Am 11.02.20 um 03:32 schrieb xinhui pan:
As root PD has no parent, we just need move its status to idle.
Suggested-by: Christian König
Signed-off-by: xinhui pan
CC: Christian König
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 34 +-
[AMD Official Use Only - Internal Distribution Only]
For patch 1/2/3/5/6
Reviewed-by: xinhui pan
From: Christian König
Sent: Monday, February 10, 2020 11:09:01 PM
To: Pan, Xinhui ; amd-gfx@lists.freedesktop.org
; dri-de...@lists.freedesktop.org
Subject: Cleanu
On Tue, Feb 11, 2020 at 11:06:53AM +, Pan, Xinhui wrote:
> [AMD Official Use Only - Internal Distribution Only]
Uh might want to fix your email setup.
-Daniel
>
> For patch 1/2/3/5/6
> Reviewed-by: xinhui pan
>
> From: Christian König
> Sent: Monday, Februa
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c| 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c | 4 +-
drivers/gpu/drm/am
So far, amdgpu_cs_submit_raw2 is mainly used for upper layer (Mesa), however,
amdgpu_cs_submit is used for current all unit tests. Our intention is that the
unit tests can actually verify the API which is really used.
Signed-off-by: Huang Rui
---
amdgpu/amdgpu_cs.c | 8 +---
1 file changed,
Use alloca instead of malloc, then we don't need free them at the end of this
function.
Signed-off-by: Huang Rui
---
amdgpu/amdgpu_cs.c | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/amdgpu/amdgpu_cs.c b/amdgpu/amdgpu_cs.c
index 1bd974f..aaa1f7b 100644
--- a/amdgpu/amd
This array won't be used.
Signed-off-by: Huang Rui
---
amdgpu/amdgpu_cs.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/amdgpu/amdgpu_cs.c b/amdgpu/amdgpu_cs.c
index 864a76a..1772dbf 100644
--- a/amdgpu/amdgpu_cs.c
+++ b/amdgpu/amdgpu_cs.c
@@ -221,7 +221,6 @@ static int amdgpu_cs_sub
So far, the amdgpu_cs_submit_raw2 is used for MesaGL, however the amdgpu tests
still use the legacy interface. So we would like to make amdgpu tests verify the
amdgpu_cs_submit_raw2 API.
Thanks,
Ray
Huang Rui (4):
amdgpu: use alloca for dependencies and sem_dependencies
amdgpu: use amdgpu_cs_
This patch is to use generic variables as the input of amdgpu_cs_submit_raw2.
Because amdgpu_cs_submit_one won't handle IOCTL directly.
Signed-off-by: Huang Rui
---
amdgpu/amdgpu_cs.c | 17 +++--
1 file changed, 7 insertions(+), 10 deletions(-)
diff --git a/amdgpu/amdgpu_cs.c b/amdg
Looks good on first glance, but Marek and/or Pierre can probably better
judge than me.
Christian.
Am 11.02.20 um 12:22 schrieb Huang Rui:
So far, the amdgpu_cs_submit_raw2 is used for MesaGL, however the amdgpu tests
still use the legacy interface. So we would like to make amdgpu tests verify
Am 11.02.20 um 12:17 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Actually for amdgpu we should be able to get completely rid of
bo->offset. All we need to do is to change amdgpu_bo
On Tue, Feb 11, 2020 at 12:17:22PM +0100, Nirmoy Das wrote:
> GPU address should belong to driver not in memory management.
> This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
>
Looks good for me. Yes, agree, the mc address should be got from amdgpu_bo.
Acked-by: Huang
Hi Christian
On 2/11/20 12:50 PM, Christian König wrote:
Am 11.02.20 um 12:17 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu
driver.
Actually for amdgpu we should be able to get completely
On 2/11/20 12:57 PM, Huang Rui wrote:
On Tue, Feb 11, 2020 at 12:17:22PM +0100, Nirmoy Das wrote:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Looks good for me. Yes, agree, the mc address should be
Am 11.02.20 um 13:06 schrieb Nirmoy:
Hi Christian
On 2/11/20 12:50 PM, Christian König wrote:
Am 11.02.20 um 12:17 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu
driver.
Actually for amdgp
Am 11.02.20 um 13:11 schrieb Nirmoy:
On 2/11/20 12:57 PM, Huang Rui wrote:
On Tue, Feb 11, 2020 at 12:17:22PM +0100, Nirmoy Das wrote:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu
driver.
Looks good for me.
On 2/11/20 1:17 PM, Christian König wrote:
Am 11.02.20 um 13:06 schrieb Nirmoy:
Hi Christian
On 2/11/20 12:50 PM, Christian König wrote:
Am 11.02.20 um 12:17 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calcul
[SNIP]
+ /*
+* Make NO_EVICT bos immediately available to
+* shrinkers, now that they are queued for
+* destruction.
+*/
+ if (bo->mem.placement & TTM_PL_FLAG_NO_EVICT) {
+ bo->mem.pl
This patch reworks the whole delayed deletion of BOs which aren't idle.
Instead of having two counters for the BO structure we resurrect the BO
when we find that a deleted BO is not idle yet.
This has many advantages, especially that we don't need to
increment/decrement the BOs reference counter
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 22 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 1 +
driv
Am 11.02.20 um 15:06 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 22 +++--
drivers/gpu
Reviewed-by: xinhui pan
> 2020年2月11日 21:43,Christian König 写道:
>
> This patch reworks the whole delayed deletion of BOs which aren't idle.
>
> Instead of having two counters for the BO structure we resurrect the BO
> when we find that a deleted BO is not idle yet.
>
> This has many advantages
On Mon, Feb 10, 2020 at 04:09:06PM +0100, Christian König wrote:
> When non-imported BOs are resurrected for delayed delete we replace
> the dma_resv object to allow for easy reclaiming of the resources.
>
> v2: move that to ttm_bo_individualize_resv
>
> Signed-off-by: Christian König
> ---
> d
[AMD Public Use]
Acked-by: Alex Deucher
From: amd-gfx on behalf of Evan Quan
Sent: Tuesday, February 11, 2020 12:41 AM
To: amd-gfx@lists.freedesktop.org
Cc: Quan, Evan
Subject: [PATCH] drm/amd/powerplay: always refetch the enabled features status
on dpm enab
[AMD Public Use]
Also, libdrm changes should go through a gitlab MR now.
Alex
From: Christian König
Sent: Tuesday, February 11, 2020 6:39 AM
To: Huang, Ray ; amd-gfx@lists.freedesktop.org
Cc: Pelloux-prayer, Pierre-eric ; Olsak,
Marek ; Liu, Aaron ; Tuikov, L
[AMD Public Use]
Hi Alex,
MR = merge request? Should I create account to send merge request in the
freedesktop gitlab?
Thanks,
Ray
From: Deucher, Alexander
Sent: Tuesday, February 11, 2020 10:30 PM
To: Christian König ; Huang, Ray
; amd-gfx@lists.freedesktop.org
Cc: Pelloux-prayer, Pierre-e
On 2/11/20 3:09 PM, Christian König wrote:
Am 11.02.20 um 15:06 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu
driver.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_obj
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 28 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 1 +
driv
Yes, correct.
Alex
From: Huang, Ray
Sent: Tuesday, February 11, 2020 9:35 AM
To: Deucher, Alexander ; Christian König
; amd-gfx@lists.freedesktop.org
Cc: Pelloux-prayer, Pierre-eric ; Olsak,
Marek ; Liu, Aaron ; Tuikov, Luben
; Koenig, Christian
Subject: RE:
> 2020年2月11日 22:14,Daniel Vetter 写道:
>
> On Mon, Feb 10, 2020 at 04:09:06PM +0100, Christian König wrote:
>> When non-imported BOs are resurrected for delayed delete we replace
>> the dma_resv object to allow for easy reclaiming of the resources.
>>
>> v2: move that to ttm_bo_individualize_res
Am 11.02.20 um 15:14 schrieb Daniel Vetter:
On Mon, Feb 10, 2020 at 04:09:06PM +0100, Christian König wrote:
When non-imported BOs are resurrected for delayed delete we replace
the dma_resv object to allow for easy reclaiming of the resources.
v2: move that to ttm_bo_individualize_resv
Signed-
Am 11.02.20 um 15:54 schrieb Nirmoy:
On 2/11/20 3:09 PM, Christian König wrote:
Am 11.02.20 um 15:06 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu
driver.
Signed-off-by: Nirmoy Das
---
Am 11.02.20 um 15:57 schrieb Nirmoy Das:
GPU address should belong to driver not in memory management.
This patch moves ttm bo.offset and gpu_offset calculation to amdgpu driver.
Signed-off-by: Nirmoy Das
Two more suggestions below, with those fixed the patch is Reviewed-by:
Christian König
Am 11.02.20 um 16:02 schrieb Pan, Xinhui:
2020年2月11日 22:14,Daniel Vetter 写道:
On Mon, Feb 10, 2020 at 04:09:06PM +0100, Christian König wrote:
When non-imported BOs are resurrected for delayed delete we replace
the dma_resv object to allow for easy reclaiming of the resources.
v2: move that
On Tue, Feb 11, 2020 at 4:23 PM Christian König
wrote:
>
> Am 11.02.20 um 16:02 schrieb Pan, Xinhui:
> >
> >> 2020年2月11日 22:14,Daniel Vetter 写道:
> >>
> >> On Mon, Feb 10, 2020 at 04:09:06PM +0100, Christian König wrote:
> >>> When non-imported BOs are resurrected for delayed delete we replace
> >
When non-imported BOs are resurrected for delayed delete we replace
the dma_resv object to allow for easy reclaiming of the resources.
v2: move that to ttm_bo_individualize_resv
v3: add a comment to explain what's going on
Signed-off-by: Christian König
Reviewed-by: xinhui pan
---
drivers/gpu/
On Tue, Feb 11, 2020 at 04:43:26PM +0100, Christian König wrote:
> When non-imported BOs are resurrected for delayed delete we replace
> the dma_resv object to allow for easy reclaiming of the resources.
>
> v2: move that to ttm_bo_individualize_resv
> v3: add a comment to explain what's going on
On 2/10/20 4:50 PM, Luben Tuikov wrote:
Hi Lucas,
Thank you for bringing awareness of this issue, publicly.
As soon as this patch showed up back in November of 2019,
I objected to it, privately.
I didn't find this objection in my mail actually
I suggested to instead use a _list_ to store
Unlike DP 1.2 edid corruption test, DP 1.4 requires to calculate
real CRC value of the last edid data block, and write it back.
Current edid CRC calculates routine adds the last CRC byte,
and check if non-zero.
This behavior is not accurate; actually, we need to return
the actual CRC value when co
Ping?
Alex
On Fri, Feb 7, 2020 at 4:17 PM Alex Deucher wrote:
>
> Split into init and register functions to avoid a segfault
> in some configs when the load/unload callbacks are removed.
>
> v2:
> - add back accidently dropped has_aux setting
> - set dev in late_register
>
> v3:
> - fix dp cec o
On Mon, Feb 10, 2020 at 10:38 AM YueHaibing wrote:
>
> drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:
> In function dcn10_post_unlock_program_front_end:
> drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:2623:29:
> warning: variable stream_status set but
On Mon, Feb 10, 2020 at 1:48 PM Rajneesh Bhardwaj
wrote:
>
> From: Alex Deucher
>
> Seems to work reliably on VI+ except for a few so enable runpm barring
> those where baco for runtime power management is not supported.
>
> [rajneesh] Picked https://patchwork.freedesktop.org/patch/335402/ to
> e
Hi Felix,
On Mon, Jan 27, 2020 at 10:34 AM Felix Kuehling wrote:
>
> I've seen hangs on a Raven AM4 system after the Ubuntu upgrade to kernel
> 5.3. I am able to work around it by disabling stutter mode with the
> module parameter amdgpu.ppfeaturemask=0xfffdbfff. If that doesn't help,
> you could
On 2/11/20 10:55 AM, Andrey Grodzovsky wrote:
On 2/10/20 4:50 PM, Luben Tuikov wrote:
Hi Lucas,
Thank you for bringing awareness of this issue, publicly.
As soon as this patch showed up back in November of 2019,
I objected to it, privately.
I didn't find this objection in my mail actually
Looks fine to me as well. Just as Christian said, would have to be approved by
Marek.
Regards,
Luben
On 2020-02-11 9:57 a.m., Deucher, Alexander wrote:
> Yes, correct.
>
> Alex
> -
On 2020-02-11 4:27 p.m., Andrey Grodzovsky wrote:
>
> On 2/11/20 10:55 AM, Andrey Grodzovsky wrote:
>> On 2/10/20 4:50 PM, Luben Tuikov wrote:
>>> Hi Lucas,
>>>
>>> Thank you for bringing awareness of this issue, publicly.
>>>
>>> As soon as this patch showed up back in November of 2019,
>>> I obj
It's used by more than just DCN2.0. Fixes missing symbol when
amdgpu is built without DCN support.
Signed-off-by: Alex Deucher
---
.../gpu/drm/amd/display/dc/core/dc_resource.c| 16
.../drm/amd/display/dc/dcn20/dcn20_resource.c| 16
.../drm/amd/display/
to cover dcn2.x related headers.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 4 ++--
drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c | 2 +-
drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
di
To cover DCN 2.x.
Signed-off-by: Alex Deucher
---
.../drm/amd/display/dc/bios/command_table_helper2.c | 13 +++--
1 file changed, 3 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c
b/drivers/gpu/drm/amd/display/dc/bios/command_tabl
> 2020年2月11日 23:43,Christian König 写道:
>
> When non-imported BOs are resurrected for delayed delete we replace
> the dma_resv object to allow for easy reclaiming of the resources.
>
> v2: move that to ttm_bo_individualize_resv
> v3: add a comment to explain what's going on
>
> Signed-off-by:
It's 25 Mhz (refclk / 4). This fixes the interpretation
of the rlc clock counter.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/soc15.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c
b/drivers/gpu/drm/amd/amdgpu/soc15.c
[AMD Official Use Only - Internal Distribution Only]
Reviewed-by: Kenneth Feng
-Original Message-
From: amd-gfx On Behalf Of Evan Quan
Sent: Tuesday, February 11, 2020 1:41 PM
To: amd-gfx@lists.freedesktop.org
Cc: Quan, Evan
Subject: [PATCH] drm/amd/powerplay: always refetch the enabl
Since 'smu_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT)' will always return
false considering the 'smu_system_features_control(smu, false)' disabled
all SMU features.
Change-Id: I73956ffa51d6da8375c7c377895a221e13d31594
Signed-off-by: Evan Quan
---
drivers/gpu/drm/amd/powerplay/amdgpu_smu.c | 1
Acked-by: Evan Quan
> -Original Message-
> From: amd-gfx On Behalf Of Alex
> Deucher
> Sent: Wednesday, February 12, 2020 2:51 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander
> Subject: [PATCH] drm/amdgpu/soc15: fix xclk for raven
>
> It's 25 Mhz (refclk / 4). This fix
54 matches
Mail list logo