public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
* Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
@ 2025-05-07 13:10 machion
  2025-05-08 13:18 ` Alex Deucher
  0 siblings, 1 reply; 6+ messages in thread
From: machion @ 2025-05-07 13:10 UTC (permalink / raw)
  To: stable; +Cc: regressions, amd-gfx, alexander.deucher, christian.koenig

Hello kernel/driver developers,

I hope, with my information it's possible to find a bug/problem in the 
kernel. Otherwise I am sorry, that I disturbed you.
I only use LTS kernels, but I can narrow it down to a hand full of them, 
where it works.

The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD 7790/8GB 
RAM
I already asked the Manjaro community, but with no luck.

The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021), 
uses vulkan

---

I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15, 
6.1 and 6.6 and and the game was always unplayable, because the frames 
where around 1fps (performance of PC is not the problem).
I asked the mesa and cinnamon team for help in the past, but also with 
no luck.
It never worked, till on 2025-03-29 when I installed 6.12.19 for the 
first time and it worked!

But it only worked with 6.12.19, 6.12.20 and 6.12.21
When I updated to 6.12.25, it was back to unplayable.

For testing I installed 6.14.4 with the same result. It doesn't work.

I also compared file /proc/config.gz of both kernels (6.12.21 <> 
6.14.4), but can't seem to see drastic changes to the graphical part.

I presume it has something to do with amdgpu.

If you need more information, I would be happy to help.

Kind regards,
Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
  2025-05-07 13:10 Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem machion
@ 2025-05-08 13:18 ` Alex Deucher
  2025-06-13 19:38   ` machion
  0 siblings, 1 reply; 6+ messages in thread
From: Alex Deucher @ 2025-05-08 13:18 UTC (permalink / raw)
  To: machion; +Cc: stable, regressions, amd-gfx, alexander.deucher, christian.koenig

On Thu, May 8, 2025 at 9:13 AM <machion@disroot.org> wrote:
>
> Hello kernel/driver developers,
>
> I hope, with my information it's possible to find a bug/problem in the
> kernel. Otherwise I am sorry, that I disturbed you.
> I only use LTS kernels, but I can narrow it down to a hand full of them,
> where it works.
>
> The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD 7790/8GB
> RAM
> I already asked the Manjaro community, but with no luck.
>
> The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021),
> uses vulkan
>
> ---
>
> I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15,
> 6.1 and 6.6 and and the game was always unplayable, because the frames
> where around 1fps (performance of PC is not the problem).
> I asked the mesa and cinnamon team for help in the past, but also with
> no luck.
> It never worked, till on 2025-03-29 when I installed 6.12.19 for the
> first time and it worked!
>
> But it only worked with 6.12.19, 6.12.20 and 6.12.21
> When I updated to 6.12.25, it was back to unplayable.

Can you bisect to see what fixed it in 6.12.19 or what broke it in
6.12.25?  For example if it was working in 6.12.21 and not working in
6.12.25, you can bisect between 6.12.21 and .25.

Alex

>
> For testing I installed 6.14.4 with the same result. It doesn't work.
>
> I also compared file /proc/config.gz of both kernels (6.12.21 <>
> 6.14.4), but can't seem to see drastic changes to the graphical part.
>
> I presume it has something to do with amdgpu.
>
> If you need more information, I would be happy to help.
>
> Kind regards,
> Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
  2025-05-08 13:18 ` Alex Deucher
@ 2025-06-13 19:38   ` machion
  2025-06-16 13:29     ` Alex Deucher
  0 siblings, 1 reply; 6+ messages in thread
From: machion @ 2025-06-13 19:38 UTC (permalink / raw)
  To: Alex Deucher
  Cc: stable, regressions, amd-gfx, alexander.deucher, christian.koenig

Hi,
sorry for the delay.
Besides less time, I had to make myself familiar with bisecting and 
again kernel compiling. Last time I compiled the kernel myself was 
around 2010 I think.

Anyway it seems I found the bad commit. The result after bisecting 10 
commits is:

a53d959fe660341788cb8dbc3ac3330d90a09ecf is the first bad commit
commit a53d959fe660341788cb8dbc3ac3330d90a09ecf
Author: Christian König <christian.koenig@amd.com>
Date:   Thu Mar 20 14:46:18 2025 +0100

     drm/amdgpu: immediately use GTT for new allocations

     commit a755906fb2b8370c43e91ba437ae1b3e228e8b02 upstream.

     Only use GTT as a fallback if we already have a backing store. This
     prevents evictions when an application constantly allocates and 
frees new
     memory.

     Partially fixes
     https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2833985.

     Signed-off-by: Christian König <christian.koenig@amd.com>
     Fixes: 216c1282dde3 ("drm/amdgpu: use GTT only as fallback for 
VRAM|GTT")
     Acked-by: Alex Deucher <alexander.deucher@amd.com>
     Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
     Cc: stable@vger.kernel.org
     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

  drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

Marion


Am 2025-05-08 15:18, schrieb Alex Deucher:
> On Thu, May 8, 2025 at 9:13 AM <machion@disroot.org> wrote:
>> 
>> Hello kernel/driver developers,
>> 
>> I hope, with my information it's possible to find a bug/problem in the
>> kernel. Otherwise I am sorry, that I disturbed you.
>> I only use LTS kernels, but I can narrow it down to a hand full of 
>> them,
>> where it works.
>> 
>> The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD 
>> 7790/8GB
>> RAM
>> I already asked the Manjaro community, but with no luck.
>> 
>> The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021),
>> uses vulkan
>> 
>> ---
>> 
>> I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15,
>> 6.1 and 6.6 and and the game was always unplayable, because the frames
>> where around 1fps (performance of PC is not the problem).
>> I asked the mesa and cinnamon team for help in the past, but also with
>> no luck.
>> It never worked, till on 2025-03-29 when I installed 6.12.19 for the
>> first time and it worked!
>> 
>> But it only worked with 6.12.19, 6.12.20 and 6.12.21
>> When I updated to 6.12.25, it was back to unplayable.
> 
> Can you bisect to see what fixed it in 6.12.19 or what broke it in
> 6.12.25?  For example if it was working in 6.12.21 and not working in
> 6.12.25, you can bisect between 6.12.21 and .25.
> 
> Alex
> 
>> 
>> For testing I installed 6.14.4 with the same result. It doesn't work.
>> 
>> I also compared file /proc/config.gz of both kernels (6.12.21 <>
>> 6.14.4), but can't seem to see drastic changes to the graphical part.
>> 
>> I presume it has something to do with amdgpu.
>> 
>> If you need more information, I would be happy to help.
>> 
>> Kind regards,
>> Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
  2025-06-13 19:38   ` machion
@ 2025-06-16 13:29     ` Alex Deucher
  2025-06-17  9:42       ` machion
  0 siblings, 1 reply; 6+ messages in thread
From: Alex Deucher @ 2025-06-16 13:29 UTC (permalink / raw)
  To: machion; +Cc: stable, regressions, amd-gfx, alexander.deucher, christian.koenig

On Fri, Jun 13, 2025 at 3:38 PM <machion@disroot.org> wrote:
>
> Hi,
> sorry for the delay.
> Besides less time, I had to make myself familiar with bisecting and
> again kernel compiling. Last time I compiled the kernel myself was
> around 2010 I think.
>
> Anyway it seems I found the bad commit. The result after bisecting 10
> commits is:
>
> a53d959fe660341788cb8dbc3ac3330d90a09ecf is the first bad commit
> commit a53d959fe660341788cb8dbc3ac3330d90a09ecf
> Author: Christian König <christian.koenig@amd.com>
> Date:   Thu Mar 20 14:46:18 2025 +0100
>
>      drm/amdgpu: immediately use GTT for new allocations
>
>      commit a755906fb2b8370c43e91ba437ae1b3e228e8b02 upstream.
>
>      Only use GTT as a fallback if we already have a backing store. This
>      prevents evictions when an application constantly allocates and
> frees new
>      memory.
>
>      Partially fixes
>      https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2833985.
>
>      Signed-off-by: Christian König <christian.koenig@amd.com>
>      Fixes: 216c1282dde3 ("drm/amdgpu: use GTT only as fallback for
> VRAM|GTT")
>      Acked-by: Alex Deucher <alexander.deucher@amd.com>
>      Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
>      Cc: stable@vger.kernel.org
>      Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>
>   drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)

Unfortunately reverting that commit will reintroduce a similar
performance issue for lots of other uses.  See:
https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2827990
for a description of the fundemental problem.

Alex

>
> Marion
>
>
> Am 2025-05-08 15:18, schrieb Alex Deucher:
> > On Thu, May 8, 2025 at 9:13 AM <machion@disroot.org> wrote:
> >>
> >> Hello kernel/driver developers,
> >>
> >> I hope, with my information it's possible to find a bug/problem in the
> >> kernel. Otherwise I am sorry, that I disturbed you.
> >> I only use LTS kernels, but I can narrow it down to a hand full of
> >> them,
> >> where it works.
> >>
> >> The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD
> >> 7790/8GB
> >> RAM
> >> I already asked the Manjaro community, but with no luck.
> >>
> >> The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021),
> >> uses vulkan
> >>
> >> ---
> >>
> >> I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15,
> >> 6.1 and 6.6 and and the game was always unplayable, because the frames
> >> where around 1fps (performance of PC is not the problem).
> >> I asked the mesa and cinnamon team for help in the past, but also with
> >> no luck.
> >> It never worked, till on 2025-03-29 when I installed 6.12.19 for the
> >> first time and it worked!
> >>
> >> But it only worked with 6.12.19, 6.12.20 and 6.12.21
> >> When I updated to 6.12.25, it was back to unplayable.
> >
> > Can you bisect to see what fixed it in 6.12.19 or what broke it in
> > 6.12.25?  For example if it was working in 6.12.21 and not working in
> > 6.12.25, you can bisect between 6.12.21 and .25.
> >
> > Alex
> >
> >>
> >> For testing I installed 6.14.4 with the same result. It doesn't work.
> >>
> >> I also compared file /proc/config.gz of both kernels (6.12.21 <>
> >> 6.14.4), but can't seem to see drastic changes to the graphical part.
> >>
> >> I presume it has something to do with amdgpu.
> >>
> >> If you need more information, I would be happy to help.
> >>
> >> Kind regards,
> >> Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
  2025-06-16 13:29     ` Alex Deucher
@ 2025-06-17  9:42       ` machion
  2025-10-10  9:06         ` machion
  0 siblings, 1 reply; 6+ messages in thread
From: machion @ 2025-06-17  9:42 UTC (permalink / raw)
  To: Alex Deucher
  Cc: stable, regressions, amd-gfx, alexander.deucher, christian.koenig

I feared that.

It seems a crazy problem, when many people are affected and in opposite 
ways.
This ReBAR/UEFI thing is also new to me, but I don't have this on my 
system either (using BIOS/MBR).

I hope, it can be fixed for all scenarios.

Marion

Am 2025-06-16 15:29, schrieb Alex Deucher:
> On Fri, Jun 13, 2025 at 3:38 PM <machion@disroot.org> wrote:
>> 
>> Hi,
>> sorry for the delay.
>> Besides less time, I had to make myself familiar with bisecting and
>> again kernel compiling. Last time I compiled the kernel myself was
>> around 2010 I think.
>> 
>> Anyway it seems I found the bad commit. The result after bisecting 10
>> commits is:
>> 
>> a53d959fe660341788cb8dbc3ac3330d90a09ecf is the first bad commit
>> commit a53d959fe660341788cb8dbc3ac3330d90a09ecf
>> Author: Christian König <christian.koenig@amd.com>
>> Date:   Thu Mar 20 14:46:18 2025 +0100
>> 
>>      drm/amdgpu: immediately use GTT for new allocations
>> 
>>      commit a755906fb2b8370c43e91ba437ae1b3e228e8b02 upstream.
>> 
>>      Only use GTT as a fallback if we already have a backing store. 
>> This
>>      prevents evictions when an application constantly allocates and
>> frees new
>>      memory.
>> 
>>      Partially fixes
>>      
>> https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2833985.
>> 
>>      Signed-off-by: Christian König <christian.koenig@amd.com>
>>      Fixes: 216c1282dde3 ("drm/amdgpu: use GTT only as fallback for
>> VRAM|GTT")
>>      Acked-by: Alex Deucher <alexander.deucher@amd.com>
>>      Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
>>      Cc: stable@vger.kernel.org
>>      Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>> 
>>   drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 ++--
>>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> Unfortunately reverting that commit will reintroduce a similar
> performance issue for lots of other uses.  See:
> https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2827990
> for a description of the fundemental problem.
> 
> Alex
> 
>> 
>> Marion
>> 
>> 
>> Am 2025-05-08 15:18, schrieb Alex Deucher:
>> > On Thu, May 8, 2025 at 9:13 AM <machion@disroot.org> wrote:
>> >>
>> >> Hello kernel/driver developers,
>> >>
>> >> I hope, with my information it's possible to find a bug/problem in the
>> >> kernel. Otherwise I am sorry, that I disturbed you.
>> >> I only use LTS kernels, but I can narrow it down to a hand full of
>> >> them,
>> >> where it works.
>> >>
>> >> The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD
>> >> 7790/8GB
>> >> RAM
>> >> I already asked the Manjaro community, but with no luck.
>> >>
>> >> The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021),
>> >> uses vulkan
>> >>
>> >> ---
>> >>
>> >> I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15,
>> >> 6.1 and 6.6 and and the game was always unplayable, because the frames
>> >> where around 1fps (performance of PC is not the problem).
>> >> I asked the mesa and cinnamon team for help in the past, but also with
>> >> no luck.
>> >> It never worked, till on 2025-03-29 when I installed 6.12.19 for the
>> >> first time and it worked!
>> >>
>> >> But it only worked with 6.12.19, 6.12.20 and 6.12.21
>> >> When I updated to 6.12.25, it was back to unplayable.
>> >
>> > Can you bisect to see what fixed it in 6.12.19 or what broke it in
>> > 6.12.25?  For example if it was working in 6.12.21 and not working in
>> > 6.12.25, you can bisect between 6.12.21 and .25.
>> >
>> > Alex
>> >
>> >>
>> >> For testing I installed 6.14.4 with the same result. It doesn't work.
>> >>
>> >> I also compared file /proc/config.gz of both kernels (6.12.21 <>
>> >> 6.14.4), but can't seem to see drastic changes to the graphical part.
>> >>
>> >> I presume it has something to do with amdgpu.
>> >>
>> >> If you need more information, I would be happy to help.
>> >>
>> >> Kind regards,
>> >> Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem
  2025-06-17  9:42       ` machion
@ 2025-10-10  9:06         ` machion
  0 siblings, 0 replies; 6+ messages in thread
From: machion @ 2025-10-10  9:06 UTC (permalink / raw)
  To: Alex Deucher
  Cc: stable, regressions, amd-gfx, alexander.deucher, christian.koenig

Hi again,

I have a question:

Would it be possible to make the "memory management/buffering" methodes 
switchable through a conf file (obviously amdgpu.conf)?
So that I don't have to keep an old kernel version installed just to 
play my game.

Marion


Am 2025-06-17 11:42, schrieb machion@disroot.org:
> I feared that.
> 
> It seems a crazy problem, when many people are affected and in opposite 
> ways.
> This ReBAR/UEFI thing is also new to me, but I don't have this on my 
> system either (using BIOS/MBR).
> 
> I hope, it can be fixed for all scenarios.
> 
> Marion
> 
> Am 2025-06-16 15:29, schrieb Alex Deucher:
>> On Fri, Jun 13, 2025 at 3:38 PM <machion@disroot.org> wrote:
>>> 
>>> Hi,
>>> sorry for the delay.
>>> Besides less time, I had to make myself familiar with bisecting and
>>> again kernel compiling. Last time I compiled the kernel myself was
>>> around 2010 I think.
>>> 
>>> Anyway it seems I found the bad commit. The result after bisecting 10
>>> commits is:
>>> 
>>> a53d959fe660341788cb8dbc3ac3330d90a09ecf is the first bad commit
>>> commit a53d959fe660341788cb8dbc3ac3330d90a09ecf
>>> Author: Christian König <christian.koenig@amd.com>
>>> Date:   Thu Mar 20 14:46:18 2025 +0100
>>> 
>>>      drm/amdgpu: immediately use GTT for new allocations
>>> 
>>>      commit a755906fb2b8370c43e91ba437ae1b3e228e8b02 upstream.
>>> 
>>>      Only use GTT as a fallback if we already have a backing store. 
>>> This
>>>      prevents evictions when an application constantly allocates and
>>> frees new
>>>      memory.
>>> 
>>>      Partially fixes
>>>      
>>> https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2833985.
>>> 
>>>      Signed-off-by: Christian König <christian.koenig@amd.com>
>>>      Fixes: 216c1282dde3 ("drm/amdgpu: use GTT only as fallback for
>>> VRAM|GTT")
>>>      Acked-by: Alex Deucher <alexander.deucher@amd.com>
>>>      Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
>>>      Cc: stable@vger.kernel.org
>>>      Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>>> 
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>> 
>> Unfortunately reverting that commit will reintroduce a similar
>> performance issue for lots of other uses.  See:
>> https://gitlab.freedesktop.org/drm/amd/-/issues/3844#note_2827990
>> for a description of the fundemental problem.
>> 
>> Alex
>> 
>>> 
>>> Marion
>>> 
>>> 
>>> Am 2025-05-08 15:18, schrieb Alex Deucher:
>>> > On Thu, May 8, 2025 at 9:13 AM <machion@disroot.org> wrote:
>>> >>
>>> >> Hello kernel/driver developers,
>>> >>
>>> >> I hope, with my information it's possible to find a bug/problem in the
>>> >> kernel. Otherwise I am sorry, that I disturbed you.
>>> >> I only use LTS kernels, but I can narrow it down to a hand full of
>>> >> them,
>>> >> where it works.
>>> >>
>>> >> The PC: Manjaro Stable/Cinnamon/X11/AMD Ryzen 5 2600/Radeon HD
>>> >> 7790/8GB
>>> >> RAM
>>> >> I already asked the Manjaro community, but with no luck.
>>> >>
>>> >> The game: Hellpoint (GOG Linux latest version, Unity3D-Engine v2021),
>>> >> uses vulkan
>>> >>
>>> >> ---
>>> >>
>>> >> I came a long road of kernels. I had many versions of 5.4, 5.10, 5.15,
>>> >> 6.1 and 6.6 and and the game was always unplayable, because the frames
>>> >> where around 1fps (performance of PC is not the problem).
>>> >> I asked the mesa and cinnamon team for help in the past, but also with
>>> >> no luck.
>>> >> It never worked, till on 2025-03-29 when I installed 6.12.19 for the
>>> >> first time and it worked!
>>> >>
>>> >> But it only worked with 6.12.19, 6.12.20 and 6.12.21
>>> >> When I updated to 6.12.25, it was back to unplayable.
>>> >
>>> > Can you bisect to see what fixed it in 6.12.19 or what broke it in
>>> > 6.12.25?  For example if it was working in 6.12.21 and not working in
>>> > 6.12.25, you can bisect between 6.12.21 and .25.
>>> >
>>> > Alex
>>> >
>>> >>
>>> >> For testing I installed 6.14.4 with the same result. It doesn't work.
>>> >>
>>> >> I also compared file /proc/config.gz of both kernels (6.12.21 <>
>>> >> 6.14.4), but can't seem to see drastic changes to the graphical part.
>>> >>
>>> >> I presume it has something to do with amdgpu.
>>> >>
>>> >> If you need more information, I would be happy to help.
>>> >>
>>> >> Kind regards,
>>> >> Marion

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2025-10-10  9:06 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-05-07 13:10 Unplayable framerates in game but specific kernel versions work, maybe amdgpu problem machion
2025-05-08 13:18 ` Alex Deucher
2025-06-13 19:38   ` machion
2025-06-16 13:29     ` Alex Deucher
2025-06-17  9:42       ` machion
2025-10-10  9:06         ` machion

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox