From 5c1e6fa49e8d8dbdd8bb457492b2bc52718df244 Mon Sep 17 00:00:00 2001 From: Huang Rui Date: Thu, 16 Dec 2021 13:35:27 -0500 Subject: drm/amdgpu: introduce new amdgpu_fence object to indicate the job embedded fence MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The job embedded fence donesn't initialize the flags at dma_fence_init(). Then we will go a wrong way in amdgpu_fence_get_timeline_name callback and trigger a null pointer panic once we enabled the trace event here. So introduce new amdgpu_fence object to indicate the job embedded fence. [ 156.131790] BUG: kernel NULL pointer dereference, address: 00000000000002a0 [ 156.131804] #PF: supervisor read access in kernel mode [ 156.131811] #PF: error_code(0x0000) - not-present page [ 156.131817] PGD 0 P4D 0 [ 156.131824] Oops: 0000 [#1] PREEMPT SMP PTI [ 156.131832] CPU: 6 PID: 1404 Comm: sdma0 Tainted: G OE 5.16.0-rc1-custom #1 [ 156.131842] Hardware name: Gigabyte Technology Co., Ltd. Z170XP-SLI/Z170XP-SLI-CF, BIOS F20 11/04/2016 [ 156.131848] RIP: 0010:strlen+0x0/0x20 [ 156.131859] Code: 89 c0 c3 0f 1f 80 00 00 00 00 48 01 fe eb 0f 0f b6 07 38 d0 74 10 48 83 c7 01 84 c0 74 05 48 39 f7 75 ec 31 c0 c3 48 89 f8 c3 <80> 3f 00 74 10 48 89 f8 48 83 c0 01 80 38 00 75 f7 48 29 f8 c3 31 [ 156.131872] RSP: 0018:ffff9bd0018dbcf8 EFLAGS: 00010206 [ 156.131880] RAX: 00000000000002a0 RBX: ffff8d0305ef01b0 RCX: 000000000000000b [ 156.131888] RDX: ffff8d03772ab924 RSI: ffff8d0305ef01b0 RDI: 00000000000002a0 [ 156.131895] RBP: ffff9bd0018dbd60 R08: ffff8d03002094d0 R09: 0000000000000000 [ 156.131901] R10: 000000000000005e R11: 0000000000000065 R12: ffff8d03002094d0 [ 156.131907] R13: 000000000000001f R14: 0000000000070018 R15: 0000000000000007 [ 156.131914] FS: 0000000000000000(0000) GS:ffff8d062ed80000(0000) knlGS:0000000000000000 [ 156.131923] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 156.131929] CR2: 00000000000002a0 CR3: 000000001120a005 CR4: 00000000003706e0 [ 156.131937] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 156.131942] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 156.131949] Call Trace: [ 156.131953] [ 156.131957] ? trace_event_raw_event_dma_fence+0xcc/0x200 [ 156.131973] ? ring_buffer_unlock_commit+0x23/0x130 [ 156.131982] dma_fence_init+0x92/0xb0 [ 156.131993] amdgpu_fence_emit+0x10d/0x2b0 [amdgpu] [ 156.132302] amdgpu_ib_schedule+0x2f9/0x580 [amdgpu] [ 156.132586] amdgpu_job_run+0xed/0x220 [amdgpu] v2: fix mismatch warning between the prototype and function name (Ray, kernel test robot) Signed-off-by: Huang Rui Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 11 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 126 ++++++++++++++++++++--------- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 4 +- 3 files changed, 90 insertions(+), 51 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a24b4c374188..bd4232691697 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4505,7 +4505,7 @@ int amdgpu_device_mode1_reset(struct amdgpu_device *adev) int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, struct amdgpu_reset_context *reset_context) { - int i, j, r = 0; + int i, r = 0; struct amdgpu_job *job = NULL; bool need_full_reset = test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); @@ -4527,15 +4527,8 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, /*clear job fence from fence drv to avoid force_completion *leave NULL and vm flush fence in fence drv */ - for (j = 0; j <= ring->fence_drv.num_fences_mask; j++) { - struct dma_fence *old, **ptr; + amdgpu_fence_driver_clear_job_fences(ring); - ptr = &ring->fence_drv.fences[j]; - old = rcu_dereference_protected(*ptr, 1); - if (old && test_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &old->flags)) { - RCU_INIT_POINTER(*ptr, NULL); - } - } /* after all hw jobs are reset, hw fence is meaningless, so force_completion */ amdgpu_fence_driver_force_completion(ring); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 3b7e86ea7167..9afd11ca2709 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -77,11 +77,13 @@ void amdgpu_fence_slab_fini(void) * Cast helper */ static const struct dma_fence_ops amdgpu_fence_ops; +static const struct dma_fence_ops amdgpu_job_fence_ops; static inline struct amdgpu_fence *to_amdgpu_fence(struct dma_fence *f) { struct amdgpu_fence *__f = container_of(f, struct amdgpu_fence, base); - if (__f->base.ops == &amdgpu_fence_ops) + if (__f->base.ops == &amdgpu_fence_ops || + __f->base.ops == &amdgpu_job_fence_ops) return __f; return NULL; @@ -158,19 +160,18 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd } seq = ++ring->fence_drv.sync_seq; - if (job != NULL && job->job_run_counter) { + if (job && job->job_run_counter) { /* reinit seq for resubmitted jobs */ fence->seqno = seq; } else { - dma_fence_init(fence, &amdgpu_fence_ops, - &ring->fence_drv.lock, - adev->fence_context + ring->idx, - seq); - } - - if (job != NULL) { - /* mark this fence has a parent job */ - set_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &fence->flags); + if (job) + dma_fence_init(fence, &amdgpu_job_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); + else + dma_fence_init(fence, &amdgpu_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); } amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, @@ -620,6 +621,25 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev) } } +/** + * amdgpu_fence_driver_clear_job_fences - clear job embedded fences of ring + * + * @ring: fence of the ring to be cleared + * + */ +void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring) +{ + int i; + struct dma_fence *old, **ptr; + + for (i = 0; i <= ring->fence_drv.num_fences_mask; i++) { + ptr = &ring->fence_drv.fences[i]; + old = rcu_dereference_protected(*ptr, 1); + if (old && old->ops == &amdgpu_job_fence_ops) + RCU_INIT_POINTER(*ptr, NULL); + } +} + /** * amdgpu_fence_driver_force_completion - force signal latest fence of ring * @@ -643,16 +663,14 @@ static const char *amdgpu_fence_get_driver_name(struct dma_fence *fence) static const char *amdgpu_fence_get_timeline_name(struct dma_fence *f) { - struct amdgpu_ring *ring; + return (const char *)to_amdgpu_fence(f)->ring->name; +} - if (test_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &f->flags)) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); +static const char *amdgpu_job_fence_get_timeline_name(struct dma_fence *f) +{ + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); - ring = to_amdgpu_ring(job->base.sched); - } else { - ring = to_amdgpu_fence(f)->ring; - } - return (const char *)ring->name; + return (const char *)to_amdgpu_ring(job->base.sched)->name; } /** @@ -665,18 +683,25 @@ static const char *amdgpu_fence_get_timeline_name(struct dma_fence *f) */ static bool amdgpu_fence_enable_signaling(struct dma_fence *f) { - struct amdgpu_ring *ring; + if (!timer_pending(&to_amdgpu_fence(f)->ring->fence_drv.fallback_timer)) + amdgpu_fence_schedule_fallback(to_amdgpu_fence(f)->ring); - if (test_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &f->flags)) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); + return true; +} - ring = to_amdgpu_ring(job->base.sched); - } else { - ring = to_amdgpu_fence(f)->ring; - } +/** + * amdgpu_job_fence_enable_signaling - enable signalling on job fence + * @f: fence + * + * This is the simliar function with amdgpu_fence_enable_signaling above, it + * only handles the job embedded fence. + */ +static bool amdgpu_job_fence_enable_signaling(struct dma_fence *f) +{ + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); - if (!timer_pending(&ring->fence_drv.fallback_timer)) - amdgpu_fence_schedule_fallback(ring); + if (!timer_pending(&to_amdgpu_ring(job->base.sched)->fence_drv.fallback_timer)) + amdgpu_fence_schedule_fallback(to_amdgpu_ring(job->base.sched)); return true; } @@ -692,19 +717,23 @@ static void amdgpu_fence_free(struct rcu_head *rcu) { struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); - if (test_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &f->flags)) { - /* free job if fence has a parent job */ - struct amdgpu_job *job; - - job = container_of(f, struct amdgpu_job, hw_fence); - kfree(job); - } else { /* free fence_slab if it's separated fence*/ - struct amdgpu_fence *fence; + kmem_cache_free(amdgpu_fence_slab, to_amdgpu_fence(f)); +} - fence = to_amdgpu_fence(f); - kmem_cache_free(amdgpu_fence_slab, fence); - } +/** + * amdgpu_job_fence_free - free up the job with embedded fence + * + * @rcu: RCU callback head + * + * Free up the job with embedded fence after the RCU grace period. + */ +static void amdgpu_job_fence_free(struct rcu_head *rcu) +{ + struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); + + /* free job if fence has a parent job */ + kfree(container_of(f, struct amdgpu_job, hw_fence)); } /** @@ -720,6 +749,19 @@ static void amdgpu_fence_release(struct dma_fence *f) call_rcu(&f->rcu, amdgpu_fence_free); } +/** + * amdgpu_job_fence_release - callback that job embedded fence can be freed + * + * @f: fence + * + * This is the simliar function with amdgpu_fence_release above, it + * only handles the job embedded fence. + */ +static void amdgpu_job_fence_release(struct dma_fence *f) +{ + call_rcu(&f->rcu, amdgpu_job_fence_free); +} + static const struct dma_fence_ops amdgpu_fence_ops = { .get_driver_name = amdgpu_fence_get_driver_name, .get_timeline_name = amdgpu_fence_get_timeline_name, @@ -727,6 +769,12 @@ static const struct dma_fence_ops amdgpu_fence_ops = { .release = amdgpu_fence_release, }; +static const struct dma_fence_ops amdgpu_job_fence_ops = { + .get_driver_name = amdgpu_fence_get_driver_name, + .get_timeline_name = amdgpu_job_fence_get_timeline_name, + .enable_signaling = amdgpu_job_fence_enable_signaling, + .release = amdgpu_job_fence_release, +}; /* * Fence debugfs diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 4d380e79752c..fae7d185ad0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -53,9 +53,6 @@ enum amdgpu_ring_priority_level { #define AMDGPU_FENCE_FLAG_INT (1 << 1) #define AMDGPU_FENCE_FLAG_TC_WB_ONLY (1 << 2) -/* fence flag bit to indicate the face is embedded in job*/ -#define AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT (DMA_FENCE_FLAG_USER_BITS + 1) - #define to_amdgpu_ring(s) container_of((s), struct amdgpu_ring, sched) #define AMDGPU_IB_POOL_SIZE (1024 * 1024) @@ -114,6 +111,7 @@ struct amdgpu_fence_driver { struct dma_fence **fences; }; +void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring); void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring); int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring, -- cgit From d4c2933fb8eece9b2703e9ddd4aa8f1a77f22450 Mon Sep 17 00:00:00 2001 From: chen gong Date: Thu, 9 Dec 2021 19:47:10 +0800 Subject: drm/amdgpu: When the VCN(1.0) block is suspended, powergating is explicitly enabled Play a video on the raven (or PCO, raven2) platform, and then do the S3 test. When resume, the following error will be reported: amdgpu 0000:02:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring vcn_dec test failed (-110) [drm:amdgpu_device_ip_resume_phase2 [amdgpu]] *ERROR* resume of IP block failed -110 amdgpu 0000:02:00.0: amdgpu: amdgpu_device_ip_resume failed (-110). PM: dpm_run_callback(): pci_pm_resume+0x0/0x90 returns -110 [why] When playing the video: The power state flag of the vcn block is set to POWER_STATE_ON. When doing suspend: There is no change to the power state flag of the vcn block, it is still POWER_STATE_ON. When doing resume: Need to open the power gate of the vcn block and set the power state flag of the VCN block to POWER_STATE_ON. But at this time, the power state flag of the vcn block is already POWER_STATE_ON. The power status flag check in the "8f2cdef drm/amd/pm: avoid duplicate powergate/ungate setting" patch will return the amdgpu_dpm_set_powergating_by_smu function directly. As a result, the gate of the power was not opened, causing the subsequent ring test to fail. [how] In the suspend function of the vcn block, explicitly change the power state flag of the vcn block to POWER_STATE_OFF. BugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1828 Signed-off-by: chen gong Reviewed-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index d54d720b3cf6..3799226defc0 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -246,6 +246,13 @@ static int vcn_v1_0_suspend(void *handle) { int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool idle_work_unexecuted; + + idle_work_unexecuted = cancel_delayed_work_sync(&adev->vcn.idle_work); + if (idle_work_unexecuted) { + if (adev->pm.dpm_enabled) + amdgpu_dpm_enable_uvd(adev, false); + } r = vcn_v1_0_hw_fini(adev); if (r) -- cgit From 892deb48269c65376f3eeb5b4c032ff2c2979bd7 Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Thu, 16 Dec 2021 17:01:45 +0000 Subject: drm/amdgpu: Separate vf2pf work item init from virt data exchange We want to be able to call virt data exchange conditionally after gmc sw init to reserve bad pages as early as possible. Since this is a conditional call, we will need to call it again unconditionally later in the init sequence. Refactor the data exchange function so it can be called multiple times without re-initializing the work item. v2: Cleaned up the code. Kept the original call to init_exchange_data() inside early init to initialize the work item, afterwards call exchange_data() when needed. Signed-off-by: Victor Skvortsov Reviewed By: Shaoyun.liu Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 ++++- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 36 ++++++++++++++++++++---------- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h | 1 + 3 files changed, 30 insertions(+), 13 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index bd4232691697..0cd2404ca63c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2317,6 +2317,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) /* need to do gmc hw init early so we can allocate gpu mem */ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC) { + /* Try to reserve bad pages early */ + if (amdgpu_sriov_vf(adev)) + amdgpu_virt_exchange_data(adev); + r = amdgpu_device_vram_scratch_init(adev); if (r) { DRM_ERROR("amdgpu_vram_scratch_init failed %d\n", r); @@ -2348,7 +2352,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) } if (amdgpu_sriov_vf(adev)) - amdgpu_virt_init_data_exchange(adev); + amdgpu_virt_exchange_data(adev); r = amdgpu_ib_pool_init(adev); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 3fc49823f527..f8e574cc0e22 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -622,17 +622,35 @@ void amdgpu_virt_fini_data_exchange(struct amdgpu_device *adev) void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) { - uint64_t bp_block_offset = 0; - uint32_t bp_block_size = 0; - struct amd_sriov_msg_pf2vf_info *pf2vf_v2 = NULL; - adev->virt.fw_reserve.p_pf2vf = NULL; adev->virt.fw_reserve.p_vf2pf = NULL; adev->virt.vf2pf_update_interval_ms = 0; - if (adev->mman.fw_vram_usage_va != NULL) { + if (adev->bios != NULL) { adev->virt.vf2pf_update_interval_ms = 2000; + adev->virt.fw_reserve.p_pf2vf = + (struct amd_sriov_msg_pf2vf_info_header *) + (adev->bios + (AMD_SRIOV_MSG_PF2VF_OFFSET_KB << 10)); + + amdgpu_virt_read_pf2vf_data(adev); + } + + if (adev->virt.vf2pf_update_interval_ms != 0) { + INIT_DELAYED_WORK(&adev->virt.vf2pf_work, amdgpu_virt_update_vf2pf_work_item); + schedule_delayed_work(&(adev->virt.vf2pf_work), msecs_to_jiffies(adev->virt.vf2pf_update_interval_ms)); + } +} + + +void amdgpu_virt_exchange_data(struct amdgpu_device *adev) +{ + uint64_t bp_block_offset = 0; + uint32_t bp_block_size = 0; + struct amd_sriov_msg_pf2vf_info *pf2vf_v2 = NULL; + + if (adev->mman.fw_vram_usage_va != NULL) { + adev->virt.fw_reserve.p_pf2vf = (struct amd_sriov_msg_pf2vf_info_header *) (adev->mman.fw_vram_usage_va + (AMD_SRIOV_MSG_PF2VF_OFFSET_KB << 10)); @@ -663,16 +681,10 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) (adev->bios + (AMD_SRIOV_MSG_PF2VF_OFFSET_KB << 10)); amdgpu_virt_read_pf2vf_data(adev); - - return; - } - - if (adev->virt.vf2pf_update_interval_ms != 0) { - INIT_DELAYED_WORK(&adev->virt.vf2pf_work, amdgpu_virt_update_vf2pf_work_item); - schedule_delayed_work(&(adev->virt.vf2pf_work), adev->virt.vf2pf_update_interval_ms); } } + void amdgpu_detect_virtualization(struct amdgpu_device *adev) { uint32_t reg; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 8d4c20bb71c5..9adfb8d63280 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -308,6 +308,7 @@ int amdgpu_virt_alloc_mm_table(struct amdgpu_device *adev); void amdgpu_virt_free_mm_table(struct amdgpu_device *adev); void amdgpu_virt_release_ras_err_handler_data(struct amdgpu_device *adev); void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev); +void amdgpu_virt_exchange_data(struct amdgpu_device *adev); void amdgpu_virt_fini_data_exchange(struct amdgpu_device *adev); void amdgpu_detect_virtualization(struct amdgpu_device *adev); -- cgit From 0cd7f378b0927e10774981eb09e388258ca925e0 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 15 Dec 2021 22:13:56 -0500 Subject: drm/amdgpu: add support for IP discovery gc_info table v2 Used on gfx9 based systems. Fixes incorrect CU counts reported in the kernel log. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1833 Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 76 +++++++++++++++++++-------- 1 file changed, 54 insertions(+), 22 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 65e1f6cc59dd..b13db855cc9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -565,10 +565,15 @@ void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev) } } +union gc_info { + struct gc_info_v1_0 v1; + struct gc_info_v2_0 v2; +}; + int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) { struct binary_header *bhdr; - struct gc_info_v1_0 *gc_info; + union gc_info *gc_info; if (!adev->mman.discovery_bin) { DRM_ERROR("ip discovery uninitialized\n"); @@ -576,28 +581,55 @@ int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) } bhdr = (struct binary_header *)adev->mman.discovery_bin; - gc_info = (struct gc_info_v1_0 *)(adev->mman.discovery_bin + + gc_info = (union gc_info *)(adev->mman.discovery_bin + le16_to_cpu(bhdr->table_list[GC].offset)); - - adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->gc_num_se); - adev->gfx.config.max_cu_per_sh = 2 * (le32_to_cpu(gc_info->gc_num_wgp0_per_sa) + - le32_to_cpu(gc_info->gc_num_wgp1_per_sa)); - adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->gc_num_sa_per_se); - adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->gc_num_rb_per_se); - adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->gc_num_gl2c); - adev->gfx.config.max_gprs = le32_to_cpu(gc_info->gc_num_gprs); - adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->gc_num_max_gs_thds); - adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->gc_gs_table_depth); - adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->gc_gsprim_buff_depth); - adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->gc_double_offchip_lds_buffer); - adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->gc_wave_size); - adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->gc_max_waves_per_simd); - adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->gc_max_scratch_slots_per_cu); - adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->gc_lds_size); - adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->gc_num_sc_per_se) / - le32_to_cpu(gc_info->gc_num_sa_per_se); - adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->gc_num_packer_per_sc); - + switch (gc_info->v1.header.version_major) { + case 1: + adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->v1.gc_num_se); + adev->gfx.config.max_cu_per_sh = 2 * (le32_to_cpu(gc_info->v1.gc_num_wgp0_per_sa) + + le32_to_cpu(gc_info->v1.gc_num_wgp1_per_sa)); + adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->v1.gc_num_sa_per_se); + adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->v1.gc_num_rb_per_se); + adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->v1.gc_num_gl2c); + adev->gfx.config.max_gprs = le32_to_cpu(gc_info->v1.gc_num_gprs); + adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->v1.gc_num_max_gs_thds); + adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->v1.gc_gs_table_depth); + adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->v1.gc_gsprim_buff_depth); + adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->v1.gc_double_offchip_lds_buffer); + adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->v1.gc_wave_size); + adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->v1.gc_max_waves_per_simd); + adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->v1.gc_max_scratch_slots_per_cu); + adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->v1.gc_lds_size); + adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v1.gc_num_sc_per_se) / + le32_to_cpu(gc_info->v1.gc_num_sa_per_se); + adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v1.gc_num_packer_per_sc); + break; + case 2: + adev->gfx.config.max_shader_engines = le32_to_cpu(gc_info->v2.gc_num_se); + adev->gfx.config.max_cu_per_sh = le32_to_cpu(gc_info->v2.gc_num_cu_per_sh); + adev->gfx.config.max_sh_per_se = le32_to_cpu(gc_info->v2.gc_num_sh_per_se); + adev->gfx.config.max_backends_per_se = le32_to_cpu(gc_info->v2.gc_num_rb_per_se); + adev->gfx.config.max_texture_channel_caches = le32_to_cpu(gc_info->v2.gc_num_tccs); + adev->gfx.config.max_gprs = le32_to_cpu(gc_info->v2.gc_num_gprs); + adev->gfx.config.max_gs_threads = le32_to_cpu(gc_info->v2.gc_num_max_gs_thds); + adev->gfx.config.gs_vgt_table_depth = le32_to_cpu(gc_info->v2.gc_gs_table_depth); + adev->gfx.config.gs_prim_buffer_depth = le32_to_cpu(gc_info->v2.gc_gsprim_buff_depth); + adev->gfx.config.double_offchip_lds_buf = le32_to_cpu(gc_info->v2.gc_double_offchip_lds_buffer); + adev->gfx.cu_info.wave_front_size = le32_to_cpu(gc_info->v2.gc_wave_size); + adev->gfx.cu_info.max_waves_per_simd = le32_to_cpu(gc_info->v2.gc_max_waves_per_simd); + adev->gfx.cu_info.max_scratch_slots_per_cu = le32_to_cpu(gc_info->v2.gc_max_scratch_slots_per_cu); + adev->gfx.cu_info.lds_size = le32_to_cpu(gc_info->v2.gc_lds_size); + adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v2.gc_num_sc_per_se) / + le32_to_cpu(gc_info->v2.gc_num_sh_per_se); + adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v2.gc_num_packer_per_sc); + break; + default: + dev_err(adev->dev, + "Unhandled GC info table %d.%d\n", + gc_info->v1.header.version_major, + gc_info->v1.header.version_minor); + return -EINVAL; + } return 0; } -- cgit From b18ff6925d84492d845debe2dbb81122b1d141fa Mon Sep 17 00:00:00 2001 From: Bokun Zhang Date: Thu, 16 Dec 2021 15:09:41 -0500 Subject: drm/amdgpu: Filter security violation registers 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 Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 83 +++++++++++++++++++--------------- 1 file changed, 46 insertions(+), 37 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index 4f546f632223..d3d6d5b045b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -542,9 +542,6 @@ static void sdma_v5_2_ctx_switch_enable(struct amdgpu_device *adev, bool enable) } for (i = 0; i < adev->sdma.num_instances; i++) { - f32_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); - f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, - AUTO_CTXSW_ENABLE, enable ? 1 : 0); if (enable && amdgpu_sdma_phase_quantum) { WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM), phase_quantum); @@ -553,7 +550,13 @@ static void sdma_v5_2_ctx_switch_enable(struct amdgpu_device *adev, bool enable) WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM), phase_quantum); } - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); + + if (!amdgpu_sriov_vf(adev)) { + f32_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL, + AUTO_CTXSW_ENABLE, enable ? 1 : 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl); + } } } @@ -576,10 +579,12 @@ static void sdma_v5_2_enable(struct amdgpu_device *adev, bool enable) sdma_v5_2_rlc_stop(adev); } - for (i = 0; i < adev->sdma.num_instances; i++) { - f32_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); - f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1); - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl); + if (!amdgpu_sriov_vf(adev)) { + for (i = 0; i < adev->sdma.num_instances; i++) { + f32_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl); + } } } @@ -608,7 +613,8 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) ring = &adev->sdma.instance[i].ring; wb_offset = (ring->rptr_offs * 4); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + if (!amdgpu_sriov_vf(adev)) + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); /* Set ring buffer size in dwords */ rb_bufsz = order_base_2(ring->ring_size / 4); @@ -683,32 +689,34 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) sdma_v5_2_ring_set_wptr(ring); /* set minor_ptr_update to 0 after wptr programed */ - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - /* set utc l1 enable flag always to 1 */ - temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); - - /* enable MCBP */ - temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); - - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); - - /* program default cache read and write policy */ - temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | - (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | - SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); + /* SRIOV VF has no control of any of registers below */ if (!amdgpu_sriov_vf(adev)) { + /* set utc l1 enable flag always to 1 */ + temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); + + /* enable MCBP */ + temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); + + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + + /* program default cache read and write policy */ + temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | + (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | + SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + /* unhalt engine */ temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); @@ -1436,13 +1444,14 @@ static int sdma_v5_2_set_trap_irq_state(struct amdgpu_device *adev, enum amdgpu_interrupt_state state) { u32 sdma_cntl; - u32 reg_offset = sdma_v5_2_get_reg_offset(adev, type, mmSDMA0_CNTL); - sdma_cntl = RREG32(reg_offset); - sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, - state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); - WREG32(reg_offset, sdma_cntl); + if (!amdgpu_sriov_vf(adev)) { + sdma_cntl = RREG32(reg_offset); + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE, + state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); + WREG32(reg_offset, sdma_cntl); + } return 0; } -- cgit From 0da6f6e5872e8c1e794bc3a705ea27a156b6a8e4 Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Mon, 13 Dec 2021 22:35:10 +0000 Subject: drm/amdgpu: Add *_SOC15_IP_NO_KIQ() macro definitions Add helper macros to change register access from direct to indirect. Signed-off-by: Victor Skvortsov Reviewed-by: David Nieto Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/soc15_common.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 8a9ca87d8663..473767e03676 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -51,6 +51,8 @@ #define RREG32_SOC15_IP(ip, reg) __RREG32_SOC15_RLC__(reg, 0, ip##_HWIP) +#define RREG32_SOC15_IP_NO_KIQ(ip, reg) __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ, ip##_HWIP) + #define RREG32_SOC15_NO_KIQ(ip, inst, reg) \ __RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \ AMDGPU_REGS_NO_KIQ, ip##_HWIP) @@ -65,6 +67,9 @@ #define WREG32_SOC15_IP(ip, reg, value) \ __WREG32_SOC15_RLC__(reg, value, 0, ip##_HWIP) +#define WREG32_SOC15_IP_NO_KIQ(ip, reg, value) \ + __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ, ip##_HWIP) + #define WREG32_SOC15_NO_KIQ(ip, inst, reg, value) \ __WREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \ value, AMDGPU_REGS_NO_KIQ, ip##_HWIP) -- cgit From 92f153bb5a4bff4df8c67dcec5a9de1dde3b300b Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Mon, 13 Dec 2021 22:40:09 +0000 Subject: drm/amdgpu: Modify indirect register access for gmc_v9_0 sriov 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 Reviewed-by: David Nieto Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 57 ++++++++++++++++++++++++++--------- 1 file changed, 43 insertions(+), 14 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index a5471923b3f6..2b86c63b032a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -478,9 +478,18 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, hub = &adev->vmhub[j]; for (i = 0; i < 16; i++) { reg = hub->vm_context0_cntl + i; - tmp = RREG32(reg); + + if (j == AMDGPU_GFXHUB_0) + tmp = RREG32_SOC15_IP(GC, reg); + else + tmp = RREG32_SOC15_IP(MMHUB, reg); + tmp &= ~bits; - WREG32(reg, tmp); + + if (j == AMDGPU_GFXHUB_0) + WREG32_SOC15_IP(GC, reg, tmp); + else + WREG32_SOC15_IP(MMHUB, reg, tmp); } } break; @@ -489,9 +498,18 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, hub = &adev->vmhub[j]; for (i = 0; i < 16; i++) { reg = hub->vm_context0_cntl + i; - tmp = RREG32(reg); + + if (j == AMDGPU_GFXHUB_0) + tmp = RREG32_SOC15_IP(GC, reg); + else + tmp = RREG32_SOC15_IP(MMHUB, reg); + tmp |= bits; - WREG32(reg, tmp); + + if (j == AMDGPU_GFXHUB_0) + WREG32_SOC15_IP(GC, reg, tmp); + else + WREG32_SOC15_IP(MMHUB, reg, tmp); } } break; @@ -788,9 +806,12 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ if (use_semaphore) { for (j = 0; j < adev->usec_timeout; j++) { - /* a read return value of 1 means semaphore acuqire */ - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_sem + - hub->eng_distance * eng); + /* a read return value of 1 means semaphore acquire */ + if (vmhub == AMDGPU_GFXHUB_0) + tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng); + else + tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng); + if (tmp & 0x1) break; udelay(1); @@ -801,8 +822,10 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } do { - WREG32_NO_KIQ(hub->vm_inv_eng0_req + - hub->eng_distance * eng, inv_req); + if (vmhub == AMDGPU_GFXHUB_0) + WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); + else + WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); /* * Issue a dummy read to wait for the ACK register to @@ -815,8 +838,11 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, hub->eng_distance * eng); for (j = 0; j < adev->usec_timeout; j++) { - tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + - hub->eng_distance * eng); + if (vmhub == AMDGPU_GFXHUB_0) + tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_ack + hub->eng_distance * eng); + else + tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_ack + hub->eng_distance * eng); + if (tmp & (1 << vmid)) break; udelay(1); @@ -827,13 +853,16 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } while (inv_req); /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (use_semaphore) + if (use_semaphore) { /* * add semaphore release after invalidation, * write with 0 means semaphore release */ - WREG32_NO_KIQ(hub->vm_inv_eng0_sem + - hub->eng_distance * eng, 0); + if (vmhub == AMDGPU_GFXHUB_0) + WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0); + else + WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0); + } spin_unlock(&adev->gmc.invalidate_lock); -- cgit From 4aa325ae541378166b3a84186bc1157a1d524f8b Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Mon, 13 Dec 2021 22:43:06 +0000 Subject: drm/amdgpu: Modify indirect register access for amdkfd_gfx_v9 sriov Modify GC register access from MMIO to RLCG if the indirect flag is set Signed-off-by: Victor Skvortsov Reviewed-by: David Nieto Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 27 +++++++++++------------ 1 file changed, 13 insertions(+), 14 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index ddfe7aff919d..1abf662a0e91 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -166,7 +166,7 @@ int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) lock_srbm(adev, mec, pipe, 0, 0); - WREG32(SOC15_REG_OFFSET(GC, 0, mmCPC_INT_CNTL), + WREG32_SOC15(GC, 0, mmCPC_INT_CNTL, CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK | CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK); @@ -279,7 +279,7 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, lower_32_bits((uintptr_t)wptr)); WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), upper_32_bits((uintptr_t)wptr)); - WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1), + WREG32_SOC15(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1, (uint32_t)get_queue_mask(adev, pipe_id, queue_id)); } @@ -488,13 +488,13 @@ bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev, uint32_t low, high; acquire_queue(adev, pipe_id, queue_id); - act = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_ACTIVE)); + act = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE); if (act) { low = lower_32_bits(queue_address >> 8); high = upper_32_bits(queue_address >> 8); - if (low == RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_BASE)) && - high == RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_BASE_HI))) + if (low == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE) && + high == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE_HI)) retval = true; } release_queue(adev); @@ -556,7 +556,7 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, end_jiffies = (utimeout * HZ / 1000) + jiffies; while (true) { - temp = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_ACTIVE)); + temp = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE); if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK)) break; if (time_after(jiffies, end_jiffies)) { @@ -645,7 +645,7 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_INDEX, gfx_index_val); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_CMD), sq_cmd); + WREG32_SOC15(GC, 0, mmSQ_CMD, sq_cmd); data = REG_SET_FIELD(data, GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES, 1); @@ -722,7 +722,7 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, pipe_idx = queue_idx / adev->gfx.mec.num_queue_per_pipe; queue_slot = queue_idx % adev->gfx.mec.num_queue_per_pipe; soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0); - reg_val = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) + + reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) + queue_slot); *wave_cnt = reg_val & SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK; if (*wave_cnt != 0) @@ -809,8 +809,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) { gfx_v9_0_select_se_sh(adev, se_idx, sh_idx, 0xffffffff); - queue_map = RREG32(SOC15_REG_OFFSET(GC, 0, - mmSPI_CSQ_WF_ACTIVE_STATUS)); + queue_map = RREG32_SOC15(GC, 0, mmSPI_CSQ_WF_ACTIVE_STATUS); /* * Assumption: queue map encodes following schema: four @@ -860,17 +859,17 @@ void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev, /* * Program TBA registers */ - WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_SHADER_TBA_LO), + WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_LO, lower_32_bits(tba_addr >> 8)); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_SHADER_TBA_HI), + WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_HI, upper_32_bits(tba_addr >> 8)); /* * Program TMA registers */ - WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_SHADER_TMA_LO), + WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_LO, lower_32_bits(tma_addr >> 8)); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_SHADER_TMA_HI), + WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_HI, upper_32_bits(tma_addr >> 8)); unlock_srbm(adev); -- cgit From 4a0165f0603f333c6b36a420b4e348b67ddf6fc8 Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Wed, 15 Dec 2021 18:42:44 +0000 Subject: drm/amdgpu: get xgmi info before ip_init 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 Reviewed-by: David Nieto Reviewed by: shaoyun.liu Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++ drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 6 ------ drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 6 ------ 3 files changed, 7 insertions(+), 12 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 0cd2404ca63c..ce820ee2e2e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3577,6 +3577,13 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (r) return r; + /* Need to get xgmi info early to decide the reset behavior*/ + if (adev->gmc.xgmi.supported) { + r = adev->gfxhub.funcs->get_xgmi_info(adev); + if (r) + return r; + } + /* enable PCIE atomic ops */ if (amdgpu_sriov_vf(adev)) adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index ae46eb35b3d7..3d5d47a799e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -914,12 +914,6 @@ static int gmc_v10_0_sw_init(void *handle) return r; } - if (adev->gmc.xgmi.supported) { - r = adev->gfxhub.funcs->get_xgmi_info(adev); - if (r) - return r; - } - r = gmc_v10_0_mc_init(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 2b86c63b032a..57f2729a7bd0 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1628,12 +1628,6 @@ static int gmc_v9_0_sw_init(void *handle) } adev->need_swiotlb = drm_need_swiotlb(44); - if (adev->gmc.xgmi.supported) { - r = adev->gfxhub.funcs->get_xgmi_info(adev); - if (r) - return r; - } - r = gmc_v9_0_mc_init(adev); if (r) return r; -- cgit From d764fb2af6cd92eee080978856f24b0cc6d555ad Mon Sep 17 00:00:00 2001 From: Victor Skvortsov Date: Wed, 15 Dec 2021 18:49:36 +0000 Subject: drm/amdgpu: Modify indirect register access for gfx9 sriov 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 Reviewed-by: David Nieto Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 117 ++++++++++++++++++++++++++-------- 1 file changed, 89 insertions(+), 28 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index edb3e3b08eed..9189fb85a4dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -63,6 +63,13 @@ #define mmGCEA_PROBE_MAP 0x070c #define mmGCEA_PROBE_MAP_BASE_IDX 0 +#define GFX9_RLCG_GC_WRITE_OLD (0x8 << 28) +#define GFX9_RLCG_GC_WRITE (0x0 << 28) +#define GFX9_RLCG_GC_READ (0x1 << 28) +#define GFX9_RLCG_VFGATE_DISABLED 0x4000000 +#define GFX9_RLCG_WRONG_OPERATION_TYPE 0x2000000 +#define GFX9_RLCG_NOT_IN_RANGE 0x1000000 + MODULE_FIRMWARE("amdgpu/vega10_ce.bin"); MODULE_FIRMWARE("amdgpu/vega10_pfp.bin"); MODULE_FIRMWARE("amdgpu/vega10_me.bin"); @@ -739,7 +746,7 @@ static const u32 GFX_RLC_SRM_INDEX_CNTL_DATA_OFFSETS[] = mmRLC_SRM_INDEX_CNTL_DATA_7 - mmRLC_SRM_INDEX_CNTL_DATA_0, }; -static void gfx_v9_0_rlcg_w(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag) +static u32 gfx_v9_0_rlcg_rw(struct amdgpu_device *adev, u32 offset, u32 v, uint32_t flag) { static void *scratch_reg0; static void *scratch_reg1; @@ -748,21 +755,20 @@ static void gfx_v9_0_rlcg_w(struct amdgpu_device *adev, u32 offset, u32 v, u32 f static void *spare_int; static uint32_t grbm_cntl; static uint32_t grbm_idx; + uint32_t i = 0; + uint32_t retries = 50000; + u32 ret = 0; + u32 tmp; scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; - scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4; - scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4; + scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG2_BASE_IDX] + mmSCRATCH_REG2)*4; + scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG3_BASE_IDX] + mmSCRATCH_REG3)*4; spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; - if (amdgpu_sriov_runtime(adev)) { - pr_err("shouldn't call rlcg write register during runtime\n"); - return; - } - if (offset == grbm_cntl || offset == grbm_idx) { if (offset == grbm_cntl) writel(v, scratch_reg2); @@ -771,41 +777,95 @@ static void gfx_v9_0_rlcg_w(struct amdgpu_device *adev, u32 offset, u32 v, u32 f writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); } else { - uint32_t i = 0; - uint32_t retries = 50000; - + /* + * SCRATCH_REG0 = read/write value + * SCRATCH_REG1[30:28] = command + * SCRATCH_REG1[19:0] = address in dword + * SCRATCH_REG1[26:24] = Error reporting + */ writel(v, scratch_reg0); - writel(offset | 0x80000000, scratch_reg1); + writel(offset | flag, scratch_reg1); writel(1, spare_int); - for (i = 0; i < retries; i++) { - u32 tmp; + for (i = 0; i < retries; i++) { tmp = readl(scratch_reg1); - if (!(tmp & 0x80000000)) + if (!(tmp & flag)) break; udelay(10); } - if (i >= retries) - pr_err("timeout: rlcg program reg:0x%05x failed !\n", offset); + + if (i >= retries) { + if (amdgpu_sriov_reg_indirect_gc(adev)) { + if (tmp & GFX9_RLCG_VFGATE_DISABLED) + pr_err("The vfgate is disabled, program reg:0x%05x failed!\n", offset); + else if (tmp & GFX9_RLCG_WRONG_OPERATION_TYPE) + pr_err("Wrong operation type, program reg:0x%05x failed!\n", offset); + else if (tmp & GFX9_RLCG_NOT_IN_RANGE) + pr_err("The register is not in range, program reg:0x%05x failed!\n", offset); + else + pr_err("Unknown error type, program reg:0x%05x failed!\n", offset); + } else + pr_err("timeout: rlcg program reg:0x%05x failed!\n", offset); + } } + ret = readl(scratch_reg0); + + return ret; +} + +static bool gfx_v9_0_get_rlcg_flag(struct amdgpu_device *adev, u32 acc_flags, u32 hwip, + int write, u32 *rlcg_flag) +{ + + switch (hwip) { + case GC_HWIP: + if (amdgpu_sriov_reg_indirect_gc(adev)) { + *rlcg_flag = write ? GFX9_RLCG_GC_WRITE : GFX9_RLCG_GC_READ; + + return true; + /* only in new version, AMDGPU_REGS_NO_KIQ and AMDGPU_REGS_RLC enabled simultaneously */ + } else if ((acc_flags & AMDGPU_REGS_RLC) && !(acc_flags & AMDGPU_REGS_NO_KIQ) && write) { + *rlcg_flag = GFX9_RLCG_GC_WRITE_OLD; + return true; + } + + break; + default: + return false; + } + + return false; +} + +static u32 gfx_v9_0_sriov_rreg(struct amdgpu_device *adev, u32 offset, u32 acc_flags, u32 hwip) +{ + u32 rlcg_flag; + + if (!amdgpu_sriov_runtime(adev) && gfx_v9_0_get_rlcg_flag(adev, acc_flags, hwip, 0, &rlcg_flag)) + return gfx_v9_0_rlcg_rw(adev, offset, 0, rlcg_flag); + + if (acc_flags & AMDGPU_REGS_NO_KIQ) + return RREG32_NO_KIQ(offset); + else + return RREG32(offset); } static void gfx_v9_0_sriov_wreg(struct amdgpu_device *adev, u32 offset, - u32 v, u32 acc_flags, u32 hwip) + u32 value, u32 acc_flags, u32 hwip) { - if ((acc_flags & AMDGPU_REGS_RLC) && - amdgpu_sriov_fullaccess(adev)) { - gfx_v9_0_rlcg_w(adev, offset, v, acc_flags); + u32 rlcg_flag; + if (!amdgpu_sriov_runtime(adev) && gfx_v9_0_get_rlcg_flag(adev, acc_flags, hwip, 1, &rlcg_flag)) { + gfx_v9_0_rlcg_rw(adev, offset, value, rlcg_flag); return; } if (acc_flags & AMDGPU_REGS_NO_KIQ) - WREG32_NO_KIQ(offset, v); + WREG32_NO_KIQ(offset, value); else - WREG32(offset, v); + WREG32(offset, value); } #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042 @@ -5135,7 +5195,7 @@ static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) if (amdgpu_sriov_is_pp_one_vf(adev)) data = RREG32_NO_KIQ(reg); else - data = RREG32(reg); + data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL); data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK; data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; @@ -5191,6 +5251,7 @@ static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .start = gfx_v9_0_rlc_start, .update_spm_vmid = gfx_v9_0_update_spm_vmid, .sriov_wreg = gfx_v9_0_sriov_wreg, + .sriov_rreg = gfx_v9_0_sriov_rreg, .is_rlcg_access_range = gfx_v9_0_is_rlcg_access_range, }; @@ -5796,16 +5857,16 @@ static void gfx_v9_0_set_compute_eop_interrupt_state(struct amdgpu_device *adev, switch (state) { case AMDGPU_IRQ_STATE_DISABLE: - mec_int_cntl = RREG32(mec_int_cntl_reg); + mec_int_cntl = RREG32_SOC15_IP(GC,mec_int_cntl_reg); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, TIME_STAMP_INT_ENABLE, 0); - WREG32(mec_int_cntl_reg, mec_int_cntl); + WREG32_SOC15_IP(GC, mec_int_cntl_reg, mec_int_cntl); break; case AMDGPU_IRQ_STATE_ENABLE: - mec_int_cntl = RREG32(mec_int_cntl_reg); + mec_int_cntl = RREG32_SOC15_IP(GC, mec_int_cntl_reg); mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, TIME_STAMP_INT_ENABLE, 1); - WREG32(mec_int_cntl_reg, mec_int_cntl); + WREG32_SOC15_IP(GC, mec_int_cntl_reg, mec_int_cntl); break; default: break; -- cgit From f4409ee84658ff95d92daace9f094fa6bf80b0e3 Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Fri, 10 Dec 2021 17:01:15 +0800 Subject: drm/amdgpu: add gpu reset control for umc page retirement Add a reset parameter for umc page retirement, let user decide whether call gpu reset in umc page retirement. Signed-off-by: Tao Zhou Reviewed-by: Hawking Zhang Acked-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 15 ++++++++++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 5 +++-- 2 files changed, 15 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 6e4bea012ea4..0c33f367a4e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -23,6 +23,13 @@ #include "amdgpu_ras.h" +static int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, + void *ras_error_status, + struct amdgpu_iv_entry *entry) +{ + return amdgpu_umc_do_page_retirement(adev, ras_error_status, entry, true); +} + int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) { int r; @@ -88,9 +95,10 @@ void amdgpu_umc_ras_fini(struct amdgpu_device *adev) } } -int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, +int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, void *ras_error_status, - struct amdgpu_iv_entry *entry) + struct amdgpu_iv_entry *entry, + bool reset) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; struct amdgpu_ras *con = amdgpu_ras_get_context(adev); @@ -164,7 +172,8 @@ int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, adev->smu.ppt_funcs->send_hbm_bad_pages_num(&adev->smu, con->eeprom_control.ras_num_recs); } - amdgpu_ras_reset_gpu(adev); + if (reset) + amdgpu_ras_reset_gpu(adev); } kfree(err_data->err_addr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index 9e40bade0a68..8d18d5121f66 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -78,9 +78,10 @@ struct amdgpu_umc { int amdgpu_umc_ras_late_init(struct amdgpu_device *adev); void amdgpu_umc_ras_fini(struct amdgpu_device *adev); -int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, +int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, void *ras_error_status, - struct amdgpu_iv_entry *entry); + struct amdgpu_iv_entry *entry, + bool reset); int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); -- cgit From b6485bed40d7859735bdbfedbd55dcc8366a88a7 Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Mon, 6 Dec 2021 15:54:54 +0800 Subject: drm/amdkfd: reset queue which consumes RAS poison (v2) 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 Reviewed-by: Hawking Zhang Acked-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 6 +++--- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 3 ++- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 46cf48b3904a..0bf09a94d944 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -721,13 +721,13 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev) return adev->have_atomics_support; } -void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev) +void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset) { struct ras_err_data err_data = {0, 0, 0, NULL}; /* CPU MCA will handle page retirement if connected_to_cpu is 1 */ if (!adev->gmc.xgmi.connected_to_cpu) - amdgpu_umc_process_ras_data_cb(adev, &err_data, NULL); - else + amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); + else if (reset) amdgpu_amdkfd_gpu_reset(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index fcbc8a9c9e06..61f899e54fd5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -296,7 +296,8 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev, uint64_t *mmap_offset); int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev, struct tile_config *config); -void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev); +void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, + bool reset); #if IS_ENABLED(CONFIG_HSA_AMD) void amdgpu_amdkfd_gpuvm_init_mem_limits(void); void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, -- cgit From fbcdbfde87509d523132b59f661a355c731139d0 Mon Sep 17 00:00:00 2001 From: Rajneesh Bhardwaj Date: Fri, 26 Nov 2021 15:15:04 -0500 Subject: drm/amdgpu: Don't inherit GEM object VMAs in child process MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 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 Reviewed-by: Christian König Signed-off-by: David Yat Sin Signed-off-by: Rajneesh Bhardwaj Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index c0d8f40a5b45..9a6507af1670 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -264,6 +264,9 @@ static int amdgpu_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_str !(vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC))) vma->vm_flags &= ~VM_MAYWRITE; + if (bo->kfd_bo) + vma->vm_flags |= VM_DONTCOPY; + return drm_gem_ttm_mmap(obj, vma); } -- cgit From 4da8b63944a4f4482303c9ad6efb18aa547d4630 Mon Sep 17 00:00:00 2001 From: sashank saye Date: Fri, 17 Dec 2021 07:50:09 -0500 Subject: drm/amdgpu: Send Message to SMU on aldebaran passthrough for sbr handling 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 Signed-off-by: sashank saye Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ce820ee2e2e6..5250c5bbbd76 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2619,11 +2619,10 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) if (r) DRM_ERROR("enable mgpu fan boost failed (%d).\n", r); - /* For XGMI + passthrough configuration on arcturus, enable light SBR */ - if (adev->asic_type == CHIP_ARCTURUS && - amdgpu_passthrough(adev) && - adev->gmc.xgmi.num_physical_nodes > 1) - smu_set_light_sbr(&adev->smu, true); + /* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */ + if (amdgpu_passthrough(adev) && ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1)|| + adev->asic_type == CHIP_ALDEBARAN )) + smu_handle_passthrough_sbr(&adev->smu, true); if (adev->gmc.xgmi.num_physical_nodes > 1) { mutex_lock(&mgpu_info.mutex); -- cgit From 11544d77e3974924c5a9c8a8320b996a3e9b2f8b Mon Sep 17 00:00:00 2001 From: Zongmin Zhou Date: Wed, 15 Dec 2021 17:23:37 +0800 Subject: drm/amdgpu: fixup bad vram size on gmc v8 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 Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 492ebed2915b..63b890f1e8af 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -515,10 +515,10 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) static int gmc_v8_0_mc_init(struct amdgpu_device *adev) { int r; + u32 tmp; adev->gmc.vram_width = amdgpu_atombios_get_vram_width(adev); if (!adev->gmc.vram_width) { - u32 tmp; int chansize, numchan; /* Get VRAM informations */ @@ -562,8 +562,15 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev) adev->gmc.vram_width = numchan * chansize; } /* size in MB on si */ - adev->gmc.mc_vram_size = RREG32(mmCONFIG_MEMSIZE) * 1024ULL * 1024ULL; - adev->gmc.real_vram_size = RREG32(mmCONFIG_MEMSIZE) * 1024ULL * 1024ULL; + tmp = RREG32(mmCONFIG_MEMSIZE); + /* some boards may have garbage in the upper 16 bits */ + if (tmp & 0xffff0000) { + DRM_INFO("Probable bad vram size: 0x%08x\n", tmp); + if (tmp & 0xffff) + tmp &= 0xffff; + } + adev->gmc.mc_vram_size = tmp * 1024ULL * 1024ULL; + adev->gmc.real_vram_size = adev->gmc.mc_vram_size; if (!(adev->flags & AMD_IS_APU)) { r = amdgpu_device_resize_fb_bar(adev); -- cgit From 87172e89dcc7b09b32a4eb5f21e35d310e3cb024 Mon Sep 17 00:00:00 2001 From: Leslie Shi Date: Thu, 16 Dec 2021 14:03:41 +0800 Subject: drm/amdgpu: Call amdgpu_device_unmap_mmio() if device is unplugged to prevent crash in GPU initialization failure MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [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 Reviewed-by: Andrey Grodzovsky Reviewed-by: Guchun Chen Acked-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5250c5bbbd76..43e9149e129a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3899,7 +3899,9 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) amdgpu_gart_dummy_page_fini(adev); - amdgpu_device_unmap_mmio(adev); + if (drm_dev_is_unplugged(adev_to_drm(adev))) + amdgpu_device_unmap_mmio(adev); + } void amdgpu_device_fini_sw(struct amdgpu_device *adev) -- cgit From fd5256cbe19609bfff0cc51e9d999a46020719da Mon Sep 17 00:00:00 2001 From: yipechai Date: Wed, 8 Dec 2021 16:04:14 +0800 Subject: drm/amdgpu: Remove the redundant code of psp bootloader functions 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 Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 78 +++++++--------------------------- 1 file changed, 15 insertions(+), 63 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 2176ef85f137..d0e76b36d4ab 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -277,13 +277,15 @@ static bool psp_v11_0_is_sos_alive(struct psp_context *psp) return sol_reg != 0x0; } -static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) +static int psp_v11_0_bootloader_load_component(struct psp_context *psp, + struct psp_bin_desc *bin_desc, + enum psp_bootloader_cmd bl_cmd) { int ret; uint32_t psp_gfxdrv_command_reg = 0; struct amdgpu_device *adev = psp->adev; - /* Check tOS sign of life register to confirm sys driver and sOS + /* Check sOS sign of life register to confirm sys driver and sOS * are already been loaded. */ if (psp_v11_0_is_sos_alive(psp)) @@ -293,13 +295,13 @@ static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) if (ret) return ret; - /* Copy PSP KDB binary to memory */ - psp_copy_fw(psp, psp->kdb.start_addr, psp->kdb.size_bytes); + /* Copy PSP System Driver binary to memory */ + psp_copy_fw(psp, bin_desc->start_addr, bin_desc->size_bytes); - /* Provide the PSP KDB to bootloader */ + /* Provide the sys driver to bootloader */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = PSP_BL__LOAD_KEY_DATABASE; + psp_gfxdrv_command_reg = bl_cmd; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, psp_gfxdrv_command_reg); @@ -308,69 +310,19 @@ static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) return ret; } -static int psp_v11_0_bootloader_load_spl(struct psp_context *psp) +static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) { - int ret; - uint32_t psp_gfxdrv_command_reg = 0; - struct amdgpu_device *adev = psp->adev; - - /* Check tOS sign of life register to confirm sys driver and sOS - * are already been loaded. - */ - if (psp_v11_0_is_sos_alive(psp)) - return 0; - - ret = psp_v11_0_wait_for_bootloader(psp); - if (ret) - return ret; - - /* Copy PSP SPL binary to memory */ - psp_copy_fw(psp, psp->spl.start_addr, psp->spl.size_bytes); - - /* Provide the PSP SPL to bootloader */ - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, - (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = PSP_BL__LOAD_TOS_SPL_TABLE; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, - psp_gfxdrv_command_reg); - - ret = psp_v11_0_wait_for_bootloader(psp); + return psp_v11_0_bootloader_load_component(psp, &psp->kdb, PSP_BL__LOAD_KEY_DATABASE); +} - return ret; +static int psp_v11_0_bootloader_load_spl(struct psp_context *psp) +{ + return psp_v11_0_bootloader_load_component(psp, &psp->spl, PSP_BL__LOAD_TOS_SPL_TABLE); } static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) { - int ret; - uint32_t psp_gfxdrv_command_reg = 0; - struct amdgpu_device *adev = psp->adev; - - /* Check sOS sign of life register to confirm sys driver and sOS - * are already been loaded. - */ - if (psp_v11_0_is_sos_alive(psp)) - return 0; - - ret = psp_v11_0_wait_for_bootloader(psp); - if (ret) - return ret; - - /* Copy PSP System Driver binary to memory */ - psp_copy_fw(psp, psp->sys.start_addr, psp->sys.size_bytes); - - /* Provide the sys driver to bootloader */ - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, - (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = PSP_BL__LOAD_SYSDRV; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, - psp_gfxdrv_command_reg); - - /* there might be handshake issue with hardware which needs delay */ - mdelay(20); - - ret = psp_v11_0_wait_for_bootloader(psp); - - return ret; + return psp_v11_0_bootloader_load_component(psp, &psp->sys, PSP_BL__LOAD_SYSDRV); } static int psp_v11_0_bootloader_load_sos(struct psp_context *psp) -- cgit From 6c92fe5fa5a1f24e50296de9d24d36a1b605b208 Mon Sep 17 00:00:00 2001 From: Kent Russell Date: Mon, 13 Dec 2021 13:14:17 -0500 Subject: drm/amdgpu: Increase potential product_name to 64 characters 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: Kent Russell Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c | 12 +++++------- 2 files changed, 7 insertions(+), 8 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index e701dedce344..8038e90daeaf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -813,6 +813,7 @@ struct amd_powerplay { #define AMDGPU_RESET_MAGIC_NUM 64 #define AMDGPU_MAX_DF_PERFMONS 4 +#define AMDGPU_PRODUCT_NAME_LEN 64 struct amdgpu_device { struct device *dev; struct pci_dev *pdev; @@ -1083,7 +1084,7 @@ struct amdgpu_device { /* Chip product information */ char product_number[16]; - char product_name[32]; + char product_name[AMDGPU_PRODUCT_NAME_LEN]; char serial[20]; atomic_t throttling_logging_enabled; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index 7709caeb233d..fdca0a447e74 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -88,7 +88,7 @@ static int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr, int amdgpu_fru_get_product_info(struct amdgpu_device *adev) { - unsigned char buff[34]; + unsigned char buff[AMDGPU_PRODUCT_NAME_LEN+2]; u32 addrptr; int size, len; @@ -131,12 +131,10 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) } len = size; - /* Product name should only be 32 characters. Any more, - * and something could be wrong. Cap it at 32 to be safe - */ - if (len >= sizeof(adev->product_name)) { - DRM_WARN("FRU Product Number is larger than 32 characters. This is likely a mistake"); - len = sizeof(adev->product_name) - 1; + if (len >= AMDGPU_PRODUCT_NAME_LEN) { + DRM_WARN("FRU Product Name is larger than %d characters. This is likely a mistake", + AMDGPU_PRODUCT_NAME_LEN); + len = AMDGPU_PRODUCT_NAME_LEN - 1; } /* Start at 2 due to buff using fields 0 and 1 for the address */ memcpy(adev->product_name, &buff[2], len); -- cgit From 67416bf85345b6990fe2adc42ab082bfefc578bd Mon Sep 17 00:00:00 2001 From: Kent Russell Date: Mon, 13 Dec 2021 13:59:43 -0500 Subject: drm/amdgpu: Access the FRU on Aldebaran 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 Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index fdca0a447e74..2a786e788627 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -56,6 +56,9 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev) return true; else return false; + case CHIP_ALDEBARAN: + /* All Aldebaran SKUs have the FRU */ + return true; default: return false; } @@ -91,10 +94,14 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) unsigned char buff[AMDGPU_PRODUCT_NAME_LEN+2]; u32 addrptr; int size, len; + int offset = 2; if (!is_fru_eeprom_supported(adev)) return 0; + if (adev->asic_type == CHIP_ALDEBARAN) + offset = 0; + /* If algo exists, it means that the i2c_adapter's initialized */ if (!adev->pm.smu_i2c.algo) { DRM_WARN("Cannot access FRU, EEPROM accessor not initialized"); @@ -137,7 +144,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) len = AMDGPU_PRODUCT_NAME_LEN - 1; } /* Start at 2 due to buff using fields 0 and 1 for the address */ - memcpy(adev->product_name, &buff[2], len); + memcpy(adev->product_name, &buff[offset], len); adev->product_name[len] = '\0'; addrptr += size + 1; @@ -155,7 +162,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) DRM_WARN("FRU Product Number is larger than 16 characters. This is likely a mistake"); len = sizeof(adev->product_number) - 1; } - memcpy(adev->product_number, &buff[2], len); + memcpy(adev->product_number, &buff[offset], len); adev->product_number[len] = '\0'; addrptr += size + 1; @@ -182,7 +189,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) DRM_WARN("FRU Serial Number is larger than 16 characters. This is likely a mistake"); len = sizeof(adev->serial) - 1; } - memcpy(adev->serial, &buff[2], len); + memcpy(adev->serial, &buff[offset], len); adev->serial[len] = '\0'; return 0; -- cgit From b6fd6e0f5eb8c6d10575d08a2c6df8ed83877e07 Mon Sep 17 00:00:00 2001 From: Surbhi Kakarya Date: Fri, 17 Dec 2021 12:01:09 -0500 Subject: drm/amdgpu: Check the memory can be accesssed by ttm_device_clear_dma_mappings. 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 Acked-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 43e9149e129a..dadfa03570d1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3895,7 +3895,8 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) amdgpu_irq_fini_hw(adev); - ttm_device_clear_dma_mappings(&adev->mman.bdev); + if (adev->mman.initialized) + ttm_device_clear_dma_mappings(&adev->mman.bdev); amdgpu_gart_dummy_page_fini(adev); -- cgit From 45e3d1db7d3ca013f1439248a268b0cdada5d5a3 Mon Sep 17 00:00:00 2001 From: Guchun Chen Date: Mon, 20 Dec 2021 22:32:29 +0800 Subject: drm/amdgpu: drop redundant semicolon A minor typo. Signed-off-by: Guchun Chen Reviewed-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index b13db855cc9a..580a5b387122 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1023,7 +1023,7 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &mes_v10_1_ip_block); break; default: - break;; + break; } return 0; } -- cgit From fec8c5244fc07b1f6a3249a8714489f594ff5c4f Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Mon, 20 Dec 2021 16:36:54 +0800 Subject: drm/amdgpu: save error count in RAS poison handler Otherwise the RAS error count couldn't be queried from sysfs. Signed-off-by: Tao Zhou Reviewed-by: Stanley.Yang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 168 ++++++++++++++++------------- drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 3 +- 3 files changed, 97 insertions(+), 76 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 0bf09a94d944..776a947b45df 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -727,7 +727,7 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bo /* CPU MCA will handle page retirement if connected_to_cpu is 1 */ if (!adev->gmc.xgmi.connected_to_cpu) - amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); + amdgpu_umc_poison_handler(adev, &err_data, reset); else if (reset) amdgpu_amdkfd_gpu_reset(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 0c33f367a4e5..46264a4002f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -23,79 +23,7 @@ #include "amdgpu_ras.h" -static int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, - void *ras_error_status, - struct amdgpu_iv_entry *entry) -{ - return amdgpu_umc_do_page_retirement(adev, ras_error_status, entry, true); -} - -int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) -{ - int r; - struct ras_fs_if fs_info = { - .sysfs_name = "umc_err_count", - }; - struct ras_ih_if ih_info = { - .cb = amdgpu_umc_process_ras_data_cb, - }; - - if (!adev->umc.ras_if) { - adev->umc.ras_if = - kmalloc(sizeof(struct ras_common_if), GFP_KERNEL); - if (!adev->umc.ras_if) - return -ENOMEM; - adev->umc.ras_if->block = AMDGPU_RAS_BLOCK__UMC; - adev->umc.ras_if->type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; - adev->umc.ras_if->sub_block_index = 0; - } - ih_info.head = fs_info.head = *adev->umc.ras_if; - - r = amdgpu_ras_late_init(adev, adev->umc.ras_if, - &fs_info, &ih_info); - if (r) - goto free; - - if (amdgpu_ras_is_supported(adev, adev->umc.ras_if->block)) { - r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0); - if (r) - goto late_fini; - } else { - r = 0; - goto free; - } - - /* ras init of specific umc version */ - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->err_cnt_init) - adev->umc.ras_funcs->err_cnt_init(adev); - - return 0; - -late_fini: - amdgpu_ras_late_fini(adev, adev->umc.ras_if, &ih_info); -free: - kfree(adev->umc.ras_if); - adev->umc.ras_if = NULL; - return r; -} - -void amdgpu_umc_ras_fini(struct amdgpu_device *adev) -{ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) && - adev->umc.ras_if) { - struct ras_common_if *ras_if = adev->umc.ras_if; - struct ras_ih_if ih_info = { - .head = *ras_if, - .cb = amdgpu_umc_process_ras_data_cb, - }; - - amdgpu_ras_late_fini(adev, ras_if, &ih_info); - kfree(ras_if); - } -} - -int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, +static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, void *ras_error_status, struct amdgpu_iv_entry *entry, bool reset) @@ -180,6 +108,100 @@ int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, return AMDGPU_RAS_SUCCESS; } +int amdgpu_umc_poison_handler(struct amdgpu_device *adev, + void *ras_error_status, + bool reset) +{ + int ret; + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + struct ras_common_if head = { + .block = AMDGPU_RAS_BLOCK__UMC, + }; + struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + + ret = + amdgpu_umc_do_page_retirement(adev, ras_error_status, NULL, reset); + + if (ret == AMDGPU_RAS_SUCCESS && obj) { + obj->err_data.ue_count += err_data->ue_count; + obj->err_data.ce_count += err_data->ce_count; + } + + return ret; +} + +static int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, + void *ras_error_status, + struct amdgpu_iv_entry *entry) +{ + return amdgpu_umc_do_page_retirement(adev, ras_error_status, entry, true); +} + +int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) +{ + int r; + struct ras_fs_if fs_info = { + .sysfs_name = "umc_err_count", + }; + struct ras_ih_if ih_info = { + .cb = amdgpu_umc_process_ras_data_cb, + }; + + if (!adev->umc.ras_if) { + adev->umc.ras_if = + kmalloc(sizeof(struct ras_common_if), GFP_KERNEL); + if (!adev->umc.ras_if) + return -ENOMEM; + adev->umc.ras_if->block = AMDGPU_RAS_BLOCK__UMC; + adev->umc.ras_if->type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; + adev->umc.ras_if->sub_block_index = 0; + } + ih_info.head = fs_info.head = *adev->umc.ras_if; + + r = amdgpu_ras_late_init(adev, adev->umc.ras_if, + &fs_info, &ih_info); + if (r) + goto free; + + if (amdgpu_ras_is_supported(adev, adev->umc.ras_if->block)) { + r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0); + if (r) + goto late_fini; + } else { + r = 0; + goto free; + } + + /* ras init of specific umc version */ + if (adev->umc.ras_funcs && + adev->umc.ras_funcs->err_cnt_init) + adev->umc.ras_funcs->err_cnt_init(adev); + + return 0; + +late_fini: + amdgpu_ras_late_fini(adev, adev->umc.ras_if, &ih_info); +free: + kfree(adev->umc.ras_if); + adev->umc.ras_if = NULL; + return r; +} + +void amdgpu_umc_ras_fini(struct amdgpu_device *adev) +{ + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) && + adev->umc.ras_if) { + struct ras_common_if *ras_if = adev->umc.ras_if; + struct ras_ih_if ih_info = { + .head = *ras_if, + .cb = amdgpu_umc_process_ras_data_cb, + }; + + amdgpu_ras_late_fini(adev, ras_if, &ih_info); + kfree(ras_if); + } +} + int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index 8d18d5121f66..b72194e8bfe5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -78,9 +78,8 @@ struct amdgpu_umc { int amdgpu_umc_ras_late_init(struct amdgpu_device *adev); void amdgpu_umc_ras_fini(struct amdgpu_device *adev); -int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, +int amdgpu_umc_poison_handler(struct amdgpu_device *adev, void *ras_error_status, - struct amdgpu_iv_entry *entry, bool reset); int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, -- cgit From 937ed9c8660a27ac4d36e8339c970a2ae7903c4d Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 21 Dec 2021 09:54:40 -0500 Subject: drm/amdgpu: fix runpm documentation It's not only supported by HG/PX laptops. It's supported by all dGPUs which supports BOCO/BACO functionality (runtime D3). BOCO - Bus Off, Chip Off. The entire chip is powered off. This is controlled by ACPI. BACO - Bus Active, Chip Off. The chip still shows up on the PCI bus, but the device itself is powered down. v2: fix missed HG/PX reference Reviewed-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index bc1355c6248d..7f94d218b74f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -331,10 +331,11 @@ module_param_named(aspm, amdgpu_aspm, int, 0444); /** * DOC: runpm (int) - * Override for runtime power management control for dGPUs in PX/HG laptops. The amdgpu driver can dynamically power down - * the dGPU on PX/HG laptops when it is idle. The default is -1 (auto enable). Setting the value to 0 disables this functionality. + * Override for runtime power management control for dGPUs. The amdgpu driver can dynamically power down + * the dGPUs when they are idle if supported. The default is -1 (auto enable). + * Setting the value to 0 disables this functionality. */ -MODULE_PARM_DESC(runpm, "PX runtime pm (2 = force enable with BAMACO, 1 = force enable with BACO, 0 = disable, -1 = PX only default)"); +MODULE_PARM_DESC(runpm, "PX runtime pm (2 = force enable with BAMACO, 1 = force enable with BACO, 0 = disable, -1 = auto)"); module_param_named(runpm, amdgpu_runtime_pm, int, 0444); /** -- cgit From 4a700546ec9b8fafedf569f363f905b9ec7f5ee8 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Fri, 17 Dec 2021 19:05:06 +0800 Subject: drm/amdgpu: put SMU into proper state on runpm suspending for BOCO capable platform By setting mp1_state as PP_MP1_STATE_UNLOAD, MP1 will do some proper cleanups and put itself into a state ready for PNP. That can workaround some random resuming failure observed on BOCO capable platforms. Signed-off-by: Evan Quan Acked-by: Alex Deucher Reviewed-by: Guchun Chen Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 7f94d218b74f..739e4d9303cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2251,12 +2251,27 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev) if (amdgpu_device_supports_px(drm_dev)) drm_dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; + /* + * By setting mp1_state as PP_MP1_STATE_UNLOAD, MP1 will do some + * proper cleanups and put itself into a state ready for PNP. That + * can address some random resuming failure observed on BOCO capable + * platforms. + * TODO: this may be also needed for PX capable platform. + */ + if (amdgpu_device_supports_boco(drm_dev)) + adev->mp1_state = PP_MP1_STATE_UNLOAD; + ret = amdgpu_device_suspend(drm_dev, false); if (ret) { adev->in_runpm = false; + if (amdgpu_device_supports_boco(drm_dev)) + adev->mp1_state = PP_MP1_STATE_NONE; return ret; } + if (amdgpu_device_supports_boco(drm_dev)) + adev->mp1_state = PP_MP1_STATE_NONE; + if (amdgpu_device_supports_px(drm_dev)) { /* Only need to handle PCI state in the driver for ATPX * PCI core handles it for _PR3. -- cgit From 6dc8265f9803ccb7e5da804e01601f0c14f270e0 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 12 Nov 2021 11:25:30 -0500 Subject: drm/amdgpu: always reset the asic in suspend (v2) If the platform suspend happens to fail and the power rail is not turned off, the GPU will be in an unknown state on resume, so reset the asic so that it will be in a known good state on resume even if the platform suspend failed. v2: handle s0ix Acked-by: Luben Tuikov Acked-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 739e4d9303cd..405b0d565ce3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2170,7 +2170,10 @@ static int amdgpu_pmops_suspend(struct device *dev) adev->in_s3 = true; r = amdgpu_device_suspend(drm_dev, true); adev->in_s3 = false; - + if (r) + return r; + if (!adev->in_s0ix) + r = amdgpu_asic_reset(adev); return r; } -- cgit From 0637d41786a3a9551f33ad8e15bdb40416362028 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 23 Dec 2021 14:13:02 -0500 Subject: drm/amdgpu: no DC support for headless chips Chips with no display hardware should return false for DC support. v2: drop Arcturus and Aldebaran Fixes: f7f12b25823c0d ("drm/amdgpu: default to true in amdgpu_device_asic_has_dc_support") Reviewed-by: Evan Quan Reviewed-by: Guchun Chen Reported-by: Tareque Md.Hanif Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index dadfa03570d1..c2a16b20de9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3185,6 +3185,12 @@ static void amdgpu_device_detect_sriov_bios(struct amdgpu_device *adev) bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) { switch (asic_type) { +#ifdef CONFIG_DRM_AMDGPU_SI + case CHIP_HAINAN: +#endif + case CHIP_TOPAZ: + /* chips with no display hardware */ + return false; #if defined(CONFIG_DRM_AMD_DC) case CHIP_TAHITI: case CHIP_PITCAIRN: -- cgit