No, that is incorrect as well :)
The mmap_sem isn't necessary taken during page table updates.
What you could do is replace get_user_pages() directly with HMM. If I'm not
completely mistaken that should work as expected.
Christian.
Am 27.09.2018 22:18 schrieb "Yang, Philip" :
I was trying to
This will break old UMDs that didn't set the flags correctly. Instead,
UMDs should stop using amdgpu_bo_va_op if they want to set the flags.
Marek
On Thu, Sep 27, 2018 at 3:05 PM Andrey Grodzovsky
wrote:
>
> Signed-off-by: Andrey Grodzovsky
> ---
> amdgpu/amdgpu_bo.c | 5 +++--
> 1 file
> -Original Message-
> From: Qu, Jim
> Sent: 2018年9月28日 11:10
> To: Cui, Flora ; Quan, Evan ;
> amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: 答复: [PATCH 1/3] drm/amdgpu: added AMD GPU instance counting
>
> See comments in line.
>
> Thanks
> JimQu
>
>
Thanks. Will get it updated.
Regards,
Evan
> -Original Message-
> From: Cui, Flora
> Sent: 2018年9月28日 10:47
> To: Quan, Evan ; amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: Re: [PATCH 1/3] drm/amdgpu: added AMD GPU instance counting
>
>
>
> On 2018年09月28日 10:03,
See comments in line.
Thanks
JimQu
发件人: amd-gfx 代表 Flora Cui
发送时间: 2018年9月28日 10:46
收件人: Quan, Evan; amd-gfx@lists.freedesktop.org
抄送: Deucher, Alexander
主题: Re: [PATCH 1/3] drm/amdgpu: added AMD GPU instance counting
On 2018年09月28日 10:03, Evan Quan
On 2018年09月28日 10:03, Evan Quan wrote:
Count all GPU instances from AMD(including iGPUs and
dGPUs) in the system.
Change-Id: If62a0873c64857a3fcdf9785557e24cb3456c12e
Signed-off-by: Evan Quan
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 18 +
Added Vega20 specific implementation for MGPU fan boost
feature.
Change-Id: I2521c5b5019261228302c51aa74fc3d9d249c87c
Signed-off-by: Evan Quan
Reviewed-by: Alex Deucher
---
.../gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c| 15 +++
1 file changed, 15 insertions(+)
diff --git
MGPU fan boost feature is enabled only when two or more dGPUs
in the system.
Change-Id: Ie1baebb04c1dfa05a090ad8a59089f072b0b0f32
Signed-off-by: Evan Quan
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c| 41 +++
Count all GPU instances from AMD(including iGPUs and
dGPUs) in the system.
Change-Id: If62a0873c64857a3fcdf9785557e24cb3456c12e
Signed-off-by: Evan Quan
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 18 +
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 7
Reviewed-by: Evan Quan
> -Original Message-
> From: amd-gfx On Behalf Of Alex
> Deucher
> Sent: 2018年9月27日 21:46
> To: amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: [PATCH] drm/amdgpu/vega20: make power profile output more
> consistent
>
> Make the profile name
Seems like AI and RV requires uncashed memory mapping to be able
to pickup value written to memory by CPU.
Enable the test for AI and RV.
Signed-off-by: Andrey Grodzovsky
---
tests/amdgpu/deadlock_tests.c | 13 -
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git
Signed-off-by: Andrey Grodzovsky
---
amdgpu/amdgpu_bo.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/amdgpu/amdgpu_bo.c b/amdgpu/amdgpu_bo.c
index c0f42e8..1892345 100644
--- a/amdgpu/amdgpu_bo.c
+++ b/amdgpu/amdgpu_bo.c
@@ -736,7 +736,7 @@ drm_public int
Signed-off-by: Andrey Grodzovsky
---
tests/amdgpu/amdgpu_test.h | 24 +---
1 file changed, 17 insertions(+), 7 deletions(-)
diff --git a/tests/amdgpu/amdgpu_test.h b/tests/amdgpu/amdgpu_test.h
index 0609a74..8024507 100644
--- a/tests/amdgpu/amdgpu_test.h
+++
Hi Dave,
More new features and fixes for 4.20:
- Add dynamic powergating support for VCN on picasso
- Scheduler cleanup
- Vega20 support for KFD
- DC cleanups and bug fixes
The following changes since commit bf78296ab1cb215d0609ac6cff4e43e941e51265:
BackMerge v4.19-rc5 into drm-next
Clang warns when one enumerated type is implicitly converted to another.
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c:315:19: warning:
implicit conversion from enumeration type 'enum
aux_channel_operation_result' to different enumeration type 'enum
aux_transaction_reply'
On Thu, Sep 27, 2018 at 11:08 AM Nathan Chancellor
wrote:
>
> On Thu, Sep 27, 2018 at 11:06:33AM -0700, Nathan Chancellor wrote:
> > Clang warns when one enumerated type is implicitly converted to another.
> >
> > drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c:315:19: warning:
> >
On Thu, Sep 27, 2018 at 11:06:33AM -0700, Nathan Chancellor wrote:
> Clang warns when one enumerated type is implicitly converted to another.
>
> drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c:315:19: warning:
> implicit conversion from enumeration type 'enum
>
Hi Dave,
Just a few fixes for 4.19:
- Couple of suspend/resume fixes
- Fix EDID emulation with DC
The following changes since commit 4fcb7f8be829d21bcbb24eef2204fb6b34ed1f80:
Merge branch 'drm-fixes-4.19' of git://people.freedesktop.org/~agd5f/linux
into drm-fixes (2018-09-21 09:52:27 +1000)
Yeah, I've read that as well.
My best guess is that we just need to add a call to hmm_vma_range_done()
after taking the lock and also replace get_user_pages() with
hmm_vma_get_pfns().
But I'm still not 100% sure how all of that is supposed to work together.
Regards,
Christian.
Am
I think the answer is here:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/vm/hmm.rst#n216
Regards,
Felix
From: Koenig, Christian
Sent: Thursday, September 27, 2018 10:30 AM
To: Kuehling, Felix
Cc: j.gli...@gmail.com; Yang, Philip ;
Applied. thanks!
Alex
On Sat, Sep 22, 2018 at 2:29 AM Nathan Chancellor
wrote:
>
> Clang generates warnings when one enumerated type is implicitly
> converted to another.
>
> drivers/gpu/drm/amd/amdgpu/../powerplay/hwmgr/ppatomfwctrl.c:532:57:
> warning: implicit conversion from enumeration
At least with get_user_pages() that is perfectly possible.
For HMM it could be that this is prevented somehow.
Christian.
Am 27.09.2018 16:27 schrieb "Kuehling, Felix" :
> In this case you can end up accessing pages which are invalidated while
> get_user_pages is in process.
What’s the
> In this case you can end up accessing pages which are invalidated while
> get_user_pages is in process.
What's the sequence of events you have in mind? Something like this?
* Page table is updated and triggers MMU notifier
* amdgpu MMU notifier runs and waits for pending CS to finish
Sorry, Please ignore this patch.
Best Regards
Rex
From: amd-gfx on behalf of Rex Zhu
Sent: Thursday, September 27, 2018 9:55 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhu, Rex
Subject: [PATCH v2] drm/amd/pp: Export load_firmware interface
Export this
Yeah I understand that, but again that won't work.
In this case you can end up accessing pages which are invalidated while
get_user_pages is in process.
Christian.
Am 27.09.2018 15:41 schrieb "Kuehling, Felix" :
> I’m not planning to change that. I don’t think there is any need to change it.
>
Export this interface for the AMDGPU_FW_LOAD_SMU type.
gfx/sdma can request smu to load firmware.
Split the smu7/8_start_smu function into two functions
1. start_smu, used for load smu firmware in smu7/8 and
check smu firmware version.
2. request_smu_load_fw, used for load other ip's firmware
Make the profile name line match previous generations more closely.
E.g.,
0 3D_FULL_SCREEN :
vs:
0(3D_FULL_SCREEN )
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
> I’m not planning to change that. I don’t think there is any need to change it.
>
> Yeah, but when HMM doesn't provide both the start and the end hock of the
> invalidation this way won't work any more.
>
> So we need to find a solution for this,
> Christian.
My whole argument is that you don’t
Am 27.09.2018 um 15:18 schrieb Kuehling, Felix:
> The problem is here:
>
> ttm_eu_fence_buffer_objects(>ticket, >validated, p->fence);
> amdgpu_mn_unlock(p->mn);
>
> We need to hold the lock until the fence is added to the reservation
object.
>
> Otherwise somebody could have changed the
> The problem is here:
>
> ttm_eu_fence_buffer_objects(>ticket, >validated, p->fence);
> amdgpu_mn_unlock(p->mn);
>
> We need to hold the lock until the fence is added to the reservation object.
>
> Otherwise somebody could have changed the page tables just in the moment
> between the check of
Reviewed-by: Feifei Xu
Thanks
Regards,
Feifei
-Original Message-
From: amd-gfx On Behalf Of Rex Zhu
Sent: 2018年9月27日 20:49
To: amd-gfx@lists.freedesktop.org
Cc: Zhu, Rex
Subject: [PATCH] drm/amdgpu: Fix vce work queue was not cancelled when suspend
The vce cancel_delayed_work_sync
Am 27.09.2018 um 14:48 schrieb Rex Zhu:
The vce cancel_delayed_work_sync never be called.
driver call the function in error path.
This caused the A+A suspend hang when runtime pm enebled.
As we will visit the smu in the idle queue. this will cause
smu hang because the dgpu has been suspend, and
The vce cancel_delayed_work_sync never be called.
driver call the function in error path.
This caused the A+A suspend hang when runtime pm enebled.
As we will visit the smu in the idle queue. this will cause
smu hang because the dgpu has been suspend, and the dgpu also
will be waked up. As the
Am 27.09.2018 um 12:00 schrieb Huang Rui:
On Wed, Sep 26, 2018 at 03:53:19PM +0200, Christian König wrote:
We ignored the return code here.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 3 +++
1 file changed, 3 insertions(+)
diff --git
Am 27.09.2018 um 13:08 schrieb Kuehling, Felix:
> We double check that there wasn't any page table modification while
we prepared the submission and restart the whole process when there
actually was some update.
>
> The reason why we need to do this is here:
>
>
> We double check that there wasn't any page table modification while we
> prepared the submission and restart the whole process when there actually was
> some update.
>
> The reason why we need to do this is here:
>
>ttm_eu_fence_buffer_objects(>ticket, >validated, p->fence);
>
On Wed, Sep 26, 2018 at 11:25:16AM -0500, Alex Deucher wrote:
> expects argument of type ‘unsigned int’ has type ‘long int’
>
> Fixes: 52e211c1f04 ("drm/amdgpu:Add error message when register failed to
> reach expected value")
> Signed-off-by: Alex Deucher
Series are Reviewed-by: Huang Rui
>
On Wed, Sep 26, 2018 at 03:53:23PM +0200, Christian König wrote:
> Not used any more.
>
> Signed-off-by: Christian König
This patch should be merged to patch 3, that totally remove prescreen_iv
interface.
Thanks,
Ray
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h | 2 --
>
On 2018-09-26 01:42 PM, sunpeng...@amd.com wrote:
> From: Eryk Brol
>
> Also add dram clock to clocks struct, for systems that uses them.
>
> Signed-off-by: Eryk Brol
> Reviewed-by: Jun Lei
> Acked-by: Leo Li
> ---
> drivers/gpu/drm/amd/display/dc/core/dc.c | 13 +
>
On 2018-09-24 06:22 PM, Nathan Chancellor wrote:
> On Mon, Sep 24, 2018 at 03:07:16PM -0700, Nick Desaulniers wrote:
>> On Fri, Sep 21, 2018 at 2:55 PM Nathan Chancellor
>> wrote:
>>>
>>> Clang warns when one enumerated type is implicitly converted to another.
>>>
>>>
On Wed, Sep 26, 2018 at 03:53:19PM +0200, Christian König wrote:
> We ignored the return code here.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
>
Am 26.09.2018 um 20:24 schrieb Jay Cornwall:
On Wed, Sep 26, 2018, at 08:53, Christian König wrote:
This allows us to filter out VM faults in the GMC code.
Signed-off-by: Christian König
The KFD needs to receive notification of unhandled VM faults; when demand
paging is disabled or the
That is correct, but take a look what we do when after calling the
amdgpu_mn_read_lock():
/* No memory allocation is allowed while holding the mn lock */
amdgpu_mn_lock(p->mn);
amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) {
struct amdgpu_bo *bo =
So back to my previous question:
>> But do we really need another lock for this? Wouldn't the
>> re-validation of userptr BOs (currently calling get_user_pages) force
>> synchronization with the ongoing page table invalidation through the
>> mmap_sem or other MM locks?
>
> No and yes. We
> -Original Message-
> From: Zhu, James
> Sent: Wednesday, September 26, 2018 9:03 PM
> To: Huang, Ray ; James Zhu ;
> amd-gfx@lists.freedesktop.org
> Cc: Zhu, James ; Gao, Likun
> Subject: Re: [PATCH v2 1/8] drm/amdgpu:Use register UVD_SCRATCH9 for
> VCN ring/ib test
>
>
>
> On
No, that won't work. We would still run into lock inversion problems.
What we could do with the scheduler is to turn submissions into dummies if we
find that the page tables are now outdated.
But that would be really hacky and I'm not sure if that would really work in
all cases.
Christian.
I had a chat with Jerome yesterday. He pointed out that the new blockable
parameter can be used to infer whether the MMU notifier is being called in a
reclaim operation. So if blockable==true, it should even be safe to take the BO
reservation lock without problems. I think with that we should
47 matches
Mail list logo