From 127ed492ad2df0aa2351a1ad32a793ae7d91161b Mon Sep 17 00:00:00 2001 From: Lijo Lazar Date: Mon, 30 Jun 2025 19:11:33 +0530 Subject: [PATCH 01/11] drm/amdgpu: Pass adev pointer to functions Pass amdgpu device context instead of drm device context to some amdgpu_device_* functions. DRM device context is not required in those functions. No functional change. Signed-off-by: Lijo Lazar Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 24 ++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 8 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 65 ++++++++++------------ drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 15 +++-- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4 +- drivers/gpu/drm/amd/pm/amdgpu_pm.c | 4 +- 6 files changed, 57 insertions(+), 63 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 7edb7ba91786..1f531b5f594d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1562,16 +1562,16 @@ void amdgpu_device_program_register_sequence(struct amdgpu_device *adev, int amdgpu_device_mode1_reset(struct amdgpu_device *adev); int amdgpu_device_link_reset(struct amdgpu_device *adev); -bool amdgpu_device_supports_atpx(struct drm_device *dev); -bool amdgpu_device_supports_px(struct drm_device *dev); -bool amdgpu_device_supports_boco(struct drm_device *dev); -bool amdgpu_device_supports_smart_shift(struct drm_device *dev); -int amdgpu_device_supports_baco(struct drm_device *dev); +bool amdgpu_device_supports_atpx(struct amdgpu_device *adev); +bool amdgpu_device_supports_px(struct amdgpu_device *adev); +bool amdgpu_device_supports_boco(struct amdgpu_device *adev); +bool amdgpu_device_supports_smart_shift(struct amdgpu_device *adev); +int amdgpu_device_supports_baco(struct amdgpu_device *adev); void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev); bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev, struct amdgpu_device *peer_adev); -int amdgpu_device_baco_enter(struct drm_device *dev); -int amdgpu_device_baco_exit(struct drm_device *dev); +int amdgpu_device_baco_enter(struct amdgpu_device *adev); +int amdgpu_device_baco_exit(struct amdgpu_device *adev); void amdgpu_device_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring); @@ -1674,7 +1674,8 @@ int amdgpu_acpi_pcie_performance_request(struct amdgpu_device *adev, u8 perf_req, bool advertise); int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, u8 dev_state, bool drv_state); -int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_state); +int amdgpu_acpi_smart_shift_update(struct amdgpu_device *adev, + enum amdgpu_ss ss_state); int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev); int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset, u64 *tmr_size); @@ -1705,8 +1706,11 @@ static inline void amdgpu_acpi_release(void) { } static inline bool amdgpu_acpi_is_power_shift_control_supported(void) { return false; } static inline int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, u8 dev_state, bool drv_state) { return 0; } -static inline int amdgpu_acpi_smart_shift_update(struct drm_device *dev, - enum amdgpu_ss ss_state) { return 0; } +static inline int amdgpu_acpi_smart_shift_update(struct amdgpu_device *adev, + enum amdgpu_ss ss_state) +{ + return 0; +} static inline void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps) { } #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index f5466c592d94..b047fdf81543 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -811,18 +811,18 @@ int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, /** * amdgpu_acpi_smart_shift_update - update dGPU device state to SBIOS * - * @dev: drm_device pointer + * @adev: amdgpu device pointer * @ss_state: current smart shift event * * returns 0 on success, * otherwise return error number. */ -int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_state) +int amdgpu_acpi_smart_shift_update(struct amdgpu_device *adev, + enum amdgpu_ss ss_state) { - struct amdgpu_device *adev = drm_to_adev(dev); int r; - if (!amdgpu_device_supports_smart_shift(dev)) + if (!amdgpu_device_supports_smart_shift(adev)) return 0; switch (ss_state) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index d282c0753b14..45b44dec0d7f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -411,19 +411,16 @@ static const struct attribute_group amdgpu_board_attrs_group = { static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev); - /** * amdgpu_device_supports_px - Is the device a dGPU with ATPX power control * - * @dev: drm_device pointer + * @adev: amdgpu device pointer * * Returns true if the device is a dGPU with ATPX power control, * otherwise return false. */ -bool amdgpu_device_supports_px(struct drm_device *dev) +bool amdgpu_device_supports_px(struct amdgpu_device *adev) { - struct amdgpu_device *adev = drm_to_adev(dev); - if ((adev->flags & AMD_IS_PX) && !amdgpu_is_atpx_hybrid()) return true; return false; @@ -432,15 +429,13 @@ bool amdgpu_device_supports_px(struct drm_device *dev) /** * amdgpu_device_supports_boco - Is the device a dGPU with ACPI power resources * - * @dev: drm_device pointer + * @adev: amdgpu device pointer * * Returns true if the device is a dGPU with ACPI power control, * otherwise return false. */ -bool amdgpu_device_supports_boco(struct drm_device *dev) +bool amdgpu_device_supports_boco(struct amdgpu_device *adev) { - struct amdgpu_device *adev = drm_to_adev(dev); - if (!IS_ENABLED(CONFIG_HOTPLUG_PCI_PCIE)) return false; @@ -453,29 +448,24 @@ bool amdgpu_device_supports_boco(struct drm_device *dev) /** * amdgpu_device_supports_baco - Does the device support BACO * - * @dev: drm_device pointer + * @adev: amdgpu device pointer * * Return: * 1 if the device supports BACO; * 3 if the device supports MACO (only works if BACO is supported) * otherwise return 0. */ -int amdgpu_device_supports_baco(struct drm_device *dev) +int amdgpu_device_supports_baco(struct amdgpu_device *adev) { - struct amdgpu_device *adev = drm_to_adev(dev); - return amdgpu_asic_supports_baco(adev); } void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev) { - struct drm_device *dev; int bamaco_support; - dev = adev_to_drm(adev); - adev->pm.rpm_mode = AMDGPU_RUNPM_NONE; - bamaco_support = amdgpu_device_supports_baco(dev); + bamaco_support = amdgpu_device_supports_baco(adev); switch (amdgpu_runtime_pm) { case 2: @@ -495,10 +485,12 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev) break; case -1: case -2: - if (amdgpu_device_supports_px(dev)) { /* enable PX as runtime mode */ + if (amdgpu_device_supports_px(adev)) { + /* enable PX as runtime mode */ adev->pm.rpm_mode = AMDGPU_RUNPM_PX; dev_info(adev->dev, "Using ATPX for runtime pm\n"); - } else if (amdgpu_device_supports_boco(dev)) { /* enable boco as runtime mode */ + } else if (amdgpu_device_supports_boco(adev)) { + /* enable boco as runtime mode */ adev->pm.rpm_mode = AMDGPU_RUNPM_BOCO; dev_info(adev->dev, "Using BOCO for runtime pm\n"); } else { @@ -547,14 +539,14 @@ void amdgpu_device_detect_runtime_pm_mode(struct amdgpu_device *adev) * amdgpu_device_supports_smart_shift - Is the device dGPU with * smart shift support * - * @dev: drm_device pointer + * @adev: amdgpu device pointer * * Returns true if the device is a dGPU with Smart Shift support, * otherwise returns false. */ -bool amdgpu_device_supports_smart_shift(struct drm_device *dev) +bool amdgpu_device_supports_smart_shift(struct amdgpu_device *adev) { - return (amdgpu_device_supports_boco(dev) && + return (amdgpu_device_supports_boco(adev) && amdgpu_acpi_is_power_shift_control_supported()); } @@ -2200,7 +2192,8 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, struct drm_device *dev = pci_get_drvdata(pdev); int r; - if (amdgpu_device_supports_px(dev) && state == VGA_SWITCHEROO_OFF) + if (amdgpu_device_supports_px(drm_to_adev(dev)) && + state == VGA_SWITCHEROO_OFF) return; if (state == VGA_SWITCHEROO_ON) { @@ -4192,13 +4185,13 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) if (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) { task_barrier_enter(&hive->tb); - adev->asic_reset_res = amdgpu_device_baco_enter(adev_to_drm(adev)); + adev->asic_reset_res = amdgpu_device_baco_enter(adev); if (adev->asic_reset_res) goto fail; task_barrier_exit(&hive->tb); - adev->asic_reset_res = amdgpu_device_baco_exit(adev_to_drm(adev)); + adev->asic_reset_res = amdgpu_device_baco_exit(adev); if (adev->asic_reset_res) goto fail; @@ -4353,7 +4346,6 @@ static void amdgpu_device_set_mcbp(struct amdgpu_device *adev) int amdgpu_device_init(struct amdgpu_device *adev, uint32_t flags) { - struct drm_device *ddev = adev_to_drm(adev); struct pci_dev *pdev = adev->pdev; int r, i; bool px = false; @@ -4814,7 +4806,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, if ((adev->pdev->class >> 8) == PCI_CLASS_DISPLAY_VGA) vga_client_register(adev->pdev, amdgpu_device_vga_set_decode); - px = amdgpu_device_supports_px(ddev); + px = amdgpu_device_supports_px(adev); if (px || (!dev_is_removable(&adev->pdev->dev) && apple_gmux_detect(NULL, NULL))) @@ -4980,7 +4972,7 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev) kfree(adev->xcp_mgr); adev->xcp_mgr = NULL; - px = amdgpu_device_supports_px(adev_to_drm(adev)); + px = amdgpu_device_supports_px(adev); if (px || (!dev_is_removable(&adev->pdev->dev) && apple_gmux_detect(NULL, NULL))) @@ -5152,7 +5144,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) return r; } - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3)) + if (amdgpu_acpi_smart_shift_update(adev, AMDGPU_SS_DEV_D3)) dev_warn(adev->dev, "smart shift update failed\n"); if (notify_clients) @@ -5321,7 +5313,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) } adev->in_suspend = false; - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0)) + if (amdgpu_acpi_smart_shift_update(adev, AMDGPU_SS_DEV_D0)) dev_warn(adev->dev, "smart shift update failed\n"); return 0; @@ -6365,7 +6357,8 @@ static int amdgpu_device_sched_resume(struct list_head *device_list, amdgpu_vf_error_put(tmp_adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r); } else { dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter)); - if (amdgpu_acpi_smart_shift_update(adev_to_drm(tmp_adev), AMDGPU_SS_DEV_D0)) + if (amdgpu_acpi_smart_shift_update(tmp_adev, + AMDGPU_SS_DEV_D0)) dev_warn(tmp_adev->dev, "smart shift update failed\n"); } @@ -6839,12 +6832,11 @@ bool amdgpu_device_is_peer_accessible(struct amdgpu_device *adev, #endif } -int amdgpu_device_baco_enter(struct drm_device *dev) +int amdgpu_device_baco_enter(struct amdgpu_device *adev) { - struct amdgpu_device *adev = drm_to_adev(dev); struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - if (!amdgpu_device_supports_baco(dev)) + if (!amdgpu_device_supports_baco(adev)) return -ENOTSUPP; if (ras && adev->ras_enabled && @@ -6854,13 +6846,12 @@ int amdgpu_device_baco_enter(struct drm_device *dev) return amdgpu_dpm_baco_enter(adev); } -int amdgpu_device_baco_exit(struct drm_device *dev) +int amdgpu_device_baco_exit(struct amdgpu_device *adev) { - struct amdgpu_device *adev = drm_to_adev(dev); struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); int ret = 0; - if (!amdgpu_device_supports_baco(dev)) + if (!amdgpu_device_supports_baco(adev)) return -ENOTSUPP; ret = amdgpu_dpm_baco_exit(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index b299e15bb5e5..4f8632737574 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2457,10 +2457,10 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) { /* only need to skip on ATPX */ - if (amdgpu_device_supports_px(ddev)) + if (amdgpu_device_supports_px(adev)) dev_pm_set_driver_flags(ddev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); /* we want direct complete for BOCO */ - if (amdgpu_device_supports_boco(ddev)) + if (amdgpu_device_supports_boco(adev)) dev_pm_set_driver_flags(ddev->dev, DPM_FLAG_SMART_PREPARE | DPM_FLAG_SMART_SUSPEND | DPM_FLAG_MAY_SKIP_RESUME); @@ -2493,9 +2493,9 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, * into D0 state. Then there will be a PMFW-aware D-state * transition(D0->D3) on runpm suspend. */ - if (amdgpu_device_supports_baco(ddev) && + if (amdgpu_device_supports_baco(adev) && !(adev->flags & AMD_IS_APU) && - (adev->asic_type >= CHIP_NAVI10)) + adev->asic_type >= CHIP_NAVI10) amdgpu_get_secondary_funcs(adev); } @@ -2560,8 +2560,7 @@ static int amdgpu_pmops_prepare(struct device *dev) /* Return a positive number here so * DPM_FLAG_SMART_SUSPEND works properly */ - if (amdgpu_device_supports_boco(drm_dev) && - pm_runtime_suspended(dev)) + if (amdgpu_device_supports_boco(adev) && pm_runtime_suspended(dev)) return 1; /* if we will not support s3 or s2i for the device @@ -2834,7 +2833,7 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev) /* nothing to do */ } else if ((adev->pm.rpm_mode == AMDGPU_RUNPM_BACO) || (adev->pm.rpm_mode == AMDGPU_RUNPM_BAMACO)) { - amdgpu_device_baco_enter(drm_dev); + amdgpu_device_baco_enter(adev); } dev_dbg(&pdev->dev, "asic/device is runtime suspended\n"); @@ -2875,7 +2874,7 @@ static int amdgpu_pmops_runtime_resume(struct device *dev) pci_set_master(pdev); } else if ((adev->pm.rpm_mode == AMDGPU_RUNPM_BACO) || (adev->pm.rpm_mode == AMDGPU_RUNPM_BAMACO)) { - amdgpu_device_baco_exit(drm_dev); + amdgpu_device_baco_exit(adev); } ret = amdgpu_device_resume(drm_dev, false); if (ret) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 195ed81d39ff..bfabb64e5535 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -91,7 +91,7 @@ void amdgpu_driver_unload_kms(struct drm_device *dev) if (adev->rmmio == NULL) return; - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DRV_UNLOAD)) + if (amdgpu_acpi_smart_shift_update(adev, AMDGPU_SS_DRV_UNLOAD)) DRM_WARN("smart shift update failed\n"); amdgpu_acpi_fini(adev); @@ -161,7 +161,7 @@ int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags) if (acpi_status) dev_dbg(dev->dev, "Error during ACPI methods call\n"); - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DRV_LOAD)) + if (amdgpu_acpi_smart_shift_update(adev, AMDGPU_SS_DRV_LOAD)) DRM_WARN("smart shift update failed\n"); out: diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index edd9895b46c0..4b151bbaffaa 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -1890,7 +1890,7 @@ static ssize_t amdgpu_set_smartshift_bias(struct device *dev, static int ss_power_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, uint32_t mask, enum amdgpu_device_attr_states *states) { - if (!amdgpu_device_supports_smart_shift(adev_to_drm(adev))) + if (!amdgpu_device_supports_smart_shift(adev)) *states = ATTR_STATE_UNSUPPORTED; return 0; @@ -1901,7 +1901,7 @@ static int ss_bias_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ { uint32_t ss_power; - if (!amdgpu_device_supports_smart_shift(adev_to_drm(adev))) + if (!amdgpu_device_supports_smart_shift(adev)) *states = ATTR_STATE_UNSUPPORTED; else if (amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE, (void *)&ss_power)) From 14b2d71a9a24727f1b9f2131ed5eb2e345840a3a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 26 Jun 2025 17:51:02 -0400 Subject: [PATCH 02/11] drm/amdgpu/gfx10: fix KGQ reset sequence MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Need to reinit the ring before remapping it and all of the KIQ handling needs to be within the kiq lock. Fixes: 1741281a157f ("drm/amdgpu/gfx10: add ring reset callbacks") Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 8c377ecbb8a7..5e099b5dc9a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -9544,7 +9544,7 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, spin_lock_irqsave(&kiq->ring_lock, flags); - if (amdgpu_ring_alloc(kiq_ring, 5 + 7 + 7 + kiq->pmf->map_queues_size)) { + if (amdgpu_ring_alloc(kiq_ring, 5 + 7 + 7)) { spin_unlock_irqrestore(&kiq->ring_lock, flags); return -ENOMEM; } @@ -9564,12 +9564,9 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, 0, 1, 0x20); gfx_v10_0_ring_emit_reg_wait(kiq_ring, SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), 0, 0xffffffff); - kiq->pmf->kiq_map_queues(kiq_ring, ring); amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - r = amdgpu_ring_test_ring(kiq_ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); if (r) return r; @@ -9579,6 +9576,19 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, return r; } + spin_lock_irqsave(&kiq->ring_lock, flags); + + if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size)) { + spin_unlock_irqrestore(&kiq->ring_lock, flags); + return -ENOMEM; + } + kiq->pmf->kiq_map_queues(kiq_ring, ring); + amdgpu_ring_commit(kiq_ring); + r = amdgpu_ring_test_ring(kiq_ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); + if (r) + return r; + r = amdgpu_ring_test_ring(ring); if (r) return r; From a54e4639c4ef37a0241bac7d2a77f2e6ffb57099 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Mon, 30 Jun 2025 23:26:17 +0300 Subject: [PATCH 03/11] drm/amd/pm/powerplay/hwmgr/smu_helper: fix order of mask and value There is a small typo in phm_wait_on_indirect_register(). Swap mask and value arguments provided to phm_wait_on_register() so that they satisfy the function signature and actual usage scheme. Found by Linux Verification Center (linuxtesting.org) with Svace static analysis tool. In practice this doesn't fix any issues because the only place this function is used uses the same value for the value and mask. Fixes: 3bace3591493 ("drm/amd/powerplay: add hardware manager sub-component") Signed-off-by: Fedor Pchelkin Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c index 79a566f3564a..c305ea4ec17d 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c @@ -149,7 +149,7 @@ int phm_wait_on_indirect_register(struct pp_hwmgr *hwmgr, } cgs_write_register(hwmgr->device, indirect_port, index); - return phm_wait_on_register(hwmgr, indirect_port + 1, mask, value); + return phm_wait_on_register(hwmgr, indirect_port + 1, value, mask); } int phm_wait_for_register_unequal(struct pp_hwmgr *hwmgr, From 9888f73679b72394bcee4f2af304456f03ceed8b Mon Sep 17 00:00:00 2001 From: Lijo Lazar Date: Tue, 1 Jul 2025 14:48:00 +0530 Subject: [PATCH 04/11] drm/amdgpu: Add a noverbose flag to psp_wait_for For extended wait with retries on a PSP register value, add a noverbose flag to avoid excessive error messages on each timeout. Signed-off-by: Lijo Lazar Reviewed-by: Asad Kamal Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 13 +++--- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 7 ++- drivers/gpu/drm/amd/amdgpu/psp_v10_0.c | 4 +- drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 35 ++++++++------- drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c | 10 ++--- drivers/gpu/drm/amd/amdgpu/psp_v12_0.c | 20 ++++----- drivers/gpu/drm/amd/amdgpu/psp_v13_0.c | 41 ++++++++++-------- drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c | 22 +++++----- drivers/gpu/drm/amd/amdgpu/psp_v14_0.c | 55 +++++++++++++----------- drivers/gpu/drm/amd/amdgpu/psp_v3_1.c | 41 +++++++++--------- 10 files changed, 130 insertions(+), 118 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index a4bede28df17..75911600d504 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -575,9 +575,11 @@ static int psp_sw_fini(struct amdgpu_ip_block *ip_block) return 0; } -int psp_wait_for(struct psp_context *psp, uint32_t reg_index, - uint32_t reg_val, uint32_t mask, bool check_changed) +int psp_wait_for(struct psp_context *psp, uint32_t reg_index, uint32_t reg_val, + uint32_t mask, uint32_t flags) { + bool check_changed = flags & PSP_WAITREG_CHANGED; + bool verbose = !(flags & PSP_WAITREG_NOVERBOSE); uint32_t val; int i; struct amdgpu_device *adev = psp->adev; @@ -597,9 +599,10 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index, udelay(1); } - dev_err(adev->dev, - "psp reg (0x%x) wait timed out, mask: %x, read: %x exp: %x", - reg_index, mask, val, reg_val); + if (verbose) + dev_err(adev->dev, + "psp reg (0x%x) wait timed out, mask: %x, read: %x exp: %x", + reg_index, mask, val, reg_val); return -ETIME; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 4bc0ec49d2e9..237b624aa51c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -134,6 +134,9 @@ enum psp_reg_prog_id { PSP_REG_LAST }; +#define PSP_WAITREG_CHANGED BIT(0) /* check if the value has changed */ +#define PSP_WAITREG_NOVERBOSE BIT(1) /* No error verbose */ + struct psp_funcs { int (*init_microcode)(struct psp_context *psp); int (*wait_for_bootloader)(struct psp_context *psp); @@ -532,8 +535,8 @@ extern const struct amdgpu_ip_block_version psp_v13_0_ip_block; extern const struct amdgpu_ip_block_version psp_v13_0_4_ip_block; extern const struct amdgpu_ip_block_version psp_v14_0_ip_block; -extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index, - uint32_t field_val, uint32_t mask, bool check_changed); +int psp_wait_for(struct psp_context *psp, uint32_t reg_index, + uint32_t field_val, uint32_t mask, uint32_t flags); extern int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index, uint32_t field_val, uint32_t mask, uint32_t msec_timeout); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index 2c4ebd98927f..3584b8c18fd9 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -94,7 +94,7 @@ static int psp_v10_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); return ret; } @@ -115,7 +115,7 @@ static int psp_v10_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 1a4a26e6ffd2..6cc05d36e359 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -152,11 +152,9 @@ static int psp_v11_0_wait_for_bootloader(struct psp_context *psp) for (retry_loop = 0; retry_loop < 10; retry_loop++) { /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, - 0x80000000, - false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) return 0; @@ -252,8 +250,8 @@ static int psp_v11_0_bootloader_load_sos(struct psp_context *psp) /* there might be handshake issue with hardware which needs delay */ mdelay(20); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), - 0, true); + RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); return ret; } @@ -279,11 +277,11 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, if (amdgpu_sriov_vf(adev)) ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); else ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); return ret; } @@ -321,13 +319,13 @@ static int psp_v11_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_101 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Wait for sOS ready for ring creation */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, 0); if (ret) { DRM_ERROR("Failed to wait for sOS ready for ring creation\n"); return ret; @@ -353,7 +351,7 @@ static int psp_v11_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -387,7 +385,7 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, - MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_MASK, 0); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -402,7 +400,7 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, - false); + 0); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); @@ -428,8 +426,9 @@ static int psp_v11_0_memory_training_send_msg(struct psp_context *psp, int msg) max_wait = MEM_TRAIN_SEND_MSG_TIMEOUT_US / adev->usec_timeout; for (i = 0; i < max_wait; i++) { - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) break; } @@ -608,7 +607,7 @@ static int psp_v11_0_load_usbc_pd_fw(struct psp_context *psp, uint64_t fw_pri_mc WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (fw_pri_mc_addr >> 20)); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -645,7 +644,7 @@ static int psp_v11_0_read_usbc_pd_fw(struct psp_context *psp, uint32_t *fw_ver) WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, C2PMSG_CMD_GFX_USB_PD_FW_VER); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (!ret) *fw_ver = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c index 338d015c0f2e..93787a90d598 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c @@ -43,7 +43,7 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, @@ -53,7 +53,7 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -91,13 +91,13 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_101 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Wait for sOS ready for ring creation */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, 0); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -123,7 +123,7 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c index d54b3e0fabaf..4c6450d62299 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -82,7 +82,7 @@ static int psp_v12_0_bootloader_load_sysdrv(struct psp_context *psp) /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -97,7 +97,7 @@ static int psp_v12_0_bootloader_load_sysdrv(struct psp_context *psp) psp_gfxdrv_command_reg); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); return ret; } @@ -118,7 +118,7 @@ static int psp_v12_0_bootloader_load_sos(struct psp_context *psp) /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -133,8 +133,8 @@ static int psp_v12_0_bootloader_load_sos(struct psp_context *psp) psp_gfxdrv_command_reg); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), - 0, true); + RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); return ret; } @@ -163,7 +163,7 @@ static int psp_v12_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); return ret; } @@ -186,11 +186,11 @@ static int psp_v12_0_ring_stop(struct psp_context *psp, if (amdgpu_sriov_vf(adev)) ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); else ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); return ret; } @@ -222,7 +222,7 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, - MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_MASK, 0); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -237,7 +237,7 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, - false); + 0); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index 58b6b64dcd68..af4a7d7c4abd 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -182,7 +182,7 @@ static int psp_v13_0_wait_for_vmbx_ready(struct psp_context *psp) ready having bit 31 of C2PMSG_33 set to 1 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_33), - 0x80000000, 0xffffffff, false); + 0x80000000, 0xffffffff, PSP_WAITREG_NOVERBOSE); if (ret == 0) break; @@ -213,7 +213,7 @@ static int psp_v13_0_wait_for_bootloader(struct psp_context *psp) for (retry_loop = 0; retry_loop < retry_cnt; retry_loop++) { ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, 0xffffffff, false); + 0x80000000, 0xffffffff, PSP_WAITREG_NOVERBOSE); if (ret == 0) return 0; @@ -362,8 +362,8 @@ static int psp_v13_0_bootloader_load_sos(struct psp_context *psp) /* there might be handshake issue with hardware which needs delay */ mdelay(20); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_81), - 0, true); + RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); if (!ret) psp_v13_0_init_sos_version(psp); @@ -386,7 +386,7 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -396,7 +396,7 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -434,13 +434,13 @@ static int psp_v13_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_101 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Wait for sOS ready for ring creation */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, 0); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -466,7 +466,7 @@ static int psp_v13_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -529,8 +529,9 @@ static int psp_v13_0_memory_training_send_msg(struct psp_context *psp, int msg) max_wait = MEM_TRAIN_SEND_MSG_TIMEOUT_US / adev->usec_timeout; for (i = 0; i < max_wait; i++) { - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) break; } @@ -682,7 +683,7 @@ static int psp_v13_0_load_usbc_pd_fw(struct psp_context *psp, uint64_t fw_pri_mc WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_36, (fw_pri_mc_addr >> 20)); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -719,7 +720,7 @@ static int psp_v13_0_read_usbc_pd_fw(struct psp_context *psp, uint32_t *fw_ver) WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_35, C2PMSG_CMD_GFX_USB_PD_FW_VER); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (!ret) *fw_ver = RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_36); @@ -744,8 +745,9 @@ static int psp_v13_0_exec_spi_cmd(struct psp_context *psp, int cmd) ret = psp_wait_for_spirom_update(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), MBOX_READY_FLAG, MBOX_READY_MASK, PSP_SPIROM_UPDATE_TIMEOUT); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, 0); if (ret) { dev_err(adev->dev, "SPI cmd %x timed out, ret = %d", cmd, ret); return ret; @@ -769,7 +771,7 @@ static int psp_v13_0_update_spirom(struct psp_context *psp, /* Confirm PSP is ready to start */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + MBOX_READY_FLAG, MBOX_READY_MASK, 0); if (ret) { dev_err(adev->dev, "PSP Not ready to start processing, ret = %d", ret); return ret; @@ -804,7 +806,7 @@ static int psp_v13_0_dump_spirom(struct psp_context *psp, /* Confirm PSP is ready to start */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + MBOX_READY_FLAG, MBOX_READY_MASK, 0); if (ret) { dev_err(adev->dev, "PSP Not ready to start processing, ret = %d", ret); return ret; @@ -931,8 +933,9 @@ static int psp_v13_0_reg_program_no_ring(struct psp_context *psp, uint32_t val, WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_102, id); WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_103, val); - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + 0x80000000, 0x80000000, 0); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c index f65af52c1c19..5f39a2edcc95 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c @@ -76,11 +76,9 @@ static int psp_v13_0_4_wait_for_bootloader(struct psp_context *psp) for (retry_loop = 0; retry_loop < 10; retry_loop++) { /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, - 0x80000000, - false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) return 0; @@ -185,8 +183,8 @@ static int psp_v13_0_4_bootloader_load_sos(struct psp_context *psp) /* there might be handshake issue with hardware which needs delay */ mdelay(20); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_81), - 0, true); + RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); return ret; } @@ -206,7 +204,7 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -216,7 +214,7 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -254,13 +252,13 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_101 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Wait for sOS ready for ring creation */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, 0); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -286,7 +284,7 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c index ffa47c7d24c9..36ef4a72ad1d 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c @@ -109,11 +109,9 @@ static int psp_v14_0_wait_for_bootloader(struct psp_context *psp) for (retry_loop = 0; retry_loop < 10; retry_loop++) { /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), - 0x80000000, - 0x80000000, - false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) return 0; @@ -228,9 +226,10 @@ static int psp_v14_0_bootloader_load_sos(struct psp_context *psp) /* there might be handshake issue with hardware which needs delay */ mdelay(20); - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_81), - 0, true); + ret = psp_wait_for(psp, + SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_81), + RREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); return ret; } @@ -250,7 +249,7 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_64, @@ -260,7 +259,7 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -298,13 +297,13 @@ static int psp_v14_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_101 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } else { /* Wait for sOS ready for ring creation */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, 0); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -330,7 +329,7 @@ static int psp_v14_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for( psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, 0); } return ret; @@ -393,8 +392,9 @@ static int psp_v14_0_memory_training_send_msg(struct psp_context *psp, int msg) max_wait = MEM_TRAIN_SEND_MSG_TIMEOUT_US / adev->usec_timeout; for (i = 0; i < max_wait; i++) { - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), + 0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE); if (ret == 0) break; } @@ -545,8 +545,9 @@ static int psp_v14_0_load_usbc_pd_fw(struct psp_context *psp, uint64_t fw_pri_mc */ WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_36, (fw_pri_mc_addr >> 20)); - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, + SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -582,8 +583,9 @@ static int psp_v14_0_read_usbc_pd_fw(struct psp_context *psp, uint32_t *fw_ver) WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_35, C2PMSG_CMD_GFX_USB_PD_FW_VER); - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, + SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_35), + 0x80000000, 0x80000000, 0); if (!ret) *fw_ver = RREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_36); @@ -607,11 +609,13 @@ static int psp_v14_0_exec_spi_cmd(struct psp_context *psp, int cmd) ret = psp_wait_for_spirom_update(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), MBOX_READY_FLAG, MBOX_READY_MASK, PSP_SPIROM_UPDATE_TIMEOUT); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, 0); - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + ret = psp_wait_for(psp, + SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, 0); if (ret) { dev_err(adev->dev, "SPI cmd %x timed out, ret = %d", cmd, ret); return ret; @@ -634,8 +638,9 @@ static int psp_v14_0_update_spirom(struct psp_context *psp, int ret; /* Confirm PSP is ready to start */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + ret = psp_wait_for(psp, + SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, 0); if (ret) { dev_err(adev->dev, "PSP Not ready to start processing, ret = %d", ret); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index f6b75e3e47ff..833830bc3e2e 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -91,7 +91,7 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp) /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -109,7 +109,7 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp) mdelay(20); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); return ret; } @@ -130,7 +130,7 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), - 0x80000000, 0x80000000, false); + 0x80000000, 0x80000000, 0); if (ret) return ret; @@ -147,8 +147,8 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) /* there might be handshake issue with hardware which needs delay */ mdelay(20); ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_81), - RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), - 0, true); + RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81), 0, + PSP_WAITREG_CHANGED); return ret; } @@ -168,7 +168,7 @@ static void psp_v3_1_reroute_ih(struct psp_context *psp) mdelay(20); psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + 0x80000000, 0x8000FFFF, 0); /* Change IH ring for UMC */ tmp = REG_SET_FIELD(0, IH_CLIENT_CFG_DATA, CREDIT_RETURN_ADDR, 0x1216b); @@ -180,7 +180,7 @@ static void psp_v3_1_reroute_ih(struct psp_context *psp) mdelay(20); psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + 0x80000000, 0x8000FFFF, 0); } static int psp_v3_1_ring_create(struct psp_context *psp, @@ -217,9 +217,9 @@ static int psp_v3_1_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, - mmMP0_SMN_C2PMSG_101), 0x80000000, - 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x8000FFFF, 0); } else { /* Write low address of the ring to C2PMSG_69 */ @@ -240,10 +240,9 @@ static int psp_v3_1_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, - mmMP0_SMN_C2PMSG_64), 0x80000000, - 0x8000FFFF, false); - + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x8000FFFF, 0); } return ret; } @@ -267,11 +266,13 @@ static int psp_v3_1_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ if (amdgpu_sriov_vf(adev)) - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x80000000, 0); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x80000000, 0); return ret; } @@ -311,7 +312,7 @@ static int psp_v3_1_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); - ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, 0); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -325,7 +326,7 @@ static int psp_v3_1_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, 0); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); From 91134e800894fc6992cb0cdadea4cc94fe21b6e2 Mon Sep 17 00:00:00 2001 From: Lijo Lazar Date: Mon, 30 Jun 2025 10:07:14 +0530 Subject: [PATCH 05/11] drm/amdkfd: Avoid queue reset if disabled If ring reset is disabled, skip resetting queues. Instead, fall back to device based reset. Signed-off-by: Lijo Lazar Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 76359c6a3f3a..500f51552038 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2339,9 +2339,18 @@ static int reset_hung_queues_sdma(struct device_queue_manager *dqm) static int reset_queues_on_hws_hang(struct device_queue_manager *dqm, bool is_sdma) { + struct amdgpu_device *adev = dqm->dev->adev; + while (halt_if_hws_hang) schedule(); + if (adev->debug_disable_gpu_ring_reset) { + dev_info_once(adev->dev, + "%s queue hung, but ring reset disabled", + is_sdma ? "sdma" : "compute"); + + return -EPERM; + } if (!amdgpu_gpu_recovery) return -ENOTRECOVERABLE; From f8410a17d377d062ef381316669653fb0ba0edc5 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 26 Jun 2025 08:49:07 -0400 Subject: [PATCH 06/11] drm/amdgpu/sdma: consolidate engine reset handling Move the force completion handling into the common engine reset function. No need to duplicate it for every IP version. Reviewed-by: Jesse Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 5 ++++- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 17 +---------------- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 6 ++---- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 6 ++---- 4 files changed, 9 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 7e26a44dcc1f..56939bb1d1a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -590,9 +590,12 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) * to be submitted to the queues after the reset is complete. */ if (!ret) { + amdgpu_fence_driver_force_completion(gfx_ring); drm_sched_wqueue_start(&gfx_ring->sched); - if (adev->sdma.has_page_queue) + if (adev->sdma.has_page_queue) { + amdgpu_fence_driver_force_completion(page_ring); drm_sched_wqueue_start(&page_ring->sched); + } } mutex_unlock(&sdma_instance->engine_reset_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index c05f3c1f50db..a7e1dbe03b29 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -1714,7 +1714,7 @@ static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring) static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u32 inst_mask, tmp_mask; + u32 inst_mask; int i, r; inst_mask = 1 << ring->me; @@ -1733,21 +1733,6 @@ static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) } r = sdma_v4_4_2_inst_start(adev, inst_mask, true); - if (r) - return r; - - tmp_mask = inst_mask; - for_each_inst(i, tmp_mask) { - ring = &adev->sdma.instance[i].ring; - - amdgpu_fence_driver_force_completion(ring); - - if (adev->sdma.has_page_queue) { - struct amdgpu_ring *page = &adev->sdma.instance[i].page; - - amdgpu_fence_driver_force_completion(page); - } - } return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 4d72b085b3dd..ed1706da7dee 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1618,10 +1618,8 @@ static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_0_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - if (r) - return r; - amdgpu_fence_driver_force_completion(ring); - return 0; + + return r; } static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index 42a25150f83a..b87a4b44fa93 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1534,10 +1534,8 @@ static int sdma_v5_2_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_2_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - if (r) - return r; - amdgpu_fence_driver_force_completion(ring); - return 0; + + return r; } static int sdma_v5_2_ring_preempt_ib(struct amdgpu_ring *ring) From 0c3c2e334c4fd00ed7a8ddb9c163a8c1138af1f1 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 26 Jun 2025 08:58:21 -0400 Subject: [PATCH 07/11] drm/amdgpu/sdma: allow caller to handle kernel rings in engine reset Add a parameter to amdgpu_sdma_reset_engine() to let the caller handle the kernel rings. This allows the kernel rings to back up their unprocessed state if the reset comes in via the drm scheduler rather than KFD. Reviewed-by: Jesse Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 44 +++++++++++-------- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 2 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 2 +- .../drm/amd/amdkfd/kfd_device_queue_manager.c | 2 +- 6 files changed, 32 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 56939bb1d1a9..8b8a04138711 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -545,10 +545,13 @@ static int amdgpu_sdma_soft_reset(struct amdgpu_device *adev, u32 instance_id) * amdgpu_sdma_reset_engine - Reset a specific SDMA engine * @adev: Pointer to the AMDGPU device * @instance_id: Logical ID of the SDMA engine instance to reset + * @caller_handles_kernel_queues: Skip kernel queue processing. Caller + * will handle it. * * Returns: 0 on success, or a negative error code on failure. */ -int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) +int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id, + bool caller_handles_kernel_queues) { int ret = 0; struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id]; @@ -556,14 +559,17 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) struct amdgpu_ring *page_ring = &sdma_instance->page; mutex_lock(&sdma_instance->engine_reset_mutex); - /* Stop the scheduler's work queue for the GFX and page rings if they are running. - * This ensures that no new tasks are submitted to the queues while - * the reset is in progress. - */ - drm_sched_wqueue_stop(&gfx_ring->sched); - if (adev->sdma.has_page_queue) - drm_sched_wqueue_stop(&page_ring->sched); + if (!caller_handles_kernel_queues) { + /* Stop the scheduler's work queue for the GFX and page rings if they are running. + * This ensures that no new tasks are submitted to the queues while + * the reset is in progress. + */ + drm_sched_wqueue_stop(&gfx_ring->sched); + + if (adev->sdma.has_page_queue) + drm_sched_wqueue_stop(&page_ring->sched); + } if (sdma_instance->funcs->stop_kernel_queue) { sdma_instance->funcs->stop_kernel_queue(gfx_ring); @@ -585,16 +591,18 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) } exit: - /* Restart the scheduler's work queue for the GFX and page rings - * if they were stopped by this function. This allows new tasks - * to be submitted to the queues after the reset is complete. - */ - if (!ret) { - amdgpu_fence_driver_force_completion(gfx_ring); - drm_sched_wqueue_start(&gfx_ring->sched); - if (adev->sdma.has_page_queue) { - amdgpu_fence_driver_force_completion(page_ring); - drm_sched_wqueue_start(&page_ring->sched); + if (!caller_handles_kernel_queues) { + /* Restart the scheduler's work queue for the GFX and page rings + * if they were stopped by this function. This allows new tasks + * to be submitted to the queues after the reset is complete. + */ + if (!ret) { + amdgpu_fence_driver_force_completion(gfx_ring); + drm_sched_wqueue_start(&gfx_ring->sched); + if (adev->sdma.has_page_queue) { + amdgpu_fence_driver_force_completion(page_ring); + drm_sched_wqueue_start(&page_ring->sched); + } } } mutex_unlock(&sdma_instance->engine_reset_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index e5f8951bbb6f..34311f32be4c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -172,7 +172,8 @@ struct amdgpu_buffer_funcs { uint32_t byte_count); }; -int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id); +int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id, + bool caller_handles_kernel_queues); #define amdgpu_emit_copy_buffer(adev, ib, s, d, b, t) (adev)->mman.buffer_funcs->emit_copy_buffer((ib), (s), (d), (b), (t)) #define amdgpu_emit_fill_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_fill_buffer((ib), (s), (d), (b)) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index a7e1dbe03b29..20fad2525969 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -1668,7 +1668,7 @@ static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, return -EOPNOTSUPP; amdgpu_amdkfd_suspend(adev, true); - r = amdgpu_sdma_reset_engine(adev, id); + r = amdgpu_sdma_reset_engine(adev, id, false); amdgpu_amdkfd_resume(adev, true); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index ed1706da7dee..5a1098bdd825 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1548,7 +1548,7 @@ static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, int r; amdgpu_amdkfd_suspend(adev, true); - r = amdgpu_sdma_reset_engine(adev, inst_id); + r = amdgpu_sdma_reset_engine(adev, inst_id, false); amdgpu_amdkfd_resume(adev, true); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index b87a4b44fa93..6843c2c3d71f 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1461,7 +1461,7 @@ static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, int r; amdgpu_amdkfd_suspend(adev, true); - r = amdgpu_sdma_reset_engine(adev, inst_id); + r = amdgpu_sdma_reset_engine(adev, inst_id, false); amdgpu_amdkfd_resume(adev, true); return r; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 500f51552038..2d91027e2a74 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2312,7 +2312,7 @@ static int reset_hung_queues_sdma(struct device_queue_manager *dqm) continue; /* Reset engine and check. */ - if (amdgpu_sdma_reset_engine(dqm->dev->adev, i) || + if (amdgpu_sdma_reset_engine(dqm->dev->adev, i, false) || dqm->dev->kfd2kgd->hqd_sdma_get_doorbell(dqm->dev->adev, i, j) || !set_sdma_queue_as_reset(dqm, doorbell_off)) { r = -ENOTRECOVERABLE; From 6d396e7ac1ce36079147933aa77ea977b03a8a9d Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 1 Jul 2025 16:09:19 -0400 Subject: [PATCH 08/11] drm/amd/display: Disable common modes for LVDS [Why] Common modes are added to LVDS for compatibility in clone mode, but not all panels support them. Non-native modes were disabled in the past but this caused problems because compositors didn't use scaling for non native modes. Now non-native modes on LVDS will enable the scaler by default. [How] Check the connector type. If the connector is LVDS avoid adding common modes. Reviewed-by: Harry Wentland Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 3dd4f9e9931d..7a4fca9a61bb 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -8374,7 +8374,8 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) drm_add_modes_noedid(connector, 1920, 1080); } else { amdgpu_dm_connector_ddc_get_modes(connector, drm_edid); - if (encoder && connector->connector_type != DRM_MODE_CONNECTOR_eDP) + if (encoder && (connector->connector_type != DRM_MODE_CONNECTOR_eDP) && + (connector->connector_type != DRM_MODE_CONNECTOR_LVDS)) amdgpu_dm_connector_add_common_modes(encoder, connector); amdgpu_dm_connector_add_freesync_modes(connector, drm_edid); } From df2719088b6a8084960ed6e2ddc7066ffc5b558d Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 1 Jul 2025 16:13:17 -0400 Subject: [PATCH 09/11] drm/amd/display: Use scaling for non-native resolutions on LVDS [Why] Common resolutions are added to supported modes to enable compatibility scenarios that compositors may use to do things like clone displays. There is no guarantee however that the panel will natively support these modes. [How] If the compositor hasn't enabled scaling but a non-native resolution has been picked for an LVDS panel turn the scaler on anyway. This will ensure compatibility. Reviewed-by: Harry Wentland Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 7a4fca9a61bb..096b23ad4845 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -7901,7 +7901,8 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, int clock, bpp = 0; bool is_y420 = false; - if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) { + if ((connector->connector_type == DRM_MODE_CONNECTOR_eDP) || + (connector->connector_type == DRM_MODE_CONNECTOR_LVDS)) { struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder); struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode; enum drm_mode_status result; From a73345b866ff8bbd93135af667c973a8fb4b2c40 Mon Sep 17 00:00:00 2001 From: Vitaly Prosyak Date: Tue, 24 Jun 2025 12:05:10 -0400 Subject: [PATCH 10/11] Revert "drm/amdgpu: fix slab-use-after-free in amdgpu_userq_mgr_fini" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 5fb90421fa0fbe0a968274912101fe917bf1c47b. The original patch moved `amdgpu_userq_mgr_fini()` to the driver's `postclose` callback, which is called after `drm_gem_release()` in the DRM file cleanup sequence.If a user application crashes or aborts without cleaning up its user queues, 'drm_gem_release()` may free GEM objects that are still referenced by active user queues, leading to use-after-free. By reverting, we ensure that user queues are disabled and cleaned up before any GEM objects are released, preventing this class of bug. However, this reintroduces a race during PCI hot-unplug, where device removal can race with per-file cleanup, leading to use-after-free in suspend/unplug paths. This will be fixed in the next patch. Fixes: 5fb90421fa0f ("drm/amdgpu: fix slab-use-after-free in amdgpu_userq_mgr_fini+0x70c") Signed-off-by: Vitaly Prosyak Acked-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 16 +++++++++++++++- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 3 --- 2 files changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 4f8632737574..1c54b2e5a225 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2911,6 +2911,20 @@ static int amdgpu_pmops_runtime_idle(struct device *dev) return ret; } +static int amdgpu_drm_release(struct inode *inode, struct file *filp) +{ + struct drm_file *file_priv = filp->private_data; + struct amdgpu_fpriv *fpriv = file_priv->driver_priv; + + if (fpriv) { + fpriv->evf_mgr.fd_closing = true; + amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); + amdgpu_userq_mgr_fini(&fpriv->userq_mgr); + } + + return drm_release(inode, filp); +} + long amdgpu_drm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -2962,7 +2976,7 @@ static const struct file_operations amdgpu_driver_kms_fops = { .owner = THIS_MODULE, .open = drm_open, .flush = amdgpu_flush, - .release = drm_release, + .release = amdgpu_drm_release, .unlocked_ioctl = amdgpu_drm_ioctl, .mmap = drm_gem_mmap, .poll = drm_poll, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index bfabb64e5535..b3c514b5e216 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -1501,9 +1501,6 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, amdgpu_vm_bo_del(adev, fpriv->prt_va); amdgpu_bo_unreserve(pd); } - fpriv->evf_mgr.fd_closing = true; - amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); - amdgpu_userq_mgr_fini(&fpriv->userq_mgr); amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr); amdgpu_vm_fini(adev, &fpriv->vm); From a886d26f2c8f9e3f3c1869ae368d09c75daac553 Mon Sep 17 00:00:00 2001 From: Vitaly Prosyak Date: Wed, 2 Jul 2025 08:35:30 -0400 Subject: [PATCH 11/11] drm/amdgpu: fix use-after-free in amdgpu_userq_suspend+0x51a/0x5a0 [ +0.000020] BUG: KASAN: slab-use-after-free in amdgpu_userq_suspend+0x51a/0x5a0 [amdgpu] [ +0.000817] Read of size 8 at addr ffff88812eec8c58 by task amd_pci_unplug/1733 [ +0.000027] CPU: 10 UID: 0 PID: 1733 Comm: amd_pci_unplug Tainted: G W 6.14.0+ #2 [ +0.000009] Tainted: [W]=WARN [ +0.000003] Hardware name: ASUS System Product Name/ROG STRIX B550-F GAMING (WI-FI), BIOS 1401 12/03/2020 [ +0.000004] Call Trace: [ +0.000004] [ +0.000003] dump_stack_lvl+0x76/0xa0 [ +0.000011] print_report+0xce/0x600 [ +0.000009] ? srso_return_thunk+0x5/0x5f [ +0.000006] ? kasan_complete_mode_report_info+0x76/0x200 [ +0.000007] ? kasan_addr_to_slab+0xd/0xb0 [ +0.000006] ? amdgpu_userq_suspend+0x51a/0x5a0 [amdgpu] [ +0.000707] kasan_report+0xbe/0x110 [ +0.000006] ? amdgpu_userq_suspend+0x51a/0x5a0 [amdgpu] [ +0.000541] __asan_report_load8_noabort+0x14/0x30 [ +0.000005] amdgpu_userq_suspend+0x51a/0x5a0 [amdgpu] [ +0.000535] ? stop_cpsch+0x396/0x600 [amdgpu] [ +0.000556] ? stop_cpsch+0x429/0x600 [amdgpu] [ +0.000536] ? __pfx_amdgpu_userq_suspend+0x10/0x10 [amdgpu] [ +0.000536] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? kgd2kfd_suspend+0x132/0x1d0 [amdgpu] [ +0.000542] amdgpu_device_fini_hw+0x581/0xe90 [amdgpu] [ +0.000485] ? down_write+0xbb/0x140 [ +0.000007] ? __mutex_unlock_slowpath.constprop.0+0x317/0x360 [ +0.000005] ? __pfx_amdgpu_device_fini_hw+0x10/0x10 [amdgpu] [ +0.000482] ? __kasan_check_write+0x14/0x30 [ +0.000004] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? up_write+0x55/0xb0 [ +0.000007] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? blocking_notifier_chain_unregister+0x6c/0xc0 [ +0.000008] amdgpu_driver_unload_kms+0x69/0x90 [amdgpu] [ +0.000484] amdgpu_pci_remove+0x93/0x130 [amdgpu] [ +0.000482] pci_device_remove+0xae/0x1e0 [ +0.000008] device_remove+0xc7/0x180 [ +0.000008] device_release_driver_internal+0x3d4/0x5a0 [ +0.000007] device_release_driver+0x12/0x20 [ +0.000004] pci_stop_bus_device+0x104/0x150 [ +0.000006] pci_stop_and_remove_bus_device_locked+0x1b/0x40 [ +0.000005] remove_store+0xd7/0xf0 [ +0.000005] ? __pfx_remove_store+0x10/0x10 [ +0.000006] ? __pfx__copy_from_iter+0x10/0x10 [ +0.000006] ? __pfx_dev_attr_store+0x10/0x10 [ +0.000006] dev_attr_store+0x3f/0x80 [ +0.000006] sysfs_kf_write+0x125/0x1d0 [ +0.000004] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? __kasan_check_write+0x14/0x30 [ +0.000005] kernfs_fop_write_iter+0x2ea/0x490 [ +0.000005] ? rw_verify_area+0x70/0x420 [ +0.000005] ? __pfx_kernfs_fop_write_iter+0x10/0x10 [ +0.000006] vfs_write+0x90d/0xe70 [ +0.000005] ? srso_return_thunk+0x5/0x5f [ +0.000005] ? __pfx_vfs_write+0x10/0x10 [ +0.000004] ? local_clock+0x15/0x30 [ +0.000008] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? __kasan_slab_free+0x5f/0x80 [ +0.000005] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? __kasan_check_read+0x11/0x20 [ +0.000004] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? fdget_pos+0x1d3/0x500 [ +0.000007] ksys_write+0x119/0x220 [ +0.000005] ? putname+0x1c/0x30 [ +0.000006] ? __pfx_ksys_write+0x10/0x10 [ +0.000007] __x64_sys_write+0x72/0xc0 [ +0.000006] x64_sys_call+0x18ab/0x26f0 [ +0.000006] do_syscall_64+0x7c/0x170 [ +0.000004] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? __pfx___x64_sys_openat+0x10/0x10 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? __kasan_check_read+0x11/0x20 [ +0.000003] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? fpregs_assert_state_consistent+0x21/0xb0 [ +0.000006] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? syscall_exit_to_user_mode+0x4e/0x240 [ +0.000005] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? do_syscall_64+0x88/0x170 [ +0.000003] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? irqentry_exit+0x43/0x50 [ +0.000004] ? srso_return_thunk+0x5/0x5f [ +0.000004] ? exc_page_fault+0x7c/0x110 [ +0.000006] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000006] RIP: 0033:0x7480c0b14887 [ +0.000005] Code: 10 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24 [ +0.000005] RSP: 002b:00007fff142b0058 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 [ +0.000006] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007480c0b14887 [ +0.000003] RDX: 0000000000000001 RSI: 00007480c0e7365a RDI: 0000000000000004 [ +0.000003] RBP: 00007fff142b0080 R08: 0000563b2e73c170 R09: 0000000000000000 [ +0.000003] R10: 0000000000000000 R11: 0000000000000246 R12: 00007fff142b02f8 [ +0.000003] R13: 0000563b159a72a9 R14: 0000563b159a9d48 R15: 00007480c0f19040 [ +0.000008] [ +0.000445] Allocated by task 427 on cpu 5 at 29.342331s: [ +0.000011] kasan_save_stack+0x28/0x60 [ +0.000006] kasan_save_track+0x18/0x70 [ +0.000006] kasan_save_alloc_info+0x38/0x60 [ +0.000005] __kasan_kmalloc+0xc1/0xd0 [ +0.000006] __kmalloc_cache_noprof+0x1bd/0x430 [ +0.000007] amdgpu_driver_open_kms+0x172/0x760 [amdgpu] [ +0.000493] drm_file_alloc+0x569/0x9a0 [ +0.000007] drm_client_init+0x1b7/0x410 [ +0.000007] drm_fbdev_client_setup+0x174/0x470 [ +0.000006] drm_client_setup+0x8a/0xf0 [ +0.000006] amdgpu_pci_probe+0x510/0x10c0 [amdgpu] [ +0.000483] local_pci_probe+0xe7/0x1b0 [ +0.000006] pci_device_probe+0x5bf/0x890 [ +0.000006] really_probe+0x1fd/0x950 [ +0.000005] __driver_probe_device+0x307/0x410 [ +0.000006] driver_probe_device+0x4e/0x150 [ +0.000005] __driver_attach+0x223/0x510 [ +0.000006] bus_for_each_dev+0x102/0x1a0 [ +0.000005] driver_attach+0x3d/0x60 [ +0.000006] bus_add_driver+0x309/0x650 [ +0.000005] driver_register+0x13d/0x490 [ +0.000006] __pci_register_driver+0x1ee/0x2b0 [ +0.000006] rfcomm_dlc_clear_state+0x69/0x220 [rfcomm] [ +0.000011] do_one_initcall+0x9c/0x3e0 [ +0.000007] do_init_module+0x29e/0x7f0 [ +0.000006] load_module+0x5c75/0x7c80 [ +0.000006] init_module_from_file+0x106/0x180 [ +0.000006] idempotent_init_module+0x377/0x740 [ +0.000006] __x64_sys_finit_module+0xd7/0x180 [ +0.000006] x64_sys_call+0x1f0b/0x26f0 [ +0.000006] do_syscall_64+0x7c/0x170 [ +0.000005] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000013] Freed by task 1733 on cpu 5 at 59.907086s: [ +0.000011] kasan_save_stack+0x28/0x60 [ +0.000006] kasan_save_track+0x18/0x70 [ +0.000005] kasan_save_free_info+0x3b/0x60 [ +0.000005] __kasan_slab_free+0x54/0x80 [ +0.000006] kfree+0x127/0x470 [ +0.000006] amdgpu_driver_postclose_kms+0x455/0x760 [amdgpu] [ +0.000493] drm_file_free.part.0+0x5b1/0xba0 [ +0.000006] drm_file_free+0x13/0x30 [ +0.000006] drm_client_release+0x1c4/0x2b0 [ +0.000006] drm_fbdev_ttm_fb_destroy+0xd2/0x120 [drm_ttm_helper] [ +0.000007] put_fb_info+0x97/0xe0 [ +0.000007] unregister_framebuffer+0x197/0x380 [ +0.000005] drm_fb_helper_unregister_info+0x94/0x100 [ +0.000005] drm_fbdev_client_unregister+0x3c/0x80 [ +0.000007] drm_client_dev_unregister+0x144/0x330 [ +0.000006] drm_dev_unregister+0x49/0x1b0 [ +0.000006] drm_dev_unplug+0x4c/0xd0 [ +0.000006] amdgpu_pci_remove+0x58/0x130 [amdgpu] [ +0.000484] pci_device_remove+0xae/0x1e0 [ +0.000008] device_remove+0xc7/0x180 [ +0.000007] device_release_driver_internal+0x3d4/0x5a0 [ +0.000006] device_release_driver+0x12/0x20 [ +0.000007] pci_stop_bus_device+0x104/0x150 [ +0.000006] pci_stop_and_remove_bus_device_locked+0x1b/0x40 [ +0.000006] remove_store+0xd7/0xf0 [ +0.000006] dev_attr_store+0x3f/0x80 [ +0.000005] sysfs_kf_write+0x125/0x1d0 [ +0.000006] kernfs_fop_write_iter+0x2ea/0x490 [ +0.000006] vfs_write+0x90d/0xe70 [ +0.000006] ksys_write+0x119/0x220 [ +0.000006] __x64_sys_write+0x72/0xc0 [ +0.000006] x64_sys_call+0x18ab/0x26f0 [ +0.000005] do_syscall_64+0x7c/0x170 [ +0.000006] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ +0.000012] The buggy address belongs to the object at ffff88812eec8000 which belongs to the cache kmalloc-rnd-07-4k of size 4096 [ +0.000016] The buggy address is located 3160 bytes inside of freed 4096-byte region [ffff88812eec8000, ffff88812eec9000) [ +0.000023] The buggy address belongs to the physical page: [ +0.000009] page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x12eec8 [ +0.000007] head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0 [ +0.000005] flags: 0x17ffffc0000040(head|node=0|zone=2|lastcpupid=0x1fffff) [ +0.000007] page_type: f5(slab) [ +0.000008] raw: 0017ffffc0000040 ffff888100054500 dead000000000122 0000000000000000 [ +0.000005] raw: 0000000000000000 0000000080040004 00000000f5000000 0000000000000000 [ +0.000006] head: 0017ffffc0000040 ffff888100054500 dead000000000122 0000000000000000 [ +0.000005] head: 0000000000000000 0000000080040004 00000000f5000000 0000000000000000 [ +0.000006] head: 0017ffffc0000003 ffffea0004bbb201 ffffffffffffffff 0000000000000000 [ +0.000005] head: 0000000000000008 0000000000000000 00000000ffffffff 0000000000000000 [ +0.000005] page dumped because: kasan: bad access detected [ +0.000010] Memory state around the buggy address: [ +0.000009] ffff88812eec8b00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ +0.000012] ffff88812eec8b80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ +0.000011] >ffff88812eec8c00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ +0.000011] ^ [ +0.000010] ffff88812eec8c80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ +0.000011] ffff88812eec8d00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ +0.000011] ================================================================== The use-after-free occurs because a delayed work item (`suspend_work`) may still be pending or running when resources it accesses are freed during device removal or file close. The previous code used `flush_work(&fpriv->evf_mgr.suspend_work.work)`, which does not wait for delayed work that has not yet started. As a result, the delayed work could run after its memory was freed, causing a use-after-free. By switching to `flush_delayed_work(&fpriv->evf_mgr.suspend_work)`, we ensure that the kernel waits for both queued and delayed work to finish before freeing memory, closing this race. Fixes: adba0929736a ("drm/amdgpu: Fix Illegal opcode in command stream Error") Signed-off-by: Vitaly Prosyak Acked-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 295e7186e156..aac0de86f3e8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -664,7 +664,7 @@ static void amdgpu_userq_restore_worker(struct work_struct *work) struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr); int ret; - flush_work(&fpriv->evf_mgr.suspend_work.work); + flush_delayed_work(&fpriv->evf_mgr.suspend_work); mutex_lock(&uq_mgr->userq_mutex);