diff options
Diffstat (limited to 'drivers/accel/ivpu')
| -rw-r--r-- | drivers/accel/ivpu/Kconfig | 9 | ||||
| -rw-r--r-- | drivers/accel/ivpu/Makefile | 2 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_drv.c | 31 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_drv.h | 16 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_fw.c | 8 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_hw.c | 10 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_hw_btrs.c | 12 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_job.c | 148 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_job.h | 2 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_jsm_msg.c | 8 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_mmu.c | 101 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_mmu.h | 4 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_mmu_context.c | 158 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_mmu_context.h | 9 | ||||
| -rw-r--r-- | drivers/accel/ivpu/ivpu_pm.c | 2 |
15 files changed, 241 insertions, 279 deletions
diff --git a/drivers/accel/ivpu/Kconfig b/drivers/accel/ivpu/Kconfig index e4d418b44626..9e055b5ce03d 100644 --- a/drivers/accel/ivpu/Kconfig +++ b/drivers/accel/ivpu/Kconfig @@ -16,3 +16,12 @@ config DRM_ACCEL_IVPU and Deep Learning applications. If "M" is selected, the module will be called intel_vpu. + +config DRM_ACCEL_IVPU_DEBUG + bool "Intel NPU debug mode" + depends on DRM_ACCEL_IVPU + help + Choose this option to enable additional + debug features for the Intel NPU driver: + - Always print debug messages regardless of dyndbg config, + - Enable unsafe module params. diff --git a/drivers/accel/ivpu/Makefile b/drivers/accel/ivpu/Makefile index e73937c86d9a..1029e0bab061 100644 --- a/drivers/accel/ivpu/Makefile +++ b/drivers/accel/ivpu/Makefile @@ -24,4 +24,6 @@ intel_vpu-$(CONFIG_DEV_COREDUMP) += ivpu_coredump.o obj-$(CONFIG_DRM_ACCEL_IVPU) += intel_vpu.o +subdir-ccflags-$(CONFIG_DRM_ACCEL_IVPU_DEBUG) += -DDEBUG + CFLAGS_ivpu_trace_points.o = -I$(src) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index e7d8967c02f2..ca2bf47ce248 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -43,8 +43,10 @@ module_param_named(dbg_mask, ivpu_dbg_mask, int, 0644); MODULE_PARM_DESC(dbg_mask, "Driver debug mask. See IVPU_DBG_* macros."); int ivpu_test_mode; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(test_mode, ivpu_test_mode, int, 0644); MODULE_PARM_DESC(test_mode, "Test mode mask. See IVPU_TEST_MODE_* macros."); +#endif u8 ivpu_pll_min_ratio; module_param_named(pll_min_ratio, ivpu_pll_min_ratio, byte, 0644); @@ -86,7 +88,7 @@ static void file_priv_unbind(struct ivpu_device *vdev, struct ivpu_file_priv *fi ivpu_cmdq_release_all_locked(file_priv); ivpu_bo_unbind_all_bos_from_context(vdev, &file_priv->ctx); - ivpu_mmu_user_context_fini(vdev, &file_priv->ctx); + ivpu_mmu_context_fini(vdev, &file_priv->ctx); file_priv->bound = false; drm_WARN_ON(&vdev->drm, !xa_erase_irq(&vdev->context_xa, file_priv->ctx.id)); } @@ -104,6 +106,8 @@ static void file_priv_release(struct kref *ref) pm_runtime_get_sync(vdev->drm.dev); mutex_lock(&vdev->context_list_lock); file_priv_unbind(vdev, file_priv); + drm_WARN_ON(&vdev->drm, !xa_empty(&file_priv->cmdq_xa)); + xa_destroy(&file_priv->cmdq_xa); mutex_unlock(&vdev->context_list_lock); pm_runtime_put_autosuspend(vdev->drm.dev); @@ -254,14 +258,14 @@ static int ivpu_open(struct drm_device *dev, struct drm_file *file) goto err_unlock; } - ret = ivpu_mmu_user_context_init(vdev, &file_priv->ctx, ctx_id); - if (ret) - goto err_xa_erase; + ivpu_mmu_context_init(vdev, &file_priv->ctx, ctx_id); + + file_priv->job_limit.min = FIELD_PREP(IVPU_JOB_ID_CONTEXT_MASK, (file_priv->ctx.id - 1)); + file_priv->job_limit.max = file_priv->job_limit.min | IVPU_JOB_ID_JOB_MASK; - file_priv->default_job_limit.min = FIELD_PREP(IVPU_JOB_ID_CONTEXT_MASK, - (file_priv->ctx.id - 1)); - file_priv->default_job_limit.max = file_priv->default_job_limit.min | IVPU_JOB_ID_JOB_MASK; - file_priv->job_limit = file_priv->default_job_limit; + xa_init_flags(&file_priv->cmdq_xa, XA_FLAGS_ALLOC1); + file_priv->cmdq_limit.min = IVPU_CMDQ_MIN_ID; + file_priv->cmdq_limit.max = IVPU_CMDQ_MAX_ID; mutex_unlock(&vdev->context_list_lock); drm_dev_exit(idx); @@ -273,8 +277,6 @@ static int ivpu_open(struct drm_device *dev, struct drm_file *file) return 0; -err_xa_erase: - xa_erase_irq(&vdev->context_xa, ctx_id); err_unlock: mutex_unlock(&vdev->context_list_lock); mutex_destroy(&file_priv->ms_lock); @@ -622,9 +624,8 @@ static int ivpu_dev_init(struct ivpu_device *vdev) lockdep_set_class(&vdev->submitted_jobs_xa.xa_lock, &submitted_jobs_xa_lock_class_key); INIT_LIST_HEAD(&vdev->bo_list); - vdev->default_db_limit.min = IVPU_MIN_DB; - vdev->default_db_limit.max = IVPU_MAX_DB; - vdev->db_limit = vdev->default_db_limit; + vdev->db_limit.min = IVPU_MIN_DB; + vdev->db_limit.max = IVPU_MAX_DB; ret = drmm_mutex_init(&vdev->drm, &vdev->context_list_lock); if (ret) @@ -652,9 +653,7 @@ static int ivpu_dev_init(struct ivpu_device *vdev) if (ret) goto err_shutdown; - ret = ivpu_mmu_global_context_init(vdev); - if (ret) - goto err_shutdown; + ivpu_mmu_global_context_init(vdev); ret = ivpu_mmu_init(vdev); if (ret) diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index f905021ac174..3fdff3f6cffd 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -49,11 +49,11 @@ #define IVPU_JOB_ID_JOB_MASK GENMASK(7, 0) #define IVPU_JOB_ID_CONTEXT_MASK GENMASK(31, 8) -#define IVPU_NUM_ENGINES 2 #define IVPU_NUM_PRIORITIES 4 -#define IVPU_NUM_CMDQS_PER_CTX (IVPU_NUM_ENGINES * IVPU_NUM_PRIORITIES) +#define IVPU_NUM_CMDQS_PER_CTX (IVPU_NUM_PRIORITIES) -#define IVPU_CMDQ_INDEX(engine, priority) ((engine) * IVPU_NUM_PRIORITIES + (priority)) +#define IVPU_CMDQ_MIN_ID 1 +#define IVPU_CMDQ_MAX_ID 255 #define IVPU_PLATFORM_SILICON 0 #define IVPU_PLATFORM_SIMICS 2 @@ -140,7 +140,7 @@ struct ivpu_device { struct xarray db_xa; struct xa_limit db_limit; - struct xa_limit default_db_limit; + u32 db_next; struct mutex bo_list_lock; /* Protects bo_list */ struct list_head bo_list; @@ -171,13 +171,15 @@ struct ivpu_file_priv { struct kref ref; struct ivpu_device *vdev; struct mutex lock; /* Protects cmdq */ - struct ivpu_cmdq *cmdq[IVPU_NUM_CMDQS_PER_CTX]; + struct xarray cmdq_xa; struct ivpu_mmu_context ctx; struct mutex ms_lock; /* Protects ms_instance_list, ms_info_bo */ struct list_head ms_instance_list; struct ivpu_bo *ms_info_bo; struct xa_limit job_limit; - struct xa_limit default_job_limit; + u32 job_id_next; + struct xa_limit cmdq_limit; + u32 cmdq_id_next; bool has_mmu_faults; bool bound; bool aborted; @@ -195,7 +197,7 @@ extern bool ivpu_force_snoop; #define IVPU_TEST_MODE_NULL_SUBMISSION BIT(2) #define IVPU_TEST_MODE_D0I3_MSG_DISABLE BIT(4) #define IVPU_TEST_MODE_D0I3_MSG_ENABLE BIT(5) -#define IVPU_TEST_MODE_PREEMPTION_DISABLE BIT(6) +#define IVPU_TEST_MODE_MIP_DISABLE BIT(6) #define IVPU_TEST_MODE_DISABLE_TIMEOUTS BIT(8) #define IVPU_TEST_MODE_TURBO BIT(9) extern int ivpu_test_mode; diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index be367465e7df..6037ec0b3096 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -46,8 +46,10 @@ #define IVPU_FOCUS_PRESENT_TIMER_MS 1000 static char *ivpu_firmware; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(firmware, ivpu_firmware, charp, 0644); MODULE_PARM_DESC(firmware, "NPU firmware binary in /lib/firmware/.."); +#endif static struct { int gen; @@ -582,8 +584,10 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params boot_params->ipc_payload_area_start = ipc_mem_rx->vpu_addr + ivpu_bo_size(ipc_mem_rx) / 2; boot_params->ipc_payload_area_size = ivpu_bo_size(ipc_mem_rx) / 2; - boot_params->global_aliased_pio_base = vdev->hw->ranges.user.start; - boot_params->global_aliased_pio_size = ivpu_hw_range_size(&vdev->hw->ranges.user); + if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { + boot_params->global_aliased_pio_base = vdev->hw->ranges.user.start; + boot_params->global_aliased_pio_size = ivpu_hw_range_size(&vdev->hw->ranges.user); + } /* Allow configuration for L2C_PAGE_TABLE with boot param value */ boot_params->autoconfig = 1; diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 1c259d717815..09ada8b500b9 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -114,14 +114,14 @@ static void memory_ranges_init(struct ivpu_device *vdev) { if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { ivpu_hw_range_init(&vdev->hw->ranges.global, 0x80000000, SZ_512M); - ivpu_hw_range_init(&vdev->hw->ranges.user, 0xc0000000, 255 * SZ_1M); + ivpu_hw_range_init(&vdev->hw->ranges.user, 0x88000000, 511 * SZ_1M); ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x180000000, SZ_2G); - ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_8G); + ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_128G); } else { ivpu_hw_range_init(&vdev->hw->ranges.global, 0x80000000, SZ_512M); - ivpu_hw_range_init(&vdev->hw->ranges.user, 0x80000000, SZ_256M); - ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x80000000 + SZ_256M, SZ_2G - SZ_256M); - ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_8G); + ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x80000000, SZ_2G); + ivpu_hw_range_init(&vdev->hw->ranges.user, 0x100000000, SZ_256G); + vdev->hw->ranges.dma = vdev->hw->ranges.user; } } diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c index 6d5f1cc71143..3212c99f3682 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.c +++ b/drivers/accel/ivpu/ivpu_hw_btrs.c @@ -141,16 +141,10 @@ static int read_tile_config_fuse(struct ivpu_device *vdev, u32 *tile_fuse_config } config = REG_GET_FLD(VPU_HW_BTRS_LNL_TILE_FUSE, CONFIG, fuse); - if (!tile_disable_check(config)) { - ivpu_err(vdev, "Fuse: Invalid tile disable config (0x%x)\n", config); - return -EIO; - } + if (!tile_disable_check(config)) + ivpu_warn(vdev, "More than 1 tile disabled, tile fuse config mask: 0x%x\n", config); - if (config) - ivpu_dbg(vdev, MISC, "Fuse: %d tiles enabled. Tile number %d disabled\n", - BTRS_LNL_TILE_MAX_NUM - 1, ffs(config) - 1); - else - ivpu_dbg(vdev, MISC, "Fuse: All %d tiles enabled\n", BTRS_LNL_TILE_MAX_NUM); + ivpu_dbg(vdev, MISC, "Tile disable config mask: 0x%x\n", config); *tile_fuse_config = config; return 0; diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 98e0b7b61407..7149312f16e1 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -35,7 +35,8 @@ static int ivpu_preemption_buffers_create(struct ivpu_device *vdev, u64 primary_size = ALIGN(vdev->fw->primary_preempt_buf_size, PAGE_SIZE); u64 secondary_size = ALIGN(vdev->fw->secondary_preempt_buf_size, PAGE_SIZE); - if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) + if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW || + ivpu_test_mode & IVPU_TEST_MODE_MIP_DISABLE) return 0; cmdq->primary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &vdev->hw->ranges.user, @@ -45,7 +46,7 @@ static int ivpu_preemption_buffers_create(struct ivpu_device *vdev, return -ENOMEM; } - cmdq->secondary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &vdev->hw->ranges.shave, + cmdq->secondary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &vdev->hw->ranges.dma, secondary_size, DRM_IVPU_BO_WC); if (!cmdq->secondary_preempt_buf) { ivpu_err(vdev, "Failed to create secondary preemption buffer\n"); @@ -72,26 +73,6 @@ static void ivpu_preemption_buffers_free(struct ivpu_device *vdev, ivpu_bo_free(cmdq->secondary_preempt_buf); } -static int ivpu_id_alloc(struct xarray *xa, u32 *id, void *entry, struct xa_limit *limit, - const struct xa_limit default_limit) -{ - int ret; - - ret = __xa_alloc(xa, id, entry, *limit, GFP_KERNEL); - if (ret) { - limit->min = default_limit.min; - ret = __xa_alloc(xa, id, entry, *limit, GFP_KERNEL); - if (ret) - return ret; - } - - limit->min = *id + 1; - if (limit->min > limit->max) - limit->min = default_limit.min; - - return ret; -} - static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) { struct ivpu_device *vdev = file_priv->vdev; @@ -102,18 +83,23 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) if (!cmdq) return NULL; - xa_lock(&vdev->db_xa); /* lock here to protect db_limit */ - ret = ivpu_id_alloc(&vdev->db_xa, &cmdq->db_id, NULL, &vdev->db_limit, - vdev->default_db_limit); - xa_unlock(&vdev->db_xa); - if (ret) { + ret = xa_alloc_cyclic(&vdev->db_xa, &cmdq->db_id, NULL, vdev->db_limit, &vdev->db_next, + GFP_KERNEL); + if (ret < 0) { ivpu_err(vdev, "Failed to allocate doorbell id: %d\n", ret); goto err_free_cmdq; } + ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, + &file_priv->cmdq_id_next, GFP_KERNEL); + if (ret < 0) { + ivpu_err(vdev, "Failed to allocate command queue id: %d\n", ret); + goto err_erase_db_xa; + } + cmdq->mem = ivpu_bo_create_global(vdev, SZ_4K, DRM_IVPU_BO_WC | DRM_IVPU_BO_MAPPABLE); if (!cmdq->mem) - goto err_erase_xa; + goto err_erase_cmdq_xa; ret = ivpu_preemption_buffers_create(vdev, file_priv, cmdq); if (ret) @@ -121,7 +107,9 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) return cmdq; -err_erase_xa: +err_erase_cmdq_xa: + xa_erase(&file_priv->cmdq_xa, cmdq->id); +err_erase_db_xa: xa_erase(&vdev->db_xa, cmdq->db_id); err_free_cmdq: kfree(cmdq); @@ -145,13 +133,13 @@ static int ivpu_hws_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq struct ivpu_device *vdev = file_priv->vdev; int ret; - ret = ivpu_jsm_hws_create_cmdq(vdev, file_priv->ctx.id, file_priv->ctx.id, cmdq->db_id, + ret = ivpu_jsm_hws_create_cmdq(vdev, file_priv->ctx.id, file_priv->ctx.id, cmdq->id, task_pid_nr(current), engine, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); if (ret) return ret; - ret = ivpu_jsm_hws_set_context_sched_properties(vdev, file_priv->ctx.id, cmdq->db_id, + ret = ivpu_jsm_hws_set_context_sched_properties(vdev, file_priv->ctx.id, cmdq->id, priority); if (ret) return ret; @@ -165,20 +153,21 @@ static int ivpu_register_db(struct ivpu_file_priv *file_priv, struct ivpu_cmdq * int ret; if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) - ret = ivpu_jsm_hws_register_db(vdev, file_priv->ctx.id, cmdq->db_id, cmdq->db_id, + ret = ivpu_jsm_hws_register_db(vdev, file_priv->ctx.id, cmdq->id, cmdq->db_id, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); else ret = ivpu_jsm_register_db(vdev, file_priv->ctx.id, cmdq->db_id, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); if (!ret) - ivpu_dbg(vdev, JOB, "DB %d registered to ctx %d\n", cmdq->db_id, file_priv->ctx.id); + ivpu_dbg(vdev, JOB, "DB %d registered to cmdq %d ctx %d\n", + cmdq->db_id, cmdq->id, file_priv->ctx.id); return ret; } static int -ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 engine, u8 priority) +ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u8 priority) { struct ivpu_device *vdev = file_priv->vdev; struct vpu_job_queue_header *jobq_header; @@ -194,7 +183,7 @@ ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 eng cmdq->jobq = (struct vpu_job_queue *)ivpu_bo_vaddr(cmdq->mem); jobq_header = &cmdq->jobq->header; - jobq_header->engine_idx = engine; + jobq_header->engine_idx = VPU_ENGINE_COMPUTE; jobq_header->head = 0; jobq_header->tail = 0; if (ivpu_test_mode & IVPU_TEST_MODE_TURBO) { @@ -205,7 +194,7 @@ ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 eng wmb(); /* Flush WC buffer for jobq->header */ if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { - ret = ivpu_hws_cmdq_init(file_priv, cmdq, engine, priority); + ret = ivpu_hws_cmdq_init(file_priv, cmdq, VPU_ENGINE_COMPUTE, priority); if (ret) return ret; } @@ -232,9 +221,9 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm cmdq->db_registered = false; if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { - ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->db_id); + ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->id); if (!ret) - ivpu_dbg(vdev, JOB, "Command queue %d destroyed\n", cmdq->db_id); + ivpu_dbg(vdev, JOB, "Command queue %d destroyed\n", cmdq->id); } ret = ivpu_jsm_unregister_db(vdev, cmdq->db_id); @@ -244,55 +233,46 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm return 0; } -static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u16 engine, - u8 priority) +static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u8 priority) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; int ret; lockdep_assert_held(&file_priv->lock); + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + if (cmdq->priority == priority) + break; + if (!cmdq) { cmdq = ivpu_cmdq_alloc(file_priv); if (!cmdq) return NULL; - file_priv->cmdq[cmdq_idx] = cmdq; + cmdq->priority = priority; } - ret = ivpu_cmdq_init(file_priv, cmdq, engine, priority); + ret = ivpu_cmdq_init(file_priv, cmdq, priority); if (ret) return NULL; return cmdq; } -static void ivpu_cmdq_release_locked(struct ivpu_file_priv *file_priv, u16 engine, u8 priority) +void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; lockdep_assert_held(&file_priv->lock); - if (cmdq) { - file_priv->cmdq[cmdq_idx] = NULL; + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) { + xa_erase(&file_priv->cmdq_xa, cmdq_id); ivpu_cmdq_fini(file_priv, cmdq); ivpu_cmdq_free(file_priv, cmdq); } } -void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) -{ - u16 engine; - u8 priority; - - lockdep_assert_held(&file_priv->lock); - - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) - ivpu_cmdq_release_locked(file_priv, engine, priority); -} - /* * Mark the doorbell as unregistered * This function needs to be called when the VPU hardware is restarted @@ -301,20 +281,13 @@ void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) */ static void ivpu_cmdq_reset(struct ivpu_file_priv *file_priv) { - u16 engine; - u8 priority; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; mutex_lock(&file_priv->lock); - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) { - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; - - if (cmdq) - cmdq->db_registered = false; - } - } + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + cmdq->db_registered = false; mutex_unlock(&file_priv->lock); } @@ -334,17 +307,11 @@ void ivpu_cmdq_reset_all_contexts(struct ivpu_device *vdev) static void ivpu_cmdq_fini_all(struct ivpu_file_priv *file_priv) { - u16 engine; - u8 priority; - - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) { - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; - if (file_priv->cmdq[cmdq_idx]) - ivpu_cmdq_fini(file_priv, file_priv->cmdq[cmdq_idx]); - } - } + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + ivpu_cmdq_fini(file_priv, cmdq); } void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) @@ -369,8 +336,8 @@ static int ivpu_cmdq_push_job(struct ivpu_cmdq *cmdq, struct ivpu_job *job) /* Check if there is space left in job queue */ if (next_entry == header->head) { - ivpu_dbg(vdev, JOB, "Job queue full: ctx %d engine %d db %d head %d tail %d\n", - job->file_priv->ctx.id, job->engine_idx, cmdq->db_id, header->head, tail); + ivpu_dbg(vdev, JOB, "Job queue full: ctx %d cmdq %d db %d head %d tail %d\n", + job->file_priv->ctx.id, cmdq->id, cmdq->db_id, header->head, tail); return -EBUSY; } @@ -381,8 +348,7 @@ static int ivpu_cmdq_push_job(struct ivpu_cmdq *cmdq, struct ivpu_job *job) if (unlikely(ivpu_test_mode & IVPU_TEST_MODE_NULL_SUBMISSION)) entry->flags = VPU_JOB_FLAGS_NULL_SUBMISSION_MASK; - if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW && - (unlikely(!(ivpu_test_mode & IVPU_TEST_MODE_PREEMPTION_DISABLE)))) { + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { if (cmdq->primary_preempt_buf) { entry->primary_preempt_buf_addr = cmdq->primary_preempt_buf->vpu_addr; entry->primary_preempt_buf_size = ivpu_bo_size(cmdq->primary_preempt_buf); @@ -557,7 +523,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) mutex_lock(&file_priv->lock); - cmdq = ivpu_cmdq_acquire(file_priv, job->engine_idx, priority); + cmdq = ivpu_cmdq_acquire(file_priv, priority); if (!cmdq) { ivpu_warn_ratelimited(vdev, "Failed to get job queue, ctx %d engine %d prio %d\n", file_priv->ctx.id, job->engine_idx, priority); @@ -567,9 +533,9 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) xa_lock(&vdev->submitted_jobs_xa); is_first_job = xa_empty(&vdev->submitted_jobs_xa); - ret = ivpu_id_alloc(&vdev->submitted_jobs_xa, &job->job_id, job, &file_priv->job_limit, - file_priv->default_job_limit); - if (ret) { + ret = __xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, + &file_priv->job_id_next, GFP_KERNEL); + if (ret < 0) { ivpu_dbg(vdev, JOB, "Too many active jobs in ctx %d\n", file_priv->ctx.id); ret = -EBUSY; @@ -699,7 +665,7 @@ int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) int idx, ret; u8 priority; - if (params->engine > DRM_IVPU_ENGINE_COPY) + if (params->engine != DRM_IVPU_ENGINE_COMPUTE) return -EINVAL; if (params->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index 6accb94028c7..8b19e3f8b4cf 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -28,8 +28,10 @@ struct ivpu_cmdq { struct ivpu_bo *secondary_preempt_buf; struct ivpu_bo *mem; u32 entry_count; + u32 id; u32 db_id; bool db_registered; + u8 priority; }; /** diff --git a/drivers/accel/ivpu/ivpu_jsm_msg.c b/drivers/accel/ivpu/ivpu_jsm_msg.c index 8f5d904ca1ed..30a40be76930 100644 --- a/drivers/accel/ivpu/ivpu_jsm_msg.c +++ b/drivers/accel/ivpu/ivpu_jsm_msg.c @@ -132,7 +132,7 @@ int ivpu_jsm_get_heartbeat(struct ivpu_device *vdev, u32 engine, u64 *heartbeat) struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.query_engine_hb.engine_idx = engine; @@ -155,7 +155,7 @@ int ivpu_jsm_reset_engine(struct ivpu_device *vdev, u32 engine) struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.engine_reset.engine_idx = engine; @@ -174,7 +174,7 @@ int ivpu_jsm_preempt_engine(struct ivpu_device *vdev, u32 engine, u32 preempt_id struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.engine_preempt.engine_idx = engine; @@ -346,7 +346,7 @@ int ivpu_jsm_hws_resume_engine(struct ivpu_device *vdev, u32 engine) struct vpu_jsm_msg resp; int ret; - if (engine >= VPU_ENGINE_NB) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.hws_resume_engine.engine_idx = engine; diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index c078e214b221..26ef52fbb93e 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -696,7 +696,7 @@ unlock: return ret; } -static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) +static int ivpu_mmu_cdtab_entry_set(struct ivpu_device *vdev, u32 ssid, u64 cd_dma, bool valid) { struct ivpu_mmu_info *mmu = vdev->mmu; struct ivpu_mmu_cdtab *cdtab = &mmu->cdtab; @@ -708,30 +708,29 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) return -EINVAL; entry = cdtab->base + (ssid * IVPU_MMU_CDTAB_ENT_SIZE); - - if (cd_dma != 0) { - cd[0] = FIELD_PREP(IVPU_MMU_CD_0_TCR_T0SZ, IVPU_MMU_T0SZ_48BIT) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_TG0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_IRGN0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_ORGN0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_SH0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_IPS, IVPU_MMU_IPS_48BIT) | - FIELD_PREP(IVPU_MMU_CD_0_ASID, ssid) | - IVPU_MMU_CD_0_TCR_EPD1 | - IVPU_MMU_CD_0_AA64 | - IVPU_MMU_CD_0_R | - IVPU_MMU_CD_0_ASET | - IVPU_MMU_CD_0_V; - cd[1] = cd_dma & IVPU_MMU_CD_1_TTB0_MASK; - cd[2] = 0; - cd[3] = 0x0000000000007444; - - /* For global context generate memory fault on VPU */ - if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) - cd[0] |= IVPU_MMU_CD_0_A; - } else { - memset(cd, 0, sizeof(cd)); - } + drm_WARN_ON(&vdev->drm, (entry[0] & IVPU_MMU_CD_0_V) == valid); + + cd[0] = FIELD_PREP(IVPU_MMU_CD_0_TCR_T0SZ, IVPU_MMU_T0SZ_48BIT) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_TG0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_IRGN0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_ORGN0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_SH0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_IPS, IVPU_MMU_IPS_48BIT) | + FIELD_PREP(IVPU_MMU_CD_0_ASID, ssid) | + IVPU_MMU_CD_0_TCR_EPD1 | + IVPU_MMU_CD_0_AA64 | + IVPU_MMU_CD_0_R | + IVPU_MMU_CD_0_ASET; + cd[1] = cd_dma & IVPU_MMU_CD_1_TTB0_MASK; + cd[2] = 0; + cd[3] = 0x0000000000007444; + + /* For global context generate memory fault on VPU */ + if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) + cd[0] |= IVPU_MMU_CD_0_A; + + if (valid) + cd[0] |= IVPU_MMU_CD_0_V; WRITE_ONCE(entry[1], cd[1]); WRITE_ONCE(entry[2], cd[2]); @@ -741,8 +740,8 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) if (!ivpu_is_force_snoop_enabled(vdev)) clflush_cache_range(entry, IVPU_MMU_CDTAB_ENT_SIZE); - ivpu_dbg(vdev, MMU, "CDTAB %s entry (SSID=%u, dma=%pad): 0x%llx, 0x%llx, 0x%llx, 0x%llx\n", - cd_dma ? "write" : "clear", ssid, &cd_dma, cd[0], cd[1], cd[2], cd[3]); + ivpu_dbg(vdev, MMU, "CDTAB set %s entry (SSID=%u, dma=%pad): 0x%llx, 0x%llx, 0x%llx, 0x%llx\n", + valid ? "valid" : "invalid", ssid, &cd_dma, cd[0], cd[1], cd[2], cd[3]); mutex_lock(&mmu->lock); if (!mmu->on) @@ -750,38 +749,18 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) ret = ivpu_mmu_cmdq_write_cfgi_all(vdev); if (ret) - goto unlock; + goto err_invalidate; ret = ivpu_mmu_cmdq_sync(vdev); + if (ret) + goto err_invalidate; unlock: mutex_unlock(&mmu->lock); - return ret; -} - -static int ivpu_mmu_cd_add_gbl(struct ivpu_device *vdev) -{ - int ret; - - ret = ivpu_mmu_cd_add(vdev, 0, vdev->gctx.pgtable.pgd_dma); - if (ret) - ivpu_err(vdev, "Failed to add global CD entry: %d\n", ret); - - return ret; -} - -static int ivpu_mmu_cd_add_user(struct ivpu_device *vdev, u32 ssid, dma_addr_t cd_dma) -{ - int ret; - - if (ssid == 0) { - ivpu_err(vdev, "Invalid SSID: %u\n", ssid); - return -EINVAL; - } - - ret = ivpu_mmu_cd_add(vdev, ssid, cd_dma); - if (ret) - ivpu_err(vdev, "Failed to add CD entry SSID=%u: %d\n", ssid, ret); + return 0; +err_invalidate: + WRITE_ONCE(entry[0], 0); + mutex_unlock(&mmu->lock); return ret; } @@ -808,12 +787,6 @@ int ivpu_mmu_init(struct ivpu_device *vdev) return ret; } - ret = ivpu_mmu_cd_add_gbl(vdev); - if (ret) { - ivpu_err(vdev, "Failed to initialize strtab: %d\n", ret); - return ret; - } - ret = ivpu_mmu_enable(vdev); if (ret) { ivpu_err(vdev, "Failed to resume MMU: %d\n", ret); @@ -966,12 +939,12 @@ void ivpu_mmu_irq_gerr_handler(struct ivpu_device *vdev) REGV_WR32(IVPU_MMU_REG_GERRORN, gerror_val); } -int ivpu_mmu_set_pgtable(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable) +int ivpu_mmu_cd_set(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable) { - return ivpu_mmu_cd_add_user(vdev, ssid, pgtable->pgd_dma); + return ivpu_mmu_cdtab_entry_set(vdev, ssid, pgtable->pgd_dma, true); } -void ivpu_mmu_clear_pgtable(struct ivpu_device *vdev, int ssid) +void ivpu_mmu_cd_clear(struct ivpu_device *vdev, int ssid) { - ivpu_mmu_cd_add_user(vdev, ssid, 0); /* 0 will clear CD entry */ + ivpu_mmu_cdtab_entry_set(vdev, ssid, 0, false); } diff --git a/drivers/accel/ivpu/ivpu_mmu.h b/drivers/accel/ivpu/ivpu_mmu.h index 6fa35c240710..7afea9cd8731 100644 --- a/drivers/accel/ivpu/ivpu_mmu.h +++ b/drivers/accel/ivpu/ivpu_mmu.h @@ -40,8 +40,8 @@ struct ivpu_mmu_info { int ivpu_mmu_init(struct ivpu_device *vdev); void ivpu_mmu_disable(struct ivpu_device *vdev); int ivpu_mmu_enable(struct ivpu_device *vdev); -int ivpu_mmu_set_pgtable(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable); -void ivpu_mmu_clear_pgtable(struct ivpu_device *vdev, int ssid); +int ivpu_mmu_cd_set(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable); +void ivpu_mmu_cd_clear(struct ivpu_device *vdev, int ssid); int ivpu_mmu_invalidate_tlb(struct ivpu_device *vdev, u16 ssid); void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_mmu_context.c b/drivers/accel/ivpu/ivpu_mmu_context.c index bbe652a7019d..891967a95bc3 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.c +++ b/drivers/accel/ivpu/ivpu_mmu_context.c @@ -90,19 +90,6 @@ static void ivpu_pgtable_free_page(struct ivpu_device *vdev, u64 *cpu_addr, dma_ } } -static int ivpu_mmu_pgtable_init(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) -{ - dma_addr_t pgd_dma; - - pgtable->pgd_dma_ptr = ivpu_pgtable_alloc_page(vdev, &pgd_dma); - if (!pgtable->pgd_dma_ptr) - return -ENOMEM; - - pgtable->pgd_dma = pgd_dma; - - return 0; -} - static void ivpu_mmu_pgtables_free(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) { int pgd_idx, pud_idx, pmd_idx; @@ -140,6 +127,27 @@ static void ivpu_mmu_pgtables_free(struct ivpu_device *vdev, struct ivpu_mmu_pgt } ivpu_pgtable_free_page(vdev, pgtable->pgd_dma_ptr, pgtable->pgd_dma); + pgtable->pgd_dma_ptr = NULL; + pgtable->pgd_dma = 0; +} + +static u64* +ivpu_mmu_ensure_pgd(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) +{ + u64 *pgd_dma_ptr = pgtable->pgd_dma_ptr; + dma_addr_t pgd_dma; + + if (pgd_dma_ptr) + return pgd_dma_ptr; + + pgd_dma_ptr = ivpu_pgtable_alloc_page(vdev, &pgd_dma); + if (!pgd_dma_ptr) + return NULL; + + pgtable->pgd_dma_ptr = pgd_dma_ptr; + pgtable->pgd_dma = pgd_dma; + + return pgd_dma_ptr; } static u64* @@ -237,6 +245,12 @@ ivpu_mmu_context_map_page(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx int pmd_idx = FIELD_GET(IVPU_MMU_PMD_INDEX_MASK, vpu_addr); int pte_idx = FIELD_GET(IVPU_MMU_PTE_INDEX_MASK, vpu_addr); + drm_WARN_ON(&vdev->drm, ctx->id == IVPU_RESERVED_CONTEXT_MMU_SSID); + + /* Allocate PGD - first level page table if needed */ + if (!ivpu_mmu_ensure_pgd(vdev, &ctx->pgtable)) + return -ENOMEM; + /* Allocate PUD - second level page table if needed */ if (!ivpu_mmu_ensure_pud(vdev, &ctx->pgtable, pgd_idx)) return -ENOMEM; @@ -418,6 +432,7 @@ int ivpu_mmu_context_map_sgt(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u64 vpu_addr, struct sg_table *sgt, bool llc_coherent) { + size_t start_vpu_addr = vpu_addr; struct scatterlist *sg; int ret; u64 prot; @@ -448,20 +463,36 @@ ivpu_mmu_context_map_sgt(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, ret = ivpu_mmu_context_map_pages(vdev, ctx, vpu_addr, dma_addr, size, prot); if (ret) { ivpu_err(vdev, "Failed to map context pages\n"); - mutex_unlock(&ctx->lock); - return ret; + goto err_unmap_pages; } vpu_addr += size; } + if (!ctx->is_cd_valid) { + ret = ivpu_mmu_cd_set(vdev, ctx->id, &ctx->pgtable); + if (ret) { + ivpu_err(vdev, "Failed to set context descriptor for context %u: %d\n", + ctx->id, ret); + goto err_unmap_pages; + } + ctx->is_cd_valid = true; + } + /* Ensure page table modifications are flushed from wc buffers to memory */ wmb(); - mutex_unlock(&ctx->lock); - ret = ivpu_mmu_invalidate_tlb(vdev, ctx->id); - if (ret) + if (ret) { ivpu_err(vdev, "Failed to invalidate TLB for ctx %u: %d\n", ctx->id, ret); + goto err_unmap_pages; + } + + mutex_unlock(&ctx->lock); + return 0; + +err_unmap_pages: + ivpu_mmu_context_unmap_pages(ctx, start_vpu_addr, vpu_addr - start_vpu_addr); + mutex_unlock(&ctx->lock); return ret; } @@ -530,65 +561,75 @@ ivpu_mmu_context_remove_node(struct ivpu_mmu_context *ctx, struct drm_mm_node *n mutex_unlock(&ctx->lock); } -static int -ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id) +void ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id) { u64 start, end; - int ret; mutex_init(&ctx->lock); - ret = ivpu_mmu_pgtable_init(vdev, &ctx->pgtable); - if (ret) { - ivpu_err(vdev, "Failed to initialize pgtable for ctx %u: %d\n", context_id, ret); - return ret; - } - if (!context_id) { start = vdev->hw->ranges.global.start; end = vdev->hw->ranges.shave.end; } else { - start = vdev->hw->ranges.user.start; - end = vdev->hw->ranges.dma.end; + start = min_t(u64, vdev->hw->ranges.user.start, vdev->hw->ranges.shave.start); + end = max_t(u64, vdev->hw->ranges.user.end, vdev->hw->ranges.dma.end); } drm_mm_init(&ctx->mm, start, end - start); ctx->id = context_id; - - return 0; } -static void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) +void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) { - if (drm_WARN_ON(&vdev->drm, !ctx->pgtable.pgd_dma_ptr)) - return; + if (ctx->is_cd_valid) { + ivpu_mmu_cd_clear(vdev, ctx->id); + ctx->is_cd_valid = false; + } mutex_destroy(&ctx->lock); ivpu_mmu_pgtables_free(vdev, &ctx->pgtable); drm_mm_takedown(&ctx->mm); - - ctx->pgtable.pgd_dma_ptr = NULL; - ctx->pgtable.pgd_dma = 0; } -int ivpu_mmu_global_context_init(struct ivpu_device *vdev) +void ivpu_mmu_global_context_init(struct ivpu_device *vdev) { - return ivpu_mmu_context_init(vdev, &vdev->gctx, IVPU_GLOBAL_CONTEXT_MMU_SSID); + ivpu_mmu_context_init(vdev, &vdev->gctx, IVPU_GLOBAL_CONTEXT_MMU_SSID); } void ivpu_mmu_global_context_fini(struct ivpu_device *vdev) { - return ivpu_mmu_context_fini(vdev, &vdev->gctx); + ivpu_mmu_context_fini(vdev, &vdev->gctx); } int ivpu_mmu_reserved_context_init(struct ivpu_device *vdev) { - return ivpu_mmu_user_context_init(vdev, &vdev->rctx, IVPU_RESERVED_CONTEXT_MMU_SSID); + int ret; + + ivpu_mmu_context_init(vdev, &vdev->rctx, IVPU_RESERVED_CONTEXT_MMU_SSID); + + mutex_lock(&vdev->rctx.lock); + + if (!ivpu_mmu_ensure_pgd(vdev, &vdev->rctx.pgtable)) { + ivpu_err(vdev, "Failed to allocate root page table for reserved context\n"); + ret = -ENOMEM; + goto unlock; + } + + ret = ivpu_mmu_cd_set(vdev, vdev->rctx.id, &vdev->rctx.pgtable); + if (ret) { + ivpu_err(vdev, "Failed to set context descriptor for reserved context\n"); + goto unlock; + } + +unlock: + mutex_unlock(&vdev->rctx.lock); + return ret; } void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev) { - return ivpu_mmu_user_context_fini(vdev, &vdev->rctx); + ivpu_mmu_cd_clear(vdev, vdev->rctx.id); + ivpu_mmu_context_fini(vdev, &vdev->rctx); } void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid) @@ -603,36 +644,3 @@ void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid) xa_unlock(&vdev->context_xa); } - -int ivpu_mmu_user_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 ctx_id) -{ - int ret; - - drm_WARN_ON(&vdev->drm, !ctx_id); - - ret = ivpu_mmu_context_init(vdev, ctx, ctx_id); - if (ret) { - ivpu_err(vdev, "Failed to initialize context %u: %d\n", ctx_id, ret); - return ret; - } - - ret = ivpu_mmu_set_pgtable(vdev, ctx_id, &ctx->pgtable); - if (ret) { - ivpu_err(vdev, "Failed to set page table for context %u: %d\n", ctx_id, ret); - goto err_context_fini; - } - - return 0; - -err_context_fini: - ivpu_mmu_context_fini(vdev, ctx); - return ret; -} - -void ivpu_mmu_user_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) -{ - drm_WARN_ON(&vdev->drm, !ctx->id); - - ivpu_mmu_clear_pgtable(vdev, ctx->id); - ivpu_mmu_context_fini(vdev, ctx); -} diff --git a/drivers/accel/ivpu/ivpu_mmu_context.h b/drivers/accel/ivpu/ivpu_mmu_context.h index 7f9aaf3d10c2..8042fc067062 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.h +++ b/drivers/accel/ivpu/ivpu_mmu_context.h @@ -23,19 +23,20 @@ struct ivpu_mmu_pgtable { }; struct ivpu_mmu_context { - struct mutex lock; /* Protects: mm, pgtable */ + struct mutex lock; /* Protects: mm, pgtable, is_cd_valid */ struct drm_mm mm; struct ivpu_mmu_pgtable pgtable; + bool is_cd_valid; u32 id; }; -int ivpu_mmu_global_context_init(struct ivpu_device *vdev); +void ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id); +void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx); +void ivpu_mmu_global_context_init(struct ivpu_device *vdev); void ivpu_mmu_global_context_fini(struct ivpu_device *vdev); int ivpu_mmu_reserved_context_init(struct ivpu_device *vdev); void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev); -int ivpu_mmu_user_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 ctx_id); -void ivpu_mmu_user_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx); void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid); int ivpu_mmu_context_insert_node(struct ivpu_mmu_context *ctx, const struct ivpu_addr_range *range, diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index e567df79a612..dbc0711e28d1 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -24,8 +24,10 @@ #include "vpu_boot_api.h" static bool ivpu_disable_recovery; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(disable_recovery, ivpu_disable_recovery, bool, 0644); MODULE_PARM_DESC(disable_recovery, "Disables recovery when NPU hang is detected"); +#endif static unsigned long ivpu_tdr_timeout_ms; module_param_named(tdr_timeout_ms, ivpu_tdr_timeout_ms, ulong, 0644); |