On 1/31/2025 3:00 AM, Vinay Belgaumkar wrote:
Functions to parse event ID and GT bit shift for PMU events.
v2: Review comments (Riana)
v3: Review comments (Lucas)
Cc: Riana Tauro
Cc: Lucas De Marchi
Cc: Kamil Konieczny
Cc: Rodrigo Vivi
Signed-off-by: Vinay Belgaumkar
Looks good to me
== Series Details ==
Series: drm/i915/gt: Use spin_lock_irqsave() in interruptible context (rev2)
URL : https://patchwork.freedesktop.org/series/143600/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_16048 -> Patchwork_143600v2
==
Hi Vinay
On 1/31/2025 3:00 AM, Vinay Belgaumkar wrote:
Simple tests for validating the PMU implementation for GT C6
residencies.
v2: Rename rc6-residency-* to gt-c6-residency and remove freq tests.
v3: Keep just gt-c6 tests, add frequency tests later.
v4: Review comments (Riana)
v5: Review comm
On Thu, 30 Jan 2025, Imre Deak wrote:
> On Thu, Jan 30, 2025 at 04:35:06PM +0200, Imre Deak wrote:
>> On Wed, Jan 29, 2025 at 04:46:38PM +0200, Jani Nikula wrote:
>> > Improve code clarity by using existing min_array() and max_array()
>> > helpers to find the lowest and highest values in an array.
On Fri, Jan 31, 2025 at 11:20:22AM +0200, Jani Nikula wrote:
> On Thu, 30 Jan 2025, Imre Deak wrote:
> > On Thu, Jan 30, 2025 at 04:35:06PM +0200, Imre Deak wrote:
> >> On Wed, Jan 29, 2025 at 04:46:38PM +0200, Jani Nikula wrote:
> >> > Improve code clarity by using existing min_array() and max_ar
Hi Vinay,
On 2025-01-30 at 13:30:28 -0800, Vinay Belgaumkar wrote:
> Simple tests for validating the PMU implementation for GT C6
> residencies.
>
> v2: Rename rc6-residency-* to gt-c6-residency and remove freq tests.
> v3: Keep just gt-c6 tests, add frequency tests later.
> v4: Review comments (R
On Fri, Jan 31, 2025 at 02:49:54PM +0200, Jani Nikula wrote:
> Commit 1c56e9a39833 ("drm/i915/dp: Get optimal link config to have best
> compressed bpp") tries to find the best compressed bpp for the
> link. However, it iterates from max to min bpp on display 13+, and from
> min to max on other pla
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups
URL : https://patchwork.freedesktop.org/series/144179/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_16049 -> Patchwork_144179v1
Summary
---
Hi Kirill,
On Thu, Jan 30, 2025 at 12:00:40PM +0200, Kirill A. Shutemov wrote:
> Use folios instead of pages.
>
> This is preparation for removing PG_reclaim.
>
> Signed-off-by: Kirill A. Shutemov
> Acked-by: David Hildenbrand
looks good:
Reviewed-by: Andi Shyti
Thanks,
Andi
On Fri, Jan 31, 2025 at 02:49:55PM +0200, Jani Nikula wrote:
> Add a platform independent helper for getting the supported DSC BPP step
> for the link.
>
> Signed-off-by: Jani Nikula
> ---
> drivers/gpu/drm/i915/display/intel_dp.c | 21 -
> 1 file changed, 16 insertions(+), 5
On Fri, 31 Jan 2025, Imre Deak wrote:
> On Fri, Jan 31, 2025 at 02:49:55PM +0200, Jani Nikula wrote:
>> Add a platform independent helper for getting the supported DSC BPP step
>> for the link.
>>
>> Signed-off-by: Jani Nikula
>> ---
>> drivers/gpu/drm/i915/display/intel_dp.c | 21 +
On Thu, Jan 30, 2025 at 12:00:40PM +0200, Kirill A. Shutemov wrote:
> Use folios instead of pages.
>
> This is preparation for removing PG_reclaim.
Well, this is a horrid little function. Rather than iterating just the
dirty folios, it iterates all folios, then locks them before checking
whether
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim sparse failed
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separately.
+./arch/x86/include/asm/bito
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim checkpatch failed
d8a74947ea11 drm/i915/dp: Iterate DSC BPP from high to low on all platforms
f3d7d3afbebc drm/i915/dp: Ad
On Fri, Jan 31, 2025 at 02:49:59PM +0200, Jani Nikula wrote:
> Instead of iterating the valid BPP array directly, switch to the same
> approach as xelpd_dsc_compute_link_config(), with a separate function to
> check if the DSC BPP is valid. This prepares us for unifying the
> platform specific func
Remove all KFD BOs from the private dma_resv object.
This prevents the KFD from being evitec unecessarily when an exported BO
is released.
Signed-off-by: Christian König
Reviewed-and-Tested-by: James Zhu
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h| 5 +-
..
One trivial TTM patch which fixes a bug for AMDKFD. As far as I can see
shouldn't have any effect on i915 nor XE, but better save than sorry.
Christian.
On Fri, Jan 31, 2025 at 02:50:00PM +0200, Jani Nikula wrote:
> Add the fractional DSC BPP force check to intel_dp_dsc_valid_bpp(), and
> use that in xelpd_dsc_compute_link_config(). This is another step closer
> towards unifying the platform specific functions.
>
> Signed-off-by: Jani Nikula
Rev
Test the fences in the private dma_resv object instead of the pointer to
a potentially shared dma_resv object.
This only matters for imported BOs with an SG table since those don't
get their dma_resv pointer replaced on release.
Signed-off-by: Christian König
Reviewed-and-Tested-by: James Zhu
R
On 1/31/2025 6:19 PM, Jani Nikula wrote:
Commit 1c56e9a39833 ("drm/i915/dp: Get optimal link config to have best
compressed bpp") tries to find the best compressed bpp for the
link. However, it iterates from max to min bpp on display 13+, and from
min to max on other platforms. This presumably
Hi,
https://patchwork.freedesktop.org/series/143990/ - Re-reported.
Thanks,
Tejasree
-Original Message-
From: I915-ci-infra On Behalf Of
Krzysztof Karas
Sent: Thursday, January 30, 2025 5:54 PM
To: i915-ci-in...@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: ✗ i
On Fri, Jan 31, 2025 at 02:50:06PM +0200, Jani Nikula wrote:
> Going forward, we'll need the connector state in
> dsc_compute_link_config(). Pass it along through the chain. Maintain the
> same parameter order where relevant.
>
> Signed-off-by: Jani Nikula
> ---
> drivers/gpu/drm/i915/display/in
== Series Details ==
Series: series starting with [1/2] drm/ttm: test private resv obj on
release/destroy
URL : https://patchwork.freedesktop.org/series/144188/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_16050 -> Patchwork_144188v1
=
On Fri, Jan 31, 2025 at 02:50:07PM +0200, Jani Nikula wrote:
> Move towards always using the fxp q4 or .4 fixed point format for
> compressed bpp. We'll need to pass the more accurate bpp to this
> function later on.
>
> Always use _x16 naming for variables that are in .4 fixed point for
> clarity
== Series Details ==
Series: drm/i915/selftests: avoid using uninitialized context (rev3)
URL : https://patchwork.freedesktop.org/series/143990/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_16041 -> Patchwork_143990v3
Summ
Hi Kirill,
On Thu, Jan 30, 2025 at 12:00:41PM +0200, Kirill A. Shutemov wrote:
> The recently introduced PG_dropbehind allows for freeing folios
> immediately after writeback. Unlike PG_reclaim, it does not need vmscan
> to be involved to get the folio freed.
>
> Instead of using folio_set_reclai
On Fri, Jan 31, 2025 at 02:49:56PM +0200, Jani Nikula wrote:
> Use the _x16 suffix for all .4 fixed point variables. Drop compressed_
> prefix, as it's implied from the precision suffix.
>
> As dsc_min_bpp and dsc_max_bpp change domain from int to .4 in the
> middle of the function, they remain th
On Fri, Jan 31, 2025 at 02:49:57PM +0200, Jani Nikula wrote:
> Try to keep the variables in the same domain a bit longer to reduce
> juggling between integers and .4 fixed point. Change parameter order to
> min, max while at it.
>
> For now, keep the juggling in dsc_compute_compressed_bpp() ensure
Disable FBC compressor clock gating before enabling FBC and
clear it after disabling FBC.
Bspec: 74212, 69741, 6
Signed-off-by: Vinod Govindapillai
---
drivers/gpu/drm/i915/display/intel_fbc.c | 15 +++
drivers/gpu/drm/i915/i915_reg.h | 6 ++
2 files changed, 21 ins
On Fri, Jan 31, 2025 at 02:50:04PM +0200, Jani Nikula wrote:
> Just use ints unless there are actual reasons to do otherwise. Here,
> there are not.
>
> Signed-off-by: Jani Nikula
> ---
> drivers/gpu/drm/i915/display/intel_dp.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
>
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev2)
URL : https://patchwork.freedesktop.org/series/144179/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_16050 -> Patchwork_144179v2
Summary
---
On Fri, Jan 31, 2025 at 02:50:05PM +0200, Jani Nikula wrote:
> The parameter is basically just a proxy for whether the function is
> being called for DP SST or DP MST. We can figure this out from crtc
> state.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Imre Deak
> ---
> drivers/gpu/drm/i915/
== Series Details ==
Series: drm/i915/display: implement wa_18038517565
URL : https://patchwork.freedesktop.org/series/144184/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_16050 -> Patchwork_144184v1
Summary
---
**F
On Fri, Jan 31, 2025 at 02:50:01PM +0200, Jani Nikula wrote:
> {icl,xelpd}_dsc_compute_link_config() are now effectively the same, and
> can be unified to a single platform independent function.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Imre Deak
> ---
> drivers/gpu/drm/i915/display/intel_d
On Fri, Jan 31, 2025 at 02:50:02PM +0200, Jani Nikula wrote:
> With just the one platform independent loop left in
> do_dsc_compute_compressed_bpp(), we don't really need the extra function
> that is simply becoming increasingly hard to even figure out a decent
> name for. Just merge the whole thin
On Fri, Jan 31, 2025 at 02:50:03PM +0200, Jani Nikula wrote:
> Drop the extra local variables and simplify the conditions. We don't
> have to try to special case the loop condition and break in the validity
> checks.
>
> Signed-off-by: Jani Nikula
Reviewed-by: Imre Deak
> ---
> drivers/gpu/dr
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev2)
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim sparse failed
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separately.
+./arch/x86/include/a
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev2)
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim checkpatch failed
749fb59743db drm/i915/dp: Iterate DSC BPP from high to low on all platforms
849813d442dd drm/i915
On Fri, Jan 31, 2025 at 02:49:58PM +0200, Jani Nikula wrote:
> Now that {icl,xelpd}_dsc_compute_link_config() take .4 fixed point as
> parameter, move the common max DSC BPP reduction one level higher. Use
> intel_dp_dsc_bpp_step() to compute the step, and pass on to both
> platform specific functi
On Fri, Jan 31, 2025 at 01:10:25AM -0500, Borah, Chaitanya Kumar wrote:
>
>
> > -Original Message-
> > From: Intel-xe On Behalf Of Ankit
> > Nautiyal
> > Sent: Friday, January 31, 2025 9:44 AM
> > To: intel-gfx@lists.freedesktop.org; intel...@lists.freedesktop.org
> > Cc: jani.nik...@lin
== Series Details ==
Series: series starting with [1/2] drm/ttm: test private resv obj on
release/destroy
URL : https://patchwork.freedesktop.org/series/144188/
State : warning
== Summary ==
Error: dim sparse failed
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separatel
== Series Details ==
Series: series starting with [1/2] drm/ttm: test private resv obj on
release/destroy
URL : https://patchwork.freedesktop.org/series/144188/
State : warning
== Summary ==
Error: dim checkpatch failed
881ca3736e5b drm/ttm: test private resv obj on release/destroy
-:16: WARN
On Fri, Jan 31, 2025 at 05:08:33PM +0200, Imre Deak wrote:
> On Fri, Jan 31, 2025 at 02:50:04PM +0200, Jani Nikula wrote:
> > Just use ints unless there are actual reasons to do otherwise. Here,
> > there are not.
> >
> > Signed-off-by: Jani Nikula
> > ---
> > drivers/gpu/drm/i915/display/intel_
This started off as preparation for UHBR SST DSC enabling, but escalated
quickly. The SST DSC code is unnecessarily complicated with the platform
differences and ints and fixed points being mixed. Clean it up quite a
bit, reducing the number of lines in the process.
BR,
Jani.
Jani Nikula (14):
With just the one platform independent loop left in
do_dsc_compute_compressed_bpp(), we don't really need the extra function
that is simply becoming increasingly hard to even figure out a decent
name for. Just merge the whole thing to
dsc_compute_compressed_bpp(). Good riddance to the short lived
d
Commit 1c56e9a39833 ("drm/i915/dp: Get optimal link config to have best
compressed bpp") tries to find the best compressed bpp for the
link. However, it iterates from max to min bpp on display 13+, and from
min to max on other platforms. This presumably leads to minimum
compressed bpp always being
Try to keep the variables in the same domain a bit longer to reduce
juggling between integers and .4 fixed point. Change parameter order to
min, max while at it.
For now, keep the juggling in dsc_compute_compressed_bpp() ensure
min/max will always have 0 fractional part. To be fixed later.
Signed
Now that {icl,xelpd}_dsc_compute_link_config() take .4 fixed point as
parameter, move the common max DSC BPP reduction one level higher. Use
intel_dp_dsc_bpp_step() to compute the step, and pass on to both
platform specific functions. (Though it's unused for now in
icl_dsc_compute_link_config()).
Use the _x16 suffix for all .4 fixed point variables. Drop compressed_
prefix, as it's implied from the precision suffix.
As dsc_min_bpp and dsc_max_bpp change domain from int to .4 in the
middle of the function, they remain the same for now.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/
{icl,xelpd}_dsc_compute_link_config() are now effectively the same, and
can be unified to a single platform independent function.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 49 +++--
1 file changed, 5 insertions(+), 44 deletions(-)
diff --git a/
Drop the extra local variables and simplify the conditions. We don't
have to try to special case the loop condition and break in the validity
checks.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
diff --gi
Instead of iterating the valid BPP array directly, switch to the same
approach as xelpd_dsc_compute_link_config(), with a separate function to
check if the DSC BPP is valid. This prepares us for unifying the
platform specific functions.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display
Add the fractional DSC BPP force check to intel_dp_dsc_valid_bpp(), and
use that in xelpd_dsc_compute_link_config(). This is another step closer
towards unifying the platform specific functions.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 10 +++---
1 file change
Add a platform independent helper for getting the supported DSC BPP step
for the link.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 21 -
1 file changed, 16 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c
b/driv
Going forward, we'll need the connector state in
dsc_compute_link_config(). Pass it along through the chain. Maintain the
same parameter order where relevant.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
di
Move towards always using the fxp q4 or .4 fixed point format for
compressed bpp. We'll need to pass the more accurate bpp to this
function later on.
Always use _x16 naming for variables that are in .4 fixed point for
clarity.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp
Just use ints unless there are actual reasons to do otherwise. Here,
there are not.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c
b/drivers/gpu/drm/i915/d
The parameter is basically just a proxy for whether the function is
being called for DP SST or DP MST. We can figure this out from crtc
state.
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 14 ++
drivers/gpu/drm/i915/display/intel_dp.h | 3 +--
dri
Quoting Krzysztof Karas (2025-01-30 11:18:28-03:00)
>Hi Gustavo,
>
>[...]
>> Let's remove that check, since it is unnecessary and causes the
>> inconsistency illustrated above.
>>
>> Signed-off-by: Gustavo Sousa
>> ---
>
>Would you add:
>Fixes: c01e78a96e12 ("drm/i915/dmc_wl: Sanitize enable_dmc_
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev3)
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim sparse failed
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separately.
+./arch/x86/include/a
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev3)
URL : https://patchwork.freedesktop.org/series/144179/
State : warning
== Summary ==
Error: dim checkpatch failed
91db7f941ad3 drm/i915/dp: Iterate DSC BPP from high to low on all platforms
b3ff9dd1b09d drm/i915
== Series Details ==
Series: drm/i915/pmu: Drop custom hotplug code (rev3)
URL : https://patchwork.freedesktop.org/series/143636/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_16053 -> Patchwork_143636v3
Summary
---
On Fri, Jan 31, 2025 at 05:16:29PM -0300, Gustavo Sousa wrote:
Quoting Krzysztof Karas (2025-01-30 11:18:28-03:00)
Hi Gustavo,
[...]
Let's remove that check, since it is unnecessary and causes the
inconsistency illustrated above.
Signed-off-by: Gustavo Sousa
---
Would you add:
Fixes: c01e7
Add a platform independent helper for getting the supported DSC BPP step
for the link.
v2: Use fxp_q4_from_int(1) (Imre)
Reviewed-by: Imre Deak
Signed-off-by: Jani Nikula
---
drivers/gpu/drm/i915/display/intel_dp.c | 21 -
1 file changed, 16 insertions(+), 5 deletions(-)
d
== Series Details ==
Series: drm/i915/pmu: Drop custom hotplug code (rev3)
URL : https://patchwork.freedesktop.org/series/143636/
State : warning
== Summary ==
Error: dim sparse failed
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separately.
Since commit 4ba4f1afb6a9 ("perf: Generic hotplug support for a PMU with
a scope"), there's generic support for system-wide counters and
integration with cpu hotplug.
The i915 counters are system-wide, even though the migration code
is using the wrong topology mask:
target = cpumask_any_b
== Series Details ==
Series: drm/i915/dp: dsc fix, refactoring and cleanups (rev3)
URL : https://patchwork.freedesktop.org/series/144179/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_16053 -> Patchwork_144179v3
Summary
---
67 matches
Mail list logo
