pandora-kernel.git
11 years agoMerge branch 'for-linux-next' of git://people.freedesktop.org/~danvet/drm-intel into...
Dave Airlie [Fri, 10 May 2013 04:35:48 +0000 (14:35 +1000)]
Merge branch 'for-linux-next' of git://people.freedesktop.org/~danvet/drm-intel into drm-next

Daniel writes:
A few intel fixes for smaller issues and one revert for an sdv hack which
we've wanted to kill anyway. Plus two drm patches included for your
convenience, both regression fixers for mine own screw-ups.

+ both fixes for stolen mem handling.

* 'for-linux-next' of git://people.freedesktop.org/~danvet/drm-intel:
  drm/i915: clear the stolen fb before resuming
  Revert "drm/i915: Calculate correct stolen size for GEN7+"
  drm/i915: hsw: fix link training for eDP on port-A
  Revert "drm/i915: revert eDP bpp clamping code changes"
  drm: don't check modeset locks in panic handler
  drm/i915: Fix pipe enabled mask for pipe C in WM calculations
  drm/mm: fix dump table BUG
  drm/i915: Always normalize return timeout for wait_timeout_ioctl

11 years agodrm/i915: clear the stolen fb before resuming
Jani Nikula [Tue, 7 May 2013 15:54:05 +0000 (18:54 +0300)]
drm/i915: clear the stolen fb before resuming

Similar to
commit 88afe715dd5469bc24ca7a19ac62dd3c241cab48
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Sun Dec 16 12:15:41 2012 +0000

    drm/i915: Clear the stolen fb before enabling

but on the resume path.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=57191
Reported-and-tested-by: Nikolay Amiantov <nikoamia@gmail.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org (3.9 only)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agoRevert "drm/i915: Calculate correct stolen size for GEN7+"
Ben Widawsky [Wed, 1 May 2013 18:00:34 +0000 (11:00 -0700)]
Revert "drm/i915: Calculate correct stolen size for GEN7+"

This reverts commit 03752f5b7b77b95d83479885040950fba1250850.

This revert requires a bit of explanation on how I understand things
work. Internally the architects/designers decide how the stolen encoding
works. We put it in a doc. BIOS writers take these docs and implement
it. Driver writers read the doc too, and read the value left by the BIOS
writers, and then we make magic.

The failing here is that in the docs we had[1] contained two different
definitions for this register for Gen7. (We have both a PCI register,
and an MMIO, and each of these were different). At the time [2] of
03752f5, we asked the architects what the correct value should be; but
that doesn't match the reality (BIOS) unfortunately.

So on all machines I can get my hands on, this revert is the right thing
to do. I've also worked with the product group to confirm that they
agree this revert is what we should do. People using HW made my "people"
who both write their own BIOS, and have access to our docs (Apple?).
Investigations are still ongoing about whether we need to add a list
of machines needing special handling, but this patch should be the
right thing for pretty much everyone.

[1] The docs are still wrong on this one. Now instead of two registers with
two definitions, we have one register with BOTH definitions, progress?
[2] The open source PRMs have the "wrong" definitions in chapter Volume
1 part6, section 1.1.12.

This digging was inspired by Paulo.

Cc: Paulo Zanoni <przanoni@gmail.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[danvet: Augment the patch saying that it's still a bit unclear
whether there are any machines out there with "wrong" firmware and
whether we need to add a list to handle them specially.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agodrm/i915: hsw: fix link training for eDP on port-A
Imre Deak [Fri, 3 May 2013 09:57:41 +0000 (12:57 +0300)]
drm/i915: hsw: fix link training for eDP on port-A

According to BSpec the link training sequence for eDP on HSW port-A
should be as follows:

1. link training: clock recovery
2. link training: equalization
3. link training: set idle transmission mode
4. display pipe enable
5. link training: disable (set normal mode)

Contrary to this at the moment we don't do step 3. and we do step 5.
before step 4. Fix this by setting idle transmission mode for eDP at
the end of intel_dp_complete_link_train and adding a new
intel_dp_stop_link_training function to disable link training. With
these changes we'll end up with the following functions corresponding
to the above steps:

intel_dp_start_link_train    -> step 1.
intel_dp_complete_link_train -> step 2., step 3.
intel_dp_stop_link_train     -> step 5.

For port-A we'll call intel_dp_stop_link_train only after enabling the
pipe, for everything else we'll call it right after
intel_dp_complete_link_train to preserve the current behavior.

Tested on HSW/HSW-ULT.

In v2:
- Due to a HW issue we must set idle transmission mode for port-A too
  before enabling the pipe. Thanks for Arthur Runyan for explaining
  this.
- Update the patch subject to make it clear that it's an eDP fix, DP is
  not affected.

v3:
- rename intel_dp_link_train() to intel_dp_set_link_train(), use 'val'
  instead 'l' as var name. (Paulo)

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agoRevert "drm/i915: revert eDP bpp clamping code changes"
Daniel Vetter [Sat, 4 May 2013 08:09:18 +0000 (10:09 +0200)]
Revert "drm/i915: revert eDP bpp clamping code changes"

This reverts commit 57c219633275c7e7413f8bc7be250dc092887458.

It's an ugly hack for a Haswell SDV platform where the vbt doesn't
seem to fully agree with the panel. Since it seems to cause issues on
real eDP platform let's just kill this hack again.

Reported-and-tested-by: Josh Boyer <jwboyer@gmail.com>
References: https://lkml.org/lkml/2013/5/3/467
Cc: Paulo Zanoni <przanoni@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agoqxl: update to new idr interfaces.
Dave Airlie [Fri, 3 May 2013 00:37:20 +0000 (10:37 +1000)]
qxl: update to new idr interfaces.

Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agoMerge branch 'server-fixes' into drm-next
Dave Airlie [Fri, 3 May 2013 00:13:08 +0000 (10:13 +1000)]
Merge branch 'server-fixes' into drm-next

Merge the fixes for the server driver dirty update paths
* server-fixes:
  drm/cirrus: deal with bo reserve fail in dirty update path
  drm/ast: deal with bo reserve fail in dirty update path
  drm/mgag200: deal with bo reserve fail in dirty update path

11 years agoMerge tag 'omapdss-for-3.10-fixes' of git://gitorious.org/linux-omap-dss2/linux into...
Dave Airlie [Fri, 3 May 2013 00:10:23 +0000 (10:10 +1000)]
Merge tag 'omapdss-for-3.10-fixes' of git://gitorious.org/linux-omap-dss2/linux into drm-next

OMAPDSS fixes for 3.10:

* Compilation fix when DSI is disabled in Kconfig
* Basic deferred probe support to fix DT boot

* tag 'omapdss-for-3.10-fixes' of git://gitorious.org/linux-omap-dss2/linux:
  OMAPDSS: TFP410: return EPROBE_DEFER if the i2c adapter not found
  OMAPDSS: VENC: Add error handling for venc_probe_pdata
  OMAPDSS: HDMI: Add error handling for hdmi_probe_pdata
  OMAPDSS: RFBI: Add error handling for rfbi_probe_pdata
  OMAPDSS: DSI: Add error handling for dsi_probe_pdata
  OMAPDSS: SDI: Add error handling for sdi_probe_pdata
  OMAPDSS: DPI: Add error handling for dpi_probe_pdata
  OMAPDSS: VENC: use platform_driver_register()
  OMAPDSS: HDMI: use platform_driver_register()
  OMAPDSS: RFBI: use platform_driver_register()
  OMAPDSS: DSI: use platform_driver_register()
  OMAPDSS: SDI: use platform_driver_register()
  OMAPDSS: DPI: use platform_driver_register()
  OMAPFB: defer probe if no displays
  OMAPFB: use module_platform_driver()
  OMAPDSS: Makefile: move omapfb after panels
  OMAPDSS: DPI: fix compilation if DSI not compiled in

11 years agoMerge branch 'drm-next-3.10-2' of git://people.freedesktop.org/~agd5f/linux into...
Dave Airlie [Fri, 3 May 2013 00:09:11 +0000 (10:09 +1000)]
Merge branch 'drm-next-3.10-2' of git://people.freedesktop.org/~agd5f/linux into drm-next

Just some fixes that have accumulated over the last couple of
weeks and some new PCI ids.

* 'drm-next-3.10-2' of git://people.freedesktop.org/~agd5f/linux:
  drm/radeon: fix handling of v6 power tables
  drm/radeon: clarify family checks in pm table parsing
  drm/radeon: consolidate UVD clock programming
  drm/radeon: fix UPLL_REF_DIV_MASK definition
  radeon: add bo tracking debugfs
  drm/radeon: add new richland pci ids
  drm/radeon: add some new SI PCI ids
  drm/radeon: fix scratch reg handling for UVD fence
  drm/radeon: allocate SA bo in the requested domain
  drm/radeon: fix possible segfault when parsing pm tables
  drm/radeon: fix endian bugs in atom_allocate_fb_scratch()

11 years agodrm: don't check modeset locks in panic handler
Daniel Vetter [Thu, 2 May 2013 07:43:05 +0000 (09:43 +0200)]
drm: don't check modeset locks in panic handler

Since we know that locking is broken in that case and it's more
important to not flood the dmesg with random gunk.

Cc: Dave Airlie <airlied@gmail.com>
Cc: Borislav Petkov <bp@alien8.de>
References: http://lkml.kernel.org/r/20130502000206.GH15623@pd.tnic
Cc: stable@vger.kernel.org
Reported-and-tested-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agodrm/nouveau: fix build with nv50->nvc0
Dave Airlie [Thu, 2 May 2013 20:40:37 +0000 (06:40 +1000)]
drm/nouveau: fix build with nv50->nvc0

Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/radeon: fix handling of v6 power tables
Alex Deucher [Wed, 1 May 2013 18:34:54 +0000 (14:34 -0400)]
drm/radeon: fix handling of v6 power tables

The code was mis-handling variable sized arrays.

Reported-by: Sylvain BERTRAND <sylware@legeek.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
11 years agodrm/radeon: clarify family checks in pm table parsing
Alex Deucher [Mon, 29 Apr 2013 14:51:26 +0000 (10:51 -0400)]
drm/radeon: clarify family checks in pm table parsing

We actually care about the chip family rather than the
DCE version although functionally they are the same.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agodrm/radeon: consolidate UVD clock programming
Christian König [Mon, 29 Apr 2013 09:55:02 +0000 (11:55 +0200)]
drm/radeon: consolidate UVD clock programming

Instead of duplicating the code over and over again, just use a single
function to handle the clock calculations.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agodrm/radeon: fix UPLL_REF_DIV_MASK definition
Christian König [Mon, 29 Apr 2013 08:20:23 +0000 (10:20 +0200)]
drm/radeon: fix UPLL_REF_DIV_MASK definition

Stupid copy & paste error over all generations.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agoradeon: add bo tracking debugfs
Jerome Glisse [Fri, 26 Apr 2013 02:29:27 +0000 (22:29 -0400)]
radeon: add bo tracking debugfs

This is to allow debugging of userspace program not freeing buffer
after, which is basicly a memory leak. This print the list of all
gem object along with their size and placement (VRAM,GTT,CPU) and
with the pid of the task that created them.

agd5f: add warning fix

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agodrm/radeon: add new richland pci ids
Alex Deucher [Thu, 25 Apr 2013 18:06:05 +0000 (14:06 -0400)]
drm/radeon: add new richland pci ids

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
11 years agodrm/radeon: add some new SI PCI ids
Alex Deucher [Thu, 25 Apr 2013 17:55:15 +0000 (13:55 -0400)]
drm/radeon: add some new SI PCI ids

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
11 years agodrm/radeon: fix scratch reg handling for UVD fence
Christian König [Wed, 24 Apr 2013 12:11:09 +0000 (14:11 +0200)]
drm/radeon: fix scratch reg handling for UVD fence

Also init the scratch reg to zero on the UVD ring.
This fixes UVD on AGP based cards.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agodrm/radeon: allocate SA bo in the requested domain
Christian König [Thu, 25 Apr 2013 16:19:08 +0000 (18:19 +0200)]
drm/radeon: allocate SA bo in the requested domain

This avoid moving the BO directly after allocating it.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
11 years agodrm/radeon: fix possible segfault when parsing pm tables
Alex Deucher [Thu, 25 Apr 2013 13:29:17 +0000 (09:29 -0400)]
drm/radeon: fix possible segfault when parsing pm tables

If we have a empty power table, bail early and allocate
the default power state.

Should fix:
https://bugs.freedesktop.org/show_bug.cgi?id=63865

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
11 years agodrm/radeon: fix endian bugs in atom_allocate_fb_scratch()
Alex Deucher [Wed, 24 Apr 2013 18:39:31 +0000 (14:39 -0400)]
drm/radeon: fix endian bugs in atom_allocate_fb_scratch()

Reviwed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
11 years agoOMAPDSS: TFP410: return EPROBE_DEFER if the i2c adapter not found
Tomi Valkeinen [Mon, 29 Apr 2013 10:08:32 +0000 (13:08 +0300)]
OMAPDSS: TFP410: return EPROBE_DEFER if the i2c adapter not found

If the I2C adapter needed by the TFP410 device is not available yet,
return EPROBE_DEFER so that the device will get probed again.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: VENC: Add error handling for venc_probe_pdata
Tomi Valkeinen [Thu, 2 May 2013 09:19:54 +0000 (12:19 +0300)]
OMAPDSS: VENC: Add error handling for venc_probe_pdata

Add proper error handling for venc_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the VENC driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: HDMI: Add error handling for hdmi_probe_pdata
Tomi Valkeinen [Fri, 26 Apr 2013 11:52:23 +0000 (14:52 +0300)]
OMAPDSS: HDMI: Add error handling for hdmi_probe_pdata

Add proper error handling for hdmi_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the HDMI driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: RFBI: Add error handling for rfbi_probe_pdata
Tomi Valkeinen [Thu, 2 May 2013 09:15:31 +0000 (12:15 +0300)]
OMAPDSS: RFBI: Add error handling for rfbi_probe_pdata

Add proper error handling for rfbi_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the RFBI driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: DSI: Add error handling for dsi_probe_pdata
Tomi Valkeinen [Fri, 26 Apr 2013 10:47:41 +0000 (13:47 +0300)]
OMAPDSS: DSI: Add error handling for dsi_probe_pdata

Add proper error handling for dsi_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the DSI driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: SDI: Add error handling for sdi_probe_pdata
Tomi Valkeinen [Thu, 2 May 2013 09:10:37 +0000 (12:10 +0300)]
OMAPDSS: SDI: Add error handling for sdi_probe_pdata

Add proper error handling for sdi_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the SDI driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: DPI: Add error handling for dpi_probe_pdata
Tomi Valkeinen [Fri, 26 Apr 2013 11:33:05 +0000 (14:33 +0300)]
OMAPDSS: DPI: Add error handling for dpi_probe_pdata

Add proper error handling for dpi_probe_pdata(). This will cause
EPROBE_DEFER to be properly passed upwards, causing the DPI driver to be
probed again later if a resource was missing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: VENC: use platform_driver_register()
Tomi Valkeinen [Thu, 2 May 2013 09:18:20 +0000 (12:18 +0300)]
OMAPDSS: VENC: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: HDMI: use platform_driver_register()
Tomi Valkeinen [Fri, 26 Apr 2013 11:48:43 +0000 (14:48 +0300)]
OMAPDSS: HDMI: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: RFBI: use platform_driver_register()
Tomi Valkeinen [Thu, 2 May 2013 09:11:56 +0000 (12:11 +0300)]
OMAPDSS: RFBI: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: DSI: use platform_driver_register()
Tomi Valkeinen [Fri, 26 Apr 2013 10:47:06 +0000 (13:47 +0300)]
OMAPDSS: DSI: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: SDI: use platform_driver_register()
Tomi Valkeinen [Thu, 2 May 2013 08:56:35 +0000 (11:56 +0300)]
OMAPDSS: SDI: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: DPI: use platform_driver_register()
Tomi Valkeinen [Fri, 26 Apr 2013 11:27:44 +0000 (14:27 +0300)]
OMAPDSS: DPI: use platform_driver_register()

Use platform_driver_register() instead of platform_driver_probe() so
that we can support EPROBE_DEFER.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPFB: defer probe if no displays
Tomi Valkeinen [Fri, 26 Apr 2013 10:40:36 +0000 (13:40 +0300)]
OMAPFB: defer probe if no displays

omapfb requires the panel drivers to have been probed when omapfb is
initialized. omapfb does not support insertion of new panels after its
probe. This causes a problem in case omapdss or the panel probes have
been deferred due to EPROBE_DEFER error, as omapfb won't find any
displays.

As a quick fix, this patch changes the omapfb probe so that if omapfb
does not find any displays, it'll return EPROBE_DEFER. This is not
perfect, as with a board with no displays, omapfb will get deferred
forever. Also, if the board has multiple displays, but only some of them
have been probed, omapfb will start and leave the unprobed displays out.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPFB: use module_platform_driver()
Tomi Valkeinen [Fri, 26 Apr 2013 10:40:25 +0000 (13:40 +0300)]
OMAPFB: use module_platform_driver()

Instead of using platform_driver_probe(), use module_platform_driver()
so that we can support deferred probing.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: Makefile: move omapfb after panels
Tomi Valkeinen [Fri, 26 Apr 2013 10:35:50 +0000 (13:35 +0300)]
OMAPDSS: Makefile: move omapfb after panels

omapfb requires the panels to have been probed before omapfb's probe. We
currently manage that by having omapfb in late initcall level. However,
a much simpler way is to just change the makefile so that omapfb is
after the panel drivers.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoOMAPDSS: DPI: fix compilation if DSI not compiled in
Tomi Valkeinen [Thu, 18 Apr 2013 09:16:39 +0000 (12:16 +0300)]
OMAPDSS: DPI: fix compilation if DSI not compiled in

Commit 100c826235793345efe06b3558cc9d36166b1e26 (OMAPDSS: DPI: use new
clock calculation code) breaks dpi.c compilation if DSI is not enabled
in the kernel configuration.

Fix compilation by adding dummy inline functions for the ones that dpi.c
references. The functions will never be called, as dpi.c knows that
there is no DSI device available.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
11 years agoMerge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux...
Dave Airlie [Thu, 2 May 2013 07:33:25 +0000 (17:33 +1000)]
Merge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-next

Add GK110 modesetting suport.

* 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6:
  drm/nve0: recognise nvf0 as a kepler board (GK110)
  drm/nouveau: force noaccel when no PFIFO support present
  drm/nvf0/disp: expose display class 2.2

11 years agodrm/nve0: recognise nvf0 as a kepler board (GK110)
Ben Skeggs [Sat, 30 Mar 2013 12:21:54 +0000 (22:21 +1000)]
drm/nve0: recognise nvf0 as a kepler board (GK110)

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau: force noaccel when no PFIFO support present
Ben Skeggs [Thu, 2 May 2013 06:20:44 +0000 (16:20 +1000)]
drm/nouveau: force noaccel when no PFIFO support present

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nvf0/disp: expose display class 2.2
Ben Skeggs [Sat, 30 Mar 2013 12:31:25 +0000 (22:31 +1000)]
drm/nvf0/disp: expose display class 2.2

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/cirrus: deal with bo reserve fail in dirty update path
Dave Airlie [Thu, 2 May 2013 06:45:02 +0000 (02:45 -0400)]
drm/cirrus: deal with bo reserve fail in dirty update path

Port over the mgag200 fix to cirrus as it suffers the same issue.

    On F19 testing, it was noticed we get a lot of errors in dmesg
    about being unable to reserve the buffer when plymouth starts,
    this is due to the buffer being in the process of migrating,
    so it makes sense we can't reserve it.

    In order to deal with it, this adds delayed updates for the dirty
    updates, when the bo is unreservable, in the normal console case
    this shouldn't ever happen, its just when plymouth or X is
    pushing the console bo to system memory.

Cc: stable@vger.kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/ast: deal with bo reserve fail in dirty update path
Dave Airlie [Thu, 2 May 2013 06:40:25 +0000 (02:40 -0400)]
drm/ast: deal with bo reserve fail in dirty update path

Port over the mgag200 fix to ast as it suffers the same issue.

    On F19 testing, it was noticed we get a lot of errors in dmesg
    about being unable to reserve the buffer when plymouth starts,
    this is due to the buffer being in the process of migrating,
    so it makes sense we can't reserve it.

    In order to deal with it, this adds delayed updates for the dirty
    updates, when the bo is unreservable, in the normal console case
    this shouldn't ever happen, its just when plymouth or X is
    pushing the console bo to system memory.

Cc: stable@vger.kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/mgag200: deal with bo reserve fail in dirty update path
Dave Airlie [Thu, 2 May 2013 04:52:01 +0000 (00:52 -0400)]
drm/mgag200: deal with bo reserve fail in dirty update path

On F19 testing, it was noticed we get a lot of errors in dmesg
about being unable to reserve the buffer when plymouth starts,
this is due to the buffer being in the process of migrating,
so it makes sense we can't reserve it.

In order to deal with it, this adds delayed updates for the dirty
updates, when the bo is unreservable, in the normal console case
this shouldn't ever happen, its just when plymouth or X is
pushing the console bo to system memory.

Cc: stable@vger.kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/i915: Fix pipe enabled mask for pipe C in WM calculations
Ville Syrjälä [Thu, 21 Mar 2013 11:10:44 +0000 (13:10 +0200)]
drm/i915: Fix pipe enabled mask for pipe C in WM calculations

Fix the incorrect enabled pipes mask for pipe C in the WM calculations.

Additionally, in an effort to make the code easier to understand,
populate the mask with 1 << PIPE_[ABC] instead of raw numbers.

v2: Use 1 << PIPE_[ABC] (ickle/danvet)
v3: Pass PIPE_[ABC] to g4x_compute_wm0() (ickle)

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agodrm/i915: fix dmabuf vmap support
Dave Airlie [Wed, 1 May 2013 04:23:41 +0000 (14:23 +1000)]
drm/i915: fix dmabuf vmap support

Sometimes that extra semicolon can really be hard to spot.

Acked-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/prime: warn for non-empty handle lookup list during drm file release
Imre Deak [Wed, 24 Apr 2013 16:04:57 +0000 (19:04 +0300)]
drm/prime: warn for non-empty handle lookup list during drm file release

drm_gem_release should release all handles connected to the drm file and
so should also release the prime lookup entries of these handles. So
just WARN if this isn't the case.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agoudl: bind the framebuffer to the correct device.
Dave Airlie [Fri, 19 Apr 2013 01:01:55 +0000 (11:01 +1000)]
udl: bind the framebuffer to the correct device.

This just moves the fb sysfs node beside the drm sysfs node which
I fixed before.

just noticed it in passing.

Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm: prime: fix refcounting on the dmabuf import error path
Imre Deak [Fri, 19 Apr 2013 01:11:56 +0000 (11:11 +1000)]
drm: prime: fix refcounting on the dmabuf import error path

In commit be8a42ae60 we inroduced a refcount problem, where on the
drm_gem_prime_fd_to_handle() error path we'll call dma_buf_put() for
self imported dma buffers.

Fix this by taking a reference on the dma buffer in the .gem_import
hook instead of assuming the caller had taken one. Besides fixing the
bug this is also more logical.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Cc: stable@vger.kernel.org
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/prime: keep a reference from the handle to exported dma-buf (v6)
Dave Airlie [Sun, 21 Apr 2013 23:54:36 +0000 (09:54 +1000)]
drm/prime: keep a reference from the handle to exported dma-buf (v6)

Currently we have a problem with this:
1. i915: create gem object
2. i915: export gem object to prime
3. radeon: import gem object
4. close prime fd
5. radeon: unref object
6. i915: unref object

i915 has an imported object reference in its file priv, that isn't
cleaned up properly until fd close. The reference gets added at step 2,
but at step 6 we don't have enough info to clean it up.

The solution is to take a reference on the dma-buf when we export it,
and drop the reference when the gem handle goes away.

So when we export a dma_buf from a gem object, we keep track of it
with the handle, we take a reference to the dma_buf. When we close
the handle (i.e. userspace is finished with the buffer), we drop
the reference to the dma_buf, and it gets collected.

This patch isn't meant to fix any other problem or bikesheds, and it doesn't
fix any races with other scenarios.

v1.1: move export symbol line back up.

v2: okay I had to do a bit more, as the first patch showed a leak
on one of my tests, that I found using the dma-buf debugfs support,
the problem case is exporting a buffer twice with the same handle,
we'd add another export handle for it unnecessarily, however
we now fail if we try to export the same object with a different gem handle,
however I'm not sure if that is a case I want to support, and I've
gotten the code to WARN_ON if we hit something like that.

v2.1: rebase this patch, write better commit msg.
v3: cleanup error handling, track import vs export in linked list,
these two patches were separate previously, but seem to work better
like this.
v4: danvet is correct, this code is no longer useful, since the buffer
better exist, so remove it.
v5: always take a reference to the dma buf object, import or export.
(Imre Deak contributed this originally)
v6: square the circle, remove import vs export tracking now
that there is no difference

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/mm: fix dump table BUG
Daniel Vetter [Sat, 20 Apr 2013 10:08:11 +0000 (12:08 +0200)]
drm/mm: fix dump table BUG

In

commit 9e8944ab564f2e3dde90a518cd32048c58918608
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Nov 15 11:32:17 2012 +0000

    drm: Introduce an iterator over holes in the drm_mm range manager

helpers and iterators for hole handling have been introduced with some
debug BUG_ONs sprinkled over. Unfortunately this broke the mm dumper
which unconditionally tried to compute the size of the very first
hole.

While at it unify the code a bit with the hole dumping in the loop.

v2: Extract a hole dump helper.

Reported-by: Christopher Harvey <charvey@matrox.com>
Cc: Christopher Harvey <charvey@matrox.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@vger.kernel.org
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agodrm/i915: Always normalize return timeout for wait_timeout_ioctl
Chris Wilson [Fri, 26 Apr 2013 13:22:46 +0000 (16:22 +0300)]
drm/i915: Always normalize return timeout for wait_timeout_ioctl

As we recompute the remaining timeout after waiting, there is a
potential for that timeout to be less than zero and so need sanitizing.
The timeout is always returned to userspace and validated, so we should
always perform the sanitation.

v2 [vsyrjala]: Only normalize the timespec if it's invalid
v3: Add a comment to clarify the situation and remove the now
    useless WARN_ON() (ickle)

Cc: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agodrm: Kill user_modes list and the associated ioctls
Ville Syrjälä [Fri, 26 Apr 2013 14:40:28 +0000 (17:40 +0300)]
drm: Kill user_modes list and the associated ioctls

There is no way to use modes added to the user_modes list. We never
look at the contents of said list in the kernel, and the only operations
userspace can do are attach and detach. So the only "benefit" of this
interface is wasting kernel memory.

Fortunately it seems no real user space application ever used these
ioctls. So just kill them.

Also remove the prototypes for the non-existing drm_mode_addmode_ioctl()
and drm_mode_rmmode_ioctl() functions.

v2: Use drm_noop instead of completely removing the ioctls

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm: Silence some sparse warnings
Ville Syrjälä [Thu, 25 Apr 2013 17:09:20 +0000 (20:09 +0300)]
drm: Silence some sparse warnings

drivers/gpu/drm/drm_pci.c:155:5: warning: symbol 'drm_pci_set_busid' was not declared. Should it be static?
drivers/gpu/drm/drm_pci.c:197:5: warning: symbol 'drm_pci_set_unique' was not declared. Should it be static?
drivers/gpu/drm/drm_pci.c:269:5: warning: symbol 'drm_pci_agp_init' was not declared. Should it be static?

drivers/gpu/drm/drm_crtc.c:181:1: warning: symbol 'drm_get_dirty_info_name' was not declared. Should it be static?
drivers/gpu/drm/drm_crtc.c:1123:5: warning: symbol 'drm_mode_group_init' was not declared. Should it be static?

drivers/gpu/drm/drm_modes.c:918:6: warning: symbol 'drm_mode_validate_clocks' was not declared. Should it be static?

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm: Make drm_ioctls const
Ville Syrjälä [Thu, 25 Apr 2013 17:09:19 +0000 (20:09 +0300)]
drm: Make drm_ioctls const

We never modify the contents of drm_ioctls, so make it const.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrivers, drm: fix qxl build error when debugfs is disabled
David Rientjes [Sun, 28 Apr 2013 21:09:25 +0000 (14:09 -0700)]
drivers, drm: fix qxl build error when debugfs is disabled

Fix build error when CONFIG_DEBUG_FS is disabled:

drivers/gpu/drm/qxl/qxl_debugfs.c: In function 'qxl_debugfs_init':
drivers/gpu/drm/qxl/qxl_debugfs.c:76:2: error: implicit declaration of function 'drm_debugfs_create_files'
drivers/gpu/drm/qxl/qxl_debugfs.c: In function 'qxl_debugfs_takedown':
drivers/gpu/drm/qxl/qxl_debugfs.c:84:2: error: implicit declaration of function 'drm_debugfs_remove_files'

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agoMerge branch 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel...
Dave Airlie [Mon, 29 Apr 2013 23:58:17 +0000 (09:58 +1000)]
Merge branch 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel into drm-next

Just a few important fixes for 3.10. 3 regression fixes, plus rectified
Haswell overclock support (the old code was correct, only docs confusing)
and improved DP data m/n selection.

* 'drm-intel-fixes' of git://people.freedesktop.org/~danvet/drm-intel:
  drm/i915: correct the calculation of first_pd_entry_in_global_pt
  Revert "drm/i915: Don't overclock on Haswell"
  drm/i915: Make data/link N value power of two
  drm/i915: avoid full modeset when changing the color range properties
  drm/i915: Fall back to bit banging mode for DVO transmitter detection

11 years agoMerge branch 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daein...
Dave Airlie [Mon, 29 Apr 2013 23:57:46 +0000 (09:57 +1000)]
Merge branch 'exynos-drm-next' of git://git./linux/kernel/git/daeinki/drm-exynos into drm-next

   This is final pull request for Exynos next and includes device tree
   support for fimc device, one revert, some code cleanups and fixup.
   The revert replaces wrong one[1] with correct one[2].
   This was my mistake and sorry for this.

* 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos:
  drm/exynos: Don't blend mixer layer 0
  drm/exynos: Remove unnecessary braces in exynos_hdmi.c
  drm/exynos: Select VIDEOMODE_HELPERS for FIMD
  drm/exynos: do not use generic flags to dumb
  drm/exynos: added ipp device registration to drm driver
  exynos/drm: hdmi: cleanup for hdmi common device registration
  drm/exynos: fix wrong return check for platform_device_register_simple
  drm/exynos: add device tree support for fimc ipp driver
  drm/exynos: rework fimc clocks handling
  drm/exynos: remove redundant devm_kfree()
  drm/exynos: enable FIMD clocks
  Revert "drm/exynos: prepare FIMD clocks"

11 years agodrm/mgag200: Remove extra variable assigns
Christopher Harvey [Wed, 20 Feb 2013 14:36:24 +0000 (09:36 -0500)]
drm/mgag200: Remove extra variable assigns

These two variables are set again immediately in 'mgag200_modeset_init'

Signed-off-by: Christopher Harvey <charvey@matrox.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/mgag200: Pass driver specific mga_device in driver functions
Christopher Harvey [Wed, 20 Feb 2013 14:34:22 +0000 (09:34 -0500)]
drm/mgag200: Pass driver specific mga_device in driver functions

Signed-off-by: Christopher Harvey <charvey@matrox.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/mgag200: Remove pointless call to drm_fb_get_bpp_depth
Christopher Harvey [Mon, 18 Feb 2013 22:33:49 +0000 (17:33 -0500)]
drm/mgag200: Remove pointless call to drm_fb_get_bpp_depth

Signed-off-by: Christopher Harvey <charvey@matrox.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
11 years agodrm/exynos: Don't blend mixer layer 0
Sean Paul [Thu, 25 Apr 2013 19:13:26 +0000 (15:13 -0400)]
drm/exynos: Don't blend mixer layer 0

This patch disables blending the mixer's layer 0 onto the background
(solid color). It doesn't make sense to blend this layer by default,
and causes color distortion if the layer is used for arbitrary content.

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: Remove unnecessary braces in exynos_hdmi.c
Sachin Kamat [Fri, 26 Apr 2013 05:59:00 +0000 (11:29 +0530)]
drm/exynos: Remove unnecessary braces in exynos_hdmi.c

Silences the following checkpatch warning:
WARNING: braces {} are not necessary for any arm of this statement

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: Select VIDEOMODE_HELPERS for FIMD
Sachin Kamat [Thu, 25 Apr 2013 11:36:06 +0000 (17:06 +0530)]
drm/exynos: Select VIDEOMODE_HELPERS for FIMD

FIMD also requires video mode helper APIs.
Without this patch we get the following build error:
drivers/gpu/drm/exynos/exynos_drm_fimd.c:895: undefined reference to
`of_get_fb_videomode'
make: *** [vmlinux] Error 1

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: do not use generic flags to dumb
Inki Dae [Thu, 18 Apr 2013 05:04:17 +0000 (14:04 +0900)]
drm/exynos: do not use generic flags to dumb

This patch removes the use of dumb flags from driver.

As Dave pointed out, the dumb flags are not driver specific
so this should be removed from driver.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
11 years agodrm/exynos: added ipp device registration to drm driver
Seung-Woo Kim [Tue, 23 Apr 2013 05:02:53 +0000 (14:02 +0900)]
drm/exynos: added ipp device registration to drm driver

This patch added exynos-drm-ipp platform device registration to the exynos drm
driver. When DT is enabled, platform devices need to be registered within the
driver code. This patch fits the requirement of both DT and Non DT based drm
drivers.

Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Donghwa Lee <dh09.lee@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agoexynos/drm: hdmi: cleanup for hdmi common device registration
Seung-Woo Kim [Mon, 22 Apr 2013 09:40:34 +0000 (18:40 +0900)]
exynos/drm: hdmi: cleanup for hdmi common device registration

The hdmi common device registration function does not need extern definition
and for error case and unregister case, exynos_drm_hdmi_pdev should be cleared.

Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: fix wrong return check for platform_device_register_simple
Seung-Woo Kim [Mon, 22 Apr 2013 08:13:13 +0000 (17:13 +0900)]
drm/exynos: fix wrong return check for platform_device_register_simple

platform_device_register_simple() never returns NULL, but IS_ERR_OR_NULL macro
is used for checking return value in exynos drm driver.

Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: add device tree support for fimc ipp driver
Sylwester Nawrocki [Tue, 23 Apr 2013 11:34:38 +0000 (13:34 +0200)]
drm/exynos: add device tree support for fimc ipp driver

This patch adds OF initialization support for the FIMC driver.
The binding documentation can be found at Documentation/devicetree/
bindings/media/samsung-fimc.txt.

The syscon regmap interface is used to serialize access to the
shared CAMBLK registers from within the V4L2 FIMC-IS and the DRM
FIMC drivers. The DRM driver uses this interface for setting up
the FIFO data link between FIMD and FIMC IP blocks, while the V4L2
one for setting up a data link between the camera ISP and FIMC for
camera capture. The CAMBLK registers are not accessed any more
through a statically mapped IO. Synchronized access to these
registers is required for simultaneous operation of the camera
ISP and the DRM IPP on Exynos4x12.

The driver data and driver_ids static data structures are removed
since Exynos4 is going to be a dt-only platform and there is
currently no board file in mainline that defines platform data
for the FIMC IPP, i.e. uses it.

Camera input signal polarities are not currently parsed from the
device tree.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: rework fimc clocks handling
Sylwester Nawrocki [Tue, 23 Apr 2013 11:34:37 +0000 (13:34 +0200)]
drm/exynos: rework fimc clocks handling

The clocks handling is refactored and a "mux" clock handling is
added to account for changes in the clocks driver. After switching
to the common clock framework the sclk_fimc clock is now split
into two clocks: a gate and a mux clock. In order to retain the
exisiting functionality two additional consumer clocks are passed
to the driver from device tree: "mux" and "parent". Then the driver
sets "parent" clock as a parent clock of the "mux" clock. These two
additional clocks are optional, and should go away when there is a
standard way of setting up parent clocks on DT platforms.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: remove redundant devm_kfree()
Sylwester Nawrocki [Tue, 23 Apr 2013 11:34:36 +0000 (13:34 +0200)]
drm/exynos: remove redundant devm_kfree()

There is no need for explicit calls of devm_kfree(), as
the allocated memory will be freed during driver's detach.
Remove the redundant devm_kfree() calls from probe() and
remove() callbacks.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agodrm/exynos: enable FIMD clocks
Vikas Sajjan [Tue, 23 Apr 2013 11:49:50 +0000 (17:19 +0530)]
drm/exynos: enable FIMD clocks

Common Clock Framework introduced the need to prepare clocks before
enabling them, otherwise clk_enable() fails. This patch adds clk_prepare_enable
and clk_disable_unprepare() calls to the driver.
This patch also removes clk_disable() from fimd_remove() as it will be done
by pm_runtime_put_sync.

Signed-off-by: Vikas Sajjan <vikas.sajjan@linaro.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
11 years agoRevert "drm/exynos: prepare FIMD clocks"
Inki Dae [Thu, 25 Apr 2013 10:21:42 +0000 (19:21 +0900)]
Revert "drm/exynos: prepare FIMD clocks"

This reverts commit b4e3a3e844a0e33cf106a1c9f27ff93340c37640.

11 years agodrm/i915: correct the calculation of first_pd_entry_in_global_pt
Zhang, Xiong Y [Sat, 27 Apr 2013 09:53:33 +0000 (09:53 +0000)]
drm/i915: correct the calculation of first_pd_entry_in_global_pt

When ppgtt is enabled, dev_priv->gtt.total has excluded the gtt space
occupied by ppgtt table in i915_gem_init_global_gtt() function. So the
calculation of first_pd_entry_in_global_pt doesn't need to subtract
I915_PPGTT_PD_ENTRIES again. Or else PPGTT directory table will be
destroyed by global gtt allocation.

This regression has been introduced in

commit a54c0c279f3864171fe53c66e769d5a137c5c651
Author: Ben Widawsky <ben@bwidawsk.net>
Date:   Thu Jan 24 14:45:00 2013 -0800

    drm/i915: remove intel_gtt structure

The breakage is pretty subtile since the old gtt_total_entries
included the pde range, whereas the new on did not.

Cc: stable@vger.kernel.org
Signed-off-by: Xiong Zhang<xiong.y.zhang@intel.com>
[danvet: Add regression citation and cc: stable. Thanks to Chris for
correcting my wrong guess about which commit broke things.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
11 years agoMerge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux...
Dave Airlie [Fri, 26 Apr 2013 05:42:02 +0000 (15:42 +1000)]
Merge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-next

"Nothing overly exciting here aside from calim's fermi/kepler vram
compression patches.  The rest is misc fixes I gathered from the list.

Most of the stuff from me is fixing issues that have come up from the
work on kepler PM, as well as a commit moving all the old-school
modesetting out of the way (no code changes here).  There's other
patches to go on top of that, but, it'll have to wait until I can rip
out the old PM code, it's a bit tangled."

* 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6: (38 commits)
  drm/nouveau/fifo: implement channel creation event generation
  drm/nouveau/core: allow non-maskable events
  drm/nouveau/timer: allow alarms to be cancelled
  drm/nouveau/device: tweak the device/subdev relationship a little
  drm/nouveau/device: enable proper constructor/destructor
  drm/nouveau/device: have engine object initialised before creation
  drm/nouveau/device: convert to engine, rather than subdev
  drm/nv50-/disp: use self as parent for subobjects
  drm/nv50-/fifo: use parent as self for subobjects
  drm/nv20-nv30/gr: use parent as self for subobjects
  drm/nvc0-/gr: use self as parent for subobjects
  drm/nv04-nv40/instmem: use self as parent for subobjects
  drm/nv04-nv40/vm: use self as parent for subobjects
  drm/nv50-/bar: use self as parent for subobjects
  drm/nv04-nv40/instmem: remove parent deref hack
  drm/nouveau/i2c: remove parent deref hack
  drm/nouveau/core: rebase object ref/use counts after ctor/init/fini events
  drm/nv50/disp: inform core when we're not creating a new context
  drm/nouveau/therm: send some messages to debug level
  drm/nve0/gr: add handling for a bunch of PGRAPH traps
  ...

11 years agodrm/nouveau/fifo: implement channel creation event generation
Ben Skeggs [Fri, 26 Apr 2013 01:44:26 +0000 (11:44 +1000)]
drm/nouveau/fifo: implement channel creation event generation

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/core: allow non-maskable events
Ben Skeggs [Fri, 26 Apr 2013 02:03:10 +0000 (12:03 +1000)]
drm/nouveau/core: allow non-maskable events

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/timer: allow alarms to be cancelled
Ben Skeggs [Thu, 25 Apr 2013 23:17:22 +0000 (09:17 +1000)]
drm/nouveau/timer: allow alarms to be cancelled

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/device: tweak the device/subdev relationship a little
Ben Skeggs [Thu, 25 Apr 2013 01:43:54 +0000 (11:43 +1000)]
drm/nouveau/device: tweak the device/subdev relationship a little

Fixes not-in-use engines not having their reset() method called on
resume.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/device: enable proper constructor/destructor
Ben Skeggs [Thu, 25 Apr 2013 01:35:18 +0000 (11:35 +1000)]
drm/nouveau/device: enable proper constructor/destructor

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/device: have engine object initialised before creation
Ben Skeggs [Thu, 25 Apr 2013 07:56:03 +0000 (17:56 +1000)]
drm/nouveau/device: have engine object initialised before creation

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/device: convert to engine, rather than subdev
Ben Skeggs [Thu, 25 Apr 2013 07:23:43 +0000 (17:23 +1000)]
drm/nouveau/device: convert to engine, rather than subdev

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv50-/disp: use self as parent for subobjects
Ben Skeggs [Wed, 24 Apr 2013 08:04:22 +0000 (18:04 +1000)]
drm/nv50-/disp: use self as parent for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv50-/fifo: use parent as self for subobjects
Ben Skeggs [Wed, 24 Apr 2013 08:02:35 +0000 (18:02 +1000)]
drm/nv50-/fifo: use parent as self for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv20-nv30/gr: use parent as self for subobjects
Ben Skeggs [Wed, 24 Apr 2013 07:55:28 +0000 (17:55 +1000)]
drm/nv20-nv30/gr: use parent as self for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nvc0-/gr: use self as parent for subobjects
Ben Skeggs [Wed, 24 Apr 2013 07:51:20 +0000 (17:51 +1000)]
drm/nvc0-/gr: use self as parent for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv04-nv40/instmem: use self as parent for subobjects
Ben Skeggs [Wed, 24 Apr 2013 07:48:56 +0000 (17:48 +1000)]
drm/nv04-nv40/instmem: use self as parent for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv04-nv40/vm: use self as parent for subobjects
Ben Skeggs [Wed, 24 Apr 2013 07:46:54 +0000 (17:46 +1000)]
drm/nv04-nv40/vm: use self as parent for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv50-/bar: use self as parent for subobjects
Ben Skeggs [Wed, 24 Apr 2013 07:44:54 +0000 (17:44 +1000)]
drm/nv50-/bar: use self as parent for subobjects

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv04-nv40/instmem: remove parent deref hack
Ben Skeggs [Wed, 24 Apr 2013 07:38:38 +0000 (17:38 +1000)]
drm/nv04-nv40/instmem: remove parent deref hack

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/i2c: remove parent deref hack
Ben Skeggs [Wed, 24 Apr 2013 07:36:57 +0000 (17:36 +1000)]
drm/nouveau/i2c: remove parent deref hack

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/core: rebase object ref/use counts after ctor/init/fini events
Ben Skeggs [Wed, 24 Apr 2013 07:34:18 +0000 (17:34 +1000)]
drm/nouveau/core: rebase object ref/use counts after ctor/init/fini events

This is intended to support named (with a handle, etc) objects having
children that don't have an outside reference.

This will replace the various hacks around the place where subdev
objects have children, and have to manually drop the self-refs so
that they can be destroyed etc when all the outside refs have gone.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv50/disp: inform core when we're not creating a new context
Ben Skeggs [Thu, 25 Apr 2013 14:12:59 +0000 (00:12 +1000)]
drm/nv50/disp: inform core when we're not creating a new context

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/therm: send some messages to debug level
Ben Skeggs [Wed, 24 Apr 2013 03:14:02 +0000 (13:14 +1000)]
drm/nouveau/therm: send some messages to debug level

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nve0/gr: add handling for a bunch of PGRAPH traps
Christoph Bumiller [Wed, 27 Mar 2013 20:59:11 +0000 (21:59 +0100)]
drm/nve0/gr: add handling for a bunch of PGRAPH traps

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/mc: handle irq-related setup ourselves
Ben Skeggs [Fri, 22 Mar 2013 00:05:03 +0000 (10:05 +1000)]
drm/nouveau/mc: handle irq-related setup ourselves

We need to be able to process interrupts before the DRM code is able to
actually enable them, set it up ourselves.  Also, it's less convoluted
to *not* use the DRM wrappers it appears...

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nv04/disp: hide all the cruft away in its own little hole
Ben Skeggs [Thu, 21 Mar 2013 05:45:11 +0000 (15:45 +1000)]
drm/nv04/disp: hide all the cruft away in its own little hole

It'd be pretty awesome if someone would care enough to port this all
properly to a class interface, perhaps submitting a command stream to
the core via a sw object on PFIFO (emulating how EVO works basically,
and also what nvidia have done forever..)..

But, this seems unlikely given how old this hardware is now, so, lets
just hide it away.

There's a heap of other bits and pieces laying around that are still
tangled.  I'll (re)move them in pieces.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
11 years agodrm/nouveau/bios: add missing newline on IO*_OR opcode debugging
Ben Skeggs [Fri, 19 Apr 2013 00:22:09 +0000 (10:22 +1000)]
drm/nouveau/bios: add missing newline on IO*_OR opcode debugging

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>