aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd
AgeCommit message (Collapse)AuthorFilesLines
2023-08-15drm/amd/display: fix incorrect stream_res allocation for older ASICWenjing Liu1-14/+13
[why] There is a recent work for developing a new pipe resource allocation policy used for new ASIC. The new code change needs to modify asic independent pipe resource allocation flow and hook up the new allocation policy in asic dependent layer. Unfortunately this change revealed a hidden bug in the old pipe resource allocation sequence used for older asics. In the older version of acquiring pipe for layer, we are always assigning otg master's opp and tg to the newly allocated secondary dpp pipe. This logic is incorrect when the secodnary dpp pipe is connected to a secondary opp head pipe in ODM combine configuration. Before the recent change, we will overwrite this wrong assignement in asic independent layer again. This covers up the issue. With the recent change, we will no longer cover up this in upper layer and therefore causes wrong tg and opp assignement to the secondary dpp pipe connected to a secondary opp head. [how] Always assign tg and opp from its own opp head instead of otg master. Reviewed-by: Martin Leung <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Wenjing Liu <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Gamut remap only changes missedKrunoslav Kovac1-0/+1
[WHY&HOW] If surface format is dynamically changed within app without changing timing / whole plane, we don't reprogram gamut remap matrix. Issue example: Linear FP16 scRGB going to PQ+BT.2020 monitor. Remap = scRGB->BT.2020 App switches swapchain format to 10-bit PQ+BT.2020. SW calculates correctly that new Remap = bypass (in=-out=BT.2020). But update not applied in HW. Reviewed-by: Jun Lei <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Krunoslav Kovac <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: PMFW to wait for DMCUB ack for FPO casesAlvin Lee3-0/+13
[Description] We want PMFW to wait for DMCUB to ACK the MCLK end message for FPO cases as well. Reviewed-by: Samson Tam <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Alvin Lee <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Enable subvp high refresh up to 175hzAlvin Lee2-3/+4
[Description] Expand the SubVP policy to include up to 175hz displays for better power saving on more display configs. Reviewed-by: Chaitanya Dhere <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Alvin Lee <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu/vcn: change end doorbell index for vcn_v4_0_3Samir Dhume1-3/+3
For sriov, doorbell index for vcn0 for AID needs to be on 32 byte boundary so we need to move the vcn end doorbell Signed-off-by: Samir Dhume <[email protected]> Acked-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: workaround address watch clearing bug for gfx v9.4.2Eric Huang1-7/+1
KFD currently relies on MEC FW to clear tcp watch control register on UNMAP_PROCESS, but FW doesn't work on it, which is a bug. So the solution is to clear the register as gfx v9 in KFD. Signed-off-by: Eric Huang <[email protected]> Reviewed-by: Jonathan Kim <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: disable clock gating logic reversed bug fixMuhammad Ahmed2-4/+4
[Why] disable clock gating logic reversed bug fix Reviewed-by: Charlene Liu <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Muhammad Ahmed <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Support Compliance Test Pattern Generation with DP2 RetimerMichael Strauss17-237/+738
[WHY] Certain retimer requires workarounds in order to correctly output test patterns. [HOW] Add vendor-specific aux sequences to program retimer's TX and pattern generator when specific compliance test patterns are requested by sink. Note: SQ128 w/a in DPMF mode only works in one flip orientation currently Reviewed-by: Hansen Dsouza <[email protected]> Reviewed-by: Wenjing Liu <[email protected]> Acked-by: Stylon Wang <[email protected]> Signed-off-by: Michael Strauss <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu/jpeg: mmsch_v4_0_3 requires doorbell on 32 byte boundarySamir Dhume1-3/+14
BASE: VCN0 unified (32 byte boundary) BASE+4: MJPEG0 BASE+5: MJPEG1 BASE+6: MJPEG2 BASE+7: MJPEG3 BASE+12: MJPEG4 BASE+13: MJPEG5 BASE+14: MJPEG6 BASE+15: MJPEG7 Signed-off-by: Samir Dhume <[email protected]> Acked-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Remove unnecessary NULL valuesRuan Jinjie2-4/+4
The NULL initialization of the pointers assigned by kzalloc() first is not necessary, because if the kzalloc() failed, the pointers will be assigned NULL, otherwise it works as usual. so remove it. Reviewed-by: Christian König <[email protected]> Signed-off-by: Ruan Jinjie <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu/vcn: mmsch_v4_0_3 requires doorbell on 32 byte boundarySamir Dhume1-3/+10
Align on 32 byte boundary. Signed-off-by: Samir Dhume <[email protected]> Acked-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: Remove unnecessary NULL valuesRuan Jinjie1-2/+2
The NULL initialization of the pointers assigned by kzalloc() first is not necessary, because if the kzalloc() failed, the pointers will be assigned NULL, otherwise it works as usual. so remove it. Reviewed-by: Christian König <[email protected]> Signed-off-by: Ruan Jinjie <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: dmub_replay: don't use kernel-doc markersRandy Dunlap1-9/+9
These functions don't use kernel-doc notation for comments so don't begin each comment block with the "/**" kernel-doc marker. This prevents a bunch of kernel-doc warnings: dmub_replay.c:37: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst dmub_replay.c:37: warning: missing initial short description on line: * Get Replay state from firmware. dmub_replay.c:66: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst dmub_replay.c:66: warning: missing initial short description on line: * Enable/Disable Replay. dmub_replay.c:116: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst dmub_replay.c:116: warning: missing initial short description on line: * Set REPLAY power optimization flags. dmub_replay.c:134: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst dmub_replay.c:134: warning: missing initial short description on line: * Setup Replay by programming phy registers and sending replay hw context values to firmware. and 10 more similar warnings. Fixes: c7ddc0a800bc ("drm/amd/display: Add Functions to enable Freesync Panel Replay") Signed-off-by: Randy Dunlap <[email protected]> Reported-by: kernel test robot <[email protected]> Link: lore.kernel.org/r/[email protected] Cc: Bhawanpreet Lakha <[email protected]> Cc: Harry Wentland <[email protected]> Cc: Alex Deucher <[email protected]> Cc: Leo Li <[email protected]> Cc: Rodrigo Siqueira <[email protected]> Cc: [email protected] Cc: Christian König <[email protected]> Cc: "Pan, Xinhui" <[email protected]> Cc: [email protected] Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd: Add amdgpu_hwmon_get_sensor_generic()Mario Limonciello1-233/+88
Many sensor function have a lot of boilerplate checks. Move these into a generic amdgpu_hwmon_get_sensor_generic() instead. No intended functional changes. Signed-off-by: Mario Limonciello <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu/vcn: Add MMSCH v4_0_3 support for sriovSamir Dhume1-0/+37
The structures are the same as v4_0 except for the init header Signed-off-by: Samir Dhume <[email protected]> Acked-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: Add I2C EEPROM support on smu v13_0_6Candice Li1-0/+2
Support I2C EEPROM on smu v13_0_6. v2: Move IP_VERSION(13, 0, 6) ahead of IP_VERSION(13, 0, 10). Signed-off-by: Candice Li <[email protected]> Reviewed-by: Yang Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: bump SMU v13.0.5 driver_if header versionTim Huang1-2/+1
Align the SMU driver interface version with PMFW to suppress the version mismatch message on driver loading. Signed-off-by: Tim Huang <[email protected]> Reviewed-by: Yifan Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: correct the logics for retreiving SMU13 OD setting limitsEvan Quan2-76/+64
To better meet the growing demainds for more OD features. Signed-off-by: Evan Quan <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: correct the way for checking custom OD settingsEvan Quan2-16/+16
`FeatureCtrlMask` should not be included in those settings interested. Signed-off-by: Evan Quan <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: Fix temperature unit of SMU v13.0.6Lijo Lazar1-3/+6
Temperature needs to be reported in millidegree Celsius. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Yang Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd: Use pci_dev_id() to simplify the codeXiongfeng Wang1-4/+4
PCI core API pci_dev_id() can be used to get the BDF number for a pci device. We don't need to compose it mannually. Use pci_dev_id() to simplify the code a little bit. Signed-off-by: Xiongfeng Wang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: fix double assign skip process context clearJonathan Kim1-1/+0
Remove redundant assignment when skipping process ctx clear. Signed-off-by: Jonathan Kim <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Update replay for clk_mgr optimizationsBhawanpreet Lakha1-0/+3
Add Replay calls to clk_mgr updates (just like PSR) Signed-off-by: Bhawanpreet Lakha <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: Fix identifier names to function definition arguments in atom.hSrinivasan Shanmugam1-10/+13
Fixes the following: WARNING: function definition argument 'struct card_info *' should also have an identifier name WARNING: function definition argument 'uint32_t' should also have an identifier name WARNING: function definition argument 'void *' should also have an identifier name WARNING: function definition argument 'struct atom_context *' should also have an identifier name WARNING: function definition argument 'int' should also have an identifier name WARNING: function definition argument 'uint32_t *' should also have an identifier name WARNING: Unnecessary space before function pointer name ERROR: space prohibited after that '*' (ctx:BxW) CHECK: Prefer kernel type 'u32' over 'uint32_t' Cc: Guchun Chen <[email protected]> Cc: Christian König <[email protected]> Cc: Alex Deucher <[email protected]> Cc: "Pan, Xinhui" <[email protected]> Signed-off-by: Srinivasan Shanmugam <[email protected]> Reviewed-by: Guchun Chen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu/pm: fix throttle_status for other than MP1 11.0.7Umio Yasuno1-7/+7
Use the right metrics table version based on the firmware. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2720 Reviewed-by: Evan Quan <[email protected]> Signed-off-by: Umio Yasuno <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: mode1 reset needs to recover mp1 for mp0 v13_0_10YiPeng Chai3-1/+30
Mode1 reset needs to recover mp1 in fatal error case for mp0 v13_0_10. v2: Define a macro to wrap psp function calls. Signed-off-by: YiPeng Chai <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: avoid driver getting empty metrics table for the first timeYang Wang1-5/+15
add metrics.AccumulationCouter check to avoid driver getting an empty metrics data since metrics table not updated completely in pmfw side. Signed-off-by: Yang Wang <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Reviewed-by: Asad Kamal <[email protected]> Tested-by: Asad Kamal <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: Use memdup_user() rather than duplicating its implementationAtul Raut1-9/+1
To prevent its redundant implementation and streamline code, use memdup_user. This fixes warnings reported by Coccinelle: ./drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c:2811:13-20: WARNING opportunity for memdup_user Signed-off-by: Atul Raut <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: Remove unnecessary ras cap checkHawking Zhang1-4/+0
RAS global isr will only be invoked by hardware interrupt. Don't need to query ras capability in isr In addition, amdgpu_ras_interrupt_fatal_error_handler ensures the isr won't be called from guest linux side by accident. The RAS cap check in isr that introduced to fix sriov crash is not needed any more Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Tao Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: fix build failure without CONFIG_DYNAMIC_DEBUGArnd Bergmann1-0/+6
When CONFIG_DYNAMIC_DEBUG is disabled altogether, calling _dynamic_func_call_no_desc() does not work: drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c: In function 'svm_range_set_attr': drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c:52:9: error: implicit declaration of function '_dynamic_func_call_no_desc' [-Werror=implicit-function-declaration] 52 | _dynamic_func_call_no_desc("svm_range_dump", svm_range_debug_dump, svms) | ^~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c:3564:9: note: in expansion of macro 'dynamic_svm_range_dump' 3564 | dynamic_svm_range_dump(svms); | ^~~~~~~~~~~~~~~~~~~~~~ Add a compile-time conditional in addition to the runtime check. Fixes: 8923137dbe4b ("drm/amdkfd: avoid svm dump when dynamic debug disabled") Signed-off-by: Arnd Bergmann <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: disable mcbp if parameter zero is setJiadong Zhu2-5/+6
The parameter amdgpu_mcbp shall have priority against the default value calculated from the chip version. User could disable mcbp by setting the parameter mcbp as zero. v2: do not trigger preemption in sw ring muxer when mcbp is disabled. Signed-off-by: Jiadong Zhu <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: Add vclk and dclk sysnode for GC 9.4.3Asad Kamal1-2/+4
Expose sysfs vclck and dclk entries for GC version 9.4.3 Signed-off-by: Asad Kamal <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/pm: disallow the fan setting if there is no fan on smu 13.0.0Kenneth Feng1-0/+4
drm/amd/pm: disallow the fan setting if there is no fan on smu 13.0.0 V2: depend on pm.no_fan to check Signed-off-by: Kenneth Feng <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Lijo Lazar <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdkfd: Add missing tba_hi programming on aldebaranJay Cornwall1-0/+1
Previously asymptomatic because high 32 bits were zero. Fixes: 96c211f1f9ef ("drm/amdkfd: Relocate TBA/TMA to opposite side of VM hole") Signed-off-by: Jay Cornwall <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amdgpu: Fix missing comment for mb() in 'amdgpu_device_aper_access'Srinivasan Shanmugam1-0/+6
This patch adds the missing code comment for memory barrier WARNING: memory barrier without comment + mb(); WARNING: memory barrier without comment + mb(); Cc: Guchun Chen <[email protected]> Cc: Christian König <[email protected]> Cc: Alex Deucher <[email protected]> Cc: "Pan, Xinhui" <[email protected]> Signed-off-by: Srinivasan Shanmugam <[email protected]> Reviewed-by: Guchun Chen <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-15drm/amd/display: Add Replay supported/enabled checksBhawanpreet Lakha3-0/+14
- Add checks for Cursor update and dirty rects (sending updates to dmub) - Add checks for dc_notify_vsync, and fbc and subvp Signed-off-by: Bhawanpreet Lakha <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdgpu: don't allow userspace to create a doorbell BOAlex Deucher1-0/+4
We need the domains in amdgpu_drm.h for the kernel driver to manage the pool, but we don't want userspace using it until the code is ready. So reject for now. Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdkfd: rename device_queue_manager_init_v10_navi10()Alex Deucher3-3/+3
Drop the navi10 in the name for consistency with other families. All gfx10 parts use the same implementation. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdkfd: drop IOMMUv2 supportAlex Deucher24-1085/+64
Now that we use the dGPU path for all APUs, drop the IOMMUv2 support. v2: drop the now unused queue manager functions for gfx7/8 APUs Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdkfd: disable IOMMUv2 support for RavenAlex Deucher1-7/+0
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdkfd: disable IOMMUv2 support for KV/CZAlex Deucher2-13/+1
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. v2: use the dGPU queue manager functions Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-11drm/amdkfd: ignore crat by defaultAlex Deucher1-4/+0
We are dropping the IOMMUv2 path, so no need to enable this. It's often buggy on consumer platforms anyway. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: Use local64_try_cmpxchg in amdgpu_perf_readUros Bizjak1-3/+2
Use local64_try_cmpxchg instead of local64_cmpxchg (*ptr, old, new) == old in amdgpu_perf_read. x86 CMPXCHG instruction returns success in ZF flag, so this change saves a compare after cmpxchg (and related move instruction in front of cmpxchg). Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg fails. There is no need to re-read the value in the loop. No functional change intended. 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]> Signed-off-by: Uros Bizjak <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdkfd: disable IOMMUv2 support for RavenAlex Deucher1-7/+0
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdkfd: disable IOMMUv2 support for KV/CZAlex Deucher2-13/+1
Use the dGPU path instead. There were a lot of platform issues with IOMMU in general on these chips due to windows not enabling IOMMU at the time. The dGPU path has been used for a long time with newer APUs and works fine. This also paves the way to simplify the driver significantly. v2: use the dGPU queue manager functions Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdkfd: ignore crat by defaultAlex Deucher1-4/+0
We are dropping the IOMMUv2 path, so no need to enable this. It's often buggy on consumer platforms anyway. Reviewed-by: Felix Kuehling <[email protected]> Acked-by: Christian König <[email protected]> Tested-by: Mike Lothian <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu/gfx11: only enable CP GFX shadowing on SR-IOVAlex Deucher1-2/+6
This is only required for SR-IOV world switches, but it adds additional latency leading to reduced performance in some benchmarks. Disable for now on bare metal. Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amd/display: Fix a regression on Polaris cardsMario Limonciello1-1/+1
DCE products don't define a `remove_stream_from_ctx` like DCN ones do. This means that when compute_mst_dsc_configs_for_state() is called it always returns -EINVAL which causes MST to fail to setup. Cc: [email protected] # 6.4.y Cc: Harry Wentland <[email protected]> Reported-by: [email protected] Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/2671 Fixes: efa4c4df864e ("drm/amd/display: call remove_stream_from_ctx from res_pool funcs") Signed-off-by: Mario Limonciello <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2023-08-09drm/amdgpu: fix possible UAF in amdgpu_cs_pass1()Alex Deucher1-1/+1
Since the gang_size check is outside of chunk parsing loop, we need to reset i before we free the chunk data. Suggested by Ye Zhang (@VAR10CK) of Baidu Security. Reviewed-by: Guchun Chen <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2023-08-09drm/amd/pm: Fix SMU v13.0.6 energy reportingLijo Lazar1-3/+2
Energy counter should be reported in units of 15.259 uJ. Don't apply any conversion. Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.