- 04 May, 2022 40 commits
-
-
Jack Xiao authored
Use per context sdma csa address for mes sdma queue. Signed-off-by:
Jack Xiao <Jack.Xiao@amd.com> Acked-by:
Christian König <christian.koenig@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Jack Xiao authored
If MES is enabled, don't use kiq to flush gpu tlb, for it would result in conflicting with mes fw. Signed-off-by:
-
Jack Xiao authored
Add mes support for gfx ib test. Signed-off-by:
-
Jack Xiao authored
From IH ring buffer, look up the coresponding kernel queue and process. Signed-off-by:
-
Jack Xiao authored
Add support for multiple mes pipes, so that reuse the existing code to initialize more mes pipe and queue. Signed-off-by:
-
Jack Xiao authored
Allocate a doorbell index for mes kiq queue. Signed-off-by:
-
Jack Xiao authored
mes_kiq parameter is used to enable mes kiq pipe. This module parameter is unneccessary or enabled by default in final version. v2: reword commit message. Signed-off-by:
-
Jack Xiao authored
Update the definitions of MES process/gang/queue. v2: add missing includes v3: rebase fix, include mm.h Signed-off-by:
-
Jack Xiao authored
Use the whole doorbell space for mes. Each queue in one process occupies one doorbell slot to ring the queue submitting. Signed-off-by:
-
Jack Xiao authored
For MES FW manages IH_VMID_x_LUT updating, skip emitting pasid mapping packet. Signed-off-by:
-
Jack Xiao authored
For MES queue VM flush, use INVALIDATE_TLBS to invalidate TLBs. This packet can let CP firmware to determine the current vmid and inv eng to invalidate. v2: unify invalidate_tlbs functions Cc: Le Ma <le.ma@amd.com> Signed-off-by:
-
Jack Xiao authored
For MES manages vmid assignment, let vmid inherit from mqd instead of ib packet setting. Signed-off-by:
-
Jack Xiao authored
Associate mes queue id with fence, so that EOP trap handler can look up which queue has issued the fence. v2: move mes queue flag to amdgpu_mes_ctx.h Signed-off-by:
-
Jack Xiao authored
As MES requires per context preemption, use per context CSA address for DE metadata to correctly enable context MCBP preemption. Signed-off-by:
-
Jack Xiao authored
As MES requires per context preemption, use per context CSA address for CE metadata to correctly enable context MCBP preemption. Signed-off-by:
-
Jack Xiao authored
Refine the existing gfx/compute mqd functions, and add them to engine mqd layer. v2: rebase fix. Signed-off-by:
-
Jack Xiao authored
assign the cpu/gpu address of fence for the normal or mes ring from ring structure. Signed-off-by:
-
Jack Xiao authored
Iniailize/finalize the ring for mes queue which submits the command stream to the mes-managed hardware queue. Signed-off-by:
-
Jack Xiao authored
Use ring structure to access the cpu/gpu address of rptr/wptr. v2: merge gfx10/sdma5/sdma5.2 patches Signed-off-by:
-
Jack Xiao authored
Define ring structure to access the cpu/gpu address of rptr/wptr/fence instead of dynamic calculation. Signed-off-by:
-
Jack Xiao authored
Add mes context data structure in amdgpu_ring. Signed-off-by:
-
Jack Xiao authored
The per-context meta data is a per-context data structure associated with a mes-managed hardware ring, which includes MCBP CSA, ring buffer and etc. v2: fix typo v3: a. use structure instead of typedef b. move amdgpu_mes_ctx_get_offs_* to amdgpu_ring.h c. use __aligned to make alignement Signed-off-by: -
Jack Xiao authored
Add the helper function to initialize mqd from ring configuration. v2: use if/else pair instead of ?/: pair v3: use simpler way to judge hqd_active v4: fix parameters to amdgpu_gfx_is_high_priority_compute_queue Signed-off-by:
-
Jack Xiao authored
Define MQD abstract layer for hw ip, for the passing mqd configuration not only from ring but more sources, like user queue. Signed-off-by:
-
Likun Gao authored
Add IMU firmware structure. Signed-off-by:
Likun Gao <Likun.Gao@amd.com> Reviewed-by:
-
Likun Gao authored
Add RLC autoload TOC header file for soc21 ASIC. v2: squash in updates Signed-off-by:
-
Evan Quan authored
Add the CG flag for Fine Grained Clock Gating. Reviewed-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Likun Gao authored
Add support for rlc v2_3 to support RLCV and RLCP fw load. Signed-off-by:
Kenneth Feng <kenneth.feng@amd.com> Signed-off-by:
-
Likun Gao authored
We need define new firmware header to support CP RS64 fw. Signed-off-by:
Wenhui Sheng <Wenhui.Sheng@amd.com> Signed-off-by:
-
Wenhui Sheng authored
Signed-off-by:
-
Hawking Zhang authored
memory queue descriptors for gfx11. Signed-off-by:
-
Hawking Zhang authored
Add gfx11 clearstate register arrays Signed-off-by:
-
Hawking Zhang authored
add soc21 enum definitions (Hawking) Signed-off-by:
-
Likun Gao authored
Set GC family for GC 11.0 IPs. Reviewed-by:
-
Hawking Zhang authored
Add GC v11_0_0 family id Signed-off-by:
-
Alice Wong authored
When psp_hw_init failed, it will set the load_type to AMDGPU_FW_LOAD_DIRECT. During amdgpu_device_ip_fini, amdgpu_ucode_free_bo checks that load_type is AMDGPU_FW_LOAD_DIRECT and skips deallocating fw_buf causing memory leak. Remove load_type check in amdgpu_ucode_free_bo. Signed-off-by:
Alice Wong <shiwei.wong@amd.com> Reviewed-by:
-
Sung Joon Kim authored
Recently we got a hard hang during the boot on DCN 3.0.1, which caused the below null pointer exception: [ +0.000426] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ +0.000003] #PF: supervisor read access in kernel mode [ +0.000003] #PF: error_code(0x0000) - not-present page [ +0.000003] PGD 0 P4D 0 [ +0.000004] Oops: 0000 [#1] PREEMPT SMP NOPTI [ +0.000005] CPU: 6 PID: 874 Comm: Xorg Not tainted 5.16.0.asdn-apr28+ #15 [ +0.000004] Hardware name: AMD Chachani-VN/Chachani-VN, BIOS WCH2303N 03/03/2022 [ +0.000003] RIP: 0010:resource_map_pool_resources+0x431/0xa70 [amdgpu] [ +0.000356] Code: c1 4d 89 c8 49 c1 e0 07 4d 01 c8 49 c1 e0 04 4d 01 f0 49 83 b8 f0 01 00 00 00 0f 85 16 02 00 00 49 8b b8 e0 02 00 00 89 45 c0 <48> 8b 17 4c 8b 92 a0 01 00 00 4d 85 d2 74 24 4c 89 4d 88 48 8d 4d [ +0.000003] RSP: 0018:ffffa92a4142f718 EFLAGS: 00010246 [ +0.000003] RAX: 0000000000000000 RBX: ffff9a0b86d93000 RCX: 0000000000000000 [ +0.000002] RDX: 0000000000000000 RSI: 000000000000554b RDI: 0000000000000000 [ +0.000002] RBP: ffffa92a4142f798 R08: ffff9a0bdb3c0000 0000000000000000 [ +0.000002] R10: 0000000000000000 R11: 000000000000f000 R12: 0000000000000000 [ +0.000001] R13: ffff9a0b88360000 R14: ffff9a0bdb3c0000 R15: ffff9a0b86273000 [ +0.000003] FS: 00007f4b5641ca40(0000) GS:ffff9a0cb7f80000(0000) knlGS:0000000000000000 [ +0.000002] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000002] CR2: 0000000000000000 CR3: 0000000102cb2000 CR4: 00000000003506e0 [ +0.000003] Call Trace: [ +0.000002] <TASK> [ +0.000004] ? kvmalloc_node+0x5c/0x90 [ +0.000009] dcn20_add_stream_to_ctx+0x1c/0x90 [amdgpu] [ +0.000330] dcn30_add_stream_to_ctx+0xe/0x10 [amdgpu] [ +0.000313] dc_add_stream_to_ctx+0x67/0x80 [amdgpu] [ +0.000300] dm_update_crtc_state+0x4dd/0x6e0 [amdgpu] [ +0.000320] amdgpu_dm_atomic_check+0x63b/0x1270 [amdgpu] [ +0.000311] ? __drm_mode_object_add+0x90/0xc0 [drm] [ +0.000043] ? preempt_count_add+0x74/0xc0 [ +0.000005] ? _raw_spin_lock_irqsave+0x2a/0x60 [ +0.000006] ? _raw_spin_unlock_irqrestore+0x29/0x3d [ +0.000003] ? drm_connector_list_iter_next+0x8e/0xb0 [drm] [ +0.000038] drm_atomic_check_only+0x5dd/0xa20 [drm] [ +0.000044] drm_atomic_commit+0x18/0x60 [drm] [ +0.000046] drm_client_modeset_commit_atomic+0x1e5/0x220 [drm] [ +0.000051] drm_client_modeset_commit_locked+0x57/0x160 [drm] [ +0.000038] __drm_fb_helper_restore_fbdev_mode_unlocked+0x60/0xd0 [drm_kms_helper] [ +0.000027] drm_fb_helper_set_par+0x40/0x50 [drm_kms_helper] [ +0.000022] fb_set_var+0x1c8/0x3d0 [ +0.000007] ? __ext4_mark_inode_dirty+0x83/0x210 [ +0.000006] ? __ext4_journal_stop+0x3c/0xb0 [ +0.000008] fbcon_blank+0x228/0x290 [ +0.000007] do_unblank_screen+0xae/0x150 [ +0.000005] vt_ioctl+0xcf4/0x1360 [ +0.000005] ? get_max_files+0x20/0x20 [ +0.000005] ? get_max_files+0x20/0x20 [ +0.000004] ? debug_smp_processor_id+0x17/0x20 [ +0.000004] tty_ioctl+0x373/0x8a0 [ +0.000005] ? __fput+0x123/0x260 [ +0.000004] ? __fget_light+0xc5/0x100 [ +0.000005] __x64_sys_ioctl+0x91/0xc0 [ +0.000005] do_syscall_64+0x3b/0xc0 [ +0.000005] entry_SYSCALL_64_after_hwframe+0x44/0xae This issue happens because "pipe_ctx->stream_res.tg" needs to be initialized first before reading its members. This commit fixes this issue by properly initializing the pointer before accessing the target data. Fixes: 663d2dae ("drm/amd/display: Add odm seamless boot support") Cc: Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by:
Sung Joon Kim <Sungjoon.Kim@amd.com> Reviewed-by:
Agustin Gutierrez <agustin.gutierrez@amd.com> Signed-off-by:
-
Likun Gao authored
Enable SMU on SMU IP version 13.0.0 Signed-off-by:
-
Evan Quan authored
Enable ATHUB and MMHUB powergating. Signed-off-by:
-
Evan Quan authored
Modify the common smu13 code and add a new smu 13.0.0 ppt file to handle the smu 13.0.0 specific configuration. v2: squash in typo fix in profile name Signed-off-by:
Lijo Lazar <lijo.lazar@amd.com> Reviewed-by:
-
