Re: release ex mode after hw_init

Yes, that's fine now On Mon, Nov 6, 2017 at 6:01 PM, Felix Kuehling wrote: > [+Oded] > > Hi Pixel, > > Thanks, this looks a lot cleaner. The series is Reviewed-by: Felix > Kuehling . Oded, are you OK with this as well? > > Regards, > Felix > > >

Re: [PATCH 1/1] drm/amdkfd: CWSR and trap handler support

Yes, that would be helpful to review. I assume the asm file provided is just a reference code and doesn't need to be in the kernel tree for this to work, correct ? On Thu, Nov 9, 2017 at 4:46 AM, Felix Kuehling wrote: > I'll probably need a v2 of this patch. #include

Re: [PATCH 2/2] drm/amdkfd: Use order_base_2 to get log2 of buffes sizes

Thanks, Applied to -next On Tue, Nov 7, 2017 at 12:04 AM, Deucher, Alexander wrote: >> -Original Message- >> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf >> Of Felix Kuehling >> Sent: Monday, November 06, 2017 2:52 PM >> To:

Re: [PATCH] drm/amdgpu: Remove check which is not valid for certain VBIOS

It fixes some headless board VBIOS which will be failed to load. From: Alex Deucher Sent: Wednesday, November 8, 2017 10:38:37 PM To: Wang, Ken Cc: amd-gfx list Subject: Re: [PATCH] drm/amdgpu: Remove check which is not valid for certain

Re: [RFC 0/7] UVD support for SI in amdgpu

On Wed, Nov 8, 2017 at 5:38 PM, Piotr Redlewski wrote: > Hi, > > Following series implements UVD support for SI in amdgpu driver. Code is based > on CIK's UVD support in amdgpu and SI's UVD support in radeon drivers. To > work, > it requires tahiti uvd firmware with added

[pull] amdgpu drm-next-4.15

Hi Dave, A few more fixes for 4.15. The following changes since commit d65d31388a23b14df9494135ad6c6549a59a3caa: Merge tag 'drm-misc-next-fixes-2017-11-07' of git://anongit.freedesktop.org/drm/drm-misc into drm-next (2017-11-08 05:22:49 +1000) are available in the git repository at:

[RFC 1/7] drm/amdgpu: remove duplicated definitions of some of the SI registers

Some of the registers are defined both in the sid.h and in the new style headers. Removed definitions from the sid.h Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/amdgpu/si.c| 1 + drivers/gpu/drm/amd/amdgpu/si_ih.c | 3 +++ drivers/gpu/drm/amd/amdgpu/sid.h

[RFC 0/7] UVD support for SI in amdgpu

Hi, Following series implements UVD support for SI in amdgpu driver. Code is based on CIK's UVD support in amdgpu and SI's UVD support in radeon drivers. To work, it requires tahiti uvd firmware with added header - I've created simple script to produce exactly this, so if anyone is interested it

[RFC 2/7] drm/amdgpu/uvd4: fix some register's mask and shift definitions

UVD_LMI_CTRL__RFU_MASK and UVD_LMI_CTRL__RFU__SHIFT are defined twice, each time with different values Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/include/asic_reg/uvd/uvd_4_0_sh_mask.h | 2 -- 1 file changed, 2 deletions(-) diff --git

[RFC 3/7] drm/amdgpu/gmc6: don't use vram location programmed by the vbios

On SI chips, fb location programmed by the vbios (40-bit address) is out of reach of the UVD memory controller, which supports only 32-bits. This partially reverts following commits: commit e4f6b39e8bcd ("drm/amdgpu: remove *_mc_access from display funcs") commit 71086a3e8470 ("drm/amdgpu/gmc6:

[RFC 7/7] drm/amdgpu: enable UVD for SI

Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/amdgpu/Makefile | 3 ++- drivers/gpu/drm/amd/amdgpu/si.c | 5 +++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index

[RFC 6/7] drm/amdgpu/uvd4: add hardware specific functions for uvd 4.0

Add logic for starting, stopping, suspending and resuming uvd block Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 7 - drivers/gpu/drm/amd/amdgpu/si.c | 250 - drivers/gpu/drm/amd/amdgpu/sid.h | 22 +-

[RFC 5/7] drm/amdgpu/uvd4: add sw init and fini stages' functions for uvd 4.0

Load firmware and initialize uvd ring Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 14 drivers/gpu/drm/amd/amdgpu/uvd_v4_0.c | 40 + 2 files changed, 54 insertions(+) diff --git

[RFC 4/7] drm/amdgpu/uvd4: add early init stage functions for uvd 4.0

Add uvd ring and interrupt functions Signed-off-by: Piotr Redlewski --- drivers/gpu/drm/amd/amdgpu/sid.h | 14 +- drivers/gpu/drm/amd/amdgpu/uvd_v4_0.c | 282 ++ drivers/gpu/drm/amd/amdgpu/uvd_v4_0.h | 29 3 files changed, 319

Re: [PATCH v2] amd/display: Fix potential null dereference in dce_calcs.c

On Wed, Nov 8, 2017 at 4:47 PM, Ernst Sjöstrand wrote: > Reported by smatch: > bw_calcs() error: potential null dereference 'data' > > Signed-off-by: Ernst Sjöstrand Reviewed and pushed. thanks! Alex > --- > I couldn't use DC_ERROR because the dc_context

Applied "ASoC: amd: use do_div rather than 64 bit division to fix 32 bit builds" to the asoc tree

The patch ASoC: amd: use do_div rather than 64 bit division to fix 32 bit builds has been applied to the asoc tree at https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git All being well this means that it will be integrated into the linux-next tree (usually sometime in the

Re: [PATCH] amd/display: Fix potential null dereference in dce_calcs.c

On 2017-11-08 03:54 PM, Deucher, Alexander wrote: >> -Original Message- >> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf >> Of Ernst Sjöstrand >> Sent: Wednesday, November 08, 2017 3:17 PM >> To: amd-gfx@lists.freedesktop.org >> Subject: [PATCH] amd/display: Fix

[PATCH] amdgpu/dm: Default PRE_VEGA ASIC support to 'y'

Even though we default PRE_VEGA support to 'n' upstream in amd-staging we want to keep it enabled by default. Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/display/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 08, 2017 at 02:35:05PM -0500, Alex Deucher wrote: > On Wed, Nov 8, 2017 at 2:30 PM, Guenter Roeck wrote: > > On Wed, Nov 8, 2017 at 11:18 AM, Deucher, Alexander > >> > Is this different to "ASoC: amd: Report accurate hw_ptr during dma" > >> > which was applied at

Re: [PATCH 1/2] drm/amdgpu: potential uninitialized variable in amdgpu_vce_ring_parse_cs()

On Wed, Nov 8, 2017 at 2:53 PM, Ernst Sjöstrand wrote: > Can't find these anywhere yet, errors still there. > > https://patchwork.freedesktop.org/series/31220/ Applied. thanks for the reminder. Alex > > Regards > //Ernst > > 2017-09-30 10:25 GMT+02:00 Christian König

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 11:35 AM, Alex Deucher wrote: > On Wed, Nov 8, 2017 at 2:30 PM, Guenter Roeck wrote: > > On Wed, Nov 8, 2017 at 11:18 AM, Deucher, Alexander > > wrote: > >> > >> > -Original Message- > >> >

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 2:49 PM, Mark Brown wrote: > On Wed, Nov 08, 2017 at 11:13:50AM -0800, Guenter Roeck wrote: >> On Wed, Nov 8, 2017 at 10:47 AM, Mark Brown wrote: > >> > Is this different to "ASoC: amd: Report accurate hw_ptr during dma" >> > which

Re: [PATCH 1/2] drm/amdgpu: potential uninitialized variable in amdgpu_vce_ring_parse_cs()

Can't find these anywhere yet, errors still there. https://patchwork.freedesktop.org/series/31220/ Regards //Ernst 2017-09-30 10:25 GMT+02:00 Christian König : > Am 30.09.2017 um 10:13 schrieb Dan Carpenter: >> >> We shifted some code around in commit 9cca0b8e5df0

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 08, 2017 at 11:13:50AM -0800, Guenter Roeck wrote: > On Wed, Nov 8, 2017 at 10:47 AM, Mark Brown wrote: > > Is this different to "ASoC: amd: Report accurate hw_ptr during dma" > > which was applied at 16:07? > I suspect we may be getting hit by chromeos-isms

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 2:30 PM, Guenter Roeck wrote: > On Wed, Nov 8, 2017 at 11:18 AM, Deucher, Alexander > wrote: >> >> > -Original Message- >> > From: Mark Brown [mailto:broo...@kernel.org] >> > Sent: Wednesday, November 08, 2017 1:48 PM

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 08, 2017 at 01:40:32PM -0500, Alex Deucher wrote: > On Wed, Nov 8, 2017 at 1:22 PM, Mark Brown wrote: > > Like I said in reply to your other mail please don't resubmit already > > applied patches. The current tip of my topic/amd branch appears to be > > this very

RE: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

> -Original Message- > From: Mark Brown [mailto:broo...@kernel.org] > Sent: Wednesday, November 08, 2017 1:48 PM > To: Alex Deucher > Cc: amd-gfx list; alsa-de...@alsa-project.org; Maling list - DRI developers; > Mukunda, Vijendar; Liam Girdwood; Takashi Iwai; Guenter Roeck; Deucher, >

Re: [PATCH 1/3] ASoC: AMD: Make the driver name consistent across files

On Wed, Nov 08, 2017 at 06:12:34PM +, Deucher, Alexander wrote: > This didn't not appear to be in your tree yet and I never got any > confirmation it being > applied. Apologies. Ah, sorry - my bad. This was one that I tried to apply but which didn't apply. signature.asc Description: PGP

Applied "ASoC: amd: Make the driver name consistent across files" to the asoc tree

The patch ASoC: amd: Make the driver name consistent across files has been applied to the asoc tree at https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours)

Re: [PATCH] ASoC: amd: fix report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 1:22 PM, Mark Brown wrote: > On Wed, Nov 08, 2017 at 01:18:41PM -0500, Alex Deucher wrote: >> From: Guenter Roeck >> >> ERROR: "__aeabi_uldivmod" [sound/soc/amd/snd-soc-acp-pcm.ko] undefined! >> >> 64-bit divides require special

[PATCH umr] Switch GRBM index when reading wave data directly.

Signed-off-by: Tom St Denis --- src/lib/wave_status.c | 4 1 file changed, 4 insertions(+) diff --git a/src/lib/wave_status.c b/src/lib/wave_status.c index fe2add779fdd..7f0130bb9347 100644 --- a/src/lib/wave_status.c +++ b/src/lib/wave_status.c @@ -116,7 +116,9 @@

[PATCH] ASoC: amd: fix report accurate hw_ptr during dma

From: Guenter Roeck ERROR: "__aeabi_uldivmod" [sound/soc/amd/snd-soc-acp-pcm.ko] undefined! 64-bit divides require special operations to avoid build errors on 32-bit systems. fixes: 61add8147942 (ASoC: amd: Report accurate hw_ptr during dma) Signed-off-by: Guenter Roeck

Re: [PATCH 2/4] drm/ttm: consistently use reservation_object_unlock

Am 08.11.2017 um 18:37 schrieb Michel Dänzer: On 08/11/17 05:41 PM, Christian König wrote: Am 08.11.2017 um 17:36 schrieb Michel Dänzer: On 08/11/17 03:59 PM, Christian König wrote: Instead of having a pointless wrapper or call the underlying ww_mutex function directly. Not sure I can agree

Re: [PATCH v2 2/3] ASoC: rt5645: Wait for 400msec before concluding on value of RT5645_VENDOR_ID2

On Wed, Nov 08, 2017 at 12:24:03PM -0500, Alex Deucher wrote: > regmap_read(regmap, RT5645_VENDOR_ID2, ); > > + /* > + * Read after 400msec, as it is the interval required between > + * read and power On. > + */ > + msleep(TIME_TO_POWER_MS); > +

RE: [PATCH 1/3] ASoC: AMD: Make the driver name consistent across files

> -Original Message- > From: Mark Brown [mailto:broo...@kernel.org] > Sent: Wednesday, November 08, 2017 1:09 PM > To: Alex Deucher > Cc: amd-gfx@lists.freedesktop.org; alsa-de...@alsa-project.org; dri- > de...@lists.freedesktop.org; Mukunda, Vijendar; lgirdw...@gmail.com; > ti...@suse.de;

Re: [PATCH 3/3] FIXUP: FROMLIST: ASoC: amd: Report accurate hw_ptr during dma

On Wed, Nov 08, 2017 at 12:45:16PM -0500, Alex Deucher wrote: > On Wed, Nov 8, 2017 at 12:42 PM, Alex Deucher wrote: > > I'm not familiar with which are chromium specific (TEST, BUG, FIXUP, > > FROMLIST I guess?). The info seems useful to have in the bug, but I > > can

[PATCH xf86-video-amdgpu] Use correct ScrnInfoPtr in redisplay_dirty

From: Michel Dänzer We used the destination pixmap's screen for flushing glamor. But when we are the master screen, the destination pixmap is from the slave screen. Fixes crash when the slave screen isn't using glamor as well. Bugzilla:

Re: [PATCH 3/3] FIXUP: FROMLIST: ASoC: amd: Report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 12:42 PM, Alex Deucher wrote: > On Wed, Nov 8, 2017 at 12:29 PM, Guenter Roeck wrote: >> On Wed, Nov 8, 2017 at 9:24 AM, Alex Deucher wrote: >>> >>> From: Guenter Roeck >>> >>> ERROR:

Re: [PATCH 3/3] FIXUP: FROMLIST: ASoC: amd: Report accurate hw_ptr during dma

On Wed, Nov 8, 2017 at 12:29 PM, Guenter Roeck wrote: > On Wed, Nov 8, 2017 at 9:24 AM, Alex Deucher wrote: >> >> From: Guenter Roeck >> >> ERROR: "__aeabi_uldivmod" [sound/soc/amd/snd-soc-acp-pcm.ko] undefined! >> >> 64-bit divides

Re: [PATCH 2/4] drm/ttm: consistently use reservation_object_unlock

On 08/11/17 05:41 PM, Christian König wrote: > Am 08.11.2017 um 17:36 schrieb Michel Dänzer: >> On 08/11/17 03:59 PM, Christian König wrote: >>> Instead of having a pointless wrapper or call the underlying ww_mutex >>> function directly. >> Not sure I can agree it's pointless, since it recently

[PATCH v2 2/3] ASoC: rt5645: Wait for 400msec before concluding on value of RT5645_VENDOR_ID2

From: Akshu Agrawal Minimum time required between power On of codec and read of RT5645_VENDOR_ID2 is 400msec. We should wait that long before reading the value. TEST=Cold boot the device and check for sound device. Signed-off-by: Akshu Agrawal

[PATCH 3/3] FIXUP: FROMLIST: ASoC: amd: Report accurate hw_ptr during dma

From: Guenter Roeck ERROR: "__aeabi_uldivmod" [sound/soc/amd/snd-soc-acp-pcm.ko] undefined! 64-bit divides require special operations to avoid build errors on 32-bit systems. BUG=b:63121716 TEST="Build i386:allmodconfig" Signed-off-by: Guenter Roeck

[PATCH v2 0/3] Fixes for AMD Stoney ACP audio

This patch set is just a couple fixes for the Audio CoProcessor (ACP) on AMD Stoney platforms and a small general fix for the rt5645 codec driver. The entire patch set can also be viewed here: https://cgit.freedesktop.org/~agd5f/linux/log/?h=sound-for-next-stoney2 Thanks! Alex v2: - Add a fix

[PATCH 1/3] ASoC: AMD: Make the driver name consistent across files

From: Akshu Agrawal This fixes the issue of driver not getting auto loaded with MODULE_ALIAS. find /sys/devices -name modalias -print0 | xargs -0 grep 'audio' /sys/devices/pci:00/:00:01.0/acp_audio_dma.0.auto/modalias:platform:acp_audio_dma BUG=b:62103837

Re: [PATCH 2/4] drm/ttm: consistently use reservation_object_unlock

Am 08.11.2017 um 17:36 schrieb Michel Dänzer: On 08/11/17 03:59 PM, Christian König wrote: Instead of having a pointless wrapper or call the underlying ww_mutex function directly. Not sure I can agree it's pointless, since it recently took me a while to realize that unlocking bo->resv is

Re: [PATCH 2/4] drm/ttm: consistently use reservation_object_unlock

On 08/11/17 03:59 PM, Christian König wrote: > Instead of having a pointless wrapper or call the underlying ww_mutex > function directly. Not sure I can agree it's pointless, since it recently took me a while to realize that unlocking bo->resv is essentially the same as unreserving the BO.

Applied "ASoC: amd: Report accurate hw_ptr during dma" to the asoc tree

The patch ASoC: amd: Report accurate hw_ptr during dma has been applied to the asoc tree at https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to

Re: [PATCH 1/3] ASoC: amd: Report accurate hw_ptr during dma

On Tue, Nov 07, 2017 at 09:34:35PM +0530, Agrawal, Akshu wrote: > On 11/7/2017 5:07 PM, Mark Brown wrote: > > > > These two URLs are different, what was being reviewed here? What is > > > > Commit-Ready supposed to mean? > Same patch is reviewed, once on 4.4 kernel (659699) and then on 4.12

Re: [PATCH 0/3] Bunch of cleanups in DM and amdgpu_crtc

Reviewed-by: Andrey Grodzovsky On 11/08/2017 10:33 AM, Harry Wentland wrote: Just removing a couple unused things I noticed while going through the code. Harry Wentland (3): amdgpu/dm: Remove fb_location form fill_plane_attributes drm/amdgpu: Remove unused

[PATCH 1/3] amdgpu/dm: Remove fb_location form fill_plane_attributes

We no longer set the framebuffer address here so this is now dead code. Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 13 - 1 file changed, 13 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

[PATCH 2/3] drm/amdgpu: Remove unused dc_stream from amdgpu_crtc

It's no longer used. In fact, there is no more dc_stream object. Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h

[PATCH 0/3] Bunch of cleanups in DM and amdgpu_crtc

Just removing a couple unused things I noticed while going through the code. Harry Wentland (3): amdgpu/dm: Remove fb_location form fill_plane_attributes drm/amdgpu: Remove unused dc_stream from amdgpu_crtc amdgpu/dm: Remove unused forward declaration

[PATCH 3/3] amdgpu/dm: Remove unused forward declaration

dc_stream has long been renamed to dc_stream_state, so this forward declaration hasn't been used at all. Change-Id: I0c03cfe600f39eb1ec4f8015be340c62a7d22eb3 Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 2 -- 1 file changed, 2

[PATCH 3/4] drm/ttm: make unlocking in ttm_bo_cleanup_refs optional

Needed for the next patch. Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_bo.c | 52 1 file changed, 28 insertions(+), 24 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c

[PATCH 4/4] drm/ttm: optimize ttm_mem_evict_first

Deleted BOs with the same reservation object can be reaped even if they can't be reserved. Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_bo.c | 35 +++ 1 file changed, 27 insertions(+), 8 deletions(-) diff --git

[PATCH 2/4] drm/ttm: consistently use reservation_object_unlock

Instead of having a pointless wrapper or call the underlying ww_mutex function directly. Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_bo.c | 13 +++-- drivers/gpu/drm/ttm/ttm_execbuf_util.c | 8 include/drm/ttm/ttm_bo_driver.h

[PATCH 1/4] drm/ttm: move unlocking out of ttm_bo_cleanup_memtype_use

Needed for the next patch and makes the code quite a bit easier to understand. Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_bo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c

Re: [PATCH] drm/amdgpu: Remove check which is not valid for certain VBIOS

On Wed, Nov 8, 2017 at 1:49 AM, wrote: > From: Ken Wang What cases does this fix? I'm guessing VFCT or some other platform method for getting the vbios? Acked-by: Alex Deucher > > Signed-off-by: Ken Wang >

Re: [PATCH] drm/amd/pp: fix dpm randomly failed on Vega10

On Wed, Nov 8, 2017 at 3:41 AM, Rex Zhu wrote: > Change-Id: I15e562ddff64cf73262ece75f08df60fedc1de1a > Signed-off-by: Rex Zhu Acked-by: Alex Deucher > --- > drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c | 29 >

Re: [PATCH] drm/amdgpu: revise retry init to fully cleanup driver

Please put that as a comment above those two lines of code. Apart from that the patch looks good to me. Regards, Christian. Am 08.11.2017 um 10:46 schrieb Ding, Pixel: When exclusive mode timeout happens, the VF is not active anymore. Exclusive requests will be ignored by host. Unload kms or

Re: [PATCH] drm/amdgpu: revise retry init to fully cleanup driver

When exclusive mode timeout happens, the VF is not active anymore. Exclusive requests will be ignored by host. Unload kms or device fini also request exclusive mode and it will get timeout again since no response received. This only happens for exclusive mode timeout, so I didn’t put them in

Re: [PATCH] drm/amdgpu:fix gpu recover missing skipping(v2)

Am 08.11.2017 um 08:08 schrieb Monk Liu: if app close CTX right after IB submit, gpu recover will fail to find out the entity behind this guilty job thus lead to no job skipping for this guilty job. to fix this corner case just move the increasement of job->karma out of the entity iteration.

Re: [PATCH 2/2] drm/amd/scheduler: add WARN_ON for s_fence->parent

Am 08.11.2017 um 07:46 schrieb Chunming Zhou: Change-Id: I1f6c81002fb2ba21c17cdc14fdde86579b28374e Signed-off-by: Chunming Zhou Reviewed-by: Christian König --- drivers/gpu/drm/amd/scheduler/gpu_scheduler.c | 1 + 1 file changed, 1

Re: [PATCH] drm/amdgpu:fix gpu recover missing skipping

Am 08.11.2017 um 07:39 schrieb Monk Liu: if app close CTX right after IB submit, gpu recover will failed to find out the entity/ctx behind the guilty job thus lead to bad job skipping in scheduler failed to fix this corner case just move the job->karma increasing out of the condition that the

Re: [PATCH] drm/amdgpu: revise retry init to fully cleanup driver

Am 08.11.2017 um 04:29 schrieb Pixel Ding: Retry at drm_dev_register instead of amdgpu_device_init. Signed-off-by: Pixel Ding --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c| 11 +--

Re: [PATCH] drm/amdgpu: bypass lru touch for KIQ ring submission

Am 08.11.2017 um 03:30 schrieb Pixel Ding: KIQ ring submission is used for register accessing on SRIOV VF that could happen both in irq enabled and irq disabled cases. Inversion lock could happen on adev->ring_lru_list_lock, while this operation is useless and just adds overhead in this use

[PATCH] drm/amd/pp: fix dpm randomly failed on Vega10

Change-Id: I15e562ddff64cf73262ece75f08df60fedc1de1a Signed-off-by: Rex Zhu --- drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c | 29 +++--- drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.h | 1 + 2 files changed, 16 insertions(+), 14 deletions(-) diff

Re: [PATCH] drm/amdgpu:remove debugfs file in amdgpu_device_finish

It’s tested and proved working correctly. Well testing doesn't prove anything. Please enable at least KASAN, lockdep and kmemleak and then try again. I'm pretty sure that at least kmemlead and lockdep will bail out quite a bunch of warnings. Regards, Christian. Am 08.11.2017 um 03:40