Age | Commit message (Collapse) | Author | Files | Lines |
|
This version brings along following fixes:
- Improvement on eDP
- PSR bug fixes
- SubVP bug fixes
- Improvements on pipe handling to address potential issues
- Freesync bug fix
- DPIA bug fix
- Fix multi-display issues
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Aric Cyr <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
The k1/k2 pixel rate dividers in dccg should only be updated on stream enable
and do not actually depend on whether odm combine is active.
This removes an on flip update of these and fixes the calculate function
to ignore odm status for dp steams.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Dmytro Laktyushkin <[email protected]>
Reviewed-by: Ariel Bernstein <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
rate ratio"
Revert commit 5f3401eeb064 ("drm/amd/display: reallocate DET for dual displays with high pixel rate ratio")
[Why]
Previously had modified DET allocation algorithm to allocate less DET
segments for lower pixel rate display and more DET segments for higher
pixel rate display. But noticed it is causing underflow when higher
pixel rate display is not displaying at higher mode
[How]
Roll back change
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Samson Tam <[email protected]>
Reviewed-by: Alvin Lee <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
DPIA doesn't support UHBR, driver should not enable UHBR
for dp tunneling
[How]
limit DPIA link rate to HBR3
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Peichen Huang <[email protected]>
Reviewed-by: Mustapha Ghaddar <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Description]
- Missed color / CSC updates in fast update path
which caused Custom Color to break.
- Add color / CSC updates to new fast update path
to fix custom color
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Alvin Lee <[email protected]>
Reviewed-by: Jun Lei <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Why:
Some EDIDs report a minimum refresh rate lower than what HW can support
How:
Add a check to calculate minimum supported refresh rate with current timing
and use that as the minimum if a lower one is passed in
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Austin Zheng <[email protected]>
Reviewed-by: Anthony Koo <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
When iterating over all pipes in the loop, the CRB allocation algorithm
may potentially skip some of the pipes. Previously, the current pipe
index didn't get updated in this case, causing incorrect outcome.
[How]
Increment the pipe index when skipping over a pipe in the loop.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Artem Grishin <[email protected]>
Reviewed-by: Dmytro Laktyushkin <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Description]
PSR implementation in FW has inline polling which can poll for up
to 1ms. This will interfere with SubVP because SubVP is timing
sensitive and can't tolerate up to 1ms worth of delay before
handling vertical or VLINE interrupts. Therefore block SubVP + DRR
cases if DRR is PSR capable
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Alvin Lee <[email protected]>
Reviewed-by: Saaem Rizvi <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why and How]
We should not disable phantom pipes in this sequence, as this should be
controlled by FW. Furthermore, the previous programming sequence would
have enabled the phantom pipe in driver as well, causing corruption.
This change should avoid this from occuring.
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Saaem Rizvi <[email protected]>
Reviewed-by: Alvin Lee <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Re-enable SubVP high refresh now that it is fixed
for displays with high refresh rates.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Alvin Lee <[email protected]>
Reviewed-by: Saaem Rizvi <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why & How]
Bugs preventing DPP/HUBP power gating have been addressed
so this should be reenabled on dcn314 for sufficient
hardware rev versions
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Daniel Miess <[email protected]>
Reviewed-by: Nicholas Kazlauskas <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Description]
- SubVP high refresh should only be enabled if all displays
are >= 120hz. We do not want to accidentally enables configs
such as 60hz[SubVP] + 120hz[SubVP]
- Ensure that the SubVP config generation code does not produce
configs such as 60hz[SubVP] + 120hz[SubVP]
- Also add admissibility checks to ensure these configs do not
pass as valid configs
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Alvin Lee <[email protected]>
Reviewed-by: Dillon Varone <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
dmub_psr_get_state() return an invalid PSR state while disable
the PSR because convert_psr_state() doesn't recognize the state
that return from DMCUB.
[How]
Add a PSR state to make the dmub_psr_get_state() return a
correct PSR state.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Tom Chung <[email protected]>
Reviewed-by: Wayne Lin <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
When the PSR enabled. If you try to adjust the timing parameters,
it may cause system hang. Because the timing mismatch with the
DMCUB settings.
[How]
Disable the PSR before adjusting timing parameters.
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Tom Chung <[email protected]>
Reviewed-by: Wayne Lin <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Aric Cyr <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Any invalid mode from hw perspective should be given corresponding
error code, otherwise it leads to confusing warning message
"[drm] Mode Validation Warning: Validation OK failed validation."
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Fangzhi Zuo <[email protected]>
Reviewed-by: Dmytro Laktyushkin <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Why:
Log DP2 data to be used for debugging purposes
How:
Check the reported link rate of the DP connection and
translate it to the DP version.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Austin Zheng <[email protected]>
Reviewed-by: Martin Leung <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why] within dc_link_detect, edp edid is read only for the first time
and saved. edid will not be read after the first time read. to run edp
edid read test, need read edp edid for each dc_link_detect. dc->config
flag allow_edp_hotplug_detection could be used for edp edid test.
[How] add debugfs for dc->config.allow_edp_hotplug_detection
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Hersen Wu <[email protected]>
Reviewed-by: Aurabindo Pillai <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why] most edp support only timings from edid. applying
non-edid timings, especially those timings out of edp
bandwidth, may damage edp.
[How] do not add non-edid timings for edp.
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Hersen Wu <[email protected]>
Reviewed-by: Roman Li <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
system"
This reverts commit c105518679b6e87232874ffc989ec403bee59664.
This patch disables the TOPDOWN flag for APU and few dGPU cards
which has the VRAM size equal to the BAR size.
When we enable the TOPDOWN flag, we get the free blocks at
the highest available memory region and we don't split the
lower order blocks. This change is required to keep off
the fragmentation related issues particularly in ASIC
which has VRAM space <= 500MiB
Hence, we are reverting this patch.
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2270
Signed-off-by: Arunpravin Paneer Selvam <[email protected]>
Reviewed-by: Christian König <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Add these two dev attrs for xgmi info details which is helpful for
developers checking the xgmi topology by catting the sys file directly.
Take 4 cards with xgmi connection as an example, get the num_hops for each
device or node through xmig_hive_info dir like,
cat /sys/bus/pci/devices/0000:41:00.0/xgmi_hive_info/node1/num_hops
will return "00 41 41 41" where "00" stands for the hops to node1 itself
and "41" is the hops in hex format to every other node in the same hive.
There are node1/node2/node3/node4 representing 4 cards in the hive.
The same for num_links dev attr.
Signed-off-by: Shiwu Zhang <[email protected]>
Acked-by: Le Ma <[email protected]>
Reviewed-by: Hawking Zhang <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Only vcn0 can process AV1 codecx. In order to use both vcn0 and
vcn1 in h264/265 transcode to AV1 cases, set vcn0 sched score to 1
at initialization time.
Signed-off-by: Sonny Jiang <[email protected]>
Reviewed-by: Leo Liu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
We have a clean build with W=1 as of
commit c168feed5d03 ("drm/amd/display/amdgpu_dm/amdgpu_dm_helpers: Move
SYNAPTICS_DEVICE_ID into CONFIG_DRM_AMD_DC_DCN ifdef"). So, let's enable
these checks unconditionally for the entire module to catch these errors
during development.
Cc: Alex Deucher <[email protected]>
Cc: Nathan Chancellor <[email protected]>
Tested-by: Nathan Chancellor <[email protected]>
Reviewed-by: Nathan Chancellor <[email protected]>
Signed-off-by: Hamza Mahfooz <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Fixes the following gcc with W=1:
drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c:76: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
* EEPROM Table structure v1
drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c:98: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
* EEPROM Table structrue v2.1
Cc: Christian König <[email protected]>
Cc: Alex Deucher <[email protected]>
Signed-off-by: Srinivasan Shanmugam <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why and How]
Add back debug bits enabling RCO for dcn314 as underflow
associated with this change has been resolved
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Daniel Miess <[email protected]>
Reviewed-by: Jun Lei <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Rename mman.entity to mman.high_pr to make the distinction
clearer that this is a high priority scheduler. Similarly,
rename the recently added mman.delayed to mman.low_pr to
make it clear it is a low priority scheduler.
No functional change in this patch.
Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Christian König <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
CONFIG_DRM_AMD_DC_DCN ifdef
Fixes the following W=1 kernel build warning(s):
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_helpers.c:48:22: warning: ‘SYNAPTICS_DEVICE_ID’ defined but not used [-Wunused-const-variable=]
Cc: Harry Wentland <[email protected]>
Cc: Leo Li <[email protected]>
Cc: Rodrigo Siqueira <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: "Christian König" <[email protected]>
Cc: "Pan, Xinhui" <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reviewed-by: Harry Wentland <[email protected]>
Signed-off-by: Lee Jones <[email protected]>
Signed-off-by: Hamza Mahfooz <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
Hardware implements root clock gating by utilizing the DPP DTO registers
with a special case of DTO enabled, phase = 0, modulo = 1. This
conflicts with our policy to always update the DPPDTO for cases where
it's expected to be disabled.
The pipes unexpectedly enter a higher power state than expected because
of this programming flow.
[How]
Guard the upper layers of HWSS against this hardware quirk with
programming the register with an internal state flag in DCCG.
While technically acting as global state for the DCCG, HWSS shouldn't be
expected to understand the hardware quirk for having DTO disabled
causing more power than DTO enabled with this specific setting.
This also prevents sequencing errors from occuring in the future if
we have to program DPP DTO in multiple locations.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Nicholas Kazlauskas <[email protected]>
Reviewed-by: Jun Lei <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.5-2023-06-02:
amdgpu:
- SR-IOV fixes
- Warning fixes
- Misc code cleanups and spelling fixes
- DCN 3.2 updates
- Improved DC FAMS support for better power management
- Improved DC SubVP support for better power management
- DCN 3.1.x fixes
- Max IB size query
- DC GPU reset fixes
- RAS updates
- DCN 3.0.x fixes
- S/G display fixes
- CP shadow buffer support
- Implement connector force callback
- Z8 power improvements
- PSP 13.0.10 vbflash support
- Mode2 reset fixes
- Store MQDs in VRAM to improve queue switch latency
- VCN 3.x fixes
- JPEG 3.x fixes
- Enable DC_FP on LoongArch
- GFXOFF fixes
- GC 9.4.3 partition support
- SDMA 4.4.2 partition support
- VCN/JPEG 4.0.3 partition support
- VCN 4.0.3 updates
- NBIO 7.9 updates
- GC 9.4.3 updates
- Take NUMA into account when allocating memory
- Handle NUMA for partitions
- SMU 13.0.6 updates
- GC 9.4.3 RAS updates
- Stop including unused swiotlb.h
- SMU 13.0.7 fixes
- Fix clock output ordering on some APUs
- Clean up DC FPGA code
- GFX9 preemption fixes
- Misc irq fixes
- S0ix fixes
- Add new DRM_AMDGPU_WERROR config parameter to help with CI
- PCIe fix for RDNA2
- kdoc fixes
- Documentation updates
amdkfd:
- Query TTM mem limit rather than hardcoding it
- GC 9.4.3 partition support
- Handle NUMA for partitions
radeon:
- Fix possible double free
- Stop including unused swiotlb.h
- Fix possible division by zero
ttm:
- Add query for TTM mem limit
- Add NUMA awareness to pools
- Export ttm_pool_fini()
UAPI:
- Add new ctx query flag to better handle GPU resets
Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22290
- Add new interface to query and set shadow buffer for RDNA3
Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21986
- Add new INFO query for max IB size
Proposed userspace: https://gitlab.freedesktop.org/bnieuwenhuizen/mesa/-/commits/ib-rejection-v3
amd-drm-next-6.5-2023-06-09:
amdgpu:
- S0ix fixes
- Initial SMU13 Overdrive support
- kdoc fixes
- Misc clode cleanups
- Flexible array fixes
- Display OTG fixes
- SMU 13.0.6 updates
- Revert some broken clock counter updates
- Misc display fixes
- GFX9 preemption fixes
- Add support for newer EEPROM bad page table format
- Add missing radeon secondary id
- Add support for new colorspace KMS API
- CSA fix
- Stable pstate fixes for APUs
- make vbl interface admin only
- Handle PCI accelerator class
amdkfd:
- Add debugger support for gdb
radeon:
- Fix possible UAF
drm:
- Add Colorspace functionality
UAPI:
- Add debugger interface for enabling gdb
Proposed userspace: https://github.com/ROCm-Developer-Tools/ROCdbgapi/tree/wip-dbgapi
- Add KMS colorspace API
Discussion: https://lists.freedesktop.org/archives/dri-devel/2023-June/408128.html
From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
[Why]
DPIA doesn't support UHBR, driver should not enable UHBR
for dp tunneling
[How]
limit DPIA link rate to HBR3
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Peichen Huang <[email protected]>
Reviewed-by: Mustapha Ghaddar <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
When the PSR enabled. If you try to adjust the timing parameters,
it may cause system hang. Because the timing mismatch with the
DMCUB settings.
[How]
Disable the PSR before adjusting timing parameters.
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Tom Chung <[email protected]>
Reviewed-by: Wayne Lin <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why] most edp support only timings from edid. applying
non-edid timings, especially those timings out of edp
bandwidth, may damage edp.
[How] do not add non-edid timings for edp.
Cc: Mario Limonciello <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: [email protected]
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Hersen Wu <[email protected]>
Reviewed-by: Roman Li <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
system"
This reverts commit c105518679b6e87232874ffc989ec403bee59664.
This patch disables the TOPDOWN flag for APU and few dGPU cards
which has the VRAM size equal to the BAR size.
When we enable the TOPDOWN flag, we get the free blocks at
the highest available memory region and we don't split the
lower order blocks. This change is required to keep off
the fragmentation related issues particularly in ASIC
which has VRAM space <= 500MiB
Hence, we are reverting this patch.
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2270
Signed-off-by: Arunpravin Paneer Selvam <[email protected]>
Reviewed-by: Christian König <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected]
|
|
Only vcn0 can process AV1 codecx. In order to use both vcn0 and
vcn1 in h264/265 transcode to AV1 cases, set vcn0 sched score to 1
at initialization time.
Signed-off-by: Sonny Jiang <[email protected]>
Reviewed-by: Leo Liu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.1.x
|
|
On smu 13.0.0, the compute workload type cannot be set on all the skus
due to some other problems. This workaround is to make sure compute workload type
can also run on some specific skus.
v2: keep the variable consistent
Signed-off-by: Kenneth Feng <[email protected]>
Acked-by: Lijo Lazar <[email protected]>
Reviewed-by: Feifei Xu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.1.x
|
|
Non-root users shouldn't be able to try to trigger a VBIOS flash
or query the flashing status. This should be reserved for users with the
appropriate permissions.
Cc: [email protected]
Fixes: 8424f2ccb3c0 ("drm/amdgpu/psp: Add vbflash sysfs interface support")
Reviewed-by: Alex Deucher <[email protected]>
Signed-off-by: Mario Limonciello <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
The VBIOS image update flow requires userspace to:
1) Write the image to `psp_vbflash`
2) Read `psp_vbflash`
3) Poll `psp_vbflash_status` to check for completion
If userspace reads `psp_vbflash` before writing an image, it's
possible that it causes problems that can put the dGPU into an invalid
state.
Explicitly check that an image has been written before letting a read
succeed.
Cc: [email protected]
Fixes: 8424f2ccb3c0 ("drm/amdgpu/psp: Add vbflash sysfs interface support")
Reviewed-by: Alex Deucher <[email protected]>
Signed-off-by: Mario Limonciello <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
0x5b70 is a missing RV370 secondary id. Add it so
we don't try and probe it with amdgpu.
Cc: [email protected]
Reviewed-by: Michel Dänzer <[email protected]>
Tested-by: Michel Dänzer <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected]
|
|
Patch the packages including CONTEXT_CONTROL and WRITE_DATA for gfx9
during the resubmission scenario.
Signed-off-by: Jiadong Zhu <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.3.x
|
|
When the preempted IB frame resubmitted to cp, we need to modify the frame
data including:
1. set PRE_RESUME 1 in CONTEXT_CONTROL.
2. use meta data(DE and CE) read from CSA in WRITE_DATA.
Add functions to save the location the first time IBs emitted and callback
to patch the package when resubmission happens.
Signed-off-by: Jiadong Zhu <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.3.x
|
|
It is firmware requirement to set gds_backup_addrlo and gds_backup_addrhi
of DE meta both zero if no gds partition is allocated for the frame.
Signed-off-by: Jiadong Zhu <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.3.x
|
|
When MEC executes unmap_queue for mid command buffer preemption, it will
kick the write pointer of the gfx ring, set CP_VMID_PREEMPT to trigger the
preemption and wait for CP_VMID_PREEMPT becomes zero after the preemption
done. There is a race condition that PFP may excute the resetting command
before MEC set CP_VMID_PREEMPT. As a result, hang happens as
CP_VMID_PREEMPT is always 0xffff.
To avoid this, we send resetting CP_VMID_PREEMPT command after the trailing
fence is siganled and update gfx write pointer explicitly.
Signed-off-by: Jiadong Zhu <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected] # 6.3.x
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2535
|
|
Why:
Limit maximum clock speeds to DC mode limits for DC mode systems
How:
Store DC mode limits when individual clocks are initialized and
cap the values when building the clock table
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Austin Zheng <[email protected]>
Reviewed-by: Alvin Lee <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[Why]
Programming register delta for DSC sub-block
[How]
Change DSC, resource files for programming register delta.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Sridevi <[email protected]>
Reviewed-by: Chris Park <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
Revert commit 257e9891db0b ("drm/amd/display: cache trace buffer size")
to fix regression found in tests.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Leo Ma <[email protected]>
Reviewed-by: Josip Pavic <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
On smu 13.0.0, the compute workload type cannot be set on all the skus
due to some other problems. This workaround is to make sure compute workload type
can also run on some specific skus.
v2: keep the variable consistent
Signed-off-by: Kenneth Feng <[email protected]>
Acked-by: Lijo Lazar <[email protected]>
Reviewed-by: Feifei Xu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
[why]
check dmub_Srv exist or not before accessing dmub.
Acked-by: Stylon Wang <[email protected]>
Signed-off-by: Charlene Liu <[email protected]>
Reviewed-by: Zhan Liu <[email protected]>
Tested-by: Daniel Wheeler <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
This patch enables following UMD stable Pstates profile levels for
power_dpm_force_performance_level interface.
- profile_peak
- profile_min_mclk
- profile_min_sclk
- profile_standard
Signed-off-by: shikaguo <[email protected]>
Reviewed-by: Tim Huang <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
and compute
enforce process isolation between graphics and compute via using the same reserved vmid.
v2: remove params "struct amdgpu_vm *vm" from
amdgpu_vmid_alloc_reserved and amdgpu_vmid_free_reserved.
Signed-off-by: Chong Li <[email protected]>
Reviewed-by: Christian Koenig <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
-Wunused-but-set-variable was only supported in clang starting with
13.0.0, so earlier versions will emit a warning, which is turned into a
hard error for the kernel to mirror GCC:
error: unknown warning option '-Wunused-but-set-variable'; did you mean '-Wunused-const-variable'? [-Werror,-Wunknown-warning-option]
The minimum supported version of clang for building the kernel is
11.0.0, so match the rest of the kernel and wrap
-Wunused-but-set-variable in a cc-option call, so that it is only used
when supported by the compiler.
Closes: https://github.com/ClangBuiltLinux/linux/issues/1869
Fixes: 1b320ad3f5a8 ("drm/amd/amdgpu: introduce DRM_AMDGPU_WERROR")
Signed-off-by: Nathan Chancellor <[email protected]>
Signed-off-by: Hamza Mahfooz <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|