reg_state[CTX_PDP0_LDW + 1] = lower_32_bits(px_dma(&ppgtt->pml4)); \
} while (0)
-enum {
- ADVANCED_CONTEXT = 0,
- LEGACY_32B_CONTEXT,
- ADVANCED_AD_CONTEXT,
- LEGACY_64B_CONTEXT
-};
-#define GEN8_CTX_ADDRESSING_MODE_SHIFT 3
-#define GEN8_CTX_ADDRESSING_MODE(dev) (USES_FULL_48BIT_PPGTT(dev) ?\
- LEGACY_64B_CONTEXT :\
- LEGACY_32B_CONTEXT)
enum {
FAULT_AND_HANG = 0,
FAULT_AND_HALT, /* Debug only */
FAULT_AND_CONTINUE /* Unsupported */
};
#define GEN8_CTX_ID_SHIFT 32
+#define GEN8_CTX_ID_WIDTH 21
#define GEN8_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x17
#define GEN9_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x26
-static int intel_lr_context_pin(struct intel_context *ctx,
+/* Typical size of the average request (2 pipecontrols and a MI_BB) */
+#define EXECLISTS_REQUEST_SIZE 64 /* bytes */
+
+static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
+ struct intel_engine_cs *engine);
+static int intel_lr_context_pin(struct i915_gem_context *ctx,
struct intel_engine_cs *engine);
/**
* intel_sanitize_enable_execlists() - sanitize i915.enable_execlists
- * @dev: DRM device.
+ * @dev_priv: i915 device private
* @enable_execlists: value of i915.enable_execlists module parameter.
*
* Only certain platforms support Execlists (the prerequisites being
*
* Return: 1 if Execlists is supported and has to be enabled.
*/
-int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists)
+int intel_sanitize_enable_execlists(struct drm_i915_private *dev_priv, int enable_execlists)
{
- WARN_ON(i915.enable_ppgtt == -1);
-
/* On platforms with execlist available, vGPU will only
* support execlist mode, no ring buffer mode.
*/
- if (HAS_LOGICAL_RING_CONTEXTS(dev) && intel_vgpu_active(dev))
+ if (HAS_LOGICAL_RING_CONTEXTS(dev_priv) && intel_vgpu_active(dev_priv))
return 1;
- if (INTEL_INFO(dev)->gen >= 9)
+ if (INTEL_GEN(dev_priv) >= 9)
return 1;
if (enable_execlists == 0)
return 0;
- if (HAS_LOGICAL_RING_CONTEXTS(dev) && USES_PPGTT(dev) &&
+ if (HAS_LOGICAL_RING_CONTEXTS(dev_priv) &&
+ USES_PPGTT(dev_priv) &&
i915.use_mmio_flip >= 0)
return 1;
static void
logical_ring_init_platform_invariants(struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
+ struct drm_i915_private *dev_priv = engine->i915;
- if (IS_GEN8(dev) || IS_GEN9(dev))
+ if (IS_GEN8(dev_priv) || IS_GEN9(dev_priv))
engine->idle_lite_restore_wa = ~0;
- engine->disable_lite_restore_wa = (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
- IS_BXT_REVID(dev, 0, BXT_REVID_A1)) &&
+ engine->disable_lite_restore_wa = (IS_SKL_REVID(dev_priv, 0, SKL_REVID_B0) ||
+ IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) &&
(engine->id == VCS || engine->id == VCS2);
engine->ctx_desc_template = GEN8_CTX_VALID;
- engine->ctx_desc_template |= GEN8_CTX_ADDRESSING_MODE(dev) <<
- GEN8_CTX_ADDRESSING_MODE_SHIFT;
- if (IS_GEN8(dev))
+ if (IS_GEN8(dev_priv))
engine->ctx_desc_template |= GEN8_CTX_L3LLC_COHERENT;
engine->ctx_desc_template |= GEN8_CTX_PRIVILEGE;
/**
* intel_lr_context_descriptor_update() - calculate & cache the descriptor
* descriptor for a pinned context
- *
* @ctx: Context to work on
- * @ring: Engine the descriptor will be used with
+ * @engine: Engine the descriptor will be used with
*
* The context descriptor encodes various attributes of a context,
* including its GTT address and some flags. Because it's fairly
* expensive to calculate, we'll just do it once and cache the result,
* which remains valid until the context is unpinned.
*
- * This is what a descriptor looks like, from LSB to MSB:
- * bits 0-11: flags, GEN8_CTX_* (cached in ctx_desc_template)
- * bits 12-31: LRCA, GTT address of (the HWSP of) this context
- * bits 32-51: ctx ID, a globally unique tag (the LRCA again!)
- * bits 52-63: reserved, may encode the engine ID (for GuC)
+ * This is what a descriptor looks like, from LSB to MSB::
+ *
+ * bits 0-11: flags, GEN8_CTX_* (cached in ctx_desc_template)
+ * bits 12-31: LRCA, GTT address of (the HWSP of) this context
+ * bits 32-52: ctx ID, a globally unique tag
+ * bits 53-54: mbz, reserved for use by hardware
+ * bits 55-63: group ID, currently unused and set to 0
*/
static void
-intel_lr_context_descriptor_update(struct intel_context *ctx,
+intel_lr_context_descriptor_update(struct i915_gem_context *ctx,
struct intel_engine_cs *engine)
{
- uint64_t lrca, desc;
+ struct intel_context *ce = &ctx->engine[engine->id];
+ u64 desc;
- lrca = ctx->engine[engine->id].lrc_vma->node.start +
- LRC_PPHWSP_PN * PAGE_SIZE;
+ BUILD_BUG_ON(MAX_CONTEXT_HW_ID > (1<<GEN8_CTX_ID_WIDTH));
- desc = engine->ctx_desc_template; /* bits 0-11 */
- desc |= lrca; /* bits 12-31 */
- desc |= (lrca >> PAGE_SHIFT) << GEN8_CTX_ID_SHIFT; /* bits 32-51 */
+ desc = ctx->desc_template; /* bits 3-4 */
+ desc |= engine->ctx_desc_template; /* bits 0-11 */
+ desc |= ce->lrc_vma->node.start + LRC_PPHWSP_PN * PAGE_SIZE;
+ /* bits 12-31 */
+ desc |= (u64)ctx->hw_id << GEN8_CTX_ID_SHIFT; /* bits 32-52 */
- ctx->engine[engine->id].lrc_desc = desc;
+ ce->lrc_desc = desc;
}
-uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
+uint64_t intel_lr_context_descriptor(struct i915_gem_context *ctx,
struct intel_engine_cs *engine)
{
return ctx->engine[engine->id].lrc_desc;
}
-/**
- * intel_execlists_ctx_id() - get the Execlists Context ID
- * @ctx: Context to get the ID for
- * @ring: Engine to get the ID for
- *
- * Do not confuse with ctx->id! Unfortunately we have a name overload
- * here: the old context ID we pass to userspace as a handler so that
- * they can refer to a context, and the new context ID we pass to the
- * ELSP so that the GPU can inform us of the context status via
- * interrupts.
- *
- * The context ID is a portion of the context descriptor, so we can
- * just extract the required part from the cached descriptor.
- *
- * Return: 20-bits globally unique context ID.
- */
-u32 intel_execlists_ctx_id(struct intel_context *ctx,
- struct intel_engine_cs *engine)
-{
- return intel_lr_context_descriptor(ctx, engine) >> GEN8_CTX_ID_SHIFT;
-}
-
static void execlists_elsp_write(struct drm_i915_gem_request *rq0,
struct drm_i915_gem_request *rq1)
{
struct intel_engine_cs *engine = rq0->engine;
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = rq0->i915;
uint64_t desc[2];
if (rq1) {
struct i915_hw_ppgtt *ppgtt = rq->ctx->ppgtt;
uint32_t *reg_state = rq->ctx->engine[engine->id].lrc_reg_state;
- reg_state[CTX_RING_TAIL+1] = rq->tail;
+ reg_state[CTX_RING_TAIL+1] = intel_ring_offset(rq->ring, rq->tail);
/* True 32b PPGTT with dynamic page allocation: update PDP
* registers and point the unallocated PDPs to scratch page.
execlists_update_context_pdps(ppgtt, reg_state);
}
-static void execlists_submit_requests(struct drm_i915_gem_request *rq0,
- struct drm_i915_gem_request *rq1)
+static void execlists_elsp_submit_contexts(struct drm_i915_gem_request *rq0,
+ struct drm_i915_gem_request *rq1)
{
struct drm_i915_private *dev_priv = rq0->i915;
unsigned int fw_domains = rq0->engine->fw_domains;
spin_unlock_irq(&dev_priv->uncore.lock);
}
-static void execlists_context_unqueue(struct intel_engine_cs *engine)
+static inline void execlists_context_status_change(
+ struct drm_i915_gem_request *rq,
+ unsigned long status)
+{
+ /*
+ * Only used when GVT-g is enabled now. When GVT-g is disabled,
+ * The compiler should eliminate this function as dead-code.
+ */
+ if (!IS_ENABLED(CONFIG_DRM_I915_GVT))
+ return;
+
+ atomic_notifier_call_chain(&rq->ctx->status_notifier, status, rq);
+}
+
+static void execlists_unqueue(struct intel_engine_cs *engine)
{
struct drm_i915_gem_request *req0 = NULL, *req1 = NULL;
struct drm_i915_gem_request *cursor, *tmp;
* If irqs are not active generate a warning as batches that finish
* without the irqs may get lost and a GPU Hang may occur.
*/
- WARN_ON(!intel_irqs_enabled(engine->dev->dev_private));
+ WARN_ON(!intel_irqs_enabled(engine->i915));
/* Try to read in pairs */
list_for_each_entry_safe(cursor, tmp, &engine->execlist_queue,
/* Same ctx: ignore first request, as second request
* will update tail past first request's workload */
cursor->elsp_submitted = req0->elsp_submitted;
- list_move_tail(&req0->execlist_link,
- &engine->execlist_retired_req_list);
+ list_del(&req0->execlist_link);
+ i915_gem_request_put(req0);
req0 = cursor;
} else {
+ if (IS_ENABLED(CONFIG_DRM_I915_GVT)) {
+ /*
+ * req0 (after merged) ctx requires single
+ * submission, stop picking
+ */
+ if (req0->ctx->execlists_force_single_submission)
+ break;
+ /*
+ * req0 ctx doesn't require single submission,
+ * but next req ctx requires, stop picking
+ */
+ if (cursor->ctx->execlists_force_single_submission)
+ break;
+ }
req1 = cursor;
WARN_ON(req1->elsp_submitted);
break;
if (unlikely(!req0))
return;
+ execlists_context_status_change(req0, INTEL_CONTEXT_SCHEDULE_IN);
+
+ if (req1)
+ execlists_context_status_change(req1,
+ INTEL_CONTEXT_SCHEDULE_IN);
+
if (req0->elsp_submitted & engine->idle_lite_restore_wa) {
/*
* WaIdleLiteRestore: make sure we never cause a lite restore
* resubmit the request. See gen8_emit_request() for where we
* prepare the padding after the end of the request.
*/
- struct intel_ringbuffer *ringbuf;
-
- ringbuf = req0->ctx->engine[engine->id].ringbuf;
req0->tail += 8;
- req0->tail &= ringbuf->size - 1;
+ req0->tail &= req0->ring->size - 1;
}
- execlists_submit_requests(req0, req1);
+ execlists_elsp_submit_contexts(req0, req1);
}
static unsigned int
-execlists_check_remove_request(struct intel_engine_cs *engine, u32 request_id)
+execlists_check_remove_request(struct intel_engine_cs *engine, u32 ctx_id)
{
struct drm_i915_gem_request *head_req;
struct drm_i915_gem_request,
execlist_link);
- if (!head_req)
- return 0;
-
- if (unlikely(intel_execlists_ctx_id(head_req->ctx, engine) != request_id))
- return 0;
+ if (WARN_ON(!head_req || (head_req->ctx_hw_id != ctx_id)))
+ return 0;
WARN(head_req->elsp_submitted == 0, "Never submitted head request\n");
if (--head_req->elsp_submitted > 0)
return 0;
- list_move_tail(&head_req->execlist_link,
- &engine->execlist_retired_req_list);
+ execlists_context_status_change(head_req, INTEL_CONTEXT_SCHEDULE_OUT);
+
+ list_del(&head_req->execlist_link);
+ i915_gem_request_put(head_req);
return 1;
}
get_context_status(struct intel_engine_cs *engine, unsigned int read_pointer,
u32 *context_id)
{
- struct drm_i915_private *dev_priv = engine->dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
u32 status;
read_pointer %= GEN8_CSB_ENTRIES;
return status;
}
-/**
- * intel_lrc_irq_handler() - handle Context Switch interrupts
- * @engine: Engine Command Streamer to handle.
- *
+/*
* Check the unread Context Status Buffers and manage the submission of new
* contexts to the ELSP accordingly.
*/
static void intel_lrc_irq_handler(unsigned long data)
{
struct intel_engine_cs *engine = (struct intel_engine_cs *)data;
- struct drm_i915_private *dev_priv = engine->dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
u32 status_pointer;
unsigned int read_pointer, write_pointer;
u32 csb[GEN8_CSB_ENTRIES][2];
if (submit_contexts) {
if (!engine->disable_lite_restore_wa ||
(csb[i][0] & GEN8_CTX_STATUS_ACTIVE_IDLE))
- execlists_context_unqueue(engine);
+ execlists_unqueue(engine);
}
spin_unlock(&engine->execlist_lock);
DRM_ERROR("More than two context complete events?\n");
}
-static void execlists_context_queue(struct drm_i915_gem_request *request)
+static void execlists_submit_request(struct drm_i915_gem_request *request)
{
struct intel_engine_cs *engine = request->engine;
struct drm_i915_gem_request *cursor;
int num_elements = 0;
- if (request->ctx != request->i915->kernel_context)
- intel_lr_context_pin(request->ctx, engine);
-
- i915_gem_request_reference(request);
-
spin_lock_bh(&engine->execlist_lock);
list_for_each_entry(cursor, &engine->execlist_queue, execlist_link)
if (request->ctx == tail_req->ctx) {
WARN(tail_req->elsp_submitted != 0,
"More than 2 already-submitted reqs queued\n");
- list_move_tail(&tail_req->execlist_link,
- &engine->execlist_retired_req_list);
+ list_del(&tail_req->execlist_link);
+ i915_gem_request_put(tail_req);
}
}
+ i915_gem_request_get(request);
list_add_tail(&request->execlist_link, &engine->execlist_queue);
+ request->ctx_hw_id = request->ctx->hw_id;
if (num_elements == 0)
- execlists_context_unqueue(engine);
+ execlists_unqueue(engine);
spin_unlock_bh(&engine->execlist_lock);
}
-static int logical_ring_invalidate_all_caches(struct drm_i915_gem_request *req)
-{
- struct intel_engine_cs *engine = req->engine;
- uint32_t flush_domains;
- int ret;
-
- flush_domains = 0;
- if (engine->gpu_caches_dirty)
- flush_domains = I915_GEM_GPU_DOMAINS;
-
- ret = engine->emit_flush(req, I915_GEM_GPU_DOMAINS, flush_domains);
- if (ret)
- return ret;
-
- engine->gpu_caches_dirty = false;
- return 0;
-}
-
-static int execlists_move_to_gpu(struct drm_i915_gem_request *req,
- struct list_head *vmas)
+int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request)
{
- const unsigned other_rings = ~intel_engine_flag(req->engine);
- struct i915_vma *vma;
- uint32_t flush_domains = 0;
- bool flush_chipset = false;
+ struct intel_engine_cs *engine = request->engine;
+ struct intel_context *ce = &request->ctx->engine[engine->id];
int ret;
- list_for_each_entry(vma, vmas, exec_list) {
- struct drm_i915_gem_object *obj = vma->obj;
-
- if (obj->active & other_rings) {
- ret = i915_gem_object_sync(obj, req->engine, &req);
- if (ret)
- return ret;
- }
-
- if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
- flush_chipset |= i915_gem_clflush_object(obj, false);
-
- flush_domains |= obj->base.write_domain;
- }
-
- if (flush_domains & I915_GEM_DOMAIN_GTT)
- wmb();
-
- /* Unconditionally invalidate gpu caches and ensure that we do flush
- * any residual writes from the previous batch.
+ /* Flush enough space to reduce the likelihood of waiting after
+ * we start building the request - in which case we will just
+ * have to repeat work.
*/
- return logical_ring_invalidate_all_caches(req);
-}
+ request->reserved_space += EXECLISTS_REQUEST_SIZE;
-int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request)
-{
- int ret = 0;
+ if (!ce->state) {
+ ret = execlists_context_deferred_alloc(request->ctx, engine);
+ if (ret)
+ return ret;
+ }
- request->ringbuf = request->ctx->engine[request->engine->id].ringbuf;
+ request->ring = ce->ring;
if (i915.enable_guc_submission) {
/*
* going any further, as the i915_add_request() call
* later on mustn't fail ...
*/
- struct intel_guc *guc = &request->i915->guc;
-
- ret = i915_guc_wq_check_space(guc->execbuf_client);
+ ret = i915_guc_wq_check_space(request);
if (ret)
return ret;
}
- if (request->ctx != request->i915->kernel_context)
- ret = intel_lr_context_pin(request->ctx, request->engine);
+ ret = intel_lr_context_pin(request->ctx, engine);
+ if (ret)
+ return ret;
+
+ ret = intel_ring_begin(request, 0);
+ if (ret)
+ goto err_unpin;
+
+ if (!ce->initialised) {
+ ret = engine->init_context(request);
+ if (ret)
+ goto err_unpin;
+
+ ce->initialised = true;
+ }
+
+ /* Note that after this point, we have committed to using
+ * this request as it is being used to both track the
+ * state of engine initialisation and liveness of the
+ * golden renderstate above. Think twice before you try
+ * to cancel/unwind this request now.
+ */
+
+ request->reserved_space -= EXECLISTS_REQUEST_SIZE;
+ return 0;
+err_unpin:
+ intel_lr_context_unpin(request->ctx, engine);
return ret;
}
/*
- * intel_logical_ring_advance_and_submit() - advance the tail and submit the workload
+ * intel_logical_ring_advance() - advance the tail and prepare for submission
* @request: Request to advance the logical ringbuffer of.
*
* The tail is updated in our logical ringbuffer struct, not in the actual context. What
* point, the tail *inside* the context is updated and the ELSP written to.
*/
static int
-intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
+intel_logical_ring_advance(struct drm_i915_gem_request *request)
{
- struct intel_ringbuffer *ringbuf = request->ringbuf;
- struct drm_i915_private *dev_priv = request->i915;
+ struct intel_ring *ring = request->ring;
struct intel_engine_cs *engine = request->engine;
- intel_logical_ring_advance(ringbuf);
- request->tail = ringbuf->tail;
+ intel_ring_advance(ring);
+ request->tail = ring->tail;
/*
* Here we add two extra NOOPs as padding to avoid
*
* Caller must reserve WA_TAIL_DWORDS for us!
*/
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- intel_logical_ring_advance(ringbuf);
-
- if (intel_engine_stopped(engine))
- return 0;
-
- if (engine->last_context != request->ctx) {
- if (engine->last_context)
- intel_lr_context_unpin(engine->last_context, engine);
- if (request->ctx != request->i915->kernel_context) {
- intel_lr_context_pin(request->ctx, engine);
- engine->last_context = request->ctx;
- } else {
- engine->last_context = NULL;
- }
- }
-
- if (dev_priv->guc.execbuf_client)
- i915_guc_submit(dev_priv->guc.execbuf_client, request);
- else
- execlists_context_queue(request);
-
- return 0;
-}
-
-int intel_logical_ring_reserve_space(struct drm_i915_gem_request *request)
-{
- /*
- * The first call merely notes the reserve request and is common for
- * all back ends. The subsequent localised _begin() call actually
- * ensures that the reservation is available. Without the begin, if
- * the request creator immediately submitted the request without
- * adding any commands to it then there might not actually be
- * sufficient room for the submission commands.
+ intel_ring_emit(ring, MI_NOOP);
+ intel_ring_emit(ring, MI_NOOP);
+ intel_ring_advance(ring);
+
+ /* We keep the previous context alive until we retire the following
+ * request. This ensures that any the context object is still pinned
+ * for any residual writes the HW makes into it on the context switch
+ * into the next object following the breadcrumb. Otherwise, we may
+ * retire the context too early.
*/
- intel_ring_reserved_space_reserve(request->ringbuf, MIN_SPACE_FOR_ADD_REQUEST);
-
- return intel_ring_begin(request, 0);
-}
-
-/**
- * execlists_submission() - submit a batchbuffer for execution, Execlists style
- * @dev: DRM device.
- * @file: DRM file.
- * @ring: Engine Command Streamer to submit to.
- * @ctx: Context to employ for this submission.
- * @args: execbuffer call arguments.
- * @vmas: list of vmas.
- * @batch_obj: the batchbuffer to submit.
- * @exec_start: batchbuffer start virtual address pointer.
- * @dispatch_flags: translated execbuffer call flags.
- *
- * This is the evil twin version of i915_gem_ringbuffer_submission. It abstracts
- * away the submission details of the execbuffer ioctl call.
- *
- * Return: non-zero if the submission fails.
- */
-int intel_execlists_submission(struct i915_execbuffer_params *params,
- struct drm_i915_gem_execbuffer2 *args,
- struct list_head *vmas)
-{
- struct drm_device *dev = params->dev;
- struct intel_engine_cs *engine = params->engine;
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_ringbuffer *ringbuf = params->ctx->engine[engine->id].ringbuf;
- u64 exec_start;
- int instp_mode;
- u32 instp_mask;
- int ret;
-
- instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
- instp_mask = I915_EXEC_CONSTANTS_MASK;
- switch (instp_mode) {
- case I915_EXEC_CONSTANTS_REL_GENERAL:
- case I915_EXEC_CONSTANTS_ABSOLUTE:
- case I915_EXEC_CONSTANTS_REL_SURFACE:
- if (instp_mode != 0 && engine != &dev_priv->engine[RCS]) {
- DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
- return -EINVAL;
- }
-
- if (instp_mode != dev_priv->relative_constants_mode) {
- if (instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
- DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
- return -EINVAL;
- }
-
- /* The HW changed the meaning on this bit on gen6 */
- instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
- }
- break;
- default:
- DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
- return -EINVAL;
- }
-
- if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
- DRM_DEBUG("sol reset is gen7 only\n");
- return -EINVAL;
- }
-
- ret = execlists_move_to_gpu(params->request, vmas);
- if (ret)
- return ret;
-
- if (engine == &dev_priv->engine[RCS] &&
- instp_mode != dev_priv->relative_constants_mode) {
- ret = intel_ring_begin(params->request, 4);
- if (ret)
- return ret;
-
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(1));
- intel_logical_ring_emit_reg(ringbuf, INSTPM);
- intel_logical_ring_emit(ringbuf, instp_mask << 16 | instp_mode);
- intel_logical_ring_advance(ringbuf);
-
- dev_priv->relative_constants_mode = instp_mode;
- }
-
- exec_start = params->batch_obj_vm_offset +
- args->batch_start_offset;
-
- ret = engine->emit_bb_start(params->request, exec_start, params->dispatch_flags);
- if (ret)
- return ret;
-
- trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
-
- i915_gem_execbuffer_move_to_active(vmas, params->request);
-
+ request->previous_context = engine->last_context;
+ engine->last_context = request->ctx;
return 0;
}
-void intel_execlists_retire_requests(struct intel_engine_cs *engine)
+void intel_execlists_cancel_requests(struct intel_engine_cs *engine)
{
struct drm_i915_gem_request *req, *tmp;
- struct list_head retired_list;
+ LIST_HEAD(cancel_list);
- WARN_ON(!mutex_is_locked(&engine->dev->struct_mutex));
- if (list_empty(&engine->execlist_retired_req_list))
- return;
+ WARN_ON(!mutex_is_locked(&engine->i915->drm.struct_mutex));
- INIT_LIST_HEAD(&retired_list);
spin_lock_bh(&engine->execlist_lock);
- list_replace_init(&engine->execlist_retired_req_list, &retired_list);
+ list_replace_init(&engine->execlist_queue, &cancel_list);
spin_unlock_bh(&engine->execlist_lock);
- list_for_each_entry_safe(req, tmp, &retired_list, execlist_link) {
- struct intel_context *ctx = req->ctx;
- struct drm_i915_gem_object *ctx_obj =
- ctx->engine[engine->id].state;
-
- if (ctx_obj && (ctx != req->i915->kernel_context))
- intel_lr_context_unpin(ctx, engine);
-
+ list_for_each_entry_safe(req, tmp, &cancel_list, execlist_link) {
list_del(&req->execlist_link);
- i915_gem_request_unreference(req);
+ i915_gem_request_put(req);
}
}
-void intel_logical_ring_stop(struct intel_engine_cs *engine)
-{
- struct drm_i915_private *dev_priv = engine->dev->dev_private;
- int ret;
-
- if (!intel_engine_initialized(engine))
- return;
-
- ret = intel_engine_idle(engine);
- if (ret)
- DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
- engine->name, ret);
-
- /* TODO: Is this correct with Execlists enabled? */
- I915_WRITE_MODE(engine, _MASKED_BIT_ENABLE(STOP_RING));
- if (wait_for((I915_READ_MODE(engine) & MODE_IDLE) != 0, 1000)) {
- DRM_ERROR("%s :timed out trying to stop ring\n", engine->name);
- return;
- }
- I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING));
-}
-
-int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
-{
- struct intel_engine_cs *engine = req->engine;
- int ret;
-
- if (!engine->gpu_caches_dirty)
- return 0;
-
- ret = engine->emit_flush(req, 0, I915_GEM_GPU_DOMAINS);
- if (ret)
- return ret;
-
- engine->gpu_caches_dirty = false;
- return 0;
-}
-
-static int intel_lr_context_do_pin(struct intel_context *ctx,
- struct intel_engine_cs *engine)
+static int intel_lr_context_pin(struct i915_gem_context *ctx,
+ struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct drm_i915_gem_object *ctx_obj = ctx->engine[engine->id].state;
- struct intel_ringbuffer *ringbuf = ctx->engine[engine->id].ringbuf;
+ struct drm_i915_private *dev_priv = ctx->i915;
+ struct intel_context *ce = &ctx->engine[engine->id];
void *vaddr;
u32 *lrc_reg_state;
int ret;
- WARN_ON(!mutex_is_locked(&engine->dev->struct_mutex));
+ lockdep_assert_held(&ctx->i915->drm.struct_mutex);
- ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
- PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
+ if (ce->pin_count++)
+ return 0;
+
+ ret = i915_gem_object_ggtt_pin(ce->state, NULL,
+ 0, GEN8_LR_CONTEXT_ALIGN,
+ PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
if (ret)
- return ret;
+ goto err;
- vaddr = i915_gem_object_pin_map(ctx_obj);
+ vaddr = i915_gem_object_pin_map(ce->state);
if (IS_ERR(vaddr)) {
ret = PTR_ERR(vaddr);
goto unpin_ctx_obj;
lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
- ret = intel_pin_and_map_ringbuffer_obj(engine->dev, ringbuf);
+ ret = intel_ring_pin(ce->ring);
if (ret)
goto unpin_map;
- ctx->engine[engine->id].lrc_vma = i915_gem_obj_to_ggtt(ctx_obj);
+ ce->lrc_vma = i915_gem_obj_to_ggtt(ce->state);
intel_lr_context_descriptor_update(ctx, engine);
- lrc_reg_state[CTX_RING_BUFFER_START+1] = ringbuf->vma->node.start;
- ctx->engine[engine->id].lrc_reg_state = lrc_reg_state;
- ctx_obj->dirty = true;
+
+ lrc_reg_state[CTX_RING_BUFFER_START+1] = ce->ring->vma->node.start;
+ ce->lrc_reg_state = lrc_reg_state;
+ ce->state->dirty = true;
/* Invalidate GuC TLB. */
if (i915.enable_guc_submission)
I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
- return ret;
+ i915_gem_context_get(ctx);
+ return 0;
unpin_map:
- i915_gem_object_unpin_map(ctx_obj);
+ i915_gem_object_unpin_map(ce->state);
unpin_ctx_obj:
- i915_gem_object_ggtt_unpin(ctx_obj);
-
+ i915_gem_object_ggtt_unpin(ce->state);
+err:
+ ce->pin_count = 0;
return ret;
}
-static int intel_lr_context_pin(struct intel_context *ctx,
- struct intel_engine_cs *engine)
+void intel_lr_context_unpin(struct i915_gem_context *ctx,
+ struct intel_engine_cs *engine)
{
- int ret = 0;
+ struct intel_context *ce = &ctx->engine[engine->id];
- if (ctx->engine[engine->id].pin_count++ == 0) {
- ret = intel_lr_context_do_pin(ctx, engine);
- if (ret)
- goto reset_pin_count;
+ lockdep_assert_held(&ctx->i915->drm.struct_mutex);
+ GEM_BUG_ON(ce->pin_count == 0);
- i915_gem_context_reference(ctx);
- }
- return ret;
+ if (--ce->pin_count)
+ return;
-reset_pin_count:
- ctx->engine[engine->id].pin_count = 0;
- return ret;
-}
+ intel_ring_unpin(ce->ring);
-void intel_lr_context_unpin(struct intel_context *ctx,
- struct intel_engine_cs *engine)
-{
- struct drm_i915_gem_object *ctx_obj = ctx->engine[engine->id].state;
+ i915_gem_object_unpin_map(ce->state);
+ i915_gem_object_ggtt_unpin(ce->state);
- WARN_ON(!mutex_is_locked(&ctx->i915->dev->struct_mutex));
- if (--ctx->engine[engine->id].pin_count == 0) {
- i915_gem_object_unpin_map(ctx_obj);
- intel_unpin_ringbuffer_obj(ctx->engine[engine->id].ringbuf);
- i915_gem_object_ggtt_unpin(ctx_obj);
- ctx->engine[engine->id].lrc_vma = NULL;
- ctx->engine[engine->id].lrc_desc = 0;
- ctx->engine[engine->id].lrc_reg_state = NULL;
+ ce->lrc_vma = NULL;
+ ce->lrc_desc = 0;
+ ce->lrc_reg_state = NULL;
- i915_gem_context_unreference(ctx);
- }
+ i915_gem_context_put(ctx);
}
static int intel_logical_ring_workarounds_emit(struct drm_i915_gem_request *req)
{
int ret, i;
- struct intel_engine_cs *engine = req->engine;
- struct intel_ringbuffer *ringbuf = req->ringbuf;
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct i915_workarounds *w = &dev_priv->workarounds;
+ struct intel_ring *ring = req->ring;
+ struct i915_workarounds *w = &req->i915->workarounds;
if (w->count == 0)
return 0;
- engine->gpu_caches_dirty = true;
- ret = logical_ring_flush_all_caches(req);
+ ret = req->engine->emit_flush(req, EMIT_BARRIER);
if (ret)
return ret;
if (ret)
return ret;
- intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(w->count));
+ intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(w->count));
for (i = 0; i < w->count; i++) {
- intel_logical_ring_emit_reg(ringbuf, w->reg[i].addr);
- intel_logical_ring_emit(ringbuf, w->reg[i].value);
+ intel_ring_emit_reg(ring, w->reg[i].addr);
+ intel_ring_emit(ring, w->reg[i].value);
}
- intel_logical_ring_emit(ringbuf, MI_NOOP);
+ intel_ring_emit(ring, MI_NOOP);
- intel_logical_ring_advance(ringbuf);
+ intel_ring_advance(ring);
- engine->gpu_caches_dirty = true;
- ret = logical_ring_flush_all_caches(req);
+ ret = req->engine->emit_flush(req, EMIT_BARRIER);
if (ret)
return ret;
* code duplication.
*/
static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine,
- uint32_t *const batch,
+ uint32_t *batch,
uint32_t index)
{
- struct drm_i915_private *dev_priv = engine->dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
uint32_t l3sqc4_flush = (0x40400000 | GEN8_LQSC_FLUSH_COHERENT_LINES);
/*
return 0;
}
-/**
- * gen8_init_indirectctx_bb() - initialize indirect ctx batch with WA
- *
- * @ring: only applicable for RCS
- * @wa_ctx: structure representing wa_ctx
- * offset: specifies start of the batch, should be cache-aligned. This is updated
- * with the offset value received as input.
- * size: size of the batch in DWORDS but HW expects in terms of cachelines
- * @batch: page in which WA are loaded
- * @offset: This field specifies the start of the batch, it should be
- * cache-aligned otherwise it is adjusted accordingly.
- * Typically we only have one indirect_ctx and per_ctx batch buffer which are
- * initialized at the beginning and shared across all contexts but this field
- * helps us to have multiple batches at different offsets and select them based
- * on a criteria. At the moment this batch always start at the beginning of the page
- * and at this point we don't have multiple wa_ctx batch buffers.
- *
- * The number of WA applied are not known at the beginning; we use this field
- * to return the no of DWORDS written.
+/*
+ * Typically we only have one indirect_ctx and per_ctx batch buffer which are
+ * initialized at the beginning and shared across all contexts but this field
+ * helps us to have multiple batches at different offsets and select them based
+ * on a criteria. At the moment this batch always start at the beginning of the page
+ * and at this point we don't have multiple wa_ctx batch buffers.
*
- * It is to be noted that this batch does not contain MI_BATCH_BUFFER_END
- * so it adds NOOPs as padding to make it cacheline aligned.
- * MI_BATCH_BUFFER_END will be added to perctx batch and both of them together
- * makes a complete batch buffer.
+ * The number of WA applied are not known at the beginning; we use this field
+ * to return the no of DWORDS written.
*
- * Return: non-zero if we exceed the PAGE_SIZE limit.
+ * It is to be noted that this batch does not contain MI_BATCH_BUFFER_END
+ * so it adds NOOPs as padding to make it cacheline aligned.
+ * MI_BATCH_BUFFER_END will be added to perctx batch and both of them together
+ * makes a complete batch buffer.
*/
-
static int gen8_init_indirectctx_bb(struct intel_engine_cs *engine,
struct i915_wa_ctx_bb *wa_ctx,
- uint32_t *const batch,
+ uint32_t *batch,
uint32_t *offset)
{
uint32_t scratch_addr;
wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
/* WaFlushCoherentL3CacheLinesAtContextSwitch:bdw */
- if (IS_BROADWELL(engine->dev)) {
+ if (IS_BROADWELL(engine->i915)) {
int rc = gen8_emit_flush_coherentl3_wa(engine, batch, index);
if (rc < 0)
return rc;
return wa_ctx_end(wa_ctx, *offset = index, CACHELINE_DWORDS);
}
-/**
- * gen8_init_perctx_bb() - initialize per ctx batch with WA
- *
- * @ring: only applicable for RCS
- * @wa_ctx: structure representing wa_ctx
- * offset: specifies start of the batch, should be cache-aligned.
- * size: size of the batch in DWORDS but HW expects in terms of cachelines
- * @batch: page in which WA are loaded
- * @offset: This field specifies the start of this batch.
- * This batch is started immediately after indirect_ctx batch. Since we ensure
- * that indirect_ctx ends on a cacheline this batch is aligned automatically.
+/*
+ * This batch is started immediately after indirect_ctx batch. Since we ensure
+ * that indirect_ctx ends on a cacheline this batch is aligned automatically.
*
- * The number of DWORDS written are returned using this field.
+ * The number of DWORDS written are returned using this field.
*
* This batch is terminated with MI_BATCH_BUFFER_END and so we need not add padding
* to align it with cacheline as padding after MI_BATCH_BUFFER_END is redundant.
*/
static int gen8_init_perctx_bb(struct intel_engine_cs *engine,
struct i915_wa_ctx_bb *wa_ctx,
- uint32_t *const batch,
+ uint32_t *batch,
uint32_t *offset)
{
uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
static int gen9_init_indirectctx_bb(struct intel_engine_cs *engine,
struct i915_wa_ctx_bb *wa_ctx,
- uint32_t *const batch,
+ uint32_t *batch,
uint32_t *offset)
{
int ret;
- struct drm_device *dev = engine->dev;
+ struct drm_i915_private *dev_priv = engine->i915;
uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
/* WaDisableCtxRestoreArbitration:skl,bxt */
- if (IS_SKL_REVID(dev, 0, SKL_REVID_D0) ||
- IS_BXT_REVID(dev, 0, BXT_REVID_A1))
+ if (IS_SKL_REVID(dev_priv, 0, SKL_REVID_D0) ||
+ IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1))
wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
/* WaFlushCoherentL3CacheLinesAtContextSwitch:skl,bxt */
return ret;
index = ret;
+ /* WaDisableGatherAtSetShaderCommonSlice:skl,bxt,kbl */
+ wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
+ wa_ctx_emit_reg(batch, index, COMMON_SLICE_CHICKEN2);
+ wa_ctx_emit(batch, index, _MASKED_BIT_DISABLE(
+ GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE));
+ wa_ctx_emit(batch, index, MI_NOOP);
+
+ /* WaClearSlmSpaceAtContextSwitch:kbl */
+ /* Actual scratch location is at 128 bytes offset */
+ if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0)) {
+ uint32_t scratch_addr
+ = engine->scratch.gtt_offset + 2*CACHELINE_BYTES;
+
+ wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6));
+ wa_ctx_emit(batch, index, (PIPE_CONTROL_FLUSH_L3 |
+ PIPE_CONTROL_GLOBAL_GTT_IVB |
+ PIPE_CONTROL_CS_STALL |
+ PIPE_CONTROL_QW_WRITE));
+ wa_ctx_emit(batch, index, scratch_addr);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ }
+
+ /* WaMediaPoolStateCmdInWABB:bxt */
+ if (HAS_POOLED_EU(engine->i915)) {
+ /*
+ * EU pool configuration is setup along with golden context
+ * during context initialization. This value depends on
+ * device type (2x6 or 3x6) and needs to be updated based
+ * on which subslice is disabled especially for 2x6
+ * devices, however it is safe to load default
+ * configuration of 3x6 device instead of masking off
+ * corresponding bits because HW ignores bits of a disabled
+ * subslice and drops down to appropriate config. Please
+ * see render_state_setup() in i915_gem_render_state.c for
+ * possible configurations, to avoid duplication they are
+ * not shown here again.
+ */
+ u32 eu_pool_config = 0x00777000;
+ wa_ctx_emit(batch, index, GEN9_MEDIA_POOL_STATE);
+ wa_ctx_emit(batch, index, GEN9_MEDIA_POOL_ENABLE);
+ wa_ctx_emit(batch, index, eu_pool_config);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ }
+
/* Pad to end of cacheline */
while (index % CACHELINE_DWORDS)
wa_ctx_emit(batch, index, MI_NOOP);
static int gen9_init_perctx_bb(struct intel_engine_cs *engine,
struct i915_wa_ctx_bb *wa_ctx,
- uint32_t *const batch,
+ uint32_t *batch,
uint32_t *offset)
{
- struct drm_device *dev = engine->dev;
uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
/* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
- if (IS_SKL_REVID(dev, 0, SKL_REVID_B0) ||
- IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
+ if (IS_SKL_REVID(engine->i915, 0, SKL_REVID_B0) ||
+ IS_BXT_REVID(engine->i915, 0, BXT_REVID_A1)) {
wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
wa_ctx_emit_reg(batch, index, GEN9_SLICE_COMMON_ECO_CHICKEN0);
wa_ctx_emit(batch, index,
}
/* WaClearTdlStateAckDirtyBits:bxt */
- if (IS_BXT_REVID(dev, 0, BXT_REVID_B0)) {
+ if (IS_BXT_REVID(engine->i915, 0, BXT_REVID_B0)) {
wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(4));
wa_ctx_emit_reg(batch, index, GEN8_STATE_ACK);
}
/* WaDisableCtxRestoreArbitration:skl,bxt */
- if (IS_SKL_REVID(dev, 0, SKL_REVID_D0) ||
- IS_BXT_REVID(dev, 0, BXT_REVID_A1))
+ if (IS_SKL_REVID(engine->i915, 0, SKL_REVID_D0) ||
+ IS_BXT_REVID(engine->i915, 0, BXT_REVID_A1))
wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_ENABLE);
wa_ctx_emit(batch, index, MI_BATCH_BUFFER_END);
{
int ret;
- engine->wa_ctx.obj = i915_gem_alloc_object(engine->dev,
- PAGE_ALIGN(size));
- if (!engine->wa_ctx.obj) {
+ engine->wa_ctx.obj = i915_gem_object_create(&engine->i915->drm,
+ PAGE_ALIGN(size));
+ if (IS_ERR(engine->wa_ctx.obj)) {
DRM_DEBUG_DRIVER("alloc LRC WA ctx backing obj failed.\n");
- return -ENOMEM;
+ ret = PTR_ERR(engine->wa_ctx.obj);
+ engine->wa_ctx.obj = NULL;
+ return ret;
}
- ret = i915_gem_obj_ggtt_pin(engine->wa_ctx.obj, PAGE_SIZE, 0);
+ ret = i915_gem_object_ggtt_pin(engine->wa_ctx.obj, NULL,
+ 0, PAGE_SIZE, 0);
if (ret) {
DRM_DEBUG_DRIVER("pin LRC WA ctx backing obj failed: %d\n",
ret);
- drm_gem_object_unreference(&engine->wa_ctx.obj->base);
+ i915_gem_object_put(engine->wa_ctx.obj);
return ret;
}
{
if (engine->wa_ctx.obj) {
i915_gem_object_ggtt_unpin(engine->wa_ctx.obj);
- drm_gem_object_unreference(&engine->wa_ctx.obj->base);
+ i915_gem_object_put(engine->wa_ctx.obj);
engine->wa_ctx.obj = NULL;
}
}
WARN_ON(engine->id != RCS);
/* update this when WA for higher Gen are added */
- if (INTEL_INFO(engine->dev)->gen > 9) {
+ if (INTEL_GEN(engine->i915) > 9) {
DRM_ERROR("WA batch buffer is not initialized for Gen%d\n",
- INTEL_INFO(engine->dev)->gen);
+ INTEL_GEN(engine->i915));
return 0;
}
batch = kmap_atomic(page);
offset = 0;
- if (INTEL_INFO(engine->dev)->gen == 8) {
+ if (IS_GEN8(engine->i915)) {
ret = gen8_init_indirectctx_bb(engine,
&wa_ctx->indirect_ctx,
batch,
&offset);
if (ret)
goto out;
- } else if (INTEL_INFO(engine->dev)->gen == 9) {
+ } else if (IS_GEN9(engine->i915)) {
ret = gen9_init_indirectctx_bb(engine,
&wa_ctx->indirect_ctx,
batch,
static void lrc_init_hws(struct intel_engine_cs *engine)
{
- struct drm_i915_private *dev_priv = engine->dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
I915_WRITE(RING_HWS_PGA(engine->mmio_base),
(u32)engine->status_page.gfx_addr);
static int gen8_init_common_ring(struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
unsigned int next_context_status_buffer_hw;
lrc_init_hws(engine);
static int gen8_init_render_ring(struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = engine->i915;
int ret;
ret = gen8_init_common_ring(engine);
static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req)
{
struct i915_hw_ppgtt *ppgtt = req->ctx->ppgtt;
+ struct intel_ring *ring = req->ring;
struct intel_engine_cs *engine = req->engine;
- struct intel_ringbuffer *ringbuf = req->ringbuf;
const int num_lri_cmds = GEN8_LEGACY_PDPES * 2;
int i, ret;
if (ret)
return ret;
- intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(num_lri_cmds));
+ intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_lri_cmds));
for (i = GEN8_LEGACY_PDPES - 1; i >= 0; i--) {
const dma_addr_t pd_daddr = i915_page_dir_dma_addr(ppgtt, i);
- intel_logical_ring_emit_reg(ringbuf,
- GEN8_RING_PDP_UDW(engine, i));
- intel_logical_ring_emit(ringbuf, upper_32_bits(pd_daddr));
- intel_logical_ring_emit_reg(ringbuf,
- GEN8_RING_PDP_LDW(engine, i));
- intel_logical_ring_emit(ringbuf, lower_32_bits(pd_daddr));
+ intel_ring_emit_reg(ring, GEN8_RING_PDP_UDW(engine, i));
+ intel_ring_emit(ring, upper_32_bits(pd_daddr));
+ intel_ring_emit_reg(ring, GEN8_RING_PDP_LDW(engine, i));
+ intel_ring_emit(ring, lower_32_bits(pd_daddr));
}
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- intel_logical_ring_advance(ringbuf);
+ intel_ring_emit(ring, MI_NOOP);
+ intel_ring_advance(ring);
return 0;
}
static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
- u64 offset, unsigned dispatch_flags)
+ u64 offset, u32 len,
+ unsigned int dispatch_flags)
{
- struct intel_ringbuffer *ringbuf = req->ringbuf;
+ struct intel_ring *ring = req->ring;
bool ppgtt = !(dispatch_flags & I915_DISPATCH_SECURE);
int ret;
if (req->ctx->ppgtt &&
(intel_engine_flag(req->engine) & req->ctx->ppgtt->pd_dirty_rings)) {
if (!USES_FULL_48BIT_PPGTT(req->i915) &&
- !intel_vgpu_active(req->i915->dev)) {
+ !intel_vgpu_active(req->i915)) {
ret = intel_logical_ring_emit_pdps(req);
if (ret)
return ret;
return ret;
/* FIXME(BDW): Address space and security selectors. */
- intel_logical_ring_emit(ringbuf, MI_BATCH_BUFFER_START_GEN8 |
- (ppgtt<<8) |
- (dispatch_flags & I915_DISPATCH_RS ?
- MI_BATCH_RESOURCE_STREAMER : 0));
- intel_logical_ring_emit(ringbuf, lower_32_bits(offset));
- intel_logical_ring_emit(ringbuf, upper_32_bits(offset));
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- intel_logical_ring_advance(ringbuf);
+ intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 |
+ (ppgtt<<8) |
+ (dispatch_flags & I915_DISPATCH_RS ?
+ MI_BATCH_RESOURCE_STREAMER : 0));
+ intel_ring_emit(ring, lower_32_bits(offset));
+ intel_ring_emit(ring, upper_32_bits(offset));
+ intel_ring_emit(ring, MI_NOOP);
+ intel_ring_advance(ring);
return 0;
}
-static bool gen8_logical_ring_get_irq(struct intel_engine_cs *engine)
+static void gen8_logical_ring_enable_irq(struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- unsigned long flags;
-
- if (WARN_ON(!intel_irqs_enabled(dev_priv)))
- return false;
-
- spin_lock_irqsave(&dev_priv->irq_lock, flags);
- if (engine->irq_refcount++ == 0) {
- I915_WRITE_IMR(engine,
- ~(engine->irq_enable_mask | engine->irq_keep_mask));
- POSTING_READ(RING_IMR(engine->mmio_base));
- }
- spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
-
- return true;
+ struct drm_i915_private *dev_priv = engine->i915;
+ I915_WRITE_IMR(engine,
+ ~(engine->irq_enable_mask | engine->irq_keep_mask));
+ POSTING_READ_FW(RING_IMR(engine->mmio_base));
}
-static void gen8_logical_ring_put_irq(struct intel_engine_cs *engine)
+static void gen8_logical_ring_disable_irq(struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- unsigned long flags;
-
- spin_lock_irqsave(&dev_priv->irq_lock, flags);
- if (--engine->irq_refcount == 0) {
- I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
- POSTING_READ(RING_IMR(engine->mmio_base));
- }
- spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+ struct drm_i915_private *dev_priv = engine->i915;
+ I915_WRITE_IMR(engine, ~engine->irq_keep_mask);
}
-static int gen8_emit_flush(struct drm_i915_gem_request *request,
- u32 invalidate_domains,
- u32 unused)
+static int gen8_emit_flush(struct drm_i915_gem_request *request, u32 mode)
{
- struct intel_ringbuffer *ringbuf = request->ringbuf;
- struct intel_engine_cs *engine = ringbuf->engine;
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- uint32_t cmd;
+ struct intel_ring *ring = request->ring;
+ u32 cmd;
int ret;
ret = intel_ring_begin(request, 4);
*/
cmd |= MI_FLUSH_DW_STORE_INDEX | MI_FLUSH_DW_OP_STOREDW;
- if (invalidate_domains & I915_GEM_GPU_DOMAINS) {
+ if (mode & EMIT_INVALIDATE) {
cmd |= MI_INVALIDATE_TLB;
- if (engine == &dev_priv->engine[VCS])
+ if (request->engine->id == VCS)
cmd |= MI_INVALIDATE_BSD;
}
- intel_logical_ring_emit(ringbuf, cmd);
- intel_logical_ring_emit(ringbuf,
- I915_GEM_HWS_SCRATCH_ADDR |
- MI_FLUSH_DW_USE_GTT);
- intel_logical_ring_emit(ringbuf, 0); /* upper addr */
- intel_logical_ring_emit(ringbuf, 0); /* value */
- intel_logical_ring_advance(ringbuf);
+ intel_ring_emit(ring, cmd);
+ intel_ring_emit(ring,
+ I915_GEM_HWS_SCRATCH_ADDR |
+ MI_FLUSH_DW_USE_GTT);
+ intel_ring_emit(ring, 0); /* upper addr */
+ intel_ring_emit(ring, 0); /* value */
+ intel_ring_advance(ring);
return 0;
}
static int gen8_emit_flush_render(struct drm_i915_gem_request *request,
- u32 invalidate_domains,
- u32 flush_domains)
+ u32 mode)
{
- struct intel_ringbuffer *ringbuf = request->ringbuf;
- struct intel_engine_cs *engine = ringbuf->engine;
+ struct intel_ring *ring = request->ring;
+ struct intel_engine_cs *engine = request->engine;
u32 scratch_addr = engine->scratch.gtt_offset + 2 * CACHELINE_BYTES;
- bool vf_flush_wa = false;
+ bool vf_flush_wa = false, dc_flush_wa = false;
u32 flags = 0;
int ret;
+ int len;
flags |= PIPE_CONTROL_CS_STALL;
- if (flush_domains) {
+ if (mode & EMIT_FLUSH) {
flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
flags |= PIPE_CONTROL_DC_FLUSH_ENABLE;
flags |= PIPE_CONTROL_FLUSH_ENABLE;
}
- if (invalidate_domains) {
+ if (mode & EMIT_INVALIDATE) {
flags |= PIPE_CONTROL_TLB_INVALIDATE;
flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
* On GEN9: before VF_CACHE_INVALIDATE we need to emit a NULL
* pipe control.
*/
- if (IS_GEN9(engine->dev))
+ if (IS_GEN9(request->i915))
vf_flush_wa = true;
+
+ /* WaForGAMHang:kbl */
+ if (IS_KBL_REVID(request->i915, 0, KBL_REVID_B0))
+ dc_flush_wa = true;
}
- ret = intel_ring_begin(request, vf_flush_wa ? 12 : 6);
+ len = 6;
+
+ if (vf_flush_wa)
+ len += 6;
+
+ if (dc_flush_wa)
+ len += 12;
+
+ ret = intel_ring_begin(request, len);
if (ret)
return ret;
if (vf_flush_wa) {
- intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
+ intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
}
- intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
- intel_logical_ring_emit(ringbuf, flags);
- intel_logical_ring_emit(ringbuf, scratch_addr);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_advance(ringbuf);
+ if (dc_flush_wa) {
+ intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+ intel_ring_emit(ring, PIPE_CONTROL_DC_FLUSH_ENABLE);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ }
- return 0;
-}
+ intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+ intel_ring_emit(ring, flags);
+ intel_ring_emit(ring, scratch_addr);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
-static u32 gen8_get_seqno(struct intel_engine_cs *engine)
-{
- return intel_read_status_page(engine, I915_GEM_HWS_INDEX);
-}
+ if (dc_flush_wa) {
+ intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+ intel_ring_emit(ring, PIPE_CONTROL_CS_STALL);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, 0);
+ }
-static void gen8_set_seqno(struct intel_engine_cs *engine, u32 seqno)
-{
- intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
+ intel_ring_advance(ring);
+
+ return 0;
}
static void bxt_a_seqno_barrier(struct intel_engine_cs *engine)
intel_flush_status_page(engine, I915_GEM_HWS_INDEX);
}
-static void bxt_a_set_seqno(struct intel_engine_cs *engine, u32 seqno)
-{
- intel_write_status_page(engine, I915_GEM_HWS_INDEX, seqno);
-
- /* See bxt_a_get_seqno() explaining the reason for the clflush. */
- intel_flush_status_page(engine, I915_GEM_HWS_INDEX);
-}
-
/*
* Reserve space for 2 NOOPs at the end of each request to be
* used as a workaround for not being allowed to do lite
*/
#define WA_TAIL_DWORDS 2
-static inline u32 hws_seqno_address(struct intel_engine_cs *engine)
-{
- return engine->status_page.gfx_addr + I915_GEM_HWS_INDEX_ADDR;
-}
-
static int gen8_emit_request(struct drm_i915_gem_request *request)
{
- struct intel_ringbuffer *ringbuf = request->ringbuf;
+ struct intel_ring *ring = request->ring;
int ret;
ret = intel_ring_begin(request, 6 + WA_TAIL_DWORDS);
/* w/a: bit 5 needs to be zero for MI_FLUSH_DW address. */
BUILD_BUG_ON(I915_GEM_HWS_INDEX_ADDR & (1 << 5));
- intel_logical_ring_emit(ringbuf,
- (MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
- intel_logical_ring_emit(ringbuf,
- hws_seqno_address(request->engine) |
- MI_FLUSH_DW_USE_GTT);
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, i915_gem_request_get_seqno(request));
- intel_logical_ring_emit(ringbuf, MI_USER_INTERRUPT);
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- return intel_logical_ring_advance_and_submit(request);
+ intel_ring_emit(ring, (MI_FLUSH_DW + 1) | MI_FLUSH_DW_OP_STOREDW);
+ intel_ring_emit(ring,
+ intel_hws_seqno_address(request->engine) |
+ MI_FLUSH_DW_USE_GTT);
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, request->fence.seqno);
+ intel_ring_emit(ring, MI_USER_INTERRUPT);
+ intel_ring_emit(ring, MI_NOOP);
+ return intel_logical_ring_advance(request);
}
static int gen8_emit_request_render(struct drm_i915_gem_request *request)
{
- struct intel_ringbuffer *ringbuf = request->ringbuf;
+ struct intel_ring *ring = request->ring;
int ret;
ret = intel_ring_begin(request, 8 + WA_TAIL_DWORDS);
* need a prior CS_STALL, which is emitted by the flush
* following the batch.
*/
- intel_logical_ring_emit(ringbuf, GFX_OP_PIPE_CONTROL(6));
- intel_logical_ring_emit(ringbuf,
- (PIPE_CONTROL_GLOBAL_GTT_IVB |
- PIPE_CONTROL_CS_STALL |
- PIPE_CONTROL_QW_WRITE));
- intel_logical_ring_emit(ringbuf, hws_seqno_address(request->engine));
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, i915_gem_request_get_seqno(request));
+ intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(6));
+ intel_ring_emit(ring,
+ (PIPE_CONTROL_GLOBAL_GTT_IVB |
+ PIPE_CONTROL_CS_STALL |
+ PIPE_CONTROL_QW_WRITE));
+ intel_ring_emit(ring, intel_hws_seqno_address(request->engine));
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, i915_gem_request_get_seqno(request));
/* We're thrashing one dword of HWS. */
- intel_logical_ring_emit(ringbuf, 0);
- intel_logical_ring_emit(ringbuf, MI_USER_INTERRUPT);
- intel_logical_ring_emit(ringbuf, MI_NOOP);
- return intel_logical_ring_advance_and_submit(request);
-}
-
-static int intel_lr_context_render_state_init(struct drm_i915_gem_request *req)
-{
- struct render_state so;
- int ret;
-
- ret = i915_gem_render_state_prepare(req->engine, &so);
- if (ret)
- return ret;
-
- if (so.rodata == NULL)
- return 0;
-
- ret = req->engine->emit_bb_start(req, so.ggtt_offset,
- I915_DISPATCH_SECURE);
- if (ret)
- goto out;
-
- ret = req->engine->emit_bb_start(req,
- (so.ggtt_offset + so.aux_batch_offset),
- I915_DISPATCH_SECURE);
- if (ret)
- goto out;
-
- i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req);
-
-out:
- i915_gem_render_state_fini(&so);
- return ret;
+ intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, MI_USER_INTERRUPT);
+ intel_ring_emit(ring, MI_NOOP);
+ return intel_logical_ring_advance(request);
}
static int gen8_init_rcs_context(struct drm_i915_gem_request *req)
if (ret)
DRM_ERROR("MOCS failed to program: expect performance issues.\n");
- return intel_lr_context_render_state_init(req);
+ return i915_gem_render_state_init(req);
}
/**
* intel_logical_ring_cleanup() - deallocate the Engine Command Streamer
- *
- * @ring: Engine Command Streamer.
- *
+ * @engine: Engine Command Streamer.
*/
void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
{
if (WARN_ON(test_bit(TASKLET_STATE_SCHED, &engine->irq_tasklet.state)))
tasklet_kill(&engine->irq_tasklet);
- dev_priv = engine->dev->dev_private;
+ dev_priv = engine->i915;
if (engine->buffer) {
- intel_logical_ring_stop(engine);
WARN_ON((I915_READ_MODE(engine) & MODE_IDLE) == 0);
}
if (engine->cleanup)
engine->cleanup(engine);
- i915_cmd_parser_fini_ring(engine);
- i915_gem_batch_pool_fini(&engine->batch_pool);
+ intel_engine_cleanup_common(engine);
if (engine->status_page.obj) {
i915_gem_object_unpin_map(engine->status_page.obj);
engine->status_page.obj = NULL;
}
+ intel_lr_context_unpin(dev_priv->kernel_context, engine);
engine->idle_lite_restore_wa = 0;
engine->disable_lite_restore_wa = false;
engine->ctx_desc_template = 0;
lrc_destroy_wa_ctx_obj(engine);
- engine->dev = NULL;
+ engine->i915 = NULL;
+}
+
+void intel_execlists_enable_submission(struct drm_i915_private *dev_priv)
+{
+ struct intel_engine_cs *engine;
+
+ for_each_engine(engine, dev_priv)
+ engine->submit_request = execlists_submit_request;
}
static void
-logical_ring_default_vfuncs(struct drm_device *dev,
- struct intel_engine_cs *engine)
+logical_ring_default_vfuncs(struct intel_engine_cs *engine)
{
/* Default vfuncs which can be overriden by each engine. */
engine->init_hw = gen8_init_common_ring;
- engine->emit_request = gen8_emit_request;
engine->emit_flush = gen8_emit_flush;
- engine->irq_get = gen8_logical_ring_get_irq;
- engine->irq_put = gen8_logical_ring_put_irq;
+ engine->emit_request = gen8_emit_request;
+ engine->submit_request = execlists_submit_request;
+
+ engine->irq_enable = gen8_logical_ring_enable_irq;
+ engine->irq_disable = gen8_logical_ring_disable_irq;
engine->emit_bb_start = gen8_emit_bb_start;
- engine->get_seqno = gen8_get_seqno;
- engine->set_seqno = gen8_set_seqno;
- if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
+ if (IS_BXT_REVID(engine->i915, 0, BXT_REVID_A1))
engine->irq_seqno_barrier = bxt_a_seqno_barrier;
- engine->set_seqno = bxt_a_set_seqno;
- }
}
static inline void
-logical_ring_default_irqs(struct intel_engine_cs *engine, unsigned shift)
+logical_ring_default_irqs(struct intel_engine_cs *engine)
{
+ unsigned shift = engine->irq_shift;
engine->irq_enable_mask = GT_RENDER_USER_INTERRUPT << shift;
engine->irq_keep_mask = GT_CONTEXT_SWITCH_INTERRUPT << shift;
}
return 0;
}
-static int
-logical_ring_init(struct drm_device *dev, struct intel_engine_cs *engine)
+static void
+logical_ring_setup(struct intel_engine_cs *engine)
{
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct intel_context *dctx = dev_priv->kernel_context;
+ struct drm_i915_private *dev_priv = engine->i915;
enum forcewake_domains fw_domains;
- int ret;
+
+ intel_engine_setup_common(engine);
/* Intentionally left blank. */
engine->buffer = NULL;
- engine->dev = dev;
- INIT_LIST_HEAD(&engine->active_list);
- INIT_LIST_HEAD(&engine->request_list);
- i915_gem_batch_pool_init(dev, &engine->batch_pool);
- init_waitqueue_head(&engine->irq_queue);
-
- INIT_LIST_HEAD(&engine->buffers);
- INIT_LIST_HEAD(&engine->execlist_queue);
- INIT_LIST_HEAD(&engine->execlist_retired_req_list);
- spin_lock_init(&engine->execlist_lock);
-
- tasklet_init(&engine->irq_tasklet,
- intel_lrc_irq_handler, (unsigned long)engine);
-
- logical_ring_init_platform_invariants(engine);
-
fw_domains = intel_uncore_forcewake_for_reg(dev_priv,
RING_ELSP(engine),
FW_REG_WRITE);
engine->fw_domains = fw_domains;
- ret = i915_cmd_parser_init_ring(engine);
+ tasklet_init(&engine->irq_tasklet,
+ intel_lrc_irq_handler, (unsigned long)engine);
+
+ logical_ring_init_platform_invariants(engine);
+ logical_ring_default_vfuncs(engine);
+ logical_ring_default_irqs(engine);
+}
+
+static int
+logical_ring_init(struct intel_engine_cs *engine)
+{
+ struct i915_gem_context *dctx = engine->i915->kernel_context;
+ int ret;
+
+ ret = intel_engine_init_common(engine);
if (ret)
goto error;
- ret = intel_lr_context_deferred_alloc(dctx, engine);
+ ret = execlists_context_deferred_alloc(dctx, engine);
if (ret)
goto error;
/* As this is the default context, always pin it */
- ret = intel_lr_context_do_pin(dctx, engine);
+ ret = intel_lr_context_pin(dctx, engine);
if (ret) {
- DRM_ERROR(
- "Failed to pin and map ringbuffer %s: %d\n",
- engine->name, ret);
+ DRM_ERROR("Failed to pin context for %s: %d\n",
+ engine->name, ret);
goto error;
}
return ret;
}
-static int logical_render_ring_init(struct drm_device *dev)
+int logical_render_ring_init(struct intel_engine_cs *engine)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine = &dev_priv->engine[RCS];
+ struct drm_i915_private *dev_priv = engine->i915;
int ret;
- engine->name = "render ring";
- engine->id = RCS;
- engine->exec_id = I915_EXEC_RENDER;
- engine->guc_id = GUC_RENDER_ENGINE;
- engine->mmio_base = RENDER_RING_BASE;
+ logical_ring_setup(engine);
- logical_ring_default_irqs(engine, GEN8_RCS_IRQ_SHIFT);
- if (HAS_L3_DPF(dev))
+ if (HAS_L3_DPF(dev_priv))
engine->irq_keep_mask |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
- logical_ring_default_vfuncs(dev, engine);
-
/* Override some for render ring. */
- if (INTEL_INFO(dev)->gen >= 9)
+ if (INTEL_GEN(dev_priv) >= 9)
engine->init_hw = gen9_init_render_ring;
else
engine->init_hw = gen8_init_render_ring;
engine->emit_flush = gen8_emit_flush_render;
engine->emit_request = gen8_emit_request_render;
- engine->dev = dev;
-
- ret = intel_init_pipe_control(engine);
+ ret = intel_init_pipe_control(engine, 4096);
if (ret)
return ret;
ret);
}
- ret = logical_ring_init(dev, engine);
+ ret = logical_ring_init(engine);
if (ret) {
lrc_destroy_wa_ctx_obj(engine);
}
return ret;
}
-static int logical_bsd_ring_init(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine = &dev_priv->engine[VCS];
-
- engine->name = "bsd ring";
- engine->id = VCS;
- engine->exec_id = I915_EXEC_BSD;
- engine->guc_id = GUC_VIDEO_ENGINE;
- engine->mmio_base = GEN6_BSD_RING_BASE;
-
- logical_ring_default_irqs(engine, GEN8_VCS1_IRQ_SHIFT);
- logical_ring_default_vfuncs(dev, engine);
-
- return logical_ring_init(dev, engine);
-}
-
-static int logical_bsd2_ring_init(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine = &dev_priv->engine[VCS2];
-
- engine->name = "bsd2 ring";
- engine->id = VCS2;
- engine->exec_id = I915_EXEC_BSD;
- engine->guc_id = GUC_VIDEO_ENGINE2;
- engine->mmio_base = GEN8_BSD2_RING_BASE;
-
- logical_ring_default_irqs(engine, GEN8_VCS2_IRQ_SHIFT);
- logical_ring_default_vfuncs(dev, engine);
-
- return logical_ring_init(dev, engine);
-}
-
-static int logical_blt_ring_init(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine = &dev_priv->engine[BCS];
-
- engine->name = "blitter ring";
- engine->id = BCS;
- engine->exec_id = I915_EXEC_BLT;
- engine->guc_id = GUC_BLITTER_ENGINE;
- engine->mmio_base = BLT_RING_BASE;
-
- logical_ring_default_irqs(engine, GEN8_BCS_IRQ_SHIFT);
- logical_ring_default_vfuncs(dev, engine);
-
- return logical_ring_init(dev, engine);
-}
-
-static int logical_vebox_ring_init(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine = &dev_priv->engine[VECS];
-
- engine->name = "video enhancement ring";
- engine->id = VECS;
- engine->exec_id = I915_EXEC_VEBOX;
- engine->guc_id = GUC_VIDEOENHANCE_ENGINE;
- engine->mmio_base = VEBOX_RING_BASE;
-
- logical_ring_default_irqs(engine, GEN8_VECS_IRQ_SHIFT);
- logical_ring_default_vfuncs(dev, engine);
-
- return logical_ring_init(dev, engine);
-}
-
-/**
- * intel_logical_rings_init() - allocate, populate and init the Engine Command Streamers
- * @dev: DRM device.
- *
- * This function inits the engines for an Execlists submission style (the equivalent in the
- * legacy ringbuffer submission world would be i915_gem_init_engines). It does it only for
- * those engines that are present in the hardware.
- *
- * Return: non-zero if the initialization failed.
- */
-int intel_logical_rings_init(struct drm_device *dev)
+int logical_xcs_ring_init(struct intel_engine_cs *engine)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
- int ret;
-
- ret = logical_render_ring_init(dev);
- if (ret)
- return ret;
-
- if (HAS_BSD(dev)) {
- ret = logical_bsd_ring_init(dev);
- if (ret)
- goto cleanup_render_ring;
- }
-
- if (HAS_BLT(dev)) {
- ret = logical_blt_ring_init(dev);
- if (ret)
- goto cleanup_bsd_ring;
- }
-
- if (HAS_VEBOX(dev)) {
- ret = logical_vebox_ring_init(dev);
- if (ret)
- goto cleanup_blt_ring;
- }
-
- if (HAS_BSD2(dev)) {
- ret = logical_bsd2_ring_init(dev);
- if (ret)
- goto cleanup_vebox_ring;
- }
+ logical_ring_setup(engine);
- return 0;
-
-cleanup_vebox_ring:
- intel_logical_ring_cleanup(&dev_priv->engine[VECS]);
-cleanup_blt_ring:
- intel_logical_ring_cleanup(&dev_priv->engine[BCS]);
-cleanup_bsd_ring:
- intel_logical_ring_cleanup(&dev_priv->engine[VCS]);
-cleanup_render_ring:
- intel_logical_ring_cleanup(&dev_priv->engine[RCS]);
-
- return ret;
+ return logical_ring_init(engine);
}
static u32
-make_rpcs(struct drm_device *dev)
+make_rpcs(struct drm_i915_private *dev_priv)
{
u32 rpcs = 0;
* No explicit RPCS request is needed to ensure full
* slice/subslice/EU enablement prior to Gen9.
*/
- if (INTEL_INFO(dev)->gen < 9)
+ if (INTEL_GEN(dev_priv) < 9)
return 0;
/*
* must make an explicit request through RPCS for full
* enablement.
*/
- if (INTEL_INFO(dev)->has_slice_pg) {
+ if (INTEL_INFO(dev_priv)->has_slice_pg) {
rpcs |= GEN8_RPCS_S_CNT_ENABLE;
- rpcs |= INTEL_INFO(dev)->slice_total <<
+ rpcs |= INTEL_INFO(dev_priv)->slice_total <<
GEN8_RPCS_S_CNT_SHIFT;
rpcs |= GEN8_RPCS_ENABLE;
}
- if (INTEL_INFO(dev)->has_subslice_pg) {
+ if (INTEL_INFO(dev_priv)->has_subslice_pg) {
rpcs |= GEN8_RPCS_SS_CNT_ENABLE;
- rpcs |= INTEL_INFO(dev)->subslice_per_slice <<
+ rpcs |= INTEL_INFO(dev_priv)->subslice_per_slice <<
GEN8_RPCS_SS_CNT_SHIFT;
rpcs |= GEN8_RPCS_ENABLE;
}
- if (INTEL_INFO(dev)->has_eu_pg) {
- rpcs |= INTEL_INFO(dev)->eu_per_subslice <<
+ if (INTEL_INFO(dev_priv)->has_eu_pg) {
+ rpcs |= INTEL_INFO(dev_priv)->eu_per_subslice <<
GEN8_RPCS_EU_MIN_SHIFT;
- rpcs |= INTEL_INFO(dev)->eu_per_subslice <<
+ rpcs |= INTEL_INFO(dev_priv)->eu_per_subslice <<
GEN8_RPCS_EU_MAX_SHIFT;
rpcs |= GEN8_RPCS_ENABLE;
}
{
u32 indirect_ctx_offset;
- switch (INTEL_INFO(engine->dev)->gen) {
+ switch (INTEL_GEN(engine->i915)) {
default:
- MISSING_CASE(INTEL_INFO(engine->dev)->gen);
+ MISSING_CASE(INTEL_GEN(engine->i915));
/* fall through */
case 9:
indirect_ctx_offset =
}
static int
-populate_lr_context(struct intel_context *ctx,
+populate_lr_context(struct i915_gem_context *ctx,
struct drm_i915_gem_object *ctx_obj,
struct intel_engine_cs *engine,
- struct intel_ringbuffer *ringbuf)
+ struct intel_ring *ring)
{
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = ctx->i915;
struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
void *vaddr;
u32 *reg_state;
RING_CONTEXT_CONTROL(engine),
_MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |
CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |
- (HAS_RESOURCE_STREAMER(dev) ?
+ (HAS_RESOURCE_STREAMER(dev_priv) ?
CTX_CTRL_RS_CTX_ENABLE : 0)));
ASSIGN_CTX_REG(reg_state, CTX_RING_HEAD, RING_HEAD(engine->mmio_base),
0);
RING_START(engine->mmio_base), 0);
ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_CONTROL,
RING_CTL(engine->mmio_base),
- ((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID);
+ ((ring->size - PAGE_SIZE) & RING_NR_PAGES) | RING_VALID);
ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_U,
RING_BBADDR_UDW(engine->mmio_base), 0);
ASSIGN_CTX_REG(reg_state, CTX_BB_HEAD_L,
if (engine->id == RCS) {
reg_state[CTX_LRI_HEADER_2] = MI_LOAD_REGISTER_IMM(1);
ASSIGN_CTX_REG(reg_state, CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE,
- make_rpcs(dev));
+ make_rpcs(dev_priv));
}
i915_gem_object_unpin_map(ctx_obj);
return 0;
}
-/**
- * intel_lr_context_free() - free the LRC specific bits of a context
- * @ctx: the LR context to free.
- *
- * The real context freeing is done in i915_gem_context_free: this only
- * takes care of the bits that are LRC related: the per-engine backing
- * objects and the logical ringbuffer.
- */
-void intel_lr_context_free(struct intel_context *ctx)
-{
- int i;
-
- for (i = I915_NUM_ENGINES; --i >= 0; ) {
- struct intel_ringbuffer *ringbuf = ctx->engine[i].ringbuf;
- struct drm_i915_gem_object *ctx_obj = ctx->engine[i].state;
-
- if (!ctx_obj)
- continue;
-
- if (ctx == ctx->i915->kernel_context) {
- intel_unpin_ringbuffer_obj(ringbuf);
- i915_gem_object_ggtt_unpin(ctx_obj);
- i915_gem_object_unpin_map(ctx_obj);
- }
-
- WARN_ON(ctx->engine[i].pin_count);
- intel_ringbuffer_free(ringbuf);
- drm_gem_object_unreference(&ctx_obj->base);
- }
-}
-
/**
* intel_lr_context_size() - return the size of the context for an engine
- * @ring: which engine to find the context size for
+ * @engine: which engine to find the context size for
*
* Each engine may require a different amount of space for a context image,
* so when allocating (or copying) an image, this function can be used to
{
int ret = 0;
- WARN_ON(INTEL_INFO(engine->dev)->gen < 8);
+ WARN_ON(INTEL_GEN(engine->i915) < 8);
switch (engine->id) {
case RCS:
- if (INTEL_INFO(engine->dev)->gen >= 9)
+ if (INTEL_GEN(engine->i915) >= 9)
ret = GEN9_LR_CONTEXT_RENDER_SIZE;
else
ret = GEN8_LR_CONTEXT_RENDER_SIZE;
return ret;
}
-/**
- * intel_lr_context_deferred_alloc() - create the LRC specific bits of a context
- * @ctx: LR context to create.
- * @ring: engine to be used with the context.
- *
- * This function can be called more than once, with different engines, if we plan
- * to use the context with them. The context backing objects and the ringbuffers
- * (specially the ringbuffer backing objects) suck a lot of memory up, and that's why
- * the creation is a deferred call: it's better to make sure first that we need to use
- * a given ring with the context.
- *
- * Return: non-zero on error.
- */
-
-int intel_lr_context_deferred_alloc(struct intel_context *ctx,
- struct intel_engine_cs *engine)
+static int execlists_context_deferred_alloc(struct i915_gem_context *ctx,
+ struct intel_engine_cs *engine)
{
- struct drm_device *dev = engine->dev;
struct drm_i915_gem_object *ctx_obj;
+ struct intel_context *ce = &ctx->engine[engine->id];
uint32_t context_size;
- struct intel_ringbuffer *ringbuf;
+ struct intel_ring *ring;
int ret;
- WARN_ON(ctx->legacy_hw_ctx.rcs_state != NULL);
- WARN_ON(ctx->engine[engine->id].state);
+ WARN_ON(ce->state);
context_size = round_up(intel_lr_context_size(engine), 4096);
/* One extra page as the sharing data between driver and GuC */
context_size += PAGE_SIZE * LRC_PPHWSP_PN;
- ctx_obj = i915_gem_alloc_object(dev, context_size);
- if (!ctx_obj) {
+ ctx_obj = i915_gem_object_create(&ctx->i915->drm, context_size);
+ if (IS_ERR(ctx_obj)) {
DRM_DEBUG_DRIVER("Alloc LRC backing obj failed.\n");
- return -ENOMEM;
+ return PTR_ERR(ctx_obj);
}
- ringbuf = intel_engine_create_ringbuffer(engine, 4 * PAGE_SIZE);
- if (IS_ERR(ringbuf)) {
- ret = PTR_ERR(ringbuf);
+ ring = intel_engine_create_ring(engine, ctx->ring_size);
+ if (IS_ERR(ring)) {
+ ret = PTR_ERR(ring);
goto error_deref_obj;
}
- ret = populate_lr_context(ctx, ctx_obj, engine, ringbuf);
+ ret = populate_lr_context(ctx, ctx_obj, engine, ring);
if (ret) {
DRM_DEBUG_DRIVER("Failed to populate LRC: %d\n", ret);
- goto error_ringbuf;
+ goto error_ring_free;
}
- ctx->engine[engine->id].ringbuf = ringbuf;
- ctx->engine[engine->id].state = ctx_obj;
+ ce->ring = ring;
+ ce->state = ctx_obj;
+ ce->initialised = engine->init_context == NULL;
- if (ctx != ctx->i915->kernel_context && engine->init_context) {
- struct drm_i915_gem_request *req;
-
- req = i915_gem_request_alloc(engine, ctx);
- if (IS_ERR(req)) {
- ret = PTR_ERR(req);
- DRM_ERROR("ring create req: %d\n", ret);
- goto error_ringbuf;
- }
-
- ret = engine->init_context(req);
- i915_add_request_no_flush(req);
- if (ret) {
- DRM_ERROR("ring init context: %d\n",
- ret);
- goto error_ringbuf;
- }
- }
return 0;
-error_ringbuf:
- intel_ringbuffer_free(ringbuf);
+error_ring_free:
+ intel_ring_free(ring);
error_deref_obj:
- drm_gem_object_unreference(&ctx_obj->base);
- ctx->engine[engine->id].ringbuf = NULL;
- ctx->engine[engine->id].state = NULL;
+ i915_gem_object_put(ctx_obj);
+ ce->ring = NULL;
+ ce->state = NULL;
return ret;
}
void intel_lr_context_reset(struct drm_i915_private *dev_priv,
- struct intel_context *ctx)
+ struct i915_gem_context *ctx)
{
struct intel_engine_cs *engine;
for_each_engine(engine, dev_priv) {
- struct drm_i915_gem_object *ctx_obj =
- ctx->engine[engine->id].state;
- struct intel_ringbuffer *ringbuf =
- ctx->engine[engine->id].ringbuf;
+ struct intel_context *ce = &ctx->engine[engine->id];
+ struct drm_i915_gem_object *ctx_obj = ce->state;
void *vaddr;
uint32_t *reg_state;
i915_gem_object_unpin_map(ctx_obj);
- ringbuf->head = 0;
- ringbuf->tail = 0;
+ ce->ring->head = 0;
+ ce->ring->tail = 0;
}
}
projects / karo-tx-linux.git / blobdiff