Commit graph

6837 commits

Author SHA1 Message Date
Paulo Zanoni
0057d8dd8d drm: make the connector properties code use the object properties code
In the future, we may want to kill the internal functions:
- drm_connector_attach_property
- drm_connector_property_set_value
- drm_connector_property_get_value

It seems the IOCTL drm_mode_connector_property_set_ioctl will have to live, but
we may change libdrm to not use it anymore, if we want.

Reviewed-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:11:28 +01:00
Paulo Zanoni
c543188afb drm: add generic ioctls to get/set properties on any object
Useless for connector properties (since they already have their own
ioctls), but useful when we add properties to CRTCs, planes and other
objects.

Reviewed-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:11:22 +01:00
Paulo Zanoni
7e3bdf4a6d drm: create struct drm_object_properties and use it
For now, only connectors have it. In the future, all objects that need
properties should use it. Since the structure is referenced inside
struct drm_mode_object, we will be able to deal with object properties
without knowing the real type of the object.

Reviewed-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:11:13 +01:00
Paulo Zanoni
afea2ad53f drm: WARN() when drm_connector_attach_property fails
Also return void instead of int. We have more than 100 callers and
no one checks for the return value.

If this function fails the property won't be exposed by the get/set
ioctls, but we should probably survive. If this starts happening,
the solution will be to increase DRM_CONNECTOR_MAX_PROPERTY and
recompile the Kernel.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:11:06 +01:00
Paulo Zanoni
26a3481586 drm: add drm_property_change_is_valid
Move code from drm_mode_connector_property_set_ioctl to a new
function, so we can reuse this code when we add crtc properties.

Reviewed-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: Rob Clark <rob.clark@linaro.org>
Tested-by: Rob Clark <rob.clark@linaro.org>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:10:55 +01:00
Alan Cox
4ab2c7f154 gma500: unload fixes
Debugging the lid problem tested various error paths which were found
wanting so start fixing them up.

There is a ton of improvement work could be done here so that every bit
of functionality agrees if its _fini, _uninit, etc, and they agree who
is responsible for deciding if the clean up is needed.

That can come later.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:05:30 +01:00
Alan Cox
6607e02401 gma500: Fix crash on D2700MUD and various other boards
The recent changes led to the lid timer code being run on various devices.
It does no harm on most but isn't needed. It also calls unconditionally
into the Poulsbo backlight code which goes bang on Cedartrail.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:05:29 +01:00
Dave Airlie
f9aa76a852 drm/kms: driver for virtual cirrus under qemu
This is the initial driver for emulated cirrus GPU found in qemu.
This driver only supports the emulated GPU and doesn't attempt
to bind to any real cirrus GPUs.

This driver is intended to be used with xf86-video-modesetting in userspace.
It requires at least version 0.3.0

This follow the same design as ast and mgag200, and is based on work
done by Matthew Garrett previously.

This GPU has no hw cursor, and it can't scanout 32-bpp, only packed 24-bpp.
i.e. it sucks.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 11:02:24 +01:00
Dave Airlie
414c453106 mgag200: initial g200se driver (v2)
This is a driver for the G200 server engines chips,
it doesn't driver any of the Matrix G series desktop cards.

It will bind to G200 SE A,B, G200EV, G200WB, G200EH and G200ER cards.

Its based on previous work done my Matthew Garrett but remodelled
to follow the same style and flow as the AST server driver. It also
works along the same lines as the AST server driver wrt memory management.

There is no userspace driver planned, xf86-video-modesetting should be used.
It also appears these GPUs have no ARGB hw cursors.

v2: add missing tagfifo reset + G200 SE memory bw setup pieces.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 10:53:41 +01:00
Dave Airlie
312fec1405 drm: Initial KMS driver for AST (ASpeed Technologies) 2000 series (v2)
This is the initial driver for the Aspeed Technologies chips found in
servers. This driver supports the AST 2000, 2100, 2200, 2150 and 2300. It
doesn't support the AST11xx due to lack of hw to test it on, and them requiring
different codepaths.

This driver is intended to be used with xf86-video-modesetting in userspace.

This driver has a slightly different design than other KMS drivers, but
future server chips will probably share similiar setup. As these GPUs commonly
have low video RAM, it doesn't make sense to put the kms console in VRAM
always. This driver places the kms console into system RAM, and does dirty
updates to a copy in video RAM. When userspace sets a new scanout buffer,
it forcefully evicts the video RAM console, and X can create a framebuffer
that can use all of of video RAM.

This driver uses TTM but in a very simple fashion to control the eviction
to system RAM of the console, and multiple servers.

v2: add s/r support, fix Kconfig.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-17 10:53:37 +01:00
Dave Airlie
f3b7e17c4b Merge branch 'topic/vga-switcheroo' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound into drm-core-next
* 'topic/vga-switcheroo' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  vga_switcheroo: Add the support for audio clients
  vga_switcheroo: Introduce struct vga_switcheroo_client_ops
  vga_switcheroo: Refactor using linked list
2012-05-13 16:58:08 +01:00
Rafał Miłecki
a366e39266 drm/radeon/hdmi: fix some coding style
Use defined macros by the way.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:19:37 +01:00
Rafał Miłecki
1c3439f228 drm/radeon/hdmi: update modesetting
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:19:36 +01:00
Rafał Miłecki
e55d3e6cb6 drm/radeon/hdmi: separate evergreen code
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:19:34 +01:00
Rafał Miłecki
1b688d0814 drm/radeon/kms/hdmi: helper getting ready ACR entry
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:19:29 +01:00
Rafał Miłecki
64fb4fb0e4 drm/radeon/kms/hdmi: clean&improve handling HDMI mode
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:19:13 +01:00
Rafał Miłecki
a273a903bf drm/radeon/kms/hdmi: enable audio packets at one place
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-13 14:18:58 +01:00
Takashi Iwai
3e9e63dbd3 vga_switcheroo: Add the support for audio clients
Add the support for audio clients to VGA-switcheroo for handling the
HDMI audio controller together with VGA switching.  The id of the
audio controller should be given explicitly at registration time
unlike the video controller.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43155

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-05-13 11:27:21 +02:00
Takashi Iwai
26ec685ff9 vga_switcheroo: Introduce struct vga_switcheroo_client_ops
This changes the API as a clean-up.  Instead of passing multiple
function pointers at each time, introduce a new struct holding the
whole callback functions and pass it to the registration.

The same struct will be used for the upcoming audio client
registration, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-05-13 11:24:09 +02:00
Takashi Iwai
79721e0a91 vga_switcheroo: Refactor using linked list
Refactor the code base a bit for the further work to adapt more clients.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-05-13 11:21:56 +02:00
Dave Airlie
218c872bf8 Merge tag 'drm-intel-next-2012-05-06-merged' of git://people.freedesktop.org/~danvet/drm-intel into drm-core-next
Daniel says

Highlights:
- sparse fixes from Ben.
- tons of little cleanups from Chris all over: tiling_changed
 clarification, deferred_free list removal, ...
- fix up irq handler on gen2 & gen3 + related cleanups from Chris
- prep work for wait_rendering_timeout from Ben with some nice
 refactorings
- first set of infoframe fixes from Paulo for doubleclocked CEA modes
- improve pch pll handling from Jesse and Chris
- gpu hangman, this also contains the reset fix for gen4
- rps sanity check from Chris - this papers over issues when the gpu fails
 to clock up on snb/ivb, and it is shockingly easy to hit. The code
 prints a big WARN backtrace and restores the hw to a sane state. The
 real fix is still in the works.

Atm I'm aware of 2 regressions in -next:
- One of the gmbus patches (not gmbus itself) regressed lvds detection on
 a MacbookPro. I've analyzed the bug already and I think I know what's
 going on, patch is awaiting test feedback.
- Just today QA reported that DP on ilk regressed. That bug is fresh of
 the press and still awaiting detailed logfiles and the bisect result.
 The only thing that's clear atm is that -fixes works and -next doesn't.
2012-05-11 17:42:41 +01:00
Rob Clark
b06d66be3b drm: pass dev to drm_vm_{open,close}_locked()
Previously these functions would assume that vma->vm_file was the
drm_file.  Although if in some cases if the drm driver needs to use
something else for the backing file (such as the tmpfs filp) then this
assumption is no longer true.  But vma->vm_private_data is still the
GEM object.

With this change, now the drm_device comes from the GEM object rather
than the drm_file so the driver is more free to play with vma->vm_file.

The scenario where this comes up is for mmap'ing of cached dmabuf's
for non-coherent systems, where the driver needs to use fault handling
and PTE shootdown to simulate coherency.  We can't use the vma->vm_file
of the dmabuf, which is using anon_inode's address_space.  The most
straightforward thing to do is to use the GEM object's obj->filp for
vma->vm_file in all cases, for which we need this patch.

Signed-off-by: Rob Clark <rob@ti.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:37:46 +01:00
Alan Cox
62363a4860 gma500: Turn on the IRQ for everything
Keep this as a patch of its own in case of bug reports.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:36:04 +01:00
Alan Cox
9c0b6fcdc9 gma500: clean up some more checks
We don't need to check these - they are always going to be the
same for any PVR based device.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:36:01 +01:00
Alan Cox
9aa65a2b9d cdv: Add all cedarview pci ids
Cover all D2xxx/N2xxx chips.

Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
[Hand applied to upstream driver]
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:54 +01:00
Alan Cox
31a0685a42 gma500: Clean up some of the noise
We have a lot of debug type stuff we don't actually need any more.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:52 +01:00
Alan Cox
213a84346f gma500: use the register map to clean up
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:51 +01:00
Alan Cox
8512e07487 gma500: introduce some register maps
All the conditional ugly register selection really wants to be
cleaned up. Use a struct describing each pipe and its registers.

This will also let us hide some of the oddments between platforms
for any future merging of bits together. In particular the way the
DPLL and FP registers randomly wander around.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:49 +01:00
Alan Cox
f693dfb72d gma500: Clean up from the psb_pipe structure
We have lots of local assignments that can now be eliminated

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:48 +01:00
Alan Cox
6256304ba3 gma500: introduce a structure describing each pipe
This starts the move away from lots of confused unions of per driver stuff
inherited when we merged the drivers together.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:47 +01:00
Alan Cox
a373bedd7e gma500: Fix build without ACPI
Reported-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-11 17:35:46 +01:00
Dave Airlie
d3029b4e03 drm/radeon/kms: fix warning on 32-bit in atomic fence printing
/ssd/git/drm-core-next/drivers/gpu/drm/radeon/radeon_fence.c: In function ‘radeon_debugfs_fence_info’:
/ssd/git/drm-core-next/drivers/gpu/drm/radeon/radeon_fence.c:606:7: warning: format ‘%lx’ expects argument of type ‘long unsigned int’, but argument 3 has type ‘long long int’ [-Wformat]

Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:56:24 +01:00
Jerome Glisse
f2e3922106 drm/radeon: make the ib an inline object
No need to malloc it any more.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:55 +01:00
Christian König
f237750f00 drm/radeon: remove r600 blit mutex v2
If we don't store local data into global variables
it isn't necessary to lock anything.

v2: rebased on new SA interface

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:54 +01:00
Jerome Glisse
68470ae7e6 drm/radeon: move the semaphore from the fence into the ib
It never really belonged there in the first place.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:49 +01:00
Christian König
7c0d409db5 drm/radeon: immediately free ttm-move semaphore
We can now protected the semaphore ram by a
fence, so free it immediately.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:46 +01:00
Jerome Glisse
c507f7ef30 drm/radeon: rip out the ib pool
It isn't necessary any more and the suballocator seems to perform
even better.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:41 +01:00
Jerome Glisse
a8c05940bd drm/radeon: simplify semaphore handling v2
Directly use the suballocator to get small chunks of memory.
It's equally fast and doesn't crash when we encounter a GPU reset.

v2: rebased on new SA interface.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:40 +01:00
Christian König
c3b7fe8b8a drm/radeon: multiple ring allocator v3
A startover with a new idea for a multiple ring allocator.
Should perform as well as a normal ring allocator as long
as only one ring does somthing, but falls back to a more
complex algorithm if more complex things start to happen.

We store the last allocated bo in last, we always try to allocate
after the last allocated bo. Principle is that in a linear GPU ring
progression was is after last is the oldest bo we allocated and thus
the first one that should no longer be in use by the GPU.

If it's not the case we skip over the bo after last to the closest
done bo if such one exist. If none exist and we are not asked to
block we report failure to allocate.

If we are asked to block we wait on all the oldest fence of all
rings. We just wait for any of those fence to complete.

v2: We need to be able to let hole point to the list_head, otherwise
    try free will never free the first allocation of the list. Also
    stop calling radeon_fence_signalled more than necessary.

v3: Don't free allocations without considering them as a hole,
    otherwise we might lose holes. Also return ENOMEM instead of ENOENT
    when running out of fences to wait for. Limit the number of holes
    we try for each ring to 3.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:39 +01:00
Jerome Glisse
0085c95061 drm/radeon: use one wait queue for all rings add fence_wait_any v2
Use one wait queue for all rings. When one ring progress, other
likely does to and we are not expecting to have a lot of waiter
anyway.

Also add a fence_wait_any that will wait until the first fence
in the fence array (one fence per ring) is signaled. This allow
to wait on all rings.

v2: some minor cleanups and improvements.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:38 +01:00
Christian König
557017a0e2 drm/radeon: define new SA interface v3
Define the interface without modifying the allocation
algorithm in any way.

v2: rebase on top of fence new uint64 patch
v3: add ring to debugfs output

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:37 +01:00
Christian König
2e0d99103e drm/radeon: make sa bo a stand alone object
Allocating and freeing it seperately.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:35 +01:00
Christian König
e6661a9664 drm/radeon: keep start and end offset in the SA
Instead of offset + size keep start and end offset directly.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:34 +01:00
Christian König
711a972933 drm/radeon: add sub allocator debugfs file
Dumping the current allocations.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:33 +01:00
Christian König
a651c55a0b drm/radeon: add proper locking to the SA v3
Make the suballocator self containing to locking.

v2: split the bugfix into a seperate patch.
v3: remove some unreleated changes.

Sig-off-by: Christian König <deathsimple@vodafone.de>

Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:32 +01:00
Christian König
dd8bea2111 drm/radeon: use inline functions to calc sa_bo addr
Instead of hacking the calculation multiple times.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:31 +01:00
Christian König
8a47cc9ec1 drm/radeon: rework locking ring emission mutex in fence deadlock detection v2
Some callers illegal called fence_wait_next/empty
while holding the ring emission mutex. So don't
relock the mutex in that cases, and move the actual
locking into the fence code.

v2: Don't try to unlock the mutex if it isn't locked.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:20 +01:00
Jerome Glisse
3b7a2b24ea drm/radeon: rework fence handling, drop fence list v7
Using 64bits fence sequence we can directly compare sequence
number to know if a fence is signaled or not. Thus the fence
list became useless, so does the fence lock that mainly
protected the fence list.

Things like ring.ready are no longer behind a lock, this should
be ok as ring.ready is initialized once and will only change
when facing lockup. Worst case is that we return an -EBUSY just
after a successfull GPU reset, or we go into wait state instead
of returning -EBUSY (thus delaying reporting -EBUSY to fence
wait caller).

v2: Remove left over comment, force using writeback on cayman and
    newer, thus not having to suffer from possibly scratch reg
    exhaustion
v3: Rebase on top of change to uint64 fence patch
v4: Change DCE5 test to force write back on cayman and newer but
    also any APU such as PALM or SUMO family
v5: Rebase on top of new uint64 fence patch
v6: Just break if seq doesn't change any more. Use radeon_fence
    prefix for all function names. Even if it's now highly optimized,
    try avoiding polling to often.
v7: We should never poll the last_seq from the hardware without
    waking the sleeping threads, otherwise we might lose events.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:19 +01:00
Jerome Glisse
bb63556729 drm/radeon: convert fence to uint64_t v4
This convert fence to use uint64_t sequence number intention is
to use the fact that uin64_t is big enough that we don't need to
care about wrap around.

Tested with and without writeback using 0xFFFFF000 as initial
fence sequence and thus allowing to test the wrap around from
32bits to 64bits.

v2: Add comment about possible race btw CPU & GPU, add comment
    stressing that we need 2 dword aligned for R600_WB_EVENT_OFFSET
    Read fence sequenc in reverse order of GPU write them so we
    mitigate the race btw CPU and GPU.

v3: Drop the need for ring to emit the 64bits fence, and just have
    each ring emit the lower 32bits of the fence sequence. We
    handle the wrap over 32bits in fence_process.

v4: Just a small optimization: Don't reread the last_seq value
    if loop restarts, since we already know its value anyway.
    Also start at zero not one for seq value and use pre instead
    of post increment in emmit, otherwise wait_empty will deadlock.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:17 +01:00
Christian König
d6999bc7b5 drm/radeon: replace the per ring mutex with a global one
A single global mutex for ring submissions seems sufficient.

Signed-off-by: Christian König <deathsimple@vodafone.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-05-09 17:22:14 +01:00