
-CONFIG_NLS_ISO8859_9=m
-CONFIG_NLS_ISO8859_13=m
-CONFIG_NLS_ISO8859_14=m
-CONFIG_NLS_ISO8859_15=m
-CONFIG_NLS_KOI8_R=m
-CONFIG_NLS_KOI8_U=m
-CONFIG_NLS_UTF8=m
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_DEBUG_BUGVERBOSE is not set
-CONFIG_DEBUG_LL=y
-CONFIG_CRYPTO_NULL=m
-CONFIG_CRYPTO_TEST=m
-CONFIG_CRYPTO_LRW=m
-CONFIG_CRYPTO_PCBC=m
-CONFIG_CRYPTO_XCBC=m
-CONFIG_CRYPTO_MD4=m
-CONFIG_CRYPTO_MICHAEL_MIC=m
-CONFIG_CRYPTO_SHA256=m
-CONFIG_CRYPTO_SHA512=m
-CONFIG_CRYPTO_TGR192=m
-CONFIG_CRYPTO_WP512=m
-CONFIG_CRYPTO_ANUBIS=m
-CONFIG_CRYPTO_BLOWFISH=m
-CONFIG_CRYPTO_CAMELLIA=m
-CONFIG_CRYPTO_CAST5=m
-CONFIG_CRYPTO_CAST6=m
-CONFIG_CRYPTO_FCRYPT=m
-CONFIG_CRYPTO_KHAZAD=m
-CONFIG_CRYPTO_SERPENT=m
-CONFIG_CRYPTO_TEA=m
-CONFIG_CRYPTO_TWOFISH=m
-CONFIG_LIBCRC32C=m
+++ /dev/null
-# CONFIG_SWAP is not set
-CONFIG_SYSVIPC=y
-CONFIG_LOG_BUF_SHIFT=14
-CONFIG_BLK_DEV_INITRD=y
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-CONFIG_ARCH_AT91=y
-CONFIG_ARCH_AT91RM9200DK=y
-CONFIG_MACH_YL9200=y
-# CONFIG_ARM_THUMB is not set
-CONFIG_ZBOOT_ROM_TEXT=0x0
-CONFIG_ZBOOT_ROM_BSS=0x0
-CONFIG_CMDLINE="mem=32M console=ttyS0,115200 initrd=0x20410000,3145728 root=/dev/ram0 rw"
-CONFIG_FPE_NWFPE=y
-CONFIG_NET=y
-CONFIG_PACKET=y
-CONFIG_UNIX=y
-CONFIG_INET=y
-CONFIG_IP_PNP=y
-CONFIG_IP_PNP_DHCP=y
-# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
-# CONFIG_INET_XFRM_MODE_TUNNEL is not set
-# CONFIG_INET_XFRM_MODE_BEET is not set
-# CONFIG_INET_LRO is not set
-# CONFIG_INET_DIAG is not set
-# CONFIG_IPV6 is not set
-CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
-CONFIG_MTD=y
-CONFIG_MTD_CONCAT=y
-CONFIG_MTD_PARTITIONS=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_CHAR=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_CFI=y
-CONFIG_MTD_JEDECPROBE=y
-CONFIG_MTD_CFI_INTELEXT=y
-CONFIG_MTD_COMPLEX_MAPPINGS=y
-CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_PLATRAM=y
-CONFIG_MTD_NAND=y
-CONFIG_MTD_NAND_ATMEL=y
-CONFIG_MTD_NAND_PLATFORM=y
-CONFIG_BLK_DEV_LOOP=y
-CONFIG_BLK_DEV_RAM=y
-CONFIG_BLK_DEV_RAM_COUNT=3
-CONFIG_BLK_DEV_RAM_SIZE=8192
-# CONFIG_MISC_DEVICES is not set
-CONFIG_BLK_DEV_SD=y
-CONFIG_ATA=y
-CONFIG_NETDEVICES=y
-CONFIG_PHYLIB=y
-CONFIG_DAVICOM_PHY=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
-# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
-CONFIG_INPUT_MOUSEDEV_SCREEN_X=640
-CONFIG_INPUT_MOUSEDEV_SCREEN_Y=480
-CONFIG_INPUT_EVDEV=y
-# CONFIG_KEYBOARD_ATKBD is not set
-CONFIG_KEYBOARD_GPIO=y
-CONFIG_INPUT_TOUCHSCREEN=y
-CONFIG_TOUCHSCREEN_ADS7846=y
-# CONFIG_SERIO_SERPORT is not set
-CONFIG_SERIAL_ATMEL=y
-CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_HW_RANDOM is not set
-CONFIG_I2C=y
-CONFIG_SPI=y
-CONFIG_SPI_DEBUG=y
-CONFIG_SPI_ATMEL=y
-CONFIG_FB=y
-CONFIG_BACKLIGHT_LCD_SUPPORT=y
-CONFIG_LCD_CLASS_DEVICE=y
-CONFIG_BACKLIGHT_CLASS_DEVICE=y
-CONFIG_DISPLAY_SUPPORT=y
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_LOGO=y
-# CONFIG_LOGO_LINUX_MONO is not set
-# CONFIG_LOGO_LINUX_VGA16 is not set
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_DEVICEFS=y
-# CONFIG_USB_DEVICE_CLASS is not set
-CONFIG_USB_MON=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_GADGET=y
-CONFIG_USB_GADGET_M66592=y
-CONFIG_USB_FILE_STORAGE=m
-CONFIG_MMC=y
-CONFIG_MMC_DEBUG=y
-# CONFIG_MMC_BLOCK_BOUNCE is not set
-CONFIG_MMC_AT91=m
-CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=y
-CONFIG_LEDS_GPIO=y
-CONFIG_LEDS_TRIGGERS=y
-CONFIG_LEDS_TRIGGER_TIMER=y
-CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_RTC_CLASS=y
-CONFIG_RTC_DRV_AT91RM9200=y
-CONFIG_EXT2_FS=y
-CONFIG_EXT2_FS_XATTR=y
-CONFIG_EXT3_FS=y
-CONFIG_REISERFS_FS=y
-CONFIG_INOTIFY=y
-CONFIG_ISO9660_FS=y
-CONFIG_JOLIET=y
-CONFIG_ZISOFS=y
-CONFIG_UDF_FS=y
-CONFIG_MSDOS_FS=y
-CONFIG_VFAT_FS=y
-CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
-CONFIG_JFFS2_FS_DEBUG=1
-CONFIG_JFFS2_COMPRESSION_OPTIONS=y
-CONFIG_JFFS2_RUBIN=y
-CONFIG_CRAMFS=y
-CONFIG_PARTITION_ADVANCED=y
-CONFIG_MAC_PARTITION=y
-CONFIG_NLS_CODEPAGE_437=y
-CONFIG_NLS_ISO8859_1=y
-# CONFIG_ENABLE_MUST_CHECK is not set
-CONFIG_DEBUG_FS=y
-CONFIG_DEBUG_KERNEL=y
-CONFIG_SLUB_DEBUG_ON=y
-CONFIG_DEBUG_KOBJECT=y
-CONFIG_DEBUG_INFO=y
-CONFIG_DEBUG_LIST=y
-CONFIG_DEBUG_USER=y
-CONFIG_DEBUG_ERRORS=y
-CONFIG_DEBUG_LL=y
@ Slightly optimised to avoid incrementing the pointer twice
usraccoff \instr, \reg, \ptr, \inc, 0, \cond, \abort
.if \rept == 2
- usraccoff \instr, \reg, \ptr, \inc, 4, \cond, \abort
+ usraccoff \instr, \reg, \ptr, \inc, \inc, \cond, \abort
.endif
add\cond \ptr, #\rept * \inc
#ifdef CONFIG_CPU_HAS_ASID
#define ASID(mm) ((mm)->context.id & 255)
+
+/* init_mm.context.id_lock should be initialized. */
+#define INIT_MM_CONTEXT(name) \
+ .context.id_lock = __SPIN_LOCK_UNLOCKED(name.context.id_lock),
#else
#define ASID(mm) (0)
#endif
#define pmd_page(pmd) pfn_to_page(__phys_to_pfn(pmd_val(pmd)))
+/* we don't need complex calculations here as the pmd is folded into the pgd */
+#define pmd_addr_end(addr,end) (end)
+
/*
* Conversion functions: convert a page and protection to a page entry,
* and a page entry and page directory to the page they refer to.
*/
.L_found:
#if __LINUX_ARM_ARCH__ >= 5
- rsb r1, r3, #0
- and r3, r3, r1
+ rsb r0, r3, #0
+ and r3, r3, r0
clz r3, r3
rsb r3, r3, #31
add r0, r2, r3
addeq r2, r2, #1
mov r0, r2
#endif
+ cmp r1, r0 @ Clamp to maxbit
+ movlo r0, r1
mov pc, lr
#ifndef __ASM_ARCH_VMALLOC_H
#define __ASM_ARCH_VMALLOC_H
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
#endif /* __ASM_ARCH_VMALLOC_H */
# AT91RM9200 board-specific support
obj-$(CONFIG_MACH_ONEARM) += board-1arm.o
-obj-$(CONFIG_ARCH_AT91RM9200DK) += board-dk.o
-obj-$(CONFIG_MACH_AT91RM9200EK) += board-ek.o
+obj-$(CONFIG_ARCH_AT91RM9200DK) += board-rm9200dk.o
+obj-$(CONFIG_MACH_AT91RM9200EK) += board-rm9200ek.o
obj-$(CONFIG_MACH_CSB337) += board-csb337.o
obj-$(CONFIG_MACH_CSB637) += board-csb637.o
obj-$(CONFIG_MACH_CARMEVA) += board-carmeva.o
static struct platform_device *__initdata at91_uarts[ATMEL_MAX_UART]; /* the UARTs to use */
struct platform_device *atmel_default_console_device; /* the serial console device */
-void __init __deprecated at91_init_serial(struct at91_uart_config *config)
-{
- int i;
-
- /* Fill in list of supported UARTs */
- for (i = 0; i < config->nr_tty; i++) {
- switch (config->tty_map[i]) {
- case 0:
- configure_usart0_pins(ATMEL_UART_CTS | ATMEL_UART_RTS);
- at91_uarts[i] = &at91rm9200_uart0_device;
- at91_clock_associate("usart0_clk", &at91rm9200_uart0_device.dev, "usart");
- break;
- case 1:
- configure_usart1_pins(ATMEL_UART_CTS | ATMEL_UART_RTS | ATMEL_UART_DSR | ATMEL_UART_DTR | ATMEL_UART_DCD | ATMEL_UART_RI);
- at91_uarts[i] = &at91rm9200_uart1_device;
- at91_clock_associate("usart1_clk", &at91rm9200_uart1_device.dev, "usart");
- break;
- case 2:
- configure_usart2_pins(0);
- at91_uarts[i] = &at91rm9200_uart2_device;
- at91_clock_associate("usart2_clk", &at91rm9200_uart2_device.dev, "usart");
- break;
- case 3:
- configure_usart3_pins(0);
- at91_uarts[i] = &at91rm9200_uart3_device;
- at91_clock_associate("usart3_clk", &at91rm9200_uart3_device.dev, "usart");
- break;
- case 4:
- configure_dbgu_pins();
- at91_uarts[i] = &at91rm9200_dbgu_device;
- at91_clock_associate("mck", &at91rm9200_dbgu_device.dev, "usart");
- break;
- default:
- continue;
- }
- at91_uarts[i]->id = i; /* update ID number to mapped ID */
- }
-
- /* Set serial console device */
- if (config->console_tty < ATMEL_MAX_UART)
- atmel_default_console_device = at91_uarts[config->console_tty];
- if (!atmel_default_console_device)
- printk(KERN_INFO "AT91: No default serial console defined.\n");
-}
-
void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins)
{
struct platform_device *pdev;
#include "generic.h"
-/*
- * Serial port configuration.
- * 0 .. 3 = USART0 .. USART3
- * 4 = DBGU
- */
-static struct at91_uart_config __initdata onearm_uart_config = {
- .console_tty = 0, /* ttyS0 */
- .nr_tty = 3,
- .tty_map = { 4, 0, 1, -1, -1 }, /* ttyS0, ..., ttyS4 */
-};
-
static void __init onearm_map_io(void)
{
/* Initialize processor: 18.432 MHz crystal */
at91rm9200_initialize(18432000, AT91RM9200_PQFP);
- /* Setup the serial ports and console */
- at91_init_serial(&onearm_uart_config);
+ /* DBGU on ttyS0. (Rx & Tx only) */
+ at91_register_uart(0, 0, 0);
+
+ /* USART0 on ttyS1 (Rx, Tx, CTS, RTS) */
+ at91_register_uart(AT91RM9200_ID_US0, 1, ATMEL_UART_CTS | ATMEL_UART_RTS);
+
+ /* USART1 on ttyS2 (Rx, Tx, CTS, RTS, DTR, DSR, DCD, RI) */
+ at91_register_uart(AT91RM9200_ID_US1, 2, ATMEL_UART_CTS | ATMEL_UART_RTS
+ | ATMEL_UART_DTR | ATMEL_UART_DSR | ATMEL_UART_DCD
+ | ATMEL_UART_RI);
+
+ /* set serial console to ttyS0 (ie, DBGU) */
+ at91_set_serial_console(0);
}
static void __init onearm_init_irq(void)
#include "generic.h"
-/*
- * Serial port configuration.
- * 0 .. 3 = USART0 .. USART3
- * 4 = DBGU
- */
-static struct at91_uart_config __initdata kafa_uart_config = {
- .console_tty = 0, /* ttyS0 */
- .nr_tty = 2,
- .tty_map = { 4, 0, -1, -1, -1 } /* ttyS0, ..., ttyS4 */
-};
-
static void __init kafa_map_io(void)
{
/* Initialize processor: 18.432 MHz crystal */
/* Set up the LEDs */
at91_init_leds(AT91_PIN_PB4, AT91_PIN_PB4);
- /* Setup the serial ports and console */
- at91_init_serial(&kafa_uart_config);
+ /* DBGU on ttyS0. (Rx & Tx only) */
+ at91_register_uart(0, 0, 0);
+
+ /* USART0 on ttyS1 (Rx, Tx, CTS, RTS) */
+ at91_register_uart(AT91RM9200_ID_US0, 1, ATMEL_UART_CTS | ATMEL_UART_RTS);
+
+ /* set serial console to ttyS0 (ie, DBGU) */
+ at91_set_serial_console(0);
}
static void __init kafa_init_irq(void)
#include "generic.h"
-/*
- * Serial port configuration.
- * 0 .. 3 = USART0 .. USART3
- * 4 = DBGU
- */
-static struct at91_uart_config __initdata picotux200_uart_config = {
- .console_tty = 0, /* ttyS0 */
- .nr_tty = 2,
- .tty_map = { 4, 1, -1, -1, -1 } /* ttyS0, ..., ttyS4 */
-};
-
static void __init picotux200_map_io(void)
{
/* Initialize processor: 18.432 MHz crystal */
at91rm9200_initialize(18432000, AT91RM9200_BGA);
- /* Setup the serial ports and console */
- at91_init_serial(&picotux200_uart_config);
+ /* DBGU on ttyS0. (Rx & Tx only) */
+ at91_register_uart(0, 0, 0);
+
+ /* USART1 on ttyS1. (Rx, Tx, CTS, RTS, DTR, DSR, DCD, RI) */
+ at91_register_uart(AT91RM9200_ID_US1, 1, ATMEL_UART_CTS | ATMEL_UART_RTS
+ | ATMEL_UART_DTR | ATMEL_UART_DSR | ATMEL_UART_DCD
+ | ATMEL_UART_RI);
+
+ /* set serial console to ttyS0 (ie, DBGU) */
+ at91_set_serial_console(0);
}
static void __init picotux200_init_irq(void)
.ports = 1,
};
-// static struct at91_udc_data __initdata picotux200_udc_data = {
-// .vbus_pin = AT91_PIN_PD4,
-// .pullup_pin = AT91_PIN_PD5,
-// };
-
static struct at91_mmc_data __initdata picotux200_mmc_data = {
.det_pin = AT91_PIN_PB27,
.slot_b = 0,
.wp_pin = AT91_PIN_PA17,
};
-// static struct spi_board_info picotux200_spi_devices[] = {
-// { /* DataFlash chip */
-// .modalias = "mtd_dataflash",
-// .chip_select = 0,
-// .max_speed_hz = 15 * 1000 * 1000,
-// },
-// #ifdef CONFIG_MTD_AT91_DATAFLASH_CARD
-// { /* DataFlash card */
-// .modalias = "mtd_dataflash",
-// .chip_select = 3,
-// .max_speed_hz = 15 * 1000 * 1000,
-// },
-// #endif
-// };
-
#define PICOTUX200_FLASH_BASE AT91_CHIPSELECT_0
#define PICOTUX200_FLASH_SIZE SZ_4M
at91_add_device_eth(&picotux200_eth_data);
/* USB Host */
at91_add_device_usbh(&picotux200_usbh_data);
- /* USB Device */
- // at91_add_device_udc(&picotux200_udc_data);
- // at91_set_multi_drive(picotux200_udc_data.pullup_pin, 1); /* pullup_pin is connected to reset */
/* I2C */
at91_add_device_i2c(NULL, 0);
- /* SPI */
- // at91_add_device_spi(picotux200_spi_devices, ARRAY_SIZE(picotux200_spi_devices));
-#ifdef CONFIG_MTD_AT91_DATAFLASH_CARD
- /* DataFlash card */
- at91_set_gpio_output(AT91_PIN_PB22, 0);
-#else
/* MMC */
at91_set_gpio_output(AT91_PIN_PB22, 1); /* this MMC card slot can optionally use SPI signaling (CS3). */
at91_add_device_mmc(0, &picotux200_mmc_data);
-#endif
/* NOR Flash */
platform_device_register(&picotux200_flash);
}
/*
- * linux/arch/arm/mach-at91/board-dk.c
+ * linux/arch/arm/mach-at91/board-rm9200dk.c
*
* Copyright (C) 2005 SAN People
*
// .vcc_pin = ... always powered
};
+#ifndef CONFIG_MTD_AT91_DATAFLASH_CARD
static struct at91_mmc_data __initdata dk_mmc_data = {
.slot_b = 0,
.wire4 = 1,
};
+#endif
static struct spi_board_info dk_spi_devices[] = {
{ /* DataFlash chip */
/*
- * linux/arch/arm/mach-at91/board-ek.c
+ * linux/arch/arm/mach-at91/board-rm9200ek.c
*
* Copyright (C) 2005 SAN People
*
.pullup_pin = AT91_PIN_PD5,
};
+#ifndef CONFIG_MTD_AT91_DATAFLASH_CARD
static struct at91_mmc_data __initdata ek_mmc_data = {
.det_pin = AT91_PIN_PB27,
.slot_b = 0,
.wire4 = 1,
.wp_pin = AT91_PIN_PA17,
};
+#endif
static struct spi_board_info ek_spi_devices[] = {
{ /* DataFlash chip */
* EPSON S1D13806 FB (discontinued chip)
* EPSON S1D13506 FB
*/
-#if defined(CONFIG_FB_S1D135XX) || defined(CONFIG_FB_S1D13XXX_MODULE)
+#if defined(CONFIG_FB_S1D13XXX) || defined(CONFIG_FB_S1D13XXX_MODULE)
#include <video/s1d13xxxfb.h>
extern void __init at91_register_uart(unsigned id, unsigned portnr, unsigned pins);
extern void __init at91_set_serial_console(unsigned portnr);
-struct at91_uart_config {
- unsigned short console_tty; /* tty number of serial console */
- unsigned short nr_tty; /* number of serial tty's */
- short tty_map[]; /* map UART to tty number */
-};
extern struct platform_device *atmel_default_console_device;
-extern void __init __deprecated at91_init_serial(struct at91_uart_config *config);
struct atmel_uart_data {
short use_dma_tx; /* use transmit DMA? */
* 0xe0000000 to 0xefffffff. This gives us 256 MB of vm space and handles
* larger physical memory designs better.
*/
-#define VMALLOC_END 0xf0000000
+#define VMALLOC_END 0xf0000000UL
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
CLK(NULL, "uart1", &uart1_clk),
CLK(NULL, "uart2", &uart2_clk),
CLK("i2c_davinci.1", NULL, &i2c_clk),
- CLK("davinci-asp.0", NULL, &asp0_clk),
- CLK("davinci-asp.1", NULL, &asp1_clk),
+ CLK("davinci-mcbsp.0", NULL, &asp0_clk),
+ CLK("davinci-mcbsp.1", NULL, &asp1_clk),
CLK("davinci_mmc.0", NULL, &mmcsd0_clk),
CLK("davinci_mmc.1", NULL, &mmcsd1_clk),
CLK("spi_davinci.0", NULL, &spi0_clk),
};
static struct platform_device dm355_asp1_device = {
- .name = "davinci-asp",
+ .name = "davinci-mcbsp",
.id = 1,
.num_resources = ARRAY_SIZE(dm355_asp1_resources),
.resource = dm355_asp1_resources,
CLK(NULL, "usb", &usb_clk),
CLK("davinci_emac.1", NULL, &emac_clk),
CLK("davinci_voicecodec", NULL, &voicecodec_clk),
- CLK("davinci-asp.0", NULL, &asp0_clk),
+ CLK("davinci-mcbsp", NULL, &asp0_clk),
CLK(NULL, "rto", &rto_clk),
CLK(NULL, "mjcp", &mjcp_clk),
CLK(NULL, NULL, NULL),
};
static struct platform_device dm365_asp_device = {
- .name = "davinci-asp",
- .id = 0,
+ .name = "davinci-mcbsp",
+ .id = -1,
.num_resources = ARRAY_SIZE(dm365_asp_resources),
.resource = dm365_asp_resources,
};
CLK("davinci_emac.1", NULL, &emac_clk),
CLK("i2c_davinci.1", NULL, &i2c_clk),
CLK("palm_bk3710", NULL, &ide_clk),
- CLK("davinci-asp", NULL, &asp_clk),
+ CLK("davinci-mcbsp", NULL, &asp_clk),
CLK("davinci_mmc.0", NULL, &mmcsd_clk),
CLK(NULL, "spi", &spi_clk),
CLK(NULL, "gpio", &gpio_clk),
};
static struct platform_device dm644x_asp_device = {
- .name = "davinci-asp",
+ .name = "davinci-mcbsp",
.id = -1,
.num_resources = ARRAY_SIZE(dm644x_asp_resources),
.resource = dm644x_asp_resources,
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
-#define VMALLOC_END 0xdf000000
+#define VMALLOC_END 0xdf000000UL
*/
-#define VMALLOC_END 0xf0000000
+#define VMALLOC_END 0xf0000000UL
#ifndef __ARCH_ARM_VMALLOC_H
#define __ARCH_ARM_VMALLOC_H
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
#endif
.flags = IMXUART_HAVE_RTSCTS,
};
-#if defined(CONFIG_TOUCHSCREEN_ADS7846) \
- || defined(CONFIG_TOUCHSCREEN_ADS7846_MODULE)
-
#define ADS7846_PENDOWN (GPIO_PORTD | 25)
static void ads7846_dev_init(void)
.get_pendown_state = ads7846_get_pendown_state,
.keep_vref_on = 1,
};
-#endif
-#if defined(CONFIG_SPI_IMX) || defined(CONFIG_SPI_IMX_MODULE)
static struct spi_board_info eukrea_mbimx27_spi_board_info[] __initdata = {
[0] = {
.modalias = "ads7846",
.chipselect = eukrea_mbimx27_spi_cs,
.num_chipselect = ARRAY_SIZE(eukrea_mbimx27_spi_cs),
};
-#endif
static struct i2c_board_info eukrea_mbimx27_i2c_devices[] = {
{
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
#ifndef __ASM_ARCH_MSM_VMALLOC_H
#define __ASM_ARCH_MSM_VMALLOC_H
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
#endif
#define imx25_add_mxc_nand(pdata) \
imx_add_mxc_nand(&imx25_mxc_nand_data, pdata)
-extern const struct imx_spi_imx_data imx25_spi_imx_data[] __initconst;
+extern const struct imx_spi_imx_data imx25_cspi_data[] __initconst;
#define imx25_add_spi_imx(id, pdata) \
- imx_add_spi_imx(&imx25_spi_imx_data[id], pdata)
+ imx_add_spi_imx(&imx25_cspi_data[id], pdata)
#define imx25_add_spi_imx0(pdata) imx25_add_spi_imx(0, pdata)
#define imx25_add_spi_imx1(pdata) imx25_add_spi_imx(1, pdata)
#define imx25_add_spi_imx2(pdata) imx25_add_spi_imx(2, pdata)
#include <mach/common.h>
#include <mach/iomux-mx3.h>
+#include <mach/spi.h>
#include <asm/mach-types.h>
};
/* Platform Data for MXC CSPI */
-#if defined(CONFIG_SPI_IMX) || defined(CONFIG_SPI_IMX_MODULE)
static int pcm037_spi1_cs[] = {MXC_SPI_CS(1), IOMUX_TO_GPIO(MX31_PIN_KEY_COL7)};
static const struct spi_imx_master pcm037_spi1_pdata __initconst = {
.chipselect = pcm037_spi1_cs,
.num_chipselect = ARRAY_SIZE(pcm037_spi1_cs),
};
-#endif
/* GPIO-keys input device */
static struct gpio_keys_button pcm037_gpio_keys[] = {
},
};
-static int eet_init_devices(void)
+static int __init eet_init_devices(void)
{
if (!machine_is_pcm037() || pcm037_variant() != PCM037_EET)
return 0;
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xd8000000
+#define VMALLOC_END 0xd8000000UL
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xf8000000
+#define VMALLOC_END 0xf8000000UL
#include <linux/irq.h>
#include <linux/time.h>
#include <linux/gpio.h>
+#include <linux/console.h>
#include <asm/mach/time.h>
#include <asm/mach/irq.h>
if (omap_irq_pending())
goto no_sleep;
+ /* Block console output in case it is on one of the OMAP UARTs */
+ if (try_acquire_console_sem())
+ goto no_sleep;
+
omap_uart_prepare_idle(0);
omap_uart_prepare_idle(1);
omap_uart_prepare_idle(2);
omap_uart_resume_idle(1);
omap_uart_resume_idle(0);
+ release_console_sem();
+
no_sleep:
if (omap2_pm_debug) {
unsigned long long tmp;
#include <linux/clk.h>
#include <linux/delay.h>
#include <linux/slab.h>
+#include <linux/console.h>
#include <plat/sram.h>
#include <plat/clockdomain.h>
omap3_enable_io_chain();
}
+ /* Block console output in case it is on one of the OMAP UARTs */
+ if (per_next_state < PWRDM_POWER_ON ||
+ core_next_state < PWRDM_POWER_ON)
+ if (try_acquire_console_sem())
+ goto console_still_active;
+
/* PER */
if (per_next_state < PWRDM_POWER_ON) {
omap_uart_prepare_idle(2);
omap_uart_resume_idle(3);
}
+ release_console_sem();
+
+console_still_active:
/* Disable IO-PAD and IO-CHAIN wakeup */
if (omap3_has_io_wakeup() &&
(per_next_state < PWRDM_POWER_ON ||
#include <linux/slab.h>
#include <linux/serial_8250.h>
#include <linux/pm_runtime.h>
+#include <linux/console.h>
#ifdef CONFIG_SERIAL_OMAP
#include <plat/omap-serial.h>
struct omap_uart_state *uart;
list_for_each_entry(uart, &uart_list, node) {
- if (num == uart->num) {
+ if (num == uart->num && uart->can_sleep) {
omap_uart_enable_clocks(uart);
/* Check for IO pad wakeup */
oh->dev_attr = uart;
+ acquire_console_sem(); /* in case the earlycon is on the UART */
+
/*
* Because of early UART probing, UART did not get idled
* on init. Now that omap_device is ready, ensure full idle
omap_uart_block_sleep(uart);
uart->timeout = DEFAULT_TIMEOUT;
+ release_console_sem();
+
if ((cpu_is_omap34xx() && uart->padconf) ||
(uart->wk_en && uart->wk_mask)) {
device_init_wakeup(&od->pdev.dev, true);
* The vmalloc() routines leaves a hole of 4kB between each vmalloced
* area for the same reason. ;)
*/
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
-#define VMALLOC_END 0xdc000000
+#define VMALLOC_END 0xdc000000UL
/* Configures BT serial port GPIOs */
s3c_gpio_cfgpin(S3C2410_GPH(0), S3C2410_GPH0_nCTS0);
- s3c_gpio_cfgpull(S3C2410_GPH(0), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPH(0), S3C_GPIO_PULL_NONE);
s3c_gpio_cfgpin(S3C2410_GPH(1), S3C2410_GPIO_OUTPUT);
- s3c_gpio_cfgpull(S3C2410_GPH(1), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPH(1), S3C_GPIO_PULL_NONE);
s3c_gpio_cfgpin(S3C2410_GPH(2), S3C2410_GPH2_TXD0);
- s3c_gpio_cfgpull(S3C2410_GPH(2), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPH(2), S3C_GPIO_PULL_NONE);
s3c_gpio_cfgpin(S3C2410_GPH(3), S3C2410_GPH3_RXD0);
- s3c_gpio_cfgpull(S3C2410_GPH(3), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPH(3), S3C_GPIO_PULL_NONE);
rfk = rfkill_alloc(DRV_NAME, &pdev->dev, RFKILL_TYPE_BLUETOOTH,
static void s3c2416_irq_demux_uart3(unsigned int irq, struct irq_desc *desc)
{
- s3c2416_irq_demux(IRQ_S3C2443_UART3, 3);
+ s3c2416_irq_demux(IRQ_S3C2443_RX3, 3);
}
#define INTMSK_UART3 (1UL << (IRQ_S3C2443_UART3 - IRQ_EINT0))
-#define SUBMSK_UART3 (0xf << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
-
+#define SUBMSK_UART3 (0x7 << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
static void s3c2416_irq_uart3_mask(unsigned int irqno)
{
static void s3c2443_irq_demux_uart3(unsigned int irq, struct irq_desc *desc)
{
- s3c2443_irq_demux(IRQ_S3C2443_UART3, 3);
+ s3c2443_irq_demux(IRQ_S3C2443_RX3, 3);
}
#define INTMSK_UART3 (1UL << (IRQ_S3C2443_UART3 - IRQ_EINT0))
-#define SUBMSK_UART3 (0xf << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
-
+#define SUBMSK_UART3 (0x7 << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
static void s3c2443_irq_uart3_mask(unsigned int irqno)
{
#include <video/platform_lcd.h>
-#define UCON (S3C2410_UCON_DEFAULT | S3C2410_UCON_UCLK)
+#define UCON S3C2410_UCON_DEFAULT
#define ULCON (S3C2410_LCON_CS8 | S3C2410_LCON_PNONE | S3C2410_LCON_STOPB)
#define UFCON (S3C2410_UFCON_RXTRIG8 | S3C2410_UFCON_FIFOMODE)
#include <video/platform_lcd.h>
-#define UCON (S3C2410_UCON_DEFAULT | S3C2410_UCON_UCLK)
+#define UCON S3C2410_UCON_DEFAULT
#define ULCON (S3C2410_LCON_CS8 | S3C2410_LCON_PNONE | S3C2410_LCON_STOPB)
#define UFCON (S3C2410_UFCON_RXTRIG8 | S3C2410_UFCON_FIFOMODE)
#include <linux/init.h>
#include <linux/serial_core.h>
#include <linux/i2c.h>
+#include <linux/sysdev.h>
#include <asm/mach/arch.h>
#include <asm/mach/map.h>
#include <linux/i2c.h>
#include <linux/init.h>
#include <linux/serial_core.h>
+#include <linux/sysdev.h>
#include <asm/mach/arch.h>
#include <asm/mach/map.h>
/*
* arch/arm/mach-shark/include/mach/vmalloc.h
*/
-#define VMALLOC_END 0xd0000000
+#define VMALLOC_END 0xd0000000UL
/* FSI */
#define IRQ_FSI evt2irq(0x1840)
+static int __fsi_set_rate(struct clk *clk, long rate, int enable)
+{
+ int ret = 0;
+
+ if (rate <= 0)
+ return ret;
+
+ if (enable) {
+ ret = clk_set_rate(clk, rate);
+ if (0 == ret)
+ ret = clk_enable(clk);
+ } else {
+ clk_disable(clk);
+ }
+
+ return ret;
+}
+
+static int __fsi_set_round_rate(struct clk *clk, long rate, int enable)
+{
+ return __fsi_set_rate(clk, clk_round_rate(clk, rate), enable);
+}
-static int fsi_set_rate(int is_porta, int rate)
+static int fsi_ak4642_set_rate(struct device *dev, int rate, int enable)
+{
+ struct clk *fsia_ick;
+ struct clk *fsiack;
+ int ret = -EIO;
+
+ fsia_ick = clk_get(dev, "icka");
+ if (IS_ERR(fsia_ick))
+ return PTR_ERR(fsia_ick);
+
+ /*
+ * FSIACK is connected to AK4642,
+ * and use external clock pin from it.
+ * it is parent of fsia_ick now.
+ */
+ fsiack = clk_get_parent(fsia_ick);
+ if (!fsiack)
+ goto fsia_ick_out;
+
+ /*
+ * we get 1/1 divided clock by setting same rate to fsiack and fsia_ick
+ *
+ ** FIXME **
+ * Because the freq_table of external clk (fsiack) are all 0,
+ * the return value of clk_round_rate became 0.
+ * So, it use __fsi_set_rate here.
+ */
+ ret = __fsi_set_rate(fsiack, rate, enable);
+ if (ret < 0)
+ goto fsiack_out;
+
+ ret = __fsi_set_round_rate(fsia_ick, rate, enable);
+ if ((ret < 0) && enable)
+ __fsi_set_round_rate(fsiack, rate, 0); /* disable FSI ACK */
+
+fsiack_out:
+ clk_put(fsiack);
+
+fsia_ick_out:
+ clk_put(fsia_ick);
+
+ return 0;
+}
+
+static int fsi_hdmi_set_rate(struct device *dev, int rate, int enable)
{
struct clk *fsib_clk;
struct clk *fdiv_clk = &sh7372_fsidivb_clk;
+ long fsib_rate = 0;
+ long fdiv_rate = 0;
+ int ackmd_bpfmd;
int ret;
- /* set_rate is not needed if port A */
- if (is_porta)
- return 0;
-
- fsib_clk = clk_get(NULL, "fsib_clk");
- if (IS_ERR(fsib_clk))
- return -EINVAL;
-
switch (rate) {
case 44100:
- clk_set_rate(fsib_clk, clk_round_rate(fsib_clk, 11283000));
- ret = SH_FSI_ACKMD_256 | SH_FSI_BPFMD_64;
+ fsib_rate = rate * 256;
+ ackmd_bpfmd = SH_FSI_ACKMD_256 | SH_FSI_BPFMD_64;
break;
case 48000:
- clk_set_rate(fsib_clk, clk_round_rate(fsib_clk, 85428000));
- clk_set_rate(fdiv_clk, clk_round_rate(fdiv_clk, 12204000));
- ret = SH_FSI_ACKMD_256 | SH_FSI_BPFMD_64;
+ fsib_rate = 85428000; /* around 48kHz x 256 x 7 */
+ fdiv_rate = rate * 256;
+ ackmd_bpfmd = SH_FSI_ACKMD_256 | SH_FSI_BPFMD_64;
break;
default:
pr_err("unsupported rate in FSI2 port B\n");
- ret = -EINVAL;
- break;
+ return -EINVAL;
}
+ /* FSI B setting */
+ fsib_clk = clk_get(dev, "ickb");
+ if (IS_ERR(fsib_clk))
+ return -EIO;
+
+ ret = __fsi_set_round_rate(fsib_clk, fsib_rate, enable);
clk_put(fsib_clk);
+ if (ret < 0)
+ return ret;
+
+ /* FSI DIV setting */
+ ret = __fsi_set_round_rate(fdiv_clk, fdiv_rate, enable);
+ if (ret < 0) {
+ /* disable FSI B */
+ if (enable)
+ __fsi_set_round_rate(fsib_clk, fsib_rate, 0);
+ return ret;
+ }
+
+ return ackmd_bpfmd;
+}
+
+static int fsi_set_rate(struct device *dev, int is_porta, int rate, int enable)
+{
+ int ret;
+
+ if (is_porta)
+ ret = fsi_ak4642_set_rate(dev, rate, enable);
+ else
+ ret = fsi_hdmi_set_rate(dev, rate, enable);
return ret;
}
goto out;
}
+ ret = clk_enable(&sh7372_pllc2_clk);
+ if (ret < 0) {
+ pr_err("Cannot enable pllc2 clock\n");
+ goto out;
+ }
pr_debug("PLLC2 set frequency %lu\n", rate);
ret = clk_set_parent(hdmi_ick, &sh7372_pllc2_clk);
device_initcall(hdmi_init_pm_clock);
-#define FSIACK_DUMMY_RATE 48000
static int __init fsi_init_pm_clock(void)
{
struct clk *fsia_ick;
int ret;
- /*
- * FSIACK is connected to AK4642,
- * and the rate is depend on playing sound rate.
- * So, set dummy rate (= 48k) here
- */
- ret = clk_set_rate(&sh7372_fsiack_clk, FSIACK_DUMMY_RATE);
- if (ret < 0) {
- pr_err("Cannot set FSIACK dummy rate: %d\n", ret);
- return ret;
- }
-
fsia_ick = clk_get(&fsi_device.dev, "icka");
if (IS_ERR(fsia_ick)) {
ret = PTR_ERR(fsia_ick);
}
ret = clk_set_parent(fsia_ick, &sh7372_fsiack_clk);
- if (ret < 0) {
- pr_err("Cannot set FSI-A parent: %d\n", ret);
- goto out;
- }
-
- ret = clk_set_rate(fsia_ick, FSIACK_DUMMY_RATE);
if (ret < 0)
- pr_err("Cannot set FSI-A rate: %d\n", ret);
+ pr_err("Cannot set FSI-A parent: %d\n", ret);
-out:
clk_put(fsia_ick);
return ret;
__raw_writel(__raw_readl(PLLC2CR) & ~0x80000000, PLLC2CR);
}
-static int pllc2_set_rate(struct clk *clk,
- unsigned long rate, int algo_id)
+static int pllc2_set_rate(struct clk *clk, unsigned long rate)
{
unsigned long value;
int idx;
if (idx < 0)
return idx;
- if (rate == clk->parent->rate) {
- pllc2_disable(clk);
- return 0;
- }
+ if (rate == clk->parent->rate)
+ return -EINVAL;
value = __raw_readl(PLLC2CR) & ~(0x3f << 24);
- if (value & 0x80000000)
- pllc2_disable(clk);
-
__raw_writel((value & ~0x80000000) | ((idx + 19) << 24), PLLC2CR);
- if (value & 0x80000000)
- return pllc2_enable(clk);
-
return 0;
}
unsigned long value;
value = __raw_readl(clk->mapping->base) >> 16;
- if (value < 2) {
- fsidiv_disable(clk);
- return -ENOENT;
- }
+ if (value < 2)
+ return -EIO;
__raw_writel((value << 16) | 0x3, clk->mapping->base);
return 0;
}
-static int fsidiv_set_rate(struct clk *clk,
- unsigned long rate, int algo_id)
+static int fsidiv_set_rate(struct clk *clk, unsigned long rate)
{
int idx;
- if (clk->parent->rate == rate) {
- fsidiv_disable(clk);
- return 0;
- }
-
idx = (clk->parent->rate / rate) & 0xffff;
if (idx < 2)
- return -ENOENT;
+ return -EINVAL;
__raw_writel(idx << 16, clk->mapping->base);
- return fsidiv_enable(clk);
+ return 0;
}
static struct clk_ops fsidiv_clk_ops = {
CLKDEV_CON_ID("vck3_clk", &div6_clks[DIV6_VCK3]),
CLKDEV_CON_ID("fmsi_clk", &div6_clks[DIV6_FMSI]),
CLKDEV_CON_ID("fmso_clk", &div6_clks[DIV6_FMSO]),
- CLKDEV_CON_ID("fsia_clk", &div6_reparent_clks[DIV6_FSIA]),
- CLKDEV_CON_ID("fsib_clk", &div6_reparent_clks[DIV6_FSIB]),
CLKDEV_CON_ID("sub_clk", &div6_clks[DIV6_SUB]),
CLKDEV_CON_ID("spu_clk", &div6_clks[DIV6_SPU]),
CLKDEV_CON_ID("vou_clk", &div6_clks[DIV6_VOU]),
CLKDEV_DEV_ID("sh_cmt.10", &mstp_clks[MSTP329]), /* CMT10 */
CLKDEV_DEV_ID("sh_fsi2", &mstp_clks[MSTP328]), /* FSI2 */
CLKDEV_DEV_ID("i2c-sh_mobile.1", &mstp_clks[MSTP323]), /* IIC1 */
- CLKDEV_DEV_ID("r8a66597_hcd.0", &mstp_clks[MSTP323]), /* USB0 */
- CLKDEV_DEV_ID("r8a66597_udc.0", &mstp_clks[MSTP323]), /* USB0 */
+ CLKDEV_DEV_ID("r8a66597_hcd.0", &mstp_clks[MSTP322]), /* USB0 */
+ CLKDEV_DEV_ID("r8a66597_udc.0", &mstp_clks[MSTP322]), /* USB0 */
CLKDEV_DEV_ID("sh_mobile_sdhi.0", &mstp_clks[MSTP314]), /* SDHI0 */
CLKDEV_DEV_ID("sh_mobile_sdhi.1", &mstp_clks[MSTP313]), /* SDHI1 */
CLKDEV_DEV_ID("sh_mmcif.0", &mstp_clks[MSTP312]), /* MMC */
static inline void ux500_cache_wait(void __iomem *reg, unsigned long mask)
{
/* wait for the operation to complete */
- while (readl(reg) & mask)
+ while (readl_relaxed(reg) & mask)
;
}
static inline void ux500_cache_sync(void)
{
void __iomem *base = __io_address(UX500_L2CC_BASE);
- writel(0, base + L2X0_CACHE_SYNC);
+ writel_relaxed(0, base + L2X0_CACHE_SYNC);
ux500_cache_wait(base + L2X0_CACHE_SYNC, 1);
}
uint32_t l2x0_way_mask = (1<<16) - 1; /* Bitmask of active ways */
/* invalidate all ways */
- writel(l2x0_way_mask, l2x0_base + L2X0_INV_WAY);
+ writel_relaxed(l2x0_way_mask, l2x0_base + L2X0_INV_WAY);
ux500_cache_wait(l2x0_base + L2X0_INV_WAY, l2x0_way_mask);
ux500_cache_sync();
}
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#define VMALLOC_END 0xd8000000
+#define VMALLOC_END 0xd8000000UL
*/
if (pfn_valid(pfn)) {
printk(KERN_WARNING "BUG: Your driver calls ioremap() on system memory. This leads\n"
- KERN_WARNING "to architecturally unpredictable behaviour on ARMv6+, and ioremap()\n"
- KERN_WARNING "will fail in the next kernel release. Please fix your driver.\n");
+ "to architecturally unpredictable behaviour on ARMv6+, and ioremap()\n"
+ "will fail in the next kernel release. Please fix your driver.\n");
WARN_ON(1);
}
#include <mach/hardware.h>
#include <mach/devices-common.h>
-#ifdef SDMA_IS_MERGED
#include <mach/sdma.h>
-#else
-struct sdma_platform_data {
- int sdma_version;
- char *cpu_name;
- int to_version;
-};
-#endif
struct imx_imx_sdma_data {
resource_size_t iobase;
imx_spi_imx_data_entry(MX21, CSPI, "imx21-cspi", _id, _hwid, SZ_4K)
imx21_cspi_data_entry(0, 1),
imx21_cspi_data_entry(1, 2),
+};
#endif
#ifdef CONFIG_ARCH_MX25
*
* Copyright (C) 2008 STMicroelectronics
* Copyright (C) 2010 Alessandro Rubini
+ * Copyright (C) 2010 Linus Walleij for ST-Ericsson
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2, as
#include <linux/clk.h>
#include <linux/jiffies.h>
#include <linux/err.h>
+#include <linux/cnt32_to_63.h>
+#include <linux/timer.h>
#include <asm/mach/time.h>
#include <plat/mtu.h>
-void __iomem *mtu_base; /* ssigned by machine code */
+void __iomem *mtu_base; /* Assigned by machine code */
/*
* Kernel assumes that sched_clock can be called early
/*
* Override the global weak sched_clock symbol with this
* local implementation which uses the clocksource to get some
- * better resolution when scheduling the kernel. We accept that
- * this wraps around for now, since it is just a relative time
- * stamp. (Inspired by OMAP implementation.)
+ * better resolution when scheduling the kernel.
+ *
+ * Because the hardware timer period may be quite short
+ * (32.3 secs on the 133 MHz MTU timer selection on ux500)
+ * and because cnt32_to_63() needs to be called at least once per
+ * half period to work properly, a kernel keepwarm() timer is set up
+ * to ensure this requirement is always met.
+ *
+ * Also the sched_clock timer will wrap around at some point,
+ * here we set it to run continously for a year.
*/
+#define SCHED_CLOCK_MIN_WRAP 3600*24*365
+static struct timer_list cnt32_to_63_keepwarm_timer;
+static u32 sched_mult;
+static u32 sched_shift;
+
unsigned long long notrace sched_clock(void)
{
- return clocksource_cyc2ns(nmdk_clksrc.read(
- &nmdk_clksrc),
- nmdk_clksrc.mult,
- nmdk_clksrc.shift);
+ u64 cycles;
+
+ if (unlikely(!mtu_base))
+ return 0;
+
+ cycles = cnt32_to_63(-readl(mtu_base + MTU_VAL(0)));
+ /*
+ * sched_mult is guaranteed to be even so will
+ * shift out bit 63
+ */
+ return (cycles * sched_mult) >> sched_shift;
+}
+
+/* Just kick sched_clock every so often */
+static void cnt32_to_63_keepwarm(unsigned long data)
+{
+ mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + data));
+ (void) sched_clock();
+}
+
+/*
+ * Set up a timer to keep sched_clock():s 32_to_63 algorithm warm
+ * once in half a 32bit timer wrap interval.
+ */
+static void __init nmdk_sched_clock_init(unsigned long rate)
+{
+ u32 v;
+ unsigned long delta;
+ u64 days;
+
+ /* Find the apropriate mult and shift factors */
+ clocks_calc_mult_shift(&sched_mult, &sched_shift,
+ rate, NSEC_PER_SEC, SCHED_CLOCK_MIN_WRAP);
+ /* We need to multiply by an even number to get rid of bit 63 */
+ if (sched_mult & 1)
+ sched_mult++;
+
+ /* Let's see what we get, take max counter and scale it */
+ days = (0xFFFFFFFFFFFFFFFFLLU * sched_mult) >> sched_shift;
+ do_div(days, NSEC_PER_SEC);
+ do_div(days, (3600*24));
+
+ pr_info("sched_clock: using %d bits @ %lu Hz wrap in %lu days\n",
+ (64 - sched_shift), rate, (unsigned long) days);
+
+ /*
+ * Program a timer to kick us at half 32bit wraparound
+ * Formula: seconds per wrap = (2^32) / f
+ */
+ v = 0xFFFFFFFFUL / rate;
+ /* We want half of the wrap time to keep cnt32_to_63 warm */
+ v /= 2;
+ pr_debug("sched_clock: prescaled timer rate: %lu Hz, "
+ "initialize keepwarm timer every %d seconds\n", rate, v);
+ /* Convert seconds to jiffies */
+ delta = msecs_to_jiffies(v*1000);
+ setup_timer(&cnt32_to_63_keepwarm_timer, cnt32_to_63_keepwarm, delta);
+ mod_timer(&cnt32_to_63_keepwarm_timer, round_jiffies(jiffies + delta));
}
/* Clockevent device: use one-shot mode */
writel(0, mtu_base + MTU_BGLR(0));
writel(cr | MTU_CRn_ENA, mtu_base + MTU_CR(0));
- /* Now the scheduling clock is ready */
+ /* Now the clock source is ready */
nmdk_clksrc.read = nmdk_read_timer;
if (clocksource_register(&nmdk_clksrc))
pr_err("timer: failed to initialize clock source %s\n",
nmdk_clksrc.name);
+ nmdk_sched_clock_init(rate);
+
/* Timer 1 is used for events */
clockevents_calc_mult_shift(&nmdk_clkevt, rate, MTU_MIN_RANGE);
/* Require clock free running */
#define PXA_FLAG_DISABLE_CLOCK_GATING (1<<0)
+/* Board design supports 8-bit data on SD/SDIO BUS */
+#define PXA_FLAG_SD_8_BIT_CAPABLE_SLOT (1<<2)
+
/*
* struct pxa_sdhci_platdata() - Platform device data for PXA SDHCI
* @max_speed: the maximum speed supported
} else {
s3c_gpio_cfgpin(S3C2410_GPE(13), S3C2410_GPIO_INPUT);
s3c_gpio_cfgpin(S3C2410_GPE(11), S3C2410_GPIO_INPUT);
- s3c_gpio_cfgpull(S3C2410_GPE(11), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPE(12), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPE(13), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPE(11), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPE(12), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPE(13), S3C_GPIO_PULL_NONE);
}
}
} else {
s3c_gpio_cfgpin(S3C2410_GPD(8), S3C2410_GPIO_INPUT);
s3c_gpio_cfgpin(S3C2410_GPD(9), S3C2410_GPIO_INPUT);
- s3c_gpio_cfgpull(S3C2410_GPD(10), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPD(9), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPD(8), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPD(10), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPD(9), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPD(8), S3C_GPIO_PULL_NONE);
}
}
} else {
s3c_gpio_cfgpin(S3C2410_GPG(7), S3C2410_GPIO_INPUT);
s3c_gpio_cfgpin(S3C2410_GPG(5), S3C2410_GPIO_INPUT);
- s3c_gpio_cfgpull(S3C2410_GPG(5), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPG(6), S3C_GPIO_PULL_NONE);
- s3c_gpio_cfgpull(S3C2410_GPG(7), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPG(5), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPG(6), S3C_GPIO_PULL_NONE);
+ s3c_gpio_setpull(S3C2410_GPG(7), S3C_GPIO_PULL_NONE);
}
}
static void pte_free_submit(struct pte_freelist_batch *batch)
{
- call_rcu(&batch->rcu, pte_free_rcu_callback);
+ call_rcu_sched(&batch->rcu, pte_free_rcu_callback);
}
void pgtable_free_tlb(struct mmu_gather *tlb, void *table, unsigned shift)
static int notrace s390_revalidate_registers(struct mci *mci)
{
int kill_task;
- u64 tmpclock;
u64 zero;
void *fpt_save_area, *fpt_creg_save_area;
: "0", "cc");
#endif
/* Revalidate clock comparator register */
- asm volatile(
- " stck 0(%1)\n"
- " sckc 0(%1)"
- : "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
-
+ if (S390_lowcore.clock_comparator == -1)
+ set_clock_comparator(S390_lowcore.mcck_clock);
+ else
+ set_clock_comparator(S390_lowcore.clock_comparator);
/* Check if old PSW is valid */
if (!mci->wp)
/*
{
unsigned long mask, cr0, cr0_saved;
u64 clock_saved;
+ u64 end;
+ mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_EXT;
+ end = get_clock() + (usecs << 12);
clock_saved = local_tick_disable();
- set_clock_comparator(get_clock() + (usecs << 12));
__ctl_store(cr0_saved, 0, 0);
cr0 = (cr0_saved & 0xffff00e0) | 0x00000800;
__ctl_load(cr0 , 0, 0);
- mask = psw_kernel_bits | PSW_MASK_WAIT | PSW_MASK_EXT;
lockdep_off();
- trace_hardirqs_on();
- __load_psw_mask(mask);
- local_irq_disable();
+ do {
+ set_clock_comparator(end);
+ trace_hardirqs_on();
+ __load_psw_mask(mask);
+ local_irq_disable();
+ } while (get_clock() < end);
lockdep_on();
__ctl_load(cr0_saved, 0, 0);
local_tick_enable(clock_saved);
#define ARCH_HAS_PREFETCHW
static inline void prefetch(void *x)
{
- __asm__ __volatile__ ("pref @%0\n\t" : : "r" (x) : "memory");
+ __builtin_prefetch(x, 0, 3);
}
-#define prefetchw(x) prefetch(x)
+static inline void prefetchw(void *x)
+{
+ __builtin_prefetch(x, 1, 3);
+}
#endif
#endif /* __KERNEL__ */
return 0;
}
-static int shoc_clk_set_rate(struct clk *clk, unsigned long rate, int algo_id)
+static int shoc_clk_set_rate(struct clk *clk, unsigned long rate)
{
unsigned long frqcr3;
unsigned int tmp;
}
if (op & CACHEFLUSH_I)
- flush_cache_all();
+ flush_icache_range(addr, addr+len);
up_read(¤t->mm->mmap_sem);
return 0;
* fill out .eh_frame -- PFM. */
.LEND_vsyscall:
.size __kernel_vsyscall,.-.LSTART_vsyscall
- .previous
.section .eh_frame,"a",@progbits
+ .previous
.LCIE:
.ualong .LCIE_end - .LCIE_start
.LCIE_start:
menu "Bus options"
+config PCI
+ bool "PCI support"
+ default y
+ select PCI_DOMAINS
+ ---help---
+ Enable PCI root complex support, so PCIe endpoint devices can
+ be attached to the Tile chip. Many, but not all, PCI devices
+ are supported under Tilera's root complex driver.
+
+config PCI_DOMAINS
+ bool
+
config NO_IOMEM
def_bool !PCI
mb_incoherent();
}
+/*
+ * Flush & invalidate a VA range that is homed remotely on a single core,
+ * waiting until the memory controller holds the flushed values.
+ */
+static inline void finv_buffer_remote(void *buffer, size_t size)
+{
+ char *p;
+ int i;
+
+ /*
+ * Flush and invalidate the buffer out of the local L1/L2
+ * and request the home cache to flush and invalidate as well.
+ */
+ __finv_buffer(buffer, size);
+
+ /*
+ * Wait for the home cache to acknowledge that it has processed
+ * all the flush-and-invalidate requests. This does not mean
+ * that the flushed data has reached the memory controller yet,
+ * but it does mean the home cache is processing the flushes.
+ */
+ __insn_mf();
+
+ /*
+ * Issue a load to the last cache line, which can't complete
+ * until all the previously-issued flushes to the same memory
+ * controller have also completed. If we weren't striping
+ * memory, that one load would be sufficient, but since we may
+ * be, we also need to back up to the last load issued to
+ * another memory controller, which would be the point where
+ * we crossed an 8KB boundary (the granularity of striping
+ * across memory controllers). Keep backing up and doing this
+ * until we are before the beginning of the buffer, or have
+ * hit all the controllers.
+ */
+ for (i = 0, p = (char *)buffer + size - 1;
+ i < (1 << CHIP_LOG_NUM_MSHIMS()) && p >= (char *)buffer;
+ ++i) {
+ const unsigned long STRIPE_WIDTH = 8192;
+
+ /* Force a load instruction to issue. */
+ *(volatile char *)p;
+
+ /* Jump to end of previous stripe. */
+ p -= STRIPE_WIDTH;
+ p = (char *)((unsigned long)p | (STRIPE_WIDTH - 1));
+ }
+
+ /* Wait for the loads (and thus flushes) to have completed. */
+ __insn_mf();
+}
+
#endif /* _ASM_TILE_CACHEFLUSH_H */
#define ioremap_writethrough(physaddr, size) ioremap(physaddr, size)
#define ioremap_fullcache(physaddr, size) ioremap(physaddr, size)
-void __iomem *ioport_map(unsigned long port, unsigned int len);
-extern inline void ioport_unmap(void __iomem *addr) {}
-
#define mmiowb()
/* Conversion between virtual and physical mappings. */
* we never run, uses them unconditionally.
*/
-static inline int ioport_panic(void)
+static inline long ioport_panic(void)
{
panic("inb/outb and friends do not exist on tile");
return 0;
}
+static inline void __iomem *ioport_map(unsigned long port, unsigned int len)
+{
+ return (void __iomem *) ioport_panic();
+}
+
+static inline void ioport_unmap(void __iomem *addr)
+{
+ ioport_panic();
+}
+
static inline u8 inb(unsigned long addr)
{
return ioport_panic();
+++ /dev/null
-/*
- * Copyright 2010 Tilera Corporation. All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation, version 2.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
- * NON INFRINGEMENT. See the GNU General Public License for
- * more details.
- */
-
-#ifndef _ASM_TILE_PCI_BRIDGE_H
-#define _ASM_TILE_PCI_BRIDGE_H
-
-#include <linux/ioport.h>
-#include <linux/pci.h>
-
-struct device_node;
-struct pci_controller;
-
-/*
- * pci_io_base returns the memory address at which you can access
- * the I/O space for PCI bus number `bus' (or NULL on error).
- */
-extern void __iomem *pci_bus_io_base(unsigned int bus);
-extern unsigned long pci_bus_io_base_phys(unsigned int bus);
-extern unsigned long pci_bus_mem_base_phys(unsigned int bus);
-
-/* Allocate a new PCI host bridge structure */
-extern struct pci_controller *pcibios_alloc_controller(void);
-
-/* Helper function for setting up resources */
-extern void pci_init_resource(struct resource *res, unsigned long start,
- unsigned long end, int flags, char *name);
-
-/* Get the PCI host controller for a bus */
-extern struct pci_controller *pci_bus_to_hose(int bus);
-
-/*
- * Structure of a PCI controller (host bridge)
- */
-struct pci_controller {
- int index; /* PCI domain number */
- struct pci_bus *root_bus;
-
- int first_busno;
- int last_busno;
-
- int hv_cfg_fd[2]; /* config{0,1} fds for this PCIe controller */
- int hv_mem_fd; /* fd to Hypervisor for MMIO operations */
-
- struct pci_ops *ops;
-
- int irq_base; /* Base IRQ from the Hypervisor */
- int plx_gen1; /* flag for PLX Gen 1 configuration */
-
- /* Address ranges that are routed to this controller/bridge. */
- struct resource mem_resources[3];
-};
-
-static inline struct pci_controller *pci_bus_to_host(struct pci_bus *bus)
-{
- return bus->sysdata;
-}
-
-extern void setup_indirect_pci_nomap(struct pci_controller *hose,
- void __iomem *cfg_addr, void __iomem *cfg_data);
-extern void setup_indirect_pci(struct pci_controller *hose,
- u32 cfg_addr, u32 cfg_data);
-extern void setup_grackle(struct pci_controller *hose);
-
-extern unsigned char common_swizzle(struct pci_dev *, unsigned char *);
-
-/*
- * The following code swizzles for exactly one bridge. The routine
- * common_swizzle below handles multiple bridges. But there are a
- * some boards that don't follow the PCI spec's suggestion so we
- * break this piece out separately.
- */
-static inline unsigned char bridge_swizzle(unsigned char pin,
- unsigned char idsel)
-{
- return (((pin-1) + idsel) % 4) + 1;
-}
-
-/*
- * The following macro is used to lookup irqs in a standard table
- * format for those PPC systems that do not already have PCI
- * interrupts properly routed.
- */
-/* FIXME - double check this */
-#define PCI_IRQ_TABLE_LOOKUP ({ \
- long _ctl_ = -1; \
- if (idsel >= min_idsel && idsel <= max_idsel && pin <= irqs_per_slot) \
- _ctl_ = pci_irq_table[idsel - min_idsel][pin-1]; \
- _ctl_; \
-})
-
-/*
- * Scan the buses below a given PCI host bridge and assign suitable
- * resources to all devices found.
- */
-extern int pciauto_bus_scan(struct pci_controller *, int);
-
-#ifdef CONFIG_PCI
-extern unsigned long pci_address_to_pio(phys_addr_t address);
-#else
-static inline unsigned long pci_address_to_pio(phys_addr_t address)
-{
- return (unsigned long)-1;
-}
-#endif
-
-#endif /* _ASM_TILE_PCI_BRIDGE_H */
#ifndef _ASM_TILE_PCI_H
#define _ASM_TILE_PCI_H
-#include <asm/pci-bridge.h>
+#include <linux/pci.h>
+
+/*
+ * Structure of a PCI controller (host bridge)
+ */
+struct pci_controller {
+ int index; /* PCI domain number */
+ struct pci_bus *root_bus;
+
+ int first_busno;
+ int last_busno;
+
+ int hv_cfg_fd[2]; /* config{0,1} fds for this PCIe controller */
+ int hv_mem_fd; /* fd to Hypervisor for MMIO operations */
+
+ struct pci_ops *ops;
+
+ int irq_base; /* Base IRQ from the Hypervisor */
+ int plx_gen1; /* flag for PLX Gen 1 configuration */
+
+ /* Address ranges that are routed to this controller/bridge. */
+ struct resource mem_resources[3];
+};
/*
* The hypervisor maps the entirety of CPA-space as bus addresses, so
*/
#define PCI_DMA_BUS_IS_PHYS 1
-struct pci_controller *pci_bus_to_hose(int bus);
-unsigned char __init common_swizzle(struct pci_dev *dev, unsigned char *pinp);
int __init tile_pci_init(void);
-void pci_iounmap(struct pci_dev *dev, void __iomem *addr);
-void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max);
-void __devinit pcibios_fixup_bus(struct pci_bus *bus);
-int __devinit _tile_cfg_read(struct pci_controller *hose,
- int bus,
- int slot,
- int function,
- int offset,
- int size,
- u32 *val);
-int __devinit _tile_cfg_write(struct pci_controller *hose,
- int bus,
- int slot,
- int function,
- int offset,
- int size,
- u32 val);
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max);
+static inline void pci_iounmap(struct pci_dev *dev, void __iomem *addr) {}
-/*
- * These are used to to config reads and writes in the early stages of
- * setup before the driver infrastructure has been set up enough to be
- * able to do config reads and writes.
- */
-#define early_cfg_read(where, size, value) \
- _tile_cfg_read(controller, \
- current_bus, \
- pci_slot, \
- pci_fn, \
- where, \
- size, \
- value)
-
-#define early_cfg_write(where, size, value) \
- _tile_cfg_write(controller, \
- current_bus, \
- pci_slot, \
- pci_fn, \
- where, \
- size, \
- value)
-
-
-
-#define PCICFG_BYTE 1
-#define PCICFG_WORD 2
-#define PCICFG_DWORD 4
+void __devinit pcibios_fixup_bus(struct pci_bus *bus);
#define TILE_NUM_PCIE 2
}
/*
- * I/O space is currently not supported.
+ * pcibios_assign_all_busses() tells whether or not the bus numbers
+ * should be reassigned, in case the BIOS didn't do it correctly, or
+ * in case we don't have a BIOS and we want to let Linux do it.
*/
+static inline int pcibios_assign_all_busses(void)
+{
+ return 1;
+}
-#define TILE_PCIE_LOWER_IO 0x0
-#define TILE_PCIE_UPPER_IO 0x10000
-#define TILE_PCIE_PCIE_IO_SIZE 0x0000FFFF
-
-#define _PAGE_NO_CACHE 0
-#define _PAGE_GUARDED 0
-
-
-#define pcibios_assign_all_busses() pci_assign_all_buses
-extern int pci_assign_all_buses;
-
+/*
+ * No special bus mastering setup handling.
+ */
static inline void pcibios_set_master(struct pci_dev *dev)
{
- /* No special bus mastering setup handling */
}
#define PCIBIOS_MIN_MEM 0
-#define PCIBIOS_MIN_IO TILE_PCIE_LOWER_IO
+#define PCIBIOS_MIN_IO 0
/*
* This flag tells if the platform is TILEmpower that needs
* special configuration for the PLX switch chip.
*/
-extern int blade_pci;
+extern int tile_plx_gen1;
+
+/* Use any cpu for PCI. */
+#define cpumask_of_pcibus(bus) cpu_online_mask
/* implement the pci_ DMA API in terms of the generic device dma_ one */
#include <asm-generic/pci-dma-compat.h>
/* generic pci stuff */
#include <asm-generic/pci.h>
-/* Use any cpu for PCI. */
-#define cpumask_of_pcibus(bus) cpu_online_mask
-
#endif /* _ASM_TILE_PCI_H */
/* Are we using huge pages in the TLB for kernel data? */
extern int kdata_huge;
+/* Support standard Linux prefetching. */
+#define ARCH_HAS_PREFETCH
+#define prefetch(x) __builtin_prefetch(x)
#define PREFETCH_STRIDE CHIP_L2_LINE_SIZE()
+/* Bring a value into the L1D, faulting the TLB if necessary. */
+#ifdef __tilegx__
+#define prefetch_L1(x) __insn_prefetch_l1_fault((void *)(x))
+#else
+#define prefetch_L1(x) __insn_prefetch_L1((void *)(x))
+#endif
+
#else /* __ASSEMBLY__ */
/* Do some slow action (e.g. read a slow SPR). */
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+/**
+ * @file drivers/xgbe/impl.h
+ * Implementation details for the NetIO library.
+ */
+
+#ifndef __DRV_XGBE_IMPL_H__
+#define __DRV_XGBE_IMPL_H__
+
+#include <hv/netio_errors.h>
+#include <hv/netio_intf.h>
+#include <hv/drv_xgbe_intf.h>
+
+
+/** How many groups we have (log2). */
+#define LOG2_NUM_GROUPS (12)
+/** How many groups we have. */
+#define NUM_GROUPS (1 << LOG2_NUM_GROUPS)
+
+/** Number of output requests we'll buffer per tile. */
+#define EPP_REQS_PER_TILE (32)
+
+/** Words used in an eDMA command without checksum acceleration. */
+#define EDMA_WDS_NO_CSUM 8
+/** Words used in an eDMA command with checksum acceleration. */
+#define EDMA_WDS_CSUM 10
+/** Total available words in the eDMA command FIFO. */
+#define EDMA_WDS_TOTAL 128
+
+
+/*
+ * FIXME: These definitions are internal and should have underscores!
+ * NOTE: The actual numeric values here are intentional and allow us to
+ * optimize the concept "if small ... else if large ... else ...", by
+ * checking for the low bit being set, and then for non-zero.
+ * These are used as array indices, so they must have the values (0, 1, 2)
+ * in some order.
+ */
+#define SIZE_SMALL (1) /**< Small packet queue. */
+#define SIZE_LARGE (2) /**< Large packet queue. */
+#define SIZE_JUMBO (0) /**< Jumbo packet queue. */
+
+/** The number of "SIZE_xxx" values. */
+#define NETIO_NUM_SIZES 3
+
+
+/*
+ * Default numbers of packets for IPP drivers. These values are chosen
+ * such that CIPP1 will not overflow its L2 cache.
+ */
+
+/** The default number of small packets. */
+#define NETIO_DEFAULT_SMALL_PACKETS 2750
+/** The default number of large packets. */
+#define NETIO_DEFAULT_LARGE_PACKETS 2500
+/** The default number of jumbo packets. */
+#define NETIO_DEFAULT_JUMBO_PACKETS 250
+
+
+/** Log2 of the size of a memory arena. */
+#define NETIO_ARENA_SHIFT 24 /* 16 MB */
+/** Size of a memory arena. */
+#define NETIO_ARENA_SIZE (1 << NETIO_ARENA_SHIFT)
+
+
+/** A queue of packets.
+ *
+ * This structure partially defines a queue of packets waiting to be
+ * processed. The queue as a whole is written to by an interrupt handler and
+ * read by non-interrupt code; this data structure is what's touched by the
+ * interrupt handler. The other part of the queue state, the read offset, is
+ * kept in user space, not in hypervisor space, so it is in a separate data
+ * structure.
+ *
+ * The read offset (__packet_receive_read in the user part of the queue
+ * structure) points to the next packet to be read. When the read offset is
+ * equal to the write offset, the queue is empty; therefore the queue must
+ * contain one more slot than the required maximum queue size.
+ *
+ * Here's an example of all 3 state variables and what they mean. All
+ * pointers move left to right.
+ *
+ * @code
+ * I I V V V V I I I I
+ * 0 1 2 3 4 5 6 7 8 9 10
+ * ^ ^ ^ ^
+ * | | |
+ * | | __last_packet_plus_one
+ * | __buffer_write
+ * __packet_receive_read
+ * @endcode
+ *
+ * This queue has 10 slots, and thus can hold 9 packets (_last_packet_plus_one
+ * = 10). The read pointer is at 2, and the write pointer is at 6; thus,
+ * there are valid, unread packets in slots 2, 3, 4, and 5. The remaining
+ * slots are invalid (do not contain a packet).
+ */
+typedef struct {
+ /** Byte offset of the next notify packet to be written: zero for the first
+ * packet on the queue, sizeof (netio_pkt_t) for the second packet on the
+ * queue, etc. */
+ volatile uint32_t __packet_write;
+
+ /** Offset of the packet after the last valid packet (i.e., when any
+ * pointer is incremented to this value, it wraps back to zero). */
+ uint32_t __last_packet_plus_one;
+}
+__netio_packet_queue_t;
+
+
+/** A queue of buffers.
+ *
+ * This structure partially defines a queue of empty buffers which have been
+ * obtained via requests to the IPP. (The elements of the queue are packet
+ * handles, which are transformed into a full netio_pkt_t when the buffer is
+ * retrieved.) The queue as a whole is written to by an interrupt handler and
+ * read by non-interrupt code; this data structure is what's touched by the
+ * interrupt handler. The other parts of the queue state, the read offset and
+ * requested write offset, are kept in user space, not in hypervisor space, so
+ * they are in a separate data structure.
+ *
+ * The read offset (__buffer_read in the user part of the queue structure)
+ * points to the next buffer to be read. When the read offset is equal to the
+ * write offset, the queue is empty; therefore the queue must contain one more
+ * slot than the required maximum queue size.
+ *
+ * The requested write offset (__buffer_requested_write in the user part of
+ * the queue structure) points to the slot which will hold the next buffer we
+ * request from the IPP, once we get around to sending such a request. When
+ * the requested write offset is equal to the write offset, no requests for
+ * new buffers are outstanding; when the requested write offset is one greater
+ * than the read offset, no more requests may be sent.
+ *
+ * Note that, unlike the packet_queue, the buffer_queue places incoming
+ * buffers at decreasing addresses. This makes the check for "is it time to
+ * wrap the buffer pointer" cheaper in the assembly code which receives new
+ * buffers, and means that the value which defines the queue size,
+ * __last_buffer, is different than in the packet queue. Also, the offset
+ * used in the packet_queue is already scaled by the size of a packet; here we
+ * use unscaled slot indices for the offsets. (These differences are
+ * historical, and in the future it's possible that the packet_queue will look
+ * more like this queue.)
+ *
+ * @code
+ * Here's an example of all 4 state variables and what they mean. Remember:
+ * all pointers move right to left.
+ *
+ * V V V I I R R V V V
+ * 0 1 2 3 4 5 6 7 8 9
+ * ^ ^ ^ ^
+ * | | | |
+ * | | | __last_buffer
+ * | | __buffer_write
+ * | __buffer_requested_write
+ * __buffer_read
+ * @endcode
+ *
+ * This queue has 10 slots, and thus can hold 9 buffers (_last_buffer = 9).
+ * The read pointer is at 2, and the write pointer is at 6; thus, there are
+ * valid, unread buffers in slots 2, 1, 0, 9, 8, and 7. The requested write
+ * pointer is at 4; thus, requests have been made to the IPP for buffers which
+ * will be placed in slots 6 and 5 when they arrive. Finally, the remaining
+ * slots are invalid (do not contain a buffer).
+ */
+typedef struct
+{
+ /** Ordinal number of the next buffer to be written: 0 for the first slot in
+ * the queue, 1 for the second slot in the queue, etc. */
+ volatile uint32_t __buffer_write;
+
+ /** Ordinal number of the last buffer (i.e., when any pointer is decremented
+ * below zero, it is reloaded with this value). */
+ uint32_t __last_buffer;
+}
+__netio_buffer_queue_t;
+
+
+/**
+ * An object for providing Ethernet packets to a process.
+ */
+typedef struct __netio_queue_impl_t
+{
+ /** The queue of packets waiting to be received. */
+ __netio_packet_queue_t __packet_receive_queue;
+ /** The intr bit mask that IDs this device. */
+ unsigned int __intr_id;
+ /** Offset to queues of empty buffers, one per size. */
+ uint32_t __buffer_queue[NETIO_NUM_SIZES];
+ /** The address of the first EPP tile, or -1 if no EPP. */
+ /* ISSUE: Actually this is always "0" or "~0". */
+ uint32_t __epp_location;
+ /** The queue ID that this queue represents. */
+ unsigned int __queue_id;
+ /** Number of acknowledgements received. */
+ volatile uint32_t __acks_received;
+ /** Last completion number received for packet_sendv. */
+ volatile uint32_t __last_completion_rcv;
+ /** Number of packets allowed to be outstanding. */
+ uint32_t __max_outstanding;
+ /** First VA available for packets. */
+ void* __va_0;
+ /** First VA in second range available for packets. */
+ void* __va_1;
+ /** Padding to align the "__packets" field to the size of a netio_pkt_t. */
+ uint32_t __padding[3];
+ /** The packets themselves. */
+ netio_pkt_t __packets[0];
+}
+netio_queue_impl_t;
+
+
+/**
+ * An object for managing the user end of a NetIO queue.
+ */
+typedef struct __netio_queue_user_impl_t
+{
+ /** The next incoming packet to be read. */
+ uint32_t __packet_receive_read;
+ /** The next empty buffers to be read, one index per size. */
+ uint8_t __buffer_read[NETIO_NUM_SIZES];
+ /** Where the empty buffer we next request from the IPP will go, one index
+ * per size. */
+ uint8_t __buffer_requested_write[NETIO_NUM_SIZES];
+ /** PCIe interface flag. */
+ uint8_t __pcie;
+ /** Number of packets left to be received before we send a credit update. */
+ uint32_t __receive_credit_remaining;
+ /** Value placed in __receive_credit_remaining when it reaches zero. */
+ uint32_t __receive_credit_interval;
+ /** First fast I/O routine index. */
+ uint32_t __fastio_index;
+ /** Number of acknowledgements expected. */
+ uint32_t __acks_outstanding;
+ /** Last completion number requested. */
+ uint32_t __last_completion_req;
+ /** File descriptor for driver. */
+ int __fd;
+}
+netio_queue_user_impl_t;
+
+
+#define NETIO_GROUP_CHUNK_SIZE 64 /**< Max # groups in one IPP request */
+#define NETIO_BUCKET_CHUNK_SIZE 64 /**< Max # buckets in one IPP request */
+
+
+/** Internal structure used to convey packet send information to the
+ * hypervisor. FIXME: Actually, it's not used for that anymore, but
+ * netio_packet_send() still uses it internally.
+ */
+typedef struct
+{
+ uint16_t flags; /**< Packet flags (__NETIO_SEND_FLG_xxx) */
+ uint16_t transfer_size; /**< Size of packet */
+ uint32_t va; /**< VA of start of packet */
+ __netio_pkt_handle_t handle; /**< Packet handle */
+ uint32_t csum0; /**< First checksum word */
+ uint32_t csum1; /**< Second checksum word */
+}
+__netio_send_cmd_t;
+
+
+/** Flags used in two contexts:
+ * - As the "flags" member in the __netio_send_cmd_t, above; used only
+ * for netio_pkt_send_{prepare,commit}.
+ * - As part of the flags passed to the various send packet fast I/O calls.
+ */
+
+/** Need acknowledgement on this packet. Note that some code in the
+ * normal send_pkt fast I/O handler assumes that this is equal to 1. */
+#define __NETIO_SEND_FLG_ACK 0x1
+
+/** Do checksum on this packet. (Only used with the __netio_send_cmd_t;
+ * normal packet sends use a special fast I/O index to denote checksumming,
+ * and multi-segment sends test the checksum descriptor.) */
+#define __NETIO_SEND_FLG_CSUM 0x2
+
+/** Get a completion on this packet. Only used with multi-segment sends. */
+#define __NETIO_SEND_FLG_COMPLETION 0x4
+
+/** Position of the number-of-extra-segments value in the flags word.
+ Only used with multi-segment sends. */
+#define __NETIO_SEND_FLG_XSEG_SHIFT 3
+
+/** Width of the number-of-extra-segments value in the flags word. */
+#define __NETIO_SEND_FLG_XSEG_WIDTH 2
+
+#endif /* __DRV_XGBE_IMPL_H__ */
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+/**
+ * @file drv_xgbe_intf.h
+ * Interface to the hypervisor XGBE driver.
+ */
+
+#ifndef __DRV_XGBE_INTF_H__
+#define __DRV_XGBE_INTF_H__
+
+/**
+ * An object for forwarding VAs and PAs to the hypervisor.
+ * @ingroup types
+ *
+ * This allows the supervisor to specify a number of areas of memory to
+ * store packet buffers.
+ */
+typedef struct
+{
+ /** The physical address of the memory. */
+ HV_PhysAddr pa;
+ /** Page table entry for the memory. This is only used to derive the
+ * memory's caching mode; the PA bits are ignored. */
+ HV_PTE pte;
+ /** The virtual address of the memory. */
+ HV_VirtAddr va;
+ /** Size (in bytes) of the memory area. */
+ int size;
+
+}
+netio_ipp_address_t;
+
+/** The various pread/pwrite offsets into the hypervisor-level driver.
+ * @ingroup types
+ */
+typedef enum
+{
+ /** Inform the Linux driver of the address of the NetIO arena memory.
+ * This offset is actually only used to convey information from netio
+ * to the Linux driver; it never makes it from there to the hypervisor.
+ * Write-only; takes a uint32_t specifying the VA address. */
+ NETIO_FIXED_ADDR = 0x5000000000000000ULL,
+
+ /** Inform the Linux driver of the size of the NetIO arena memory.
+ * This offset is actually only used to convey information from netio
+ * to the Linux driver; it never makes it from there to the hypervisor.
+ * Write-only; takes a uint32_t specifying the VA size. */
+ NETIO_FIXED_SIZE = 0x5100000000000000ULL,
+
+ /** Register current tile with IPP. Write then read: write, takes a
+ * netio_input_config_t, read returns a pointer to a netio_queue_impl_t. */
+ NETIO_IPP_INPUT_REGISTER_OFF = 0x6000000000000000ULL,
+
+ /** Unregister current tile from IPP. Write-only, takes a dummy argument. */
+ NETIO_IPP_INPUT_UNREGISTER_OFF = 0x6100000000000000ULL,
+
+ /** Start packets flowing. Write-only, takes a dummy argument. */
+ NETIO_IPP_INPUT_INIT_OFF = 0x6200000000000000ULL,
+
+ /** Stop packets flowing. Write-only, takes a dummy argument. */
+ NETIO_IPP_INPUT_UNINIT_OFF = 0x6300000000000000ULL,
+
+ /** Configure group (typically we group on VLAN). Write-only: takes an
+ * array of netio_group_t's, low 24 bits of the offset is the base group
+ * number times the size of a netio_group_t. */
+ NETIO_IPP_INPUT_GROUP_CFG_OFF = 0x6400000000000000ULL,
+
+ /** Configure bucket. Write-only: takes an array of netio_bucket_t's, low
+ * 24 bits of the offset is the base bucket number times the size of a
+ * netio_bucket_t. */
+ NETIO_IPP_INPUT_BUCKET_CFG_OFF = 0x6500000000000000ULL,
+
+ /** Get/set a parameter. Read or write: read or write data is the parameter
+ * value, low 32 bits of the offset is a __netio_getset_offset_t. */
+ NETIO_IPP_PARAM_OFF = 0x6600000000000000ULL,
+
+ /** Get fast I/O index. Read-only; returns a 4-byte base index value. */
+ NETIO_IPP_GET_FASTIO_OFF = 0x6700000000000000ULL,
+
+ /** Configure hijack IP address. Packets with this IPv4 dest address
+ * go to bucket NETIO_NUM_BUCKETS - 1. Write-only: takes an IP address
+ * in some standard form. FIXME: Define the form! */
+ NETIO_IPP_INPUT_HIJACK_CFG_OFF = 0x6800000000000000ULL,
+
+ /**
+ * Offsets beyond this point are reserved for the supervisor (although that
+ * enforcement must be done by the supervisor driver itself).
+ */
+ NETIO_IPP_USER_MAX_OFF = 0x6FFFFFFFFFFFFFFFULL,
+
+ /** Register I/O memory. Write-only, takes a netio_ipp_address_t. */
+ NETIO_IPP_IOMEM_REGISTER_OFF = 0x7000000000000000ULL,
+
+ /** Unregister I/O memory. Write-only, takes a netio_ipp_address_t. */
+ NETIO_IPP_IOMEM_UNREGISTER_OFF = 0x7100000000000000ULL,
+
+ /* Offsets greater than 0x7FFFFFFF can't be used directly from Linux
+ * userspace code due to limitations in the pread/pwrite syscalls. */
+
+ /** Drain LIPP buffers. */
+ NETIO_IPP_DRAIN_OFF = 0xFA00000000000000ULL,
+
+ /** Supply a netio_ipp_address_t to be used as shared memory for the
+ * LEPP command queue. */
+ NETIO_EPP_SHM_OFF = 0xFB00000000000000ULL,
+
+ /* 0xFC... is currently unused. */
+
+ /** Stop IPP/EPP tiles. Write-only, takes a dummy argument. */
+ NETIO_IPP_STOP_SHIM_OFF = 0xFD00000000000000ULL,
+
+ /** Start IPP/EPP tiles. Write-only, takes a dummy argument. */
+ NETIO_IPP_START_SHIM_OFF = 0xFE00000000000000ULL,
+
+ /** Supply packet arena. Write-only, takes an array of
+ * netio_ipp_address_t values. */
+ NETIO_IPP_ADDRESS_OFF = 0xFF00000000000000ULL,
+} netio_hv_offset_t;
+
+/** Extract the base offset from an offset */
+#define NETIO_BASE_OFFSET(off) ((off) & 0xFF00000000000000ULL)
+/** Extract the local offset from an offset */
+#define NETIO_LOCAL_OFFSET(off) ((off) & 0x00FFFFFFFFFFFFFFULL)
+
+
+/**
+ * Get/set offset.
+ */
+typedef union
+{
+ struct
+ {
+ uint64_t addr:48; /**< Class-specific address */
+ unsigned int class:8; /**< Class (e.g., NETIO_PARAM) */
+ unsigned int opcode:8; /**< High 8 bits of NETIO_IPP_PARAM_OFF */
+ }
+ bits; /**< Bitfields */
+ uint64_t word; /**< Aggregated value to use as the offset */
+}
+__netio_getset_offset_t;
+
+/**
+ * Fast I/O index offsets (must be contiguous).
+ */
+typedef enum
+{
+ NETIO_FASTIO_ALLOCATE = 0, /**< Get empty packet buffer */
+ NETIO_FASTIO_FREE_BUFFER = 1, /**< Give buffer back to IPP */
+ NETIO_FASTIO_RETURN_CREDITS = 2, /**< Give credits to IPP */
+ NETIO_FASTIO_SEND_PKT_NOCK = 3, /**< Send a packet, no checksum */
+ NETIO_FASTIO_SEND_PKT_CK = 4, /**< Send a packet, with checksum */
+ NETIO_FASTIO_SEND_PKT_VEC = 5, /**< Send a vector of packets */
+ NETIO_FASTIO_SENDV_PKT = 6, /**< Sendv one packet */
+ NETIO_FASTIO_NUM_INDEX = 7, /**< Total number of fast I/O indices */
+} netio_fastio_index_t;
+
+/** 3-word return type for Fast I/O call. */
+typedef struct
+{
+ int err; /**< Error code. */
+ uint32_t val0; /**< Value. Meaning depends upon the specific call. */
+ uint32_t val1; /**< Value. Meaning depends upon the specific call. */
+} netio_fastio_rv3_t;
+
+/** 0-argument fast I/O call */
+int __netio_fastio0(uint32_t fastio_index);
+/** 1-argument fast I/O call */
+int __netio_fastio1(uint32_t fastio_index, uint32_t arg0);
+/** 3-argument fast I/O call, 2-word return value */
+netio_fastio_rv3_t __netio_fastio3_rv3(uint32_t fastio_index, uint32_t arg0,
+ uint32_t arg1, uint32_t arg2);
+/** 4-argument fast I/O call */
+int __netio_fastio4(uint32_t fastio_index, uint32_t arg0, uint32_t arg1,
+ uint32_t arg2, uint32_t arg3);
+/** 6-argument fast I/O call */
+int __netio_fastio6(uint32_t fastio_index, uint32_t arg0, uint32_t arg1,
+ uint32_t arg2, uint32_t arg3, uint32_t arg4, uint32_t arg5);
+/** 9-argument fast I/O call */
+int __netio_fastio9(uint32_t fastio_index, uint32_t arg0, uint32_t arg1,
+ uint32_t arg2, uint32_t arg3, uint32_t arg4, uint32_t arg5,
+ uint32_t arg6, uint32_t arg7, uint32_t arg8);
+
+/** Allocate an empty packet.
+ * @param fastio_index Fast I/O index.
+ * @param size Size of the packet to allocate.
+ */
+#define __netio_fastio_allocate(fastio_index, size) \
+ __netio_fastio1((fastio_index) + NETIO_FASTIO_ALLOCATE, size)
+
+/** Free a buffer.
+ * @param fastio_index Fast I/O index.
+ * @param handle Handle for the packet to free.
+ */
+#define __netio_fastio_free_buffer(fastio_index, handle) \
+ __netio_fastio1((fastio_index) + NETIO_FASTIO_FREE_BUFFER, handle)
+
+/** Increment our receive credits.
+ * @param fastio_index Fast I/O index.
+ * @param credits Number of credits to add.
+ */
+#define __netio_fastio_return_credits(fastio_index, credits) \
+ __netio_fastio1((fastio_index) + NETIO_FASTIO_RETURN_CREDITS, credits)
+
+/** Send packet, no checksum.
+ * @param fastio_index Fast I/O index.
+ * @param ackflag Nonzero if we want an ack.
+ * @param size Size of the packet.
+ * @param va Virtual address of start of packet.
+ * @param handle Packet handle.
+ */
+#define __netio_fastio_send_pkt_nock(fastio_index, ackflag, size, va, handle) \
+ __netio_fastio4((fastio_index) + NETIO_FASTIO_SEND_PKT_NOCK, ackflag, \
+ size, va, handle)
+
+/** Send packet, calculate checksum.
+ * @param fastio_index Fast I/O index.
+ * @param ackflag Nonzero if we want an ack.
+ * @param size Size of the packet.
+ * @param va Virtual address of start of packet.
+ * @param handle Packet handle.
+ * @param csum0 Shim checksum header.
+ * @param csum1 Checksum seed.
+ */
+#define __netio_fastio_send_pkt_ck(fastio_index, ackflag, size, va, handle, \
+ csum0, csum1) \
+ __netio_fastio6((fastio_index) + NETIO_FASTIO_SEND_PKT_CK, ackflag, \
+ size, va, handle, csum0, csum1)
+
+
+/** Format for the "csum0" argument to the __netio_fastio_send routines
+ * and LEPP. Note that this is currently exactly identical to the
+ * ShimProtocolOffloadHeader.
+ */
+typedef union
+{
+ struct
+ {
+ unsigned int start_byte:7; /**< The first byte to be checksummed */
+ unsigned int count:14; /**< Number of bytes to be checksummed. */
+ unsigned int destination_byte:7; /**< The byte to write the checksum to. */
+ unsigned int reserved:4; /**< Reserved. */
+ } bits; /**< Decomposed method of access. */
+ unsigned int word; /**< To send out the IDN. */
+} __netio_checksum_header_t;
+
+
+/** Sendv packet with 1 or 2 segments.
+ * @param fastio_index Fast I/O index.
+ * @param flags Ack/csum/notify flags in low 3 bits; number of segments minus
+ * 1 in next 2 bits; expected checksum in high 16 bits.
+ * @param confno Confirmation number to request, if notify flag set.
+ * @param csum0 Checksum descriptor; if zero, no checksum.
+ * @param va_F Virtual address of first segment.
+ * @param va_L Virtual address of last segment, if 2 segments.
+ * @param len_F_L Length of first segment in low 16 bits; length of last
+ * segment, if 2 segments, in high 16 bits.
+ */
+#define __netio_fastio_sendv_pkt_1_2(fastio_index, flags, confno, csum0, \
+ va_F, va_L, len_F_L) \
+ __netio_fastio6((fastio_index) + NETIO_FASTIO_SENDV_PKT, flags, confno, \
+ csum0, va_F, va_L, len_F_L)
+
+/** Send packet on PCIe interface.
+ * @param fastio_index Fast I/O index.
+ * @param flags Ack/csum/notify flags in low 3 bits.
+ * @param confno Confirmation number to request, if notify flag set.
+ * @param csum0 Checksum descriptor; Hard wired 0, not needed for PCIe.
+ * @param va_F Virtual address of the packet buffer.
+ * @param va_L Virtual address of last segment, if 2 segments. Hard wired 0.
+ * @param len_F_L Length of the packet buffer in low 16 bits.
+ */
+#define __netio_fastio_send_pcie_pkt(fastio_index, flags, confno, csum0, \
+ va_F, va_L, len_F_L) \
+ __netio_fastio6((fastio_index) + PCIE_FASTIO_SENDV_PKT, flags, confno, \
+ csum0, va_F, va_L, len_F_L)
+
+/** Sendv packet with 3 or 4 segments.
+ * @param fastio_index Fast I/O index.
+ * @param flags Ack/csum/notify flags in low 3 bits; number of segments minus
+ * 1 in next 2 bits; expected checksum in high 16 bits.
+ * @param confno Confirmation number to request, if notify flag set.
+ * @param csum0 Checksum descriptor; if zero, no checksum.
+ * @param va_F Virtual address of first segment.
+ * @param va_L Virtual address of last segment (third segment if 3 segments,
+ * fourth segment if 4 segments).
+ * @param len_F_L Length of first segment in low 16 bits; length of last
+ * segment in high 16 bits.
+ * @param va_M0 Virtual address of "middle 0" segment; this segment is sent
+ * second when there are three segments, and third if there are four.
+ * @param va_M1 Virtual address of "middle 1" segment; this segment is sent
+ * second when there are four segments.
+ * @param len_M0_M1 Length of middle 0 segment in low 16 bits; length of middle
+ * 1 segment, if 4 segments, in high 16 bits.
+ */
+#define __netio_fastio_sendv_pkt_3_4(fastio_index, flags, confno, csum0, va_F, \
+ va_L, len_F_L, va_M0, va_M1, len_M0_M1) \
+ __netio_fastio9((fastio_index) + NETIO_FASTIO_SENDV_PKT, flags, confno, \
+ csum0, va_F, va_L, len_F_L, va_M0, va_M1, len_M0_M1)
+
+/** Send vector of packets.
+ * @param fastio_index Fast I/O index.
+ * @param seqno Number of packets transmitted so far on this interface;
+ * used to decide which packets should be acknowledged.
+ * @param nentries Number of entries in vector.
+ * @param va Virtual address of start of vector entry array.
+ * @return 3-word netio_fastio_rv3_t structure. The structure's err member
+ * is an error code, or zero if no error. The val0 member is the
+ * updated value of seqno; it has been incremented by 1 for each
+ * packet sent. That increment may be less than nentries if an
+ * error occured, or if some of the entries in the vector contain
+ * handles equal to NETIO_PKT_HANDLE_NONE. The val1 member is the
+ * updated value of nentries; it has been decremented by 1 for each
+ * vector entry processed. Again, that decrement may be less than
+ * nentries (leaving the returned value positive) if an error
+ * occurred.
+ */
+#define __netio_fastio_send_pkt_vec(fastio_index, seqno, nentries, va) \
+ __netio_fastio3_rv3((fastio_index) + NETIO_FASTIO_SEND_PKT_VEC, seqno, \
+ nentries, va)
+
+
+/** An egress DMA command for LEPP. */
+typedef struct
+{
+ /** Is this a TSO transfer?
+ *
+ * NOTE: This field is always 0, to distinguish it from
+ * lepp_tso_cmd_t. It must come first!
+ */
+ uint8_t tso : 1;
+
+ /** Unused padding bits. */
+ uint8_t _unused : 3;
+
+ /** Should this packet be sent directly from caches instead of DRAM,
+ * using hash-for-home to locate the packet data?
+ */
+ uint8_t hash_for_home : 1;
+
+ /** Should we compute a checksum? */
+ uint8_t compute_checksum : 1;
+
+ /** Is this the final buffer for this packet?
+ *
+ * A single packet can be split over several input buffers (a "gather"
+ * operation). This flag indicates that this is the last buffer
+ * in a packet.
+ */
+ uint8_t end_of_packet : 1;
+
+ /** Should LEPP advance 'comp_busy' when this DMA is fully finished? */
+ uint8_t send_completion : 1;
+
+ /** High bits of Client Physical Address of the start of the buffer
+ * to be egressed.
+ *
+ * NOTE: Only 6 bits are actually needed here, as CPAs are
+ * currently 38 bits. So two bits could be scavenged from this.
+ */
+ uint8_t cpa_hi;
+
+ /** The number of bytes to be egressed. */
+ uint16_t length;
+
+ /** Low 32 bits of Client Physical Address of the start of the buffer
+ * to be egressed.
+ */
+ uint32_t cpa_lo;
+
+ /** Checksum information (only used if 'compute_checksum'). */
+ __netio_checksum_header_t checksum_data;
+
+} lepp_cmd_t;
+
+
+/** A chunk of physical memory for a TSO egress. */
+typedef struct
+{
+ /** The low bits of the CPA. */
+ uint32_t cpa_lo;
+ /** The high bits of the CPA. */
+ uint16_t cpa_hi : 15;
+ /** Should this packet be sent directly from caches instead of DRAM,
+ * using hash-for-home to locate the packet data?
+ */
+ uint16_t hash_for_home : 1;
+ /** The length in bytes. */
+ uint16_t length;
+} lepp_frag_t;
+
+
+/** An LEPP command that handles TSO. */
+typedef struct
+{
+ /** Is this a TSO transfer?
+ *
+ * NOTE: This field is always 1, to distinguish it from
+ * lepp_cmd_t. It must come first!
+ */
+ uint8_t tso : 1;
+
+ /** Unused padding bits. */
+ uint8_t _unused : 7;
+
+ /** Size of the header[] array in bytes. It must be in the range
+ * [40, 127], which are the smallest header for a TCP packet over
+ * Ethernet and the maximum possible prepend size supported by
+ * hardware, respectively. Note that the array storage must be
+ * padded out to a multiple of four bytes so that the following
+ * LEPP command is aligned properly.
+ */
+ uint8_t header_size;
+
+ /** Byte offset of the IP header in header[]. */
+ uint8_t ip_offset;
+
+ /** Byte offset of the TCP header in header[]. */
+ uint8_t tcp_offset;
+
+ /** The number of bytes to use for the payload of each packet,
+ * except of course the last one, which may not have enough bytes.
+ * This means that each Ethernet packet except the last will have a
+ * size of header_size + payload_size.
+ */
+ uint16_t payload_size;
+
+ /** The length of the 'frags' array that follows this struct. */
+ uint16_t num_frags;
+
+ /** The actual frags. */
+ lepp_frag_t frags[0 /* Variable-sized; num_frags entries. */];
+
+ /*
+ * The packet header template logically follows frags[],
+ * but you can't declare that in C.
+ *
+ * uint32_t header[header_size_in_words_rounded_up];
+ */
+
+} lepp_tso_cmd_t;
+
+
+/** An LEPP completion ring entry. */
+typedef void* lepp_comp_t;
+
+
+/** Maximum number of frags for one TSO command. This is adapted from
+ * linux's "MAX_SKB_FRAGS", and presumably over-estimates by one, for
+ * our page size of exactly 65536. We add one for a "body" fragment.
+ */
+#define LEPP_MAX_FRAGS (65536 / HV_PAGE_SIZE_SMALL + 2 + 1)
+
+/** Total number of bytes needed for an lepp_tso_cmd_t. */
+#define LEPP_TSO_CMD_SIZE(num_frags, header_size) \
+ (sizeof(lepp_tso_cmd_t) + \
+ (num_frags) * sizeof(lepp_frag_t) + \
+ (((header_size) + 3) & -4))
+
+/** The size of the lepp "cmd" queue. */
+#define LEPP_CMD_QUEUE_BYTES \
+ (((CHIP_L2_CACHE_SIZE() - 2 * CHIP_L2_LINE_SIZE()) / \
+ (sizeof(lepp_cmd_t) + sizeof(lepp_comp_t))) * sizeof(lepp_cmd_t))
+
+/** The largest possible command that can go in lepp_queue_t::cmds[]. */
+#define LEPP_MAX_CMD_SIZE LEPP_TSO_CMD_SIZE(LEPP_MAX_FRAGS, 128)
+
+/** The largest possible value of lepp_queue_t::cmd_{head, tail} (inclusive).
+ */
+#define LEPP_CMD_LIMIT \
+ (LEPP_CMD_QUEUE_BYTES - LEPP_MAX_CMD_SIZE)
+
+/** The maximum number of completions in an LEPP queue. */
+#define LEPP_COMP_QUEUE_SIZE \
+ ((LEPP_CMD_LIMIT + sizeof(lepp_cmd_t) - 1) / sizeof(lepp_cmd_t))
+
+/** Increment an index modulo the queue size. */
+#define LEPP_QINC(var) \
+ (var = __insn_mnz(var - (LEPP_COMP_QUEUE_SIZE - 1), var + 1))
+
+/** A queue used to convey egress commands from the client to LEPP. */
+typedef struct
+{
+ /** Index of first completion not yet processed by user code.
+ * If this is equal to comp_busy, there are no such completions.
+ *
+ * NOTE: This is only read/written by the user.
+ */
+ unsigned int comp_head;
+
+ /** Index of first completion record not yet completed.
+ * If this is equal to comp_tail, there are no such completions.
+ * This index gets advanced (modulo LEPP_QUEUE_SIZE) whenever
+ * a command with the 'completion' bit set is finished.
+ *
+ * NOTE: This is only written by LEPP, only read by the user.
+ */
+ volatile unsigned int comp_busy;
+
+ /** Index of the first empty slot in the completion ring.
+ * Entries from this up to but not including comp_head (in ring order)
+ * can be filled in with completion data.
+ *
+ * NOTE: This is only read/written by the user.
+ */
+ unsigned int comp_tail;
+
+ /** Byte index of first command enqueued for LEPP but not yet processed.
+ *
+ * This is always divisible by sizeof(void*) and always <= LEPP_CMD_LIMIT.
+ *
+ * NOTE: LEPP advances this counter as soon as it no longer needs
+ * the cmds[] storage for this entry, but the transfer is not actually
+ * complete (i.e. the buffer pointed to by the command is no longer
+ * needed) until comp_busy advances.
+ *
+ * If this is equal to cmd_tail, the ring is empty.
+ *
+ * NOTE: This is only written by LEPP, only read by the user.
+ */
+ volatile unsigned int cmd_head;
+
+ /** Byte index of first empty slot in the command ring. This field can
+ * be incremented up to but not equal to cmd_head (because that would
+ * mean the ring is empty).
+ *
+ * This is always divisible by sizeof(void*) and always <= LEPP_CMD_LIMIT.
+ *
+ * NOTE: This is read/written by the user, only read by LEPP.
+ */
+ volatile unsigned int cmd_tail;
+
+ /** A ring of variable-sized egress DMA commands.
+ *
+ * NOTE: Only written by the user, only read by LEPP.
+ */
+ char cmds[LEPP_CMD_QUEUE_BYTES]
+ __attribute__((aligned(CHIP_L2_LINE_SIZE())));
+
+ /** A ring of user completion data.
+ * NOTE: Only read/written by the user.
+ */
+ lepp_comp_t comps[LEPP_COMP_QUEUE_SIZE]
+ __attribute__((aligned(CHIP_L2_LINE_SIZE())));
+} lepp_queue_t;
+
+
+/** An internal helper function for determining the number of entries
+ * available in a ring buffer, given that there is one sentinel.
+ */
+static inline unsigned int
+_lepp_num_free_slots(unsigned int head, unsigned int tail)
+{
+ /*
+ * One entry is reserved for use as a sentinel, to distinguish
+ * "empty" from "full". So we compute
+ * (head - tail - 1) % LEPP_QUEUE_SIZE, but without using a slow % operation.
+ */
+ return (head - tail - 1) + ((head <= tail) ? LEPP_COMP_QUEUE_SIZE : 0);
+}
+
+
+/** Returns how many new comp entries can be enqueued. */
+static inline unsigned int
+lepp_num_free_comp_slots(const lepp_queue_t* q)
+{
+ return _lepp_num_free_slots(q->comp_head, q->comp_tail);
+}
+
+static inline int
+lepp_qsub(int v1, int v2)
+{
+ int delta = v1 - v2;
+ return delta + ((delta >> 31) & LEPP_COMP_QUEUE_SIZE);
+}
+
+
+/** FIXME: Check this from linux, via a new "pwrite()" call. */
+#define LIPP_VERSION 1
+
+
+/** We use exactly two bytes of alignment padding. */
+#define LIPP_PACKET_PADDING 2
+
+/** The minimum size of a "small" buffer (including the padding). */
+#define LIPP_SMALL_PACKET_SIZE 128
+
+/*
+ * NOTE: The following two values should total to less than around
+ * 13582, to keep the total size used for "lipp_state_t" below 64K.
+ */
+
+/** The maximum number of "small" buffers.
+ * This is enough for 53 network cpus with 128 credits. Note that
+ * if these are exhausted, we will fall back to using large buffers.
+ */
+#define LIPP_SMALL_BUFFERS 6785
+
+/** The maximum number of "large" buffers.
+ * This is enough for 53 network cpus with 128 credits.
+ */
+#define LIPP_LARGE_BUFFERS 6785
+
+#endif /* __DRV_XGBE_INTF_H__ */
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+/**
+ * Error codes returned from NetIO routines.
+ */
+
+#ifndef __NETIO_ERRORS_H__
+#define __NETIO_ERRORS_H__
+
+/**
+ * @addtogroup error
+ *
+ * @brief The error codes returned by NetIO functions.
+ *
+ * NetIO functions return 0 (defined as ::NETIO_NO_ERROR) on success, and
+ * a negative value if an error occurs.
+ *
+ * In cases where a NetIO function failed due to a error reported by
+ * system libraries, the error code will be the negation of the
+ * system errno at the time of failure. The @ref netio_strerror()
+ * function will deliver error strings for both NetIO and system error
+ * codes.
+ *
+ * @{
+ */
+
+/** The set of all NetIO errors. */
+typedef enum
+{
+ /** Operation successfully completed. */
+ NETIO_NO_ERROR = 0,
+
+ /** A packet was successfully retrieved from an input queue. */
+ NETIO_PKT = 0,
+
+ /** Largest NetIO error number. */
+ NETIO_ERR_MAX = -701,
+
+ /** The tile is not registered with the IPP. */
+ NETIO_NOT_REGISTERED = -701,
+
+ /** No packet was available to retrieve from the input queue. */
+ NETIO_NOPKT = -702,
+
+ /** The requested function is not implemented. */
+ NETIO_NOT_IMPLEMENTED = -703,
+
+ /** On a registration operation, the target queue already has the maximum
+ * number of tiles registered for it, and no more may be added. On a
+ * packet send operation, the output queue is full and nothing more can
+ * be queued until some of the queued packets are actually transmitted. */
+ NETIO_QUEUE_FULL = -704,
+
+ /** The calling process or thread is not bound to exactly one CPU. */
+ NETIO_BAD_AFFINITY = -705,
+
+ /** Cannot allocate memory on requested controllers. */
+ NETIO_CANNOT_HOME = -706,
+
+ /** On a registration operation, the IPP specified is not configured
+ * to support the options requested; for instance, the application
+ * wants a specific type of tagged headers which the configured IPP
+ * doesn't support. Or, the supplied configuration information is
+ * not self-consistent, or is out of range; for instance, specifying
+ * both NETIO_RECV and NETIO_NO_RECV, or asking for more than
+ * NETIO_MAX_SEND_BUFFERS to be preallocated. On a VLAN or bucket
+ * configure operation, the number of items, or the base item, was
+ * out of range.
+ */
+ NETIO_BAD_CONFIG = -707,
+
+ /** Too many tiles have registered to transmit packets. */
+ NETIO_TOOMANY_XMIT = -708,
+
+ /** Packet transmission was attempted on a queue which was registered
+ with transmit disabled. */
+ NETIO_UNREG_XMIT = -709,
+
+ /** This tile is already registered with the IPP. */
+ NETIO_ALREADY_REGISTERED = -710,
+
+ /** The Ethernet link is down. The application should try again later. */
+ NETIO_LINK_DOWN = -711,
+
+ /** An invalid memory buffer has been specified. This may be an unmapped
+ * virtual address, or one which does not meet alignment requirements.
+ * For netio_input_register(), this error may be returned when multiple
+ * processes specify different memory regions to be used for NetIO
+ * buffers. That can happen if these processes specify explicit memory
+ * regions with the ::NETIO_FIXED_BUFFER_VA flag, or if tmc_cmem_init()
+ * has not been called by a common ancestor of the processes.
+ */
+ NETIO_FAULT = -712,
+
+ /** Cannot combine user-managed shared memory and cache coherence. */
+ NETIO_BAD_CACHE_CONFIG = -713,
+
+ /** Smallest NetIO error number. */
+ NETIO_ERR_MIN = -713,
+
+#ifndef __DOXYGEN__
+ /** Used internally to mean that no response is needed; never returned to
+ * an application. */
+ NETIO_NO_RESPONSE = 1
+#endif
+} netio_error_t;
+
+/** @} */
+
+#endif /* __NETIO_ERRORS_H__ */
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+/**
+ * NetIO interface structures and macros.
+ */
+
+#ifndef __NETIO_INTF_H__
+#define __NETIO_INTF_H__
+
+#include <hv/netio_errors.h>
+
+#ifdef __KERNEL__
+#include <linux/types.h>
+#else
+#include <stdint.h>
+#endif
+
+#if !defined(__HV__) && !defined(__BOGUX__) && !defined(__KERNEL__)
+#include <assert.h>
+#define netio_assert assert /**< Enable assertions from macros */
+#else
+#define netio_assert(...) ((void)(0)) /**< Disable assertions from macros */
+#endif
+
+/*
+ * If none of these symbols are defined, we're building libnetio in an
+ * environment where we have pthreads, so we'll enable locking.
+ */
+#if !defined(__HV__) && !defined(__BOGUX__) && !defined(__KERNEL__) && \
+ !defined(__NEWLIB__)
+#define _NETIO_PTHREAD /**< Include a mutex in netio_queue_t below */
+
+/*
+ * If NETIO_UNLOCKED is defined, we don't do use per-cpu locks on
+ * per-packet NetIO operations. We still do pthread locking on things
+ * like netio_input_register, though. This is used for building
+ * libnetio_unlocked.
+ */
+#ifndef NETIO_UNLOCKED
+
+/* Avoid PLT overhead by using our own inlined per-cpu lock. */
+#include <sched.h>
+typedef int _netio_percpu_mutex_t;
+
+static __inline int
+_netio_percpu_mutex_init(_netio_percpu_mutex_t* lock)
+{
+ *lock = 0;
+ return 0;
+}
+
+static __inline int
+_netio_percpu_mutex_lock(_netio_percpu_mutex_t* lock)
+{
+ while (__builtin_expect(__insn_tns(lock), 0))
+ sched_yield();
+ return 0;
+}
+
+static __inline int
+_netio_percpu_mutex_unlock(_netio_percpu_mutex_t* lock)
+{
+ *lock = 0;
+ return 0;
+}
+
+#else /* NETIO_UNLOCKED */
+
+/* Don't do any locking for per-packet NetIO operations. */
+typedef int _netio_percpu_mutex_t;
+#define _netio_percpu_mutex_init(L)
+#define _netio_percpu_mutex_lock(L)
+#define _netio_percpu_mutex_unlock(L)
+
+#endif /* NETIO_UNLOCKED */
+#endif /* !__HV__, !__BOGUX, !__KERNEL__, !__NEWLIB__ */
+
+/** How many tiles can register for a given queue.
+ * @ingroup setup */
+#define NETIO_MAX_TILES_PER_QUEUE 64
+
+
+/** Largest permissible queue identifier.
+ * @ingroup setup */
+#define NETIO_MAX_QUEUE_ID 255
+
+
+#ifndef __DOXYGEN__
+
+/* Metadata packet checksum/ethertype flags. */
+
+/** The L4 checksum has not been calculated. */
+#define _NETIO_PKT_NO_L4_CSUM_SHIFT 0
+#define _NETIO_PKT_NO_L4_CSUM_RMASK 1
+#define _NETIO_PKT_NO_L4_CSUM_MASK \
+ (_NETIO_PKT_NO_L4_CSUM_RMASK << _NETIO_PKT_NO_L4_CSUM_SHIFT)
+
+/** The L3 checksum has not been calculated. */
+#define _NETIO_PKT_NO_L3_CSUM_SHIFT 1
+#define _NETIO_PKT_NO_L3_CSUM_RMASK 1
+#define _NETIO_PKT_NO_L3_CSUM_MASK \
+ (_NETIO_PKT_NO_L3_CSUM_RMASK << _NETIO_PKT_NO_L3_CSUM_SHIFT)
+
+/** The L3 checksum is incorrect (or perhaps has not been calculated). */
+#define _NETIO_PKT_BAD_L3_CSUM_SHIFT 2
+#define _NETIO_PKT_BAD_L3_CSUM_RMASK 1
+#define _NETIO_PKT_BAD_L3_CSUM_MASK \
+ (_NETIO_PKT_BAD_L3_CSUM_RMASK << _NETIO_PKT_BAD_L3_CSUM_SHIFT)
+
+/** The Ethernet packet type is unrecognized. */
+#define _NETIO_PKT_TYPE_UNRECOGNIZED_SHIFT 3
+#define _NETIO_PKT_TYPE_UNRECOGNIZED_RMASK 1
+#define _NETIO_PKT_TYPE_UNRECOGNIZED_MASK \
+ (_NETIO_PKT_TYPE_UNRECOGNIZED_RMASK << \
+ _NETIO_PKT_TYPE_UNRECOGNIZED_SHIFT)
+
+/* Metadata packet type flags. */
+
+/** Where the packet type bits are; this field is the index into
+ * _netio_pkt_info. */
+#define _NETIO_PKT_TYPE_SHIFT 4
+#define _NETIO_PKT_TYPE_RMASK 0x3F
+
+/** How many VLAN tags the packet has, and, if we have two, which one we
+ * actually grouped on. A VLAN within a proprietary (Marvell or Broadcom)
+ * tag is counted here. */
+#define _NETIO_PKT_VLAN_SHIFT 4
+#define _NETIO_PKT_VLAN_RMASK 0x3
+#define _NETIO_PKT_VLAN_MASK \
+ (_NETIO_PKT_VLAN_RMASK << _NETIO_PKT_VLAN_SHIFT)
+#define _NETIO_PKT_VLAN_NONE 0 /* No VLAN tag. */
+#define _NETIO_PKT_VLAN_ONE 1 /* One VLAN tag. */
+#define _NETIO_PKT_VLAN_TWO_OUTER 2 /* Two VLAN tags, outer one used. */
+#define _NETIO_PKT_VLAN_TWO_INNER 3 /* Two VLAN tags, inner one used. */
+
+/** Which proprietary tags the packet has. */
+#define _NETIO_PKT_TAG_SHIFT 6
+#define _NETIO_PKT_TAG_RMASK 0x3
+#define _NETIO_PKT_TAG_MASK \
+ (_NETIO_PKT_TAG_RMASK << _NETIO_PKT_TAG_SHIFT)
+#define _NETIO_PKT_TAG_NONE 0 /* No proprietary tags. */
+#define _NETIO_PKT_TAG_MRVL 1 /* Marvell HyperG.Stack tags. */
+#define _NETIO_PKT_TAG_MRVL_EXT 2 /* HyperG.Stack extended tags. */
+#define _NETIO_PKT_TAG_BRCM 3 /* Broadcom HiGig tags. */
+
+/** Whether a packet has an LLC + SNAP header. */
+#define _NETIO_PKT_SNAP_SHIFT 8
+#define _NETIO_PKT_SNAP_RMASK 0x1
+#define _NETIO_PKT_SNAP_MASK \
+ (_NETIO_PKT_SNAP_RMASK << _NETIO_PKT_SNAP_SHIFT)
+
+/* NOTE: Bits 9 and 10 are unused. */
+
+/** Length of any custom data before the L2 header, in words. */
+#define _NETIO_PKT_CUSTOM_LEN_SHIFT 11
+#define _NETIO_PKT_CUSTOM_LEN_RMASK 0x1F
+#define _NETIO_PKT_CUSTOM_LEN_MASK \
+ (_NETIO_PKT_CUSTOM_LEN_RMASK << _NETIO_PKT_CUSTOM_LEN_SHIFT)
+
+/** The L4 checksum is incorrect (or perhaps has not been calculated). */
+#define _NETIO_PKT_BAD_L4_CSUM_SHIFT 16
+#define _NETIO_PKT_BAD_L4_CSUM_RMASK 0x1
+#define _NETIO_PKT_BAD_L4_CSUM_MASK \
+ (_NETIO_PKT_BAD_L4_CSUM_RMASK << _NETIO_PKT_BAD_L4_CSUM_SHIFT)
+
+/** Length of the L2 header, in words. */
+#define _NETIO_PKT_L2_LEN_SHIFT 17
+#define _NETIO_PKT_L2_LEN_RMASK 0x1F
+#define _NETIO_PKT_L2_LEN_MASK \
+ (_NETIO_PKT_L2_LEN_RMASK << _NETIO_PKT_L2_LEN_SHIFT)
+
+
+/* Flags in minimal packet metadata. */
+
+/** We need an eDMA checksum on this packet. */
+#define _NETIO_PKT_NEED_EDMA_CSUM_SHIFT 0
+#define _NETIO_PKT_NEED_EDMA_CSUM_RMASK 1
+#define _NETIO_PKT_NEED_EDMA_CSUM_MASK \
+ (_NETIO_PKT_NEED_EDMA_CSUM_RMASK << _NETIO_PKT_NEED_EDMA_CSUM_SHIFT)
+
+/* Data within the packet information table. */
+
+/* Note that, for efficiency, code which uses these fields assumes that none
+ * of the shift values below are zero. See uses below for an explanation. */
+
+/** Offset within the L2 header of the innermost ethertype (in halfwords). */
+#define _NETIO_PKT_INFO_ETYPE_SHIFT 6
+#define _NETIO_PKT_INFO_ETYPE_RMASK 0x1F
+
+/** Offset within the L2 header of the VLAN tag (in halfwords). */
+#define _NETIO_PKT_INFO_VLAN_SHIFT 11
+#define _NETIO_PKT_INFO_VLAN_RMASK 0x1F
+
+#endif
+
+
+/** The size of a memory buffer representing a small packet.
+ * @ingroup egress */
+#define SMALL_PACKET_SIZE 256
+
+/** The size of a memory buffer representing a large packet.
+ * @ingroup egress */
+#define LARGE_PACKET_SIZE 2048
+
+/** The size of a memory buffer representing a jumbo packet.
+ * @ingroup egress */
+#define JUMBO_PACKET_SIZE (12 * 1024)
+
+
+/* Common ethertypes.
+ * @ingroup ingress */
+/** @{ */
+/** The ethertype of IPv4. */
+#define ETHERTYPE_IPv4 (0x0800)
+/** The ethertype of ARP. */
+#define ETHERTYPE_ARP (0x0806)
+/** The ethertype of VLANs. */
+#define ETHERTYPE_VLAN (0x8100)
+/** The ethertype of a Q-in-Q header. */
+#define ETHERTYPE_Q_IN_Q (0x9100)
+/** The ethertype of IPv6. */
+#define ETHERTYPE_IPv6 (0x86DD)
+/** The ethertype of MPLS. */
+#define ETHERTYPE_MPLS (0x8847)
+/** @} */
+
+
+/** The possible return values of NETIO_PKT_STATUS.
+ * @ingroup ingress
+ */
+typedef enum
+{
+ /** No problems were detected with this packet. */
+ NETIO_PKT_STATUS_OK,
+ /** The packet is undersized; this is expected behavior if the packet's
+ * ethertype is unrecognized, but otherwise the packet is likely corrupt. */
+ NETIO_PKT_STATUS_UNDERSIZE,
+ /** The packet is oversized and some trailing bytes have been discarded.
+ This is expected behavior for short packets, since it's impossible to
+ precisely determine the amount of padding which may have been added to
+ them to make them meet the minimum Ethernet packet size. */
+ NETIO_PKT_STATUS_OVERSIZE,
+ /** The packet was judged to be corrupt by hardware (for instance, it had
+ a bad CRC, or part of it was discarded due to lack of buffer space in
+ the I/O shim) and should be discarded. */
+ NETIO_PKT_STATUS_BAD
+} netio_pkt_status_t;
+
+
+/** Log2 of how many buckets we have. */
+#define NETIO_LOG2_NUM_BUCKETS (10)
+
+/** How many buckets we have.
+ * @ingroup ingress */
+#define NETIO_NUM_BUCKETS (1 << NETIO_LOG2_NUM_BUCKETS)
+
+
+/**
+ * @brief A group-to-bucket identifier.
+ *
+ * @ingroup setup
+ *
+ * This tells us what to do with a given group.
+ */
+typedef union {
+ /** The header broken down into bits. */
+ struct {
+ /** Whether we should balance on L4, if available */
+ unsigned int __balance_on_l4:1;
+ /** Whether we should balance on L3, if available */
+ unsigned int __balance_on_l3:1;
+ /** Whether we should balance on L2, if available */
+ unsigned int __balance_on_l2:1;
+ /** Reserved for future use */
+ unsigned int __reserved:1;
+ /** The base bucket to use to send traffic */
+ unsigned int __bucket_base:NETIO_LOG2_NUM_BUCKETS;
+ /** The mask to apply to the balancing value. This must be one less
+ * than a power of two, e.g. 0x3 or 0xFF.
+ */
+ unsigned int __bucket_mask:NETIO_LOG2_NUM_BUCKETS;
+ /** Pad to 32 bits */
+ unsigned int __padding:(32 - 4 - 2 * NETIO_LOG2_NUM_BUCKETS);
+ } bits;
+ /** To send out the IDN. */
+ unsigned int word;
+}
+netio_group_t;
+
+
+/**
+ * @brief A VLAN-to-bucket identifier.
+ *
+ * @ingroup setup
+ *
+ * This tells us what to do with a given VLAN.
+ */
+typedef netio_group_t netio_vlan_t;
+
+
+/**
+ * A bucket-to-queue mapping.
+ * @ingroup setup
+ */
+typedef unsigned char netio_bucket_t;
+
+
+/**
+ * A packet size can always fit in a netio_size_t.
+ * @ingroup setup
+ */
+typedef unsigned int netio_size_t;
+
+
+/**
+ * @brief Ethernet standard (ingress) packet metadata.
+ *
+ * @ingroup ingress
+ *
+ * This is additional data associated with each packet.
+ * This structure is opaque and accessed through the @ref ingress.
+ *
+ * Also, the buffer population operation currently assumes that standard
+ * metadata is at least as large as minimal metadata, and will need to be
+ * modified if that is no longer the case.
+ */
+typedef struct
+{
+#ifdef __DOXYGEN__
+ /** This structure is opaque. */
+ unsigned char opaque[24];
+#else
+ /** The overall ordinal of the packet */
+ unsigned int __packet_ordinal;
+ /** The ordinal of the packet within the group */
+ unsigned int __group_ordinal;
+ /** The best flow hash IPP could compute. */
+ unsigned int __flow_hash;
+ /** Flags pertaining to checksum calculation, packet type, etc. */
+ unsigned int __flags;
+ /** The first word of "user data". */
+ unsigned int __user_data_0;
+ /** The second word of "user data". */
+ unsigned int __user_data_1;
+#endif
+}
+netio_pkt_metadata_t;
+
+
+/** To ensure that the L3 header is aligned mod 4, the L2 header should be
+ * aligned mod 4 plus 2, since every supported L2 header is 4n + 2 bytes
+ * long. The standard way to do this is to simply add 2 bytes of padding
+ * before the L2 header.
+ */
+#define NETIO_PACKET_PADDING 2
+
+
+
+/**
+ * @brief Ethernet minimal (egress) packet metadata.
+ *
+ * @ingroup egress
+ *
+ * This structure represents information about packets which have
+ * been processed by @ref netio_populate_buffer() or
+ * @ref netio_populate_prepend_buffer(). This structure is opaque
+ * and accessed through the @ref egress.
+ *
+ * @internal This structure is actually copied into the memory used by
+ * standard metadata, which is assumed to be large enough.
+ */
+typedef struct
+{
+#ifdef __DOXYGEN__
+ /** This structure is opaque. */
+ unsigned char opaque[14];
+#else
+ /** The offset of the L2 header from the start of the packet data. */
+ unsigned short l2_offset;
+ /** The offset of the L3 header from the start of the packet data. */
+ unsigned short l3_offset;
+ /** Where to write the checksum. */
+ unsigned char csum_location;
+ /** Where to start checksumming from. */
+ unsigned char csum_start;
+ /** Flags pertaining to checksum calculation etc. */
+ unsigned short flags;
+ /** The L2 length of the packet. */
+ unsigned short l2_length;
+ /** The checksum with which to seed the checksum generator. */
+ unsigned short csum_seed;
+ /** How much to checksum. */
+ unsigned short csum_length;
+#endif
+}
+netio_pkt_minimal_metadata_t;
+
+
+#ifndef __DOXYGEN__
+
+/**
+ * @brief An I/O notification header.
+ *
+ * This is the first word of data received from an I/O shim in a notification
+ * packet. It contains framing and status information.
+ */
+typedef union
+{
+ unsigned int word; /**< The whole word. */
+ /** The various fields. */
+ struct
+ {
+ unsigned int __channel:7; /**< Resource channel. */
+ unsigned int __type:4; /**< Type. */
+ unsigned int __ack:1; /**< Whether an acknowledgement is needed. */
+ unsigned int __reserved:1; /**< Reserved. */
+ unsigned int __protocol:1; /**< A protocol-specific word is added. */
+ unsigned int __status:2; /**< Status of the transfer. */
+ unsigned int __framing:2; /**< Framing of the transfer. */
+ unsigned int __transfer_size:14; /**< Transfer size in bytes (total). */
+ } bits;
+}
+__netio_pkt_notif_t;
+
+
+/**
+ * Returns the base address of the packet.
+ */
+#define _NETIO_PKT_HANDLE_BASE(p) \
+ ((unsigned char*)((p).word & 0xFFFFFFC0))
+
+/**
+ * Returns the base address of the packet.
+ */
+#define _NETIO_PKT_BASE(p) \
+ _NETIO_PKT_HANDLE_BASE(p->__packet)
+
+/**
+ * @brief An I/O notification packet (second word)
+ *
+ * This is the second word of data received from an I/O shim in a notification
+ * packet. This is the virtual address of the packet buffer, plus some flag
+ * bits. (The virtual address of the packet is always 256-byte aligned so we
+ * have room for 8 bits' worth of flags in the low 8 bits.)
+ *
+ * @internal
+ * NOTE: The low two bits must contain "__queue", so the "packet size"
+ * (SIZE_SMALL, SIZE_LARGE, or SIZE_JUMBO) can be determined quickly.
+ *
+ * If __addr or __offset are moved, _NETIO_PKT_BASE
+ * (defined right below this) must be changed.
+ */
+typedef union
+{
+ unsigned int word; /**< The whole word. */
+ /** The various fields. */
+ struct
+ {
+ /** Which queue the packet will be returned to once it is sent back to
+ the IPP. This is one of the SIZE_xxx values. */
+ unsigned int __queue:2;
+
+ /** The IPP handle of the sending IPP. */
+ unsigned int __ipp_handle:2;
+
+ /** Reserved for future use. */
+ unsigned int __reserved:1;
+
+ /** If 1, this packet has minimal (egress) metadata; otherwise, it
+ has standard (ingress) metadata. */
+ unsigned int __minimal:1;
+
+ /** Offset of the metadata within the packet. This value is multiplied
+ * by 64 and added to the base packet address to get the metadata
+ * address. Note that this field is aligned within the word such that
+ * you can easily extract the metadata address with a 26-bit mask. */
+ unsigned int __offset:2;
+
+ /** The top 24 bits of the packet's virtual address. */
+ unsigned int __addr:24;
+ } bits;
+}
+__netio_pkt_handle_t;
+
+#endif /* !__DOXYGEN__ */
+
+
+/**
+ * @brief A handle for an I/O packet's storage.
+ * @ingroup ingress
+ *
+ * netio_pkt_handle_t encodes the concept of a ::netio_pkt_t with its
+ * packet metadata removed. It is a much smaller type that exists to
+ * facilitate applications where the full ::netio_pkt_t type is too
+ * large, such as those that cache enormous numbers of packets or wish
+ * to transmit packet descriptors over the UDN.
+ *
+ * Because there is no metadata, most ::netio_pkt_t operations cannot be
+ * performed on a netio_pkt_handle_t. It supports only
+ * netio_free_handle() (to free the buffer) and
+ * NETIO_PKT_CUSTOM_DATA_H() (to access a pointer to its contents).
+ * The application must acquire any additional metadata it wants from the
+ * original ::netio_pkt_t and record it separately.
+ *
+ * A netio_pkt_handle_t can be extracted from a ::netio_pkt_t by calling
+ * NETIO_PKT_HANDLE(). An invalid handle (analogous to NULL) can be
+ * created by assigning the value ::NETIO_PKT_HANDLE_NONE. A handle can
+ * be tested for validity with NETIO_PKT_HANDLE_IS_VALID().
+ */
+typedef struct
+{
+ unsigned int word; /**< Opaque bits. */
+} netio_pkt_handle_t;
+
+/**
+ * @brief A packet descriptor.
+ *
+ * @ingroup ingress
+ * @ingroup egress
+ *
+ * This data structure represents a packet. The structure is manipulated
+ * through the @ref ingress and the @ref egress.
+ *
+ * While the contents of a netio_pkt_t are opaque, the structure itself is
+ * portable. This means that it may be shared between all tiles which have
+ * done a netio_input_register() call for the interface on which the pkt_t
+ * was initially received (via netio_get_packet()) or retrieved (via
+ * netio_get_buffer()). The contents of a netio_pkt_t can be transmitted to
+ * another tile via shared memory, or via a UDN message, or by other means.
+ * The destination tile may then use the pkt_t as if it had originally been
+ * received locally; it may read or write the packet's data, read its
+ * metadata, free the packet, send the packet, transfer the netio_pkt_t to
+ * yet another tile, and so forth.
+ *
+ * Once a netio_pkt_t has been transferred to a second tile, the first tile
+ * should not reference the original copy; in particular, if more than one
+ * tile frees or sends the same netio_pkt_t, the IPP's packet free lists will
+ * become corrupted. Note also that each tile which reads or modifies
+ * packet data must obey the memory coherency rules outlined in @ref input.
+ */
+typedef struct
+{
+#ifdef __DOXYGEN__
+ /** This structure is opaque. */
+ unsigned char opaque[32];
+#else
+ /** For an ingress packet (one with standard metadata), this is the
+ * notification header we got from the I/O shim. For an egress packet
+ * (one with minimal metadata), this word is zero if the packet has not
+ * been populated, and nonzero if it has. */
+ __netio_pkt_notif_t __notif_header;
+
+ /** Virtual address of the packet buffer, plus state flags. */
+ __netio_pkt_handle_t __packet;
+
+ /** Metadata associated with the packet. */
+ netio_pkt_metadata_t __metadata;
+#endif
+}
+netio_pkt_t;
+
+
+#ifndef __DOXYGEN__
+
+#define __NETIO_PKT_NOTIF_HEADER(pkt) ((pkt)->__notif_header)
+#define __NETIO_PKT_IPP_HANDLE(pkt) ((pkt)->__packet.bits.__ipp_handle)
+#define __NETIO_PKT_QUEUE(pkt) ((pkt)->__packet.bits.__queue)
+#define __NETIO_PKT_NOTIF_HEADER_M(mda, pkt) ((pkt)->__notif_header)
+#define __NETIO_PKT_IPP_HANDLE_M(mda, pkt) ((pkt)->__packet.bits.__ipp_handle)
+#define __NETIO_PKT_MINIMAL(pkt) ((pkt)->__packet.bits.__minimal)
+#define __NETIO_PKT_QUEUE_M(mda, pkt) ((pkt)->__packet.bits.__queue)
+#define __NETIO_PKT_FLAGS_M(mda, pkt) ((mda)->__flags)
+
+/* Packet information table, used by the attribute access functions below. */
+extern const uint16_t _netio_pkt_info[];
+
+#endif /* __DOXYGEN__ */
+
+
+#ifndef __DOXYGEN__
+/* These macros are deprecated and will disappear in a future MDE release. */
+#define NETIO_PKT_GOOD_CHECKSUM(pkt) \
+ NETIO_PKT_L4_CSUM_CORRECT(pkt)
+#define NETIO_PKT_GOOD_CHECKSUM_M(mda, pkt) \
+ NETIO_PKT_L4_CSUM_CORRECT_M(mda, pkt)
+#endif /* __DOXYGEN__ */
+
+
+/* Packet attribute access functions. */
+
+/** Return a pointer to the metadata for a packet.
+ * @ingroup ingress
+ *
+ * Calling this function once and passing the result to other retrieval
+ * functions with a "_M" suffix usually improves performance. This
+ * function must be called on an 'ingress' packet (i.e. one retrieved
+ * by @ref netio_get_packet(), on which @ref netio_populate_buffer() or
+ * @ref netio_populate_prepend_buffer have not been called). Use of this
+ * function on an 'egress' packet will cause an assertion failure.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to the packet's standard metadata.
+ */
+static __inline netio_pkt_metadata_t*
+NETIO_PKT_METADATA(netio_pkt_t* pkt)
+{
+ netio_assert(!pkt->__packet.bits.__minimal);
+ return &pkt->__metadata;
+}
+
+
+/** Return a pointer to the minimal metadata for a packet.
+ * @ingroup egress
+ *
+ * Calling this function once and passing the result to other retrieval
+ * functions with a "_MM" suffix usually improves performance. This
+ * function must be called on an 'egress' packet (i.e. one on which
+ * @ref netio_populate_buffer() or @ref netio_populate_prepend_buffer()
+ * have been called, or one retrieved by @ref netio_get_buffer()). Use of
+ * this function on an 'ingress' packet will cause an assertion failure.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to the packet's standard metadata.
+ */
+static __inline netio_pkt_minimal_metadata_t*
+NETIO_PKT_MINIMAL_METADATA(netio_pkt_t* pkt)
+{
+ netio_assert(pkt->__packet.bits.__minimal);
+ return (netio_pkt_minimal_metadata_t*) &pkt->__metadata;
+}
+
+
+/** Determine whether a packet has 'minimal' metadata.
+ * @ingroup pktfuncs
+ *
+ * This function will return nonzero if the packet is an 'egress'
+ * packet (i.e. one on which @ref netio_populate_buffer() or
+ * @ref netio_populate_prepend_buffer() have been called, or one
+ * retrieved by @ref netio_get_buffer()), and zero if the packet
+ * is an 'ingress' packet (i.e. one retrieved by @ref netio_get_packet(),
+ * which has not been converted into an 'egress' packet).
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the packet has minimal metadata.
+ */
+static __inline unsigned int
+NETIO_PKT_IS_MINIMAL(netio_pkt_t* pkt)
+{
+ return pkt->__packet.bits.__minimal;
+}
+
+
+/** Return a handle for a packet's storage.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A handle for the packet's storage.
+ */
+static __inline netio_pkt_handle_t
+NETIO_PKT_HANDLE(netio_pkt_t* pkt)
+{
+ netio_pkt_handle_t h;
+ h.word = pkt->__packet.word;
+ return h;
+}
+
+
+/** A special reserved value indicating the absence of a packet handle.
+ *
+ * @ingroup pktfuncs
+ */
+#define NETIO_PKT_HANDLE_NONE ((netio_pkt_handle_t) { 0 })
+
+
+/** Test whether a packet handle is valid.
+ *
+ * Applications may wish to use the reserved value NETIO_PKT_HANDLE_NONE
+ * to indicate no packet at all. This function tests to see if a packet
+ * handle is a real handle, not this special reserved value.
+ *
+ * @ingroup pktfuncs
+ *
+ * @param[in] handle Handle on which to operate.
+ * @return One if the packet handle is valid, else zero.
+ */
+static __inline unsigned int
+NETIO_PKT_HANDLE_IS_VALID(netio_pkt_handle_t handle)
+{
+ return handle.word != 0;
+}
+
+
+
+/** Return a pointer to the start of the packet's custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup ingress
+ *
+ * @param[in] handle Handle on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_CUSTOM_DATA_H(netio_pkt_handle_t handle)
+{
+ return _NETIO_PKT_HANDLE_BASE(handle) + NETIO_PACKET_PADDING;
+}
+
+
+/** Return the length of the packet's custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ *
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet's custom header, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_CUSTOM_HEADER_LENGTH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ /*
+ * Note that we effectively need to extract a quantity from the flags word
+ * which is measured in words, and then turn it into bytes by shifting
+ * it left by 2. We do this all at once by just shifting right two less
+ * bits, and shifting the mask up two bits.
+ */
+ return ((mda->__flags >> (_NETIO_PKT_CUSTOM_LEN_SHIFT - 2)) &
+ (_NETIO_PKT_CUSTOM_LEN_RMASK << 2));
+}
+
+
+/** Return the length of the packet, starting with the custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_CUSTOM_LENGTH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (__NETIO_PKT_NOTIF_HEADER(pkt).bits.__transfer_size -
+ NETIO_PACKET_PADDING);
+}
+
+
+/** Return a pointer to the start of the packet's custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_CUSTOM_DATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return NETIO_PKT_CUSTOM_DATA_H(NETIO_PKT_HANDLE(pkt));
+}
+
+
+/** Return the length of the packet's L2 (Ethernet plus VLAN or SNAP) header.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet's L2 header, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_HEADER_LENGTH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ /*
+ * Note that we effectively need to extract a quantity from the flags word
+ * which is measured in words, and then turn it into bytes by shifting
+ * it left by 2. We do this all at once by just shifting right two less
+ * bits, and shifting the mask up two bits. We then add two bytes.
+ */
+ return ((mda->__flags >> (_NETIO_PKT_L2_LEN_SHIFT - 2)) &
+ (_NETIO_PKT_L2_LEN_RMASK << 2)) + 2;
+}
+
+
+/** Return the length of the packet, starting with the L2 (Ethernet) header.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_LENGTH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (NETIO_PKT_CUSTOM_LENGTH_M(mda, pkt) -
+ NETIO_PKT_CUSTOM_HEADER_LENGTH_M(mda,pkt));
+}
+
+
+/** Return a pointer to the start of the packet's L2 (Ethernet) header.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_L2_DATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (NETIO_PKT_CUSTOM_DATA_M(mda, pkt) +
+ NETIO_PKT_CUSTOM_HEADER_LENGTH_M(mda, pkt));
+}
+
+
+/** Retrieve the length of the packet, starting with the L3 (generally,
+ * the IP) header.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Length of the packet's L3 header and data, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L3_LENGTH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (NETIO_PKT_L2_LENGTH_M(mda, pkt) -
+ NETIO_PKT_L2_HEADER_LENGTH_M(mda,pkt));
+}
+
+
+/** Return a pointer to the packet's L3 (generally, the IP) header.
+ * @ingroup ingress
+ *
+ * Note that we guarantee word alignment of the L3 header.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to the packet's L3 header.
+ */
+static __inline unsigned char*
+NETIO_PKT_L3_DATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (NETIO_PKT_L2_DATA_M(mda, pkt) +
+ NETIO_PKT_L2_HEADER_LENGTH_M(mda, pkt));
+}
+
+
+/** Return the ordinal of the packet.
+ * @ingroup ingress
+ *
+ * Each packet is given an ordinal number when it is delivered by the IPP.
+ * In the medium term, the ordinal is unique and monotonically increasing,
+ * being incremented by 1 for each packet; the ordinal of the first packet
+ * delivered after the IPP starts is zero. (Since the ordinal is of finite
+ * size, given enough input packets, it will eventually wrap around to zero;
+ * in the long term, therefore, ordinals are not unique.) The ordinals
+ * handed out by different IPPs are not disjoint, so two packets from
+ * different IPPs may have identical ordinals. Packets dropped by the
+ * IPP or by the I/O shim are not assigned ordinals.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's per-IPP packet ordinal.
+ */
+static __inline unsigned int
+NETIO_PKT_ORDINAL_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return mda->__packet_ordinal;
+}
+
+
+/** Return the per-group ordinal of the packet.
+ * @ingroup ingress
+ *
+ * Each packet is given a per-group ordinal number when it is
+ * delivered by the IPP. By default, the group is the packet's VLAN,
+ * although IPP can be recompiled to use different values. In
+ * the medium term, the ordinal is unique and monotonically
+ * increasing, being incremented by 1 for each packet; the ordinal of
+ * the first packet distributed to a particular group is zero.
+ * (Since the ordinal is of finite size, given enough input packets,
+ * it will eventually wrap around to zero; in the long term,
+ * therefore, ordinals are not unique.) The ordinals handed out by
+ * different IPPs are not disjoint, so two packets from different IPPs
+ * may have identical ordinals; similarly, packets distributed to
+ * different groups may have identical ordinals. Packets dropped by
+ * the IPP or by the I/O shim are not assigned ordinals.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's per-IPP, per-group ordinal.
+ */
+static __inline unsigned int
+NETIO_PKT_GROUP_ORDINAL_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return mda->__group_ordinal;
+}
+
+
+/** Return the VLAN ID assigned to the packet.
+ * @ingroup ingress
+ *
+ * This value is usually contained within the packet header.
+ *
+ * This value will be zero if the packet does not have a VLAN tag, or if
+ * this value was not extracted from the packet.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's VLAN ID.
+ */
+static __inline unsigned short
+NETIO_PKT_VLAN_ID_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ int vl = (mda->__flags >> _NETIO_PKT_VLAN_SHIFT) & _NETIO_PKT_VLAN_RMASK;
+ unsigned short* pkt_p;
+ int index;
+ unsigned short val;
+
+ if (vl == _NETIO_PKT_VLAN_NONE)
+ return 0;
+
+ pkt_p = (unsigned short*) NETIO_PKT_L2_DATA_M(mda, pkt);
+ index = (mda->__flags >> _NETIO_PKT_TYPE_SHIFT) & _NETIO_PKT_TYPE_RMASK;
+
+ val = pkt_p[(_netio_pkt_info[index] >> _NETIO_PKT_INFO_VLAN_SHIFT) &
+ _NETIO_PKT_INFO_VLAN_RMASK];
+
+#ifdef __TILECC__
+ return (__insn_bytex(val) >> 16) & 0xFFF;
+#else
+ return (__builtin_bswap32(val) >> 16) & 0xFFF;
+#endif
+}
+
+
+/** Return the ethertype of the packet.
+ * @ingroup ingress
+ *
+ * This value is usually contained within the packet header.
+ *
+ * This value is reliable if @ref NETIO_PKT_ETHERTYPE_RECOGNIZED_M()
+ * returns true, and otherwise, may not be well defined.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's ethertype.
+ */
+static __inline unsigned short
+NETIO_PKT_ETHERTYPE_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ unsigned short* pkt_p = (unsigned short*) NETIO_PKT_L2_DATA_M(mda, pkt);
+ int index = (mda->__flags >> _NETIO_PKT_TYPE_SHIFT) & _NETIO_PKT_TYPE_RMASK;
+
+ unsigned short val =
+ pkt_p[(_netio_pkt_info[index] >> _NETIO_PKT_INFO_ETYPE_SHIFT) &
+ _NETIO_PKT_INFO_ETYPE_RMASK];
+
+ return __builtin_bswap32(val) >> 16;
+}
+
+
+/** Return the flow hash computed on the packet.
+ * @ingroup ingress
+ *
+ * For TCP and UDP packets, this hash is calculated by hashing together
+ * the "5-tuple" values, specifically the source IP address, destination
+ * IP address, protocol type, source port and destination port.
+ * The hash value is intended to be helpful for millions of distinct
+ * flows.
+ *
+ * For IPv4 or IPv6 packets which are neither TCP nor UDP, the flow hash is
+ * derived by hashing together the source and destination IP addresses.
+ *
+ * For MPLS-encapsulated packets, the flow hash is derived by hashing
+ * the first MPLS label.
+ *
+ * For all other packets the flow hash is computed from the source
+ * and destination Ethernet addresses.
+ *
+ * The hash is symmetric, meaning it produces the same value if the
+ * source and destination are swapped. The only exceptions are
+ * tunneling protocols 0x04 (IP in IP Encapsulation), 0x29 (Simple
+ * Internet Protocol), 0x2F (General Routing Encapsulation) and 0x32
+ * (Encap Security Payload), which use only the destination address
+ * since the source address is not meaningful.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's 32-bit flow hash.
+ */
+static __inline unsigned int
+NETIO_PKT_FLOW_HASH_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return mda->__flow_hash;
+}
+
+
+/** Return the first word of "user data" for the packet.
+ *
+ * The contents of the user data words depend on the IPP.
+ *
+ * When using the standard ipp1, ipp2, or ipp4 sub-drivers, the first
+ * word of user data contains the least significant bits of the 64-bit
+ * arrival cycle count (see @c get_cycle_count_low()).
+ *
+ * See the <em>System Programmer's Guide</em> for details.
+ *
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's first word of "user data".
+ */
+static __inline unsigned int
+NETIO_PKT_USER_DATA_0_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return mda->__user_data_0;
+}
+
+
+/** Return the second word of "user data" for the packet.
+ *
+ * The contents of the user data words depend on the IPP.
+ *
+ * When using the standard ipp1, ipp2, or ipp4 sub-drivers, the second
+ * word of user data contains the most significant bits of the 64-bit
+ * arrival cycle count (see @c get_cycle_count_high()).
+ *
+ * See the <em>System Programmer's Guide</em> for details.
+ *
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's second word of "user data".
+ */
+static __inline unsigned int
+NETIO_PKT_USER_DATA_1_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return mda->__user_data_1;
+}
+
+
+/** Determine whether the L4 (TCP/UDP) checksum was calculated.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the L4 checksum was calculated.
+ */
+static __inline unsigned int
+NETIO_PKT_L4_CSUM_CALCULATED_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return !(mda->__flags & _NETIO_PKT_NO_L4_CSUM_MASK);
+}
+
+
+/** Determine whether the L4 (TCP/UDP) checksum was calculated and found to
+ * be correct.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the checksum was calculated and is correct.
+ */
+static __inline unsigned int
+NETIO_PKT_L4_CSUM_CORRECT_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return !(mda->__flags &
+ (_NETIO_PKT_BAD_L4_CSUM_MASK | _NETIO_PKT_NO_L4_CSUM_MASK));
+}
+
+
+/** Determine whether the L3 (IP) checksum was calculated.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the L3 (IP) checksum was calculated.
+*/
+static __inline unsigned int
+NETIO_PKT_L3_CSUM_CALCULATED_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return !(mda->__flags & _NETIO_PKT_NO_L3_CSUM_MASK);
+}
+
+
+/** Determine whether the L3 (IP) checksum was calculated and found to be
+ * correct.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the checksum was calculated and is correct.
+ */
+static __inline unsigned int
+NETIO_PKT_L3_CSUM_CORRECT_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return !(mda->__flags &
+ (_NETIO_PKT_BAD_L3_CSUM_MASK | _NETIO_PKT_NO_L3_CSUM_MASK));
+}
+
+
+/** Determine whether the ethertype was recognized and L3 packet data was
+ * processed.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the ethertype was recognized and L3 packet data was
+ * processed.
+ */
+static __inline unsigned int
+NETIO_PKT_ETHERTYPE_RECOGNIZED_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return !(mda->__flags & _NETIO_PKT_TYPE_UNRECOGNIZED_MASK);
+}
+
+
+/** Retrieve the status of a packet and any errors that may have occurred
+ * during ingress processing (length mismatches, CRC errors, etc.).
+ * @ingroup ingress
+ *
+ * Note that packets for which @ref NETIO_PKT_ETHERTYPE_RECOGNIZED()
+ * returns zero are always reported as underlength, as there is no a priori
+ * means to determine their length. Normally, applications should use
+ * @ref NETIO_PKT_BAD_M() instead of explicitly checking status with this
+ * function.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's status.
+ */
+static __inline netio_pkt_status_t
+NETIO_PKT_STATUS_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (netio_pkt_status_t) __NETIO_PKT_NOTIF_HEADER(pkt).bits.__status;
+}
+
+
+/** Report whether a packet is bad (i.e., was shorter than expected based on
+ * its headers, or had a bad CRC).
+ * @ingroup ingress
+ *
+ * Note that this function does not verify L3 or L4 checksums.
+ *
+ * @param[in] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the packet is bad and should be discarded.
+ */
+static __inline unsigned int
+NETIO_PKT_BAD_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return ((NETIO_PKT_STATUS_M(mda, pkt) & 1) &&
+ (NETIO_PKT_ETHERTYPE_RECOGNIZED_M(mda, pkt) ||
+ NETIO_PKT_STATUS_M(mda, pkt) == NETIO_PKT_STATUS_BAD));
+}
+
+
+/** Return the length of the packet, starting with the L2 (Ethernet) header.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_LENGTH_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt)
+{
+ return mmd->l2_length;
+}
+
+
+/** Return the length of the L2 (Ethernet) header.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet's L2 header, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_HEADER_LENGTH_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt)
+{
+ return mmd->l3_offset - mmd->l2_offset;
+}
+
+
+/** Return the length of the packet, starting with the L3 (IP) header.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return Length of the packet's L3 header and data, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L3_LENGTH_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt)
+{
+ return (NETIO_PKT_L2_LENGTH_MM(mmd, pkt) -
+ NETIO_PKT_L2_HEADER_LENGTH_MM(mmd, pkt));
+}
+
+
+/** Return a pointer to the packet's L3 (generally, the IP) header.
+ * @ingroup egress
+ *
+ * Note that we guarantee word alignment of the L3 header.
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to the packet's L3 header.
+ */
+static __inline unsigned char*
+NETIO_PKT_L3_DATA_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt)
+{
+ return _NETIO_PKT_BASE(pkt) + mmd->l3_offset;
+}
+
+
+/** Return a pointer to the packet's L2 (Ethernet) header.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_L2_DATA_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt)
+{
+ return _NETIO_PKT_BASE(pkt) + mmd->l2_offset;
+}
+
+
+/** Retrieve the status of a packet and any errors that may have occurred
+ * during ingress processing (length mismatches, CRC errors, etc.).
+ * @ingroup ingress
+ *
+ * Note that packets for which @ref NETIO_PKT_ETHERTYPE_RECOGNIZED()
+ * returns zero are always reported as underlength, as there is no a priori
+ * means to determine their length. Normally, applications should use
+ * @ref NETIO_PKT_BAD() instead of explicitly checking status with this
+ * function.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's status.
+ */
+static __inline netio_pkt_status_t
+NETIO_PKT_STATUS(netio_pkt_t* pkt)
+{
+ netio_assert(!pkt->__packet.bits.__minimal);
+
+ return (netio_pkt_status_t) __NETIO_PKT_NOTIF_HEADER(pkt).bits.__status;
+}
+
+
+/** Report whether a packet is bad (i.e., was shorter than expected based on
+ * its headers, or had a bad CRC).
+ * @ingroup ingress
+ *
+ * Note that this function does not verify L3 or L4 checksums.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the packet is bad and should be discarded.
+ */
+static __inline unsigned int
+NETIO_PKT_BAD(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_BAD_M(mda, pkt);
+}
+
+
+/** Return the length of the packet's custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet's custom header, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_CUSTOM_HEADER_LENGTH(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_CUSTOM_HEADER_LENGTH_M(mda, pkt);
+}
+
+
+/** Return the length of the packet, starting with the custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_CUSTOM_LENGTH(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_CUSTOM_LENGTH_M(mda, pkt);
+}
+
+
+/** Return a pointer to the packet's custom header.
+ * A custom header may or may not be present, depending upon the IPP; its
+ * contents and alignment are also IPP-dependent. Currently, none of the
+ * standard IPPs supplied by Tilera produce a custom header. If present,
+ * the custom header precedes the L2 header in the packet buffer.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_CUSTOM_DATA(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_CUSTOM_DATA_M(mda, pkt);
+}
+
+
+/** Return the length of the packet's L2 (Ethernet plus VLAN or SNAP) header.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet's L2 header, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_HEADER_LENGTH(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_L2_HEADER_LENGTH_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L2_HEADER_LENGTH_M(mda, pkt);
+ }
+}
+
+
+/** Return the length of the packet, starting with the L2 (Ethernet) header.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The length of the packet, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L2_LENGTH(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_L2_LENGTH_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L2_LENGTH_M(mda, pkt);
+ }
+}
+
+
+/** Return a pointer to the packet's L2 (Ethernet) header.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to start of the packet.
+ */
+static __inline unsigned char*
+NETIO_PKT_L2_DATA(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_L2_DATA_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L2_DATA_M(mda, pkt);
+ }
+}
+
+
+/** Retrieve the length of the packet, starting with the L3 (generally, the IP)
+ * header.
+ * @ingroup pktfuncs
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Length of the packet's L3 header and data, in bytes.
+ */
+static __inline netio_size_t
+NETIO_PKT_L3_LENGTH(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_L3_LENGTH_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L3_LENGTH_M(mda, pkt);
+ }
+}
+
+
+/** Return a pointer to the packet's L3 (generally, the IP) header.
+ * @ingroup pktfuncs
+ *
+ * Note that we guarantee word alignment of the L3 header.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return A pointer to the packet's L3 header.
+ */
+static __inline unsigned char*
+NETIO_PKT_L3_DATA(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_L3_DATA_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L3_DATA_M(mda, pkt);
+ }
+}
+
+
+/** Return the ordinal of the packet.
+ * @ingroup ingress
+ *
+ * Each packet is given an ordinal number when it is delivered by the IPP.
+ * In the medium term, the ordinal is unique and monotonically increasing,
+ * being incremented by 1 for each packet; the ordinal of the first packet
+ * delivered after the IPP starts is zero. (Since the ordinal is of finite
+ * size, given enough input packets, it will eventually wrap around to zero;
+ * in the long term, therefore, ordinals are not unique.) The ordinals
+ * handed out by different IPPs are not disjoint, so two packets from
+ * different IPPs may have identical ordinals. Packets dropped by the
+ * IPP or by the I/O shim are not assigned ordinals.
+ *
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's per-IPP packet ordinal.
+ */
+static __inline unsigned int
+NETIO_PKT_ORDINAL(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_ORDINAL_M(mda, pkt);
+}
+
+
+/** Return the per-group ordinal of the packet.
+ * @ingroup ingress
+ *
+ * Each packet is given a per-group ordinal number when it is
+ * delivered by the IPP. By default, the group is the packet's VLAN,
+ * although IPP can be recompiled to use different values. In
+ * the medium term, the ordinal is unique and monotonically
+ * increasing, being incremented by 1 for each packet; the ordinal of
+ * the first packet distributed to a particular group is zero.
+ * (Since the ordinal is of finite size, given enough input packets,
+ * it will eventually wrap around to zero; in the long term,
+ * therefore, ordinals are not unique.) The ordinals handed out by
+ * different IPPs are not disjoint, so two packets from different IPPs
+ * may have identical ordinals; similarly, packets distributed to
+ * different groups may have identical ordinals. Packets dropped by
+ * the IPP or by the I/O shim are not assigned ordinals.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's per-IPP, per-group ordinal.
+ */
+static __inline unsigned int
+NETIO_PKT_GROUP_ORDINAL(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_GROUP_ORDINAL_M(mda, pkt);
+}
+
+
+/** Return the VLAN ID assigned to the packet.
+ * @ingroup ingress
+ *
+ * This is usually also contained within the packet header. If the packet
+ * does not have a VLAN tag, the VLAN ID returned by this function is zero.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's VLAN ID.
+ */
+static __inline unsigned short
+NETIO_PKT_VLAN_ID(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_VLAN_ID_M(mda, pkt);
+}
+
+
+/** Return the ethertype of the packet.
+ * @ingroup ingress
+ *
+ * This value is reliable if @ref NETIO_PKT_ETHERTYPE_RECOGNIZED()
+ * returns true, and otherwise, may not be well defined.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's ethertype.
+ */
+static __inline unsigned short
+NETIO_PKT_ETHERTYPE(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_ETHERTYPE_M(mda, pkt);
+}
+
+
+/** Return the flow hash computed on the packet.
+ * @ingroup ingress
+ *
+ * For TCP and UDP packets, this hash is calculated by hashing together
+ * the "5-tuple" values, specifically the source IP address, destination
+ * IP address, protocol type, source port and destination port.
+ * The hash value is intended to be helpful for millions of distinct
+ * flows.
+ *
+ * For IPv4 or IPv6 packets which are neither TCP nor UDP, the flow hash is
+ * derived by hashing together the source and destination IP addresses.
+ *
+ * For MPLS-encapsulated packets, the flow hash is derived by hashing
+ * the first MPLS label.
+ *
+ * For all other packets the flow hash is computed from the source
+ * and destination Ethernet addresses.
+ *
+ * The hash is symmetric, meaning it produces the same value if the
+ * source and destination are swapped. The only exceptions are
+ * tunneling protocols 0x04 (IP in IP Encapsulation), 0x29 (Simple
+ * Internet Protocol), 0x2F (General Routing Encapsulation) and 0x32
+ * (Encap Security Payload), which use only the destination address
+ * since the source address is not meaningful.
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's 32-bit flow hash.
+ */
+static __inline unsigned int
+NETIO_PKT_FLOW_HASH(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_FLOW_HASH_M(mda, pkt);
+}
+
+
+/** Return the first word of "user data" for the packet.
+ *
+ * The contents of the user data words depend on the IPP.
+ *
+ * When using the standard ipp1, ipp2, or ipp4 sub-drivers, the first
+ * word of user data contains the least significant bits of the 64-bit
+ * arrival cycle count (see @c get_cycle_count_low()).
+ *
+ * See the <em>System Programmer's Guide</em> for details.
+ *
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's first word of "user data".
+ */
+static __inline unsigned int
+NETIO_PKT_USER_DATA_0(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_USER_DATA_0_M(mda, pkt);
+}
+
+
+/** Return the second word of "user data" for the packet.
+ *
+ * The contents of the user data words depend on the IPP.
+ *
+ * When using the standard ipp1, ipp2, or ipp4 sub-drivers, the second
+ * word of user data contains the most significant bits of the 64-bit
+ * arrival cycle count (see @c get_cycle_count_high()).
+ *
+ * See the <em>System Programmer's Guide</em> for details.
+ *
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return The packet's second word of "user data".
+ */
+static __inline unsigned int
+NETIO_PKT_USER_DATA_1(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_USER_DATA_1_M(mda, pkt);
+}
+
+
+/** Determine whether the L4 (TCP/UDP) checksum was calculated.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the L4 checksum was calculated.
+ */
+static __inline unsigned int
+NETIO_PKT_L4_CSUM_CALCULATED(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L4_CSUM_CALCULATED_M(mda, pkt);
+}
+
+
+/** Determine whether the L4 (TCP/UDP) checksum was calculated and found to
+ * be correct.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the checksum was calculated and is correct.
+ */
+static __inline unsigned int
+NETIO_PKT_L4_CSUM_CORRECT(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L4_CSUM_CORRECT_M(mda, pkt);
+}
+
+
+/** Determine whether the L3 (IP) checksum was calculated.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the L3 (IP) checksum was calculated.
+*/
+static __inline unsigned int
+NETIO_PKT_L3_CSUM_CALCULATED(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L3_CSUM_CALCULATED_M(mda, pkt);
+}
+
+
+/** Determine whether the L3 (IP) checksum was calculated and found to be
+ * correct.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the checksum was calculated and is correct.
+ */
+static __inline unsigned int
+NETIO_PKT_L3_CSUM_CORRECT(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_L3_CSUM_CORRECT_M(mda, pkt);
+}
+
+
+/** Determine whether the Ethertype was recognized and L3 packet data was
+ * processed.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ * @return Nonzero if the Ethertype was recognized and L3 packet data was
+ * processed.
+ */
+static __inline unsigned int
+NETIO_PKT_ETHERTYPE_RECOGNIZED(netio_pkt_t* pkt)
+{
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_ETHERTYPE_RECOGNIZED_M(mda, pkt);
+}
+
+
+/** Set an egress packet's L2 length, using a metadata pointer to speed the
+ * computation.
+ * @ingroup egress
+ *
+ * @param[in,out] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @param[in] len Packet L2 length, in bytes.
+ */
+static __inline void
+NETIO_PKT_SET_L2_LENGTH_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt,
+ int len)
+{
+ mmd->l2_length = len;
+}
+
+
+/** Set an egress packet's L2 length.
+ * @ingroup egress
+ *
+ * @param[in,out] pkt Packet on which to operate.
+ * @param[in] len Packet L2 length, in bytes.
+ */
+static __inline void
+NETIO_PKT_SET_L2_LENGTH(netio_pkt_t* pkt, int len)
+{
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ NETIO_PKT_SET_L2_LENGTH_MM(mmd, pkt, len);
+}
+
+
+/** Set an egress packet's L2 header length, using a metadata pointer to
+ * speed the computation.
+ * @ingroup egress
+ *
+ * It is not normally necessary to call this routine; only the L2 length,
+ * not the header length, is needed to transmit a packet. It may be useful if
+ * the egress packet will later be processed by code which expects to use
+ * functions like @ref NETIO_PKT_L3_DATA() to get a pointer to the L3 payload.
+ *
+ * @param[in,out] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @param[in] len Packet L2 header length, in bytes.
+ */
+static __inline void
+NETIO_PKT_SET_L2_HEADER_LENGTH_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt, int len)
+{
+ mmd->l3_offset = mmd->l2_offset + len;
+}
+
+
+/** Set an egress packet's L2 header length.
+ * @ingroup egress
+ *
+ * It is not normally necessary to call this routine; only the L2 length,
+ * not the header length, is needed to transmit a packet. It may be useful if
+ * the egress packet will later be processed by code which expects to use
+ * functions like @ref NETIO_PKT_L3_DATA() to get a pointer to the L3 payload.
+ *
+ * @param[in,out] pkt Packet on which to operate.
+ * @param[in] len Packet L2 header length, in bytes.
+ */
+static __inline void
+NETIO_PKT_SET_L2_HEADER_LENGTH(netio_pkt_t* pkt, int len)
+{
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ NETIO_PKT_SET_L2_HEADER_LENGTH_MM(mmd, pkt, len);
+}
+
+
+/** Set up an egress packet for hardware checksum computation, using a
+ * metadata pointer to speed the operation.
+ * @ingroup egress
+ *
+ * NetIO provides the ability to automatically calculate a standard
+ * 16-bit Internet checksum on transmitted packets. The application
+ * may specify the point in the packet where the checksum starts, the
+ * number of bytes to be checksummed, and the two bytes in the packet
+ * which will be replaced with the completed checksum. (If the range
+ * of bytes to be checksummed includes the bytes to be replaced, the
+ * initial values of those bytes will be included in the checksum.)
+ *
+ * For some protocols, the packet checksum covers data which is not present
+ * in the packet, or is at least not contiguous to the main data payload.
+ * For instance, the TCP checksum includes a "pseudo-header" which includes
+ * the source and destination IP addresses of the packet. To accommodate
+ * this, the checksum engine may be "seeded" with an initial value, which
+ * the application would need to compute based on the specific protocol's
+ * requirements. Note that the seed is given in host byte order (little-
+ * endian), not network byte order (big-endian); code written to compute a
+ * pseudo-header checksum in network byte order will need to byte-swap it
+ * before use as the seed.
+ *
+ * Note that the checksum is computed as part of the transmission process,
+ * so it will not be present in the packet upon completion of this routine.
+ *
+ * @param[in,out] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ * @param[in] start Offset within L2 packet of the first byte to include in
+ * the checksum.
+ * @param[in] length Number of bytes to include in the checksum.
+ * the checksum.
+ * @param[in] location Offset within L2 packet of the first of the two bytes
+ * to be replaced with the calculated checksum.
+ * @param[in] seed Initial value of the running checksum before any of the
+ * packet data is added.
+ */
+static __inline void
+NETIO_PKT_DO_EGRESS_CSUM_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt, int start, int length,
+ int location, uint16_t seed)
+{
+ mmd->csum_start = start;
+ mmd->csum_length = length;
+ mmd->csum_location = location;
+ mmd->csum_seed = seed;
+ mmd->flags |= _NETIO_PKT_NEED_EDMA_CSUM_MASK;
+}
+
+
+/** Set up an egress packet for hardware checksum computation.
+ * @ingroup egress
+ *
+ * NetIO provides the ability to automatically calculate a standard
+ * 16-bit Internet checksum on transmitted packets. The application
+ * may specify the point in the packet where the checksum starts, the
+ * number of bytes to be checksummed, and the two bytes in the packet
+ * which will be replaced with the completed checksum. (If the range
+ * of bytes to be checksummed includes the bytes to be replaced, the
+ * initial values of those bytes will be included in the checksum.)
+ *
+ * For some protocols, the packet checksum covers data which is not present
+ * in the packet, or is at least not contiguous to the main data payload.
+ * For instance, the TCP checksum includes a "pseudo-header" which includes
+ * the source and destination IP addresses of the packet. To accommodate
+ * this, the checksum engine may be "seeded" with an initial value, which
+ * the application would need to compute based on the specific protocol's
+ * requirements. Note that the seed is given in host byte order (little-
+ * endian), not network byte order (big-endian); code written to compute a
+ * pseudo-header checksum in network byte order will need to byte-swap it
+ * before use as the seed.
+ *
+ * Note that the checksum is computed as part of the transmission process,
+ * so it will not be present in the packet upon completion of this routine.
+ *
+ * @param[in,out] pkt Packet on which to operate.
+ * @param[in] start Offset within L2 packet of the first byte to include in
+ * the checksum.
+ * @param[in] length Number of bytes to include in the checksum.
+ * the checksum.
+ * @param[in] location Offset within L2 packet of the first of the two bytes
+ * to be replaced with the calculated checksum.
+ * @param[in] seed Initial value of the running checksum before any of the
+ * packet data is added.
+ */
+static __inline void
+NETIO_PKT_DO_EGRESS_CSUM(netio_pkt_t* pkt, int start, int length,
+ int location, uint16_t seed)
+{
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ NETIO_PKT_DO_EGRESS_CSUM_MM(mmd, pkt, start, length, location, seed);
+}
+
+
+/** Return the number of bytes which could be prepended to a packet, using a
+ * metadata pointer to speed the operation.
+ * See @ref netio_populate_prepend_buffer() to get a full description of
+ * prepending.
+ *
+ * @param[in,out] mda Pointer to packet's standard metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline int
+NETIO_PKT_PREPEND_AVAIL_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+ return (pkt->__packet.bits.__offset << 6) +
+ NETIO_PKT_CUSTOM_HEADER_LENGTH_M(mda, pkt);
+}
+
+
+/** Return the number of bytes which could be prepended to a packet, using a
+ * metadata pointer to speed the operation.
+ * See @ref netio_populate_prepend_buffer() to get a full description of
+ * prepending.
+ * @ingroup egress
+ *
+ * @param[in,out] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline int
+NETIO_PKT_PREPEND_AVAIL_MM(netio_pkt_minimal_metadata_t* mmd, netio_pkt_t* pkt)
+{
+ return (pkt->__packet.bits.__offset << 6) + mmd->l2_offset;
+}
+
+
+/** Return the number of bytes which could be prepended to a packet.
+ * See @ref netio_populate_prepend_buffer() to get a full description of
+ * prepending.
+ * @ingroup egress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline int
+NETIO_PKT_PREPEND_AVAIL(netio_pkt_t* pkt)
+{
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd = NETIO_PKT_MINIMAL_METADATA(pkt);
+
+ return NETIO_PKT_PREPEND_AVAIL_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = NETIO_PKT_METADATA(pkt);
+
+ return NETIO_PKT_PREPEND_AVAIL_M(mda, pkt);
+ }
+}
+
+
+/** Flush a packet's minimal metadata from the cache, using a metadata pointer
+ * to speed the operation.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_MINIMAL_METADATA_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt)
+{
+}
+
+
+/** Invalidate a packet's minimal metadata from the cache, using a metadata
+ * pointer to speed the operation.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_INV_MINIMAL_METADATA_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush and then invalidate a packet's minimal metadata from the cache,
+ * using a metadata pointer to speed the operation.
+ * @ingroup egress
+ *
+ * @param[in] mmd Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_INV_MINIMAL_METADATA_MM(netio_pkt_minimal_metadata_t* mmd,
+ netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush a packet's metadata from the cache, using a metadata pointer
+ * to speed the operation.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's minimal metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_METADATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+}
+
+
+/** Invalidate a packet's metadata from the cache, using a metadata
+ * pointer to speed the operation.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_INV_METADATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush and then invalidate a packet's metadata from the cache,
+ * using a metadata pointer to speed the operation.
+ * @ingroup ingress
+ *
+ * @param[in] mda Pointer to packet's metadata.
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_INV_METADATA_M(netio_pkt_metadata_t* mda, netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush a packet's minimal metadata from the cache.
+ * @ingroup egress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_MINIMAL_METADATA(netio_pkt_t* pkt)
+{
+}
+
+
+/** Invalidate a packet's minimal metadata from the cache.
+ * @ingroup egress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_INV_MINIMAL_METADATA(netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush and then invalidate a packet's minimal metadata from the cache.
+ * @ingroup egress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_INV_MINIMAL_METADATA(netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush a packet's metadata from the cache.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_METADATA(netio_pkt_t* pkt)
+{
+}
+
+
+/** Invalidate a packet's metadata from the cache.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_INV_METADATA(netio_pkt_t* pkt)
+{
+}
+
+
+/** Flush and then invalidate a packet's metadata from the cache.
+ * @ingroup ingress
+ *
+ * @param[in] pkt Packet on which to operate.
+ */
+static __inline void
+NETIO_PKT_FLUSH_INV_METADATA(netio_pkt_t* pkt)
+{
+}
+
+/** Number of NUMA nodes we can distribute buffers to.
+ * @ingroup setup */
+#define NETIO_NUM_NODE_WEIGHTS 16
+
+/**
+ * @brief An object for specifying the characteristics of NetIO communication
+ * endpoint.
+ *
+ * @ingroup setup
+ *
+ * The @ref netio_input_register() function uses this structure to define
+ * how an application tile will communicate with an IPP.
+ *
+ *
+ * Future updates to NetIO may add new members to this structure,
+ * which can affect the success of the registration operation. Thus,
+ * if dynamically initializing the structure, applications are urged to
+ * zero it out first, for example:
+ *
+ * @code
+ * netio_input_config_t config;
+ * memset(&config, 0, sizeof (config));
+ * config.flags = NETIO_RECV | NETIO_XMIT_CSUM | NETIO_TAG_NONE;
+ * config.num_receive_packets = NETIO_MAX_RECEIVE_PKTS;
+ * config.queue_id = 0;
+ * .
+ * .
+ * .
+ * @endcode
+ *
+ * since that guarantees that any unused structure members, including
+ * members which did not exist when the application was first developed,
+ * will not have unexpected values.
+ *
+ * If statically initializing the structure, we strongly recommend use of
+ * C99-style named initializers, for example:
+ *
+ * @code
+ * netio_input_config_t config = {
+ * .flags = NETIO_RECV | NETIO_XMIT_CSUM | NETIO_TAG_NONE,
+ * .num_receive_packets = NETIO_MAX_RECEIVE_PKTS,
+ * .queue_id = 0,
+ * },
+ * @endcode
+ *
+ * instead of the old-style structure initialization:
+ *
+ * @code
+ * // Bad example! Currently equivalent to the above, but don't do this.
+ * netio_input_config_t config = {
+ * NETIO_RECV | NETIO_XMIT_CSUM | NETIO_TAG_NONE, NETIO_MAX_RECEIVE_PKTS, 0
+ * },
+ * @endcode
+ *
+ * since the C99 style requires no changes to the code if elements of the
+ * config structure are rearranged. (It also makes the initialization much
+ * easier to understand.)
+ *
+ * Except for items which address a particular tile's transmit or receive
+ * characteristics, such as the ::NETIO_RECV flag, applications are advised
+ * to specify the same set of configuration data on all registrations.
+ * This prevents differing results if multiple tiles happen to do their
+ * registration operations in a different order on different invocations of
+ * the application. This is particularly important for things like link
+ * management flags, and buffer size and homing specifications.
+ *
+ * Unless the ::NETIO_FIXED_BUFFER_VA flag is specified in flags, the NetIO
+ * buffer pool is automatically created and mapped into the application's
+ * virtual address space at an address chosen by the operating system,
+ * using the common memory (cmem) facility in the Tilera Multicore
+ * Components library. The cmem facility allows multiple processes to gain
+ * access to shared memory which is mapped into each process at an
+ * identical virtual address. In order for this to work, the processes
+ * must have a common ancestor, which must create the common memory using
+ * tmc_cmem_init().
+ *
+ * In programs using the iLib process creation API, or in programs which use
+ * only one process (which include programs using the pthreads library),
+ * tmc_cmem_init() is called automatically. All other applications
+ * must call it explicitly, before any child processes which might call
+ * netio_input_register() are created.
+ */
+typedef struct
+{
+ /** Registration characteristics.
+
+ This value determines several characteristics of the registration;
+ flags for different types of behavior are ORed together to make the
+ final flag value. Generally applications should specify exactly
+ one flag from each of the following categories:
+
+ - Whether the application will be receiving packets on this queue
+ (::NETIO_RECV or ::NETIO_NO_RECV).
+
+ - Whether the application will be transmitting packets on this queue,
+ and if so, whether it will request egress checksum calculation
+ (::NETIO_XMIT, ::NETIO_XMIT_CSUM, or ::NETIO_NO_XMIT). It is
+ legal to call netio_get_buffer() without one of the XMIT flags,
+ as long as ::NETIO_RECV is specified; in this case, the retrieved
+ buffers must be passed to another tile for transmission.
+
+ - Whether the application expects any vendor-specific tags in
+ its packets' L2 headers (::NETIO_TAG_NONE, ::NETIO_TAG_BRCM,
+ or ::NETIO_TAG_MRVL). This must match the configuration of the
+ target IPP.
+
+ To accommodate applications written to previous versions of the NetIO
+ interface, none of the flags above are currently required; if omitted,
+ NetIO behaves more or less as if ::NETIO_RECV | ::NETIO_XMIT_CSUM |
+ ::NETIO_TAG_NONE were used. However, explicit specification of
+ the relevant flags allows NetIO to do a better job of resource
+ allocation, allows earlier detection of certain configuration errors,
+ and may enable advanced features or higher performance in the future,
+ so their use is strongly recommended.
+
+ Note that specifying ::NETIO_NO_RECV along with ::NETIO_NO_XMIT
+ is a special case, intended primarily for use by programs which
+ retrieve network statistics or do link management operations.
+ When these flags are both specified, the resulting queue may not
+ be used with NetIO routines other than netio_get(), netio_set(),
+ and netio_input_unregister(). See @ref link for more information
+ on link management.
+
+ Other flags are optional; their use is described below.
+ */
+ int flags;
+
+ /** Interface name. This is a string which identifies the specific
+ Ethernet controller hardware to be used. The format of the string
+ is a device type and a device index, separated by a slash; so,
+ the first 10 Gigabit Ethernet controller is named "xgbe/0", while
+ the second 10/100/1000 Megabit Ethernet controller is named "gbe/1".
+ */
+ const char* interface;
+
+ /** Receive packet queue size. This specifies the maximum number
+ of ingress packets that can be received on this queue without
+ being retrieved by @ref netio_get_packet(). If the IPP's distribution
+ algorithm calls for a packet to be sent to this queue, and this
+ number of packets are already pending there, the new packet
+ will either be discarded, or sent to another tile registered
+ for the same queue_id (see @ref drops). This value must
+ be at least ::NETIO_MIN_RECEIVE_PKTS, can always be at least
+ ::NETIO_MAX_RECEIVE_PKTS, and may be larger than that on certain
+ interfaces.
+ */
+ int num_receive_packets;
+
+ /** The queue ID being requested. Legal values for this range from 0
+ to ::NETIO_MAX_QUEUE_ID, inclusive. ::NETIO_MAX_QUEUE_ID is always
+ greater than or equal to the number of tiles; this allows one queue
+ for each tile, plus at least one additional queue. Some applications
+ may wish to use the additional queue as a destination for unwanted
+ packets, since packets delivered to queues for which no tiles have
+ registered are discarded.
+ */
+ unsigned int queue_id;
+
+ /** Maximum number of small send buffers to be held in the local empty
+ buffer cache. This specifies the size of the area which holds
+ empty small egress buffers requested from the IPP but not yet
+ retrieved via @ref netio_get_buffer(). This value must be greater
+ than zero if the application will ever use @ref netio_get_buffer()
+ to allocate empty small egress buffers; it may be no larger than
+ ::NETIO_MAX_SEND_BUFFERS. See @ref epp for more details on empty
+ buffer caching.
+ */
+ int num_send_buffers_small_total;
+
+ /** Number of small send buffers to be preallocated at registration.
+ If this value is nonzero, the specified number of empty small egress
+ buffers will be requested from the IPP during the netio_input_register
+ operation; this may speed the execution of @ref netio_get_buffer().
+ This may be no larger than @ref num_send_buffers_small_total. See @ref
+ epp for more details on empty buffer caching.
+ */
+ int num_send_buffers_small_prealloc;
+
+ /** Maximum number of large send buffers to be held in the local empty
+ buffer cache. This specifies the size of the area which holds empty
+ large egress buffers requested from the IPP but not yet retrieved via
+ @ref netio_get_buffer(). This value must be greater than zero if the
+ application will ever use @ref netio_get_buffer() to allocate empty
+ large egress buffers; it may be no larger than ::NETIO_MAX_SEND_BUFFERS.
+ See @ref epp for more details on empty buffer caching.
+ */
+ int num_send_buffers_large_total;
+
+ /** Number of large send buffers to be preallocated at registration.
+ If this value is nonzero, the specified number of empty large egress
+ buffers will be requested from the IPP during the netio_input_register
+ operation; this may speed the execution of @ref netio_get_buffer().
+ This may be no larger than @ref num_send_buffers_large_total. See @ref
+ epp for more details on empty buffer caching.
+ */
+ int num_send_buffers_large_prealloc;
+
+ /** Maximum number of jumbo send buffers to be held in the local empty
+ buffer cache. This specifies the size of the area which holds empty
+ jumbo egress buffers requested from the IPP but not yet retrieved via
+ @ref netio_get_buffer(). This value must be greater than zero if the
+ application will ever use @ref netio_get_buffer() to allocate empty
+ jumbo egress buffers; it may be no larger than ::NETIO_MAX_SEND_BUFFERS.
+ See @ref epp for more details on empty buffer caching.
+ */
+ int num_send_buffers_jumbo_total;
+
+ /** Number of jumbo send buffers to be preallocated at registration.
+ If this value is nonzero, the specified number of empty jumbo egress
+ buffers will be requested from the IPP during the netio_input_register
+ operation; this may speed the execution of @ref netio_get_buffer().
+ This may be no larger than @ref num_send_buffers_jumbo_total. See @ref
+ epp for more details on empty buffer caching.
+ */
+ int num_send_buffers_jumbo_prealloc;
+
+ /** Total packet buffer size. This determines the total size, in bytes,
+ of the NetIO buffer pool. Note that the maximum number of available
+ buffers of each size is determined during hypervisor configuration
+ (see the <em>System Programmer's Guide</em> for details); this just
+ influences how much host memory is allocated for those buffers.
+
+ The buffer pool is allocated from common memory, which will be
+ automatically initialized if needed. If your buffer pool is larger
+ than 240 MB, you might need to explicitly call @c tmc_cmem_init(),
+ as described in the Application Libraries Reference Manual (UG227).
+
+ Packet buffers are currently allocated in chunks of 16 MB; this
+ value will be rounded up to the next larger multiple of 16 MB.
+ If this value is zero, a default of 32 MB will be used; this was
+ the value used by previous versions of NetIO. Note that taking this
+ default also affects the placement of buffers on Linux NUMA nodes.
+ See @ref buffer_node_weights for an explanation of buffer placement.
+
+ In order to successfully allocate packet buffers, Linux must have
+ available huge pages on the relevant Linux NUMA nodes. See the
+ <em>System Programmer's Guide</em> for information on configuring
+ huge page support in Linux.
+ */
+ uint64_t total_buffer_size;
+
+ /** Buffer placement weighting factors.
+
+ This array specifies the relative amount of buffering to place
+ on each of the available Linux NUMA nodes. This array is
+ indexed by the NUMA node, and the values in the array are
+ proportional to the amount of buffer space to allocate on that
+ node.
+
+ If memory striping is enabled in the Hypervisor, then there is
+ only one logical NUMA node (node 0). In that case, NetIO will by
+ default ignore the suggested buffer node weights, and buffers
+ will be striped across the physical memory controllers. See
+ UG209 System Programmer's Guide for a description of the
+ hypervisor option that controls memory striping.
+
+ If memory striping is disabled, then there are up to four NUMA
+ nodes, corresponding to the four DDRAM controllers in the TILE
+ processor architecture. See UG100 Tile Processor Architecture
+ Overview for a diagram showing the location of each of the DDRAM
+ controllers relative to the tile array.
+
+ For instance, if memory striping is disabled, the following
+ configuration strucure:
+
+ @code
+ netio_input_config_t config = {
+ .
+ .
+ .
+ .total_buffer_size = 4 * 16 * 1024 * 1024;
+ .buffer_node_weights = { 1, 0, 1, 0 },
+ },
+ @endcode
+
+ would result in 32 MB of buffers being placed on controller 0, and
+ 32 MB on controller 2. (Since buffers are allocated in units of
+ 16 MB, some sets of weights will not be able to be matched exactly.)
+
+ For the weights to be effective, @ref total_buffer_size must be
+ nonzero. If @ref total_buffer_size is zero, causing the default
+ 32 MB of buffer space to be used, then any specified weights will
+ be ignored, and buffers will positioned as they were in previous
+ versions of NetIO:
+
+ - For xgbe/0 and gbe/0, 16 MB of buffers will be placed on controller 1,
+ and the other 16 MB will be placed on controller 2.
+
+ - For xgbe/1 and gbe/1, 16 MB of buffers will be placed on controller 2,
+ and the other 16 MB will be placed on controller 3.
+
+ If @ref total_buffer_size is nonzero, but all weights are zero,
+ then all buffer space will be allocated on Linux NUMA node zero.
+
+ By default, the specified buffer placement is treated as a hint;
+ if sufficient free memory is not available on the specified
+ controllers, the buffers will be allocated elsewhere. However,
+ if the ::NETIO_STRICT_HOMING flag is specified in @ref flags, then a
+ failure to allocate buffer space exactly as requested will cause the
+ registration operation to fail with an error of ::NETIO_CANNOT_HOME.
+
+ Note that maximal network performance cannot be achieved with
+ only one memory controller.
+ */
+ uint8_t buffer_node_weights[NETIO_NUM_NODE_WEIGHTS];
+
+ /** Fixed virtual address for packet buffers. Only valid when
+ ::NETIO_FIXED_BUFFER_VA is specified in @ref flags; see the
+ description of that flag for details.
+ */
+ void* fixed_buffer_va;
+
+ /**
+ Maximum number of outstanding send packet requests. This value is
+ only relevant when an EPP is in use; it determines the number of
+ slots in the EPP's outgoing packet queue which this tile is allowed
+ to consume, and thus the number of packets which may be sent before
+ the sending tile must wait for an acknowledgment from the EPP.
+ Modifying this value is generally only helpful when using @ref
+ netio_send_packet_vector(), where it can help improve performance by
+ allowing a single vector send operation to process more packets.
+ Typically it is not specified, and the default, which divides the
+ outgoing packet slots evenly between all tiles on the chip, is used.
+
+ If a registration asks for more outgoing packet queue slots than are
+ available, ::NETIO_TOOMANY_XMIT will be returned. The total number
+ of packet queue slots which are available for all tiles for each EPP
+ is subject to change, but is currently ::NETIO_TOTAL_SENDS_OUTSTANDING.
+
+
+ This value is ignored if ::NETIO_XMIT is not specified in flags.
+ If you want to specify a large value here for a specific tile, you are
+ advised to specify NETIO_NO_XMIT on other, non-transmitting tiles so
+ that they do not consume a default number of packet slots. Any tile
+ transmitting is required to have at least ::NETIO_MIN_SENDS_OUTSTANDING
+ slots allocated to it; values less than that will be silently
+ increased by the NetIO library.
+ */
+ int num_sends_outstanding;
+}
+netio_input_config_t;
+
+
+/** Registration flags; used in the @ref netio_input_config_t structure.
+ * @addtogroup setup
+ */
+/** @{ */
+
+/** Fail a registration request if we can't put packet buffers
+ on the specified memory controllers. */
+#define NETIO_STRICT_HOMING 0x00000002
+
+/** This application expects no tags on its L2 headers. */
+#define NETIO_TAG_NONE 0x00000004
+
+/** This application expects Marvell extended tags on its L2 headers. */
+#define NETIO_TAG_MRVL 0x00000008
+
+/** This application expects Broadcom tags on its L2 headers. */
+#define NETIO_TAG_BRCM 0x00000010
+
+/** This registration may call routines which receive packets. */
+#define NETIO_RECV 0x00000020
+
+/** This registration may not call routines which receive packets. */
+#define NETIO_NO_RECV 0x00000040
+
+/** This registration may call routines which transmit packets. */
+#define NETIO_XMIT 0x00000080
+
+/** This registration may call routines which transmit packets with
+ checksum acceleration. */
+#define NETIO_XMIT_CSUM 0x00000100
+
+/** This registration may not call routines which transmit packets. */
+#define NETIO_NO_XMIT 0x00000200
+
+/** This registration wants NetIO buffers mapped at an application-specified
+ virtual address.
+
+ NetIO buffers are by default created by the TMC common memory facility,
+ which must be configured by a common ancestor of all processes sharing
+ a network interface. When this flag is specified, NetIO buffers are
+ instead mapped at an address chosen by the application (and specified
+ in @ref netio_input_config_t::fixed_buffer_va). This allows multiple
+ unrelated but cooperating processes to share a NetIO interface.
+ All processes sharing the same interface must specify this flag,
+ and all must specify the same fixed virtual address.
+
+ @ref netio_input_config_t::fixed_buffer_va must be a
+ multiple of 16 MB, and the packet buffers will occupy @ref
+ netio_input_config_t::total_buffer_size bytes of virtual address
+ space, beginning at that address. If any of those virtual addresses
+ are currently occupied by other memory objects, like application or
+ shared library code or data, @ref netio_input_register() will return
+ ::NETIO_FAULT. While it is impossible to provide a fixed_buffer_va
+ which will work for all applications, a good first guess might be to
+ use 0xb0000000 minus @ref netio_input_config_t::total_buffer_size.
+ If that fails, it might be helpful to consult the running application's
+ virtual address description file (/proc/<em>pid</em>/maps) to see
+ which regions of virtual address space are available.
+ */
+#define NETIO_FIXED_BUFFER_VA 0x00000400
+
+/** This registration call will not complete unless the network link
+ is up. The process will wait several seconds for this to happen (the
+ precise interval is link-dependent), but if the link does not come up,
+ ::NETIO_LINK_DOWN will be returned. This flag is the default if
+ ::NETIO_NOREQUIRE_LINK_UP is not specified. Note that this flag by
+ itself does not request that the link be brought up; that can be done
+ with the ::NETIO_AUTO_LINK_UPDN or ::NETIO_AUTO_LINK_UP flags (the
+ latter is the default if no NETIO_AUTO_LINK_xxx flags are specified),
+ or by explicitly setting the link's desired state via netio_set().
+ If the link is not brought up by one of those methods, and this flag
+ is specified, the registration operation will return ::NETIO_LINK_DOWN.
+ This flag is ignored if it is specified along with ::NETIO_NO_XMIT and
+ ::NETIO_NO_RECV. See @ref link for more information on link
+ management.
+ */
+#define NETIO_REQUIRE_LINK_UP 0x00000800
+
+/** This registration call will complete even if the network link is not up.
+ Whenever the link is not up, packets will not be sent or received:
+ netio_get_packet() will return ::NETIO_NOPKT once all queued packets
+ have been drained, and netio_send_packet() and similar routines will
+ return NETIO_QUEUE_FULL once the outgoing packet queue in the EPP
+ or the I/O shim is full. See @ref link for more information on link
+ management.
+ */
+#define NETIO_NOREQUIRE_LINK_UP 0x00001000
+
+#ifndef __DOXYGEN__
+/*
+ * These are part of the implementation of the NETIO_AUTO_LINK_xxx flags,
+ * but should not be used directly by applications, and are thus not
+ * documented.
+ */
+#define _NETIO_AUTO_UP 0x00002000
+#define _NETIO_AUTO_DN 0x00004000
+#define _NETIO_AUTO_PRESENT 0x00008000
+#endif
+
+/** Set the desired state of the link to up, allowing any speeds which are
+ supported by the link hardware, as part of this registration operation.
+ Do not take down the link automatically. This is the default if
+ no other NETIO_AUTO_LINK_xxx flags are specified. This flag is ignored
+ if it is specified along with ::NETIO_NO_XMIT and ::NETIO_NO_RECV.
+ See @ref link for more information on link management.
+ */
+#define NETIO_AUTO_LINK_UP (_NETIO_AUTO_PRESENT | _NETIO_AUTO_UP)
+
+/** Set the desired state of the link to up, allowing any speeds which are
+ supported by the link hardware, as part of this registration operation.
+ Set the desired state of the link to down the next time no tiles are
+ registered for packet reception or transmission. This flag is ignored
+ if it is specified along with ::NETIO_NO_XMIT and ::NETIO_NO_RECV.
+ See @ref link for more information on link management.
+ */
+#define NETIO_AUTO_LINK_UPDN (_NETIO_AUTO_PRESENT | _NETIO_AUTO_UP | \
+ _NETIO_AUTO_DN)
+
+/** Set the desired state of the link to down the next time no tiles are
+ registered for packet reception or transmission. This flag is ignored
+ if it is specified along with ::NETIO_NO_XMIT and ::NETIO_NO_RECV.
+ See @ref link for more information on link management.
+ */
+#define NETIO_AUTO_LINK_DN (_NETIO_AUTO_PRESENT | _NETIO_AUTO_DN)
+
+/** Do not bring up the link automatically as part of this registration
+ operation. Do not take down the link automatically. This flag
+ is ignored if it is specified along with ::NETIO_NO_XMIT and
+ ::NETIO_NO_RECV. See @ref link for more information on link management.
+ */
+#define NETIO_AUTO_LINK_NONE _NETIO_AUTO_PRESENT
+
+
+/** Minimum number of receive packets. */
+#define NETIO_MIN_RECEIVE_PKTS 16
+
+/** Lower bound on the maximum number of receive packets; may be higher
+ than this on some interfaces. */
+#define NETIO_MAX_RECEIVE_PKTS 128
+
+/** Maximum number of send buffers, per packet size. */
+#define NETIO_MAX_SEND_BUFFERS 16
+
+/** Number of EPP queue slots, and thus outstanding sends, per EPP. */
+#define NETIO_TOTAL_SENDS_OUTSTANDING 2015
+
+/** Minimum number of EPP queue slots, and thus outstanding sends, per
+ * transmitting tile. */
+#define NETIO_MIN_SENDS_OUTSTANDING 16
+
+
+/**@}*/
+
+#ifndef __DOXYGEN__
+
+/**
+ * An object for providing Ethernet packets to a process.
+ */
+struct __netio_queue_impl_t;
+
+/**
+ * An object for managing the user end of a NetIO queue.
+ */
+struct __netio_queue_user_impl_t;
+
+#endif /* !__DOXYGEN__ */
+
+
+/** A netio_queue_t describes a NetIO communications endpoint.
+ * @ingroup setup
+ */
+typedef struct
+{
+#ifdef __DOXYGEN__
+ uint8_t opaque[8]; /**< This is an opaque structure. */
+#else
+ struct __netio_queue_impl_t* __system_part; /**< The system part. */
+ struct __netio_queue_user_impl_t* __user_part; /**< The user part. */
+#ifdef _NETIO_PTHREAD
+ _netio_percpu_mutex_t lock; /**< Queue lock. */
+#endif
+#endif
+}
+netio_queue_t;
+
+
+/**
+ * @brief Packet send context.
+ *
+ * @ingroup egress
+ *
+ * Packet send context for use with netio_send_packet_prepare and _commit.
+ */
+typedef struct
+{
+#ifdef __DOXYGEN__
+ uint8_t opaque[44]; /**< This is an opaque structure. */
+#else
+ uint8_t flags; /**< Defined below */
+ uint8_t datalen; /**< Number of valid words pointed to by data. */
+ uint32_t request[9]; /**< Request to be sent to the EPP or shim. Note
+ that this is smaller than the 11-word maximum
+ request size, since some constant values are
+ not saved in the context. */
+ uint32_t *data; /**< Data to be sent to the EPP or shim via IDN. */
+#endif
+}
+netio_send_pkt_context_t;
+
+
+#ifndef __DOXYGEN__
+#define SEND_PKT_CTX_USE_EPP 1 /**< We're sending to an EPP. */
+#define SEND_PKT_CTX_SEND_CSUM 2 /**< Request includes a checksum. */
+#endif
+
+/**
+ * @brief Packet vector entry.
+ *
+ * @ingroup egress
+ *
+ * This data structure is used with netio_send_packet_vector() to send multiple
+ * packets with one NetIO call. The structure should be initialized by
+ * calling netio_pkt_vector_set(), rather than by setting the fields
+ * directly.
+ *
+ * This structure is guaranteed to be a power of two in size, no
+ * bigger than one L2 cache line, and to be aligned modulo its size.
+ */
+typedef struct
+#ifndef __DOXYGEN__
+__attribute__((aligned(8)))
+#endif
+{
+ /** Reserved for use by the user application. When initialized with
+ * the netio_set_pkt_vector_entry() function, this field is guaranteed
+ * to be visible to readers only after all other fields are already
+ * visible. This way it can be used as a valid flag or generation
+ * counter. */
+ uint8_t user_data;
+
+ /* Structure members below this point should not be accessed directly by
+ * applications, as they may change in the future. */
+
+ /** Low 8 bits of the packet address to send. The high bits are
+ * acquired from the 'handle' field. */
+ uint8_t buffer_address_low;
+
+ /** Number of bytes to transmit. */
+ uint16_t size;
+
+ /** The raw handle from a netio_pkt_t. If this is NETIO_PKT_HANDLE_NONE,
+ * this vector entry will be skipped and no packet will be transmitted. */
+ netio_pkt_handle_t handle;
+}
+netio_pkt_vector_entry_t;
+
+
+/**
+ * @brief Initialize fields in a packet vector entry.
+ *
+ * @ingroup egress
+ *
+ * @param[out] v Pointer to the vector entry to be initialized.
+ * @param[in] pkt Packet to be transmitted when the vector entry is passed to
+ * netio_send_packet_vector(). Note that the packet's attributes
+ * (e.g., its L2 offset and length) are captured at the time this
+ * routine is called; subsequent changes in those attributes will not
+ * be reflected in the packet which is actually transmitted.
+ * Changes in the packet's contents, however, will be so reflected.
+ * If this is NULL, no packet will be transmitted.
+ * @param[in] user_data User data to be set in the vector entry.
+ * This function guarantees that the "user_data" field will become
+ * visible to a reader only after all other fields have become visible.
+ * This allows a structure in a ring buffer to be written and read
+ * by a polling reader without any locks or other synchronization.
+ */
+static __inline void
+netio_pkt_vector_set(volatile netio_pkt_vector_entry_t* v, netio_pkt_t* pkt,
+ uint8_t user_data)
+{
+ if (pkt)
+ {
+ if (NETIO_PKT_IS_MINIMAL(pkt))
+ {
+ netio_pkt_minimal_metadata_t* mmd =
+ (netio_pkt_minimal_metadata_t*) &pkt->__metadata;
+ v->buffer_address_low = (uintptr_t) NETIO_PKT_L2_DATA_MM(mmd, pkt) & 0xFF;
+ v->size = NETIO_PKT_L2_LENGTH_MM(mmd, pkt);
+ }
+ else
+ {
+ netio_pkt_metadata_t* mda = &pkt->__metadata;
+ v->buffer_address_low = (uintptr_t) NETIO_PKT_L2_DATA_M(mda, pkt) & 0xFF;
+ v->size = NETIO_PKT_L2_LENGTH_M(mda, pkt);
+ }
+ v->handle.word = pkt->__packet.word;
+ }
+ else
+ {
+ v->handle.word = 0; /* Set handle to NETIO_PKT_HANDLE_NONE. */
+ }
+
+ __asm__("" : : : "memory");
+
+ v->user_data = user_data;
+}
+
+
+/**
+ * Flags and structures for @ref netio_get() and @ref netio_set().
+ * @ingroup config
+ */
+
+/** @{ */
+/** Parameter class; addr is a NETIO_PARAM_xxx value. */
+#define NETIO_PARAM 0
+/** Interface MAC address. This address is only valid with @ref netio_get().
+ * The value is a 6-byte MAC address. Depending upon the overall system
+ * design, a MAC address may or may not be available for each interface. */
+#define NETIO_PARAM_MAC 0
+
+/** Determine whether to suspend output on the receipt of pause frames.
+ * If the value is nonzero, the I/O shim will suspend output when a pause
+ * frame is received. If the value is zero, pause frames will be ignored. */
+#define NETIO_PARAM_PAUSE_IN 1
+
+/** Determine whether to send pause frames if the I/O shim packet FIFOs are
+ * nearly full. If the value is zero, pause frames are not sent. If
+ * the value is nonzero, it is the delay value which will be sent in any
+ * pause frames which are output, in units of 512 bit times. */
+#define NETIO_PARAM_PAUSE_OUT 2
+
+/** Jumbo frame support. The value is a 4-byte integer. If the value is
+ * nonzero, the MAC will accept frames of up to 10240 bytes. If the value
+ * is zero, the MAC will only accept frames of up to 1544 bytes. */
+#define NETIO_PARAM_JUMBO 3
+
+/** I/O shim's overflow statistics register. The value is two 16-bit integers.
+ * The first 16-bit value (or the low 16 bits, if the value is treated as a
+ * 32-bit number) is the count of packets which were completely dropped and
+ * not delivered by the shim. The second 16-bit value (or the high 16 bits,
+ * if the value is treated as a 32-bit number) is the count of packets
+ * which were truncated and thus only partially delivered by the shim. This
+ * register is automatically reset to zero after it has been read.
+ */
+#define NETIO_PARAM_OVERFLOW 4
+
+/** IPP statistics. This address is only valid with @ref netio_get(). The
+ * value is a netio_stat_t structure. Unlike the I/O shim statistics, the
+ * IPP statistics are not all reset to zero on read; see the description
+ * of the netio_stat_t for details. */
+#define NETIO_PARAM_STAT 5
+
+/** Possible link state. The value is a combination of "NETIO_LINK_xxx"
+ * flags. With @ref netio_get(), this will indicate which flags are
+ * actually supported by the hardware.
+ *
+ * For historical reasons, specifying this value to netio_set() will have
+ * the same behavior as using ::NETIO_PARAM_LINK_CONFIG, but this usage is
+ * discouraged.
+ */
+#define NETIO_PARAM_LINK_POSSIBLE_STATE 6
+
+/** Link configuration. The value is a combination of "NETIO_LINK_xxx" flags.
+ * With @ref netio_set(), this will attempt to immediately bring up the
+ * link using whichever of the requested flags are supported by the
+ * hardware, or take down the link if the flags are zero; if this is
+ * not possible, an error will be returned. Many programs will want
+ * to use ::NETIO_PARAM_LINK_DESIRED_STATE instead.
+ *
+ * For historical reasons, specifying this value to netio_get() will
+ * have the same behavior as using ::NETIO_PARAM_LINK_POSSIBLE_STATE,
+ * but this usage is discouraged.
+ */
+#define NETIO_PARAM_LINK_CONFIG NETIO_PARAM_LINK_POSSIBLE_STATE
+
+/** Current link state. This address is only valid with @ref netio_get().
+ * The value is zero or more of the "NETIO_LINK_xxx" flags, ORed together.
+ * If the link is down, the value ANDed with NETIO_LINK_SPEED will be
+ * zero; if the link is up, the value ANDed with NETIO_LINK_SPEED will
+ * result in exactly one of the NETIO_LINK_xxx values, indicating the
+ * current speed. */
+#define NETIO_PARAM_LINK_CURRENT_STATE 7
+
+/** Variant symbol for current state, retained for compatibility with
+ * pre-MDE-2.1 programs. */
+#define NETIO_PARAM_LINK_STATUS NETIO_PARAM_LINK_CURRENT_STATE
+
+/** Packet Coherence protocol. This address is only valid with @ref netio_get().
+ * The value is nonzero if the interface is configured for cache-coherent DMA.
+ */
+#define NETIO_PARAM_COHERENT 8
+
+/** Desired link state. The value is a conbination of "NETIO_LINK_xxx"
+ * flags, which specify the desired state for the link. With @ref
+ * netio_set(), this will, in the background, attempt to bring up the link
+ * using whichever of the requested flags are reasonable, or take down the
+ * link if the flags are zero. The actual link up or down operation may
+ * happen after this call completes. If the link state changes in the
+ * future, the system will continue to try to get back to the desired link
+ * state; for instance, if the link is brought up successfully, and then
+ * the network cable is disconnected, the link will go down. However, the
+ * desired state of the link is still up, so if the cable is reconnected,
+ * the link will be brought up again.
+ *
+ * With @ref netio_get(), this will indicate the desired state for the
+ * link, as set with a previous netio_set() call, or implicitly by a
+ * netio_input_register() or netio_input_unregister() operation. This may
+ * not reflect the current state of the link; to get that, use
+ * ::NETIO_PARAM_LINK_CURRENT_STATE. */
+#define NETIO_PARAM_LINK_DESIRED_STATE 9
+
+/** NetIO statistics structure. Retrieved using the ::NETIO_PARAM_STAT
+ * address passed to @ref netio_get(). */
+typedef struct
+{
+ /** Number of packets which have been received by the IPP and forwarded
+ * to a tile's receive queue for processing. This value wraps at its
+ * maximum, and is not cleared upon read. */
+ uint32_t packets_received;
+
+ /** Number of packets which have been dropped by the IPP, because they could
+ * not be received, or could not be forwarded to a tile. The former happens
+ * when the IPP does not have a free packet buffer of suitable size for an
+ * incoming frame. The latter happens when all potential destination tiles
+ * for a packet, as defined by the group, bucket, and queue configuration,
+ * have full receive queues. This value wraps at its maximum, and is not
+ * cleared upon read. */
+ uint32_t packets_dropped;
+
+ /*
+ * Note: the #defines after each of the following four one-byte values
+ * denote their location within the third word of the netio_stat_t. They
+ * are intended for use only by the IPP implementation and are thus omitted
+ * from the Doxygen output.
+ */
+
+ /** Number of packets dropped because no worker was able to accept a new
+ * packet. This value saturates at its maximum, and is cleared upon
+ * read. */
+ uint8_t drops_no_worker;
+#ifndef __DOXYGEN__
+#define NETIO_STAT_DROPS_NO_WORKER 0
+#endif
+
+ /** Number of packets dropped because no small buffers were available.
+ * This value saturates at its maximum, and is cleared upon read. */
+ uint8_t drops_no_smallbuf;
+#ifndef __DOXYGEN__
+#define NETIO_STAT_DROPS_NO_SMALLBUF 1
+#endif
+
+ /** Number of packets dropped because no large buffers were available.
+ * This value saturates at its maximum, and is cleared upon read. */
+ uint8_t drops_no_largebuf;
+#ifndef __DOXYGEN__
+#define NETIO_STAT_DROPS_NO_LARGEBUF 2
+#endif
+
+ /** Number of packets dropped because no jumbo buffers were available.
+ * This value saturates at its maximum, and is cleared upon read. */
+ uint8_t drops_no_jumbobuf;
+#ifndef __DOXYGEN__
+#define NETIO_STAT_DROPS_NO_JUMBOBUF 3
+#endif
+}
+netio_stat_t;
+
+
+/** Link can run, should run, or is running at 10 Mbps. */
+#define NETIO_LINK_10M 0x01
+
+/** Link can run, should run, or is running at 100 Mbps. */
+#define NETIO_LINK_100M 0x02
+
+/** Link can run, should run, or is running at 1 Gbps. */
+#define NETIO_LINK_1G 0x04
+
+/** Link can run, should run, or is running at 10 Gbps. */
+#define NETIO_LINK_10G 0x08
+
+/** Link should run at the highest speed supported by the link and by
+ * the device connected to the link. Only usable as a value for
+ * the link's desired state; never returned as a value for the current
+ * or possible states. */
+#define NETIO_LINK_ANYSPEED 0x10
+
+/** All legal link speeds. */
+#define NETIO_LINK_SPEED (NETIO_LINK_10M | \
+ NETIO_LINK_100M | \
+ NETIO_LINK_1G | \
+ NETIO_LINK_10G | \
+ NETIO_LINK_ANYSPEED)
+
+
+/** MAC register class. Addr is a register offset within the MAC.
+ * Registers within the XGbE and GbE MACs are documented in the Tile
+ * Processor I/O Device Guide (UG104). MAC registers start at address
+ * 0x4000, and do not include the MAC_INTERFACE registers. */
+#define NETIO_MAC 1
+
+/** MDIO register class (IEEE 802.3 clause 22 format). Addr is the "addr"
+ * member of a netio_mdio_addr_t structure. */
+#define NETIO_MDIO 2
+
+/** MDIO register class (IEEE 802.3 clause 45 format). Addr is the "addr"
+ * member of a netio_mdio_addr_t structure. */
+#define NETIO_MDIO_CLAUSE45 3
+
+/** NetIO MDIO address type. Retrieved or provided using the ::NETIO_MDIO
+ * address passed to @ref netio_get() or @ref netio_set(). */
+typedef union
+{
+ struct
+ {
+ unsigned int reg:16; /**< MDIO register offset. For clause 22 access,
+ must be less than 32. */
+ unsigned int phy:5; /**< Which MDIO PHY to access. */
+ unsigned int dev:5; /**< Which MDIO device to access within that PHY.
+ Applicable for clause 45 access only; ignored
+ for clause 22 access. */
+ }
+ bits; /**< Container for bitfields. */
+ uint64_t addr; /**< Value to pass to @ref netio_get() or
+ * @ref netio_set(). */
+}
+netio_mdio_addr_t;
+
+/** @} */
+
+#endif /* __NETIO_INTF_H__ */
obj-$(CONFIG_MODULES) += module.o
obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_PCI) += pci.o
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/capability.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/bootmem.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <linux/uaccess.h>
+
+#include <asm/processor.h>
+#include <asm/sections.h>
+#include <asm/byteorder.h>
+#include <asm/hv_driver.h>
+#include <hv/drv_pcie_rc_intf.h>
+
+
+/*
+ * Initialization flow and process
+ * -------------------------------
+ *
+ * This files containes the routines to search for PCI buses,
+ * enumerate the buses, and configure any attached devices.
+ *
+ * There are two entry points here:
+ * 1) tile_pci_init
+ * This sets up the pci_controller structs, and opens the
+ * FDs to the hypervisor. This is called from setup_arch() early
+ * in the boot process.
+ * 2) pcibios_init
+ * This probes the PCI bus(es) for any attached hardware. It's
+ * called by subsys_initcall. All of the real work is done by the
+ * generic Linux PCI layer.
+ *
+ */
+
+/*
+ * This flag tells if the platform is TILEmpower that needs
+ * special configuration for the PLX switch chip.
+ */
+int __write_once tile_plx_gen1;
+
+static struct pci_controller controllers[TILE_NUM_PCIE];
+static int num_controllers;
+
+static struct pci_ops tile_cfg_ops;
+
+
+/*
+ * We don't need to worry about the alignment of resources.
+ */
+resource_size_t pcibios_align_resource(void *data, const struct resource *res,
+ resource_size_t size, resource_size_t align)
+{
+ return res->start;
+}
+EXPORT_SYMBOL(pcibios_align_resource);
+
+/*
+ * Open a FD to the hypervisor PCI device.
+ *
+ * controller_id is the controller number, config type is 0 or 1 for
+ * config0 or config1 operations.
+ */
+static int __init tile_pcie_open(int controller_id, int config_type)
+{
+ char filename[32];
+ int fd;
+
+ sprintf(filename, "pcie/%d/config%d", controller_id, config_type);
+
+ fd = hv_dev_open((HV_VirtAddr)filename, 0);
+
+ return fd;
+}
+
+
+/*
+ * Get the IRQ numbers from the HV and set up the handlers for them.
+ */
+static int __init tile_init_irqs(int controller_id,
+ struct pci_controller *controller)
+{
+ char filename[32];
+ int fd;
+ int ret;
+ int x;
+ struct pcie_rc_config rc_config;
+
+ sprintf(filename, "pcie/%d/ctl", controller_id);
+ fd = hv_dev_open((HV_VirtAddr)filename, 0);
+ if (fd < 0) {
+ pr_err("PCI: hv_dev_open(%s) failed\n", filename);
+ return -1;
+ }
+ ret = hv_dev_pread(fd, 0, (HV_VirtAddr)(&rc_config),
+ sizeof(rc_config), PCIE_RC_CONFIG_MASK_OFF);
+ hv_dev_close(fd);
+ if (ret != sizeof(rc_config)) {
+ pr_err("PCI: wanted %zd bytes, got %d\n",
+ sizeof(rc_config), ret);
+ return -1;
+ }
+ /* Record irq_base so that we can map INTx to IRQ # later. */
+ controller->irq_base = rc_config.intr;
+
+ for (x = 0; x < 4; x++)
+ tile_irq_activate(rc_config.intr + x,
+ TILE_IRQ_HW_CLEAR);
+
+ if (rc_config.plx_gen1)
+ controller->plx_gen1 = 1;
+
+ return 0;
+}
+
+/*
+ * First initialization entry point, called from setup_arch().
+ *
+ * Find valid controllers and fill in pci_controller structs for each
+ * of them.
+ *
+ * Returns the number of controllers discovered.
+ */
+int __init tile_pci_init(void)
+{
+ int i;
+
+ pr_info("PCI: Searching for controllers...\n");
+
+ /* Do any configuration we need before using the PCIe */
+
+ for (i = 0; i < TILE_NUM_PCIE; i++) {
+ int hv_cfg_fd0 = -1;
+ int hv_cfg_fd1 = -1;
+ int hv_mem_fd = -1;
+ char name[32];
+ struct pci_controller *controller;
+
+ /*
+ * Open the fd to the HV. If it fails then this
+ * device doesn't exist.
+ */
+ hv_cfg_fd0 = tile_pcie_open(i, 0);
+ if (hv_cfg_fd0 < 0)
+ continue;
+ hv_cfg_fd1 = tile_pcie_open(i, 1);
+ if (hv_cfg_fd1 < 0) {
+ pr_err("PCI: Couldn't open config fd to HV "
+ "for controller %d\n", i);
+ goto err_cont;
+ }
+
+ sprintf(name, "pcie/%d/mem", i);
+ hv_mem_fd = hv_dev_open((HV_VirtAddr)name, 0);
+ if (hv_mem_fd < 0) {
+ pr_err("PCI: Could not open mem fd to HV!\n");
+ goto err_cont;
+ }
+
+ pr_info("PCI: Found PCI controller #%d\n", i);
+
+ controller = &controllers[num_controllers];
+
+ if (tile_init_irqs(i, controller)) {
+ pr_err("PCI: Could not initialize "
+ "IRQs, aborting.\n");
+ goto err_cont;
+ }
+
+ controller->index = num_controllers;
+ controller->hv_cfg_fd[0] = hv_cfg_fd0;
+ controller->hv_cfg_fd[1] = hv_cfg_fd1;
+ controller->hv_mem_fd = hv_mem_fd;
+ controller->first_busno = 0;
+ controller->last_busno = 0xff;
+ controller->ops = &tile_cfg_ops;
+
+ num_controllers++;
+ continue;
+
+err_cont:
+ if (hv_cfg_fd0 >= 0)
+ hv_dev_close(hv_cfg_fd0);
+ if (hv_cfg_fd1 >= 0)
+ hv_dev_close(hv_cfg_fd1);
+ if (hv_mem_fd >= 0)
+ hv_dev_close(hv_mem_fd);
+ continue;
+ }
+
+ /*
+ * Before using the PCIe, see if we need to do any platform-specific
+ * configuration, such as the PLX switch Gen 1 issue on TILEmpower.
+ */
+ for (i = 0; i < num_controllers; i++) {
+ struct pci_controller *controller = &controllers[i];
+
+ if (controller->plx_gen1)
+ tile_plx_gen1 = 1;
+ }
+
+ return num_controllers;
+}
+
+/*
+ * (pin - 1) converts from the PCI standard's [1:4] convention to
+ * a normal [0:3] range.
+ */
+static int tile_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+ struct pci_controller *controller =
+ (struct pci_controller *)dev->sysdata;
+ return (pin - 1) + controller->irq_base;
+}
+
+
+static void __init fixup_read_and_payload_sizes(void)
+{
+ struct pci_dev *dev = NULL;
+ int smallest_max_payload = 0x1; /* Tile maxes out at 256 bytes. */
+ int max_read_size = 0x2; /* Limit to 512 byte reads. */
+ u16 new_values;
+
+ /* Scan for the smallest maximum payload size. */
+ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ int pcie_caps_offset;
+ u32 devcap;
+ int max_payload;
+
+ pcie_caps_offset = pci_find_capability(dev, PCI_CAP_ID_EXP);
+ if (pcie_caps_offset == 0)
+ continue;
+
+ pci_read_config_dword(dev, pcie_caps_offset + PCI_EXP_DEVCAP,
+ &devcap);
+ max_payload = devcap & PCI_EXP_DEVCAP_PAYLOAD;
+ if (max_payload < smallest_max_payload)
+ smallest_max_payload = max_payload;
+ }
+
+ /* Now, set the max_payload_size for all devices to that value. */
+ new_values = (max_read_size << 12) | (smallest_max_payload << 5);
+ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ int pcie_caps_offset;
+ u16 devctl;
+
+ pcie_caps_offset = pci_find_capability(dev, PCI_CAP_ID_EXP);
+ if (pcie_caps_offset == 0)
+ continue;
+
+ pci_read_config_word(dev, pcie_caps_offset + PCI_EXP_DEVCTL,
+ &devctl);
+ devctl &= ~(PCI_EXP_DEVCTL_PAYLOAD | PCI_EXP_DEVCTL_READRQ);
+ devctl |= new_values;
+ pci_write_config_word(dev, pcie_caps_offset + PCI_EXP_DEVCTL,
+ devctl);
+ }
+}
+
+
+/*
+ * Second PCI initialization entry point, called by subsys_initcall.
+ *
+ * The controllers have been set up by the time we get here, by a call to
+ * tile_pci_init.
+ */
+static int __init pcibios_init(void)
+{
+ int i;
+
+ pr_info("PCI: Probing PCI hardware\n");
+
+ /*
+ * Delay a bit in case devices aren't ready. Some devices are
+ * known to require at least 20ms here, but we use a more
+ * conservative value.
+ */
+ mdelay(250);
+
+ /* Scan all of the recorded PCI controllers. */
+ for (i = 0; i < num_controllers; i++) {
+ struct pci_controller *controller = &controllers[i];
+ struct pci_bus *bus;
+
+ pr_info("PCI: initializing controller #%d\n", i);
+
+ /*
+ * This comes from the generic Linux PCI driver.
+ *
+ * It reads the PCI tree for this bus into the Linux
+ * data structures.
+ *
+ * This is inlined in linux/pci.h and calls into
+ * pci_scan_bus_parented() in probe.c.
+ */
+ bus = pci_scan_bus(0, controller->ops, controller);
+ controller->root_bus = bus;
+ controller->last_busno = bus->subordinate;
+
+ }
+
+ /* Do machine dependent PCI interrupt routing */
+ pci_fixup_irqs(pci_common_swizzle, tile_map_irq);
+
+ /*
+ * This comes from the generic Linux PCI driver.
+ *
+ * It allocates all of the resources (I/O memory, etc)
+ * associated with the devices read in above.
+ */
+
+ pci_assign_unassigned_resources();
+
+ /* Configure the max_read_size and max_payload_size values. */
+ fixup_read_and_payload_sizes();
+
+ /* Record the I/O resources in the PCI controller structure. */
+ for (i = 0; i < num_controllers; i++) {
+ struct pci_bus *root_bus = controllers[i].root_bus;
+ struct pci_bus *next_bus;
+ struct pci_dev *dev;
+
+ list_for_each_entry(dev, &root_bus->devices, bus_list) {
+ /* Find the PCI host controller, ie. the 1st bridge. */
+ if ((dev->class >> 8) == PCI_CLASS_BRIDGE_PCI &&
+ (PCI_SLOT(dev->devfn) == 0)) {
+ next_bus = dev->subordinate;
+ controllers[i].mem_resources[0] =
+ *next_bus->resource[0];
+ controllers[i].mem_resources[1] =
+ *next_bus->resource[1];
+ controllers[i].mem_resources[2] =
+ *next_bus->resource[2];
+
+ break;
+ }
+ }
+
+ }
+
+ return 0;
+}
+subsys_initcall(pcibios_init);
+
+/*
+ * No bus fixups needed.
+ */
+void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+{
+ /* Nothing needs to be done. */
+}
+
+/*
+ * This can be called from the generic PCI layer, but doesn't need to
+ * do anything.
+ */
+char __devinit *pcibios_setup(char *str)
+{
+ /* Nothing needs to be done. */
+ return str;
+}
+
+/*
+ * This is called from the generic Linux layer.
+ */
+void __init pcibios_update_irq(struct pci_dev *dev, int irq)
+{
+ pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
+}
+
+/*
+ * Enable memory and/or address decoding, as appropriate, for the
+ * device described by the 'dev' struct.
+ *
+ * This is called from the generic PCI layer, and can be called
+ * for bridges or endpoints.
+ */
+int pcibios_enable_device(struct pci_dev *dev, int mask)
+{
+ u16 cmd, old_cmd;
+ u8 header_type;
+ int i;
+ struct resource *r;
+
+ pci_read_config_byte(dev, PCI_HEADER_TYPE, &header_type);
+
+ pci_read_config_word(dev, PCI_COMMAND, &cmd);
+ old_cmd = cmd;
+ if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+ /*
+ * For bridges, we enable both memory and I/O decoding
+ * in call cases.
+ */
+ cmd |= PCI_COMMAND_IO;
+ cmd |= PCI_COMMAND_MEMORY;
+ } else {
+ /*
+ * For endpoints, we enable memory and/or I/O decoding
+ * only if they have a memory resource of that type.
+ */
+ for (i = 0; i < 6; i++) {
+ r = &dev->resource[i];
+ if (r->flags & IORESOURCE_UNSET) {
+ pr_err("PCI: Device %s not available "
+ "because of resource collisions\n",
+ pci_name(dev));
+ return -EINVAL;
+ }
+ if (r->flags & IORESOURCE_IO)
+ cmd |= PCI_COMMAND_IO;
+ if (r->flags & IORESOURCE_MEM)
+ cmd |= PCI_COMMAND_MEMORY;
+ }
+ }
+
+ /*
+ * We only write the command if it changed.
+ */
+ if (cmd != old_cmd)
+ pci_write_config_word(dev, PCI_COMMAND, cmd);
+ return 0;
+}
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max)
+{
+ unsigned long start = pci_resource_start(dev, bar);
+ unsigned long len = pci_resource_len(dev, bar);
+ unsigned long flags = pci_resource_flags(dev, bar);
+
+ if (!len)
+ return NULL;
+ if (max && len > max)
+ len = max;
+
+ if (!(flags & IORESOURCE_MEM)) {
+ pr_info("PCI: Trying to map invalid resource %#lx\n", flags);
+ start = 0;
+ }
+
+ return (void __iomem *)start;
+}
+EXPORT_SYMBOL(pci_iomap);
+
+
+/****************************************************************
+ *
+ * Tile PCI config space read/write routines
+ *
+ ****************************************************************/
+
+/*
+ * These are the normal read and write ops
+ * These are expanded with macros from pci_bus_read_config_byte() etc.
+ *
+ * devfn is the combined PCI slot & function.
+ *
+ * offset is in bytes, from the start of config space for the
+ * specified bus & slot.
+ */
+
+static int __devinit tile_cfg_read(struct pci_bus *bus,
+ unsigned int devfn,
+ int offset,
+ int size,
+ u32 *val)
+{
+ struct pci_controller *controller = bus->sysdata;
+ int busnum = bus->number & 0xff;
+ int slot = (devfn >> 3) & 0x1f;
+ int function = devfn & 0x7;
+ u32 addr;
+ int config_mode = 1;
+
+ /*
+ * There is no bridge between the Tile and bus 0, so we
+ * use config0 to talk to bus 0.
+ *
+ * If we're talking to a bus other than zero then we
+ * must have found a bridge.
+ */
+ if (busnum == 0) {
+ /*
+ * We fake an empty slot for (busnum == 0) && (slot > 0),
+ * since there is only one slot on bus 0.
+ */
+ if (slot) {
+ *val = 0xFFFFFFFF;
+ return 0;
+ }
+ config_mode = 0;
+ }
+
+ addr = busnum << 20; /* Bus in 27:20 */
+ addr |= slot << 15; /* Slot (device) in 19:15 */
+ addr |= function << 12; /* Function is in 14:12 */
+ addr |= (offset & 0xFFF); /* byte address in 0:11 */
+
+ return hv_dev_pread(controller->hv_cfg_fd[config_mode], 0,
+ (HV_VirtAddr)(val), size, addr);
+}
+
+
+/*
+ * See tile_cfg_read() for relevent comments.
+ * Note that "val" is the value to write, not a pointer to that value.
+ */
+static int __devinit tile_cfg_write(struct pci_bus *bus,
+ unsigned int devfn,
+ int offset,
+ int size,
+ u32 val)
+{
+ struct pci_controller *controller = bus->sysdata;
+ int busnum = bus->number & 0xff;
+ int slot = (devfn >> 3) & 0x1f;
+ int function = devfn & 0x7;
+ u32 addr;
+ int config_mode = 1;
+ HV_VirtAddr valp = (HV_VirtAddr)&val;
+
+ /*
+ * For bus 0 slot 0 we use config 0 accesses.
+ */
+ if (busnum == 0) {
+ /*
+ * We fake an empty slot for (busnum == 0) && (slot > 0),
+ * since there is only one slot on bus 0.
+ */
+ if (slot)
+ return 0;
+ config_mode = 0;
+ }
+
+ addr = busnum << 20; /* Bus in 27:20 */
+ addr |= slot << 15; /* Slot (device) in 19:15 */
+ addr |= function << 12; /* Function is in 14:12 */
+ addr |= (offset & 0xFFF); /* byte address in 0:11 */
+
+#ifdef __BIG_ENDIAN
+ /* Point to the correct part of the 32-bit "val". */
+ valp += 4 - size;
+#endif
+
+ return hv_dev_pwrite(controller->hv_cfg_fd[config_mode], 0,
+ valp, size, addr);
+}
+
+
+static struct pci_ops tile_cfg_ops = {
+ .read = tile_cfg_read,
+ .write = tile_cfg_write,
+};
+
+
+/*
+ * In the following, each PCI controller's mem_resources[1]
+ * represents its (non-prefetchable) PCI memory resource.
+ * mem_resources[0] and mem_resources[2] refer to its PCI I/O and
+ * prefetchable PCI memory resources, respectively.
+ * For more details, see pci_setup_bridge() in setup-bus.c.
+ * By comparing the target PCI memory address against the
+ * end address of controller 0, we can determine the controller
+ * that should accept the PCI memory access.
+ */
+#define TILE_READ(size, type) \
+type _tile_read##size(unsigned long addr) \
+{ \
+ type val; \
+ int idx = 0; \
+ if (addr > controllers[0].mem_resources[1].end && \
+ addr > controllers[0].mem_resources[2].end) \
+ idx = 1; \
+ if (hv_dev_pread(controllers[idx].hv_mem_fd, 0, \
+ (HV_VirtAddr)(&val), sizeof(type), addr)) \
+ pr_err("PCI: read %zd bytes at 0x%lX failed\n", \
+ sizeof(type), addr); \
+ return val; \
+} \
+EXPORT_SYMBOL(_tile_read##size)
+
+TILE_READ(b, u8);
+TILE_READ(w, u16);
+TILE_READ(l, u32);
+TILE_READ(q, u64);
+
+#define TILE_WRITE(size, type) \
+void _tile_write##size(type val, unsigned long addr) \
+{ \
+ int idx = 0; \
+ if (addr > controllers[0].mem_resources[1].end && \
+ addr > controllers[0].mem_resources[2].end) \
+ idx = 1; \
+ if (hv_dev_pwrite(controllers[idx].hv_mem_fd, 0, \
+ (HV_VirtAddr)(&val), sizeof(type), addr)) \
+ pr_err("PCI: write %zd bytes at 0x%lX failed\n", \
+ sizeof(type), addr); \
+} \
+EXPORT_SYMBOL(_tile_write##size)
+
+TILE_WRITE(b, u8);
+TILE_WRITE(w, u16);
+TILE_WRITE(l, u32);
+TILE_WRITE(q, u64);
for_each_online_node(i)
register_one_node(i);
- for_each_present_cpu(i)
+ for (i = 0; i < smp_height * smp_width; ++i)
register_cpu(&cpu_devices[i], i);
return 0;
void *memchr(const void *s, int c, size_t n)
{
+ const uint32_t *last_word_ptr;
+ const uint32_t *p;
+ const char *last_byte_ptr;
+ uintptr_t s_int;
+ uint32_t goal, before_mask, v, bits;
+ char *ret;
+
+ if (__builtin_expect(n == 0, 0)) {
+ /* Don't dereference any memory if the array is empty. */
+ return NULL;
+ }
+
/* Get an aligned pointer. */
- const uintptr_t s_int = (uintptr_t) s;
- const uint32_t *p = (const uint32_t *)(s_int & -4);
+ s_int = (uintptr_t) s;
+ p = (const uint32_t *)(s_int & -4);
/* Create four copies of the byte for which we are looking. */
- const uint32_t goal = 0x01010101 * (uint8_t) c;
+ goal = 0x01010101 * (uint8_t) c;
/* Read the first word, but munge it so that bytes before the array
* will not match goal.
* Note that this shift count expression works because we know
* shift counts are taken mod 32.
*/
- const uint32_t before_mask = (1 << (s_int << 3)) - 1;
- uint32_t v = (*p | before_mask) ^ (goal & before_mask);
+ before_mask = (1 << (s_int << 3)) - 1;
+ v = (*p | before_mask) ^ (goal & before_mask);
/* Compute the address of the last byte. */
- const char *const last_byte_ptr = (const char *)s + n - 1;
+ last_byte_ptr = (const char *)s + n - 1;
/* Compute the address of the word containing the last byte. */
- const uint32_t *const last_word_ptr =
- (const uint32_t *)((uintptr_t) last_byte_ptr & -4);
-
- uint32_t bits;
- char *ret;
-
- if (__builtin_expect(n == 0, 0)) {
- /* Don't dereference any memory if the array is empty. */
- return NULL;
- }
+ last_word_ptr = (const uint32_t *)((uintptr_t) last_byte_ptr & -4);
while ((bits = __insn_seqb(v, goal)) == 0) {
if (__builtin_expect(p == last_word_ptr, 0)) {
* when we compare them.
*/
u32 my_ticket_;
+ u32 iterations = 0;
- /* Take out the next ticket; this will also stop would-be readers. */
- if (val & 1)
- val = get_rwlock(rwlock);
- rwlock->lock = __insn_addb(val, 1 << WR_NEXT_SHIFT);
+ /*
+ * Wait until there are no readers, then bump up the next
+ * field and capture the ticket value.
+ */
+ for (;;) {
+ if (!(val & 1)) {
+ if ((val >> RD_COUNT_SHIFT) == 0)
+ break;
+ rwlock->lock = val;
+ }
+ delay_backoff(iterations++);
+ val = __insn_tns((int *)&rwlock->lock);
+ }
- /* Extract my ticket value from the original word. */
+ /* Take out the next ticket and extract my ticket value. */
+ rwlock->lock = __insn_addb(val, 1 << WR_NEXT_SHIFT);
my_ticket_ = val >> WR_NEXT_SHIFT;
- /*
- * Wait until the "current" field matches our ticket, and
- * there are no remaining readers.
- */
+ /* Wait until the "current" field matches our ticket. */
for (;;) {
u32 curr_ = val >> WR_CURR_SHIFT;
- u32 readers = val >> RD_COUNT_SHIFT;
- u32 delta = ((my_ticket_ - curr_) & WR_MASK) + !!readers;
+ u32 delta = ((my_ticket_ - curr_) & WR_MASK);
if (likely(delta == 0))
break;
static void free_winch(struct winch *winch, int free_irq_ok)
{
+ if (free_irq_ok)
+ free_irq(WINCH_IRQ, winch);
+
list_del(&winch->list);
if (winch->pid != -1)
os_close_file(winch->fd);
if (winch->stack != 0)
free_stack(winch->stack, 0);
- if (free_irq_ok)
- free_irq(WINCH_IRQ, winch);
kfree(winch);
}
select HAVE_UNSTABLE_SCHED_CLOCK
select HAVE_IDE
select HAVE_OPROFILE
- select HAVE_PERF_EVENTS if (!M386 && !M486)
+ select HAVE_PERF_EVENTS
select HAVE_IRQ_WORK
select HAVE_IOREMAP_PROT
select HAVE_KPROBES
}
/* Return an pointer with offset calculated */
-static inline unsigned long __set_fixmap_offset(enum fixed_addresses idx,
- phys_addr_t phys, pgprot_t flags)
+static __always_inline unsigned long
+__set_fixmap_offset(enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags)
{
__set_fixmap(idx, phys, flags);
return fix_to_virt(idx) + (phys & (PAGE_SIZE - 1));
#define FAM10H_MMIO_CONF_ENABLE (1<<0)
#define FAM10H_MMIO_CONF_BUSRANGE_MASK 0xf
#define FAM10H_MMIO_CONF_BUSRANGE_SHIFT 2
-#define FAM10H_MMIO_CONF_BASE_MASK 0xfffffff
+#define FAM10H_MMIO_CONF_BASE_MASK 0xfffffffULL
#define FAM10H_MMIO_CONF_BASE_SHIFT 20
#define MSR_FAM10H_NODE_ID 0xc001100c
#define __PV_IS_CALLEE_SAVE(func) \
((struct paravirt_callee_save) { func })
-static inline unsigned long arch_local_save_flags(void)
+static inline notrace unsigned long arch_local_save_flags(void)
{
return PVOP_CALLEE0(unsigned long, pv_irq_ops.save_fl);
}
-static inline void arch_local_irq_restore(unsigned long f)
+static inline notrace void arch_local_irq_restore(unsigned long f)
{
PVOP_VCALLEE1(pv_irq_ops.restore_fl, f);
}
-static inline void arch_local_irq_disable(void)
+static inline notrace void arch_local_irq_disable(void)
{
PVOP_VCALLEE0(pv_irq_ops.irq_disable);
}
-static inline void arch_local_irq_enable(void)
+static inline notrace void arch_local_irq_enable(void)
{
PVOP_VCALLEE0(pv_irq_ops.irq_enable);
}
-static inline unsigned long arch_local_irq_save(void)
+static inline notrace unsigned long arch_local_irq_save(void)
{
unsigned long f;
#define UVH_APICID 0x002D0E00L
#define UV_APIC_PNODE_SHIFT 6
+#define UV_APICID_HIBIT_MASK 0xffff0000
+
/* Local Bus from cpu's perspective */
#define LOCAL_BUS_BASE 0x1c00000
#define LOCAL_BUS_SIZE (4 * 1024 * 1024)
}
}
+extern unsigned int uv_apicid_hibits;
static unsigned long uv_hub_ipi_value(int apicid, int vector, int mode)
{
+ apicid |= uv_apicid_hibits;
return (1UL << UVH_IPI_INT_SEND_SHFT) |
((apicid) << UVH_IPI_INT_APIC_ID_SHFT) |
(mode << UVH_IPI_INT_DELIVERY_MODE_SHFT) |
*
* SGI UV MMR definitions
*
- * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2007-2010 Silicon Graphics, Inc. All rights reserved.
*/
#ifndef _ASM_X86_UV_UV_MMRS_H
} s;
};
+/* ========================================================================= */
+/* UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK */
+/* ========================================================================= */
+#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK 0x320130UL
+#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_32 0x009f0
+
+#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_SHFT 0
+#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_MASK 0x00000000ffffffffUL
+
+union uvh_lb_target_physical_apic_id_mask_u {
+ unsigned long v;
+ struct uvh_lb_target_physical_apic_id_mask_s {
+ unsigned long bit_enables : 32; /* RW */
+ unsigned long rsvd_32_63 : 32; /* */
+ } s;
+};
+
/* ========================================================================= */
/* UVH_NODE_ID */
/* ========================================================================= */
#define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START)
#endif
-#ifndef machine_to_phys_mapping
-#define machine_to_phys_mapping ((unsigned long *)HYPERVISOR_VIRT_START)
-#endif
+#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START)
+#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END)
+#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>__MACH2PHYS_SHIFT)
/* Maximum number of virtual CPUs in multi-processor guests. */
#define MAX_VIRT_CPUS 32
/* And the trap vector is... */
#define TRAP_INSTR "int $0x82"
+#define __MACH2PHYS_VIRT_START 0xF5800000
+#define __MACH2PHYS_VIRT_END 0xF6800000
+
+#define __MACH2PHYS_SHIFT 2
+
/*
* Virtual addresses beyond this are not modifiable by guest OSes. The
* machine->physical mapping table starts at this address, read-only.
#define __HYPERVISOR_VIRT_END 0xFFFF880000000000
#define __MACH2PHYS_VIRT_START 0xFFFF800000000000
#define __MACH2PHYS_VIRT_END 0xFFFF804000000000
-
-#ifndef HYPERVISOR_VIRT_START
-#define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START)
-#define HYPERVISOR_VIRT_END mk_unsigned_long(__HYPERVISOR_VIRT_END)
-#endif
-
-#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START)
-#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END)
-#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>3)
-#ifndef machine_to_phys_mapping
-#define machine_to_phys_mapping ((unsigned long *)HYPERVISOR_VIRT_START)
-#endif
+#define __MACH2PHYS_SHIFT 3
/*
* int HYPERVISOR_set_segment_base(unsigned int which, unsigned long base)
#include <linux/types.h>
#include <linux/spinlock.h>
#include <linux/pfn.h>
+#include <linux/mm.h>
#include <asm/uaccess.h>
#include <asm/page.h>
#define MAX_DOMAIN_PAGES \
((unsigned long)((u64)CONFIG_XEN_MAX_DOMAIN_MEMORY * 1024 * 1024 * 1024 / PAGE_SIZE))
+extern unsigned long *machine_to_phys_mapping;
+extern unsigned int machine_to_phys_order;
extern unsigned long get_phys_to_machine(unsigned long pfn);
extern bool set_phys_to_machine(unsigned long pfn, unsigned long mfn);
if (xen_feature(XENFEAT_auto_translated_physmap))
return mfn;
-#if 0
if (unlikely((mfn >> machine_to_phys_order) != 0))
- return max_mapnr;
-#endif
+ return ~0;
pfn = 0;
/*
#include <linux/nmi.h>
#include <linux/module.h>
-/* For reliability, we're prepared to waste bits here. */
-static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly;
-
u64 hw_nmi_get_sample_period(void)
{
return (u64)(cpu_khz) * 1000 * 60;
}
#ifdef ARCH_HAS_NMI_WATCHDOG
+
+/* For reliability, we're prepared to waste bits here. */
+static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly;
+
void arch_trigger_all_cpu_backtrace(void)
{
int i;
static union uvh_apicid uvh_apicid;
int uv_min_hub_revision_id;
EXPORT_SYMBOL_GPL(uv_min_hub_revision_id);
+unsigned int uv_apicid_hibits;
+EXPORT_SYMBOL_GPL(uv_apicid_hibits);
static DEFINE_SPINLOCK(uv_nmi_lock);
static inline bool is_GRU_range(u64 start, u64 end)
uvh_apicid.s.pnode_shift = UV_APIC_PNODE_SHIFT;
}
+/*
+ * Add an extra bit as dictated by bios to the destination apicid of
+ * interrupts potentially passing through the UV HUB. This prevents
+ * a deadlock between interrupts and IO port operations.
+ */
+static void __init uv_set_apicid_hibit(void)
+{
+ union uvh_lb_target_physical_apic_id_mask_u apicid_mask;
+ unsigned long *mmr;
+
+ mmr = early_ioremap(UV_LOCAL_MMR_BASE |
+ UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK, sizeof(*mmr));
+ apicid_mask.v = *mmr;
+ early_iounmap(mmr, sizeof(*mmr));
+ uv_apicid_hibits = apicid_mask.s.bit_enables & UV_APICID_HIBIT_MASK;
+}
+
static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
{
int nodeid;
__get_cpu_var(x2apic_extra_bits) =
nodeid << (uvh_apicid.s.pnode_shift - 1);
uv_system_type = UV_NON_UNIQUE_APIC;
+ uv_set_apicid_hibit();
return 1;
}
}
int pnode;
pnode = uv_apicid_to_pnode(phys_apicid);
+ phys_apicid |= uv_apicid_hibits;
val = (1UL << UVH_IPI_INT_SEND_SHFT) |
(phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
int cpu = cpumask_first(cpumask);
if ((unsigned)cpu < nr_cpu_ids)
- return per_cpu(x86_cpu_to_apicid, cpu);
+ return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits;
else
return BAD_APICID;
}
if (cpumask_test_cpu(cpu, cpu_online_mask))
break;
}
- return per_cpu(x86_cpu_to_apicid, cpu);
+ return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits;
}
static unsigned int x2apic_get_apic_id(unsigned long x)
#endif
+static bool check_hw_exists(void)
+{
+ u64 val, val_new = 0;
+ int ret = 0;
+
+ val = 0xabcdUL;
+ ret |= checking_wrmsrl(x86_pmu.perfctr, val);
+ ret |= rdmsrl_safe(x86_pmu.perfctr, &val_new);
+ if (ret || val != val_new)
+ return false;
+
+ return true;
+}
+
static void reserve_ds_buffers(void);
static void release_ds_buffers(void);
pmu_check_apic();
+ /* sanity check that the hardware exists or is emulated */
+ if (!check_hw_exists()) {
+ pr_cont("Broken PMU hardware detected, software events only.\n");
+ return;
+ }
+
pr_cont("%s PMU driver.\n", x86_pmu.name);
if (x86_pmu.quirks)
* A tiny bit of offset fixup is necessary - 4*4 means the 4 words
* pushed above; +8 corresponds to copy_thread's esp0 setting.
*/
- pushl_cfi (TI_sysenter_return-THREAD_SIZE_asm+8+4*4)(%esp)
+ pushl_cfi ((TI_sysenter_return)-THREAD_SIZE_asm+8+4*4)(%esp)
CFI_REL_OFFSET eip, 0
pushl_cfi %eax
.endm
/* save partial stack frame */
+ .pushsection .kprobes.text, "ax"
ENTRY(save_args)
XCPT_FRAME
cld
ret
CFI_ENDPROC
END(save_args)
+ .popsection
ENTRY(save_rest)
PARTIAL_FRAME 1 REST_SKIP+8
dr6_p = (unsigned long *)ERR_PTR(args->err);
dr6 = *dr6_p;
+ /* If it's a single step, TRAP bits are random */
+ if (dr6 & DR_STEP)
+ return NOTIFY_DONE;
+
/* Do an early return if no trap bits are set in DR6 */
if ((dr6 & DR_TRAP_BITS) == 0)
return NOTIFY_DONE;
};
static u64 __cpuinitdata fam10h_pci_mmconf_base;
-static int __cpuinitdata fam10h_pci_mmconf_base_status;
static struct pci_hostbridge_probe pci_probes[] __cpuinitdata = {
{ 0, 0x18, PCI_VENDOR_ID_AMD, 0x1200 },
return start1 - start2;
}
-/*[47:0] */
-/* need to avoid (0xfd<<32) and (0xfe<<32), ht used space */
+#define MMCONF_UNIT (1ULL << FAM10H_MMIO_CONF_BASE_SHIFT)
+#define MMCONF_MASK (~(MMCONF_UNIT - 1))
+#define MMCONF_SIZE (MMCONF_UNIT << 8)
+/* need to avoid (0xfd<<32), (0xfe<<32), and (0xff<<32), ht used space */
#define FAM10H_PCI_MMCONF_BASE (0xfcULL<<32)
-#define BASE_VALID(b) ((b != (0xfdULL << 32)) && (b != (0xfeULL << 32)))
+#define BASE_VALID(b) ((b) + MMCONF_SIZE <= (0xfdULL<<32) || (b) >= (1ULL<<40))
static void __cpuinit get_fam10h_pci_mmconf_base(void)
{
int i;
struct range range[8];
/* only try to get setting from BSP */
- /* -1 or 1 */
- if (fam10h_pci_mmconf_base_status)
+ if (fam10h_pci_mmconf_base)
return;
if (!early_pci_allowed())
- goto fail;
+ return;
found = 0;
for (i = 0; i < ARRAY_SIZE(pci_probes); i++) {
}
if (!found)
- goto fail;
+ return;
/* SYS_CFG */
address = MSR_K8_SYSCFG;
/* TOP_MEM2 is not enabled? */
if (!(val & (1<<21))) {
- tom2 = 0;
+ tom2 = 1ULL << 32;
} else {
/* TOP_MEM2 */
address = MSR_K8_TOP_MEM2;
rdmsrl(address, val);
- tom2 = val & (0xffffULL<<32);
+ tom2 = max(val & 0xffffff800000ULL, 1ULL << 32);
}
if (base <= tom2)
- base = tom2 + (1ULL<<32);
+ base = (tom2 + 2 * MMCONF_UNIT - 1) & MMCONF_MASK;
/*
* need to check if the range is in the high mmio range that is
if (!(reg & 3))
continue;
- start = (((u64)reg) << 8) & (0xffULL << 32); /* 39:16 on 31:8*/
+ start = (u64)(reg & 0xffffff00) << 8; /* 39:16 on 31:8*/
reg = read_pci_config(bus, slot, 1, 0x84 + (i << 3));
- end = (((u64)reg) << 8) & (0xffULL << 32); /* 39:16 on 31:8*/
+ end = ((u64)(reg & 0xffffff00) << 8) | 0xffff; /* 39:16 on 31:8*/
- if (!end)
+ if (end < tom2)
continue;
range[hi_mmio_num].start = start;
if (range[hi_mmio_num - 1].end < base)
goto out;
- if (range[0].start > base)
+ if (range[0].start > base + MMCONF_SIZE)
goto out;
/* need to find one window */
- base = range[0].start - (1ULL << 32);
+ base = (range[0].start & MMCONF_MASK) - MMCONF_UNIT;
if ((base > tom2) && BASE_VALID(base))
goto out;
- base = range[hi_mmio_num - 1].end + (1ULL << 32);
- if ((base > tom2) && BASE_VALID(base))
+ base = (range[hi_mmio_num - 1].end + MMCONF_UNIT) & MMCONF_MASK;
+ if (BASE_VALID(base))
goto out;
/* need to find window between ranges */
- if (hi_mmio_num > 1)
- for (i = 0; i < hi_mmio_num - 1; i++) {
- if (range[i + 1].start > (range[i].end + (1ULL << 32))) {
- base = range[i].end + (1ULL << 32);
- if ((base > tom2) && BASE_VALID(base))
- goto out;
- }
+ for (i = 1; i < hi_mmio_num; i++) {
+ base = (range[i - 1].end + MMCONF_UNIT) & MMCONF_MASK;
+ val = range[i].start & MMCONF_MASK;
+ if (val >= base + MMCONF_SIZE && BASE_VALID(base))
+ goto out;
}
-
-fail:
- fam10h_pci_mmconf_base_status = -1;
return;
+
out:
fam10h_pci_mmconf_base = base;
- fam10h_pci_mmconf_base_status = 1;
}
void __cpuinit fam10h_check_enable_mmcfg(void)
/* only trust the one handle 256 buses, if acpi=off */
if (!acpi_pci_disabled || busnbits >= 8) {
- u64 base;
- base = val & (0xffffULL << 32);
- if (fam10h_pci_mmconf_base_status <= 0) {
+ u64 base = val & MMCONF_MASK;
+
+ if (!fam10h_pci_mmconf_base) {
fam10h_pci_mmconf_base = base;
- fam10h_pci_mmconf_base_status = 1;
return;
} else if (fam10h_pci_mmconf_base == base)
return;
* with 256 buses
*/
get_fam10h_pci_mmconf_base();
- if (fam10h_pci_mmconf_base_status <= 0)
+ if (!fam10h_pci_mmconf_base) {
+ pci_probe &= ~PCI_CHECK_ENABLE_AMD_MMCONF;
return;
+ }
printk(KERN_INFO "Enable MMCONFIG on AMD Family 10h\n");
val &= ~((FAM10H_MMIO_CONF_BASE_MASK<<FAM10H_MMIO_CONF_BASE_SHIFT) |
static void __cpuinit calculate_tlb_offset(void)
{
- int cpu, node, nr_node_vecs;
+ int cpu, node, nr_node_vecs, idx = 0;
/*
* we are changing tlb_vector_offset for each CPU in runtime, but this
* will not cause inconsistency, as the write is atomic under X86. we
nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
for_each_online_node(node) {
- int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
+ int node_offset = (idx % NUM_INVALIDATE_TLB_VECTORS) *
nr_node_vecs;
int cpu_offset = 0;
for_each_cpu(cpu, cpumask_of_node(node)) {
cpu_offset++;
cpu_offset = cpu_offset % nr_node_vecs;
}
+ idx++;
}
}
* the below initialization can't be in firmware because the
* messaging IRQ will be determined by the OS
*/
- apicid = uvhub_to_first_apicid(uvhub);
+ apicid = uvhub_to_first_apicid(uvhub) | uv_apicid_hibits;
uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG,
((apicid << 32) | vector));
}
apicid = cpu_physical_id(cpu);
pnode = uv_apicid_to_pnode(apicid);
+ apicid |= uv_apicid_hibits;
val = (1UL << UVH_IPI_INT_SEND_SHFT) |
(apicid << UVH_IPI_INT_APIC_ID_SHFT) |
(X86_PLATFORM_IPI_VECTOR << UVH_IPI_INT_VECTOR_SHFT);
static int uv_setup_intr(int cpu, u64 expires)
{
u64 val;
+ unsigned long apicid = cpu_physical_id(cpu) | uv_apicid_hibits;
int pnode = uv_cpu_to_pnode(cpu);
uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG,
UVH_EVENT_OCCURRED0_RTC1_MASK);
val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) |
- ((u64)cpu_physical_id(cpu) << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT);
+ ((u64)apicid << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT);
/* Set configuration */
uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, val);
enum xen_domain_type xen_domain_type = XEN_NATIVE;
EXPORT_SYMBOL_GPL(xen_domain_type);
+unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
+EXPORT_SYMBOL(machine_to_phys_mapping);
+unsigned int machine_to_phys_order;
+EXPORT_SYMBOL(machine_to_phys_order);
+
struct start_info *xen_start_info;
EXPORT_SYMBOL_GPL(xen_start_info);
/* First C function to be called on Xen boot */
asmlinkage void __init xen_start_kernel(void)
{
+ struct physdev_set_iopl set_iopl;
+ int rc;
pgd_t *pgd;
if (!xen_start_info)
xen_domain_type = XEN_PV_DOMAIN;
+ xen_setup_machphys_mapping();
+
/* Install Xen paravirt ops */
pv_info = xen_info;
pv_init_ops = xen_init_ops;
/* Allocate and initialize top and mid mfn levels for p2m structure */
xen_build_mfn_list_list();
- init_mm.pgd = pgd;
-
/* keep using Xen gdt for now; no urgent need to change it */
#ifdef CONFIG_X86_32
#else
pv_info.kernel_rpl = 0;
#endif
-
/* set the limit of our address space */
xen_reserve_top();
+ /* We used to do this in xen_arch_setup, but that is too late on AMD
+ * were early_cpu_init (run before ->arch_setup()) calls early_amd_init
+ * which pokes 0xcf8 port.
+ */
+ set_iopl.iopl = 1;
+ rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
+ if (rc != 0)
+ xen_raw_printk("physdev_op failed %d\n", rc);
+
#ifdef CONFIG_X86_32
/* set up basic CPUID stuff */
cpu_detect(&new_cpu_data);
set_page_prot(pmd, PAGE_KERNEL_RO);
}
+void __init xen_setup_machphys_mapping(void)
+{
+ struct xen_machphys_mapping mapping;
+ unsigned long machine_to_phys_nr_ents;
+
+ if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
+ machine_to_phys_mapping = (unsigned long *)mapping.v_start;
+ machine_to_phys_nr_ents = mapping.max_mfn + 1;
+ } else {
+ machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
+ }
+ machine_to_phys_order = fls(machine_to_phys_nr_ents - 1);
+}
+
#ifdef CONFIG_X86_64
static void convert_pfn_mfn(void *v)
{
return pgd;
}
#else /* !CONFIG_X86_64 */
-static RESERVE_BRK_ARRAY(pmd_t, level2_kernel_pgt, PTRS_PER_PMD);
+static RESERVE_BRK_ARRAY(pmd_t, initial_kernel_pmd, PTRS_PER_PMD);
+static RESERVE_BRK_ARRAY(pmd_t, swapper_kernel_pmd, PTRS_PER_PMD);
+
+static __init void xen_write_cr3_init(unsigned long cr3)
+{
+ unsigned long pfn = PFN_DOWN(__pa(swapper_pg_dir));
+
+ BUG_ON(read_cr3() != __pa(initial_page_table));
+ BUG_ON(cr3 != __pa(swapper_pg_dir));
+
+ /*
+ * We are switching to swapper_pg_dir for the first time (from
+ * initial_page_table) and therefore need to mark that page
+ * read-only and then pin it.
+ *
+ * Xen disallows sharing of kernel PMDs for PAE
+ * guests. Therefore we must copy the kernel PMD from
+ * initial_page_table into a new kernel PMD to be used in
+ * swapper_pg_dir.
+ */
+ swapper_kernel_pmd =
+ extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
+ memcpy(swapper_kernel_pmd, initial_kernel_pmd,
+ sizeof(pmd_t) * PTRS_PER_PMD);
+ swapper_pg_dir[KERNEL_PGD_BOUNDARY] =
+ __pgd(__pa(swapper_kernel_pmd) | _PAGE_PRESENT);
+ set_page_prot(swapper_kernel_pmd, PAGE_KERNEL_RO);
+
+ set_page_prot(swapper_pg_dir, PAGE_KERNEL_RO);
+ xen_write_cr3(cr3);
+ pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, pfn);
+
+ pin_pagetable_pfn(MMUEXT_UNPIN_TABLE,
+ PFN_DOWN(__pa(initial_page_table)));
+ set_page_prot(initial_page_table, PAGE_KERNEL);
+ set_page_prot(initial_kernel_pmd, PAGE_KERNEL);
+
+ pv_mmu_ops.write_cr3 = &xen_write_cr3;
+}
__init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd,
unsigned long max_pfn)
{
pmd_t *kernel_pmd;
- level2_kernel_pgt = extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
+ initial_kernel_pmd =
+ extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->pt_base) +
xen_start_info->nr_pt_frames * PAGE_SIZE +
512*1024);
kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd);
- memcpy(level2_kernel_pgt, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
+ memcpy(initial_kernel_pmd, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
- xen_map_identity_early(level2_kernel_pgt, max_pfn);
+ xen_map_identity_early(initial_kernel_pmd, max_pfn);
- memcpy(swapper_pg_dir, pgd, sizeof(pgd_t) * PTRS_PER_PGD);
- set_pgd(&swapper_pg_dir[KERNEL_PGD_BOUNDARY],
- __pgd(__pa(level2_kernel_pgt) | _PAGE_PRESENT));
+ memcpy(initial_page_table, pgd, sizeof(pgd_t) * PTRS_PER_PGD);
+ initial_page_table[KERNEL_PGD_BOUNDARY] =
+ __pgd(__pa(initial_kernel_pmd) | _PAGE_PRESENT);
- set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
- set_page_prot(swapper_pg_dir, PAGE_KERNEL_RO);
+ set_page_prot(initial_kernel_pmd, PAGE_KERNEL_RO);
+ set_page_prot(initial_page_table, PAGE_KERNEL_RO);
set_page_prot(empty_zero_page, PAGE_KERNEL_RO);
pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
- xen_write_cr3(__pa(swapper_pg_dir));
-
- pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(swapper_pg_dir)));
+ pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE,
+ PFN_DOWN(__pa(initial_page_table)));
+ xen_write_cr3(__pa(initial_page_table));
memblock_x86_reserve_range(__pa(xen_start_info->pt_base),
__pa(xen_start_info->pt_base +
xen_start_info->nr_pt_frames * PAGE_SIZE),
"XEN PAGETABLES");
- return swapper_pg_dir;
+ return initial_page_table;
}
#endif /* CONFIG_X86_64 */
.write_cr2 = xen_write_cr2,
.read_cr3 = xen_read_cr3,
+#ifdef CONFIG_X86_32
+ .write_cr3 = xen_write_cr3_init,
+#else
.write_cr3 = xen_write_cr3,
+#endif
.flush_tlb_user = xen_flush_tlb,
.flush_tlb_kernel = xen_flush_tlb,
prot = __pgprot(pgprot_val(prot) | _PAGE_IOMAP);
- vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
+ BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_RESERVED | VM_IO)) ==
+ (VM_PFNMAP | VM_RESERVED | VM_IO)));
rmd.mfn = mfn;
rmd.prot = prot;
#include <xen/interface/callback.h>
#include <xen/interface/memory.h>
#include <xen/interface/physdev.h>
-#include <xen/interface/memory.h>
#include <xen/features.h>
#include "xen-ops.h"
else
extra_pages = 0;
- if (!xen_initial_domain())
- xen_add_extra_mem(extra_pages);
+ xen_add_extra_mem(extra_pages);
return "Xen";
}
void __init xen_arch_setup(void)
{
- struct physdev_set_iopl set_iopl;
- int rc;
-
xen_panic_handler_init();
HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments);
xen_enable_sysenter();
xen_enable_syscall();
- set_iopl.iopl = 1;
- rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
- if (rc != 0)
- printk(KERN_INFO "physdev_op failed %d\n", rc);
-
#ifdef CONFIG_ACPI
if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
{
unsigned int nr_reads = 0, nr_writes = 0;
unsigned int max_nr_reads = throtl_grp_quantum*3/4;
- unsigned int max_nr_writes = throtl_grp_quantum - nr_reads;
+ unsigned int max_nr_writes = throtl_grp_quantum - max_nr_reads;
struct bio *bio;
/* Try to dispatch 75% READS and 25% WRITES */
{
struct request_queue *q;
int cnt = FD_MAX_UNITS;
- struct request *rq;
+ struct request *rq = NULL;
/* Find next queue we can dispatch from */
fdc_queue = fdc_queue + 1;
{
struct request_queue *q;
int old_pos = fdc_queue;
- struct request *rq;
+ struct request *rq = NULL;
do {
q = unit[fdc_queue].disk->queue;
MODULE_LICENSE("GPL");
static DEFINE_MUTEX(cciss_mutex);
+static struct proc_dir_entry *proc_cciss;
#include "cciss_cmd.h"
#include "cciss.h"
#define ENG_GIG_FACTOR (ENG_GIG/512)
#define ENGAGE_SCSI "engage scsi"
-static struct proc_dir_entry *proc_cciss;
-
static void cciss_seq_show_header(struct seq_file *seq)
{
ctlr_info_t *h = seq->private;
struct blk_shadow {
struct blkif_request req;
- unsigned long request;
+ struct request *request;
unsigned long frame[BLKIF_MAX_SEGMENTS_PER_REQUEST];
};
unsigned long id)
{
info->shadow[id].req.id = info->shadow_free;
- info->shadow[id].request = 0;
+ info->shadow[id].request = NULL;
info->shadow_free = id;
}
}
/*
- * blkif_queue_request
+ * Generate a Xen blkfront IO request from a blk layer request. Reads
+ * and writes are handled as expected. Since we lack a loose flush
+ * request, we map flushes into a full ordered barrier.
*
- * request block io
- *
- * id: for guest use only.
- * operation: BLKIF_OP_{READ,WRITE,PROBE}
- * buffer: buffer to read/write into. this should be a
- * virtual address in the guest os.
+ * @req: a request struct
*/
static int blkif_queue_request(struct request *req)
{
/* Fill out a communications ring structure. */
ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
id = get_id_from_freelist(info);
- info->shadow[id].request = (unsigned long)req;
+ info->shadow[id].request = req;
ring_req->id = id;
ring_req->sector_number = (blkif_sector_t)blk_rq_pos(req);
ring_req->operation = rq_data_dir(req) ?
BLKIF_OP_WRITE : BLKIF_OP_READ;
+ if (req->cmd_flags & (REQ_FLUSH | REQ_FUA)) {
+ /*
+ * Ideally we could just do an unordered
+ * flush-to-disk, but all we have is a full write
+ * barrier at the moment. However, a barrier write is
+ * a superset of FUA, so we can implement it the same
+ * way. (It's also a FLUSH+FUA, since it is
+ * guaranteed ordered WRT previous writes.)
+ */
+ ring_req->operation = BLKIF_OP_WRITE_BARRIER;
+ }
+
ring_req->nr_segments = blk_rq_map_sg(req->q, req, info->sg);
BUG_ON(ring_req->nr_segments > BLKIF_MAX_SEGMENTS_PER_REQUEST);
bret = RING_GET_RESPONSE(&info->ring, i);
id = bret->id;
- req = (struct request *)info->shadow[id].request;
+ req = info->shadow[id].request;
blkif_completion(&info->shadow[id]);
printk(KERN_WARNING "blkfront: %s: write barrier op failed\n",
info->gd->disk_name);
error = -EOPNOTSUPP;
+ }
+ if (unlikely(bret->status == BLKIF_RSP_ERROR &&
+ info->shadow[id].req.nr_segments == 0)) {
+ printk(KERN_WARNING "blkfront: %s: empty write barrier op failed\n",
+ info->gd->disk_name);
+ error = -EOPNOTSUPP;
+ }
+ if (unlikely(error)) {
+ if (error == -EOPNOTSUPP)
+ error = 0;
info->feature_flush = 0;
xlvbd_flush(info);
}
/* Stage 3: Find pending requests and requeue them. */
for (i = 0; i < BLK_RING_SIZE; i++) {
/* Not in use? */
- if (copy[i].request == 0)
+ if (!copy[i].request)
continue;
/* Grab a request slot and copy shadow state into it. */
req->seg[j].gref,
info->xbdev->otherend_id,
pfn_to_mfn(info->shadow[req->id].frame[j]),
- rq_data_dir(
- (struct request *)
- info->shadow[req->id].request));
+ rq_data_dir(info->shadow[req->id].request));
info->shadow[req->id].req = *req;
info->ring.req_prod_pvt++;
*/
info->feature_flush = 0;
- /*
- * The driver doesn't properly handled empty flushes, so
- * lets disable barrier support for now.
- */
-#if 0
if (!err && barrier)
- info->feature_flush = REQ_FLUSH;
-#endif
+ info->feature_flush = REQ_FLUSH | REQ_FUA;
err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size);
if (err) {
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/wait.h>
+#include <linux/acpi.h>
#include "tpm.h"
#define TPM_HEADER_SIZE 10
static LIST_HEAD(tis_chips);
static DEFINE_SPINLOCK(tis_lock);
+#ifdef CONFIG_ACPI
+static int is_itpm(struct pnp_dev *dev)
+{
+ struct acpi_device *acpi = pnp_acpi_device(dev);
+ struct acpi_hardware_id *id;
+
+ list_for_each_entry(id, &acpi->pnp.ids, list) {
+ if (!strcmp("INTC0102", id->id))
+ return 1;
+ }
+
+ return 0;
+}
+#else
+static int is_itpm(struct pnp_dev *dev)
+{
+ return 0;
+}
+#endif
+
static int check_locality(struct tpm_chip *chip, int l)
{
if ((ioread8(chip->vendor.iobase + TPM_ACCESS(l)) &
"1.2 TPM (device-id 0x%X, rev-id %d)\n",
vendor >> 16, ioread8(chip->vendor.iobase + TPM_RID(0)));
+ if (is_itpm(to_pnp_dev(dev)))
+ itpm = 1;
+
if (itpm)
dev_info(dev, "Intel iTPM workaround enabled\n");
nr_queues = use_multiport(portdev) ? (nr_ports + 1) * 2 : 2;
vqs = kmalloc(nr_queues * sizeof(struct virtqueue *), GFP_KERNEL);
- if (!vqs) {
- err = -ENOMEM;
- goto fail;
- }
io_callbacks = kmalloc(nr_queues * sizeof(vq_callback_t *), GFP_KERNEL);
- if (!io_callbacks) {
- err = -ENOMEM;
- goto free_vqs;
- }
io_names = kmalloc(nr_queues * sizeof(char *), GFP_KERNEL);
- if (!io_names) {
- err = -ENOMEM;
- goto free_callbacks;
- }
portdev->in_vqs = kmalloc(nr_ports * sizeof(struct virtqueue *),
GFP_KERNEL);
- if (!portdev->in_vqs) {
- err = -ENOMEM;
- goto free_names;
- }
portdev->out_vqs = kmalloc(nr_ports * sizeof(struct virtqueue *),
GFP_KERNEL);
- if (!portdev->out_vqs) {
+ if (!vqs || !io_callbacks || !io_names || !portdev->in_vqs ||
+ !portdev->out_vqs) {
err = -ENOMEM;
- goto free_invqs;
+ goto free;
}
/*
io_callbacks,
(const char **)io_names);
if (err)
- goto free_outvqs;
+ goto free;
j = 0;
portdev->in_vqs[0] = vqs[0];
portdev->out_vqs[i] = vqs[j + 1];
}
}
- kfree(io_callbacks);
kfree(io_names);
+ kfree(io_callbacks);
kfree(vqs);
return 0;
-free_names:
- kfree(io_names);
-free_callbacks:
- kfree(io_callbacks);
-free_outvqs:
+free:
kfree(portdev->out_vqs);
-free_invqs:
kfree(portdev->in_vqs);
-free_vqs:
+ kfree(io_names);
+ kfree(io_callbacks);
kfree(vqs);
-fail:
+
return err;
}
obj-$(CONFIG_EDAC_MM_EDAC) += edac_core.o
obj-$(CONFIG_EDAC_MCE) += edac_mce.o
-edac_core-objs := edac_mc.o edac_device.o edac_mc_sysfs.o edac_pci_sysfs.o
-edac_core-objs += edac_module.o edac_device_sysfs.o
+edac_core-y := edac_mc.o edac_device.o edac_mc_sysfs.o edac_pci_sysfs.o
+edac_core-y += edac_module.o edac_device_sysfs.o
ifdef CONFIG_PCI
-edac_core-objs += edac_pci.o edac_pci_sysfs.o
+edac_core-y += edac_pci.o edac_pci_sysfs.o
endif
obj-$(CONFIG_EDAC_MCE_INJ) += mce_amd_inj.o
-edac_mce_amd-objs := mce_amd.o
+edac_mce_amd-y := mce_amd.o
obj-$(CONFIG_EDAC_DECODE_MCE) += edac_mce_amd.o
obj-$(CONFIG_EDAC_AMD76X) += amd76x_edac.o
return 0;
err_sysfs_create:
- while (i-- >= 0)
+ while (--i >= 0)
sysfs_remove_file(mce_kobj, &sysfs_attrs[i]->attr);
kobject_del(mce_kobj);
*/
#include <linux/bug.h>
+#include <linux/delay.h>
#include <linux/device.h>
#include <linux/firewire.h>
#include <linux/firewire-constants.h>
#include <asm/unaligned.h>
#include <net/arp.h>
-#define FWNET_MAX_FRAGMENTS 25 /* arbitrary limit */
-#define FWNET_ISO_PAGE_COUNT (PAGE_SIZE < 16 * 1024 ? 4 : 2)
+/* rx limits */
+#define FWNET_MAX_FRAGMENTS 30 /* arbitrary, > TX queue depth */
+#define FWNET_ISO_PAGE_COUNT (PAGE_SIZE < 16*1024 ? 4 : 2)
+
+/* tx limits */
+#define FWNET_MAX_QUEUED_DATAGRAMS 20 /* < 64 = number of tlabels */
+#define FWNET_MIN_QUEUED_DATAGRAMS 10 /* should keep AT DMA busy enough */
+#define FWNET_TX_QUEUE_LEN FWNET_MAX_QUEUED_DATAGRAMS /* ? */
#define IEEE1394_BROADCAST_CHANNEL 31
#define IEEE1394_ALL_NODES (0xffc0 | 0x003f)
struct fw_address_handler handler;
u64 local_fifo;
- /* List of packets to be sent */
- struct list_head packet_list;
- /*
- * List of packets that were broadcasted. When we get an ISO interrupt
- * one of them has been sent
- */
- struct list_head broadcasted_list;
- /* List of packets that have been sent but not yet acked */
- struct list_head sent_list;
+ /* Number of tx datagrams that have been queued but not yet acked */
+ int queued_datagrams;
struct list_head peer_list;
struct fw_card *card;
unsigned pdg_size; /* pd_list size */
u16 datagram_label; /* outgoing datagram label */
- unsigned max_payload; /* includes RFC2374_FRAG_HDR_SIZE overhead */
+ u16 max_payload; /* includes RFC2374_FRAG_HDR_SIZE overhead */
int node_id;
int generation;
unsigned speed;
/* This is our task struct. It's used for the packet complete callback. */
struct fwnet_packet_task {
- /*
- * ptask can actually be on dev->packet_list, dev->broadcasted_list,
- * or dev->sent_list depending on its current state.
- */
- struct list_head pt_link;
struct fw_transaction transaction;
struct rfc2734_header hdr;
struct sk_buff *skb;
struct fwnet_device *dev;
int outstanding_pkts;
- unsigned max_payload;
u64 fifo_addr;
u16 dest_node;
+ u16 max_payload;
u8 generation;
u8 speed;
+ u8 enqueued;
};
/*
net->stats.rx_packets++;
net->stats.rx_bytes += skb->len;
}
- if (netif_queue_stopped(net))
- netif_wake_queue(net);
return 0;
net->stats.rx_dropped++;
dev_kfree_skb_any(skb);
- if (netif_queue_stopped(net))
- netif_wake_queue(net);
return -ENOENT;
}
* Datagram is not complete, we're done for the
* moment.
*/
- spin_unlock_irqrestore(&dev->lock, flags);
-
- return 0;
+ retval = 0;
fail:
spin_unlock_irqrestore(&dev->lock, flags);
- if (netif_queue_stopped(net))
- netif_wake_queue(net);
-
return retval;
}
kmem_cache_free(fwnet_packet_task_cache, ptask);
}
+/* Caller must hold dev->lock. */
+static void dec_queued_datagrams(struct fwnet_device *dev)
+{
+ if (--dev->queued_datagrams == FWNET_MIN_QUEUED_DATAGRAMS)
+ netif_wake_queue(dev->netdev);
+}
+
static int fwnet_send_packet(struct fwnet_packet_task *ptask);
static void fwnet_transmit_packet_done(struct fwnet_packet_task *ptask)
{
struct fwnet_device *dev = ptask->dev;
+ struct sk_buff *skb = ptask->skb;
unsigned long flags;
bool free;
ptask->outstanding_pkts--;
/* Check whether we or the networking TX soft-IRQ is last user. */
- free = (ptask->outstanding_pkts == 0 && !list_empty(&ptask->pt_link));
+ free = (ptask->outstanding_pkts == 0 && ptask->enqueued);
+ if (free)
+ dec_queued_datagrams(dev);
- if (ptask->outstanding_pkts == 0)
- list_del(&ptask->pt_link);
+ if (ptask->outstanding_pkts == 0) {
+ dev->netdev->stats.tx_packets++;
+ dev->netdev->stats.tx_bytes += skb->len;
+ }
spin_unlock_irqrestore(&dev->lock, flags);
u16 fg_off;
u16 datagram_label;
u16 lf;
- struct sk_buff *skb;
/* Update the ptask to point to the next fragment and send it */
lf = fwnet_get_hdr_lf(&ptask->hdr);
datagram_label = fwnet_get_hdr_dgl(&ptask->hdr);
break;
}
- skb = ptask->skb;
+
skb_pull(skb, ptask->max_payload);
if (ptask->outstanding_pkts > 1) {
fwnet_make_sf_hdr(&ptask->hdr, RFC2374_HDR_INTFRAG,
fwnet_free_ptask(ptask);
}
+static void fwnet_transmit_packet_failed(struct fwnet_packet_task *ptask)
+{
+ struct fwnet_device *dev = ptask->dev;
+ unsigned long flags;
+ bool free;
+
+ spin_lock_irqsave(&dev->lock, flags);
+
+ /* One fragment failed; don't try to send remaining fragments. */
+ ptask->outstanding_pkts = 0;
+
+ /* Check whether we or the networking TX soft-IRQ is last user. */
+ free = ptask->enqueued;
+ if (free)
+ dec_queued_datagrams(dev);
+
+ dev->netdev->stats.tx_dropped++;
+ dev->netdev->stats.tx_errors++;
+
+ spin_unlock_irqrestore(&dev->lock, flags);
+
+ if (free)
+ fwnet_free_ptask(ptask);
+}
+
static void fwnet_write_complete(struct fw_card *card, int rcode,
void *payload, size_t length, void *data)
{
ptask = data;
- if (rcode == RCODE_COMPLETE)
+ if (rcode == RCODE_COMPLETE) {
fwnet_transmit_packet_done(ptask);
- else
+ } else {
fw_error("fwnet_write_complete: failed: %x\n", rcode);
- /* ??? error recovery */
+ fwnet_transmit_packet_failed(ptask);
+ }
}
static int fwnet_send_packet(struct fwnet_packet_task *ptask)
spin_lock_irqsave(&dev->lock, flags);
/* If the AT tasklet already ran, we may be last user. */
- free = (ptask->outstanding_pkts == 0 && list_empty(&ptask->pt_link));
+ free = (ptask->outstanding_pkts == 0 && !ptask->enqueued);
if (!free)
- list_add_tail(&ptask->pt_link, &dev->broadcasted_list);
+ ptask->enqueued = true;
+ else
+ dec_queued_datagrams(dev);
spin_unlock_irqrestore(&dev->lock, flags);
spin_lock_irqsave(&dev->lock, flags);
/* If the AT tasklet already ran, we may be last user. */
- free = (ptask->outstanding_pkts == 0 && list_empty(&ptask->pt_link));
+ free = (ptask->outstanding_pkts == 0 && !ptask->enqueued);
if (!free)
- list_add_tail(&ptask->pt_link, &dev->sent_list);
+ ptask->enqueued = true;
+ else
+ dec_queued_datagrams(dev);
spin_unlock_irqrestore(&dev->lock, flags);
struct fwnet_peer *peer;
unsigned long flags;
+ spin_lock_irqsave(&dev->lock, flags);
+
+ /* Can this happen? */
+ if (netif_queue_stopped(dev->netdev)) {
+ spin_unlock_irqrestore(&dev->lock, flags);
+
+ return NETDEV_TX_BUSY;
+ }
+
ptask = kmem_cache_alloc(fwnet_packet_task_cache, GFP_ATOMIC);
if (ptask == NULL)
goto fail;
proto = hdr_buf.h_proto;
dg_size = skb->len;
- /* serialize access to peer, including peer->datagram_label */
- spin_lock_irqsave(&dev->lock, flags);
-
/*
* Set the transmission type for the packet. ARP packets and IP
* broadcast packets are sent via GASP.
peer = fwnet_peer_find_by_guid(dev, be64_to_cpu(guid));
if (!peer || peer->fifo == FWNET_NO_FIFO_ADDR)
- goto fail_unlock;
+ goto fail;
generation = peer->generation;
dest_node = peer->node_id;
max_payload += RFC2374_FRAG_HDR_SIZE;
}
+ if (++dev->queued_datagrams == FWNET_MAX_QUEUED_DATAGRAMS)
+ netif_stop_queue(dev->netdev);
+
spin_unlock_irqrestore(&dev->lock, flags);
ptask->max_payload = max_payload;
- INIT_LIST_HEAD(&ptask->pt_link);
+ ptask->enqueued = 0;
fwnet_send_packet(ptask);
return NETDEV_TX_OK;
- fail_unlock:
- spin_unlock_irqrestore(&dev->lock, flags);
fail:
+ spin_unlock_irqrestore(&dev->lock, flags);
+
if (ptask)
kmem_cache_free(fwnet_packet_task_cache, ptask);
net->addr_len = FWNET_ALEN;
net->hard_header_len = FWNET_HLEN;
net->type = ARPHRD_IEEE1394;
- net->tx_queue_len = 10;
+ net->tx_queue_len = FWNET_TX_QUEUE_LEN;
}
/* caller must hold fwnet_device_mutex */
dev->broadcast_rcv_context = NULL;
dev->broadcast_xmt_max_payload = 0;
dev->broadcast_xmt_datagramlabel = 0;
-
dev->local_fifo = FWNET_NO_FIFO_ADDR;
-
- INIT_LIST_HEAD(&dev->packet_list);
- INIT_LIST_HEAD(&dev->broadcasted_list);
- INIT_LIST_HEAD(&dev->sent_list);
+ dev->queued_datagrams = 0;
INIT_LIST_HEAD(&dev->peer_list);
-
dev->card = card;
dev->netdev = net;
struct fwnet_peer *peer = dev_get_drvdata(_dev);
struct fwnet_device *dev = peer->dev;
struct net_device *net;
- struct fwnet_packet_task *ptask, *pt_next;
+ int i;
mutex_lock(&fwnet_device_mutex);
dev->card);
fw_iso_context_destroy(dev->broadcast_rcv_context);
}
- list_for_each_entry_safe(ptask, pt_next,
- &dev->packet_list, pt_link) {
- dev_kfree_skb_any(ptask->skb);
- kmem_cache_free(fwnet_packet_task_cache, ptask);
- }
- list_for_each_entry_safe(ptask, pt_next,
- &dev->broadcasted_list, pt_link) {
- dev_kfree_skb_any(ptask->skb);
- kmem_cache_free(fwnet_packet_task_cache, ptask);
- }
- list_for_each_entry_safe(ptask, pt_next,
- &dev->sent_list, pt_link) {
- dev_kfree_skb_any(ptask->skb);
- kmem_cache_free(fwnet_packet_task_cache, ptask);
- }
+ for (i = 0; dev->queued_datagrams && i < 5; i++)
+ ssleep(1);
+ WARN_ON(dev->queued_datagrams);
list_del(&dev->dev_link);
free_netdev(net);
int count = 0, ro, fail = 0;
struct drm_crtc_helper_funcs *crtc_funcs;
int ret = 0;
+ int i;
DRM_DEBUG_KMS("\n");
if (ret != 0)
goto fail;
}
+ DRM_DEBUG_KMS("Setting connector DPMS state to on\n");
+ for (i = 0; i < set->num_connectors; i++) {
+ DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
+ drm_get_connector_name(set->connectors[i]));
+ set->connectors[i]->dpms = DRM_MODE_DPMS_ON;
+ }
kfree(save_connectors);
kfree(save_encoders);
struct delayed_work *delayed_work = to_delayed_work(work);
struct drm_device *dev = container_of(delayed_work, struct drm_device, mode_config.output_poll_work);
struct drm_connector *connector;
- enum drm_connector_status old_status, status;
+ enum drm_connector_status old_status;
bool repoll = false, changed = false;
if (!drm_kms_helper_poll)
!(connector->polled & DRM_CONNECTOR_POLL_HPD))
continue;
- status = connector->funcs->detect(connector, false);
- if (old_status != status)
+ connector->status = connector->funcs->detect(connector, false);
+ DRM_DEBUG_KMS("connector status updated to %d\n", connector->status);
+ if (old_status != connector->status)
changed = true;
}
static uint32_t i915_gem_get_gtt_alignment(struct drm_gem_object *obj);
-static int i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj,
- bool pipelined);
+static int i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj);
static void i915_gem_object_flush_gtt_write_domain(struct drm_gem_object *obj);
static void i915_gem_object_flush_cpu_write_domain(struct drm_gem_object *obj);
static int i915_gem_object_set_to_cpu_domain(struct drm_gem_object *obj,
if (reg->gpu) {
int ret;
- ret = i915_gem_object_flush_gpu_write_domain(obj, true);
+ ret = i915_gem_object_flush_gpu_write_domain(obj);
if (ret)
return ret;
/** Flushes any GPU write domain for the object if it's dirty. */
static int
-i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj,
- bool pipelined)
+i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj)
{
struct drm_device *dev = obj->dev;
uint32_t old_write_domain;
obj->read_domains,
old_write_domain);
- if (pipelined)
- return 0;
-
- return i915_gem_object_wait_rendering(obj, true);
+ return 0;
}
/** Flushes the GTT write domain for the object if it's dirty. */
if (obj_priv->gtt_space == NULL)
return -EINVAL;
- ret = i915_gem_object_flush_gpu_write_domain(obj, false);
+ ret = i915_gem_object_flush_gpu_write_domain(obj);
if (ret != 0)
return ret;
+ ret = i915_gem_object_wait_rendering(obj, true);
+ if (ret)
+ return ret;
i915_gem_object_flush_cpu_write_domain(obj);
- if (write) {
- ret = i915_gem_object_wait_rendering(obj, true);
- if (ret)
- return ret;
- }
-
old_write_domain = obj->write_domain;
old_read_domains = obj->read_domains;
if (obj_priv->gtt_space == NULL)
return -EINVAL;
- ret = i915_gem_object_flush_gpu_write_domain(obj, true);
+ ret = i915_gem_object_flush_gpu_write_domain(obj);
if (ret)
return ret;
uint32_t old_write_domain, old_read_domains;
int ret;
- ret = i915_gem_object_flush_gpu_write_domain(obj, false);
+ ret = i915_gem_object_flush_gpu_write_domain(obj);
if (ret != 0)
return ret;
+ ret = i915_gem_object_wait_rendering(obj, true);
+ if (ret)
+ return ret;
i915_gem_object_flush_gtt_write_domain(obj);
*/
i915_gem_object_set_to_full_cpu_read_domain(obj);
- if (write) {
- ret = i915_gem_object_wait_rendering(obj, true);
- if (ret)
- return ret;
- }
-
old_write_domain = obj->write_domain;
old_read_domains = obj->read_domains;
if (offset == 0 && size == obj->size)
return i915_gem_object_set_to_cpu_domain(obj, 0);
- ret = i915_gem_object_flush_gpu_write_domain(obj, false);
+ ret = i915_gem_object_flush_gpu_write_domain(obj);
if (ret != 0)
return ret;
+ ret = i915_gem_object_wait_rendering(obj, true);
+ if (ret)
+ return ret;
+
i915_gem_object_flush_gtt_write_domain(obj);
/* If we're already fully in the CPU read domain, we're done. */
return 0;
}
-/**
- * Pin an object to the GTT and evaluate the relocations landing in it.
- */
static int
-i915_gem_execbuffer_relocate(struct drm_i915_gem_object *obj,
- struct drm_file *file_priv,
- struct drm_i915_gem_exec_object2 *entry)
+i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
+ struct drm_file *file_priv,
+ struct drm_i915_gem_exec_object2 *entry,
+ struct drm_i915_gem_relocation_entry *reloc)
{
struct drm_device *dev = obj->base.dev;
- drm_i915_private_t *dev_priv = dev->dev_private;
- struct drm_i915_gem_relocation_entry __user *user_relocs;
- struct drm_gem_object *target_obj = NULL;
- uint32_t target_handle = 0;
- int i, ret = 0;
+ struct drm_gem_object *target_obj;
+ uint32_t target_offset;
+ int ret = -EINVAL;
- user_relocs = (void __user *)(uintptr_t)entry->relocs_ptr;
- for (i = 0; i < entry->relocation_count; i++) {
- struct drm_i915_gem_relocation_entry reloc;
- uint32_t target_offset;
+ target_obj = drm_gem_object_lookup(dev, file_priv,
+ reloc->target_handle);
+ if (target_obj == NULL)
+ return -ENOENT;
- if (__copy_from_user_inatomic(&reloc,
- user_relocs+i,
- sizeof(reloc))) {
- ret = -EFAULT;
- break;
- }
+ target_offset = to_intel_bo(target_obj)->gtt_offset;
- if (reloc.target_handle != target_handle) {
- drm_gem_object_unreference(target_obj);
+#if WATCH_RELOC
+ DRM_INFO("%s: obj %p offset %08x target %d "
+ "read %08x write %08x gtt %08x "
+ "presumed %08x delta %08x\n",
+ __func__,
+ obj,
+ (int) reloc->offset,
+ (int) reloc->target_handle,
+ (int) reloc->read_domains,
+ (int) reloc->write_domain,
+ (int) target_offset,
+ (int) reloc->presumed_offset,
+ reloc->delta);
+#endif
- target_obj = drm_gem_object_lookup(dev, file_priv,
- reloc.target_handle);
- if (target_obj == NULL) {
- ret = -ENOENT;
- break;
- }
+ /* The target buffer should have appeared before us in the
+ * exec_object list, so it should have a GTT space bound by now.
+ */
+ if (target_offset == 0) {
+ DRM_ERROR("No GTT space found for object %d\n",
+ reloc->target_handle);
+ goto err;
+ }
- target_handle = reloc.target_handle;
- }
- target_offset = to_intel_bo(target_obj)->gtt_offset;
+ /* Validate that the target is in a valid r/w GPU domain */
+ if (reloc->write_domain & (reloc->write_domain - 1)) {
+ DRM_ERROR("reloc with multiple write domains: "
+ "obj %p target %d offset %d "
+ "read %08x write %08x",
+ obj, reloc->target_handle,
+ (int) reloc->offset,
+ reloc->read_domains,
+ reloc->write_domain);
+ goto err;
+ }
+ if (reloc->write_domain & I915_GEM_DOMAIN_CPU ||
+ reloc->read_domains & I915_GEM_DOMAIN_CPU) {
+ DRM_ERROR("reloc with read/write CPU domains: "
+ "obj %p target %d offset %d "
+ "read %08x write %08x",
+ obj, reloc->target_handle,
+ (int) reloc->offset,
+ reloc->read_domains,
+ reloc->write_domain);
+ goto err;
+ }
+ if (reloc->write_domain && target_obj->pending_write_domain &&
+ reloc->write_domain != target_obj->pending_write_domain) {
+ DRM_ERROR("Write domain conflict: "
+ "obj %p target %d offset %d "
+ "new %08x old %08x\n",
+ obj, reloc->target_handle,
+ (int) reloc->offset,
+ reloc->write_domain,
+ target_obj->pending_write_domain);
+ goto err;
+ }
-#if WATCH_RELOC
- DRM_INFO("%s: obj %p offset %08x target %d "
- "read %08x write %08x gtt %08x "
- "presumed %08x delta %08x\n",
- __func__,
- obj,
- (int) reloc.offset,
- (int) reloc.target_handle,
- (int) reloc.read_domains,
- (int) reloc.write_domain,
- (int) target_offset,
- (int) reloc.presumed_offset,
- reloc.delta);
-#endif
+ target_obj->pending_read_domains |= reloc->read_domains;
+ target_obj->pending_write_domain |= reloc->write_domain;
- /* The target buffer should have appeared before us in the
- * exec_object list, so it should have a GTT space bound by now.
- */
- if (target_offset == 0) {
- DRM_ERROR("No GTT space found for object %d\n",
- reloc.target_handle);
- ret = -EINVAL;
- break;
- }
+ /* If the relocation already has the right value in it, no
+ * more work needs to be done.
+ */
+ if (target_offset == reloc->presumed_offset)
+ goto out;
- /* Validate that the target is in a valid r/w GPU domain */
- if (reloc.write_domain & (reloc.write_domain - 1)) {
- DRM_ERROR("reloc with multiple write domains: "
- "obj %p target %d offset %d "
- "read %08x write %08x",
- obj, reloc.target_handle,
- (int) reloc.offset,
- reloc.read_domains,
- reloc.write_domain);
- ret = -EINVAL;
- break;
- }
- if (reloc.write_domain & I915_GEM_DOMAIN_CPU ||
- reloc.read_domains & I915_GEM_DOMAIN_CPU) {
- DRM_ERROR("reloc with read/write CPU domains: "
- "obj %p target %d offset %d "
- "read %08x write %08x",
- obj, reloc.target_handle,
- (int) reloc.offset,
- reloc.read_domains,
- reloc.write_domain);
- ret = -EINVAL;
- break;
- }
- if (reloc.write_domain && target_obj->pending_write_domain &&
- reloc.write_domain != target_obj->pending_write_domain) {
- DRM_ERROR("Write domain conflict: "
- "obj %p target %d offset %d "
- "new %08x old %08x\n",
- obj, reloc.target_handle,
- (int) reloc.offset,
- reloc.write_domain,
- target_obj->pending_write_domain);
- ret = -EINVAL;
- break;
- }
+ /* Check that the relocation address is valid... */
+ if (reloc->offset > obj->base.size - 4) {
+ DRM_ERROR("Relocation beyond object bounds: "
+ "obj %p target %d offset %d size %d.\n",
+ obj, reloc->target_handle,
+ (int) reloc->offset,
+ (int) obj->base.size);
+ goto err;
+ }
+ if (reloc->offset & 3) {
+ DRM_ERROR("Relocation not 4-byte aligned: "
+ "obj %p target %d offset %d.\n",
+ obj, reloc->target_handle,
+ (int) reloc->offset);
+ goto err;
+ }
- target_obj->pending_read_domains |= reloc.read_domains;
- target_obj->pending_write_domain |= reloc.write_domain;
+ /* and points to somewhere within the target object. */
+ if (reloc->delta >= target_obj->size) {
+ DRM_ERROR("Relocation beyond target object bounds: "
+ "obj %p target %d delta %d size %d.\n",
+ obj, reloc->target_handle,
+ (int) reloc->delta,
+ (int) target_obj->size);
+ goto err;
+ }
- /* If the relocation already has the right value in it, no
- * more work needs to be done.
- */
- if (target_offset == reloc.presumed_offset)
- continue;
+ reloc->delta += target_offset;
+ if (obj->base.write_domain == I915_GEM_DOMAIN_CPU) {
+ uint32_t page_offset = reloc->offset & ~PAGE_MASK;
+ char *vaddr;
- /* Check that the relocation address is valid... */
- if (reloc.offset > obj->base.size - 4) {
- DRM_ERROR("Relocation beyond object bounds: "
- "obj %p target %d offset %d size %d.\n",
- obj, reloc.target_handle,
- (int) reloc.offset, (int) obj->base.size);
- ret = -EINVAL;
- break;
- }
- if (reloc.offset & 3) {
- DRM_ERROR("Relocation not 4-byte aligned: "
- "obj %p target %d offset %d.\n",
- obj, reloc.target_handle,
- (int) reloc.offset);
- ret = -EINVAL;
- break;
- }
+ vaddr = kmap_atomic(obj->pages[reloc->offset >> PAGE_SHIFT]);
+ *(uint32_t *)(vaddr + page_offset) = reloc->delta;
+ kunmap_atomic(vaddr);
+ } else {
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ uint32_t __iomem *reloc_entry;
+ void __iomem *reloc_page;
- /* and points to somewhere within the target object. */
- if (reloc.delta >= target_obj->size) {
- DRM_ERROR("Relocation beyond target object bounds: "
- "obj %p target %d delta %d size %d.\n",
- obj, reloc.target_handle,
- (int) reloc.delta, (int) target_obj->size);
- ret = -EINVAL;
- break;
- }
+ ret = i915_gem_object_set_to_gtt_domain(&obj->base, 1);
+ if (ret)
+ goto err;
- reloc.delta += target_offset;
- if (obj->base.write_domain == I915_GEM_DOMAIN_CPU) {
- uint32_t page_offset = reloc.offset & ~PAGE_MASK;
- char *vaddr;
+ /* Map the page containing the relocation we're going to perform. */
+ reloc->offset += obj->gtt_offset;
+ reloc_page = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping,
+ reloc->offset & PAGE_MASK);
+ reloc_entry = (uint32_t __iomem *)
+ (reloc_page + (reloc->offset & ~PAGE_MASK));
+ iowrite32(reloc->delta, reloc_entry);
+ io_mapping_unmap_atomic(reloc_page);
+ }
- vaddr = kmap_atomic(obj->pages[reloc.offset >> PAGE_SHIFT]);
- *(uint32_t *)(vaddr + page_offset) = reloc.delta;
- kunmap_atomic(vaddr);
- } else {
- uint32_t __iomem *reloc_entry;
- void __iomem *reloc_page;
+ /* and update the user's relocation entry */
+ reloc->presumed_offset = target_offset;
- ret = i915_gem_object_set_to_gtt_domain(&obj->base, 1);
- if (ret)
- break;
+out:
+ ret = 0;
+err:
+ drm_gem_object_unreference(target_obj);
+ return ret;
+}
- /* Map the page containing the relocation we're going to perform. */
- reloc.offset += obj->gtt_offset;
- reloc_page = io_mapping_map_atomic_wc(dev_priv->mm.gtt_mapping,
- reloc.offset & PAGE_MASK);
- reloc_entry = (uint32_t __iomem *)
- (reloc_page + (reloc.offset & ~PAGE_MASK));
- iowrite32(reloc.delta, reloc_entry);
- io_mapping_unmap_atomic(reloc_page);
- }
+static int
+i915_gem_execbuffer_relocate_object(struct drm_i915_gem_object *obj,
+ struct drm_file *file_priv,
+ struct drm_i915_gem_exec_object2 *entry)
+{
+ struct drm_i915_gem_relocation_entry __user *user_relocs;
+ int i, ret;
+
+ user_relocs = (void __user *)(uintptr_t)entry->relocs_ptr;
+ for (i = 0; i < entry->relocation_count; i++) {
+ struct drm_i915_gem_relocation_entry reloc;
+
+ if (__copy_from_user_inatomic(&reloc,
+ user_relocs+i,
+ sizeof(reloc)))
+ return -EFAULT;
+
+ ret = i915_gem_execbuffer_relocate_entry(obj, file_priv, entry, &reloc);
+ if (ret)
+ return ret;
- /* and update the user's relocation entry */
- reloc.presumed_offset = target_offset;
if (__copy_to_user_inatomic(&user_relocs[i].presumed_offset,
- &reloc.presumed_offset,
- sizeof(reloc.presumed_offset))) {
- ret = -EFAULT;
- break;
- }
+ &reloc.presumed_offset,
+ sizeof(reloc.presumed_offset)))
+ return -EFAULT;
}
- drm_gem_object_unreference(target_obj);
- return ret;
+ return 0;
+}
+
+static int
+i915_gem_execbuffer_relocate_object_slow(struct drm_i915_gem_object *obj,
+ struct drm_file *file_priv,
+ struct drm_i915_gem_exec_object2 *entry,
+ struct drm_i915_gem_relocation_entry *relocs)
+{
+ int i, ret;
+
+ for (i = 0; i < entry->relocation_count; i++) {
+ ret = i915_gem_execbuffer_relocate_entry(obj, file_priv, entry, &relocs[i]);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
}
static int
-i915_gem_execbuffer_pin(struct drm_device *dev,
- struct drm_file *file,
- struct drm_gem_object **object_list,
- struct drm_i915_gem_exec_object2 *exec_list,
- int count)
+i915_gem_execbuffer_relocate(struct drm_device *dev,
+ struct drm_file *file,
+ struct drm_gem_object **object_list,
+ struct drm_i915_gem_exec_object2 *exec_list,
+ int count)
+{
+ int i, ret;
+
+ for (i = 0; i < count; i++) {
+ struct drm_i915_gem_object *obj = to_intel_bo(object_list[i]);
+ obj->base.pending_read_domains = 0;
+ obj->base.pending_write_domain = 0;
+ ret = i915_gem_execbuffer_relocate_object(obj, file,
+ &exec_list[i]);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static int
+i915_gem_execbuffer_reserve(struct drm_device *dev,
+ struct drm_file *file,
+ struct drm_gem_object **object_list,
+ struct drm_i915_gem_exec_object2 *exec_list,
+ int count)
{
struct drm_i915_private *dev_priv = dev->dev_private;
int ret, i, retry;
return 0;
}
+static int
+i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
+ struct drm_file *file,
+ struct drm_gem_object **object_list,
+ struct drm_i915_gem_exec_object2 *exec_list,
+ int count)
+{
+ struct drm_i915_gem_relocation_entry *reloc;
+ int i, total, ret;
+
+ for (i = 0; i < count; i++) {
+ struct drm_i915_gem_object *obj = to_intel_bo(object_list[i]);
+ obj->in_execbuffer = false;
+ }
+
+ mutex_unlock(&dev->struct_mutex);
+
+ total = 0;
+ for (i = 0; i < count; i++)
+ total += exec_list[i].relocation_count;
+
+ reloc = drm_malloc_ab(total, sizeof(*reloc));
+ if (reloc == NULL) {
+ mutex_lock(&dev->struct_mutex);
+ return -ENOMEM;
+ }
+
+ total = 0;
+ for (i = 0; i < count; i++) {
+ struct drm_i915_gem_relocation_entry __user *user_relocs;
+
+ user_relocs = (void __user *)(uintptr_t)exec_list[i].relocs_ptr;
+
+ if (copy_from_user(reloc+total, user_relocs,
+ exec_list[i].relocation_count *
+ sizeof(*reloc))) {
+ ret = -EFAULT;
+ mutex_lock(&dev->struct_mutex);
+ goto err;
+ }
+
+ total += exec_list[i].relocation_count;
+ }
+
+ ret = i915_mutex_lock_interruptible(dev);
+ if (ret) {
+ mutex_lock(&dev->struct_mutex);
+ goto err;
+ }
+
+ ret = i915_gem_execbuffer_reserve(dev, file,
+ object_list, exec_list,
+ count);
+ if (ret)
+ goto err;
+
+ total = 0;
+ for (i = 0; i < count; i++) {
+ struct drm_i915_gem_object *obj = to_intel_bo(object_list[i]);
+ obj->base.pending_read_domains = 0;
+ obj->base.pending_write_domain = 0;
+ ret = i915_gem_execbuffer_relocate_object_slow(obj, file,
+ &exec_list[i],
+ reloc + total);
+ if (ret)
+ goto err;
+
+ total += exec_list[i].relocation_count;
+ }
+
+ /* Leave the user relocations as are, this is the painfully slow path,
+ * and we want to avoid the complication of dropping the lock whilst
+ * having buffers reserved in the aperture and so causing spurious
+ * ENOSPC for random operations.
+ */
+
+err:
+ drm_free_large(reloc);
+ return ret;
+}
+
static int
i915_gem_execbuffer_move_to_gpu(struct drm_device *dev,
struct drm_file *file,
for (i = 0; i < count; i++) {
char __user *ptr = (char __user *)(uintptr_t)exec[i].relocs_ptr;
- size_t length = exec[i].relocation_count * sizeof(struct drm_i915_gem_relocation_entry);
+ int length; /* limited by fault_in_pages_readable() */
+
+ /* First check for malicious input causing overflow */
+ if (exec[i].relocation_count >
+ INT_MAX / sizeof(struct drm_i915_gem_relocation_entry))
+ return -EINVAL;
+ length = exec[i].relocation_count *
+ sizeof(struct drm_i915_gem_relocation_entry);
if (!access_ok(VERIFY_READ, ptr, length))
return -EFAULT;
}
/* Move the objects en-masse into the GTT, evicting if necessary. */
- ret = i915_gem_execbuffer_pin(dev, file,
- object_list, exec_list,
- args->buffer_count);
+ ret = i915_gem_execbuffer_reserve(dev, file,
+ object_list, exec_list,
+ args->buffer_count);
if (ret)
goto err;
/* The objects are in their final locations, apply the relocations. */
- for (i = 0; i < args->buffer_count; i++) {
- struct drm_i915_gem_object *obj = to_intel_bo(object_list[i]);
- obj->base.pending_read_domains = 0;
- obj->base.pending_write_domain = 0;
- ret = i915_gem_execbuffer_relocate(obj, file, &exec_list[i]);
+ ret = i915_gem_execbuffer_relocate(dev, file,
+ object_list, exec_list,
+ args->buffer_count);
+ if (ret) {
+ if (ret == -EFAULT) {
+ ret = i915_gem_execbuffer_relocate_slow(dev, file,
+ object_list,
+ exec_list,
+ args->buffer_count);
+ BUG_ON(!mutex_is_locked(&dev->struct_mutex));
+ }
if (ret)
goto err;
}
if (drm_core_check_feature(dev, DRIVER_MODESET))
return;
+ /* Cursor state */
+ dev_priv->saveCURACNTR = I915_READ(CURACNTR);
+ dev_priv->saveCURAPOS = I915_READ(CURAPOS);
+ dev_priv->saveCURABASE = I915_READ(CURABASE);
+ dev_priv->saveCURBCNTR = I915_READ(CURBCNTR);
+ dev_priv->saveCURBPOS = I915_READ(CURBPOS);
+ dev_priv->saveCURBBASE = I915_READ(CURBBASE);
+ if (IS_GEN2(dev))
+ dev_priv->saveCURSIZE = I915_READ(CURSIZE);
+
if (HAS_PCH_SPLIT(dev)) {
dev_priv->savePCH_DREF_CONTROL = I915_READ(PCH_DREF_CONTROL);
dev_priv->saveDISP_ARB_CTL = I915_READ(DISP_ARB_CTL);
I915_WRITE(DSPBCNTR, dev_priv->saveDSPBCNTR);
I915_WRITE(DSPBADDR, I915_READ(DSPBADDR));
+ /* Cursor state */
+ I915_WRITE(CURAPOS, dev_priv->saveCURAPOS);
+ I915_WRITE(CURACNTR, dev_priv->saveCURACNTR);
+ I915_WRITE(CURABASE, dev_priv->saveCURABASE);
+ I915_WRITE(CURBPOS, dev_priv->saveCURBPOS);
+ I915_WRITE(CURBCNTR, dev_priv->saveCURBCNTR);
+ I915_WRITE(CURBBASE, dev_priv->saveCURBBASE);
+ if (IS_GEN2(dev))
+ I915_WRITE(CURSIZE, dev_priv->saveCURSIZE);
+
return;
}
/* Don't save them in KMS mode */
i915_save_modeset_reg(dev);
- /* Cursor state */
- dev_priv->saveCURACNTR = I915_READ(CURACNTR);
- dev_priv->saveCURAPOS = I915_READ(CURAPOS);
- dev_priv->saveCURABASE = I915_READ(CURABASE);
- dev_priv->saveCURBCNTR = I915_READ(CURBCNTR);
- dev_priv->saveCURBPOS = I915_READ(CURBPOS);
- dev_priv->saveCURBBASE = I915_READ(CURBBASE);
- if (IS_GEN2(dev))
- dev_priv->saveCURSIZE = I915_READ(CURSIZE);
-
/* CRT state */
if (HAS_PCH_SPLIT(dev)) {
dev_priv->saveADPA = I915_READ(PCH_ADPA);
/* Don't restore them in KMS mode */
i915_restore_modeset_reg(dev);
- /* Cursor state */
- I915_WRITE(CURAPOS, dev_priv->saveCURAPOS);
- I915_WRITE(CURACNTR, dev_priv->saveCURACNTR);
- I915_WRITE(CURABASE, dev_priv->saveCURABASE);
- I915_WRITE(CURBPOS, dev_priv->saveCURBPOS);
- I915_WRITE(CURBCNTR, dev_priv->saveCURBCNTR);
- I915_WRITE(CURBBASE, dev_priv->saveCURBBASE);
- if (IS_GEN2(dev))
- I915_WRITE(CURSIZE, dev_priv->saveCURSIZE);
-
/* CRT state */
if (HAS_PCH_SPLIT(dev))
I915_WRITE(PCH_ADPA, dev_priv->saveADPA);
struct drm_i915_private *dev_priv = dev->dev_private;
struct intel_encoder *encoder;
bool dpd_is_edp = false;
+ bool has_lvds = false;
if (IS_MOBILE(dev) && !IS_I830(dev))
- intel_lvds_init(dev);
+ has_lvds = intel_lvds_init(dev);
+ if (!has_lvds && !HAS_PCH_SPLIT(dev)) {
+ /* disable the panel fitter on everything but LVDS */
+ I915_WRITE(PFIT_CONTROL, 0);
+ }
if (HAS_PCH_SPLIT(dev)) {
dpd_is_edp = intel_dpd_is_edp(dev);
mode->clock = dev_priv->panel_fixed_mode->clock;
}
- /* Just use VBT values for eDP */
- if (is_edp(intel_dp)) {
- intel_dp->lane_count = dev_priv->edp.lanes;
- intel_dp->link_bw = dev_priv->edp.rate;
- adjusted_mode->clock = intel_dp_link_clock(intel_dp->link_bw);
- DRM_DEBUG_KMS("eDP link bw %02x lane count %d clock %d\n",
- intel_dp->link_bw, intel_dp->lane_count,
- adjusted_mode->clock);
- return true;
- }
-
for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
for (clock = 0; clock <= max_clock; clock++) {
int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count);
}
}
+ if (is_edp(intel_dp)) {
+ /* okay we failed just pick the highest */
+ intel_dp->lane_count = max_lane_count;
+ intel_dp->link_bw = bws[max_clock];
+ adjusted_mode->clock = intel_dp_link_clock(intel_dp->link_bw);
+ DRM_DEBUG_KMS("Force picking display port link bw %02x lane "
+ "count %d clock %d\n",
+ intel_dp->link_bw, intel_dp->lane_count,
+ adjusted_mode->clock);
+
+ return true;
+ }
+
return false;
}
}
static uint32_t
-intel_dp_signal_levels(struct intel_dp *intel_dp)
+intel_dp_signal_levels(uint8_t train_set, int lane_count)
{
- struct drm_device *dev = intel_dp->base.base.dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- uint32_t signal_levels = 0;
- u8 train_set = intel_dp->train_set[0];
- u32 vswing = train_set & DP_TRAIN_VOLTAGE_SWING_MASK;
- u32 preemphasis = train_set & DP_TRAIN_PRE_EMPHASIS_MASK;
+ uint32_t signal_levels = 0;
- if (is_edp(intel_dp)) {
- vswing = dev_priv->edp.vswing;
- preemphasis = dev_priv->edp.preemphasis;
- }
-
- switch (vswing) {
+ switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
case DP_TRAIN_VOLTAGE_SWING_400:
default:
signal_levels |= DP_VOLTAGE_0_4;
signal_levels |= DP_VOLTAGE_1_2;
break;
}
- switch (preemphasis) {
+ switch (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) {
case DP_TRAIN_PRE_EMPHASIS_0:
default:
signal_levels |= DP_PRE_EMPHASIS_0;
return true;
}
-static bool
-intel_dp_aux_handshake_required(struct intel_dp *intel_dp)
-{
- struct drm_device *dev = intel_dp->base.base.dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
-
- if (is_edp(intel_dp) && dev_priv->no_aux_handshake)
- return false;
-
- return true;
-}
-
static bool
intel_dp_set_link_train(struct intel_dp *intel_dp,
uint32_t dp_reg_value,
I915_WRITE(intel_dp->output_reg, dp_reg_value);
POSTING_READ(intel_dp->output_reg);
- if (!intel_dp_aux_handshake_required(intel_dp))
- return true;
-
intel_dp_aux_native_write_1(intel_dp,
DP_TRAINING_PATTERN_SET,
dp_train_pat);
POSTING_READ(intel_dp->output_reg);
intel_wait_for_vblank(dev, intel_crtc->pipe);
- if (intel_dp_aux_handshake_required(intel_dp))
- /* Write the link configuration data */
- intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET,
- intel_dp->link_configuration,
- DP_LINK_CONFIGURATION_SIZE);
+ /* Write the link configuration data */
+ intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET,
+ intel_dp->link_configuration,
+ DP_LINK_CONFIGURATION_SIZE);
DP |= DP_PORT_EN;
if (HAS_PCH_CPT(dev) && !is_edp(intel_dp))
signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
} else {
- signal_levels = intel_dp_signal_levels(intel_dp);
+ signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
}
break;
/* Set training pattern 1 */
- udelay(500);
- if (intel_dp_aux_handshake_required(intel_dp)) {
+ udelay(100);
+ if (!intel_dp_get_link_status(intel_dp))
break;
- } else {
- if (!intel_dp_get_link_status(intel_dp))
- break;
- if (intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
- clock_recovery = true;
- break;
- }
+ if (intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
+ clock_recovery = true;
+ break;
+ }
- /* Check to see if we've tried the max voltage */
- for (i = 0; i < intel_dp->lane_count; i++)
- if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
- break;
- if (i == intel_dp->lane_count)
+ /* Check to see if we've tried the max voltage */
+ for (i = 0; i < intel_dp->lane_count; i++)
+ if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
break;
+ if (i == intel_dp->lane_count)
+ break;
- /* Check to see if we've tried the same voltage 5 times */
- if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
- ++tries;
- if (tries == 5)
- break;
- } else
- tries = 0;
- voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
+ /* Check to see if we've tried the same voltage 5 times */
+ if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
+ ++tries;
+ if (tries == 5)
+ break;
+ } else
+ tries = 0;
+ voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
- /* Compute new intel_dp->train_set as requested by target */
- intel_get_adjust_train(intel_dp);
- }
+ /* Compute new intel_dp->train_set as requested by target */
+ intel_get_adjust_train(intel_dp);
}
intel_dp->DP = DP;
signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
} else {
- signal_levels = intel_dp_signal_levels(intel_dp);
+ signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
}
DP_TRAINING_PATTERN_2))
break;
- udelay(500);
-
- if (!intel_dp_aux_handshake_required(intel_dp)) {
+ udelay(400);
+ if (!intel_dp_get_link_status(intel_dp))
break;
- } else {
- if (!intel_dp_get_link_status(intel_dp))
- break;
- if (intel_channel_eq_ok(intel_dp)) {
- channel_eq = true;
- break;
- }
+ if (intel_channel_eq_ok(intel_dp)) {
+ channel_eq = true;
+ break;
+ }
- /* Try 5 times */
- if (tries > 5)
- break;
+ /* Try 5 times */
+ if (tries > 5)
+ break;
- /* Compute new intel_dp->train_set as requested by target */
- intel_get_adjust_train(intel_dp);
- ++tries;
- }
+ /* Compute new intel_dp->train_set as requested by target */
+ intel_get_adjust_train(intel_dp);
+ ++tries;
}
+
if (HAS_PCH_CPT(dev) && !is_edp(intel_dp))
reg = DP | DP_LINK_TRAIN_OFF_CPT;
else
extern void intel_dvo_init(struct drm_device *dev);
extern void intel_tv_init(struct drm_device *dev);
extern void intel_mark_busy(struct drm_device *dev, struct drm_gem_object *obj);
-extern void intel_lvds_init(struct drm_device *dev);
+extern bool intel_lvds_init(struct drm_device *dev);
extern void intel_dp_init(struct drm_device *dev, int dp_reg);
void
intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
* Create the connector, register the LVDS DDC bus, and try to figure out what
* modes we can display on the LVDS panel (if present).
*/
-void intel_lvds_init(struct drm_device *dev)
+bool intel_lvds_init(struct drm_device *dev)
{
struct drm_i915_private *dev_priv = dev->dev_private;
struct intel_lvds *intel_lvds;
/* Skip init on machines we know falsely report LVDS */
if (dmi_check_system(intel_no_lvds))
- return;
+ return false;
pin = GMBUS_PORT_PANEL;
if (!lvds_is_present_in_vbt(dev, &pin)) {
DRM_DEBUG_KMS("LVDS is not present in VBT\n");
- return;
+ return false;
}
if (HAS_PCH_SPLIT(dev)) {
if ((I915_READ(PCH_LVDS) & LVDS_DETECTED) == 0)
- return;
+ return false;
if (dev_priv->edp.support) {
DRM_DEBUG_KMS("disable LVDS for eDP support\n");
- return;
+ return false;
}
}
if (!intel_lvds_ddc_probe(dev, pin)) {
DRM_DEBUG_KMS("LVDS did not respond to DDC probe\n");
- return;
+ return false;
}
intel_lvds = kzalloc(sizeof(struct intel_lvds), GFP_KERNEL);
if (!intel_lvds) {
- return;
+ return false;
}
intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
if (!intel_connector) {
kfree(intel_lvds);
- return;
+ return false;
}
if (!HAS_PCH_SPLIT(dev)) {
/* keep the LVDS connector */
dev_priv->int_lvds_connector = connector;
drm_sysfs_connector_add(connector);
- return;
+ return true;
failed:
DRM_DEBUG_KMS("No LVDS modes found, disabling.\n");
drm_encoder_cleanup(encoder);
kfree(intel_lvds);
kfree(intel_connector);
+ return false;
}
* This is set if we treat the device as HDMI, instead of DVI.
*/
bool is_hdmi;
- bool has_audio;
+ bool has_hdmi_monitor;
+ bool has_hdmi_audio;
/**
* This is set if we detect output of sdvo device as LVDS and
if (!intel_sdvo_set_target_input(intel_sdvo))
return;
- if (intel_sdvo->is_hdmi &&
+ if (intel_sdvo->has_hdmi_monitor &&
!intel_sdvo_set_avi_infoframe(intel_sdvo))
return;
}
if (intel_crtc->pipe == 1)
sdvox |= SDVO_PIPE_B_SELECT;
- if (intel_sdvo->has_audio)
+ if (intel_sdvo->has_hdmi_audio)
sdvox |= SDVO_AUDIO_ENABLE;
if (INTEL_INFO(dev)->gen >= 4) {
return drm_get_edid(connector, &sdvo->ddc);
}
-static struct drm_connector *
-intel_find_analog_connector(struct drm_device *dev)
-{
- struct drm_connector *connector;
- struct intel_sdvo *encoder;
-
- list_for_each_entry(encoder,
- &dev->mode_config.encoder_list,
- base.base.head) {
- if (encoder->base.type == INTEL_OUTPUT_ANALOG) {
- list_for_each_entry(connector,
- &dev->mode_config.connector_list,
- head) {
- if (&encoder->base ==
- intel_attached_encoder(connector))
- return connector;
- }
- }
- }
-
- return NULL;
-}
-
-static int
-intel_analog_is_connected(struct drm_device *dev)
-{
- struct drm_connector *analog_connector;
-
- analog_connector = intel_find_analog_connector(dev);
- if (!analog_connector)
- return false;
-
- if (analog_connector->funcs->detect(analog_connector, false) ==
- connector_status_disconnected)
- return false;
-
- return true;
-}
-
/* Mac mini hack -- use the same DDC as the analog connector */
static struct edid *
intel_sdvo_get_analog_edid(struct drm_connector *connector)
{
struct drm_i915_private *dev_priv = connector->dev->dev_private;
- if (!intel_analog_is_connected(connector->dev))
- return NULL;
-
- return drm_get_edid(connector, &dev_priv->gmbus[dev_priv->crt_ddc_pin].adapter);
+ return drm_get_edid(connector,
+ &dev_priv->gmbus[dev_priv->crt_ddc_pin].adapter);
}
enum drm_connector_status
/* DDC bus is shared, match EDID to connector type */
if (edid->input & DRM_EDID_INPUT_DIGITAL) {
status = connector_status_connected;
- intel_sdvo->is_hdmi = drm_detect_hdmi_monitor(edid);
- intel_sdvo->has_audio = drm_detect_monitor_audio(edid);
+ if (intel_sdvo->is_hdmi) {
+ intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid);
+ intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
+ }
}
connector->display_info.raw_edid = NULL;
kfree(edid);
if (status == connector_status_connected) {
struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector);
if (intel_sdvo_connector->force_audio)
- intel_sdvo->has_audio = intel_sdvo_connector->force_audio > 0;
+ intel_sdvo->has_hdmi_audio = intel_sdvo_connector->force_audio > 0;
}
return status;
if (!intel_sdvo_write_cmd(intel_sdvo,
SDVO_CMD_GET_ATTACHED_DISPLAYS, NULL, 0))
return connector_status_unknown;
- if (intel_sdvo->is_tv) {
- /* add 30ms delay when the output type is SDVO-TV */
+
+ /* add 30ms delay when the output type might be TV */
+ if (intel_sdvo->caps.output_flags &
+ (SDVO_OUTPUT_SVID0 | SDVO_OUTPUT_CVBS0))
mdelay(30);
- }
+
if (!intel_sdvo_read_response(intel_sdvo, &response, 2))
return connector_status_unknown;
edid = intel_sdvo_get_analog_edid(connector);
if (edid != NULL) {
- drm_mode_connector_update_edid_property(connector, edid);
- drm_add_edid_modes(connector, edid);
+ if (edid->input & DRM_EDID_INPUT_DIGITAL) {
+ drm_mode_connector_update_edid_property(connector, edid);
+ drm_add_edid_modes(connector, edid);
+ }
connector->display_info.raw_edid = NULL;
kfree(edid);
}
intel_sdvo_connector->force_audio = val;
- if (val > 0 && intel_sdvo->has_audio)
+ if (val > 0 && intel_sdvo->has_hdmi_audio)
return 0;
- if (val < 0 && !intel_sdvo->has_audio)
+ if (val < 0 && !intel_sdvo->has_hdmi_audio)
return 0;
- intel_sdvo->has_audio = val > 0;
+ intel_sdvo->has_hdmi_audio = val > 0;
goto done;
}
intel_sdvo_set_colorimetry(intel_sdvo,
SDVO_COLORIMETRY_RGB256);
connector->connector_type = DRM_MODE_CONNECTOR_HDMIA;
+
+ intel_sdvo_add_hdmi_properties(intel_sdvo_connector);
intel_sdvo->is_hdmi = true;
}
intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) |
intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo);
- intel_sdvo_add_hdmi_properties(intel_sdvo_connector);
-
return true;
}
base += 3;
break;
case ATOM_IIO_WRITE:
+ (void)ctx->card->ioreg_read(ctx->card, CU16(base + 1));
ctx->card->ioreg_write(ctx->card, CU16(base + 1), temp);
base += 3;
break;
if (array_mode == V_0280A0_ARRAY_LINEAR_GENERAL) {
/* the initial DDX does bad things with the CB size occasionally */
/* it rounds up height too far for slice tile max but the BO is smaller */
- tmp = (height - 7) * pitch * bpe;
+ tmp = (height - 7) * 8 * bpe;
if ((tmp + track->cb_color_bo_offset[i]) > radeon_bo_size(track->cb_color_bo[i])) {
dev_warn(p->dev, "%s offset[%d] %d %d %lu too big\n", __func__, i, track->cb_color_bo_offset[i], tmp, radeon_bo_size(track->cb_color_bo[i]));
return -EINVAL;
#define R600_HDP_NONSURFACE_BASE 0x2c04
#define R600_BUS_CNTL 0x5420
+# define R600_BIOS_ROM_DIS (1 << 1)
#define R600_CONFIG_CNTL 0x5424
#define R600_CONFIG_MEMSIZE 0x5428
#define R600_CONFIG_F0_BASE 0x542C
}
}
+ /* some DCE3 boards have bad data for this entry */
+ if (ASIC_IS_DCE3(rdev)) {
+ if ((i == 4) &&
+ (gpio->usClkMaskRegisterIndex == 0x1fda) &&
+ (gpio->sucI2cId.ucAccess == 0x94))
+ gpio->sucI2cId.ucAccess = 0x14;
+ }
+
if (gpio->sucI2cId.ucAccess == id) {
i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
}
}
+ /* some DCE3 boards have bad data for this entry */
+ if (ASIC_IS_DCE3(rdev)) {
+ if ((i == 4) &&
+ (gpio->usClkMaskRegisterIndex == 0x1fda) &&
+ (gpio->sucI2cId.ucAccess == 0x94))
+ gpio->sucI2cId.ucAccess = 0x14;
+ }
+
i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
}
return true;
}
+
static bool r700_read_disabled_bios(struct radeon_device *rdev)
{
uint32_t viph_control;
bool r;
viph_control = RREG32(RADEON_VIPH_CONTROL);
- bus_cntl = RREG32(RADEON_BUS_CNTL);
+ bus_cntl = RREG32(R600_BUS_CNTL);
d1vga_control = RREG32(AVIVO_D1VGA_CONTROL);
d2vga_control = RREG32(AVIVO_D2VGA_CONTROL);
vga_render_control = RREG32(AVIVO_VGA_RENDER_CONTROL);
/* disable VIP */
WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN));
/* enable the rom */
- WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM));
+ WREG32(R600_BUS_CNTL, (bus_cntl & ~R600_BIOS_ROM_DIS));
/* Disable VGA mode */
WREG32(AVIVO_D1VGA_CONTROL,
(d1vga_control & ~(AVIVO_DVGA_CONTROL_MODE_ENABLE |
cg_spll_status = RREG32(R600_CG_SPLL_STATUS);
}
WREG32(RADEON_VIPH_CONTROL, viph_control);
- WREG32(RADEON_BUS_CNTL, bus_cntl);
+ WREG32(R600_BUS_CNTL, bus_cntl);
WREG32(AVIVO_D1VGA_CONTROL, d1vga_control);
WREG32(AVIVO_D2VGA_CONTROL, d2vga_control);
WREG32(AVIVO_VGA_RENDER_CONTROL, vga_render_control);
bool r;
viph_control = RREG32(RADEON_VIPH_CONTROL);
- bus_cntl = RREG32(RADEON_BUS_CNTL);
+ bus_cntl = RREG32(R600_BUS_CNTL);
d1vga_control = RREG32(AVIVO_D1VGA_CONTROL);
d2vga_control = RREG32(AVIVO_D2VGA_CONTROL);
vga_render_control = RREG32(AVIVO_VGA_RENDER_CONTROL);
/* disable VIP */
WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN));
/* enable the rom */
- WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM));
+ WREG32(R600_BUS_CNTL, (bus_cntl & ~R600_BIOS_ROM_DIS));
/* Disable VGA mode */
WREG32(AVIVO_D1VGA_CONTROL,
(d1vga_control & ~(AVIVO_DVGA_CONTROL_MODE_ENABLE |
/* restore regs */
WREG32(RADEON_VIPH_CONTROL, viph_control);
- WREG32(RADEON_BUS_CNTL, bus_cntl);
+ WREG32(R600_BUS_CNTL, bus_cntl);
WREG32(AVIVO_D1VGA_CONTROL, d1vga_control);
WREG32(AVIVO_D2VGA_CONTROL, d2vga_control);
WREG32(AVIVO_VGA_RENDER_CONTROL, vga_render_control);
clk = RBIOS8(offset + 3 + (i * 5) + 3);
data = RBIOS8(offset + 3 + (i * 5) + 4);
i2c = combios_setup_i2c_bus(rdev, DDC_MONID,
- clk, data);
+ (1 << clk), (1 << data));
rdev->i2c_bus[4] = radeon_i2c_create(dev, &i2c, "GPIOPAD_MASK");
break;
}
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+ connector->interlace_allowed = true;
+ connector->doublescan_allowed = true;
break;
case DRM_MODE_CONNECTOR_DVIA:
drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type);
1);
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
+ connector->interlace_allowed = true;
+ connector->doublescan_allowed = true;
break;
case DRM_MODE_CONNECTOR_DVII:
case DRM_MODE_CONNECTOR_DVID:
rdev->mode_info.load_detect_property,
1);
}
+ connector->interlace_allowed = true;
+ if (connector_type == DRM_MODE_CONNECTOR_DVII)
+ connector->doublescan_allowed = true;
+ else
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_HDMIA:
case DRM_MODE_CONNECTOR_HDMIB:
0);
}
subpixel_order = SubPixelHorizontalRGB;
+ connector->interlace_allowed = true;
+ if (connector_type == DRM_MODE_CONNECTOR_HDMIB)
+ connector->doublescan_allowed = true;
+ else
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_DisplayPort:
case DRM_MODE_CONNECTOR_eDP:
rdev->mode_info.underscan_vborder_property,
0);
}
+ connector->interlace_allowed = true;
+ /* in theory with a DP to VGA converter... */
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_SVIDEO:
case DRM_MODE_CONNECTOR_Composite:
radeon_atombios_get_tv_info(rdev));
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
+ connector->interlace_allowed = false;
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_LVDS:
radeon_dig_connector = kzalloc(sizeof(struct radeon_connector_atom_dig), GFP_KERNEL);
dev->mode_config.scaling_mode_property,
DRM_MODE_SCALE_FULLSCREEN);
subpixel_order = SubPixelHorizontalRGB;
+ connector->interlace_allowed = false;
+ connector->doublescan_allowed = false;
break;
}
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+ connector->interlace_allowed = true;
+ connector->doublescan_allowed = true;
break;
case DRM_MODE_CONNECTOR_DVIA:
drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type);
1);
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
+ connector->interlace_allowed = true;
+ connector->doublescan_allowed = true;
break;
case DRM_MODE_CONNECTOR_DVII:
case DRM_MODE_CONNECTOR_DVID:
1);
}
subpixel_order = SubPixelHorizontalRGB;
+ connector->interlace_allowed = true;
+ if (connector_type == DRM_MODE_CONNECTOR_DVII)
+ connector->doublescan_allowed = true;
+ else
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_SVIDEO:
case DRM_MODE_CONNECTOR_Composite:
radeon_combios_get_tv_info(rdev));
/* no HPD on analog connectors */
radeon_connector->hpd.hpd = RADEON_HPD_NONE;
+ connector->interlace_allowed = false;
+ connector->doublescan_allowed = false;
break;
case DRM_MODE_CONNECTOR_LVDS:
drm_connector_init(dev, &radeon_connector->base, &radeon_lvds_connector_funcs, connector_type);
dev->mode_config.scaling_mode_property,
DRM_MODE_SCALE_FULLSCREEN);
subpixel_order = SubPixelHorizontalRGB;
+ connector->interlace_allowed = false;
+ connector->doublescan_allowed = false;
break;
}
0
};
+#ifdef MODULE
static struct pci_device_id i5k_amb_ids[] __devinitdata = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_5000_ERR) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_5400_ERR) },
{ 0, }
};
MODULE_DEVICE_TABLE(pci, i5k_amb_ids);
+#endif
static int __devinit i5k_amb_probe(struct platform_device *pdev)
{
return 0;
}
-#ifdef CONFIG_PM
+#ifdef CONFIG_PM_SLEEP
static int lis3lv02d_i2c_suspend(struct device *dev)
{
struct i2c_client *client = container_of(dev, struct i2c_client, dev);
return 0;
}
-#else
-#define lis3lv02d_i2c_suspend NULL
-#define lis3lv02d_i2c_resume NULL
-#define lis3lv02d_i2c_shutdown NULL
-#endif
+#endif /* CONFIG_PM_SLEEP */
+#ifdef CONFIG_PM_RUNTIME
static int lis3_i2c_runtime_suspend(struct device *dev)
{
struct i2c_client *client = container_of(dev, struct i2c_client, dev);
lis3lv02d_poweron(lis3);
return 0;
}
+#endif /* CONFIG_PM_RUNTIME */
static const struct i2c_device_id lis3lv02d_id[] = {
{"lis3lv02d", 0 },
static int __init icn_init(void)
{
char *p;
- char rev[10];
+ char rev[20];
memset(&dev, 0, sizeof(icn_dev));
dev.memaddr = (membase & 0x0ffc000);
spin_lock_init(&dev.devlock);
if ((p = strchr(revision, ':'))) {
- strcpy(rev, p + 1);
+ strncpy(rev, p + 1, 20);
p = strchr(rev, '$');
- *p = 0;
+ if (p)
+ *p = 0;
} else
strcpy(rev, " ??? ");
printk(KERN_NOTICE "ICN-ISDN-driver Rev%smem=0x%08lx\n", rev,
u8 num_leds;
};
-#define cdev_to_led(c) container_of(c, struct lp5521_led, cdev)
-#define engine_to_lp5521(eng) container_of((eng), struct lp5521_chip, \
- engines[(eng)->id - 1])
-#define led_to_lp5521(led) container_of((led), struct lp5521_chip, \
- leds[(led)->id])
+static inline struct lp5521_led *cdev_to_led(struct led_classdev *cdev)
+{
+ return container_of(cdev, struct lp5521_led, cdev);
+}
+
+static inline struct lp5521_chip *engine_to_lp5521(struct lp5521_engine *engine)
+{
+ return container_of(engine, struct lp5521_chip,
+ engines[engine->id - 1]);
+}
+
+static inline struct lp5521_chip *led_to_lp5521(struct lp5521_led *led)
+{
+ return container_of(led, struct lp5521_chip,
+ leds[led->id]);
+}
static void lp5521_led_brightness_work(struct work_struct *work);
/* move current engine to direct mode and remember the state */
ret = lp5521_set_engine_mode(eng, LP5521_CMD_DIRECT);
- usleep_range(1000, 10000);
+ /* Mode change requires min 500 us delay. 1 - 2 ms with margin */
+ usleep_range(1000, 2000);
ret |= lp5521_read(client, LP5521_REG_OP_MODE, &mode);
/* For loading, all the engines to load mode */
lp5521_write(client, LP5521_REG_OP_MODE, LP5521_CMD_DIRECT);
- usleep_range(1000, 10000);
+ /* Mode change requires min 500 us delay. 1 - 2 ms with margin */
+ usleep_range(1000, 2000);
lp5521_write(client, LP5521_REG_OP_MODE, LP5521_CMD_LOAD);
- usleep_range(1000, 10000);
+ /* Mode change requires min 500 us delay. 1 - 2 ms with margin */
+ usleep_range(1000, 2000);
addr = LP5521_PROG_MEM_BASE + eng->prog_page * LP5521_PROG_MEM_SIZE;
i2c_smbus_write_i2c_block_data(client,
lp5521_init_engine(chip, attr_group);
- lp5521_write(client, LP5521_REG_RESET, 0xff);
-
- usleep_range(10000, 20000);
-
/* Set all PWMs to direct control mode */
ret = lp5521_write(client, LP5521_REG_OP_MODE, 0x3F);
ret |= lp5521_write(client, LP5521_REG_ENABLE,
LP5521_MASTER_ENABLE | LP5521_LOGARITHMIC_PWM |
LP5521_EXEC_RUN);
- /* enable takes 500us */
- usleep_range(500, 20000);
+ /* enable takes 500us. 1 - 2 ms leaves some margin */
+ usleep_range(1000, 2000);
return ret;
}
LP5521_MASTER_ENABLE | LP5521_LOGARITHMIC_PWM);
if (ret)
return ret;
- usleep_range(1000, 10000);
+ /* enable takes 500us. 1 - 2 ms leaves some margin */
+ usleep_range(1000, 2000);
ret = lp5521_read(client, LP5521_REG_ENABLE, &buf);
if (ret)
return ret;
if (pdata->enable) {
pdata->enable(0);
- usleep_range(1000, 10000);
+ usleep_range(1000, 2000); /* Keep enable down at least 1ms */
pdata->enable(1);
- usleep_range(1000, 10000); /* Spec says min 500us */
+ usleep_range(1000, 2000); /* 500us abs min. */
}
+ lp5521_write(client, LP5521_REG_RESET, 0xff);
+ usleep_range(10000, 20000); /*
+ * Exact value is not available. 10 - 20ms
+ * appears to be enough for reset.
+ */
ret = lp5521_detect(client);
if (ret) {
u8 num_leds;
};
-#define cdev_to_led(c) container_of(c, struct lp5523_led, cdev)
+static inline struct lp5523_led *cdev_to_led(struct led_classdev *cdev)
+{
+ return container_of(cdev, struct lp5523_led, cdev);
+}
-static struct lp5523_chip *engine_to_lp5523(struct lp5523_engine *engine)
+static inline struct lp5523_chip *engine_to_lp5523(struct lp5523_engine *engine)
{
return container_of(engine, struct lp5523_chip,
engines[engine->id - 1]);
}
-static struct lp5523_chip *led_to_lp5523(struct lp5523_led *led)
+static inline struct lp5523_chip *led_to_lp5523(struct lp5523_led *led)
{
return container_of(led, struct lp5523_chip,
leds[led->id]);
{ 0x9c, 0x50, 0x9c, 0xd0, 0x9d, 0x80, 0xd8, 0x00, 0},
};
- lp5523_write(client, LP5523_REG_RESET, 0xff);
-
- usleep_range(10000, 100000);
-
ret |= lp5523_write(client, LP5523_REG_ENABLE, LP5523_ENABLE);
- /* Chip startup time after reset is 500 us */
- usleep_range(1000, 10000);
+ /* Chip startup time is 500 us, 1 - 2 ms gives some margin */
+ usleep_range(1000, 2000);
ret |= lp5523_write(client, LP5523_REG_CONFIG,
LP5523_AUTO_INC | LP5523_PWR_SAVE |
return -1;
}
- /* Wait 3ms and check the engine status */
- usleep_range(3000, 20000);
+ /* Let the programs run for couple of ms and check the engine status */
+ usleep_range(3000, 6000);
lp5523_read(client, LP5523_REG_STATUS, &status);
status &= LP5523_ENG_STATUS_MASK;
/* Measure VDD (i.e. VBAT) first (channel 16 corresponds to VDD) */
lp5523_write(chip->client, LP5523_REG_LED_TEST_CTRL,
LP5523_EN_LEDTEST | 16);
- usleep_range(3000, 10000);
+ usleep_range(3000, 6000); /* ADC conversion time is typically 2.7 ms */
ret = lp5523_read(chip->client, LP5523_REG_STATUS, &status);
if (!(status & LP5523_LEDTEST_DONE))
- usleep_range(3000, 10000);
+ usleep_range(3000, 6000); /* Was not ready. Wait little bit */
ret |= lp5523_read(chip->client, LP5523_REG_LED_TEST_ADC, &vdd);
vdd--; /* There may be some fluctuation in measurement */
chip->pdata->led_config[i].led_current);
lp5523_write(chip->client, LP5523_REG_LED_PWM_BASE + i, 0xff);
- /* let current stabilize 2ms before measurements start */
- usleep_range(2000, 10000);
+ /* let current stabilize 2 - 4ms before measurements start */
+ usleep_range(2000, 4000);
lp5523_write(chip->client,
LP5523_REG_LED_TEST_CTRL,
LP5523_EN_LEDTEST | i);
- /* ledtest takes 2.7ms */
- usleep_range(3000, 10000);
+ /* ADC conversion time is 2.7 ms typically */
+ usleep_range(3000, 6000);
ret = lp5523_read(chip->client, LP5523_REG_STATUS, &status);
if (!(status & LP5523_LEDTEST_DONE))
- usleep_range(3000, 10000);
+ usleep_range(3000, 6000);/* Was not ready. Wait. */
ret |= lp5523_read(chip->client, LP5523_REG_LED_TEST_ADC, &adc);
if (adc >= vdd || adc < LP5523_ADC_SHORTCIRC_LIM)
if (pdata->enable) {
pdata->enable(0);
- usleep_range(1000, 10000);
+ usleep_range(1000, 2000); /* Keep enable down at least 1ms */
pdata->enable(1);
- usleep_range(1000, 10000); /* Spec says min 500us */
+ usleep_range(1000, 2000); /* 500us abs min. */
}
+ lp5523_write(client, LP5523_REG_RESET, 0xff);
+ usleep_range(10000, 20000); /*
+ * Exact value is not available. 10 - 20ms
+ * appears to be enough for reset.
+ */
ret = lp5523_detect(client);
if (ret)
goto fail2;
DMI_MATCH(DMI_PRODUCT_VERSION, "1.00.00")
}
},
+ {}
};
/*
md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
rdev->sb_page);
md_super_wait(rdev->mddev);
- return num_sectors / 2; /* kB for sysfs */
+ return num_sectors;
}
md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
rdev->sb_page);
md_super_wait(rdev->mddev);
- return num_sectors / 2; /* kB for sysfs */
+ return num_sectors;
}
static struct super_type super_types[] = {
if (mddev->kobj.sd &&
sysfs_create_group(&mddev->kobj, &md_bitmap_group))
printk(KERN_DEBUG "pointless warning\n");
+
+ blk_queue_flush(mddev->queue, REQ_FLUSH | REQ_FUA);
abort:
mutex_unlock(&disks_mutex);
if (!error && mddev->kobj.sd) {
* is not possible.
*/
if (!test_bit(Faulty, &rdev->flags) &&
+ !mddev->recovery_disabled &&
mddev->degraded < conf->raid_disks) {
err = -EBUSY;
goto abort;
static int isl29020_remove(struct i2c_client *client)
{
- struct als_data *data = i2c_get_clientdata(client);
sysfs_remove_group(&client->dev.kobj, &m_als_gr);
- kfree(data);
return 0;
}
module_init(sensor_isl29020_init);
module_exit(sensor_isl29020_exit);
-MODULE_AUTHOR("Kalhan Trisal <kalhan.trisal@intel.com");
+MODULE_AUTHOR("Kalhan Trisal <kalhan.trisal@intel.com>");
MODULE_DESCRIPTION("Intersil isl29020 ALS Driver");
MODULE_LICENSE("GPL v2");
* nodes that can comprise an access protection grouping. The access
* protection is in regards to memory, IOI and IPI.
*/
- max_regions = 64;
region_size = xp_region_size;
- switch (region_size) {
- case 128:
- max_regions *= 2;
- case 64:
- max_regions *= 2;
- case 32:
- max_regions *= 2;
- region_size = 16;
- DBUG_ON(!is_shub2());
+ if (is_uv())
+ max_regions = 256;
+ else {
+ max_regions = 64;
+
+ switch (region_size) {
+ case 128:
+ max_regions *= 2;
+ case 64:
+ max_regions *= 2;
+ case 32:
+ max_regions *= 2;
+ region_size = 16;
+ DBUG_ON(!is_shub2());
+ }
}
for (region = 0; region < max_regions; region++) {
if (host->caps & MMC_CAP_DISABLE)
cancel_delayed_work(&host->disable);
- cancel_delayed_work(&host->detect);
+ cancel_delayed_work_sync(&host->detect);
mmc_flush_scheduled_work();
/* clear pm flags now and let card drivers set them as needed */
struct mmc_card *oldcard)
{
struct mmc_card *card;
- int err, ddr = MMC_SDR_MODE;
+ int err, ddr = 0;
u32 cid[4];
unsigned int max_dtr;
1 << bus_width, ddr);
err = 0;
} else {
- mmc_card_set_ddr_mode(card);
+ if (ddr)
+ mmc_card_set_ddr_mode(card);
+ else
+ ddr = MMC_SDR_MODE;
+
mmc_set_bus_width_ddr(card->host, bus_width, ddr);
}
}
BUG_ON(!host->card);
/* Make sure card is powered before detecting it */
- err = pm_runtime_get_sync(&host->card->dev);
- if (err < 0)
- goto out;
+ if (host->caps & MMC_CAP_POWER_OFF_CARD) {
+ err = pm_runtime_get_sync(&host->card->dev);
+ if (err < 0)
+ goto out;
+ }
mmc_claim_host(host);
mmc_release_host(host);
+ /*
+ * Tell PM core it's OK to power off the card now.
+ *
+ * The _sync variant is used in order to ensure that the card
+ * is left powered off in case an error occurred, and the card
+ * is going to be removed.
+ *
+ * Since there is no specific reason to believe a new user
+ * is about to show up at this point, the _sync variant is
+ * desirable anyway.
+ */
+ if (host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_put_sync(&host->card->dev);
+
out:
if (err) {
mmc_sdio_remove(host);
mmc_detach_bus(host);
mmc_release_host(host);
}
-
- /* Tell PM core that we're done */
- pm_runtime_put(&host->card->dev);
}
/*
card = host->card;
/*
- * Let runtime PM core know our card is active
+ * Enable runtime PM only if supported by host+card+board
*/
- err = pm_runtime_set_active(&card->dev);
- if (err)
- goto remove;
+ if (host->caps & MMC_CAP_POWER_OFF_CARD) {
+ /*
+ * Let runtime PM core know our card is active
+ */
+ err = pm_runtime_set_active(&card->dev);
+ if (err)
+ goto remove;
- /*
- * Enable runtime PM for this card
- */
- pm_runtime_enable(&card->dev);
+ /*
+ * Enable runtime PM for this card
+ */
+ pm_runtime_enable(&card->dev);
+ }
/*
* The number of functions on the card is encoded inside
goto remove;
/*
- * Enable Runtime PM for this func
+ * Enable Runtime PM for this func (if supported)
*/
- pm_runtime_enable(&card->sdio_func[i]->dev);
+ if (host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_enable(&card->sdio_func[i]->dev);
}
mmc_release_host(host);
#include <linux/pm_runtime.h>
#include <linux/mmc/card.h>
+#include <linux/mmc/host.h>
#include <linux/mmc/sdio_func.h>
#include "sdio_cis.h"
* it should call pm_runtime_put_noidle() in its probe routine and
* pm_runtime_get_noresume() in its remove routine.
*/
- ret = pm_runtime_get_sync(dev);
- if (ret < 0)
- goto out;
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD) {
+ ret = pm_runtime_get_sync(dev);
+ if (ret < 0)
+ goto out;
+ }
/* Set the default block size so the driver is sure it's something
* sensible. */
return 0;
disable_runtimepm:
- pm_runtime_put_noidle(dev);
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_put_noidle(dev);
out:
return ret;
}
{
struct sdio_driver *drv = to_sdio_driver(dev->driver);
struct sdio_func *func = dev_to_sdio_func(dev);
- int ret;
+ int ret = 0;
/* Make sure card is powered before invoking ->remove() */
- ret = pm_runtime_get_sync(dev);
- if (ret < 0)
- goto out;
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD) {
+ ret = pm_runtime_get_sync(dev);
+ if (ret < 0)
+ goto out;
+ }
drv->remove(func);
}
/* First, undo the increment made directly above */
- pm_runtime_put_noidle(dev);
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_put_noidle(dev);
/* Then undo the runtime PM settings in sdio_bus_probe() */
- pm_runtime_put_noidle(dev);
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_put_noidle(dev);
out:
return ret;
static int sdio_bus_pm_prepare(struct device *dev)
{
+ struct sdio_func *func = dev_to_sdio_func(dev);
+
/*
* Resume an SDIO device which was suspended at run time at this
* point, in order to allow standard SDIO suspend/resume paths
* since there is little point in failing system suspend if a
* device can't be resumed.
*/
- pm_runtime_resume(dev);
+ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD)
+ pm_runtime_resume(dev);
return 0;
}
* Monitor a 0->1 transition first
*/
if (mmc_slot(host).features & HSMMC_HAS_UPDATED_RESET) {
- while ((!(OMAP_HSMMC_READ(host, SYSCTL) & bit))
+ while ((!(OMAP_HSMMC_READ(host->base, SYSCTL) & bit))
&& (i++ < limit))
cpu_relax();
}
#include <linux/clk.h>
#include <linux/mmc/host.h>
#include <linux/mmc/sdhci-pltfm.h>
+#include <mach/hardware.h>
#include "sdhci.h"
#include "sdhci-pltfm.h"
#include "sdhci-esdhc.h"
clk_enable(clk);
pltfm_host->clk = clk;
+ if (cpu_is_mx35() || cpu_is_mx51())
+ host->quirks |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;
+
+ /* Fix errata ENGcm07207 which is present on i.MX25 and i.MX35 */
+ if (cpu_is_mx25() || cpu_is_mx35())
+ host->quirks |= SDHCI_QUIRK_NO_MULTIBLOCK;
+
return 0;
}
};
struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
- .quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_NO_MULTIBLOCK
- | SDHCI_QUIRK_BROKEN_ADMA,
+ .quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_BROKEN_ADMA,
/* ADMA has issues. Might be fixable */
- /* NO_MULTIBLOCK might be MX35 only (Errata: ENGcm07207) */
.ops = &sdhci_esdhc_ops,
.init = esdhc_pltfm_init,
.exit = esdhc_pltfm_exit,
* ADMA operation is disabled for Moorestown platform due to
* hardware bugs.
*/
-static int mrst_hc1_probe(struct sdhci_pci_chip *chip)
+static int mrst_hc_probe(struct sdhci_pci_chip *chip)
{
/*
- * slots number is fixed here for MRST as SDIO3 is never used and has
- * hardware bugs.
+ * slots number is fixed here for MRST as SDIO3/5 are never used and
+ * have hardware bugs.
*/
chip->num_slots = 1;
return 0;
.quirks = SDHCI_QUIRK_BROKEN_ADMA | SDHCI_QUIRK_NO_HISPD_BIT,
};
-static const struct sdhci_pci_fixes sdhci_intel_mrst_hc1 = {
+static const struct sdhci_pci_fixes sdhci_intel_mrst_hc1_hc2 = {
.quirks = SDHCI_QUIRK_BROKEN_ADMA | SDHCI_QUIRK_NO_HISPD_BIT,
- .probe = mrst_hc1_probe,
+ .probe = mrst_hc_probe,
};
static const struct sdhci_pci_fixes sdhci_intel_mfd_sd = {
.device = PCI_DEVICE_ID_INTEL_MRST_SD1,
.subvendor = PCI_ANY_ID,
.subdevice = PCI_ANY_ID,
- .driver_data = (kernel_ulong_t)&sdhci_intel_mrst_hc1,
+ .driver_data = (kernel_ulong_t)&sdhci_intel_mrst_hc1_hc2,
+ },
+
+ {
+ .vendor = PCI_VENDOR_ID_INTEL,
+ .device = PCI_DEVICE_ID_INTEL_MRST_SD2,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = PCI_ANY_ID,
+ .driver_data = (kernel_ulong_t)&sdhci_intel_mrst_hc1_hc2,
},
{
{
struct sdhci_pci_chip *chip;
struct sdhci_pci_slot *slot;
+ mmc_pm_flag_t slot_pm_flags;
mmc_pm_flag_t pm_flags = 0;
int i, ret;
return ret;
}
- pm_flags |= slot->host->mmc->pm_flags;
+ slot_pm_flags = slot->host->mmc->pm_flags;
+ if (slot_pm_flags & MMC_PM_WAKE_SDIO_IRQ)
+ sdhci_enable_irq_wakeups(slot->host);
+
+ pm_flags |= slot_pm_flags;
}
if (chip->fixes && chip->fixes->suspend) {
pci_save_state(pdev);
if (pm_flags & MMC_PM_KEEP_POWER) {
- if (pm_flags & MMC_PM_WAKE_SDIO_IRQ)
+ if (pm_flags & MMC_PM_WAKE_SDIO_IRQ) {
+ pci_pme_active(pdev, true);
pci_enable_wake(pdev, PCI_D3hot, 1);
+ }
pci_set_power_state(pdev, PCI_D3hot);
} else {
pci_enable_wake(pdev, pci_choose_state(pdev, state), 0);
if (pdata->quirks)
host->quirks |= pdata->quirks;
+ /* If slot design supports 8 bit data, indicate this to MMC. */
+ if (pdata->flags & PXA_FLAG_SD_8_BIT_CAPABLE_SLOT)
+ host->mmc->caps |= MMC_CAP_8_BIT_DATA;
+
ret = sdhci_add_host(host);
if (ret) {
dev_err(&pdev->dev, "failed to add host\n");
if (host->ops->platform_send_init_74_clocks)
host->ops->platform_send_init_74_clocks(host, ios->power_mode);
- ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
-
- if (ios->bus_width == MMC_BUS_WIDTH_8)
- ctrl |= SDHCI_CTRL_8BITBUS;
- else
- ctrl &= ~SDHCI_CTRL_8BITBUS;
+ /*
+ * If your platform has 8-bit width support but is not a v3 controller,
+ * or if it requires special setup code, you should implement that in
+ * platform_8bit_width().
+ */
+ if (host->ops->platform_8bit_width)
+ host->ops->platform_8bit_width(host, ios->bus_width);
+ else {
+ ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
+ if (ios->bus_width == MMC_BUS_WIDTH_8) {
+ ctrl &= ~SDHCI_CTRL_4BITBUS;
+ if (host->version >= SDHCI_SPEC_300)
+ ctrl |= SDHCI_CTRL_8BITBUS;
+ } else {
+ if (host->version >= SDHCI_SPEC_300)
+ ctrl &= ~SDHCI_CTRL_8BITBUS;
+ if (ios->bus_width == MMC_BUS_WIDTH_4)
+ ctrl |= SDHCI_CTRL_4BITBUS;
+ else
+ ctrl &= ~SDHCI_CTRL_4BITBUS;
+ }
+ sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
+ }
- if (ios->bus_width == MMC_BUS_WIDTH_4)
- ctrl |= SDHCI_CTRL_4BITBUS;
- else
- ctrl &= ~SDHCI_CTRL_4BITBUS;
+ ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
if ((ios->timing == MMC_TIMING_SD_HS ||
ios->timing == MMC_TIMING_MMC_HS)
EXPORT_SYMBOL_GPL(sdhci_resume_host);
+void sdhci_enable_irq_wakeups(struct sdhci_host *host)
+{
+ u8 val;
+ val = sdhci_readb(host, SDHCI_WAKE_UP_CONTROL);
+ val |= SDHCI_WAKE_ON_INT;
+ sdhci_writeb(host, val, SDHCI_WAKE_UP_CONTROL);
+}
+
+EXPORT_SYMBOL_GPL(sdhci_enable_irq_wakeups);
+
#endif /* CONFIG_PM */
/*****************************************************************************\
mmc->f_min = host->max_clk / SDHCI_MAX_DIV_SPEC_300;
else
mmc->f_min = host->max_clk / SDHCI_MAX_DIV_SPEC_200;
+
mmc->f_max = host->max_clk;
mmc->caps |= MMC_CAP_SDIO_IRQ;
+ /*
+ * A controller may support 8-bit width, but the board itself
+ * might not have the pins brought out. Boards that support
+ * 8-bit width must set "mmc->caps |= MMC_CAP_8_BIT_DATA;" in
+ * their platform code before calling sdhci_add_host(), and we
+ * won't assume 8-bit width for hosts without that CAP.
+ */
if (!(host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA))
- mmc->caps |= MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA;
+ mmc->caps |= MMC_CAP_4_BIT_DATA;
if (caps & SDHCI_CAN_DO_HISPD)
mmc->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED;
#define SDHCI_CTRL_ADMA1 0x08
#define SDHCI_CTRL_ADMA32 0x10
#define SDHCI_CTRL_ADMA64 0x18
-#define SDHCI_CTRL_8BITBUS 0x20
+#define SDHCI_CTRL_8BITBUS 0x20
#define SDHCI_POWER_CONTROL 0x29
#define SDHCI_POWER_ON 0x01
#define SDHCI_BLOCK_GAP_CONTROL 0x2A
#define SDHCI_WAKE_UP_CONTROL 0x2B
+#define SDHCI_WAKE_ON_INT 0x01
+#define SDHCI_WAKE_ON_INSERT 0x02
+#define SDHCI_WAKE_ON_REMOVE 0x04
#define SDHCI_CLOCK_CONTROL 0x2C
#define SDHCI_DIVIDER_SHIFT 8
#define SDHCI_CLOCK_BASE_SHIFT 8
#define SDHCI_MAX_BLOCK_MASK 0x00030000
#define SDHCI_MAX_BLOCK_SHIFT 16
+#define SDHCI_CAN_DO_8BIT 0x00040000
#define SDHCI_CAN_DO_ADMA2 0x00080000
#define SDHCI_CAN_DO_ADMA1 0x00100000
#define SDHCI_CAN_DO_HISPD 0x00200000
unsigned int (*get_max_clock)(struct sdhci_host *host);
unsigned int (*get_min_clock)(struct sdhci_host *host);
unsigned int (*get_timeout_clock)(struct sdhci_host *host);
+ int (*platform_8bit_width)(struct sdhci_host *host,
+ int width);
void (*platform_send_init_74_clocks)(struct sdhci_host *host,
u8 power_mode);
unsigned int (*get_ro)(struct sdhci_host *host);
#ifdef CONFIG_PM
extern int sdhci_suspend_host(struct sdhci_host *host, pm_message_t state);
extern int sdhci_resume_host(struct sdhci_host *host);
+extern void sdhci_enable_irq_wakeups(struct sdhci_host *host);
#endif
#endif /* __SDHCI_HW_H */
struct usb_device *usb_dev = interface_to_usbdev(intf);
struct mmc_host *mmc;
struct ushc_data *ushc;
- int ret = -ENOMEM;
+ int ret;
mmc = mmc_alloc_host(sizeof(struct ushc_data), &intf->dev);
if (mmc == NULL)
mmc->max_blk_count = 511;
ushc->int_urb = usb_alloc_urb(0, GFP_KERNEL);
- if (ushc->int_urb == NULL)
+ if (ushc->int_urb == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
ushc->int_data = kzalloc(sizeof(struct ushc_int_data), GFP_KERNEL);
- if (ushc->int_data == NULL)
+ if (ushc->int_data == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
usb_fill_int_urb(ushc->int_urb, ushc->usb_dev,
usb_rcvintpipe(usb_dev,
intf->cur_altsetting->endpoint[0].desc.bEndpointAddress),
intf->cur_altsetting->endpoint[0].desc.bInterval);
ushc->cbw_urb = usb_alloc_urb(0, GFP_KERNEL);
- if (ushc->cbw_urb == NULL)
+ if (ushc->cbw_urb == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
ushc->cbw = kzalloc(sizeof(struct ushc_cbw), GFP_KERNEL);
- if (ushc->cbw == NULL)
+ if (ushc->cbw == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
ushc->cbw->signature = USHC_CBW_SIGNATURE;
usb_fill_bulk_urb(ushc->cbw_urb, ushc->usb_dev, usb_sndbulkpipe(usb_dev, 2),
cbw_callback, ushc);
ushc->data_urb = usb_alloc_urb(0, GFP_KERNEL);
- if (ushc->data_urb == NULL)
+ if (ushc->data_urb == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
ushc->csw_urb = usb_alloc_urb(0, GFP_KERNEL);
- if (ushc->csw_urb == NULL)
+ if (ushc->csw_urb == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
ushc->csw = kzalloc(sizeof(struct ushc_cbw), GFP_KERNEL);
- if (ushc->csw == NULL)
+ if (ushc->csw == NULL) {
+ ret = -ENOMEM;
goto err;
+ }
usb_fill_bulk_urb(ushc->csw_urb, ushc->usb_dev, usb_rcvbulkpipe(usb_dev, 6),
ushc->csw, sizeof(struct ushc_csw),
csw_callback, ushc);
* erased, so it became unstable and corrupted, and should be
* erased.
*/
- return 0;
+ err = 0;
+ goto out_unlock;
}
if (err)
- return err;
+ goto out_unlock;
- if (ubi_check_pattern(ubi->peb_buf1, 0xFF, ubi->leb_size)) {
- mutex_unlock(&ubi->buf_mutex);
- return 0;
- }
+ if (ubi_check_pattern(ubi->peb_buf1, 0xFF, ubi->leb_size))
+ goto out_unlock;
ubi_err("PEB %d contains corrupted VID header, and the data does not "
"contain all 0xFF, this may be a non-UBI PEB or a severe VID "
pnum, ubi->leb_start, ubi->leb_size);
ubi_dbg_print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
ubi->peb_buf1, ubi->leb_size, 1);
+ err = 1;
+
+out_unlock:
mutex_unlock(&ubi->buf_mutex);
- return 1;
+ return err;
}
/**
depends on PCI
select MII
---help---
- This is a gigabit ethernet driver for Topcliff PCH.
- Topcliff PCH is the platform controller hub that is used in Intel's
+ This is a gigabit ethernet driver for EG20T PCH.
+ EG20T PCH is the platform controller hub that is used in Intel's
general embedded platform.
- Topcliff PCH has Gigabit Ethernet interface.
+ EG20T PCH has Gigabit Ethernet interface.
Using this interface, it is able to access system devices connected
to Gigabit Ethernet.
This driver enables Gigabit Ethernet function.
source "drivers/net/caif/Kconfig"
+config TILE_NET
+ tristate "Tilera GBE/XGBE network driver support"
+ depends on TILE
+ default y
+ select CRC32
+ help
+ This is a standard Linux network device driver for the
+ on-chip Tilera Gigabit Ethernet and XAUI interfaces.
+
+ To compile this driver as a module, choose M here: the module
+ will be called tile_net.
+
config XEN_NETDEV_FRONTEND
tristate "Xen network device frontend driver"
depends on XEN
obj-$(CONFIG_OCTEON_MGMT_ETHERNET) += octeon/
obj-$(CONFIG_PCH_GBE) += pch_gbe/
+obj-$(CONFIG_TILE_NET) += tile/
addr[0] = addr[1] = 0;
AT_READ_REG(hw, REG_OTP_CTRL, &otp_ctrl_data);
if (atl1c_check_eeprom_exist(hw)) {
- if (hw->nic_type == athr_l1c || hw->nic_type == athr_l2c_b) {
+ if (hw->nic_type == athr_l1c || hw->nic_type == athr_l2c) {
/* Enable OTP CLK */
if (!(otp_ctrl_data & OTP_CTRL_CLK_EN)) {
otp_ctrl_data |= OTP_CTRL_CLK_EN;
spin_lock_irqsave(&aup->lock, flags);
if (force_reset || (!aup->mac_enabled)) {
- writel(MAC_EN_CLOCK_ENABLE, &aup->enable);
+ writel(MAC_EN_CLOCK_ENABLE, aup->enable);
au_sync_delay(2);
writel((MAC_EN_RESET0 | MAC_EN_RESET1 | MAC_EN_RESET2
- | MAC_EN_CLOCK_ENABLE), &aup->enable);
+ | MAC_EN_CLOCK_ENABLE), aup->enable);
au_sync_delay(2);
aup->mac_enabled = 1;
au1000_hard_stop(dev);
- writel(MAC_EN_CLOCK_ENABLE, &aup->enable);
+ writel(MAC_EN_CLOCK_ENABLE, aup->enable);
au_sync_delay(2);
- writel(0, &aup->enable);
+ writel(0, aup->enable);
au_sync_delay(2);
aup->tx_full = 0;
/* set a random MAC now in case platform_data doesn't provide one */
random_ether_addr(dev->dev_addr);
- writel(0, &aup->enable);
+ writel(0, aup->enable);
aup->mac_enabled = 0;
pd = pdev->dev.platform_data;
}
/*
- * Collect up to maxaddrs worth of a netdevice's unicast addresses into an
- * array of addrss pointers and return the number collected.
+ * Collect up to maxaddrs worth of a netdevice's unicast addresses, starting
+ * at a specified offset within the list, into an array of addrss pointers and
+ * return the number collected.
*/
-static inline int collect_netdev_uc_list_addrs(const struct net_device *dev,
- const u8 **addr,
- unsigned int maxaddrs)
+static inline unsigned int collect_netdev_uc_list_addrs(const struct net_device *dev,
+ const u8 **addr,
+ unsigned int offset,
+ unsigned int maxaddrs)
{
+ unsigned int index = 0;
unsigned int naddr = 0;
const struct netdev_hw_addr *ha;
- for_each_dev_addr(dev, ha) {
- addr[naddr++] = ha->addr;
- if (naddr >= maxaddrs)
- break;
- }
+ for_each_dev_addr(dev, ha)
+ if (index++ >= offset) {
+ addr[naddr++] = ha->addr;
+ if (naddr >= maxaddrs)
+ break;
+ }
return naddr;
}
/*
- * Collect up to maxaddrs worth of a netdevice's multicast addresses into an
- * array of addrss pointers and return the number collected.
+ * Collect up to maxaddrs worth of a netdevice's multicast addresses, starting
+ * at a specified offset within the list, into an array of addrss pointers and
+ * return the number collected.
*/
-static inline int collect_netdev_mc_list_addrs(const struct net_device *dev,
- const u8 **addr,
- unsigned int maxaddrs)
+static inline unsigned int collect_netdev_mc_list_addrs(const struct net_device *dev,
+ const u8 **addr,
+ unsigned int offset,
+ unsigned int maxaddrs)
{
+ unsigned int index = 0;
unsigned int naddr = 0;
const struct netdev_hw_addr *ha;
- netdev_for_each_mc_addr(ha, dev) {
- addr[naddr++] = ha->addr;
- if (naddr >= maxaddrs)
- break;
- }
+ netdev_for_each_mc_addr(ha, dev)
+ if (index++ >= offset) {
+ addr[naddr++] = ha->addr;
+ if (naddr >= maxaddrs)
+ break;
+ }
return naddr;
}
u64 mhash = 0;
u64 uhash = 0;
bool free = true;
- u16 filt_idx[7];
+ unsigned int offset, naddr;
const u8 *addr[7];
- int ret, naddr = 0;
+ int ret;
const struct port_info *pi = netdev_priv(dev);
/* first do the secondary unicast addresses */
- naddr = collect_netdev_uc_list_addrs(dev, addr, ARRAY_SIZE(addr));
- if (naddr > 0) {
+ for (offset = 0; ; offset += naddr) {
+ naddr = collect_netdev_uc_list_addrs(dev, addr, offset,
+ ARRAY_SIZE(addr));
+ if (naddr == 0)
+ break;
+
ret = t4vf_alloc_mac_filt(pi->adapter, pi->viid, free,
- naddr, addr, filt_idx, &uhash, sleep);
+ naddr, addr, NULL, &uhash, sleep);
if (ret < 0)
return ret;
}
/* next set up the multicast addresses */
- naddr = collect_netdev_mc_list_addrs(dev, addr, ARRAY_SIZE(addr));
- if (naddr > 0) {
+ for (offset = 0; ; offset += naddr) {
+ naddr = collect_netdev_mc_list_addrs(dev, addr, offset,
+ ARRAY_SIZE(addr));
+ if (naddr == 0)
+ break;
+
ret = t4vf_alloc_mac_filt(pi->adapter, pi->viid, free,
- naddr, addr, filt_idx, &mhash, sleep);
+ naddr, addr, NULL, &mhash, sleep);
if (ret < 0)
return ret;
+ free = false;
}
return t4vf_set_addr_hash(pi->adapter, pi->viid, uhash != 0,
unsigned int naddr, const u8 **addr, u16 *idx,
u64 *hash, bool sleep_ok)
{
- int i, ret;
+ int offset, ret = 0;
+ unsigned nfilters = 0;
+ unsigned int rem = naddr;
struct fw_vi_mac_cmd cmd, rpl;
- struct fw_vi_mac_exact *p;
- size_t len16;
- if (naddr > ARRAY_SIZE(cmd.u.exact))
+ if (naddr > FW_CLS_TCAM_NUM_ENTRIES)
return -EINVAL;
- len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd,
- u.exact[naddr]), 16);
- memset(&cmd, 0, sizeof(cmd));
- cmd.op_to_viid = cpu_to_be32(FW_CMD_OP(FW_VI_MAC_CMD) |
- FW_CMD_REQUEST |
- FW_CMD_WRITE |
- (free ? FW_CMD_EXEC : 0) |
- FW_VI_MAC_CMD_VIID(viid));
- cmd.freemacs_to_len16 = cpu_to_be32(FW_VI_MAC_CMD_FREEMACS(free) |
- FW_CMD_LEN16(len16));
+ for (offset = 0; offset < naddr; /**/) {
+ unsigned int fw_naddr = (rem < ARRAY_SIZE(cmd.u.exact)
+ ? rem
+ : ARRAY_SIZE(cmd.u.exact));
+ size_t len16 = DIV_ROUND_UP(offsetof(struct fw_vi_mac_cmd,
+ u.exact[fw_naddr]), 16);
+ struct fw_vi_mac_exact *p;
+ int i;
- for (i = 0, p = cmd.u.exact; i < naddr; i++, p++) {
- p->valid_to_idx =
- cpu_to_be16(FW_VI_MAC_CMD_VALID |
- FW_VI_MAC_CMD_IDX(FW_VI_MAC_ADD_MAC));
- memcpy(p->macaddr, addr[i], sizeof(p->macaddr));
- }
+ memset(&cmd, 0, sizeof(cmd));
+ cmd.op_to_viid = cpu_to_be32(FW_CMD_OP(FW_VI_MAC_CMD) |
+ FW_CMD_REQUEST |
+ FW_CMD_WRITE |
+ (free ? FW_CMD_EXEC : 0) |
+ FW_VI_MAC_CMD_VIID(viid));
+ cmd.freemacs_to_len16 =
+ cpu_to_be32(FW_VI_MAC_CMD_FREEMACS(free) |
+ FW_CMD_LEN16(len16));
+
+ for (i = 0, p = cmd.u.exact; i < fw_naddr; i++, p++) {
+ p->valid_to_idx = cpu_to_be16(
+ FW_VI_MAC_CMD_VALID |
+ FW_VI_MAC_CMD_IDX(FW_VI_MAC_ADD_MAC));
+ memcpy(p->macaddr, addr[offset+i], sizeof(p->macaddr));
+ }
+
+
+ ret = t4vf_wr_mbox_core(adapter, &cmd, sizeof(cmd), &rpl,
+ sleep_ok);
+ if (ret && ret != -ENOMEM)
+ break;
- ret = t4vf_wr_mbox_core(adapter, &cmd, sizeof(cmd), &rpl, sleep_ok);
- if (ret)
- return ret;
-
- for (i = 0, p = rpl.u.exact; i < naddr; i++, p++) {
- u16 index = FW_VI_MAC_CMD_IDX_GET(be16_to_cpu(p->valid_to_idx));
-
- if (idx)
- idx[i] = (index >= FW_CLS_TCAM_NUM_ENTRIES
- ? 0xffff
- : index);
- if (index < FW_CLS_TCAM_NUM_ENTRIES)
- ret++;
- else if (hash)
- *hash |= (1 << hash_mac_addr(addr[i]));
+ for (i = 0, p = rpl.u.exact; i < fw_naddr; i++, p++) {
+ u16 index = FW_VI_MAC_CMD_IDX_GET(
+ be16_to_cpu(p->valid_to_idx));
+
+ if (idx)
+ idx[offset+i] =
+ (index >= FW_CLS_TCAM_NUM_ENTRIES
+ ? 0xffff
+ : index);
+ if (index < FW_CLS_TCAM_NUM_ENTRIES)
+ nfilters++;
+ else if (hash)
+ *hash |= (1ULL << hash_mac_addr(addr[offset+i]));
+ }
+
+ free = false;
+ offset += fw_naddr;
+ rem -= fw_naddr;
}
+
+ /*
+ * If there were no errors or we merely ran out of room in our MAC
+ * address arena, return the number of filters actually written.
+ */
+ if (ret == 0 || ret == -ENOMEM)
+ ret = nfilters;
return ret;
}
char e1000_driver_name[] = "e1000";
static char e1000_driver_string[] = "Intel(R) PRO/1000 Network Driver";
-#define DRV_VERSION "7.3.21-k6-NAPI"
+#define DRV_VERSION "7.3.21-k8-NAPI"
const char e1000_driver_version[] = DRV_VERSION;
static const char e1000_copyright[] = "Copyright (c) 1999-2006 Intel Corporation.";
struct net_device *netdev = adapter->netdev;
u32 rctl, tctl;
- /* signal that we're down so the interrupt handler does not
- * reschedule our watchdog timer */
- set_bit(__E1000_DOWN, &adapter->flags);
/* disable receives in the hardware */
rctl = er32(RCTL);
e1000_irq_disable(adapter);
+ /*
+ * Setting DOWN must be after irq_disable to prevent
+ * a screaming interrupt. Setting DOWN also prevents
+ * timers and tasks from rescheduling.
+ */
+ set_bit(__E1000_DOWN, &adapter->flags);
+
del_timer_sync(&adapter->tx_fifo_stall_timer);
del_timer_sync(&adapter->watchdog_timer);
del_timer_sync(&adapter->phy_info_timer);
skb_arr_rq1[index] = netdev_alloc_skb(dev,
EHEA_L_PKT_SIZE);
if (!skb_arr_rq1[index]) {
+ ehea_info("Unable to allocate enough skb in the array\n");
pr->rq1_skba.os_skbs = fill_wqes - i;
break;
}
struct net_device *dev = pr->port->netdev;
int i;
- for (i = 0; i < pr->rq1_skba.len; i++) {
+ if (nr_rq1a > pr->rq1_skba.len) {
+ ehea_error("NR_RQ1A bigger than skb array len\n");
+ return;
+ }
+
+ for (i = 0; i < nr_rq1a; i++) {
skb_arr_rq1[i] = netdev_alloc_skb(dev, EHEA_L_PKT_SIZE);
- if (!skb_arr_rq1[i])
+ if (!skb_arr_rq1[i]) {
+ ehea_info("No enough memory to allocate skb array\n");
break;
+ }
}
/* Ring doorbell */
- ehea_update_rq1a(pr->qp, nr_rq1a);
+ ehea_update_rq1a(pr->qp, i);
}
static int ehea_refill_rq_def(struct ehea_port_res *pr,
skb = netdev_alloc_skb(dev,
EHEA_L_PKT_SIZE);
- if (!skb)
+ if (!skb) {
+ ehea_info("Not enough memory to allocate skb\n");
break;
+ }
}
skb_copy_to_linear_data(skb, ((char *)cqe) + 64,
cqe->num_bytes_transfered - 4);
/* Baud Rate Error Correction x 10000 */
u32 rate_err_array[] = {
- 0000, 0625, 1250, 1875,
+ 0, 625, 1250, 1875,
2500, 3125, 3750, 4375,
5000, 5625, 6250, 6875,
7500, 8125, 8750, 9375,
/*
* Copyright (C) 1999 - 2010 Intel Corporation.
- * Copyright (C) 2010 OKI SEMICONDUCTOR Co., LTD.
+ * Copyright (C) 2010 OKI SEMICONDUCTOR CO., LTD.
*
* This code was derived from the Intel e1000e Linux driver.
*
module_init(pch_gbe_init_module);
module_exit(pch_gbe_exit_module);
-MODULE_DESCRIPTION("OKI semiconductor PCH Gigabit ethernet Driver");
-MODULE_AUTHOR("OKI semiconductor, <masa-korg@dsn.okisemi.com>");
+MODULE_DESCRIPTION("EG20T PCH Gigabit ethernet Driver");
+MODULE_AUTHOR("OKI SEMICONDUCTOR, <toshiharu-linux@dsn.okisemi.com>");
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_VERSION);
MODULE_DEVICE_TABLE(pci, pch_gbe_pcidev_id);
.err = "using default of "
__MODULE_STRING(PCH_GBE_DEFAULT_TXD),
.def = PCH_GBE_DEFAULT_TXD,
- .arg = { .r = { .min = PCH_GBE_MIN_TXD } },
- .arg = { .r = { .max = PCH_GBE_MAX_TXD } }
+ .arg = { .r = { .min = PCH_GBE_MIN_TXD,
+ .max = PCH_GBE_MAX_TXD } }
};
struct pch_gbe_tx_ring *tx_ring = adapter->tx_ring;
tx_ring->count = TxDescriptors;
.err = "using default of "
__MODULE_STRING(PCH_GBE_DEFAULT_RXD),
.def = PCH_GBE_DEFAULT_RXD,
- .arg = { .r = { .min = PCH_GBE_MIN_RXD } },
- .arg = { .r = { .max = PCH_GBE_MAX_RXD } }
+ .arg = { .r = { .min = PCH_GBE_MIN_RXD,
+ .max = PCH_GBE_MAX_RXD } }
};
struct pch_gbe_rx_ring *rx_ring = adapter->rx_ring;
rx_ring->count = RxDescriptors;
#include <linux/ethtool.h>
#include <linux/phy.h>
#include <linux/marvell_phy.h>
+#include <linux/of.h>
#include <asm/io.h>
#include <asm/irq.h>
#include <asm/uaccess.h>
+#define MII_MARVELL_PHY_PAGE 22
+
#define MII_M1011_IEVENT 0x13
#define MII_M1011_IEVENT_CLEAR 0x0000
#define MII_88E1121_PHY_LED_CTRL 16
#define MII_88E1121_PHY_LED_PAGE 3
#define MII_88E1121_PHY_LED_DEF 0x0030
-#define MII_88E1121_PHY_PAGE 22
#define MII_M1011_PHY_STATUS 0x11
#define MII_M1011_PHY_STATUS_1000 0x8000
return 0;
}
+#ifdef CONFIG_OF_MDIO
+/*
+ * Set and/or override some configuration registers based on the
+ * marvell,reg-init property stored in the of_node for the phydev.
+ *
+ * marvell,reg-init = <reg-page reg mask value>,...;
+ *
+ * There may be one or more sets of <reg-page reg mask value>:
+ *
+ * reg-page: which register bank to use.
+ * reg: the register.
+ * mask: if non-zero, ANDed with existing register value.
+ * value: ORed with the masked value and written to the regiser.
+ *
+ */
+static int marvell_of_reg_init(struct phy_device *phydev)
+{
+ const __be32 *paddr;
+ int len, i, saved_page, current_page, page_changed, ret;
+
+ if (!phydev->dev.of_node)
+ return 0;
+
+ paddr = of_get_property(phydev->dev.of_node, "marvell,reg-init", &len);
+ if (!paddr || len < (4 * sizeof(*paddr)))
+ return 0;
+
+ saved_page = phy_read(phydev, MII_MARVELL_PHY_PAGE);
+ if (saved_page < 0)
+ return saved_page;
+ page_changed = 0;
+ current_page = saved_page;
+
+ ret = 0;
+ len /= sizeof(*paddr);
+ for (i = 0; i < len - 3; i += 4) {
+ u16 reg_page = be32_to_cpup(paddr + i);
+ u16 reg = be32_to_cpup(paddr + i + 1);
+ u16 mask = be32_to_cpup(paddr + i + 2);
+ u16 val_bits = be32_to_cpup(paddr + i + 3);
+ int val;
+
+ if (reg_page != current_page) {
+ current_page = reg_page;
+ page_changed = 1;
+ ret = phy_write(phydev, MII_MARVELL_PHY_PAGE, reg_page);
+ if (ret < 0)
+ goto err;
+ }
+
+ val = 0;
+ if (mask) {
+ val = phy_read(phydev, reg);
+ if (val < 0) {
+ ret = val;
+ goto err;
+ }
+ val &= mask;
+ }
+ val |= val_bits;
+
+ ret = phy_write(phydev, reg, val);
+ if (ret < 0)
+ goto err;
+
+ }
+err:
+ if (page_changed) {
+ i = phy_write(phydev, MII_MARVELL_PHY_PAGE, saved_page);
+ if (ret == 0)
+ ret = i;
+ }
+ return ret;
+}
+#else
+static int marvell_of_reg_init(struct phy_device *phydev)
+{
+ return 0;
+}
+#endif /* CONFIG_OF_MDIO */
+
static int m88e1121_config_aneg(struct phy_device *phydev)
{
int err, oldpage, mscr;
- oldpage = phy_read(phydev, MII_88E1121_PHY_PAGE);
+ oldpage = phy_read(phydev, MII_MARVELL_PHY_PAGE);
- err = phy_write(phydev, MII_88E1121_PHY_PAGE,
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
MII_88E1121_PHY_MSCR_PAGE);
if (err < 0)
return err;
return err;
}
- phy_write(phydev, MII_88E1121_PHY_PAGE, oldpage);
+ phy_write(phydev, MII_MARVELL_PHY_PAGE, oldpage);
err = phy_write(phydev, MII_BMCR, BMCR_RESET);
if (err < 0)
if (err < 0)
return err;
- oldpage = phy_read(phydev, MII_88E1121_PHY_PAGE);
+ oldpage = phy_read(phydev, MII_MARVELL_PHY_PAGE);
- phy_write(phydev, MII_88E1121_PHY_PAGE, MII_88E1121_PHY_LED_PAGE);
+ phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_88E1121_PHY_LED_PAGE);
phy_write(phydev, MII_88E1121_PHY_LED_CTRL, MII_88E1121_PHY_LED_DEF);
- phy_write(phydev, MII_88E1121_PHY_PAGE, oldpage);
+ phy_write(phydev, MII_MARVELL_PHY_PAGE, oldpage);
err = genphy_config_aneg(phydev);
{
int err, oldpage, mscr;
- oldpage = phy_read(phydev, MII_88E1121_PHY_PAGE);
+ oldpage = phy_read(phydev, MII_MARVELL_PHY_PAGE);
- err = phy_write(phydev, MII_88E1121_PHY_PAGE,
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE,
MII_88E1121_PHY_MSCR_PAGE);
if (err < 0)
return err;
if (err < 0)
return err;
- err = phy_write(phydev, MII_88E1121_PHY_PAGE, oldpage);
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, oldpage);
if (err < 0)
return err;
return err;
}
+ err = marvell_of_reg_init(phydev);
+ if (err < 0)
+ return err;
err = phy_write(phydev, MII_BMCR, BMCR_RESET);
if (err < 0)
int err;
/* Change address */
- err = phy_write(phydev, 0x16, 0x0002);
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x0002);
if (err < 0)
return err;
return err;
/* Change address */
- err = phy_write(phydev, 0x16, 0x0003);
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x0003);
if (err < 0)
return err;
if (err < 0)
return err;
+ err = marvell_of_reg_init(phydev);
+ if (err < 0)
+ return err;
+
/* Reset address */
- err = phy_write(phydev, 0x16, 0x0);
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x0);
+ if (err < 0)
+ return err;
+
+ err = phy_write(phydev, MII_BMCR, BMCR_RESET);
+ if (err < 0)
+ return err;
+
+ return 0;
+}
+
+static int m88e1149_config_init(struct phy_device *phydev)
+{
+ int err;
+
+ /* Change address */
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x0002);
+ if (err < 0)
+ return err;
+
+ /* Enable 1000 Mbit */
+ err = phy_write(phydev, 0x15, 0x1048);
+ if (err < 0)
+ return err;
+
+ err = marvell_of_reg_init(phydev);
+ if (err < 0)
+ return err;
+
+ /* Reset address */
+ err = phy_write(phydev, MII_MARVELL_PHY_PAGE, 0x0);
if (err < 0)
return err;
}
}
+ err = marvell_of_reg_init(phydev);
+ if (err < 0)
+ return err;
+
return 0;
}
.config_intr = &marvell_config_intr,
.driver = { .owner = THIS_MODULE },
},
+ {
+ .phy_id = MARVELL_PHY_ID_88E1149R,
+ .phy_id_mask = MARVELL_PHY_ID_MASK,
+ .name = "Marvell 88E1149R",
+ .features = PHY_GBIT_FEATURES,
+ .flags = PHY_HAS_INTERRUPT,
+ .config_init = &m88e1149_config_init,
+ .config_aneg = &m88e1118_config_aneg,
+ .read_status = &genphy_read_status,
+ .ack_interrupt = &marvell_ack_interrupt,
+ .config_intr = &marvell_config_intr,
+ .driver = { .owner = THIS_MODULE },
+ },
{
.phy_id = MARVELL_PHY_ID_88E1240,
.phy_id_mask = MARVELL_PHY_ID_MASK,
{ 0x01410e10, 0xfffffff0 },
{ 0x01410cb0, 0xfffffff0 },
{ 0x01410cd0, 0xfffffff0 },
+ { 0x01410e50, 0xfffffff0 },
{ 0x01410e30, 0xfffffff0 },
{ 0x01410e90, 0xfffffff0 },
{ }
*/
dev_net_set(dev, net);
- ret = -EEXIST;
mutex_lock(&pn->all_ppp_mutex);
if (unit < 0) {
unit = unit_get(&pn->units_idr, ppp);
if (unit < 0) {
- *retp = unit;
+ ret = unit;
goto out2;
}
} else {
+ ret = -EEXIST;
if (unit_find(&pn->units_idr, unit))
goto out2; /* unit already exists */
/*
ppp->closing = 1;
ppp_unlock(ppp);
unregister_netdev(ppp->dev);
+ unit_put(&pn->units_idr, ppp->file.index);
} else
ppp_unlock(ppp);
- unit_put(&pn->units_idr, ppp->file.index);
ppp->file.dead = 1;
ppp->owner = NULL;
wake_up_interruptible(&ppp->file.rwait);
* by holding all_ppp_mutex
*/
-/* associate pointer with specified number */
-static int unit_set(struct idr *p, void *ptr, int n)
+static int __unit_alloc(struct idr *p, void *ptr, int n)
{
int unit, err;
}
err = idr_get_new_above(p, ptr, n, &unit);
- if (err == -EAGAIN)
- goto again;
+ if (err < 0) {
+ if (err == -EAGAIN)
+ goto again;
+ return err;
+ }
+
+ return unit;
+}
+
+/* associate pointer with specified number */
+static int unit_set(struct idr *p, void *ptr, int n)
+{
+ int unit;
- if (unit != n) {
+ unit = __unit_alloc(p, ptr, n);
+ if (unit < 0)
+ return unit;
+ else if (unit != n) {
idr_remove(p, unit);
return -EINVAL;
}
/* get new free unit number and associate pointer with it */
static int unit_get(struct idr *p, void *ptr)
{
- int unit, err;
-
-again:
- if (!idr_pre_get(p, GFP_KERNEL)) {
- printk(KERN_ERR "PPP: No free memory for idr\n");
- return -ENOMEM;
- }
-
- err = idr_get_new_above(p, ptr, 0, &unit);
- if (err == -EAGAIN)
- goto again;
-
- return unit;
+ return __unit_alloc(p, ptr, 0);
}
/* put unit number back to a pool */
/* NETIF_MSG_PKTDATA | */
NETIF_MSG_HW | NETIF_MSG_WOL | 0;
-static int debug = 0x00007fff; /* defaults above */
-module_param(debug, int, 0);
+static int debug = -1; /* defaults above */
+module_param(debug, int, 0664);
MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
#define MSIX_IRQ 0
#define MSI_IRQ 1
#define LEG_IRQ 2
static int qlge_irq_type = MSIX_IRQ;
-module_param(qlge_irq_type, int, MSIX_IRQ);
+module_param(qlge_irq_type, int, 0664);
MODULE_PARM_DESC(qlge_irq_type, "0 = MSI-X, 1 = MSI, 2 = Legacy.");
static int qlge_mpi_coredump;
--- /dev/null
+#
+# Makefile for the TILE on-chip networking support.
+#
+
+obj-$(CONFIG_TILE_NET) += tile_net.o
+ifdef CONFIG_TILEGX
+tile_net-objs := tilegx.o mpipe.o iorpc_mpipe.o dma_queue.o
+else
+tile_net-objs := tilepro.o
+endif
--- /dev/null
+/*
+ * Copyright 2010 Tilera Corporation. All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, version 2.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for
+ * more details.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/moduleparam.h>
+#include <linux/sched.h>
+#include <linux/kernel.h> /* printk() */
+#include <linux/slab.h> /* kmalloc() */
+#include <linux/errno.h> /* error codes */
+#include <linux/types.h> /* size_t */
+#include <linux/interrupt.h>
+#include <linux/in.h>
+#include <linux/netdevice.h> /* struct device, and other headers */
+#include <linux/etherdevice.h> /* eth_type_trans */
+#include <linux/skbuff.h>
+#include <linux/ioctl.h>
+#include <linux/cdev.h>
+#include <linux/hugetlb.h>
+#include <linux/in6.h>
+#include <linux/timer.h>
+#include <linux/io.h>
+#include <asm/checksum.h>
+#include <asm/homecache.h>
+
+#include <hv/drv_xgbe_intf.h>
+#include <hv/drv_xgbe_impl.h>
+#include <hv/hypervisor.h>
+#include <hv/netio_intf.h>
+
+/* For TSO */
+#include <linux/ip.h>
+#include <linux/tcp.h>
+
+
+/* There is no singlethread_cpu, so schedule work on the current cpu. */
+#define singlethread_cpu -1
+
+
+/*
+ * First, "tile_net_init_module()" initializes all four "devices" which
+ * can be used by linux.
+ *
+ * Then, "ifconfig DEVICE up" calls "tile_net_open()", which analyzes
+ * the network cpus, then uses "tile_net_open_aux()" to initialize
+ * LIPP/LEPP, and then uses "tile_net_open_inner()" to register all
+ * the tiles, provide buffers to LIPP, allow ingress to start, and
+ * turn on hypervisor interrupt handling (and NAPI) on all tiles.
+ *
+ * If registration fails due to the link being down, then "retry_work"
+ * is used to keep calling "tile_net_open_inner()" until it succeeds.
+ *
+ * If "ifconfig DEVICE down" is called, it uses "tile_net_stop()" to
+ * stop egress, drain the LIPP buffers, unregister all the tiles, stop
+ * LIPP/LEPP, and wipe the LEPP queue.
+ *
+ * We start out with the ingress interrupt enabled on each CPU. When
+ * this interrupt fires, we disable it, and call "napi_schedule()".
+ * This will cause "tile_net_poll()" to be called, which will pull
+ * packets from the netio queue, filtering them out, or passing them
+ * to "netif_receive_skb()". If our budget is exhausted, we will
+ * return, knowing we will be called again later. Otherwise, we
+ * reenable the ingress interrupt, and call "napi_complete()".
+ *
+ *
+ * NOTE: The use of "native_driver" ensures that EPP exists, and that
+ * "epp_sendv" is legal, and that "LIPP" is being used.
+ *
+ * NOTE: Failing to free completions for an arbitrarily long time
+ * (which is defined to be illegal) does in fact cause bizarre
+ * problems. The "egress_timer" helps prevent this from happening.
+ *
+ * NOTE: The egress code can be interrupted by the interrupt handler.
+ */
+
+
+/* HACK: Allow use of "jumbo" packets. */
+/* This should be 1500 if "jumbo" is not set in LIPP. */
+/* This should be at most 10226 (10240 - 14) if "jumbo" is set in LIPP. */
+/* ISSUE: This has not been thoroughly tested (except at 1500). */
+#define TILE_NET_MTU 1500
+
+/* HACK: Define to support GSO. */
+/* ISSUE: This may actually hurt performance of the TCP blaster. */
+/* #define TILE_NET_GSO */
+
+/* Define this to collapse "duplicate" acks. */
+/* #define IGNORE_DUP_ACKS */
+
+/* HACK: Define this to verify incoming packets. */
+/* #define TILE_NET_VERIFY_INGRESS */
+
+/* Use 3000 to enable the Linux Traffic Control (QoS) layer, else 0. */
+#define TILE_NET_TX_QUEUE_LEN 0
+
+/* Define to dump packets (prints out the whole packet on tx and rx). */
+/* #define TILE_NET_DUMP_PACKETS */
+
+/* Define to enable debug spew (all PDEBUG's are enabled). */
+/* #define TILE_NET_DEBUG */
+
+
+/* Define to activate paranoia checks. */
+/* #define TILE_NET_PARANOIA */
+
+/* Default transmit lockup timeout period, in jiffies. */
+#define TILE_NET_TIMEOUT (5 * HZ)
+
+/* Default retry interval for bringing up the NetIO interface, in jiffies. */
+#define TILE_NET_RETRY_INTERVAL (5 * HZ)
+
+/* Number of ports (xgbe0, xgbe1, gbe0, gbe1). */
+#define TILE_NET_DEVS 4
+
+
+
+/* Paranoia. */
+#if NET_IP_ALIGN != LIPP_PACKET_PADDING
+#error "NET_IP_ALIGN must match LIPP_PACKET_PADDING."
+#endif
+
+
+/* Debug print. */
+#ifdef TILE_NET_DEBUG
+#define PDEBUG(fmt, args...) net_printk(fmt, ## args)
+#else
+#define PDEBUG(fmt, args...)
+#endif
+
+
+MODULE_AUTHOR("Tilera");
+MODULE_LICENSE("GPL");
+
+
+#define IS_MULTICAST(mac_addr) \
+ (((u8 *)(mac_addr))[0] & 0x01)
+
+#define IS_BROADCAST(mac_addr) \
+ (((u16 *)(mac_addr))[0] == 0xffff)
+
+
+/*
+ * Queue of incoming packets for a specific cpu and device.
+ *
+ * Includes a pointer to the "system" data, and the actual "user" data.
+ */
+struct tile_netio_queue {
+ netio_queue_impl_t *__system_part;
+ netio_queue_user_impl_t __user_part;
+
+};
+
+
+/*
+ * Statistics counters for a specific cpu and device.
+ */
+struct tile_net_stats_t {
+ u32 rx_packets;
+ u32 rx_bytes;
+ u32 tx_packets;
+ u32 tx_bytes;
+};
+
+
+/*
+ * Info for a specific cpu and device.
+ *
+ * ISSUE: There is a "dev" pointer in "napi" as well.
+ */
+struct tile_net_cpu {
+ /* The NAPI struct. */
+ struct napi_struct napi;
+ /* Packet queue. */
+ struct tile_netio_queue queue;
+ /* Statistics. */
+ struct tile_net_stats_t stats;
+ /* ISSUE: Is this needed? */
+ bool napi_enabled;
+ /* True if this tile has succcessfully registered with the IPP. */
+ bool registered;
+ /* True if the link was down last time we tried to register. */
+ bool link_down;
+ /* True if "egress_timer" is scheduled. */
+ bool egress_timer_scheduled;
+ /* Number of small sk_buffs which must still be provided. */
+ unsigned int num_needed_small_buffers;
+ /* Number of large sk_buffs which must still be provided. */
+ unsigned int num_needed_large_buffers;
+ /* A timer for handling egress completions. */
+ struct timer_list egress_timer;
+};
+
+
+/*
+ * Info for a specific device.
+ */
+struct tile_net_priv {
+ /* Our network device. */
+ struct net_device *dev;
+ /* The actual egress queue. */
+ lepp_queue_t *epp_queue;
+ /* Protects "epp_queue->cmd_tail" and "epp_queue->comp_tail" */
+ spinlock_t cmd_lock;
+ /* Protects "epp_queue->comp_head". */
+ spinlock_t comp_lock;
+ /* The hypervisor handle for this interface. */
+ int hv_devhdl;
+ /* The intr bit mask that IDs this device. */
+ u32 intr_id;
+ /* True iff "tile_net_open_aux()" has succeeded. */
+ int partly_opened;
+ /* True iff "tile_net_open_inner()" has succeeded. */
+ int fully_opened;
+ /* Effective network cpus. */
+ struct cpumask network_cpus_map;
+ /* Number of network cpus. */
+ int network_cpus_count;
+ /* Credits per network cpu. */
+ int network_cpus_credits;
+ /* Network stats. */
+ struct net_device_stats stats;
+ /* For NetIO bringup retries. */
+ struct delayed_work retry_work;
+ /* Quick access to per cpu data. */
+ struct tile_net_cpu *cpu[NR_CPUS];
+};
+
+
+/*
+ * The actual devices (xgbe0, xgbe1, gbe0, gbe1).
+ */
+static struct net_device *tile_net_devs[TILE_NET_DEVS];
+
+/*
+ * The "tile_net_cpu" structures for each device.
+ */
+static DEFINE_PER_CPU(struct tile_net_cpu, hv_xgbe0);
+static DEFINE_PER_CPU(struct tile_net_cpu, hv_xgbe1);
+static DEFINE_PER_CPU(struct tile_net_cpu, hv_gbe0);
+static DEFINE_PER_CPU(struct tile_net_cpu, hv_gbe1);
+
+
+/*
+ * True if "network_cpus" was specified.
+ */
+static bool network_cpus_used;
+
+/*
+ * The actual cpus in "network_cpus".
+ */
+static struct cpumask network_cpus_map;
+
+
+
+#ifdef TILE_NET_DEBUG
+/*
+ * printk with extra stuff.
+ *
+ * We print the CPU we're running in brackets.
+ */
+static void net_printk(char *fmt, ...)
+{
+ int i;
+ int len;
+ va_list args;
+ static char buf[256];
+
+ len = sprintf(buf, "tile_net[%2.2d]: ", smp_processor_id());
+ va_start(args, fmt);
+ i = vscnprintf(buf + len, sizeof(buf) - len - 1, fmt, args);
+ va_end(args);
+ buf[255] = '\0';
+ pr_notice(buf);
+}
+#endif
+
+
+#ifdef TILE_NET_DUMP_PACKETS
+/*
+ * Dump a packet.
+ */
+static void dump_packet(unsigned char *data, unsigned long length, char *s)
+{
+ unsigned long i;
+ static unsigned int count;
+
+ pr_info("dump_packet(data %p, length 0x%lx s %s count 0x%x)\n",
+ data, length, s, count++);
+
+ pr_info("\n");
+
+ for (i = 0; i < length; i++) {
+ if ((i & 0xf) == 0)
+ sprintf(buf, "%8.8lx:", i);
+ sprintf(buf + strlen(buf), " %2.2x", data[i]);
+ if ((i & 0xf) == 0xf || i == length - 1)
+ pr_info("%s\n", buf);
+ }
+}
+#endif
+
+
+/*
+ * Provide support for the __netio_fastio1() swint
+ * (see <hv/drv_xgbe_intf.h> for how it is used).
+ *
+ * The fastio swint2 call may clobber all the caller-saved registers.
+ * It rarely clobbers memory, but we allow for the possibility in
+ * the signature just to be on the safe side.
+ *
+ * Also, gcc doesn't seem to allow an input operand to be
+ * clobbered, so we fake it with dummy outputs.
+ *
+ * This function can't be static because of the way it is declared
+ * in the netio header.
+ */
+inline int __netio_fastio1(u32 fastio_index, u32 arg0)
+{
+ long result, clobber_r1, clobber_r10;
+ asm volatile("swint2"
+ : "=R00" (result),
+ "=R01" (clobber_r1), "=R10" (clobber_r10)
+ : "R10" (fastio_index), "R01" (arg0)
+ : "memory", "r2", "r3", "r4",
+ "r5", "r6", "r7", "r8", "r9",
+ "r11", "r12", "r13", "r14",
+ "r15", "r16", "r17", "r18", "r19",
+ "r20", "r21", "r22", "r23", "r24",
+ "r25", "r26", "r27", "r28", "r29");
+ return result;
+}
+
+
+/*
+ * Provide a linux buffer to LIPP.
+ */
+static void tile_net_provide_linux_buffer(struct tile_net_cpu *info,
+ void *va, bool small)
+{
+ struct tile_netio_queue *queue = &info->queue;
+
+ /* Convert "va" and "small" to "linux_buffer_t". */
+ unsigned int buffer = ((unsigned int)(__pa(va) >> 7) << 1) + small;
+
+ __netio_fastio_free_buffer(queue->__user_part.__fastio_index, buffer);
+}
+
+
+/*
+ * Provide a linux buffer for LIPP.
+ */
+static bool tile_net_provide_needed_buffer(struct tile_net_cpu *info,
+ bool small)
+{
+ /* ISSUE: What should we use here? */
+ unsigned int large_size = NET_IP_ALIGN + TILE_NET_MTU + 100;
+
+ /* Round up to ensure to avoid "false sharing" with last cache line. */
+ unsigned int buffer_size =
+ (((small ? LIPP_SMALL_PACKET_SIZE : large_size) +
+ CHIP_L2_LINE_SIZE() - 1) & -CHIP_L2_LINE_SIZE());
+
+ /*
+ * ISSUE: Since CPAs are 38 bits, and we can only encode the
+ * high 31 bits in a "linux_buffer_t", the low 7 bits must be
+ * zero, and thus, we must align the actual "va" mod 128.
+ */
+ const unsigned long align = 128;
+
+ struct sk_buff *skb;
+ void *va;
+
+ struct sk_buff **skb_ptr;
+
+ /* Note that "dev_alloc_skb()" adds NET_SKB_PAD more bytes, */
+ /* and also "reserves" that many bytes. */
+ /* ISSUE: Can we "share" the NET_SKB_PAD bytes with "skb_ptr"? */
+ int len = sizeof(*skb_ptr) + align + buffer_size;
+
+ while (1) {
+
+ /* Allocate (or fail). */
+ skb = dev_alloc_skb(len);
+ if (skb == NULL)
+ return false;
+
+ /* Make room for a back-pointer to 'skb'. */
+ skb_reserve(skb, sizeof(*skb_ptr));
+
+ /* Make sure we are aligned. */
+ skb_reserve(skb, -(long)skb->data & (align - 1));
+
+ /* This address is given to IPP. */
+ va = skb->data;
+
+ if (small)
+ break;
+
+ /* ISSUE: This has never been observed! */
+ /* Large buffers must not span a huge page. */
+ if (((((long)va & ~HPAGE_MASK) + 1535) & HPAGE_MASK) == 0)
+ break;
+ pr_err("Leaking unaligned linux buffer at %p.\n", va);
+ }
+
+ /* Skip two bytes to satisfy LIPP assumptions. */
+ /* Note that this aligns IP on a 16 byte boundary. */
+ /* ISSUE: Do this when the packet arrives? */
+ skb_reserve(skb, NET_IP_ALIGN);
+
+ /* Save a back-pointer to 'skb'. */
+ skb_ptr = va - sizeof(*skb_ptr);
+ *skb_ptr = skb;
+
+ /* Invalidate the packet buffer. */
+ if (!hash_default)
+ __inv_buffer(skb->data, buffer_size);
+
+ /* Make sure "skb_ptr" has been flushed. */
+ __insn_mf();
+
+#ifdef TILE_NET_PARANOIA
+#if CHIP_HAS_CBOX_HOME_MAP()
+ if (hash_default) {
+ HV_PTE pte = *virt_to_pte(current->mm, (unsigned long)va);
+ if (hv_pte_get_mode(pte) != HV_PTE_MODE_CACHE_HASH_L3)
+ panic("Non-coherent ingress buffer!");
+ }
+#endif
+#endif
+
+ /* Provide the new buffer. */
+ tile_net_provide_linux_buffer(info, va, small);
+
+ return true;
+}
+
+
+/*
+ * Provide linux buffers for LIPP.
+ */
+static void tile_net_provide_needed_buffers(struct tile_net_cpu *info)
+{
+ while (info->num_needed_small_buffers != 0) {
+ if (!tile_net_provide_needed_buffer(info, true))
+ goto oops;
+ info->num_needed_small_buffers--;
+ }
+
+ while (info->num_needed_large_buffers != 0) {
+ if (!tile_net_provide_needed_buffer(info, false))
+ goto oops;
+ info->num_needed_large_buffers--;
+ }
+
+ return;
+
+oops:
+
+ /* Add a description to the page allocation failure dump. */
+ pr_notice("Could not provide a linux buffer to LIPP.\n");
+}
+
+
+/*
+ * Grab some LEPP completions, and store them in "comps", of size
+ * "comps_size", and return the number of completions which were
+ * stored, so the caller can free them.
+ *
+ * If "pending" is not NULL, it will be set to true if there might
+ * still be some pending completions caused by this tile, else false.
+ */
+static unsigned int tile_net_lepp_grab_comps(struct net_device *dev,
+ struct sk_buff *comps[],
+ unsigned int comps_size,
+ bool *pending)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ lepp_queue_t *eq = priv->epp_queue;
+
+ unsigned int n = 0;
+
+ unsigned int comp_head;
+ unsigned int comp_busy;
+ unsigned int comp_tail;
+
+ spin_lock(&priv->comp_lock);
+
+ comp_head = eq->comp_head;
+ comp_busy = eq->comp_busy;
+ comp_tail = eq->comp_tail;
+
+ while (comp_head != comp_busy && n < comps_size) {
+ comps[n++] = eq->comps[comp_head];
+ LEPP_QINC(comp_head);
+ }
+
+ if (pending != NULL)
+ *pending = (comp_head != comp_tail);
+
+ eq->comp_head = comp_head;
+
+ spin_unlock(&priv->comp_lock);
+
+ return n;
+}
+
+
+/*
+ * Make sure the egress timer is scheduled.
+ *
+ * Note that we use "schedule if not scheduled" logic instead of the more
+ * obvious "reschedule" logic, because "reschedule" is fairly expensive.
+ */
+static void tile_net_schedule_egress_timer(struct tile_net_cpu *info)
+{
+ if (!info->egress_timer_scheduled) {
+ mod_timer_pinned(&info->egress_timer, jiffies + 1);
+ info->egress_timer_scheduled = true;
+ }
+}
+
+
+/*
+ * The "function" for "info->egress_timer".
+ *
+ * This timer will reschedule itself as long as there are any pending
+ * completions expected (on behalf of any tile).
+ *
+ * ISSUE: Realistically, will the timer ever stop scheduling itself?
+ *
+ * ISSUE: This timer is almost never actually needed, so just use a global
+ * timer that can run on any tile.
+ *
+ * ISSUE: Maybe instead track number of expected completions, and free
+ * only that many, resetting to zero if "pending" is ever false.
+ */
+static void tile_net_handle_egress_timer(unsigned long arg)
+{
+ struct tile_net_cpu *info = (struct tile_net_cpu *)arg;
+ struct net_device *dev = info->napi.dev;
+
+ struct sk_buff *olds[32];
+ unsigned int wanted = 32;
+ unsigned int i, nolds = 0;
+ bool pending;
+
+ /* The timer is no longer scheduled. */
+ info->egress_timer_scheduled = false;
+
+ nolds = tile_net_lepp_grab_comps(dev, olds, wanted, &pending);
+
+ for (i = 0; i < nolds; i++)
+ kfree_skb(olds[i]);
+
+ /* Reschedule timer if needed. */
+ if (pending)
+ tile_net_schedule_egress_timer(info);
+}
+
+
+#ifdef IGNORE_DUP_ACKS
+
+/*
+ * Help detect "duplicate" ACKs. These are sequential packets (for a
+ * given flow) which are exactly 66 bytes long, sharing everything but
+ * ID=2@0x12, Hsum=2@0x18, Ack=4@0x2a, WinSize=2@0x30, Csum=2@0x32,
+ * Tstamps=10@0x38. The ID's are +1, the Hsum's are -1, the Ack's are
+ * +N, and the Tstamps are usually identical.
+ *
+ * NOTE: Apparently truly duplicate acks (with identical "ack" values),
+ * should not be collapsed, as they are used for some kind of flow control.
+ */
+static bool is_dup_ack(char *s1, char *s2, unsigned int len)
+{
+ int i;
+
+ unsigned long long ignorable = 0;
+
+ /* Identification. */
+ ignorable |= (1ULL << 0x12);
+ ignorable |= (1ULL << 0x13);
+
+ /* Header checksum. */
+ ignorable |= (1ULL << 0x18);
+ ignorable |= (1ULL << 0x19);
+
+ /* ACK. */
+ ignorable |= (1ULL << 0x2a);
+ ignorable |= (1ULL << 0x2b);
+ ignorable |= (1ULL << 0x2c);
+ ignorable |= (1ULL << 0x2d);
+
+ /* WinSize. */
+ ignorable |= (1ULL << 0x30);
+ ignorable |= (1ULL << 0x31);
+
+ /* Checksum. */
+ ignorable |= (1ULL << 0x32);
+ ignorable |= (1ULL << 0x33);
+
+ for (i = 0; i < len; i++, ignorable >>= 1) {
+
+ if ((ignorable & 1) || (s1[i] == s2[i]))
+ continue;
+
+#ifdef TILE_NET_DEBUG
+ /* HACK: Mention non-timestamp diffs. */
+ if (i < 0x38 && i != 0x2f &&
+ net_ratelimit())
+ pr_info("Diff at 0x%x\n", i);
+#endif
+
+ return false;
+ }
+
+#ifdef TILE_NET_NO_SUPPRESS_DUP_ACKS
+ /* HACK: Do not suppress truly duplicate ACKs. */
+ /* ISSUE: Is this actually necessary or helpful? */
+ if (s1[0x2a] == s2[0x2a] &&
+ s1[0x2b] == s2[0x2b] &&
+ s1[0x2c] == s2[0x2c] &&
+ s1[0x2d] == s2[0x2d]) {
+ return false;
+ }
+#endif
+
+ return true;
+}
+
+#endif
+
+
+
+/*
+ * Like "tile_net_handle_packets()", but just discard packets.
+ */
+static void tile_net_discard_packets(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+ struct tile_netio_queue *queue = &info->queue;
+ netio_queue_impl_t *qsp = queue->__system_part;
+ netio_queue_user_impl_t *qup = &queue->__user_part;
+
+ while (qup->__packet_receive_read !=
+ qsp->__packet_receive_queue.__packet_write) {
+
+ int index = qup->__packet_receive_read;
+
+ int index2_aux = index + sizeof(netio_pkt_t);
+ int index2 =
+ ((index2_aux ==
+ qsp->__packet_receive_queue.__last_packet_plus_one) ?
+ 0 : index2_aux);
+
+ netio_pkt_t *pkt = (netio_pkt_t *)
+ ((unsigned long) &qsp[1] + index);
+
+ /* Extract the "linux_buffer_t". */
+ unsigned int buffer = pkt->__packet.word;
+
+ /* Convert "linux_buffer_t" to "va". */
+ void *va = __va((phys_addr_t)(buffer >> 1) << 7);
+
+ /* Acquire the associated "skb". */
+ struct sk_buff **skb_ptr = va - sizeof(*skb_ptr);
+ struct sk_buff *skb = *skb_ptr;
+
+ kfree_skb(skb);
+
+ /* Consume this packet. */
+ qup->__packet_receive_read = index2;
+ }
+}
+
+
+/*
+ * Handle the next packet. Return true if "processed", false if "filtered".
+ */
+static bool tile_net_poll_aux(struct tile_net_cpu *info, int index)
+{
+ struct net_device *dev = info->napi.dev;
+
+ struct tile_netio_queue *queue = &info->queue;
+ netio_queue_impl_t *qsp = queue->__system_part;
+ netio_queue_user_impl_t *qup = &queue->__user_part;
+ struct tile_net_stats_t *stats = &info->stats;
+
+ int filter;
+
+ int index2_aux = index + sizeof(netio_pkt_t);
+ int index2 =
+ ((index2_aux ==
+ qsp->__packet_receive_queue.__last_packet_plus_one) ?
+ 0 : index2_aux);
+
+ netio_pkt_t *pkt = (netio_pkt_t *)((unsigned long) &qsp[1] + index);
+
+ netio_pkt_metadata_t *metadata = NETIO_PKT_METADATA(pkt);
+
+ /* Extract the packet size. */
+ unsigned long len =
+ (NETIO_PKT_CUSTOM_LENGTH(pkt) +
+ NET_IP_ALIGN - NETIO_PACKET_PADDING);
+
+ /* Extract the "linux_buffer_t". */
+ unsigned int buffer = pkt->__packet.word;
+
+ /* Extract "small" (vs "large"). */
+ bool small = ((buffer & 1) != 0);
+
+ /* Convert "linux_buffer_t" to "va". */
+ void *va = __va((phys_addr_t)(buffer >> 1) << 7);
+
+ /* Extract the packet data pointer. */
+ /* Compare to "NETIO_PKT_CUSTOM_DATA(pkt)". */
+ unsigned char *buf = va + NET_IP_ALIGN;
+
+#ifdef IGNORE_DUP_ACKS
+
+ static int other;
+ static int final;
+ static int keep;
+ static int skip;
+
+#endif
+
+ /* Invalidate the packet buffer. */
+ if (!hash_default)
+ __inv_buffer(buf, len);
+
+ /* ISSUE: Is this needed? */
+ dev->last_rx = jiffies;
+
+#ifdef TILE_NET_DUMP_PACKETS
+ dump_packet(buf, len, "rx");
+#endif /* TILE_NET_DUMP_PACKETS */
+
+#ifdef TILE_NET_VERIFY_INGRESS
+ if (!NETIO_PKT_L4_CSUM_CORRECT_M(metadata, pkt) &&
+ NETIO_PKT_L4_CSUM_CALCULATED_M(metadata, pkt)) {
+ /*
+ * FIXME: This complains about UDP packets
+ * with a "zero" checksum (bug 6624).
+ */
+#ifdef TILE_NET_PANIC_ON_BAD
+ dump_packet(buf, len, "rx");
+ panic("Bad L4 checksum.");
+#else
+ pr_warning("Bad L4 checksum on %d byte packet.\n", len);
+#endif
+ }
+ if (!NETIO_PKT_L3_CSUM_CORRECT_M(metadata, pkt) &&
+ NETIO_PKT_L3_CSUM_CALCULATED_M(metadata, pkt)) {
+ dump_packet(buf, len, "rx");
+ panic("Bad L3 checksum.");
+ }
+ switch (NETIO_PKT_STATUS_M(metadata, pkt)) {
+ case NETIO_PKT_STATUS_OVERSIZE:
+ if (len >= 64) {
+ dump_packet(buf, len, "rx");
+ panic("Unexpected OVERSIZE.");
+ }
+ break;
+ case NETIO_PKT_STATUS_BAD:
+#ifdef TILE_NET_PANIC_ON_BAD
+ dump_packet(buf, len, "rx");
+ panic("Unexpected BAD packet.");
+#else
+ pr_warning("Unexpected BAD %d byte packet.\n", len);
+#endif
+ }
+#endif
+
+ filter = 0;
+
+ if (!(dev->flags & IFF_UP)) {
+ /* Filter packets received before we're up. */
+ filter = 1;
+ } else if (!(dev->flags & IFF_PROMISC)) {
+ /*
+ * FIXME: Implement HW multicast filter.
+ */
+ if (!IS_MULTICAST(buf) && !IS_BROADCAST(buf)) {
+ /* Filter packets not for our address. */
+ const u8 *mine = dev->dev_addr;
+ filter = compare_ether_addr(mine, buf);
+ }
+ }
+
+#ifdef IGNORE_DUP_ACKS
+
+ if (len != 66) {
+ /* FIXME: Must check "is_tcp_ack(buf, len)" somehow. */
+
+ other++;
+
+ } else if (index2 ==
+ qsp->__packet_receive_queue.__packet_write) {
+
+ final++;
+
+ } else {
+
+ netio_pkt_t *pkt2 = (netio_pkt_t *)
+ ((unsigned long) &qsp[1] + index2);
+
+ netio_pkt_metadata_t *metadata2 =
+ NETIO_PKT_METADATA(pkt2);
+
+ /* Extract the packet size. */
+ unsigned long len2 =
+ (NETIO_PKT_CUSTOM_LENGTH(pkt2) +
+ NET_IP_ALIGN - NETIO_PACKET_PADDING);
+
+ if (len2 == 66 &&
+ NETIO_PKT_FLOW_HASH_M(metadata, pkt) ==
+ NETIO_PKT_FLOW_HASH_M(metadata2, pkt2)) {
+
+ /* Extract the "linux_buffer_t". */
+ unsigned int buffer2 = pkt2->__packet.word;
+
+ /* Convert "linux_buffer_t" to "va". */
+ void *va2 =
+ __va((phys_addr_t)(buffer2 >> 1) << 7);
+
+ /* Extract the packet data pointer. */
+ /* Compare to "NETIO_PKT_CUSTOM_DATA(pkt)". */
+ unsigned char *buf2 = va2 + NET_IP_ALIGN;
+
+ /* Invalidate the packet buffer. */
+ if (!hash_default)
+ __inv_buffer(buf2, len2);
+
+ if (is_dup_ack(buf, buf2, len)) {
+ skip++;
+ filter = 1;
+ } else {
+ keep++;
+ }
+ }
+ }
+
+ if (net_ratelimit())
+ pr_info("Other %d Final %d Keep %d Skip %d.\n",
+ other, final, keep, skip);
+
+#endif
+
+ if (filter) {
+
+ /* ISSUE: Update "drop" statistics? */
+
+ tile_net_provide_linux_buffer(info, va, small);
+
+ } else {
+
+ /* Acquire the associated "skb". */
+ struct sk_buff **skb_ptr = va - sizeof(*skb_ptr);
+ struct sk_buff *skb = *skb_ptr;
+
+ /* Paranoia. */
+ if (skb->data != buf)
+ panic("Corrupt linux buffer from LIPP! "
+ "VA=%p, skb=%p, skb->data=%p\n",
+ va, skb, skb->data);
+
+ /* Encode the actual packet length. */
+ skb_put(skb, len);
+
+ /* NOTE: This call also sets "skb->dev = dev". */
+ skb->protocol = eth_type_trans(skb, dev);
+
+ /* ISSUE: Discard corrupt packets? */
+ /* ISSUE: Discard packets with bad checksums? */
+
+ /* Avoid recomputing TCP/UDP checksums. */
+ if (NETIO_PKT_L4_CSUM_CORRECT_M(metadata, pkt))
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+ netif_receive_skb(skb);
+
+ stats->rx_packets++;
+ stats->rx_bytes += len;
+
+ if (small)
+ info->num_needed_small_buffers++;
+ else
+ info->num_needed_large_buffers++;
+ }
+
+ /* Return four credits after every fourth packet. */
+ if (--qup->__receive_credit_remaining == 0) {
+ u32 interval = qup->__receive_credit_interval;
+ qup->__receive_credit_remaining = interval;
+ __netio_fastio_return_credits(qup->__fastio_index, interval);
+ }
+
+ /* Consume this packet. */
+ qup->__packet_receive_read = index2;
+
+ return !filter;
+}
+
+
+/*
+ * Handle some packets for the given device on the current CPU.
+ *
+ * ISSUE: The "rotting packet" race condition occurs if a packet
+ * arrives after the queue appears to be empty, and before the
+ * hypervisor interrupt is re-enabled.
+ */
+static int tile_net_poll(struct napi_struct *napi, int budget)
+{
+ struct net_device *dev = napi->dev;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+ struct tile_netio_queue *queue = &info->queue;
+ netio_queue_impl_t *qsp = queue->__system_part;
+ netio_queue_user_impl_t *qup = &queue->__user_part;
+
+ unsigned int work = 0;
+
+ while (1) {
+ int index = qup->__packet_receive_read;
+ if (index == qsp->__packet_receive_queue.__packet_write)
+ break;
+
+ if (tile_net_poll_aux(info, index)) {
+ if (++work >= budget)
+ goto done;
+ }
+ }
+
+ napi_complete(&info->napi);
+
+ /* Re-enable hypervisor interrupts. */
+ enable_percpu_irq(priv->intr_id);
+
+ /* HACK: Avoid the "rotting packet" problem. */
+ if (qup->__packet_receive_read !=
+ qsp->__packet_receive_queue.__packet_write)
+ napi_schedule(&info->napi);
+
+ /* ISSUE: Handle completions? */
+
+done:
+
+ tile_net_provide_needed_buffers(info);
+
+ return work;
+}
+
+
+/*
+ * Handle an ingress interrupt for the given device on the current cpu.
+ */
+static irqreturn_t tile_net_handle_ingress_interrupt(int irq, void *dev_ptr)
+{
+ struct net_device *dev = (struct net_device *)dev_ptr;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+
+ /* Disable hypervisor interrupt. */
+ disable_percpu_irq(priv->intr_id);
+
+ napi_schedule(&info->napi);
+
+ return IRQ_HANDLED;
+}
+
+
+/*
+ * One time initialization per interface.
+ */
+static int tile_net_open_aux(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ int ret;
+ int dummy;
+ unsigned int epp_lotar;
+
+ /*
+ * Find out where EPP memory should be homed.
+ */
+ ret = hv_dev_pread(priv->hv_devhdl, 0,
+ (HV_VirtAddr)&epp_lotar, sizeof(epp_lotar),
+ NETIO_EPP_SHM_OFF);
+ if (ret < 0) {
+ pr_err("could not read epp_shm_queue lotar.\n");
+ return -EIO;
+ }
+
+ /*
+ * Home the page on the EPP.
+ */
+ {
+ int epp_home = hv_lotar_to_cpu(epp_lotar);
+ struct page *page = virt_to_page(priv->epp_queue);
+ homecache_change_page_home(page, 0, epp_home);
+ }
+
+ /*
+ * Register the EPP shared memory queue.
+ */
+ {
+ netio_ipp_address_t ea = {
+ .va = 0,
+ .pa = __pa(priv->epp_queue),
+ .pte = hv_pte(0),
+ .size = PAGE_SIZE,
+ };
+ ea.pte = hv_pte_set_lotar(ea.pte, epp_lotar);
+ ea.pte = hv_pte_set_mode(ea.pte, HV_PTE_MODE_CACHE_TILE_L3);
+ ret = hv_dev_pwrite(priv->hv_devhdl, 0,
+ (HV_VirtAddr)&ea,
+ sizeof(ea),
+ NETIO_EPP_SHM_OFF);
+ if (ret < 0)
+ return -EIO;
+ }
+
+ /*
+ * Start LIPP/LEPP.
+ */
+ if (hv_dev_pwrite(priv->hv_devhdl, 0, (HV_VirtAddr)&dummy,
+ sizeof(dummy), NETIO_IPP_START_SHIM_OFF) < 0) {
+ pr_warning("Failed to start LIPP/LEPP.\n");
+ return -EIO;
+ }
+
+ return 0;
+}
+
+
+/*
+ * Register with hypervisor on each CPU.
+ *
+ * Strangely, this function does important things even if it "fails",
+ * which is especially common if the link is not up yet. Hopefully
+ * these things are all "harmless" if done twice!
+ */
+static void tile_net_register(void *dev_ptr)
+{
+ struct net_device *dev = (struct net_device *)dev_ptr;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info;
+
+ struct tile_netio_queue *queue;
+
+ /* Only network cpus can receive packets. */
+ int queue_id =
+ cpumask_test_cpu(my_cpu, &priv->network_cpus_map) ? 0 : 255;
+
+ netio_input_config_t config = {
+ .flags = 0,
+ .num_receive_packets = priv->network_cpus_credits,
+ .queue_id = queue_id
+ };
+
+ int ret = 0;
+ netio_queue_impl_t *queuep;
+
+ PDEBUG("tile_net_register(queue_id %d)\n", queue_id);
+
+ if (!strcmp(dev->name, "xgbe0"))
+ info = &__get_cpu_var(hv_xgbe0);
+ else if (!strcmp(dev->name, "xgbe1"))
+ info = &__get_cpu_var(hv_xgbe1);
+ else if (!strcmp(dev->name, "gbe0"))
+ info = &__get_cpu_var(hv_gbe0);
+ else if (!strcmp(dev->name, "gbe1"))
+ info = &__get_cpu_var(hv_gbe1);
+ else
+ BUG();
+
+ /* Initialize the egress timer. */
+ init_timer(&info->egress_timer);
+ info->egress_timer.data = (long)info;
+ info->egress_timer.function = tile_net_handle_egress_timer;
+
+ priv->cpu[my_cpu] = info;
+
+ /*
+ * Register ourselves with the IPP.
+ */
+ ret = hv_dev_pwrite(priv->hv_devhdl, 0,
+ (HV_VirtAddr)&config,
+ sizeof(netio_input_config_t),
+ NETIO_IPP_INPUT_REGISTER_OFF);
+ PDEBUG("hv_dev_pwrite(NETIO_IPP_INPUT_REGISTER_OFF) returned %d\n",
+ ret);
+ if (ret < 0) {
+ printk(KERN_DEBUG "hv_dev_pwrite NETIO_IPP_INPUT_REGISTER_OFF"
+ " failure %d\n", ret);
+ info->link_down = (ret == NETIO_LINK_DOWN);
+ return;
+ }
+
+ /*
+ * Get the pointer to our queue's system part.
+ */
+
+ ret = hv_dev_pread(priv->hv_devhdl, 0,
+ (HV_VirtAddr)&queuep,
+ sizeof(netio_queue_impl_t *),
+ NETIO_IPP_INPUT_REGISTER_OFF);
+ PDEBUG("hv_dev_pread(NETIO_IPP_INPUT_REGISTER_OFF) returned %d\n",
+ ret);
+ PDEBUG("queuep %p\n", queuep);
+ if (ret <= 0) {
+ /* ISSUE: Shouldn't this be a fatal error? */
+ pr_err("hv_dev_pread NETIO_IPP_INPUT_REGISTER_OFF failure\n");
+ return;
+ }
+
+ queue = &info->queue;
+
+ queue->__system_part = queuep;
+
+ memset(&queue->__user_part, 0, sizeof(netio_queue_user_impl_t));
+
+ /* This is traditionally "config.num_receive_packets / 2". */
+ queue->__user_part.__receive_credit_interval = 4;
+ queue->__user_part.__receive_credit_remaining =
+ queue->__user_part.__receive_credit_interval;
+
+ /*
+ * Get a fastio index from the hypervisor.
+ * ISSUE: Shouldn't this check the result?
+ */
+ ret = hv_dev_pread(priv->hv_devhdl, 0,
+ (HV_VirtAddr)&queue->__user_part.__fastio_index,
+ sizeof(queue->__user_part.__fastio_index),
+ NETIO_IPP_GET_FASTIO_OFF);
+ PDEBUG("hv_dev_pread(NETIO_IPP_GET_FASTIO_OFF) returned %d\n", ret);
+
+ netif_napi_add(dev, &info->napi, tile_net_poll, 64);
+
+ /* Now we are registered. */
+ info->registered = true;
+}
+
+
+/*
+ * Unregister with hypervisor on each CPU.
+ */
+static void tile_net_unregister(void *dev_ptr)
+{
+ struct net_device *dev = (struct net_device *)dev_ptr;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+
+ int ret = 0;
+ int dummy = 0;
+
+ /* Do nothing if never registered. */
+ if (info == NULL)
+ return;
+
+ /* Do nothing if already unregistered. */
+ if (!info->registered)
+ return;
+
+ /*
+ * Unregister ourselves with LIPP.
+ */
+ ret = hv_dev_pwrite(priv->hv_devhdl, 0, (HV_VirtAddr)&dummy,
+ sizeof(dummy), NETIO_IPP_INPUT_UNREGISTER_OFF);
+ PDEBUG("hv_dev_pwrite(NETIO_IPP_INPUT_UNREGISTER_OFF) returned %d\n",
+ ret);
+ if (ret < 0) {
+ /* FIXME: Just panic? */
+ pr_err("hv_dev_pwrite NETIO_IPP_INPUT_UNREGISTER_OFF"
+ " failure %d\n", ret);
+ }
+
+ /*
+ * Discard all packets still in our NetIO queue. Hopefully,
+ * once the unregister call is complete, there will be no
+ * packets still in flight on the IDN.
+ */
+ tile_net_discard_packets(dev);
+
+ /* Reset state. */
+ info->num_needed_small_buffers = 0;
+ info->num_needed_large_buffers = 0;
+
+ /* Cancel egress timer. */
+ del_timer(&info->egress_timer);
+ info->egress_timer_scheduled = false;
+
+ netif_napi_del(&info->napi);
+
+ /* Now we are unregistered. */
+ info->registered = false;
+}
+
+
+/*
+ * Helper function for "tile_net_stop()".
+ *
+ * Also used to handle registration failure in "tile_net_open_inner()",
+ * when "fully_opened" is known to be false, and the various extra
+ * steps in "tile_net_stop()" are not necessary. ISSUE: It might be
+ * simpler if we could just call "tile_net_stop()" anyway.
+ */
+static void tile_net_stop_aux(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ int dummy = 0;
+
+ /* Unregister all tiles, so LIPP will stop delivering packets. */
+ on_each_cpu(tile_net_unregister, (void *)dev, 1);
+
+ /* Stop LIPP/LEPP. */
+ if (hv_dev_pwrite(priv->hv_devhdl, 0, (HV_VirtAddr)&dummy,
+ sizeof(dummy), NETIO_IPP_STOP_SHIM_OFF) < 0)
+ panic("Failed to stop LIPP/LEPP!\n");
+
+ priv->partly_opened = 0;
+}
+
+
+/*
+ * Disable ingress interrupts for the given device on the current cpu.
+ */
+static void tile_net_disable_intr(void *dev_ptr)
+{
+ struct net_device *dev = (struct net_device *)dev_ptr;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+
+ /* Disable hypervisor interrupt. */
+ disable_percpu_irq(priv->intr_id);
+
+ /* Disable NAPI if needed. */
+ if (info != NULL && info->napi_enabled) {
+ napi_disable(&info->napi);
+ info->napi_enabled = false;
+ }
+}
+
+
+/*
+ * Enable ingress interrupts for the given device on the current cpu.
+ */
+static void tile_net_enable_intr(void *dev_ptr)
+{
+ struct net_device *dev = (struct net_device *)dev_ptr;
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+
+ /* Enable hypervisor interrupt. */
+ enable_percpu_irq(priv->intr_id);
+
+ /* Enable NAPI. */
+ napi_enable(&info->napi);
+ info->napi_enabled = true;
+}
+
+
+/*
+ * tile_net_open_inner does most of the work of bringing up the interface.
+ * It's called from tile_net_open(), and also from tile_net_retry_open().
+ * The return value is 0 if the interface was brought up, < 0 if
+ * tile_net_open() should return the return value as an error, and > 0 if
+ * tile_net_open() should return success and schedule a work item to
+ * periodically retry the bringup.
+ */
+static int tile_net_open_inner(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info;
+ struct tile_netio_queue *queue;
+ unsigned int irq;
+ int i;
+
+ /*
+ * First try to register just on the local CPU, and handle any
+ * semi-expected "link down" failure specially. Note that we
+ * do NOT call "tile_net_stop_aux()", unlike below.
+ */
+ tile_net_register(dev);
+ info = priv->cpu[my_cpu];
+ if (!info->registered) {
+ if (info->link_down)
+ return 1;
+ return -EAGAIN;
+ }
+
+ /*
+ * Now register everywhere else. If any registration fails,
+ * even for "link down" (which might not be possible), we
+ * clean up using "tile_net_stop_aux()".
+ */
+ smp_call_function(tile_net_register, (void *)dev, 1);
+ for_each_online_cpu(i) {
+ if (!priv->cpu[i]->registered) {
+ tile_net_stop_aux(dev);
+ return -EAGAIN;
+ }
+ }
+
+ queue = &info->queue;
+
+ /*
+ * Set the device intr bit mask.
+ * The tile_net_register above sets per tile __intr_id.
+ */
+ priv->intr_id = queue->__system_part->__intr_id;
+ BUG_ON(!priv->intr_id);
+
+ /*
+ * Register the device interrupt handler.
+ * The __ffs() function returns the index into the interrupt handler
+ * table from the interrupt bit mask which should have one bit
+ * and one bit only set.
+ */
+ irq = __ffs(priv->intr_id);
+ tile_irq_activate(irq, TILE_IRQ_PERCPU);
+ BUG_ON(request_irq(irq, tile_net_handle_ingress_interrupt,
+ 0, dev->name, (void *)dev) != 0);
+
+ /* ISSUE: How could "priv->fully_opened" ever be "true" here? */
+
+ if (!priv->fully_opened) {
+
+ int dummy = 0;
+
+ /* Allocate initial buffers. */
+
+ int max_buffers =
+ priv->network_cpus_count * priv->network_cpus_credits;
+
+ info->num_needed_small_buffers =
+ min(LIPP_SMALL_BUFFERS, max_buffers);
+
+ info->num_needed_large_buffers =
+ min(LIPP_LARGE_BUFFERS, max_buffers);
+
+ tile_net_provide_needed_buffers(info);
+
+ if (info->num_needed_small_buffers != 0 ||
+ info->num_needed_large_buffers != 0)
+ panic("Insufficient memory for buffer stack!");
+
+ /* Start LIPP/LEPP and activate "ingress" at the shim. */
+ if (hv_dev_pwrite(priv->hv_devhdl, 0, (HV_VirtAddr)&dummy,
+ sizeof(dummy), NETIO_IPP_INPUT_INIT_OFF) < 0)
+ panic("Failed to activate the LIPP Shim!\n");
+
+ priv->fully_opened = 1;
+ }
+
+ /* On each tile, enable the hypervisor to trigger interrupts. */
+ /* ISSUE: Do this before starting LIPP/LEPP? */
+ on_each_cpu(tile_net_enable_intr, (void *)dev, 1);
+
+ /* Start our transmit queue. */
+ netif_start_queue(dev);
+
+ return 0;
+}
+
+
+/*
+ * Called periodically to retry bringing up the NetIO interface,
+ * if it doesn't come up cleanly during tile_net_open().
+ */
+static void tile_net_open_retry(struct work_struct *w)
+{
+ struct delayed_work *dw =
+ container_of(w, struct delayed_work, work);
+
+ struct tile_net_priv *priv =
+ container_of(dw, struct tile_net_priv, retry_work);
+
+ /*
+ * Try to bring the NetIO interface up. If it fails, reschedule
+ * ourselves to try again later; otherwise, tell Linux we now have
+ * a working link. ISSUE: What if the return value is negative?
+ */
+ if (tile_net_open_inner(priv->dev))
+ schedule_delayed_work_on(singlethread_cpu, &priv->retry_work,
+ TILE_NET_RETRY_INTERVAL);
+ else
+ netif_carrier_on(priv->dev);
+}
+
+
+/*
+ * Called when a network interface is made active.
+ *
+ * Returns 0 on success, negative value on failure.
+ *
+ * The open entry point is called when a network interface is made
+ * active by the system (IFF_UP). At this point all resources needed
+ * for transmit and receive operations are allocated, the interrupt
+ * handler is registered with the OS, the watchdog timer is started,
+ * and the stack is notified that the interface is ready.
+ *
+ * If the actual link is not available yet, then we tell Linux that
+ * we have no carrier, and we keep checking until the link comes up.
+ */
+static int tile_net_open(struct net_device *dev)
+{
+ int ret = 0;
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ /*
+ * We rely on priv->partly_opened to tell us if this is the
+ * first time this interface is being brought up. If it is
+ * set, the IPP was already initialized and should not be
+ * initialized again.
+ */
+ if (!priv->partly_opened) {
+
+ int count;
+ int credits;
+
+ /* Initialize LIPP/LEPP, and start the Shim. */
+ ret = tile_net_open_aux(dev);
+ if (ret < 0) {
+ pr_err("tile_net_open_aux failed: %d\n", ret);
+ return ret;
+ }
+
+ /* Analyze the network cpus. */
+
+ if (network_cpus_used)
+ cpumask_copy(&priv->network_cpus_map,
+ &network_cpus_map);
+ else
+ cpumask_copy(&priv->network_cpus_map, cpu_online_mask);
+
+
+ count = cpumask_weight(&priv->network_cpus_map);
+
+ /* Limit credits to available buffers, and apply min. */
+ credits = max(16, (LIPP_LARGE_BUFFERS / count) & ~1);
+
+ /* Apply "GBE" max limit. */
+ /* ISSUE: Use higher limit for XGBE? */
+ credits = min(NETIO_MAX_RECEIVE_PKTS, credits);
+
+ priv->network_cpus_count = count;
+ priv->network_cpus_credits = credits;
+
+#ifdef TILE_NET_DEBUG
+ pr_info("Using %d network cpus, with %d credits each\n",
+ priv->network_cpus_count, priv->network_cpus_credits);
+#endif
+
+ priv->partly_opened = 1;
+ }
+
+ /*
+ * Attempt to bring up the link.
+ */
+ ret = tile_net_open_inner(dev);
+ if (ret <= 0) {
+ if (ret == 0)
+ netif_carrier_on(dev);
+ return ret;
+ }
+
+ /*
+ * We were unable to bring up the NetIO interface, but we want to
+ * try again in a little bit. Tell Linux that we have no carrier
+ * so it doesn't try to use the interface before the link comes up
+ * and then remember to try again later.
+ */
+ netif_carrier_off(dev);
+ schedule_delayed_work_on(singlethread_cpu, &priv->retry_work,
+ TILE_NET_RETRY_INTERVAL);
+
+ return 0;
+}
+
+
+/*
+ * Disables a network interface.
+ *
+ * Returns 0, this is not allowed to fail.
+ *
+ * The close entry point is called when an interface is de-activated
+ * by the OS. The hardware is still under the drivers control, but
+ * needs to be disabled. A global MAC reset is issued to stop the
+ * hardware, and all transmit and receive resources are freed.
+ *
+ * ISSUE: Can this can be called while "tile_net_poll()" is running?
+ */
+static int tile_net_stop(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ bool pending = true;
+
+ PDEBUG("tile_net_stop()\n");
+
+ /* ISSUE: Only needed if not yet fully open. */
+ cancel_delayed_work_sync(&priv->retry_work);
+
+ /* Can't transmit any more. */
+ netif_stop_queue(dev);
+
+ /*
+ * Disable hypervisor interrupts on each tile.
+ */
+ on_each_cpu(tile_net_disable_intr, (void *)dev, 1);
+
+ /*
+ * Unregister the interrupt handler.
+ * The __ffs() function returns the index into the interrupt handler
+ * table from the interrupt bit mask which should have one bit
+ * and one bit only set.
+ */
+ if (priv->intr_id)
+ free_irq(__ffs(priv->intr_id), dev);
+
+ /*
+ * Drain all the LIPP buffers.
+ */
+
+ while (true) {
+ int buffer;
+
+ /* NOTE: This should never fail. */
+ if (hv_dev_pread(priv->hv_devhdl, 0, (HV_VirtAddr)&buffer,
+ sizeof(buffer), NETIO_IPP_DRAIN_OFF) < 0)
+ break;
+
+ /* Stop when done. */
+ if (buffer == 0)
+ break;
+
+ {
+ /* Convert "linux_buffer_t" to "va". */
+ void *va = __va((phys_addr_t)(buffer >> 1) << 7);
+
+ /* Acquire the associated "skb". */
+ struct sk_buff **skb_ptr = va - sizeof(*skb_ptr);
+ struct sk_buff *skb = *skb_ptr;
+
+ kfree_skb(skb);
+ }
+ }
+
+ /* Stop LIPP/LEPP. */
+ tile_net_stop_aux(dev);
+
+
+ priv->fully_opened = 0;
+
+
+ /*
+ * XXX: ISSUE: It appears that, in practice anyway, by the
+ * time we get here, there are no pending completions.
+ */
+ while (pending) {
+
+ struct sk_buff *olds[32];
+ unsigned int wanted = 32;
+ unsigned int i, nolds = 0;
+
+ nolds = tile_net_lepp_grab_comps(dev, olds,
+ wanted, &pending);
+
+ /* ISSUE: We have never actually seen this debug spew. */
+ if (nolds != 0)
+ pr_info("During tile_net_stop(), grabbed %d comps.\n",
+ nolds);
+
+ for (i = 0; i < nolds; i++)
+ kfree_skb(olds[i]);
+ }
+
+
+ /* Wipe the EPP queue. */
+ memset(priv->epp_queue, 0, sizeof(lepp_queue_t));
+
+ /* Evict the EPP queue. */
+ finv_buffer(priv->epp_queue, PAGE_SIZE);
+
+ return 0;
+}
+
+
+/*
+ * Prepare the "frags" info for the resulting LEPP command.
+ *
+ * If needed, flush the memory used by the frags.
+ */
+static unsigned int tile_net_tx_frags(lepp_frag_t *frags,
+ struct sk_buff *skb,
+ void *b_data, unsigned int b_len)
+{
+ unsigned int i, n = 0;
+
+ struct skb_shared_info *sh = skb_shinfo(skb);
+
+ phys_addr_t cpa;
+
+ if (b_len != 0) {
+
+ if (!hash_default)
+ finv_buffer_remote(b_data, b_len);
+
+ cpa = __pa(b_data);
+ frags[n].cpa_lo = cpa;
+ frags[n].cpa_hi = cpa >> 32;
+ frags[n].length = b_len;
+ frags[n].hash_for_home = hash_default;
+ n++;
+ }
+
+ for (i = 0; i < sh->nr_frags; i++) {
+
+ skb_frag_t *f = &sh->frags[i];
+ unsigned long pfn = page_to_pfn(f->page);
+
+ /* FIXME: Compute "hash_for_home" properly. */
+ /* ISSUE: The hypervisor checks CHIP_HAS_REV1_DMA_PACKETS(). */
+ int hash_for_home = hash_default;
+
+ /* FIXME: Hmmm. */
+ if (!hash_default) {
+ void *va = pfn_to_kaddr(pfn) + f->page_offset;
+ BUG_ON(PageHighMem(f->page));
+ finv_buffer_remote(va, f->size);
+ }
+
+ cpa = ((phys_addr_t)pfn << PAGE_SHIFT) + f->page_offset;
+ frags[n].cpa_lo = cpa;
+ frags[n].cpa_hi = cpa >> 32;
+ frags[n].length = f->size;
+ frags[n].hash_for_home = hash_for_home;
+ n++;
+ }
+
+ return n;
+}
+
+
+/*
+ * This function takes "skb", consisting of a header template and a
+ * payload, and hands it to LEPP, to emit as one or more segments,
+ * each consisting of a possibly modified header, plus a piece of the
+ * payload, via a process known as "tcp segmentation offload".
+ *
+ * Usually, "data" will contain the header template, of size "sh_len",
+ * and "sh->frags" will contain "skb->data_len" bytes of payload, and
+ * there will be "sh->gso_segs" segments.
+ *
+ * Sometimes, if "sendfile()" requires copying, we will be called with
+ * "data" containing the header and payload, with "frags" being empty.
+ *
+ * In theory, "sh->nr_frags" could be 3, but in practice, it seems
+ * that this will never actually happen.
+ *
+ * See "emulate_large_send_offload()" for some reference code, which
+ * does not handle checksumming.
+ *
+ * ISSUE: How do we make sure that high memory DMA does not migrate?
+ */
+static int tile_net_tx_tso(struct sk_buff *skb, struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+ struct tile_net_stats_t *stats = &info->stats;
+
+ struct skb_shared_info *sh = skb_shinfo(skb);
+
+ unsigned char *data = skb->data;
+
+ /* The ip header follows the ethernet header. */
+ struct iphdr *ih = ip_hdr(skb);
+ unsigned int ih_len = ih->ihl * 4;
+
+ /* Note that "nh == ih", by definition. */
+ unsigned char *nh = skb_network_header(skb);
+ unsigned int eh_len = nh - data;
+
+ /* The tcp header follows the ip header. */
+ struct tcphdr *th = (struct tcphdr *)(nh + ih_len);
+ unsigned int th_len = th->doff * 4;
+
+ /* The total number of header bytes. */
+ /* NOTE: This may be less than skb_headlen(skb). */
+ unsigned int sh_len = eh_len + ih_len + th_len;
+
+ /* The number of payload bytes at "skb->data + sh_len". */
+ /* This is non-zero for sendfile() without HIGHDMA. */
+ unsigned int b_len = skb_headlen(skb) - sh_len;
+
+ /* The total number of payload bytes. */
+ unsigned int d_len = b_len + skb->data_len;
+
+ /* The maximum payload size. */
+ unsigned int p_len = sh->gso_size;
+
+ /* The total number of segments. */
+ unsigned int num_segs = sh->gso_segs;
+
+ /* The temporary copy of the command. */
+ u32 cmd_body[(LEPP_MAX_CMD_SIZE + 3) / 4];
+ lepp_tso_cmd_t *cmd = (lepp_tso_cmd_t *)cmd_body;
+
+ /* Analyze the "frags". */
+ unsigned int num_frags =
+ tile_net_tx_frags(cmd->frags, skb, data + sh_len, b_len);
+
+ /* The size of the command, including frags and header. */
+ size_t cmd_size = LEPP_TSO_CMD_SIZE(num_frags, sh_len);
+
+ /* The command header. */
+ lepp_tso_cmd_t cmd_init = {
+ .tso = true,
+ .header_size = sh_len,
+ .ip_offset = eh_len,
+ .tcp_offset = eh_len + ih_len,
+ .payload_size = p_len,
+ .num_frags = num_frags,
+ };
+
+ unsigned long irqflags;
+
+ lepp_queue_t *eq = priv->epp_queue;
+
+ struct sk_buff *olds[4];
+ unsigned int wanted = 4;
+ unsigned int i, nolds = 0;
+
+ unsigned int cmd_head, cmd_tail, cmd_next;
+ unsigned int comp_tail;
+
+ unsigned int free_slots;
+
+
+ /* Paranoia. */
+ BUG_ON(skb->protocol != htons(ETH_P_IP));
+ BUG_ON(ih->protocol != IPPROTO_TCP);
+ BUG_ON(skb->ip_summed != CHECKSUM_PARTIAL);
+ BUG_ON(num_frags > LEPP_MAX_FRAGS);
+ /*--BUG_ON(num_segs != (d_len + (p_len - 1)) / p_len); */
+ BUG_ON(num_segs <= 1);
+
+
+ /* Finish preparing the command. */
+
+ /* Copy the command header. */
+ *cmd = cmd_init;
+
+ /* Copy the "header". */
+ memcpy(&cmd->frags[num_frags], data, sh_len);
+
+
+ /* Prefetch and wait, to minimize time spent holding the spinlock. */
+ prefetch_L1(&eq->comp_tail);
+ prefetch_L1(&eq->cmd_tail);
+ mb();
+
+
+ /* Enqueue the command. */
+
+ spin_lock_irqsave(&priv->cmd_lock, irqflags);
+
+ /*
+ * Handle completions if needed to make room.
+ * HACK: Spin until there is sufficient room.
+ */
+ free_slots = lepp_num_free_comp_slots(eq);
+ if (free_slots < 1) {
+spin:
+ nolds += tile_net_lepp_grab_comps(dev, olds + nolds,
+ wanted - nolds, NULL);
+ if (lepp_num_free_comp_slots(eq) < 1)
+ goto spin;
+ }
+
+ cmd_head = eq->cmd_head;
+ cmd_tail = eq->cmd_tail;
+
+ /* NOTE: The "gotos" below are untested. */
+
+ /* Prepare to advance, detecting full queue. */
+ cmd_next = cmd_tail + cmd_size;
+ if (cmd_tail < cmd_head && cmd_next >= cmd_head)
+ goto spin;
+ if (cmd_next > LEPP_CMD_LIMIT) {
+ cmd_next = 0;
+ if (cmd_next == cmd_head)
+ goto spin;
+ }
+
+ /* Copy the command. */
+ memcpy(&eq->cmds[cmd_tail], cmd, cmd_size);
+
+ /* Advance. */
+ cmd_tail = cmd_next;
+
+ /* Record "skb" for eventual freeing. */
+ comp_tail = eq->comp_tail;
+ eq->comps[comp_tail] = skb;
+ LEPP_QINC(comp_tail);
+ eq->comp_tail = comp_tail;
+
+ /* Flush before allowing LEPP to handle the command. */
+ __insn_mf();
+
+ eq->cmd_tail = cmd_tail;
+
+ spin_unlock_irqrestore(&priv->cmd_lock, irqflags);
+
+ if (nolds == 0)
+ nolds = tile_net_lepp_grab_comps(dev, olds, wanted, NULL);
+
+ /* Handle completions. */
+ for (i = 0; i < nolds; i++)
+ kfree_skb(olds[i]);
+
+ /* Update stats. */
+ stats->tx_packets += num_segs;
+ stats->tx_bytes += (num_segs * sh_len) + d_len;
+
+ /* Make sure the egress timer is scheduled. */
+ tile_net_schedule_egress_timer(info);
+
+ return NETDEV_TX_OK;
+}
+
+
+/*
+ * Transmit a packet (called by the kernel via "hard_start_xmit" hook).
+ */
+static int tile_net_tx(struct sk_buff *skb, struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+ int my_cpu = smp_processor_id();
+ struct tile_net_cpu *info = priv->cpu[my_cpu];
+ struct tile_net_stats_t *stats = &info->stats;
+
+ unsigned long irqflags;
+
+ struct skb_shared_info *sh = skb_shinfo(skb);
+
+ unsigned int len = skb->len;
+ unsigned char *data = skb->data;
+
+ unsigned int csum_start = skb->csum_start - skb_headroom(skb);
+
+ lepp_frag_t frags[LEPP_MAX_FRAGS];
+
+ unsigned int num_frags;
+
+ lepp_queue_t *eq = priv->epp_queue;
+
+ struct sk_buff *olds[4];
+ unsigned int wanted = 4;
+ unsigned int i, nolds = 0;
+
+ unsigned int cmd_size = sizeof(lepp_cmd_t);
+
+ unsigned int cmd_head, cmd_tail, cmd_next;
+ unsigned int comp_tail;
+
+ lepp_cmd_t cmds[LEPP_MAX_FRAGS];
+
+ unsigned int free_slots;
+
+
+ /*
+ * This is paranoia, since we think that if the link doesn't come
+ * up, telling Linux we have no carrier will keep it from trying
+ * to transmit. If it does, though, we can't execute this routine,
+ * since data structures we depend on aren't set up yet.
+ */
+ if (!info->registered)
+ return NETDEV_TX_BUSY;
+
+
+ /* Save the timestamp. */
+ dev->trans_start = jiffies;
+
+
+#ifdef TILE_NET_PARANOIA
+#if CHIP_HAS_CBOX_HOME_MAP()
+ if (hash_default) {
+ HV_PTE pte = *virt_to_pte(current->mm, (unsigned long)data);
+ if (hv_pte_get_mode(pte) != HV_PTE_MODE_CACHE_HASH_L3)
+ panic("Non-coherent egress buffer!");
+ }
+#endif
+#endif
+
+
+#ifdef TILE_NET_DUMP_PACKETS
+ /* ISSUE: Does not dump the "frags". */
+ dump_packet(data, skb_headlen(skb), "tx");
+#endif /* TILE_NET_DUMP_PACKETS */
+
+
+ if (sh->gso_size != 0)
+ return tile_net_tx_tso(skb, dev);
+
+
+ /* Prepare the commands. */
+
+ num_frags = tile_net_tx_frags(frags, skb, data, skb_headlen(skb));
+
+ for (i = 0; i < num_frags; i++) {
+
+ bool final = (i == num_frags - 1);
+
+ lepp_cmd_t cmd = {
+ .cpa_lo = frags[i].cpa_lo,
+ .cpa_hi = frags[i].cpa_hi,
+ .length = frags[i].length,
+ .hash_for_home = frags[i].hash_for_home,
+ .send_completion = final,
+ .end_of_packet = final
+ };
+
+ if (i == 0 && skb->ip_summed == CHECKSUM_PARTIAL) {
+ cmd.compute_checksum = 1;
+ cmd.checksum_data.bits.start_byte = csum_start;
+ cmd.checksum_data.bits.count = len - csum_start;
+ cmd.checksum_data.bits.destination_byte =
+ csum_start + skb->csum_offset;
+ }
+
+ cmds[i] = cmd;
+ }
+
+
+ /* Prefetch and wait, to minimize time spent holding the spinlock. */
+ prefetch_L1(&eq->comp_tail);
+ prefetch_L1(&eq->cmd_tail);
+ mb();
+
+
+ /* Enqueue the commands. */
+
+ spin_lock_irqsave(&priv->cmd_lock, irqflags);
+
+ /*
+ * Handle completions if needed to make room.
+ * HACK: Spin until there is sufficient room.
+ */
+ free_slots = lepp_num_free_comp_slots(eq);
+ if (free_slots < 1) {
+spin:
+ nolds += tile_net_lepp_grab_comps(dev, olds + nolds,
+ wanted - nolds, NULL);
+ if (lepp_num_free_comp_slots(eq) < 1)
+ goto spin;
+ }
+
+ cmd_head = eq->cmd_head;
+ cmd_tail = eq->cmd_tail;
+
+ /* NOTE: The "gotos" below are untested. */
+
+ /* Copy the commands, or fail. */
+ for (i = 0; i < num_frags; i++) {
+
+ /* Prepare to advance, detecting full queue. */
+ cmd_next = cmd_tail + cmd_size;
+ if (cmd_tail < cmd_head && cmd_next >= cmd_head)
+ goto spin;
+ if (cmd_next > LEPP_CMD_LIMIT) {
+ cmd_next = 0;
+ if (cmd_next == cmd_head)
+ goto spin;
+ }
+
+ /* Copy the command. */
+ *(lepp_cmd_t *)&eq->cmds[cmd_tail] = cmds[i];
+
+ /* Advance. */
+ cmd_tail = cmd_next;
+ }
+
+ /* Record "skb" for eventual freeing. */
+ comp_tail = eq->comp_tail;
+ eq->comps[comp_tail] = skb;
+ LEPP_QINC(comp_tail);
+ eq->comp_tail = comp_tail;
+
+ /* Flush before allowing LEPP to handle the command. */
+ __insn_mf();
+
+ eq->cmd_tail = cmd_tail;
+
+ spin_unlock_irqrestore(&priv->cmd_lock, irqflags);
+
+ if (nolds == 0)
+ nolds = tile_net_lepp_grab_comps(dev, olds, wanted, NULL);
+
+ /* Handle completions. */
+ for (i = 0; i < nolds; i++)
+ kfree_skb(olds[i]);
+
+ /* HACK: Track "expanded" size for short packets (e.g. 42 < 60). */
+ stats->tx_packets++;
+ stats->tx_bytes += ((len >= ETH_ZLEN) ? len : ETH_ZLEN);
+
+ /* Make sure the egress timer is scheduled. */
+ tile_net_schedule_egress_timer(info);
+
+ return NETDEV_TX_OK;
+}
+
+
+/*
+ * Deal with a transmit timeout.
+ */
+static void tile_net_tx_timeout(struct net_device *dev)
+{
+ PDEBUG("tile_net_tx_timeout()\n");
+ PDEBUG("Transmit timeout at %ld, latency %ld\n", jiffies,
+ jiffies - dev->trans_start);
+
+ /* XXX: ISSUE: This doesn't seem useful for us. */
+ netif_wake_queue(dev);
+}
+
+
+/*
+ * Ioctl commands.
+ */
+static int tile_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+{
+ return -EOPNOTSUPP;
+}
+
+
+/*
+ * Get System Network Statistics.
+ *
+ * Returns the address of the device statistics structure.
+ */
+static struct net_device_stats *tile_net_get_stats(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+ u32 rx_packets = 0;
+ u32 tx_packets = 0;
+ u32 rx_bytes = 0;
+ u32 tx_bytes = 0;
+ int i;
+
+ for_each_online_cpu(i) {
+ if (priv->cpu[i]) {
+ rx_packets += priv->cpu[i]->stats.rx_packets;
+ rx_bytes += priv->cpu[i]->stats.rx_bytes;
+ tx_packets += priv->cpu[i]->stats.tx_packets;
+ tx_bytes += priv->cpu[i]->stats.tx_bytes;
+ }
+ }
+
+ priv->stats.rx_packets = rx_packets;
+ priv->stats.rx_bytes = rx_bytes;
+ priv->stats.tx_packets = tx_packets;
+ priv->stats.tx_bytes = tx_bytes;
+
+ return &priv->stats;
+}
+
+
+/*
+ * Change the "mtu".
+ *
+ * The "change_mtu" method is usually not needed.
+ * If you need it, it must be like this.
+ */
+static int tile_net_change_mtu(struct net_device *dev, int new_mtu)
+{
+ PDEBUG("tile_net_change_mtu()\n");
+
+ /* Check ranges. */
+ if ((new_mtu < 68) || (new_mtu > 1500))
+ return -EINVAL;
+
+ /* Accept the value. */
+ dev->mtu = new_mtu;
+
+ return 0;
+}
+
+
+/*
+ * Change the Ethernet Address of the NIC.
+ *
+ * The hypervisor driver does not support changing MAC address. However,
+ * the IPP does not do anything with the MAC address, so the address which
+ * gets used on outgoing packets, and which is accepted on incoming packets,
+ * is completely up to the NetIO program or kernel driver which is actually
+ * handling them.
+ *
+ * Returns 0 on success, negative on failure.
+ */
+static int tile_net_set_mac_address(struct net_device *dev, void *p)
+{
+ struct sockaddr *addr = p;
+
+ if (!is_valid_ether_addr(addr->sa_data))
+ return -EINVAL;
+
+ /* ISSUE: Note that "dev_addr" is now a pointer. */
+ memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+
+ return 0;
+}
+
+
+/*
+ * Obtain the MAC address from the hypervisor.
+ * This must be done before opening the device.
+ */
+static int tile_net_get_mac(struct net_device *dev)
+{
+ struct tile_net_priv *priv = netdev_priv(dev);
+
+ char hv_dev_name[32];
+ int len;
+
+ __netio_getset_offset_t offset = { .word = NETIO_IPP_PARAM_OFF };
+
+ int ret;
+
+ /* For example, "xgbe0". */
+ strcpy(hv_dev_name, dev->name);
+ len = strlen(hv_dev_name);
+
+ /* For example, "xgbe/0". */
+ hv_dev_name[len] = hv_dev_name[len - 1];
+ hv_dev_name[len - 1] = '/';
+ len++;
+
+ /* For example, "xgbe/0/native_hash". */
+ strcpy(hv_dev_name + len, hash_default ? "/native_hash" : "/native");
+
+ /* Get the hypervisor handle for this device. */
+ priv->hv_devhdl = hv_dev_open((HV_VirtAddr)hv_dev_name, 0);
+ PDEBUG("hv_dev_open(%s) returned %d %p\n",
+ hv_dev_name, priv->hv_devhdl, &priv->hv_devhdl);
+ if (priv->hv_devhdl < 0) {
+ if (priv->hv_devhdl == HV_ENODEV)
+ printk(KERN_DEBUG "Ignoring unconfigured device %s\n",
+ hv_dev_name);
+ else
+ printk(KERN_DEBUG "hv_dev_open(%s) returned %d\n",
+ hv_dev_name, priv->hv_devhdl);
+ return -1;
+ }
+
+ /*
+ * Read the hardware address from the hypervisor.
+ * ISSUE: Note that "dev_addr" is now a pointer.
+ */
+ offset.bits.class = NETIO_PARAM;
+ offset.bits.addr = NETIO_PARAM_MAC;
+ ret = hv_dev_pread(priv->hv_devhdl, 0,
+ (HV_VirtAddr)dev->dev_addr, dev->addr_len,
+ offset.word);
+ PDEBUG("hv_dev_pread(NETIO_PARAM_MAC) returned %d\n", ret);
+ if (ret <= 0) {
+ printk(KERN_DEBUG "hv_dev_pread(NETIO_PARAM_MAC) %s failed\n",
+ dev->name);
+ /*
+ * Since the device is configured by the hypervisor but we
+ * can't get its MAC address, we are most likely running
+ * the simulator, so let's generate a random MAC address.
+ */
+ random_ether_addr(dev->dev_addr);
+ }
+
+ return 0;
+}
+
+
+static struct net_device_ops tile_net_ops = {
+ .ndo_open = tile_net_open,
+ .ndo_stop = tile_net_stop,
+ .ndo_start_xmit = tile_net_tx,
+ .ndo_do_ioctl = tile_net_ioctl,
+ .ndo_get_stats = tile_net_get_stats,
+ .ndo_change_mtu = tile_net_change_mtu,
+ .ndo_tx_timeout = tile_net_tx_timeout,
+ .ndo_set_mac_address = tile_net_set_mac_address
+};
+
+
+/*
+ * The setup function.
+ *
+ * This uses ether_setup() to assign various fields in dev, including
+ * setting IFF_BROADCAST and IFF_MULTICAST, then sets some extra fields.
+ */
+static void tile_net_setup(struct net_device *dev)
+{
+ PDEBUG("tile_net_setup()\n");
+
+ ether_setup(dev);
+
+ dev->netdev_ops = &tile_net_ops;
+
+ dev->watchdog_timeo = TILE_NET_TIMEOUT;
+
+ /* We want lockless xmit. */
+ dev->features |= NETIF_F_LLTX;
+
+ /* We support hardware tx checksums. */
+ dev->features |= NETIF_F_HW_CSUM;
+
+ /* We support scatter/gather. */
+ dev->features |= NETIF_F_SG;
+
+ /* We support TSO. */
+ dev->features |= NETIF_F_TSO;
+
+#ifdef TILE_NET_GSO
+ /* We support GSO. */
+ dev->features |= NETIF_F_GSO;
+#endif
+
+ if (hash_default)
+ dev->features |= NETIF_F_HIGHDMA;
+
+ /* ISSUE: We should support NETIF_F_UFO. */
+
+ dev->tx_queue_len = TILE_NET_TX_QUEUE_LEN;
+
+ dev->mtu = TILE_NET_MTU;
+}
+
+
+/*
+ * Allocate the device structure, register the device, and obtain the
+ * MAC address from the hypervisor.
+ */
+static struct net_device *tile_net_dev_init(const char *name)
+{
+ int ret;
+ struct net_device *dev;
+ struct tile_net_priv *priv;
+ struct page *page;
+
+ /*
+ * Allocate the device structure. This allocates "priv", calls
+ * tile_net_setup(), and saves "name". Normally, "name" is a
+ * template, instantiated by register_netdev(), but not for us.
+ */
+ dev = alloc_netdev(sizeof(*priv), name, tile_net_setup);
+ if (!dev) {
+ pr_err("alloc_netdev(%s) failed\n", name);
+ return NULL;
+ }
+
+ priv = netdev_priv(dev);
+
+ /* Initialize "priv". */
+
+ memset(priv, 0, sizeof(*priv));
+
+ /* Save "dev" for "tile_net_open_retry()". */
+ priv->dev = dev;
+
+ INIT_DELAYED_WORK(&priv->retry_work, tile_net_open_retry);
+
+ spin_lock_init(&priv->cmd_lock);
+ spin_lock_init(&priv->comp_lock);
+
+ /* Allocate "epp_queue". */
+ BUG_ON(get_order(sizeof(lepp_queue_t)) != 0);
+ page = alloc_pages(GFP_KERNEL | __GFP_ZERO, 0);
+ if (!page) {
+ free_netdev(dev);
+ return NULL;
+ }
+ priv->epp_queue = page_address(page);
+
+ /* Register the network device. */
+ ret = register_netdev(dev);
+ if (ret) {
+ pr_err("register_netdev %s failed %d\n", dev->name, ret);
+ free_page((unsigned long)priv->epp_queue);
+ free_netdev(dev);
+ return NULL;
+ }
+
+ /* Get the MAC address. */
+ ret = tile_net_get_mac(dev);
+ if (ret < 0) {
+ unregister_netdev(dev);
+ free_page((unsigned long)priv->epp_queue);
+ free_netdev(dev);
+ return NULL;
+ }
+
+ return dev;
+}
+
+
+/*
+ * Module cleanup.
+ */
+static void tile_net_cleanup(void)
+{
+ int i;
+
+ for (i = 0; i < TILE_NET_DEVS; i++) {
+ if (tile_net_devs[i]) {
+ struct net_device *dev = tile_net_devs[i];
+ struct tile_net_priv *priv = netdev_priv(dev);
+ unregister_netdev(dev);
+ finv_buffer(priv->epp_queue, PAGE_SIZE);
+ free_page((unsigned long)priv->epp_queue);
+ free_netdev(dev);
+ }
+ }
+}
+
+
+/*
+ * Module initialization.
+ */
+static int tile_net_init_module(void)
+{
+ pr_info("Tilera IPP Net Driver\n");
+
+ tile_net_devs[0] = tile_net_dev_init("xgbe0");
+ tile_net_devs[1] = tile_net_dev_init("xgbe1");
+ tile_net_devs[2] = tile_net_dev_init("gbe0");
+ tile_net_devs[3] = tile_net_dev_init("gbe1");
+
+ return 0;
+}
+
+
+#ifndef MODULE
+/*
+ * The "network_cpus" boot argument specifies the cpus that are dedicated
+ * to handle ingress packets.
+ *
+ * The parameter should be in the form "network_cpus=m-n[,x-y]", where
+ * m, n, x, y are integer numbers that represent the cpus that can be
+ * neither a dedicated cpu nor a dataplane cpu.
+ */
+static int __init network_cpus_setup(char *str)
+{
+ int rc = cpulist_parse_crop(str, &network_cpus_map);
+ if (rc != 0) {
+ pr_warning("network_cpus=%s: malformed cpu list\n",
+ str);
+ } else {
+
+ /* Remove dedicated cpus. */
+ cpumask_and(&network_cpus_map, &network_cpus_map,
+ cpu_possible_mask);
+
+
+ if (cpumask_empty(&network_cpus_map)) {
+ pr_warning("Ignoring network_cpus='%s'.\n",
+ str);
+ } else {
+ char buf[1024];
+ cpulist_scnprintf(buf, sizeof(buf), &network_cpus_map);
+ pr_info("Linux network CPUs: %s\n", buf);
+ network_cpus_used = true;
+ }
+ }
+
+ return 0;
+}
+__setup("network_cpus=", network_cpus_setup);
+#endif
+
+
+module_init(tile_net_init_module);
+module_exit(tile_net_cleanup);
#define UCC_GETH_UTFS_INIT 512 /* Tx virtual FIFO size
*/
#define UCC_GETH_UTFET_INIT 256 /* 1/2 utfs */
-#define UCC_GETH_UTFTT_INIT 512
+#define UCC_GETH_UTFTT_INIT 256 /* 1/2 utfs
+ due to errata */
/* Gigabit Ethernet (1000 Mbps) */
#define UCC_GETH_URFS_GIGA_INIT 4096/*2048*/ /* Rx virtual
FIFO size */
case HSO_INTF_BULK:
/* It's a regular bulk interface */
- if (((port_spec & HSO_PORT_MASK) == HSO_PORT_NETWORK) &&
- !disable_net)
- hso_dev = hso_create_net_device(interface, port_spec);
- else
+ if ((port_spec & HSO_PORT_MASK) == HSO_PORT_NETWORK) {
+ if (!disable_net)
+ hso_dev =
+ hso_create_net_device(interface, port_spec);
+ } else {
hso_dev =
hso_create_bulk_serial_device(interface, port_spec);
+ }
if (!hso_dev)
goto exit;
break;
static int x25_asy_close(struct net_device *dev)
{
struct x25_asy *sl = netdev_priv(dev);
- int err;
spin_lock(&sl->lock);
if (sl->tty)
netif_stop_queue(dev);
sl->rcount = 0;
sl->xleft = 0;
- err = lapb_unregister(dev);
- if (err != LAPB_OK)
- printk(KERN_ERR "x25_asy_close: lapb_unregister error -%d\n",
- err);
spin_unlock(&sl->lock);
return 0;
}
static void x25_asy_close_tty(struct tty_struct *tty)
{
struct x25_asy *sl = tty->disc_data;
+ int err;
/* First make sure we're connected. */
if (!sl || sl->magic != X25_ASY_MAGIC)
dev_close(sl->dev);
rtnl_unlock();
+ err = lapb_unregister(sl->dev);
+ if (err != LAPB_OK)
+ printk(KERN_ERR "x25_asy_close: lapb_unregister error -%d\n",
+ err);
+
tty->disc_data = NULL;
sl->tty = NULL;
x25_asy_free(sl);
bool stopped;
spin_lock_bh(&sc->rx.rxbuflock);
- ath9k_hw_stoppcurecv(ah);
+ ath9k_hw_abortpcurecv(ah);
ath9k_hw_setrxfilter(ah, 0);
stopped = ath9k_hw_stopdmarecv(ah);
}
unlock:
- if (err && (vif_id != -1)) {
+ if (err && (vif_id >= 0)) {
vif_priv->active = false;
bitmap_release_region(&ar->vif_bitmap, vif_id, 0);
ar->vifs--;
err_free_ssb:
kfree(sdio);
err_disable_func:
+ sdio_claim_host(func);
sdio_disable_func(func);
err_release_host:
sdio_release_host(func);
obj-$(CONFIG_X86_VISWS) += setup-irq.o
obj-$(CONFIG_MN10300) += setup-bus.o
obj-$(CONFIG_MICROBLAZE) += setup-bus.o
+obj-$(CONFIG_TILE) += setup-bus.o setup-irq.o
#
# ACPI Related PCI FW Functions
DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82875_HB,
quirk_unhide_mch_dev6);
+#ifdef CONFIG_TILE
+/*
+ * The Tilera TILEmpower platform needs to set the link speed
+ * to 2.5GT(Giga-Transfers)/s (Gen 1). The default link speed
+ * setting is 5GT/s (Gen 2). 0x98 is the Link Control2 PCIe
+ * capability register of the PEX8624 PCIe switch. The switch
+ * supports link speed auto negotiation, but falsely sets
+ * the link speed to 5GT/s.
+ */
+static void __devinit quirk_tile_plx_gen1(struct pci_dev *dev)
+{
+ if (tile_plx_gen1) {
+ pci_write_config_dword(dev, 0x98, 0x1);
+ mdelay(50);
+ }
+}
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_PLX, 0x8624, quirk_tile_plx_gen1);
+#endif /* CONFIG_TILE */
#ifdef CONFIG_PCI_MSI
/* Some chipsets do not support MSI. We cannot easily rely on setting
return;
/* reset adapter interrupt indicators */
- put_indicator(irq_ptr->dsci);
set_subchannel_ind(irq_ptr, 1);
+ put_indicator(irq_ptr->dsci);
}
void __exit tiqdio_unregister_thinints(void)
* : SCpnt - Command to queue
* Returns : 0 - success, else error
*/
-extern int fas216_noqueue_command(struct Scsi_Host *, struct scsi_cmnd *)
+extern int fas216_noqueue_command(struct Scsi_Host *, struct scsi_cmnd *);
/* Function: irqreturn_t fas216_intr (FAS216_Info *info)
* Purpose : handle interrupts from the interface to progress a command
list_add(&clk->sibling, &root_clks);
list_add(&clk->node, &clock_list);
+
+#ifdef CONFIG_SH_CLK_CPG_LEGACY
if (clk->ops && clk->ops->init)
clk->ops->init(clk);
+#endif
out_unlock:
mutex_unlock(&clock_list_sem);
EXPORT_SYMBOL_GPL(clk_get_rate);
int clk_set_rate(struct clk *clk, unsigned long rate)
-{
- return clk_set_rate_ex(clk, rate, 0);
-}
-EXPORT_SYMBOL_GPL(clk_set_rate);
-
-int clk_set_rate_ex(struct clk *clk, unsigned long rate, int algo_id)
{
int ret = -EOPNOTSUPP;
unsigned long flags;
spin_lock_irqsave(&clock_lock, flags);
if (likely(clk->ops && clk->ops->set_rate)) {
- ret = clk->ops->set_rate(clk, rate, algo_id);
+ ret = clk->ops->set_rate(clk, rate);
if (ret != 0)
goto out_unlock;
} else {
return ret;
}
-EXPORT_SYMBOL_GPL(clk_set_rate_ex);
+EXPORT_SYMBOL_GPL(clk_set_rate);
int clk_set_parent(struct clk *clk, struct clk *parent)
{
clkp->ops->set_parent(clkp,
clkp->parent);
if (likely(clkp->ops->set_rate))
- clkp->ops->set_rate(clkp,
- rate, NO_CHANGE);
+ clkp->ops->set_rate(clkp, rate);
else if (likely(clkp->ops->recalc))
clkp->rate = clkp->ops->recalc(clkp);
}
return 0;
}
-static int sh_clk_div6_set_rate(struct clk *clk,
- unsigned long rate, int algo_id)
+static int sh_clk_div6_set_rate(struct clk *clk, unsigned long rate)
{
unsigned long value;
int idx;
unsigned long value;
int ret;
- ret = sh_clk_div6_set_rate(clk, clk->rate, 0);
+ ret = sh_clk_div6_set_rate(clk, clk->rate);
if (ret == 0) {
value = __raw_readl(clk->enable_reg);
value &= ~0x100; /* clear stop bit to enable clock */
return 0;
}
-static int sh_clk_div4_set_rate(struct clk *clk, unsigned long rate, int algo_id)
+static int sh_clk_div4_set_rate(struct clk *clk, unsigned long rate)
{
struct clk_div4_table *d4t = clk->priv;
unsigned long value;
xfer->tx_dma = xfer->rx_dma = INVALID_DMA_ADDRESS;
if (xfer->tx_buf) {
+ /* tx_buf is a const void* where we need a void * for the dma
+ * mapping */
+ void *nonconst_tx = (void *)xfer->tx_buf;
+
xfer->tx_dma = dma_map_single(dev,
- (void *) xfer->tx_buf, xfer->len,
+ nonconst_tx, xfer->len,
DMA_TO_DEVICE);
if (dma_mapping_error(dev, xfer->tx_dma))
return -ENOMEM;
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4312) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4315) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4318) },
+ { PCI_DEVICE(PCI_VENDOR_ID_BCM_GVC, 0x4318) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4319) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4320) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4321) },
return ret;
}
-static DEVICE_ATTR(stat_status, S_IWUGO | S_IRUGO, read_status, reboot);
+static DEVICE_ATTR(stat_status, S_IWUSR | S_IRUGO, read_status, reboot);
static ssize_t read_human_status(struct device *dev,
struct device_attribute *attr, char *buf)
return ret;
}
-static DEVICE_ATTR(stat_human_status, S_IWUGO | S_IRUGO,
- read_human_status, NULL);
+static DEVICE_ATTR(stat_human_status, S_IRUGO, read_human_status, NULL);
static ssize_t read_delin(struct device *dev, struct device_attribute *attr,
char *buf)
return ret;
}
-static DEVICE_ATTR(stat_delin, S_IWUGO | S_IRUGO, read_delin, NULL);
+static DEVICE_ATTR(stat_delin, S_IRUGO, read_delin, NULL);
#define UEA_ATTR(name, reset) \
\
}
} else {
/* gpio_request fail so use -EINVAL for gpio_is_valid */
- ubc->vbus_pin = -EINVAL;
+ udc->vbus_pin = -EINVAL;
}
}
&debug_registers_fops))
goto file_error;
- if (!debugfs_create_file("lpm", S_IRUGO|S_IWUGO, ehci->debug_dir, bus,
+ if (!debugfs_create_file("lpm", S_IRUGO|S_IWUSR, ehci->debug_dir, bus,
&debug_lpm_fops))
goto file_error;
tmp && tmp != qh;
tmp = tmp->qh_next.qh)
continue;
- /* periodic qh self-unlinks on empty */
- if (!tmp)
- goto nogood;
- unlink_async (ehci, qh);
+ /* periodic qh self-unlinks on empty, and a COMPLETING qh
+ * may already be unlinked.
+ */
+ if (tmp)
+ unlink_async(ehci, qh);
/* FALL THROUGH */
case QH_STATE_UNLINK: /* wait for hw to finish? */
case QH_STATE_UNLINK_WAIT:
}
/* else FALL THROUGH */
default:
-nogood:
/* caller was supposed to have unlinked any requests;
* that's not our job. just leak this memory.
*/
qh_put (ehci->async);
ehci->async = NULL;
+ if (ehci->dummy)
+ qh_put(ehci->dummy);
+ ehci->dummy = NULL;
+
/* DMA consistent memory and pools */
if (ehci->qtd_pool)
dma_pool_destroy (ehci->qtd_pool);
if (ehci->periodic == NULL) {
goto fail;
}
- for (i = 0; i < ehci->periodic_size; i++)
- ehci->periodic [i] = EHCI_LIST_END(ehci);
+
+ if (ehci->use_dummy_qh) {
+ struct ehci_qh_hw *hw;
+ ehci->dummy = ehci_qh_alloc(ehci, flags);
+ if (!ehci->dummy)
+ goto fail;
+
+ hw = ehci->dummy->hw;
+ hw->hw_next = EHCI_LIST_END(ehci);
+ hw->hw_qtd_next = EHCI_LIST_END(ehci);
+ hw->hw_alt_next = EHCI_LIST_END(ehci);
+ hw->hw_token &= ~QTD_STS_ACTIVE;
+ ehci->dummy->hw = hw;
+
+ for (i = 0; i < ehci->periodic_size; i++)
+ ehci->periodic[i] = ehci->dummy->qh_dma;
+ } else {
+ for (i = 0; i < ehci->periodic_size; i++)
+ ehci->periodic[i] = EHCI_LIST_END(ehci);
+ }
/* software shadow of hardware table */
ehci->pshadow = kcalloc(ehci->periodic_size, sizeof(void *), flags);
if (retval)
return retval;
+ if ((pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == 0x7808) ||
+ (pdev->vendor == PCI_VENDOR_ID_ATI && pdev->device == 0x4396)) {
+ /* EHCI controller on AMD SB700/SB800/Hudson-2/3 platforms may
+ * read/write memory space which does not belong to it when
+ * there is NULL pointer with T-bit set to 1 in the frame list
+ * table. To avoid the issue, the frame list link pointer
+ * should always contain a valid pointer to a inactive qh.
+ */
+ ehci->use_dummy_qh = 1;
+ ehci_info(ehci, "applying AMD SB700/SB800/Hudson-2/3 EHCI "
+ "dummy qh workaround\n");
+ }
+
/* data structure init */
retval = ehci_init(hcd);
if (retval)
*/
*prev_p = *periodic_next_shadow(ehci, &here,
Q_NEXT_TYPE(ehci, *hw_p));
- *hw_p = *shadow_next_periodic(ehci, &here, Q_NEXT_TYPE(ehci, *hw_p));
+
+ if (!ehci->use_dummy_qh ||
+ *shadow_next_periodic(ehci, &here, Q_NEXT_TYPE(ehci, *hw_p))
+ != EHCI_LIST_END(ehci))
+ *hw_p = *shadow_next_periodic(ehci, &here,
+ Q_NEXT_TYPE(ehci, *hw_p));
+ else
+ *hw_p = ehci->dummy->qh_dma;
}
/* how many of the uframe's 125 usecs are allocated? */
* pointer for much longer, if at all.
*/
*q_p = q.itd->itd_next;
- *hw_p = q.itd->hw_next;
+ if (!ehci->use_dummy_qh ||
+ q.itd->hw_next != EHCI_LIST_END(ehci))
+ *hw_p = q.itd->hw_next;
+ else
+ *hw_p = ehci->dummy->qh_dma;
type = Q_NEXT_TYPE(ehci, q.itd->hw_next);
wmb();
modified = itd_complete (ehci, q.itd);
* URB completion.
*/
*q_p = q.sitd->sitd_next;
- *hw_p = q.sitd->hw_next;
+ if (!ehci->use_dummy_qh ||
+ q.sitd->hw_next != EHCI_LIST_END(ehci))
+ *hw_p = q.sitd->hw_next;
+ else
+ *hw_p = ehci->dummy->qh_dma;
type = Q_NEXT_TYPE(ehci, q.sitd->hw_next);
wmb();
modified = sitd_complete (ehci, q.sitd);
/* async schedule support */
struct ehci_qh *async;
+ struct ehci_qh *dummy; /* For AMD quirk use */
struct ehci_qh *reclaim;
unsigned scanning : 1;
unsigned need_io_watchdog:1;
unsigned broken_periodic:1;
unsigned fs_i_thresh:1; /* Intel iso scheduling */
+ unsigned use_dummy_qh:1; /* AMD Frame List table quirk*/
/* required for usb32 quirk */
#define OHCI_CTRL_HCFS (3 << 6)
return 0;
}
-static int __init isp1362_probe(struct platform_device *pdev)
+static int __devinit isp1362_probe(struct platform_device *pdev)
{
struct usb_hcd *hcd;
struct isp1362_hcd *isp1362_hcd;
if (udev->speed == USB_SPEED_SUPER)
return ep->ss_ep_comp.wBytesPerInterval;
- max_packet = ep->desc.wMaxPacketSize & 0x3ff;
+ max_packet = GET_MAX_PACKET(ep->desc.wMaxPacketSize);
max_burst = (ep->desc.wMaxPacketSize & 0x1800) >> 11;
/* A 0 in max burst means 1 transfer per ESIT */
return max_packet * (max_burst + 1);
/* Fall through */
case USB_SPEED_FULL:
case USB_SPEED_LOW:
- max_packet = ep->desc.wMaxPacketSize & 0x3ff;
+ max_packet = GET_MAX_PACKET(ep->desc.wMaxPacketSize);
ep_ctx->ep_info2 |= MAX_PACKET(max_packet);
break;
default:
if (!(status & STS_EINT)) {
spin_unlock(&xhci->lock);
- xhci_warn(xhci, "Spurious interrupt.\n");
return IRQ_NONE;
}
xhci_dbg(xhci, "op reg status = %08x\n", status);
xhci_write_64(xhci, xhci->s3.erst_base, &xhci->ir_set->erst_base);
}
+static void xhci_set_cmd_ring_deq(struct xhci_hcd *xhci)
+{
+ u64 val_64;
+
+ /* step 2: initialize command ring buffer */
+ val_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring);
+ val_64 = (val_64 & (u64) CMD_RING_RSVD_BITS) |
+ (xhci_trb_virt_to_dma(xhci->cmd_ring->deq_seg,
+ xhci->cmd_ring->dequeue) &
+ (u64) ~CMD_RING_RSVD_BITS) |
+ xhci->cmd_ring->cycle_state;
+ xhci_dbg(xhci, "// Setting command ring address to 0x%llx\n",
+ (long unsigned long) val_64);
+ xhci_write_64(xhci, val_64, &xhci->op_regs->cmd_ring);
+}
+
+/*
+ * The whole command ring must be cleared to zero when we suspend the host.
+ *
+ * The host doesn't save the command ring pointer in the suspend well, so we
+ * need to re-program it on resume. Unfortunately, the pointer must be 64-byte
+ * aligned, because of the reserved bits in the command ring dequeue pointer
+ * register. Therefore, we can't just set the dequeue pointer back in the
+ * middle of the ring (TRBs are 16-byte aligned).
+ */
+static void xhci_clear_command_ring(struct xhci_hcd *xhci)
+{
+ struct xhci_ring *ring;
+ struct xhci_segment *seg;
+
+ ring = xhci->cmd_ring;
+ seg = ring->deq_seg;
+ do {
+ memset(seg->trbs, 0, SEGMENT_SIZE);
+ seg = seg->next;
+ } while (seg != ring->deq_seg);
+
+ /* Reset the software enqueue and dequeue pointers */
+ ring->deq_seg = ring->first_seg;
+ ring->dequeue = ring->first_seg->trbs;
+ ring->enq_seg = ring->deq_seg;
+ ring->enqueue = ring->dequeue;
+
+ /*
+ * Ring is now zeroed, so the HW should look for change of ownership
+ * when the cycle bit is set to 1.
+ */
+ ring->cycle_state = 1;
+
+ /*
+ * Reset the hardware dequeue pointer.
+ * Yes, this will need to be re-written after resume, but we're paranoid
+ * and want to make sure the hardware doesn't access bogus memory
+ * because, say, the BIOS or an SMI started the host without changing
+ * the command ring pointers.
+ */
+ xhci_set_cmd_ring_deq(xhci);
+}
+
/*
* Stop HC (not bus-specific)
*
spin_unlock_irq(&xhci->lock);
return -ETIMEDOUT;
}
+ xhci_clear_command_ring(xhci);
/* step 3: save registers */
xhci_save_registers(xhci);
u32 command, temp = 0;
struct usb_hcd *hcd = xhci_to_hcd(xhci);
struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
- u64 val_64;
int old_state, retval;
old_state = hcd->state;
/* step 1: restore register */
xhci_restore_registers(xhci);
/* step 2: initialize command ring buffer */
- val_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring);
- val_64 = (val_64 & (u64) CMD_RING_RSVD_BITS) |
- (xhci_trb_virt_to_dma(xhci->cmd_ring->deq_seg,
- xhci->cmd_ring->dequeue) &
- (u64) ~CMD_RING_RSVD_BITS) |
- xhci->cmd_ring->cycle_state;
- xhci_dbg(xhci, "// Setting command ring address to 0x%llx\n",
- (long unsigned long) val_64);
- xhci_write_64(xhci, val_64, &xhci->op_regs->cmd_ring);
+ xhci_set_cmd_ring_deq(xhci);
/* step 3: restore state and start state*/
/* step 3: set CRS flag */
command = xhci_readl(xhci, &xhci->op_regs->command);
return retval;
}
+ spin_unlock_irq(&xhci->lock);
/* Re-setup MSI-X */
if (hcd->irq)
free_irq(hcd->irq, hcd);
hcd->irq = pdev->irq;
}
+ spin_lock_irq(&xhci->lock);
/* step 4: set Run/Stop bit */
command = xhci_readl(xhci, &xhci->op_regs->command);
command |= CMD_RUN;
#define MAX_PACKET_MASK (0xffff << 16)
#define MAX_PACKET_DECODED(p) (((p) >> 16) & 0xffff)
+/* Get max packet size from ep desc. Bit 10..0 specify the max packet size.
+ * USB2.0 spec 9.6.6.
+ */
+#define GET_MAX_PACKET(p) ((p) & 0x7ff)
+
/* tx_info bitmasks */
#define AVG_TRB_LENGTH_FOR_EP(p) ((p) & 0xffff)
#define MAX_ESIT_PAYLOAD_FOR_EP(p) (((p) & 0xffff) << 16)
return read_port(dev, attr, buf, 1, CYPRESS_READ_PORT_ID1);
}
-static DEVICE_ATTR(port0, S_IWUGO | S_IRUGO,
- get_port0_handler, set_port0_handler);
+static DEVICE_ATTR(port0, S_IRUGO | S_IWUSR, get_port0_handler, set_port0_handler);
-static DEVICE_ATTR(port1, S_IWUGO | S_IRUGO,
- get_port1_handler, set_port1_handler);
+static DEVICE_ATTR(port1, S_IRUGO | S_IWUSR, get_port1_handler, set_port1_handler);
static int cypress_probe(struct usb_interface *interface,
return count;
}
-static DEVICE_ATTR(speed, S_IWUGO | S_IRUGO, show_speed, set_speed);
+static DEVICE_ATTR(speed, S_IRUGO | S_IWUSR, show_speed, set_speed);
static int tv_probe(struct usb_interface *interface,
const struct usb_device_id *id)
change_color(led); \
return count; \
} \
-static DEVICE_ATTR(value, S_IWUGO | S_IRUGO, show_##value, set_##value);
+static DEVICE_ATTR(value, S_IRUGO | S_IWUSR, show_##value, set_##value);
show_set(blue);
show_set(red);
show_set(green);
\
return count; \
} \
-static DEVICE_ATTR(name, S_IWUGO | S_IRUGO, show_attr_##name, set_attr_##name);
+static DEVICE_ATTR(name, S_IRUGO | S_IWUSR, show_attr_##name, set_attr_##name);
static ssize_t show_attr_text(struct device *dev,
struct device_attribute *attr, char *buf)
return count;
}
-static DEVICE_ATTR(text, S_IWUGO | S_IRUGO, show_attr_text, set_attr_text);
+static DEVICE_ATTR(text, S_IRUGO | S_IWUSR, show_attr_text, set_attr_text);
static ssize_t show_attr_decimals(struct device *dev,
struct device_attribute *attr, char *buf)
return count;
}
-static DEVICE_ATTR(decimals, S_IWUGO | S_IRUGO,
- show_attr_decimals, set_attr_decimals);
+static DEVICE_ATTR(decimals, S_IRUGO | S_IWUSR, show_attr_decimals, set_attr_decimals);
static ssize_t show_attr_textmode(struct device *dev,
struct device_attribute *attr, char *buf)
return -EINVAL;
}
-static DEVICE_ATTR(textmode, S_IWUGO | S_IRUGO,
- show_attr_textmode, set_attr_textmode);
+static DEVICE_ATTR(textmode, S_IRUGO | S_IWUSR, show_attr_textmode, set_attr_textmode);
MYDEV_ATTR_SIMPLE_UNSIGNED(powered, update_display_powered);
}
return count;
}
-static DEVICE_ATTR(a_bus_req, S_IRUGO | S_IWUGO, get_a_bus_req, set_a_bus_req);
+static DEVICE_ATTR(a_bus_req, S_IRUGO | S_IWUSR, get_a_bus_req, set_a_bus_req);
static ssize_t
get_a_bus_drop(struct device *dev, struct device_attribute *attr, char *buf)
}
return count;
}
-static DEVICE_ATTR(a_bus_drop, S_IRUGO | S_IWUGO,
- get_a_bus_drop, set_a_bus_drop);
+static DEVICE_ATTR(a_bus_drop, S_IRUGO | S_IWUSR, get_a_bus_drop, set_a_bus_drop);
static ssize_t
get_b_bus_req(struct device *dev, struct device_attribute *attr, char *buf)
}
return count;
}
-static DEVICE_ATTR(b_bus_req, S_IRUGO | S_IWUGO, get_b_bus_req, set_b_bus_req);
+static DEVICE_ATTR(b_bus_req, S_IRUGO | S_IWUSR, get_b_bus_req, set_b_bus_req);
static ssize_t
set_a_clr_err(struct device *dev, struct device_attribute *attr,
}
return count;
}
-static DEVICE_ATTR(a_clr_err, S_IWUGO, NULL, set_a_clr_err);
+static DEVICE_ATTR(a_clr_err, S_IWUSR, NULL, set_a_clr_err);
static struct attribute *inputs_attrs[] = {
&dev_attr_a_bus_req.attr,
}
return result;
}
-static DEVICE_ATTR(truinst, S_IWUGO | S_IRUGO, show_truinst, NULL);
+static DEVICE_ATTR(truinst, S_IRUGO, show_truinst, NULL);
int sierra_ms_init(struct us_data *us)
{
size_t hdr_size;
struct socket *sock;
- sock = rcu_dereference_check(vq->private_data,
- lockdep_is_held(&vq->mutex));
+ /* TODO: check that we are running from vhost_worker?
+ * Not sure it's worth it, it's straight-forward enough. */
+ sock = rcu_dereference_check(vq->private_data, 1);
if (!sock)
return;
{
struct backlight_device *bd = to_backlight_device(dev);
- if (bd->ops->options & BL_CORE_SUSPENDRESUME) {
- mutex_lock(&bd->ops_lock);
+ mutex_lock(&bd->ops_lock);
+ if (bd->ops && bd->ops->options & BL_CORE_SUSPENDRESUME) {
bd->props.state |= BL_CORE_SUSPENDED;
backlight_update_status(bd);
- mutex_unlock(&bd->ops_lock);
}
+ mutex_unlock(&bd->ops_lock);
return 0;
}
{
struct backlight_device *bd = to_backlight_device(dev);
- if (bd->ops->options & BL_CORE_SUSPENDRESUME) {
- mutex_lock(&bd->ops_lock);
+ mutex_lock(&bd->ops_lock);
+ if (bd->ops && bd->ops->options & BL_CORE_SUSPENDRESUME) {
bd->props.state &= ~BL_CORE_SUSPENDED;
backlight_update_status(bd);
- mutex_unlock(&bd->ops_lock);
}
+ mutex_unlock(&bd->ops_lock);
return 0;
}
*
*/
-int fb_alloc_cmap(struct fb_cmap *cmap, int len, int transp)
+int fb_alloc_cmap_gfp(struct fb_cmap *cmap, int len, int transp, gfp_t flags)
{
- int size = len*sizeof(u16);
-
- if (cmap->len != len) {
- fb_dealloc_cmap(cmap);
- if (!len)
- return 0;
- if (!(cmap->red = kmalloc(size, GFP_ATOMIC)))
- goto fail;
- if (!(cmap->green = kmalloc(size, GFP_ATOMIC)))
- goto fail;
- if (!(cmap->blue = kmalloc(size, GFP_ATOMIC)))
- goto fail;
- if (transp) {
- if (!(cmap->transp = kmalloc(size, GFP_ATOMIC)))
+ int size = len * sizeof(u16);
+ int ret = -ENOMEM;
+
+ if (cmap->len != len) {
+ fb_dealloc_cmap(cmap);
+ if (!len)
+ return 0;
+
+ cmap->red = kmalloc(size, flags);
+ if (!cmap->red)
+ goto fail;
+ cmap->green = kmalloc(size, flags);
+ if (!cmap->green)
+ goto fail;
+ cmap->blue = kmalloc(size, flags);
+ if (!cmap->blue)
+ goto fail;
+ if (transp) {
+ cmap->transp = kmalloc(size, flags);
+ if (!cmap->transp)
+ goto fail;
+ } else {
+ cmap->transp = NULL;
+ }
+ }
+ cmap->start = 0;
+ cmap->len = len;
+ ret = fb_copy_cmap(fb_default_cmap(len), cmap);
+ if (ret)
goto fail;
- } else
- cmap->transp = NULL;
- }
- cmap->start = 0;
- cmap->len = len;
- fb_copy_cmap(fb_default_cmap(len), cmap);
- return 0;
+ return 0;
fail:
- fb_dealloc_cmap(cmap);
- return -ENOMEM;
+ fb_dealloc_cmap(cmap);
+ return ret;
+}
+
+int fb_alloc_cmap(struct fb_cmap *cmap, int len, int transp)
+{
+ return fb_alloc_cmap_gfp(cmap, len, transp, GFP_ATOMIC);
}
/**
int rc, size = cmap->len * sizeof(u16);
struct fb_cmap umap;
+ if (size < 0 || size < cmap->len)
+ return -E2BIG;
+
memset(&umap, 0, sizeof(struct fb_cmap));
- rc = fb_alloc_cmap(&umap, cmap->len, cmap->transp != NULL);
+ rc = fb_alloc_cmap_gfp(&umap, cmap->len, cmap->transp != NULL,
+ GFP_KERNEL);
if (rc)
return rc;
if (copy_from_user(umap.red, cmap->red, size) ||
write_fp(par, FP_PT1, 0);
temp = FP_PT2_SCRC;
- if (info->var.sync & FB_SYNC_HOR_HIGH_ACT)
+ if (!(info->var.sync & FB_SYNC_HOR_HIGH_ACT))
temp |= FP_PT2_HSP;
- if (info->var.sync & FB_SYNC_VERT_HIGH_ACT)
+ if (!(info->var.sync & FB_SYNC_VERT_HIGH_ACT))
temp |= FP_PT2_VSP;
write_fp(par, FP_PT2, temp);
#include <linux/clk.h>
#include <linux/mutex.h>
+#include <mach/dma.h>
#include <mach/hardware.h>
#include <mach/ipu.h>
#include <mach/mx3fb.h>
struct device *dev;
struct mx3fb_platform_data *mx3fb_pdata;
+ if (!imx_dma_is_ipu(chan))
+ return false;
+
if (!rq)
return false;
/* Couldn't reconfigure, hopefully, can continue as before */
return;
- info->fix.line_length = mode2.xres * (ch->cfg.bpp / 8);
+ info->fix.line_length = mode1.xres * (ch->cfg.bpp / 8);
/*
* fb_set_var() calls the notifier change internally, only if
* user event, we have to call the chain ourselves.
*/
event.info = info;
- event.data = &mode2;
+ event.data = &mode1;
fb_notifier_call_chain(evnt, &event);
}
#include "init.h"
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
#include "300vtbl.h"
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
#include "310vtbl.h"
#endif
/* POINTER INITIALIZATION */
/*********************************************/
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
static void
InitCommonPointer(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static void
InitTo300Pointer(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
InitTo310Pointer(struct SiS_Private *SiS_Pr)
{
SiSInitPtr(struct SiS_Private *SiS_Pr)
{
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
InitTo300Pointer(SiS_Pr);
#else
return false;
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
InitTo310Pointer(SiS_Pr);
#else
return false;
/* HELPER: Get ModeID */
/*********************************************/
-#ifndef SIS_XORG_XF86
static
-#endif
unsigned short
SiS_GetModeID(int VGAEngine, unsigned int VBFlags, int HDisplay, int VDisplay,
int Depth, bool FSTN, int LCDwidth, int LCDheight)
void
SiS_SetReg(SISIOADDRESS port, unsigned short index, unsigned short data)
{
- OutPortByte(port, index);
- OutPortByte(port + 1, data);
+ outb((u8)index, port);
+ outb((u8)data, port + 1);
}
void
SiS_SetRegByte(SISIOADDRESS port, unsigned short data)
{
- OutPortByte(port, data);
+ outb((u8)data, port);
}
void
SiS_SetRegShort(SISIOADDRESS port, unsigned short data)
{
- OutPortWord(port, data);
+ outw((u16)data, port);
}
void
SiS_SetRegLong(SISIOADDRESS port, unsigned int data)
{
- OutPortLong(port, data);
+ outl((u32)data, port);
}
unsigned char
SiS_GetReg(SISIOADDRESS port, unsigned short index)
{
- OutPortByte(port, index);
- return(InPortByte(port + 1));
+ outb((u8)index, port);
+ return inb(port + 1);
}
unsigned char
SiS_GetRegByte(SISIOADDRESS port)
{
- return(InPortByte(port));
+ return inb(port);
}
unsigned short
SiS_GetRegShort(SISIOADDRESS port)
{
- return(InPortWord(port));
+ return inw(port);
}
unsigned int
SiS_GetRegLong(SISIOADDRESS port)
{
- return(InPortLong(port));
+ return inl(port);
}
void
SiSInitPCIetc(struct SiS_Private *SiS_Pr)
{
switch(SiS_Pr->ChipType) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
case SIS_300:
case SIS_540:
case SIS_630:
SiS_SetRegOR(SiS_Pr->SiS_P3c4,0x1E,0x5A);
break;
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
case SIS_315H:
case SIS_315:
case SIS_315PRO:
/* HELPER: SetLVDSetc */
/*********************************************/
-#ifdef SIS_LINUX_KERNEL
static
-#endif
void
SiSSetLVDSetc(struct SiS_Private *SiS_Pr)
{
if((temp == 1) || (temp == 2)) return;
switch(SiS_Pr->ChipType) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
case SIS_540:
case SIS_630:
case SIS_730:
}
break;
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
case SIS_550:
case SIS_650:
case SIS_740:
/* HELPER: GetVBType */
/*********************************************/
-#ifdef SIS_LINUX_KERNEL
static
-#endif
void
SiS_GetVBType(struct SiS_Private *SiS_Pr)
{
/* HELPER: Check RAM size */
/*********************************************/
-#ifdef SIS_LINUX_KERNEL
static bool
SiS_CheckMemorySize(struct SiS_Private *SiS_Pr, unsigned short ModeNo,
unsigned short ModeIdIndex)
if(AdapterMemSize < memorysize) return false;
return true;
}
-#endif
/*********************************************/
/* HELPER: Get DRAM type */
/*********************************************/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static unsigned char
SiS_Get310DRAMType(struct SiS_Private *SiS_Pr)
{
/* HELPER: ClearBuffer */
/*********************************************/
-#ifdef SIS_LINUX_KERNEL
static void
SiS_ClearBuffer(struct SiS_Private *SiS_Pr, unsigned short ModeNo)
{
if(SiS_Pr->SiS_ModeType >= ModeEGA) {
if(ModeNo > 0x13) {
- SiS_SetMemory(memaddr, memsize, 0);
+ memset_io(memaddr, 0, memsize);
} else {
pBuffer = (unsigned short SISIOMEMTYPE *)memaddr;
for(i = 0; i < 0x4000; i++) writew(0x0000, &pBuffer[i]);
pBuffer = (unsigned short SISIOMEMTYPE *)memaddr;
for(i = 0; i < 0x4000; i++) writew(0x0720, &pBuffer[i]);
} else {
- SiS_SetMemory(memaddr, 0x8000, 0);
+ memset_io(memaddr, 0, 0x8000);
}
}
-#endif
/*********************************************/
/* HELPER: SearchModeID */
SiS_SetReg(SiS_Pr->SiS_P3d4,0x14,0x4F);
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType == XGI_20) {
SiS_SetReg(SiS_Pr->SiS_P3d4,0x04,crt1data[4] - 1);
if(!(temp = crt1data[5] & 0x1f)) {
SiS_SetReg(SiS_Pr->SiS_P3c4,0x2c,clkb);
if(SiS_Pr->ChipType >= SIS_315H) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_SetReg(SiS_Pr->SiS_P3c4,0x2D,0x01);
if(SiS_Pr->ChipType == XGI_20) {
unsigned short mf = SiS_GetModeFlag(SiS_Pr, ModeNo, ModeIdIndex);
/* FIFO */
/*********************************************/
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
void
SiS_GetFIFOThresholdIndex300(struct SiS_Private *SiS_Pr, unsigned short *idx1,
unsigned short *idx2)
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x09,0x80,data);
/* Write foreground and background queue */
-#ifdef SIS_LINUX_KERNEL
templ = sisfb_read_nbridge_pci_dword(SiS_Pr, 0x50);
-#else
- templ = pciReadLong(0x00000000, 0x50);
-#endif
if(SiS_Pr->ChipType == SIS_730) {
}
-#ifdef SIS_LINUX_KERNEL
sisfb_write_nbridge_pci_dword(SiS_Pr, 0x50, templ);
templ = sisfb_read_nbridge_pci_dword(SiS_Pr, 0xA0);
-#else
- pciWriteLong(0x00000000, 0x50, templ);
- templ = pciReadLong(0x00000000, 0xA0);
-#endif
/* GUI grant timer (PCI config 0xA3) */
if(SiS_Pr->ChipType == SIS_730) {
}
-#ifdef SIS_LINUX_KERNEL
sisfb_write_nbridge_pci_dword(SiS_Pr, 0xA0, templ);
-#else
- pciWriteLong(0x00000000, 0xA0, templ);
-#endif
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
SiS_SetCRT1FIFO_310(struct SiS_Private *SiS_Pr, unsigned short ModeNo, unsigned short ModeIdIndex)
{
}
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(VCLK > 150) data |= 0x80;
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x07,0x7B,data);
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x32,0xF7,data);
#endif
} else if(SiS_Pr->ChipType < XGI_20) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(VCLK >= 166) data |= 0x0c;
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x32,0xf3,data);
}
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(VCLK >= 200) data |= 0x0c;
if(SiS_Pr->ChipType == XGI_20) data &= ~0x04;
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x32,0xf3,data);
unsigned short ModeIdIndex, unsigned short RRTI)
{
unsigned short data, infoflag = 0, modeflag, resindex;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
unsigned short data2, data3;
#endif
SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x0F,0xB7,data);
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
SiS_SetRegAND(SiS_Pr->SiS_P3c4,0x31,0xfb);
}
SiS_SetVCLKState(SiS_Pr, ModeNo, RRTI, ModeIdIndex);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(((SiS_Pr->ChipType >= SIS_315H) && (SiS_Pr->ChipType < SIS_661)) ||
(SiS_Pr->ChipType == XGI_40)) {
if(SiS_GetReg(SiS_Pr->SiS_P3d4,0x31) & 0x40) {
#endif
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
SiS_SetupDualChip(struct SiS_Private *SiS_Pr)
{
SiS_Pr->SiS_SelectCRT2Rate = 0;
SiS_Pr->SiS_SetFlag &= (~ProgrammingCRT2);
-#ifdef SIS_XORG_XF86
- xf86DrvMsgVerb(0, X_PROBED, 4, "(init: VBType=0x%04x, VBInfo=0x%04x)\n",
- SiS_Pr->SiS_VBType, SiS_Pr->SiS_VBInfo);
-#endif
-
if(SiS_Pr->SiS_VBInfo & SetSimuScanMode) {
if(SiS_Pr->SiS_VBInfo & SetInSlaveMode) {
SiS_Pr->SiS_SetFlag |= ProgrammingCRT2;
}
switch(SiS_Pr->ChipType) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
case SIS_300:
SiS_SetCRT1FIFO_300(SiS_Pr, ModeNo, RefreshRateTableIndex);
break;
break;
#endif
default:
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType == XGI_20) {
unsigned char sr2b = 0, sr2c = 0;
switch(ModeNo) {
SiS_SetCRT1ModeRegs(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType == XGI_40) {
SiS_SetupDualChip(SiS_Pr);
}
SiS_LoadDAC(SiS_Pr, ModeNo, ModeIdIndex);
-#ifdef SIS_LINUX_KERNEL
if(SiS_Pr->SiS_flag_clearbuffer) {
SiS_ClearBuffer(SiS_Pr, ModeNo);
}
-#endif
if(!(SiS_Pr->SiS_VBInfo & (SetSimuScanMode | SwitchCRT2 | SetCRT2ToLCDA))) {
SiS_WaitRetrace1(SiS_Pr);
static void
SiS_ResetVB(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
unsigned short temp;
* which locks CRT2 in some way to CRT1 timing. Disable
* this here.
*/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if((IS_SIS651) || (IS_SISM650) ||
SiS_Pr->ChipType == SIS_340 ||
SiS_Pr->ChipType == XGI_40) {
static void
SiS_Handle760(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned int somebase;
unsigned char temp1, temp2, temp3;
(!(SiS_Pr->SiS_SysFlags & SF_760UMA)) )
return;
-#ifdef SIS_LINUX_KERNEL
somebase = sisfb_read_mio_pci_word(SiS_Pr, 0x74);
-#else
- somebase = pciReadWord(0x00001000, 0x74);
-#endif
somebase &= 0xffff;
if(somebase == 0) return;
temp2 = 0x0b;
}
-#ifdef SIS_LINUX_KERNEL
sisfb_write_nbridge_pci_byte(SiS_Pr, 0x7e, temp1);
sisfb_write_nbridge_pci_byte(SiS_Pr, 0x8d, temp2);
-#else
- pciWriteByte(0x00000000, 0x7e, temp1);
- pciWriteByte(0x00000000, 0x8d, temp2);
-#endif
SiS_SetRegByte((somebase + 0x85), temp3);
#endif
}
-/*********************************************/
-/* X.org/XFree86: SET SCREEN PITCH */
-/*********************************************/
-
-#ifdef SIS_XORG_XF86
-static void
-SiS_SetPitchCRT1(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn)
-{
- SISPtr pSiS = SISPTR(pScrn);
- unsigned short HDisplay = pSiS->scrnPitch >> 3;
-
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x13,(HDisplay & 0xFF));
- SiS_SetRegANDOR(SiS_Pr->SiS_P3c4,0x0E,0xF0,(HDisplay >> 8));
-}
-
-static void
-SiS_SetPitchCRT2(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn)
-{
- SISPtr pSiS = SISPTR(pScrn);
- unsigned short HDisplay = pSiS->scrnPitch2 >> 3;
-
- /* Unlock CRT2 */
- if(pSiS->VGAEngine == SIS_315_VGA)
- SiS_SetRegOR(SiS_Pr->SiS_Part1Port,0x2F, 0x01);
- else
- SiS_SetRegOR(SiS_Pr->SiS_Part1Port,0x24, 0x01);
-
- SiS_SetReg(SiS_Pr->SiS_Part1Port,0x07,(HDisplay & 0xFF));
- SiS_SetRegANDOR(SiS_Pr->SiS_Part1Port,0x09,0xF0,(HDisplay >> 8));
-}
-
-static void
-SiS_SetPitch(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn)
-{
- SISPtr pSiS = SISPTR(pScrn);
- bool isslavemode = false;
-
- if( (pSiS->VBFlags2 & VB2_VIDEOBRIDGE) &&
- ( ((pSiS->VGAEngine == SIS_300_VGA) &&
- (SiS_GetReg(SiS_Pr->SiS_Part1Port,0x00) & 0xa0) == 0x20) ||
- ((pSiS->VGAEngine == SIS_315_VGA) &&
- (SiS_GetReg(SiS_Pr->SiS_Part1Port,0x00) & 0x50) == 0x10) ) ) {
- isslavemode = true;
- }
-
- /* We need to set pitch for CRT1 if bridge is in slave mode, too */
- if((pSiS->VBFlags & DISPTYPE_DISP1) || (isslavemode)) {
- SiS_SetPitchCRT1(SiS_Pr, pScrn);
- }
- /* We must not set the pitch for CRT2 if bridge is in slave mode */
- if((pSiS->VBFlags & DISPTYPE_DISP2) && (!isslavemode)) {
- SiS_SetPitchCRT2(SiS_Pr, pScrn);
- }
-}
-#endif
-
/*********************************************/
/* SiSSetMode() */
/*********************************************/
-#ifdef SIS_XORG_XF86
-/* We need pScrn for setting the pitch correctly */
-bool
-SiSSetMode(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn, unsigned short ModeNo, bool dosetpitch)
-#else
bool
SiSSetMode(struct SiS_Private *SiS_Pr, unsigned short ModeNo)
-#endif
{
SISIOADDRESS BaseAddr = SiS_Pr->IOAddress;
unsigned short RealModeNo, ModeIdIndex;
unsigned char backupreg = 0;
-#ifdef SIS_LINUX_KERNEL
unsigned short KeepLockReg;
SiS_Pr->UseCustomMode = false;
SiS_Pr->CRT1UsesCustomMode = false;
-#endif
SiS_Pr->SiS_flag_clearbuffer = 0;
if(SiS_Pr->UseCustomMode) {
ModeNo = 0xfe;
} else {
-#ifdef SIS_LINUX_KERNEL
if(!(ModeNo & 0x80)) SiS_Pr->SiS_flag_clearbuffer = 1;
-#endif
ModeNo &= 0x7f;
}
SiS_GetSysFlags(SiS_Pr);
SiS_Pr->SiS_VGAINFO = 0x11;
-#if defined(SIS_XORG_XF86) && (defined(i386) || defined(__i386) || defined(__i386__) || defined(__AMD64__) || defined(__amd64__) || defined(__x86_64__))
- if(pScrn) SiS_Pr->SiS_VGAINFO = SiS_GetSetBIOSScratch(pScrn, 0x489, 0xff);
-#endif
-#ifdef SIS_LINUX_KERNEL
KeepLockReg = SiS_GetReg(SiS_Pr->SiS_P3c4,0x05);
-#endif
SiS_SetReg(SiS_Pr->SiS_P3c4,0x05,0x86);
SiSInitPCIetc(SiS_Pr);
SiS_GetLCDResInfo(SiS_Pr, ModeNo, ModeIdIndex);
SiS_SetLowModeTest(SiS_Pr, ModeNo);
-#ifdef SIS_LINUX_KERNEL
/* Check memory size (kernel framebuffer driver only) */
if(!SiS_CheckMemorySize(SiS_Pr, ModeNo, ModeIdIndex)) {
return false;
}
-#endif
SiS_OpenCRTC(SiS_Pr);
SiS_DisplayOn(SiS_Pr);
SiS_SetRegByte(SiS_Pr->SiS_P3c6,0xFF);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if(SiS_Pr->SiS_IF_DEF_LVDS == 1) {
if(!(SiS_IsDualEdge(SiS_Pr))) {
if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
if(SiS_Pr->ChipType >= SIS_315H) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(!SiS_Pr->SiS_ROMNew) {
if(SiS_IsVAMode(SiS_Pr)) {
SiS_SetRegOR(SiS_Pr->SiS_P3d4,0x35,0x01);
}
}
-#ifdef SIS_XORG_XF86
- if(pScrn) {
- /* SetPitch: Adapt to virtual size & position */
- if((ModeNo > 0x13) && (dosetpitch)) {
- SiS_SetPitch(SiS_Pr, pScrn);
- }
-
- /* Backup/Set ModeNo in BIOS scratch area */
- SiS_GetSetModeID(pScrn, ModeNo);
- }
-#endif
-
SiS_CloseCRTC(SiS_Pr);
SiS_Handle760(SiS_Pr);
-#ifdef SIS_LINUX_KERNEL
/* We never lock registers in XF86 */
if(KeepLockReg != 0xA1) SiS_SetReg(SiS_Pr->SiS_P3c4,0x05,0x00);
-#endif
return true;
}
-/*********************************************/
-/* X.org/XFree86: SiSBIOSSetMode() */
-/* for non-Dual-Head mode */
-/*********************************************/
-
-#ifdef SIS_XORG_XF86
-bool
-SiSBIOSSetMode(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom)
-{
- SISPtr pSiS = SISPTR(pScrn);
- unsigned short ModeNo = 0;
-
- SiS_Pr->UseCustomMode = false;
-
- if((IsCustom) && (SiS_CheckBuildCustomMode(pScrn, mode, pSiS->VBFlags))) {
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3, "Setting custom mode %dx%d\n",
- SiS_Pr->CHDisplay,
- (mode->Flags & V_INTERLACE ? SiS_Pr->CVDisplay * 2 :
- (mode->Flags & V_DBLSCAN ? SiS_Pr->CVDisplay / 2 :
- SiS_Pr->CVDisplay)));
-
- } else {
-
- /* Don't need vbflags here; checks done earlier */
- ModeNo = SiS_GetModeNumber(pScrn, mode, pSiS->VBFlags);
- if(!ModeNo) return false;
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3, "Setting standard mode 0x%x\n", ModeNo);
-
- }
-
- return(SiSSetMode(SiS_Pr, pScrn, ModeNo, true));
-}
-
-/*********************************************/
-/* X.org/XFree86: SiSBIOSSetModeCRT2() */
-/* for Dual-Head modes */
-/*********************************************/
-
-bool
-SiSBIOSSetModeCRT2(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom)
-{
- SISIOADDRESS BaseAddr = SiS_Pr->IOAddress;
- SISPtr pSiS = SISPTR(pScrn);
-#ifdef SISDUALHEAD
- SISEntPtr pSiSEnt = pSiS->entityPrivate;
-#endif
- unsigned short ModeIdIndex;
- unsigned short ModeNo = 0;
- unsigned char backupreg = 0;
-
- SiS_Pr->UseCustomMode = false;
-
- /* Remember: Custom modes for CRT2 are ONLY supported
- * -) on the 30x/B/C, and
- * -) if CRT2 is LCD or VGA, or CRT1 is LCDA
- */
-
- if((IsCustom) && (SiS_CheckBuildCustomMode(pScrn, mode, pSiS->VBFlags))) {
-
- ModeNo = 0xfe;
-
- } else {
-
- ModeNo = SiS_GetModeNumber(pScrn, mode, pSiS->VBFlags);
- if(!ModeNo) return false;
-
- }
-
- SiSRegInit(SiS_Pr, BaseAddr);
- SiSInitPtr(SiS_Pr);
- SiS_GetSysFlags(SiS_Pr);
-#if defined(i386) || defined(__i386) || defined(__i386__) || defined(__AMD64__) || defined(__amd64__) || defined(__x86_64__)
- SiS_Pr->SiS_VGAINFO = SiS_GetSetBIOSScratch(pScrn, 0x489, 0xff);
-#else
- SiS_Pr->SiS_VGAINFO = 0x11;
-#endif
-
- SiS_SetReg(SiS_Pr->SiS_P3c4,0x05,0x86);
-
- SiSInitPCIetc(SiS_Pr);
- SiSSetLVDSetc(SiS_Pr);
- SiSDetermineROMUsage(SiS_Pr);
-
- /* Save mode info so we can set it from within SetMode for CRT1 */
-#ifdef SISDUALHEAD
- if(pSiS->DualHeadMode) {
- pSiSEnt->CRT2ModeNo = ModeNo;
- pSiSEnt->CRT2DMode = mode;
- pSiSEnt->CRT2IsCustom = IsCustom;
- pSiSEnt->CRT2CR30 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x30);
- pSiSEnt->CRT2CR31 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x31);
- pSiSEnt->CRT2CR35 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x35);
- pSiSEnt->CRT2CR38 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x38);
-#if 0
- /* We can't set CRT2 mode before CRT1 mode is set - says who...? */
- if(pSiSEnt->CRT1ModeNo == -1) {
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "Setting CRT2 mode delayed until after setting CRT1 mode\n");
- return true;
- }
-#endif
- pSiSEnt->CRT2ModeSet = true;
- }
-#endif
-
- if(SiS_Pr->UseCustomMode) {
-
- unsigned short temptemp = SiS_Pr->CVDisplay;
-
- if(SiS_Pr->CModeFlag & DoubleScanMode) temptemp >>= 1;
- else if(SiS_Pr->CInfoFlag & InterlaceMode) temptemp <<= 1;
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "Setting custom mode %dx%d on CRT2\n",
- SiS_Pr->CHDisplay, temptemp);
-
- } else {
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "Setting standard mode 0x%x on CRT2\n", ModeNo);
-
- }
-
- SiS_UnLockCRT2(SiS_Pr);
-
- if(!SiS_Pr->UseCustomMode) {
- if(!(SiS_SearchModeID(SiS_Pr, &ModeNo, &ModeIdIndex))) return false;
- } else {
- ModeIdIndex = 0;
- }
-
- SiS_GetVBType(SiS_Pr);
-
- SiS_InitVB(SiS_Pr);
- if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
- if(SiS_Pr->ChipType >= SIS_315H) {
- SiS_ResetVB(SiS_Pr);
- SiS_SetRegOR(SiS_Pr->SiS_P3c4,0x32,0x10);
- SiS_SetRegOR(SiS_Pr->SiS_Part2Port,0x00,0x0c);
- backupreg = SiS_GetReg(SiS_Pr->SiS_P3d4,0x38);
- } else {
- backupreg = SiS_GetReg(SiS_Pr->SiS_P3d4,0x35);
- }
- }
-
- /* Get VB information (connectors, connected devices) */
- if(!SiS_Pr->UseCustomMode) {
- SiS_GetVBInfo(SiS_Pr, ModeNo, ModeIdIndex, 1);
- } else {
- /* If this is a custom mode, we don't check the modeflag for CRT2Mode */
- SiS_GetVBInfo(SiS_Pr, ModeNo, ModeIdIndex, 0);
- }
- SiS_SetYPbPr(SiS_Pr);
- SiS_SetTVMode(SiS_Pr, ModeNo, ModeIdIndex);
- SiS_GetLCDResInfo(SiS_Pr, ModeNo, ModeIdIndex);
- SiS_SetLowModeTest(SiS_Pr, ModeNo);
-
- SiS_ResetSegmentRegisters(SiS_Pr);
-
- /* Set mode on CRT2 */
- if( (SiS_Pr->SiS_VBType & VB_SISVB) ||
- (SiS_Pr->SiS_IF_DEF_LVDS == 1) ||
- (SiS_Pr->SiS_IF_DEF_CH70xx != 0) ||
- (SiS_Pr->SiS_IF_DEF_TRUMPION != 0) ) {
- SiS_SetCRT2Group(SiS_Pr, ModeNo);
- }
-
- SiS_StrangeStuff(SiS_Pr);
-
- SiS_DisplayOn(SiS_Pr);
- SiS_SetRegByte(SiS_Pr->SiS_P3c6,0xFF);
-
- if(SiS_Pr->ChipType >= SIS_315H) {
- if(SiS_Pr->SiS_IF_DEF_LVDS == 1) {
- if(!(SiS_IsDualEdge(SiS_Pr))) {
- SiS_SetRegAND(SiS_Pr->SiS_Part1Port,0x13,0xfb);
- }
- }
- }
-
- if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
- if(SiS_Pr->ChipType >= SIS_315H) {
- if(!SiS_Pr->SiS_ROMNew) {
- if(SiS_IsVAMode(SiS_Pr)) {
- SiS_SetRegOR(SiS_Pr->SiS_P3d4,0x35,0x01);
- } else {
- SiS_SetRegAND(SiS_Pr->SiS_P3d4,0x35,0xFE);
- }
- }
-
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x38,backupreg);
-
- if(SiS_GetReg(SiS_Pr->SiS_P3d4,0x30) & SetCRT2ToLCD) {
- SiS_SetRegAND(SiS_Pr->SiS_P3d4,0x38,0xfc);
- }
- } else if((SiS_Pr->ChipType == SIS_630) ||
- (SiS_Pr->ChipType == SIS_730)) {
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x35,backupreg);
- }
- }
-
- /* SetPitch: Adapt to virtual size & position */
- SiS_SetPitchCRT2(SiS_Pr, pScrn);
-
- SiS_Handle760(SiS_Pr);
-
- return true;
-}
-
-/*********************************************/
-/* X.org/XFree86: SiSBIOSSetModeCRT1() */
-/* for Dual-Head modes */
-/*********************************************/
-
-bool
-SiSBIOSSetModeCRT1(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom)
-{
- SISIOADDRESS BaseAddr = SiS_Pr->IOAddress;
- SISPtr pSiS = SISPTR(pScrn);
- unsigned short ModeIdIndex, ModeNo = 0;
- unsigned char backupreg = 0;
-#ifdef SISDUALHEAD
- SISEntPtr pSiSEnt = pSiS->entityPrivate;
- unsigned char backupcr30, backupcr31, backupcr38, backupcr35, backupp40d=0;
- bool backupcustom;
-#endif
-
- SiS_Pr->UseCustomMode = false;
-
- if((IsCustom) && (SiS_CheckBuildCustomMode(pScrn, mode, pSiS->VBFlags))) {
-
- unsigned short temptemp = SiS_Pr->CVDisplay;
-
- if(SiS_Pr->CModeFlag & DoubleScanMode) temptemp >>= 1;
- else if(SiS_Pr->CInfoFlag & InterlaceMode) temptemp <<= 1;
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "Setting custom mode %dx%d on CRT1\n",
- SiS_Pr->CHDisplay, temptemp);
- ModeNo = 0xfe;
-
- } else {
-
- ModeNo = SiS_GetModeNumber(pScrn, mode, 0); /* don't give VBFlags */
- if(!ModeNo) return false;
-
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "Setting standard mode 0x%x on CRT1\n", ModeNo);
- }
-
- SiSInitPtr(SiS_Pr);
- SiSRegInit(SiS_Pr, BaseAddr);
- SiS_GetSysFlags(SiS_Pr);
-#if defined(i386) || defined(__i386) || defined(__i386__) || defined(__AMD64__) || defined(__amd64__) || defined(__x86_64__)
- SiS_Pr->SiS_VGAINFO = SiS_GetSetBIOSScratch(pScrn, 0x489, 0xff);
-#else
- SiS_Pr->SiS_VGAINFO = 0x11;
-#endif
-
- SiS_SetReg(SiS_Pr->SiS_P3c4,0x05,0x86);
-
- SiSInitPCIetc(SiS_Pr);
- SiSSetLVDSetc(SiS_Pr);
- SiSDetermineROMUsage(SiS_Pr);
-
- SiS_UnLockCRT2(SiS_Pr);
-
- if(!SiS_Pr->UseCustomMode) {
- if(!(SiS_SearchModeID(SiS_Pr, &ModeNo, &ModeIdIndex))) return false;
- } else {
- ModeIdIndex = 0;
- }
-
- /* Determine VBType */
- SiS_GetVBType(SiS_Pr);
-
- SiS_InitVB(SiS_Pr);
- if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
- if(SiS_Pr->ChipType >= SIS_315H) {
- backupreg = SiS_GetReg(SiS_Pr->SiS_P3d4,0x38);
- } else {
- backupreg = SiS_GetReg(SiS_Pr->SiS_P3d4,0x35);
- }
- }
-
- /* Get VB information (connectors, connected devices) */
- /* (We don't care if the current mode is a CRT2 mode) */
- SiS_GetVBInfo(SiS_Pr, ModeNo, ModeIdIndex, 0);
- SiS_SetYPbPr(SiS_Pr);
- SiS_SetTVMode(SiS_Pr, ModeNo, ModeIdIndex);
- SiS_GetLCDResInfo(SiS_Pr, ModeNo, ModeIdIndex);
- SiS_SetLowModeTest(SiS_Pr, ModeNo);
-
- SiS_OpenCRTC(SiS_Pr);
-
- /* Set mode on CRT1 */
- SiS_SetCRT1Group(SiS_Pr, ModeNo, ModeIdIndex);
- if(SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA) {
- SiS_SetCRT2Group(SiS_Pr, ModeNo);
- }
-
- /* SetPitch: Adapt to virtual size & position */
- SiS_SetPitchCRT1(SiS_Pr, pScrn);
-
- SiS_HandleCRT1(SiS_Pr);
-
- SiS_StrangeStuff(SiS_Pr);
-
- SiS_CloseCRTC(SiS_Pr);
-
-#ifdef SISDUALHEAD
- if(pSiS->DualHeadMode) {
- pSiSEnt->CRT1ModeNo = ModeNo;
- pSiSEnt->CRT1DMode = mode;
- }
-#endif
-
- if(SiS_Pr->UseCustomMode) {
- SiS_Pr->CRT1UsesCustomMode = true;
- SiS_Pr->CSRClock_CRT1 = SiS_Pr->CSRClock;
- SiS_Pr->CModeFlag_CRT1 = SiS_Pr->CModeFlag;
- } else {
- SiS_Pr->CRT1UsesCustomMode = false;
- }
-
- /* Reset CRT2 if changing mode on CRT1 */
-#ifdef SISDUALHEAD
- if(pSiS->DualHeadMode) {
- if(pSiSEnt->CRT2ModeNo != -1) {
- xf86DrvMsgVerb(pScrn->scrnIndex, X_INFO, 3,
- "(Re-)Setting mode for CRT2\n");
- backupcustom = SiS_Pr->UseCustomMode;
- backupcr30 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x30);
- backupcr31 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x31);
- backupcr35 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x35);
- backupcr38 = SiS_GetReg(SiS_Pr->SiS_P3d4,0x38);
- if(SiS_Pr->SiS_VBType & VB_SISVB) {
- /* Backup LUT-enable */
- if(pSiSEnt->CRT2ModeSet) {
- backupp40d = SiS_GetReg(SiS_Pr->SiS_Part4Port,0x0d) & 0x08;
- }
- }
- if(SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA) {
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x30,pSiSEnt->CRT2CR30);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x31,pSiSEnt->CRT2CR31);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x35,pSiSEnt->CRT2CR35);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x38,pSiSEnt->CRT2CR38);
- }
-
- SiSBIOSSetModeCRT2(SiS_Pr, pSiSEnt->pScrn_1,
- pSiSEnt->CRT2DMode, pSiSEnt->CRT2IsCustom);
-
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x30,backupcr30);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x31,backupcr31);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x35,backupcr35);
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x38,backupcr38);
- if(SiS_Pr->SiS_VBType & VB_SISVB) {
- SiS_SetRegANDOR(SiS_Pr->SiS_Part4Port,0x0d, ~0x08, backupp40d);
- }
- SiS_Pr->UseCustomMode = backupcustom;
- }
- }
-#endif
-
- /* Warning: From here, the custom mode entries in SiS_Pr are
- * possibly overwritten
- */
-
- SiS_DisplayOn(SiS_Pr);
- SiS_SetRegByte(SiS_Pr->SiS_P3c6,0xFF);
-
- if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
- if(SiS_Pr->ChipType >= SIS_315H) {
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x38,backupreg);
- } else if((SiS_Pr->ChipType == SIS_630) ||
- (SiS_Pr->ChipType == SIS_730)) {
- SiS_SetReg(SiS_Pr->SiS_P3d4,0x35,backupreg);
- }
- }
-
- SiS_Handle760(SiS_Pr);
-
- /* Backup/Set ModeNo in BIOS scratch area */
- SiS_GetSetModeID(pScrn,ModeNo);
-
- return true;
-}
-#endif /* Linux_XF86 */
-
#ifndef GETBITSTR
#define BITMASK(h,l) (((unsigned)(1U << ((h)-(l)+1))-1)<<(l))
#define GENMASK(mask) BITMASK(1?mask,0?mask)
SiS_Pr->CVBlankStart = SiS_Pr->SiS_VGAVDE;
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
tempbx = SiS_Pr->SiS_VGAHT;
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) {
tempbx = SiS_Pr->PanelHT;
remaining = tempbx % 8;
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
/* OK for LCDA, LVDS */
tempbx = SiS_Pr->PanelHT - SiS_Pr->PanelXRes;
tempax = SiS_Pr->SiS_VGAHDE; /* not /2 ! */
SiS_Pr->CHTotal = SiS_Pr->CHBlankEnd = tempbx;
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_VGAHDE == SiS_Pr->PanelXRes) {
SiS_Pr->CHSyncStart = SiS_Pr->SiS_VGAHDE + ((SiS_Pr->PanelHRS + 1) & ~1);
SiS_Pr->CHSyncEnd = SiS_Pr->CHSyncStart + SiS_Pr->PanelHRE;
}
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
tempax = VGAHDE;
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) {
tempbx = SiS_Pr->PanelXRes;
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) {
tempax = SiS_Pr->PanelYRes;
} else if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
/* Stupid hack for 640x400/320x200 */
if(SiS_Pr->SiS_LCDResInfo == Panel_1024x768) {
if((tempax + tempbx) == 438) tempbx += 16;
if(modeflag & DoubleScanMode) tempax |= 0x80;
SiS_SetRegANDOR(SiS_Pr->SiS_P3d4,0x09,0x5F,tempax);
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "%d %d %d %d %d %d %d %d (%d %d %d %d)\n",
- SiS_Pr->CHDisplay, SiS_Pr->CHSyncStart, SiS_Pr->CHSyncEnd, SiS_Pr->CHTotal,
- SiS_Pr->CVDisplay, SiS_Pr->CVSyncStart, SiS_Pr->CVSyncEnd, SiS_Pr->CVTotal,
- SiS_Pr->CHBlankStart, SiS_Pr->CHBlankEnd, SiS_Pr->CVBlankStart, SiS_Pr->CVBlankEnd);
- xf86DrvMsg(0, X_INFO, " {{0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,\n",
- SiS_Pr->CCRT1CRTC[0], SiS_Pr->CCRT1CRTC[1],
- SiS_Pr->CCRT1CRTC[2], SiS_Pr->CCRT1CRTC[3],
- SiS_Pr->CCRT1CRTC[4], SiS_Pr->CCRT1CRTC[5],
- SiS_Pr->CCRT1CRTC[6], SiS_Pr->CCRT1CRTC[7]);
- xf86DrvMsg(0, X_INFO, " 0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,\n",
- SiS_Pr->CCRT1CRTC[8], SiS_Pr->CCRT1CRTC[9],
- SiS_Pr->CCRT1CRTC[10], SiS_Pr->CCRT1CRTC[11],
- SiS_Pr->CCRT1CRTC[12], SiS_Pr->CCRT1CRTC[13],
- SiS_Pr->CCRT1CRTC[14], SiS_Pr->CCRT1CRTC[15]);
- xf86DrvMsg(0, X_INFO, " 0x%02x}},\n", SiS_Pr->CCRT1CRTC[16]);
-#endif
-#endif
}
void
SiS_Generic_ConvertCRData(struct SiS_Private *SiS_Pr, unsigned char *crdata,
int xres, int yres,
-#ifdef SIS_XORG_XF86
- DisplayModePtr current
-#endif
-#ifdef SIS_LINUX_KERNEL
struct fb_var_screeninfo *var, bool writeres
-#endif
)
{
unsigned short HRE, HBE, HRS, HBS, HDE, HT;
D = B - F - C;
-#ifdef SIS_XORG_XF86
- current->HDisplay = (E * 8);
- current->HSyncStart = (E * 8) + (F * 8);
- current->HSyncEnd = (E * 8) + (F * 8) + (C * 8);
- current->HTotal = (E * 8) + (F * 8) + (C * 8) + (D * 8);
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO,
- "H: A %d B %d C %d D %d E %d F %d HT %d HDE %d HRS %d HBS %d HBE %d HRE %d\n",
- A, B, C, D, E, F, HT, HDE, HRS, HBS, HBE, HRE);
-#else
- (void)VBS; (void)HBS; (void)A;
-#endif
-#endif
-#ifdef SIS_LINUX_KERNEL
if(writeres) var->xres = xres = E * 8;
var->left_margin = D * 8;
var->right_margin = F * 8;
var->hsync_len = C * 8;
-#endif
/* Vertical */
sr_data = crdata[13];
D = B - F - C;
-#ifdef SIS_XORG_XF86
- current->VDisplay = VDE + 1;
- current->VSyncStart = VRS + 1;
- current->VSyncEnd = ((VRS & ~0x1f) | VRE) + 1;
- if(VRE <= (VRS & 0x1f)) current->VSyncEnd += 32;
- current->VTotal = E + D + C + F;
-#if 0
- current->VDisplay = E;
- current->VSyncStart = E + D;
- current->VSyncEnd = E + D + C;
- current->VTotal = E + D + C + F;
-#endif
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO,
- "V: A %d B %d C %d D %d E %d F %d VT %d VDE %d VRS %d VBS %d VBE %d VRE %d\n",
- A, B, C, D, E, F, VT, VDE, VRS, VBS, VBE, VRE);
-#endif
-#endif
-#ifdef SIS_LINUX_KERNEL
if(writeres) var->yres = yres = E;
var->upper_margin = D;
var->lower_margin = F;
var->vsync_len = C;
-#endif
if((xres == 320) && ((yres == 200) || (yres == 240))) {
/* Terrible hack, but correct CRTC data for
* a negative D. The CRT controller does not
* seem to like correcting HRE to 50)
*/
-#ifdef SIS_XORG_XF86
- current->HDisplay = 320;
- current->HSyncStart = 328;
- current->HSyncEnd = 376;
- current->HTotal = 400;
-#endif
-#ifdef SIS_LINUX_KERNEL
var->left_margin = (400 - 376);
var->right_margin = (328 - 320);
var->hsync_len = (376 - 328);
-#endif
}
#ifndef _INIT_H_
#define _INIT_H_
-#include "osdef.h"
#include "initdef.h"
-#ifdef SIS_XORG_XF86
-#include "sis.h"
-#define SIS_NEED_inSISREG
-#define SIS_NEED_inSISREGW
-#define SIS_NEED_inSISREGL
-#define SIS_NEED_outSISREG
-#define SIS_NEED_outSISREGW
-#define SIS_NEED_outSISREGL
-#include "sis_regs.h"
-#endif
-
-#ifdef SIS_LINUX_KERNEL
#include "vgatypes.h"
#include "vstruct.h"
#ifdef SIS_CP
#include <linux/fb.h>
#include "sis.h"
#include <video/sisfb.h>
-#endif
/* Mode numbers */
static const unsigned short ModeIndex_320x200[] = {0x59, 0x41, 0x00, 0x4f};
{ 1280, 854, 8,16} /* 0x22 */
};
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
static const struct SiS_StandTable_S SiS_StandTable[]=
{
/* 0x00: MD_0_200 */
};
bool SiSInitPtr(struct SiS_Private *SiS_Pr);
-#ifdef SIS_XORG_XF86
-unsigned short SiS_GetModeID(int VGAEngine, unsigned int VBFlags, int HDisplay, int VDisplay,
- int Depth, bool FSTN, int LCDwith, int LCDheight);
-#endif
unsigned short SiS_GetModeID_LCD(int VGAEngine, unsigned int VBFlags, int HDisplay,
int VDisplay, int Depth, bool FSTN,
unsigned short CustomT, int LCDwith, int LCDheight,
void SiS_DisplayOn(struct SiS_Private *SiS_Pr);
void SiS_DisplayOff(struct SiS_Private *SiS_Pr);
void SiSRegInit(struct SiS_Private *SiS_Pr, SISIOADDRESS BaseAddr);
-#ifndef SIS_LINUX_KERNEL
-void SiSSetLVDSetc(struct SiS_Private *SiS_Pr);
-#endif
void SiS_SetEnableDstn(struct SiS_Private *SiS_Pr, int enable);
void SiS_SetEnableFstn(struct SiS_Private *SiS_Pr, int enable);
unsigned short SiS_GetModeFlag(struct SiS_Private *SiS_Pr, unsigned short ModeNo,
unsigned short ModeIdIndex);
bool SiSDetermineROMLayout661(struct SiS_Private *SiS_Pr);
-#ifndef SIS_LINUX_KERNEL
-void SiS_GetVBType(struct SiS_Private *SiS_Pr);
-#endif
bool SiS_SearchModeID(struct SiS_Private *SiS_Pr, unsigned short *ModeNo,
unsigned short *ModeIdIndex);
unsigned short ModeIdIndex);
unsigned short SiS_GetOffset(struct SiS_Private *SiS_Pr,unsigned short ModeNo,
unsigned short ModeIdIndex, unsigned short RRTI);
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
void SiS_GetFIFOThresholdIndex300(struct SiS_Private *SiS_Pr, unsigned short *idx1,
unsigned short *idx2);
unsigned short SiS_GetFIFOThresholdB300(unsigned short idx1, unsigned short idx2);
unsigned short SiS_GetLatencyFactor630(struct SiS_Private *SiS_Pr, unsigned short index);
#endif
void SiS_LoadDAC(struct SiS_Private *SiS_Pr, unsigned short ModeNo, unsigned short ModeIdIndex);
-#ifdef SIS_XORG_XF86
-bool SiSSetMode(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn, unsigned short ModeNo,
- bool dosetpitch);
-bool SiSBIOSSetMode(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom);
-bool SiSBIOSSetModeCRT2(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom);
-bool SiSBIOSSetModeCRT1(struct SiS_Private *SiS_Pr, ScrnInfoPtr pScrn,
- DisplayModePtr mode, bool IsCustom);
-#endif
-#ifdef SIS_LINUX_KERNEL
bool SiSSetMode(struct SiS_Private *SiS_Pr, unsigned short ModeNo);
-#endif
void SiS_CalcCRRegisters(struct SiS_Private *SiS_Pr, int depth);
void SiS_CalcLCDACRT1Timing(struct SiS_Private *SiS_Pr, unsigned short ModeNo,
unsigned short ModeIdIndex);
-#ifdef SIS_XORG_XF86
-void SiS_Generic_ConvertCRData(struct SiS_Private *SiS_Pr, unsigned char *crdata, int xres,
- int yres, DisplayModePtr current);
-#endif
-#ifdef SIS_LINUX_KERNEL
void SiS_Generic_ConvertCRData(struct SiS_Private *SiS_Pr, unsigned char *crdata, int xres,
int yres, struct fb_var_screeninfo *var, bool writeres);
-#endif
/* From init301.c: */
extern void SiS_GetVBInfo(struct SiS_Private *SiS_Pr, unsigned short ModeNo,
extern bool SiS_IsVAMode(struct SiS_Private *);
extern bool SiS_IsDualEdge(struct SiS_Private *);
-#ifdef SIS_XORG_XF86
-/* From other modules: */
-extern unsigned short SiS_CheckBuildCustomMode(ScrnInfoPtr pScrn, DisplayModePtr mode,
- unsigned int VBFlags);
-extern unsigned char SiS_GetSetBIOSScratch(ScrnInfoPtr pScrn, unsigned short offset,
- unsigned char value);
-extern unsigned char SiS_GetSetModeID(ScrnInfoPtr pScrn, unsigned char id);
-extern unsigned short SiS_GetModeNumber(ScrnInfoPtr pScrn, DisplayModePtr mode,
- unsigned int VBFlags);
-#endif
-
-#ifdef SIS_LINUX_KERNEL
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
extern unsigned int sisfb_read_nbridge_pci_dword(struct SiS_Private *SiS_Pr, int reg);
extern void sisfb_write_nbridge_pci_dword(struct SiS_Private *SiS_Pr, int reg,
unsigned int val);
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
extern void sisfb_write_nbridge_pci_byte(struct SiS_Private *SiS_Pr, int reg,
unsigned char val);
extern unsigned int sisfb_read_mio_pci_word(struct SiS_Private *SiS_Pr, int reg);
#endif
-#endif
#endif
#include "init301.h"
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
#include "oem300.h"
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
#include "oem310.h"
#endif
#define SiS_I2CDELAYSHORT 150
static unsigned short SiS_GetBIOSLCDResInfo(struct SiS_Private *SiS_Pr);
-#ifdef SIS_LINUX_KERNEL
static void SiS_SetCH70xx(struct SiS_Private *SiS_Pr, unsigned short reg, unsigned char val);
-#endif
/*********************************************/
/* HELPER: Lock/Unlock CRT2 */
SiS_SetRegOR(SiS_Pr->SiS_Part1Port,0x24,0x01);
}
-#ifdef SIS_LINUX_KERNEL
static
-#endif
void
SiS_LockCRT2(struct SiS_Private *SiS_Pr)
{
/* HELPER: Get Pointer to LCD structure */
/*********************************************/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static unsigned char *
GetLCDStructPtr661(struct SiS_Private *SiS_Pr)
{
/* HELPER: GET SOME DATA FROM BIOS ROM */
/*********************************************/
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static bool
SiS_CR36BIOSWord23b(struct SiS_Private *SiS_Pr)
{
SiS_GetReg(SiS_Pr->SiS_P3c4, 0x05);
}
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
static void
SiS_GenericDelay(struct SiS_Private *SiS_Pr, unsigned short delay)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
SiS_LongDelay(struct SiS_Private *SiS_Pr, unsigned short delay)
{
}
#endif
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
static void
SiS_ShortDelay(struct SiS_Private *SiS_Pr, unsigned short delay)
{
static void
SiS_PanelDelay(struct SiS_Private *SiS_Pr, unsigned short DelayTime)
{
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
unsigned short PanelID, DelayIndex, Delay=0;
#endif
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
PanelID = SiS_GetReg(SiS_Pr->SiS_P3d4,0x36);
if(SiS_Pr->SiS_VBType & VB_SISVB) {
}
SiS_ShortDelay(SiS_Pr, Delay);
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if((SiS_Pr->ChipType >= SIS_661) ||
(SiS_Pr->ChipType <= SIS_315PRO) ||
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
SiS_PanelDelayLoop(struct SiS_Private *SiS_Pr, unsigned short DelayTime, unsigned short DelayLoop)
{
while((!(SiS_GetRegByte(SiS_Pr->SiS_P3da) & 0x08)) && --watchdog);
}
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
static void
SiS_WaitRetrace2(struct SiS_Private *SiS_Pr, unsigned short reg)
{
SiS_WaitVBRetrace(struct SiS_Private *SiS_Pr)
{
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
if(!(SiS_GetReg(SiS_Pr->SiS_Part1Port,0x00) & 0x20)) return;
}
}
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(!(SiS_GetReg(SiS_Pr->SiS_Part1Port,0x00) & 0x40)) {
SiS_WaitRetrace1(SiS_Pr);
} else {
/* HELPER: MISC */
/*********************************************/
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static bool
SiS_Is301B(struct SiS_Private *SiS_Pr)
{
bool
SiS_IsDualEdge(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if((SiS_Pr->ChipType != SIS_650) || (SiS_GetReg(SiS_Pr->SiS_P3d4,0x5f) & 0xf0)) {
if(SiS_GetReg(SiS_Pr->SiS_P3d4,0x38) & EnableDualEdge) return true;
bool
SiS_IsVAMode(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short flag;
if(SiS_Pr->ChipType >= SIS_315H) {
return false;
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsVAorLCD(struct SiS_Private *SiS_Pr)
{
static bool
SiS_IsDualLink(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if((SiS_CRT2IsLCD(SiS_Pr)) ||
(SiS_IsVAMode(SiS_Pr))) {
return false;
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_TVEnabled(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_LCDAEnabled(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_WeHaveBacklightCtrl(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsNotM650orLater(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsYPbPr(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsChScart(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsTVOrYPbPrOrScart(struct SiS_Private *SiS_Pr)
{
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static bool
SiS_IsLCDOrLCDA(struct SiS_Private *SiS_Pr)
{
/*********************************************/
/* Setup general purpose IO for Chrontel communication */
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
void
SiS_SetChrontelGPIO(struct SiS_Private *SiS_Pr, unsigned short myvbinfo)
{
if(!(SiS_Pr->SiS_ChSW)) return;
-#ifdef SIS_LINUX_KERNEL
acpibase = sisfb_read_lpc_pci_dword(SiS_Pr, 0x74);
-#else
- acpibase = pciReadLong(0x00000800, 0x74);
-#endif
acpibase &= 0xFFFF;
if(!acpibase) return;
temp = SiS_GetRegShort((acpibase + 0x3c)); /* ACPI register 0x3c: GP Event 1 I/O mode select */
tempax &= (DriverMode | LoadDACFlag | SetNotSimuMode | SetPALTV);
tempbx |= tempax;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if(SiS_Pr->SiS_VBType & VB_SISLCDA) {
if(ModeNo == 0x03) {
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
if(!(SiS_Pr->SiS_VBType & VB_SISVGA2)) {
tempbx &= ~(SetCRT2ToRAMDAC);
SiS_Pr->SiS_VBInfo = tempbx;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->ChipType == SIS_630) {
SiS_SetChrontelGPIO(SiS_Pr, SiS_Pr->SiS_VBInfo);
}
#endif
-#ifdef SIS_LINUX_KERNEL
#if 0
printk(KERN_DEBUG "sisfb: (init301: VBInfo= 0x%04x, SetFlag=0x%04x)\n",
SiS_Pr->SiS_VBInfo, SiS_Pr->SiS_SetFlag);
#endif
-#endif
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_PROBED, "(init301: VBInfo=0x%04x, SetFlag=0x%04x)\n",
- SiS_Pr->SiS_VBInfo, SiS_Pr->SiS_SetFlag);
-#endif
-#endif
}
/*********************************************/
}
SiS_Pr->SiS_VBInfo &= ~SetPALTV;
-
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "(init301: TVMode %x, VBInfo %x)\n", SiS_Pr->SiS_TVMode, SiS_Pr->SiS_VBInfo);
-#endif
-#endif
}
/*********************************************/
static void
SiS_GetLCDInfoBIOS(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned char *ROMAddr;
unsigned short temp;
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Paneldata driver: [%d %d] [H %d %d] [V %d %d] [C %d 0x%02x 0x%02x]\n",
- SiS_Pr->PanelHT, SiS_Pr->PanelVT,
- SiS_Pr->PanelHRS, SiS_Pr->PanelHRE,
- SiS_Pr->PanelVRS, SiS_Pr->PanelVRE,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].CLOCK,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].Part4_A,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].Part4_B);
-#endif
-#endif
-
if((ROMAddr = GetLCDStructPtr661(SiS_Pr))) {
if((temp = SISGETROMW(6)) != SiS_Pr->PanelHT) {
SiS_Pr->SiS_NeedRomModeData = true;
SiS_Pr->SiS_VCLKData[VCLK_CUSTOM_315].SR2C =
SiS_Pr->SiS_VBVCLKData[VCLK_CUSTOM_315].Part4_B = ROMAddr[20];
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Paneldata BIOS: [%d %d] [H %d %d] [V %d %d] [C %d 0x%02x 0x%02x]\n",
- SiS_Pr->PanelHT, SiS_Pr->PanelVT,
- SiS_Pr->PanelHRS, SiS_Pr->PanelHRE,
- SiS_Pr->PanelVRS, SiS_Pr->PanelVRE,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].CLOCK,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].Part4_A,
- SiS_Pr->SiS_VBVCLKData[SiS_Pr->PanelVCLKIdx315].Part4_B);
-#endif
-#endif
-
}
#endif
}
{
unsigned short temp,modeflag,resinfo=0,modexres=0,modeyres=0;
bool panelcanscale = false;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
static const unsigned char SiS300SeriesLCDRes[] =
{ 0, 1, 2, 3, 7, 4, 5, 8,
0, 0, 10, 0, 0, 0, 0, 15 };
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned char *myptr = NULL;
#endif
SiS_Pr->SiS_LCDTypeInfo = (temp & 0x0F) - 1;
}
temp &= 0x0f;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->ChipType < SIS_315H) {
/* Very old BIOSes only know 7 sizes (NetVista 2179, 1.01g) */
if(SiS_Pr->SiS_VBType & VB_SIS301) {
#endif
/* Translate to our internal types */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType == SIS_550) {
if (temp == Panel310_1152x768) temp = Panel_320x240_2; /* Verified working */
else if(temp == Panel310_320x240_2) temp = Panel_320x240_2;
SiS_Pr->SiS_LCDResInfo = temp;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_IF_DEF_LVDS == 1) {
if(SiS_Pr->SiS_CustomT == CUT_BARCO1366) {
SiS_Pr->SiS_LCDResInfo = Panel_Barco1366;
else if(SiS_Pr->UsePanelScaler == 1) SiS_Pr->SiS_LCDInfo |= DontExpandLCD;
/* Dual link, Pass 1:1 BIOS default, etc. */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_661) {
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) {
if(temp & 0x08) SiS_Pr->SiS_LCDInfo |= LCDPass11;
}
}
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_IF_DEF_LVDS == 1) {
if(SiS_Pr->SiS_CustomT == CUT_PANEL848 || SiS_Pr->SiS_CustomT == CUT_PANEL856) {
SiS_Pr->SiS_LCDInfo = 0x80 | 0x40 | 0x20; /* neg h/v sync, RGB24(D0 = 0) */
SiS_Pr->SiS_SetFlag |= LCDVESATiming;
}
-#ifdef SIS_LINUX_KERNEL
#if 0
printk(KERN_DEBUG "sisfb: (LCDInfo=0x%04x LCDResInfo=0x%02x LCDTypeInfo=0x%02x)\n",
SiS_Pr->SiS_LCDInfo, SiS_Pr->SiS_LCDResInfo, SiS_Pr->SiS_LCDTypeInfo);
#endif
-#endif
-#ifdef SIS_XORG_XF86
- xf86DrvMsgVerb(0, X_PROBED, 4,
- "(init301: LCDInfo=0x%04x LCDResInfo=0x%02x LCDTypeInfo=0x%02x SetFlag=0x%04x)\n",
- SiS_Pr->SiS_LCDInfo, SiS_Pr->SiS_LCDResInfo, SiS_Pr->SiS_LCDTypeInfo, SiS_Pr->SiS_SetFlag);
-#endif
}
/*********************************************/
VCLKIndex = SiS_Pr->PanelVCLKIdx315;
}
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
/* Special Timing: Barco iQ Pro R series */
if(SiS_Pr->SiS_CustomT == CUT_BARCO1366) VCLKIndex = 0x44;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "VCLKIndex %d (0x%x)\n", VCLKIndex, VCLKIndex);
-#endif
-#endif
-
return VCLKIndex;
}
{
unsigned short i, j, modeflag, tempah=0;
short tempcl;
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
unsigned short tempbl;
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
unsigned short tempah2, tempbl2;
#endif
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* ---- 300 series ---- */
+#ifdef CONFIG_FB_SIS_300 /* ---- 300 series ---- */
/* For 301BDH: (with LCD via LVDS) */
if(SiS_Pr->SiS_VBType & VB_NoLCD) {
if(SiS_Pr->SiS_VBInfo & SetInSlaveMode) tempah ^= 0xA0;
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* ------- 315/330 series ------ */
+#ifdef CONFIG_FB_SIS_315 /* ------- 315/330 series ------ */
if(ModeNo > 0x13) {
tempcl -= ModeVGA;
if(SiS_Pr->SiS_VBInfo & SetInSlaveMode) tempah ^= 0x50;
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
if(SiS_Pr->ChipType < SIS_315H) {
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x00,tempah);
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->SiS_IF_DEF_LVDS == 1) {
SiS_SetRegANDOR(SiS_Pr->SiS_Part1Port,0x00,0xa0,tempah);
} else if(SiS_Pr->SiS_VBType & VB_SISVB) {
if(SiS_Pr->ChipType >= SIS_315H) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
/* LVDS can only be slave in 8bpp modes */
tempah = 0x80;
if((modeflag & CRT2Mode) && (SiS_Pr->SiS_ModeType > ModeVGA)) {
} else {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
tempah = 0;
if( (!(SiS_Pr->SiS_VBInfo & SetInSlaveMode)) && (SiS_Pr->SiS_ModeType > ModeVGA) ) {
tempah |= 0x02;
if(SiS_Pr->ChipType >= SIS_315H) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
/* unsigned char bridgerev = SiS_GetReg(SiS_Pr->SiS_Part4Port,0x01); */
/* The following is nearly unpreditable and varies from machine
SiS_SetRegANDOR(SiS_Pr->SiS_Part4Port,0x23,tempbl,tempah);
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
} else if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
SiS_SetRegAND(SiS_Pr->SiS_Part4Port,0x21,0x3f);
if((SiS_Pr->SiS_VBInfo & DisableCRT2Display) ||
} else { /* LVDS */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if(SiS_Pr->SiS_IF_DEF_CH70xx != 0) {
}
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->SiS_CustomT == CUT_COMPAQ1280) {
if(SiS_Pr->SiS_LCDResInfo == Panel_1280x1024) {
if(!(SiS_Pr->SiS_LCDInfo & DontExpandLCD)) {
case Panel_1280x1024: tempbx = 24; break;
case Panel_1400x1050: tempbx = 26; break;
case Panel_1600x1200: tempbx = 28; break;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
case Panel_Barco1366: tempbx = 80; break;
#endif
}
if(SiS_Pr->SiS_LCDInfo & LCDPass11) tempbx = 30;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_CustomT == CUT_BARCO1024) {
tempbx = 82;
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) tempbx++;
if((SiS_Pr->SiS_VBType & VB_SISVB) && (SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA)) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_CalcPanelLinkTiming(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
SiS_CalcLCDACRT1Timing(SiS_Pr, ModeNo, ModeIdIndex);
#endif
case 16: LVDSData = SiS_Pr->SiS_LVDS800x600Data_1; break;
case 18: LVDSData = SiS_Pr->SiS_LVDS1024x600Data_1; break;
case 20: LVDSData = SiS_Pr->SiS_LVDS1024x768Data_1; break;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
case 80: LVDSData = SiS_Pr->SiS_LVDSBARCO1366Data_1; break;
case 81: LVDSData = SiS_Pr->SiS_LVDSBARCO1366Data_2; break;
case 82: LVDSData = SiS_Pr->SiS_LVDSBARCO1024Data_1; break;
(SiS_Pr->SiS_SetFlag & SetDOSMode) ) {
SiS_Pr->SiS_HDE = SiS_Pr->PanelXRes;
SiS_Pr->SiS_VDE = SiS_Pr->PanelYRes;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_CustomT == CUT_BARCO1366) {
if(ResIndex < 0x08) {
SiS_Pr->SiS_HDE = 1280;
unsigned short resinfo, CRT2Index, ResIndex;
const struct SiS_LCDData *LCDPtr = NULL;
const struct SiS_TVData *TVPtr = NULL;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
short resinfo661;
#endif
} else {
modeflag = SiS_Pr->SiS_EModeIDTable[ModeIdIndex].Ext_ModeFlag;
resinfo = SiS_Pr->SiS_EModeIDTable[ModeIdIndex].Ext_RESINFO;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
resinfo661 = SiS_Pr->SiS_EModeIDTable[ModeIdIndex].ROMMODEIDX661;
if( (SiS_Pr->SiS_VBInfo & SetCRT2ToLCD) &&
(SiS_Pr->SiS_SetFlag & LCDVESATiming) &&
} else if( (!(SiS_Pr->SiS_LCDInfo & DontExpandLCD)) && (romptr) && (ROMAddr) ) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_Pr->SiS_RVBHCMAX = ROMAddr[romptr];
SiS_Pr->SiS_RVBHCFACT = ROMAddr[romptr+1];
SiS_Pr->SiS_VGAHT = ROMAddr[romptr+2] | ((ROMAddr[romptr+3] & 0x0f) << 8);
case Panel_1680x1050 :
case Panel_1680x1050 + 32: LCDPtr = SiS_Pr->SiS_LCD1680x1050Data; break;
case 100 : LCDPtr = SiS_Pr->SiS_NoScaleData; break;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
case 200 : LCDPtr = SiS310_ExtCompaq1280x1024Data; break;
case 201 : LCDPtr = SiS_Pr->SiS_St2LCD1280x1024Data; break;
#endif
default : LCDPtr = SiS_Pr->SiS_ExtLCD1024x768Data; break;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "GetCRT2Data: Index %d ResIndex %d\n", CRT2Index, ResIndex);
-#endif
-#endif
-
SiS_Pr->SiS_RVBHCMAX = (LCDPtr+ResIndex)->RVBHCMAX;
SiS_Pr->SiS_RVBHCFACT = (LCDPtr+ResIndex)->RVBHCFACT;
SiS_Pr->SiS_VGAHT = (LCDPtr+ResIndex)->VGAHT;
{
const struct SiS_LVDSDes *PanelDesPtr = NULL;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_VBInfo & SetCRT2ToLCD) {
if(SiS_Pr->ChipType < SIS_315H) {
if((SiS_Pr->SiS_VBType & VB_SIS30xBLV) && (SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA)) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->SiS_LCDInfo & DontExpandLCD) {
/* non-pass 1:1 only, see above */
if(SiS_Pr->SiS_VGAHDE != SiS_Pr->PanelXRes) {
} else {
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
switch(SiS_Pr->SiS_LCDResInfo) {
case Panel_800x600:
if(SiS_Pr->SiS_VGAVDE == SiS_Pr->PanelYRes) {
}
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
switch(SiS_Pr->SiS_LCDResInfo) {
case Panel_1024x768:
case Panel_1280x1024:
if(SiS_Pr->ChipType < SIS_315H) {
if(!(modeflag & HalfDCLK)) SiS_Pr->SiS_LCDHDES = 320;
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->SiS_LCDResInfo == Panel_1024x768) SiS_Pr->SiS_LCDHDES = 480;
if(SiS_Pr->SiS_LCDResInfo == Panel_1400x1050) SiS_Pr->SiS_LCDHDES = 804;
if(SiS_Pr->SiS_LCDResInfo == Panel_1600x1200) SiS_Pr->SiS_LCDHDES = 704;
/* DISABLE VIDEO BRIDGE */
/*********************************************/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static int
SiS_HandlePWD(struct SiS_Private *SiS_Pr)
{
ret = 1;
}
SiS_SetRegANDOR(SiS_Pr->SiS_Part4Port,0x27,0x7f,temp);
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, 0, "Setting PWD %x\n", temp);
-#endif
-#endif
}
#endif
return ret;
void
SiS_DisableBridge(struct SiS_Private *SiS_Pr)
{
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short tempah, pushax=0, modenum;
#endif
unsigned short temp=0;
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* 300 series */
+#ifdef CONFIG_FB_SIS_300 /* 300 series */
if(!(SiS_CR36BIOSWord23b(SiS_Pr))) {
if(SiS_Pr->SiS_VBType & VB_SISLVDS) {
}
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* 315 series */
+#ifdef CONFIG_FB_SIS_315 /* 315 series */
int didpwd = 0;
bool custom1 = (SiS_Pr->SiS_CustomT == CUT_COMPAQ1280) ||
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
} else { /* ============ For 301 ================ */
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(!(SiS_CR36BIOSWord23b(SiS_Pr))) {
SiS_SetRegSR11ANDOR(SiS_Pr,0xF7,0x08);
SiS_PanelDelay(SiS_Pr, 3);
SiS_SetRegOR(SiS_Pr->SiS_P3c4,0x1E,0x20);
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x00,temp);
} else {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
SiS_SetRegAND(SiS_Pr->SiS_P3c4,0x1E,0xDF); /* disable CRT2 */
if( (!(SiS_CRT2IsLCD(SiS_Pr))) ||
(!(SiS_CR36BIOSWord23d(SiS_Pr))) ) {
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* 300 series */
+#ifdef CONFIG_FB_SIS_300 /* 300 series */
if(SiS_Pr->SiS_IF_DEF_CH70xx == 1) {
SiS_SetCH700x(SiS_Pr,0x0E,0x09);
SiS_SetRegSR11ANDOR(SiS_Pr,0xFB,0x04);
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* 315 series */
+#ifdef CONFIG_FB_SIS_315 /* 315 series */
if(!(SiS_IsNotM650orLater(SiS_Pr))) {
/*if(SiS_Pr->ChipType < SIS_340) { */ /* XGI needs this */
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
} /* 315 series */
* from outside the context of a mode switch!
* MUST call getVBType before calling this
*/
-#ifdef SIS_LINUX_KERNEL
static
-#endif
void
SiS_EnableBridge(struct SiS_Private *SiS_Pr)
{
unsigned short temp=0, tempah;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short temp1, pushax=0;
bool delaylong = false;
#endif
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* 300 series */
+#ifdef CONFIG_FB_SIS_300 /* 300 series */
if(SiS_CRT2IsLCD(SiS_Pr)) {
if(SiS_Pr->SiS_VBType & VB_SISLVDS) {
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* 315 series */
+#ifdef CONFIG_FB_SIS_315 /* 315 series */
#ifdef SET_EMI
unsigned char r30=0, r31=0, r32=0, r33=0, cr36=0;
SiS_SetRegAND(SiS_Pr->SiS_Part1Port,0x00,0x7f);
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* 300 series */
+#ifdef CONFIG_FB_SIS_300 /* 300 series */
if(SiS_CRT2IsLCD(SiS_Pr)) {
if(SiS_Pr->ChipType == SIS_730) {
}
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* 315 series */
+#ifdef CONFIG_FB_SIS_315 /* 315 series */
if(!(SiS_IsNotM650orLater(SiS_Pr))) {
/*if(SiS_Pr->ChipType < SIS_340) {*/ /* XGI needs this */
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
} /* 310 series */
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* ---- 300 series --- */
+#ifdef CONFIG_FB_SIS_300 /* ---- 300 series --- */
if(SiS_Pr->SiS_VBType & VB_SIS30xBLV) { /* 630 - 301B(-DH) */
}
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* ------- 315 series ------ */
+#ifdef CONFIG_FB_SIS_315 /* ------- 315 series ------ */
if(SiS_Pr->SiS_VBType & VB_SISLVDS) { /* 315 - LVDS */
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
}
}
/* Set CRT2 FIFO on 300/540/630/730 */
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static void
SiS_SetCRT2FIFO_300(struct SiS_Private *SiS_Pr,unsigned short ModeNo)
{
} else {
-#ifdef SIS_LINUX_KERNEL
pci50 = sisfb_read_nbridge_pci_dword(SiS_Pr, 0x50);
pciA0 = sisfb_read_nbridge_pci_dword(SiS_Pr, 0xa0);
-#else
- pci50 = pciReadLong(0x00000000, 0x50);
- pciA0 = pciReadLong(0x00000000, 0xA0);
-#endif
if(SiS_Pr->ChipType == SIS_730) {
#endif
/* Set CRT2 FIFO on 315/330 series */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void
SiS_SetCRT2FIFO_310(struct SiS_Private *SiS_Pr)
{
temp = SiS_GetRegByte((SiS_Pr->SiS_P3ca+0x02));
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x1b,temp); /* ? */
-
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "%d %d %d %d %d %d %d %d (%d %d %d %d)\n",
- SiS_Pr->CHDisplay, SiS_Pr->CHSyncStart, SiS_Pr->CHSyncEnd, SiS_Pr->CHTotal,
- SiS_Pr->CVDisplay, SiS_Pr->CVSyncStart, SiS_Pr->CVSyncEnd, SiS_Pr->CVTotal,
- SiS_Pr->CHBlankStart, SiS_Pr->CHBlankEnd, SiS_Pr->CVBlankStart, SiS_Pr->CVBlankEnd);
-
- xf86DrvMsg(0, X_INFO, " {{0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,\n",
- SiS_Pr->CCRT1CRTC[0], SiS_Pr->CCRT1CRTC[1],
- SiS_Pr->CCRT1CRTC[2], SiS_Pr->CCRT1CRTC[3],
- SiS_Pr->CCRT1CRTC[4], SiS_Pr->CCRT1CRTC[5],
- SiS_Pr->CCRT1CRTC[6], SiS_Pr->CCRT1CRTC[7]);
- xf86DrvMsg(0, X_INFO, " 0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,0x%02x,\n",
- SiS_Pr->CCRT1CRTC[8], SiS_Pr->CCRT1CRTC[9],
- SiS_Pr->CCRT1CRTC[10], SiS_Pr->CCRT1CRTC[11],
- SiS_Pr->CCRT1CRTC[12], SiS_Pr->CCRT1CRTC[13],
- SiS_Pr->CCRT1CRTC[14], SiS_Pr->CCRT1CRTC[15]);
- xf86DrvMsg(0, X_INFO, " 0x%02x}},\n", SiS_Pr->CCRT1CRTC[16]);
-#endif
-#endif
}
/* Setup panel link
unsigned short push2, tempax, tempbx, tempcx, temp;
unsigned int tempeax = 0, tempebx, tempecx, tempvcfact = 0;
bool islvds = false, issis = false, chkdclkfirst = false;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
unsigned short crt2crtc = 0;
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short pushcx;
#endif
if(ModeNo <= 0x13) {
modeflag = SiS_Pr->SiS_SModeIDTable[ModeIdIndex].St_ModeFlag;
resinfo = SiS_Pr->SiS_SModeIDTable[ModeIdIndex].St_ResInfo;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
crt2crtc = SiS_Pr->SiS_SModeIDTable[ModeIdIndex].St_CRT2CRTC;
#endif
} else if(SiS_Pr->UseCustomMode) {
} else {
modeflag = SiS_Pr->SiS_EModeIDTable[ModeIdIndex].Ext_ModeFlag;
resinfo = SiS_Pr->SiS_EModeIDTable[ModeIdIndex].Ext_RESINFO;
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
crt2crtc = SiS_Pr->SiS_RefIndex[RefreshRateTableIndex].Ext_CRT2CRTC;
#endif
}
}
}
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if((SiS_Pr->ChipType >= SIS_315H) && (SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA)) {
if(IS_SIS330) {
SiS_SetRegOR(SiS_Pr->SiS_Part1Port,0x2D,0x10);
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* 300 series */
+#ifdef CONFIG_FB_SIS_300 /* 300 series */
tempeax = SiS_Pr->SiS_VGAVDE << 6;
temp = (tempeax % (unsigned int)SiS_Pr->SiS_VDE);
tempeax = tempeax / (unsigned int)SiS_Pr->SiS_VDE;
temp = (unsigned short)(tempeax & 0x00FF);
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x1E,temp); /* BPLVCFACT */
tempvcfact = temp;
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* 315 series */
+#ifdef CONFIG_FB_SIS_315 /* 315 series */
tempeax = SiS_Pr->SiS_VGAVDE << 18;
tempebx = SiS_Pr->SiS_VDE;
temp = (tempeax % tempebx);
temp = (unsigned short)(tempecx & 0x00FF);
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x23,temp);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if(SiS_Pr->SiS_VBInfo & SetCRT2ToLCDA) {
if((islvds) || (SiS_Pr->SiS_VBInfo & VB_SISLVDS)) {
}
#endif
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->SiS_IF_DEF_TRUMPION) {
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
unsigned char *trumpdata;
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->SiS_IF_DEF_FSTN || SiS_Pr->SiS_IF_DEF_DSTN) {
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x25,0x00);
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x26,0x00);
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x45,0x0a);
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
/* Set Part 1 */
SiS_SetGroup1(struct SiS_Private *SiS_Pr, unsigned short ModeNo, unsigned short ModeIdIndex,
unsigned short RefreshRateTableIndex)
{
-#if defined(SIS300) || defined(SIS315H)
+#if defined(CONFIG_FB_SIS_300) || defined(CONFIG_FB_SIS_315)
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
#endif
unsigned short temp=0, tempax=0, tempbx=0, tempcx=0, bridgeadd=0;
unsigned short pushbx=0, CRT1Index=0, modeflag, resinfo=0;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short tempbl=0;
#endif
(SiS_Pr->SiS_VBInfo & SetInSlaveMode)) ) {
if(SiS_Pr->ChipType < SIS_315H ) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
SiS_SetCRT2FIFO_300(SiS_Pr, ModeNo);
#endif
} else {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_SetCRT2FIFO_310(SiS_Pr);
#endif
}
if(SiS_Pr->ChipType < SIS_315H ) {
-#ifdef SIS300 /* ------------- 300 series --------------*/
+#ifdef CONFIG_FB_SIS_300 /* ------------- 300 series --------------*/
temp = (SiS_Pr->SiS_VGAHT - 1) & 0x0FF; /* BTVGA2HT 0x08,0x09 */
SiS_SetReg(SiS_Pr->SiS_Part1Port,0x08,temp); /* CRT2 Horizontal Total */
bridgeadd = 12;
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* ------------------- 315/330 series --------------- */
+#ifdef CONFIG_FB_SIS_315 /* ------------------- 315/330 series --------------- */
tempcx = SiS_Pr->SiS_VGAHT; /* BTVGA2HT 0x08,0x09 */
if(modeflag & HalfDCLK) {
}
}
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
} /* 315/330 series */
if(SiS_Pr->ChipType < SIS_315H) {
-#ifdef SIS300 /* ---------- 300 series -------------- */
+#ifdef CONFIG_FB_SIS_300 /* ---------- 300 series -------------- */
if(SiS_Pr->SiS_VBType & VB_SISVB) {
temp = 0x20;
SiS_SetRegANDOR(SiS_Pr->SiS_Part1Port,0x13,~0x3C,temp); /* Panel Link Delay Compensation; (Software Command Reset; Power Saving) */
-#endif /* SIS300 */
+#endif /* CONFIG_FB_SIS_300 */
} else {
-#ifdef SIS315H /* --------------- 315/330 series ---------------*/
+#ifdef CONFIG_FB_SIS_315 /* --------------- 315/330 series ---------------*/
if(SiS_Pr->ChipType < SIS_661) {
if(modeflag & HalfDCLK) tempax |= 0x40;
SiS_SetRegANDOR(SiS_Pr->SiS_Part1Port,0x2C,0x3f,tempax);
-#endif /* SIS315H */
+#endif /* CONFIG_FB_SIS_315 */
}
/* SET PART 2 REGISTER GROUP */
/*********************************************/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static unsigned char *
SiS_GetGroup2CLVXPtr(struct SiS_Private *SiS_Pr, int tabletype)
{
}
#endif
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static void
SiS_Group2LCDSpecial(struct SiS_Private *SiS_Pr, unsigned short ModeNo, unsigned short crt2crtc)
{
unsigned int longtemp, PhaseIndex;
bool newtvphase;
const unsigned char *TimingPoint;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short resindex, CRT2Index;
const struct SiS_Part2PortTbl *CRT2Part2Ptr = NULL;
SiS_SetRegAND(SiS_Pr->SiS_Part2Port,0x17,0xFB);
SiS_SetRegAND(SiS_Pr->SiS_Part2Port,0x18,0xDF);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_GetCRT2Part2Ptr(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex,
&CRT2Index, &resindex)) {
switch(CRT2Index) {
/* Non-expanding: lcdvdes = tempcx = VT-1; lcdvdee = tempbx = VDE-1 */
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdvdes 0x%x lcdvdee 0x%x\n", tempcx, tempbx);
-#endif
-#endif
-
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x05,tempcx); /* lcdvdes */
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x06,tempbx); /* lcdvdee */
tempbx = SiS_Pr->CVSyncStart;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdvrs 0x%x\n", tempbx);
-#endif
-#endif
-
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x04,tempbx); /* lcdvrs */
temp = (tempbx >> 4) & 0xF0;
temp |= (SiS_Pr->CVSyncEnd & 0x0f);
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdvre[3:0] 0x%x\n", (temp & 0x0f));
-#endif
-#endif
-
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x01,temp);
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
SiS_Group2LCDSpecial(SiS_Pr, ModeNo, crt2crtc);
#endif
tempax >>= 1;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdhdee 0x%x\n", tempbx);
-#endif
-#endif
-
tempbx += bridgeoffset;
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x23,tempbx); /* lcdhdee */
tempbx += bridgeoffset;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdhrs 0x%x\n", tempbx);
-#endif
-#endif
-
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x1C,tempbx); /* lcdhrs */
SiS_SetRegANDOR(SiS_Pr->SiS_Part2Port,0x1D,0x0F,((tempbx >> 4) & 0xf0));
tempbx += bridgeoffset;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "lcdhre 0x%x\n", tempbx);
-#endif
-#endif
-
SiS_SetReg(SiS_Pr->SiS_Part2Port,0x21,tempbx); /* lcdhre */
SiS_SetGroup2_Tail(SiS_Pr, ModeNo);
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
SiS_Set300Part2Regs(SiS_Pr, ModeIdIndex, RefreshRateTableIndex, ModeNo);
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
} /* CRT2-LCD from table */
#endif
}
/* SET PART 4 REGISTER GROUP */
/*********************************************/
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
#if 0
static void
SiS_ShiftXPos(struct SiS_Private *SiS_Pr, int shift)
if(SiS_Pr->SiS_IF_DEF_CH70xx == 1) {
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
/* Chrontel 7005 - I assume that it does not come with a 315 series chip */
/* Chrontel 7019 - assumed that it does not come with a 300 series chip */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
unsigned short temp;
}
-#ifdef SIS315H /* ----------- 315 series only ---------- */
+#ifdef CONFIG_FB_SIS_315 /* ----------- 315 series only ---------- */
void
SiS_Chrontel701xBLOn(struct SiS_Private *SiS_Pr)
bool
SiS_SetCRT2Group(struct SiS_Private *SiS_Pr, unsigned short ModeNo)
{
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
unsigned char *ROMAddr = SiS_Pr->VirtualRomBase;
#endif
unsigned short ModeIdIndex, RefreshRateTableIndex;
SiS_GetLVDSDesData(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "(init301: LCDHDES 0x%03x LCDVDES 0x%03x)\n", SiS_Pr->SiS_LCDHDES, SiS_Pr->SiS_LCDVDES);
- xf86DrvMsg(0, X_INFO, "(init301: HDE 0x%03x VDE 0x%03x)\n", SiS_Pr->SiS_HDE, SiS_Pr->SiS_VDE);
- xf86DrvMsg(0, X_INFO, "(init301: VGAHDE 0x%03x VGAVDE 0x%03x)\n", SiS_Pr->SiS_VGAHDE, SiS_Pr->SiS_VGAVDE);
- xf86DrvMsg(0, X_INFO, "(init301: HT 0x%03x VT 0x%03x)\n", SiS_Pr->SiS_HT, SiS_Pr->SiS_VT);
- xf86DrvMsg(0, X_INFO, "(init301: VGAHT 0x%03x VGAVT 0x%03x)\n", SiS_Pr->SiS_VGAHT, SiS_Pr->SiS_VGAVT);
-#endif
-#endif
-
if(SiS_Pr->SiS_SetFlag & LowModeTests) {
SiS_SetGroup1(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
}
if(SiS_Pr->SiS_SetFlag & LowModeTests) {
SiS_SetGroup2(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_SetGroup2_C_ELV(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
#endif
SiS_SetGroup3(SiS_Pr, ModeNo, ModeIdIndex);
SiS_SetGroup4(SiS_Pr, ModeNo, ModeIdIndex, RefreshRateTableIndex);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_SetGroup4_C_ELV(SiS_Pr, ModeNo, ModeIdIndex);
#endif
SiS_SetGroup5(SiS_Pr, ModeNo, ModeIdIndex);
if(SiS_Pr->SiS_IF_DEF_CH70xx != 0) {
if(SiS_Pr->SiS_VBInfo & (SetCRT2ToLCD | SetCRT2ToLCDA)) {
if(SiS_Pr->SiS_IF_DEF_CH70xx == 2) {
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
SiS_SetCH701xForLCD(SiS_Pr);
#endif
}
}
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
if(SiS_Pr->ChipType < SIS_315H) {
if(SiS_Pr->SiS_SetFlag & LowModeTests) {
if(SiS_Pr->SiS_UseOEM) {
}
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
if(SiS_Pr->ChipType >= SIS_315H) {
if(SiS_Pr->SiS_SetFlag & LowModeTests) {
if(SiS_Pr->ChipType < SIS_661) {
}
}
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static unsigned char *
SiS_SetTrumpBlockLoop(struct SiS_Private *SiS_Pr, unsigned char *dataptr)
{
dataptr = SiS_SetTrumpBlockLoop(SiS_Pr, dataptr);
if(!dataptr) return false;
}
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Trumpion block success\n");
-#endif
-#endif
return true;
}
#endif
SiS_SetChReg(SiS_Pr, reg, val, 0);
}
-#ifdef SIS_LINUX_KERNEL
static
-#endif
void
SiS_SetCH70xx(struct SiS_Private *SiS_Pr, unsigned short reg, unsigned char val)
{
/* Read from Chrontel 70xx */
/* Parameter is [Register no (S7-S0)] */
-#ifdef SIS_LINUX_KERNEL
static
-#endif
unsigned short
SiS_GetCH70xx(struct SiS_Private *SiS_Pr, unsigned short tempbx)
{
}
/* Our own DDC functions */
-#ifndef SIS_XORG_XF86
static
-#endif
unsigned short
SiS_InitDDCRegs(struct SiS_Private *SiS_Pr, unsigned int VBFlags, int VGAEngine,
unsigned short adaptnum, unsigned short DDCdatatype, bool checkcr32,
SiS_SetupDDCN(SiS_Pr);
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "DDC Port %x Index %x Shift %d\n",
- SiS_Pr->SiS_DDC_Port, SiS_Pr->SiS_DDC_Index, temp);
-#endif
-#endif
return 0;
}
SiS_SetSwitchDDC2(SiS_Pr);
if(SiS_PrepareDDC(SiS_Pr)) {
SiS_SetStop(SiS_Pr);
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Probe: Prepare failed\n");
-#endif
-#endif
return 0xFFFF;
}
mask = 0xf0;
} else {
failed = true;
ret = 0xFFFF;
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Probe: Read 1 failed\n");
-#endif
-#endif
}
}
if(!failed) {
if(temp == value) ret = 0;
else {
ret = 0xFFFF;
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "Probe: Read 2 failed\n");
-#endif
-#endif
if(SiS_Pr->SiS_DDC_DeviceAddr == 0xa0) {
if(temp == 0x30) ret = 0;
}
return ret;
}
-#ifndef SIS_XORG_XF86
static
-#endif
unsigned short
SiS_ProbeDDC(struct SiS_Private *SiS_Pr)
{
return flag;
}
-#ifndef SIS_XORG_XF86
static
-#endif
unsigned short
SiS_ReadDDC(struct SiS_Private *SiS_Pr, unsigned short DDCdatatype, unsigned char *buffer)
{
temp = SiS_GetReg(SiS_Pr->SiS_DDC_Port,SiS_Pr->SiS_DDC_Index);
} while((!(temp & SiS_Pr->SiS_DDC_Clk)) && --watchdog);
if (!watchdog) {
-#ifdef SIS_XORG_XF86
-#ifdef TWDEBUG
- xf86DrvMsg(0, X_INFO, "SetClkHigh failed\n");
-#endif
-#endif
return 0xFFFF;
}
SiS_DDC2Delay(SiS_Pr,SiS_I2CDELAYSHORT);
/* =============== SiS 315/330 O.E.M. ================= */
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static unsigned short
GetRAMDACromptr(struct SiS_Private *SiS_Pr)
/* ================= SiS 300 O.E.M. ================== */
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static void
SetOEMLCDData2(struct SiS_Private *SiS_Pr, unsigned short ModeNo,unsigned short ModeIdIndex,
#ifndef _INIT301_H_
#define _INIT301_H_
-#include "osdef.h"
#include "initdef.h"
-#ifdef SIS_XORG_XF86
-#include "sis.h"
-#include "sis_regs.h"
-#endif
-
-#ifdef SIS_LINUX_KERNEL
#include "vgatypes.h"
#include "vstruct.h"
#ifdef SIS_CP
#include <linux/fb.h>
#include "sis.h"
#include <video/sisfb.h>
-#endif
static const unsigned char SiS_YPbPrTable[3][64] = {
{
0xFF,0xFF,
};
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
/* 661 et al LCD data structure (2.03.00) */
static const unsigned char SiS_LCDStruct661[] = {
/* 1024x768 */
};
#endif
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static unsigned char SiS300_TrumpionData[14][80] = {
{ 0x02,0x0A,0x0A,0x01,0x04,0x01,0x00,0x03,0x0D,0x00,0x0D,0x10,0x7F,0x00,0x80,0x02,
0x20,0x03,0x0B,0x00,0x90,0x01,0xC1,0x01,0x60,0x0C,0x30,0x10,0x00,0x00,0x04,0x23,
#endif
void SiS_UnLockCRT2(struct SiS_Private *SiS_Pr);
-#ifndef SIS_LINUX_KERNEL
-void SiS_LockCRT2(struct SiS_Private *SiS_Pr);
-#endif
void SiS_EnableCRT2(struct SiS_Private *SiS_Pr);
unsigned short SiS_GetRatePtr(struct SiS_Private *SiS_Pr, unsigned short ModeNo, unsigned short ModeIdIndex);
void SiS_WaitRetrace1(struct SiS_Private *SiS_Pr);
unsigned short RefreshRateTableIndex);
unsigned short SiS_GetResInfo(struct SiS_Private *SiS_Pr,unsigned short ModeNo,unsigned short ModeIdIndex);
void SiS_DisableBridge(struct SiS_Private *SiS_Pr);
-#ifndef SIS_LINUX_KERNEL
-void SiS_EnableBridge(struct SiS_Private *SiS_Pr);
-#endif
bool SiS_SetCRT2Group(struct SiS_Private *SiS_Pr, unsigned short ModeNo);
void SiS_SiS30xBLOn(struct SiS_Private *SiS_Pr);
void SiS_SiS30xBLOff(struct SiS_Private *SiS_Pr);
unsigned short SiS_GetCH700x(struct SiS_Private *SiS_Pr, unsigned short tempax);
void SiS_SetCH701x(struct SiS_Private *SiS_Pr, unsigned short reg, unsigned char val);
unsigned short SiS_GetCH701x(struct SiS_Private *SiS_Pr, unsigned short tempax);
-#ifndef SIS_LINUX_KERNEL
-void SiS_SetCH70xx(struct SiS_Private *SiS_Pr, unsigned short reg, unsigned char val);
-unsigned short SiS_GetCH70xx(struct SiS_Private *SiS_Pr, unsigned short tempax);
-#endif
void SiS_SetCH70xxANDOR(struct SiS_Private *SiS_Pr, unsigned short reg,
unsigned char orval,unsigned short andval);
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void SiS_Chrontel701xOn(struct SiS_Private *SiS_Pr);
static void SiS_Chrontel701xOff(struct SiS_Private *SiS_Pr);
static void SiS_ChrontelInitTVVSync(struct SiS_Private *SiS_Pr);
void SiS_Chrontel701xBLOff(struct SiS_Private *SiS_Pr);
#endif /* 315 */
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static bool SiS_SetTrumpionBlock(struct SiS_Private *SiS_Pr, unsigned char *dataptr);
void SiS_SetChrontelGPIO(struct SiS_Private *SiS_Pr, unsigned short myvbinfo);
#endif
unsigned short adaptnum, unsigned short DDCdatatype,
unsigned char *buffer, unsigned int VBFlags2);
-#ifdef SIS_XORG_XF86
-unsigned short SiS_InitDDCRegs(struct SiS_Private *SiS_Pr, unsigned int VBFlags,
- int VGAEngine, unsigned short adaptnum, unsigned short DDCdatatype,
- bool checkcr32, unsigned int VBFlags2);
-unsigned short SiS_ProbeDDC(struct SiS_Private *SiS_Pr);
-unsigned short SiS_ReadDDC(struct SiS_Private *SiS_Pr, unsigned short DDCdatatype,
- unsigned char *buffer);
-#else
static unsigned short SiS_InitDDCRegs(struct SiS_Private *SiS_Pr, unsigned int VBFlags,
int VGAEngine, unsigned short adaptnum, unsigned short DDCdatatype,
bool checkcr32, unsigned int VBFlags2);
static unsigned short SiS_ProbeDDC(struct SiS_Private *SiS_Pr);
static unsigned short SiS_ReadDDC(struct SiS_Private *SiS_Pr, unsigned short DDCdatatype,
unsigned char *buffer);
-#endif
static void SiS_SetSwitchDDC2(struct SiS_Private *SiS_Pr);
static unsigned short SiS_SetStart(struct SiS_Private *SiS_Pr);
static unsigned short SiS_SetStop(struct SiS_Private *SiS_Pr);
static void SiS_SendACK(struct SiS_Private *SiS_Pr, unsigned short yesno);
static unsigned short SiS_DoProbeDDC(struct SiS_Private *SiS_Pr);
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
static void SiS_OEM300Setting(struct SiS_Private *SiS_Pr,
unsigned short ModeNo, unsigned short ModeIdIndex, unsigned short RefTabindex);
static void SetOEMLCDData2(struct SiS_Private *SiS_Pr,
unsigned short ModeNo, unsigned short ModeIdIndex,unsigned short RefTableIndex);
#endif
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
static void SiS_OEM310Setting(struct SiS_Private *SiS_Pr,
unsigned short ModeNo,unsigned short ModeIdIndex, unsigned short RRTI);
static void SiS_OEM661Setting(struct SiS_Private *SiS_Pr,
extern void SiS_CalcCRRegisters(struct SiS_Private *SiS_Pr, int depth);
extern unsigned short SiS_GetRefCRTVCLK(struct SiS_Private *SiS_Pr, unsigned short Index, int UseWide);
extern unsigned short SiS_GetRefCRT1CRTC(struct SiS_Private *SiS_Pr, unsigned short Index, int UseWide);
-#ifdef SIS300
+#ifdef CONFIG_FB_SIS_300
extern void SiS_GetFIFOThresholdIndex300(struct SiS_Private *SiS_Pr, unsigned short *tempbx,
unsigned short *tempcl);
extern unsigned short SiS_GetFIFOThresholdB300(unsigned short tempbx, unsigned short tempcl);
extern unsigned short SiS_GetLatencyFactor630(struct SiS_Private *SiS_Pr, unsigned short index);
-#ifdef SIS_LINUX_KERNEL
extern unsigned int sisfb_read_nbridge_pci_dword(struct SiS_Private *SiS_Pr, int reg);
extern unsigned int sisfb_read_lpc_pci_dword(struct SiS_Private *SiS_Pr, int reg);
#endif
-#endif
#endif
* Author: Thomas Winischhofer <thomas@winischhofer.net>
*/
-#include "osdef.h"
#include "initdef.h"
#include "vgatypes.h"
#include "vstruct.h"
if(rateindex > 0) rateindex--;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
switch(ModeNo) {
case 0x5a: ModeNo = 0x50; break;
case 0x5b: ModeNo = 0x56;
if(rateindex > 0) rateindex--;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
switch(ModeNo) {
case 0x5a: ModeNo = 0x50; break;
case 0x5b: ModeNo = 0x56;
if(rateindex > 0) rateindex--;
-#ifdef SIS315H
+#ifdef CONFIG_FB_SIS_315
switch(ModeNo) {
case 0x5a: ModeNo = 0x50; break;
case 0x5b: ModeNo = 0x56;
+++ /dev/null
-/* $XFree86$ */
-/* $XdotOrg$ */
-/*
- * OS depending defines
- *
- * Copyright (C) 2001-2005 by Thomas Winischhofer, Vienna, Austria
- *
- * If distributed as part of the Linux kernel, the following license terms
- * apply:
- *
- * * This program is free software; you can redistribute it and/or modify
- * * it under the terms of the GNU General Public License as published by
- * * the Free Software Foundation; either version 2 of the named License,
- * * or any later version.
- * *
- * * This program is distributed in the hope that it will be useful,
- * * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * * GNU General Public License for more details.
- * *
- * * You should have received a copy of the GNU General Public License
- * * along with this program; if not, write to the Free Software
- * * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA
- *
- * Otherwise, the following license terms apply:
- *
- * * Redistribution and use in source and binary forms, with or without
- * * modification, are permitted provided that the following conditions
- * * are met:
- * * 1) Redistributions of source code must retain the above copyright
- * * notice, this list of conditions and the following disclaimer.
- * * 2) Redistributions in binary form must reproduce the above copyright
- * * notice, this list of conditions and the following disclaimer in the
- * * documentation and/or other materials provided with the distribution.
- * * 3) The name of the author may not be used to endorse or promote products
- * * derived from this software without specific prior written permission.
- * *
- * * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
- * * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
- * * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
- * * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
- * * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- * * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
- * * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
- * * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- * * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * Author: Thomas Winischhofer <thomas@winischhofer.net>
- * Silicon Integrated Systems, Inc. (used by permission)
- *
- */
-
-#ifndef _SIS_OSDEF_H_
-#define _SIS_OSDEF_H_
-
-/* The choices are: */
-#define SIS_LINUX_KERNEL /* Linux kernel framebuffer */
-#undef SIS_XORG_XF86 /* XFree86/X.org */
-
-#ifdef OutPortByte
-#undef OutPortByte
-#endif
-
-#ifdef OutPortWord
-#undef OutPortWord
-#endif
-
-#ifdef OutPortLong
-#undef OutPortLong
-#endif
-
-#ifdef InPortByte
-#undef InPortByte
-#endif
-
-#ifdef InPortWord
-#undef InPortWord
-#endif
-
-#ifdef InPortLong
-#undef InPortLong
-#endif
-
-/**********************************************************************/
-/* LINUX KERNEL */
-/**********************************************************************/
-
-#ifdef SIS_LINUX_KERNEL
-
-#ifdef CONFIG_FB_SIS_300
-#define SIS300
-#endif
-
-#ifdef CONFIG_FB_SIS_315
-#define SIS315H
-#endif
-
-#if !defined(SIS300) && !defined(SIS315H)
-#warning Neither CONFIG_FB_SIS_300 nor CONFIG_FB_SIS_315 is set
-#warning sisfb will not work!
-#endif
-
-#define OutPortByte(p,v) outb((u8)(v),(SISIOADDRESS)(p))
-#define OutPortWord(p,v) outw((u16)(v),(SISIOADDRESS)(p))
-#define OutPortLong(p,v) outl((u32)(v),(SISIOADDRESS)(p))
-#define InPortByte(p) inb((SISIOADDRESS)(p))
-#define InPortWord(p) inw((SISIOADDRESS)(p))
-#define InPortLong(p) inl((SISIOADDRESS)(p))
-#define SiS_SetMemory(MemoryAddress,MemorySize,value) memset_io(MemoryAddress, value, MemorySize)
-
-#endif /* LINUX_KERNEL */
-
-/**********************************************************************/
-/* XFree86/X.org */
-/**********************************************************************/
-
-#ifdef SIS_XORG_XF86
-
-#define SIS300
-#define SIS315H
-
-#define OutPortByte(p,v) outSISREG((IOADDRESS)(p),(CARD8)(v))
-#define OutPortWord(p,v) outSISREGW((IOADDRESS)(p),(CARD16)(v))
-#define OutPortLong(p,v) outSISREGL((IOADDRESS)(p),(CARD32)(v))
-#define InPortByte(p) inSISREG((IOADDRESS)(p))
-#define InPortWord(p) inSISREGW((IOADDRESS)(p))
-#define InPortLong(p) inSISREGL((IOADDRESS)(p))
-#define SiS_SetMemory(MemoryAddress,MemorySize,value) memset(MemoryAddress, value, MemorySize)
-
-#endif /* XF86 */
-
-#endif /* _OSDEF_H_ */
#ifndef _SIS_H_
#define _SIS_H_
-#include "osdef.h"
#include <video/sisfb.h>
#include "vgatypes.h"
#include "sis.h"
#include "sis_main.h"
+#if !defined(CONFIG_FB_SIS_300) && !defined(CONFIG_FB_SIS_315)
+#warning Neither CONFIG_FB_SIS_300 nor CONFIG_FB_SIS_315 is set
+#warning sisfb will not work!
+#endif
+
static void sisfb_handle_command(struct sis_video_info *ivideo,
struct sisfb_cmd *sisfb_command);
if(sisfb_check_rom(rom_base, ivideo)) {
if((myrombase = vmalloc(65536))) {
-
- /* Work around bug in pci/rom.c: Folks forgot to check
- * whether the size retrieved from the BIOS image eventually
- * is larger than the mapped size
- */
- if(pci_resource_len(pdev, PCI_ROM_RESOURCE) < romsize)
- romsize = pci_resource_len(pdev, PCI_ROM_RESOURCE);
-
memcpy_fromio(myrombase, rom_base,
(romsize > 65536) ? 65536 : romsize);
}
}
-#else
-
- pci_read_config_dword(pdev, PCI_ROM_ADDRESS, &temp);
- pci_write_config_dword(pdev, PCI_ROM_ADDRESS,
- (ivideo->video_base & PCI_ROM_ADDRESS_MASK) | PCI_ROM_ADDRESS_ENABLE);
-
- rom_base = ioremap(ivideo->video_base, 65536);
- if(rom_base) {
- if(sisfb_check_rom(rom_base, ivideo)) {
- if((myrombase = vmalloc(65536)))
- memcpy_fromio(myrombase, rom_base, 65536);
- }
- iounmap(rom_base);
- }
-
- pci_write_config_dword(pdev, PCI_ROM_ADDRESS, temp);
-
#endif
return myrombase;
#define SISIOMEMTYPE
-#ifdef SIS_LINUX_KERNEL
typedef unsigned long SISIOADDRESS;
#include <linux/types.h> /* Need __iomem */
#undef SISIOMEMTYPE
#define SISIOMEMTYPE __iomem
-#endif
-
-#ifdef SIS_XORG_XF86
-#if XF86_VERSION_CURRENT < XF86_VERSION_NUMERIC(4,2,0,0,0)
-typedef unsigned long IOADDRESS;
-typedef unsigned long SISIOADDRESS;
-#else
-typedef IOADDRESS SISIOADDRESS;
-#endif
-#endif
typedef enum _SIS_CHIP_TYPE {
SIS_VGALegacy = 0,
{
unsigned char ChipType;
unsigned char ChipRevision;
-#ifdef SIS_XORG_XF86
- PCITAG PciTag;
-#endif
-#ifdef SIS_LINUX_KERNEL
void *ivideo;
-#endif
unsigned char *VirtualRomBase;
bool UseROM;
-#ifdef SIS_LINUX_KERNEL
unsigned char SISIOMEMTYPE *VideoMemoryAddress;
unsigned int VideoMemorySize;
-#endif
SISIOADDRESS IOAddress;
SISIOADDRESS IOAddress2; /* For dual chip XGI volari */
-#ifdef SIS_LINUX_KERNEL
SISIOADDRESS RelIO;
-#endif
SISIOADDRESS SiS_P3c4;
SISIOADDRESS SiS_P3d4;
SISIOADDRESS SiS_P3c0;
unsigned short SiS_IF_DEF_FSTN;
unsigned short SiS_SysFlags;
unsigned char SiS_VGAINFO;
-#ifdef SIS_XORG_XF86
- unsigned short SiS_CP1, SiS_CP2, SiS_CP3, SiS_CP4;
-#endif
bool SiS_UseROM;
bool SiS_ROMNew;
bool SiS_XGIROM;
struct device_attribute *attr, char *buf)
{
struct virtio_device *dev = container_of(_d,struct virtio_device,dev);
- return sprintf(buf, "%hu", dev->id.device);
+ return sprintf(buf, "0x%04x\n", dev->id.device);
}
static ssize_t vendor_show(struct device *_d,
struct device_attribute *attr, char *buf)
{
struct virtio_device *dev = container_of(_d,struct virtio_device,dev);
- return sprintf(buf, "%hu", dev->id.vendor);
+ return sprintf(buf, "0x%04x\n", dev->id.vendor);
}
static ssize_t status_show(struct device *_d,
struct device_attribute *attr, char *buf)
{
struct virtio_device *dev = container_of(_d,struct virtio_device,dev);
- return sprintf(buf, "0x%08x", dev->config->get_status(dev));
+ return sprintf(buf, "0x%08x\n", dev->config->get_status(dev));
}
static ssize_t modalias_show(struct device *_d,
struct device_attribute *attr, char *buf)
pr_debug("Added buffer head %i to %p\n", head, vq);
END_USE(vq);
- /* If we're indirect, we can fit many (assuming not OOM). */
- if (vq->indirect)
- return vq->num_free ? vq->vring.num : 0;
return vq->num_free;
}
EXPORT_SYMBOL_GPL(virtqueue_add_buf_gfp);
obj-$(CONFIG_HOTPLUG_CPU) += cpu_hotplug.o
obj-$(CONFIG_XEN_XENCOMM) += xencomm.o
obj-$(CONFIG_XEN_BALLOON) += balloon.o
-obj-$(CONFIG_XEN_DEV_EVTCHN) += evtchn.o
+obj-$(CONFIG_XEN_DEV_EVTCHN) += xen-evtchn.o
obj-$(CONFIG_XENFS) += xenfs/
obj-$(CONFIG_XEN_SYS_HYPERVISOR) += sys-hypervisor.o
obj-$(CONFIG_XEN_PLATFORM_PCI) += platform-pci.o
obj-$(CONFIG_SWIOTLB_XEN) += swiotlb-xen.o
obj-$(CONFIG_XEN_DOM0) += pci.o
+
+xen-evtchn-y := evtchn.o
+
#include <asm/pgtable.h>
#include <asm/uaccess.h>
#include <asm/tlb.h>
+#include <asm/e820.h>
#include <asm/xen/hypervisor.h>
#include <asm/xen/hypercall.h>
}
/* balloon_append: add the given page to the balloon. */
-static void balloon_append(struct page *page)
+static void __balloon_append(struct page *page)
{
/* Lowmem is re-populated first, so highmem pages go at list tail. */
if (PageHighMem(page)) {
list_add(&page->lru, &ballooned_pages);
balloon_stats.balloon_low++;
}
+}
+static void balloon_append(struct page *page)
+{
+ __balloon_append(page);
totalram_pages--;
}
static int increase_reservation(unsigned long nr_pages)
{
- unsigned long pfn, i, flags;
+ unsigned long pfn, i;
struct page *page;
long rc;
struct xen_memory_reservation reservation = {
if (nr_pages > ARRAY_SIZE(frame_list))
nr_pages = ARRAY_SIZE(frame_list);
- spin_lock_irqsave(&xen_reservation_lock, flags);
-
page = balloon_first_page();
for (i = 0; i < nr_pages; i++) {
BUG_ON(page == NULL);
balloon_stats.current_pages += rc;
out:
- spin_unlock_irqrestore(&xen_reservation_lock, flags);
-
return rc < 0 ? rc : rc != nr_pages;
}
static int decrease_reservation(unsigned long nr_pages)
{
- unsigned long pfn, i, flags;
+ unsigned long pfn, i;
struct page *page;
int need_sleep = 0;
int ret;
kmap_flush_unused();
flush_tlb_all();
- spin_lock_irqsave(&xen_reservation_lock, flags);
-
/* No more mappings: invalidate P2M and add to balloon. */
for (i = 0; i < nr_pages; i++) {
pfn = mfn_to_pfn(frame_list[i]);
balloon_stats.current_pages -= nr_pages;
- spin_unlock_irqrestore(&xen_reservation_lock, flags);
-
return need_sleep;
}
static int __init balloon_init(void)
{
- unsigned long pfn;
+ unsigned long pfn, extra_pfn_end;
struct page *page;
if (!xen_pv_domain())
register_balloon(&balloon_sysdev);
/* Initialise the balloon with excess memory space. */
- for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) {
+ extra_pfn_end = min(e820_end_of_ram_pfn(),
+ (unsigned long)PFN_DOWN(xen_extra_mem_start + xen_extra_mem_size));
+ for (pfn = PFN_UP(xen_extra_mem_start);
+ pfn < extra_pfn_end;
+ pfn++) {
page = pfn_to_page(pfn);
- if (!PageReserved(page))
- balloon_append(page);
+ /* totalram_pages doesn't include the boot-time
+ balloon extension, so don't subtract from it. */
+ __balloon_append(page);
}
target_watch.callback = watch_target;
cpumask_copy(irq_to_desc(irq)->affinity, cpumask_of(cpu));
#endif
- __clear_bit(chn, cpu_evtchn_mask(cpu_from_irq(irq)));
- __set_bit(chn, cpu_evtchn_mask(cpu));
+ clear_bit(chn, cpu_evtchn_mask(cpu_from_irq(irq)));
+ set_bit(chn, cpu_evtchn_mask(cpu));
irq_info[irq].cpu = cpu;
}
static void init_evtchn_cpu_bindings(void)
{
+ int i;
#ifdef CONFIG_SMP
struct irq_desc *desc;
- int i;
/* By default all event channels notify CPU#0. */
for_each_irq_desc(i, desc) {
}
#endif
- memset(cpu_evtchn_mask(0), ~0, sizeof(struct cpu_evtchn_s));
+ for_each_possible_cpu(i)
+ memset(cpu_evtchn_mask(i),
+ (i == 0) ? ~0 : 0, sizeof(struct cpu_evtchn_s));
+
}
static inline void clear_evtchn(int port)
goto out;
if (xen_initial_domain()) {
- unmap_irq.pirq = info->u.pirq.gsi;
+ unmap_irq.pirq = info->u.pirq.pirq;
unmap_irq.domid = DOMID_SELF;
rc = HYPERVISOR_physdev_op(PHYSDEVOP_unmap_pirq, &unmap_irq);
if (rc) {
const char *name;
};
-/* Who's bound to each port? */
-static struct per_user_data *port_user[NR_EVENT_CHANNELS];
+/*
+ * Who's bound to each port? This is logically an array of struct
+ * per_user_data *, but we encode the current enabled-state in bit 0.
+ */
+static unsigned long *port_user;
static DEFINE_SPINLOCK(port_user_lock); /* protects port_user[] and ring_prod */
-irqreturn_t evtchn_interrupt(int irq, void *data)
+static inline struct per_user_data *get_port_user(unsigned port)
+{
+ return (struct per_user_data *)(port_user[port] & ~1);
+}
+
+static inline void set_port_user(unsigned port, struct per_user_data *u)
+{
+ port_user[port] = (unsigned long)u;
+}
+
+static inline bool get_port_enabled(unsigned port)
+{
+ return port_user[port] & 1;
+}
+
+static inline void set_port_enabled(unsigned port, bool enabled)
+{
+ if (enabled)
+ port_user[port] |= 1;
+ else
+ port_user[port] &= ~1;
+}
+
+static irqreturn_t evtchn_interrupt(int irq, void *data)
{
unsigned int port = (unsigned long)data;
struct per_user_data *u;
spin_lock(&port_user_lock);
- u = port_user[port];
+ u = get_port_user(port);
+
+ WARN(!get_port_enabled(port),
+ "Interrupt for port %d, but apparently not enabled; per-user %p\n",
+ port, u);
disable_irq_nosync(irq);
+ set_port_enabled(port, false);
if ((u->ring_prod - u->ring_cons) < EVTCHN_RING_SIZE) {
u->ring[EVTCHN_RING_MASK(u->ring_prod)] = port;
kill_fasync(&u->evtchn_async_queue,
SIGIO, POLL_IN);
}
- } else {
+ } else
u->ring_overflow = 1;
- }
spin_unlock(&port_user_lock);
goto out;
spin_lock_irq(&port_user_lock);
- for (i = 0; i < (count/sizeof(evtchn_port_t)); i++)
- if ((kbuf[i] < NR_EVENT_CHANNELS) && (port_user[kbuf[i]] == u))
- enable_irq(irq_from_evtchn(kbuf[i]));
+
+ for (i = 0; i < (count/sizeof(evtchn_port_t)); i++) {
+ unsigned port = kbuf[i];
+
+ if (port < NR_EVENT_CHANNELS &&
+ get_port_user(port) == u &&
+ !get_port_enabled(port)) {
+ set_port_enabled(port, true);
+ enable_irq(irq_from_evtchn(port));
+ }
+ }
+
spin_unlock_irq(&port_user_lock);
rc = count;
* interrupt handler yet, and our caller has already
* serialized bind operations.)
*/
- BUG_ON(port_user[port] != NULL);
- port_user[port] = u;
+ BUG_ON(get_port_user(port) != NULL);
+ set_port_user(port, u);
+ set_port_enabled(port, true); /* start enabled */
rc = bind_evtchn_to_irqhandler(port, evtchn_interrupt, IRQF_DISABLED,
u->name, (void *)(unsigned long)port);
unbind_from_irqhandler(irq, (void *)(unsigned long)port);
- /* make sure we unbind the irq handler before clearing the port */
- barrier();
-
- port_user[port] = NULL;
+ set_port_user(port, NULL);
}
static long evtchn_ioctl(struct file *file,
spin_lock_irq(&port_user_lock);
rc = -ENOTCONN;
- if (port_user[unbind.port] != u) {
+ if (get_port_user(unbind.port) != u) {
spin_unlock_irq(&port_user_lock);
break;
}
- evtchn_unbind_from_user(u, unbind.port);
+ disable_irq(irq_from_evtchn(unbind.port));
spin_unlock_irq(&port_user_lock);
+ evtchn_unbind_from_user(u, unbind.port);
+
rc = 0;
break;
}
if (notify.port >= NR_EVENT_CHANNELS) {
rc = -EINVAL;
- } else if (port_user[notify.port] != u) {
+ } else if (get_port_user(notify.port) != u) {
rc = -ENOTCONN;
} else {
notify_remote_via_evtchn(notify.port);
filp->private_data = u;
- return 0;
+ return nonseekable_open(inode, filp);;
}
static int evtchn_release(struct inode *inode, struct file *filp)
free_page((unsigned long)u->ring);
for (i = 0; i < NR_EVENT_CHANNELS; i++) {
- if (port_user[i] != u)
+ if (get_port_user(i) != u)
continue;
- evtchn_unbind_from_user(port_user[i], i);
+ disable_irq(irq_from_evtchn(i));
}
spin_unlock_irq(&port_user_lock);
+ for (i = 0; i < NR_EVENT_CHANNELS; i++) {
+ if (get_port_user(i) != u)
+ continue;
+
+ evtchn_unbind_from_user(get_port_user(i), i);
+ }
+
kfree(u->name);
kfree(u);
.fasync = evtchn_fasync,
.open = evtchn_open,
.release = evtchn_release,
- .llseek = noop_llseek,
+ .llseek = no_llseek,
};
static struct miscdevice evtchn_miscdev = {
.minor = MISC_DYNAMIC_MINOR,
- .name = "evtchn",
+ .name = "xen/evtchn",
.fops = &evtchn_fops,
};
static int __init evtchn_init(void)
if (!xen_domain())
return -ENODEV;
+ port_user = kcalloc(NR_EVENT_CHANNELS, sizeof(*port_user), GFP_KERNEL);
+ if (port_user == NULL)
+ return -ENOMEM;
+
spin_lock_init(&port_user_lock);
- memset(port_user, 0, sizeof(port_user));
/* Create '/dev/misc/evtchn'. */
err = misc_register(&evtchn_miscdev);
static void __exit evtchn_cleanup(void)
{
+ kfree(port_user);
+ port_user = NULL;
+
misc_deregister(&evtchn_miscdev);
}
xen_pfn_t *mfnp = data;
struct mmap_batch_state *st = state;
- put_user(*mfnp, st->user++);
-
- return 0;
+ return put_user(*mfnp, st->user++);
}
static struct vm_operations_struct privcmd_vm_ops;
up_write(&mm->mmap_sem);
if (state.err > 0) {
- ret = 0;
-
state.user = m.arr;
- traverse_pages(m.num, sizeof(xen_pfn_t),
+ ret = traverse_pages(m.num, sizeof(xen_pfn_t),
&pagelist,
mmap_return_errors, &state);
}
if (xen_feature(XENFEAT_auto_translated_physmap))
return -ENOSYS;
- /* DONTCOPY is essential for Xen as copy_page_range is broken. */
- vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY;
+ /* DONTCOPY is essential for Xen because copy_page_range doesn't know
+ * how to recreate these mappings */
+ vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY | VM_PFNMAP;
vma->vm_ops = &privcmd_vm_ops;
vma->vm_private_data = NULL;
#include <linux/module.h>
#include <linux/fs.h>
#include <linux/magic.h>
-#include <linux/mm.h>
-#include <linux/backing-dev.h>
#include <xen/xen.h>
MODULE_DESCRIPTION("Xen filesystem");
MODULE_LICENSE("GPL");
-static int xenfs_set_page_dirty(struct page *page)
-{
- return !TestSetPageDirty(page);
-}
-
-static const struct address_space_operations xenfs_aops = {
- .set_page_dirty = xenfs_set_page_dirty,
-};
-
-static struct backing_dev_info xenfs_backing_dev_info = {
- .ra_pages = 0, /* No readahead */
- .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK,
-};
-
static struct inode *xenfs_make_inode(struct super_block *sb, int mode)
{
struct inode *ret = new_inode(sb);
if (ret) {
ret->i_mode = mode;
- ret->i_mapping->a_ops = &xenfs_aops;
- ret->i_mapping->backing_dev_info = &xenfs_backing_dev_info;
ret->i_uid = ret->i_gid = 0;
ret->i_blocks = 0;
ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
return rc;
}
-static int xenfs_mount(struct file_system_type *fs_type,
- int flags, const char *dev_name,
- void *data)
+static struct dentry *xenfs_mount(struct file_system_type *fs_type,
+ int flags, const char *dev_name,
+ void *data)
{
return mount_single(fs_type, flags, data, xenfs_fill_super);
}
static int __init xenfs_init(void)
{
- int err;
- if (!xen_domain()) {
- printk(KERN_INFO "xenfs: not registering filesystem on non-xen platform\n");
- return 0;
- }
-
- err = register_filesystem(&xenfs_type);
- if (err) {
- printk(KERN_ERR "xenfs: Unable to register filesystem!\n");
- goto out;
- }
-
- err = bdi_init(&xenfs_backing_dev_info);
- if (err)
- unregister_filesystem(&xenfs_type);
-
- out:
+ if (xen_domain())
+ return register_filesystem(&xenfs_type);
- return err;
+ printk(KERN_INFO "XENFS: not registering filesystem on non-xen platform\n");
+ return 0;
}
static void __exit xenfs_exit(void)
static struct bio *compressed_bio_alloc(struct block_device *bdev,
u64 first_byte, gfp_t gfp_flags)
{
- struct bio *bio;
int nr_vecs;
nr_vecs = bio_get_nr_vecs(bdev);
- bio = bio_alloc(gfp_flags, nr_vecs);
-
- if (bio == NULL && (current->flags & PF_MEMALLOC)) {
- while (!bio && (nr_vecs /= 2))
- bio = bio_alloc(gfp_flags, nr_vecs);
- }
-
- if (bio) {
- bio->bi_size = 0;
- bio->bi_bdev = bdev;
- bio->bi_sector = first_byte >> 9;
- }
- return bio;
+ return btrfs_bio_alloc(bdev, first_byte >> 9, nr_vecs, gfp_flags);
}
static int check_compressed_csum(struct inode *inode,
int extents_thresh;
int free_extents;
int total_bitmaps;
- int ro:1;
- int dirty:1;
- int iref:1;
+ unsigned int ro:1;
+ unsigned int dirty:1;
+ unsigned int iref:1;
int disk_cache_state;
#include <linux/freezer.h>
#include <linux/crc32c.h>
#include <linux/slab.h>
+#include <linux/migrate.h>
#include "compat.h"
#include "ctree.h"
#include "disk-io.h"
ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE,
btrfs_header_generation(eb));
BUG_ON(ret);
+ WARN_ON(!btrfs_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN));
+
found_start = btrfs_header_bytenr(eb);
if (found_start != start) {
WARN_ON(1);
__btree_submit_bio_done);
}
+static int btree_migratepage(struct address_space *mapping,
+ struct page *newpage, struct page *page)
+{
+ /*
+ * we can't safely write a btree page from here,
+ * we haven't done the locking hook
+ */
+ if (PageDirty(page))
+ return -EAGAIN;
+ /*
+ * Buffers may be managed in a filesystem specific way.
+ * We must have no buffers or drop them.
+ */
+ if (page_has_private(page) &&
+ !try_to_release_page(page, GFP_KERNEL))
+ return -EAGAIN;
+#ifdef CONFIG_MIGRATION
+ return migrate_page(mapping, newpage, page);
+#else
+ return -ENOSYS;
+#endif
+}
+
static int btree_writepage(struct page *page, struct writeback_control *wbc)
{
struct extent_io_tree *tree;
}
redirty_page_for_writepage(wbc, page);
- eb = btrfs_find_tree_block(root, page_offset(page),
- PAGE_CACHE_SIZE);
+ eb = btrfs_find_tree_block(root, page_offset(page), PAGE_CACHE_SIZE);
WARN_ON(!eb);
was_dirty = test_and_set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags);
.releasepage = btree_releasepage,
.invalidatepage = btree_invalidatepage,
.sync_page = block_sync_page,
+#ifdef CONFIG_MIGRATION
+ .migratepage = btree_migratepage,
+#endif
};
int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize,
GFP_NOFS);
struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
GFP_NOFS);
- struct btrfs_root *tree_root = kzalloc(sizeof(struct btrfs_root),
- GFP_NOFS);
- struct btrfs_fs_info *fs_info = kzalloc(sizeof(*fs_info),
- GFP_NOFS);
+ struct btrfs_root *tree_root = btrfs_sb(sb);
+ struct btrfs_fs_info *fs_info = tree_root->fs_info;
struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
GFP_NOFS);
struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
return ERR_PTR(ret);
}
+static int btrfs_get_name(struct dentry *parent, char *name,
+ struct dentry *child)
+{
+ struct inode *inode = child->d_inode;
+ struct inode *dir = parent->d_inode;
+ struct btrfs_path *path;
+ struct btrfs_root *root = BTRFS_I(dir)->root;
+ struct btrfs_inode_ref *iref;
+ struct btrfs_root_ref *rref;
+ struct extent_buffer *leaf;
+ unsigned long name_ptr;
+ struct btrfs_key key;
+ int name_len;
+ int ret;
+
+ if (!dir || !inode)
+ return -EINVAL;
+
+ if (!S_ISDIR(dir->i_mode))
+ return -EINVAL;
+
+ path = btrfs_alloc_path();
+ if (!path)
+ return -ENOMEM;
+ path->leave_spinning = 1;
+
+ if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+ key.objectid = BTRFS_I(inode)->root->root_key.objectid;
+ key.type = BTRFS_ROOT_BACKREF_KEY;
+ key.offset = (u64)-1;
+ root = root->fs_info->tree_root;
+ } else {
+ key.objectid = inode->i_ino;
+ key.offset = dir->i_ino;
+ key.type = BTRFS_INODE_REF_KEY;
+ }
+
+ ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+ if (ret < 0) {
+ btrfs_free_path(path);
+ return ret;
+ } else if (ret > 0) {
+ if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+ path->slots[0]--;
+ } else {
+ btrfs_free_path(path);
+ return -ENOENT;
+ }
+ }
+ leaf = path->nodes[0];
+
+ if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) {
+ rref = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_root_ref);
+ name_ptr = (unsigned long)(rref + 1);
+ name_len = btrfs_root_ref_name_len(leaf, rref);
+ } else {
+ iref = btrfs_item_ptr(leaf, path->slots[0],
+ struct btrfs_inode_ref);
+ name_ptr = (unsigned long)(iref + 1);
+ name_len = btrfs_inode_ref_name_len(leaf, iref);
+ }
+
+ read_extent_buffer(leaf, name, name_ptr, name_len);
+ btrfs_free_path(path);
+
+ /*
+ * have to add the null termination to make sure that reconnect_path
+ * gets the right len for strlen
+ */
+ name[name_len] = '\0';
+
+ return 0;
+}
+
const struct export_operations btrfs_export_ops = {
.encode_fh = btrfs_encode_fh,
.fh_to_dentry = btrfs_fh_to_dentry,
.fh_to_parent = btrfs_fh_to_parent,
.get_parent = btrfs_get_parent,
+ .get_name = btrfs_get_name,
};
* our reservation.
*/
if (unused <= space_info->total_bytes) {
- unused -= space_info->total_bytes;
+ unused = space_info->total_bytes - unused;
if (unused >= num_bytes) {
if (!reserved)
space_info->bytes_reserved += orig_bytes;
bio_put(bio);
}
-static struct bio *
-extent_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
- gfp_t gfp_flags)
+struct bio *
+btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
+ gfp_t gfp_flags)
{
struct bio *bio;
else
nr = bio_get_nr_vecs(bdev);
- bio = extent_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH);
+ bio = btrfs_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH);
bio_add_page(bio, page, page_size, offset);
bio->bi_end_io = end_io_func;
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
__u64 start, __u64 len, get_extent_t *get_extent)
{
- int ret;
+ int ret = 0;
u64 off = start;
u64 max = start + len;
u32 flags = 0;
+ u32 found_type;
+ u64 last;
u64 disko = 0;
+ struct btrfs_key found_key;
struct extent_map *em = NULL;
struct extent_state *cached_state = NULL;
+ struct btrfs_path *path;
+ struct btrfs_file_extent_item *item;
int end = 0;
u64 em_start = 0, em_len = 0;
unsigned long emflags;
- ret = 0;
+ int hole = 0;
if (len == 0)
return -EINVAL;
+ path = btrfs_alloc_path();
+ if (!path)
+ return -ENOMEM;
+ path->leave_spinning = 1;
+
+ ret = btrfs_lookup_file_extent(NULL, BTRFS_I(inode)->root,
+ path, inode->i_ino, -1, 0);
+ if (ret < 0) {
+ btrfs_free_path(path);
+ return ret;
+ }
+ WARN_ON(!ret);
+ path->slots[0]--;
+ item = btrfs_item_ptr(path->nodes[0], path->slots[0],
+ struct btrfs_file_extent_item);
+ btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]);
+ found_type = btrfs_key_type(&found_key);
+
+ /* No extents, just return */
+ if (found_key.objectid != inode->i_ino ||
+ found_type != BTRFS_EXTENT_DATA_KEY) {
+ btrfs_free_path(path);
+ return 0;
+ }
+ last = found_key.offset;
+ btrfs_free_path(path);
+
lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0,
&cached_state, GFP_NOFS);
em = get_extent(inode, NULL, 0, off, max - off, 0);
ret = PTR_ERR(em);
goto out;
}
+
while (!end) {
+ hole = 0;
off = em->start + em->len;
if (off >= max)
end = 1;
+ if (em->block_start == EXTENT_MAP_HOLE) {
+ hole = 1;
+ goto next;
+ }
+
em_start = em->start;
em_len = em->len;
if (em->block_start == EXTENT_MAP_LAST_BYTE) {
end = 1;
flags |= FIEMAP_EXTENT_LAST;
- } else if (em->block_start == EXTENT_MAP_HOLE) {
- flags |= FIEMAP_EXTENT_UNWRITTEN;
} else if (em->block_start == EXTENT_MAP_INLINE) {
flags |= (FIEMAP_EXTENT_DATA_INLINE |
FIEMAP_EXTENT_NOT_ALIGNED);
if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
flags |= FIEMAP_EXTENT_ENCODED;
+next:
emflags = em->flags;
free_extent_map(em);
em = NULL;
-
if (!end) {
em = get_extent(inode, NULL, 0, off, max - off, 0);
if (!em)
}
emflags = em->flags;
}
+
if (test_bit(EXTENT_FLAG_VACANCY, &emflags)) {
flags |= FIEMAP_EXTENT_LAST;
end = 1;
}
- ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
- em_len, flags);
- if (ret)
- goto out_free;
+ if (em_start == last) {
+ flags |= FIEMAP_EXTENT_LAST;
+ end = 1;
+ }
+
+ if (!hole) {
+ ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
+ em_len, flags);
+ if (ret)
+ goto out_free;
+ }
}
out_free:
free_extent_map(em);
spin_lock(&tree->buffer_lock);
eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT);
- if (!eb)
- goto out;
+ if (!eb) {
+ spin_unlock(&tree->buffer_lock);
+ return ret;
+ }
if (test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) {
ret = 0;
struct extent_io_tree *tree,
u64 start, u64 end, struct page *locked_page,
unsigned long op);
+struct bio *
+btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
+ gfp_t gfp_flags);
#endif
if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
trans = btrfs_start_transaction(root, 0);
+ if (IS_ERR(trans)) {
+ num_written = PTR_ERR(trans);
+ goto done;
+ }
+ mutex_lock(&inode->i_mutex);
ret = btrfs_log_dentry_safe(trans, root,
file->f_dentry);
+ mutex_unlock(&inode->i_mutex);
if (ret == 0) {
ret = btrfs_sync_log(trans, root);
if (ret == 0)
(start_pos + num_written - 1) >> PAGE_CACHE_SHIFT);
}
}
+done:
current->backing_dev_info = NULL;
return num_written ? num_written : err;
}
BTRFS_I(inode)->index_cnt = 2;
BTRFS_I(inode)->root = root;
BTRFS_I(inode)->generation = trans->transid;
+ inode->i_generation = BTRFS_I(inode)->generation;
btrfs_set_inode_space_info(root, inode);
if (mode & S_IFDIR)
}
static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
- struct dentry *dentry, struct inode *inode,
- int backref, u64 index)
+ struct inode *dir, struct dentry *dentry,
+ struct inode *inode, int backref, u64 index)
{
- int err = btrfs_add_link(trans, dentry->d_parent->d_inode,
- inode, dentry->d_name.name,
- dentry->d_name.len, backref, index);
+ int err = btrfs_add_link(trans, dir, inode,
+ dentry->d_name.name, dentry->d_name.len,
+ backref, index);
if (!err) {
d_instantiate(dentry, inode);
return 0;
btrfs_set_trans_block_group(trans, dir);
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len,
- dentry->d_parent->d_inode->i_ino, objectid,
+ dentry->d_name.len, dir->i_ino, objectid,
BTRFS_I(dir)->block_group, mode, &index);
err = PTR_ERR(inode);
if (IS_ERR(inode))
}
btrfs_set_trans_block_group(trans, inode);
- err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
if (err)
drop_inode = 1;
else {
btrfs_set_trans_block_group(trans, dir);
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len,
- dentry->d_parent->d_inode->i_ino,
- objectid, BTRFS_I(dir)->block_group, mode,
- &index);
+ dentry->d_name.len, dir->i_ino, objectid,
+ BTRFS_I(dir)->block_group, mode, &index);
err = PTR_ERR(inode);
if (IS_ERR(inode))
goto out_unlock;
}
btrfs_set_trans_block_group(trans, inode);
- err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
if (err)
drop_inode = 1;
else {
return -EPERM;
btrfs_inc_nlink(inode);
+ inode->i_ctime = CURRENT_TIME;
err = btrfs_set_inode_index(dir, &index);
if (err)
btrfs_set_trans_block_group(trans, dir);
ihold(inode);
- err = btrfs_add_nondir(trans, dentry, inode, 1, index);
+ err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index);
if (err) {
drop_inode = 1;
} else {
+ struct dentry *parent = dget_parent(dentry);
btrfs_update_inode_block_group(trans, dir);
err = btrfs_update_inode(trans, root, inode);
BUG_ON(err);
- btrfs_log_new_name(trans, inode, NULL, dentry->d_parent);
+ btrfs_log_new_name(trans, inode, NULL, parent);
+ dput(parent);
}
nr = trans->blocks_used;
btrfs_set_trans_block_group(trans, dir);
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len,
- dentry->d_parent->d_inode->i_ino, objectid,
+ dentry->d_name.len, dir->i_ino, objectid,
BTRFS_I(dir)->block_group, S_IFDIR | mode,
&index);
if (IS_ERR(inode)) {
if (err)
goto out_fail;
- err = btrfs_add_link(trans, dentry->d_parent->d_inode,
- inode, dentry->d_name.name,
- dentry->d_name.len, 0, index);
+ err = btrfs_add_link(trans, dir, inode, dentry->d_name.name,
+ dentry->d_name.len, 0, index);
if (err)
goto out_fail;
u64 bytes;
u32 *csums;
void *private;
+
+ /* number of bios pending for this dio */
+ atomic_t pending_bios;
+
+ /* IO errors */
+ int errors;
+
+ struct bio *orig_bio;
};
static void btrfs_endio_direct_read(struct bio *bio, int err)
{
+ struct btrfs_dio_private *dip = bio->bi_private;
struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1;
struct bio_vec *bvec = bio->bi_io_vec;
- struct btrfs_dio_private *dip = bio->bi_private;
struct inode *inode = dip->inode;
struct btrfs_root *root = BTRFS_I(inode)->root;
u64 start;
struct btrfs_trans_handle *trans;
struct btrfs_ordered_extent *ordered = NULL;
struct extent_state *cached_state = NULL;
+ u64 ordered_offset = dip->logical_offset;
+ u64 ordered_bytes = dip->bytes;
int ret;
if (err)
goto out_done;
-
- ret = btrfs_dec_test_ordered_pending(inode, &ordered,
- dip->logical_offset, dip->bytes);
+again:
+ ret = btrfs_dec_test_first_ordered_pending(inode, &ordered,
+ &ordered_offset,
+ ordered_bytes);
if (!ret)
- goto out_done;
+ goto out_test;
BUG_ON(!ordered);
out:
btrfs_delalloc_release_metadata(inode, ordered->len);
btrfs_end_transaction(trans, root);
+ ordered_offset = ordered->file_offset + ordered->len;
btrfs_put_ordered_extent(ordered);
btrfs_put_ordered_extent(ordered);
+
+out_test:
+ /*
+ * our bio might span multiple ordered extents. If we haven't
+ * completed the accounting for the whole dio, go back and try again
+ */
+ if (ordered_offset < dip->logical_offset + dip->bytes) {
+ ordered_bytes = dip->logical_offset + dip->bytes -
+ ordered_offset;
+ goto again;
+ }
out_done:
bio->bi_private = dip->private;
return 0;
}
+static void btrfs_end_dio_bio(struct bio *bio, int err)
+{
+ struct btrfs_dio_private *dip = bio->bi_private;
+
+ if (err) {
+ printk(KERN_ERR "btrfs direct IO failed ino %lu rw %lu "
+ "disk_bytenr %lu len %u err no %d\n",
+ dip->inode->i_ino, bio->bi_rw, bio->bi_sector,
+ bio->bi_size, err);
+ dip->errors = 1;
+
+ /*
+ * before atomic variable goto zero, we must make sure
+ * dip->errors is perceived to be set.
+ */
+ smp_mb__before_atomic_dec();
+ }
+
+ /* if there are more bios still pending for this dio, just exit */
+ if (!atomic_dec_and_test(&dip->pending_bios))
+ goto out;
+
+ if (dip->errors)
+ bio_io_error(dip->orig_bio);
+ else {
+ set_bit(BIO_UPTODATE, &dip->orig_bio->bi_flags);
+ bio_endio(dip->orig_bio, 0);
+ }
+out:
+ bio_put(bio);
+}
+
+static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev,
+ u64 first_sector, gfp_t gfp_flags)
+{
+ int nr_vecs = bio_get_nr_vecs(bdev);
+ return btrfs_bio_alloc(bdev, first_sector, nr_vecs, gfp_flags);
+}
+
+static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
+ int rw, u64 file_offset, int skip_sum,
+ u32 *csums)
+{
+ int write = rw & REQ_WRITE;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ int ret;
+
+ bio_get(bio);
+ ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0);
+ if (ret)
+ goto err;
+
+ if (write && !skip_sum) {
+ ret = btrfs_wq_submit_bio(root->fs_info,
+ inode, rw, bio, 0, 0,
+ file_offset,
+ __btrfs_submit_bio_start_direct_io,
+ __btrfs_submit_bio_done);
+ goto err;
+ } else if (!skip_sum)
+ btrfs_lookup_bio_sums_dio(root, inode, bio,
+ file_offset, csums);
+
+ ret = btrfs_map_bio(root, rw, bio, 0, 1);
+err:
+ bio_put(bio);
+ return ret;
+}
+
+static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
+ int skip_sum)
+{
+ struct inode *inode = dip->inode;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
+ struct bio *bio;
+ struct bio *orig_bio = dip->orig_bio;
+ struct bio_vec *bvec = orig_bio->bi_io_vec;
+ u64 start_sector = orig_bio->bi_sector;
+ u64 file_offset = dip->logical_offset;
+ u64 submit_len = 0;
+ u64 map_length;
+ int nr_pages = 0;
+ u32 *csums = dip->csums;
+ int ret = 0;
+
+ bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
+ if (!bio)
+ return -ENOMEM;
+ bio->bi_private = dip;
+ bio->bi_end_io = btrfs_end_dio_bio;
+ atomic_inc(&dip->pending_bios);
+
+ map_length = orig_bio->bi_size;
+ ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+ &map_length, NULL, 0);
+ if (ret) {
+ bio_put(bio);
+ return -EIO;
+ }
+
+ while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) {
+ if (unlikely(map_length < submit_len + bvec->bv_len ||
+ bio_add_page(bio, bvec->bv_page, bvec->bv_len,
+ bvec->bv_offset) < bvec->bv_len)) {
+ /*
+ * inc the count before we submit the bio so
+ * we know the end IO handler won't happen before
+ * we inc the count. Otherwise, the dip might get freed
+ * before we're done setting it up
+ */
+ atomic_inc(&dip->pending_bios);
+ ret = __btrfs_submit_dio_bio(bio, inode, rw,
+ file_offset, skip_sum,
+ csums);
+ if (ret) {
+ bio_put(bio);
+ atomic_dec(&dip->pending_bios);
+ goto out_err;
+ }
+
+ if (!skip_sum)
+ csums = csums + nr_pages;
+ start_sector += submit_len >> 9;
+ file_offset += submit_len;
+
+ submit_len = 0;
+ nr_pages = 0;
+
+ bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev,
+ start_sector, GFP_NOFS);
+ if (!bio)
+ goto out_err;
+ bio->bi_private = dip;
+ bio->bi_end_io = btrfs_end_dio_bio;
+
+ map_length = orig_bio->bi_size;
+ ret = btrfs_map_block(map_tree, READ, start_sector << 9,
+ &map_length, NULL, 0);
+ if (ret) {
+ bio_put(bio);
+ goto out_err;
+ }
+ } else {
+ submit_len += bvec->bv_len;
+ nr_pages ++;
+ bvec++;
+ }
+ }
+
+ ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum,
+ csums);
+ if (!ret)
+ return 0;
+
+ bio_put(bio);
+out_err:
+ dip->errors = 1;
+ /*
+ * before atomic variable goto zero, we must
+ * make sure dip->errors is perceived to be set.
+ */
+ smp_mb__before_atomic_dec();
+ if (atomic_dec_and_test(&dip->pending_bios))
+ bio_io_error(dip->orig_bio);
+
+ /* bio_end_io() will handle error, so we needn't return it */
+ return 0;
+}
+
static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
loff_t file_offset)
{
dip->disk_bytenr = (u64)bio->bi_sector << 9;
bio->bi_private = dip;
+ dip->errors = 0;
+ dip->orig_bio = bio;
+ atomic_set(&dip->pending_bios, 0);
if (write)
bio->bi_end_io = btrfs_endio_direct_write;
else
bio->bi_end_io = btrfs_endio_direct_read;
- ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0);
- if (ret)
- goto out_err;
-
- if (write && !skip_sum) {
- ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
- inode, rw, bio, 0, 0,
- dip->logical_offset,
- __btrfs_submit_bio_start_direct_io,
- __btrfs_submit_bio_done);
- if (ret)
- goto out_err;
+ ret = btrfs_submit_direct_hook(rw, dip, skip_sum);
+ if (!ret)
return;
- } else if (!skip_sum)
- btrfs_lookup_bio_sums_dio(root, inode, bio,
- dip->logical_offset, dip->csums);
-
- ret = btrfs_map_bio(root, rw, bio, 0, 1);
- if (ret)
- goto out_err;
- return;
-out_err:
- kfree(dip->csums);
- kfree(dip);
free_ordered:
/*
* If this is a write, we need to clean up the reserved space and kill
BUG_ON(ret);
if (old_inode->i_ino != BTRFS_FIRST_FREE_OBJECTID) {
- btrfs_log_new_name(trans, old_inode, old_dir,
- new_dentry->d_parent);
+ struct dentry *parent = dget_parent(new_dentry);
+ btrfs_log_new_name(trans, old_inode, old_dir, parent);
+ dput(parent);
btrfs_end_log_trans(root);
}
out_fail:
btrfs_set_trans_block_group(trans, dir);
inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
- dentry->d_name.len,
- dentry->d_parent->d_inode->i_ino, objectid,
+ dentry->d_name.len, dir->i_ino, objectid,
BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO,
&index);
err = PTR_ERR(inode);
}
btrfs_set_trans_block_group(trans, inode);
- err = btrfs_add_nondir(trans, dentry, inode, 0, index);
+ err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
if (err)
drop_inode = 1;
else {
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_key ins;
u64 cur_offset = start;
+ u64 i_size;
int ret = 0;
bool own_trans = true;
(actual_len > inode->i_size) &&
(cur_offset > inode->i_size)) {
if (cur_offset > actual_len)
- i_size_write(inode, actual_len);
+ i_size = actual_len;
else
- i_size_write(inode, cur_offset);
- i_size_write(inode, cur_offset);
- btrfs_ordered_update_i_size(inode, cur_offset, NULL);
+ i_size = cur_offset;
+ i_size_write(inode, i_size);
+ btrfs_ordered_update_i_size(inode, i_size, NULL);
}
ret = btrfs_update_inode(trans, root, inode);
btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start);
mutex_lock(&inode->i_mutex);
+ ret = inode_newsize_ok(inode, alloc_end);
+ if (ret)
+ goto out;
+
if (alloc_start > inode->i_size) {
ret = btrfs_cont_expand(inode, alloc_start);
if (ret)
.readlink = generic_readlink,
.follow_link = page_follow_link_light,
.put_link = page_put_link,
+ .getattr = btrfs_getattr,
.permission = btrfs_permission,
.setxattr = btrfs_setxattr,
.getxattr = btrfs_getxattr,
struct btrfs_inode_item *inode_item;
struct extent_buffer *leaf;
struct btrfs_root *new_root;
- struct inode *dir = dentry->d_parent->d_inode;
+ struct dentry *parent = dget_parent(dentry);
+ struct inode *dir;
int ret;
int err;
u64 objectid;
ret = btrfs_find_free_objectid(NULL, root->fs_info->tree_root,
0, &objectid);
- if (ret)
+ if (ret) {
+ dput(parent);
return ret;
+ }
+
+ dir = parent->d_inode;
+
/*
* 1 - inode item
* 2 - refs
* 2 - dir items
*/
trans = btrfs_start_transaction(root, 6);
- if (IS_ERR(trans))
+ if (IS_ERR(trans)) {
+ dput(parent);
return PTR_ERR(trans);
+ }
leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
0, objectid, NULL, 0, 0, 0);
d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
fail:
+ dput(parent);
if (async_transid) {
*async_transid = trans->transid;
err = btrfs_commit_transaction_async(trans, root, 1);
char *name, int namelen, u64 *async_transid)
{
struct inode *inode;
+ struct dentry *parent;
struct btrfs_pending_snapshot *pending_snapshot;
struct btrfs_trans_handle *trans;
int ret;
btrfs_orphan_cleanup(pending_snapshot->snap);
- inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
+ parent = dget_parent(dentry);
+ inode = btrfs_lookup_dentry(parent->d_inode, dentry);
+ dput(parent);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto fail;
olen = len = src->i_size - off;
/* if we extend to eof, continue to block boundary */
if (off + len == src->i_size)
- len = ((src->i_size + bs-1) & ~(bs-1))
- - off;
+ len = ALIGN(src->i_size, bs) - off;
/* verify the end result is block aligned */
- if ((off & (bs-1)) ||
- ((off + len) & (bs-1)))
+ if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) ||
+ !IS_ALIGNED(destoff, bs))
goto out_unlock;
/* do any pending delalloc/csum calc on src, one way or
* but shouldn't round up the file size
*/
endoff = new_key.offset + datal;
- if (endoff > off+olen)
- endoff = off+olen;
+ if (endoff > destoff+olen)
+ endoff = destoff+olen;
if (endoff > inode->i_size)
btrfs_i_size_write(inode, endoff);
return 0;
}
+/*
+ * this is used to account for finished IO across a given range
+ * of the file. The IO may span ordered extents. If
+ * a given ordered_extent is completely done, 1 is returned, otherwise
+ * 0.
+ *
+ * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used
+ * to make sure this function only returns 1 once for a given ordered extent.
+ *
+ * file_offset is updated to one byte past the range that is recorded as
+ * complete. This allows you to walk forward in the file.
+ */
+int btrfs_dec_test_first_ordered_pending(struct inode *inode,
+ struct btrfs_ordered_extent **cached,
+ u64 *file_offset, u64 io_size)
+{
+ struct btrfs_ordered_inode_tree *tree;
+ struct rb_node *node;
+ struct btrfs_ordered_extent *entry = NULL;
+ int ret;
+ u64 dec_end;
+ u64 dec_start;
+ u64 to_dec;
+
+ tree = &BTRFS_I(inode)->ordered_tree;
+ spin_lock(&tree->lock);
+ node = tree_search(tree, *file_offset);
+ if (!node) {
+ ret = 1;
+ goto out;
+ }
+
+ entry = rb_entry(node, struct btrfs_ordered_extent, rb_node);
+ if (!offset_in_entry(entry, *file_offset)) {
+ ret = 1;
+ goto out;
+ }
+
+ dec_start = max(*file_offset, entry->file_offset);
+ dec_end = min(*file_offset + io_size, entry->file_offset +
+ entry->len);
+ *file_offset = dec_end;
+ if (dec_start > dec_end) {
+ printk(KERN_CRIT "bad ordering dec_start %llu end %llu\n",
+ (unsigned long long)dec_start,
+ (unsigned long long)dec_end);
+ }
+ to_dec = dec_end - dec_start;
+ if (to_dec > entry->bytes_left) {
+ printk(KERN_CRIT "bad ordered accounting left %llu size %llu\n",
+ (unsigned long long)entry->bytes_left,
+ (unsigned long long)to_dec);
+ }
+ entry->bytes_left -= to_dec;
+ if (entry->bytes_left == 0)
+ ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
+ else
+ ret = 1;
+out:
+ if (!ret && cached && entry) {
+ *cached = entry;
+ atomic_inc(&entry->refs);
+ }
+ spin_unlock(&tree->lock);
+ return ret == 0;
+}
+
/*
* this is used to account for finished IO across a given range
* of the file. The IO should not span ordered extents. If
int btrfs_dec_test_ordered_pending(struct inode *inode,
struct btrfs_ordered_extent **cached,
u64 file_offset, u64 io_size);
+int btrfs_dec_test_first_ordered_pending(struct inode *inode,
+ struct btrfs_ordered_extent **cached,
+ u64 *file_offset, u64 io_size);
int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
u64 start, u64 len, u64 disk_len, int type);
int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset,
case Opt_space_cache:
printk(KERN_INFO "btrfs: enabling disk space caching\n");
btrfs_set_opt(info->mount_opt, SPACE_CACHE);
+ break;
case Opt_clear_cache:
printk(KERN_INFO "btrfs: force clearing of disk cache\n");
btrfs_set_opt(info->mount_opt, CLEAR_CACHE);
static int btrfs_test_super(struct super_block *s, void *data)
{
- struct btrfs_fs_devices *test_fs_devices = data;
+ struct btrfs_root *test_root = data;
struct btrfs_root *root = btrfs_sb(s);
- return root->fs_info->fs_devices == test_fs_devices;
+ /*
+ * If this super block is going away, return false as it
+ * can't match as an existing super block.
+ */
+ if (!atomic_read(&s->s_active))
+ return 0;
+ return root->fs_info->fs_devices == test_root->fs_info->fs_devices;
+}
+
+static int btrfs_set_super(struct super_block *s, void *data)
+{
+ s->s_fs_info = data;
+
+ return set_anon_super(s, data);
}
+
/*
* Find a superblock for the given device / mount point.
*
struct super_block *s;
struct dentry *root;
struct btrfs_fs_devices *fs_devices = NULL;
+ struct btrfs_root *tree_root = NULL;
+ struct btrfs_fs_info *fs_info = NULL;
fmode_t mode = FMODE_READ;
char *subvol_name = NULL;
u64 subvol_objectid = 0;
goto error_close_devices;
}
+ /*
+ * Setup a dummy root and fs_info for test/set super. This is because
+ * we don't actually fill this stuff out until open_ctree, but we need
+ * it for searching for existing supers, so this lets us do that and
+ * then open_ctree will properly initialize everything later.
+ */
+ fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS);
+ tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+ if (!fs_info || !tree_root) {
+ error = -ENOMEM;
+ goto error_close_devices;
+ }
+ fs_info->tree_root = tree_root;
+ fs_info->fs_devices = fs_devices;
+ tree_root->fs_info = fs_info;
+
bdev = fs_devices->latest_bdev;
- s = sget(fs_type, btrfs_test_super, set_anon_super, fs_devices);
+ s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root);
if (IS_ERR(s))
goto error_s;
error = PTR_ERR(s);
error_close_devices:
btrfs_close_devices(fs_devices);
+ kfree(fs_info);
+ kfree(tree_root);
error_free_subvol_name:
kfree(subvol_name);
return ERR_PTR(error);
struct btrfs_root *root = pending->root;
struct btrfs_root *parent_root;
struct inode *parent_inode;
+ struct dentry *parent;
struct dentry *dentry;
struct extent_buffer *tmp;
struct extent_buffer *old;
trans->block_rsv = &pending->block_rsv;
dentry = pending->dentry;
- parent_inode = dentry->d_parent->d_inode;
+ parent = dget_parent(dentry);
+ parent_inode = parent->d_inode;
parent_root = BTRFS_I(parent_inode)->root;
record_root_in_trans(trans, parent_root);
parent_inode->i_ino, index,
dentry->d_name.name, dentry->d_name.len);
BUG_ON(ret);
+ dput(parent);
key.offset = (u64)-1;
pending->snap = btrfs_read_fs_root_no_name(root->fs_info, &key);
{
int ret = 0;
struct btrfs_root *root;
+ struct dentry *old_parent = NULL;
/*
* for regular files, if its inode is already on disk, we don't
if (IS_ROOT(parent))
break;
- parent = parent->d_parent;
+ parent = dget_parent(parent);
+ dput(old_parent);
+ old_parent = parent;
inode = parent->d_inode;
}
+ dput(old_parent);
out:
return ret;
}
{
int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL;
struct super_block *sb;
+ struct dentry *old_parent = NULL;
int ret = 0;
u64 last_committed = root->fs_info->last_trans_committed;
if (IS_ROOT(parent))
break;
- parent = parent->d_parent;
+ parent = dget_parent(parent);
+ dput(old_parent);
+ old_parent = parent;
}
ret = 0;
end_trans:
+ dput(old_parent);
if (ret < 0) {
BUG_ON(ret != -ENOSPC);
root->fs_info->last_trans_log_full_commit = trans->transid;
int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct dentry *dentry)
{
- return btrfs_log_inode_parent(trans, root, dentry->d_inode,
- dentry->d_parent, 0);
+ struct dentry *parent = dget_parent(dentry);
+ int ret;
+
+ ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent, 0);
+ dput(parent);
+
+ return ret;
}
/*
select NLS
select CRYPTO
select CRYPTO_MD5
+ select CRYPTO_HMAC
select CRYPTO_ARC4
help
This is the client VFS module for the Common Internet File System
to be cached locally on disk through the general filesystem cache
manager. If unsure, say N.
+config CIFS_ACL
+ bool "Provide CIFS ACL support (EXPERIMENTAL)"
+ depends on EXPERIMENTAL && CIFS_XATTR
+ help
+ Allows to fetch CIFS/NTFS ACL from the server. The DACL blob
+ is handed over to the application/caller.
+
config CIFS_EXPERIMENTAL
bool "CIFS Experimental Features (EXPERIMENTAL)"
depends on CIFS && EXPERIMENTAL
struct tcon_link *tlink = cifs_sb_tlink(cifs_sb);
if (IS_ERR(tlink))
- return NULL;
+ return ERR_CAST(tlink);
xid = GetXid();
rc = CIFSSMBGetCIFSACL(xid, tlink_tcon(tlink), fid, &pntsd, pacllen);
cifs_put_tlink(tlink);
- cFYI(1, "GetCIFSACL rc = %d ACL len %d", rc, *pacllen);
+ cFYI(1, "%s: rc = %d ACL len %d", __func__, rc, *pacllen);
+ if (rc)
+ return ERR_PTR(rc);
return pntsd;
}
struct tcon_link *tlink = cifs_sb_tlink(cifs_sb);
if (IS_ERR(tlink))
- return NULL;
+ return ERR_CAST(tlink);
tcon = tlink_tcon(tlink);
xid = GetXid();
rc = CIFSSMBOpen(xid, tcon, path, FILE_OPEN, READ_CONTROL, 0,
&fid, &oplock, NULL, cifs_sb->local_nls,
cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
- if (rc) {
- cERROR(1, "Unable to open file to get ACL");
- goto out;
+ if (!rc) {
+ rc = CIFSSMBGetCIFSACL(xid, tcon, fid, &pntsd, pacllen);
+ CIFSSMBClose(xid, tcon, fid);
}
- rc = CIFSSMBGetCIFSACL(xid, tcon, fid, &pntsd, pacllen);
- cFYI(1, "GetCIFSACL rc = %d ACL len %d", rc, *pacllen);
-
- CIFSSMBClose(xid, tcon, fid);
- out:
cifs_put_tlink(tlink);
FreeXid(xid);
+
+ cFYI(1, "%s: rc = %d ACL len %d", __func__, rc, *pacllen);
+ if (rc)
+ return ERR_PTR(rc);
return pntsd;
}
/* Retrieve an ACL from the server */
-static struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *cifs_sb,
+struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *cifs_sb,
struct inode *inode, const char *path,
u32 *pacllen)
{
}
/* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */
-void
+int
cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr,
struct inode *inode, const char *path, const __u16 *pfid)
{
pntsd = get_cifs_acl(cifs_sb, inode, path, &acllen);
/* if we can retrieve the ACL, now parse Access Control Entries, ACEs */
- if (pntsd)
+ if (IS_ERR(pntsd)) {
+ rc = PTR_ERR(pntsd);
+ cERROR(1, "%s: error %d getting sec desc", __func__, rc);
+ } else {
rc = parse_sec_desc(pntsd, acllen, fattr);
- if (rc)
- cFYI(1, "parse sec desc failed rc = %d", rc);
+ kfree(pntsd);
+ if (rc)
+ cERROR(1, "parse sec desc failed rc = %d", rc);
+ }
- kfree(pntsd);
- return;
+ return rc;
}
/* Convert mode bits to an ACL so we can update the ACL on the server */
-int mode_to_acl(struct inode *inode, const char *path, __u64 nmode)
+int mode_to_cifs_acl(struct inode *inode, const char *path, __u64 nmode)
{
int rc = 0;
__u32 secdesclen = 0;
/* Add three ACEs for owner, group, everyone getting rid of
other ACEs as chmod disables ACEs and set the security descriptor */
- if (pntsd) {
+ if (IS_ERR(pntsd)) {
+ rc = PTR_ERR(pntsd);
+ cERROR(1, "%s: error %d getting sec desc", __func__, rc);
+ } else {
/* allocate memory for the smb header,
set security descriptor request security descriptor
parameters, and secuirty descriptor itself */
seq_printf(s, ",acl");
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS)
seq_printf(s, ",mfsymlinks");
+ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE)
+ seq_printf(s, ",fsc");
seq_printf(s, ",rsize=%d", cifs_sb->rsize);
seq_printf(s, ",wsize=%d", cifs_sb->wsize);
extern int cifs_get_inode_info_unix(struct inode **pinode,
const unsigned char *search_path,
struct super_block *sb, int xid);
-extern void cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb,
+extern int cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb,
struct cifs_fattr *fattr, struct inode *inode,
const char *path, const __u16 *pfid);
-extern int mode_to_acl(struct inode *inode, const char *path, __u64);
+extern int mode_to_cifs_acl(struct inode *inode, const char *path, __u64);
+extern struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *, struct inode *,
+ const char *, u32 *);
extern int cifs_mount(struct super_block *, struct cifs_sb_info *, char *,
const char *);
"supported. Instead set "
"/proc/fs/cifs/LookupCacheEnabled to 0\n");
} else if (strnicmp(data, "fsc", 3) == 0) {
+#ifndef CONFIG_CIFS_FSCACHE
+ cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE"
+ "kernel config option set");
+ return 1;
+#endif
vol->fsc = true;
} else if (strnicmp(data, "mfsymlinks", 10) == 0) {
vol->mfsymlinks = true;
/* Search for server name delimiter */
sep = memchr(hostname, '\\', len);
if (sep)
- len = sep - unc;
+ len = sep - hostname;
else
cFYI(1, "%s: probably server name is whole unc: %s",
__func__, unc);
void cifs_oplock_break_put(struct cifsFileInfo *cfile)
{
+ struct super_block *sb = cfile->dentry->d_sb;
+
cifsFileInfo_put(cfile);
- cifs_sb_deactive(cfile->dentry->d_sb);
+ cifs_sb_deactive(sb);
}
const struct address_space_operations cifs_addr_ops = {
* fs/cifs/fscache.c - CIFS filesystem cache interface
*
* Copyright (c) 2010 Novell, Inc.
- * Author(s): Suresh Jayaraman (sjayaraman@suse.de>
+ * Author(s): Suresh Jayaraman <sjayaraman@suse.de>
*
* This library is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published
if (cifsi->fscache)
return;
- cifsi->fscache = fscache_acquire_cookie(tcon->fscache,
+ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) {
+ cifsi->fscache = fscache_acquire_cookie(tcon->fscache,
&cifs_fscache_inode_object_def, cifsi);
- cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache,
+ cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache,
cifsi->fscache);
+ }
}
void cifs_fscache_release_inode_cookie(struct inode *inode)
{
if ((filp->f_flags & O_ACCMODE) != O_RDONLY)
cifs_fscache_disable_inode_cookie(inode);
- else {
+ else
cifs_fscache_enable_inode_cookie(inode);
- cFYI(1, "CIFS: fscache inode cookie set");
- }
}
void cifs_fscache_reset_inode_cookie(struct inode *inode)
#ifdef CONFIG_CIFS_EXPERIMENTAL
/* fill in 0777 bits from ACL */
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) {
- cFYI(1, "Getting mode bits from ACL");
- cifs_acl_to_fattr(cifs_sb, &fattr, *pinode, full_path, pfid);
+ rc = cifs_acl_to_fattr(cifs_sb, &fattr, *pinode, full_path,
+ pfid);
+ if (rc) {
+ cFYI(1, "%s: Getting ACL failed with error: %d",
+ __func__, rc);
+ goto cgii_exit;
+ }
}
#endif
rc = cifs_get_inode_info(&inode, full_path, NULL, sb,
xid, NULL);
- if (!inode)
- return ERR_PTR(rc);
+ if (!inode) {
+ inode = ERR_PTR(rc);
+ goto out;
+ }
#ifdef CONFIG_CIFS_FSCACHE
/* populate tcon->resource_id */
inode->i_uid = cifs_sb->mnt_uid;
inode->i_gid = cifs_sb->mnt_gid;
} else if (rc) {
- kfree(full_path);
- _FreeXid(xid);
iget_failed(inode);
- return ERR_PTR(rc);
+ inode = ERR_PTR(rc);
}
-
+out:
kfree(full_path);
/* can not call macro FreeXid here since in a void func
* TODO: This is no longer true
return false;
}
-/* check invalid_mapping flag and zap the cache if it's set */
+/*
+ * Zap the cache. Called when invalid_mapping flag is set.
+ */
static void
cifs_invalidate_mapping(struct inode *inode)
{
if (attrs->ia_valid & ATTR_MODE) {
rc = 0;
#ifdef CONFIG_CIFS_EXPERIMENTAL
- if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL)
- rc = mode_to_acl(inode, full_path, mode);
- else
+ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) {
+ rc = mode_to_cifs_acl(inode, full_path, mode);
+ if (rc) {
+ cFYI(1, "%s: Setting ACL failed with error: %d",
+ __func__, rc);
+ goto cifs_setattr_exit;
+ }
+ } else
#endif
if (((mode & S_IWUGO) == 0) &&
(cifsInode->cifsAttrs & ATTR_READONLY) == 0) {
char *full_path = NULL;
struct cifsFileInfo *cifsFile;
struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
- struct tcon_link *tlink;
+ struct tcon_link *tlink = NULL;
struct cifsTconInfo *pTcon;
- tlink = cifs_sb_tlink(cifs_sb);
- if (IS_ERR(tlink))
- return PTR_ERR(tlink);
- pTcon = tlink_tcon(tlink);
-
- if (file->private_data == NULL)
- file->private_data =
- kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL);
if (file->private_data == NULL) {
- rc = -ENOMEM;
- goto error_exit;
+ tlink = cifs_sb_tlink(cifs_sb);
+ if (IS_ERR(tlink))
+ return PTR_ERR(tlink);
+
+ cifsFile = kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL);
+ if (cifsFile == NULL) {
+ rc = -ENOMEM;
+ goto error_exit;
+ }
+ file->private_data = cifsFile;
+ cifsFile->tlink = cifs_get_tlink(tlink);
+ pTcon = tlink_tcon(tlink);
+ } else {
+ cifsFile = file->private_data;
+ pTcon = tlink_tcon(cifsFile->tlink);
}
- cifsFile = file->private_data;
cifsFile->invalidHandle = true;
cifsFile->srch_inf.endOfSearch = false;
- cifsFile->tlink = cifs_get_tlink(tlink);
full_path = build_path_from_dentry(file->f_path.dentry);
if (full_path == NULL) {
#define MAX_EA_VALUE_SIZE 65535
#define CIFS_XATTR_DOS_ATTRIB "user.DosAttrib"
+#define CIFS_XATTR_CIFS_ACL "system.cifs_acl"
#define CIFS_XATTR_USER_PREFIX "user."
#define CIFS_XATTR_SYSTEM_PREFIX "system."
#define CIFS_XATTR_OS2_PREFIX "os2."
-#define CIFS_XATTR_SECURITY_PREFIX ".security"
+#define CIFS_XATTR_SECURITY_PREFIX "security."
#define CIFS_XATTR_TRUSTED_PREFIX "trusted."
#define XATTR_TRUSTED_PREFIX_LEN 8
#define XATTR_SECURITY_PREFIX_LEN 9
cifs_sb->local_nls,
cifs_sb->mnt_cifs_flags &
CIFS_MOUNT_MAP_SPECIAL_CHR);
-#ifdef CONFIG_CIFS_EXPERIMENTAL
- else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) {
- __u16 fid;
- int oplock = 0;
- struct cifs_ntsd *pacl = NULL;
- __u32 buflen = 0;
- if (experimEnabled)
- rc = CIFSSMBOpen(xid, pTcon, full_path,
- FILE_OPEN, GENERIC_READ, 0, &fid,
- &oplock, NULL, cifs_sb->local_nls,
- cifs_sb->mnt_cifs_flags &
- CIFS_MOUNT_MAP_SPECIAL_CHR);
- /* else rc is EOPNOTSUPP from above */
-
- if (rc == 0) {
- rc = CIFSSMBGetCIFSACL(xid, pTcon, fid, &pacl,
- &buflen);
- CIFSSMBClose(xid, pTcon, fid);
- }
- }
-#endif /* EXPERIMENTAL */
#else
- cFYI(1, "query POSIX ACL not supported yet");
+ cFYI(1, "Query POSIX ACL not supported yet");
#endif /* CONFIG_CIFS_POSIX */
} else if (strncmp(ea_name, POSIX_ACL_XATTR_DEFAULT,
strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) {
cifs_sb->mnt_cifs_flags &
CIFS_MOUNT_MAP_SPECIAL_CHR);
#else
- cFYI(1, "query POSIX default ACL not supported yet");
-#endif
+ cFYI(1, "Query POSIX default ACL not supported yet");
+#endif /* CONFIG_CIFS_POSIX */
+ } else if (strncmp(ea_name, CIFS_XATTR_CIFS_ACL,
+ strlen(CIFS_XATTR_CIFS_ACL)) == 0) {
+#ifdef CONFIG_CIFS_ACL
+ u32 acllen;
+ struct cifs_ntsd *pacl;
+
+ pacl = get_cifs_acl(cifs_sb, direntry->d_inode,
+ full_path, &acllen);
+ if (IS_ERR(pacl)) {
+ rc = PTR_ERR(pacl);
+ cERROR(1, "%s: error %zd getting sec desc",
+ __func__, rc);
+ } else {
+ if (ea_value) {
+ if (acllen > buf_size)
+ acllen = -ERANGE;
+ else
+ memcpy(ea_value, pacl, acllen);
+ }
+ rc = acllen;
+ kfree(pacl);
+ }
+#else
+ cFYI(1, "Query CIFS ACL not supported yet");
+#endif /* CONFIG_CIFS_ACL */
} else if (strncmp(ea_name,
CIFS_XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) == 0) {
cFYI(1, "Trusted xattr namespace not supported yet");
argv++;
if (i++ >= max)
return -E2BIG;
+
+ if (fatal_signal_pending(current))
+ return -ERESTARTNOHAND;
+ cond_resched();
}
}
return i;
while (len > 0) {
int offset, bytes_to_copy;
+ if (fatal_signal_pending(current)) {
+ ret = -ERESTARTNOHAND;
+ goto out;
+ }
+ cond_resched();
+
offset = pos % PAGE_SIZE;
if (offset == 0)
offset = PAGE_SIZE;
if (!kmapped_page || kpos != (pos & PAGE_MASK)) {
struct page *page;
-#ifdef CONFIG_STACK_GROWSUP
- ret = expand_stack_downwards(bprm->vma, pos);
- if (ret < 0) {
- /* We've exceed the stack rlimit. */
- ret = -E2BIG;
- goto out;
- }
-#endif
- ret = get_user_pages(current, bprm->mm, pos,
- 1, 1, 1, &page, NULL);
- if (ret <= 0) {
- /* We've exceed the stack rlimit. */
+ page = get_arg_page(bprm, pos, 1);
+ if (!page) {
ret = -E2BIG;
goto out;
}
return retval;
out:
- if (bprm->mm)
+ if (bprm->mm) {
+ acct_arg_size(bprm, 0);
mmput(bprm->mm);
+ }
out_file:
if (bprm->file) {
#ifdef CONFIG_MMU
-static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
+void acct_arg_size(struct linux_binprm *bprm, unsigned long pages)
+{
+ struct mm_struct *mm = current->mm;
+ long diff = (long)(pages - bprm->vma_pages);
+
+ if (!mm || !diff)
+ return;
+
+ bprm->vma_pages = pages;
+
+#ifdef SPLIT_RSS_COUNTING
+ add_mm_counter(mm, MM_ANONPAGES, diff);
+#else
+ spin_lock(&mm->page_table_lock);
+ add_mm_counter(mm, MM_ANONPAGES, diff);
+ spin_unlock(&mm->page_table_lock);
+#endif
+}
+
+struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
int write)
{
struct page *page;
unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start;
struct rlimit *rlim;
+ acct_arg_size(bprm, size / PAGE_SIZE);
+
/*
* We've historically supported up to 32 pages (ARG_MAX)
* of argument strings even with small stacks
#else
-static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
+void acct_arg_size(struct linux_binprm *bprm, unsigned long pages)
+{
+}
+
+struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
int write)
{
struct page *page;
/*
* Release all of the old mmap stuff
*/
+ acct_arg_size(bprm, 0);
retval = exec_mmap(bprm->mm);
if (retval)
goto out;
return retval;
out:
- if (bprm->mm)
- mmput (bprm->mm);
+ if (bprm->mm) {
+ acct_arg_size(bprm, 0);
+ mmput(bprm->mm);
+ }
out_file:
if (bprm->file) {
void fuse_finish_open(struct inode *inode, struct file *file)
{
struct fuse_file *ff = file->private_data;
+ struct fuse_conn *fc = get_fuse_conn(inode);
if (ff->open_flags & FOPEN_DIRECT_IO)
file->f_op = &fuse_direct_io_file_operations;
invalidate_inode_pages2(inode->i_mapping);
if (ff->open_flags & FOPEN_NONSEEKABLE)
nonseekable_open(inode, file);
+ if (fc->atomic_o_trunc && (file->f_flags & O_TRUNC)) {
+ struct fuse_inode *fi = get_fuse_inode(inode);
+
+ spin_lock(&fc->lock);
+ fi->attr_version = ++fc->attr_version;
+ i_size_write(inode, 0);
+ spin_unlock(&fc->lock);
+ fuse_invalidate_attr(inode);
+ }
}
int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
struct fs_disk_quota *fdq)
{
struct inode *inode = &ip->i_inode;
+ struct gfs2_sbd *sdp = GFS2_SB(inode);
struct address_space *mapping = inode->i_mapping;
unsigned long index = loc >> PAGE_CACHE_SHIFT;
unsigned offset = loc & (PAGE_CACHE_SIZE - 1);
qd->qd_qb.qb_value = qp->qu_value;
if (fdq) {
if (fdq->d_fieldmask & FS_DQ_BSOFT) {
- qp->qu_warn = cpu_to_be64(fdq->d_blk_softlimit);
+ qp->qu_warn = cpu_to_be64(fdq->d_blk_softlimit >> sdp->sd_fsb2bb_shift);
qd->qd_qb.qb_warn = qp->qu_warn;
}
if (fdq->d_fieldmask & FS_DQ_BHARD) {
- qp->qu_limit = cpu_to_be64(fdq->d_blk_hardlimit);
+ qp->qu_limit = cpu_to_be64(fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift);
qd->qd_qb.qb_limit = qp->qu_limit;
}
}
fdq->d_version = FS_DQUOT_VERSION;
fdq->d_flags = (type == QUOTA_USER) ? FS_USER_QUOTA : FS_GROUP_QUOTA;
fdq->d_id = id;
- fdq->d_blk_hardlimit = be64_to_cpu(qlvb->qb_limit);
- fdq->d_blk_softlimit = be64_to_cpu(qlvb->qb_warn);
- fdq->d_bcount = be64_to_cpu(qlvb->qb_value);
+ fdq->d_blk_hardlimit = be64_to_cpu(qlvb->qb_limit) << sdp->sd_fsb2bb_shift;
+ fdq->d_blk_softlimit = be64_to_cpu(qlvb->qb_warn) << sdp->sd_fsb2bb_shift;
+ fdq->d_bcount = be64_to_cpu(qlvb->qb_value) << sdp->sd_fsb2bb_shift;
gfs2_glock_dq_uninit(&q_gh);
out:
/* If nothing has changed, this is a no-op */
if ((fdq->d_fieldmask & FS_DQ_BSOFT) &&
- (fdq->d_blk_softlimit == be64_to_cpu(qd->qd_qb.qb_warn)))
+ ((fdq->d_blk_softlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_warn)))
fdq->d_fieldmask ^= FS_DQ_BSOFT;
if ((fdq->d_fieldmask & FS_DQ_BHARD) &&
- (fdq->d_blk_hardlimit == be64_to_cpu(qd->qd_qb.qb_limit)))
+ ((fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_limit)))
fdq->d_fieldmask ^= FS_DQ_BHARD;
if (fdq->d_fieldmask == 0)
goto out_i;
}
ret = -ESRCH;
- /*
- * We want IOPRIO_WHO_PGRP/IOPRIO_WHO_USER to be "atomic",
- * so we can't use rcu_read_lock(). See re-copy of ->ioprio
- * in copy_process().
- */
- read_lock(&tasklist_lock);
+ rcu_read_lock();
switch (which) {
case IOPRIO_WHO_PROCESS:
- rcu_read_lock();
if (!who)
p = current;
else
p = find_task_by_vpid(who);
if (p)
ret = set_task_ioprio(p, ioprio);
- rcu_read_unlock();
break;
case IOPRIO_WHO_PGRP:
if (!who)
break;
do_each_thread(g, p) {
- int match;
-
- rcu_read_lock();
- match = __task_cred(p)->uid == who;
- rcu_read_unlock();
- if (!match)
+ if (__task_cred(p)->uid != who)
continue;
ret = set_task_ioprio(p, ioprio);
if (ret)
ret = -EINVAL;
}
- read_unlock(&tasklist_lock);
+ rcu_read_unlock();
return ret;
}
int ret = -ESRCH;
int tmpio;
- read_lock(&tasklist_lock);
+ rcu_read_lock();
switch (which) {
case IOPRIO_WHO_PROCESS:
- rcu_read_lock();
if (!who)
p = current;
else
p = find_task_by_vpid(who);
if (p)
ret = get_task_ioprio(p);
- rcu_read_unlock();
break;
case IOPRIO_WHO_PGRP:
if (!who)
break;
do_each_thread(g, p) {
- int match;
-
- rcu_read_lock();
- match = __task_cred(p)->uid == user->uid;
- rcu_read_unlock();
- if (!match)
+ if (__task_cred(p)->uid != user->uid)
continue;
tmpio = get_task_ioprio(p);
if (tmpio < 0)
ret = -EINVAL;
}
- read_unlock(&tasklist_lock);
+ rcu_read_unlock();
return ret;
}
u64 cookie;
u64 ino;
struct qstr string;
+ unsigned char d_type;
};
struct nfs_cache_array {
struct nfs_cache_array_entry array[0];
};
-#define MAX_READDIR_ARRAY ((PAGE_SIZE - sizeof(struct nfs_cache_array)) / sizeof(struct nfs_cache_array_entry))
-
typedef __be32 * (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, struct nfs_server *, int);
typedef struct {
struct file *file;
if (IS_ERR(array))
return PTR_ERR(array);
+
+ cache_entry = &array->array[array->size];
+
+ /* Check that this entry lies within the page bounds */
ret = -ENOSPC;
- if (array->size >= MAX_READDIR_ARRAY)
+ if ((char *)&cache_entry[1] - (char *)page_address(page) > PAGE_SIZE)
goto out;
- cache_entry = &array->array[array->size];
cache_entry->cookie = entry->prev_cookie;
cache_entry->ino = entry->ino;
+ cache_entry->d_type = entry->d_type;
ret = nfs_readdir_make_qstr(&cache_entry->string, entry->name, entry->len);
if (ret)
goto out;
static
int nfs_same_file(struct dentry *dentry, struct nfs_entry *entry)
{
- struct nfs_inode *node;
if (dentry->d_inode == NULL)
goto different;
- node = NFS_I(dentry->d_inode);
- if (node->fh.size != entry->fh->size)
- goto different;
- if (strncmp(node->fh.data, entry->fh->data, node->fh.size) != 0)
+ if (nfs_compare_fh(entry->fh, NFS_FH(dentry->d_inode)) != 0)
goto different;
return 1;
different:
struct xdr_stream stream;
struct xdr_buf buf;
__be32 *ptr = xdr_page;
- int status;
struct nfs_cache_array *array;
+ unsigned int count = 0;
+ int status;
buf.head->iov_base = xdr_page;
buf.head->iov_len = buflen;
break;
}
+ count++;
+
if (desc->plus == 1)
nfs_prime_dcache(desc->file->f_path.dentry, entry);
break;
} while (!entry->eof);
- if (status == -EBADCOOKIE && entry->eof) {
+ if (count == 0 || (status == -EBADCOOKIE && entry->eof == 1)) {
array = nfs_readdir_get_array(page);
if (!IS_ERR(array)) {
array->eof_index = array->size;
status = 0;
nfs_readdir_release_array(page);
- }
+ } else
+ status = PTR_ERR(array);
}
return status;
}
int i = 0;
int res = 0;
struct nfs_cache_array *array = NULL;
- unsigned int d_type = DT_UNKNOWN;
- struct dentry *dentry = NULL;
array = nfs_readdir_get_array(desc->page);
- if (IS_ERR(array))
- return PTR_ERR(array);
+ if (IS_ERR(array)) {
+ res = PTR_ERR(array);
+ goto out;
+ }
for (i = desc->cache_entry_index; i < array->size; i++) {
- d_type = DT_UNKNOWN;
+ struct nfs_cache_array_entry *ent;
- res = filldir(dirent, array->array[i].string.name,
- array->array[i].string.len, file->f_pos,
- nfs_compat_user_ino64(array->array[i].ino), d_type);
- if (res < 0)
+ ent = &array->array[i];
+ if (filldir(dirent, ent->string.name, ent->string.len,
+ file->f_pos, nfs_compat_user_ino64(ent->ino),
+ ent->d_type) < 0) {
+ desc->eof = 1;
break;
+ }
file->f_pos++;
desc->cache_entry_index = i;
if (i < (array->size-1))
desc->eof = 1;
nfs_readdir_release_array(desc->page);
+out:
cache_page_release(desc);
- if (dentry != NULL)
- dput(dentry);
dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n",
(unsigned long long)*desc->dir_cookie, res);
return res;
goto out;
}
- if (nfs_readdir_xdr_to_array(desc, page, inode) == -1) {
- status = -EIO;
- goto out_release;
- }
-
desc->page_index = 0;
desc->page = page;
+
+ status = nfs_readdir_xdr_to_array(desc, page, inode);
+ if (status < 0)
+ goto out_release;
+
status = nfs_do_filldir(desc, dirent, filldir);
out:
res = readdir_search_pagecache(desc);
if (res == -EBADCOOKIE) {
+ res = 0;
/* This means either end of directory */
if (*desc->dir_cookie && desc->eof == 0) {
/* Or that the server has 'lost' a cookie */
res = uncached_readdir(desc, dirent, filldir);
- if (res >= 0)
+ if (res == 0)
continue;
}
- res = 0;
break;
}
if (res == -ETOOSMALL && desc->plus) {
break;
res = nfs_do_filldir(desc, dirent, filldir);
- if (res < 0) {
- res = 0;
+ if (res < 0)
break;
- }
}
out:
nfs_unblock_sillyrename(dentry);
goto out;
nfs_alloc_commit_data(dreq);
- if (dreq->commit_data == NULL || count < wsize)
+ if (dreq->commit_data == NULL || count <= wsize)
sync = NFS_FILE_SYNC;
dreq->inode = inode;
return 0;
}
+/*
+ * Convert a umode to a dirent->d_type
+ */
+static inline
+unsigned char nfs_umode_to_dtype(umode_t mode)
+{
+ return (mode >> 12) & 15;
+}
+
/*
* Determine the number of pages in an array of length 'len' and
* with a base offset of 'base'
entry->prev_cookie = entry->cookie;
entry->cookie = ntohl(*p++);
+ entry->d_type = DT_UNKNOWN;
+
p = xdr_inline_peek(xdr, 8);
if (p != NULL)
entry->eof = !p[0] && p[1];
out_overflow:
print_overflow_msg(__func__, xdr);
- return ERR_PTR(-EIO);
+ return ERR_PTR(-EAGAIN);
}
/*
entry->prev_cookie = entry->cookie;
p = xdr_decode_hyper(p, &entry->cookie);
+ entry->d_type = DT_UNKNOWN;
if (plus) {
entry->fattr->valid = 0;
p = xdr_decode_post_op_attr_stream(xdr, entry->fattr);
if (IS_ERR(p))
goto out_overflow_exit;
+ entry->d_type = nfs_umode_to_dtype(entry->fattr->mode);
/* In fact, a post_op_fh3: */
p = xdr_inline_decode(xdr, 4);
if (unlikely(!p))
out_overflow:
print_overflow_msg(__func__, xdr);
out_overflow_exit:
- return ERR_PTR(-EIO);
+ return ERR_PTR(-EAGAIN);
}
/*
if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID)
entry->ino = entry->fattr->fileid;
+ entry->d_type = DT_UNKNOWN;
+ if (entry->fattr->valid & NFS_ATTR_FATTR_TYPE)
+ entry->d_type = nfs_umode_to_dtype(entry->fattr->mode);
+
if (verify_attr_len(xdr, p, len) < 0)
goto out_overflow;
out_overflow:
print_overflow_msg(__func__, xdr);
- return ERR_PTR(-EIO);
+ return ERR_PTR(-EAGAIN);
}
/*
* the device at this point.
*
* To prevent nilfs_dat_translate() from returning the
- * uncommited block number, this makes a copy of the entry
+ * uncommitted block number, this makes a copy of the entry
* buffer and redirects nilfs_dat_translate() to the copy.
*/
if (!buffer_nilfs_redirected(entry_bh)) {
ino = vdesc->vd_ino;
cno = vdesc->vd_cno;
inode = nilfs_iget_for_gc(sb, ino, cno);
- if (unlikely(inode == NULL)) {
- ret = -ENOMEM;
+ if (IS_ERR(inode)) {
+ ret = PTR_ERR(inode);
goto failed;
}
do {
return ret;
}
+/*
+ * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same
+ * location, so checking ->i_pipe is not enough to verify that this is a
+ * pipe.
+ */
+struct pipe_inode_info *get_pipe_info(struct file *file)
+{
+ struct inode *i = file->f_path.dentry->d_inode;
+
+ return S_ISFIFO(i->i_mode) ? i->i_pipe : NULL;
+}
+
long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg)
{
struct pipe_inode_info *pipe;
long ret;
- pipe = file->f_path.dentry->d_inode->i_pipe;
+ pipe = get_pipe_info(file);
if (!pipe)
return -EBADF;
* skip over unmapped regions.
*/
#define PAGEMAP_WALK_SIZE (PMD_SIZE)
+#define PAGEMAP_WALK_MASK (PMD_MASK)
static ssize_t pagemap_read(struct file *file, char __user *buf,
size_t count, loff_t *ppos)
{
unsigned long end;
pm.pos = 0;
- end = start_vaddr + PAGEMAP_WALK_SIZE;
+ end = (start_vaddr + PAGEMAP_WALK_SIZE) & PAGEMAP_WALK_MASK;
/* overflow ? */
if (end < start_vaddr || end > end_vaddr)
end = end_vaddr;
return 0;
}
- /* we need to make sure nobody is changing the file size beneath
- ** us
- */
- reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb);
depth = reiserfs_write_lock_once(inode->i_sb);
+ /* we need to make sure nobody is changing the file size beneath us */
+ reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb);
+
write_from = inode->i_size & (blocksize - 1);
/* if we are on a block boundary, we are already unpacked. */
if (write_from == 0) {
static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe,
struct pipe_inode_info *opipe,
size_t len, unsigned int flags);
-/*
- * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same
- * location, so checking ->i_pipe is not enough to verify that this is a
- * pipe.
- */
-static inline struct pipe_inode_info *pipe_info(struct inode *inode)
-{
- if (S_ISFIFO(inode->i_mode))
- return inode->i_pipe;
-
- return NULL;
-}
/*
* Determine where to splice to/from.
loff_t offset, *off;
long ret;
- ipipe = pipe_info(in->f_path.dentry->d_inode);
- opipe = pipe_info(out->f_path.dentry->d_inode);
+ ipipe = get_pipe_info(in);
+ opipe = get_pipe_info(out);
if (ipipe && opipe) {
if (off_in || off_out)
int error;
long ret;
- pipe = pipe_info(file->f_path.dentry->d_inode);
+ pipe = get_pipe_info(file);
if (!pipe)
return -EBADF;
};
long ret;
- pipe = pipe_info(file->f_path.dentry->d_inode);
+ pipe = get_pipe_info(file);
if (!pipe)
return -EBADF;
static long do_tee(struct file *in, struct file *out, size_t len,
unsigned int flags)
{
- struct pipe_inode_info *ipipe = pipe_info(in->f_path.dentry->d_inode);
- struct pipe_inode_info *opipe = pipe_info(out->f_path.dentry->d_inode);
+ struct pipe_inode_info *ipipe = get_pipe_info(in);
+ struct pipe_inode_info *opipe = get_pipe_info(out);
int ret = -EINVAL;
/*
char buf[BINPRM_BUF_SIZE];
#ifdef CONFIG_MMU
struct vm_area_struct *vma;
+ unsigned long vma_pages;
#else
# define MAX_ARG_PAGES 32
struct page *page[MAX_ARG_PAGES];
unsigned long loader, exec;
};
+extern void acct_arg_size(struct linux_binprm *bprm, unsigned long pages);
+extern struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
+ int write);
+
#define BINPRM_FLAGS_ENFORCE_NONDUMP_BIT 0
#define BINPRM_FLAGS_ENFORCE_NONDUMP (1 << BINPRM_FLAGS_ENFORCE_NONDUMP_BIT)
return 0;
}
-#define enable_intr_remapping(mode) (-1)
-#define disable_intr_remapping() (0)
-#define reenable_intr_remapping(mode) (0)
#define intr_remapping_enabled (0)
+
+static inline int enable_intr_remapping(int eim)
+{
+ return -1;
+}
+
+static inline void disable_intr_remapping(void)
+{
+}
+
+static inline int reenable_intr_remapping(int eim)
+{
+ return 0;
+}
#endif
/* Can't use the common MSI interrupt functions
/* drivers/video/fbcmap.c */
extern int fb_alloc_cmap(struct fb_cmap *cmap, int len, int transp);
+extern int fb_alloc_cmap_gfp(struct fb_cmap *cmap, int len, int transp, gfp_t flags);
extern void fb_dealloc_cmap(struct fb_cmap *cmap);
extern int fb_copy_cmap(const struct fb_cmap *from, struct fb_cmap *to);
extern int fb_cmap_to_user(const struct fb_cmap *from, struct fb_cmap_user *to);
#define SEEK_MAX SEEK_END
struct fstrim_range {
- uint64_t start;
- uint64_t len;
- uint64_t minlen;
+ __u64 start;
+ __u64 len;
+ __u64 minlen;
};
/* And dynamically-tunable limits and defaults: */
#ifdef CONFIG_HAVE_HW_BREAKPOINT
+extern int __init init_hw_breakpoint(void);
+
static inline void hw_breakpoint_init(struct perf_event_attr *attr)
{
memset(attr, 0, sizeof(*attr));
#else /* !CONFIG_HAVE_HW_BREAKPOINT */
+static inline int __init init_hw_breakpoint(void) { return 0; }
+
static inline struct perf_event *
register_user_hw_breakpoint(struct perf_event_attr *attr,
perf_overflow_handler_t triggered,
#define MARVELL_PHY_ID_88E1118 0x01410e10
#define MARVELL_PHY_ID_88E1121R 0x01410cb0
#define MARVELL_PHY_ID_88E1145 0x01410cd0
+#define MARVELL_PHY_ID_88E1149R 0x01410e50
#define MARVELL_PHY_ID_88E1240 0x01410e30
#define MARVELL_PHY_ID_88E1318S 0x01410e90
#define WM8350_MCLK_SEL_PLL_32K 3
#define WM8350_MCLK_SEL_MCLK 5
-#define WM8350_MCLK_DIR_OUT 0
-#define WM8350_MCLK_DIR_IN 1
-
/* clock divider id's */
#define WM8350_ADC_CLKDIV 0
#define WM8350_DAC_CLKDIV 1
/* DDR mode at 1.8V */
#define MMC_CAP_1_2V_DDR (1 << 12) /* can support */
/* DDR mode at 1.2V */
+#define MMC_CAP_POWER_OFF_CARD (1 << 13) /* Can power off after boot */
mmc_pm_flag_t pm_caps; /* supported pm features */
#define symbol_put_addr(p) do { } while(0)
#endif /* CONFIG_MODULE_UNLOAD */
-int use_module(struct module *a, struct module *b);
+int ref_module(struct module *a, struct module *b);
/* This is a #define so the string doesn't get put in every .o file */
#define module_name(mod) \
int eof;
struct nfs_fh * fh;
struct nfs_fattr * fattr;
+ unsigned char d_type;
};
/*
static inline int TestClearPageCgroup##uname(struct page_cgroup *pc) \
{ return test_and_clear_bit(PCG_##lname, &pc->flags); }
-TESTPCGFLAG(Locked, LOCK)
-
/* Cache flag is set only once (at allocation) */
TESTPCGFLAG(Cache, CACHE)
CLEARPCGFLAG(Cache, CACHE)
bit_spin_unlock(PCG_LOCK, &pc->flags);
}
+static inline int page_is_cgroup_locked(struct page_cgroup *pc)
+{
+ return bit_spin_is_locked(PCG_LOCK, &pc->flags);
+}
+
#else /* CONFIG_CGROUP_MEM_RES_CTLR */
struct page_cgroup;
#define PCI_DEVICE_ID_AFAVLAB_P030 0x2182
#define PCI_SUBDEVICE_ID_AFAVLAB_P061 0x2150
+#define PCI_VENDOR_ID_BCM_GVC 0x14a4
#define PCI_VENDOR_ID_BROADCOM 0x14e4
#define PCI_DEVICE_ID_TIGON3_5752 0x1600
#define PCI_DEVICE_ID_TIGON3_5752M 0x1601
#define PCI_DEVICE_ID_INTEL_MFD_SDIO2 0x0822
#define PCI_DEVICE_ID_INTEL_MFD_EMMC0 0x0823
#define PCI_DEVICE_ID_INTEL_MFD_EMMC1 0x0824
+#define PCI_DEVICE_ID_INTEL_MRST_SD2 0x084F
#define PCI_DEVICE_ID_INTEL_I960 0x0960
#define PCI_DEVICE_ID_INTEL_I960RM 0x0962
#define PCI_DEVICE_ID_INTEL_8257X_SOL 0x1062
int nr_active;
int is_active;
int nr_stat;
+ int rotate_disable;
atomic_t refcount;
struct task_struct *task;
extern const char *perf_pmu_name(void);
extern void __perf_event_task_sched_in(struct task_struct *task);
extern void __perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
-
-extern atomic_t perf_task_events;
-
-static inline void perf_event_task_sched_in(struct task_struct *task)
-{
- COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
-}
-
-static inline
-void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
-{
- COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
-}
-
extern int perf_event_init_task(struct task_struct *child);
extern void perf_event_exit_task(struct task_struct *child);
extern void perf_event_free_task(struct task_struct *task);
__perf_sw_event(event_id, nr, nmi, regs, addr);
}
+extern atomic_t perf_task_events;
+
+static inline void perf_event_task_sched_in(struct task_struct *task)
+{
+ COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
+}
+
+static inline
+void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
+{
+ perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+
+ COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
+}
+
extern void perf_event_mmap(struct vm_area_struct *vma);
extern struct perf_guest_info_callbacks *perf_guest_cbs;
extern int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks *callbacks);
/* for F_SETPIPE_SZ and F_GETPIPE_SZ */
long pipe_fcntl(struct file *, unsigned int, unsigned long arg);
+struct pipe_inode_info *get_pipe_info(struct file *file);
#endif
};
struct clk_ops {
+#ifdef CONFIG_SH_CLK_CPG_LEGACY
void (*init)(struct clk *clk);
+#endif
int (*enable)(struct clk *clk);
void (*disable)(struct clk *clk);
unsigned long (*recalc)(struct clk *clk);
- int (*set_rate)(struct clk *clk, unsigned long rate, int algo_id);
+ int (*set_rate)(struct clk *clk, unsigned long rate);
int (*set_parent)(struct clk *clk, struct clk *parent);
long (*round_rate)(struct clk *clk, unsigned long rate);
};
void clk_unregister(struct clk *);
void clk_enable_init_clocks(void);
-/**
- * clk_set_rate_ex - set the clock rate for a clock source, with additional parameter
- * @clk: clock source
- * @rate: desired clock rate in Hz
- * @algo_id: algorithm id to be passed down to ops->set_rate
- *
- * Returns success (0) or negative errno.
- */
-int clk_set_rate_ex(struct clk *clk, unsigned long rate, int algo_id);
-
-enum clk_sh_algo_id {
- NO_CHANGE = 0,
-
- IUS_N1_N1,
- IUS_322,
- IUS_522,
- IUS_N11,
-
- SB_N1,
-
- SB3_N1,
- SB3_32,
- SB3_43,
- SB3_54,
-
- BP_N1,
-
- IP_N1,
-};
-
struct clk_div_mult_table {
unsigned int *divisors;
unsigned int nr_divisors;
extern void unix_notinflight(struct file *fp);
extern void unix_gc(void);
extern void wait_for_unix_gc(void);
+extern struct sock *unix_get_socket(struct file *filp);
#define UNIX_HASH_SIZE 256
spinlock_t lock;
unsigned int gc_candidate : 1;
unsigned int gc_maybe_cycle : 1;
+ unsigned char recursion_level;
struct socket_wq peer_wq;
};
#define unix_sk(__sk) ((struct unix_sock *)__sk)
* ACK_MD (FSI2)
* CKG1 (FSI)
*
- * err: return value < 0
+ * err : return value < 0
+ * no change : return value == 0
+ * change xMD : return value > 0
*
* 0x-00000AB
*
struct sh_fsi_platform_info {
unsigned long porta_flags;
unsigned long portb_flags;
- int (*set_rate)(int is_porta, int rate); /* for master mode */
+ int (*set_rate)(struct device *dev, int is_porta, int rate, int enable);
};
#endif /* __SOUND_FSI_H */
#define FBIPUT_COLOR _IOW('F', 6, int)
#define FBIPUT_HSYNC _IOW('F', 9, int)
#define FBIPUT_VSYNC _IOW('F', 10, int)
-#define FBIO_WAITFORVSYNC _IOW('F', 0x20, u_int32_t)
#endif /* ifndef DA8XX_FB_H */
};
DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mfn_list);
+/*
+ * Returns the location in virtual address space of the machine_to_phys
+ * mapping table. Architectures which do not have a m2p table, or which do not
+ * map it by default into guest address space, do not implement this command.
+ * arg == addr of xen_machphys_mapping_t.
+ */
+#define XENMEM_machphys_mapping 12
+struct xen_machphys_mapping {
+ unsigned long v_start, v_end; /* Start and end virtual addresses. */
+ unsigned long max_mfn; /* Maximum MFN that can be looked up. */
+};
+DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mapping_t);
+
/*
* Sets the GPFN at which a particular page appears in the specified guest's
* pseudophysical address space.
+#ifndef _XEN_PAGE_H
+#define _XEN_PAGE_H
+
#include <asm/xen/page.h>
+
+extern phys_addr_t xen_extra_mem_start, xen_extra_mem_size;
+
+#endif /* _XEN_PAGE_H */
#define __LINUX_PUBLIC_PRIVCMD_H__
#include <linux/types.h>
+#include <linux/compiler.h>
typedef unsigned long xen_pfn_t;
-#ifndef __user
-#define __user
-#endif
-
struct privcmd_hypercall {
__u64 op;
__u64 arg[5];
if boot option "noswapaccount" is set, swap will not be accounted.
Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
size is 4096bytes, 512k per 1Gbytes of swap.
+config CGROUP_MEM_RES_CTLR_SWAP_ENABLED
+ bool "Memory Resource Controller Swap Extension enabled by default"
+ depends on CGROUP_MEM_RES_CTLR_SWAP
+ default y
+ help
+ Memory Resource Controller Swap Extension comes with its price in
+ a bigger memory consumption. General purpose distribution kernels
+ which want to enable the feautre but keep it disabled by default
+ and let the user enable it by swapaccount boot command line
+ parameter should have this option unselected.
+ For those who want to have the feature enabled by default should
+ select this option (if, for some reason, they need to disable it
+ then noswapaccount does the trick).
menuconfig CGROUP_SCHED
bool "Group CPU scheduler"
.read = hw_breakpoint_pmu_read,
};
-static int __init init_hw_breakpoint(void)
+int __init init_hw_breakpoint(void)
{
unsigned int **task_bp_pinned;
int cpu, err_cpu;
return -ENOMEM;
}
-core_initcall(init_hw_breakpoint);
* Clear the BUSY bit and return to the free state if
* no-one else claimed it meanwhile.
*/
- cmpxchg(&entry->next, next_flags(NULL, IRQ_WORK_BUSY), NULL);
+ (void)cmpxchg(&entry->next,
+ next_flags(NULL, IRQ_WORK_BUSY),
+ NULL);
}
}
EXPORT_SYMBOL_GPL(irq_work_run);
kmemleak_scan_area(mod->trace_events, sizeof(*mod->trace_events) *
mod->num_trace_events, GFP_KERNEL);
#endif
+#ifdef CONFIG_TRACING
+ mod->trace_bprintk_fmt_start = section_objs(info, "__trace_printk_fmt",
+ sizeof(*mod->trace_bprintk_fmt_start),
+ &mod->num_trace_bprintk_fmt);
+ /*
+ * This section contains pointers to allocated objects in the trace
+ * code and not scanning it leads to false positives.
+ */
+ kmemleak_scan_area(mod->trace_bprintk_fmt_start,
+ sizeof(*mod->trace_bprintk_fmt_start) *
+ mod->num_trace_bprintk_fmt, GFP_KERNEL);
+#endif
#ifdef CONFIG_FTRACE_MCOUNT_RECORD
/* sechdrs[0].sh_size is always zero */
mod->ftrace_callsites = section_objs(info, "__mcount_loc",
#include <linux/kernel_stat.h>
#include <linux/perf_event.h>
#include <linux/ftrace_event.h>
+#include <linux/hw_breakpoint.h>
#include <asm/irq_regs.h>
{
int ctxn;
- perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
-
for_each_task_context_nr(ctxn)
perf_event_context_sched_out(task, ctxn, next);
}
{
raw_spin_lock(&ctx->lock);
- /* Rotate the first entry last of non-pinned groups */
- list_rotate_left(&ctx->flexible_groups);
+ /*
+ * Rotate the first entry last of non-pinned groups. Rotation might be
+ * disabled by the inheritance code.
+ */
+ if (!ctx->rotate_disable)
+ list_rotate_left(&ctx->flexible_groups);
raw_spin_unlock(&ctx->lock);
}
raw_spin_unlock_irq(&ctx->lock);
mutex_unlock(&ctx->mutex);
- mutex_lock(&event->owner->perf_event_mutex);
- list_del_init(&event->owner_entry);
- mutex_unlock(&event->owner->perf_event_mutex);
- put_task_struct(event->owner);
-
free_event(event);
return 0;
static int perf_release(struct inode *inode, struct file *file)
{
struct perf_event *event = file->private_data;
+ struct task_struct *owner;
file->private_data = NULL;
+ rcu_read_lock();
+ owner = ACCESS_ONCE(event->owner);
+ /*
+ * Matches the smp_wmb() in perf_event_exit_task(). If we observe
+ * !owner it means the list deletion is complete and we can indeed
+ * free this event, otherwise we need to serialize on
+ * owner->perf_event_mutex.
+ */
+ smp_read_barrier_depends();
+ if (owner) {
+ /*
+ * Since delayed_put_task_struct() also drops the last
+ * task reference we can safely take a new reference
+ * while holding the rcu_read_lock().
+ */
+ get_task_struct(owner);
+ }
+ rcu_read_unlock();
+
+ if (owner) {
+ mutex_lock(&owner->perf_event_mutex);
+ /*
+ * We have to re-check the event->owner field, if it is cleared
+ * we raced with perf_event_exit_task(), acquiring the mutex
+ * ensured they're done, and we can proceed with freeing the
+ * event.
+ */
+ if (event->owner)
+ list_del_init(&event->owner_entry);
+ mutex_unlock(&owner->perf_event_mutex);
+ put_task_struct(owner);
+ }
+
return perf_event_release_kernel(event);
}
mutex_unlock(&ctx->mutex);
event->owner = current;
- get_task_struct(current);
+
mutex_lock(¤t->perf_event_mutex);
list_add_tail(&event->owner_entry, ¤t->perf_event_list);
mutex_unlock(¤t->perf_event_mutex);
++ctx->generation;
mutex_unlock(&ctx->mutex);
- event->owner = current;
- get_task_struct(current);
- mutex_lock(¤t->perf_event_mutex);
- list_add_tail(&event->owner_entry, ¤t->perf_event_list);
- mutex_unlock(¤t->perf_event_mutex);
-
return event;
err_free:
*/
void perf_event_exit_task(struct task_struct *child)
{
+ struct perf_event *event, *tmp;
int ctxn;
+ mutex_lock(&child->perf_event_mutex);
+ list_for_each_entry_safe(event, tmp, &child->perf_event_list,
+ owner_entry) {
+ list_del_init(&event->owner_entry);
+
+ /*
+ * Ensure the list deletion is visible before we clear
+ * the owner, closes a race against perf_release() where
+ * we need to serialize on the owner->perf_event_mutex.
+ */
+ smp_wmb();
+ event->owner = NULL;
+ }
+ mutex_unlock(&child->perf_event_mutex);
+
for_each_task_context_nr(ctxn)
perf_event_exit_task_context(child, ctxn);
}
struct perf_event *event;
struct task_struct *parent = current;
int inherited_all = 1;
+ unsigned long flags;
int ret = 0;
child->perf_event_ctxp[ctxn] = NULL;
break;
}
+ /*
+ * We can't hold ctx->lock when iterating the ->flexible_group list due
+ * to allocations, but we need to prevent rotation because
+ * rotate_ctx() will change the list from interrupt context.
+ */
+ raw_spin_lock_irqsave(&parent_ctx->lock, flags);
+ parent_ctx->rotate_disable = 1;
+ raw_spin_unlock_irqrestore(&parent_ctx->lock, flags);
+
list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) {
ret = inherit_task_group(event, parent, parent_ctx,
child, ctxn, &inherited_all);
break;
}
+ raw_spin_lock_irqsave(&parent_ctx->lock, flags);
+ parent_ctx->rotate_disable = 0;
+ raw_spin_unlock_irqrestore(&parent_ctx->lock, flags);
+
child_ctx = child->perf_event_ctxp[ctxn];
if (child_ctx && inherited_all) {
void __init perf_event_init(void)
{
+ int ret;
+
perf_event_init_all_cpus();
init_srcu_struct(&pmus_srcu);
perf_pmu_register(&perf_swevent);
perf_pmu_register(&perf_task_clock);
perf_tp_register();
perf_cpu_notifier(perf_cpu_notify);
+
+ ret = init_hw_breakpoint();
+ WARN(ret, "hw_breakpoint initialization failed with: %d", ret);
}
if (pid == 0)
return 0;
- read_lock(&tasklist_lock);
+ rcu_read_lock();
p = find_task_by_vpid(pid);
if (!p || !(CPUCLOCK_PERTHREAD(which_clock) ?
- same_thread_group(p, current) : thread_group_leader(p))) {
+ same_thread_group(p, current) : has_group_leader_pid(p))) {
error = -EINVAL;
}
- read_unlock(&tasklist_lock);
+ rcu_read_unlock();
return error;
}
INIT_LIST_HEAD(&new_timer->it.cpu.entry);
- read_lock(&tasklist_lock);
+ rcu_read_lock();
if (CPUCLOCK_PERTHREAD(new_timer->it_clock)) {
if (pid == 0) {
p = current;
p = current->group_leader;
} else {
p = find_task_by_vpid(pid);
- if (p && !thread_group_leader(p))
+ if (p && !has_group_leader_pid(p))
p = NULL;
}
}
} else {
ret = -EINVAL;
}
- read_unlock(&tasklist_lock);
+ rcu_read_unlock();
return ret;
}
set_task_cpu(p, this_cpu);
activate_task(this_rq, p, 0);
check_preempt_curr(this_rq, p, 0);
-
- /* re-arm NEWIDLE balancing when moving tasks */
- src_rq->avg_idle = this_rq->avg_idle = 2*sysctl_sched_migration_cost;
- this_rq->idle_stamp = 0;
}
/*
interval = msecs_to_jiffies(sd->balance_interval);
if (time_after(next_balance, sd->last_balance + interval))
next_balance = sd->last_balance + interval;
- if (pulled_task)
+ if (pulled_task) {
+ this_rq->idle_stamp = 0;
break;
+ }
}
raw_spin_lock(&this_rq->lock);
__ftrace_trace_stack(global_trace.buffer, flags, 3, preempt_count());
}
+static DEFINE_PER_CPU(int, user_stack_count);
+
void
ftrace_trace_userstack(struct ring_buffer *buffer, unsigned long flags, int pc)
{
if (unlikely(in_nmi()))
return;
+ /*
+ * prevent recursion, since the user stack tracing may
+ * trigger other kernel events.
+ */
+ preempt_disable();
+ if (__this_cpu_read(user_stack_count))
+ goto out;
+
+ __this_cpu_inc(user_stack_count);
+
+
+
event = trace_buffer_lock_reserve(buffer, TRACE_USER_STACK,
sizeof(*entry), flags, pc);
if (!event)
save_stack_trace_user(&trace);
if (!filter_check_discard(call, entry, buffer, event))
ring_buffer_unlock_commit(buffer, event);
+
+ __this_cpu_dec(user_stack_count);
+
+ out:
+ preempt_enable();
}
#ifdef UNUSED
*
* Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
*/
-#include <linux/kernel.h>
#include <linux/rwsem.h>
#include <linux/mutex.h>
#include <linux/module.h>
{
if (__debug_locks_off()) {
if (!debug_locks_silent) {
- oops_in_progress = 1;
console_verbose();
return 1;
}
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
/* Turned on only when memory cgroup is enabled && really_do_swap_account = 1 */
int do_swap_account __read_mostly;
-static int really_do_swap_account __initdata = 1; /* for remember boot option*/
+
+/* for remember boot option*/
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP_ENABLED
+static int really_do_swap_account __initdata = 1;
+#else
+static int really_do_swap_account __initdata = 0;
+#endif
+
#else
#define do_swap_account (0)
#endif
/* "mc" and its members are protected by cgroup_mutex */
static struct move_charge_struct {
- spinlock_t lock; /* for from, to, moving_task */
+ spinlock_t lock; /* for from, to */
struct mem_cgroup *from;
struct mem_cgroup *to;
unsigned long precharge;
unsigned long moved_charge;
unsigned long moved_swap;
struct task_struct *moving_task; /* a task moving charges */
+ struct mm_struct *mm;
wait_queue_head_t waitq; /* a waitq for other context */
} mc = {
.lock = __SPIN_LOCK_UNLOCKED(mc.lock),
{
VM_BUG_ON(from == to);
VM_BUG_ON(PageLRU(pc->page));
- VM_BUG_ON(!PageCgroupLocked(pc));
+ VM_BUG_ON(!page_is_cgroup_locked(pc));
VM_BUG_ON(!PageCgroupUsed(pc));
VM_BUG_ON(pc->mem_cgroup != from);
unsigned long precharge;
struct vm_area_struct *vma;
- down_read(&mm->mmap_sem);
+ /* We've already held the mmap_sem */
for (vma = mm->mmap; vma; vma = vma->vm_next) {
struct mm_walk mem_cgroup_count_precharge_walk = {
.pmd_entry = mem_cgroup_count_precharge_pte_range,
walk_page_range(vma->vm_start, vma->vm_end,
&mem_cgroup_count_precharge_walk);
}
- up_read(&mm->mmap_sem);
precharge = mc.precharge;
mc.precharge = 0;
mc.moved_swap = 0;
}
+ if (mc.mm) {
+ up_read(&mc.mm->mmap_sem);
+ mmput(mc.mm);
+ }
spin_lock(&mc.lock);
mc.from = NULL;
mc.to = NULL;
- mc.moving_task = NULL;
spin_unlock(&mc.lock);
+ mc.moving_task = NULL;
+ mc.mm = NULL;
mem_cgroup_end_move(from);
memcg_oom_recover(from);
memcg_oom_recover(to);
return 0;
/* We move charges only when we move a owner of the mm */
if (mm->owner == p) {
+ /*
+ * We do all the move charge works under one mmap_sem to
+ * avoid deadlock with down_write(&mmap_sem)
+ * -> try_charge() -> if (mc.moving_task) -> sleep.
+ */
+ down_read(&mm->mmap_sem);
+
VM_BUG_ON(mc.from);
VM_BUG_ON(mc.to);
VM_BUG_ON(mc.precharge);
VM_BUG_ON(mc.moved_charge);
VM_BUG_ON(mc.moved_swap);
VM_BUG_ON(mc.moving_task);
+ VM_BUG_ON(mc.mm);
+
mem_cgroup_start_move(from);
spin_lock(&mc.lock);
mc.from = from;
mc.precharge = 0;
mc.moved_charge = 0;
mc.moved_swap = 0;
- mc.moving_task = current;
spin_unlock(&mc.lock);
+ mc.moving_task = current;
+ mc.mm = mm;
ret = mem_cgroup_precharge_mc(mm);
if (ret)
mem_cgroup_clear_mc();
- }
- mmput(mm);
+ /* We call up_read() and mmput() in clear_mc(). */
+ } else
+ mmput(mm);
}
return ret;
}
struct vm_area_struct *vma;
lru_add_drain_all();
- down_read(&mm->mmap_sem);
+ /* We've already held the mmap_sem */
for (vma = mm->mmap; vma; vma = vma->vm_next) {
int ret;
struct mm_walk mem_cgroup_move_charge_walk = {
*/
break;
}
- up_read(&mm->mmap_sem);
}
static void mem_cgroup_move_task(struct cgroup_subsys *ss,
struct task_struct *p,
bool threadgroup)
{
- struct mm_struct *mm;
-
- if (!mc.to)
+ if (!mc.mm)
/* no need to move charge */
return;
- mm = get_task_mm(p);
- if (mm) {
- mem_cgroup_move_charge(mm);
- mmput(mm);
- }
+ mem_cgroup_move_charge(mc.mm);
mem_cgroup_clear_mc();
}
#else /* !CONFIG_MMU */
};
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+static int __init enable_swap_account(char *s)
+{
+ /* consider enabled if no parameter or 1 is given */
+ if (!s || !strcmp(s, "1"))
+ really_do_swap_account = 1;
+ else if (!strcmp(s, "0"))
+ really_do_swap_account = 0;
+ return 1;
+}
+__setup("swapaccount", enable_swap_account);
static int __init disable_swap_account(char *s)
{
- really_do_swap_account = 0;
+ enable_swap_account("0");
return 1;
}
__setup("noswapaccount", disable_swap_account);
mm->mmap = vma->vm_next;
delete_vma_from_mm(vma);
delete_vma(mm, vma);
+ cond_resched();
}
kleave("");
build_zonelist_cache(pgdat);
}
-#ifdef CONFIG_MEMORY_HOTPLUG
- /* Setup real pagesets for the new zone */
- if (data) {
- struct zone *zone = data;
- setup_zone_pageset(zone);
- }
-#endif
-
/*
* Initialize the boot_pagesets that are going to be used
* for bootstrapping processors. The real pagesets for
} else {
/* we have to stop all cpus to guarantee there is no user
of zonelist */
- stop_machine(__build_all_zonelists, data, NULL);
+#ifdef CONFIG_MEMORY_HOTPLUG
+ if (data)
+ setup_zone_pageset((struct zone *)data);
+#endif
+ stop_machine(__build_all_zonelists, NULL, NULL);
/* cpuset refresh routine should be here */
}
vm_total_pages = nr_free_pagecache_pages();
pgd_t *pgd;
unsigned long next;
int err = 0;
- struct vm_area_struct *vma;
if (addr >= end)
return err;
pgd = pgd_offset(walk->mm, addr);
do {
+ struct vm_area_struct *uninitialized_var(vma);
+
next = pgd_addr_end(addr, end);
+#ifdef CONFIG_HUGETLB_PAGE
/*
* handle hugetlb vma individually because pagetable walk for
* the hugetlb page is dependent on the architecture and
* we can't handled it in the same manner as non-huge pages.
*/
vma = find_vma(walk->mm, addr);
-#ifdef CONFIG_HUGETLB_PAGE
if (vma && is_vm_hugetlb_page(vma)) {
if (vma->vm_end < next)
next = vma->vm_end;
#
# Makefile for CEPH filesystem.
#
-
-ifneq ($(KERNELRELEASE),)
-
obj-$(CONFIG_CEPH_LIB) += libceph.o
libceph-objs := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \
ceph_fs.o ceph_strings.o ceph_hash.o \
pagevec.o
-else
-#Otherwise we were called directly from the command
-# line; invoke the kernel build system.
-
-KERNELDIR ?= /lib/modules/$(shell uname -r)/build
-PWD := $(shell pwd)
-
-default: all
-
-all:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules
-
-modules_install:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules_install
-
-clean:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) clean
-
-endif
if (b->vec.iov_base) {
b->is_vmalloc = false;
} else {
- b->vec.iov_base = __vmalloc(len, gfp, PAGE_KERNEL);
+ b->vec.iov_base = __vmalloc(len, gfp | __GFP_HIGHMEM, PAGE_KERNEL);
if (!b->vec.iov_base) {
kfree(b);
return NULL;
nr_table_entries = roundup_pow_of_two(nr_table_entries + 1);
lopt_size += nr_table_entries * sizeof(struct request_sock *);
if (lopt_size > PAGE_SIZE)
- lopt = __vmalloc(lopt_size,
- GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
- PAGE_KERNEL);
+ lopt = vzalloc(lopt_size);
else
lopt = kzalloc(lopt_size, GFP_KERNEL);
if (lopt == NULL)
dccp_update_gsr(sk, seqno);
if (dh->dccph_type != DCCP_PKT_SYNC &&
- (ackno != DCCP_PKT_WITHOUT_ACK_SEQ))
+ ackno != DCCP_PKT_WITHOUT_ACK_SEQ &&
+ after48(ackno, dp->dccps_gar))
dp->dccps_gar = ackno;
} else {
unsigned long now = jiffies;
if (r_len > sizeof(struct linkinfo_dn))
r_len = sizeof(struct linkinfo_dn);
+ memset(&link, 0, sizeof(link));
+
switch(sock->state) {
case SS_CONNECTING:
link.idn_linkstate = LL_CONNECTING;
#include <linux/skbuff.h>
#include <linux/udp.h>
#include <linux/slab.h>
+#include <linux/vmalloc.h>
#include <net/sock.h>
#include <net/inet_common.h>
#include <linux/stat.h>
#endif
#ifdef CONFIG_ECONET_AUNUDP
struct msghdr udpmsg;
- struct iovec iov[msg->msg_iovlen+1];
+ struct iovec iov[2];
struct aunhdr ah;
struct sockaddr_in udpdest;
__kernel_size_t size;
- int i;
mm_segment_t oldfs;
+ char *userbuf;
#endif
/*
mutex_lock(&econet_mutex);
- if (saddr == NULL) {
- struct econet_sock *eo = ec_sk(sk);
-
- addr.station = eo->station;
- addr.net = eo->net;
- port = eo->port;
- cb = eo->cb;
- } else {
- if (msg->msg_namelen < sizeof(struct sockaddr_ec)) {
- mutex_unlock(&econet_mutex);
- return -EINVAL;
- }
- addr.station = saddr->addr.station;
- addr.net = saddr->addr.net;
- port = saddr->port;
- cb = saddr->cb;
- }
+ if (saddr == NULL || msg->msg_namelen < sizeof(struct sockaddr_ec)) {
+ mutex_unlock(&econet_mutex);
+ return -EINVAL;
+ }
+ addr.station = saddr->addr.station;
+ addr.net = saddr->addr.net;
+ port = saddr->port;
+ cb = saddr->cb;
/* Look for a device with the right network number. */
dev = net2dev_map[addr.net];
}
}
- if (len + 15 > dev->mtu) {
- mutex_unlock(&econet_mutex);
- return -EMSGSIZE;
- }
-
if (dev->type == ARPHRD_ECONET) {
/* Real hardware Econet. We're not worthy etc. */
#ifdef CONFIG_ECONET_NATIVE
unsigned short proto = 0;
int res;
+ if (len + 15 > dev->mtu) {
+ mutex_unlock(&econet_mutex);
+ return -EMSGSIZE;
+ }
+
dev_hold(dev);
skb = sock_alloc_send_skb(sk, len+LL_ALLOCATED_SPACE(dev),
eb = (struct ec_cb *)&skb->cb;
- /* BUG: saddr may be NULL */
eb->cookie = saddr->cookie;
eb->sec = *saddr;
eb->sent = ec_tx_done;
return -ENETDOWN; /* No socket - can't send */
}
+ if (len > 32768) {
+ err = -E2BIG;
+ goto error;
+ }
+
/* Make up a UDP datagram and hand it off to some higher intellect. */
memset(&udpdest, 0, sizeof(udpdest));
/* tack our header on the front of the iovec */
size = sizeof(struct aunhdr);
- /*
- * XXX: that is b0rken. We can't mix userland and kernel pointers
- * in iovec, since on a lot of platforms copy_from_user() will
- * *not* work with the kernel and userland ones at the same time,
- * regardless of what we do with set_fs(). And we are talking about
- * econet-over-ethernet here, so "it's only ARM anyway" doesn't
- * apply. Any suggestions on fixing that code? -- AV
- */
iov[0].iov_base = (void *)&ah;
iov[0].iov_len = size;
- for (i = 0; i < msg->msg_iovlen; i++) {
- void __user *base = msg->msg_iov[i].iov_base;
- size_t iov_len = msg->msg_iov[i].iov_len;
- /* Check it now since we switch to KERNEL_DS later. */
- if (!access_ok(VERIFY_READ, base, iov_len)) {
- mutex_unlock(&econet_mutex);
- return -EFAULT;
- }
- iov[i+1].iov_base = base;
- iov[i+1].iov_len = iov_len;
- size += iov_len;
+
+ userbuf = vmalloc(len);
+ if (userbuf == NULL) {
+ err = -ENOMEM;
+ goto error;
}
+ iov[1].iov_base = userbuf;
+ iov[1].iov_len = len;
+ err = memcpy_fromiovec(userbuf, msg->msg_iov, len);
+ if (err)
+ goto error_free_buf;
+
/* Get a skbuff (no data, just holds our cb information) */
if ((skb = sock_alloc_send_skb(sk, 0,
msg->msg_flags & MSG_DONTWAIT,
- &err)) == NULL) {
- mutex_unlock(&econet_mutex);
- return err;
- }
+ &err)) == NULL)
+ goto error_free_buf;
eb = (struct ec_cb *)&skb->cb;
udpmsg.msg_name = (void *)&udpdest;
udpmsg.msg_namelen = sizeof(udpdest);
udpmsg.msg_iov = &iov[0];
- udpmsg.msg_iovlen = msg->msg_iovlen + 1;
+ udpmsg.msg_iovlen = 2;
udpmsg.msg_control = NULL;
udpmsg.msg_controllen = 0;
udpmsg.msg_flags=0;
oldfs = get_fs(); set_fs(KERNEL_DS); /* More privs :-) */
err = sock_sendmsg(udpsock, &udpmsg, size);
set_fs(oldfs);
+
+error_free_buf:
+ vfree(userbuf);
#else
err = -EPROTOTYPE;
#endif
+ error:
mutex_unlock(&econet_mutex);
return err;
err = 0;
switch (cmd) {
case SIOCSIFADDR:
+ if (!capable(CAP_NET_ADMIN))
+ return -EPERM;
+
edev = dev->ec_ptr;
if (edev == NULL) {
/* Magic up a new one. */
if (size <= PAGE_SIZE)
return kzalloc(size, GFP_KERNEL);
else
- return __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+ return vzalloc(size);
}
static void __tnode_vfree(struct work_struct *arg)
}
}
}
- sk_add_bind_node(child, &tb->owners);
- inet_csk(child)->icsk_bind_hash = tb;
+ inet_bind_hash(child, tb, port);
spin_unlock(&head->lock);
return 0;
static int tcp_retr1_max = 255;
static int ip_local_port_range_min[] = { 1, 1 };
static int ip_local_port_range_max[] = { 65535, 65535 };
+static int tcp_adv_win_scale_min = -31;
+static int tcp_adv_win_scale_max = 31;
/* Update system visible IP port range */
static void set_local_port_range(int range[2])
.data = &sysctl_tcp_adv_win_scale,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = proc_dointvec
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &tcp_adv_win_scale_min,
+ .extra2 = &tcp_adv_win_scale_max,
},
{
.procname = "tcp_tw_reuse",
/* Values greater than interface MTU won't take effect. However
* at the point when this call is done we typically don't yet
* know which interface is going to be used */
- if (val < 64 || val > MAX_TCP_WINDOW) {
+ if (val < TCP_MIN_MSS || val > MAX_TCP_WINDOW) {
err = -EINVAL;
break;
}
}
get_sk:
sk_nulls_for_each_from(sk, node) {
- if (sk->sk_family == st->family && net_eq(sock_net(sk), net)) {
+ if (!net_eq(sock_net(sk), net))
+ continue;
+ if (sk->sk_family == st->family) {
cur = sk;
goto out;
}
ifa->state = INET6_IFADDR_STATE_DEAD;
spin_unlock_bh(&ifa->state_lock);
- if (state == INET6_IFADDR_STATE_DEAD) {
- in6_ifa_put(ifa);
- } else {
+ if (state != INET6_IFADDR_STATE_DEAD) {
__ipv6_ifa_notify(RTM_DELADDR, ifa);
atomic_notifier_call_chain(&inet6addr_chain,
NETDEV_DOWN, ifa);
}
+
+ in6_ifa_put(ifa);
write_lock_bh(&idev->lock);
}
}
dprint_status(task);
task->tk_status = 0;
- task->tk_action = call_allocate;
- if (status >= 0 && rpcauth_uptodatecred(task))
- return;
+ task->tk_action = call_refresh;
switch (status) {
- case -EACCES:
- rpc_exit(task, -EACCES);
- return;
- case -ENOMEM:
- rpc_exit(task, -ENOMEM);
+ case 0:
+ if (rpcauth_uptodatecred(task))
+ task->tk_action = call_allocate;
return;
case -ETIMEDOUT:
rpc_delay(task, 3*HZ);
+ case -EAGAIN:
+ status = -EACCES;
+ if (!task->tk_cred_retry)
+ break;
+ task->tk_cred_retry--;
+ dprintk("RPC: %5u %s: retry refresh creds\n",
+ task->tk_pid, __func__);
+ return;
}
- task->tk_action = call_refresh;
+ dprintk("RPC: %5u %s: refresh creds failed with error %d\n",
+ task->tk_pid, __func__, status);
+ rpc_exit(task, status);
}
/*
sock_wfree(skb);
}
+#define MAX_RECURSION_LEVEL 4
+
static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
{
int i;
+ unsigned char max_level = 0;
+ int unix_sock_count = 0;
+
+ for (i = scm->fp->count - 1; i >= 0; i--) {
+ struct sock *sk = unix_get_socket(scm->fp->fp[i]);
+
+ if (sk) {
+ unix_sock_count++;
+ max_level = max(max_level,
+ unix_sk(sk)->recursion_level);
+ }
+ }
+ if (unlikely(max_level > MAX_RECURSION_LEVEL))
+ return -ETOOMANYREFS;
/*
* Need to duplicate file references for the sake of garbage
if (!UNIXCB(skb).fp)
return -ENOMEM;
- for (i = scm->fp->count-1; i >= 0; i--)
- unix_inflight(scm->fp->fp[i]);
- return 0;
+ if (unix_sock_count) {
+ for (i = scm->fp->count - 1; i >= 0; i--)
+ unix_inflight(scm->fp->fp[i]);
+ }
+ return max_level;
}
static int unix_scm_to_skb(struct scm_cookie *scm, struct sk_buff *skb, bool send_fds)
struct sk_buff *skb;
long timeo;
struct scm_cookie tmp_scm;
+ int max_level;
if (NULL == siocb->scm)
siocb->scm = &tmp_scm;
goto out;
err = unix_scm_to_skb(siocb->scm, skb, true);
- if (err)
+ if (err < 0)
goto out_free;
+ max_level = err + 1;
unix_get_secdata(siocb->scm, skb);
skb_reset_transport_header(skb);
if (sock_flag(other, SOCK_RCVTSTAMP))
__net_timestamp(skb);
skb_queue_tail(&other->sk_receive_queue, skb);
+ if (max_level > unix_sk(other)->recursion_level)
+ unix_sk(other)->recursion_level = max_level;
unix_state_unlock(other);
other->sk_data_ready(other, len);
sock_put(other);
int sent = 0;
struct scm_cookie tmp_scm;
bool fds_sent = false;
+ int max_level;
if (NULL == siocb->scm)
siocb->scm = &tmp_scm;
/* Only send the fds in the first buffer */
err = unix_scm_to_skb(siocb->scm, skb, !fds_sent);
- if (err) {
+ if (err < 0) {
kfree_skb(skb);
goto out_err;
}
+ max_level = err + 1;
fds_sent = true;
err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size);
goto pipe_err_free;
skb_queue_tail(&other->sk_receive_queue, skb);
+ if (max_level > unix_sk(other)->recursion_level)
+ unix_sk(other)->recursion_level = max_level;
unix_state_unlock(other);
other->sk_data_ready(other, size);
sent += size;
unix_state_lock(sk);
skb = skb_dequeue(&sk->sk_receive_queue);
if (skb == NULL) {
+ unix_sk(sk)->recursion_level = 0;
if (copied >= target)
goto unlock;
unsigned int unix_tot_inflight;
-static struct sock *unix_get_socket(struct file *filp)
+struct sock *unix_get_socket(struct file *filp)
{
struct sock *u_sock = NULL;
struct inode *inode = filp->f_path.dentry->d_inode;
}
static bool gc_in_progress = false;
+#define UNIX_INFLIGHT_TRIGGER_GC 16000
void wait_for_unix_gc(void)
{
+ /*
+ * If number of inflight sockets is insane,
+ * force a garbage collect right now.
+ */
+ if (unix_tot_inflight > UNIX_INFLIGHT_TRIGGER_GC && !gc_in_progress)
+ unix_gc();
wait_event(unix_gc_wait, gc_in_progress == false);
}
if (sz <= PAGE_SIZE)
n = kzalloc(sz, GFP_KERNEL);
else if (hashdist)
- n = __vmalloc(sz, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL);
+ n = vzalloc(sz);
else
n = (struct hlist_head *)
__get_free_pages(GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO,
# Extract GFP flags from the kernel source
TMPFILE=`mktemp -t gfptranslate-XXXXXX` || exit 1
-grep "^#define __GFP" $SOURCE/include/linux/gfp.h | sed -e 's/(__force gfp_t)//' | sed -e 's/u)/)/' | grep -v GFP_BITS | sed -e 's/)\//) \//' > $TMPFILE
+grep -q ___GFP $SOURCE/include/linux/gfp.h
+if [ $? -eq 0 ]; then
+ grep "^#define ___GFP" $SOURCE/include/linux/gfp.h | sed -e 's/u$//' | grep -v GFP_BITS > $TMPFILE
+else
+ grep "^#define __GFP" $SOURCE/include/linux/gfp.h | sed -e 's/(__force gfp_t)//' | sed -e 's/u)/)/' | grep -v GFP_BITS | sed -e 's/)\//) \//' > $TMPFILE
+fi
# Parse the flags
IFS="
return PTR_ERR(pclk);
}
sample_clk = clk_get(&pdev->dev, "sample_clk");
- if (IS_ERR(pclk)) {
+ if (IS_ERR(sample_clk)) {
dev_dbg(&pdev->dev, "no sample clock\n");
- retval = PTR_ERR(pclk);
+ retval = PTR_ERR(sample_clk);
goto out_put_pclk;
}
clk_enable(pclk);
entry->jiffies = jiffies;
entry->pos = pos;
entry->period_size = runtime->period_size;
- entry->buffer_size = runtime->buffer_size;;
+ entry->buffer_size = runtime->buffer_size;
entry->old_hw_ptr = runtime->status->hw_ptr;
entry->hw_ptr_base = runtime->hw_ptr_base;
log->idx = (log->idx + 1) % XRUN_LOG_CNT;
if (sound_nblocks >= MAX_MEM_BLOCKS)
sound_nblocks = MAX_MEM_BLOCKS - 1;
- op = (struct audio_operations *) (sound_mem_blocks[sound_nblocks] = vmalloc(sizeof(struct audio_operations)));
+ op = (struct audio_operations *) (sound_mem_blocks[sound_nblocks] = vzalloc(sizeof(struct audio_operations)));
sound_nblocks++;
if (sound_nblocks >= MAX_MEM_BLOCKS)
sound_nblocks = MAX_MEM_BLOCKS - 1;
sound_unload_audiodev(num);
return -(ENOMEM);
}
- memset((char *) op, 0, sizeof(struct audio_operations));
init_waitqueue_head(&op->in_sleeper);
init_waitqueue_head(&op->out_sleeper);
init_waitqueue_head(&op->poll_sleeper);
/* FIXME: This leaks a mixer_operations struct every time its called
until you unload sound! */
- op = (struct mixer_operations *) (sound_mem_blocks[sound_nblocks] = vmalloc(sizeof(struct mixer_operations)));
+ op = (struct mixer_operations *) (sound_mem_blocks[sound_nblocks] = vzalloc(sizeof(struct mixer_operations)));
sound_nblocks++;
if (sound_nblocks >= MAX_MEM_BLOCKS)
sound_nblocks = MAX_MEM_BLOCKS - 1;
printk(KERN_ERR "Sound: Can't allocate mixer driver for (%s)\n", name);
return -ENOMEM;
}
- memset((char *) op, 0, sizeof(struct mixer_operations));
memcpy((char *) op, (char *) driver, driver_size);
strlcpy(op->name, name, sizeof(op->name));
return err;
parms[dev].prech_timeout = MAX_SCHEDULE_TIMEOUT;
- midi_in_buf[dev] = (struct midi_buf *) vmalloc(sizeof(struct midi_buf));
+ midi_in_buf[dev] = vmalloc(sizeof(struct midi_buf));
if (midi_in_buf[dev] == NULL)
{
}
midi_in_buf[dev]->len = midi_in_buf[dev]->head = midi_in_buf[dev]->tail = 0;
- midi_out_buf[dev] = (struct midi_buf *) vmalloc(sizeof(struct midi_buf));
+ midi_out_buf[dev] = vmalloc(sizeof(struct midi_buf));
if (midi_out_buf[dev] == NULL)
{
return 0;
case SNDCTL_COPR_LOAD:
- buf = (copr_buffer *) vmalloc(sizeof(copr_buffer));
+ buf = vmalloc(sizeof(copr_buffer));
if (buf == NULL)
return -ENOSPC;
if (copy_from_user(buf, arg, sizeof(copr_buffer))) {
return err;
case SNDCTL_COPR_SENDMSG:
- mbuf = (copr_msg *)vmalloc(sizeof(copr_msg));
+ mbuf = vmalloc(sizeof(copr_msg));
if (mbuf == NULL)
return -ENOSPC;
if (copy_from_user(mbuf, arg, sizeof(copr_msg))) {
case SNDCTL_COPR_RCVMSG:
err = 0;
- mbuf = (copr_msg *)vmalloc(sizeof(copr_msg));
+ mbuf = vmalloc(sizeof(copr_msg));
if (mbuf == NULL)
return -ENOSPC;
data = (unsigned short *)mbuf->data;
{
if (sequencer_ok)
return;
- queue = (unsigned char *)vmalloc(SEQ_MAX_QUEUE * EV_SZ);
+ queue = vmalloc(SEQ_MAX_QUEUE * EV_SZ);
if (queue == NULL)
{
printk(KERN_ERR "sequencer: Can't allocate memory for sequencer output queue\n");
return;
}
- iqueue = (unsigned char *)vmalloc(SEQ_MAX_QUEUE * IEV_SZ);
+ iqueue = vmalloc(SEQ_MAX_QUEUE * IEV_SZ);
if (iqueue == NULL)
{
printk(KERN_ERR "sequencer: Can't allocate memory for sequencer input queue\n");
struct hpi_message hm;
struct hpi_response hr;
struct hpi_adapter *pa;
- pa = (struct hpi_adapter *)pci_get_drvdata(pci_dev);
+ pa = pci_get_drvdata(pci_dev);
hpi_init_message_response(&hm, &hr, HPI_OBJ_SUBSYSTEM,
HPI_SUBSYS_DELETE_ADAPTER);
count_areas = size/2;
addr_area2 = addr+count_areas;
- count_areas--; /* max. index */
snd_azf3328_dbgcodec("setdma: buffers %08lx[%u] / %08lx[%u]\n",
addr, count_areas, addr_area2, count_areas);
+ count_areas--; /* max. index */
+
/* build combined I/O buffer length word */
lengths = (count_areas << 16) | (count_areas);
spin_lock_irqsave(&chip->reg_lock, flags);
.rate_max = AZF_FREQ_66200,
.channels_min = 1,
.channels_max = 2,
- .buffer_bytes_max = 65536,
- .period_bytes_min = 64,
- .period_bytes_max = 65536,
- .periods_min = 1,
- .periods_max = 1024,
+ .buffer_bytes_max = (64*1024),
+ .period_bytes_min = 1024,
+ .period_bytes_max = (32*1024),
+ /* We simply have two DMA areas (instead of a list of descriptors
+ such as other cards); I believe that this is a fixed hardware
+ attribute and there isn't much driver magic to be done to expand it.
+ Thus indicate that we have at least and at most 2 periods. */
+ .periods_min = 2,
+ .periods_max = 2,
/* FIXME: maybe that card actually has a FIFO?
* Hmm, it seems newer revisions do have one, but we still don't know
* its size... */
chip = snd_timer_chip(timer);
spin_lock_irqsave(&chip->reg_lock, flags);
/* disable timer countdown and interrupt */
- /* FIXME: should we write TIMER_IRQ_ACK here? */
- snd_azf3328_ctrl_outb(chip, IDX_IO_TIMER_VALUE + 3, 0);
+ /* Hmm, should we write TIMER_IRQ_ACK here?
+ YES indeed, otherwise a rogue timer operation - which prompts
+ ALSA(?) to call repeated stop() in vain, but NOT start() -
+ will never end (value 0x03 is kept shown in control byte).
+ Simply manually poking 0x04 _once_ immediately successfully stops
+ the hardware/ALSA interrupt activity. */
+ snd_azf3328_ctrl_outb(chip, IDX_IO_TIMER_VALUE + 3, 0x04);
spin_unlock_irqrestore(&chip->reg_lock, flags);
snd_azf3328_dbgcallleave();
return 0;
apcm->substream = substream;
apcm->interrupt = ct_atc_pcm_interrupt;
- runtime->private_data = apcm;
- runtime->private_free = ct_atc_pcm_free_substream;
if (IEC958 == substream->pcm->device) {
runtime->hw = ct_spdif_passthru_playback_hw;
atc->spdif_out_passthru(atc, 1);
}
apcm->timer = ct_timer_instance_new(atc->timer, apcm);
- if (!apcm->timer)
+ if (!apcm->timer) {
+ kfree(apcm);
return -ENOMEM;
+ }
+ runtime->private_data = apcm;
+ runtime->private_free = ct_atc_pcm_free_substream;
return 0;
}
apcm->started = 0;
apcm->substream = substream;
apcm->interrupt = ct_atc_pcm_interrupt;
- runtime->private_data = apcm;
- runtime->private_free = ct_atc_pcm_free_substream;
runtime->hw = ct_pcm_capture_hw;
runtime->hw.rate_max = atc->rsr * atc->msr;
}
apcm->timer = ct_timer_instance_new(atc->timer, apcm);
- if (!apcm->timer)
+ if (!apcm->timer) {
+ kfree(apcm);
return -ENOMEM;
+ }
+ runtime->private_data = apcm;
+ runtime->private_free = ct_atc_pcm_free_substream;
return 0;
}
SND_PCI_QUIRK(0x1028, 0x0402, "Dell Vostro", CXT5066_DELL_VOSTRO),
SND_PCI_QUIRK(0x1028, 0x0408, "Dell Inspiron One 19T", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x103c, 0x360b, "HP G60", CXT5066_HP_LAPTOP),
+ SND_PCI_QUIRK(0x1043, 0x13f3, "Asus A52J", CXT5066_HP_LAPTOP),
SND_PCI_QUIRK(0x1179, 0xff1e, "Toshiba Satellite C650D", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x1179, 0xff50, "Toshiba Satellite P500-PSPGSC-01800T", CXT5066_OLPC_XO_1_5),
SND_PCI_QUIRK(0x1179, 0xffe0, "Toshiba Satellite Pro T130-15F", CXT5066_OLPC_XO_1_5),
SND_PCI_QUIRK(0x17aa, 0x21b2, "Thinkpad X100e", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x21b3, "Thinkpad Edge 13 (197)", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x21b4, "Thinkpad Edge", CXT5066_IDEAPAD),
+ SND_PCI_QUIRK(0x17aa, 0x21c8, "Thinkpad Edge 11", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x215e, "Lenovo Thinkpad", CXT5066_THINKPAD),
SND_PCI_QUIRK(0x17aa, 0x38af, "Lenovo G series", CXT5066_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x390a, "Lenovo S10-3t", CXT5066_IDEAPAD),
spec->init_amp = ALC_INIT_GPIO3;
break;
case 5:
+ default:
spec->init_amp = ALC_INIT_DEFAULT;
break;
}
{ }
};
+/*
+ *ALC888 Acer Aspire 7730G model
+ */
+
+static struct hda_verb alc888_acer_aspire_7730G_verbs[] = {
+/* Bias voltage on for external mic port */
+ {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN | PIN_VREF80},
+/* Front Mic: set to PIN_IN (empty by default) */
+ {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN},
+/* Unselect Front Mic by default in input mixer 3 */
+ {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0xb)},
+/* Enable unsolicited event for HP jack */
+ {0x15, AC_VERB_SET_UNSOLICITED_ENABLE, ALC880_HP_EVENT | AC_USRSP_EN},
+/* Enable speaker output */
+ {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT},
+ {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE},
+ {0x14, AC_VERB_SET_EAPD_BTLENABLE, 2},
+/* Enable headphone output */
+ {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT | PIN_HP},
+ {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE},
+ {0x15, AC_VERB_SET_CONNECT_SEL, 0x00},
+ {0x15, AC_VERB_SET_EAPD_BTLENABLE, 2},
+/*Enable internal subwoofer */
+ {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT},
+ {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE},
+ {0x17, AC_VERB_SET_CONNECT_SEL, 0x02},
+ {0x17, AC_VERB_SET_EAPD_BTLENABLE, 2},
+ { }
+};
+
/*
* ALC889 Acer Aspire 8930G model
*/
spec->autocfg.speaker_pins[2] = 0x17;
}
+static void alc888_acer_aspire_7730g_setup(struct hda_codec *codec)
+{
+ struct alc_spec *spec = codec->spec;
+
+ spec->autocfg.hp_pins[0] = 0x15;
+ spec->autocfg.speaker_pins[0] = 0x14;
+ spec->autocfg.speaker_pins[1] = 0x16;
+ spec->autocfg.speaker_pins[2] = 0x17;
+}
+
static void alc889_acer_aspire_8930g_setup(struct hda_codec *codec)
{
struct alc_spec *spec = codec->spec;
{ }
};
-static struct hda_verb alc888_acer_aspire_7730G_verbs[] = {
- {0x15, AC_VERB_SET_CONNECT_SEL, 0x00},
- {0x17, AC_VERB_SET_CONNECT_SEL, 0x02},
- {0x15, AC_VERB_SET_UNSOLICITED_ENABLE, ALC880_HP_EVENT | AC_USRSP_EN},
- { } /* end */
-};
-
static void alc888_6st_dell_setup(struct hda_codec *codec)
{
struct alc_spec *spec = codec->spec;
.const_channel_count = 6,
.input_mux = &alc883_capture_source,
.unsol_event = alc_automute_amp_unsol_event,
- .setup = alc888_acer_aspire_6530g_setup,
+ .setup = alc888_acer_aspire_7730g_setup,
.init_hook = alc_automute_amp,
},
[ALC883_MEDION] = {
/* different alc269-variants */
enum {
ALC269_TYPE_NORMAL,
+ ALC269_TYPE_ALC258,
ALC269_TYPE_ALC259,
+ ALC269_TYPE_ALC269VB,
+ ALC269_TYPE_ALC270,
ALC269_TYPE_ALC271X,
};
static int patch_alc269(struct hda_codec *codec)
{
struct alc_spec *spec;
- int board_config;
+ int board_config, coef;
int err;
spec = kzalloc(sizeof(*spec), GFP_KERNEL);
alc_auto_parse_customize_define(codec);
- if ((alc_read_coef_idx(codec, 0) & 0x00f0) == 0x0010){
+ coef = alc_read_coef_idx(codec, 0);
+ if ((coef & 0x00f0) == 0x0010) {
if (codec->bus->pci->subsystem_vendor == 0x1025 &&
spec->cdefine.platform_type == 1) {
alc_codec_rename(codec, "ALC271X");
spec->codec_variant = ALC269_TYPE_ALC271X;
- } else {
+ } else if ((coef & 0xf000) == 0x1000) {
+ spec->codec_variant = ALC269_TYPE_ALC270;
+ } else if ((coef & 0xf000) == 0x2000) {
alc_codec_rename(codec, "ALC259");
spec->codec_variant = ALC269_TYPE_ALC259;
+ } else if ((coef & 0xf000) == 0x3000) {
+ alc_codec_rename(codec, "ALC258");
+ spec->codec_variant = ALC269_TYPE_ALC258;
+ } else {
+ alc_codec_rename(codec, "ALC269VB");
+ spec->codec_variant = ALC269_TYPE_ALC269VB;
}
} else
alc_fix_pll_init(codec, 0x20, 0x04, 15);
spec->stream_digital_capture = &alc269_pcm_digital_capture;
if (!spec->adc_nids) { /* wasn't filled automatically? use default */
- if (spec->codec_variant != ALC269_TYPE_NORMAL) {
+ if (spec->codec_variant == ALC269_TYPE_NORMAL) {
spec->adc_nids = alc269_adc_nids;
spec->num_adc_nids = ARRAY_SIZE(alc269_adc_nids);
spec->capsrc_nids = alc269_capsrc_nids;
static int alc861vd_auto_create_input_ctls(struct hda_codec *codec,
const struct auto_pin_cfg *cfg)
{
- return alc_auto_create_input_ctls(codec, cfg, 0x15, 0x09, 0);
+ return alc_auto_create_input_ctls(codec, cfg, 0x0b, 0x22, 0);
}
return 0x02;
else if (nid >= 0x0c && nid <= 0x0e)
return nid - 0x0c + 0x02;
+ else if (nid == 0x26) /* ALC887-VD has this DAC too */
+ return 0x25;
else
return 0;
}
static hda_nid_t alc662_dac_to_mix(struct hda_codec *codec, hda_nid_t pin,
hda_nid_t dac)
{
- hda_nid_t mix[4];
+ hda_nid_t mix[5];
int i, num;
num = snd_hda_get_connections(codec, pin, mix, ARRAY_SIZE(mix));
static struct snd_pci_quirk alc662_fixup_tbl[] = {
SND_PCI_QUIRK(0x1025, 0x038b, "Acer Aspire 8943G", ALC662_FIXUP_ASPIRE),
+ SND_PCI_QUIRK(0x144d, 0xc051, "Samsung R720", ALC662_FIXUP_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x38af, "Lenovo Ideapad Y550P", ALC662_FIXUP_IDEAPAD),
SND_PCI_QUIRK(0x17aa, 0x3a0d, "Lenovo Ideapad Y550", ALC662_FIXUP_IDEAPAD),
{}
{
if ((alc_read_coef_idx(codec, 0) & 0x00f0)==0x0030){
kfree(codec->chip_name);
- codec->chip_name = kstrdup("ALC888-VD", GFP_KERNEL);
+ if (codec->vendor_id == 0x10ec0887)
+ codec->chip_name = kstrdup("ALC887-VD", GFP_KERNEL);
+ else
+ codec->chip_name = kstrdup("ALC888-VD", GFP_KERNEL);
if (!codec->chip_name) {
alc_free(codec);
return -ENOMEM;
{ .id = 0x10ec0885, .rev = 0x100103, .name = "ALC889A",
.patch = patch_alc882 },
{ .id = 0x10ec0885, .name = "ALC885", .patch = patch_alc882 },
- { .id = 0x10ec0887, .name = "ALC887", .patch = patch_alc882 },
+ { .id = 0x10ec0887, .name = "ALC887", .patch = patch_alc888 },
{ .id = 0x10ec0888, .rev = 0x100101, .name = "ALC1200",
.patch = patch_alc882 },
{ .id = 0x10ec0888, .name = "ALC888", .patch = patch_alc888 },
0x11, 0x20, 0
};
+#define STAC92HD87B_NUM_DMICS 1
+static hda_nid_t stac92hd87b_dmic_nids[STAC92HD87B_NUM_DMICS + 1] = {
+ 0x11, 0
+};
+
#define STAC92HD83XXX_NUM_CAPS 2
static unsigned long stac92hd83xxx_capvols[] = {
HDA_COMPOSE_AMP_VAL(0x17, 3, 0, HDA_OUTPUT),
return err;
}
- if (snd_hda_get_bool_hint(codec, "separate_dmux") != 1) {
+ if (snd_hda_get_bool_hint(codec, "separate_dmux") != 1)
snd_hda_add_imux_item(imux, label, index, NULL);
- spec->num_analog_muxes++;
- }
}
return 0;
stac92hd83xxx_brd_tbl[spec->board_config]);
switch (codec->vendor_id) {
+ case 0x111d76d1:
+ case 0x111d76d9:
+ spec->dmic_nids = stac92hd87b_dmic_nids;
+ spec->num_dmics = stac92xx_connected_ports(codec,
+ stac92hd87b_dmic_nids,
+ STAC92HD87B_NUM_DMICS);
+ /* Fall through */
case 0x111d7666:
case 0x111d7667:
case 0x111d7668:
case 0x111d7669:
- case 0x111d76d1:
- case 0x111d76d9:
spec->num_pins = ARRAY_SIZE(stac92hd88xxx_pin_nids);
spec->pin_nids = stac92hd88xxx_pin_nids;
spec->mono_nid = 0;
.name = "Dell Inspiron 8600", /* STAC9750/51 */
.type = AC97_TUNE_HP_ONLY
},
+ {
+ .subvendor = 0x1028,
+ .subdevice = 0x0182,
+ .name = "Dell Latitude D610", /* STAC9750/51 */
+ .type = AC97_TUNE_HP_ONLY
+ },
{
.subvendor = 0x1028,
.subdevice = 0x0186,
#include <sound/hwdep.h>
+#ifndef readl_be
#define readl_be(x) be32_to_cpu(__raw_readl(x))
+#endif
+
+#ifndef writel_be
#define writel_be(data,addr) __raw_writel(cpu_to_be32(data),addr)
+#endif
+#ifndef readl_le
#define readl_le(x) le32_to_cpu(__raw_readl(x))
+#endif
+
+#ifndef writel_le
#define writel_le(data,addr) __raw_writel(cpu_to_le32(data),addr)
+#endif
#define MIXART_MEM(mgr,x) ((mgr)->mem[0].virt + (x))
#define MIXART_REG(mgr,x) ((mgr)->mem[1].virt + (x))
chip->rsrc[i].start + 1,
rnames[i]) == NULL) {
printk(KERN_ERR "snd: can't request rsrc "
- " %d (%s: 0x%016llx:%016llx)\n",
- i, rnames[i],
- (unsigned long long)chip->rsrc[i].start,
- (unsigned long long)chip->rsrc[i].end);
+ " %d (%s: %pR)\n",
+ i, rnames[i], &chip->rsrc[i]);
err = -ENODEV;
goto __error;
}
chip->rsrc[i].start + 1,
rnames[i]) == NULL) {
printk(KERN_ERR "snd: can't request rsrc "
- " %d (%s: 0x%016llx:%016llx)\n",
- i, rnames[i],
- (unsigned long long)chip->rsrc[i].start,
- (unsigned long long)chip->rsrc[i].end);
+ " %d (%s: %pR)\n",
+ i, rnames[i], &chip->rsrc[i]);
err = -ENODEV;
goto __error;
}
config SND_AT91_SOC_SAM9G20_WM8731
tristate "SoC Audio support for WM8731-based At91sam9g20 evaluation board"
- depends on ATMEL_SSC && ARCH_AT91SAM9G20 && SND_ATMEL_SOC
+ depends on ATMEL_SSC && ARCH_AT91SAM9G20 && SND_ATMEL_SOC && \
+ AT91_PROGRAMMABLE_CLOCKS
select SND_ATMEL_SOC_SSC
select SND_SOC_WM8731
help
config SND_AT32_SOC_PLAYPAQ
tristate "SoC Audio support for PlayPaq with WM8510"
- depends on SND_ATMEL_SOC && BOARD_PLAYPAQ
+ depends on SND_ATMEL_SOC && BOARD_PLAYPAQ && AT91_PROGRAMMABLE_CLOCKS
select SND_ATMEL_SOC_SSC
select SND_SOC_WM8510
help
}
pllb = clk_get(NULL, "pllb");
- if (IS_ERR(mclk)) {
+ if (IS_ERR(pllb)) {
printk(KERN_ERR "ASoC: Failed to get PLLB\n");
- ret = PTR_ERR(mclk);
+ ret = PTR_ERR(pllb);
goto err_mclk;
}
ret = clk_set_parent(mclk, pllb);
if (!at91sam9g20ek_snd_device) {
printk(KERN_ERR "ASoC: Platform device allocation failed\n");
ret = -ENOMEM;
+ goto err_mclk;
}
platform_set_drvdata(at91sam9g20ek_snd_device,
ret = platform_device_add(at91sam9g20ek_snd_device);
if (ret) {
printk(KERN_ERR "ASoC: Platform device allocation failed\n");
- platform_device_put(at91sam9g20ek_snd_device);
+ goto err_device_add;
}
return ret;
+err_device_add:
+ platform_device_put(at91sam9g20ek_snd_device);
err_mclk:
clk_put(mclk);
mclk = NULL;
return 0;
err1:
- platform_device_del(afeb9260_snd_device);
platform_device_put(afeb9260_snd_device);
return err;
}
#include <sound/max98088.h>
#include "max98088.h"
+enum max98088_type {
+ MAX98088,
+ MAX98089,
+};
+
struct max98088_cdata {
unsigned int rate;
unsigned int fmt;
struct max98088_priv {
u8 reg_cache[M98088_REG_CNT];
+ enum max98088_type devtype;
void *control_data;
struct max98088_pdata *pdata;
unsigned int sysclk;
static int max98088_remove(struct snd_soc_codec *codec)
{
+ struct max98088_priv *max98088 = snd_soc_codec_get_drvdata(codec);
+
max98088_set_bias_level(codec, SND_SOC_BIAS_OFF);
+ kfree(max98088->eq_texts);
return 0;
}
if (max98088 == NULL)
return -ENOMEM;
+ max98088->devtype = id->driver_data;
+
i2c_set_clientdata(i2c, max98088);
max98088->control_data = i2c;
max98088->pdata = i2c->dev.platform_data;
}
static const struct i2c_device_id max98088_i2c_id[] = {
- { "max98088", 0 },
+ { "max98088", MAX98088 },
+ { "max98089", MAX98089 },
{ }
};
MODULE_DEVICE_TABLE(i2c, max98088_i2c_id);
.reg_cache_size = sizeof(stac9766_reg),
.reg_word_size = sizeof(u16),
.reg_cache_step = 2,
+ .reg_cache_default = stac9766_reg,
};
static __devinit int stac9766_probe(struct platform_device *pdev)
int aic3x_get_gpio(struct snd_soc_codec *codec, int gpio)
{
u8 reg = gpio ? AIC3X_GPIO2_REG : AIC3X_GPIO1_REG;
- u8 val, bit = gpio ? 2: 1;
+ u8 val = 0, bit = gpio ? 2 : 1;
aic3x_read(codec, reg, &val);
return (val >> bit) & 1;
int aic3x_headset_detected(struct snd_soc_codec *codec)
{
- u8 val;
+ u8 val = 0;
aic3x_read(codec, AIC3X_HEADSET_DETECT_CTRL_B, &val);
return (val >> 4) & 1;
}
int aic3x_button_pressed(struct snd_soc_codec *codec)
{
- u8 val;
+ u8 val = 0;
aic3x_read(codec, AIC3X_HEADSET_DETECT_CTRL_B, &val);
return (val >> 5) & 1;
}
if (data->power_state) {
val = i2c_smbus_write_byte_data(tpa6130a2_client, reg, value);
- if (val < 0)
+ if (val < 0) {
dev_err(&tpa6130a2_client->dev, "Write failed\n");
+ return val;
+ }
}
/* Either powered on or off, we save the context */
.resume = uda134x_soc_resume,
.reg_cache_size = sizeof(uda134x_reg),
.reg_word_size = sizeof(u8),
+ .reg_cache_default = uda134x_reg,
.reg_cache_step = 1,
.read = uda134x_read_reg_cache,
.write = uda134x_write,
}
/* MCLK direction */
- if (dir == WM8350_MCLK_DIR_OUT)
+ if (dir == SND_SOC_CLOCK_OUT)
wm8350_set_bits(wm8350, WM8350_CLOCK_CONTROL_2,
WM8350_MCLK_DIR);
else
wm8350_set_bits(wm8350, WM8350_ROUT2_VOLUME,
WM8350_OUT2_VU | WM8350_OUT2R_MUTE);
+ /* Make sure AIF tristating is disabled by default */
+ wm8350_clear_bits(wm8350, WM8350_AI_FORMATING, WM8350_AIF_TRI);
+
+ /* Make sure we've got a sane companding setup too */
+ wm8350_clear_bits(wm8350, WM8350_ADC_DAC_COMP,
+ WM8350_DAC_COMP | WM8350_LOOPBACK);
+
/* Make sure jack detect is disabled to start off with */
wm8350_clear_bits(wm8350, WM8350_JACK_DETECT,
WM8350_JDL_ENA | WM8350_JDR_ENA);
return -EINVAL;
}
- return 0;
snd_pcm_hw_constraint_list(substream->runtime, 0,
SNDRV_PCM_HW_PARAM_RATE,
&wm8523->rate_constraint);
/* codec private data */
struct wm8776_priv {
enum snd_soc_control_type control_type;
- u16 reg_cache[WM8776_CACHEREGNUM];
int sysclk[2];
};
wm8904_set_bias_level(codec, SND_SOC_BIAS_OFF);
regulator_bulk_free(ARRAY_SIZE(wm8904->supplies), wm8904->supplies);
+ kfree(wm8904->retune_mobile_texts);
+ kfree(wm8904->drc_texts);
return 0;
}
if (fs <= 24000)
reg |= WM8961_DACSLOPE;
else
- reg &= WM8961_DACSLOPE;
+ reg &= ~WM8961_DACSLOPE;
snd_soc_write(codec, WM8961_ADC_DAC_CONTROL_2, reg);
return 0;
freq /= 2;
} else {
dev_dbg(codec->dev, "Using MCLK/1 for %dHz MCLK\n", freq);
- reg &= WM8961_MCLKDIV;
+ reg &= ~WM8961_MCLKDIV;
}
snd_soc_write(codec, WM8961_CLOCKING1, reg);
{
struct wm8962_priv *wm8962 = dev_get_drvdata(dev);
long int time;
+ int ret;
- strict_strtol(buf, 10, &time);
+ ret = strict_strtol(buf, 10, &time);
+ if (ret != 0)
+ return ret;
input_event(wm8962->beep, EV_SND, SND_TONE, time);
return -ENOMEM;
snd_soc_codec_set_drvdata(codec, wm8994);
+ codec->reg_cache = &wm8994->reg_cache;
+
wm8994->pdata = dev_get_platdata(codec->dev->parent);
wm8994->codec = codec;
wm8994_free_irq(codec->control_data, WM8994_IRQ_MIC2_DET, wm8994);
wm8994_free_irq(codec->control_data, WM8994_IRQ_MIC1_SHRT, wm8994);
wm8994_free_irq(codec->control_data, WM8994_IRQ_MIC1_DET, wm8994);
+ kfree(wm8994->retune_mobile_texts);
+ kfree(wm8994->drc_texts);
kfree(wm8994);
return 0;
.resume = wm8994_resume,
.read = wm8994_read,
.write = wm8994_write,
+ .readable_register = wm8994_readable,
+ .volatile_register = wm8994_volatile,
.set_bias_level = wm8994_set_bias_level,
};
}
/* davinci-evm digital audio interface glue - connects codec <--> CPU */
-static struct snd_soc_dai_link evm_dai = {
+static struct snd_soc_dai_link dm6446_evm_dai = {
.name = "TLV320AIC3X",
.stream_name = "AIC3X",
- .cpu_dai_name = "davinci-mcasp.0",
+ .cpu_dai_name = "davinci-mcbsp",
.codec_dai_name = "tlv320aic3x-hifi",
- .codec_name = "tlv320aic3x-codec.0-001a",
+ .codec_name = "tlv320aic3x-codec.1-001b",
+ .platform_name = "davinci-pcm-audio",
+ .init = evm_aic3x_init,
+ .ops = &evm_ops,
+};
+
+static struct snd_soc_dai_link dm355_evm_dai = {
+ .name = "TLV320AIC3X",
+ .stream_name = "AIC3X",
+ .cpu_dai_name = "davinci-mcbsp.1",
+ .codec_dai_name = "tlv320aic3x-hifi",
+ .codec_name = "tlv320aic3x-codec.1-001b",
.platform_name = "davinci-pcm-audio",
.init = evm_aic3x_init,
.ops = &evm_ops,
#ifdef CONFIG_SND_DM365_AIC3X_CODEC
.name = "TLV320AIC3X",
.stream_name = "AIC3X",
- .cpu_dai_name = "davinci-i2s",
+ .cpu_dai_name = "davinci-mcbsp",
.codec_dai_name = "tlv320aic3x-hifi",
.init = evm_aic3x_init,
- .codec_name = "tlv320aic3x-codec.0-001a",
+ .codec_name = "tlv320aic3x-codec.1-0018",
.ops = &evm_ops,
#elif defined(CONFIG_SND_DM365_VOICE_CODEC)
.name = "Voice Codec - CQ93VC",
.ops = &evm_ops,
};
-/* davinci dm6446, dm355 evm audio machine driver */
-static struct snd_soc_card snd_soc_card_evm = {
- .name = "DaVinci EVM",
- .dai_link = &evm_dai,
+/* davinci dm6446 evm audio machine driver */
+static struct snd_soc_card dm6446_snd_soc_card_evm = {
+ .name = "DaVinci DM6446 EVM",
+ .dai_link = &dm6446_evm_dai,
+ .num_links = 1,
+};
+
+/* davinci dm355 evm audio machine driver */
+static struct snd_soc_card dm355_snd_soc_card_evm = {
+ .name = "DaVinci DM355 EVM",
+ .dai_link = &dm355_evm_dai,
.num_links = 1,
};
int ret;
if (machine_is_davinci_evm()) {
- evm_snd_dev_data = &snd_soc_card_evm;
+ evm_snd_dev_data = &dm6446_snd_soc_card_evm;
index = 0;
} else if (machine_is_davinci_dm355_evm()) {
- evm_snd_dev_data = &snd_soc_card_evm;
+ evm_snd_dev_data = &dm355_snd_soc_card_evm;
index = 1;
} else if (machine_is_davinci_dm365_evm()) {
evm_snd_dev_data = &dm365_snd_soc_card_evm;
snd_pcm_format_t fmt;
unsigned element_cnt = 1;
- dai->capture_dma_data = dev->dma_params;
- dai->playback_dma_data = dev->dma_params;
-
/* general line settings */
spcr = davinci_mcbsp_read_reg(dev, DAVINCI_MCBSP_SPCR_REG);
if (substream->stream == SNDRV_PCM_STREAM_CAPTURE) {
return ret;
}
+static int davinci_i2s_startup(struct snd_pcm_substream *substream,
+ struct snd_soc_dai *dai)
+{
+ struct davinci_mcbsp_dev *dev = snd_soc_dai_get_drvdata(dai);
+
+ snd_soc_dai_set_dma_data(dai, substream, dev->dma_params);
+ return 0;
+}
+
static void davinci_i2s_shutdown(struct snd_pcm_substream *substream,
struct snd_soc_dai *dai)
{
#define DAVINCI_I2S_RATES SNDRV_PCM_RATE_8000_96000
static struct snd_soc_dai_ops davinci_i2s_dai_ops = {
+ .startup = davinci_i2s_startup,
.shutdown = davinci_i2s_shutdown,
.prepare = davinci_i2s_prepare,
.trigger = davinci_i2s_trigger,
.probe = davinci_i2s_probe,
.remove = davinci_i2s_remove,
.driver = {
- .name = "davinci-i2s",
+ .name = "davinci-mcbsp",
.owner = THIS_MODULE,
},
};
int word_length;
u8 fifo_level;
- cpu_dai->capture_dma_data = dev->dma_params;
- cpu_dai->playback_dma_data = dev->dma_params;
-
davinci_hw_common_param(dev, substream->stream);
if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
fifo_level = dev->txnumevt;
return ret;
}
+static int davinci_mcasp_startup(struct snd_pcm_substream *substream,
+ struct snd_soc_dai *dai)
+{
+ struct davinci_audio_dev *dev = snd_soc_dai_get_drvdata(dai);
+
+ snd_soc_dai_set_dma_data(dai, substream, dev->dma_params);
+ return 0;
+}
+
static struct snd_soc_dai_ops davinci_mcasp_dai_ops = {
+ .startup = davinci_mcasp_startup,
.trigger = davinci_mcasp_trigger,
.hw_params = davinci_mcasp_hw_params,
.set_fmt = davinci_mcasp_set_dai_fmt,
static struct snd_soc_dai_link sffsdr_dai = {
.name = "PCM3008", /* Codec name */
.stream_name = "PCM3008 HiFi",
- .cpu_dai_name = "davinci-asp.0",
+ .cpu_dai_name = "davinci-mcbsp",
.codec_dai_name = "pcm3008-hifi",
.codec_name = "pcm3008-codec",
.platform_name = "davinci-pcm-audio",
&davinci_vcif_dev->dma_params[substream->stream];
u32 w;
- dai->capture_dma_data = davinci_vcif_dev->dma_params;
- dai->playback_dma_data = davinci_vcif_dev->dma_params;
-
/* Restart the codec before setup */
davinci_vcif_stop(substream);
davinci_vcif_start(substream);
return ret;
}
+static int davinci_vcif_startup(struct snd_pcm_substream *substream,
+ struct snd_soc_dai *dai)
+{
+ struct davinci_vcif_dev *dev = snd_soc_dai_get_drvdata(dai);
+
+ snd_soc_dai_set_dma_data(dai, substream, dev->dma_params);
+ return 0;
+}
+
#define DAVINCI_VCIF_RATES SNDRV_PCM_RATE_8000_48000
static struct snd_soc_dai_ops davinci_vcif_dai_ops = {
+ .startup = davinci_vcif_startup,
.trigger = davinci_vcif_trigger,
.hw_params = davinci_vcif_hw_params,
};
static int davinci_vcif_remove(struct platform_device *pdev)
{
+ struct davinci_vcif_dev *davinci_vcif_dev = dev_get_drvdata(&pdev->dev);
+
snd_soc_unregister_dai(&pdev->dev);
+ kfree(davinci_vcif_dev);
return 0;
}
ret = platform_device_add(simone_snd_ac97_device);
if (ret)
- goto fail;
+ goto fail1;
simone_snd_device = platform_device_alloc("soc-audio", -1);
if (!simone_snd_device) {
ret = -ENOMEM;
- goto fail;
+ goto fail2;
}
platform_set_drvdata(simone_snd_device, &snd_soc_simone);
ret = platform_device_add(simone_snd_device);
- if (ret) {
- platform_device_put(simone_snd_device);
- goto fail;
- }
+ if (ret)
+ goto fail3;
- return ret;
+ return 0;
-fail:
+fail3:
+ platform_device_put(simone_snd_device);
+fail2:
+ platform_device_del(simone_snd_ac97_device);
+fail1:
platform_device_put(simone_snd_ac97_device);
return ret;
}
rc = platform_device_add(pdev);
if (rc) {
pr_err("efika_fabric_init: platform_device_add() failed\n");
+ platform_device_put(pdev);
return -ENODEV;
}
return 0;
#include <linux/module.h>
#include <linux/of_device.h>
#include <linux/slab.h>
-#include <linux/of_device.h>
#include <linux/of_platform.h>
#include <sound/soc.h>
rc = snd_soc_register_dais(&op->dev, psc_i2s_dai, ARRAY_SIZE(psc_i2s_dai));
if (rc != 0) {
pr_err("Failed to register DAI\n");
- return 0;
+ return rc;
}
psc_dma = dev_get_drvdata(&op->dev);
rc = platform_device_add(pdev);
if (rc) {
pr_err("pcm030_fabric_init: platform_device_add() failed\n");
+ platform_device_put(pdev);
return -ENODEV;
}
return 0;
struct snd_pcm_hw_params *params)
{
struct snd_soc_pcm_runtime *rtd = substream->private_data;
- struct snd_soc_dai *codec_dai = rtd->dai->codec_dai;
- struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
+ struct snd_soc_dai *codec_dai = rtd->codec_dai;
+ struct snd_soc_dai *cpu_dai = rtd->cpu_dai;
int ret;
ret = snd_soc_dai_set_fmt(cpu_dai, SND_SOC_DAIFMT_I2S |
static struct snd_soc_dai_link eukrea_tlv320_dai = {
.name = "tlv320aic23",
.stream_name = "TLV320AIC23",
- .codec_dai = "tlv320aic23-hifi",
+ .codec_dai_name = "tlv320aic23-hifi",
.platform_name = "imx-pcm-audio.0",
.codec_name = "tlv320aic23-codec.0-001a",
- .cpu_dai = "imx-ssi.0",
+ .cpu_dai_name = "imx-ssi.0",
.ops = &eukrea_tlv320_snd_ops,
};
#include <linux/module.h>
#include <linux/platform_device.h>
#include <linux/slab.h>
+#include <linux/dmaengine.h>
#include <sound/core.h>
#include <sound/initval.h>
#include <sound/pcm_params.h>
#include <sound/soc.h>
-#include <mach/dma-mx1-mx2.h>
+#include <mach/dma.h>
#include "imx-ssi.h"
struct imx_pcm_runtime_data {
- int sg_count;
- struct scatterlist *sg_list;
- int period;
+ int period_bytes;
int periods;
- unsigned long dma_addr;
int dma;
- struct snd_pcm_substream *substream;
unsigned long offset;
unsigned long size;
- unsigned long period_cnt;
void *buf;
int period_time;
+ struct dma_async_tx_descriptor *desc;
+ struct dma_chan *dma_chan;
+ struct imx_dma_data dma_data;
};
-/* Called by the DMA framework when a period has elapsed */
-static void imx_ssi_dma_progression(int channel, void *data,
- struct scatterlist *sg)
+static void audio_dma_irq(void *data)
{
- struct snd_pcm_substream *substream = data;
+ struct snd_pcm_substream *substream = (struct snd_pcm_substream *)data;
struct snd_pcm_runtime *runtime = substream->runtime;
struct imx_pcm_runtime_data *iprtd = runtime->private_data;
- if (!sg)
- return;
-
- runtime = iprtd->substream->runtime;
+ iprtd->offset += iprtd->period_bytes;
+ iprtd->offset %= iprtd->period_bytes * iprtd->periods;
- iprtd->offset = sg->dma_address - runtime->dma_addr;
-
- snd_pcm_period_elapsed(iprtd->substream);
+ snd_pcm_period_elapsed(substream);
}
-static void imx_ssi_dma_callback(int channel, void *data)
+static bool filter(struct dma_chan *chan, void *param)
{
- pr_err("%s shouldn't be called\n", __func__);
-}
+ struct imx_pcm_runtime_data *iprtd = param;
-static void snd_imx_dma_err_callback(int channel, void *data, int err)
-{
- struct snd_pcm_substream *substream = data;
- struct snd_soc_pcm_runtime *rtd = substream->private_data;
- struct imx_pcm_dma_params *dma_params =
- snd_soc_dai_get_dma_data(rtd->dai->cpu_dai, substream);
- struct snd_pcm_runtime *runtime = substream->runtime;
- struct imx_pcm_runtime_data *iprtd = runtime->private_data;
- int ret;
+ if (!imx_dma_is_general_purpose(chan))
+ return false;
- pr_err("DMA timeout on channel %d -%s%s%s%s\n",
- channel,
- err & IMX_DMA_ERR_BURST ? " burst" : "",
- err & IMX_DMA_ERR_REQUEST ? " request" : "",
- err & IMX_DMA_ERR_TRANSFER ? " transfer" : "",
- err & IMX_DMA_ERR_BUFFER ? " buffer" : "");
+ chan->private = &iprtd->dma_data;
- imx_dma_disable(iprtd->dma);
- ret = imx_dma_setup_sg(iprtd->dma, iprtd->sg_list, iprtd->sg_count,
- IMX_DMA_LENGTH_LOOP, dma_params->dma_addr,
- substream->stream == SNDRV_PCM_STREAM_PLAYBACK ?
- DMA_MODE_WRITE : DMA_MODE_READ);
- if (!ret)
- imx_dma_enable(iprtd->dma);
+ return true;
}
-static int imx_ssi_dma_alloc(struct snd_pcm_substream *substream)
+static int imx_ssi_dma_alloc(struct snd_pcm_substream *substream,
+ struct snd_pcm_hw_params *params)
{
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct imx_pcm_dma_params *dma_params;
struct snd_pcm_runtime *runtime = substream->runtime;
struct imx_pcm_runtime_data *iprtd = runtime->private_data;
+ struct dma_slave_config slave_config;
+ dma_cap_mask_t mask;
+ enum dma_slave_buswidth buswidth;
int ret;
dma_params = snd_soc_dai_get_dma_data(rtd->cpu_dai, substream);
- iprtd->dma = imx_dma_request_by_prio(DRV_NAME, DMA_PRIO_HIGH);
- if (iprtd->dma < 0) {
- pr_err("Failed to claim the audio DMA\n");
- return -ENODEV;
- }
+ iprtd->dma_data.peripheral_type = IMX_DMATYPE_SSI;
+ iprtd->dma_data.priority = DMA_PRIO_HIGH;
+ iprtd->dma_data.dma_request = dma_params->dma;
- ret = imx_dma_setup_handlers(iprtd->dma,
- imx_ssi_dma_callback,
- snd_imx_dma_err_callback, substream);
- if (ret)
- goto out;
+ /* Try to grab a DMA channel */
+ dma_cap_zero(mask);
+ dma_cap_set(DMA_SLAVE, mask);
+ iprtd->dma_chan = dma_request_channel(mask, filter, iprtd);
+ if (!iprtd->dma_chan)
+ return -EINVAL;
- ret = imx_dma_setup_progression_handler(iprtd->dma,
- imx_ssi_dma_progression);
- if (ret) {
- pr_err("Failed to setup the DMA handler\n");
- goto out;
+ switch (params_format(params)) {
+ case SNDRV_PCM_FORMAT_S16_LE:
+ buswidth = DMA_SLAVE_BUSWIDTH_2_BYTES;
+ break;
+ case SNDRV_PCM_FORMAT_S20_3LE:
+ case SNDRV_PCM_FORMAT_S24_LE:
+ buswidth = DMA_SLAVE_BUSWIDTH_4_BYTES;
+ break;
+ default:
+ return 0;
}
- ret = imx_dma_config_channel(iprtd->dma,
- IMX_DMA_MEMSIZE_16 | IMX_DMA_TYPE_FIFO,
- IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR,
- dma_params->dma, 1);
- if (ret < 0) {
- pr_err("Cannot configure DMA channel: %d\n", ret);
- goto out;
+ if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+ slave_config.direction = DMA_TO_DEVICE;
+ slave_config.dst_addr = dma_params->dma_addr;
+ slave_config.dst_addr_width = buswidth;
+ slave_config.dst_maxburst = dma_params->burstsize;
+ } else {
+ slave_config.direction = DMA_FROM_DEVICE;
+ slave_config.src_addr = dma_params->dma_addr;
+ slave_config.src_addr_width = buswidth;
+ slave_config.src_maxburst = dma_params->burstsize;
}
- imx_dma_config_burstlen(iprtd->dma, dma_params->burstsize * 2);
+ ret = dmaengine_slave_config(iprtd->dma_chan, &slave_config);
+ if (ret)
+ return ret;
return 0;
-out:
- imx_dma_free(iprtd->dma);
- return ret;
}
static int snd_imx_pcm_hw_params(struct snd_pcm_substream *substream,
struct snd_pcm_hw_params *params)
{
+ struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_pcm_runtime *runtime = substream->runtime;
struct imx_pcm_runtime_data *iprtd = runtime->private_data;
- int i;
unsigned long dma_addr;
+ struct dma_chan *chan;
+ struct imx_pcm_dma_params *dma_params;
+ int ret;
- imx_ssi_dma_alloc(substream);
+ dma_params = snd_soc_dai_get_dma_data(rtd->cpu_dai, substream);
+ ret = imx_ssi_dma_alloc(substream, params);
+ if (ret)
+ return ret;
+ chan = iprtd->dma_chan;
iprtd->size = params_buffer_bytes(params);
iprtd->periods = params_periods(params);
- iprtd->period = params_period_bytes(params);
+ iprtd->period_bytes = params_period_bytes(params);
iprtd->offset = 0;
iprtd->period_time = HZ / (params_rate(params) /
params_period_size(params));
snd_pcm_set_runtime_buffer(substream, &substream->dma_buffer);
- if (iprtd->sg_count != iprtd->periods) {
- kfree(iprtd->sg_list);
-
- iprtd->sg_list = kcalloc(iprtd->periods + 1,
- sizeof(struct scatterlist), GFP_KERNEL);
- if (!iprtd->sg_list)
- return -ENOMEM;
- iprtd->sg_count = iprtd->periods + 1;
- }
-
- sg_init_table(iprtd->sg_list, iprtd->sg_count);
dma_addr = runtime->dma_addr;
- for (i = 0; i < iprtd->periods; i++) {
- iprtd->sg_list[i].page_link = 0;
- iprtd->sg_list[i].offset = 0;
- iprtd->sg_list[i].dma_address = dma_addr;
- iprtd->sg_list[i].length = iprtd->period;
- dma_addr += iprtd->period;
+ iprtd->buf = (unsigned int *)substream->dma_buffer.area;
+
+ iprtd->desc = chan->device->device_prep_dma_cyclic(chan, dma_addr,
+ iprtd->period_bytes * iprtd->periods,
+ iprtd->period_bytes,
+ substream->stream == SNDRV_PCM_STREAM_PLAYBACK ?
+ DMA_TO_DEVICE : DMA_FROM_DEVICE);
+ if (!iprtd->desc) {
+ dev_err(&chan->dev->device, "cannot prepare slave dma\n");
+ return -EINVAL;
}
- /* close the loop */
- iprtd->sg_list[iprtd->sg_count - 1].offset = 0;
- iprtd->sg_list[iprtd->sg_count - 1].length = 0;
- iprtd->sg_list[iprtd->sg_count - 1].page_link =
- ((unsigned long) iprtd->sg_list | 0x01) & ~0x02;
+ iprtd->desc->callback = audio_dma_irq;
+ iprtd->desc->callback_param = substream;
+
return 0;
}
struct snd_pcm_runtime *runtime = substream->runtime;
struct imx_pcm_runtime_data *iprtd = runtime->private_data;
- if (iprtd->dma >= 0) {
- imx_dma_free(iprtd->dma);
- iprtd->dma = -EINVAL;
+ if (iprtd->dma_chan) {
+ dma_release_channel(iprtd->dma_chan);
+ iprtd->dma_chan = NULL;
}
- kfree(iprtd->sg_list);
- iprtd->sg_list = NULL;
-
return 0;
}
static int snd_imx_pcm_prepare(struct snd_pcm_substream *substream)
{
- struct snd_pcm_runtime *runtime = substream->runtime;
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct imx_pcm_dma_params *dma_params;
- struct imx_pcm_runtime_data *iprtd = runtime->private_data;
- int err;
dma_params = snd_soc_dai_get_dma_data(rtd->cpu_dai, substream);
- iprtd->substream = substream;
- iprtd->buf = (unsigned int *)substream->dma_buffer.area;
- iprtd->period_cnt = 0;
-
- pr_debug("%s: buf: %p period: %d periods: %d\n",
- __func__, iprtd->buf, iprtd->period, iprtd->periods);
-
- err = imx_dma_setup_sg(iprtd->dma, iprtd->sg_list, iprtd->sg_count,
- IMX_DMA_LENGTH_LOOP, dma_params->dma_addr,
- substream->stream == SNDRV_PCM_STREAM_PLAYBACK ?
- DMA_MODE_WRITE : DMA_MODE_READ);
- if (err)
- return err;
-
return 0;
}
case SNDRV_PCM_TRIGGER_START:
case SNDRV_PCM_TRIGGER_RESUME:
case SNDRV_PCM_TRIGGER_PAUSE_RELEASE:
- imx_dma_enable(iprtd->dma);
+ dmaengine_submit(iprtd->desc);
break;
case SNDRV_PCM_TRIGGER_STOP:
case SNDRV_PCM_TRIGGER_SUSPEND:
case SNDRV_PCM_TRIGGER_PAUSE_PUSH:
- imx_dma_disable(iprtd->dma);
+ dmaengine_terminate_all(iprtd->dma_chan);
break;
default:
struct snd_pcm_runtime *runtime = substream->runtime;
struct imx_pcm_runtime_data *iprtd = runtime->private_data;
+ pr_debug("%s: %ld %ld\n", __func__, iprtd->offset,
+ bytes_to_frames(substream->runtime, iprtd->offset));
+
return bytes_to_frames(substream->runtime, iprtd->offset);
}
.channels_max = 2,
.buffer_bytes_max = IMX_SSI_DMABUF_SIZE,
.period_bytes_min = 128,
- .period_bytes_max = 16 * 1024,
+ .period_bytes_max = 65535, /* Limited by SDMA engine */
.periods_min = 2,
.periods_max = 255,
.fifo_size = 0,
}
snd_soc_set_runtime_hwparams(substream, &snd_imx_hardware);
+
+ return 0;
+}
+
+static int snd_imx_close(struct snd_pcm_substream *substream)
+{
+ struct snd_pcm_runtime *runtime = substream->runtime;
+ struct imx_pcm_runtime_data *iprtd = runtime->private_data;
+
+ kfree(iprtd);
+
return 0;
}
static struct snd_pcm_ops imx_pcm_ops = {
.open = snd_imx_open,
+ .close = snd_imx_close,
.ioctl = snd_pcm_lib_ioctl,
.hw_params = snd_imx_pcm_hw_params,
.hw_free = snd_imx_pcm_hw_free,
.name = "imx-pcm-audio",
.owner = THIS_MODULE,
},
-
.probe = imx_soc_platform_probe,
.remove = __devexit_p(imx_soc_platform_remove),
};
platform_driver_unregister(&imx_pcm_driver);
}
module_exit(snd_imx_pcm_exit);
-
}
EXPORT_SYMBOL_GPL(imx_pcm_free);
+static int imx_ssi_dai_probe(struct snd_soc_dai *dai)
+{
+ struct imx_ssi *ssi = dev_get_drvdata(dai->dev);
+ uint32_t val;
+
+ snd_soc_dai_set_drvdata(dai, ssi);
+
+ val = SSI_SFCSR_TFWM0(ssi->dma_params_tx.burstsize) |
+ SSI_SFCSR_RFWM0(ssi->dma_params_rx.burstsize);
+ writel(val, ssi->base + SSI_SFCSR);
+
+ return 0;
+}
+
static struct snd_soc_dai_driver imx_ssi_dai = {
+ .probe = imx_ssi_dai_probe,
.playback = {
.channels_min = 2,
.channels_max = 2,
.ops = &imx_ssi_pcm_dai_ops,
};
-static int imx_ssi_dai_probe(struct snd_soc_dai *dai)
-{
- struct imx_ssi *ssi = dev_get_drvdata(dai->dev);
- uint32_t val;
-
- snd_soc_dai_set_drvdata(dai, ssi);
-
- val = SSI_SFCSR_TFWM0(ssi->dma_params_tx.burstsize) |
- SSI_SFCSR_RFWM0(ssi->dma_params_rx.burstsize);
- writel(val, ssi->base + SSI_SFCSR);
-
- return 0;
-}
-
static struct snd_soc_dai_driver imx_ac97_dai = {
.probe = imx_ssi_dai_probe,
.ac97_control = 1,
goto failed_register;
}
- ssi->soc_platform_pdev = platform_device_alloc("imx-fiq-pcm-audio", pdev->id);
- if (!ssi->soc_platform_pdev)
+ ssi->soc_platform_pdev_fiq = platform_device_alloc("imx-fiq-pcm-audio", pdev->id);
+ if (!ssi->soc_platform_pdev_fiq) {
+ ret = -ENOMEM;
+ goto failed_pdev_fiq_alloc;
+ }
+
+ platform_set_drvdata(ssi->soc_platform_pdev_fiq, ssi);
+ ret = platform_device_add(ssi->soc_platform_pdev_fiq);
+ if (ret) {
+ dev_err(&pdev->dev, "failed to add platform device\n");
+ goto failed_pdev_fiq_add;
+ }
+
+ ssi->soc_platform_pdev = platform_device_alloc("imx-pcm-audio", pdev->id);
+ if (!ssi->soc_platform_pdev) {
+ ret = -ENOMEM;
goto failed_pdev_alloc;
+ }
+
platform_set_drvdata(ssi->soc_platform_pdev, ssi);
ret = platform_device_add(ssi->soc_platform_pdev);
if (ret) {
failed_pdev_add:
platform_device_put(ssi->soc_platform_pdev);
failed_pdev_alloc:
+ platform_device_del(ssi->soc_platform_pdev_fiq);
+failed_pdev_fiq_add:
+ platform_device_put(ssi->soc_platform_pdev_fiq);
+failed_pdev_fiq_alloc:
snd_soc_unregister_dai(&pdev->dev);
failed_register:
failed_ac97:
struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
struct imx_ssi *ssi = platform_get_drvdata(pdev);
- platform_device_del(ssi->soc_platform_pdev);
- platform_device_put(ssi->soc_platform_pdev);
+ platform_device_unregister(ssi->soc_platform_pdev);
+ platform_device_unregister(ssi->soc_platform_pdev_fiq);
snd_soc_unregister_dai(&pdev->dev);
#define DRV_NAME "imx-ssi"
+#include <linux/dmaengine.h>
+#include <mach/dma.h>
+
struct imx_pcm_dma_params {
int dma;
unsigned long dma_addr;
int enabled;
struct platform_device *soc_platform_pdev;
+ struct platform_device *soc_platform_pdev_fiq;
};
struct snd_soc_platform *imx_ssi_fiq_init(struct platform_device *pdev,
#include <sound/soc-dapm.h>
#include <asm/mach-types.h>
-#include "../codecs/wm9712.h"
-#include "imx-ssi.h"
-
static struct snd_soc_card imx_phycore;
static struct snd_soc_ops imx_phycore_hifi_ops = {
};
static struct snd_soc_card imx_phycore = {
- .name = "PhyCORE-audio",
+ .name = "PhyCORE-ac97-audio",
.dai_link = imx_phycore_dai_ac97,
.num_links = ARRAY_SIZE(imx_phycore_dai_ac97),
};
+static struct platform_device *imx_phycore_snd_ac97_device;
static struct platform_device *imx_phycore_snd_device;
static int __init imx_phycore_init(void)
/* return happy. We might run on a totally different machine */
return 0;
- imx_phycore_snd_device = platform_device_alloc("soc-audio", -1);
- if (!imx_phycore_snd_device)
+ imx_phycore_snd_ac97_device = platform_device_alloc("soc-audio", -1);
+ if (!imx_phycore_snd_ac97_device)
return -ENOMEM;
- platform_set_drvdata(imx_phycore_snd_device, &imx_phycore);
- ret = platform_device_add(imx_phycore_snd_device);
+ platform_set_drvdata(imx_phycore_snd_ac97_device, &imx_phycore);
+ ret = platform_device_add(imx_phycore_snd_ac97_device);
+ if (ret)
+ goto fail1;
imx_phycore_snd_device = platform_device_alloc("wm9712-codec", -1);
- if (!imx_phycore_snd_device)
- return -ENOMEM;
+ if (!imx_phycore_snd_device) {
+ ret = -ENOMEM;
+ goto fail2;
+ }
ret = platform_device_add(imx_phycore_snd_device);
if (ret) {
printk(KERN_ERR "ASoC: Platform device allocation failed\n");
- platform_device_put(imx_phycore_snd_device);
+ goto fail3;
}
+ return 0;
+
+fail3:
+ platform_device_put(imx_phycore_snd_device);
+fail2:
+ platform_device_del(imx_phycore_snd_ac97_device);
+fail1:
+ platform_device_put(imx_phycore_snd_ac97_device);
return ret;
}
static void __exit imx_phycore_exit(void)
{
platform_device_unregister(imx_phycore_snd_device);
+ platform_device_unregister(imx_phycore_snd_ac97_device);
}
late_initcall(imx_phycore_init);
static int __devexit nuc900_ac97_drvremove(struct platform_device *pdev)
{
-
snd_soc_unregister_dai(&pdev->dev);
clk_put(nuc900_ac97_data->clk);
release_mem_region(nuc900_ac97_data->res->start,
resource_size(nuc900_ac97_data->res));
+ kfree(nuc900_ac97_data);
nuc900_ac97_data = NULL;
return 0;
case OMAP_MCBSP_CLKR_SRC_CLKR:
+ if (cpu_class_is_omap1())
+ break;
omap2_mcbsp1_mux_clkr_src(CLKR_SRC_CLKR);
break;
case OMAP_MCBSP_CLKR_SRC_CLKX:
+ if (cpu_class_is_omap1())
+ break;
omap2_mcbsp1_mux_clkr_src(CLKR_SRC_CLKX);
break;
case OMAP_MCBSP_FSR_SRC_FSR:
+ if (cpu_class_is_omap1())
+ break;
omap2_mcbsp1_mux_fsr_src(FSR_SRC_FSR);
break;
case OMAP_MCBSP_FSR_SRC_FSX:
+ if (cpu_class_is_omap1())
+ break;
omap2_mcbsp1_mux_fsr_src(FSR_SRC_FSX);
break;
default:
pr_err(PREFIX "Failed to get DAC regulator from %s: %ld\n",
dev_name(&omap3pandora_snd_device->dev),
PTR_ERR(omap3pandora_dac_reg));
+ ret = PTR_ERR(omap3pandora_dac_reg);
goto fail3;
}
tlv320aic23_mclk = clk_get(dev, "mclk");
if (IS_ERR(tlv320aic23_mclk)) {
printk(KERN_ERR "Could not get mclk clock\n");
- return -ENODEV;
+ err = PTR_ERR(tlv320aic23_mclk);
+ goto err2;
}
/*
if (clk_set_rate(tlv320aic23_mclk, CODEC_CLOCK)) {
printk(KERN_ERR "Cannot set MCLK for AIC23 CODEC\n");
err = -ECANCELED;
- goto err1;
+ goto err3;
}
}
(uint) clk_get_rate(tlv320aic23_mclk), CODEC_CLOCK);
return 0;
-err1:
+
+err3:
clk_put(tlv320aic23_mclk);
+err2:
platform_device_del(osk_snd_device);
+err1:
platform_device_put(osk_snd_device);
return err;
static void __exit osk_soc_exit(void)
{
+ clk_put(tlv320aic23_mclk);
platform_device_unregister(osk_snd_device);
}
config SND_PXA2XX_SOC
tristate "SoC Audio for the Intel PXA2xx chip"
depends on ARCH_PXA
+ select SND_ARM
select SND_PXA2XX_LIB
help
Say Y or M if you want to add support for codecs attached to
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_soc_codec *codec = rtd->codec;
+ mutex_lock(&codec->mutex);
+
/* check the jack status at stream startup */
corgi_ext_control(codec);
+
+ mutex_unlock(&codec->mutex);
+
return 0;
}
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_soc_codec *codec = rtd->codec;
+ mutex_lock(&codec->mutex);
+
/* check the jack status at stream startup */
magician_ext_control(codec);
+ mutex_unlock(&codec->mutex);
+
return 0;
}
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_soc_codec *codec = rtd->codec;
+ mutex_lock(&codec->mutex);
+
/* check the jack status at stream startup */
poodle_ext_control(codec);
+
+ mutex_unlock(&codec->mutex);
+
return 0;
}
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_soc_codec *codec = rtd->codec;
+ mutex_lock(&codec->mutex);
+
/* check the jack status at stream startup */
spitz_ext_control(codec);
+
+ mutex_unlock(&codec->mutex);
+
return 0;
}
struct snd_soc_pcm_runtime *rtd = substream->private_data;
struct snd_soc_codec *codec = rtd->codec;
+ mutex_lock(&codec->mutex);
+
/* check the jack status at stream startup */
tosa_ext_control(codec);
+
+ mutex_unlock(&codec->mutex);
+
return 0;
}
tristate "SoC Audio for the Samsung S3CXXXX chips"
depends on ARCH_S3C2410 || ARCH_S3C64XX || ARCH_S5PC100 || ARCH_S5PV210
select S3C64XX_DMA if ARCH_S3C64XX
+ select S3C2410_DMA if ARCH_S3C2410
help
Say Y or M if you want to add support for codecs attached to
the S3C24XX AC97 or I2S interfaces. You will also need to
16000,
44100,
48000,
- 88200,
};
static struct snd_pcm_hw_constraint_list hw_rates = {
};
static struct platform_device *s3c24xx_snd_device;
-static struct clk *xtal;
static int rx1950_startup(struct snd_pcm_substream *substream)
{
case 44100:
case 88200:
clk_source = S3C24XX_CLKSRC_MPLL;
- fs_mode = S3C2410_IISMOD_256FS;
- div = clk_get_rate(xtal) / (256 * rate);
- if (clk_get_rate(xtal) % (256 * rate) > (128 * rate))
- div++;
+ fs_mode = S3C2410_IISMOD_384FS;
+ div = 1;
break;
default:
printk(KERN_ERR "%s: rate %d is not supported\n",
/* set MCLK division for sample rate */
ret = snd_soc_dai_set_clkdiv(cpu_dai, S3C24XX_DIV_MCLK,
- S3C2410_IISMOD_384FS);
+ fs_mode);
if (ret < 0)
return ret;
goto err_plat_add;
}
- xtal = clk_get(&s3c24xx_snd_device->dev, "xtal");
-
- if (IS_ERR(xtal)) {
- ret = PTR_ERR(xtal);
- platform_device_unregister(s3c24xx_snd_device);
- goto err_clk;
- }
-
return 0;
-err_clk:
err_plat_add:
err_plat_alloc:
err_gpio_conf:
platform_device_unregister(s3c24xx_snd_device);
snd_soc_jack_free_gpios(&hp_jack, ARRAY_SIZE(hp_jack_gpios),
hp_jack_gpios);
- clk_put(xtal);
gpio_free(S3C2410_GPA(1));
}
}
mout_epll = clk_get(NULL, "mout_epll");
- if (IS_ERR(fout_epll)) {
+ if (IS_ERR(mout_epll)) {
printk(KERN_WARNING "%s: Cannot find mout_epll.\n",
__func__);
ret = -EINVAL;
}
sclk_spdif = clk_get(NULL, "sclk_spdif");
- if (IS_ERR(fout_epll)) {
+ if (IS_ERR(sclk_spdif)) {
printk(KERN_WARNING "%s: Cannot find sclk_spdif.\n",
__func__);
ret = -EINVAL;
struct fsi_stream playback;
struct fsi_stream capture;
+ long rate;
+
u32 mst_ctrl;
};
{
struct fsi_priv *fsi = fsi_get_priv(substream);
int is_play = fsi_is_play(substream);
+ struct fsi_master *master = fsi_get_master(fsi);
+ int (*set_rate)(struct device *dev, int is_porta, int rate, int enable);
fsi_irq_disable(fsi, is_play);
fsi_clk_ctrl(fsi, 0);
+ set_rate = master->info->set_rate;
+ if (set_rate && fsi->rate)
+ set_rate(dai->dev, fsi_is_port_a(fsi), fsi->rate, 0);
+ fsi->rate = 0;
+
pm_runtime_put_sync(dai->dev);
}
{
struct fsi_priv *fsi = fsi_get_priv(substream);
struct fsi_master *master = fsi_get_master(fsi);
- int (*set_rate)(int is_porta, int rate) = master->info->set_rate;
+ int (*set_rate)(struct device *dev, int is_porta, int rate, int enable);
int fsi_ver = master->core->ver;
- int is_play = fsi_is_play(substream);
+ long rate = params_rate(params);
int ret;
- /* if slave mode, set_rate is not needed */
- if (!fsi_is_master_mode(fsi, is_play))
+ set_rate = master->info->set_rate;
+ if (!set_rate)
return 0;
- /* it is error if no set_rate */
- if (!set_rate)
- return -EIO;
+ ret = set_rate(dai->dev, fsi_is_port_a(fsi), rate, 1);
+ if (ret < 0) /* error */
+ return ret;
- ret = set_rate(fsi_is_port_a(fsi), params_rate(params));
+ fsi->rate = rate;
if (ret > 0) {
u32 data = 0;
static int __devexit sh4_soc_dai_remove(struct platform_device *pdev)
{
- snd_soc_unregister_dai(&pdev->dev, ARRAY_SIZE(sh4_ssi_dai));
+ snd_soc_unregister_dais(&pdev->dev, ARRAY_SIZE(sh4_ssi_dai));
return 0;
}
for (i = 0; i < count; i++) {
dai = kzalloc(sizeof(struct snd_soc_dai), GFP_KERNEL);
- if (dai == NULL)
- return -ENOMEM;
+ if (dai == NULL) {
+ ret = -ENOMEM;
+ goto err;
+ }
/* create DAI component name */
dai->name = fmt_multiple_name(dev, &dai_drv[i]);
return 0;
error:
- for (i--; i >= 0; i--)
- snd_soc_unregister_dai(dev);
-
if (codec->reg_cache)
kfree(codec->reg_cache);
kfree(codec->name);
struct snd_soc_dapm_widget *b,
int sort[])
{
- if (a->codec != b->codec)
- return (unsigned long)a - (unsigned long)b;
if (sort[a->id] != sort[b->id])
return sort[a->id] - sort[b->id];
if (a->reg != b->reg)
return a->reg - b->reg;
+ if (a->codec != b->codec)
+ return (unsigned long)a->codec - (unsigned long)b->codec;
return 0;
}
if (max_tries < 1)
max_tries = 1;
- /* ssc_div must be a power of 2. */
+ /* ssc_div must be even. */
ssc_div = (ssc_div + 1) & ~1UL;
if ((ssc_rate / (ssc_div * 2 * 16)) < BITRATE_MIN) {
if (err < 0)
err = event__synthesize_kernel_mmap(process_synthesized_event,
session, machine, "_stext");
- if (err < 0) {
- pr_err("Couldn't record kernel reference relocation symbol.\n");
- return err;
- }
+ if (err < 0)
+ pr_err("Couldn't record kernel reference relocation symbol\n"
+ "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+ "Check /proc/kallsyms permission or run as root.\n");
err = event__synthesize_modules(process_synthesized_event,
session, machine);
- if (err < 0) {
- pr_err("Couldn't record kernel reference relocation symbol.\n");
- return err;
- }
+ if (err < 0)
+ pr_err("Couldn't record kernel module information.\n"
+ "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+ "Check /proc/modules permission or run as root.\n");
+
if (perf_guest)
perf_session__process_machines(session, event__synthesize_guest_os);
{
struct rb_node **p = &self->rb_node;
struct rb_node *parent = NULL;
- struct symbol_name_rb_node *symn = ((void *)sym) - sizeof(*parent), *s;
+ struct symbol_name_rb_node *symn, *s;
+
+ symn = container_of(sym, struct symbol_name_rb_node, sym);
while (*p != NULL) {
parent = *p;
git.openpandora.org / pandora-kernel.git / commitdiff