[PATCH 16/46] drm/i915: Move common active lists from engine to i915_scheduler
Chris Wilson
chris at chris-wilson.co.uk
Sat Jan 30 12:52:07 UTC 2021
Extract the scheduler lists into a related structure, stop sprawling
over struct intel_engine_cs. Also transfer the responsibility of tracing
from ENGINE_TRACE() to SCHED_TRACE().
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
drivers/gpu/drm/i915/gt/intel_engine_cs.c | 30 ++------
drivers/gpu/drm/i915/gt/intel_engine_types.h | 8 +--
.../drm/i915/gt/intel_execlists_submission.c | 7 +-
drivers/gpu/drm/i915/gt/mock_engine.c | 6 +-
drivers/gpu/drm/i915/i915_request.h | 8 ++-
drivers/gpu/drm/i915/i915_scheduler.c | 72 +++++++++++++++----
drivers/gpu/drm/i915/i915_scheduler.h | 13 +++-
drivers/gpu/drm/i915/i915_scheduler_types.h | 18 +++++
.../gpu/drm/i915/selftests/i915_scheduler.c | 1 +
9 files changed, 111 insertions(+), 52 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 1b7661386dee..7c627e8d54d1 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -576,8 +576,6 @@ void intel_engine_init_execlists(struct intel_engine_cs *engine)
execlists->queue_priority_hint = INT_MIN;
execlists->queue = RB_ROOT_CACHED;
-
- i915_sched_init_ipi(&execlists->ipi);
}
static void cleanup_status_page(struct intel_engine_cs *engine)
@@ -693,7 +691,11 @@ static int engine_setup_common(struct intel_engine_cs *engine)
goto err_status;
}
- intel_engine_init_active(engine, ENGINE_PHYSICAL);
+ i915_sched_init_engine(&engine->active,
+ engine->i915->drm.dev,
+ engine->name,
+ engine->mask,
+ ENGINE_PHYSICAL);
intel_engine_init_execlists(engine);
intel_engine_init_cmd_parser(engine);
intel_engine_init__pm(engine);
@@ -761,28 +763,6 @@ static int measure_breadcrumb_dw(struct intel_context *ce)
return dw;
}
-void
-intel_engine_init_active(struct intel_engine_cs *engine, unsigned int subclass)
-{
- INIT_LIST_HEAD(&engine->active.requests);
- INIT_LIST_HEAD(&engine->active.hold);
-
- spin_lock_init(&engine->active.lock);
- lockdep_set_subclass(&engine->active.lock, subclass);
-
- /*
- * Due to an interesting quirk in lockdep's internal debug tracking,
- * after setting a subclass we must ensure the lock is used. Otherwise,
- * nr_unused_locks is incremented once too often.
- */
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
- local_irq_disable();
- lock_map_acquire(&engine->active.lock.dep_map);
- lock_map_release(&engine->active.lock.dep_map);
- local_irq_enable();
-#endif
-}
-
static struct intel_context *
create_pinned_context(struct intel_engine_cs *engine,
unsigned int hwsp,
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 9105b7769635..36bcd85cc73b 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -258,8 +258,6 @@ struct intel_engine_execlists {
struct rb_root_cached queue;
struct rb_root_cached virtual;
- struct i915_sched_ipi ipi;
-
/**
* @csb_write: control register for Context Switch buffer
*
@@ -329,11 +327,7 @@ struct intel_engine_cs {
struct intel_sseu sseu;
- struct {
- spinlock_t lock;
- struct list_head requests;
- struct list_head hold; /* ready requests, but on hold */
- } active;
+ struct i915_sched_engine active;
/* keep a request in reserve for a [pm] barrier under oom */
struct i915_request *request_pool;
diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 195d7a21b4cd..21a5375af399 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -3347,7 +3347,6 @@ intel_execlists_create_virtual(struct intel_engine_cs **siblings,
snprintf(ve->base.name, sizeof(ve->base.name), "virtual");
- intel_engine_init_active(&ve->base, ENGINE_VIRTUAL);
intel_engine_init_execlists(&ve->base);
ve->base.cops = &virtual_context_ops;
@@ -3433,6 +3432,12 @@ intel_execlists_create_virtual(struct intel_engine_cs **siblings,
ve->base.flags |= I915_ENGINE_IS_VIRTUAL;
+ i915_sched_init_engine(&ve->base.active,
+ ve->base.i915->drm.dev,
+ ve->base.name,
+ ve->base.mask,
+ ENGINE_VIRTUAL);
+
virtual_engine_initial_hint(ve);
return &ve->context;
diff --git a/drivers/gpu/drm/i915/gt/mock_engine.c b/drivers/gpu/drm/i915/gt/mock_engine.c
index cf1269e74998..350191479c30 100644
--- a/drivers/gpu/drm/i915/gt/mock_engine.c
+++ b/drivers/gpu/drm/i915/gt/mock_engine.c
@@ -327,7 +327,11 @@ int mock_engine_init(struct intel_engine_cs *engine)
{
struct intel_context *ce;
- intel_engine_init_active(engine, ENGINE_MOCK);
+ i915_sched_init_engine(&engine->active,
+ engine->i915->drm.dev,
+ engine->name,
+ engine->mask,
+ ENGINE_MOCK);
intel_engine_init_execlists(engine);
intel_engine_init__pm(engine);
intel_engine_init_retire(engine);
diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index c0bd4cb8786a..a8dd7120e182 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -51,11 +51,13 @@ struct i915_capture_list {
struct i915_vma *vma;
};
+#define RQ_FMT "%llx:%lld"
+#define RQ_ARG(rq) (rq) ? (rq)->fence.context : 0, (rq) ? (rq)->fence.seqno : 0
+
#define RQ_TRACE(rq, fmt, ...) do { \
const struct i915_request *rq__ = (rq); \
- ENGINE_TRACE(rq__->engine, "fence %llx:%lld, current %d " fmt, \
- rq__->fence.context, rq__->fence.seqno, \
- hwsp_seqno(rq__), ##__VA_ARGS__); \
+ ENGINE_TRACE(rq__->engine, "fence " RQ_FMT ", current %d " fmt, \
+ RQ_ARG(rq__), hwsp_seqno(rq__), ##__VA_ARGS__); \
} while (0)
enum {
diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c
index 694ca3a3b563..0a86f16f3cc8 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.c
+++ b/drivers/gpu/drm/i915/i915_scheduler.c
@@ -85,12 +85,46 @@ static void ipi_schedule(struct work_struct *wrk)
} while (rq);
}
-void i915_sched_init_ipi(struct i915_sched_ipi *ipi)
+static void i915_sched_init_ipi(struct i915_sched_ipi *ipi)
{
INIT_WORK(&ipi->work, ipi_schedule);
ipi->list = NULL;
}
+void i915_sched_init_engine(struct i915_sched_engine *se,
+ struct device *dev,
+ const char *name,
+ unsigned long mask,
+ unsigned int subclass)
+{
+ spin_lock_init(&se->lock);
+ lockdep_set_subclass(&se->lock, subclass);
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+ se->dbg.dev = dev;
+ se->dbg.name = name;
+#endif
+
+ se->mask = mask;
+
+ INIT_LIST_HEAD(&se->requests);
+ INIT_LIST_HEAD(&se->hold);
+
+ i915_sched_init_ipi(&se->ipi);
+
+ /*
+ * Due to an interesting quirk in lockdep's internal debug tracking,
+ * after setting a subclass we must ensure the lock is used. Otherwise,
+ * nr_unused_locks is incremented once too often.
+ */
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+ local_irq_disable();
+ lock_map_acquire(&se->lock.dep_map);
+ lock_map_release(&se->lock.dep_map);
+ local_irq_enable();
+#endif
+}
+
static void __ipi_add(struct i915_request *rq)
{
#define STUB ((struct i915_request *)1)
@@ -114,13 +148,13 @@ static void __ipi_add(struct i915_request *rq)
}
/* Carefully insert ourselves into the head of the llist */
- first = READ_ONCE(engine->execlists.ipi.list);
+ first = READ_ONCE(engine->active.ipi.list);
do {
rq->sched.ipi_link = ptr_pack_bits(first, 1, 1);
- } while (!try_cmpxchg(&engine->execlists.ipi.list, &first, rq));
+ } while (!try_cmpxchg(&engine->active.ipi.list, &first, rq));
if (!first)
- queue_work(system_unbound_wq, &engine->execlists.ipi.work);
+ queue_work(system_unbound_wq, &engine->active.ipi.work);
}
/*
@@ -302,12 +336,11 @@ static void kick_submission(struct intel_engine_cs *engine,
if (inflight->context == rq->context)
return;
- ENGINE_TRACE(engine,
- "bumping queue-priority-hint:%d for rq:%llx:%lld, inflight:%llx:%lld prio %d\n",
- prio,
- rq->fence.context, rq->fence.seqno,
- inflight->fence.context, inflight->fence.seqno,
- inflight->sched.attr.priority);
+ SCHED_TRACE(&engine->active,
+ "bumping queue-priority-hint:%d for rq:" RQ_FMT ", inflight:" RQ_FMT " prio %d\n",
+ prio,
+ RQ_ARG(rq), RQ_ARG(inflight),
+ inflight->sched.attr.priority);
engine->execlists.queue_priority_hint = prio;
if (need_preempt(prio, rq_prio(inflight)))
@@ -332,6 +365,9 @@ static void __i915_request_set_priority(struct i915_request *rq, int prio)
struct list_head *pos = &rq->sched.signalers_list;
struct list_head *plist;
+ SCHED_TRACE(&engine->active, "PI for " RQ_FMT ", prio:%d\n",
+ RQ_FMT(rq), prio);
+
plist = lookup_priolist(engine, prio);
/*
@@ -471,6 +507,8 @@ void __i915_sched_defer_request(struct intel_engine_cs *engine,
struct i915_request *rn;
LIST_HEAD(dfs);
+ SCHED_TRACE(&engine->active, "defer request " RQ_FMT "\n", RQ_ARG(rq));
+
lockdep_assert_held(&engine->active.lock);
GEM_BUG_ON(!test_bit(I915_FENCE_FLAG_PQUEUE, &rq->fence.flags));
@@ -598,6 +636,8 @@ void i915_request_enqueue(struct i915_request *rq)
unsigned long flags;
bool kick = false;
+ SCHED_TRACE(&engine->active, "queue request " RQ_FMT "\n", RQ_ARG(rq));
+
/* Will be called from irq-context when using foreign fences. */
spin_lock_irqsave(&engine->active.lock, flags);
GEM_BUG_ON(test_bit(I915_FENCE_FLAG_PQUEUE, &rq->fence.flags));
@@ -658,6 +698,10 @@ __i915_sched_rewind_requests(struct intel_engine_cs *engine)
active = rq;
}
+ SCHED_TRACE(&engine->active,
+ "rewind requests, active request " RQ_FMT "\n",
+ RQ_ARG(active));
+
return active;
}
@@ -675,8 +719,8 @@ bool __i915_sched_suspend_request(struct intel_engine_cs *engine,
if (i915_request_on_hold(rq))
return false;
- ENGINE_TRACE(engine, "suspending request %llx:%lld\n",
- rq->fence.context, rq->fence.seqno);
+ SCHED_TRACE(&engine->active, "suspending request " RQ_FMT "\n",
+ RQ_ARG(rq));
/*
* Transfer this request onto the hold queue to prevent it
@@ -756,8 +800,8 @@ void __i915_sched_resume_request(struct intel_engine_cs *engine,
if (!i915_request_on_hold(rq))
return;
- ENGINE_TRACE(engine, "resuming request %llx:%lld\n",
- rq->fence.context, rq->fence.seqno);
+ SCHED_TRACE(&engine->active, "resuming request " RQ_FMT "\n",
+ RQ_ARG(rq));
/*
* Move this request back to the priority queue, and all of its
diff --git a/drivers/gpu/drm/i915/i915_scheduler.h b/drivers/gpu/drm/i915/i915_scheduler.h
index 00ce0a9d519d..346a738c86b5 100644
--- a/drivers/gpu/drm/i915/i915_scheduler.h
+++ b/drivers/gpu/drm/i915/i915_scheduler.h
@@ -16,6 +16,13 @@
struct drm_printer;
struct intel_engine_cs;
+#define SCHED_TRACE(se, fmt, ...) do { \
+ const struct i915_sched_engine *se__ __maybe_unused = (se); \
+ GEM_TRACE("%s sched:%s: " fmt, \
+ dev_name(se__->dbg.dev), se__->dbg.name, \
+ ##__VA_ARGS__); \
+} while (0)
+
#define priolist_for_each_request(it, plist) \
list_for_each_entry(it, &(plist)->requests, sched.link)
@@ -36,7 +43,11 @@ int i915_sched_node_add_dependency(struct i915_sched_node *node,
void i915_sched_node_retire(struct i915_sched_node *node);
-void i915_sched_init_ipi(struct i915_sched_ipi *ipi);
+void i915_sched_init_engine(struct i915_sched_engine *se,
+ struct device *dev,
+ const char *name,
+ unsigned long mask,
+ unsigned int subclass);
void i915_request_set_priority(struct i915_request *request, int prio);
diff --git a/drivers/gpu/drm/i915/i915_scheduler_types.h b/drivers/gpu/drm/i915/i915_scheduler_types.h
index 60a3aa743ab8..3cf60e60dc29 100644
--- a/drivers/gpu/drm/i915/i915_scheduler_types.h
+++ b/drivers/gpu/drm/i915/i915_scheduler_types.h
@@ -20,6 +20,24 @@ struct i915_sched_ipi {
struct work_struct work;
};
+struct i915_sched_engine {
+ spinlock_t lock; /* protects the scheduling lists and queue */
+
+ unsigned long mask; /* available scheduling channels */
+
+ struct list_head requests; /* active request, on HW */
+ struct list_head hold; /* ready requests, but on hold */
+
+ struct i915_sched_ipi ipi;
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+ struct {
+ struct device *dev;
+ const char *name;
+ } dbg;
+#endif
+};
+
struct i915_sched_attr {
/**
* @priority: execution and service priority
diff --git a/drivers/gpu/drm/i915/selftests/i915_scheduler.c b/drivers/gpu/drm/i915/selftests/i915_scheduler.c
index acc666f755d7..a70b6d035d0a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_scheduler.c
+++ b/drivers/gpu/drm/i915/selftests/i915_scheduler.c
@@ -884,6 +884,7 @@ int i915_scheduler_perf_selftests(struct drm_i915_private *i915)
T(i915_priolist),
T(i915_sched_attr),
T(i915_sched_node),
+ T(i915_sched_engine),
T(i915_dependency),
#undef T
{}
--
2.20.1
More information about the Intel-gfx-trybot
mailing list