pandora-kernel.git
18 years ago[PATCH] USB UHCI: improved reset handling
Alan Stern [Fri, 22 Apr 2005 18:39:12 +0000 (14:39 -0400)]
[PATCH] USB UHCI: improved reset handling

This patch improves the strategy uhci-hcd uses for performing controller
resets and checking whether they are needed.

The HCRESET command doesn't affect the Suspend, Resume,
or Reset bits in the port status & control registers, so
the driver must clear them by itself.  This means the
code to figure out how many ports there are has to be moved
to an earlier spot in the driver.

The R/WC bits in the USBLEGSUP register can be set by the
hardware even in the absence of BIOS meddling with legacy
support features.  Hence it's not a good idea to check them
while trying to determine whether the BIOS has altered the
controller's state.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: Use root-hub IRQs while suspended
Alan Stern [Thu, 21 Apr 2005 20:04:58 +0000 (16:04 -0400)]
[PATCH] USB UHCI: Use root-hub IRQs while suspended

This patch, which has as478b as a prerequisite, enables the uhci-hcd
driver to take advantage of root-hub IRQs rather than polling during the
time it is suspended.  (Unfortunately the hardware doesn't support
port-change interrupts while the controller is running.)  It also turns
off the driver's private timer while the controller is suspended, as it
isn't needed then.  The combined elimination of polling interrupts and
timer interrupts ought to be enough to allow some systems to save a
noticeable amount of power while they are otherwise idle.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: Fix up loose ends
Alan Stern [Sat, 9 Apr 2005 21:30:08 +0000 (17:30 -0400)]
[PATCH] USB UHCI: Fix up loose ends

This patch tidies up a few loose ends left by the preceding patches.
It indicates the controller supports remote wakeup whenever the PM
capability is present -- which shouldn't cause any harm if the
assumption turns out to be wrong.  It refuses to suspend the
controller if the root hub is still active, and it refuses to resume
the root hub if the controller is suspended.  It adds checks for a
dead controller in several spots, and it adds memory barriers as
needed to insure that I/O operations are completed before moving on.

Actually I'm not certain the last part is being done correctly.  With
code like this:

outw(..., ...);
mb();
udelay(5);

do we know for certain that the outw() will complete _before_ the
delay begins?  If not, how should this be written?

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: Add root-hub suspend/resume support
Alan Stern [Sat, 9 Apr 2005 21:29:00 +0000 (17:29 -0400)]
[PATCH] USB UHCI: Add root-hub suspend/resume support

This patch implements (finally!) separate suspend and resume routines
for the root hub and the controller in the UHCI driver.  It also
changes the sequence used to reset the controller during initial
probing, so as to preserve the existing state during a Resume-From-Disk.
(This new sequence is what should be used in the PCI Quirks code for
early USB handoffs, incidentally.)  Lastly it adds a notion of the
controller being "inaccessible" while in a PCI low-power state, when
normal I/O operations shouldn't be allowed.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: Add root hub states
Alan Stern [Sat, 9 Apr 2005 21:27:32 +0000 (17:27 -0400)]
[PATCH] USB UHCI: Add root hub states

This patch starts making some serious changes to the UHCI driver.
There's a set of private states for the root hub, and the internal
routines for suspending and resuming work completely differently, with
transitions based on the new states.  Now the driver distinguishes
between a privately auto-stopped state and a publicly suspended state,
and it will properly suspend controllers with broken resume-detect
interrupts instead of resetting them.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: Minor improvements
Alan Stern [Sat, 9 Apr 2005 21:26:00 +0000 (17:26 -0400)]
[PATCH] USB UHCI: Minor improvements

This patch makes a few small improvements in the UHCI driver.  Some
code is moved between different source files and a more useful pointer
is passed to a callback routine.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB UHCI: subroutine reordering
Alan Stern [Sat, 9 Apr 2005 21:24:42 +0000 (17:24 -0400)]
[PATCH] USB UHCI: subroutine reordering

This patch moves a few subroutines around in the uhci-hcd source file.
Nothing else is changed.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB: turn a user mode driver error into a hard error
David Brownell [Tue, 12 Apr 2005 01:34:17 +0000 (18:34 -0700)]
[PATCH] USB: turn a user mode driver error into a hard error

This patch turns a user mode driver error into a hard error, and updates
the relevant diagnostic slightly to help troubleshooting.  gphoto was
known to have this problem, hopefully it is now fixed (they have had
plenty of warning...)

This had been left as a soft error to give various user mode drivers a
change to be properly fixed, with the statement that starting in about
2.6.10 it would be changed.  It had been mostly safe as a soft error ...
but that can not be guaranteed.  Now that a year has passed, it's time to
really insist that the user mode drivers finally fix their relevant bugs.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB: Fix oops at rmmod after failed probe in isp116x-hcd
Olav Kongas [Thu, 21 Apr 2005 14:12:59 +0000 (17:12 +0300)]
[PATCH] USB: Fix oops at rmmod after failed probe in isp116x-hcd

This patch fixes an oops triggered at rmmod of isp116x-hcd
after the probe() has failed.

Also, it extends the error message printed, if the driver
cannot detect "Chip's Clock Ready" after a software reset.
As Ian Campbell recently reported, this happens if the
chip's H_WAKEUP pin is not pulled low during software reset.
Several people have already had this issue, hence the update
to the error message.

Also, extend the error message about the failed clock
detection after the software reset.

Signed-off-by: Olav Kongas <ok@artecdesign.ee>
18 years ago[PATCH] USB: Add isp116x-hcd USB host controller driver
Olav Kongas [Sat, 9 Apr 2005 19:57:39 +0000 (22:57 +0300)]
[PATCH] USB: Add isp116x-hcd USB host controller driver

This patch provides an "isp116x-hcd" driver for Philips'
ISP1160/ISP1161 USB host controllers.

The driver:
 - is relatively small, meant for use on embedded platforms.
 - runs usbtests 1-14 without problems for days.
 - has been in use by 6-7 different people on ARM and PPC platforms,
   running a range of devices including USB hubs.
 - supports suspend/resume of both the platform device and the root hub;
   supports remote wakeup of the root hub (but NOT the platform device)
   by USB devices.
 - does NOT support ISO transfers (nobody has asked for them).
 - is PIO-only.

Signed-off-by: Olav Kongas <ok@artecdesign.ee>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB: omap_udc updates (mostly cleanups)
David Brownell [Mon, 11 Apr 2005 22:38:25 +0000 (15:38 -0700)]
[PATCH] USB: omap_udc updates (mostly cleanups)

Various USB patches, mostly for portability:

  - Fifo mode 1 didn't work previously (oopsed), so now it's fixed and
    (why not) defines even more endpoints for composite devices.

  - OMAP 1710 doesn't have an internal transceiver.

  - Small PM update:  if the USB link is suspended, don't disconnect on
    entry to deep sleep.

  - Be more correct about handling zero length control reads.  OMAP
    seems to mis-handle that protocol peculiarity though; best avoided.

  - Platform device resources (for UDC and OTG controllers) now use
    physical addresses, so /proc/iomem is more consistent.

  - Minor cleanups, notably (by volume) for "sparse" NULL warnings.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
18 years ago[PATCH] USB: g_file_storage: export "stall" parameter
Alan Stern [Mon, 18 Apr 2005 16:43:25 +0000 (12:43 -0400)]
[PATCH] USB: g_file_storage: export "stall" parameter

This patch changes the g_file_storage driver to make the "stall" module
parameter generally available; currently it is available only if the
testing version of the module has been configured.  It also fixes a typo
in a comment -- thanks, Pat!

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] USB: g_file_storage: Consolidate min()s
Alan Stern [Tue, 5 Apr 2005 15:56:54 +0000 (11:56 -0400)]
[PATCH] USB: g_file_storage: Consolidate min()s

This patch simplifies the g_file_storage driver by consolidating a bunch
of min() calculations at a single spot.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
18 years ago[PATCH] fix silly config option.
Dave Jones [Mon, 27 Jun 2005 05:39:28 +0000 (01:39 -0400)]
[PATCH] fix silly config option.

CONFIG_CONFIG_TUNER_MULTI_I2C probably isn't what the
author meant to create.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] CFQ io scheduler: scheduler switch oops
Jens Axboe [Mon, 27 Jun 2005 12:49:39 +0000 (14:49 +0200)]
[PATCH] CFQ io scheduler: scheduler switch oops

If cfq is managing a queue and a new scheduler is later selected, it is
possible for the cfqd unplug_work work to be queued after the kblockd
work struct has been flushed.  The problem is the ordering of
cfq_shutdown_timer_wq() and blk_put_queue() in cfq_put_cfqd().  The
latter may rearm the work, leaving cfq_kick_queue() with dead data.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] CFQ io scheduler, add ioprio documentation
Jens Axboe [Mon, 27 Jun 2005 08:56:58 +0000 (10:56 +0200)]
[PATCH] CFQ io scheduler, add ioprio documentation

Add ioprio documentation

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] CFQ io scheduler updates
Jens Axboe [Mon, 27 Jun 2005 08:56:24 +0000 (10:56 +0200)]
[PATCH] CFQ io scheduler updates

- Adjust slice values

- Instead of one async queue, one is defined per priority level. This
  prevents kernel threads (such as reiserfs/x and others) that run at
  higher io priority from conflicting with others. Previously, it was a
  coin toss what io prio the async queue got, it was defined by who
  first set up the queue.

- Let a time slice only begin, when the previous slice is completely
  done. Previously we could be somewhat unfair to a new sync slice, if
  the previous slice was async and had several ios queued. This might
  need a little tweaking if throughput suffers a little due to this,
  allowing perhaps an overlap of a single request or so.

- Optimize the calling of kblockd_schedule_work() by doing it only when
  it is strictly necessary (no requests in driver and work left to do).

- Correct sync vs async logic. A 'normal' process can be purely async as
  well, and a flusher can be purely sync as well. Sync or async is now a
  property of the class defined and requests pending. Previously writers
  could be considered sync, when they were really async.

- Get rid of the bit fields in cfqq and crq, use flags instead.

- Various other cleanups and fixes

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Fix cfq_find_next_crq()
Jens Axboe [Mon, 27 Jun 2005 08:55:49 +0000 (10:55 +0200)]
[PATCH] Fix cfq_find_next_crq()

In cfq_find_next_crq(), cfq tries to find the next request by choosing
one of two requests before and after the current one.  Currently, when
choosing the next request, if there's no next request, the next
candidate is NULL, resulting in selection of the previous request.  This
results in weird scheduling.  Once we reach the end, we always seek
backward.

The correct behavior is using the first request as the next candidate.
cfq_choose_req() already has logics for handling wrapped requests.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] Update cfq io scheduler to time sliced design
Jens Axboe [Mon, 27 Jun 2005 08:55:12 +0000 (10:55 +0200)]
[PATCH] Update cfq io scheduler to time sliced design

This updates the CFQ io scheduler to the new time sliced design (cfq
v3).  It provides full process fairness, while giving excellent
aggregate system throughput even for many competing processes.  It
supports io priorities, either inherited from the cpu nice value or set
directly with the ioprio_get/set syscalls.  The latter closely mimic
set/getpriority.

This import is based on my latest from -mm.

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
18 years ago[PATCH] ARM: 2759/1: Fix IXP4xx debug code (again)
Deepak Saxena [Mon, 27 Jun 2005 20:48:48 +0000 (21:48 +0100)]
[PATCH] ARM: 2759/1: Fix IXP4xx debug code (again)

Patch from Deepak Saxena

Accidently swapped the order of movne and orrne. Bad.

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: 2758/1: Fix comment in file header to read "ARM" instead i386
Michael Burian [Mon, 27 Jun 2005 14:58:39 +0000 (15:58 +0100)]
[PATCH] ARM: 2758/1: Fix comment in file header to read "ARM" instead i386

Patch from Michael Burian

This does not look like an include file for "i386", so use "ARM" instead.

Signed-off-by: Michael Burian <dynmail1@gassner-waagen.at>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Update mach-types
Russell King [Mon, 27 Jun 2005 13:49:10 +0000 (14:49 +0100)]
[PATCH] ARM: Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Fix speeling eroor in io.h
Russell King [Mon, 27 Jun 2005 13:23:38 +0000 (14:23 +0100)]
[PATCH] ARM: Fix speeling eroor in io.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Move memmap freeing into init.c
Russell King [Mon, 27 Jun 2005 13:16:47 +0000 (14:16 +0100)]
[PATCH] ARM: Move memmap freeing into init.c

It doesn't make sense for this to be in mm-armv.c now that 26-bit
ARM support is no longer integrated into arch/arm.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Move PGD kernel page table initialisation
Russell King [Mon, 27 Jun 2005 13:08:56 +0000 (14:08 +0100)]
[PATCH] ARM: Move PGD kernel page table initialisation

It doesn't make sense to have the PGD kernel pointers initialisation
separate from the PGD user pointers, especially when we clean the
data cache over the whole range.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Add VST idle loop call
Russell King [Mon, 27 Jun 2005 13:04:05 +0000 (14:04 +0100)]
[PATCH] ARM: Add VST idle loop call

This call allows the dynamic tick support to reprogram the timer
immediately before the CPU idles.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] ARM: Add missed AAEC2000 file
Russell King [Mon, 27 Jun 2005 12:59:43 +0000 (13:59 +0100)]
[PATCH] ARM: Add missed AAEC2000 file

My scripts missed committing this file.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] Serial: Split 8250 port table
Russell King [Mon, 27 Jun 2005 10:12:54 +0000 (11:12 +0100)]
[PATCH] Serial: Split 8250 port table

Add separate files for the different 8250 ISA-based serial boards.

Looking across all the various architectures, it seems reasonable that
we can key the availability of the configuration options for these
beasts to the bus-related symbols (iow, CONFIG_ISA).  We also standardise
the base baud/uart clock rate for these boards - I'm sure that isn't
architecture specific, but is solely dependent on the crystal fitted
on the board (which should be the same no matter what type of machine
its fitted into.)

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] Serial: Remove {un,}register_serial from 68328serial.c
Russell King [Mon, 27 Jun 2005 09:29:26 +0000 (10:29 +0100)]
[PATCH] Serial: Remove {un,}register_serial from 68328serial.c

68328serial.c does not make use of register_serial/unregister_serial,
which is traditionally used to register 8250-compatible UARTs with
the 8250-compatible serial driver.

Acked-by: David McCullough
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
18 years ago[PATCH] sound/oss/via82cxxx_audio: Use the DMA_32BIT_MASK constant
Tobias Klauser [Sat, 28 May 2005 23:17:29 +0000 (01:17 +0200)]
[PATCH] sound/oss/via82cxxx_audio: Use the DMA_32BIT_MASK constant

Use the DMA_32BIT_MASK constant from dma-mapping.h
when calling pci_set_dma_mask() or pci_set_consistent_dma_mask()
See http://marc.theaimsgroup.com/?t=108001993000001&r=1&w=2 for details

Signed-off-by: Tobias Klauser <tklauser@nuerscht.ch>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] libata fix read capacity handling for more than 2TB
Philip Pokorny [Sat, 28 May 2005 08:24:47 +0000 (01:24 -0700)]
[PATCH] libata fix read capacity handling for more than 2TB

This is a multi-part message in MIME format.

18 years ago[PATCH] b44: check link state during open
John W. Linville [Wed, 8 Jun 2005 19:11:57 +0000 (15:11 -0400)]
[PATCH] b44: check link state during open

Check the link state during b44_open.  This closes a 1 HZ window
that existed after b44_open ran but before the b44_timer handler ran,
during which ethtool would report "Link detected: yes" no matter what
the link state actually was.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
18 years ago[PATCH] fix tulip suspend/resume #2
Adam Belay [Mon, 20 Jun 2005 21:28:41 +0000 (14:28 -0700)]
[PATCH] fix tulip suspend/resume #2

This patch allows the tulip driver to suspend and resume properly.  It was
originally written by Karsten Keil and then modified by Adam Belay.

Signed-off-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Adam Belay <abelay@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] uninitialized variable in prism54 isl38xx_trigger_device
Olaf Hering [Wed, 25 May 2005 23:16:51 +0000 (01:16 +0200)]
[PATCH] uninitialized variable in prism54 isl38xx_trigger_device

drivers/net/wireless/prism54/isl_38xx.c:131: warning: 'current_time.tv_sec' is used uninitialized in this function
drivers/net/wireless/prism54/isl_38xx.c:131: warning: 'current_time.tv_usec' is used uninitialized in this function

Signed-off-by: Olaf Hering <olh@suse.de>
18 years ago[PATCH] NETDEV: fix receiving multicast frames.
YOSHIFUJI Hideaki [Wed, 25 May 2005 07:07:04 +0000 (16:07 +0900)]
[PATCH] NETDEV: fix receiving multicast frames.

Some USB ethernet drivers did not accept multicast frames appropriately.
IPv6 did not work with those drivers without this patch.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
18 years ago[PATCH] NETDEV: Elecom (Laneed) LD-USBL/TX support.
YOSHIFUJI Hideaki [Wed, 25 May 2005 07:06:59 +0000 (16:06 +0900)]
[PATCH] NETDEV: Elecom (Laneed) LD-USBL/TX support.

Elecom (Laneed) LD-USBL/TX support.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
18 years ago[PATCH] 3c523: needs a license
Randy Dunlap [Mon, 13 Jun 2005 19:31:53 +0000 (12:31 -0700)]
[PATCH] 3c523: needs a license

Module needs a license to prevent kernel tainting.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
diffstat:=
 drivers/net/3c523.c |    1 +
 1 files changed, 1 insertion(+)

18 years ago[PATCH] gianfar: Add support enhanced TSEC features on the MPC 8548
Kumar Gala [Mon, 20 Jun 2005 15:54:21 +0000 (10:54 -0500)]
[PATCH] gianfar: Add support enhanced TSEC features on the MPC 8548

Jeff,

Just incase this got lost in the recent netdev mailing list transition
here is a nicer version of Andy's patch for gianfar.

- kumar

* TCP/IP/UDP checksumming and verification
* VLAN tag insertion/extraction
* Larger multicast hash-table
* Padding to align IP headers

Also added:
* msg lvl support
* Some whitespace cleanup

Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
18 years ago[PATCH] smc91x: plug race between TX tasklet and driver reset
Nicolas Pitre [Mon, 20 Jun 2005 03:56:21 +0000 (23:56 -0400)]
[PATCH] smc91x: plug race between TX tasklet and driver reset

The race causes a kernel oops when smc_hardware_send_pkt() tries to
dereference pending_tx_skb which would have been freed from one of the
driver reset paths just after the tx_task tasklet has been scheduled.
This race is possible on SMP but was uncovered by the kernel RT work.

Signed-off-by: Nicolas Pitre <nico@cam.org>
18 years ago[PATCH] via-rhine trivial whitespace patch
Denis Vlasenko [Fri, 17 Jun 2005 05:23:17 +0000 (08:23 +0300)]
[PATCH] via-rhine trivial whitespace patch

--Boundary-00=_F5lsC5eH1wGW5o9
Content-Type: text/plain;
  charset="koi8-r"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Hi Jeff,

In some messages in via-rhine.c there is a leading space
for no apparent reason. This patch removes it.
--
vda

--Boundary-00=_F5lsC5eH1wGW5o9
Content-Type: text/x-diff;
  charset="koi8-r";
  name="via-rhine.c.diff"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="via-rhine.c.diff"

18 years ago[PATCH] e1000: Driver version, white space, comments & other
Malli Chilakala [Sat, 18 Jun 2005 00:46:06 +0000 (17:46 -0700)]
[PATCH] e1000: Driver version, white space, comments & other

Driver version, white space, comments & other

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Included proposals to false late collisions due to H/W latencies
Malli Chilakala [Sat, 18 Jun 2005 00:44:49 +0000 (17:44 -0700)]
[PATCH] e1000: Included proposals to false late collisions due to H/W latencies

Included proposals to false late collisions due to H/W latencies

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Ethtool set speed/duplex validates parameters for consistency
Malli Chilakala [Sat, 18 Jun 2005 00:44:20 +0000 (17:44 -0700)]
[PATCH] e1000: Ethtool set speed/duplex validates parameters for consistency

Ethtool set speed/duplex validates parameters for consistency

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Enable ethtool phys_id feature for 82573 controllers
Malli Chilakala [Sat, 18 Jun 2005 00:43:56 +0000 (17:43 -0700)]
[PATCH] e1000: Enable ethtool phys_id feature for 82573 controllers

Enable ethtool phys_id feature for 82573 controllers

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Fixed VLAN tag processing error for big-endian architectures
Malli Chilakala [Sat, 18 Jun 2005 00:43:25 +0000 (17:43 -0700)]
[PATCH] e1000: Fixed VLAN tag processing error for big-endian architectures

Fixed VLAN tag processing error for big-endian architectures

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Ethtool cleanup patch from Stephen Hemminger
Malli Chilakala [Sat, 18 Jun 2005 00:43:06 +0000 (17:43 -0700)]
[PATCH] e1000: Ethtool cleanup patch from Stephen Hemminger

Ethtool cleanup patch from Stephen Hemminger
   * use ADVERTISED_xxx fields when setting advertised fields
   * don't hardcode constant for advertised field

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Fixed the loopback test failure for 82573 based adapters
Malli Chilakala [Sat, 18 Jun 2005 00:42:42 +0000 (17:42 -0700)]
[PATCH] e1000: Fixed the loopback test failure for 82573 based adapters

Fixed the loopback test failure for 82573 based adapters

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Fixed register and loopback test failures with 82573 controllers
Malli Chilakala [Sat, 18 Jun 2005 00:42:29 +0000 (17:42 -0700)]
[PATCH] e1000: Fixed register and loopback test failures with 82573 controllers

Fixed register and loopback test failures with 82573 controllers

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Cleanup debug message printed when Tx Unit hang is detected
Malli Chilakala [Sat, 18 Jun 2005 00:42:07 +0000 (17:42 -0700)]
[PATCH] e1000: Cleanup debug message printed when Tx Unit hang is detected

Cleanup debug message printed when Tx Unit hang is detected

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Use netdev_priv() to get to netdev->priv
Malli Chilakala [Sat, 18 Jun 2005 00:41:45 +0000 (17:41 -0700)]
[PATCH] e1000: Use netdev_priv() to get to netdev->priv

Use netdev_priv() to get to netdev->priv - from shemminger@osdl.org

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Use correct WOL settings for 82544 adapters
Malli Chilakala [Sat, 18 Jun 2005 00:40:41 +0000 (17:40 -0700)]
[PATCH] e1000: Use correct WOL settings for 82544 adapters

Use correct WOL settings for 82544 adapters

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e1000: Synchronize phy access between stats update and MII_IOCTL
Malli Chilakala [Sat, 18 Jun 2005 00:40:19 +0000 (17:40 -0700)]
[PATCH] e1000: Synchronize phy access between stats update and MII_IOCTL

Synchronize phy access between stats update and MII_IOCTL

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e100: NAPI performance enhancements
Malli Chilakala [Sat, 18 Jun 2005 00:31:44 +0000 (17:31 -0700)]
[PATCH] e100: NAPI performance enhancements

NAPI performance enhancements - Fixed issues with shared interrupts and NAPI resulting in bad performance.

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e100: e100_eeprom_load was called after e100_phy_init
Malli Chilakala [Sat, 18 Jun 2005 00:30:22 +0000 (17:30 -0700)]
[PATCH] e100: e100_eeprom_load was called after e100_phy_init

e100_eeprom_load was called after e100_phy_init causing phy_init
not to use values set in EEPROM - from emann@mrv.com

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years ago[PATCH] e100: fixed e100 MDI/MDI-X issues
Malli Chilakala [Sat, 18 Jun 2005 00:27:41 +0000 (17:27 -0700)]
[PATCH] e100: fixed e100 MDI/MDI-X issues

Added patch from Eran Mann to fix following e100 MDI/MDI-X issues
 * MDI/MDI-X autodetection should never be enabled for 82551ER/QM chips
 * enabling this feature based on eeprom settings

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Ganesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: John Ronciak <john.ronciak@intel.com>
18 years agoMerge /spare/repo/netdev-2.6/ branch 'orinoco'
Jeff Garzik [Mon, 27 Jun 2005 04:27:47 +0000 (00:27 -0400)]
Merge /spare/repo/netdev-2.6/ branch 'orinoco'

18 years agowireless/orinoco: remove changelog, bump version
Jeff Garzik [Mon, 27 Jun 2005 04:27:07 +0000 (00:27 -0400)]
wireless/orinoco: remove changelog, bump version

18 years ago[PATCH] orinoco: monitor mode support
Christoph Hellwig [Sat, 18 Jun 2005 23:28:06 +0000 (01:28 +0200)]
[PATCH] orinoco: monitor mode support

Patch from Pavel Roskin

18 years ago[PATCH] orinoco: always use 802.11 header for rx processing
Christoph Hellwig [Sat, 18 Jun 2005 23:28:02 +0000 (01:28 +0200)]
[PATCH] orinoco: always use 802.11 header for rx processing

If the frame has ToDS flag set, mark it by setting skb->pkt_type to
PACKET_OTHERHOST, so that applications unaware of promiscous mode won't get
uplink (STA->AP) packets for STA->STA transmissions relayed by the AP.
Thanks to John Denker and David Gibson for finding the problem and the
solution.

Patch from Pavel Roskin

18 years ago[PATCH] orinoco: scanning support
Christoph Hellwig [Sat, 18 Jun 2005 23:27:56 +0000 (01:27 +0200)]
[PATCH] orinoco: scanning support

Patch from Pavel Roskin

18 years ago[PATCH] orinoco: manual roaming for Symbol and Intersilfirmware
Christoph Hellwig [Sat, 18 Jun 2005 23:27:51 +0000 (01:27 +0200)]
[PATCH] orinoco: manual roaming for Symbol and Intersilfirmware

Patch from Pavel Roskin

18 years ago[PATCH] orinoco: basic ethtool support
Christoph Hellwig [Sat, 18 Jun 2005 23:27:40 +0000 (01:27 +0200)]
[PATCH] orinoco: basic ethtool support

I completely reimplemented this based on ethtool_ops, CVS has an
ioctl-based version.

18 years ago[PATCH] orinoco: wireless API 15 support
Christoph Hellwig [Sat, 18 Jun 2005 23:27:33 +0000 (01:27 +0200)]
[PATCH] orinoco: wireless API 15 support

(patch from Moustafa Youssef, updated by Jim Carter and Pavel Roskin).

18 years ago[PATCH] orinoco: include <net/ieee80211.h>
Christoph Hellwig [Sat, 18 Jun 2005 23:27:28 +0000 (01:27 +0200)]
[PATCH] orinoco: include <net/ieee80211.h>

We need constants from this header in the next patches.

18 years ago[PATCH] bring over ieee80211.h from mainline
Christoph Hellwig [Sat, 18 Jun 2005 23:27:20 +0000 (01:27 +0200)]
[PATCH] bring over ieee80211.h from mainline

the prototypes and inlines aren't actually needed, but let's not diverge
from -mm too far.

18 years ago[PATCH] cs89x0.c: support for Philips' pnx0105 network adapter
dmitry pervushin [Mon, 20 Jun 2005 22:32:54 +0000 (15:32 -0700)]
[PATCH] cs89x0.c: support for Philips' pnx0105 network adapter

This patch is to provide support for cs89x0-based network device on
Philips' pnx0105 board.

Signed-off-by: dmitry pervushin <dpervushin@ru.mvista.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] drivers/net/skfp/: cleanups
Adrian Bunk [Tue, 21 Jun 2005 08:47:06 +0000 (01:47 -0700)]
[PATCH] drivers/net/skfp/: cleanups

This patch contains the following cleanups:
- make needlessly global code static
- remove the completely unused smtparse.c
- remove the following unused global functions:
  - drvfbi.c: init_dma
  - drvfbi.c: dis_dma
  - drvfbi.c: get_rom_byte
  - drvfbi.c: mac_drv_vpd_read
  - drvfbi.c: mac_drv_pci_fix
  - fplustm.c: mac_set_func_addr
  - fplustm.c: mac_del_multicast
  - hwmtm.c: mac_drv_rx_frag
  - pcmplc.c: pcm_set_lct_short
  - smt.c: smt_please_reconnect
  - smt.c: smt_change_t_neg
  - smtdef.c: smt_set_defaults

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[netdrvr] tulip: add pci id
Hideki Yamane [Mon, 27 Jun 2005 04:18:32 +0000 (00:18 -0400)]
[netdrvr] tulip: add pci id

18 years ago[PATCH] forcedeth: Add support for new device id
Manfred Spraul [Sun, 26 Jun 2005 09:36:52 +0000 (11:36 +0200)]
[PATCH] forcedeth: Add support for new device id

This is a multi-part message in MIME format.

18 years ago[PATCH] forcedeth: Poll for link changes
Manfred Spraul [Sat, 18 Jun 2005 14:27:19 +0000 (16:27 +0200)]
[PATCH] forcedeth: Poll for link changes

This is a multi-part message in MIME format.

18 years ago[PATCH] forcedeth: add two new pci ids
Manfred Spraul [Sun, 5 Jun 2005 15:36:11 +0000 (17:36 +0200)]
[PATCH] forcedeth: add two new pci ids

This is a multi-part message in MIME format.

18 years ago[PATCH] drivers/block/sx8.c: Use the DMA_{64, 32}BIT_MASK constants
Tobias Klauser [Mon, 20 Jun 2005 21:49:08 +0000 (23:49 +0200)]
[PATCH] drivers/block/sx8.c: Use the DMA_{64, 32}BIT_MASK constants

Use the DMA_{64,32}BIT_MASK constants from dma-mapping.h when calling
pci_set_dma_mask() or pci_set_consistent_dma_mask()
These patches include dma-mapping.h explicitly because it caused errors
on some architectures otherwise.
See http://marc.theaimsgroup.com/?t=108001993000001&r=1&w=2 for details

Signed-off-by: Tobias Klauser <tklauser@nuerscht.ch>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] net/lanstreamer: replace schedule_timeout() with ssleep()/msleep_interruptible()
Nishanth Aravamudan [Mon, 20 Jun 2005 21:53:40 +0000 (23:53 +0200)]
[PATCH] net/lanstreamer: replace schedule_timeout() with ssleep()/msleep_interruptible()

Use ssleep() / msleep_interruptible() [as appropriate]
instead of schedule_timeout() to guarantee the task delays as expected.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] net/pcnet32: replace schedule_timeout() with msleep_interruptible()
Nishanth Aravamudan [Mon, 20 Jun 2005 21:54:12 +0000 (23:54 +0200)]
[PATCH] net/pcnet32: replace schedule_timeout() with msleep_interruptible()

Use msleep_interruptible() instead of schedule_timeout() to
guarantee the task delays as expected.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] net/farsync: add set_current_state() before schedule_timeout()
Nishanth Aravamudan [Mon, 20 Jun 2005 21:54:25 +0000 (23:54 +0200)]
[PATCH] net/farsync: add set_current_state() before schedule_timeout()

Insert set_current_state() before schedule_timeout() so the
function delays as expected. Without the addition, schedule_timeout()
will return immediately.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Maximilian Attems <janitor@sternwelten.at>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] net/sb1000: replace nicedelay() with ssleep()
Nishanth Aravamudan [Mon, 20 Jun 2005 21:54:27 +0000 (23:54 +0200)]
[PATCH] net/sb1000: replace nicedelay() with ssleep()

Use ssleep() instead of nicedelay()
to guarantee the task delays as expected. Remove the prototype and
definition of nicedelay(). This is a very weird function, because it is
called to sleep in terms of usecs, but always sleeps for 1 second,
completely ignoring the parameter. I have gone ahead and followed suit,
just sleeping for a second in all cases, but maybe someone with the
hardware could tell me if perhaps the paramter *should* matter. Additionally,
nicedelay() is called in TASK_INTERRUPTIBLE state, but doesn't deal with signals
in case these longer delays do not complete, so I believe ssleep() is more
appropriate.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[PATCH] wireless: char* -> char[] conversion in airo.c
Domen Puncer [Mon, 20 Jun 2005 21:54:37 +0000 (23:54 +0200)]
[PATCH] wireless: char* -> char[] conversion in airo.c

This conversion makes code from line 7101 right:
if (copy_to_user(com.data, swversion, sizeof(swversion)))

size output (before, after):
  55416    2228     160   57804    e1cc drivers/net/wireless/airo.o
  55412    2228     160   57800    e1c8 drivers/net/wireless/airo.o

more outputs from Alexey Dobriyan:
2.95.3:
           text    data     bss     dec     hex filename
before    51118    2156     160   53434    d0ba drivers/net/wireless/airo.o
after     51118    2156     160   53434    d0ba drivers/net/wireless/airo.o

3.3.5-20050130:
before    46999    2156     160   49315    c0a3 drivers/net/wireless/airo.o
after     46994    2156     160   49310    c09e drivers/net/wireless/airo.o

4.1.0-20050522:
before    45555    2220     160   47935    bb3f drivers/net/wireless/airo.o
after     45550    2220     160   47930    bb3a drivers/net/wireless/airo.o

Signed-off-by: Domen Puncer <domen@coderock.org>
18 years agoMerge /spare/repo/linux-2.6/
Jeff Garzik [Mon, 27 Jun 2005 03:38:58 +0000 (23:38 -0400)]
Merge /spare/repo/linux-2.6/

18 years agoMerge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Mon, 27 Jun 2005 02:16:33 +0000 (19:16 -0700)]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

18 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Mon, 27 Jun 2005 00:28:24 +0000 (17:28 -0700)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm

18 years ago[IPVS]: Fix for overflows
pageexec [Sun, 26 Jun 2005 23:00:19 +0000 (16:00 -0700)]
[IPVS]: Fix for overflows

From: <pageexec@freemail.hu>

$subject was fixed in 2.4 already, 2.6 needs it as well.

The impact of the bugs is a kernel stack overflow and privilege escalation
from CAP_NET_ADMIN via the IP_VS_SO_SET_STARTDAEMON/IP_VS_SO_GET_DAEMON
ioctls.  People running with 'root=all caps' (i.e., most users) are not
really affected (there's nothing to escalate), but SELinux and similar
users should take it seriously if they grant CAP_NET_ADMIN to other users.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PATCH] DM9000 network driver bugfix
Sascha Hauer [Mon, 20 Jun 2005 22:32:51 +0000 (15:32 -0700)]
[PATCH] DM9000 network driver bugfix

This patch fixes two bugs in the dm9000 network driver:

- Don't read one byte too much in 8bit mode.
- release correct resource

Signed-off-by: Jochen Karrer <j.karrer@lightmaze.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] Documentation/networking/dmfe.txt: Make documentation nicer
Ismail Donmez [Mon, 20 Jun 2005 22:32:52 +0000 (15:32 -0700)]
[PATCH] Documentation/networking/dmfe.txt: Make documentation nicer

Patch indents dmfe.txt to look like other docs.  It adds a tip about CNET
cards using Davicom chipsets.  Also it removes parts where it refers to how
to build driver out-of-kernel which seems to be cruft from times where the
driver was out of the kernel.

Signed-off-by: Ismail Donmez <ismail@kde.org.tr>
Signed-off-by: Domen Puncer <domen@coderock.org>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] fealnx.c calls dev_kfree_skb from atomic context
Denis Vlasenko [Mon, 20 Jun 2005 22:33:04 +0000 (15:33 -0700)]
[PATCH] fealnx.c calls dev_kfree_skb from atomic context

Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] fix int vs. pm_message_t confusion in airo
Pavel Machek [Mon, 20 Jun 2005 22:33:04 +0000 (15:33 -0700)]
[PATCH] fix int vs. pm_message_t confusion in airo

Fix int vs. pm_message_t confusion in airo. Should change no code.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] defxx: Use irqreturn_t for the interrupt handler
Maciej W. Rozycki [Mon, 20 Jun 2005 22:33:03 +0000 (15:33 -0700)]
[PATCH] defxx: Use irqreturn_t for the interrupt handler

This is a fix for the interrupt handler in the defxx driver to use
irqreturn_t.  Beside the obvious fix of returning a proper status at all,
it actually checks board registers as appropriate for determining if an
interrupt has been recorded in the bus-specific interface logic.

The patch also includes an obvious one-line fix for SET_NETDEV_DEV needed
for the EISA variation, for which I've decided there is no point in sending
separately.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] dmfe warning fix
Andrew Morton [Mon, 20 Jun 2005 22:32:59 +0000 (15:32 -0700)]
[PATCH] dmfe warning fix

drivers/net/tulip/dmfe.c: In function `dmfe_parse_srom':
drivers/net/tulip/dmfe.c:1805: warning: passing arg 1 of `__le16_to_cpup' from incompatible pointer type
drivers/net/tulip/dmfe.c:1817: warning: passing arg 1 of `__le32_to_cpup' from incompatible pointer type
drivers/net/tulip/dmfe.c:1817: warning: passing arg 1 of `__le32_to_cpup' from incompatible pointer type

This is basically a guess:

Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years ago[PATCH] document that 8139TOO supports 8129/8130
Adrian Bunk [Mon, 16 May 2005 19:13:03 +0000 (21:13 +0200)]
[PATCH] document that 8139TOO supports 8129/8130

The 8129/8130 support is a sub-option that is not visible if the user
hasn't enabled the 8139 support.

Let's make it a bit easier for users to find the driver for their nic.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
18 years ago[NETLINK]: Fix two socket hashing bugs.
David S. Miller [Sun, 26 Jun 2005 22:31:51 +0000 (15:31 -0700)]
[NETLINK]: Fix two socket hashing bugs.

1) netlink_release() should only decrement the hash entry
   count if the socket was actually hashed.

   This was causing hash->entries to underflow, which
   resulting in all kinds of troubles.

   On 64-bit systems, this would cause the following
   conditional to erroneously trigger:

err = -ENOMEM;
if (BITS_PER_LONG > 32 && unlikely(hash->entries >= UINT_MAX))
goto err;

2) netlink_autobind() needs to propagate the error return from
   netlink_insert().  Otherwise, callers will not see the error
   as they should and thus try to operate on a socket with a zero pid,
   which is very bad.

   However, it should not propagate -EBUSY.  If two threads race
   to autobind the socket, that is fine.  This is consistent with the
   autobind behavior in other protocols.

   So bug #1 above, combined with this one, resulted in hangs
   on netlink_sendmsg() calls to the rtnetlink socket.  We'd try
   to do the user sendmsg() with the socket's pid set to zero,
   later we do a socket lookup using that pid (via the value we
   stashed away in NETLINK_CB(skb).pid), but that won't give us the
   user socket, it will give us the rtnetlink socket.  So when we
   try to wake up the receive queue, we dive back into rtnetlink_rcv()
   which tries to recursively take the rtnetlink semaphore.

Thanks to Jakub Jelink for providing backtraces.  Also, thanks to
Herbert Xu for supplying debugging patches to help track this down,
and also finding a mistake in an earlier version of this fix.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PATCH] net/slip: replace schedule_timeout() with msleep_interruptible()
Nishanth Aravamudan [Mon, 2 May 2005 06:34:57 +0000 (23:34 -0700)]
[PATCH] net/slip: replace schedule_timeout() with msleep_interruptible()

Use msleep_interruptible() instead of schedule_timeout() to guarantee
the task delays as expected.

18 years ago[PATCH] drivers/net/ewrk3.c: remove dead code
Adrian Bunk [Mon, 2 May 2005 01:47:00 +0000 (03:47 +0200)]
[PATCH] drivers/net/ewrk3.c: remove dead code

This patch removes some obviously dead code found by the Coverity
checker.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
18 years ago[PATCH] drivers/net/skfp/: fix LITTLE_ENDIAN
Adrian Bunk [Mon, 2 May 2005 01:46:52 +0000 (03:46 +0200)]
[PATCH] drivers/net/skfp/: fix LITTLE_ENDIAN

This patch fixes the LITTLE_ENDIAN #define and a function prototype.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
18 years ago[PATCH] drivers/net/tokenring/: cleanups
Adrian Bunk [Mon, 2 May 2005 01:46:43 +0000 (03:46 +0200)]
[PATCH] drivers/net/tokenring/: cleanups

This patch contains the follwing cleanups:
- make needlessly global code static
- remove obsolete Emacs settings

Signed-off-by: Adrian Bunk <bunk@stusta.de>
18 years ago[ATALK]: Include asm/byteorder.h in linux/atalk.h
David S. Miller [Sun, 26 Jun 2005 22:28:10 +0000 (15:28 -0700)]
[ATALK]: Include asm/byteorder.h in linux/atalk.h

We're using __be16 in userland visible types, so we
have to include asm/byteorder.h so that works.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PKTGEN]: Fix random packet sizes causing panic
Robert Olsson [Sun, 26 Jun 2005 22:27:10 +0000 (15:27 -0700)]
[PKTGEN]: Fix random packet sizes causing panic

Signed-off-by: Robert Olsson <robert.olsson@its.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PATCH] pcnet_cs.c: IRQ handler optimization
Andreas Mohr [Mon, 11 Apr 2005 23:47:43 +0000 (16:47 -0700)]
[PATCH] pcnet_cs.c: IRQ handler optimization

During some performance diagnostics I stumbled on this slightly wasteful
code in pcnet_cs.c which I made the patch included at the bottom for (two
minor comment fixes included).

Improvement:
instead of *always* calculating
lea    0x2c0(%edx),%ebx
and then additionally doing the
mov    %edx,0xc0(%ebx)
addition *if we need it*,

we now do the *whole* calculation of
mov    %edx,0x380(%ebx)
*only* if we need it.
This even manages to save us a whole 16-byte alignment buffer loss
in this compilation case.

Result: slightly improves IRQ handler performance in both shared and
non-shared IRQ case, which should make my rusty P3/700 a slight bit happier.

Thank you for your support,

Andreas Mohr

old asm result (using gcc 3.3.5):

000015a0 <ei_irq_wrapper>:
    15a0:       55                      push   %ebp
    15a1:       89 e5                   mov    %esp,%ebp
    15a3:       53                      push   %ebx
    15a4:       8d 9a c0 02 00 00       lea    0x2c0(%edx),%ebx
    15aa:       e8 fc ff ff ff          call   15ab <ei_irq_wrapper+0xb>
    15af:       83 f8 01                cmp    $0x1,%eax
    15b2:       74 03                   je     15b7 <ei_irq_wrapper+0x17>
    15b4:       5b                      pop    %ebx
    15b5:       5d                      pop    %ebp
    15b6:       c3                      ret
    15b7:       31 d2                   xor    %edx,%edx
    15b9:       89 93 c0 00 00 00       mov    %edx,0xc0(%ebx)
    15bf:       eb f3                   jmp    15b4 <ei_irq_wrapper+0x14>
    15c1:       eb 0d                   jmp    15d0 <ei_watchdog>
    15c3:       90                      nop
    15c4:       90                      nop
    15c5:       90                      nop
    15c6:       90                      nop
    15c7:       90                      nop
    15c8:       90                      nop
    15c9:       90                      nop
    15ca:       90                      nop
    15cb:       90                      nop
    15cc:       90                      nop
    15cd:       90                      nop
    15ce:       90                      nop
    15cf:       90                      nop

000015d0 <ei_watchdog>:

new asm result:

000015a0 <ei_irq_wrapper>:
    15a0:       55                      push   %ebp
    15a1:       89 e5                   mov    %esp,%ebp
    15a3:       53                      push   %ebx
    15a4:       89 d3                   mov    %edx,%ebx
    15a6:       e8 fc ff ff ff          call   15a7 <ei_irq_wrapper+0x7>
    15ab:       83 f8 01                cmp    $0x1,%eax
    15ae:       74 03                   je     15b3 <ei_irq_wrapper+0x13>
    15b0:       5b                      pop    %ebx
    15b1:       5d                      pop    %ebp
    15b2:       c3                      ret
    15b3:       31 d2                   xor    %edx,%edx
    15b5:       89 93 80 03 00 00       mov    %edx,0x380(%ebx)
    15bb:       eb f3                   jmp    15b0 <ei_irq_wrapper+0x10>
    15bd:       8d 76 00                lea    0x0(%esi),%esi

000015c0 <ei_watchdog>:

Signed-off-by: Andrew Morton <akpm@osdl.org>
18 years agodrivers/net/: Use the DMA_{64,32}BIT_MASK constants
Domen Puncer [Sun, 26 Jun 2005 22:22:14 +0000 (18:22 -0400)]
drivers/net/: Use the DMA_{64,32}BIT_MASK constants

Use the DMA_{64,32}BIT_MASK constants from dma-mapping.h when calling
pci_set_dma_mask() or pci_set_consistent_dma_mask()

This patch includes dma-mapping.h explicitly because it caused errors
on some architectures otherwise.

See http://marc.theaimsgroup.com/?t=108001993000001&r=1&w=2 for details

Signed-off-by: Tobias Klauser <tklauser@nuerscht.ch>
Signed-off-by: Domen Puncer <domen@coderock.org>
18 years ago[TCP]: Let TCP_CONG_ADVANCED default to n
Adrian Bunk [Sun, 26 Jun 2005 22:21:15 +0000 (15:21 -0700)]
[TCP]: Let TCP_CONG_ADVANCED default to n

It doesn't seem to make much sense to let an "If unsure, say N." option
default to y.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[IPV4]: Fix thinko in TCP_CONG_BIC default.
David S. Miller [Sun, 26 Jun 2005 22:20:20 +0000 (15:20 -0700)]
[IPV4]: Fix thinko in TCP_CONG_BIC default.

Since it is tristate when we offer it as a choice, we should
definte it also as tristate when forcing it as the default.
Otherwise kconfig warns.

Signed-off-by: David S. Miller <davem@davemloft.net>
18 years ago[PATCH] arlan: module parameter fixes
Magnus Damm [Tue, 5 Apr 2005 18:05:50 +0000 (20:05 +0200)]
[PATCH] arlan: module parameter fixes

Make sure the code compiles with and without ARLAN_ENTRY_EXIT_DEBUGGING.
Only provide parameter descriptions when parameters are defined.
Remove "arlan_"-prefix to shape up built-in parameter names:

arlan.arlan_debug -> arlan.debug
arlan.arlan_EEPROM_bad -> arlan.EEPROM_bad
arlan.arlan_entry_and_exit_debug -> arlan.entry_and_exit_debug
arlan.arlan_entry_debug -> arlan.entry_debug
arlan.arlan_exit_debug -> arlan.exit_debug

Signed-off-by: Magnus Damm <damm@opensource.se>