On 2018年09月11日 11:37, zhoucm1 wrote:
On 2018年09月11日 11:32, Deng, Emily wrote:
-Original Message-
From: amd-gfx On Behalf Of
zhoucm1
Sent: Tuesday, September 11, 2018 11:28 AM
To: Deng, Emily ; Zhou, David(ChunMing)
; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu:
The bulk moving mechanism still has bug on some corner cases. So disable it by
default till it is fixed. We can use the module parameter to enable it for
debugging.
Signed-off-by: Huang Rui
Cc: Christian K??nig
Cc: Tom StDenis
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
On 2018年09月11日 11:32, Deng, Emily wrote:
-Original Message-
From: amd-gfx On Behalf Of
zhoucm1
Sent: Tuesday, September 11, 2018 11:28 AM
To: Deng, Emily ; Zhou, David(ChunMing)
; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
On 2018年09月11日
>-Original Message-
>From: amd-gfx On Behalf Of
>zhoucm1
>Sent: Tuesday, September 11, 2018 11:28 AM
>To: Deng, Emily ; Zhou, David(ChunMing)
>; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
>
>
>
>On 2018年09月11日 11:23, Deng, Emily wrote:
>>>
On 2018年09月11日 11:23, Deng, Emily wrote:
-Original Message-
From: Zhou, David(ChunMing)
Sent: Tuesday, September 11, 2018 11:03 AM
To: Deng, Emily ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
On 2018年09月11日 10:51, Emily Deng wrote:
It
>-Original Message-
>From: Zhou, David(ChunMing)
>Sent: Tuesday, September 11, 2018 11:03 AM
>To: Deng, Emily ; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
>
>
>
>On 2018年09月11日 10:51, Emily Deng wrote:
>> It will ramdomly have the dead lock
On 2018年09月11日 10:51, Emily Deng wrote:
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4.
On 09/11/2018 10:51 AM, Emily Deng wrote:
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4.
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4. amdgpu_device_recover_vram_from_shadow is waiting for the bo's
Apart from Felix comments,
Looks good for me, patch 2 ~ 8 are
Reviewed-by: Junwei Zhang
Patch 9 ~ 11 are
Acked-by: Junwei Zhang
On 09/10/2018 02:03 AM, Christian König wrote:
Hi everyone,
Especially on Vega and Raven VM handling is rather inefficient while creating
PTEs because we
On 09/10/2018 02:03 AM, Christian König wrote:
Both a leaf as well as dfs iterator to walk over all the PDs/PTs.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 221 +
1 file changed, 221 insertions(+)
diff --git
Patches 2, 3, 5, 6, 8, 9, 11 are Reviewed-by: Felix Kuehling
I replied with comments to 1, 4, 7, 10.
On another thread, some of the machine learning guys found that the main
overhead of our memory allocator is clearing of BOs. I'm thinking about
a way to avoid that, but your patch 1 interferes
>-Original Message-
>From: amd-gfx On Behalf Of Deng,
>Emily
>Sent: Monday, September 10, 2018 6:33 PM
>To: Koenig, Christian ; amd-
>g...@lists.freedesktop.org
>Subject: RE: [PATCH v2] drm/amdgpu: Fix the dead lock issue.
>
>>-Original Message-
>>From: Koenig, Christian
>>Sent:
On 09/10/2018 05:33 PM, Christian König wrote:
Am 10.09.2018 um 04:44 schrieb Zhang, Jerry (Junwei):
On 09/10/2018 02:04 AM, Christian König wrote:
Make a VM mapping which is as unaligned as possible.
Is it going to test unaligned address between BO allocation and BO mapping
and skip huge
On 2018-09-09 02:03 PM, Christian König wrote:
> Both a leaf as well as dfs iterator to walk over all the PDs/PTs.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 221
> +
> 1 file changed, 221 insertions(+)
>
> diff --git
On 2018-09-09 02:03 PM, Christian König wrote:
> The fragment size controls only the L1 on Vega/Raven and we now don't
> have any extra overhead any more because of larger fragments.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 ++-
> 1 file changed,
On 2018-09-09 02:03 PM, Christian König wrote:
> Less code and easier to maintain.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 28 ++--
> 1 file changed, 6 insertions(+), 22 deletions(-)
>
> diff --git
This looks good. But it complicates something I've been looking at:
Remembering which process drm_mm_nodes last belonged to, so that they
don't need to be cleared next time they are allocated by the same
process. Having most nodes the same size (vram_page_split pages) would
make this very easy and
Attached patch fixes SDMA TO after GPU reset, it's a regression caused
by cbd5285 drm/amdgpu: move setting the GART addr into TTM.
But to me it looks safer just to revert the original patch all together
since we never can predict for sure if VM flush will take place and so
it's safer to just
Hi Christian,
Ok, then can I get a reviewed-by for this change?
I will drop patch 2 for now, as Philip said offline that he will still need to
use that kfifo. I will follow up with Philip anyway.
I will re-work on the translation retry patches, after this one.
Probably also follow up the idea
From: Gert Wollny
On evergreen depth-stencil textures are allocated as two objects, and
when using the eg_surface_init_1d_miptrees code path the size evaluation
uses the generalized surf_minify function. Here when allocating the
depth texture the alignment takes the depth bpe value into account,
Signed-off-by: James Zhu
Add error message when register failed to reach expected value, It will
help discover potential issue.
---
drivers/gpu/drm/amd/amdgpu/soc15_common.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
Signed-off-by: James Zhu
When VCN PG state is unchanged, it is unnecessary to reset
power gate state again.
---
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
Signed-off-by: James Zhu
Add error message when register failed to reach expected value, It will
help discover potential issue.
---
drivers/gpu/drm/amd/amdgpu/soc15_common.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
pirmd., 2018. g. 10. sept., plkst. 12:04 — lietotājs Michel Dänzer
() rakstīja:
>
> On 2018-09-08 7:24 p.m., Dāvis Mosāns wrote:
> > Hello,
> >
> > With Radeon RX Vega 64 when launching LibreOffice (6.1.0.3) on Arch
> > Linux with 4.18.5 kernel then display freezes.
> > dmesg contains:
> >
> >
I'm not sure why you need to distinguish current and future state when
dealing with page faults. When you get a page fault, you know that the
GPU is trying to access memory right now, in the present. So you're
always working with the current state. When the CPU page table changes,
you get an MMU
Hi Jérôme,
I love your patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.19-rc3 next-20180910]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux
Hi Jérôme,
I love your patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.19-rc3 next-20180910]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux
Hi Jérôme,
I love your patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.19-rc3 next-20180910]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux
Hi Jérôme,
I love your patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.19-rc3 next-20180910]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux
Am 10.09.2018 um 19:48 schrieb shaoyunl:
Fix : 5c777a5 'Adjust GART and AGP location with xgmi offset'
Change-Id: I2d78024fbe44a37f46a35d34c1e64dbd3937fdf1
Signed-off-by: shaoyunl
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 5 +
1 file changed, 5
Fix : 5c777a5 'Adjust GART and AGP location with xgmi offset'
Change-Id: I2d78024fbe44a37f46a35d34c1e64dbd3937fdf1
Signed-off-by: shaoyunl
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
Here a commit message is missing.
Especially something like Fix: so
that Alex can squash it into the other patch during upstreaming.
Apart from that the patch looks good to me,
Christian.
Am 10.09.2018 um 18:06 schrieb shaoyunl:
Change-Id: I2d78024fbe44a37f46a35d34c1e64dbd3937fdf1
Change-Id: I2d78024fbe44a37f46a35d34c1e64dbd3937fdf1
Signed-off-by: shaoyunl
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index cf97c1c..ae44671 100644
---
Hi Aaron,
thanks for the patch. Moving to the amd-gfx mailing list, where
xf86-video-amdgpu patches are reviewed. Comments inline below.
On 2018-09-10 1:14 p.m., Aaron Liu wrote:
> Add gamma_lut/degamma_lut/ctm checking before pushing
> staged color management properties on the CRTC.
> If
Sorry , miss this comments and already submitted the serial of the patch .
It's a good catch . I will send out another review to initialize
fb_start/fb_end with vram_start/vram_end for older chips
Regards
Shaoyun.liu
-Original Message-
From: Christian König
Reviewed-by: Andrey Grodzovsky
Andrey
On 09/10/2018 09:54 AM, Christian König wrote:
Slowly leaking memory one page at a time :)
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 23 +++
1 file changed, 15 insertions(+), 8 deletions(-)
diff
Slowly leaking memory one page at a time :)
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 23 +++
1 file changed, 15 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
Am 10.09.2018 um 15:05 schrieb Tom St Denis:
On 2018-09-10 9:04 a.m., Christian König wrote:
Hi Tom,
I'm talking about adding new printks to figure out what the heck is
going wrong here.
Thanks,
Christian.
Hi Christian,
Sure, if you want to send me a simple patch that adds more printk
On 2018-09-10 9:04 a.m., Christian König wrote:
Hi Tom,
I'm talking about adding new printks to figure out what the heck is
going wrong here.
Thanks,
Christian.
Hi Christian,
Sure, if you want to send me a simple patch that adds more printk I'll
gladly give it a try (doubly so since my
Hi Tom,
I'm talking about adding new printks to figure out what the heck is
going wrong here.
Thanks,
Christian.
Am 10.09.2018 um 14:59 schrieb Tom St Denis:
Hi Christian,
Are you adding new traces or turning on existing ones? Would you like
me to try them out in my setup?
Tom
On
Hi Christian,
Are you adding new traces or turning on existing ones? Would you like
me to try them out in my setup?
Tom
On 2018-09-10 8:49 a.m., Christian König wrote:
Am 10.09.2018 um 14:05 schrieb Huang Rui:
On Mon, Sep 10, 2018 at 05:25:48PM +0800, Koenig, Christian wrote:
Am
Am 10.09.2018 um 14:05 schrieb Huang Rui:
On Mon, Sep 10, 2018 at 05:25:48PM +0800, Koenig, Christian wrote:
Am 10.09.2018 um 11:23 schrieb Huang Rui:
On Mon, Sep 10, 2018 at 11:00:04AM +0200, Christian König wrote:
Hi Ray,
well those patches doesn't make sense, the pointer is only local to
On Mon, Sep 10, 2018 at 05:25:48PM +0800, Koenig, Christian wrote:
> Am 10.09.2018 um 11:23 schrieb Huang Rui:
> > On Mon, Sep 10, 2018 at 11:00:04AM +0200, Christian König wrote:
> >> Hi Ray,
> >>
> >> well those patches doesn't make sense, the pointer is only local to
> >> the function.
> >
On Mon, Sep 10, 2018 at 05:25:48PM +0800, Koenig, Christian wrote:
> Am 10.09.2018 um 11:23 schrieb Huang Rui:
> > On Mon, Sep 10, 2018 at 11:00:04AM +0200, Christian König wrote:
> >> Hi Ray,
> >>
> >> well those patches doesn't make sense, the pointer is only local to
> >> the function.
> >
Am 10.09.2018 um 11:55 schrieb Deng, Emily:
-Original Message-
From: amd-gfx On Behalf Of
Christian König
Sent: Monday, September 10, 2018 5:49 PM
To: Deng, Emily ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH v2] drm/amdgpu: Fix the dead lock issue.
Am 10.09.2018 um 11:47 schrieb
>-Original Message-
>From: amd-gfx On Behalf Of
>Christian König
>Sent: Monday, September 10, 2018 5:49 PM
>To: Deng, Emily ; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH v2] drm/amdgpu: Fix the dead lock issue.
>
>Am 10.09.2018 um 11:47 schrieb Deng, Emily:
>>> -Original
Am 10.09.2018 um 11:47 schrieb Deng, Emily:
-Original Message-
From: Christian König
Sent: Monday, September 10, 2018 5:41 PM
To: Deng, Emily ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH v2] drm/amdgpu: Fix the dead lock issue.
Am 10.09.2018 um 11:34 schrieb Emily Deng:
It will
>-Original Message-
>From: Christian König
>Sent: Monday, September 10, 2018 5:41 PM
>To: Deng, Emily ; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH v2] drm/amdgpu: Fix the dead lock issue.
>
>Am 10.09.2018 um 11:34 schrieb Emily Deng:
>> It will ramdomly have the dead lock issue
On Mon, Sep 10, 2018 at 02:13:09PM +0800, Emily Deng wrote:
> since we use PSP to program IH regs now
>
> Signed-off-by: Monk Liu
Reviewed-by: Huang Rui
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git
Hi Felix & Oak,
over the weekend I had the idea that we could just use the shadow BOs to
have the current state in a page fault. They are GTT BOs and CPU
accessible anyway.
Regards,
Christian.
Am 08.09.2018 um 09:34 schrieb Christian König:
Hi Felix,
But why do you want to update page
Am 10.09.2018 um 11:34 schrieb Emily Deng:
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4.
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4. amdgpu_device_recover_vram_from_shadow is waiting for the bo's
Am 10.09.2018 um 04:44 schrieb Zhang, Jerry (Junwei):
On 09/10/2018 02:04 AM, Christian König wrote:
Make a VM mapping which is as unaligned as possible.
Is it going to test unaligned address between BO allocation and BO
mapping
and skip huge page mapping?
Yes and no.
Huge page handling
Am 10.09.2018 um 11:23 schrieb Huang Rui:
On Mon, Sep 10, 2018 at 11:00:04AM +0200, Christian König wrote:
Hi Ray,
well those patches doesn't make sense, the pointer is only local to
the function.
You're right.
I narrowed it with gdb dump from ttm_bo_bulk_move_lru_tail+0x2b, the
On Mon, Sep 10, 2018 at 11:00:04AM +0200, Christian König wrote:
> Hi Ray,
>
> well those patches doesn't make sense, the pointer is only local to
> the function.
You're right.
I narrowed it with gdb dump from ttm_bo_bulk_move_lru_tail+0x2b, the
use-after-free should be in below codes:
man =
On 2018-09-08 7:24 p.m., Dāvis Mosāns wrote:
> Hello,
>
> With Radeon RX Vega 64 when launching LibreOffice (6.1.0.3) on Arch
> Linux with 4.18.5 kernel then display freezes.
> dmesg contains:
>
> [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout, signaled
> seq=8322, emitted seq=8325
Hi Ray,
well those patches doesn't make sense, the pointer is only local to the
function.
Regards,
Christian.
Am 10.09.2018 um 10:57 schrieb Huang Rui:
It avoids to be refered again after freed.
Signed-off-by: Huang Rui
Cc: Christian König
Cc: Tom StDenis
---
It avoids to be refered again after freed.
Signed-off-by: Huang Rui
Cc: Christian K??nig
Cc: Tom StDenis
---
drivers/gpu/drm/ttm/ttm_bo.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 138c989..d3ef5f8 100644
---
The tbo pointer will still have value even the amdgpu bo is freed.
It doesn't make sense that it still points a freed memory. It could be refered
mistakenly, so set it as null.
Signed-off-by: Huang Rui
Cc: Christian K??nig
Cc: Tom StDenis
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 1 +
>-Original Message-
>From: Koenig, Christian
>Sent: Monday, September 10, 2018 3:23 PM
>To: Deng, Emily ; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
>
>Am 10.09.2018 um 09:19 schrieb Deng, Emily:
>>> -Original Message-
>>> From:
Am 10.09.2018 um 09:19 schrieb Deng, Emily:
-Original Message-
From: Christian König
Sent: Monday, September 10, 2018 3:06 PM
To: Deng, Emily ; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
Am 10.09.2018 um 06:07 schrieb Emily Deng:
It will
>-Original Message-
>From: Christian König
>Sent: Monday, September 10, 2018 3:06 PM
>To: Deng, Emily ; amd-gfx@lists.freedesktop.org
>Subject: Re: [PATCH] drm/amdgpu: Fix the dead lock issue.
>
>Am 10.09.2018 um 06:07 schrieb Emily Deng:
>> It will ramdomly have the dead lock issue when
Am 10.09.2018 um 08:13 schrieb Emily Deng:
since we use PSP to program IH regs now
Signed-off-by: Monk Liu
Acked-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Am 10.09.2018 um 06:07 schrieb Emily Deng:
It will ramdomly have the dead lock issue when test TDR:
1. amdgpu_device_handle_vram_lost gets the lock shadow_list_lock
2. amdgpu_bo_create locked the bo's resv lock
3. amdgpu_bo_create_shadow is waiting for the shadow_list_lock
4.
Am 10.09.2018 um 02:57 schrieb jgli...@redhat.com:
From: Jérôme Glisse
[This depends on some HMM patchset queued upstream see branch [1]]
This is simple change to switch to use HMM for user ptr buffer object
which conveniently avoid to pin pages. I have more things in the pipe
to make HMM
Am 10.09.2018 um 08:19 schrieb Huang Rui:
On Sun, Sep 09, 2018 at 06:38:13PM +0800, StDenis, Tom wrote:
On 2018-09-08 5:12 a.m., Huang Rui wrote:
On Wed, Sep 05, 2018 at 05:08:26PM +0200, Christian König wrote:
Otherwise we might run into a use after free during bulk move.
Signed-off-by:
On Sun, Sep 09, 2018 at 06:38:13PM +0800, StDenis, Tom wrote:
> On 2018-09-08 5:12 a.m., Huang Rui wrote:
> > On Wed, Sep 05, 2018 at 05:08:26PM +0200, Christian König wrote:
> >> Otherwise we might run into a use after free during bulk move.
> >>
> >> Signed-off-by: Christian König
> >
> > Is
since we use PSP to program IH regs now
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 93476b8..acfc63e
69 matches
Mail list logo