After a full device powerdown via the optimus power switch, we seem
to lose the HDMI device completely on power on, this keep track of
whether we had a hdmi audio sub function device at power on, and
pokes a magic register to make it reappear after the optimus power
switch is thrown.
This at least
Am Mittwoch, den 24.07.2013, 17:13 +1000 schrieb Dave Airlie:
> After a full device powerdown via the optimus power switch, we seem
> to lose the HDMI device completely on power on, this keep track of
keep*s*
> whether we had a hdmi audio sub function device at power on, and
> pokes a magic regis
On Wed, Jul 24, 2013 at 8:04 AM, Daniel Vetter wrote:
> This is the 2nd attempt, I've always been a bit dissatisified with the
> tricky nature of the first one:
>
> http://lists.freedesktop.org/archives/dri-devel/2012-July/025451.html
>
> The issue is that the flink ioctl can race with calling gem
https://bugs.freedesktop.org/show_bug.cgi?id=60879
--- Comment #39 from Hristo Venev ---
What is the difference between OpenCL integer division and OpenGL shader
integer division?
--
You are receiving this mail because:
You are the assignee for the bug.
_
Hi Al
Any comment on this?
Regards
David
On Thu, Jul 11, 2013 at 1:45 AM, David Herrmann wrote:
> DRM core shares a single address_space across all inodes that belong to
> the same DRM device. This allows efficient unmapping of user-space
> mappings during buffer destruction. However, there is
https://bugs.freedesktop.org/show_bug.cgi?id=67002
--- Comment #8 from jackdac...@gmail.com ---
ok,so the atombios error is fixed now
what would the next step be to gradually fix this ? (so the uvd error after
resume & gpu lock/hang)
switch to gcc 4.7* ? would that make a change ?
the gcc 4.8*/
https://bugs.freedesktop.org/show_bug.cgi?id=63599
--- Comment #22 from Paul Wolneykien ---
(In reply to comment #21)
> That just disables the use of vram for exa pixmaps. You can accomplish the
> same thing by adding:
> Option "EXAPixmaps" "false"
> to the device section of your xorg.conf
Ye
The usage of strict_strtol() is not preferred, because
strict_strtol() is obsolete. Thus, kstrtol() should be
used.
Signed-off-by: Jingoo Han
---
drivers/gpu/drm/nouveau/nouveau_pm.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=47941
BugLink: https://bugs.launchpad.net/bugs/1163720
BugLink: https://bugs.launchpad.net/bugs/1162026
Some machines suffer from non-functional backlight controls if
BLM_PCH_PWM_ENABLE is set, so provide a quirk to avoid doing so.
Apply this qu
This patch adds a dt-binding document for exynos rotator. It describes which
nodes should be defined to use the rotator.
Signed-off-by: Chanho Park
Signed-off-by: Kyungmin Park
---
.../bindings/drm/exynos/samsung-rotator.txt| 35
1 file changed, 35 insertions(+)
This patch adds a device node of rotator for exynos4 platform. It has proper
register and clock information. It also has limit table to get restrictions of
the image size.
Signed-off-by: Chanho Park
Signed-off-by: Kyungmin Park
---
arch/arm/boot/dts/exynos4.dtsi | 23 +++
This patchset includes dt support for exynos rotator.
The exynos4 platform is only dt-based since 3.10, we should convert driver data
and ids to dt-based parsing methods. The rotator driver has a limit table to
get size limit. The minimum size of RGB888 format is 8 x 8 and maximum size is
8K x 8K.
The exynos4 platform is only dt-based since 3.10, we should convert driver data
and ids to dt-based parsing methods. The rotator driver has a limit table to get
size limit. The minimum size of RGB888 format is 8 x 8 and maximum size is 8K x
8K. The other format, YCbCr420 2-Plane has 32 x 32 min siz
On Mon, Jul 22, 2013 at 07:49:26AM +0100, Chanho Park wrote:
> This patch adds a dt-binding document for exynos rotator. It describes which
> nodes should be defined to use the rotator.
>
> Signed-off-by: Chanho Park
> Signed-off-by: Kyungmin Park
> ---
> .../bindings/drm/exynos/samsung-rotator
Hi Mark,
Thank you for your review.
> -Original Message-
> From: Mark Rutland [mailto:mark.rutl...@arm.com]
> Sent: Monday, July 22, 2013 5:48 PM
> To: Chanho Park
> Cc: inki@samsung.com; kgene@samsung.com; linux-samsung-
> s...@vger.kernel.org; jy0922.s...@samsung.com; devicetree-
> -Original Message-
> From: Inki Dae [mailto:inki@samsung.com]
> Sent: Tuesday, July 23, 2013 10:36 AM
> To: 'Chanho Park'; 'Mark Rutland'; 'Chanho Park'
> Cc: kgene@samsung.com; linux-samsung-...@vger.kernel.org;
> jy0922.s...@samsung.com; sw0312@samsung.com; dri-
> de...@list
devm_ioremap_resource does sanity checks on the given resource. No need to
duplicate this in the driver.
Signed-off-by: Wolfram Sang
---
Please apply via the subsystem-tree.
drivers/gpu/host1x/drm/hdmi.c |3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/gpu/host1x/drm/hdmi.c b/dr
The patch below does not apply to the 3.10-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to .
thanks,
greg k-h
-- original commit in Linus's tree --
>
Here is another bit of cleaning up the devm usage. It is again removing the
resource check with devm_ioremap_resource, because a) new drivers came in and
b) coccinelle had a bug and missed to find a couple of occasions. Unlike last
time, I think it is better if these patches go in via the subsystem
On 17 July 2013 11:19, Michel Dänzer wrote:
> On Mit, 2013-07-17 at 10:58 +0100, Sami Kerola wrote:
>> On 15 July 2013 11:35, Michel Dänzer wrote:
>> > On Son, 2013-07-14 at 13:26 +0100, Sami Kerola wrote:
>> >>
>> >> Jul 14 12:51:31 kerolasa-home kernel: radeon :00:01.0: GPU lockup
>> >> CP
nr_to_scan needs to be (unsigned) long, not int, otherwise we get negative
values on 32bit systems during shrink resulting in lots of messages like one
below.
[ 4078.989603] shrink_slab: i915_gem_inactive_scan+0x0/0xc0 negative objects to
delete nr=-289580136
Introduced in "drivers: convert shri
On Tue, Jul 23, 2013 at 11:18:06PM +0400, Artem Savkov wrote:
> nr_to_scan needs to be (unsigned) long, not int, otherwise we get negative
> values on 32bit systems during shrink resulting in lots of messages like one
> below.
>
> [ 4078.989603] shrink_slab: i915_gem_inactive_scan+0x0/0xc0 negativ
Since Exynos DRM drivers do not support module build,
remove module.h header file inclusion from files that do
not have any users.
Signed-off-by: Sachin Kamat
---
drivers/gpu/drm/exynos/exynos_ddc.c |1 -
drivers/gpu/drm/exynos/exynos_drm_fimc.c|1 -
drivers/gpu/drm/exynos/ex
On Wed, Jul 24, 2013 at 11:02:02AM +0200, Daniel Vetter wrote:
> On Wed, Jul 24, 2013 at 8:04 AM, Daniel Vetter wrote:
> > This is the 2nd attempt, I've always been a bit dissatisified with the
> > tricky nature of the first one:
> >
> > http://lists.freedesktop.org/archives/dri-devel/2012-July/02
On Wed, Jul 24, 2013 at 05:13:42PM +1000, Dave Airlie wrote:
> After a full device powerdown via the optimus power switch, we seem
> to lose the HDMI device completely on power on, this keep track of
> whether we had a hdmi audio sub function device at power on, and
> pokes a magic register to make
On Tue, Jul 16, 2013 at 09:12:09AM +0200, Daniel Vetter wrote:
> The gem flink name holds a reference onto the object itself, and this
> self-reference would prevent an flink'ed object from every being
> freed. To break that loop we remove the flink name when the last
> userspace handle disappears,
On Tue, Jul 23, 2013 at 6:22 PM, Antti Palosaari wrote:
> Hello
> I just upgraded Kernel from 3.10.0-rc6+ to 3.11.0-rc2+ and ran serious
> problems. That Kernel is current Linux Media master, which contains some
> upcoming media stuff that should not still has any effect, but lets mention
> for su
On Tue, Jul 23, 2013 at 2:59 PM, Sami Kerola wrote:
> On 17 July 2013 11:19, Michel Dänzer wrote:
>> On Mit, 2013-07-17 at 10:58 +0100, Sami Kerola wrote:
>>> On 15 July 2013 11:35, Michel Dänzer wrote:
>>> > On Son, 2013-07-14 at 13:26 +0100, Sami Kerola wrote:
>>> >>
>>> >> Jul 14 12:51:31 ker
Hi
This series fixes unplugging of DRM devices. It introduces four new helpers
which replace the old drm_put_dev(), drm_unplug_dev(), drm_fill_in_dev()
helpers:
- drm_dev_alloc()
Allocate a DRM device, initialize all static objects and fill in driver data.
The device is not registered nor a
Instead of managing device allocation+initialization in each bus-driver,
we should do that in a central place. drm_fill_in_dev() already does most
of it, but also requires the global drm lock for partial AGP device
registration.
Split both apart so we have a clean device initialization/allocation
All bus drivers do device setup themselves. This requires us to adjust all
of them if we introduce new core features. Thus, merge all these into a
uniform drm_dev_register() helper.
Note that this removes the drm_lastclose() error path for AGP as it is
horribly broken. Moreover, no bus driver call
Try to keep all functions that handle DRM file_operations into drm_fops.c
so internal helpers can be marked static later.
This makes the split between the 3 core files more obvious:
- drm_stub.c: DRM device allocation/destruction and management
- drm_fops.c: DRM file_operations (except for ioctl
The error paths in DRM bus drivers currently leak memory as they don't
correctly revert drm_dev_alloc(). Introduce drm_dev_free() to free DRM
devices which haven't been registered, yet.
We must be careful not to introduce any side-effects with cleanups done in
drm_dev_free(). drm_ht_remove(), drm_
Analog to drm_dev_register(), we now provide drm_dev_unregister() which
does the reverse. drm_dev_put() is still in place and combines the calls
to drm_dev_unregister() and drm_dev_free() so buses don't have to change.
*_get() and *_put() are used for reference-counting in the kernel.
However, drm
DRM could make great use of percpu-rwsems to track active users and wait
for them during hw hotplugging. Export symbols and allow using them in
runtime loadable modules.
Signed-off-by: David Herrmann
---
lib/Makefile | 2 +-
lib/percpu-rwsem.c | 7 +++
2 files changed, 8 insertions(+),
If we want to support real hotplugging, we need to block new syscalls from
entering any drm_* fops. We also need to wait for these to finish before
unregistering a device.
This patch introduces drm_dev_get/put_active() which mark a device as
active during file-ops. If a device is unplugged, drm_de
Introduce two new helpers, drm_agp_clear() and drm_agp_destroy() which
clear all AGP mappings and destroy the AGP head. This allows to reduce the
AGP code in core DRM and move it all to drm_agpsupport.c.
Also use these new helpers to clean up AGP allocations in the
drm_dev_register() error path.
Our current unplug-helpers remove all nodes from user-space and mark the
device as unplugged. However, any pending user-space call is still
continued. We wait for the last close() call to actually unload the
device.
This patch uses the drm_dev_get_active() infrastructure to allow immediate
unplugg
Op 24-07-13 15:43, David Herrmann schreef:
> If we want to support real hotplugging, we need to block new syscalls from
> entering any drm_* fops. We also need to wait for these to finish before
> unregistering a device.
>
> This patch introduces drm_dev_get/put_active() which mark a device as
> ac
Hi
On Wed, Jul 24, 2013 at 4:03 PM, Maarten Lankhorst
wrote:
> Op 24-07-13 15:43, David Herrmann schreef:
>> If we want to support real hotplugging, we need to block new syscalls from
>> entering any drm_* fops. We also need to wait for these to finish before
>> unregistering a device.
>>
>> This
Op 24-07-13 16:24, David Herrmann schreef:
> Hi
>
> On Wed, Jul 24, 2013 at 4:03 PM, Maarten Lankhorst
> wrote:
>> Op 24-07-13 15:43, David Herrmann schreef:
>>> If we want to support real hotplugging, we need to block new syscalls from
>>> entering any drm_* fops. We also need to wait for these t
https://bugs.freedesktop.org/show_bug.cgi?id=67002
--- Comment #9 from Andre Heider ---
Maybe unrelated to this bug report, but with today's Linus' master
(04012e3076943f17c8cfc77a196701c8f0b8964f) the only suspend/resume issue left
is that I get:
[drm:rv770_stop_dpm] *ERROR* Could not force DPM
On Tue, Jul 23, 2013 at 09:54:57AM -0700, gre...@linuxfoundation.org wrote:
>
> The patch below does not apply to the 3.10-stable tree.
You can drop it. The patch is already in 3.10 and it does not need to
be backported to earlier trees.
There was a snaffu were I thought David would post this in
On Tue, Jul 23, 2013 at 02:47:13PM +0200, David Herrmann wrote:
> If we want to map GPU memory into user-space, we need to linearize the
> addresses to not confuse mm-core. Currently, GEM and TTM both implement
> their own offset-managers to assign a pgoff to each object for user-space
> CPU access
https://bugs.freedesktop.org/show_bug.cgi?id=61182
--- Comment #51 from Nikoli ---
This bug is not fixed, still happens with kwin-4.10.4 and mesa-9.1.4
https://bugs.kde.org/show_bug.cgi?id=322773
https://bugs.gentoo.org/show_bug.cgi?id=476606
--
You are receiving this mail because:
You are the
https://bugs.freedesktop.org/show_bug.cgi?id=61182
--- Comment #52 from Alex Deucher ---
(In reply to comment #51)
> This bug is not fixed, still happens with kwin-4.10.4 and mesa-9.1.4
> https://bugs.kde.org/show_bug.cgi?id=322773
> https://bugs.gentoo.org/show_bug.cgi?id=476606
I think you are
Realized this today while booting a ThinkPad T420 with integrated intel graphic
:
Jul 24 17:36:10 n22 kernel: [ cut here ]
Jul 24 17:36:10 n22 kernel: WARNING: at
drivers/gpu/drm/i915/intel_display.c:1656 ironlake_crtc_disable+0x7ce/0x800
[i915]()
Jul 24 17:36:10 n22 ker
On Tue, Jul 23, 2013 at 02:47:14PM +0200, David Herrmann wrote:
> Use the new vma manager instead of the old hashtable. Also convert all
> drivers to use the new convenience helpers. This drops all the
> (map_list.hash.key << PAGE_SHIFT) non-sense.
>
> Locking and access-management is exactly the
Hello Maarten Lankhorst,
This is a semi-automatic email about new static checker warnings.
The patch 0108bc808107: "drm/nouveau: do not allow negative sizes for
now" from Jul 7, 2013, leads to the following Smatch complaint:
drivers/gpu/drm/nouveau/nouveau_bo.c:222 nouveau_bo_new()
war
On Tue, Jul 23, 2013 at 02:47:15PM +0200, David Herrmann wrote:
> Use the new vma-manager infrastructure. This doesn't change any
> implementation details as the vma-offset-manager is nearly copied 1-to-1
> from TTM.
>
> The vm_lock is moved into the offset manager so we can drop it from TTM.
> Du
On Tue, Jul 23, 2013 at 02:47:16PM +0200, David Herrmann wrote:
> Instead of unmapping the nodes in TTM and GEM users manually, we provide
> a generic wrapper which does the correct thing for all vma-nodes.
>
> v2: remove bdev->dev_mapping test in ttm_bo_unmap_virtual_unlocked() as
> ttm_mem_io_fr
Hello Ben Skeggs,
The patch 060810d7abaa: "drm/nouveau: fix locking issues in page
flipping paths" from Jul 8, 2013, leads to the following Smatch warning:
"drivers/gpu/drm/nouveau/nouveau_display.c:599
nouveau_crtc_page_flip()
error: double unlock 'mutex:&chan->cli->mutex'"
drivers/gp
Hi
On Wed, Jul 24, 2013 at 5:35 PM, Daniel Vetter wrote:
> On Tue, Jul 23, 2013 at 02:47:13PM +0200, David Herrmann wrote:
>> If we want to map GPU memory into user-space, we need to linearize the
>> addresses to not confuse mm-core. Currently, GEM and TTM both implement
>> their own offset-manag
Hello Daniel Vetter,
The patch 25f397a429df: "drm/crtc-helper: explicit DPMS on after
modeset" from Jul 19, 2013, leads to the following warning:
"drivers/gpu/drm/drm_crtc_helper.c:681 drm_crtc_helper_set_config()
info: ignoring unreachable code."
drivers/gpu/drm/drm_crtc_helper.c
67
Hi
On Wed, Jul 24, 2013 at 5:52 PM, Daniel Vetter wrote:
> On Tue, Jul 23, 2013 at 02:47:14PM +0200, David Herrmann wrote:
>> Use the new vma manager instead of the old hashtable. Also convert all
>> drivers to use the new convenience helpers. This drops all the
>> (map_list.hash.key << PAGE_SHIF
https://bugs.freedesktop.org/show_bug.cgi?id=67265
Priority: medium
Bug ID: 67265
Assignee: dri-devel@lists.freedesktop.org
Summary: r600g causes KWin crashes with kernel 3.9
Severity: normal
Classification: Unclassified
OS:
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #1 from Nikoli ---
Created attachment 82948
--> https://bugs.freedesktop.org/attachment.cgi?id=82948&action=edit
kwin gdb
--
You are receiving this mail because:
You are the assignee for the bug.
__
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #2 from Nikoli ---
Created attachment 82949
--> https://bugs.freedesktop.org/attachment.cgi?id=82949&action=edit
glxgears gdb
--
You are receiving this mail because:
You are the assignee for the bug.
__
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #3 from Nikoli ---
Created attachment 82950
--> https://bugs.freedesktop.org/attachment.cgi?id=82950&action=edit
/var/log/Xorg.0.log.old (created during crash)
--
You are receiving this mail because:
You are the assignee for the b
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #4 from Nikoli ---
Created attachment 82951
--> https://bugs.freedesktop.org/attachment.cgi?id=82951&action=edit
dmesg
--
You are receiving this mail because:
You are the assignee for the bug.
_
https://bugs.freedesktop.org/show_bug.cgi?id=61182
--- Comment #53 from Nikoli ---
Done, bug #67265
--
You are receiving this mail because:
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freed
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #5 from Alex Deucher ---
Is this a regression? Did kwin work properly with an older kernel or mesa? If
so, can you narrow down what component caused the problem and then bisect?
--
You are receiving this mail because:
You are the
https://bugs.freedesktop.org/show_bug.cgi?id=67265
--- Comment #6 from Nikoli ---
No, not regression. Older mesa and kernels had same problem.
--
You are receiving this mail because:
You are the assignee for the bug.
___
dri-devel mailing list
dri-dev
If we want to map GPU memory into user-space, we need to linearize the
addresses to not confuse mm-core. Currently, GEM and TTM both implement
their own offset-managers to assign a pgoff to each object for user-space
CPU access. GEM uses a hash-table, TTM uses an rbtree.
This patch provides a unif
Use the new vma manager instead of the old hashtable. Also convert all
drivers to use the new convenience helpers. This drops all the
(map_list.hash.key << PAGE_SHIFT) non-sense.
Locking and access-management is exactly the same as before with an
additional lock inside of the vma-manager, which st
Use the new vma-manager infrastructure. This doesn't change any
implementation details as the vma-offset-manager is nearly copied 1-to-1
from TTM.
The vm_lock is moved into the offset manager so we can drop it from TTM.
During lookup, we use the vma locking helpers to take a reference to the
found
Instead of unmapping the nodes in TTM and GEM users manually, we provide
a generic wrapper which does the correct thing for all vma-nodes.
v2: remove bdev->dev_mapping test in ttm_bo_unmap_virtual_unlocked() as
ttm_mem_io_free_vm() does nothing in that case (io_reserved_vm is 0).
v4: Fix docbook c
On 07/24/2013 04:24 PM, Alex Deucher wrote:
On Tue, Jul 23, 2013 at 6:22 PM, Antti Palosaari wrote:
Hello
I just upgraded Kernel from 3.10.0-rc6+ to 3.11.0-rc2+ and ran serious
problems. That Kernel is current Linux Media master, which contains some
upcoming media stuff that should not still ha
https://bugs.freedesktop.org/show_bug.cgi?id=67002
--- Comment #10 from jackdac...@gmail.com ---
(In reply to comment #9)
> Maybe unrelated to this bug report, but with today's Linus' master
> (04012e3076943f17c8cfc77a196701c8f0b8964f) the only suspend/resume issue
> left is that I get:
>
> [drm:
Daniel Vetter writes:
> Matching tiling modes is actually already a requirement on gen4+ (since
> the tiling bit and the linear/tiled offset registers can't be changed with
> a MI_DISPLAY_FLIP command).
Async flip has a harder requirement -- you must use X tiling, both
before and after the flip.
https://bugs.freedesktop.org/show_bug.cgi?id=67276
Priority: medium
Bug ID: 67276
Assignee: dri-devel@lists.freedesktop.org
Summary: kernel-3.11 [drm:r600_uvd_ring_test] *ERROR* radeon:
ring 5 test failed (0xCAFEDEAD)
Severit
On Wed, Jul 24, 2013 at 01:26:32PM -0700, Keith Packard wrote:
> Daniel Vetter writes:
>
> > Matching tiling modes is actually already a requirement on gen4+ (since
> > the tiling bit and the linear/tiled offset registers can't be changed with
> > a MI_DISPLAY_FLIP command).
>
> Async flip has a
https://bugs.freedesktop.org/show_bug.cgi?id=67002
--- Comment #11 from jackdac...@gmail.com ---
I had to restart due to some other issue
and it wouldn't "successfully" resume anymore:
the lag/slowdown was there again
forcing performance_level to "low" or "auto" didn't seem to make a differenc
https://bugs.freedesktop.org/show_bug.cgi?id=67276
--- Comment #1 from Alex Deucher ---
Does it only happen when you enable dpm?
--
You are receiving this mail because:
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freed
https://bugs.freedesktop.org/show_bug.cgi?id=67276
--- Comment #2 from Joshua Cov. ---
(In reply to comment #1)
> Does it only happen when you enable dpm?
I'm not sure. I haven't tried with dpm disabled
--
You are receiving this mail because:
You are the assignee for the bug.
_
https://bugs.freedesktop.org/show_bug.cgi?id=66963
--- Comment #7 from Sergey ---
Some findings:
The actual hand happens in
drivers/gpu/drm/radeon/rv6xx_dpm.c:
int rv6xx_dpm_enable(struct radeon_device *rdev)
on 'r600_start_dpm(rdev)' call.
Makes sense that configurations are harmless before act
https://bugs.freedesktop.org/show_bug.cgi?id=67283
Priority: medium
Bug ID: 67283
Assignee: dri-devel@lists.freedesktop.org
Summary: VDPAU doesn't work on hybrid laptop through DRI_PRIME
Severity: normal
Classification: Unclassified
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #1 from Hohahiu ---
Created attachment 82969
--> https://bugs.freedesktop.org/attachment.cgi?id=82969&action=edit
Xorg.0.log
--
You are receiving this mail because:
You are the assignee for the bug.
___
Okay as I warned, the qxl driver was running a bit free and loose with its
ttm object reservations and the new lockdep enabled reservation tracking
shone a bright light into it, it also with the new reservations mutexes
hits a possible deadlock during boot.
the first patch is a real fix to rende
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #2 from Martin Peres ---
For the records, we have the exact same problem with nouveau.
--
You are receiving this mail because:
You are the assignee for the bug.
___
dri-devel mailing list
Daniel Vetter writes:
> We could just unconditionally increase the alignement in
> intel_pin_and_fence_fb_obj - we already have more strict requirements due
> to a bunch of w/a in other places. So shouldn't hurt at all really.
That seems like a fine plan; 32kB isn't that onerous. Do you want the
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #3 from Dave Airlie ---
I can make vdpauinfo work with two patches, but I haven't looked at making
anything play yet!
one patch is for libvdpau and one patch for mesa.
--
You are receiving this mail because:
You are the assignee fo
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #4 from Dave Airlie ---
Created attachment 82970
--> https://bugs.freedesktop.org/attachment.cgi?id=82970&action=edit
mesa patch to make vl pick the right device
--
You are receiving this mail because:
You are the assignee for the
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #5 from Dave Airlie ---
Created attachment 82971
--> https://bugs.freedesktop.org/attachment.cgi?id=82971&action=edit
libvdpau patch to pick the correct driver.
--
You are receiving this mail because:
You are the assignee for the
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #6 from Martin Peres ---
Thanks airlied, it works for me.
However, on KWin, I have to minimize and then maximize again the video in order
to have the video rolling. Otherwise, I only see an empty/black window.
--
You are receiving
https://bugs.freedesktop.org/show_bug.cgi?id=67283
--- Comment #7 from Hohahiu ---
I've tried both patches. It seems like everything works pretty well. So
vdpainfo now shows the supported formats. The 1080p video plays without any
problem (I also use kwin, xrender in particular) in mplayer. Top s
On Tue, Jul 23, 2013 at 10:43 PM, Rob Clark wrote:
> On Tue, Jul 23, 2013 at 8:31 AM, Daniel Vetter
> wrote:
>> On Tue, Jul 23, 2013 at 2:07 PM, Rob Clark wrote:
>>> On Tue, Jul 16, 2013 at 3:11 AM, Daniel Vetter
>>> wrote:
Only one callsite and since ->handle_count is not a simple refer
https://bugzilla.kernel.org/show_bug.cgi?id=60381
--- Comment #30 from rafael castillo ---
ok i tried the latest patches in drm-fixes and the crashes seemed to stop but i
can't get the gpu to reclock
i tried /sys/class/drm/card0/device/power_dpm_state to performance alone and it
never scales fro
https://bugzilla.kernel.org/show_bug.cgi?id=60381
--- Comment #31 from rafael castillo ---
uvd reclocks fine but the desktop flicker when it does
--
You are receiving this mail because:
You are watching the assignee of the bug.
Okay these are a bit large but I'd like to put them into -fixes ASAP,
as qxl totally fails on -rc1/2 due to the lockdep and reservation warnings,
The first patch is just a fix for broken rendering while rendering from
an irq context, it also fixes sleep while atomic issues.
The second patch is ju
From: Dave Airlie
Due to the nature of qxl hw we cannot queue operations while in an irq
context, so we queue these operations as best we can until atomic allocations
fail, and dequeue them later in a work queue.
Daniel looked over the locking on the list and agrees it should be sufficent.
The
From: Dave Airlie
In order to fix an issue with reservations we need to create the releases
as pre-pinned objects, this changes the placement interface and bo creation
interface to allow creating pinned objects to save nested reservations later.
This is just a stepping stone to main fix which fo
From: Dave Airlie
The recent addition of lockdep support to reservations and their subsequent
use by TTM showed up a number of potential problems with the way qxl was using
TTM objects.
a) it was allocating objects, and reserving them later without validating
underneath the reservation, which me
https://bugzilla.kernel.org/show_bug.cgi?id=60381
--- Comment #32 from Alex Deucher ---
The dynamic re-clocking doesn't currently work reliably on SI asics.
--
You are receiving this mail because:
You are watching the assignee of the bug.
https://bugzilla.kernel.org/show_bug.cgi?id=60381
--- Comment #33 from rafael castillo ---
well step by step is getting better, im happy enough to get stable desktop now
and since my card render KDE like an monster reclocking is not uber important
for me right now.
thanks for an awesome job ;)
bug.
-- next part --
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/dri-devel/attachments/20130724/98027fa8/attachment.html>
On Wed, Jul 24, 2013 at 2:00 AM, Dave Airlie wrote:
> On Tue, Jul 23, 2013 at 10:43 PM, Rob Clark wrote:
>> On Tue, Jul 23, 2013 at 8:31 AM, Daniel Vetter
>> wrote:
>>> On Tue, Jul 23, 2013 at 2:07 PM, Rob Clark wrote:
On Tue, Jul 16, 2013 at 3:11 AM, Daniel Vetter
wrote:
> Onl
On Wed, Jul 24, 2013 at 4:00 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> Due to the nature of qxl hw we cannot queue operations while in an irq
> context, so we queue these operations as best we can until atomic allocations
> fail, and dequeue them later in a work queue.
>
> Daniel looked over
This is the 2nd attempt, I've always been a bit dissatisified with the
tricky nature of the first one:
http://lists.freedesktop.org/archives/dri-devel/2012-July/025451.html
The issue is that the flink ioctl can race with calling gem_close on
the last gem handle. In that case we'll end up with a z
This patch makes g2d power domain and clock to be controlled
through pm runtime interfaces instead of controlling them
respectively.
Signed-off-by: Inki Dae
Signed-off-by: Kyungmin Park
---
drivers/gpu/drm/exynos/exynos_drm_g2d.c | 40 +-
1 files changed, 28 insert
1 - 100 of 165 matches
Mail list logo
