[Intel-gfx] [PATCH v5 4/8] drm/i915: introduce mem->reserved
Matthew Auld
matthew.auld at intel.com
Wed Jan 27 13:14:13 UTC 2021
From: Abdiel Janulgue <abdiel.janulgue at linux.intel.com>
In the following patch we need to reserve regions unaccessible to the
driver during initialization, so add mem->reserved for collecting such
regions.
v2: turn into an actual intel_memory_region_reserve api
Cc: Imre Deak <imre.deak at intel.com>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue at linux.intel.com>
Signed-off-by: Matthew Auld <matthew.auld at intel.com>
Reviewed-by: Chris Wilson <chris at chris-wilson.co.uk>
---
drivers/gpu/drm/i915/intel_memory_region.c | 14 ++++
drivers/gpu/drm/i915/intel_memory_region.h | 5 ++
.../drm/i915/selftests/intel_memory_region.c | 77 +++++++++++++++++++
3 files changed, 96 insertions(+)
diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c
index b1b610bfff09..49d306b5532f 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/intel_memory_region.c
@@ -156,9 +156,22 @@ int intel_memory_region_init_buddy(struct intel_memory_region *mem)
void intel_memory_region_release_buddy(struct intel_memory_region *mem)
{
+ i915_buddy_free_list(&mem->mm, &mem->reserved);
i915_buddy_fini(&mem->mm);
}
+int intel_memory_region_reserve(struct intel_memory_region *mem,
+ u64 offset, u64 size)
+{
+ int ret;
+
+ mutex_lock(&mem->mm_lock);
+ ret = i915_buddy_alloc_range(&mem->mm, &mem->reserved, offset, size);
+ mutex_unlock(&mem->mm_lock);
+
+ return ret;
+}
+
struct intel_memory_region *
intel_memory_region_create(struct drm_i915_private *i915,
resource_size_t start,
@@ -185,6 +198,7 @@ intel_memory_region_create(struct drm_i915_private *i915,
mutex_init(&mem->objects.lock);
INIT_LIST_HEAD(&mem->objects.list);
INIT_LIST_HEAD(&mem->objects.purgeable);
+ INIT_LIST_HEAD(&mem->reserved);
mutex_init(&mem->mm_lock);
diff --git a/drivers/gpu/drm/i915/intel_memory_region.h b/drivers/gpu/drm/i915/intel_memory_region.h
index 6ffc0673f005..d17e4fe3123c 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.h
+++ b/drivers/gpu/drm/i915/intel_memory_region.h
@@ -89,6 +89,8 @@ struct intel_memory_region {
unsigned int id;
char name[8];
+ struct list_head reserved;
+
dma_addr_t remap_addr;
struct {
@@ -113,6 +115,9 @@ void __intel_memory_region_put_pages_buddy(struct intel_memory_region *mem,
struct list_head *blocks);
void __intel_memory_region_put_block_buddy(struct i915_buddy_block *block);
+int intel_memory_region_reserve(struct intel_memory_region *mem,
+ u64 offset, u64 size);
+
struct intel_memory_region *
intel_memory_region_create(struct drm_i915_private *i915,
resource_size_t start,
diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
index ce7adfa3bca0..64348528e1d5 100644
--- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
@@ -144,6 +144,82 @@ static bool is_contiguous(struct drm_i915_gem_object *obj)
return true;
}
+static int igt_mock_reserve(void *arg)
+{
+ struct intel_memory_region *mem = arg;
+ resource_size_t avail = resource_size(&mem->region);
+ struct drm_i915_gem_object *obj;
+ const u32 chunk_size = SZ_32M;
+ u32 i, offset, count, *order;
+ u64 allocated, cur_avail;
+ I915_RND_STATE(prng);
+ LIST_HEAD(objects);
+ int err = 0;
+
+ if (!list_empty(&mem->reserved)) {
+ pr_err("%s region reserved list is not empty\n", __func__);
+ return -EINVAL;
+ }
+
+ count = avail / chunk_size;
+ order = i915_random_order(count, &prng);
+ if (!order)
+ return 0;
+
+ /* Reserve a bunch of ranges within the region */
+ for (i = 0; i < count; ++i) {
+ u64 start = order[i] * chunk_size;
+ u64 size = i915_prandom_u32_max_state(chunk_size, &prng);
+
+ /* Allow for some really big holes */
+ if (!size)
+ continue;
+
+ size = round_up(size, PAGE_SIZE);
+ offset = igt_random_offset(&prng, 0, chunk_size, size,
+ PAGE_SIZE);
+
+ err = intel_memory_region_reserve(mem, start + offset, size);
+ if (err) {
+ pr_err("%s failed to reserve range", __func__);
+ goto out_close;
+ }
+
+ /* XXX: maybe sanity check the block range here? */
+ avail -= size;
+ }
+
+ /* Try to see if we can allocate from the remaining space */
+ allocated = 0;
+ cur_avail = avail;
+ do {
+ u32 size = i915_prandom_u32_max_state(cur_avail, &prng);
+
+ size = max_t(u32, round_up(size, PAGE_SIZE), PAGE_SIZE);
+ obj = igt_object_create(mem, &objects, size, 0);
+ if (IS_ERR(obj)) {
+ if (PTR_ERR(obj) == -ENXIO)
+ break;
+
+ err = PTR_ERR(obj);
+ goto out_close;
+ }
+ cur_avail -= size;
+ allocated += size;
+ } while (1);
+
+ if (allocated != avail) {
+ pr_err("%s mismatch between allocation and free space", __func__);
+ err = -EINVAL;
+ }
+
+out_close:
+ kfree(order);
+ close_objects(mem, &objects);
+ i915_buddy_free_list(&mem->mm, &mem->reserved);
+ return err;
+}
+
static int igt_mock_contiguous(void *arg)
{
struct intel_memory_region *mem = arg;
@@ -930,6 +1006,7 @@ static int perf_memcpy(void *arg)
int intel_memory_region_mock_selftests(void)
{
static const struct i915_subtest tests[] = {
+ SUBTEST(igt_mock_reserve),
SUBTEST(igt_mock_fill),
SUBTEST(igt_mock_contiguous),
SUBTEST(igt_mock_splintered_region),
--
2.26.2
More information about the Intel-gfx
mailing list