7bcce66a1d
Fix kernel crash when we hit OOM conditions.
...
(Alan Hourihane)
2008-03-12 10:07:56 +01:00
f1a681ebe5
Avoid duplicate calls to drm_ttm_bind in some cases.
2008-03-12 10:02:09 +01:00
fa1d88e3b2
Make sure other TTM memory types than TT is really unbound when evicted.
2008-03-12 09:56:06 +01:00
8a18d123f5
Avoid large kmallocs.
2008-03-12 09:49:27 +01:00
981f515e2b
drm: fix fd closing ordering.
...
If the master fd goes away before the aiglx fd, we try and get a lock
that actually doesn't exist.
2008-03-12 14:48:01 +10:00
cf1a2499ed
global hotplug events happen in the pipe A stat register,
...
they are not pipe A specific. Remove pipe B code.
2008-03-11 21:24:29 +00:00
903d9231d6
Add support for monitor hotplug signals/waits
...
Also adjust i915 irq handling as it follows the 16bit'ism's
of the i8xx series.
2008-03-11 20:30:25 +00:00
f13936f7fc
nouveau: move AGP reset to mem_init_agp
...
Also, power cycle PGRAPH when resetting AGP -- it seems to fix problems encountered by p0g on nv25
2008-03-11 16:45:35 +00:00
5a7f4b3074
drm: fix oops on unload.
...
if we are unloading the module, there is no master so therefore no lock
2008-03-11 16:05:26 +10:00
52748d1792
drm: hopefully fix cursors on 965
2008-03-11 13:23:33 +10:00
a7e6ca62ad
Merge branch 'modesetting-101' of ssh://git.freedesktop.org/git/mesa/drm into modesetting-101
2008-03-10 23:36:27 +01:00
a7dc4d08b9
rradeon_ms: rework fence code and bring radeon ms up to date
2008-03-10 23:35:07 +01:00
07ba3b7193
remove unneeded load call
2008-03-10 18:30:17 +10:00
2848f04861
Switch from PIPE_VBLANK to PIPE_EVENT interrupts.
...
My 965GM gets interrupts stuck when using the old PIPE_VBLANK interrupt.
Switch to the PIPE_EVENT interrupt mechanism, and set the PIPE*STAT
registers to use START_VBLANK on 965 and VBLANK on previous chips.
2008-03-08 00:04:30 -08:00
ce3733572e
drm/radeon: check sarea_priv exists
2008-03-08 08:30:30 +10:00
9f19e79f95
drm: we already worked out the pitch. multiplying by 4 is just madness..
2008-03-07 17:09:51 +11:00
1ccccbd4ce
nouveau: redo channel idle detection
...
Will hopefully work a bit better than previous code, which depended on
knowing the channel's most recent PUT value. Some chips always return
0 on reading these regs, and currently userspace is the only other entity
which knows the value.
2008-03-07 15:18:34 +11:00
cd924de029
nouveau: don't touch NV_USER regs on channel destroy.
...
Not only was this entirely pointless, it actually causes my NV30GL to
die randomly when channels are destroyed.
2008-03-07 15:18:34 +11:00
33cb42a9f7
make startup of Xorg smoother if the mode doesn't change.
...
just flip the framebuffer in when required.
2008-03-07 13:03:42 +11:00
cf28ca4212
actually turn the irq off
2008-03-07 13:03:32 +11:00
348d95e00b
worst merge effort ever
2008-03-07 12:25:26 +11:00
47b7ec71fe
drm/modesetting: fixup irq removal on exit
2008-03-07 12:15:38 +11:00
f96baf1e1e
Merge branch 'master' of ../../drm into modesetting-101
2008-03-07 11:59:21 +11:00
2540ea7dc6
flush_agp_mappings commit
2008-03-07 09:29:35 +10:00
ccae12a837
I really screwed up that merge somehow
2008-03-07 08:58:24 +10:00
add7d21c79
drm: fixup for new sysfs API
2008-03-07 08:56:20 +10:00
48a166af14
woah somehow got these upstream
2008-03-07 08:49:27 +10:00
92a30dd608
drm/bo: allow non-suser priv to add kernel BOs.
...
modprobe can be run with dropped capabilities we still want the kernel bos
to work.
2008-03-06 14:43:23 +10:00
fef1c93aa8
build fix
2008-03-05 10:35:47 +00:00
b87c7ff79e
Add property info.
...
fix bo handle
2008-03-05 10:35:46 +00:00
1a959a2095
Check mode before adding to EDID
2008-03-05 10:35:46 +00:00
8bfe29d9e4
Use ARRAY_SIZE
2008-03-05 10:35:46 +00:00
3ea1902be9
propogate failed fixups back up
2008-03-05 10:35:46 +00:00
5662934ee4
Fix connector description table
2008-03-05 10:35:46 +00:00
44a2209790
Merge branch 'master' of ../../drm into modesetting-101
...
Conflicts:
shared-core/drm.h
2008-03-06 05:39:07 +10:00
d5c0101252
ttm: make sure userspace can't destroy kernel create memory managers
...
this adds something to say the kernel initialised the memory region not
the userspace. and blocks userspace from deallocating kernel areas
2008-03-06 05:37:54 +10:00
180c9188f4
drm/ttm: add ioctl to get back memory managed area sized
...
taken from modesetting branch but could be useful outside it.
2008-03-06 05:31:50 +10:00
e00dea812d
Merge branch 'master' of ../../drm into modesetting-101
...
Conflicts:
linux-core/drmP.h
linux-core/drm_drv.c
linux-core/drm_proc.c
linux-core/drm_stub.c
linux-core/drm_sysfs.c
2008-03-06 05:26:23 +10:00
12574590cd
drm: reorganise minor number handling using code from modesetting branch
...
Rip out the whole head thing and replace it with an idr and drm_minor
structure.
2008-03-06 05:21:50 +10:00
a875821f7b
drm: Fix for non-coherent DMA PowerPC
...
This patch fixes bits of the DRM so to make the radeon DRI work on
non-cache coherent PCI DMA variants of the PowerPC processors.
It moves the few places that needs change to wrappers to that
other architectures with similar issues can easily add their
own changes to those wrappers, at least until we have more useful
generic kernel API.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2008-03-05 19:25:03 +10:00
638353103d
i915: Evict if relocatee buffer is CACHED_MAPPED before
...
writting relocations, otherwise the GPU probably sees some
inconsistent data. Fix fd.o bug#14656
2008-03-05 15:09:17 +08:00
f78cdac8e5
fixup previous merge
2008-03-05 15:28:59 +10:00
4dbf447f43
drm: fixup compat with old x.org drivers
2008-03-05 15:28:38 +10:00
4aa7efe398
libdrm: fix warnings in mode code
2008-03-05 10:41:54 +10:00
43891ff2d0
Merge remote branch 'origin/master' into modesetting-101
...
Conflicts:
linux-core/drm_compat.c
2008-03-05 10:37:02 +10:00
81db48536c
remove unused functions + include header file
2008-03-05 10:36:42 +10:00
a6a2f2c8c4
Clarify when WAIT_LAZY is relevant to users.
2008-03-04 13:45:41 -08:00
3332a0add6
Remove unused DRM_FENCE_FLAG_WAIT_IGNORE_SIGNALS.
2008-03-04 13:41:30 -08:00
d41846adb7
Clarify through the names what drm_ttm_alloc_pages() and friend actually did.
...
These are all about the page directory (pointers to pages) rather than the
actual pages backing the allocation.
2008-03-04 13:40:46 -08:00
eedf3fa2f0
Don't shortcut the info syscall for drmBOBusy on nonshareable objects.
...
This broke the results when you're trying to check if a buffer you dispatched
some time ago is done being rendered from.
2008-03-04 12:16:51 -08:00
Thomas Hellstrom
Dave Airlie
Alan Hourihane
Stuart Bennett
Jerome Glisse
Dave Airlie
Keith Packard
Dave Airlie
Ben Skeggs
Alan Hourihane
Benjamin Herrenschmidt
Xiang, Haihao
Eric Anholt