Am 19.10.21 um 20:14 schrieb Nirmoy Das:
Get rid off pin/unpin of gart BO at resume/suspend and
instead pin only once and try to recover gart content
at resume time. This is much more stable in case there
is OOM situation at 2nd call to amdgpu_device_evict_resources()
while evicting GART table.
Am 19.10.21 um 20:14 schrieb Nirmoy Das:
Do not allow exported amdgpu_gtt_mgr_*() to accept
any ttm_resource_manager pointer. Also there is no need
to force other module to call a ttm function just to
eventually call gtt_mgr functions.
That's a rather bad idea I think.
The GTT and VRAM manager
Thanks!
Remove drm_mm references and add DRM buddy
functions
Signed-off-by: Arunpravin
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 4 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 233 +++
2 files changed, 138 insertions(+), 99 deletions(-)
diff --git a/drivers/gpu/drm/amd/
- Add res cursor support for drm buddy
- Replace if..else statement with switch case statement
Signed-off-by: Arunpravin
---
.../gpu/drm/amd/amdgpu/amdgpu_res_cursor.h| 97 +++
1 file changed, 78 insertions(+), 19 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_r
Move vram defines and inline functions into
a header file
Signed-off-by: Arunpravin
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 18 +
drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h | 72
2 files changed, 73 insertions(+), 17 deletions(-)
create mode 100644 drivers/
i915 buddy selftests will be moved to drm selftest folder,
hence the config condition check may be removed.
Signed-off-by: Arunpravin
---
drivers/gpu/drm/drm_buddy.c | 4
1 file changed, 4 deletions(-)
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index 5eb7c418700
Remove i915 buddy references and add DRM buddy
functions
Signed-off-by: Arunpravin
---
drivers/gpu/drm/i915/Makefile | 1 -
drivers/gpu/drm/i915/i915_module.c| 3 -
drivers/gpu/drm/i915/i915_scatterlist.c | 11 +--
drivers/gpu/drm/i915/i915_ttm_buddy_manager.c
Export functions and write kerneldoc description
Signed-off-by: Arunpravin
---
drivers/gpu/drm/drm_buddy.c | 89 ++---
1 file changed, 83 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index 3e3303dd6658..5eb7
On contiguous allocation, we round up the size
to the nearest power of 2, implement a function
to free unused pages.
Signed-off-by: Arunpravin
---
drivers/gpu/drm/drm_buddy.c | 87 +
include/drm/drm_buddy.h | 4 ++
2 files changed, 91 insertions(+)
diff
Implemented a function which walk through the order list,
compares the offset and returns the maximum offset block,
this method is unpredictable in obtaining the high range
address blocks which depends on allocation and deallocation.
for instance, if driver requests address at a low specific
range,
This function becomes obsolete and may be removed.
Signed-off-by: Arunpravin
---
drivers/gpu/drm/drm_buddy.c | 101
include/drm/drm_buddy.h | 4 --
2 files changed, 105 deletions(-)
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
- Make drm_buddy_alloc a single function to handle
range allocation and non-range allocation demands.
- Implemented a new function alloc_range() which allocates
the requested order (in bytes) comply with range limitations
- Moved memory alignment logic from i915 driver
Signed-off-by: Arunpra
- Include drm buddy to DRM root Makefile
- Add drm buddy init and exit function calls
to drm core
Signed-off-by: Arunpravin
---
drivers/gpu/drm/Makefile | 2 +-
drivers/gpu/drm/drm_drv.c | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/Makefile b/drivers/
- Move i915_buddy.c to drm root folder
- Rename "i915" string with "drm" string wherever applicable
- Rename "I915" string with "DRM" string wherever applicable
- Fix header file dependencies
- Fix alignment issues
Signed-off-by: Arunpravin
---
.../drm/{i915/i915_buddy.c => drm_buddy.c}| 193
- Move i915_buddy.h to include/drm
- rename "i915" string to "drm"
- rename "I915" string to "DRM"
Signed-off-by: Arunpravin
---
drivers/gpu/drm/i915/i915_buddy.h | 143 --
include/drm/drm_buddy.h | 143 ++
2 files changed, 143 in
This series of patches implemented to move i915 buddy allocator
to drm root, and introduce new features include
- make drm_buddy_alloc a prime vehicle for allocation
- TOPDOWN range of address allocation support
- a function to free unused pages on contiguous allocation
- a function to allocate re
drm_get_edid() internally calls to drm_connector_update_edid_property()
and then drm_add_display_info(), which parses the EDID.
This happens in the function intel_hdmi_set_edid() and
intel_sdvo_tmds_sink_detect() (via intel_sdvo_get_edid()).
Once EDID is parsed, the monitor HDMI support informatio
According to the documentation, drm_add_edid_modes
"... Also fills out the &drm_display_info structure and ELD in @connector
with any information which can be derived from the edid."
drm_add_edid_modes accepts a struct edid *edid parameter which may have a
value or may be null. When it is not nu
On Tue, Oct 19, 2021 at 09:35:08PM +0300, Ville Syrjälä wrote:
> On Sat, Oct 16, 2021 at 08:42:14PM +0200, Claudio Suarez wrote:
> > According to the documentation, drm_add_edid_modes
> > "... Also fills out the &drm_display_info structure and ELD in @connector
> > with any information which can be
On Tue, Oct 19, 2021 at 5:49 PM Vignesh Chander wrote:
>
Please add a patch description, something like:
Enable GPU metrics feature in one VF mode. These are only possible in
one VF mode because the VF
is dedicated in that case.
With that fixed:
Reviewed-by: Alex Deucher
> Signed-off-by: Vig
Signed-off-by: Vignesh Chander
Change-Id: I14a5c4d6b9d790b7f298b67cece2c501a003e2a7
---
drivers/gpu/drm/amd/pm/amdgpu_pm.c | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index c255b4b8e685
Modified definitions:
- amdgpu_amdkfd_get_fw_version
- amdgpu_amdkfd_get_local_mem_info
- amdgpu_amdkfd_get_gpu_clock_counter
- amdgpu_amdkfd_get_max_engine_clock_in_mhz
- amdgpu_amdkfd_get_cu_info
- amdgpu_amdkfd_get_dmabuf_info
- amdgpu_amdkfd_get_vram_usage
- amdgpu_amdkfd_get_hive_id
- amdgpu_
Modified definitions:
- program_sh_mem_settings
- set_pasid_vmid_mapping
- init_interrupts
- address_watch_disable
- address_watch_execute
- wave_control_execute
- address_watch_get_offset
- get_atc_vmid_pasid_mapping_info
- set_scratch_backing_va
- set_vm_context_page_table_base
- read_vmid_from_
Static funcs in amdgpu_amdkfd_gfx_v9.c now using amdgpu_device.
Signed-off-by: Graham Sider
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 52 ---
1 file changed, 23 insertions(+), 29 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
b/drivers/gpu
Modified definitions:
- hqd_load
- hiq_mqd_load
- hqd_sdma_load
- hqd_dump
- hqd_sdma_dump
- hqd_is_occupied
- hqd_destroy
- hqd_sdma_is_occupied
- hqd_sdma_destroy
---
.../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 13 +++
.../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.h | 9 +++--
.../drm/am
Signed-off-by: Graham Sider
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 4 ++--
drivers/gpu/drm/amd/amdkfd/kfd_device.c | 4 +---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 -
drivers/gpu/drm/amd/include/kgd_kfd_in
Signed-off-by: Graham Sider
---
.../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 5 ---
.../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c| 5 ---
.../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c | 5 ---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 5 ---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_
Modified definitions:
- amdgpu_amdkfd_submit_ib
- amdgpu_amdkfd_set_compute_idle
- amdgpu_amdkfd_have_atomics_support
- amdgpu_amdkfd_flush_gpu_tlb_pasid
- amdgpu_amdkfd_flush_gpu_tlb_pasid
- amdgpu_amdkfd_gpu_reset
- amdgpu_amdkfd_alloc_gtt_mem
- amdgpu_amdkfd_free_gtt_mem
- amdgpu_amdkfd_alloc_g
Modified definitions:
- amdgpu_amdkfd_gpuvm_acquire_process_vm
- amdgpu_amdkfd_gpuvm_release_process_vm
- amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu
- amdgpu_amdkfd_gpuvm_free_memory_of_gpu
- amdgpu_amdkfd_gpuvm_map_memory_to_gpu
- amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu
- amdgpu_amdkfd_gpuvm_sync_
Static funcs in amdgpu_amdkfd_gfx_v10_3.c now using amdgpu_device.
Signed-off-by: Graham Sider
---
.../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c | 52 ---
1 file changed, 23 insertions(+), 29 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
b/drive
Static funcs in amdgpu_amdkfd_gfx_v10.c now using amdgpu_device.
Signed-off-by: Graham Sider
---
.../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c| 52 ---
1 file changed, 23 insertions(+), 29 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
b/drivers/g
Static funcs in amdgpu_amdkfd_gfx_v7.c now using amdgpu_device.
Signed-off-by: Graham Sider
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 51 +--
1 file changed, 23 insertions(+), 28 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
b/drivers/gpu
Static funcs in amdgpu_amdkfd_gfx_v8.c now using amdgpu_device.
Signed-off-by: Graham Sider
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c | 51 +--
1 file changed, 23 insertions(+), 28 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
b/drivers/gpu
Patch series to remove kgd_dev struct and replace all instances with
amdgpu_device objects.
amdgpu_device needs to be declared in kgd_kfd_interface.h to be visible
to kfd2kgd_calls.
Signed-off-by: Graham Sider
---
drivers/gpu/drm/amd/amdkfd/kfd_device.c | 1 +
drivers/gpu/drm/amd/amdkfd
Unused. Remove it.
Fixes: d1065882691179 ("Revert "drm/amd/display: Add helper for blanking all dp
displays"")
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.
Applied. thanks!
On Fri, Oct 15, 2021 at 2:48 AM Qing Wang wrote:
>
> show() must not use snprintf() when formatting the value to be
> returned to user space.
>
> Fix the following coccicheck warning:
> drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c:427:
> WARNING: use scnprintf or sprintf.
>
> Signe
No need to use the id variable, just use the constant
plus instance offset directly.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
b/drivers/gpu/drm/amd/
No need to use the tmp variable, just use the constant
directly.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 388
Roughly the same code was present in all VCN versions.
Consolidate it into a single function.
v2: use AMDGPU_UCODE_ID_VCN + i, check if num_inst >= 2
Signed-off-by: Alex Deucher
Reviewed-by: James Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 27 +
drivers/gpu/drm/a
Only enable firmware for the instance that is enabled.
v2: use AMDGPU_UCODE_ID_VCN + i
Fixes: 1b592d00b4ac83 ("drm/amdgpu/vcn: remove manual instance setting")
Reviewed-by: James Zhu
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 16
1 file changed, 8
On Tue, Oct 19, 2021 at 9:02 AM Jason Gunthorpe wrote:
>
> On Tue, Oct 19, 2021 at 04:13:34PM +0100, Joao Martins wrote:
> > On 10/19/21 00:06, Jason Gunthorpe wrote:
> > > On Mon, Oct 18, 2021 at 12:37:30PM -0700, Dan Williams wrote:
> > >
> > >>> device-dax uses PUD, along with TTM, they are the
Reviewed-by: Luben Tuikov
Regards,
Luben
On 2021-10-19 13:50, Kent Russell wrote:
> Change the error message when the bad_page_threshold is reached,
> explicitly stating that the GPU will not be initialized.
>
> Cc: Luben Tuikov
> Cc: Mukul Joshi
> Signed-off-by: Kent Russell
> ---
> drivers
On 2021-10-19 14:22, Russell, Kent wrote:
> [AMD Official Use Only]
>
>
>
>> -Original Message-
>> From: Kuehling, Felix
>> Sent: Tuesday, October 19, 2021 2:09 PM
>> To: Russell, Kent ; amd-gfx@lists.freedesktop.org
>> Cc: Tuikov, Luben ; Joshi, Mukul
>> Subject: Re: [PATCH 1/4] drm/amdg
On Sat, Oct 16, 2021 at 08:42:26PM +0200, Claudio Suarez wrote:
> Once EDID is parsed, the monitor HDMI support information is available
> through drm_display_info.is_hdmi. Retriving the same information with
> drm_detect_hdmi_monitor() is less efficient. Change to
> drm_display_info.is_hdmi where
On Sat, Oct 16, 2021 at 08:42:14PM +0200, Claudio Suarez wrote:
> According to the documentation, drm_add_edid_modes
> "... Also fills out the &drm_display_info structure and ELD in @connector
> with any information which can be derived from the edid."
>
> drm_add_edid_modes accepts a struct edid
[AMD Official Use Only]
> -Original Message-
> From: Kuehling, Felix
> Sent: Tuesday, October 19, 2021 2:13 PM
> To: Russell, Kent ; amd-gfx@lists.freedesktop.org
> Cc: Tuikov, Luben ; Joshi, Mukul
> Subject: Re: [PATCH 3/4] drm/amdgpu: Add kernel parameter for ignoring bad
> page
> t
[AMD Official Use Only]
> -Original Message-
> From: Kuehling, Felix
> Sent: Tuesday, October 19, 2021 2:09 PM
> To: Russell, Kent ; amd-gfx@lists.freedesktop.org
> Cc: Tuikov, Luben ; Joshi, Mukul
> Subject: Re: [PATCH 1/4] drm/amdgpu: Warn when bad pages approaches threshold
>
> Am
Do not allow exported amdgpu_gtt_mgr_*() to accept
any ttm_resource_manager pointer. Also there is no need
to force other module to call a ttm function just to
eventually call gtt_mgr functions.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 +--
drivers/gpu/drm/a
Get rid off pin/unpin of gart BO at resume/suspend and
instead pin only once and try to recover gart content
at resume time. This is much more stable in case there
is OOM situation at 2nd call to amdgpu_device_evict_resources()
while evicting GART table.
Signed-off-by: Nirmoy Das
---
drivers/gpu
Do not allow exported amdgpu_vram_mgr_*() to accept
any ttm_resource_manager pointer. Also there is no need
to force other module to call a ttm function just to
eventually call vram_mgr functions.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 3 +-
drivers/gpu/drm
Am 2021-10-19 um 1:50 p.m. schrieb Kent Russell:
> When a GPU hits the bad_page_threshold, it will not be initialized by
> the amdgpu driver. This means that the table cannot be cleared, nor can
> information gathering be performed (getting serial number, BDF, etc).
> Add an override called ignor
Am 2021-10-19 um 1:50 p.m. schrieb Kent Russell:
> Currently dmesg doesn't warn when the number of bad pages approaches the
> threshold for page retirement. WARN when the number of bad pages
> is at 90% or greater for easier checks and planning, instead of waiting
> until the GPU is full of bad pag
If the ignore_bad_page_threshold kernel parameter is set to true,
continue to post the GPU. Print an warning to dmesg that this action has
been done, and that page retirement will obviously not work for said GPU
Cc: Luben Tuikov
Cc: Mukul Joshi
Signed-off-by: Kent Russell
---
drivers/gpu/drm/a
Change the error message when the bad_page_threshold is reached,
explicitly stating that the GPU will not be initialized.
Cc: Luben Tuikov
Cc: Mukul Joshi
Signed-off-by: Kent Russell
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
When a GPU hits the bad_page_threshold, it will not be initialized by
the amdgpu driver. This means that the table cannot be cleared, nor can
information gathering be performed (getting serial number, BDF, etc).
Add an override called ignore_bad_page_threshold that can be set to true
to still initi
Currently dmesg doesn't warn when the number of bad pages approaches the
threshold for page retirement. WARN when the number of bad pages
is at 90% or greater for easier checks and planning, instead of waiting
until the GPU is full of bad pages
Cc: Luben Tuikov
Cc: Mukul Joshi
Signed-off-by: Ken
With two nit-pick below.
ThispatchisReviewed-by:JamesZhu
On 2021-10-19 11:56 a.m., Alex Deucher wrote:
Roughly the same code was present in all VCN versions.
Consolidate it into a single function.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 25 ++
On 10/19/2021 5:43 PM, Christian König wrote:
Am 19.10.21 um 15:22 schrieb Nirmoy Das:
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
Also remove 2nd call to amdgpu_device_evict_resources()
as we don't need it.
Signed-off-by: Nirm
On 2021-10-19 11:54 a.m., Christian König wrote:
Am 19.10.21 um 17:41 schrieb Andrey Grodzovsky:
On 2021-10-19 9:22 a.m., Nirmoy Das wrote:
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
+Christian
Could you guys also clarify w
On Tue, Oct 19, 2021 at 04:13:34PM +0100, Joao Martins wrote:
> On 10/19/21 00:06, Jason Gunthorpe wrote:
> > On Mon, Oct 18, 2021 at 12:37:30PM -0700, Dan Williams wrote:
> >
> >>> device-dax uses PUD, along with TTM, they are the only places. I'm not
> >>> sure TTM is a real place though.
> >>
>
Roughly the same code was present in all VCN versions.
Consolidate it into a single function.
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 25 +
drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 ++
drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 10 +---
Am 19.10.21 um 17:41 schrieb Andrey Grodzovsky:
On 2021-10-19 9:22 a.m., Nirmoy Das wrote:
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
+Christian
Could you guys also clarify what exactly are the stability issues this
fixes ?
On 2021-10-19 11:13 a.m., Alex Deucher wrote:
Only enable firmware for the instance that is enabled.
Fixes: 1b592d00b4ac83 ("drm/amdgpu/vcn: remove manual instance setting")
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 19 +++
1 file changed, 11 ins
Am 19.10.21 um 15:22 schrieb Nirmoy Das:
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
Also remove 2nd call to amdgpu_device_evict_resources()
as we don't need it.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_d
On 2021-10-19 9:22 a.m., Nirmoy Das wrote:
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
+Christian
Could you guys also clarify what exactly are the stability issues this
fixes ?
Andrey
Also remove 2nd call to amdgpu_devic
Add missing check in smu_v11_0_init_display_count(),
Fixes: af3b89d3a639d5 ("drm/amdgpu/smu11.0: convert to IP version checking")
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/
On 10/19/21 00:06, Jason Gunthorpe wrote:
> On Mon, Oct 18, 2021 at 12:37:30PM -0700, Dan Williams wrote:
>
>>> device-dax uses PUD, along with TTM, they are the only places. I'm not
>>> sure TTM is a real place though.
>>
>> I was setting device-dax aside because it can use Joao's changes to
>> g
Only enable firmware for the instance that is enabled.
Fixes: 1b592d00b4ac83 ("drm/amdgpu/vcn: remove manual instance setting")
Signed-off-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/drivers/g
It again fails with the same message!
But this time it is different!
Here's why:
openat(AT_FDCWD,
"/sys/class/drm/card0/device/pp_dpm_fclk", O_RDONLY) = 3
read(3, "0: 571Mhz \n1: 1274Mhz *\n2: 1221M"...,
8191) = 36
read(3, "", 8191)
[AMD Official Use Only]
It was the rocm-smi -c flag. Maybe some work was done to make it more robust,
that would be nice. But the -c flag is supposed to show the current frequency
for each clock type. -g would do the same, but just for SCLK.
Kent
From: Tuikov, Luben
Sent: Tuesday, October 19,
Get rid off pin/unpin and evict and swap back gart
page table which should make things less likely to break.
Also remove 2nd call to amdgpu_device_evict_resources()
as we don't need it.
Signed-off-by: Nirmoy Das
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 -
drivers/gpu/drm/amd/amdg
On Mon, Oct 18, 2021 at 11:22 PM Aaron Liu wrote:
>
> B0 internal rev_id is 0x01, B1 internal rev_id is 0x02.
> The external rev_id for B0 and B1 is 0x20.
> The original expression is not suitable for B1.
Are you sure about this? We'll be losing the difference between B0
and B1. I think 0x19 is
PSP firmware will be responsible for applying the GRBM CAM remapping in
the production. And the GRBM_CAM_INDEX / GRBM_CAM_DATA registers will be
protected by PSP under security policy. So remove it according to the
new security policy.
Signed-off-by: Huang Rui
---
drivers/gpu/drm/amd/amdgpu/gfx_
[AMD Official Use Only]
> -邮件原件-
> 发件人: Christian König
> 发送时间: Tuesday, October 19, 2021 4:46 PM
> 收件人: Yang, Stanley ; Das, Nirmoy
> ; amd-gfx@lists.freedesktop.org
> 主题: Re: 回复: [PATCH Review 1/1] drm/ttm: fix debugfs node create failed
>
> Am 19.10.21 um 10:02 schrieb Yang, Stanley
Am 19.10.21 um 10:02 schrieb Yang, Stanley:
[AMD Official Use Only]
-邮件原件-
发件人: amd-gfx 代表 Das,
Nirmoy
发送时间: Thursday, October 14, 2021 2:11 AM
收件人: Christian König ; amd-
g...@lists.freedesktop.org
主题: Re: [PATCH Review 1/1] drm/ttm: fix debugfs node create failed
On 10/13/2021 2:2
Dear Simon,
Am 19.10.21 um 10:10 schrieb Simon Ser:
On Tuesday, October 19th, 2021 at 01:21, Paul Menzel
wrote:
Am 19.10.21 um 01:06 schrieb Simon Ser:
On Tuesday, October 19th, 2021 at 01:03, Paul Menzel wrote:
Excuse my ignorance. Reading the commit message, there was a Linux
kernel ch
On Tuesday, October 19th, 2021 at 01:21, Paul Menzel
wrote:
> Am 19.10.21 um 01:06 schrieb Simon Ser:
> > On Tuesday, October 19th, 2021 at 01:03, Paul Menzel wrote:
> >
> >> Excuse my ignorance. Reading the commit message, there was a Linux
> >> kernel change, that broke Chrome OS userspace, ri
+AlexD
+ChrisianK
+LKML
On 2021-10-19 03:44, Paul Menzel wrote:
> Dear Luben,
>
>
> Am 19.10.21 um 06:50 schrieb Luben Tuikov:
>> On 2021-10-19 00:38, Lazar, Lijo wrote:
>>> On 10/19/2021 9:45 AM, Luben Tuikov wrote:
On 2021-10-18 23:38, Lazar, Lijo wrote:
> On 10/19/2021 5:19 AM, Luben T
[AMD Official Use Only]
> -邮件原件-
> 发件人: amd-gfx 代表 Das,
> Nirmoy
> 发送时间: Thursday, October 14, 2021 2:11 AM
> 收件人: Christian König ; amd-
> g...@lists.freedesktop.org
> 主题: Re: [PATCH Review 1/1] drm/ttm: fix debugfs node create failed
>
>
> On 10/13/2021 2:29 PM, Christian König wrote
Dear Luben,
Thank you for your quick reply.
Am 19.10.21 um 09:43 schrieb Luben Tuikov:
On 2021-10-19 03:23, Paul Menzel wrote:
Sorry, two more style nits.
1. Could you please use 75 characters per line for the text width of
the commit messages. Currently, especially 4/5, are hard to read
Dear Luben,
Am 19.10.21 um 06:50 schrieb Luben Tuikov:
On 2021-10-19 00:38, Lazar, Lijo wrote:
On 10/19/2021 9:45 AM, Luben Tuikov wrote:
On 2021-10-18 23:38, Lazar, Lijo wrote:
On 10/19/2021 5:19 AM, Luben Tuikov wrote:
[…]
- if (ret)
-
On 2021-10-19 03:23, Paul Menzel wrote:
> Dear Luben,
>
>
> Sorry, two more style nits.
>
> 1. Could you please use 75 characters per line for the text width of
> the commit messages. Currently, especially 4/5, are hard to read being
> so short.
This is the default we use--I've not made any cha
Dear Luben,
Sorry, two more style nits.
1. Could you please use 75 characters per line for the text width of
the commit messages. Currently, especially 4/5, are hard to read being
so short.
2. No idea, what is done in amd-gfx, but for me it is more common to
put the iteration number (rer
84 matches
Mail list logo
