On 2018年07月25日 16:20, Takashi Iwai wrote:
On Wed, 25 Jul 2018 10:02:34 +0200,
jimqu wrote:
On 2018年07月25日 13:46, Takashi Iwai wrote:
On Wed, 25 Jul 2018 07:38:37 +0200,
Qu, Jim wrote:
Jim: Just like Alex said, we want driver can get eld info when hotplug in new
device. amdgpu driver is a
On Thu, Jul 26, 2018 at 08:12:05PM +0800, Koenig, Christian wrote:
> Am 26.07.2018 um 13:59 schrieb Bas Nieuwenhuizen:
> > On Thu, Jul 26, 2018 at 1:52 PM, Huang Rui wrote:
> >> Hi,
> >>
> >> As suggested by Christian, we should move non-x86 definitions into one
> >> common
> >> header, and it
On 07/26/2018 07:07 PM, Christian König wrote:
Am 26.07.2018 um 12:02 schrieb Junwei Zhang:
pass the evict flag instead of hard code
Signed-off-by: Junwei Zhang
I can't come up with a reason why we shouldn't do this, but please test
suspend/resume with this change just to be extra sure.
On 2018-07-25 09:54 AM, Alex Deucher wrote:
> On Wed, Jul 25, 2018 at 9:45 AM, Harry Wentland
> wrote:
>> This structure needs to align with structure in atomfirmware table.
>> Update it.
>>
>> Signed-off-by: Harry Wentland
>
> Acked-by: Alex Deucher
>
Looks like patchwork messed up the
On Thu, Jul 26, 2018 at 11:39 AM Huang Rui wrote:
>
> On Thu, Jul 26, 2018 at 05:28:53PM +0800, Christian König wrote:
> > Am 26.07.2018 um 11:14 schrieb Huang Rui:
> > > From: Bas Nieuwenhuizen
> > >
> > > Every set_pages_array_wb call resulted in cross-core
> > > interrupts and TLB flushes.
Looks good
Reviewed-by: Dmytro Laktyushkin
-Original Message-
From: Harry Wentland [mailto:harry.wentl...@amd.com]
Sent: Wednesday, July 25, 2018 09:46
To: amd-gfx@lists.freedesktop.org; Laktyushkin, Dmytro
; Cheng, Tony
Cc: Wentland, Harry
Subject: [PATCH] drm/amd: Add missing
On Wed, Jul 11, 2018 at 3:41 PM Arnd Bergmann wrote:
>
> getrawmonotonic64() and get_monotonic_boottime64() are deprecated
> because of the nonstandard naming.
>
> The replacement functions ktime_get_raw_ns() and ktime_get_boot_ns()
> also simplify the callers.
>
> Reviewed-by: Felix Kuehling .
>
Am 26.07.2018 um 13:59 schrieb Bas Nieuwenhuizen:
On Thu, Jul 26, 2018 at 1:52 PM, Huang Rui wrote:
Hi,
As suggested by Christian, we should move non-x86 definitions into one common
header, and it will make the codes readable. They are based on the improvement
fix of Bas (already rebase Bas's
On Thu, Jul 26, 2018 at 1:52 PM, Huang Rui wrote:
> Hi,
>
> As suggested by Christian, we should move non-x86 definitions into one common
> header, and it will make the codes readable. They are based on the improvement
> fix of Bas (already rebase Bas's patch to drm-next).
>
> Changes from V1 ->
All non-x86 definitions are moved to ttm_set_memory header, so remove it from
ttm_page_alloc.c.
Suggested-by: Christian König
Signed-off-by: Huang Rui
---
drivers/gpu/drm/ttm/ttm_page_alloc.c | 62 +++-
1 file changed, 5 insertions(+), 57 deletions(-)
diff
From: Bas Nieuwenhuizen
Every set_pages_array_wb call resulted in cross-core
interrupts and TLB flushes. Merge more of them for
less overhead.
This reduces the time needed to free a 1.6 GiB GTT WC
buffer as part of Vulkan CTS from ~2 sec to < 0.25 sec.
(Allocation still takes more than 2 sec
All non-x86 definitions are moved to ttm_set_memory header, so remove it from
ttm_page_alloc_dma.c.
Suggested-by: Christian König
Signed-off-by: Huang Rui
---
drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 48 +++-
1 file changed, 4 insertions(+), 44 deletions(-)
diff
Hi,
As suggested by Christian, we should move non-x86 definitions into one common
header, and it will make the codes readable. They are based on the improvement
fix of Bas (already rebase Bas's patch to drm-next).
Changes from V1 -> V2:
- add ttm_ prefix at this header.
- use set_pages_wb
This patch moves all non-x86 abstraction to the ttm_set_memory header.
It is to make function calling more clearly.
(v2): add ttm_ prefix.
Suggested-by: Christian König
Signed-off-by: Huang Rui
---
include/drm/ttm/ttm_set_memory.h | 128 +++
1 file changed,
Note which task is using the entity and only kill it if the last user of
the entity is killed. This should prevent problems when entities are leaked to
child processes.
Signed-off-by: Christian König
---
drivers/gpu/drm/scheduler/gpu_scheduler.c | 6 +-
include/drm/gpu_scheduler.h
Am 26.07.2018 um 12:02 schrieb Junwei Zhang:
pass the evict flag instead of hard code
Signed-off-by: Junwei Zhang
I can't come up with a reason why we shouldn't do this, but please test
suspend/resume with this change just to be extra sure.
Apart from that patch is Reviewed-by: Christian
On Thu, Jul 26, 2018 at 05:43:10PM +0800, Bas Nieuwenhuizen wrote:
> On Thu, Jul 26, 2018 at 11:39 AM Huang Rui wrote:
> >
> > On Thu, Jul 26, 2018 at 05:28:53PM +0800, Christian König wrote:
> > > Am 26.07.2018 um 11:14 schrieb Huang Rui:
> > > > From: Bas Nieuwenhuizen
> > > >
> > > > Every
pass the evict flag instead of hard code
Signed-off-by: Junwei Zhang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 8c4358e..c1ae528
On Thu, Jul 26, 2018 at 05:28:53PM +0800, Christian König wrote:
> Am 26.07.2018 um 11:14 schrieb Huang Rui:
> > From: Bas Nieuwenhuizen
> >
> > Every set_pages_array_wb call resulted in cross-core
> > interrupts and TLB flushes. Merge more of them for
> > less overhead.
> >
> > This reduces the
From: Bas Nieuwenhuizen
Every set_pages_array_wb call resulted in cross-core
interrupts and TLB flushes. Merge more of them for
less overhead.
This reduces the time needed to free a 1.6 GiB GTT WC
buffer as part of Vulkan CTS from ~2 sec to < 0.25 sec.
(Allocation still takes more than 2 sec
Hi,
As suggested by Christian, we should move non-x86 definitions into one common
header, and it will make the codes readable. They are based on the improvement
fix of Bas (already rebase Bas's patch to drm-next).
Thanks,
Ray
Bas Nieuwenhuizen (1):
drm/ttm: Merge hugepage attr changes in
Am 26.07.2018 um 11:14 schrieb Huang Rui:
This patch moves all non-x86 abstraction to the ttm_set_memory header.
It is to make function calling more clearly.
Suggested-by: Christian König
Signed-off-by: Huang Rui
---
include/drm/ttm/ttm_set_memory.h | 127
Am 26.07.2018 um 11:14 schrieb Huang Rui:
From: Bas Nieuwenhuizen
Every set_pages_array_wb call resulted in cross-core
interrupts and TLB flushes. Merge more of them for
less overhead.
This reduces the time needed to free a 1.6 GiB GTT WC
buffer as part of Vulkan CTS from ~2 sec to < 0.25
On Thu, Jul 26, 2018 at 05:18:46PM +0800, Michel Dänzer wrote:
>
> Hi Ray,
>
>
> On 2018-07-26 11:14 AM, Huang Rui wrote:
> > Hi,
> >
> > As suggested by Christian, we should move non-x86 definitions into one
> > common
> > header, and it will make the codes readable. They are based on the
>
Hi Ray,
On 2018-07-26 11:14 AM, Huang Rui wrote:
> Hi,
>
> As suggested by Christian, we should move non-x86 definitions into one common
> header, and it will make the codes readable. They are based on the improvement
> fix of Bas (already rebase Bas's patch to drm-next).
TTM patches need to
This patch moves all non-x86 abstraction to the ttm_set_memory header.
It is to make function calling more clearly.
Suggested-by: Christian König
Signed-off-by: Huang Rui
---
include/drm/ttm/ttm_set_memory.h | 127 +++
1 file changed, 127 insertions(+)
From: Bas Nieuwenhuizen
Every set_pages_array_wb call resulted in cross-core
interrupts and TLB flushes. Merge more of them for
less overhead.
This reduces the time needed to free a 1.6 GiB GTT WC
buffer as part of Vulkan CTS from ~2 sec to < 0.25 sec.
(Allocation still takes more than 2 sec
Hi,
As suggested by Christian, we should move non-x86 definitions into one common
header, and it will make the codes readable. They are based on the improvement
fix of Bas (already rebase Bas's patch to drm-next).
Thanks,
Ray
Bas Nieuwenhuizen (1):
drm/ttm: Merge hugepage attr changes in
On Thu, Jul 26, 2018 at 7:52 AM, Zhang, Jerry (Junwei)
wrote:
> On 07/26/2018 04:29 AM, Bas Nieuwenhuizen wrote:
>>
>> Every set_pages_array_wb call resulted in cross-core
>> interrupts and TLB flushes. Merge more of them for
>> less overhead.
>>
>> This reduces the time needed to free a 1.6 GiB
On Thu, Jul 26, 2018 at 08:37:45AM +0200, Christian König wrote:
> Am 25.07.2018 um 22:29 schrieb Bas Nieuwenhuizen:
> >Every set_pages_array_wb call resulted in cross-core
> >interrupts and TLB flushes. Merge more of them for
> >less overhead.
> >
> >This reduces the time needed to free a 1.6 GiB
Am 26.07.2018 um 08:11 schrieb Huang Rui:
This patch cleans up spaces and algin the text to refine the comment for
amdgpu_ttm.
Signed-off-by: Huang Rui
Yeah, that was also on my TODO list for a while.
Patch is Reviewed-by: Christian König , but
please make sure that the result of "make
Am 25.07.2018 um 22:29 schrieb Bas Nieuwenhuizen:
Every set_pages_array_wb call resulted in cross-core
interrupts and TLB flushes. Merge more of them for
less overhead.
This reduces the time needed to free a 1.6 GiB GTT WC
buffer as part of Vulkan CTS from ~2 sec to < 0.25 sec.
(Allocation
Am 25.07.2018 um 21:43 schrieb James Zhu:
On 2018-07-25 07:05 AM, Christian König wrote:
Patch the IBs for the second UVD instance so that userspace don't need
to care about the instance they submit to.
v2: use direct IB patching
Signed-off-by: Christian König
Reviewed-by: Alex Deucher
---
This patch cleans up spaces and algin the text to refine the comment for
amdgpu_ttm.
Signed-off-by: Huang Rui
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 107 +++-
1 file changed, 51 insertions(+), 56 deletions(-)
diff --git
34 matches
Mail list logo