On Thu, Mar 27, 2025 at 4:48 PM Mario Limonciello wrote:
>
> From: Mario Limonciello
>
> The errors for power consumption in amdgpu_acpi_is_s0ix_active() are
> under device scope. As they're drm errors, adjust to drm scope.
Is there an advantage to drm vs dev scope? I guess just consistency
wit
On Thu, Jan 9, 2025 at 12:07 PM Felix Kuehling wrote:
>
> From: Christian König
>
> Try pinning into VRAM to allow P2P with RDMA NICs without ODP
> support if all attachments can do P2P. If any attachment can't do
> P2P just pin into GTT instead.
>
> Signed-off-by: Christian König
> Signed-off-b
Applied. Thanks!
Alex
On Fri, Mar 28, 2025 at 9:03 AM Christian König
wrote:
>
> Am 27.03.25 um 20:51 schrieb Natalie Vock:
> > From: Maarten Lankhorst
> >
> > Similar to xe, enable some simple management of VRAM only.
> >
> > Co-developed-by: Maxime Ripard
> > Signed-off-by: Maxime Ripard
>
On Thu, Mar 27, 2025 at 4:39 PM Mario Limonciello wrote:
>
> From: Mario Limonciello
>
> Messages emitted from amdgpu_acpi are not device specific nor DRM
> specific, but rather operate on ACPI handles. Adjust the messages
> to use ACPI macros instead.
>
> Signed-off-by: Mario Limonciello
Acke
This patch series adds support for CRIU checkpointing of processes that
share memory with the amdgpu dmabuf interface.
In this v3, the drm interfaces have been changed from creating buffer objects
with specified gem handles to changign the gem handle of an existing buffer
object.
In the accompany
On Fri, Mar 28, 2025 at 1:10 AM Alexandre Demers
wrote:
>
> They will be used later when switching away from sid.h/si_enums.h.
>
> Signed-off-by: Alexandre Demers
Applied the series. Thanks!
> ---
> .../drm/amd/include/asic_reg/smu/smu_6_0_d.h | 44
> .../include/asic_reg/smu/smu_6_0_s
On 2025-03-26 21:40, Huacai Chen wrote:
Hi, Alex,
On Thu, Mar 27, 2025 at 8:10 AM Alex Hung wrote:
The following error messages showed up on an APU and a dGPU during testing.
<3> [100.231411] BUG: sleeping function called from invalid context at
include/linux/sched/mm.h:321
<3> [100.23141
From: Boyuan Zhang
Enabling VCN FW workaround for drm key injection through shared
memory for vcn 4_0_5
Signed-off-by: Boyuan Zhang
---
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
b/drivers/gpu/drm/amd
From: Kenneth Feng
pause the workload settign in dm when doing idle optimization
Signed-off-by: Kenneth Feng
Signed-off-by: Alex Deucher
---
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c| 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/am
To be used for display idle optimizations when
we want to pause non-default profiles.
Signed-off-by: Alex Deucher
---
.../gpu/drm/amd/include/kgd_pp_interface.h| 1 +
drivers/gpu/drm/amd/pm/amdgpu_dpm.c | 19 +++
drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h | 2 +
On 3/28/2025 13:10, Gergo Koteles wrote:
Hi Mario,
Thanks for the suggestions!
On Fri, 2025-03-28 at 08:42 -0500, Mario Limonciello wrote:
On 3/28/2025 06:12, Rafael J. Wysocki wrote:
CC: Hans
On Fri, Mar 28, 2025 at 3:51 AM Gergo Koteles wrote:
Some Lenovo laptops incorrectly return EDID
CRIU restore of drm buffer objects requires the ability to create
a buffer object with a specific gem handle.
Add function drm_prime_change_buf_handle to change the gem handle
of a specified buffer object.
This function calls drm_gem_object_release_handle.
drm_gem_object_release_handle is an inte
The kfd CRIU checkpoint ioctl would return an error if trying
to checkpoint a process with no kfd buffer objects.
This is a normal case and should not be an error.
Signed-off-by: David Francis
---
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
di
A nitpick below. With that addressed,
Reviewed-by: Amber Lin
Regards,
Amber
On 3/27/25 13:47, Apurv Mishra wrote:
remove workaround code for the early engineering samples
GC v9.4.3 SOCs with revID 0 - GFX 940 & 941 - from driver
Remove "- GFX 940 & 941 - from driver" and simply "Remove workar
On 3/27/2025 23:05, Lazar, Lijo wrote:
On 3/28/2025 2:08 AM, Mario Limonciello wrote:
From: Mario Limonciello
checkpatch.pl complains about unnecessary error messages for failing
to allocate memory. These aren't needed when the return code is -ENOMEM.
It's not about the error code. It conv
[AMD Official Use Only - AMD Internal Distribution Only]
> -Original Message-
> From: Skvortsov, Victor
> Sent: Thursday, March 27, 2025 11:58 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Rehman, Ahmad ; Chan, Hing Pong
> ; Luo, Zhigang ; Skvortsov,
> Victor
> Subject: [PATCH] drm/amdgpu
On 3/28/2025 8:28 PM, Alex Deucher wrote:
On Thu, Mar 27, 2025 at 9:50 AM Christian König
wrote:
Am 27.03.25 um 10:37 schrieb SRINIVASAN SHANMUGAM:
On 3/27/2025 2:54 PM, Christian König wrote:
Over all this change doesn't seem to make much sense to me.
Why exactly is isolation->spearhead n
On Thu, Mar 27, 2025 at 9:50 AM Christian König
wrote:
>
> Am 27.03.25 um 10:37 schrieb SRINIVASAN SHANMUGAM:
>
> On 3/27/2025 2:54 PM, Christian König wrote:
>
> Over all this change doesn't seem to make much sense to me.
>
> Why exactly is isolation->spearhead not pointing to the dummy kernel jo
On Thu, Mar 27, 2025 at 10:37 PM Feng, Kenneth wrote:
>
> [AMD Official Use Only - AMD Internal Distribution Only]
>
> -Original Message-
> From: Alex Deucher
> Sent: Thursday, March 27, 2025 10:09 PM
> To: Feng, Kenneth
> Cc: amd-gfx@lists.freedesktop.org; Wang, Yang(Kevin)
> ; Deucher
On 3/28/2025 06:12, Rafael J. Wysocki wrote:
CC: Hans
On Fri, Mar 28, 2025 at 3:51 AM Gergo Koteles wrote:
Some Lenovo laptops incorrectly return EDID as
buffer in ACPI package (instead of just a buffer)
when calling _DDC.
Calling _DDC generates this ACPI Warning:
ACPI Warning: \_SB.PCI0.GP1
[Public]
Reviewed-by: Kent Russell
> -Original Message-
> From: Cornwall, Jay
> Sent: Friday, March 28, 2025 12:05 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Cornwall, Jay ; Russell, Kent
>
> Subject: [PATCH] Revert "drm/amdgpu: Increase KIQ invalidate_tlbs timeout"
>
> This revert
Am 27.03.25 um 20:51 schrieb Natalie Vock:
> From: Maarten Lankhorst
>
> Similar to xe, enable some simple management of VRAM only.
>
> Co-developed-by: Maxime Ripard
> Signed-off-by: Maxime Ripard
> Signed-off-by: Maarten Lankhorst
Reviewed-by: Christian König
> ---
>
> Resending this one s
On Tue, Mar 25, 2025 at 11:18:45AM -0600, Rodrigo Siqueira wrote:
> +.. kernel-figure:: pipe_and_queue_abstraction.svg
> +
> +In the central part of this figure, you can see two elements, one called
Did you mean hardware block?
> +**Pipe** and another named **Queues**; it is important to highlight
Split amdgpu_device_gpu_recover into the following stages:
halt activities,asic reset,schedule resume and amdgpu resume.
The reason is that the subsequent addition of dpc recover
code will have a high similarity with gpu reset
Signed-off-by: Ce Sun
Reviewed-by: Hawking Zhang
---
drivers/gpu/drm
Don't fetch it again if we already have it. It seems the
registers don't reliably have the value at resume in some
cases.
Fixes: 785f0f9fe742 ("drm/amdgpu: Add mes v12_0 ip block support (v4)")
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 21 -
1
Don't fetch it again if we already have it. It seems the
registers don't reliably have the value at resume in some
cases.
Fixes: 028c3fb37e70 ("drm/amdgpu/mes11: initiate mes v11 support")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4083
Signed-off-by: Alex Deucher
---
drivers/gpu/d
++ dri-devel
On 28-03-2025 15:57, Aravind Iddamsetty wrote:
> Hi,
>
> Based on the discussions around using Netlink for RAS purposes, as
> summarized in this blog post [1] by Dave Airlie. I had proposed a series
> regarding RAS infrastructure in DRM [2].
>
> I came across your work, which appears
The MES queue deactivation and active status are already set in
mes_userq_unmap|map(), so the caller needn't set the queue_active
bit again.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/
The MES queue deactivation and active status are already set in
mes_userq_unmap|map(), so the caller needn't set the queue_active
bit again.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/
[Public]
There requires a similar change on the userq resume, and a new patch will
include the change.
Regards,
Prike
> -Original Message-
> From: Liang, Prike
> Sent: Friday, March 28, 2025 6:46 PM
> To: amd-gfx@lists.freedesktop.org
> Cc: Deucher, Alexander ; Liang, Prike
>
>
CC: Hans
On Fri, Mar 28, 2025 at 3:51 AM Gergo Koteles wrote:
>
> Some Lenovo laptops incorrectly return EDID as
> buffer in ACPI package (instead of just a buffer)
> when calling _DDC.
>
> Calling _DDC generates this ACPI Warning:
> ACPI Warning: \_SB.PCI0.GP17.VGA.LCD._DDC: Return type mismatch
The MES queue deactivation status is already set in mes_userq_unmap(),
so the caller needn't set the queue_active bit again.
Signed-off-by: Prike Liang
---
drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu
[Public]
> -Original Message-
> From: Emily Deng
> Sent: Friday, March 28, 2025 6:19 AM
> To: amd-gfx@lists.freedesktop.org; Kim, Jonathan
> Cc: Deng, Emily
> Subject: [PATCH] drm/amdkfd: sriov doesn't support per queue reset
>
> Disable per queue reset for sriov.
>
> Signed-off-by: Emi
Hi,
Based on the discussions around using Netlink for RAS purposes, as
summarized in this blog post [1] by Dave Airlie. I had proposed a series
regarding RAS infrastructure in DRM [2].
I came across your work, which appears to address related areas and I'm
particularly interested in understanding
Disable per queue reset for sriov.
Signed-off-by: Emily Deng
---
drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e477d7509646..2c4711c67d8a
[AMD Official Use Only - AMD Internal Distribution Only]
Reviewed-by: Hawking Zhang
Regards,
Hawking
-Original Message-
From: Lazar, Lijo
Sent: Wednesday, March 26, 2025 16:03
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking ; Kamal, Asad ;
Zhou1, Tao
Subject: [PATCH] drm/amdgpu:
On 3/26/2025 1:33 PM, Lijo Lazar wrote:
> If RAS header read from EEPROM is corrupted, it could result in trying
> to allocate huge memory for reading the records. Add some validation to
> header fields.
>
> Signed-off-by: Lijo Lazar
> ---
> .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c| 22 +
Some Lenovo laptops incorrectly return EDID as
buffer in ACPI package (instead of just a buffer)
when calling _DDC.
Calling _DDC generates this ACPI Warning:
ACPI Warning: \_SB.PCI0.GP17.VGA.LCD._DDC: Return type mismatch - \
found Package, expected Integer/Buffer (20240827/nspredef-254)
Use the
Hi Huacai,
Tested successfully with `glmark2` on both x86 and Loongson platforms, using
AMD Radeon RX 9070 XT.
---
**Intel i5-10400F Platform:**
- **Board / CPU**: Intel i5-10400F
- **Firmware Vendor**: American Megatrends International, LLC
- **Kernel**:
https://lore.kernel.org/all/202503270
From: Alex Hung
[ Upstream commit ea79068d4073bf303f8203f2625af7d9185a1bc6 ]
[WHAT & HOW]
A denominator cannot be 0, and is checked before used.
This fixes 2 DIVIDE_BY_ZERO issues reported by Coverity.
Reviewed-by: Harry Wentland
Signed-off-by: Jerry Zuo
Signed-off-by: Alex Hung
Tested-by:
From: Maarten Lankhorst
Similar to xe, enable some simple management of VRAM only.
Co-developed-by: Maxime Ripard
Signed-off-by: Maxime Ripard
Signed-off-by: Maarten Lankhorst
---
Resending this one since it got lost in the initial dmem cgroup
submission. Probably needs to be merged through
old_bits arg is currently a pointer to the input bits, but this could
allow inadvertent changes to the input by the fn. Disallow this.
And constify new_bits while here.
Signed-off-by: Jim Cromie
Reviewed-by: Louis Chauvet
---
lib/dynamic_debug.c | 21 +++--
1 file changed, 11 i
42 matches
Mail list logo
