From 29d48b87db64b6697ddad007548e51d032081c59 Mon Sep 17 00:00:00 2001 From: Philip Yang Date: Wed, 14 Dec 2022 10:15:17 -0500 Subject: drm/amdkfd: Fix kfd_process_device_init_vm error handling Should only destroy the ib_mem and let process cleanup worker to free the outstanding BOs. Reset the pointer in pdd->qpd structure, to avoid NULL pointer access in process destroy worker. BUG: kernel NULL pointer dereference, address: 0000000000000010 Call Trace: amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel+0x46/0xb0 [amdgpu] kfd_process_device_destroy_cwsr_dgpu+0x40/0x70 [amdgpu] kfd_process_destroy_pdds+0x71/0x190 [amdgpu] kfd_process_wq_release+0x2a2/0x3b0 [amdgpu] process_one_work+0x2a1/0x600 worker_thread+0x39/0x3d0 Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index a26257171ab7..6caa9dd57ff1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -689,13 +689,13 @@ void kfd_process_destroy_wq(void) } static void kfd_process_free_gpuvm(struct kgd_mem *mem, - struct kfd_process_device *pdd, void *kptr) + struct kfd_process_device *pdd, void **kptr) { struct kfd_dev *dev = pdd->dev; - if (kptr) { + if (kptr && *kptr) { amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(mem); - kptr = NULL; + *kptr = NULL; } amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(dev->adev, mem, pdd->drm_priv); @@ -795,7 +795,7 @@ static void kfd_process_device_destroy_ib_mem(struct kfd_process_device *pdd) if (!qpd->ib_kaddr || !qpd->ib_base) return; - kfd_process_free_gpuvm(qpd->ib_mem, pdd, qpd->ib_kaddr); + kfd_process_free_gpuvm(qpd->ib_mem, pdd, &qpd->ib_kaddr); } struct kfd_process *kfd_create_process(struct file *filep) @@ -1277,7 +1277,7 @@ static void kfd_process_device_destroy_cwsr_dgpu(struct kfd_process_device *pdd) if (!dev->cwsr_enabled || !qpd->cwsr_kaddr || !qpd->cwsr_base) return; - kfd_process_free_gpuvm(qpd->cwsr_mem, pdd, qpd->cwsr_kaddr); + kfd_process_free_gpuvm(qpd->cwsr_mem, pdd, &qpd->cwsr_kaddr); } void kfd_process_set_trap_handler(struct qcm_process_device *qpd, @@ -1598,8 +1598,8 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, return 0; err_init_cwsr: + kfd_process_device_destroy_ib_mem(pdd); err_reserve_ib_mem: - kfd_process_device_free_bos(pdd); pdd->drm_priv = NULL; return ret; -- cgit v1.2.3 From 1a799c4c190ea9f0e81028e3eb3037ed0ab17ff5 Mon Sep 17 00:00:00 2001 From: Philip Yang Date: Tue, 13 Dec 2022 00:50:03 -0500 Subject: drm/amdkfd: Fix double release compute pasid If kfd_process_device_init_vm returns failure after vm is converted to compute vm and vm->pasid set to compute pasid, KFD will not take pdd->drm_file reference. As a result, drm close file handler maybe called to release the compute pasid before KFD process destroy worker to release the same pasid and set vm->pasid to zero, this generates below WARNING backtrace and NULL pointer access. Add helper amdgpu_amdkfd_gpuvm_set_vm_pasid and call it at the last step of kfd_process_device_init_vm, to ensure vm pasid is the original pasid if acquiring vm failed or is the compute pasid with pdd->drm_file reference taken to avoid double release same pasid. amdgpu: Failed to create process VM object ida_free called for id=32770 which is not allocated. WARNING: CPU: 57 PID: 72542 at ../lib/idr.c:522 ida_free+0x96/0x140 RIP: 0010:ida_free+0x96/0x140 Call Trace: amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu] amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu] drm_file_free.part.13+0x216/0x270 [drm] drm_close_helper.isra.14+0x60/0x70 [drm] drm_release+0x6e/0xf0 [drm] __fput+0xcc/0x280 ____fput+0xe/0x20 task_work_run+0x96/0xc0 do_exit+0x3d0/0xc10 BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: 0010:ida_free+0x76/0x140 Call Trace: amdgpu_pasid_free_delayed+0xe1/0x2a0 [amdgpu] amdgpu_driver_postclose_kms+0x2d8/0x340 [amdgpu] drm_file_free.part.13+0x216/0x270 [drm] drm_close_helper.isra.14+0x60/0x70 [drm] drm_release+0x6e/0xf0 [drm] __fput+0xcc/0x280 ____fput+0xe/0x20 task_work_run+0x96/0xc0 do_exit+0x3d0/0xc10 Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 4 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 39 +++++++++++++++++------- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 12 ++++++-- 3 files changed, 40 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 589939631ed4..0040deaf8a83 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -270,8 +270,10 @@ int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_ (&((struct amdgpu_fpriv *) \ ((struct drm_file *)(drm_priv))->driver_priv)->vm) +int amdgpu_amdkfd_gpuvm_set_vm_pasid(struct amdgpu_device *adev, + struct file *filp, u32 pasid); int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, - struct file *filp, u32 pasid, + struct file *filp, void **process_info, struct dma_fence **ef); void amdgpu_amdkfd_gpuvm_release_process_vm(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 0a854bb8b47e..b15091d8310d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -1429,10 +1429,9 @@ static void amdgpu_amdkfd_gpuvm_unpin_bo(struct amdgpu_bo *bo) amdgpu_bo_unreserve(bo); } -int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, - struct file *filp, u32 pasid, - void **process_info, - struct dma_fence **ef) +int amdgpu_amdkfd_gpuvm_set_vm_pasid(struct amdgpu_device *adev, + struct file *filp, u32 pasid) + { struct amdgpu_fpriv *drv_priv; struct amdgpu_vm *avm; @@ -1443,10 +1442,6 @@ int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, return ret; avm = &drv_priv->vm; - /* Already a compute VM? */ - if (avm->process_info) - return -EINVAL; - /* Free the original amdgpu allocated pasid, * will be replaced with kfd allocated pasid. */ @@ -1455,14 +1450,36 @@ int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, amdgpu_vm_set_pasid(adev, avm, 0); } - /* Convert VM into a compute VM */ - ret = amdgpu_vm_make_compute(adev, avm); + ret = amdgpu_vm_set_pasid(adev, avm, pasid); if (ret) return ret; - ret = amdgpu_vm_set_pasid(adev, avm, pasid); + return 0; +} + +int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, + struct file *filp, + void **process_info, + struct dma_fence **ef) +{ + struct amdgpu_fpriv *drv_priv; + struct amdgpu_vm *avm; + int ret; + + ret = amdgpu_file_to_fpriv(filp, &drv_priv); if (ret) return ret; + avm = &drv_priv->vm; + + /* Already a compute VM? */ + if (avm->process_info) + return -EINVAL; + + /* Convert VM into a compute VM */ + ret = amdgpu_vm_make_compute(adev, avm); + if (ret) + return ret; + /* Initialize KFD part of the VM and process info */ ret = init_kfd_vm(avm, process_info, ef); if (ret) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 6caa9dd57ff1..51b1683ac5c1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -1576,9 +1576,9 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, p = pdd->process; dev = pdd->dev; - ret = amdgpu_amdkfd_gpuvm_acquire_process_vm( - dev->adev, drm_file, p->pasid, - &p->kgd_process_info, &p->ef); + ret = amdgpu_amdkfd_gpuvm_acquire_process_vm(dev->adev, drm_file, + &p->kgd_process_info, + &p->ef); if (ret) { pr_err("Failed to create process VM object\n"); return ret; @@ -1593,10 +1593,16 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, if (ret) goto err_init_cwsr; + ret = amdgpu_amdkfd_gpuvm_set_vm_pasid(dev->adev, drm_file, p->pasid); + if (ret) + goto err_set_pasid; + pdd->drm_file = drm_file; return 0; +err_set_pasid: + kfd_process_device_destroy_cwsr_dgpu(pdd); err_init_cwsr: kfd_process_device_destroy_ib_mem(pdd); err_reserve_ib_mem: -- cgit v1.2.3 From d118b18fb1da02b41df2da78cb2794b3638d89cd Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 15 Dec 2022 17:36:31 +0100 Subject: drm/amd/pm: avoid large variable on kernel stack The activity_monitor_external[] array is too big to fit on the kernel stack, resulting in this warning with clang: drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_7_ppt.c:1438:12: error: stack frame size (1040) exceeds limit (1024) in 'smu_v13_0_7_get_power_profile_mode' [-Werror,-Wframe-larger-than] Use dynamic allocation instead. It should also be possible to have single element here instead of the array, but this seems easier. v2: fix up argument to sizeof() (Alex) Fixes: 334682ae8151 ("drm/amd/pm: enable workload type change on smu_v13_0_7") Signed-off-by: Arnd Bergmann Signed-off-by: Alex Deucher --- .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index ab1c004606be..07e3dc18c8b8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -1440,7 +1440,7 @@ static int smu_v13_0_7_get_power_limit(struct smu_context *smu, static int smu_v13_0_7_get_power_profile_mode(struct smu_context *smu, char *buf) { - DpmActivityMonitorCoeffIntExternal_t activity_monitor_external[PP_SMC_POWER_PROFILE_COUNT]; + DpmActivityMonitorCoeffIntExternal_t *activity_monitor_external; uint32_t i, j, size = 0; int16_t workload_type = 0; int result = 0; @@ -1448,6 +1448,12 @@ static int smu_v13_0_7_get_power_profile_mode(struct smu_context *smu, char *buf if (!buf) return -EINVAL; + activity_monitor_external = kcalloc(PP_SMC_POWER_PROFILE_COUNT, + sizeof(*activity_monitor_external), + GFP_KERNEL); + if (!activity_monitor_external) + return -ENOMEM; + size += sysfs_emit_at(buf, size, " "); for (i = 0; i <= PP_SMC_POWER_PROFILE_WINDOW3D; i++) size += sysfs_emit_at(buf, size, "%-14s%s", amdgpu_pp_profile_name[i], @@ -1460,15 +1466,17 @@ static int smu_v13_0_7_get_power_profile_mode(struct smu_context *smu, char *buf workload_type = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_WORKLOAD, i); - if (workload_type < 0) - return -EINVAL; + if (workload_type < 0) { + result = -EINVAL; + goto out; + } result = smu_cmn_update_table(smu, SMU_TABLE_ACTIVITY_MONITOR_COEFF, workload_type, (void *)(&activity_monitor_external[i]), false); if (result) { dev_err(smu->adev->dev, "[%s] Failed to get activity monitor!", __func__); - return result; + goto out; } } @@ -1496,7 +1504,10 @@ do { \ PRINT_DPM_MONITOR(Fclk_BoosterFreq); #undef PRINT_DPM_MONITOR - return size; + result = size; +out: + kfree(activity_monitor_external); + return result; } static int smu_v13_0_7_set_power_profile_mode(struct smu_context *smu, long *input, uint32_t size) -- cgit v1.2.3 From afa6646b1c5d3affd541f76bd7476e4b835a9174 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 16 Dec 2022 11:42:20 -0500 Subject: drm/amdgpu: skip MES for S0ix as well since it's part of GFX It's also part of gfxoff. Cc: stable@vger.kernel.org # 6.0, 6.1 Reviewed-by: Mario Limonciello Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 64660a41d53c..afe6af9c0138 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3016,14 +3016,15 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) continue; } - /* skip suspend of gfx and psp for S0ix + /* skip suspend of gfx/mes and psp for S0ix * gfx is in gfxoff state, so on resume it will exit gfxoff just * like at runtime. PSP is also part of the always on hardware * so no need to suspend it. */ if (adev->in_s0ix && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX)) + adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX || + adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_MES)) continue; /* XXX handle errors */ -- cgit v1.2.3 From 272b981416f8be0180c4d8066f90635fa7c1c501 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Thu, 15 Dec 2022 13:38:46 +0800 Subject: drm/amd/pm: bump SMU13.0.0 driver_if header to version 0x34 To fit the latest PMFW and suppress the warning emerged on driver loading. Signed-off-by: Evan Quan Acked-by: Alex Deucher Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0, 6.1 --- drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h | 2 +- drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h | 1 + drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 2 ++ 3 files changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h index b76f0f7e4299..d6b964cf73bd 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h @@ -522,9 +522,9 @@ typedef enum { TEMP_HOTSPOT_M, TEMP_MEM, TEMP_VR_GFX, + TEMP_VR_SOC, TEMP_VR_MEM0, TEMP_VR_MEM1, - TEMP_VR_SOC, TEMP_VR_U, TEMP_LIQUID0, TEMP_LIQUID1, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index ea29ac6a80e6..e8c6febb8b64 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -28,6 +28,7 @@ #define SMU13_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU13_DRIVER_IF_VERSION_YELLOW_CARP 0x04 #define SMU13_DRIVER_IF_VERSION_ALDE 0x08 +#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0 0x34 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x07 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04 #define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10 0x32 diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index e3a80ac987df..e54b760b875b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -290,6 +290,8 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_ALDE; break; case IP_VERSION(13, 0, 0): + smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0; + break; case IP_VERSION(13, 0, 10): smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10; break; -- cgit v1.2.3 From e73fc71e8f015d61f3adca7659cb209fd5117aa5 Mon Sep 17 00:00:00 2001 From: Evan Quan Date: Fri, 16 Dec 2022 17:04:24 +0800 Subject: drm/amd/pm: correct the fan speed retrieving in PWM for some SMU13 asics For SMU 13.0.0 and 13.0.7, the output from PMFW is in percent. Driver need to convert that into correct PMW(255) based. Signed-off-by: Evan Quan Acked-by: Alex Deucher Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0, 6.1 --- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 17 ++++++++++++++--- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 17 ++++++++++++++--- 2 files changed, 28 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 713fb6ad39f6..9643b21c636a 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -1441,12 +1441,23 @@ out: static int smu_v13_0_0_get_fan_speed_pwm(struct smu_context *smu, uint32_t *speed) { + int ret; + if (!speed) return -EINVAL; - return smu_v13_0_0_get_smu_metrics_data(smu, - METRICS_CURR_FANPWM, - speed); + ret = smu_v13_0_0_get_smu_metrics_data(smu, + METRICS_CURR_FANPWM, + speed); + if (ret) { + dev_err(smu->adev->dev, "Failed to get fan speed(PWM)!"); + return ret; + } + + /* Convert the PMFW output which is in percent to pwm(255) based */ + *speed = MIN(*speed * 255 / 100, 255); + + return 0; } static int smu_v13_0_0_get_fan_speed_rpm(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 07e3dc18c8b8..5c6c6ad011ca 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -1363,12 +1363,23 @@ static int smu_v13_0_7_populate_umd_state_clk(struct smu_context *smu) static int smu_v13_0_7_get_fan_speed_pwm(struct smu_context *smu, uint32_t *speed) { + int ret; + if (!speed) return -EINVAL; - return smu_v13_0_7_get_smu_metrics_data(smu, - METRICS_CURR_FANPWM, - speed); + ret = smu_v13_0_7_get_smu_metrics_data(smu, + METRICS_CURR_FANPWM, + speed); + if (ret) { + dev_err(smu->adev->dev, "Failed to get fan speed(PWM)!"); + return ret; + } + + /* Convert the PMFW output which is in percent to pwm(255) based */ + *speed = MIN(*speed * 255 / 100, 255); + + return 0; } static int smu_v13_0_7_get_fan_speed_rpm(struct smu_context *smu, -- cgit v1.2.3 From 8660495a9c5b9afeec4cc006b3b75178f0fb2f10 Mon Sep 17 00:00:00 2001 From: Tim Huang Date: Mon, 19 Dec 2022 18:32:32 +0800 Subject: drm/amdgpu: skip mes self test after s0i3 resume for MES IP v11.0 MES is part of gfxoff and MES suspend and resume are skipped for S0i3. But the mes_self_test call path is still in the amdgpu_device_ip_late_init. it's should also be skipped for s0ix as no hardware re-initialization happened. Besides, mes_self_test will free the BO that triggers a lot of warning messages while in the suspend state. [ 81.656085] WARNING: CPU: 2 PID: 1550 at drivers/gpu/drm/amd/amdgpu/amdgpu_object.c:425 amdgpu_bo_free_kernel+0xfc/0x110 [amdgpu] [ 81.679435] Call Trace: [ 81.679726] [ 81.679981] amdgpu_mes_remove_hw_queue+0x17a/0x230 [amdgpu] [ 81.680857] amdgpu_mes_self_test+0x390/0x430 [amdgpu] [ 81.681665] mes_v11_0_late_init+0x37/0x50 [amdgpu] [ 81.682423] amdgpu_device_ip_late_init+0x53/0x280 [amdgpu] [ 81.683257] amdgpu_device_resume+0xae/0x2a0 [amdgpu] [ 81.684043] amdgpu_pmops_resume+0x37/0x70 [amdgpu] [ 81.684818] pci_pm_resume+0x5c/0xa0 [ 81.685247] ? pci_pm_thaw+0x90/0x90 [ 81.685658] dpm_run_callback+0x4e/0x160 [ 81.686110] device_resume+0xad/0x210 [ 81.686529] async_resume+0x1e/0x40 [ 81.686931] async_run_entry_fn+0x33/0x120 [ 81.687405] process_one_work+0x21d/0x3f0 [ 81.687869] worker_thread+0x4a/0x3c0 [ 81.688293] ? process_one_work+0x3f0/0x3f0 [ 81.688777] kthread+0xff/0x130 [ 81.689157] ? kthread_complete_and_exit+0x20/0x20 [ 81.689707] ret_from_fork+0x22/0x30 [ 81.690118] [ 81.690380] ---[ end trace 0000000000000000 ]--- v2: make the comment clean and use adev->in_s0ix instead of adev->suspend Signed-off-by: Tim Huang Reviewed-by: Mario Limonciello Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0, 6.1 --- drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 5459366f49ff..970b066b37bb 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -1342,7 +1342,8 @@ static int mes_v11_0_late_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!amdgpu_in_reset(adev) && + /* it's only intended for use in mes_self_test case, not for s0ix and reset */ + if (!amdgpu_in_reset(adev) && !adev->in_s0ix && (adev->ip_versions[GC_HWIP][0] != IP_VERSION(11, 0, 3))) amdgpu_mes_self_test(adev); -- cgit v1.2.3 From e1d900df63adcb748905131dd6258e570e11aed1 Mon Sep 17 00:00:00 2001 From: Saleemkhan Jamadar Date: Tue, 20 Dec 2022 13:21:44 +0530 Subject: drm/amdgpu: enable VCN DPG for GC IP v11.0.4 Enable VCN Dynamic Power Gating control for GC IP v11.0.4. Signed-off-by: Saleemkhan Jamadar Reviewed-by: Veerabadhran Gopalakrishnan Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0, 6.1 --- drivers/gpu/drm/amd/amdgpu/soc21.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index 7d5fdf450d0c..5562670b7b52 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -666,6 +666,7 @@ static int soc21_common_early_init(void *handle) AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG; adev->pg_flags = AMD_PG_SUPPORT_VCN | + AMD_PG_SUPPORT_VCN_DPG | AMD_PG_SUPPORT_GFX_PG | AMD_PG_SUPPORT_JPEG; adev->external_rev_id = adev->rev_id + 0x1; -- cgit v1.2.3 From c1c4a8b217213c1924eabf4f28385bbee9cc50c1 Mon Sep 17 00:00:00 2001 From: Christian König Date: Mon, 19 Dec 2022 11:47:18 +0100 Subject: drm/amdgpu: grab extra fence reference for drm_sched_job_add_dependency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit That function consumes the reference. Reviewed-by: Luben Tuikov Reported-by: Borislav Petkov (AMD) Tested-by: Borislav Petkov (AMD) Signed-off-by: Christian König Fixes: aab9cf7b6954 ("drm/amdgpu: use scheduler dependencies for VM updates") Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c index 59cf64216fbb..535cd6569bcc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c @@ -238,8 +238,10 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p, /* Wait for PD/PT moves to be completed */ dma_resv_iter_begin(&cursor, bo->tbo.base.resv, DMA_RESV_USAGE_KERNEL); dma_resv_for_each_fence_unlocked(&cursor, fence) { + dma_fence_get(fence); r = drm_sched_job_add_dependency(&p->job->base, fence); if (r) { + dma_fence_put(fence); dma_resv_iter_end(&cursor); return r; } -- cgit v1.2.3