- 09 Apr, 2021 40 commits
-
-
Wyatt Wood authored
[Why] Today a parameter is passed to dmub_srv_send_gpint_command which specifies the max amount of time to wait for the command to be acked. This parameter instead specifies the number of times to check if the command was acked. We wish to add a lower bound to the timeout, which can be accomplished by using a delay or a timestamp. It has been decided to use udelay instead of a timestamp simply because we don't want to expose a high precision counter if it doesn't yield large benefits. [How] Add udelay(1) between each for loop iteration. Signed-off-by:
Wyatt Wood <wyatt.wood@amd.com> Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
Qingqing Zhuo <qingqing.zhuo@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Wyatt Wood authored
[Why] GPINT command to get PSR state from FW times out. [How] Add retry to get valid PSR state. Signed-off-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Hawking Zhang authored
gfx ras is only available in cerntain ip generations. Signed-off-by:
Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by:
Dennis Li <Dennis.Li@amd.com> Reviewed-by:
John Clements <John.Clements@amd.com> Signed-off-by:
-
Hawking Zhang authored
mmhub ras is only avaiable in cerntain mmhub ip generation. Signed-off-by:
-
Hawking Zhang authored
df/mca ras is not managed by gpu driver when gpu is connected to cpu through xgmi. gpu driver should register x86 mca notifier for umc ras error notification Signed-off-by:
-
Hawking Zhang authored
umc ras is not managed by gpu driver when gpu is connected to cpu through xgmi. split umc callbacks into ras and non-ras ones so gpu driver only initializes umc ras callbacks when it manages umc ras. Signed-off-by:
-
Hawking Zhang authored
xgmi ras is not managed by gpu driver when gpu is connected to cpu through xgmi. move all xgmi ras functions to xgmi_ras_funcs so gpu driver only initializes xgmi ras functions when it manages xgmi ras. Signed-off-by:
-
Hawking Zhang authored
nbio ras is not managed by gpu driver when gpu is connected to cpu through xgmi. split nbio callbacks into ras and non-ras ones so gpu driver only initializes nbio ras callbacks when it manages nbio ras. Signed-off-by:
-
Hawking Zhang authored
query_ras_error_address will be invoked to query bad page address when there is poison data in HBM consumed by GPU engines. Signed-off-by:
-
Hawking Zhang authored
umc query_ras_error_count will be invoked to query umc correctable and uncorrectable error. It will reset the umc ras error counter after the query. Signed-off-by:
-
Hawking Zhang authored
Add helper functions to query correctable and uncorrectable umc ras error. Signed-off-by:
-
Hawking Zhang authored
umc_v6_7_funcs are callbacks to support umc ras functionalities in aldebaran Signed-off-by:
-
Hawking Zhang authored
Driver only manages GFX/SDMA/MMHUB RAS in platforms that gpu node is connected to cpu through XGMI, other than that, it queries VBIOS for RAS capabilities. Signed-off-by:
-
Amber Lin authored
Power Management IP is initialized/enabled before KFD init. When a thermal throttling happens before kfd_smi_init is done, calling the KFD SMI update function causes a stack dump by referring a NULL pointer ( smi_clients list). Check if kfd_init is completed before calling the function. Signed-off-by:
Amber Lin <Amber.Lin@amd.com> Reviewed-by:
Mukul Joshi <mukul.joshi@amd.com> Signed-off-by:
-
Wan Jiabing authored
struct mod_hdcp is declared twice. One is declared at 33rd line. The blew one is not needed. Remove the duplicate. Signed-off-by:
Wan Jiabing <wanjiabing@vivo.com> Signed-off-by:
-
Alex Deucher authored
These were leftover from the old CI dpm code which was retired a while ago. Acked-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Bernard Zhao authored
Fix patch check warning: WARNING: suspect code indent for conditional statements (8, 17) + if (obj && obj->use < 0) { + DRM_ERROR("RAS ERROR: Unbalance obj(%s) use\n", obj->head.name); WARNING: braces {} are not necessary for single statement blocks + if (obj && obj->use < 0) { + DRM_ERROR("RAS ERROR: Unbalance obj(%s) use\n", obj->head.name); + } Signed-off-by:Bernard Zhao <bernard@vivo.com> Signed-off-by:
-
Bernard Zhao authored
Fix patch check warning: WARNING: Prefer kmalloc_array over kmalloc with multiply + buf = kmalloc(MAX_KFIFO_SIZE * sizeof(*buf), GFP_KERNEL); Signed-off-by:
-
Stanley.Yang authored
Signed-off-by:
Stanley.Yang <Stanley.Yang@amd.com> Reivewed-by:
-
Philip Yang authored
Forgot to reserve a fence slot to use sdma to update page table, cause below kernel BUG backtrace to handle vm retry fault while application is exiting. [ 133.048143] kernel BUG at /home/yangp/git/compute_staging/kernel/drivers/dma-buf/dma-resv.c:281! [ 133.048487] Workqueue: events amdgpu_irq_handle_ih1 [amdgpu] [ 133.048506] RIP: 0010:dma_resv_add_shared_fence+0x204/0x280 [ 133.048672] amdgpu_vm_sdma_commit+0x134/0x220 [amdgpu] [ 133.048788] amdgpu_vm_bo_update_range+0x220/0x250 [amdgpu] [ 133.048905] amdgpu_vm_handle_fault+0x202/0x370 [amdgpu] [ 133.049031] gmc_v9_0_process_interrupt+0x1ab/0x310 [amdgpu] [ 133.049165] ? kgd2kfd_interrupt+0x9a/0x180 [amdgpu] [ 133.049289] ? amdgpu_irq_dispatch+0xb6/0x240 [amdgpu] [ 133.049408] amdgpu_irq_dispatch+0xb6/0x240 [amdgpu] [ 133.049534] amdgpu_ih_process+0x9b/0x1c0 [amdgpu] [ 133.049657] amdgpu_irq_handle_ih1+0x21/0x60 [amdgpu] [ 133.049669] process_one_work+0x29f/0x640 [ 133.049678] worker_thread+0x39/0x3f0 [ 133.049685] ? process_one_work+0x640/0x640 Signed-off-by:
Philip Yang <Philip.Yang@amd.com> Signed-off-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
-
Peng Ju Zhou authored
1. expand rlcg interface for gc & mmhub indirect access 2. add rlcg interface for no kiq v2: squash in fix for gfx9 (Changfeng) Signed-off-by:
Peng Ju Zhou <PengJu.Zhou@amd.com> Reviewed-by:
Emily.Deng <Emily.Deng@amd.com> Signed-off-by:
-
Peng Ju Zhou authored
using the control bits got from host to control registers access. Signed-off-by:
-
Peng Ju Zhou authored
get pf2vf msg info at it's earliest time so that guest driver can use these info to decide whether register indirect access enabled. Signed-off-by:
-
Peng Ju Zhou authored
unify host driver and guest driver indirect access control bits names Signed-off-by:
-
Alex Deucher authored
Use uintptr_t rather than uint64_t. Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
-
Xℹ Ruoyao authored
The page table of AMDGPU requires an alignment to CPU page so we should check ioctl parameters for it. Return -EINVAL if some parameter is unaligned to CPU page, instead of corrupt the page table sliently. Reviewed-by:
Xi Ruoyao <xry111@mengyan1223.wang> Signed-off-by:
-
Huacai Chen authored
In Mesa, dev_info.gart_page_size is used for alignment and it was set to AMDGPU_GPU_PAGE_SIZE(4KB). However, the page table of AMDGPU driver requires an alignment on CPU pages. So, for non-4KB page system, gart_page_size should be max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE). Signed-off-by:
Rui Wang <wangr@lemote.com> Signed-off-by:
Huacai Chen <chenhc@lemote.com> Link: https://github.com/loongson-community/linux-stable/commit/caa9c0a1 [Xi: rebased for drm-next, use max_t for checkpatch, and reworded commit message.] Signed-off-by:
Dan Horák <dan@danny.cz> Reviewed-by:
-
Guchun Chen authored
Leverage the same logic from amdgpu_ttm_tt_unpin_userptr. Signed-off-by:
Guchun Chen <guchun.chen@amd.com> Reviewed-by:
-
Guchun Chen authored
warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement] int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); v2: put short variable declaration last Signed-off-by:
-
Guchun Chen authored
ttm->sg needs to be checked before accessing its child member. Call Trace: amdgpu_ttm_backend_destroy+0x12/0x70 [amdgpu] ttm_bo_cleanup_memtype_use+0x3a/0x60 [ttm] ttm_bo_release+0x17d/0x300 [ttm] amdgpu_bo_unref+0x1a/0x30 [amdgpu] amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu+0x78b/0x8b0 [amdgpu] kfd_ioctl_alloc_memory_of_gpu+0x118/0x220 [amdgpu] kfd_ioctl+0x222/0x400 [amdgpu] ? kfd_dev_is_large_bar+0x90/0x90 [amdgpu] __x64_sys_ioctl+0x8e/0xd0 ? __context_tracking_exit+0x52/0x90 do_syscall_64+0x33/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7f97f264d317 Code: b3 66 90 48 8b 05 71 4b 2d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 41 4b 2d 00 f7 d8 64 89 01 48 RSP: 002b:00007ffdb402c338 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007f97f3cc63a0 RCX: 00007f97f264d317 RDX: 00007ffdb402c380 RSI: 00000000c0284b16 RDI: 0000000000000003 RBP: 00007ffdb402c380 R08: 00007ffdb402c428 R09: 00000000c4000004 R10: 00000000c4000004 R11: 0000000000000246 R12: 00000000c0284b16 R13: 0000000000000003 R14: 00007f97f3cc63a0 R15: 00007f8836200000 Signed-off-by:
-
Victor Lu authored
[why] Our CI enables drm.debug=0x4 logs and the dmesg is flooded with cursor updates. We probably want to avoid spamming the log with DRM_DEBUG_KMS. [how] Define and use pr_debug macros instead of a few spammy DRM_DEBUG_*'s. Signed-off-by:
Victor Lu <victorchengchi.lu@amd.com> Reviewed-by:
-
Guchun Chen authored
Otherwise, below errors will be found on DIMGREY_CAVEFISH with DCN302. Error log observed in driver load: [drm:amdgpu_dm_irq_register_interrupt [amdgpu]] *ERROR* DM_IRQ: invalid irq_source: 0! Error observed in mode1_rest sequence: [ 27.265920] #PF: supervisor read access in kernel mode [ 27.265941] #PF: error_code(0x0000) - not-present page [ 27.265956] PGD 0 P4D 0 [ 27.265967] Oops: 0000 [#1] SMP NOPTI [ 27.265979] CPU: 0 PID: 1360 Comm: cat Tainted: G W 5.11.0-99b3786c1770 #20210323 [ 27.266005] Hardware name: System manufacturer System Product Name/PRIME Z390-A, BIOS 1401 11/26/2019 [ 27.266033] RIP: 0010:dal_irq_service_ack+0x25/0x60 [amdgpu] [ 27.266203] Code: 5d 5d c3 66 90 0f 1f 44 00 00 55 83 fe 61 48 89 e5 77 27 89 f0 48 8d 04 40 48 c1 e0 04 48 03 47 08 74 17 48 8b 50 28 48 89 c6 <48> 8b 52 08 48 85 d2 74 20 e8 1d 64 45 c9 5d c3 89 f2 48 c7 c7 f0 [ 27.266248] RSP: 0018:ffffa115824a3c08 EFLAGS: 00010082 [ 27.266270] RAX: ffffffffc0942c10 RBX: ffffffffc0942c10 RCX: 0000000000000000 [ 27.266288] RDX: 0000000000000000 RSI: ffffffffc0942c10 RDI: ffff88d509cba7a0 [ 27.266312] RBP: ffffa115824a3c08 R08: 0000000000000000 R09: 0000000000000001 [ 27.266335] R10: ffffa115824a3b20 R11: ffffa115824a3b58 R12: ffff88d509cba7a0 [ 27.266353] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000246 [ 27.266377] FS: 00007fb3e2438580(0000) GS:ffff88d50dc00000(0000) knlGS:0000000000000000 [ 27.266402] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 27.266417] CR2: 0000000000000008 CR3: 000000083e5ac006 CR4: 00000000003706f0 [ 27.266441] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 27.266464] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 27.266483] Call Trace: [ 27.266491] dal_irq_service_set+0x31/0x80 [amdgpu] [ 27.266639] dc_interrupt_set+0x24/0x30 [amdgpu] [ 27.266775] amdgpu_dm_set_dmub_trace_irq_state+0x22/0x30 [amdgpu] [ 27.266920] amdgpu_irq_update+0x57/0xa0 [amdgpu] [ 27.267030] amdgpu_irq_gpu_reset_resume_helper+0x64/0xa0 [amdgpu] [ 27.267135] amdgpu_do_asic_reset+0x211/0x420 [amdgpu] [ 27.267232] amdgpu_device_gpu_recover+0x517/0xa70 [amdgpu] [ 27.267325] gpu_recover_get+0x2e/0x60 [amdgpu] [ 27.267421] simple_attr_read+0x6b/0x130 [ 27.267441] debugfs_attr_read+0x49/0x70 [ 27.267454] full_proxy_read+0x5c/0x90 [ 27.267474] vfs_read+0xa1/0x190 [ 27.267486] ksys_read+0xa7/0xe0 [ 27.267501] __x64_sys_read+0x1a/0x20 [ 27.267521] do_syscall_64+0x37/0x80 [ 27.267541] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 27.267562] RIP: 0033:0x7fb3e2356142 Signed-off-by:
Leo (Hanghong) Ma <hanghong.ma@amd.com> Signed-off-by:
-
Rohit Khaire authored
Add 3 sub flags to notify guest for indirect reg access of gc, mmhub and ih The host sets these flags depending on L1 RAP version, asic and other scenarios. These flags ensure that there is compatibility between different guest/host/vbios versions. Signed-off-by:
Rohit Khaire <rohit.khaire@amd.com> Reviewed-by:
Monk Liu <monk.liu@amd.com> Acked-by:
Luben Tuikov <luben.tuikov@amd.com> Signed-off-by:
-
Alex Deucher authored
Do the same thing we do for Renoir. We can check, but since the sbios has started DPM, it will always return true which causes the driver to skip some of the SMU init when it shouldn't. Reviewed-by:
Zhan Liu <zhan.liu@amd.com> Acked-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Diego Viola authored
Signed-off-by:
Diego Viola <diego.viola@gmail.com> Signed-off-by:
-
Werner Sembach authored
When encoder validation of a display mode fails, retry with less bandwidth heavy YCbCr420 color mode, if available. This enables some HDMI 1.4 setups to support 4k60Hz output, which previously failed silently. On some setups, while the monitor and the gpu support display modes with pixel clocks of up to 600MHz, the link encoder might not. This prevents YCbCr444 and RGB encoding for 4k60Hz, but YCbCr420 encoding might still be possible. However, which color mode is used is decided before the link encoder capabilities are checked. This patch fixes the problem by retrying to find a display mode with YCbCr420 enforced and using it, if it is valid. Reviewed-by:
Werner Sembach <wse@tuxedocomputers.com> Signed-off-by:
-
Aric Cyr authored
This version brings along following features/fixes: - LTTPR improvements - Disable MALL when SMU not present - Fix bug in HW that causes P-State to hang when DPG is enabled in certain conditions - Update code path for enabling DPG - Update display endpoint control path - Add dynamic link encoder selection - Fix debugfs MST topology to dump from the root MST node - Enable DP DSC Compliance automation for Linux - ASSR is enabled only when DPCD is supported and the display connected is internal - Added kernel trace event to print real-time refresh rate value to debug VRR issues Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Acked-by:
Anson Jacob <Anson.Jacob@amd.com> Tested-by:
Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by:
-
Chris Park authored
[Why] Bring-up purpose code to disable DMUB calling into SMU and timeout for MALL when SMU is not present. [How] Disable MALL when SMU is not present. Signed-off-by:
Chris Park <Chris.Park@amd.com> Reviewed-by:
-
Anthony Koo authored
Signed-off-by:
-
Qingqing Zhuo authored
[Why&How] Add log for easier debug purposes. Signed-off-by:
-
