From: Michel Dänzer <hidden> Date: 2018-08-16 10:56:09
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
thanks,
greg k-h
From: Michel Dänzer <hidden> Date: 2018-08-16 18:35:32
On 2018-08-16 05:20 PM, Greg KH wrote:
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel Dänzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
That's a bit unfortunate in this case, as there can be spurious WARN
splats with only the first fix.
quoted
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
All active trees where they apply (and so do the patches they depend on).
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
On Thu, Aug 16, 2018 at 05:36:16PM +0200, Michel D�nzer wrote:
On 2018-08-16 05:20 PM, Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
That's a bit unfortunate in this case, as there can be spurious WARN
splats with only the first fix.
That's fine, no one is going to only apply one of them.
quoted
quoted
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
All active trees where they apply (and so do the patches they depend on).
A hint might be nice here, what trees did you test this on?
thanks,
greg k-h
From: Michel Dänzer <hidden> Date: 2018-08-16 19:15:56
On 2018-08-16 05:54 PM, Greg KH wrote:
On Thu, Aug 16, 2018 at 05:36:16PM +0200, Michel Dänzer wrote:
quoted
On 2018-08-16 05:20 PM, Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel Dänzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
That's a bit unfortunate in this case, as there can be spurious WARN
splats with only the first fix.
That's fine, no one is going to only apply one of them.
quoted
quoted
quoted
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
All active trees where they apply (and so do the patches they depend on).
A hint might be nice here, what trees did you test this on?
I didn't test it on any stable trees, but if the patches apply on a
branch, it means they're needed. :)
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
From: Michel Dänzer <hidden> Date: 2018-09-12 13:43:30
Hi Greg,
On 2018-08-16 5:54 p.m., Greg KH wrote:
On Thu, Aug 16, 2018 at 05:36:16PM +0200, Michel Dänzer wrote:
quoted
On 2018-08-16 05:20 PM, Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel Dänzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
That's a bit unfortunate in this case, as there can be spurious WARN
splats with only the first fix.
That's fine, no one is going to only apply one of them.
quoted
quoted
quoted
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
All active trees where they apply (and so do the patches they depend on).
A hint might be nice here, what trees did you test this on?
Is there a reason why these 4 fixes haven't been backported yet? People
are running into the issues fixed by them.
There seem to be many other fixes under drivers/gpu/drm/ with Cc: stable
which haven't been backported yet.
Thanks,
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
On Wed, Sep 12, 2018 at 10:39:56AM +0200, Michel D�nzer wrote:
Hi Greg,
On 2018-08-16 5:54 p.m., Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 05:36:16PM +0200, Michel D�nzer wrote:
quoted
On 2018-08-16 05:20 PM, Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
Nothing gets squashed, we apply individual patches :)
That's a bit unfortunate in this case, as there can be spurious WARN
splats with only the first fix.
That's fine, no one is going to only apply one of them.
quoted
quoted
quoted
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
What stable tree(s) do you want these applied to?
All active trees where they apply (and so do the patches they depend on).
A hint might be nice here, what trees did you test this on?
Is there a reason why these 4 fixes haven't been backported yet? People
are running into the issues fixed by them.
There seem to be many other fixes under drivers/gpu/drm/ with Cc: stable
which haven't been backported yet.
Ah, our emails crossed. I just tried to apply these, but the original
ones all failed to apply :(
And I normally wait for the drm patches to go "last" for the stable
trees due to all of the merge issues they normally cause, like this :(
greg k-h
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
Those commits did not apply to the stable tree, so can you please send a
series of patches that do apply so that I can queue all of these up
properly?
thanks,
greg k-h
From: Michel Dänzer <hidden> Date: 2018-09-12 14:15:35
On 2018-09-12 10:55 a.m., Greg KH wrote:
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel Dänzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
Those commits did not apply to the stable tree, so can you please send a
series of patches that do apply so that I can queue all of these up
properly?
Just keep the
if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
line in the backport of ddc21af4d0f37f42b33c54cb69b215997fe5b082. The
remaining 3 should apply fine AFAICT.
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
On Wed, Sep 12, 2018 at 11:11:54AM +0200, Michel D�nzer wrote:
On 2018-09-12 10:55 a.m., Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
Those commits did not apply to the stable tree, so can you please send a
series of patches that do apply so that I can queue all of these up
properly?
Just keep the
if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
line in the backport of ddc21af4d0f37f42b33c54cb69b215997fe5b082. The
remaining 3 should apply fine AFAICT.
Can you provide a patch that does that? I don't have the resources to
hand-edit diffs in order to get them to apply. I also, almost always,
when I have tried to do this, get it wrong :)
thanks,
greg k-h
From: Michel Dänzer <hidden> Date: 2018-09-12 14:31:46
On 2018-09-12 11:16 a.m., Greg KH wrote:
On Wed, Sep 12, 2018 at 11:11:54AM +0200, Michel Dänzer wrote:
quoted
On 2018-09-12 10:55 a.m., Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel Dänzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
Those commits did not apply to the stable tree, so can you please send a
series of patches that do apply so that I can queue all of these up
properly?
Just keep the
if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
line in the backport of ddc21af4d0f37f42b33c54cb69b215997fe5b082. The
remaining 3 should apply fine AFAICT.
Can you provide a patch that does that? I don't have the resources to
hand-edit diffs in order to get them to apply. I also, almost always,
when I have tried to do this, get it wrong :)
Sure, attached. Sorry about the trouble, the conflict was due to a
helper function added upstream.
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
On Wed, Sep 12, 2018 at 11:28:01AM +0200, Michel D�nzer wrote:
On 2018-09-12 11:16 a.m., Greg KH wrote:
quoted
On Wed, Sep 12, 2018 at 11:11:54AM +0200, Michel D�nzer wrote:
quoted
On 2018-09-12 10:55 a.m., Greg KH wrote:
quoted
On Thu, Aug 16, 2018 at 09:53:54AM +0200, Michel D�nzer wrote:
quoted
Hi stable kernel maintainers,
please squash these amdgpu fixes together and backport them to all
applicable stable branches:
15e6b76880e65be24250e30986084b5569b7a06f "drm/amdgpu: Warn and update
pin_size values when
destroying a pinned BO"
456607d816d89a442a3d5ec98b02c8bc950b5228 "drm/amdgpu: Don't warn on
destroying a pinned BO"
(These depend on commits a5ccfe5c20740f2fbf00291490cdf8d2373ec255 and
ddc21af4d0f37f42b33c54cb69b215997fe5b082, which already have Cc: stable)
Those commits did not apply to the stable tree, so can you please send a
series of patches that do apply so that I can queue all of these up
properly?
Just keep the
if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
line in the backport of ddc21af4d0f37f42b33c54cb69b215997fe5b082. The
remaining 3 should apply fine AFAICT.
Can you provide a patch that does that? I don't have the resources to
hand-edit diffs in order to get them to apply. I also, almost always,
when I have tried to do this, get it wrong :)
Sure, attached. Sorry about the trouble, the conflict was due to a
helper function added upstream.
Thanks for the backport, all are now queued up.
greg k-h