On Wed, Nov 20, 2024 at 5:17 PM Petr Vorel wrote:
>
> > On Thu, Nov 21, 2024 at 5:41 AM Petr Vorel wrote:
>
> > > It will be used in the next commit for DRM_MSM.
>
> > > Suggested-by: Rob Clark
> > > Signed-off-by: Petr Vorel
> > > ---
> > > Changes v3->v4:
> > > * Move definition to the end of
On Thu, Nov 21, 2024 at 5:41 AM Petr Vorel wrote:
>
> It will be used in the next commit for DRM_MSM.
>
> Suggested-by: Rob Clark
> Signed-off-by: Petr Vorel
> ---
> Changes v3->v4:
> * Move definition to the end of the file
I prefer to not check the tool.
Why don't you install python3?
>
On Wed, Nov 20, 2024 at 3:18 AM Dmitry Baryshkov
wrote:
>
> On Tue, Nov 19, 2024 at 06:56:38PM +0100, Neil Armstrong wrote:
> > Now the features defines have the right name, introduce a features
> > bitfield and move the features defines in it, fixing all code checking
> > for them.
> >
> > No fun
On Wed, Nov 20, 2024 at 7:49 AM Petr Vorel wrote:
>
> 0fddd045f88e introduced python3 dependency, require it to quick early.
>
> Signed-off-by: Petr Vorel
> ---
> Hi all,
>
> RFC because I'm not sure if previous failed build wasn't better:
>
> GENHDR drivers/gpu/drm/msm/generated/a2xx.xm
The Adreno GMU Management Unit (GNU) can also scale the DDR Bandwidth
along the Frequency and Power Domain level, but by default we leave the
OPP core vote for the interconnect ddr path.
While scaling via the interconnect path was sufficient, newer GPUs
like the A750 requires specific vote paremet
On Tue, Nov 19, 2024 at 06:56:45PM +0100, Neil Armstrong wrote:
> Each GPU OPP requires a specific peak DDR bandwidth, let's add
> those to each OPP and also the related interconnect path.
>
> Signed-off-by: Neil Armstrong
> ---
> arch/arm64/boot/dts/qcom/sm8550.dtsi | 11 +++
> 1 file c
The variable pixel_div can remain zero due to an invalid rate input,
leading to a potential division by zero issue. This patch fixes it and
the function now logs an error and returns early.
Additionally, this patch resolves trailing whitespace issues detected
by checkpatch.pl in the same file.
Si
Half of the current "Quirks" are in fact features, so rename
the defines with FEAT instead of QUIRK.
They will be moved in a separate bitfield in a second time.
No functional changes.
Signed-off-by: Neil Armstrong
---
drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 62 +++-
On Wed, Nov 20, 2024 at 01:02:32PM +0200, Dmitry Baryshkov wrote:
> On Wed, Nov 20, 2024 at 10:34:51AM +0530, Dheeraj Reddy Jonnalagadda wrote:
> > The variable pixel_div can remain zero due to an invalid rate input,
>
> No, it can not. Rate is set by the driver, which knowns which rates are
> sup
On 20/11/2024 12:47, Dmitry Baryshkov wrote:
On Tue, Nov 19, 2024 at 06:56:45PM +0100, Neil Armstrong wrote:
Each GPU OPP requires a specific peak DDR bandwidth, let's add
those to each OPP and also the related interconnect path.
Signed-off-by: Neil Armstrong
---
arch/arm64/boot/dts/qcom/sm8
On 20/11/2024 12:19, Dmitry Baryshkov wrote:
On Tue, Nov 19, 2024 at 06:56:39PM +0100, Neil Armstrong wrote:
The Adreno GMU Management Unit (GNU) can also scale the DDR Bandwidth
along the Frequency and Power Domain level, but by default we leave the
OPP core vote for the interconnect ddr path.
On 20/11/2024 12:42, Dmitry Baryshkov wrote:
On Tue, Nov 19, 2024 at 06:56:40PM +0100, Neil Armstrong wrote:
The Adreno GMU Management Unit (GMU) can also scale DDR Bandwidth along
the Frequency and Power Domain level, but by default we leave the
OPP core scale the interconnect ddr path.
In ord
On Wed, Nov 20, 2024 at 05:24:20PM +0530, Dheeraj Reddy Jonnalagadda wrote:
> On Wed, Nov 20, 2024 at 01:02:32PM +0200, Dmitry Baryshkov wrote:
> > On Wed, Nov 20, 2024 at 10:34:51AM +0530, Dheeraj Reddy Jonnalagadda wrote:
> > > The variable pixel_div can remain zero due to an invalid rate input,
On Tue, Nov 19, 2024 at 06:56:39PM +0100, Neil Armstrong wrote:
> The Adreno GMU Management Unit (GNU) can also scale the DDR Bandwidth
> along the Frequency and Power Domain level, but by default we leave the
> OPP core vote for the interconnect ddr path.
>
> While scaling via the interconnect pa
On Tue, Nov 19, 2024 at 06:56:40PM +0100, Neil Armstrong wrote:
> The Adreno GMU Management Unit (GMU) can also scale DDR Bandwidth along
> the Frequency and Power Domain level, but by default we leave the
> OPP core scale the interconnect ddr path.
>
> In order to calculate vote values used by th
On Tue, Nov 19, 2024 at 06:56:46PM +0100, Neil Armstrong wrote:
> Each GPU OPP requires a specific peak DDR bandwidth, let's add
> those to each OPP and also the related interconnect path.
>
> Signed-off-by: Neil Armstrong
> ---
> arch/arm64/boot/dts/qcom/sm8650.dtsi | 14 ++
> 1 fil
On Tue, Nov 19, 2024 at 06:56:44PM +0100, Neil Armstrong wrote:
> Now all the DDR bandwidth voting via the GPU Management Unit (GMU)
> is in place, declare the Bus Control Modules (BCMs) and the
> corresponding parameters in the GPU info struct and add the
> GMU_BW_VOTE feature bit to enable it.
>
On Tue, Nov 19, 2024 at 06:56:42PM +0100, Neil Armstrong wrote:
> The Adreno GMU Management Unit (GMU) can also scale the DDR Bandwidth
> along the Frequency and Power Domain level, until now we left the OPP
> core scale the OPP bandwidth via the interconnect path.
>
> In order to enable bandwidth
On Tue, Nov 19, 2024 at 06:56:38PM +0100, Neil Armstrong wrote:
> Now the features defines have the right name, introduce a features
> bitfield and move the features defines in it, fixing all code checking
> for them.
>
> No functional changes intended.
I think it might be better to squahs this p
On Wed, Nov 20, 2024 at 10:34:51AM +0530, Dheeraj Reddy Jonnalagadda wrote:
> The variable pixel_div can remain zero due to an invalid rate input,
No, it can not. Rate is set by the driver, which knowns which rates are
supported.
> leading to a potential division by zero issue. This patch fixes
On Fri, Aug 02, 2024 at 10:47:33PM +0300, Dmitry Baryshkov wrote:
> During suspend/resume process all connectors are explicitly disabled and
> then reenabled. However resume fails because of the connector_status check:
>
> [ 1185.831970] [dpu error]connector not connected 3
Please also include th
On Tue, Nov 19, 2024 at 10:02:33PM -0500, Leonard Lausen wrote:
> The finding is that while 6.10.14 with this patch applied still suffers from
> that regression, 6.11.9 and 6.12 do not face the CRTC state regression.
> Therefore, whatever issue the patch uncovered in older kernels and which
> just
22 matches
Mail list logo
