[PATCH 5/5] drm/i915/gt: Pipelined clear
Chris Wilson
chris at chris-wilson.co.uk
Tue Dec 1 11:51:43 UTC 2020
Update the PTE and emit a clear within a single unpreemptible packet
such that we can schedule and pipeline clears.
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
drivers/gpu/drm/i915/gt/intel_migrate.c | 135 +++++++++++++++++++++
drivers/gpu/drm/i915/gt/intel_migrate.h | 13 ++
drivers/gpu/drm/i915/gt/selftest_migrate.c | 62 ++++++++++
3 files changed, 210 insertions(+)
diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c
index 1ab9c569448e..53a33636cfe3 100644
--- a/drivers/gpu/drm/i915/gt/intel_migrate.c
+++ b/drivers/gpu/drm/i915/gt/intel_migrate.c
@@ -450,6 +450,117 @@ intel_context_migrate_copy(struct intel_context *ce,
return err;
}
+static int emit_clear(struct i915_request *rq, int size)
+{
+ const int gen = INTEL_GEN(rq->engine->i915);
+ u32 *cs;
+
+ GEM_BUG_ON(size >> PAGE_SHIFT > S16_MAX);
+
+ cs = intel_ring_begin(rq, gen >= 8 ? 8 : 6);
+ if (IS_ERR(cs))
+ return PTR_ERR(cs);
+
+ if (gen >= 8) {
+ *cs++ = XY_COLOR_BLT_CMD | BLT_WRITE_RGBA | (7 - 2);
+ *cs++ = BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | PAGE_SIZE;
+ *cs++ = 0;
+ *cs++ = size >> PAGE_SHIFT << 16 | PAGE_SIZE / 4;
+ *cs++ = 0; /* offset */
+ *cs++ = 0;
+ *cs++ = 0; /* value */
+ *cs++ = MI_NOOP;
+ } else {
+ *cs++ = XY_COLOR_BLT_CMD | BLT_WRITE_RGBA | (6 - 2);
+ *cs++ = BLT_DEPTH_32 | BLT_ROP_COLOR_COPY | PAGE_SIZE;
+ *cs++ = 0;
+ *cs++ = size >> PAGE_SHIFT << 16 | PAGE_SIZE / 4;
+ *cs++ = 0;
+ *cs++ = 0; /* value */
+ }
+
+ intel_ring_advance(rq, cs);
+ return 0;
+}
+
+int
+intel_context_migrate_clear(struct intel_context *ce,
+ struct dma_fence *await,
+ struct scatterlist *sg,
+ unsigned long flags,
+ struct i915_request **out)
+{
+ struct sgt_dma it = sg_sgt(sg);
+ struct i915_request *rq;
+ int err;
+
+ *out = NULL;
+
+ /* GEM_BUG_ON(ce->vm != migrate_vm); */
+
+ err = intel_context_pin(ce);
+ if (err)
+ return err;
+
+ GEM_BUG_ON(ce->ring->size < SZ_64K);
+
+ do {
+ int len;
+
+ rq = i915_request_create(ce);
+ if (IS_ERR(rq)) {
+ err = PTR_ERR(rq);
+ goto out_ce;
+ }
+
+ if (await) {
+ err = i915_request_await_dma_fence(rq, await);
+ if (err)
+ goto out_rq;
+
+ if (rq->engine->emit_init_breadcrumb) {
+ err = rq->engine->emit_init_breadcrumb(rq);
+ if (err)
+ goto out_rq;
+ }
+
+ await = NULL;
+ }
+
+ /* The PTE updates + clear must not be interrupted. */
+ err = emit_no_arbitration(rq);
+ if (err)
+ goto out_rq;
+
+ len = emit_pte(rq, &it, flags, 0, CHUNK_SZ);
+ if (len <= 0) {
+ err = len;
+ goto out_rq;
+ }
+
+ err = rq->engine->emit_flush(rq, EMIT_INVALIDATE);
+ if (err)
+ goto out_rq;
+
+ err = emit_clear(rq, len);
+
+ /* Arbitration is re-enabled between requests. */
+out_rq:
+ if (*out)
+ i915_request_put(*out);
+ *out = i915_request_get(rq);
+ i915_request_add(rq);
+ if (err || !it.sg)
+ break;
+
+ cond_resched();
+ } while (1);
+
+out_ce:
+ intel_context_unpin(ce);
+ return err;
+}
+
int
intel_migrate_copy(struct intel_migrate *m,
struct dma_fence *await,
@@ -477,6 +588,30 @@ intel_migrate_copy(struct intel_migrate *m,
return err;
}
+int
+intel_migrate_clear(struct intel_migrate *m,
+ struct dma_fence *await,
+ struct scatterlist *sg,
+ unsigned long flags,
+ struct i915_request **out)
+{
+ struct intel_context *ce;
+ int err;
+
+ if (!m->ce)
+ return -ENODEV;
+
+ ce = intel_migrate_create_context(m);
+ if (IS_ERR(ce))
+ ce = intel_context_get(m->ce);
+ GEM_BUG_ON(IS_ERR(ce));
+
+ err = intel_context_migrate_clear(ce, await, sg, flags, out);
+
+ intel_context_put(ce);
+ return err;
+}
+
void intel_migrate_fini(struct intel_migrate *m)
{
if (!m->ce)
diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.h b/drivers/gpu/drm/i915/gt/intel_migrate.h
index ce0168c89f2b..540c3f20917f 100644
--- a/drivers/gpu/drm/i915/gt/intel_migrate.h
+++ b/drivers/gpu/drm/i915/gt/intel_migrate.h
@@ -31,6 +31,19 @@ int intel_context_migrate_copy(struct intel_context *ce,
struct scatterlist *dst, unsigned long dst_flags,
struct i915_request **out);
+int
+intel_migrate_clear(struct intel_migrate *m,
+ struct dma_fence *await,
+ struct scatterlist *sg,
+ unsigned long flags,
+ struct i915_request **out);
+int
+intel_context_migrate_clear(struct intel_context *ce,
+ struct dma_fence *await,
+ struct scatterlist *sg,
+ unsigned long flags,
+ struct i915_request **out);
+
void intel_migrate_fini(struct intel_migrate *m);
#endif /* __INTEL_MIGRATE__ */
diff --git a/drivers/gpu/drm/i915/gt/selftest_migrate.c b/drivers/gpu/drm/i915/gt/selftest_migrate.c
index 179ab19ac659..5ba313f2ae82 100644
--- a/drivers/gpu/drm/i915/gt/selftest_migrate.c
+++ b/drivers/gpu/drm/i915/gt/selftest_migrate.c
@@ -95,10 +95,72 @@ static int live_migrate_copy(void *arg)
return err;
}
+static int live_migrate_clear(void *arg)
+{
+ struct intel_migrate *m = arg;
+ struct drm_i915_private *i915 = m->ce->engine->i915;
+ I915_RND_STATE(prng);
+ int i, j, k;
+ int err = 0;
+
+ for (i = 0; i < ARRAY_SIZE(sizes); i++) {
+ struct drm_i915_gem_object *obj;
+ struct i915_request *rq;
+ u32 *vaddr;
+
+ obj = i915_gem_object_create_internal(i915, sizes[i]);
+ if (IS_ERR(obj))
+ break;
+
+ vaddr = i915_gem_object_pin_map(obj, I915_MAP_WC);
+ if (IS_ERR(vaddr)) {
+ i915_gem_object_put(obj);
+ break;
+ }
+
+ for (j = 0; j < sizes[i] / sizeof(u32); j++)
+ vaddr[j] = ~j;
+ i915_gem_object_flush_map(obj);
+
+ err = intel_migrate_clear(m, NULL,
+ obj->mm.pages->sgl,
+ obj->cache_level,
+ &rq);
+ if (err)
+ pr_err("Clear failed, size: %u\n", sizes[i]);
+
+ if (rq) {
+ if (i915_request_wait(rq, 0, HZ) < 0) {
+ pr_err("Clear timed out, size: %u\n", sizes[i]);
+ err = -ETIME;
+ }
+ i915_request_put(rq);
+ }
+
+ for (j = 0; !err && j < sizes[i] / PAGE_SIZE; j++) {
+ k = i915_prandom_u32_max_state(1024, &prng);
+ if (vaddr[j * 1024 + k] != 0) {
+ pr_err("Clear failed, size: %u, offset: %zu\n",
+ sizes[i], (j * 1024 + k) * sizeof(u32));
+ igt_hexdump(vaddr + j * 1024, 4096);
+ err = -EINVAL;
+ }
+ }
+
+ i915_gem_object_put(obj);
+ i915_gem_drain_freed_objects(i915);
+ if (err)
+ break;
+ }
+
+ return err;
+}
+
int intel_migrate_live_selftests(struct drm_i915_private *i915)
{
static const struct i915_subtest tests[] = {
SUBTEST(live_migrate_copy),
+ SUBTEST(live_migrate_clear),
};
struct intel_migrate m;
int err;
--
2.20.1
More information about the Intel-gfx-trybot
mailing list