Re: GTT memory usage overflow

Hi, no that is certainly not normal. We temporary allow the kernel to allocate more pages than available when we bind stuff into the GART, but that shouldn't have this effect. How do you allocate the GTT space? Regards, Christian. Am 12.07.2018 um 17:50 schrieb sunnanyong: Hi All, When t

Re: [PATCH v2] drm/amdgpu: Allow to create BO lists in CS ioctl v3

Am 12.07.2018 um 18:43 schrieb Andrey Grodzovsky: This change is to support MESA performace optimization. Modify CS IOCTL to allow its input as command buffer and an array of buffer handles to create a temporay bo list and then destroy it when IOCTL completes. This saves on calling for BO_LIST cr

Re: [PATCH 3/3] drm/scheduler: modify args of drm_sched_entity_init

Am 12.07.2018 um 08:36 schrieb Nayan Deshmukh: replace run queue by a list of run queues and remove the sched arg as that is part of run queue itself Signed-off-by: Nayan Deshmukh Going over this I've found one more thing which needs to be fixed: [SNIP] diff --git a/drivers/gpu/drm/etnaviv/

Re: [PATCH 3/3] drm/scheduler: modify args of drm_sched_entity_init

On Thu, Jul 12, 2018 at 11:21 PM Eric Anholt wrote: > > Nayan Deshmukh writes: > > diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler.c > > b/drivers/gpu/drm/scheduler/gpu_scheduler.c > > index 3dc1a4f07e3f..b2dbd1c1ba69 100644 > > --- a/drivers/gpu/drm/scheduler/gpu_scheduler.c > > +++ b/driv

Re: [PATCH 3/3] drm/scheduler: modify args of drm_sched_entity_init

Am 13.07.2018 um 09:20 schrieb Nayan Deshmukh: On Thu, Jul 12, 2018 at 11:21 PM Eric Anholt wrote: Nayan Deshmukh writes: diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler.c b/drivers/gpu/drm/scheduler/gpu_scheduler.c index 3dc1a4f07e3f..b2dbd1c1ba69 100644 --- a/drivers/gpu/drm/scheduler

Re: [PATCH 1/3] drm/amdgpu/pp: remove dead vega12 code

Am 12.07.2018 um 22:23 schrieb Alex Deucher: Commented out. Signed-off-by: Alex Deucher Acked-by: Christian König --- .../gpu/drm/amd/powerplay/smumgr/vega12_smumgr.c | 29 -- 1 file changed, 29 deletions(-) diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vega12

Re: GTT memory usage overflow

On 2018-07-13 09:00 AM, Christian König wrote: > Hi, > > no that is certainly not normal. > > We temporary allow the kernel to allocate more pages than available when > we bind stuff into the GART, but that shouldn't have this effect. mgr->available can happen to be negative when amdgpu_gtt_mgr_

Re: GTT memory usage overflow

Am 13.07.2018 um 09:40 schrieb Michel Dänzer: On 2018-07-13 09:00 AM, Christian König wrote: Hi, no that is certainly not normal. We temporary allow the kernel to allocate more pages than available when we bind stuff into the GART, but that shouldn't have this effect. mgr->available can happe

[PATCH 1/2] ALSA: HDA: use PCI_BASE_CLASS_DISPLAY to cover more display adapters

Signed-off-by: Jim Qu --- sound/pci/hda/hda_intel.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 1ae1850..e0064567 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -31,7 +31,7 @@

[PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP output. For some dGPU which is no output, audio codec usually is disabled. In currect HDA audio driver, it will set all codec as VGA_SWITCHEROO_DIS, the audio which is binded to U

Re: [PATCH libdrm] amdgpu: add amdgpu_bo_handle_type_kms_noimport

On 2018-07-12 07:03 PM, Marek Olšák wrote: > On Thu, Jul 12, 2018, 3:31 AM Michel Dänzer wrote: >> On 2018-07-12 02:47 AM, Marek Olšák wrote: >>> From: Marek Olšák >>> >>> diff --git a/amdgpu/amdgpu_bo.c b/amdgpu/amdgpu_bo.c >>> index 9e37b149..d29be244 100644 >>> --- a/amdgpu/amdgpu_bo.c >>> +++

Re: [PATCH 1/2] ALSA: HDA: use PCI_BASE_CLASS_DISPLAY to cover more display adapters

On Fri, Jul 13, 2018 at 04:06:01PM +0800, Jim Qu wrote: > Signed-off-by: Jim Qu Empty commit message. Please add an explanation why the change is necessary so folks browsing the git history understand it's motivation. I think here the motivation is that the PCI class is sometimes PCI_CLASS_DISPL

Re: [alsa-devel] [PATCH 1/2] ALSA: HDA: use PCI_BASE_CLASS_DISPLAY to cover more display adapters

On Fri, 13 Jul 2018 10:06:01 +0200, Jim Qu wrote: > > Signed-off-by: Jim Qu More explanations please. e.g. why is this needed? The code change itself looks good, but ... > @@ -31,7 +31,7 @@ > * CHANGES: > * > * 2004.12.01 Major rewrite by tiwai, merged the work of pshou > - * >

[PATCH xf86-video-amdgpu] Hardcode "non-desktop" RandR property name

From: Michel Dänzer It's a bit silly to require current randrproto just for this definition, which can't really change anyway. Suggested-by: Qiang Yu Signed-off-by: Michel Dänzer --- configure.ac | 2 +- src/drmmode_display.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)

Re: [PATCH xf86-video-amdgpu 1/3] modesetting: Record non-desktop kernel property at PreInit time

On 2018-07-13 04:26 AM, Yu, Qiang wrote: > Hi Michel, > > Could randrproto >= 1.6.0 also be optional? randrproto 1.6 is pretty new so > most old xserver doesn't compile with it. When compile latest ddx with > various OS we support, it will fail. > > Maybe we can define RR_PROPERTY_NON_DESKTOP in

Re: [alsa-devel] [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

On Fri, 13 Jul 2018 10:06:02 +0200, Jim Qu wrote: > > On modern laptop, there are more and more platforms > have two GPUs, and each of them maybe have audio codec > for HDMP/DP output. For some dGPU which is no output, > audio codec usually is disabled. > > In currect HDA audio driver, it will se

Re: [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

On Fri, Jul 13, 2018 at 04:06:02PM +0800, Jim Qu wrote: > On modern laptop, there are more and more platforms > have two GPUs, and each of them maybe have audio codec > for HDMP/DP output. For some dGPU which is no output, > audio codec usually is disabled. > > In currect HDA audio driver, it will

Re: [alsa-devel] [PATCH 1/2] ALSA: HDA: use PCI_BASE_CLASS_DISPLAY to cover more display adapters

See comments in line thanks JimQu 获取 Outlook for Android 发件人: Takashi Iwai 发送时间: 7月13日星期五 16:33 主题: Re: [alsa-devel] [PATCH 1/2] ALSA: HDA: use PCI_BASE_CLASS_DISPLAY tocover more display adapters 收件人: Qu, Jim 抄送: alsa-de...@alsa-project.org, dri-de...@lists.freedesktop

[PATCH v2 1/2] drm/scheduler: add a pointer to scheduler in the rq

This patch is in preparation for a better load balancing in scheduler. It allows us to associate entities with the run queues instead of binding them to a scheduler. Signed-off-by: Nayan Deshmukh --- drivers/gpu/drm/scheduler/gpu_scheduler.c | 6 -- include/drm/gpu_scheduler.h

[PATCH v2 2/2] drm/scheduler: modify args of drm_sched_entity_init

replace run queue by a list of run queues and remove the sched arg as that is part of run queue itself Signed-off-by: Nayan Deshmukh Reviewed-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 +-- drivers/gpu/drm/amd/amdg

RE: DEBUG_LOCKS_WARN_ON(sem->owner != ((struct task_struct *)(1UL << 0)))

Reproduced on my side. This backtrace only shows one time after boot up unless reboot again. This backtrace is introduced by KFDIPCTest.BasicTest(other tests also can introduce it). This test defines some buffers which are mapped to gpu, and the fork later causes this warning. The code commen

RE: DEBUG_LOCKS_WARN_ON(sem->owner != ((struct task_struct *)(1UL << 0)))

I think this is a kernel bug. See patch https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/kernel/locking?h=v4.18-rc4&id=03eeafdd9ab06a770d42c2b264d50dff7e2f4eee -Original Message- From: Pan, Xinhui Sent: 2018年7月13日 18:17 To: Kuehling, Felix ; Michel Dänzer Cc: am

Re: [PATCH v2 1/2] drm/scheduler: add a pointer to scheduler in the rq

I forgot to add these: On Fri, Jul 13, 2018 at 3:21 PM Nayan Deshmukh wrote: > > This patch is in preparation for a better load balancing in > scheduler. It allows us to associate entities with the > run queues instead of binding them to a scheduler. > > Signed-off-by: Nayan Deshmukh Reviewed-by:

Re: [PATCH v2 1/2] drm/scheduler: add a pointer to scheduler in the rq

Am 13.07.2018 um 15:40 schrieb Nayan Deshmukh: I forgot to add these: Not a problem at all. It is actually my job to do this while committing the patches. BTW: I've pushed you patches to amd-staging-drm-next just a few minutes ago. Thanks for the help. I'm preparing quite a cleanup for am

RE: DEBUG_LOCKS_WARN_ON(sem->owner != ((struct task_struct *)(1UL << 0)))

Already fixed on upstream. https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/kernel/locking?h=v4.18-rc4&id=03eeafdd9ab06a770d42c2b264d50dff7e2f4eee From: Pan, Xinhui Sent: Friday, July 13, 2018 6:17:06 PM To: Kuehling, Felix; Michel Dänzer Cc

Re: DEBUG_LOCKS_WARN_ON(sem->owner != ((struct task_struct *)(1UL << 0)))

On 2018-07-13 01:13 PM, Pan, Xinhui wrote: > I think this is a kernel bug. See patch > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/kernel/locking?h=v4.18-rc4&id=03eeafdd9ab06a770d42c2b264d50dff7e2f4eee Nice find, thanks! -- Earthling Michel Dänzer |

Re: 答复: 答复: [alsa-devel] 答复: [PATCH] vgaswitchroo: set audio client id according to bound gpu client id

On Wed, 11 Jul 2018 13:12:01 +0200, Takashi Iwai wrote: > > And the forced runtime PM is still an issue, and this would need the > other notification mechanism than the HD-audio unsolicited event as > AMD HDMI controller doesn't honor the HD-audio WAKEEN bit. Since we had a nice "hack week" in th

[PATCH 2/7] drm/amdgpu: remove fence context from the job

Can be obtained directly from the fence as well. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 1 - 4 files changed, 1 inse

[PATCH 4/7] drm/amdgpu: remove job->ring

We can easily get that from the scheduler. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu.h| 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 18 +- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c|

[PATCH 7/7] drm/amdgpu: minor cleanup in amdgpu_job.c

Remove superflous NULL check, fix coding style a bit, shorten error messages. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 11 --- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/

[PATCH 1/7] drm/amdgpu: cleanup job header

Move job related defines, structure and function declarations to amdgpu_job.h Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 46 +--- drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 74 + 2 files changed, 75 insertions(+), 4

[PATCH 6/7] drm/amdgpu: remove job->adev

We can get that from the ring. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 13 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 1 - 2 files changed, 6 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/

[PATCH 3/7] drm/amdgpu: remove ring parameter from amdgpu_job_submit

We know the ring through the entity anyway. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 9 - drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 5 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 6 +++--- dri

[PATCH 5/7] drm/amdgpu: add amdgpu_job_submit_direct helper

Make sure that we properly initialize at least the sched member. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 15 +++ drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 4 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 17 ++--- drivers/gpu/drm/amd

Re: [alsa-devel] [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

在 2018/7/13 16:46, Takashi Iwai 写道: On Fri, 13 Jul 2018 10:06:02 +0200, Jim Qu wrote: On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP output. For some dGPU which is no output, audio codec usually is disabled. In currect HD

Re: [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

在 2018/7/13 17:27, Lukas Wunner 写道: On Fri, Jul 13, 2018 at 04:06:02PM +0800, Jim Qu wrote: On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP output. For some dGPU which is no output, audio codec usually is disabled. In curr

[PATCH] drm/amdgpu: Make sure IB tests flushed after IP resume

Fixes: 2c773de2 (drm/amdgpu: defer test IBs on the rings at boot (V3)) Signed-off-by: Leo Liu --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index aeb1b

Re: [PATCH] drm/amdgpu: Make sure IB tests flushed after IP resume

Am 13.07.2018 um 17:41 schrieb Leo Liu: Fixes: 2c773de2 (drm/amdgpu: defer test IBs on the rings at boot (V3)) Signed-off-by: Leo Liu Reviewed-by: Christian König for now, but we might want to move that even further down. Christian. --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 +

Re: [alsa-devel] [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

On Fri, 13 Jul 2018 17:25:39 +0200, jimqu wrote: > > > > 在 2018/7/13 16:46, Takashi Iwai 写道: > > On Fri, 13 Jul 2018 10:06:02 +0200, > > Jim Qu wrote: > >> On modern laptop, there are more and more platforms > >> have two GPUs, and each of them maybe have audio codec > >> for HDMP/DP output. For

Re: [alsa-devel] [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

在 2018/7/14 0:38, Takashi Iwai 写道: On Fri, 13 Jul 2018 17:25:39 +0200, jimqu wrote: 在 2018/7/13 16:46, Takashi Iwai 写道: On Fri, 13 Jul 2018 10:06:02 +0200, Jim Qu wrote: On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP o

[pull] amdgpu drm-fixes-4.18

Hi Dave, A few more fixes for 4.18. Two display fixes and a fix to avoid a segfault if the GPU does not power up properly on resume. These are on top of my pull from earlier this week. The following changes since commit 9d4a0d4cdc8b5904ec7c9b9e04bab3e9e60d7a74: drm/amdgpu: Verify root PD is m

[PATCH] drm/amd/display: Retry link training if preferred settings not used

Some receivers seem to fail the first link training but are good on subsequent tries. Check if the max setting was good otherwise retry link training. This fixes HTC vive pro issue where it would fail to light up after being disabled. Signed-off-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/dis

Re: [PATCH] drm/amd/display: Retry link training if preferred settings not used

On 2018-07-13 01:40 PM, Bhawanpreet Lakha wrote: > Some receivers seem to fail the first link training but are good on > subsequent tries. Check if the max setting was good otherwise retry > link training. > > This fixes HTC vive pro issue where it would fail to light up after > being disabled. >

Re: [PATCH libdrm] amdgpu: add amdgpu_bo_handle_type_kms_noimport

On Fri, Jul 13, 2018 at 4:28 AM, Michel Dänzer wrote: > On 2018-07-12 07:03 PM, Marek Olšák wrote: >> On Thu, Jul 12, 2018, 3:31 AM Michel Dänzer wrote: >>> On 2018-07-12 02:47 AM, Marek Olšák wrote: From: Marek Olšák diff --git a/amdgpu/amdgpu_bo.c b/amdgpu/amdgpu_bo.c index

Re: [PATCH 0/6] Raven support for KFD

This patch series applies on top of my previous 25-patch series on top of Oded's amdkfd-next branch. On the user mode side, you need my drm-next-wip branch in https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/tree/fxkamd/drm-next-wip. The rest of ROCm user mode should work unmodified. Reg

[PATCH] Revert "drm/amd/powerplay: fix performance drop on Vega10"

This reverts commit b87079ec7b4d38efee015367315958ce5495ba93. SMU FW team ask to remove this version specific code. Signed-off-by: Eric Huang --- drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c | 5 - 1 file changed, 5 deletions(-) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10

Re: [PATCH 1/6] drm/amdkfd: Consolidate duplicate memory banks info in topology

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao > > If there are several memory banks that has the same properties in CRAT, > we aggregate them into one memory bank. This cleans up memory banks on > APUs (e.g. Raven) where the CRAT reports each memory channel as a > separ

Re: [PATCH 2/6] drm/amdkfd: Make SDMA engine number an ASIC-dependent variable

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao > > On Raven there is only one SDMA engine instead of previously assumed two, > so we need to adapt our code to this new scenario. > > Signed-off-by: Yong Zhao > Reviewed-by: Felix Kuehling > Signed-off-by: Felix Kuehling

Re: [PATCH 3/6] drm/amdkfd: Avoid flooding dmesg on Raven due to IOMMU issues

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao > > On Raven Invalid PPRs can be reported because multiple PPRs can be > still queued when memory is freed. Apply a rate limit to avoid > flooding the log in this case. > > Signed-off-by: Yong Zhao > Reviewed-by: Felix Kueh

Re: [PATCH 4/6] drm/amdkfd: Workaround to accommodate Raven too many PPR issue

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao Please add a patch description explaining why this is needed for Raven. Alex > > Signed-off-by: Yong Zhao > Reviewed-by: Felix Kuehling > Signed-off-by: Felix Kuehling > --- > drivers/gpu/drm/amd/amdkfd/kfd_events.c

Re: [PATCH 5/6] drm/amdkfd: Optimize out some duplicated code in kfd_signal_iommu_event()

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao Please add a patch description. Alex > > Signed-off-by: Yong Zhao > Reviewed-by: Felix Kuehling > Signed-off-by: Felix Kuehling > --- > drivers/gpu/drm/amd/amdkfd/kfd_events.c | 26 +++--- > 1 file

Re: [PATCH 6/6] drm/amdkfd: Enable Raven for KFD

On Thu, Jul 12, 2018 at 5:24 PM, Felix Kuehling wrote: > From: Yong Zhao > > Signed-off-by: Yong Zhao > Reviewed-by: Felix Kuehling > Signed-off-by: Felix Kuehling Please add a patch description. With that fixed: Acked-by: Alex Deucher > --- > drivers/gpu/drm/amd/amdkfd/kfd_device.c | 16

Re: [PATCH] Revert "drm/amd/powerplay: fix performance drop on Vega10"

Assuming there are no ill effects: Acked-by: Alex Deucher From: amd-gfx on behalf of Eric Huang Sent: Friday, July 13, 2018 3:22:30 PM To: amd-gfx@lists.freedesktop.org Cc: Huang, JinHuiEric Subject: [PATCH] Revert "drm/amd/powerplay: fix performance drop on V

[PATCH 1/6] drm/amdkfd: Consolidate duplicate memory banks info in topology

From: Yong Zhao If there are several memory banks that has the same properties in CRAT, we aggregate them into one memory bank. This cleans up memory banks on APUs (e.g. Raven) where the CRAT reports each memory channel as a separate bank. This only confuses user mode, which only deals with virtu

[PATCH 2/6] drm/amdkfd: Make SDMA engine number an ASIC-dependent variable

From: Yong Zhao On Raven there is only one SDMA engine instead of previously assumed two, so we need to adapt our code to this new scenario. Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Acked-by: Alex Deucher --- drivers/gpu/drm/amd/amdkfd/kfd_device.c

[PATCH 6/6] drm/amdkfd: Enable Raven for KFD

From: Yong Zhao Add DID and kfd_device_info for Raven. Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Acked-by: Alex Deucher --- drivers/gpu/drm/amd/amdkfd/kfd_device.c | 16 1 file changed, 16 insertions(+) diff --git a/drivers/gpu/drm/

[PATCH 3/6] drm/amdkfd: Avoid flooding dmesg on Raven due to IOMMU issues

From: Yong Zhao On Raven Invalid PPRs (peripheral page requests) can be reported because multiple PPRs can be still queued when memory is freed. Apply a rate limit to avoid flooding the log in this case. Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Acked-

[PATCH 4/6] drm/amdkfd: Workaround to accommodate Raven too many PPR issue

From: Yong Zhao On Raven multiple PPRs can be queued up by the hardware. When the first of those requests is handled by the IOMMU driver, the memory access succeeds. After that the application may be done with the memory and unmap it. At that point the page table entries are invalidated, but ther

[PATCH 5/6] drm/amdkfd: Optimize out some duplicated code in kfd_signal_iommu_event()

From: Yong Zhao memory_exception_data is already initialized for not-present faults. It only needs to be overridden for permission faults. Signed-off-by: Yong Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling --- drivers/gpu/drm/amd/amdkfd/kfd_events.c | 26 +++--

[PATCH 0/6] Raven support for KFD v2

Raven refers to Ryzen APUs with integrated GFXv9 GPU. This patch series completes Raven support for KFD: * fix up memory banks info from CRAT * support different number of SDMA engines * workaround IOMMUv2 PPR issues * add device info v2: Updated patch descriptions, no code changes Yong Zhao (6)

Re: [PATCH 0/6] Raven support for KFD v2

On Fri, Jul 13, 2018 at 4:17 PM, Felix Kuehling wrote: > Raven refers to Ryzen APUs with integrated GFXv9 GPU. > This patch series completes Raven support for KFD: > > * fix up memory banks info from CRAT > * support different number of SDMA engines > * workaround IOMMUv2 PPR issues > * add device

Re: [PATCH 2/2] vgaswitchreoo: set audio client id in vgaswitchreoo enable function

在 2018/7/13 17:27, Lukas Wunner 写道: On Fri, Jul 13, 2018 at 04:06:02PM +0800, Jim Qu wrote: On modern laptop, there are more and more platforms have two GPUs, and each of them maybe have audio codec for HDMP/DP output. For some dGPU which is no output, audio codec usually is disabled. In curr