Merge tag 'amd-drm-next-5.9-2020-07-17' of git://people.freedesktop.org/~agd5f/linux into drm-next

amd-drm-next-5.9-2020-07-17:

amdgpu:
- SI UVD/VCE clock support
- Updates for Sienna Cichlid
- Expose drm rotation property
- Atomfirmware updates for renoir
- updates to GPUVM hub handling for different register layouts
- swSMU restructuring and cleanups
- RAS fixes
- DC fixes
- mode1 reset support for Sienna Cichlid
- Add support for Navy Flounder GPUs

amdkfd:
- Add SMI events watch interface

UAPI:
- Add amdkfd SMI events watch interface
  Userspace which uses this interface:
  ROCm/rocm_smi_lib@2235ede

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200717132022.4014-1-alexander.deucher@amd.com

Author: airlied

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -193,6 +193,7 @@ static const bool debug_evictions; /* = false */
 #endif
 
 extern int amdgpu_tmz;
+extern int amdgpu_reset_method;
 
 #ifdef CONFIG_DRM_AMDGPU_SI
 extern int amdgpu_si_support;
@@ -1010,10 +1011,10 @@ int amdgpu_gpu_wait_for_idle(struct amdgpu_device *adev);
 
 void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
 			       uint32_t *buf, size_t size, bool write);
-uint32_t amdgpu_device_rreg(struct amdgpu_device *adev, uint32_t reg,
-			    uint32_t acc_flags);
-void amdgpu_device_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
+uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg,
 			uint32_t acc_flags);
+void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
+		    uint32_t acc_flags);
 void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
 		    uint32_t acc_flags);
 void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value);
@@ -1032,18 +1033,18 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
  */
 #define AMDGPU_REGS_NO_KIQ    (1<<1)
 
-#define RREG32_NO_KIQ(reg) amdgpu_device_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
-#define WREG32_NO_KIQ(reg, v) amdgpu_device_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
+#define RREG32_NO_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
+#define WREG32_NO_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
 
 #define RREG32_KIQ(reg) amdgpu_kiq_rreg(adev, (reg))
 #define WREG32_KIQ(reg, v) amdgpu_kiq_wreg(adev, (reg), (v))
 
 #define RREG8(reg) amdgpu_mm_rreg8(adev, (reg))
 #define WREG8(reg, v) amdgpu_mm_wreg8(adev, (reg), (v))
 
-#define RREG32(reg) amdgpu_device_rreg(adev, (reg), 0)
-#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_device_rreg(adev, (reg), 0))
-#define WREG32(reg, v) amdgpu_device_wreg(adev, (reg), (v), 0)
+#define RREG32(reg) amdgpu_mm_rreg(adev, (reg), 0)
+#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_mm_rreg(adev, (reg), 0))
+#define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0)
 #define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
 #define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
 #define RREG32_PCIE(reg) adev->pcie_rreg(adev, (reg))
@@ -1080,7 +1081,16 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 		tmp_ |= ((val) & ~(mask));			\
 		WREG32_PLL(reg, tmp_);				\
 	} while (0)
-#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_device_rreg((adev), (reg), false))
+
+#define WREG32_SMC_P(_Reg, _Val, _Mask)                         \
+	do {                                                    \
+		u32 tmp = RREG32_SMC(_Reg);                     \
+		tmp &= (_Mask);                                 \
+		tmp |= ((_Val) & ~(_Mask));                     \
+		WREG32_SMC(_Reg, tmp);                          \
+	} while (0)
+
+#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_mm_rreg((adev), (reg), false))
 #define RREG32_IO(reg) amdgpu_io_rreg(adev, (reg))
 #define WREG32_IO(reg, v) amdgpu_io_wreg(adev, (reg), (v))
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c

@@ -31,8 +31,6 @@
 #include "amdgpu_xgmi.h"
 #include <uapi/linux/kfd_ioctl.h>
 
-static const unsigned int compute_vmid_bitmap = 0xFF00;
-
 /* Total memory size in system memory and all GPU VRAM. Used to
  * estimate worst case amount of memory to reserve for page tables
  */
@@ -113,7 +111,9 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 
 	if (adev->kfd.dev) {
 		struct kgd2kfd_shared_resources gpu_resources = {
-			.compute_vmid_bitmap = compute_vmid_bitmap,
+			.compute_vmid_bitmap =
+				((1 << AMDGPU_NUM_VMID) - 1) -
+				((1 << adev->vm_manager.first_kfd_vmid) - 1),
 			.num_pipe_per_mec = adev->gfx.mec.num_pipe_per_mec,
 			.num_queue_per_pipe = adev->gfx.mec.num_queue_per_pipe,
 			.gpuvm_size = min(adev->vm_manager.max_pfn
@@ -637,10 +637,8 @@ void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle)
 
 bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
 {
-	if (adev->kfd.dev) {
-		if ((1 << vmid) & compute_vmid_bitmap)
-			return true;
-	}
+	if (adev->kfd.dev)
+		return vmid >= adev->vm_manager.first_kfd_vmid;
 
 	return false;
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c

@@ -111,6 +111,7 @@ int amdgpu_atomfirmware_allocate_fb_scratch(struct amdgpu_device *adev)
 
 union igp_info {
 	struct atom_integrated_system_info_v1_11 v11;
+	struct atom_integrated_system_info_v1_12 v12;
 };
 
 union umc_info {
@@ -214,6 +215,15 @@ amdgpu_atomfirmware_get_vram_info(struct amdgpu_device *adev,
 				if (vram_type)
 					*vram_type = convert_atom_mem_type_to_vram_type(adev, mem_type);
 				break;
+			case 12:
+				mem_channel_number = igp_info->v12.umachannelnumber;
+				/* channel width is 64 */
+				if (vram_width)
+					*vram_width = mem_channel_number * 64;
+				mem_type = igp_info->v12.memorytype;
+				if (vram_type)
+					*vram_type = convert_atom_mem_type_to_vram_type(adev, mem_type);
+				break;
 			default:
 				return -EINVAL;
 			}

drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c

@@ -1343,27 +1343,37 @@ static void amdgpu_ib_preempt_job_recovery(struct drm_gpu_scheduler *sched)
 static void amdgpu_ib_preempt_mark_partial_job(struct amdgpu_ring *ring)
 {
 	struct amdgpu_job *job;
-	struct drm_sched_job *s_job;
+	struct drm_sched_job *s_job, *tmp;
 	uint32_t preempt_seq;
 	struct dma_fence *fence, **ptr;
 	struct amdgpu_fence_driver *drv = &ring->fence_drv;
 	struct drm_gpu_scheduler *sched = &ring->sched;
+	bool preempted = true;
 
 	if (ring->funcs->type != AMDGPU_RING_TYPE_GFX)
 		return;
 
 	preempt_seq = le32_to_cpu(*(drv->cpu_addr + 2));
-	if (preempt_seq <= atomic_read(&drv->last_seq))
-		return;
+	if (preempt_seq <= atomic_read(&drv->last_seq)) {
+		preempted = false;
+		goto no_preempt;
+	}
 
 	preempt_seq &= drv->num_fences_mask;
 	ptr = &drv->fences[preempt_seq];
 	fence = rcu_dereference_protected(*ptr, 1);
 
+no_preempt:
 	spin_lock(&sched->job_list_lock);
-	list_for_each_entry(s_job, &sched->ring_mirror_list, node) {
+	list_for_each_entry_safe(s_job, tmp, &sched->ring_mirror_list, node) {
+		if (dma_fence_is_signaled(&s_job->s_fence->finished)) {
+			/* remove job from ring_mirror_list */
+			list_del_init(&s_job->node);
+			sched->ops->free_job(s_job);
+			continue;
+		}
 		job = to_amdgpu_job(s_job);
-		if (job->fence == fence)
+		if (preempted && job->fence == fence)
 			/* mark the job as preempted */
 			job->preemption_status |= AMDGPU_IB_PREEMPTED;
 	}
@@ -1461,10 +1471,12 @@ static int amdgpu_debugfs_sclk_set(void *data, u64 val)
 	}
 
 	if (is_support_sw_smu(adev)) {
-		ret = smu_get_dpm_freq_range(&adev->smu, SMU_SCLK, &min_freq, &max_freq, true);
+		ret = smu_get_dpm_freq_range(&adev->smu, SMU_SCLK, &min_freq, &max_freq);
 		if (ret || val > max_freq || val < min_freq)
 			return -EINVAL;
-		ret = smu_set_soft_freq_range(&adev->smu, SMU_SCLK, (uint32_t)val, (uint32_t)val, true);
+		ret = smu_set_soft_freq_range(&adev->smu, SMU_SCLK, (uint32_t)val, (uint32_t)val);
+	} else {
+		return 0;
 	}
 
 	pm_runtime_mark_last_busy(adev->ddev->dev);