- 30 Dec, 2021 25 commits
-
-
Wenjing Liu authored
[why] Update all accesses to use hpo dp link encoder through link resource only. Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Reviewed-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Signed-off-by:
Wenjing Liu <wenjing.liu@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Wenjing Liu authored
[why] This commit is to populate link res in preparation of the next commit. The next commit will replace all existing code to use link res instead Tested-by:
-
Wenjing Liu authored
[why] There will be a series of re-arch changes in Link Resource Management. They are more and more muxable link resource objects and the resource is insufficient for a one to one allocation to all links created. Therefore a link resource sharing logic is required to determine which link should use certain link resource. This commit is the first one in this series that starts by defining a link resource struct, this struct will be available to all interfaces that need to perform link programming sequence. In later commits, we will granduately decouple link resource objects out of dc link. So instead of access a link resource from dc link. Current link's resource can be accessible through pipe_ctx->link_res during commit, or by calling dc_link_get_cur_link_res function with current link passed in after commit. Tested-by:
-
Aric Cyr authored
This version brings along the following: - Fixes and improvements in the LTTPR code - Improve z-state - Fix null pointer check - Improve communication with s0i2 - Update multiple-display split policy - Add missing registers Tested-by:
Aric Cyr <aric.cyr@amd.com> Signed-off-by:
-
Anthony Koo authored
Tested-by:
Anthony Koo <Anthony.Koo@amd.com> Signed-off-by:
-
Martin Leung authored
Undo ODM Combine regression causing causing pipe allocation issues. Tested-by:
Martin Leung <Martin.Leung@amd.com> Signed-off-by:
-
Wesley Chalmers authored
[WHY] These registers are currently missing from the DCN303 header files Tested-by:
George Shen <George.Shen@amd.com> Acked-by:
Wesley Chalmers <Wesley.Chalmers@amd.com> Signed-off-by:
-
Angus Wang authored
[WHY] Current implementation of pipe split policy prevents pipe split with multiple displays connected, which caused the MCLK speed to be stuck at max [HOW] Changed the pipe split policies so that pipe split is allowed for multi-display configurations Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1522 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1709 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1655 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1403Tested-by:
Angus Wang <angus.wang@amd.com> Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] We'll exit optimized power state to do link detection but we won't enter back into the optimized power state. This could potentially block s2idle entry depending on the sequencing, but it also means we're losing some power during the transition period. [How] Hook up the handler like DCN21. It was also missed like the exit_optimized_pwr_state callback. Fixes: 64b1d0e8 ("drm/amd/display: Add DCN3.1 HWSEQ") Tested-by:
Eric Yang <Eric.Yang2@amd.com> Acked-by:
Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by:
-
George Shen authored
[Why] DP spec specifies that DPRX shall use the read interval in the TRAINING_AUX_RD_INTERVAL_PHY_REPEATER LTTPR DPCD register. This register's bit definition is the same as the AUX read interval register for DPRX. [How} Remove logic which forces AUX read interval to 100us for repeaters when in LTTPR non-transparent mode. Tested-by:
-
Nicholas Kazlauskas authored
[Why] Otherwise SMU won't mark Display as idle when trying to perform s2idle. [How] Mark the bit in the dcn31 codepath, doesn't apply to older ASIC. It needed to be split from phy refclk off to prevent entering s2idle when PSR was engaged but driver was not ready. Fixes: 118a3315 ("drm/amd/display: Add DCN3.1 clock manager support") Tested-by:
-
Alvin Lee authored
[Why] Bug fix for null function ptr (should check for NULL instead of not NULL) [How] Fix if condition Tested-by:
Samson Tam <samson.tam@amd.com> Acked-by:
Alvin Lee <Alvin.Lee2@amd.com> Signed-off-by:
-
Lai, Derek authored
[Why] The change of setting a timer callback on boot for 10 seconds is still working, just lacked power down for DCN10. [How] Added power down for DCN10. Tested-by:
Derek Lai <Derek.Lai@amd.com> Signed-off-by:
-
Nicholas Kazlauskas authored
[Why] Stutter period won't be less than 5000.0, but if PSR is enabled then we can potentially enter Z9 when MPO is enabled. SMU will try to enter Z9 too early in these cases (before PSR is enabled) and we'll see underflow. [How] Block z-states (z9, z10) until we can add a new interface to SMU to signal when we can support z10 but not z9. We can revert this once the interface change is in. Tested-by:
-
Shen, George authored
[Why] Current implementation is not scalable and retrofits the existing standard link training code for purposes outside of its original design. [How] Refactor vendor specific link training sequence into its own separate function to be called instead of the standard link training function. Tested-by:
-
George Shen authored
[Why] Max link rate should be limited to the maximum link rate support by any LTTPR that are connected, including when operating in transparent mode. [How] Include transparent mode when factoring in LTTPR max supported link rate. Tested-by:
-
Charlene Liu authored
[why] B0 PHY C map to F, D map to G driver use logic instance, dmub does the remap. Driver still need use the right PHY instance to access right HW. [how] use phyical instance when program PHY register. [note] could move resync_control programming to dmub next. Tested-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Signed-off-by:
-
Surbhi Kakarya authored
If the event guard is enabled and VF doesn't receive an ack from PF for full access, the guest driver load crashes. This is caused due to the call to ttm_device_clear_dma_mappings with non-initialized mman during driver tear down. This patch adds the necessary condition to check if the mman initialization passed or not and takes the path based on the condition output. Signed-off-by:
Surbhi Kakarya <Surbhi.Kakarya@amd.com> Acked-by:
-
Guchun Chen authored
This patch keeps the setting of sdma queue number to the same after recent KFD code refactor. Additionally, improve code to use switch case to list IP version to complete kfd device_info structure filling for IH version assignment. This makes consistency with the IP parse code in amdgpu_discovery.c. v2: use dev_warn for the default switch case; set default sdma queue per engine(8) and IH handler to v9. (Jonathan) v3: Fix missed IP version check of Raven. Fixes: f0dc99a6 ("drm/amdkfd: add kfd_device_info_init function") Signed-off-by:Guchun Chen <guchun.chen@amd.com> Reviewed-by:
Jonathan Kim <jonathan.kim@amd.com> Reviewed-by:
Graham Sider <Graham.Sider@amd.com> Signed-off-by:
-
Kent Russell authored
This is supported, although the offset is different from VG20, so fix that with a variable and enable getting the product name and serial number from the FRU. Do this for all SKUs since all SKUs have the FRU Signed-off-by:
Kent Russell <kent.russell@amd.com> Reviewed-by:
-
Kent Russell authored
On Aldebaran, the serial may be obtained from the FRU. Only overwrite the serial with the unique_id if the serial is empty. This will support printing serial numbers for mGPU devices where there are 2 unique_ids for the 2 GPUs, but only one serial number for the board Signed-off-by:
-
Kent Russell authored
It's supported, so support the unique_id sysfs file Signed-off-by:
-
Kent Russell authored
Having seen at least 1 42-character product_name, bump the number up to 64, and put that definition into amdgpu.h to make future adjustments simpler. Signed-off-by:
-
yipechai authored
The psp bootloader functions code of psp_v13_0.c had been optimized before. According the code style of psp_v13_0.c to remove the redundant code of psp_v11_0.c. v2: squash in drop unused variable (Alex) Signed-off-by:
yipechai <YiPeng.Chai@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
Leslie Shi authored
drm/amdgpu: Call amdgpu_device_unmap_mmio() if device is unplugged to prevent crash in GPU initialization failure [Why] In amdgpu_driver_load_kms, when amdgpu_device_init returns error during driver modprobe, it will start the error handle path immediately and call into amdgpu_device_unmap_mmio as well to release mapped VRAM. However, in the following release callback, driver stills visits the unmapped memory like vcn.inst[i].fw_shared_cpu_addr in vcn_v3_0_sw_fini. So a kernel crash occurs. [How] call amdgpu_device_unmap_mmio() if device is unplugged to prevent invalid memory address in vcn_v3_0_sw_fini() when GPU initialization failure. Signed-off-by:
Leslie Shi <Yuliang.Shi@amd.com> Reviewed-by:
Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
- 28 Dec, 2021 14 commits
-
-
Zongmin Zhou authored
Some boards(like RX550) seem to have garbage in the upper 16 bits of the vram size register. Check for this and clamp the size properly. Fixes boards reporting bogus amounts of vram. after add this patch,the maximum GPU VRAM size is 64GB, otherwise only 64GB vram size will be used. Signed-off-by: Zongmin Zhou<zhouzongmin@kylinos.cn> Signed-off-by: -
Nicholas Kazlauskas authored
[Why] A porting error on a previous patch left the block of code that causes the crash from a NULL pointer dereference. More specifically, we try to access link_enc before it's assigned in the USB4 case in the following assignment: config.dio_output_idx = link_enc->transmitter - TRANSMITTER_UNIPHY_A; [How] That assignment occurs later depending on the ASIC version. It's only needed on DCN31 and only after link_enc is already assigned. Fixes: 98643044 ("drm/amd/display: fix a crash on USB4 over C20 PHY") Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
sashank saye authored
For Aldebaran chip passthrough case we need to intimate SMU about special handling for SBR.On older chips we send LightSBR to SMU, enabling the same for Aldebaran. Slight difference, compared to previous chips, is on Aldebaran, SMU would do a heavy reset on SBR. Hence, the word Heavy instead of Light SBR is used for SMU to differentiate. Reviewed by: Shaoyun.liu <Shaoyun.liu@amd.com> Signed-off-by:
sashank saye <sashank.saye@amd.com> Signed-off-by:
-
Rajneesh Bhardwaj authored
When an application having open file access to a node forks, its shared mappings also get reflected in the address space of child process even though it cannot access them with the object permissions applied. With the existing permission checks on the gem objects, it might be reasonable to also create the VMAs with VM_DONTCOPY flag so a user space application doesn't need to explicitly call the madvise(addr, len, MADV_DONTFORK) system call to prevent the pages in the mapped range to appear in the address space of the child process. It also prevents the memory leaks due to additional reference counts on the mapped BOs in the child process that prevented freeing the memory in the parent for which we had worked around earlier in the user space inside the thunk library. Additionally, we faced this issue when using CRIU to checkpoint restore an application that had such inherited mappings in the child which confuse CRIU when it mmaps on restore. Having this flag set for the render node VMAs helps. VMAs mapped via KFD already take care of this so this is needed only for the render nodes. To limit the impact of the change to user space consumers such as OpenGL etc, limit it to KFD BOs only. Acked-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
David Yat Sin <david.yatsin@amd.com> Signed-off-by:
Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by:
-
Tao Zhou authored
CP supports unmap queue with reset mode which only destroys specific queue without affecting others. Replacing whole gpu reset with reset queue mode for RAS poison consumption saves much time, and we can also fallback to gpu reset solution if reset queue fails. v2: Return directly if process is NULL; Reset queue solution is not applicable to SDMA, fallback to legacy way; Call kfd_unref_process after lookup process. Signed-off-by:Tao Zhou <tao.zhou1@amd.com> Reviewed-by:
-
Tao Zhou authored
The new interface unmaps queues with reset mode for the process consumes RAS poison, it's only for compute queue. v2: rename the function to reset_queues. Signed-off-by:
-
Tao Zhou authored
So we can set reset mode for unmap operation, no functional change. Signed-off-by:
-
Tao Zhou authored
Add a reset parameter for umc page retirement, let user decide whether call gpu reset in umc page retirement. Signed-off-by:
-
Victor Skvortsov authored
Expand RLCG interface for new GC read & write commands. New interface will only be used if the PF enables the flag in pf2vf msg. v2: Added a description for the scratch registers Signed-off-by:
Victor Skvortsov <victor.skvortsov@amd.com> Reviewed-by:
David Nieto <david.nieto@amd.com> Signed-off-by:
-
Victor Skvortsov authored
Driver needs to call get_xgmi_info() before ip_init to determine whether it needs to handle a pending hive reset. Signed-off-by:
-
Victor Skvortsov authored
Modify GC register access from MMIO to RLCG if the indirect flag is set Signed-off-by:
-
Victor Skvortsov authored
Modify GC register access from MMIO to RLCG if the indirect flag is set v2: Replaced ternary operator with if-else for better readability Signed-off-by:
-
Victor Skvortsov authored
Add helper macros to change register access from direct to indirect. Signed-off-by:
-
Bokun Zhang authored
Recently, there is security policy update under SRIOV. We need to filter the registers that hit the violation and move the code to the host driver side so that the guest driver can execute correctly. Signed-off-by:
Bokun Zhang <Bokun.Zhang@amd.com> Reviewed-by:
-
- 16 Dec, 2021 1 commit
-
-
Alex Deucher authored
Used on gfx9 based systems. Fixes incorrect CU counts reported in the kernel log. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1833Reviewed-by:
-
