Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1598365 > unrolled thread

[PATCH] drm/radeon: Fix GPU lockups for the R7 M270

Started byUmang Raghuvanshi <u@umangis.me>
First post2017-03-11 17:30 +0100
Last post2017-03-14 15:40 +0100
Articles 4 — 2 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH] drm/radeon: Fix GPU lockups for the R7 M270 Umang Raghuvanshi <u@umangis.me> - 2017-03-11 17:30 +0100
    Re: [PATCH] drm/radeon: Fix GPU lockups for the R7 M270 Alex Deucher <alexdeucher@gmail.com> - 2017-03-13 20:40 +0100
      Re: [PATCH] drm/radeon: Fix GPU lockups for the R7 M270 Umang Raghuvanshi <u@umangis.me> - 2017-03-14 03:50 +0100
        Re: [PATCH] drm/radeon: Fix GPU lockups for the R7 M270 Alex Deucher <alexdeucher@gmail.com> - 2017-03-14 15:40 +0100

#1598365 — [PATCH] drm/radeon: Fix GPU lockups for the R7 M270

FromUmang Raghuvanshi <u@umangis.me>
Date2017-03-11 17:30 +0100
Subject[PATCH] drm/radeon: Fix GPU lockups for the R7 M270
Message-ID<tjS42-5lj-19@gated-at.bofh.it>
Commit 3a69adfe5617 ("drm/radeon: drop oland quirks") removed
quirks for Oland but also caused a regression where M270 GPUs
would go into a lock-up when OpenGL intensive applications were used.
This reverts the change only for the M270 and fixes the lock-ups.

Signed-off-by: Umang Raghuvanshi <u@umangis.me>
---
 drivers/gpu/drm/radeon/si_dpm.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index d12b8978142f..c2c129eef3c4 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -2984,6 +2984,11 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		    (rdev->pdev->device == 0x6667)) {
 			max_sclk = 75000;
 		}
+	} else if (rdev->family == CHIP_OLAND && rdev->pdev->device == 0x6604 &&
+		   rdev->pdev->revision == 0) {
+		// Fix max_sclk and max_mclk for the Radeon R7 M270
+		max_sclk = 75000;
+		max_mclk = 80000;
 	}
 
 	if (rps->vce_active) {
-- 
2.12.0

[toc] | [next] | [standalone]


#1599734

FromAlex Deucher <alexdeucher@gmail.com>
Date2017-03-13 20:40 +0100
Message-ID<tkDZ1-5cJ-39@gated-at.bofh.it>
In reply to#1598365
On Sat, Mar 11, 2017 at 11:21 AM, Umang Raghuvanshi <u@umangis.me> wrote:
> Commit 3a69adfe5617 ("drm/radeon: drop oland quirks") removed
> quirks for Oland but also caused a regression where M270 GPUs
> would go into a lock-up when OpenGL intensive applications were used.
> This reverts the change only for the M270 and fixes the lock-ups.

Does your kernel have this patch:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ef736d394e85b1bf1fd65ba5e5257b85f6c82325

Alex

>
> Signed-off-by: Umang Raghuvanshi <u@umangis.me>
> ---
>  drivers/gpu/drm/radeon/si_dpm.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
> index d12b8978142f..c2c129eef3c4 100644
> --- a/drivers/gpu/drm/radeon/si_dpm.c
> +++ b/drivers/gpu/drm/radeon/si_dpm.c
> @@ -2984,6 +2984,11 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
>                     (rdev->pdev->device == 0x6667)) {
>                         max_sclk = 75000;
>                 }
> +       } else if (rdev->family == CHIP_OLAND && rdev->pdev->device == 0x6604 &&
> +                  rdev->pdev->revision == 0) {
> +               // Fix max_sclk and max_mclk for the Radeon R7 M270
> +               max_sclk = 75000;
> +               max_mclk = 80000;
>         }
>
>         if (rps->vce_active) {
> --
> 2.12.0
>
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel

[toc] | [prev] | [next] | [standalone]


#1600001

FromUmang Raghuvanshi <u@umangis.me>
Date2017-03-14 03:50 +0100
Message-ID<tkKH7-1Sd-9@gated-at.bofh.it>
In reply to#1599734

[Multipart message — attachments visible in raw view] — view raw

On 03/14/2017 01:00 AM, Alex Deucher wrote:

> Does your kernel have this patch:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ef736d394e85b1bf1fd65ba5e5257b85f6c82325

Yes, my kernel has this patch (this issue first occurred in v4.10).

-- 
Cheers,
Umang Raghuvanshi.

[toc] | [prev] | [next] | [standalone]


#1600500

FromAlex Deucher <alexdeucher@gmail.com>
Date2017-03-14 15:40 +0100
Message-ID<tkVMd-1qV-1@gated-at.bofh.it>
In reply to#1600001
On Mon, Mar 13, 2017 at 10:42 PM, Umang Raghuvanshi <u@umangis.me> wrote:
>
>
> On 03/14/2017 01:00 AM, Alex Deucher wrote:
>
>> Does your kernel have this patch:
>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ef736d394e85b1bf1fd65ba5e5257b85f6c82325
>
> Yes, my kernel has this patch (this issue first occurred in v4.10).


Can you send me the dmesg output from your system?

Thanks,

Alex

>
> --
> Cheers,
> Umang Raghuvanshi.
>

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web