Re: [PATCH v11] drm/amdgpu: add drm buddy support to amdgpu

Am 29.03.22 um 21:18 schrieb Arunpravin Paneer Selvam: On 29/03/22 9:30 pm, Arunpravin Paneer Selvam wrote: On 29/03/22 4:54 pm, Christian König wrote: Am 29.03.22 um 13:19 schrieb Arunpravin Paneer Selvam: [SNIP] + pages_left = node->base.num_pages; i = 0; - spin_loc

Re: [PATCH V2] drm/amdgpu: fix incorrect GCR_GENERAL_CNTL address

Dear Ruili, Am 29.03.22 um 11:17 schrieb Ji, Ruili: This is not related to any issue. I didn’t mean an issue (where I’d use Resolves to differentiate the two cases), but the commit introducing the incorrect address. Kind regards, Paul PS: Please use interleaved style when replying ins

Re: [PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

[Cc: Remove undeliverable Chun Ming Zhou ] Am 30.03.22 um 08:34 schrieb Paul Menzel: Dear Tsung-Hua, Thank you for your patch. Am 30.03.22 um 04:46 schrieb Ryan Lin: The commit message summary is quite long and confusing. Maybe: Use 3 CRTC for 3250c to get intern

Re: [PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

Dear Tsung-Hua, Thank you for your patch. Am 30.03.22 um 04:46 schrieb Ryan Lin: The commit message summary is quite long and confusing. Maybe: Use 3 CRTC for 3250c to get internal display working [Why] External displays take priority over internal display when there are fewer display contr

RE: [PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

[AMD Official Use Only] Hi Chandan, This issue we found on the Zork project which uses the kernel 5.4 on. So I just implemented it on kernel 5.4. For finding out which is 3250c, I referenced the function which is implemented from another function. Below is the part where I found it. drivers/gp

RE: [PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

Hi Ryan, Is this change applicable on a specific kernel version? On latest I see IP DISCOVERY based impl for CHIP_RAVEN. >[Why] >External displays take priority over internal display when there are fewer >display controllers than displays. > > [How] >The root cause is because of that number of

Re: [PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

On Tue, Mar 29, 2022 at 10:57 PM Ryan Lin wrote: > > [Why] > External displays take priority over internal display when there are fewer > display controllers than displays. > > [How] > The root cause is because of that number of the crtc is not correct. > The number of the crtc on the 3250c is 3,

[PATCH v2] drm/amdgpu: fix that issue that the number of the crtc of the 3250c is not correct

[Why] External displays take priority over internal display when there are fewer display controllers than displays. [How] The root cause is because of that number of the crtc is not correct. The number of the crtc on the 3250c is 3, but on the 3500c is 4. On the source code, we can see that number

RE: [PATCH] drm/amd/display: Fix by adding FPU protection for dcn30_internal_validate_bw

Hi Paul, >Am 29.03.22 um 10:29 schrieb CHANDAN VURDIGERE NATARAJ: > >Is it common to spell your name all uppercase? If not, please use Chandan >nVurdigere Nataraj. > >> [WHY] > >The [] already emphasize the word, so Why could be used. > >> Below general protection fault observed when WebGL Aquari

[PATCH v2] drm/amd/display: don't ignore alpha property on pre-multiplied mode

"Pre-multiplied" is the default pixel blend mode for KMS/DRM, as documented in supported_modes of drm_plane_create_blend_mode_property(): https://cgit.freedesktop.org/drm/drm-misc/tree/drivers/gpu/drm/drm_blend.c In this mode, both 'pixel alpha' and 'plane alpha' participate in the calculation, as

Re: [PATCH v11] drm/amdgpu: add drm buddy support to amdgpu

On 29/03/22 9:30 pm, Arunpravin Paneer Selvam wrote: > > > On 29/03/22 4:54 pm, Christian König wrote: >> Am 29.03.22 um 13:19 schrieb Arunpravin Paneer Selvam: >>> [SNIP] > + pages_left = node->base.num_pages; > > i = 0; > - spin_lock(&mgr->lock); >

RE: [PATCH 4/4] drm/amdgpu: Add unique_id support for sienna cichlid

[AMD Official Use Only] > -Original Message- > From: Alex Deucher > Sent: Tuesday, March 29, 2022 11:28 AM > To: Russell, Kent > Cc: amd-gfx list > Subject: Re: [PATCH 4/4] drm/amdgpu: Add unique_id support for sienna cichlid > > On Tue, Mar 29, 2022 at 11:10 AM Kent Russell wrote: > >

Re: [PATCH] drm/amd/display: Fix by adding FPU protection for dcn30_internal_validate_bw

Dear Chandan, Am 29.03.22 um 10:29 schrieb CHANDAN VURDIGERE NATARAJ: Is it common to spell your name all uppercase? If not, please use Chandan nVurdigere Nataraj. [WHY] The [] already emphasize the word, so Why could be used. Below general protection fault observed when WebGL Aquarium

Re: [RFC PATCH] drm/amd/display: dont ignore alpha property

On 03/28, Melissa Wen wrote: > On 03/28, Kazlauskas, Nicholas wrote: > > [AMD Official Use Only] > > > > > -Original Message- > > > From: Melissa Wen > > > Sent: Friday, March 25, 2022 4:45 PM > > > To: amd-gfx@lists.freedesktop.org; Wentland, Harry > > > ; Deucher, Alexander > > > ; Siqu

Re: [PATCH v2 1/2] drm: Add GPU reset sysfs event

I don't know what iris does, but I would guess that the same problems as with AMD GPUs apply, making GPUs resets very fragile. Marek On Tue., Mar. 29, 2022, 08:14 Christian König, wrote: > My main question is what does the iris driver better than radeonsi when > the client doesn't support the r

Re: [PATCH v11] drm/amdgpu: add drm buddy support to amdgpu

On 29/03/22 4:54 pm, Christian König wrote: > Am 29.03.22 um 13:19 schrieb Arunpravin Paneer Selvam: >> [SNIP] + pages_left = node->base.num_pages; i = 0; - spin_lock(&mgr->lock); while (pages_left) { - uint32_t alignment = tbo->

Re: [PATCH 18/23] drm/amdgpu: remove dma_resv workaround

On Mon, Mar 21, 2022 at 02:58:51PM +0100, Christian König wrote: > We can now add multiple writers to the dma_resv object. > > Also enable the check for not adding containers in dma_resv.c again. > > Signed-off-by: Christian König > Cc: amd-gfx@lists.freedesktop.org It's a bit much magic, but t

Re: [PATCH 4/4] drm/amdgpu: Add unique_id support for sienna cichlid

On Tue, Mar 29, 2022 at 11:10 AM Kent Russell wrote: > > This is being added to SMU Metrics, so add the required tie-ins in the > kernel. Also create the corresponding unique_id sysfs file. > > v2: Add FW version check, remove SMU mutex > v3: Fix style warning > v4: Add MP1 IP_VERSION check to FW

Re: [PATCH] drm/amd/display: Fix by adding FPU protection for dcn30_internal_validate_bw

[Public] Acked-by: Alex Deucher From: amd-gfx on behalf of CHANDAN VURDIGERE NATARAJ Sent: Tuesday, March 29, 2022 4:29 AM To: Wentland, Harry ; Li, Sun peng (Leo) ; Deucher, Alexander Cc: VURDIGERENATARAJ, CHANDAN ; amd-gfx@lists.freedesktop.org Subject: [

[PATCH 4/4] drm/amdgpu: Add unique_id support for sienna cichlid

This is being added to SMU Metrics, so add the required tie-ins in the kernel. Also create the corresponding unique_id sysfs file. v2: Add FW version check, remove SMU mutex v3: Fix style warning v4: Add MP1 IP_VERSION check to FW version check Signed-off-by: Kent Russell --- drivers/gpu/drm/am

[PATCH 3/4] drm/amdgpu: Use metrics data function to get unique_id for Aldebaran

This is abstracted well enough in the get_metrics_data function, so use the function Signed-off-by: Kent Russell --- .../gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 16 +--- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_

[PATCH 1/4] drm/amdgpu: Use switch case for unique_id

To ease readability, use switch to set unique_id as supported for the supported IP_VERSIONs, and set it to unsupported by default for all other ASICs. This makes it easier to add IP_VERSIONs later on, and makes it obvious that it is not supported by default, instead of the current logic that assume

[PATCH 2/4] drm/amdgpu: Add UNIQUE_ID to MetricsMember_t

This will allow us to use the generic *_get_metrics_data functions for ASICs that support unique_id Signed-off-by: Kent Russell --- drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm

Re: [PATCH] UPSTREAM: drm/amdgpu: Disable ABM when AC mode

On 2022-03-29 10:33, Harry Wentland wrote: > > > On 2022-03-24 19:10, Ryan Lin wrote: >> Disable ABM feature when the system is running on AC mode to get >> the more perfect contrast of the display. >> >> Signed-off-by: Ryan Lin >> >> --- >> drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4

Re: [PATCH] UPSTREAM: drm/amdgpu: Disable ABM when AC mode

On 2022-03-24 19:10, Ryan Lin wrote: > Disable ABM feature when the system is running on AC mode to get > the more perfect contrast of the display. > > Signed-off-by: Ryan Lin > > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4 ++ > drivers/gpu/drm/amd/amdgpu/amdgpu_device.c|

Re: drm/amdgpu: Disable ABM when AC mode

On Tue, Mar 29, 2022 at 4:56 AM Ryan Lin wrote: > > Disable ABM feature when the system is running on AC mode to get > the more perfect contrast of the display. > > v2: remove "UPSTREAM" from the subject. > > Signed-off-by: Ryan Lin > > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4 ++

Re: [PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

Well that is a really bad idea. Tools should only use sysfs functionality, not debugfs since that isn't stable. And I totally agree to Hawking here. As I wrote on the other mail this patch is a no-go. Regards, Christian. Am 29.03.22 um 15:18 schrieb Chai, Thomas: [AMD Official Use Only]

RE: [PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

[AMD Official Use Only] Sorry for the confusing commit message。 This interface is only for amdgpu ras tool new function. There is no impact on currently existing tools and scripts。 -Original Message- From: Zhang, Hawking Sent: Tuesday, March 29, 2022 4:33 PM To: Chai, Thomas ; amd-gfx

Re: [PATCH v2 1/2] drm: Add GPU reset sysfs event

My main question is what does the iris driver better than radeonsi when the client doesn't support the robustness extension? From Daniels description it sounds like they have at least a partial recovery mechanism in place. Apart from that I completely agree to what you said below. Christian.

RE: [PATCH 2/2] drm/amdgpu: Add unique_id support for sienna cichlid

[AMD Official Use Only] Ah I didn’t see that we added a Metrics v3 in there between now and when I made my first version of this patch. I'll test it out and make sure that things are still looking alright. Thanks Lijo! Kent > -Original Message- > From: Lazar, Lijo > Sent: Tuesday, Ma

Re: [PATCH v2] drm: add a check to verify the size alignment

Am 29.03.22 um 13:28 schrieb Arunpravin Paneer Selvam: On 23/03/22 1:15 pm, Christian König wrote: Am 23.03.22 um 08:34 schrieb Arunpravin Paneer Selvam: Add a simple check to reject any size not aligned to the min_page_size. handle instances when size is not aligned with the min_page_size. Un

Re: [PATCH v2] drm: add a check to verify the size alignment

On Tue, 29 Mar 2022 at 12:17, Arunpravin Paneer Selvam wrote: > > > > On 23/03/22 1:15 pm, Christian König wrote: > > Am 23.03.22 um 08:34 schrieb Arunpravin Paneer Selvam: > >> Add a simple check to reject any size not aligned to the > >> min_page_size. > >> > >> handle instances when size is not

Re: [PATCH v11] drm/amdgpu: add drm buddy support to amdgpu

Am 29.03.22 um 13:19 schrieb Arunpravin Paneer Selvam: [SNIP] + pages_left = node->base.num_pages; i = 0; - spin_lock(&mgr->lock); while (pages_left) { - uint32_t alignment = tbo->page_alignment; + if (tbo->page_alignment) +

Re: [PATCH v2] drm: add a check to verify the size alignment

On 23/03/22 1:15 pm, Christian König wrote: > Am 23.03.22 um 08:34 schrieb Arunpravin Paneer Selvam: >> Add a simple check to reject any size not aligned to the >> min_page_size. >> >> handle instances when size is not aligned with the min_page_size. >> Unigine Heaven has allocation requests for

Re: [PATCH v11] drm/amdgpu: add drm buddy support to amdgpu

> -Original Message- > From: amd-gfx On Behalf Of Christian > König > Sent: Wednesday, March 23, 2022 1:07 PM > To: Paneer Selvam, Arunpravin ; > intel-...@lists.freedesktop.org; dri-de...@lists.freedesktop.org; > amd-gfx@lists.freedesktop.org > Cc: Deucher, Alexander ; matthew.a...@i

RE: [PATCH V2] drm/amdgpu: fix incorrect GCR_GENERAL_CNTL address

[AMD Official Use Only] Hi Paul, This is not related to any issue. Kind regards, Ruili -Original Message- From: Paul Menzel Sent: 2022年3月29日 16:16 To: Ji, Ruili Cc: amd-gfx@lists.freedesktop.org; Zhang, Yifan ; Liu, Aaron ; Liang, Prike ; Huang, Ray ; Deucher, Alexander ; Ji, Ruili

drm/amdgpu: Disable ABM when AC mode

Disable ABM feature when the system is running on AC mode to get the more perfect contrast of the display. v2: remove "UPSTREAM" from the subject. Signed-off-by: Ryan Lin --- drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c| 1 + drivers/gp

Re: [PATCH next, v2] kernel: Add 1 ms delay to init handler to fix s3 resume hang

On Tue, Mar 29, 2022 at 08:20:24AM +0200, Christian König wrote: > Am 29.03.22 um 05:05 schrieb Zhenneng Li: > > This is a workaround for s3 resume hang for r7 340(amdgpu). > > When we test s3 with r7 340 on arm64 platform, graphics card will hang up, > > the error message are as follows: > > Mar

RE: [PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

[AMD Official Use Only] I'm not sure I understand the fix correctly - It seems to me it is trying to stop user/test cases that initiate error injection request back-to-back? But anyway, we shouldn't make the change or leverage debugfs for that purpose, and there is no guarantee test scripts/app

[PATCH] drm/amd/display: Fix by adding FPU protection for dcn30_internal_validate_bw

[WHY] Below general protection fault observed when WebGL Aquarium is run for longer duration. If drm debug logs are enabled and set to 0x1f then the issue is observed within 10 minutes of run. [ 100.717056] general protection fault, probably for non-canonical address 0x2d33302d32323032: [#1

Re: [PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

Dear Yi Peng, Thank you for the patch. Two nits regarding the commit message. Am 29.03.22 um 09:38 schrieb yipechai: Some AMDGPU RAS debugfs operations like UE injection can cause gpu reset. Before doing the next debugfs operation, the application should call poll to check if the gpu has finis

Re: [PATCH V2] drm/amdgpu: fix incorrect GCR_GENERAL_CNTL address

Dear Ruili, Thank you for your patch. Am 28.03.22 um 06:58 schrieb Ji, Ruili: From: Ruili Ji gfx10.3.3/gfx10.3.6/gfx10.3.7 shall use 0x1580 address for GCR_GENERAL_CNTL Is any “user-visible“ problem fixed by this? Please add a Fixes tag. Kind regards, Paul Signed-off-by: Ruili Ji -

Re: [PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

Am 29.03.22 um 09:38 schrieb yipechai: Some AMDGPU RAS debugfs operations like UE injection can cause gpu reset. Before doing the next debugfs operation, the application should call poll to check if the gpu has finished recovering. Well NAK. debugfs files are designed to be used from the comman

Re: [Intel-gfx] [PATCH v2 03/11] drm/edid: slightly restructure timing and non-timing descriptor structs

I think I'm just going to revert back to my original plan of leaving the struct restructuring to another time in the future. BR, Jani. On Mon, 28 Mar 2022, kernel test robot wrote: > Hi Jani, > > I love your patch! Perhaps something to improve: > > [auto build test WARNING on drm/drm-next] >

[PATCH next, v2] kernel: Add 1 ms delay to init handler to fix s3 resume hang

This is a workaround for s3 resume hang for r7 340(amdgpu). When we test s3 with r7 340 on arm64 platform, graphics card will hang up, the error message are as follows: Mar 4 01:14:11 greatwall-GW-XX-XXX kernel: [1.599374][ 7] [ T291] amdgpu :02:00.0: fb0: amdgpudrmfb frame buffer de

回复: Re: [PATCH] drm/amdgpu: resolve s3 hang for r7340

               主　题：Re: [PATCH] drm/amdgpu: resolve s3 hang for r7340            日　期：2022-03-28 15:38            发件人：Paul Menzel            收件人：Zhenneng Li                            [Cc: -Jack Zhang (invalid address)Am 28.03.22 um 09:36 schrieb Paul Menzel:> Dear Zhenneng,> > > Thank

[PATCH] drm/amdgpu: Support AMDGPU RAS debugfs poll interface

Some AMDGPU RAS debugfs operations like UE injection can cause gpu reset. Before doing the next debugfs operation, the application should call poll to check if the gpu has finished recovering. Signed-off-by: yipechai --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 38 - dri

RE: [PATCH V2] drm/amdgpu: fix incorrect GCR_GENERAL_CNTL address

[AMD Official Use Only] Reviewed-by: Yifan Zhang -Original Message- From: amd-gfx On Behalf Of Ji, Ruili Sent: Monday, March 28, 2022 12:59 PM To: amd-gfx@lists.freedesktop.org Cc: Zhang, Yifan ; Liu, Aaron ; Liang, Prike ; Huang, Ray ; Deucher, Alexander ; Ji, Ruili Subject: [PATCH