From: Andy Yan
Use the common dp link power up/down helpers to avoid duplicating code.
Signed-off-by: Andy Yan
Reviewed-by: Dmitry Baryshkov
---
(no changes since v1)
.../drm/bridge/cadence/cdns-mhdp8546-core.c | 74 +--
1 file changed, 2 insertions(+), 72 deletions(-)
di
From: Andy Yan
Use the common dp link power up/down helpers to avoid duplicating code.
Signed-off-by: Andy Yan
Reviewed-by: Dmitry Baryshkov
---
(no changes since v1)
.../drm/bridge/analogix/analogix-anx78xx.c| 30 +--
1 file changed, 1 insertion(+), 29 deletions(-)
dif
From: Andy Yan
Use the common dp link power up/down helpers to avoid duplicating code.
Signed-off-by: Andy Yan
Reviewed-by: Dmitry Baryshkov
---
(no changes since v1)
.../drm/bridge/analogix/analogix-anx6345.c| 30 +--
1 file changed, 1 insertion(+), 29 deletions(-)
dif
From: Andy Yan
The helper functions drm_dp_link_power_up/down were moved to Tegra
DRM in commit 9a42c7c647a9 ("drm/tegra: Move drm_dp_link helpers to Tegra
DRM")".
Now since more and more users are duplicating the same code in their
own drivers, it's time to make them as DRM DP common helpers a
From: Andy Yan
Use the common dp link power up/down helpers to avoid duplicating code.
Signed-off-by: Andy Yan
Reviewed-by: Dmitry Baryshkov
---
(no changes since v1)
drivers/gpu/drm/bridge/ite-it6505.c | 46 +++--
1 file changed, 4 insertions(+), 42 deletions(-)
di
From: Andy Yan
Because the layer/window enable/disable is take effect by vsync, if the
overlay configuration of these layers does not follow vsync and
takes effect immediately instead, when multiple layers are dynamically
enable/disable, inconsistent display contents may be seen on the screen.
S
Variable "bridge" is allocated by agp_alloc_bridge() and
have to be released by agp_put_bridge() if something goes
wrong. In this patch, add the missing call of agp_put_bridge()
in agp_amdk7_probe() to prevent potential memory leak bug.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: sta...@vger.kern
Gently ping！
Hi, Dmitry. There is no more comment for this patch set for one week, I wonder,
could this patch be applied, or wait for more comments.
Thanks,
Baihan
From: Baihan Li
To support DP HPD, edid printing, and colorbar display features based on
the Hisislcon DP devices.
---
ChangeLo
On Sun, Mar 16, 2025 at 7:32 PM Dmitry Baryshkov <
dmitry.barysh...@oss.qualcomm.com> wrote:
> On Sun, Mar 16, 2025 at 02:58:22PM -0400, Anusha Srivatsa wrote:
> > Move away from using deprecated API and use _multi
> > variants if available. Use mipi_dsi_msleep()
> > and mipi_dsi_usleep_range() in
In order to enforce per-SoC register rules, add per-compatible
restrictions. For example, V3D 3.3 (used in brcm,7268-v3d) has a cache
controller (GCA), which is not present in other V3D generations. Declaring
these differences helps ensure the DTB accurately reflect the hardware
design.
The exampl
On Thu, Mar 6, 2025 at 2:52 AM Huang, Honglei1
wrote:
>
> On 2025/3/1 5:21, Demi Marie Obenour wrote:
> > On 2/28/25 12:36 AM, Honglei Huang wrote:
> >> From: Honglei Huang
> >>
> >> Add a new resource for blob resource, called userptr, used for let
> >> host access guest user space memory, to a
In addition to the standard reset controller, V3D 7.x requires configuring
the V3D_SMS registers for proper power on/off and reset. Add the new
registers to `v3d_regs.h` and ensure they are properly configured during
device probing, removal, and reset.
This change fixes GPU reset issues on the Ras
As established in commit 89d04995f76c ("MAINTAINERS: Drop Emma Anholt
from all M lines."), Emma is no longer active in the Linux kernel and
dropped the V3D maintainership. Therefore, remove Emma as one of the DT
maintainers and add the current V3D driver maintainer.
Acked-by: Emma Anholt
Acked-by
The V3D driver currently determines the GPU tech version (33, 41...)
by reading a register. This approach has worked so far since this
information wasn’t needed before powering on the GPU.
V3D 7.1 introduces new registers that must be written to power on the
GPU, requiring us to know the V3D versi
Dave Airlie writes:
>> Dave, you're the only one with fingerprints on this document; is the
>> change OK with you?
>
> LGTM,
>
> Acked-by: Dave Airlie
OK, applied - thanks.
jon
On Tue, 18 Mar 2025 at 08:53, Jonathan Corbet wrote:
>
> Jacek Lawrynowicz writes:
>
> > The guidelines mention that firmware updates can't break the kernel,
> > but it doesn't state directly that they can't break userspace programs.
> > Make it explicit that firmware updates cannot break UAPI.
>
Jacek Lawrynowicz writes:
> The guidelines mention that firmware updates can't break the kernel,
> but it doesn't state directly that they can't break userspace programs.
> Make it explicit that firmware updates cannot break UAPI.
>
> Signed-off-by: Jacek Lawrynowicz
> ---
> .../driver-api/firm
On Tue, 11 Mar 2025 01:49:52 +, li...@treblig.org wrote:
> The pcf50633 was used as part of the OpenMoko devices but
> the support for its main chip was recently removed in:
> commit 61b7f8920b17 ("ARM: s3c: remove all s3c24xx support")
>
> See https://lore.kernel.org/all/Z8z236h4B5A6Ki3D@gall
-Original Message-
From: Wajdeczko, Michal
Sent: Saturday, March 15, 2025 7:45 AM
To: Cavitt, Jonathan ; intel...@lists.freedesktop.org
Cc: Gupta, saurabhg ; Zuo, Alex ;
joonas.lahti...@linux.intel.com; Brost, Matthew ;
Zhang, Jianxun ; Lin, Shuicheng
; dri-devel@lists.freedesktop.org;
soc/qcom/qcom,sps-sic.yaml | 38 ++
> arch/arm/boot/dts/qcom/qcom-apq8064.dtsi | 85
> +++---
> 5 files changed, 136 insertions(+), 44 deletions(-)
> ---
> base-commit: 2bb3cf7805af73af62ebfcbb55e350549b6a0912
> change-id: 20250317-fix-nexus-4-dd0fe3bca98
On 2025-03-17 15:07, Deucher, Alexander wrote:
> [Public]
>
>> -Original Message-
>> From: Daisuke Matsuda
>> Sent: Thursday, March 13, 2025 9:18 PM
>> To: amd-...@lists.freedesktop.org; dri-devel@lists.freedesktop.org; Deucher,
>> Alexander ; Koenig, Christian
>>
>> Cc: airl...@gmail.c
On 17/03/2025 20:20, Kaustabh Chakraborty wrote:
> Exynos7870 SoC uses the ARM Mali T830 GPU, document its compatible
> string with the appropriate fallback. The T830 compatible is already
> defined in the panfrost driver, but was commented out as it was unused.
>
> Signed-off-by: Kaustabh Chakrab
aximum', 'exclusiveMinimum', 'minimum', 'maximum',
'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#
doc reference errors (make refc
Exynos7870 SoC uses the ARM Mali T830 GPU, document its compatible
string with the appropriate fallback. The T830 compatible is already
defined in the panfrost driver, but was commented out as it was unused.
Signed-off-by: Kaustabh Chakraborty
---
This patch series is a part of Exynos7870 upstrea
[Public]
> -Original Message-
> From: Daisuke Matsuda
> Sent: Thursday, March 13, 2025 9:18 PM
> To: amd-...@lists.freedesktop.org; dri-devel@lists.freedesktop.org; Deucher,
> Alexander ; Koenig, Christian
>
> Cc: airl...@gmail.com; sim...@ffwll.ch; Daisuke Matsuda dais...@fujitsu.com>
In order to fix DT schema warning and describe hardware properly, add
missing sleep clock to the timer node.
Fixes: f335b8af4fd5 ("ARM: dts: qcom: Add initial APQ8064 SoC and IFC6410 board
device trees")
Signed-off-by: Dmitry Baryshkov
---
arch/arm/boot/dts/qcom/qcom-apq8064.dtsi | 2 ++
1 file
ommit: 2bb3cf7805af73af62ebfcbb55e350549b6a0912
change-id: 20250317-fix-nexus-4-dd0fe3bca987
Best regards,
--
Dmitry Baryshkov
The CoreSight replicator device isn't a part of the system MMIO bus, as
such it should not be a part of the soc node. Follow the example of
other platforms and move it out of the soc bus to the top-level.
Fixes: 7a5c275fd821 ("ARM: dts: qcom: Add apq8064 CoreSight components")
Signed-off-by: Dmitr
Use new SoC-specific compatible to the SPS SIC in addition to the
"syscon" compatible and rename the node to follow the purpose of it.
Signed-off-by: Dmitry Baryshkov
---
arch/arm/boot/dts/qcom/qcom-apq8064.dtsi | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/boo
Use new SoC-specific compatible for the SFPB device node in addition to
the "syscon" compatible.
Signed-off-by: Dmitry Baryshkov
---
arch/arm/boot/dts/qcom/qcom-apq8064.dtsi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm/boot/dts/qcom/qcom-apq8064.dtsi
b/arch/arm/b
Follow up the expected way of describing the SFPB hwspinlock and merge
hwspinlock node into corresponding syscon node, fixing several dt-schema
warnings.
Fixes: 24a9baf933dc ("ARM: dts: qcom: apq8064: Add hwmutex and SMEM nodes")
Signed-off-by: Dmitry Baryshkov
---
arch/arm/boot/dts/qcom/qcom-ap
Add DT schema for the MultiMedia SubSystem System FPB device, which
provides several registers to control interface between multimedia
devices (primarily display) and system busses.
Signed-off-by: Dmitry Baryshkov
---
.../devicetree/bindings/display/msm/qcom,sfpb.yaml | 39 ++
As most other CoreSight devices the replicator can use either of the
optional clocks (or both). Document those optional clocks in the schema.
Fixes: 3c15fddf3121 ("dt-bindings: arm: Convert CoreSight bindings to DT
schema")
Signed-off-by: Dmitry Baryshkov
---
.../bindings/arm/arm,coresight-stat
Add schema to properly describe the SPS SIC block present on APQ8064 and
several similar platforms.
Signed-off-by: Dmitry Baryshkov
---
.../devicetree/bindings/soc/qcom/qcom,sps-sic.yaml | 38 ++
1 file changed, 38 insertions(+)
diff --git a/Documentation/devicetree/bindings
On Qualcomm platforms in addition to regulators the RPM also provides
clocks via the child clock-controller node. Describe it properly in the
schema.
Fixes: 6b42133d2189 ("dt-bindings: soc: qcom: convert non-smd RPM bindings to
dt-schema")
Signed-off-by: Dmitry Baryshkov
---
Documentation/devic
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Boris Brezillon
Cc: Ste
Avoid dereferencing struct drm_gem_object.import_attach for the
imported dma-buf. The dma_buf field in the GEM object instance refers
to the same buffer. Prepares to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Lucas Stach
Cc: Russell King
Cc: Christian Gmeiner
Cc: etna...
This series enables jack detection for HDMI audio on vc4 devices. This
means that userspace audio applications can detect when HDMI devices are
connected or disconnected, for example to play/pause or switch to
another audio device.
Changes in v4:
- Split the update of the comment on
drm_atomic_
On Mon, Mar 17, 2025 at 09:42:11AM +0100, Thomas Zimmermann wrote:
> Hi
>
> Am 13.03.25 um 17:51 schrieb Lee Jones:
> > On Thu, 06 Mar 2025, Thomas Zimmermann wrote:
> >
> > > This series removes the remaining dependencies on fbdev from the
> > > backlight, lcd and led subsystems. Each depends on f
Add ALSA jack detection to the vc4-hdmi audio driver so userspace knows
when to add/remove HDMI audio devices.
Signed-off-by: Stefan Wahren
Reviewed-by: Maxime Ripard
Signed-off-by: David Turner
---
drivers/gpu/drm/vc4/vc4_hdmi.c | 18 ++
drivers/gpu/drm/vc4/vc4_hdmi.h | 7 +++
From: Stefan Wahren
drm_atomic_helper_connector_hdmi_hotplug() must be called
regardless of the connection status, otherwise the HDMI audio
disconnect event won't be notified.
Fixes: 2ea9ec5d2c20 ("drm/vc4: hdmi: use
drm_atomic_helper_connector_hdmi_hotplug()")
Suggested-by: Dmitry Baryshkov
S
Update the comment on drm_atomic_helper_connector_hdmi_hotplug() to
clarify that it must be called for all status updates.
Signed-off-by: David Turner
---
drivers/gpu/drm/display/drm_hdmi_state_helper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/display/d
On Fri, 14 Mar 2025 19:08:22 +0100
Maxime Ripard wrote:
> On Fri, Mar 14, 2025 at 11:31:18AM +0100, Luca Ceresoli wrote:
> > Many functions get a drm_bridge pointer, only use it in the function body
> > (or a smaller scope such as a loop body), and don't store it. In these
> > cases they always n
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Lucas Stach
Cc: Russell
Hi Janusz,
On 2025-03-14 at 21:38:35 GMT, Janusz Krzysztofik wrote:
> Starting with commit ec3e00b4ee27 ("drm/i915: stop registering if
> drm_dev_register() fails"), we return from i915_driver_register()
> immediately if drm_dev_register() fails, skipping remaining registration
> steps, and contin
Hi Janusz,
On 2025-03-14 at 21:38:34 GMT, Janusz Krzysztofik wrote:
> Commit f67986b0119c04 ("drm/i915/pxp: Promote pxp subsystem to top-level
> of i915") added PXP initialization to driver probe path, but didn't add a
> respective PXP cleanup on probe error. That lack of cleanup seems
> harmless
Hi Janusz,
thanks for the series.
On 2025-03-14 at 21:38:33 GMT, Janusz Krzysztofik wrote:
> Commit 8f460e2c78f2 ("drm/i915: Demidlayer driver loading") which
> introduced manual device registration also added a message that is
> submitted on device registration failure as an error. If that fail
`std::ffi::CStr` was moved to `core::ffi::CStr` in Rust 1.64. Replace
`kernel::str::CStr` with `core::ffi::CStr` now that we can.
C-String literals were added in Rust 1.77. Opportunistically replace
instances of `kernel::c_str!` with C-String literals where other code
changes were already necessar
Ignore, managed to mangle two commits into one.
On 17/03/2025 14:46, Colin Ian King wrote:
Don't populate the const read-only array dflt_rules on the stack at run
time, instead make it static.
Signed-off-by: Colin Ian King
---
drivers/gpu/drm/i915/intel_memory_region.c| 2 +-
driv
From: Mario Limonciello
[ Upstream commit 4afacc9948e1f8fdbca401d259ae65ad93d298c0 ]
When userspace invokes S4 the flow is:
1) amdgpu_pmops_prepare()
2) amdgpu_pmops_freeze()
3) Create hibernation image
4) amdgpu_pmops_thaw()
5) Write out image to disk
6) Turn off system
Then on resume amdgpu_
From: Mario Limonciello
[ Upstream commit 4afacc9948e1f8fdbca401d259ae65ad93d298c0 ]
When userspace invokes S4 the flow is:
1) amdgpu_pmops_prepare()
2) amdgpu_pmops_freeze()
3) Create hibernation image
4) amdgpu_pmops_thaw()
5) Write out image to disk
6) Turn off system
Then on resume amdgpu_
From: Mario Limonciello
[ Upstream commit 4afacc9948e1f8fdbca401d259ae65ad93d298c0 ]
When userspace invokes S4 the flow is:
1) amdgpu_pmops_prepare()
2) amdgpu_pmops_freeze()
3) Create hibernation image
4) amdgpu_pmops_thaw()
5) Write out image to disk
6) Turn off system
Then on resume amdgpu_
Am 15.03.25 um 05:40 schrieb feng.w...@zte.com.cn:
> From: FengWei
>
> Use max3() macro instead of nesting max() to simplify the return
> statement.
Please add a "dma-buf: heaps: " prefix to the subject line, apart from that
looks good to me.
Regards,
Christian.
>
> Signed-off-by: FengWei
> -
From: Mario Limonciello
[ Upstream commit 4afacc9948e1f8fdbca401d259ae65ad93d298c0 ]
When userspace invokes S4 the flow is:
1) amdgpu_pmops_prepare()
2) amdgpu_pmops_freeze()
3) Create hibernation image
4) amdgpu_pmops_thaw()
5) Write out image to disk
6) Turn off system
Then on resume amdgpu_
On Tue, Mar 11, 2025 at 06:37:13PM -0700, Dan Williams wrote:
> > There is a use case for using TDISP and getting devices up into an
> > ecrypted/attested state on pure bare metal without any KVM, VFIO
> > should work in that use case too.
>
> Are you sure you are not confusing the use case for n
From: Rodrigo Vivi
[ Upstream commit c605acb53f449f6289f042790307d7dc9e62d03d ]
In a rare situation of thermal limit during resume, GuC can
be slow and run into delays like this:
xe :00:02.0: [drm] GT1: excessive init time: 667ms! \
[status = 0x8002F034, timeouts = 0]
xe 00
From: Mario Limonciello
[ Upstream commit 4afacc9948e1f8fdbca401d259ae65ad93d298c0 ]
When userspace invokes S4 the flow is:
1) amdgpu_pmops_prepare()
2) amdgpu_pmops_freeze()
3) Create hibernation image
4) amdgpu_pmops_thaw()
5) Write out image to disk
6) Turn off system
Then on resume amdgpu_
On Mon, Mar 17, 2025 at 02:37:03PM +, David Turner wrote:
> Hi Maxime,
>
> Thanks for the reviews.
>
> On Fri, 14 Mar 2025 at 13:33, Maxime Ripard wrote:
> > I also think that it's wrong, it's not just about audio. It just needs
> > to be called as part of detect no matter what the status is
Hi,
On Sat, Mar 15, 2025 at 9:50 PM Tejas Vipin wrote:
>
> @@ -62,93 +62,66 @@ static void s6d7aa0_reset(struct s6d7aa0 *ctx)
> msleep(50);
> }
>
> -static int s6d7aa0_lock(struct s6d7aa0 *ctx, bool lock)
> +static void s6d7aa0_lock(struct s6d7aa0 *ctx, struct mipi_dsi_multi_context
> *
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Zack Rusin
Cc: Broadcom
On Mon, Mar 17, 2025 at 9:22 AM Thomas Zimmermann wrote:
>
> Avoid dereferencing struct drm_gem_object.import_attach for the
> imported dma-buf. The dma_buf field in the GEM object instance refers
> to the same buffer. Prepares to make import_attach optional.
>
> Signed-off-by: Thomas Zimmermann
On Mon, Mar 17, 2025 at 9:22 AM Thomas Zimmermann wrote:
>
> Instead of testing import_attach for imported GEM buffers, invoke
> drm_gem_is_imported() to do the test. The helper tests the dma_buf
> itself while import_attach is just an artifact of the import. Prepares
> to make import_attach optio
On Fri, Mar 14, 2025 at 10:20 AM Doug Anderson
wrote:
> Hi,
>
> On Thu, Mar 13, 2025 at 9:47 PM Anusha Srivatsa
> wrote:
> >
> > @@ -181,24 +162,15 @@ static int r63353_panel_prepare(struct drm_panel
> *panel)
> > static int r63353_panel_deactivate(struct r63353_panel *rpanel)
> > {
> >
> where leaving the dangling devices isn't really a problem. For
> instance: if somehow you fail to link up to a MIPI source then it
> feels OK that the GPIO / PWM and DP-AUX bus parts of the bridge
> continue to dangle. You can still use them, right?
Ok, convinced. Thanks for the input. Will sen
Remove `impl Display for CStr` in preparation for replacing `CStr` with
`core::ffi::CStr` which doesn't impl `Display`. Add `CStr::display`
returning a helper struct to replace the lost functionality; this
matches the APIs exposed by `std::ffi::OSstr` and `std::path::Path` for
printing non-Unicode
Hi,
On Sat, Mar 15, 2025 at 2:00 AM Wolfram Sang
wrote:
>
> Hi Doug,
>
> > Seems reasonable to me. I would probably put an error message in this
> > case, though? I don't think regmap_read() necessarily prints an error
> > so it would just be a mysterious failure for why things didn't probe,
> >
On 17/03/2025 10:48, Barnabás Czémán wrote:
>
>
> On March 17, 2025 10:21:50 AM GMT+01:00, Krzysztof Kozlowski
> wrote:
>> On Sat, Mar 15, 2025 at 03:57:37PM +0100, Barnabás Czémán wrote:
>>> From: Adam Skladowski
>>>
>>> Adreno 505 (MSM8937), Adreno 506(MSM8953) and Adreno 510(MSM8976)
>>> re
C-String literals were added in Rust 1.77. Replace instances of
`kernel::c_str!` with C-String literals where possible and rename
`kernel::c_str!` to `c_str_avoid_literals` to clarify its intended use.
Closes: https://github.com/Rust-for-Linux/linux/issues/1075
Signed-off-by: Tamir Duberstein
---
Clean up references to `kernel::str::CStr`.
Signed-off-by: Tamir Duberstein
---
drivers/gpu/drm/drm_panic_qr.rs | 4 ++--
drivers/net/phy/ax88796b_rust.rs | 1 +
drivers/net/phy/qt2025.rs| 1 +
rust/kernel/device.rs| 3 +--
rust/kernel/driver.rs| 4 ++--
rus
This picks up from Michal Rostecki's work[0]. Per Michal's guidance I
have omitted Co-authored tags, as the end result is quite different.
Link:
https://lore.kernel.org/rust-for-linux/20240819153656.28807-2-vadorov...@protonmail.com/t/#u
[0]
Closes: https://github.com/Rust-for-Linux/linux/issues
Hi all,
This is a new iteration on Panfrost support for AARCH64_4K page table
format. The main reason behind this patchset is that MediaTek MT8188 SoC
(ARM Mali-G57 MC3 GPU) constantly faults due to the actual Panfrost cache
configuration.
Currently, Panfrost only supports MMU configuration in "L
On 17/03/2025 10:57, Barnabás Czémán wrote:
>
>
> On March 17, 2025 10:17:46 AM GMT+01:00, Krzysztof Kozlowski
> wrote:
>> On Sat, Mar 15, 2025 at 03:57:35PM +0100, Barnabás Czémán wrote:
>>> Add device tree bindings for the global clock controller on Qualcomm
>>> MSM8937 platform.
>>>
>>> Sign
On Mon, Mar 17, 2025 at 6:12 AM Neil Armstrong
wrote:
> On 13/03/2025 01:54, Anusha Srivatsa wrote:
> > Start using the new helper that does the refcounted
> > allocations.
> >
> > Signed-off-by: Anusha Srivatsa
> > ---
> > drivers/gpu/drm/panel/panel-simple.c | 4 +---
> > 1 file changed, 1
Hi Thomas,
Am Montag, dem 17.03.2025 um 14:06 +0100 schrieb Thomas Zimmermann:
> Avoid dereferencing struct drm_gem_object.import_attach for the
> imported dma-buf. The dma_buf field in the GEM object instance refers
> to the same buffer. Prepares to make import_attach optional.
>
Both etnaviv pa
From: Rob Clark
IB_SIZE is only b0..b19. Starting with a6xx gen3, additional fields
were added above the IB_SIZE. Accidentially setting them can cause
badness. Fix this by properly defining the CP_INDIRECT_BUFFER packet
and using the generated builder macro to ensure unintended bits are not
se
On Fri, 14 Mar 2025 19:10:41 +0100
Maxime Ripard wrote:
> On Fri, Mar 14, 2025 at 11:31:19AM +0100, Luca Ceresoli wrote:
> > drm_bridge_chain_get_first_bridge() returns a bridge pointer that the
> > caller could hold for a long time. Increment the refcount of the returned
> > bridge and document
Hello Maxime,
thanks for the very prompt feedback!
On Fri, 14 Mar 2025 19:04:40 +0100
Maxime Ripard wrote:
> On Fri, Mar 14, 2025 at 11:31:15AM +0100, Luca Ceresoli wrote:
> > DRM bridges are currently considered as a fixed element of a DRM card, and
> > thus their lifetime is assumed to extend
Hello Maxime,
On Fri, 14 Mar 2025 19:21:01 +0100
Maxime Ripard wrote:
> Hi,
>
> On Fri, Mar 14, 2025 at 11:31:13AM +0100, Luca Ceresoli wrote:
> > This series improves the way DRM bridges are allocated and
> > initialized and makes them reference-counted. The goal of reference
> > counting is t
MediaTek MT8188 SoC has an ARM Mali-G57 MC3 GPU (Valhall-JM), which
constantly faults with the current panfrost support.
For instance, running `glmark2-es2-drm` benchmark test:
```
[ 79.617461] panfrost 1300.gpu: js fault, js=1, status=JOB_BUS_FAULT,
head=0xaadc380, tail=0xaadc380
[ 80.11
MediaTek MT8192 SoC has an ARM Mali-G57 MC5 GPU (Valhall-JM). Now that
Panfrost supports AARCH64_4K page table format, let's enable it on this
SoC.
Running glmark2-es2-drm [0] benchmark, reported the same performance
score on both modes Mali LPAE (LEGACY) vs. AARCH64_4K, before and after
this comm
Currently, Panfrost only supports MMU configuration in "LEGACY" (as
Bifrost calls it) mode, a (modified) version of LPAE "Large Physical
Address Extension", which in Linux we've called "mali_lpae".
This commit adds support for conditionally enabling AARCH64_4K page
table format. To achieve that, a
Panfrost does not support uncached mappings, so flag them properly. Also
flag the pages that are mapped as response to a page fault as cached.
Signed-off-by: Boris Brezillon
Signed-off-by: Ariel D'Alessandro
Reviewed-by: AngeloGioacchino Del Regno
Reviewed-by: Steven Price
---
drivers/gpu/dr
Set this feature flag on all Mali Bifrost platforms as the MMU supports
AARCH64 4K page table format.
Signed-off-by: Ariel D'Alessandro
Reviewed-by: Boris Brezillon
Reviewed-by: Steven Price
Reviewed-by: AngeloGioacchino Del Regno
---
drivers/gpu/drm/panfrost/panfrost_features.h | 3 +++
1 f
As done in panthor, define and use these GPU_MMU_FEATURES_* macros,
which makes code easier to read and reuse.
Signed-off-by: Ariel D'Alessandro
Reviewed-by: Boris Brezillon
Reviewed-by: Steven Price
Reviewed-by: AngeloGioacchino Del Regno
---
drivers/gpu/drm/panfrost/panfrost_mmu.c | 6 +++
Don't populate the const read-only array dflt_rules on the stack at run
time, instead make it static.
Signed-off-by: Colin Ian King
---
drivers/gpu/drm/i915/intel_memory_region.c| 2 +-
drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c | 2 +-
2 files changed, 2 insertions(+), 2 deletion
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Inki Dae
Cc: Seung-Woo
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
---
drivers/gpu/drm/gud/gud
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Russell King
---
drive
Hi Maxime,
Thanks for the reviews.
On Fri, 14 Mar 2025 at 13:33, Maxime Ripard wrote:
> I also think that it's wrong, it's not just about audio. It just needs
> to be called as part of detect no matter what the status is. Audio is
> one thing that it will possibly handle depending on the driver,
Hi,
On Sat, Mar 15, 2025 at 1:17 PM Wolfram Sang
wrote:
>
> Do not create a custom directory in debugfs-root, but use the
> debugfs_init callback to create a custom directory at the given place
> for the bridge. The new directory layout looks like this on a Renesas
> GrayHawk-Single with a R-Car
Boris,
On 3/17/25 10:44 AM, Boris Brezillon wrote:
On Mon, 17 Mar 2025 09:40:42 -0300
Ariel D'Alessandro wrote:
+static int panfrost_mmu_cfg_init(struct panfrost_mmu *mmu,
+ enum io_pgtable_fmt fmt)
+{
+ struct panfrost_device *pfdev = mmu->pfdev;
+
+
(Cc: dri-devel@lists.freedesktop.org, Boris)
Hi Himal,
Please make sure to copy in dri-devel for such patches.
On Fri, Mar 14, 2025 at 01:32:10PM +0530, Himal Prasad Ghimiray wrote:
> Introduce MADVISE operations that do not unmap the GPU VMA. These
> operations split VMAs if the start or end ad
On Mon, Mar 17, 2025 at 02:06:44PM +0100, Thomas Zimmermann wrote:
> Instead of testing import_attach for imported GEM buffers, invoke
> drm_gem_is_imported() to do the test. The helper tests the dma_buf
> itself while import_attach is just an artifact of the import. Prepares
> to make import_attac
On Mon, Mar 17, 2025 at 02:06:47PM +0100, Thomas Zimmermann wrote:
> Instead of testing import_attach for imported GEM buffers, invoke
> drm_gem_is_imported() to do the test. The helper tests the dma_buf
> itself while import_attach is just an artifact of the import. Prepares
> to make import_attac
Hi
Am 13.03.25 um 10:42 schrieb Jocelyn Falempe:
Some drivers like virtio-gpu, don't map the scanout buffer in the
kernel. Calling vmap() in a panic handler is not safe, and writing an
atomic_vmap() API is more complex than expected [1].
So instead, pass the array of pages of the scanout buffer
On Mon, 17 Mar 2025 09:40:42 -0300
Ariel D'Alessandro wrote:
> +static int panfrost_mmu_cfg_init(struct panfrost_mmu *mmu,
> + enum io_pgtable_fmt fmt)
> +{
> + struct panfrost_device *pfdev = mmu->pfdev;
> +
> + switch (fmt) {
> + case ARM_64_LPAE_S1:
>
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Sandy Huang
Cc: "Heiko
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Oleksandr Andrushchenko
Avoid dereferencing struct drm_gem_object.import_attach for the
imported dma-buf. The dma_buf field in the GEM object instance refers
to the same buffer. Prepares to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: Zack Rusin
Cc: Broadcom internal kernel review list
---
driver
Instead of testing import_attach for imported GEM buffers, invoke
drm_gem_is_imported() to do the test. The helper tests the dma_buf
itself while import_attach is just an artifact of the import. Prepares
to make import_attach optional.
Signed-off-by: Thomas Zimmermann
Cc: David Airlie
Cc: Gerd H
1 - 100 of 160 matches
Mail list logo
