Pekka Paalanen [Sat, 28 Feb 2009 14:09:56 +0000 (16:09 +0200)]
drm: build fix for Linux drm_vblank_cleanup()
The commit
957b10695b619d6ed2f1098b00502395d9a3c149, "Move vblank_init
to driver load time." forgot to add the function declaration in
linux-core/drmP.h.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Robert Noland [Fri, 27 Feb 2009 23:52:08 +0000 (17:52 -0600)]
FreeBSD: Introduce a kernel tuneable to disable msi at boot time.
Robert Noland [Wed, 25 Feb 2009 23:59:26 +0000 (17:59 -0600)]
Move vblank_init to driver load time.
Eric Anholt [Fri, 27 Feb 2009 21:46:31 +0000 (13:46 -0800)]
intel: Update reloc_tree_size of the first buffer when we count the tree size.
This helps avoid the n^2 performance cost of counting tree size when we
get a lot of relocations into our batch buffer. rgb10text on keithp's laptop
went from 136k glyphs/sec to 234k glyphs/sec.
Robert Noland [Fri, 27 Feb 2009 05:51:57 +0000 (23:51 -0600)]
FreeBSD: Fix up some ioctl permissions issues missed many times over.
This was somehow hit with r600 demo.
Submitted by: Jung-uk Kim <jkim@FreeBSD.org>
Robert Noland [Wed, 25 Feb 2009 20:19:01 +0000 (14:19 -0600)]
i915: Backport jbarnes gm45 vblank counter patch.
Eric Anholt [Wed, 25 Feb 2009 05:36:56 +0000 (21:36 -0800)]
intel: Don't copy dirty data out when freeing a BO in the fake bufmgr.
Robert Noland [Tue, 24 Feb 2009 21:49:45 +0000 (15:49 -0600)]
FreeBSD: Drop the Giant lock.
Robert Noland [Tue, 24 Feb 2009 21:47:35 +0000 (15:47 -0600)]
FreeBSD: Turn on msi
There is a blacklist for devices that advertise the capability, but
don't work properly.
Robert Noland [Tue, 24 Feb 2009 20:21:10 +0000 (14:21 -0600)]
FreeBSD: Add some vblank related debugging and replace DRM_WAIT_ON with a local instance.
Robert Noland [Tue, 24 Feb 2009 20:00:21 +0000 (14:00 -0600)]
i915: A few whitespace cleanups.
Robert Noland [Tue, 24 Feb 2009 18:59:08 +0000 (12:59 -0600)]
linux: Add msi_enabled to the drm_device so the build should be happy.
Robert Noland [Tue, 24 Feb 2009 18:28:42 +0000 (12:28 -0600)]
radeon: Prepare radeon for msi support.
Robert Noland [Tue, 24 Feb 2009 08:00:42 +0000 (02:00 -0600)]
i915: This was part of a sync to the intel driver at some point
-Remove the old TTM interface
-Move register definitions to i915_reg.h
-Rework the irq handler
Kristian Høgsberg [Tue, 24 Feb 2009 16:33:34 +0000 (11:33 -0500)]
nouveau: Dist nouvea_dma.h as well.
Oops. Disting is hard.
Kristian Høgsberg [Tue, 24 Feb 2009 16:19:41 +0000 (11:19 -0500)]
nouveau: Also dist nouveau_private.h.
Robert Noland [Tue, 24 Feb 2009 06:22:56 +0000 (00:22 -0600)]
FreeBSD: Rip out the locked task support now that i915 no longer uses it.
Robert Noland [Tue, 24 Feb 2009 04:39:07 +0000 (22:39 -0600)]
i915: Rip out the use of vblank_swap
Robert Noland [Tue, 24 Feb 2009 02:22:44 +0000 (20:22 -0600)]
FreeBSD: There is no compelling reason to hold the lock here.
When I was lock profiling, this was high up on the list and I
see no reason to do it.
Robert Noland [Tue, 24 Feb 2009 02:16:15 +0000 (20:16 -0600)]
FreeBSD: Don't set PZERO in mtx_sleep.
Robert Noland [Tue, 24 Feb 2009 02:01:26 +0000 (20:01 -0600)]
FreeBSD: Set MAP_NOSYNC on mmaps.
There is no reason to gratuitously sync these maps to swap.
Eric Anholt [Mon, 23 Feb 2009 21:30:20 +0000 (13:30 -0800)]
Add libdrm_intel.pc by popular demand.
Ivan Kokshaysky [Mon, 23 Feb 2009 20:54:18 +0000 (15:54 -0500)]
Fix DRM_CAS() on Alpha. (#16549)
Kristian Høgsberg [Mon, 23 Feb 2009 20:08:03 +0000 (15:08 -0500)]
modetest: Print names of properties.
Pekka Paalanen [Sun, 22 Feb 2009 10:40:47 +0000 (12:40 +0200)]
Fix fix distcheck for optional nouveau stuff.
Tylo.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Eric Anholt [Sat, 21 Feb 2009 18:01:40 +0000 (10:01 -0800)]
Fix distcheck for optional nouveau stuff.
Eric Anholt [Sat, 21 Feb 2009 17:57:19 +0000 (09:57 -0800)]
Bump version to 2.4.5 for new API additions.
Eric Anholt [Wed, 18 Feb 2009 21:06:35 +0000 (13:06 -0800)]
intel: Add a new bufmgr alloc function to get BOs ready for rendering to.
This avoids using the oldest BO in the BO cache and waiting for it to be
idle before we turn around and render to it with the GPU. Thanks to
Chris Wilson for pointing out how silly we were being.
Ben Skeggs [Thu, 19 Feb 2009 23:25:35 +0000 (09:25 +1000)]
libdrm/nouveau: free drmVersion after we're done with it
Ben Skeggs [Thu, 19 Feb 2009 09:40:19 +0000 (19:40 +1000)]
libdrm/nouveau: fix dma debugging
Pekka Paalanen [Wed, 18 Feb 2009 20:46:40 +0000 (22:46 +0200)]
nouveau: support backlight only when kernel does
Loading nouveau.ko would fail with unknown symbols, if the backlight
class device support is not provided in the kernel. Let's make the
backlight support dependant on the kernel configuration.
This is a bit ugly, the proper way would be to check for the config in
Makefile.kernel whether to build nouveau_backlight.o at all, and if not,
nouveau_drv.h should provide the stubs.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Matthew Garrett [Wed, 18 Feb 2009 13:21:26 +0000 (13:21 +0000)]
nv40, nv50: fix backlight build for <2.6.29 kernels
Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org>
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Matthew Garrett [Wed, 18 Feb 2009 02:08:25 +0000 (02:08 +0000)]
nouveau: Add in-kernel backlight control support
Several nvidia-based systems don't support backlight control via the
standard ACPI control mechanisms. Instead, it's necessary for the driver
to modify the backlight control registers directly. This patch adds
support for determining whether the registers appear to be in use, and
if so registers a kernel backlight device to control them. The backlight
can then be controlled via existing userspace tools.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Pekka Paalanen [Sat, 14 Feb 2009 20:22:39 +0000 (22:22 +0200)]
drm_compat: remove kmap_atomic_prot_pfn()
This function is unused, and yet creates build problems: the symbol
init_mm is not exported by the latest -rc kernels and I don't believe it
ever will be. Even CONFIG_UNUSED_SYMBOLS does not provide it anymore.
If this function is needed in the future, it needs to be reinvented in
any case. So remove it.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Pekka Paalanen [Mon, 16 Feb 2009 18:43:02 +0000 (20:43 +0200)]
Remove i915 driver sources from linux-core
Intel developers have stated, that their DRM development continues
elsewhere in some Linux kernel trees. This makes the code in drm.git
just dead weight. This removal allows further cleanup of compatibility
code.
shared-core and bsd-core are left untouched this time.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Acked-by: Eric Anholt <eric@anholt.net>
Eric Anholt [Mon, 16 Feb 2009 02:35:03 +0000 (18:35 -0800)]
intel: Clear ioctl parameter outvalues and pads that confuse valgrind.
The minor CPU cost here is probably outweighed by bothering us with noise in
the tool.
Ben Skeggs [Sun, 15 Feb 2009 12:06:18 +0000 (22:06 +1000)]
nv40: fail completely if we don't have a ctxprog for the chipset
Ben Skeggs [Sun, 15 Feb 2009 11:52:19 +0000 (21:52 +1000)]
nv50: context info for chipset 0xa0
Ben Skeggs [Wed, 11 Feb 2009 04:18:03 +0000 (14:18 +1000)]
libdrm: make libdrm_nouveau conditional, disabled by default
libdrm isn't supposed to ship APIs not present in a released kernel.
Jakob Bornecrantz [Wed, 11 Feb 2009 15:43:20 +0000 (16:43 +0100)]
mode: Make xfdrmMode.[c|h] not depend on drm_mode.h
Pekka Paalanen [Wed, 4 Feb 2009 19:44:50 +0000 (21:44 +0200)]
nouveau: fix type warnings: void* arith, un/signed
nouveau_notifier.c had two places where void* was used in arithmetic,
fixed by using char*.
nouveau_dma_wait(), nouveau_notifier_wait_status() and
nouveau_resource_alloc() had signed/unsigned comparison warnings, fixed
by changing the function parameter into an unsigned type.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Ben Skeggs [Wed, 11 Feb 2009 00:48:36 +0000 (10:48 +1000)]
drm/nv50: fix nv9x chipsets
NVIDIA do this fun little sequence after updating the PRAMIN page tables.
On 9xxx chips, none of the PRAMIN BAR bindings (except the initial one)
worked, hence the majority of the setup needed to create a channel
ended up in the wrong place, causing all sorts of fun.
This is done by NVIDIA on nv8x chips also, so we'll do it for them too,
even though they appear to work without it.
Ben Skeggs [Wed, 11 Feb 2009 00:12:43 +0000 (10:12 +1000)]
drm/nv50: add context info for nv98
It won't work yet, just like the other 9xxx chips. Real soon now :)
Ben Skeggs [Mon, 9 Feb 2009 23:05:09 +0000 (09:05 +1000)]
drm/nv50: use a slightly different initial context for nv96
I'm not 100% sure that the nv94 one we were using won't work. The context
layouts are identical (well.. same ctxprog, so of course!), only a couple
of registers differ. But, be safe until we actually get some 9xxx chips
working.
Ben Skeggs [Mon, 9 Feb 2009 23:02:17 +0000 (09:02 +1000)]
drm/nv50: correct ramfc pointer in channel header
Suprisingly the card still worked without this...
Ben Skeggs [Mon, 9 Feb 2009 22:57:21 +0000 (08:57 +1000)]
drm/nv50: let the card handle the initial context switch
Our PFIFO/PGRAPH context save/load functions don't really work well
(at all?) on nv5x yet. Depending on what random state the card is
in before the drm loads, fbcon probably won't work correctly.
Luckily we've setup the GPU in such a way that it'll actually do a
hw context switch for the first context. Not sure of how successful
this'd be currently on the older chips (actually, pretty sure it won't
work), so NV50 only for now.
Stuart Bennett [Sat, 7 Feb 2009 21:20:17 +0000 (21:20 +0000)]
nouveau: don't try to traverse non-existent lists
Fixes nouveau_ioctl_mem_free Oops
Ben Skeggs [Thu, 5 Feb 2009 13:03:29 +0000 (23:03 +1000)]
nouveau: fix some issues where buffer objects never get freed
Kristian Høgsberg [Wed, 4 Feb 2009 17:17:13 +0000 (12:17 -0500)]
modetest: Add syntax to also specify a crtc when setting a mode.
Ben Skeggs [Fri, 30 Jan 2009 01:25:35 +0000 (11:25 +1000)]
nouveau: install libdrm_nouveau with libdrm
Ben Skeggs [Fri, 30 Jan 2009 01:18:54 +0000 (11:18 +1000)]
nouveau: bring in new mm api definitions, without the actual mm code
Use of the new bits is guarded with a mm_enabled=0 hardcode.
Kristian Høgsberg [Tue, 3 Feb 2009 20:03:41 +0000 (15:03 -0500)]
modetest: Use cairo (if available) to paint a better pattern.
Kristian Høgsberg [Tue, 3 Feb 2009 19:02:50 +0000 (14:02 -0500)]
modetest: Don't sleep when just dumping state, wait for key for modeset.
Kristian Høgsberg [Tue, 3 Feb 2009 19:00:00 +0000 (14:00 -0500)]
modetest: Handle setting modes on multiple crtcs with one fb.
Kristian Høgsberg [Tue, 3 Feb 2009 18:26:22 +0000 (13:26 -0500)]
modetest: Use a more interesting test pattern.
Stephane Marchesin [Mon, 2 Feb 2009 22:47:11 +0000 (23:47 +0100)]
Remove the "nv" driver.
Kristian Høgsberg [Mon, 2 Feb 2009 20:47:09 +0000 (15:47 -0500)]
No need to pin buffer anymore in modetest.c
Stuart Bennett [Thu, 29 Jan 2009 23:39:50 +0000 (23:39 +0000)]
nouveau: don't save channel context if it has recently become invalid
Bug exposed by DDX change
d9da090c
Stuart Bennett [Sat, 25 Oct 2008 19:59:38 +0000 (20:59 +0100)]
nouveau: no suspend support for nv50+
Eric Anholt [Wed, 28 Jan 2009 02:00:45 +0000 (18:00 -0800)]
Clean up README for the current state of kernel module affairs.
Bug #19608.
Eric Anholt [Wed, 28 Jan 2009 01:16:11 +0000 (17:16 -0800)]
intel: don't count fences on 965 and later, as they don't use them.
Eric Anholt [Wed, 28 Jan 2009 00:54:11 +0000 (16:54 -0800)]
intel: Fix under-counting of fences registers required in check_aperture.
Jesse Barnes [Fri, 23 Jan 2009 22:13:45 +0000 (14:13 -0800)]
intel: libdrm support for fence management in execbuf
This patch tries to use the available fence count to figure out whether a
given batch can succeed or not (just like the aperture check).
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Pekka Paalanen [Tue, 27 Jan 2009 20:39:50 +0000 (22:39 +0200)]
drm compat: fix euid for >=2.6.28
drm_fops.c reads the current process' EUID directly from task_struct.
Apparently starting in 2.6.28-rc4 this fails to build.
In Linus' tree, commit
b6dff3ec5e116e3af6f537d4caedcad6b9e5082a
"CRED: Separate task security context from task_struct"
moves the euid field from task_struct to another struct.
Earlier commit
9e2b2dc4133f65272a6d3c5dcb2ce63f8a87cae9
"CRED: Introduce credential access wrappers" implements the wrapper
macros to access e.g. euid. This is in 2.6.27-rc4, and this contains the
definition of current_euid() that will be used in the DRM compatibility header
for kernels before 2.6.27. That commit also creates <linux/cred.h>, which
contains the upstream definition of current_euid().
drm_fops.c is fixed to use current_euid(), and drm_compat.h will offer
the compatibility definition for kernels <2.6.27.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Ben Skeggs [Mon, 26 Jan 2009 22:36:33 +0000 (08:36 +1000)]
nv50: support chipset NV96
ctxprog seen in okias' trace identical to one we use on NV94, assuming
the initial context values for NV94 will work here too.
Ben Skeggs [Thu, 13 Nov 2008 02:04:21 +0000 (13:04 +1100)]
nv04-nv40: correct RAMHT size
Dave Airlie [Sun, 18 Jan 2009 20:46:23 +0000 (06:46 +1000)]
drm: remove drmstat/dristat from linux-core build
pointed out by pq
Eric Anholt [Thu, 15 Jan 2009 00:11:32 +0000 (16:11 -0800)]
bump version to 2.4.4
Eric Anholt [Mon, 5 Jan 2009 01:37:18 +0000 (17:37 -0800)]
intel: Retry pin ioctl on -EINTR.
Owain G. Ainsworth [Sat, 10 Jan 2009 18:48:46 +0000 (18:48 +0000)]
Don't use DRM_BO_FLAG_NO_MOVE in bufmgr fake. It's a ttm flag.
It's also unused, so worthless.
Owain G. Ainsworth [Sun, 11 Jan 2009 19:02:07 +0000 (19:02 +0000)]
Remove drmModeReplaceFb after it was removed from the kernel.
It is impossible to replace the original semantics of this call purely
in userland, since the fb_id would change.
after discussion with Dr_Jakob
Signed-Off-By: Owain Ainsworth <oga@openbsd.org>
Acked-By: Jakob Bornecrantz <jakob@vmware.com>
Ben Skeggs [Mon, 12 Jan 2009 00:16:07 +0000 (10:16 +1000)]
nv50: ack nsource to prevent continuous protection fault irqs
Jesse Barnes [Wed, 7 Jan 2009 19:47:52 +0000 (11:47 -0800)]
libdrm: add autoconf check for clock_gettime
Should be more portable this way.
Jesse Barnes [Wed, 7 Jan 2009 18:48:26 +0000 (10:48 -0800)]
libdrm: only check for vblank timeout if we caught EINTR
Michel caught a case where we might overwrite a success or other return
value with EBUSY, so check the return value before checking for the
timeout condition.
Jesse Barnes [Wed, 7 Jan 2009 18:18:08 +0000 (10:18 -0800)]
libdrm: add timeout handling to drmWaitVBlank
In some cases, vblank interrupts may be disabled or otherwise broken.
The kernel has a 3s timeout builtin to handle these cases, but the X
server's SIGALM for cursor handling may interrupt vblank wait ioctls,
causing libdrm to restart the ioctl, making the kernel's timeout
useless.
This change tracks time across ioctl restarts and returns EBUSY to the
caller if the expected vblank sequence doesn't occur within 1s of the
first call.
Fixes fdo bz #18041, which is caused by a drmWaitVBlank hanging due to
the corresponding pipe getting disabled (thus preventing further events
from coming in).
Jesse Barnes [Mon, 29 Dec 2008 17:21:58 +0000 (09:21 -0800)]
Remove executable from modeprint test dir
Accidentally committed it with the mode setting changes.
Robert Noland [Tue, 23 Dec 2008 18:56:23 +0000 (13:56 -0500)]
[FreeBSD] Fix build on FreeBSD after modesetting import.
Dave Airlie [Tue, 23 Dec 2008 01:22:44 +0000 (11:22 +1000)]
radeon: only write irq regs if irq is enabled
Eric Anholt [Wed, 17 Dec 2008 21:55:53 +0000 (13:55 -0800)]
intel: Rename plane[AB]* back to pipe[AB]*.
The values are really going to continue meaning pipe, not plane, and that's
what they're called in the kernel copy of the header. Userland hasn't ever
made the switch to pipe!=plane, since userland checks are based on DRM
version, which is still stuck at 1.6. However, Mesa did start using
plane[AB] names, so provide a compat define.
Eric Anholt [Wed, 17 Dec 2008 21:47:59 +0000 (13:47 -0800)]
intel: Remove linux build of i915 DRM, as it's unmaintained and a user trap.
The code's in the linux kernel.
Eric Anholt [Wed, 17 Dec 2008 21:45:01 +0000 (13:45 -0800)]
intel: Sync GEM ioctl comments for easier diffing against the kernel.
Kristian Høgsberg [Thu, 18 Dec 2008 05:02:43 +0000 (00:02 -0500)]
modetest: Print more stuff. Prettier.
Eric Anholt [Wed, 17 Dec 2008 18:41:21 +0000 (10:41 -0800)]
libdrm: Fix modetest/modeprint to use automake stuff.
Keith Packard [Mon, 15 Dec 2008 23:08:12 +0000 (15:08 -0800)]
intel: Cache tiling/swizzle state in user mode. Reset tiling on reuse.
Remember tiling mode values provided by appplications, and
record tiling mode when creating a buffer from another application. This
eliminates any need to ask the kernel for tiling values and also makes
reused buffers get the right tiling.
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Mon, 15 Dec 2008 23:08:17 +0000 (15:08 -0800)]
intel: Debug output %u vs uint64_t warning fix
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Sat, 13 Dec 2008 19:06:11 +0000 (11:06 -0800)]
intel: return error status from drm_intel_gem_bo_map
Applications may actually care if the mapping operation failed, so when
it happens, return an error indication. errno is probably trashed by
fprintf though.
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Fri, 21 Nov 2008 09:49:39 +0000 (01:49 -0800)]
intel: Dump out memory usage information when the kernel fails to pin
The execbuffer ioctl returns ENOMEM when it fails to pin all of the buffers
in the GTT. This is usually caused by the DRM client attempting to use too
much memory in a single request. Dumping out the requested and available
memory values should help point out failures in the DRM code to catch over
commitments of this form.
Signed-off-by: Keith Packard <keithp@keithp.com>
Jesse Barnes [Wed, 17 Dec 2008 18:09:49 +0000 (10:09 -0800)]
libdrm: add mode setting files
Add mode setting files to libdrm, including xf86drmMode.* and the new
drm_mode.h header. Also add a couple of tests to sanity check the
kernel interfaces and update code to support them.
Eric Anholt [Sun, 14 Dec 2008 22:39:24 +0000 (14:39 -0800)]
intel: don't skip set_domain on mapping of shared buffers.
Eric Anholt [Sun, 14 Dec 2008 22:35:48 +0000 (14:35 -0800)]
intel: don't let named buffers into the BO cache.
We wouldn't want some remaining 3D rendering to scribble on our batchbuffer.
Eric Anholt [Sun, 14 Dec 2008 22:32:09 +0000 (14:32 -0800)]
intel: Remove the mapped flag, which is adequately covered by bo_gem->virtual.
Jesse Barnes [Wed, 10 Dec 2008 23:47:28 +0000 (15:47 -0800)]
Revert "Merge branch 'modesetting-gem'"
This reverts commit
6656db10551bbb8770dd945b6d81d5138521f208.
We really just want the libdrm and ioctl bits, not all the driver
stuff.
Jesse Barnes [Wed, 10 Dec 2008 23:37:45 +0000 (15:37 -0800)]
Bump version to 2.4.3 to reflect addition of mode setting header
Jesse Barnes [Wed, 10 Dec 2008 23:37:29 +0000 (15:37 -0800)]
Update .gitignore with new modetest program
Jesse Barnes [Tue, 9 Dec 2008 18:23:43 +0000 (10:23 -0800)]
Merge branch 'modesetting-gem'
Dave Airlie [Mon, 8 Dec 2008 04:05:32 +0000 (14:05 +1000)]
libdrm: bump to 2.4 for lib version
we already have a 2.3.x version out there
Jesse Barnes [Wed, 3 Dec 2008 19:54:07 +0000 (11:54 -0800)]
Merge branch 'master' into modesetting-gem
Jesse Barnes [Wed, 3 Dec 2008 19:53:36 +0000 (11:53 -0800)]
Merge branch 'master' into modesetting-gem
Eric Anholt [Mon, 1 Dec 2008 19:49:11 +0000 (11:49 -0800)]
Bump version to 2.4.2 to signal new libdrm_intel functions.
Eric Anholt [Thu, 13 Nov 2008 19:44:22 +0000 (11:44 -0800)]
intel: Add a function for setting (GTT,GTT) domain, for use by UXA.
This function can also serve the role that the bo_wait_rendering did, when
write_enable is unset.