RE: [PATCH] drm/amdgpu: Field type update in drm_amdgpu_info_vbios

[AMD Official Use Only - Internal Distribution Only] Done. UMR patch sent out. Hi Tom, Please help review it to bring the umr part back. Thanks in advance! Jiawei From: Deucher, Alexander Sent: Tuesday, May 11, 2021 9:16 PM To: Gu, JiaWei (Will) ; amd-gfx@lists.freedesktop.org Cc: StDenis,

Re: [PATCH] drm/amdgpu: Add compile flag for securedisplay

On Wed, May 12, 2021 at 09:51:21AM +0800, Su, Jinzhou (Joe) wrote: > Add compile flag CONFIG_DEBUG_FS to clear the warning: > unused variable 'amdgpu_securedisplay_debugfs_ops' > > Signed-off-by: Jinzhou Su According to the 0 day's robot mail, please feel free to add below reported-by:

[PATCH] add vbios info query

Signed-off-by: Jiawei Gu --- src/app/CMakeLists.txt | 1 + src/app/main.c | 8 + src/app/vbios.c| 58 ++ src/lib/lowlevel/linux/query_drm.c | 11 ++ src/umr.h | 15

Re: [PATCH] drm/amdgpu: Add compile flag for securedisplay

[AMD Official Use Only - Internal Distribution Only] Acked-by: Alex Deucher From: amd-gfx on behalf of Jinzhou Su Sent: Tuesday, May 11, 2021 9:51 PM To: amd-gfx@lists.freedesktop.org Cc: Su, Jinzhou (Joe) ; Huang, Ray Subject: [PATCH] drm/amdgpu: Add

[PATCH] drm/amd/amdgpu: psp program IH_RB_CTRL on navi12 and sienna_cichlid

[Why] IH_RB_CNTL is blocked by PSP so we need to ask psp to help config it. [How] Move psp ip block before ih, and use psp to program IH_RB_CNTL under sriov. Signed-off-by: YuBiao Wang --- drivers/gpu/drm/amd/amdgpu/navi10_ih.c | 20 ++-- drivers/gpu/drm/amd/amdgpu/nv.c

[PATCH] drm/amdgpu: Add compile flag for securedisplay

Add compile flag CONFIG_DEBUG_FS to clear the warning: unused variable 'amdgpu_securedisplay_debugfs_ops' Signed-off-by: Jinzhou Su --- drivers/gpu/drm/amd/amdgpu/amdgpu_securedisplay.c | 4 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_securedisplay.c

[PATCH] drm/amdgpu: Register bad page handler for Aldebaran

On Aldebaran, GPU driver will handle bad page retirement even though UMC is host managed. As a result, register a bad page retirement handler on the mce notifier chain to retire bad pages on Aldebaran. Signed-off-by: Mukul Joshi Reviewed-by: John Clements Acked-by: Felix Kuehling ---

Re: [PATCH v2] drm/radeon/dpm: Disable sclk switching on Oland when two 4K 60Hz monitors are connected

On Mon, May 10, 2021 at 11:33 PM Kai-Heng Feng wrote: > > On Fri, Apr 30, 2021 at 12:57 PM Kai-Heng Feng > wrote: > > > > Screen flickers rapidly when two 4K 60Hz monitors are in use. This issue > > doesn't happen when one monitor is 4K 60Hz (pixelclock 594MHz) and > > another one is 4K 30Hz

[PATCH 2/2] drm/amdgpu: fix fence calculation

The proper metric for fence utilization over several contexts is an harmonic mean, but such calculation is prohibitive in kernel space, so the code approximates it. Because the approximation diverges when one context has a very small ratio compared with the other context, this change filter out

[PATCH 1/2] drm/amdgpu: free resources on fence usage query

Free the resources if the fence needs to be ignored during the ratio calculation Signed-off-by: David M Nieto Change-Id: Ibfc55a94c53d4b3a1dba8fff4c53fd893195bb96 --- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git

Re: [PATCH 2/2] drm/amdgpu: fix fence calculation

[AMD Official Use Only - Internal Distribution Only] The local variables need to be initialized to zero, since amdgpu_ctx_fence_time accumulates and does not initialize David From: Christian König Sent: Tuesday, May 11, 2021 12:53 AM To: Nieto, David M ;

Re: [PATCH] drm/amdgpu: fix uninitialized return value

On Tue, May 11, 2021 at 1:25 PM ts8060 wrote: > > The amdgpu_display_suspend_helper() function uses an uninitialized > variable as its return value, causing a failure to suspend/resume on a > Radeon R7 240/340 GPU, with the following error in dmesg: > [drm:amdgpu_device_ip_suspend_phase1

Re: [PATCH v6 10/16] drm/amdgpu: Guard against write accesses after device removal

On 2021-05-11 2:50 a.m., Christian König wrote: Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: This should prevent writing to memory or IO ranges possibly already allocated for other uses after our device is removed. v5: Protect more places wher memcopy_to/form_io takes place Protect IB

[PATCH] drm/amdgpu: fix uninitialized return value

The amdgpu_display_suspend_helper() function uses an uninitialized variable as its return value, causing a failure to suspend/resume on a Radeon R7 240/340 GPU, with the following error in dmesg: [drm:amdgpu_device_ip_suspend_phase1 [amdgpu]] *ERROR* suspend of IP block failed 118489088

Re: [PATCH] drm/amdgpu: Align serial size in drm_amdgpu_info_vbios

[AMD Public Use] The point of having the device ID in the structure is because we are reading it from the VBIOS header, not the asic registers. They should match, but an user may flash a VBIOS for a different devid and they may not match. Regarding sysfs vs ioctl I see value in providing it in

Re: [PATCH v6 06/16] drm/amdgpu: Handle IOMMU enabled case.

On 2021-05-11 11:56 a.m., Alex Deucher wrote: On Mon, May 10, 2021 at 12:37 PM Andrey Grodzovsky wrote: Handle all DMA IOMMU gropup related dependencies before the group is removed. v5: Drop IOMMU notifier and switch to lockless call to ttm_tt_unpopulate v6: Drop the BO unamp list

Re: [PATCH v6 06/16] drm/amdgpu: Handle IOMMU enabled case.

On Mon, May 10, 2021 at 12:37 PM Andrey Grodzovsky wrote: > > Handle all DMA IOMMU gropup related dependencies before the > group is removed. > > v5: Drop IOMMU notifier and switch to lockless call to ttm_tt_unpopulate > v6: Drop the BO unamp list > > Signed-off-by: Andrey Grodzovsky > --- >

Re: [RFC] Add BPF_PROG_TYPE_CGROUP_IOCTL

On Fri, May 7, 2021 at 7:45 PM Tejun Heo wrote: > > Hello, > > On Fri, May 07, 2021 at 06:30:56PM -0400, Alex Deucher wrote: > > Maybe we are speaking past each other. I'm not following. We got > > here because a device specific cgroup didn't make sense. With my > > Linux user hat on, that

Re: [PATCH v6 06/16] drm/amdgpu: Handle IOMMU enabled case.

On 2021-05-11 2:44 a.m., Christian König wrote: Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Handle all DMA IOMMU gropup related dependencies before the group is removed. v5: Drop IOMMU notifier and switch to lockless call to ttm_tt_unpopulate v6: Drop the BO unamp list Signed-off-by:

[PATCH 2/2] drm/amdgpu/display: restore the backlight on modeset (v2)

To stay consistent with the user's setting. v2: rebase on multi-eDP support Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1337 Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 6 ++ 1 file changed, 6 insertions(+) diff --git

[PATCH 1/2] drm/amdgpu/display: add helper functions to get/set backlight (v2)

And cache the value. These can be used by the backlight callbacks and modesetting functions. v2: rebase on latest backlight changes. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1337 Signed-off-by: Alex Deucher --- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 42 ++-

Re: [PATCH v2 2/2] drm/amdkfd: new range accessible by all GPUs

Am 2021-05-11 um 11:15 a.m. schrieb Philip Yang: > If xnack is on, new range is created to recover retry vm fault or > created by SVM API calls, set all GPUs have access to the range. > > Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling > --- > drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 12

[PATCH v2 2/2] drm/amdkfd: new range accessible by all GPUs

If xnack is on, new range is created to recover retry vm fault or created by SVM API calls, set all GPUs have access to the range. Signed-off-by: Philip Yang --- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 12 1 file changed, 8 insertions(+), 4 deletions(-) diff --git

Re: [PATCH 2/2] drm/amdkfd: unregistered range accessible by all GPUs

On 2021-05-10 8:56 a.m., Felix Kuehling wrote: Am 2021-05-07 um 3:07 p.m. schrieb Philip Yang: New range is created to recover retry vm fault, set all GPUs have access to the range. The new range preferred_loc is default value

Re: [PATCH v6 01/16] drm/ttm: Remap all page faults to per process dummy page.

Am 11.05.21 um 16:44 schrieb Andrey Grodzovsky: On 2021-05-11 2:38 a.m., Christian König wrote: Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: On device removal reroute all CPU mappings to dummy page. v3: Remove loop to find DRM file and instead access it by vma->vm_file->private_data.

Re: [PATCH v6 04/16] drm/amdkfd: Split kfd suspend from devie exit

On 2021-05-11 2:40 a.m., Christian König wrote: Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Helps to expdite HW related stuff to amdgpu_pci_remove Signed-off-by: Andrey Grodzovsky ---   drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +-   drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 2 +-

Re: [PATCH v6 01/16] drm/ttm: Remap all page faults to per process dummy page.

On 2021-05-11 2:38 a.m., Christian König wrote: Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: On device removal reroute all CPU mappings to dummy page. v3: Remove loop to find DRM file and instead access it by vma->vm_file->private_data. Move dummy page installation into a separate

Re: [PATCH] drm/amdgpu: Align serial size in drm_amdgpu_info_vbios

Yeah, but umr is making strong use of sysfs as well. The only justification of this interface would be if we want to use it in Mesa. And I agree with Marek that looks redundant with the device structure to me as well. Christian. Am 11.05.21 um 16:04 schrieb Deucher, Alexander: [AMD

Re: [PATCH] drm/amdgpu: Align serial size in drm_amdgpu_info_vbios

[AMD Public Use] It's being used by umr and some other smi tools to provide vbios information for debugging. Alex From: amd-gfx on behalf of Marek Olšák Sent: Tuesday, May 11, 2021 4:18 AM To: Christian König Cc: Kees Cook ; Gu, JiaWei (Will) ; amd-gfx

Re: [PATCH] drm/amdkfd: refine the poison data consumption handling

Am 2021-05-11 um 4:06 a.m. schrieb Dennis Li: > The user applications maybe register the KFD_EVENT_TYPE_HW_EXCEPTION and I guess the HW exception event is sent because the current handling of poison consumption triggers a mode2 reset. If that can be removed in the future, then we should not send

Re: [PATCH] drm/ttm: use dma_alloc_pages for the page pool

On Tue, May 11, 2021 at 09:35:20AM +0200, Christian König wrote: > We certainly going to need the drm_need_swiotlb() for userptr support > (unless we add some approach for drivers to opt out of swiotlb). swiotlb use is driven by three things: 1) addressing limitations of the device 2)

Re: [PATCH v6 04/16] drm/amdkfd: Split kfd suspend from devie exit

[AMD Public Use] Typo in the subject: devie > device Alex From: Grodzovsky, Andrey Sent: Monday, May 10, 2021 12:36 PM To: dri-de...@lists.freedesktop.org ; amd-gfx@lists.freedesktop.org ; linux-...@vger.kernel.org ; ckoenig.leichtzumer...@gmail.com ;

Re: [PATCH] drm/amdgpu: Field type update in drm_amdgpu_info_vbios

[AMD Official Use Only - Internal Distribution Only] Please update the umr patch as well. Acked-by: Alex Deucher From: amd-gfx on behalf of Jiawei Gu Sent: Tuesday, May 11, 2021 1:31 AM To: amd-gfx@lists.freedesktop.org Cc: StDenis, Tom ; Gu, JiaWei (Will)

RE: [PATCH] drm/amd/pm: Update aldebaran pmfw interface

[AMD Public Use] Reviewed-by: Hawking Zhang Regards, Hawking _ From: Lazar, Lijo Sent: Tuesday, May 11, 2021 20:30 To: amd-gfx@lists.freedesktop.org Cc: Zhang, Hawking ; Xu, Feifei ; Wang, Kevin(Yang) Subject: [PATCH] drm/amd/pm: Update aldebaran

[PATCH] drm/amd/pm: Update aldebaran pmfw interface

[AMD Public Use] Update aldebaran driver-PMFW interface to version 0x07 Signed-off-by: Lijo Lazar lijo.la...@amd.com --- drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 7 +-- drivers/gpu/drm/amd/pm/inc/smu_v13_0.h | 2 +- 2 files changed,

Re: [PATCH] drm/amd/display: Expose active display color configurations to userspace

On Tue, 11 May 2021 12:03:30 +0200 Werner Sembach wrote: > Am 11.05.21 um 10:07 schrieb Pekka Paalanen: > > On Mon, 10 May 2021 17:47:01 -0400 > > Alex Deucher wrote: > > > >> On Fri, May 7, 2021 at 3:27 PM Werner Sembach > >> wrote: > >>> xrandr --prop and other userspace info tools have

[PATCH] drm/amd/amdgpu: psp program IH_RB_CTRL on navi12 and navi21

[Why] IH_RB_CNTL is blocked by PSP so we need to ask psp to help config it. [How] Move psp ip block before ih, and use psp to program IH_RB_CNTL under sriov. Signed-off-by: YuBiao Wang --- drivers/gpu/drm/amd/amdgpu/navi10_ih.c | 20 ++-- drivers/gpu/drm/amd/amdgpu/nv.c

Re: [PATCH] drm/amd/display: Expose active display color configurations to userspace

Am 11.05.21 um 10:07 schrieb Pekka Paalanen: > On Mon, 10 May 2021 17:47:01 -0400 > Alex Deucher wrote: > >> On Fri, May 7, 2021 at 3:27 PM Werner Sembach >> wrote: >>> xrandr --prop and other userspace info tools have currently no way of >>> telling which color configuration is used on HDMI

Re: [PATCH] drm/ttm: use dma_alloc_pages for the page pool

Am 11.05.21 um 10:50 schrieb Christoph Hellwig: On Tue, May 11, 2021 at 09:35:20AM +0200, Christian König wrote: We certainly going to need the drm_need_swiotlb() for userptr support (unless we add some approach for drivers to opt out of swiotlb). swiotlb use is driven by three things: 1)

Re: [PATCH] drm/amdgpu: Align serial size in drm_amdgpu_info_vbios

Mesa doesn't use sysfs. Note that this is a uapi, meaning that once it's in the kernel, it can't be changed like that. What's the use case for this new interface? Isn't it partially redundant with the current device info structure, which seems to have the equivalent of dev_id and rev_id? Marek

Re: [PATCH] drm/amd/display: Expose active display color configurations to userspace

On Mon, 10 May 2021 17:47:01 -0400 Alex Deucher wrote: > On Fri, May 7, 2021 at 3:27 PM Werner Sembach > wrote: > > > > xrandr --prop and other userspace info tools have currently no way of > > telling which color configuration is used on HDMI and DP ports. > > > > The ongoing transsition from

[PATCH] drm/amdkfd: refine the poison data consumption handling

The user applications maybe register the KFD_EVENT_TYPE_HW_EXCEPTION and KFD_EVENT_TYPE_MEMORY events, driver could notify them when poison data consumed. Beside that, some applications maybe register SIGBUS signal hander. These applications will handle poison data by themselves, exit or re-create

Re: [PATCH 2/2] drm/amdgpu: fix fence calculation

Am 10.05.21 um 22:29 schrieb David M Nieto: The proper metric for fence utilization over several contexts is an harmonic mean, but such calculation is prohibitive in kernel space, so the code approximates it. Because the approximation diverges when one context has a very small ratio compared

Re: [PATCH] drm/amdgpu: Align serial size in drm_amdgpu_info_vbios

Marek and other userspace folks need to decide that. Basic question here is if Mesa is already accessing sysfs nodes for OpenGL or RADV. If that is the case then we should probably expose the information there as well. If that isn't the case (which I think it is) then we should implement it

RFC: use dma_alloc_noncoherent in ttm_pool_alloc_page

Hi all, the memory allocation for the TTM pool is a big mess with two allocation methods that both have issues, a layering violation and odd guessing of pools in the callers. This patch switches to the dma_alloc_noncoherent API instead fixing all of the above issues. Warning: i don't have any

Re: [PATCH v2] drm/amd/display: Fix two cursor duplication when using overlay

I using linux kernel 5.13 rc1, has "drm/amd/display: Reject non-zero src_y and src_x for video planes" patch, git bisect bad commit is "drm/amd/display: Fix two cursor duplication when using overlay", i build kernel 5.13 many times, reboot test many times. 2021-05-11 8:37 GMT+08:00, Rodrigo

[PATCH] drm/ttm: use dma_alloc_pages for the page pool

Use the dma_alloc_pages allocator for the TTM pool allocator. This allocator is a front end to the page allocator which takes the DMA mask of the device into account, thus offering the best of both worlds of the two existing allocator versions. This conversion also removes the ugly layering

Re: [PATCH v2] drm/radeon/dpm: Disable sclk switching on Oland when two 4K 60Hz monitors are connected

On Fri, Apr 30, 2021 at 12:57 PM Kai-Heng Feng wrote: > > Screen flickers rapidly when two 4K 60Hz monitors are in use. This issue > doesn't happen when one monitor is 4K 60Hz (pixelclock 594MHz) and > another one is 4K 30Hz (pixelclock 297MHz). > > The issue is gone after setting

Re: [PATCH] drm/amd/pm: enable ASPM by default

You could even write why it was disabled before and why you can enable it now again in the commit message. Regards //Ernst Den tis 11 maj 2021 kl 09:25 skrev Chen, Jiansong (Simon) < jiansong.c...@amd.com>: > [AMD Official Use Only - Internal Distribution Only] > > Better to make the commit msg

Re: [PATCH] drm/ttm: use dma_alloc_pages for the page pool

Am 11.05.21 um 08:05 schrieb Christoph Hellwig: Use the dma_alloc_pages allocator for the TTM pool allocator. This allocator is a front end to the page allocator which takes the DMA mask of the device into account, thus offering the best of both worlds of the two existing allocator versions.

RE: [PATCH] drm/amd/pm: enable ASPM by default

[AMD Official Use Only - Internal Distribution Only] Better to make the commit msg more specific, eg. Change predicate accordingly since aspm is enabled by default. Either way, Reviewed-by: Jiansong Chen Regards, Jiansong -Original Message- From: amd-gfx On Behalf Of Kenneth Feng

Re: [PATCH v6 16/16] drm/amdgpu: Verify DMA opearations from device are done

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: In case device remove is just simualted by sysfs then verify device doesn't keep doing DMA to the released memory after pci_remove is done. Signed-off-by: Andrey Grodzovsky Acked-by: Christian König ---

Re: [PATCH v6 13/16] drm/amdgpu: Fix hang on device removal.

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: If removing while commands in flight you cannot wait to flush the HW fences on a ring since the device is gone. Signed-off-by: Andrey Grodzovsky --- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 16 ++-- 1 file changed, 10

Re: [PATCH v6 12/16] drm/amdgpu: Prevent any job recoveries after device is unplugged.

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Return DRM_TASK_STATUS_ENODEV back to the scheduler when device is not present so they timeout timer will not be rearmed. v5: Update to match updated return values in enum drm_gpu_sched_stat Signed-off-by: Andrey Grodzovsky Reviewed-by:

Re: [PATCH v6 11/16] drm/sched: Make timeout timer rearm conditional.

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: We don't want to rearm the timer if driver hook reports that the device is gone. v5: Update drm_gpu_sched_stat values in code. Signed-off-by: Andrey Grodzovsky Reviewed-by: Christian König --- drivers/gpu/drm/scheduler/sched_main.c | 11

Re: [PATCH v6 10/16] drm/amdgpu: Guard against write accesses after device removal

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: This should prevent writing to memory or IO ranges possibly already allocated for other uses after our device is removed. v5: Protect more places wher memcopy_to/form_io takes place Protect IB submissions v6: Switch to !drm_dev_enter instead of

Re: [PATCH v6 06/16] drm/amdgpu: Handle IOMMU enabled case.

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Handle all DMA IOMMU gropup related dependencies before the group is removed. v5: Drop IOMMU notifier and switch to lockless call to ttm_tt_unpopulate v6: Drop the BO unamp list Signed-off-by: Andrey Grodzovsky ---

Re: [PATCH v6 05/16] drm/amdgpu: Add early fini callback

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Use it to call disply code dependent on device->drv_data before it's set to NULL on device unplug v5: Move HW finilization into this callback to prevent MMIO accesses post cpi remove. Signed-off-by: Andrey Grodzovsky Acked-by: Christian

Re: [PATCH v6 04/16] drm/amdkfd: Split kfd suspend from devie exit

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: Helps to expdite HW related stuff to amdgpu_pci_remove Signed-off-by: Andrey Grodzovsky --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 2 +- drivers/gpu/drm/amd/amdkfd/kfd_device.c| 3

Re: [PATCH v6 01/16] drm/ttm: Remap all page faults to per process dummy page.

Am 10.05.21 um 18:36 schrieb Andrey Grodzovsky: On device removal reroute all CPU mappings to dummy page. v3: Remove loop to find DRM file and instead access it by vma->vm_file->private_data. Move dummy page installation into a separate function. v4: Map the entire BOs VA space into on demand

RE: [PATCH] drm/amdgpu: add synchronization among waves in the same threadgroup

[AMD Public Use] Reviewed-by: Hawking Zhang Regards, Hawking -Original Message- From: Dennis Li Sent: Tuesday, May 11, 2021 14:04 To: amd-gfx@lists.freedesktop.org; Deucher, Alexander ; Kuehling, Felix ; Zhang, Hawking ; Koenig, Christian Cc: Li, Dennis Subject: [PATCH]

[PATCH] drm/amdgpu: add synchronization among waves in the same threadgroup

It is possible that the previous waves have exited before others are created, so the other waves maybe reuse pyhsical resouces left by previous ones. Therefore add barrier instruction to synchronize waves within the same threadgroup. Signed-off-by: Dennis Li diff --git