Re: [PATCH 2/6] drm/amdgpu: Enable amdgpu_ttm_tt_get_user_pages in worker threads

Hi Felix, The patch looks fine to me and I can ACK it, but I would prefer that Alex or Christian will review it as well before pushing it. Thanks, Oded On Fri, Mar 23, 2018 at 10:32 PM, Felix Kuehling wrote: > This commit allows amdgpu_ttm_tt_get_user_pages to work in a

Re: [PATCH 2/2] drm/amdgpu: set ttm bo priority before initialization

Looks like a good idea to me as well. Reviewed-by: Christian König for the series. Regards, Christian. Am 11.05.2018 um 07:27 schrieb Zhou, David(ChunMing): The series is OK to me, Reviewed-by: Chunming Zhou It is better to wait Christian to

Re: [PATCH 3/6] drm/amdgpu: Avoid reclaim while holding locks taken in MMU notifier

Hi Felix, Same as patch 2. ACK but I want an extra review from amdgpu people. Thanks, Oded On Fri, Mar 23, 2018 at 10:32 PM, Felix Kuehling wrote: > When an MMU notifier runs in memory reclaim context, it can deadlock > trying to take locks that are already held in the

Re: [PATCH 4/6] drm/amdkfd: GFP_NOIO while holding locks taken in MMU notifier

On Fri, Mar 23, 2018 at 10:32 PM, Felix Kuehling wrote: > When an MMU notifier runs in memory reclaim context, it can deadlock > trying to take locks that are already held in the thread causing the > memory reclaim. The solution is to avoid memory reclaim while holding >

Re: [PATCH 0/6] Userptr memory mapping support for KFD

On Fri, Mar 23, 2018 at 10:32 PM, Felix Kuehling wrote: > Update of remaining patches from the GPUVM patch series. This should apply > on top of the fixes I just sent out. > > Felix Kuehling (6): > drm/amdgpu: Add MMU notifier type for KFD userptr > drm/amdgpu: Enable

Re: [PATCH 2/6] drm/amdgpu: Enable amdgpu_ttm_tt_get_user_pages in worker threads

Adding Alex, Christian On Fri, May 11, 2018 at 10:36 AM, Oded Gabbay wrote: > Hi Felix, > The patch looks fine to me and I can ACK it, but I would prefer that > Alex or Christian will review it as well before pushing it. > Thanks, > Oded > > On Fri, Mar 23, 2018 at 10:32

Re: [PATCH 11/21] drm/amdkfd: Add GFXv9 MQD manager

On Wed, Apr 11, 2018 at 12:33 AM, Felix Kuehling wrote: > Signed-off-by: John Bridgman > Signed-off-by: Jay Cornwall > Signed-off-by: Felix Kuehling > --- > drivers/gpu/drm/amd/amdkfd/Makefile

[PATCH] drm/amdgpu: fix null pointer for bo unmap trace function

Signed-off-by: Junwei Zhang --- drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index 532263a..e96e26d 100644 ---

Re: [PATCH 07/21] drm/amdkfd: Clean up KFD_MMAP_ offset handling

On Wed, Apr 11, 2018 at 12:33 AM, Felix Kuehling wrote: > From: Harish Kasiviswanathan > > Use bit-rotate for better clarity and remove _MASK from the #defines as > these represent mmap types. > > Centralize all the parsing of the mmap

Re: GPU hang trying to run OpenCL kernels on x86_64

Hi Michel, Just made a fresh Ubuntu 18.04 LTS install for armhf with ubuntu default mesa, libdrm and libclc, with the same 4.16.7 kernel and it runs fine with the OpenCL applications that cause the hang on x86_64. So it is surely a kernel or mesa issue exclusive to x86_64, libraries versions are

Re: [PATCH] drm/amdgpu: fix null pointer for bo unmap trace function

Am 11.05.2018 um 11:30 schrieb Junwei Zhang: Signed-off-by: Junwei Zhang Reviewed-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

Re: [PATCH 3/6] drm/amdgpu: Avoid reclaim while holding locks taken in MMU notifier

Hi Oded, it looks sane to me and is Reviewed-by: Christian König . Christian. Am 11.05.2018 um 09:38 schrieb Oded Gabbay: Hi Felix, Same as patch 2. ACK but I want an extra review from amdgpu people. Thanks, Oded On Fri, Mar 23, 2018 at 10:32 PM, Felix Kuehling

[PATCH 1/2] drm/amdgpu: drop printing the BO offset in the gem debugfs

It is meaningless anyway. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 4 1 file changed, 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index

[PATCH 2/2] drm/amdgpu: print the BO flags in the gem debugfs entry

Quite usfull to know. Signed-off-by: Christian König --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 15 +++ 1 file changed, 15 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index

Re: [PATCH 2/2] drm/amdgpu: print the BO flags in the gem debugfs entry

On Fri, May 11, 2018 at 11:40 AM, Christian König wrote: > Quite usfull to know. "useful" With that fixed: Reviewed-by: Alex Deucher > > Signed-off-by: Christian König > --- >

Re: [PATCH 07/21] drm/amdkfd: Clean up KFD_MMAP_ offset handling

On 2018-05-11 04:52 AM, Oded Gabbay wrote: >> --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h >> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h >> @@ -41,9 +41,33 @@ >> >> #define KFD_SYSFS_FILE_MODE 0444 >> >> -#define KFD_MMAP_DOORBELL_MASK 0x8ull >> -#define KFD_MMAP_EVENTS_MASK

Re: [PATCH 1/2] drm/amdgpu: drop printing the BO offset in the gem debugfs

On Fri, May 11, 2018 at 11:40 AM, Christian König wrote: > It is meaningless anyway. > > Signed-off-by: Christian König Reviewed-by: Alex Deucher > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 4 >

[PATCH] drm/amdgpu/vi: don't enable DCE on a specific Polaris sku

It's fused off. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vi.c | 12 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 4ac1288ab7df..4f844beba0db 100644

Fix underflow on Hawaii on 18.20 release branch

Thanks ___ amd-gfx mailing list amd-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/amd-gfx

[PATCH] drm/amdgpu: Fix display corruption on CI with dpm enabled

From: Rex Zhu with dpm enabled, need to get active crtcs in dc/no-dc mode. caused by 'commit ebb649667a31 ("drm/amdgpu: Set pm_display_cfg in non-dc mode")' Reviewed-by: Huang Rui Signed-off-by: Rex Zhu ---

[PATCH v2 4/4] drm/nouveau: Switch to the generic underscan props

Now that underscan props can be parsed by the core and assigned to conn_state->underscan.xxx, we can rely on this implementation and get rid of the nouveau-specific underscan props. Signed-off-by: Boris Brezillon --- drivers/gpu/drm/nouveau/nouveau_connector.c | 39

[PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

Applying an underscan setup is just a matter of scaling all planes appropriately and adjusting the CRTC X/Y offset to account for the horizontal and vertical border. Create an vc4_plane_underscan_adj() function doing that and call it from vc4_plane_setup_clipping_and_scaling() so that we are

[PATCH v2 3/4] drm/vc4: Attach underscan props to the HDMI connector

Now that the plane code takes the underscan setup into account, we can safely attach the underscan props to the HDMI connector. We also take care of filling AVI infoframes correctly to expose the top/botton/left/right bar. Note that these underscan props match pretty well the

[PATCH v2 0/4] drm/connector: Provide generic support for underscan

Hello, This is an attempt at providing generic support for underscan connector props. We already have 3 drivers defining the same underscan, underscan vborder and underscan hborder properties (amd, radeon and nouveau) and I am about to add a new one, hence my proposal to put the prop parsing code

[PATCH v2 1/4] drm/connector: Add generic underscan properties

We have 3 drivers defining the "underscan", "underscan hborder" and "underscan vborder" properties (radeon, amd and nouveau) and we are about to add the same kind of thing in VC4. Define generic underscan props and add new fields to the drm_connector state so that the property parsing logic can

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, May 11, 2018 at 04:59:17PM +0200, Boris Brezillon wrote: > Applying an underscan setup is just a matter of scaling all planes > appropriately and adjusting the CRTC X/Y offset to account for the > horizontal and vertical border. > > Create an vc4_plane_underscan_adj() function doing that

Re: [PATCH] drm/amdgpu: add HDP flush dummy for UVD 6/7

Just did so. Sorry for the delay yesterday was a holiday here and well you know how long it takes to catch up on mails :) Christian. Am 11.05.2018 um 16:09 schrieb Alex Deucher: Are you going to push this or should I? Alex On Tue, May 8, 2018 at 11:46 AM, Deucher, Alexander

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, 11 May 2018 18:34:50 +0300 Ville Syrjälä wrote: > On Fri, May 11, 2018 at 04:59:17PM +0200, Boris Brezillon wrote: > > Applying an underscan setup is just a matter of scaling all planes > > appropriately and adjusting the CRTC X/Y offset to account for the

Fix underflow on Hawaii on 18.20 release branch

Rex Zhu's patch fixes an underflow issue observed on Hawaii. The patch is in drm-next-amd-staging branch, need an approval to port the patch into 18.20 release branch in order to resolve a ticket SWDEV-152336. Thanks ___ amd-gfx mailing list

Re: [PATCH 1/3] drm/connector: Add generic underscan properties

On Mon, 7 May 2018 17:25:30 +0200 Daniel Vetter wrote: > On Mon, May 07, 2018 at 05:15:33PM +0200, Daniel Vetter wrote: > > On Mon, May 07, 2018 at 04:44:32PM +0200, Boris Brezillon wrote: > > > We have 3 drivers defining the "underscan", "underscan hborder" and > > >

Re: [PATCH 1/3] drm/connector: Add generic underscan properties

On Tue, 8 May 2018 10:18:10 +1000 Ben Skeggs wrote: > On 8 May 2018 at 04:26, Harry Wentland wrote: > > > > > > On 2018-05-07 12:19 PM, Boris Brezillon wrote: > >> On Mon, 7 May 2018 18:01:44 +0300 > >> Ville Syrjälä

Re: [PATCH] drm/amdgpu: add HDP flush dummy for UVD 6/7

Are you going to push this or should I? Alex On Tue, May 8, 2018 at 11:46 AM, Deucher, Alexander wrote: > Acked-by: Alex Deucher > > > From: amd-gfx on behalf of

Re: [PATCH] drm/amdgpu/gmc9: remove unused register defs

Am 10.05.2018 um 22:21 schrieb Alex Deucher: These got moved to the new df module so no longer used in this file. Signed-off-by: Alex Deucher Reviewed-by: Christian König --- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 14 --

Re: [PATCH 2/6] drm/amdgpu: Enable amdgpu_ttm_tt_get_user_pages in worker threads

On Fri, May 11, 2018 at 3:36 AM, Oded Gabbay wrote: > Hi Felix, > The patch looks fine to me and I can ACK it, but I would prefer that > Alex or Christian will review it as well before pushing it. Not my area of expertise, but it looks correct to me. Acked-by: Alex Deucher

Re: [PATCH 11/21] drm/amdkfd: Add GFXv9 MQD manager

This patch series was meant to be applied after the userptr changes. I haven't tested this without the userptr changes. I think your main concern about userptr is the use of GFP_NOIO. I remember considering memalloc_noio_save/restore when I worked on this over a year ago. I found an old email

Re: [PATCH 07/21] drm/amdkfd: Clean up KFD_MMAP_ offset handling

On Fri, May 11, 2018 at 6:57 PM, Felix Kuehling wrote: > On 2018-05-11 04:52 AM, Oded Gabbay wrote: >>> --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h >>> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h >>> @@ -41,9 +41,33 @@ >>> >>> #define KFD_SYSFS_FILE_MODE 0444 >>> >>>

Re: [PATCH 11/21] drm/amdkfd: Add GFXv9 MQD manager

On Fri, May 11, 2018 at 9:15 PM, Felix Kuehling wrote: > This patch series was meant to be applied after the userptr changes. I > haven't tested this without the userptr changes. > > I think your main concern about userptr is the use of GFP_NOIO. I > remember considering

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, May 11, 2018 at 05:52:56PM +0200, Boris Brezillon wrote: > On Fri, 11 May 2018 18:34:50 +0300 > Ville Syrjälä wrote: > > > On Fri, May 11, 2018 at 04:59:17PM +0200, Boris Brezillon wrote: > > > Applying an underscan setup is just a matter of scaling all

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, May 11, 2018 at 07:12:21PM +0200, Boris Brezillon wrote: > On Fri, 11 May 2018 19:54:02 +0300 > Ville Syrjälä wrote: > > > On Fri, May 11, 2018 at 05:52:56PM +0200, Boris Brezillon wrote: > > > On Fri, 11 May 2018 18:34:50 +0300 > > > Ville Syrjälä

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, 11 May 2018 19:54:02 +0300 Ville Syrjälä wrote: > On Fri, May 11, 2018 at 05:52:56PM +0200, Boris Brezillon wrote: > > On Fri, 11 May 2018 18:34:50 +0300 > > Ville Syrjälä wrote: > > > > > On Fri, May 11, 2018 at

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, 11 May 2018 20:29:48 +0300 Ville Syrjälä wrote: > On Fri, May 11, 2018 at 07:12:21PM +0200, Boris Brezillon wrote: > > On Fri, 11 May 2018 19:54:02 +0300 > > Ville Syrjälä wrote: > > > > > On Fri, May 11, 2018 at

Re: [PATCH 00/21] GFXv9/Vega10 support for KFD

On Wed, Apr 11, 2018 at 12:32 AM, Felix Kuehling wrote: > This patch series adds support for GFXv9 GPUs to KFD. In this series it > enables support for Vega10. Raven support requires some extra work that > will follow shortly, but Raven support is already included and I

Re: [PATCH 16/21] drm/amdkfd: Add 64-bit doorbell and wptr support to kernel queue

applied this patch instead of original, thanks. On Wed, Apr 25, 2018 at 12:42 AM, Felix Kuehling wrote: > A minor update to this patch is attached. The rest of the series is > unchanged and rebased cleanly on 4.17-rc2 on my system. > > Regards, > Felix > > > On

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

On Fri, May 11, 2018 at 09:47:49PM +0200, Boris Brezillon wrote: > On Fri, 11 May 2018 20:29:48 +0300 > Ville Syrjälä wrote: > > > On Fri, May 11, 2018 at 07:12:21PM +0200, Boris Brezillon wrote: > > > On Fri, 11 May 2018 19:54:02 +0300 > > > Ville Syrjälä

Re: [PATCH v2 2/4] drm/vc4: Take underscan setup into account when updating planes

Ville Syrjälä writes: > On Fri, May 11, 2018 at 07:12:21PM +0200, Boris Brezillon wrote: >> On Fri, 11 May 2018 19:54:02 +0300 >> Ville Syrjälä wrote: >> >> > On Fri, May 11, 2018 at 05:52:56PM +0200, Boris Brezillon wrote: >> > >

Re: [PATCH] drm/amdkfd: Integer overflows in ioctl

On Tue, Apr 24, 2018 at 9:58 PM, Felix Kuehling wrote: > Reviewed-by: Felix Kuehling > > We could probably add a sanity check for n_devices to avoid user mode > causing excessive memory allocations in the kernel. There is no good > reason for this

Re: [PATCH 11/21] drm/amdkfd: Add GFXv9 MQD manager

On 2018-05-11 03:06 PM, Oded Gabbay wrote: > On Fri, May 11, 2018 at 9:15 PM, Felix Kuehling > wrote: >> This patch series was meant to be applied after the userptr changes. I >> haven't tested this without the userptr changes. >> >> I think your main concern about