On 9/10/25 4:27 PM, Matthew Schwartz wrote:
On clients that utilize AMD_PRIVATE_COLOR properties for HDR support,
brightness sliders can include a hardware controlled portion and a
gamma-based portion. This is the case on the Steam Deck OLED when using
gamescope with Steam as a client.
When a us
Replace the previous O(N^2) implementation of remove_duplicates() with
a O(N) version using a fast/slow pointer approach. The new version
keeps only the first occurrence of each element and compacts the array
in place, improving efficiency without changing functionality.
Signed-off-by: Kuan-Wei Ch
[Public]
Regards,
Prike
> -Original Message-
> From: Chen, Xiaogang
> Sent: Thursday, September 11, 2025 6:01 AM
> To: Liang, Prike ; amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander ; Kuehling, Felix
>
> Subject: Re: [PATCH] drm/amdkfd: pass the PD BOs directly for PA
>
>
>
On Tue, Sep 9, 2025 at 9:16 PM Mario Limonciello (AMD)
wrote:
>
> Suspend resume actions will check the state of the device and whether
> bus PM should be skipped. These same actions make sense during hibernation
> image restore.
Not really (see below).
They kind of would have made sense in the
It's probably better to use 0x3fff00 to match Mesa and PAL. There is no
benefit in using a different limit, not even a perf benefit, and it's
better to be consistent with all UMDs.
Marek
On Wed, Sep 10, 2025 at 7:54 AM Christian König
wrote:
> On 10.09.25 11:34, Timur Kristóf wrote:
> > On Wed,
On Mon, Sep 08, 2025 at 07:35:08PM +0200, Christian König wrote:
> On 08.09.25 19:05, Alex Hung wrote:
> >
> >
> > On 8/24/25 12:23, Kuan-Wei Chiu wrote:
> >> Replace the custom bubble sort used for sorting reserved time
> >> candidates in with the kernel's standard sort() helper. The previous
>
On 9/4/2025 9:45 PM, David Yat Sin wrote:
Add look-up for the BO by address. This is required for memory
allocations where the virtual address are reserved outside of
KFD and then later mapped to GPU memory.
Signed-off-by: David Yat Sin
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 41 +
On 9/9/2025 8:45 PM, Liang, Prike wrote:
[Public]
Ping..
Regards,
Prike
-Original Message-
From: Liang, Prike
Sent: Monday, September 8, 2025 10:40 AM
To: amd-gfx@lists.freedesktop.org
Cc: Deucher, Alexander ; Kuehling, Felix
; Liang, Prike
Subject: [PATCH] drm/amdkfd: pass
[AMD Official Use Only - AMD Internal Distribution Only]
Acked-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Asad Kamal
Sent: Friday, September 5, 2025 00:43
To: amd-gfx@lists.freedesktop.org; Lazar, Lijo
Cc: Zhang, Hawking ; Ma, Le ; Zhang,
Morris ;
On 2025-09-09 16:43, James Zhu wrote:
if destination is on system ram. migrate_vma_pages can fail if a CPU
thread faults on the same page. However, the page table is locked and
only one of the new pages will be inserted. The device driver will see
that the MIGRATE_PFN_MIGRATE bit is cleared if
On 2025-06-18 11:19, Melissa Wen wrote:
> Add Linux opaque object to dc_sink for storing EDID data cross driver,
> drm_edid. Also include the Linux call to free this object, the
> drm_edid_free()
>
> Signed-off-by: Melissa Wen
>
> ---
>
> v3:
> - remove uneccessary include (jani)
>
> v5:
>
On 2025-09-09 16:42, James Zhu wrote:
to get migration pages. dst MIGRATE_PFN_VALID bit and src
MIGRATE_PFN_MIGRATE bit should always be set when migration success.
cpage includes src MIGRATE_PFN_MIGRATE bit set and MIGRATE_PFN_VALID
bit unset pages for both RAM and VRAM when memory is only al
On 2025-06-18 11:19, Melissa Wen wrote:
> From: Rodrigo Siqueira
>
> As part of the effort of stopping using raw edid, this commit move the
> copy of the edid in DC to a dedicated function that will allow the usage
> of drm_edid in the next steps.
>
> Signed-off-by: Rodrigo Siqueira
> Co-dev
On 9/10/25 12:11 PM, Bjorn Helgaas wrote:
On Wed, Sep 10, 2025 at 11:52:00AM -0500, Mario Limonciello wrote:
On 9/10/25 10:06 AM, Bjorn Helgaas wrote:
On Tue, Sep 09, 2025 at 02:16:12PM -0500, Mario Limonciello (AMD) wrote:
PCI devices can be configured as wakeup sources from low power states.
[AMD Official Use Only - AMD Internal Distribution Only]
The patch is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Kamal, Asad
Sent: Friday, September 5, 2025 00:43
To: amd-gfx@lists.freedesktop.org; Lazar, Lijo
Cc: Zhang, Hawking ; Ma, Le ; Zhang,
Morris ; Ka
[AMD Official Use Only - AMD Internal Distribution Only]
The patch is
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: amd-gfx On Behalf Of Asad Kamal
Sent: Friday, September 5, 2025 00:43
To: amd-gfx@lists.freedesktop.org; Lazar, Lijo
Cc: Zhang, Hawking ; Ma, Le ;
On Tue, Sep 09, 2025 at 03:36:44PM +0530, Suraj Kandpal wrote:
> Now with drm_writeback_connector moved to drm_connector it makes
> more sense use drm_connector as an argument rather than drm_connector.
than drm_writeback_connector
> The writeback connector can easily be derived from drm_connecto
On Tue, Sep 09, 2025 at 03:36:43PM +0530, Suraj Kandpal wrote:
> Some drivers cannot work with the current design where the connector
> is embedded within the drm_writeback_connector such as Intel and
> some drivers that can get it working end up adding a lot of checks
> all around the code to chec
On Wed, Sep 10, 2025 at 07:32:51AM +0200, Christophe JAILLET wrote:
> Le 10/09/2025 à 05:47, Dmitry Baryshkov a écrit :
> > On Mon, Sep 08, 2025 at 11:38:44PM +0200, Christophe JAILLET wrote:
> > > Le 08/09/2025 à 23:26, Dmitry Baryshkov a écrit :
> > > > On Mon, Sep 08, 2025 at 11:09:07PM +0200, C
The I2C init for SMU_V13 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that SMU_V13 init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load
The I2C init for Sienna Cichlid uses i2c_add_adapter() and
i2c_del_adapter(), this commit replaces the use of these two functions
with devm_i2c_add_adapter(). Notice that Sienna Cichlid init initializes
multiple I2C buses in a loop; if something goes wrong, the previous
adapters are removed, and th
On Wed, Sep 10, 2025 at 8:19 PM Mario Limonciello wrote:
>
> On 9/10/25 1:11 PM, Rafael J. Wysocki wrote:
> > Hi Mario,
> >
> > On Tue, Sep 9, 2025 at 9:16 PM Mario Limonciello (AMD)
> > wrote:
> >>
> >> A variety of issues both in function and in power consumption have been
> >> raised as a resu
On Wed, Sep 10, 2025 at 7:48 PM Mario Limonciello wrote:
>
> On 9/10/25 8:58 AM, Rafael J. Wysocki wrote:
> > On Tue, Sep 9, 2025 at 9:16 PM Mario Limonciello (AMD)
> > wrote:
> >>
> >> PMSG_POWEROFF will be used for the PM core to allow differentiating between
> >> a hibernation or shutdown sequ
On 9/10/25 1:11 PM, Rafael J. Wysocki wrote:
Hi Mario,
On Tue, Sep 9, 2025 at 9:16 PM Mario Limonciello (AMD)
wrote:
A variety of issues both in function and in power consumption have been
raised as a result of devices not being put into a low power state when
the system is powered off.
Ther
Hi Mario,
On Tue, Sep 9, 2025 at 9:16 PM Mario Limonciello (AMD)
wrote:
>
> A variety of issues both in function and in power consumption have been
> raised as a result of devices not being put into a low power state when
> the system is powered off.
>
> There have been some localized changes[1]
The I2C init for V13_0_6 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that V13_0_6 init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load
The I2C init for Navi10 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that Navi10 init initializes multiple I2C
buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is
Instead of using i2c_add_adapter() and i2c_del_adapter() in the SMU V11,
use devm_i2c_add_adapter() to simplify the code path.
Signed-off-by: Rodrigo Siqueira
---
drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/am
This commit replaces i2c_add_adapter() with devm_i2c_add_adapter() and
removes part of the cleanup logic since the new function handles the i2c
removal.
Signed-off-by: Rodrigo Siqueira
---
drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c | 18 +++---
1 file changed, 3 insertions(+), 15 deleti
This commit replaces the utilization of i2c_add/del_adapter() with
devm_i2c_add_adapter() to reduce the amount of boilerplate. Using
devm_i2c_add_adapter() has the advantage of removing the manual
manipulation of the I2C adapter.
Suggested-by: Robert Beckett
Signed-off-by: Rodrigo Siqueira
---
On 9/10/25 9:46 AM, Bjorn Helgaas wrote:
On Tue, Sep 09, 2025 at 02:16:13PM -0500, Mario Limonciello (AMD) wrote:
In order to unify suspend and hibernate codepaths without code duplication
the common code should be in common helpers. Move it from
pci_pm_suspend_noirq() into a helper. No intend
On 9/10/25 10:06 AM, Bjorn Helgaas wrote:
On Tue, Sep 09, 2025 at 02:16:12PM -0500, Mario Limonciello (AMD) wrote:
PCI devices can be configured as wakeup sources from low power states.
However, when the system is halting or powering off such wakeups are
not expected and may lead to spurious beh
Hi Dave, Simona,
Fixes for 6.17.
The following changes since commit 8b556ddeee8da9420699ce221b6267f395e7d72b:
Merge tag 'amd-drm-fixes-6.17-2025-09-03' of
https://gitlab.freedesktop.org/agd5f/linux into drm-fixes (2025-09-05 08:06:34
+1000)
are available in the Git repository at:
https:/
On 9/10/25 10:18 AM, Bjorn Helgaas wrote:
On Tue, Sep 09, 2025 at 02:16:18PM -0500, Mario Limonciello (AMD) wrote:
When the system is powered off the kernel will call device_shutdown()
which will issue callbacks into PCI core to wake up a device and call
it's shutdown() callback. This will leav
On Tue, Sep 09, 2025 at 02:16:18PM -0500, Mario Limonciello (AMD) wrote:
> When the system is powered off the kernel will call device_shutdown()
> which will issue callbacks into PCI core to wake up a device and call
> it's shutdown() callback. This will leave devices in ACPI D0 which can
> cause
On 10.09.25 16:52, Marek Olšák wrote:
> It's probably better to use 0x3fff00 to match Mesa and PAL. There is no
> benefit in using a different limit, not even a perf benefit, and it's better
> to be consistent with all UMDs.
Unification with Mesa is certainly a valid argument, but just using an
I added the comment into Mesa that 0x3fff00 is the limit. I did research on
that bug separately from PAL, but I don't remember the details.
There is no performance to gain here. It's only about consistency and clear
communication to the public what the recommended SDMA programming is.
Marek
On W
On Tue, Sep 09, 2025 at 02:16:13PM -0500, Mario Limonciello (AMD) wrote:
> In order to unify suspend and hibernate codepaths without code duplication
> the common code should be in common helpers. Move it from
> pci_pm_suspend_noirq() into a helper. No intended functional changes.
>
> Tested-by:
On 10.09.25 14:37, Arunpravin Paneer Selvam wrote:
> Hi Christian,
>
> On 9/9/2025 9:55 PM, Christian König wrote:
>> On 09.09.25 16:05, Peter Zijlstra wrote:
>>> On Tue, Sep 09, 2025 at 02:04:30PM +0200, Christian König wrote:
Hi Arun,
On 09.09.25 11:56, Arunpravin Paneer Selvam wr
On Wed, Sep 10, 2025 at 4:34 AM Lijo Lazar wrote:
>
> Read VRAM vendor information from scratch register for GC v9.4.3 and
> GC v9.5.0 SOCs.
>
> Signed-off-by: Lijo Lazar
Acked-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8
> 1 file changed, 8 insertions(+)
>
> d
[Public]
Regards,
Prike
> -Original Message-
> From: Alex Deucher
> Sent: Tuesday, September 9, 2025 9:14 PM
> To: Liang, Prike
> Cc: amd-gfx@lists.freedesktop.org; Deucher, Alexander
> ; Koenig, Christian
> Subject: Re: [PATCH v10 08/14] drm/amdgpu: add userq object va track hel
On Wed, 2025-09-10 at 10:34 +0200, Christian König wrote:
> On 09.09.25 18:56, Timur Kristóf wrote:
> > > Even when we apply it I think we should drop that, the value the
> > > kernel uses is correct.
> >
> > Hi Christian,
> >
> > The kernel and Mesa disagree on the limits for almost all SDMA
> >
[Public]
Maybe we need a PTE schedule check in the amdgpu_vm_generation() for the userq
only case.
This could be a sperate patch for handling that, for this patch is
Reviewed-by: Prike Liang
Regards,
Prike
> -Original Message-
> From: Liang, Prike
> Sent: Wednesday, September
On 10.09.25 11:34, Timur Kristóf wrote:
> On Wed, 2025-09-10 at 10:34 +0200, Christian König wrote:
>> On 09.09.25 18:56, Timur Kristóf wrote:
Even when we apply it I think we should drop that, the value the
kernel uses is correct.
>>>
>>> Hi Christian,
>>>
>>> The kernel and Mesa disagre
[Public]
Reviewed-by: Prike Liang
Regards,
Prike
> -Original Message-
> From: amd-gfx On Behalf Of Prike Liang
> Sent: Wednesday, September 10, 2025 7:37 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander ; Koenig, Christian
> ; Deucher, Alexander
> ; Koenig, Christi
[Public]
-Original Message-
From: Wang, Yang(Kevin)
Sent: Wednesday, September 10, 2025 8:14 AM
To: Lazar, Lijo ; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking ; Deucher, Alexander
; Kamal, Asad
Subject: RE: [PATCH v3 1/7] drm/amdgpu/pm: Add definition for gpu_metrics v1.9
[AMD Offi
From: Alex Deucher
Add an API to query queue status such as whether the
queue is hung or whether vram is lost.
Reviewed-by: Christian König
Reviewed-by: Sunil Khatri
Signed-off-by: Alex Deucher
---
include/uapi/drm/amdgpu_drm.h | 14 ++
1 file changed, 14 insertions(+)
diff --gi
It requires validating the userq VA whether is mapped before
trying to resume the queue.
Signed-off-by: Prike Liang
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 31 ---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 2 ++
2 files changed, 29 ins
Query the status of the user queue, currently whether
the queue is hung and whether or not VRAM is lost.
v2: Misc cleanups
Reviewed-by: Sunil Khatri
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 35 ++-
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
Track the userq obj for its life time, and reference and
dereference the buffer counter at its creating and destroying
period.
Suggested-by: Alex Deucher
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 38 +--
drivers/gpu/drm/amd/amdgpu/amdgpu_user
Extend the userq state for identifying the
userq invalid cases.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 1 +
2 files changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
b/drivers/gp
On 09/09/2025 14:36, Alex Deucher wrote:
On Tue, Sep 9, 2025 at 4:17 AM Tvrtko Ursulin wrote:
On 08/09/2025 22:15, Thorsten Blum wrote:
Replace kzalloc() followed by copy_from_user() with memdup_user() to
improve and simplify ta_if_load_debugfs_write() and
ta_if_invoke_debugfs_write().
No
On 09.09.25 18:56, Timur Kristóf wrote:
>> Even when we apply it I think we should drop that, the value the
>> kernel uses is correct.
>
> Hi Christian,
>
> The kernel and Mesa disagree on the limits for almost all SDMA
> versions, so it would be nice to actually understand what the limits of
> t
Replace kmalloc() followed by copy_from_user() with memdup_user() to
improve and simplify kfd_criu_restore_queue().
No functional changes intended.
Signed-off-by: Thorsten Blum
---
.../amd/amdkfd/kfd_process_queue_manager.c| 22 +--
1 file changed, 6 insertions(+), 16 deleti
On Mon, 2025-09-08 at 14:18 -0600, Alex Hung wrote:
>
>
> On 8/25/25 02:52, Xi Ruoyao wrote:
> > dml21_map_dc_state_into_dml_display_cfg calls (the call is usually
> > inlined by the compiler) populate_dml21_surface_config_from_plane_state
> > and populate_dml21_plane_config_from_plane_state whic
hi Louis, thx for the nudge.
On Mon, Sep 8, 2025 at 8:38 AM Louis Chauvet
wrote:
> \
>
> Le 03/08/2025 à 05:57, Jim Cromie a écrit :
> > Describe the 3 API macros providing dynamic_debug's classmaps
> >
> > DYNDBG_CLASSMAP_DEFINE - create & export a classmap
> > DYNDBG_CLASSMAP_USE- refer to
Le 19/08/2025 à 22:32, Dmitry Baryshkov a écrit :
Use drmm_plain_encoder_alloc() to allocate simple encoder and
drmm_writeback_connector_init() in order to initialize writeback
connector instance.
Reviewed-by: Louis Chauvet
Reviewed-by: Suraj Kandpal
Reviewed-by: Jessica Zhang
Signed-off-b
Le 08/09/2025 à 23:26, Dmitry Baryshkov a écrit :
On Mon, Sep 08, 2025 at 11:09:07PM +0200, Christophe JAILLET wrote:
Le 19/08/2025 à 22:32, Dmitry Baryshkov a écrit :
Use drmm_plain_encoder_alloc() to allocate simple encoder and
drmm_writeback_connector_init() in order to initialize writeback
On Mon, Sep 08, 2025 at 07:05:17PM +0200, Michel Dänzer wrote:
> These messages are primarily intended for developers, not users
But everybody sees them! And they're flooding the console.
And most people seeing them are users, not developers.
And if those messages are only for developers, they b
On Mon, Sep 08, 2025 at 11:10:30AM -0600, Alex Hung wrote:
>
>
> On 8/24/25 12:23, Kuan-Wei Chiu wrote:
> > Replace the previous O(N^2) implementation of remove_duplicates() in
> > with a O(N) version using a fast/slow pointer approach. The new version
> > keeps only the first occurrence of each
60 matches
Mail list logo
