From: zhengbin
Fixes coccicheck warning:
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c:723:2-50: WARNING: Assignment
of 0/1 to bool variable
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c:733:3-52: WARNING: Assignment
of 0/1 to bool variable
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.
Hi Luben,
Thanks for reviewing. Was expecting lot from vim auto-indent :/
On 1/22/20 1:07 AM, Luben Tuikov wrote:
On 2020-01-21 11:50 a.m., Nirmoy Das wrote:
- switch (i) {
+ priority = (ctx->override_priority == DRM_SCHED_PRIORITY_UNSET) ?
+
Do not allocate all the entity at once. This is required for
dynamic amdgpu_ctx_entity creation.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 143 +---
1 file changed, 76 insertions(+), 67 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu
On 1/22/20 12:34 AM, Luben Tuikov wrote:
+ for (j = 0; j < amdgpu_ctx_num_entities[i]; ++j) {
+ struct amdgpu_ctx_entity *entity = &ctx->entities[i][j];
- kfree(entity->fences);
- entity->fences = NULL;
+ kfree(entity
Currently we pre-allocate entities and fences for all the HW IPs on
context creation and some of which are might never be used.
This patch tries to resolve entity/fences wastage by creating entities
for a HW IP only when it is required.
v2: consolidated priority checking at amdgpu_ctx_priority_pe
On 1/21/20 3:31 PM, William Lewis wrote:
+ * @sched_list: A list of schedulers (drm_gpu_schedulers).
+ * Jobs from this entity, can be scheduled on any scheduler
There's no reason to separate the subject from the verb here with a comma.
Thanks William.
+ * on th
expand sched_list definition for better understanding.
Also fix a typo atleast -> at least
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/scheduler/sched_entity.c | 2 +-
include/drm/gpu_scheduler.h | 5 +++--
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu
Fixes: 445a7d54 (drm/amdgpu/vcn2.5: fix a bug for the 2nd vcn instance)
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index de32e96a8409..70fae7
Am 22.01.20 um 10:38 schrieb Nirmoy Das:
expand sched_list definition for better understanding.
Also fix a typo atleast -> at least
Signed-off-by: Nirmoy Das
Reviewed-by: Christian König
---
drivers/gpu/drm/scheduler/sched_entity.c | 2 +-
include/drm/gpu_scheduler.h | 5 ++
Am 22.01.20 um 10:33 schrieb Nirmoy Das:
Currently we pre-allocate entities and fences for all the HW IPs on
context creation and some of which are might never be used.
This patch tries to resolve entity/fences wastage by creating entities
for a HW IP only when it is required.
v2: consolidated
Am 22.01.20 um 10:44 schrieb Nirmoy Das:
Fixes: 445a7d54 (drm/amdgpu/vcn2.5: fix a bug for the 2nd vcn instance)
Signed-off-by: Nirmoy Das
Acked-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vc
On 1/22/20 12:39 PM, Christian König wrote:
Am 22.01.20 um 10:33 schrieb Nirmoy Das:
Currently we pre-allocate entities and fences for all the HW IPs on
context creation and some of which are might never be used.
This patch tries to resolve entity/fences wastage by creating entities
for a HW I
Allow partial invalidation on unallocated PDs. This is useful when we
need to silence faults to stop interrupt floods on Vega.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 23 ++-
1 file changed, 18 insertions(+), 5 deletions(-)
diff --git a/dr
That we can't find a PD above the root is expected can only happen if
we try to update a larger range than actually managed by the VM.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd
On 2020-01-22 4:25 a.m., Nirmoy wrote:
> Hi Luben,
>
> Thanks for reviewing. Was expecting lot from vim auto-indent :/
Yes, no problem. It's a learning experience for me as well.
If you use Emacs, pressing TAB anywhere on a line, will indent
it according to the mode. (It runs the (c-indent-line
On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
> [Why]
> We need to support SRM
Why do we need to support SRM?
Please refer to section 5 "Renewability" of the HDCP 2.x spec and its
requirement to store the SRM in non-volatile memory.
Why does this involve PSP?
Describe that for AMD's content
On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
> These patches adds support for SRM loading. By providing an interface to the
> usermode
>
> SRM has to be persistent and since the kernel cannot directly write to system
> storage we need to provide an interface so that the usermode can do it for
On Wed, Jan 22, 2020 at 11:12 AM Harry Wentland wrote:
>
> On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
> > These patches adds support for SRM loading. By providing an interface to the
> > usermode
> >
> > SRM has to be persistent and since the kernel cannot directly write to
> > system
> >
On 2020-01-22 4:32 a.m., Nirmoy wrote:
>
> On 1/22/20 12:34 AM, Luben Tuikov wrote:
>>
>>> + for (j = 0; j < amdgpu_ctx_num_entities[i]; ++j) {
>>> + struct amdgpu_ctx_entity *entity = &ctx->entities[i][j];
>>>
>>> - kfree(entity->fences);
>>> - e
On 2020-01-22 11:23 a.m., Alex Deucher wrote:
On Wed, Jan 22, 2020 at 11:12 AM Harry Wentland wrote:
On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
These patches adds support for SRM loading. By providing an interface to the
usermode
SRM has to be persistent and since the kernel cannot di
Should work properly with the latest sbios on 5.5 and newer
kernels.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
inde
Hello all,
I recently started playing the game "0.A.D." and found after a while
the display go blank. Music continues playing and the machine can still be
reached through ssh.
It looks to me same as problem reported here (dmesg output is almost
equal): https://gitlab.freedesktop.org/mesa/mesa/iss
Do we have user-space/libs rely on this errno in particular?
On 2020-01-22 9:03 a.m., Christian König wrote:
> That we can't find a PD above the root is expected can only happen if
> we try to update a larger range than actually managed by the VM.
>
> Signed-off-by: Christian König
> ---
> driv
Am 22.01.20 um 19:11 schrieb Luben Tuikov:
Do we have user-space/libs rely on this errno in particular?
No, userspace can't even directly call this function.
The problem was rather that during debugging I called the function with
fixed parameters and wondered for quite a while why it can't fi
On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
> [Why]
> We need to set/get SRM and linux kernel is not suppose to write to the
"We need to set/get SRM from/to PSP FW..."
> storage, so we need to provide a interface.
>
an interface.
Highlight that we expect these to be exercised by usermode
On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
> Call the cmd ids for set/get srm according to the sysfs call
>
> Signed-off-by: Bhawanpreet Lakha
> Reviewed-by: Rodrigo Siqueira
> ---
> .../amd/display/amdgpu_dm/amdgpu_dm_hdcp.c| 49 ++-
> 1 file changed, 48 insertions(+
On 2020-01-22 11:44 a.m., Bhawanpreet Lakha wrote:
>
> On 2020-01-22 11:23 a.m., Alex Deucher wrote:
>> On Wed, Jan 22, 2020 at 11:12 AM Harry Wentland wrote:
>>> On 2020-01-16 3:29 p.m., Bhawanpreet Lakha wrote:
These patches adds support for SRM loading. By providing an
interface to
[AMD Public Use]
If you resend the patch set, it might be good to include the script in the
series for reference as well.
Alex
From: amd-gfx on behalf of Harry
Wentland
Sent: Wednesday, January 22, 2020 2:27 PM
To: Lakha, Bhawanpreet ; Alex Deucher
Cc: Went
On Wed, 2020-01-22 at 04:48 +, Lin, Wayne wrote:
> [AMD Public Use]
>
> Sorry for any inconvenience I brought.
Nothing to be sorry about! This happens from time to time, it's my fault for
not noticing it in the first place anyway :P. Ville from Intel is able to
review it, so there's no rush
On Tue, Jan 21, 2020 at 11:08 AM Chen Zhou wrote:
>
> This patch series remove unnecessary conversion to bool in dir
> drivers/gpu/drm/amd/amdgpu/, which is detected by coccicheck.
Thanks for the patches. Already applied this patch:
https://patchwork.freedesktop.org/series/72281/#rev2
which cove
On Wed, Jan 22, 2020 at 3:22 AM Zheng Bin wrote:
>
> From: zhengbin
>
> Fixes coccicheck warning:
>
> drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c:723:2-50: WARNING:
> Assignment of 0/1 to bool variable
> drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c:733:3-52: WARNING:
> Assignment of 0/
These patches adds support for SRM loading. SRM has to be saved in a
non-volatile memory(spec) and PSP can't do that directly so we need to
use the driver to do this
Since the kernel cannot directly write to system storage we need to provide an
interface so that the usermode can do it for us
v2
-
[Why]
We need to support SRM(System Renewability Message)
As per hdcp spec (5.Renewability) SRM needs to be storage in a non-volatile
memory.
PSP owns the checking of SRM but doesn't have the ability to store it in a
non-volatile memory. So we need the kernel driver to facilitate it using the
inte
Call the cmd ids for set/get srm according to the sysfs call
v2: Use define for the magic number
Signed-off-by: Bhawanpreet Lakha
Reviewed-by: Rodrigo Siqueira
---
.../amd/display/amdgpu_dm/amdgpu_dm_hdcp.c| 50 ++-
1 file changed, 49 insertions(+), 1 deletion(-)
diff --gi
This is just a reference for the patches. not to be merged
Signed-off-by: Bhawanpreet Lakha
---
REFERENCE | 49 +
1 file changed, 49 insertions(+)
create mode 100644 REFERENCE
diff --git a/REFERENCE b/REFERENCE
new file mode 100644
index
[Why]
We need this to create sysfs (followup patch)
[How]
Change the parameter
Signed-off-by: Bhawanpreet Lakha
Reviewed-by: Rodrigo Siqueira
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c | 4 ++--
drivers/gpu/drm/amd/
[Why]
we need to load SRM before we start HDCP. Because for S3 case the sysfs call
will be
after we have already enabled HDCP, so we might not be using the latest SRM
[How]
Set srm before starting HDCP.
Signed-off-by: Bhawanpreet Lakha
Reviewed-by: Rodrigo Siqueira
---
drivers/gpu/drm/amd/dis
[Why]
PSP doesn't have the ability to store SRM in a non-volatile memory. And since
the kernel cannot write to the storage directly, we need usermode to facilitate
this
As per spec the SRM needs to be persistent so this interface is to be
called by the usermode anytime the system goes down/powers
Two style nit-picks inline. Otherwise the series is
Reviewed-by: Felix Kuehling
On 2020-01-22 9:03 a.m., Christian König wrote:
Allow partial invalidation on unallocated PDs. This is useful when we
need to silence faults to stop interrupt floods on Vega.
Signed-off-by: Christian König
---
39 matches
Mail list logo
