* [PATCH v6 1/8] drm/i915/gem: Break out some shmem backend utils
@ 2021-10-05 18:23 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:23 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
From: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Break out some shmem backend utils for future reuse by the TTM backend:
shmem_alloc_st(), shmem_free_st() and __shmem_writeback() which we can
use to provide a shmem-backed TTM page pool for cached-only TTM
buffer objects.
Main functional change here is that we now compute the page sizes using
the dma segments rather than using the physical page address segments.
v2(Reported-by: kernel test robot <lkp@intel.com>)
- Make sure we initialise the mapping on the error path in
shmem_get_pages()
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 181 +++++++++++++---------
1 file changed, 106 insertions(+), 75 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 11f072193f3b..36b711ae9e28 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -25,46 +25,61 @@ static void check_release_pagevec(struct pagevec *pvec)
cond_resched();
}
-static int shmem_get_pages(struct drm_i915_gem_object *obj)
+static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup)
{
- struct drm_i915_private *i915 = to_i915(obj->base.dev);
- struct intel_memory_region *mem = obj->mm.region;
- const unsigned long page_count = obj->base.size / PAGE_SIZE;
+ struct sgt_iter sgt_iter;
+ struct pagevec pvec;
+ struct page *page;
+
+ mapping_clear_unevictable(mapping);
+
+ pagevec_init(&pvec);
+ for_each_sgt_page(page, sgt_iter, st) {
+ if (dirty)
+ set_page_dirty(page);
+
+ if (backup)
+ mark_page_accessed(page);
+
+ if (!pagevec_add(&pvec, page))
+ check_release_pagevec(&pvec);
+ }
+ if (pagevec_count(&pvec))
+ check_release_pagevec(&pvec);
+
+ sg_free_table(st);
+ kfree(st);
+}
+
+static struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment)
+{
+ const unsigned long page_count = size / PAGE_SIZE;
unsigned long i;
- struct address_space *mapping;
struct sg_table *st;
struct scatterlist *sg;
- struct sgt_iter sgt_iter;
struct page *page;
unsigned long last_pfn = 0; /* suppress gcc warning */
- unsigned int max_segment = i915_sg_segment_size();
- unsigned int sg_page_sizes;
gfp_t noreclaim;
int ret;
- /*
- * Assert that the object is not currently in any GPU domain. As it
- * wasn't in the GTT, there shouldn't be any way it could have been in
- * a GPU cache
- */
- GEM_BUG_ON(obj->read_domains & I915_GEM_GPU_DOMAINS);
- GEM_BUG_ON(obj->write_domain & I915_GEM_GPU_DOMAINS);
-
/*
* If there's no chance of allocating enough pages for the whole
* object, bail early.
*/
- if (obj->base.size > resource_size(&mem->region))
- return -ENOMEM;
+ if (size > resource_size(&mr->region))
+ return ERR_PTR(-ENOMEM);
st = kmalloc(sizeof(*st), GFP_KERNEL);
if (!st)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
-rebuild_st:
if (sg_alloc_table(st, page_count, GFP_KERNEL)) {
kfree(st);
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
}
/*
@@ -73,14 +88,12 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
*
* Fail silently without starting the shrinker
*/
- mapping = obj->base.filp->f_mapping;
mapping_set_unevictable(mapping);
noreclaim = mapping_gfp_constraint(mapping, ~__GFP_RECLAIM);
noreclaim |= __GFP_NORETRY | __GFP_NOWARN;
sg = st->sgl;
st->nents = 0;
- sg_page_sizes = 0;
for (i = 0; i < page_count; i++) {
const unsigned int shrink[] = {
I915_SHRINK_BOUND | I915_SHRINK_UNBOUND,
@@ -135,10 +148,9 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
if (!i ||
sg->length >= max_segment ||
page_to_pfn(page) != last_pfn + 1) {
- if (i) {
- sg_page_sizes |= sg->length;
+ if (i)
sg = sg_next(sg);
- }
+
st->nents++;
sg_set_page(sg, page, PAGE_SIZE, 0);
} else {
@@ -149,14 +161,65 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
/* Check that the i965g/gm workaround works. */
GEM_BUG_ON(gfp & __GFP_DMA32 && last_pfn >= 0x00100000UL);
}
- if (sg) { /* loop terminated early; short sg table */
- sg_page_sizes |= sg->length;
+ if (sg) /* loop terminated early; short sg table */
sg_mark_end(sg);
- }
/* Trim unused sg entries to avoid wasting memory. */
i915_sg_trim(st);
+ return st;
+err_sg:
+ sg_mark_end(sg);
+ if (sg != st->sgl) {
+ shmem_free_st(st, mapping, false, false);
+ } else {
+ mapping_clear_unevictable(mapping);
+ sg_free_table(st);
+ kfree(st);
+ }
+
+ /*
+ * shmemfs first checks if there is enough memory to allocate the page
+ * and reports ENOSPC should there be insufficient, along with the usual
+ * ENOMEM for a genuine allocation failure.
+ *
+ * We use ENOSPC in our driver to mean that we have run out of aperture
+ * space and so want to translate the error from shmemfs back to our
+ * usual understanding of ENOMEM.
+ */
+ if (ret == -ENOSPC)
+ ret = -ENOMEM;
+
+ return ERR_PTR(ret);
+}
+
+static int shmem_get_pages(struct drm_i915_gem_object *obj)
+{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
+ struct intel_memory_region *mem = obj->mm.region;
+ struct address_space *mapping = obj->base.filp->f_mapping;
+ const unsigned long page_count = obj->base.size / PAGE_SIZE;
+ unsigned int max_segment = i915_sg_segment_size();
+ struct sg_table *st;
+ struct sgt_iter sgt_iter;
+ struct page *page;
+ int ret;
+
+ /*
+ * Assert that the object is not currently in any GPU domain. As it
+ * wasn't in the GTT, there shouldn't be any way it could have been in
+ * a GPU cache
+ */
+ GEM_BUG_ON(obj->read_domains & I915_GEM_GPU_DOMAINS);
+ GEM_BUG_ON(obj->write_domain & I915_GEM_GPU_DOMAINS);
+
+rebuild_st:
+ st = shmem_alloc_st(i915, obj->base.size, mem, mapping, max_segment);
+ if (IS_ERR(st)) {
+ ret = PTR_ERR(st);
+ goto err_st;
+ }
+
ret = i915_gem_gtt_prepare_pages(obj, st);
if (ret) {
/*
@@ -168,6 +231,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
for_each_sgt_page(page, sgt_iter, st)
put_page(page);
sg_free_table(st);
+ kfree(st);
max_segment = PAGE_SIZE;
goto rebuild_st;
@@ -200,28 +264,12 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
if (IS_JSL_EHL(i915) && obj->flags & I915_BO_ALLOC_USER)
obj->cache_dirty = true;
- __i915_gem_object_set_pages(obj, st, sg_page_sizes);
+ __i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
return 0;
-err_sg:
- sg_mark_end(sg);
err_pages:
- mapping_clear_unevictable(mapping);
- if (sg != st->sgl) {
- struct pagevec pvec;
-
- pagevec_init(&pvec);
- for_each_sgt_page(page, sgt_iter, st) {
- if (!pagevec_add(&pvec, page))
- check_release_pagevec(&pvec);
- }
- if (pagevec_count(&pvec))
- check_release_pagevec(&pvec);
- }
- sg_free_table(st);
- kfree(st);
-
+ shmem_free_st(st, mapping, false, false);
/*
* shmemfs first checks if there is enough memory to allocate the page
* and reports ENOSPC should there be insufficient, along with the usual
@@ -231,6 +279,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
* space and so want to translate the error from shmemfs back to our
* usual understanding of ENOMEM.
*/
+err_st:
if (ret == -ENOSPC)
ret = -ENOMEM;
@@ -251,10 +300,8 @@ shmem_truncate(struct drm_i915_gem_object *obj)
obj->mm.pages = ERR_PTR(-EFAULT);
}
-static void
-shmem_writeback(struct drm_i915_gem_object *obj)
+static void __shmem_writeback(size_t size, struct address_space *mapping)
{
- struct address_space *mapping;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_NONE,
.nr_to_write = SWAP_CLUSTER_MAX,
@@ -270,10 +317,9 @@ shmem_writeback(struct drm_i915_gem_object *obj)
* instead of invoking writeback so they are aged and paged out
* as normal.
*/
- mapping = obj->base.filp->f_mapping;
/* Begin writeback on each dirty page */
- for (i = 0; i < obj->base.size >> PAGE_SHIFT; i++) {
+ for (i = 0; i < size >> PAGE_SHIFT; i++) {
struct page *page;
page = find_lock_page(mapping, i);
@@ -296,6 +342,12 @@ shmem_writeback(struct drm_i915_gem_object *obj)
}
}
+static void
+shmem_writeback(struct drm_i915_gem_object *obj)
+{
+ __shmem_writeback(obj->base.size, obj->base.filp->f_mapping);
+}
+
void
__i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
struct sg_table *pages,
@@ -316,11 +368,6 @@ __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
void i915_gem_object_put_pages_shmem(struct drm_i915_gem_object *obj, struct sg_table *pages)
{
- struct sgt_iter sgt_iter;
- struct pagevec pvec;
- struct page *page;
-
- GEM_WARN_ON(IS_DGFX(to_i915(obj->base.dev)));
__i915_gem_object_release_shmem(obj, pages, true);
i915_gem_gtt_finish_pages(obj, pages);
@@ -328,25 +375,9 @@ void i915_gem_object_put_pages_shmem(struct drm_i915_gem_object *obj, struct sg_
if (i915_gem_object_needs_bit17_swizzle(obj))
i915_gem_object_save_bit_17_swizzle(obj, pages);
- mapping_clear_unevictable(file_inode(obj->base.filp)->i_mapping);
-
- pagevec_init(&pvec);
- for_each_sgt_page(page, sgt_iter, pages) {
- if (obj->mm.dirty)
- set_page_dirty(page);
-
- if (obj->mm.madv == I915_MADV_WILLNEED)
- mark_page_accessed(page);
-
- if (!pagevec_add(&pvec, page))
- check_release_pagevec(&pvec);
- }
- if (pagevec_count(&pvec))
- check_release_pagevec(&pvec);
+ shmem_free_st(pages, file_inode(obj->base.filp)->i_mapping,
+ obj->mm.dirty, obj->mm.madv == I915_MADV_WILLNEED);
obj->mm.dirty = false;
-
- sg_free_table(pages);
- kfree(pages);
}
static void
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 1/8] drm/i915/gem: Break out some shmem backend utils
@ 2021-10-05 18:23 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:23 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
From: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Break out some shmem backend utils for future reuse by the TTM backend:
shmem_alloc_st(), shmem_free_st() and __shmem_writeback() which we can
use to provide a shmem-backed TTM page pool for cached-only TTM
buffer objects.
Main functional change here is that we now compute the page sizes using
the dma segments rather than using the physical page address segments.
v2(Reported-by: kernel test robot <lkp@intel.com>)
- Make sure we initialise the mapping on the error path in
shmem_get_pages()
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 181 +++++++++++++---------
1 file changed, 106 insertions(+), 75 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 11f072193f3b..36b711ae9e28 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -25,46 +25,61 @@ static void check_release_pagevec(struct pagevec *pvec)
cond_resched();
}
-static int shmem_get_pages(struct drm_i915_gem_object *obj)
+static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup)
{
- struct drm_i915_private *i915 = to_i915(obj->base.dev);
- struct intel_memory_region *mem = obj->mm.region;
- const unsigned long page_count = obj->base.size / PAGE_SIZE;
+ struct sgt_iter sgt_iter;
+ struct pagevec pvec;
+ struct page *page;
+
+ mapping_clear_unevictable(mapping);
+
+ pagevec_init(&pvec);
+ for_each_sgt_page(page, sgt_iter, st) {
+ if (dirty)
+ set_page_dirty(page);
+
+ if (backup)
+ mark_page_accessed(page);
+
+ if (!pagevec_add(&pvec, page))
+ check_release_pagevec(&pvec);
+ }
+ if (pagevec_count(&pvec))
+ check_release_pagevec(&pvec);
+
+ sg_free_table(st);
+ kfree(st);
+}
+
+static struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment)
+{
+ const unsigned long page_count = size / PAGE_SIZE;
unsigned long i;
- struct address_space *mapping;
struct sg_table *st;
struct scatterlist *sg;
- struct sgt_iter sgt_iter;
struct page *page;
unsigned long last_pfn = 0; /* suppress gcc warning */
- unsigned int max_segment = i915_sg_segment_size();
- unsigned int sg_page_sizes;
gfp_t noreclaim;
int ret;
- /*
- * Assert that the object is not currently in any GPU domain. As it
- * wasn't in the GTT, there shouldn't be any way it could have been in
- * a GPU cache
- */
- GEM_BUG_ON(obj->read_domains & I915_GEM_GPU_DOMAINS);
- GEM_BUG_ON(obj->write_domain & I915_GEM_GPU_DOMAINS);
-
/*
* If there's no chance of allocating enough pages for the whole
* object, bail early.
*/
- if (obj->base.size > resource_size(&mem->region))
- return -ENOMEM;
+ if (size > resource_size(&mr->region))
+ return ERR_PTR(-ENOMEM);
st = kmalloc(sizeof(*st), GFP_KERNEL);
if (!st)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
-rebuild_st:
if (sg_alloc_table(st, page_count, GFP_KERNEL)) {
kfree(st);
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
}
/*
@@ -73,14 +88,12 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
*
* Fail silently without starting the shrinker
*/
- mapping = obj->base.filp->f_mapping;
mapping_set_unevictable(mapping);
noreclaim = mapping_gfp_constraint(mapping, ~__GFP_RECLAIM);
noreclaim |= __GFP_NORETRY | __GFP_NOWARN;
sg = st->sgl;
st->nents = 0;
- sg_page_sizes = 0;
for (i = 0; i < page_count; i++) {
const unsigned int shrink[] = {
I915_SHRINK_BOUND | I915_SHRINK_UNBOUND,
@@ -135,10 +148,9 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
if (!i ||
sg->length >= max_segment ||
page_to_pfn(page) != last_pfn + 1) {
- if (i) {
- sg_page_sizes |= sg->length;
+ if (i)
sg = sg_next(sg);
- }
+
st->nents++;
sg_set_page(sg, page, PAGE_SIZE, 0);
} else {
@@ -149,14 +161,65 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
/* Check that the i965g/gm workaround works. */
GEM_BUG_ON(gfp & __GFP_DMA32 && last_pfn >= 0x00100000UL);
}
- if (sg) { /* loop terminated early; short sg table */
- sg_page_sizes |= sg->length;
+ if (sg) /* loop terminated early; short sg table */
sg_mark_end(sg);
- }
/* Trim unused sg entries to avoid wasting memory. */
i915_sg_trim(st);
+ return st;
+err_sg:
+ sg_mark_end(sg);
+ if (sg != st->sgl) {
+ shmem_free_st(st, mapping, false, false);
+ } else {
+ mapping_clear_unevictable(mapping);
+ sg_free_table(st);
+ kfree(st);
+ }
+
+ /*
+ * shmemfs first checks if there is enough memory to allocate the page
+ * and reports ENOSPC should there be insufficient, along with the usual
+ * ENOMEM for a genuine allocation failure.
+ *
+ * We use ENOSPC in our driver to mean that we have run out of aperture
+ * space and so want to translate the error from shmemfs back to our
+ * usual understanding of ENOMEM.
+ */
+ if (ret == -ENOSPC)
+ ret = -ENOMEM;
+
+ return ERR_PTR(ret);
+}
+
+static int shmem_get_pages(struct drm_i915_gem_object *obj)
+{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
+ struct intel_memory_region *mem = obj->mm.region;
+ struct address_space *mapping = obj->base.filp->f_mapping;
+ const unsigned long page_count = obj->base.size / PAGE_SIZE;
+ unsigned int max_segment = i915_sg_segment_size();
+ struct sg_table *st;
+ struct sgt_iter sgt_iter;
+ struct page *page;
+ int ret;
+
+ /*
+ * Assert that the object is not currently in any GPU domain. As it
+ * wasn't in the GTT, there shouldn't be any way it could have been in
+ * a GPU cache
+ */
+ GEM_BUG_ON(obj->read_domains & I915_GEM_GPU_DOMAINS);
+ GEM_BUG_ON(obj->write_domain & I915_GEM_GPU_DOMAINS);
+
+rebuild_st:
+ st = shmem_alloc_st(i915, obj->base.size, mem, mapping, max_segment);
+ if (IS_ERR(st)) {
+ ret = PTR_ERR(st);
+ goto err_st;
+ }
+
ret = i915_gem_gtt_prepare_pages(obj, st);
if (ret) {
/*
@@ -168,6 +231,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
for_each_sgt_page(page, sgt_iter, st)
put_page(page);
sg_free_table(st);
+ kfree(st);
max_segment = PAGE_SIZE;
goto rebuild_st;
@@ -200,28 +264,12 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
if (IS_JSL_EHL(i915) && obj->flags & I915_BO_ALLOC_USER)
obj->cache_dirty = true;
- __i915_gem_object_set_pages(obj, st, sg_page_sizes);
+ __i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
return 0;
-err_sg:
- sg_mark_end(sg);
err_pages:
- mapping_clear_unevictable(mapping);
- if (sg != st->sgl) {
- struct pagevec pvec;
-
- pagevec_init(&pvec);
- for_each_sgt_page(page, sgt_iter, st) {
- if (!pagevec_add(&pvec, page))
- check_release_pagevec(&pvec);
- }
- if (pagevec_count(&pvec))
- check_release_pagevec(&pvec);
- }
- sg_free_table(st);
- kfree(st);
-
+ shmem_free_st(st, mapping, false, false);
/*
* shmemfs first checks if there is enough memory to allocate the page
* and reports ENOSPC should there be insufficient, along with the usual
@@ -231,6 +279,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
* space and so want to translate the error from shmemfs back to our
* usual understanding of ENOMEM.
*/
+err_st:
if (ret == -ENOSPC)
ret = -ENOMEM;
@@ -251,10 +300,8 @@ shmem_truncate(struct drm_i915_gem_object *obj)
obj->mm.pages = ERR_PTR(-EFAULT);
}
-static void
-shmem_writeback(struct drm_i915_gem_object *obj)
+static void __shmem_writeback(size_t size, struct address_space *mapping)
{
- struct address_space *mapping;
struct writeback_control wbc = {
.sync_mode = WB_SYNC_NONE,
.nr_to_write = SWAP_CLUSTER_MAX,
@@ -270,10 +317,9 @@ shmem_writeback(struct drm_i915_gem_object *obj)
* instead of invoking writeback so they are aged and paged out
* as normal.
*/
- mapping = obj->base.filp->f_mapping;
/* Begin writeback on each dirty page */
- for (i = 0; i < obj->base.size >> PAGE_SHIFT; i++) {
+ for (i = 0; i < size >> PAGE_SHIFT; i++) {
struct page *page;
page = find_lock_page(mapping, i);
@@ -296,6 +342,12 @@ shmem_writeback(struct drm_i915_gem_object *obj)
}
}
+static void
+shmem_writeback(struct drm_i915_gem_object *obj)
+{
+ __shmem_writeback(obj->base.size, obj->base.filp->f_mapping);
+}
+
void
__i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
struct sg_table *pages,
@@ -316,11 +368,6 @@ __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
void i915_gem_object_put_pages_shmem(struct drm_i915_gem_object *obj, struct sg_table *pages)
{
- struct sgt_iter sgt_iter;
- struct pagevec pvec;
- struct page *page;
-
- GEM_WARN_ON(IS_DGFX(to_i915(obj->base.dev)));
__i915_gem_object_release_shmem(obj, pages, true);
i915_gem_gtt_finish_pages(obj, pages);
@@ -328,25 +375,9 @@ void i915_gem_object_put_pages_shmem(struct drm_i915_gem_object *obj, struct sg_
if (i915_gem_object_needs_bit17_swizzle(obj))
i915_gem_object_save_bit_17_swizzle(obj, pages);
- mapping_clear_unevictable(file_inode(obj->base.filp)->i_mapping);
-
- pagevec_init(&pvec);
- for_each_sgt_page(page, sgt_iter, pages) {
- if (obj->mm.dirty)
- set_page_dirty(page);
-
- if (obj->mm.madv == I915_MADV_WILLNEED)
- mark_page_accessed(page);
-
- if (!pagevec_add(&pvec, page))
- check_release_pagevec(&pvec);
- }
- if (pagevec_count(&pvec))
- check_release_pagevec(&pvec);
+ shmem_free_st(pages, file_inode(obj->base.filp)->i_mapping,
+ obj->mm.dirty, obj->mm.madv == I915_MADV_WILLNEED);
obj->mm.dirty = false;
-
- sg_free_table(pages);
- kfree(pages);
}
static void
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 2/8] drm/i915/ttm: add tt shmem backend
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:23 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:23 UTC (permalink / raw)
To: intel-gfx
Cc: dri-devel, Thomas Hellström, Christian König, Oak Zeng
For cached objects we can allocate our pages directly in shmem. This
should make it possible(in a later patch) to utilise the existing
i915-gem shrinker code for such objects. For now this is still disabled.
v2(Thomas):
- Add optional try_to_writeback hook for objects. Importantly we need
to check if the object is even still shrinkable; in between us
dropping the shrinker LRU lock and acquiring the object lock it could for
example have been moved. Also we need to differentiate between
"lazy" shrinking and the immediate writeback mode. Also later we need to
handle objects which don't even have mm.pages, so bundling this into
put_pages() would require somehow handling that edge case, hence
just letting the ttm backend handle everything in try_to_writeback
doesn't seem too bad.
v3(Thomas):
- Likely a bad idea to touch the object from the unpopulate hook,
since it's not possible to hold a reference, without also creating
circular dependency, so likely this is too fragile. For now just
ensure we at least mark the pages as dirty/accessed when called from the
shrinker on WILLNEED objects.
- s/try_to_writeback/shrinker_release_pages, since this can do more
than just writeback.
- Get rid of do_backup boolean and just set the SWAPPED flag prior to
calling unpopulate.
- Keep shmem_tt as lowest priority for the TTM LRU bo_swapout walk, since
these just get skipped anyway. We can try to come up with something
better later.
v4(Thomas):
- s/PCI_DMA/DMA/. Also drop NO_KERNEL_MAPPING and NO_WARN, which
apparently doesn't do anything with streaming mappings.
- Just pass along the error for ->truncate, and assume nothing.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.h | 11 +-
.../gpu/drm/i915/gem/i915_gem_object_types.h | 4 +-
drivers/gpu/drm/i915/gem/i915_gem_pages.c | 6 +-
drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 18 +-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 17 +-
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 233 ++++++++++++++++--
6 files changed, 247 insertions(+), 42 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 9df3ee60604e..e641db297e0e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -93,7 +93,6 @@ void i915_gem_flush_free_objects(struct drm_i915_private *i915);
struct sg_table *
__i915_gem_object_unset_pages(struct drm_i915_gem_object *obj);
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
/**
* i915_gem_object_lookup_rcu - look up a temporary GEM object from its handle
@@ -449,7 +448,7 @@ i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
}
int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
+int i915_gem_object_truncate(struct drm_i915_gem_object *obj);
void i915_gem_object_writeback(struct drm_i915_gem_object *obj);
/**
@@ -612,6 +611,14 @@ int i915_gem_object_wait_migration(struct drm_i915_gem_object *obj,
bool i915_gem_object_placement_possible(struct drm_i915_gem_object *obj,
enum intel_memory_type type);
+struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment);
+void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup);
+void __shmem_writeback(size_t size, struct address_space *mapping);
+
#ifdef CONFIG_MMU_NOTIFIER
static inline bool
i915_gem_object_is_userptr(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 7c3da4e3e737..7dd5f804aab3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -54,8 +54,10 @@ struct drm_i915_gem_object_ops {
int (*get_pages)(struct drm_i915_gem_object *obj);
void (*put_pages)(struct drm_i915_gem_object *obj,
struct sg_table *pages);
- void (*truncate)(struct drm_i915_gem_object *obj);
+ int (*truncate)(struct drm_i915_gem_object *obj);
void (*writeback)(struct drm_i915_gem_object *obj);
+ int (*shrinker_release_pages)(struct drm_i915_gem_object *obj,
+ bool should_writeback);
int (*pread)(struct drm_i915_gem_object *obj,
const struct drm_i915_gem_pread *arg);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index 8eb1c3a6fc9c..ea6d9b3d2d6b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -158,11 +158,13 @@ int i915_gem_object_pin_pages_unlocked(struct drm_i915_gem_object *obj)
}
/* Immediately discard the backing storage */
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj)
+int i915_gem_object_truncate(struct drm_i915_gem_object *obj)
{
drm_gem_free_mmap_offset(&obj->base);
if (obj->ops->truncate)
- obj->ops->truncate(obj);
+ return obj->ops->truncate(obj);
+
+ return 0;
}
/* Try to discard unwanted pages */
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 36b711ae9e28..8375dc23f233 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -25,8 +25,8 @@ static void check_release_pagevec(struct pagevec *pvec)
cond_resched();
}
-static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
- bool dirty, bool backup)
+void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup)
{
struct sgt_iter sgt_iter;
struct pagevec pvec;
@@ -52,10 +52,10 @@ static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
kfree(st);
}
-static struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
- size_t size, struct intel_memory_region *mr,
- struct address_space *mapping,
- unsigned int max_segment)
+struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment)
{
const unsigned long page_count = size / PAGE_SIZE;
unsigned long i;
@@ -286,7 +286,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
return ret;
}
-static void
+static int
shmem_truncate(struct drm_i915_gem_object *obj)
{
/*
@@ -298,9 +298,11 @@ shmem_truncate(struct drm_i915_gem_object *obj)
shmem_truncate_range(file_inode(obj->base.filp), 0, (loff_t)-1);
obj->mm.madv = __I915_MADV_PURGED;
obj->mm.pages = ERR_PTR(-EFAULT);
+
+ return 0;
}
-static void __shmem_writeback(size_t size, struct address_space *mapping)
+void __shmem_writeback(size_t size, struct address_space *mapping)
{
struct writeback_control wbc = {
.sync_mode = WB_SYNC_NONE,
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index 5ab136ffdeb2..ae2a8d54b7a4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -56,19 +56,24 @@ static bool unsafe_drop_pages(struct drm_i915_gem_object *obj,
return false;
}
-static void try_to_writeback(struct drm_i915_gem_object *obj,
- unsigned int flags)
+static int try_to_writeback(struct drm_i915_gem_object *obj, unsigned int flags)
{
+ if (obj->ops->shrinker_release_pages)
+ return obj->ops->shrinker_release_pages(obj,
+ flags & I915_SHRINK_WRITEBACK);
+
switch (obj->mm.madv) {
case I915_MADV_DONTNEED:
i915_gem_object_truncate(obj);
- return;
+ return 0;
case __I915_MADV_PURGED:
- return;
+ return 0;
}
if (flags & I915_SHRINK_WRITEBACK)
i915_gem_object_writeback(obj);
+
+ return 0;
}
/**
@@ -222,8 +227,8 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww,
}
if (!__i915_gem_object_put_pages(obj)) {
- try_to_writeback(obj, shrink);
- count += obj->base.size >> PAGE_SHIFT;
+ if (!try_to_writeback(obj, shrink))
+ count += obj->base.size >> PAGE_SHIFT;
}
if (!ww)
i915_gem_object_unlock(obj);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 74a1ffd0d7dd..0fe1eb8f38e9 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -35,6 +35,8 @@
* @ttm: The base TTM page vector.
* @dev: The struct device used for dma mapping and unmapping.
* @cached_st: The cached scatter-gather table.
+ * @is_shmem: Set if using shmem.
+ * @filp: The shmem file, if using shmem backend.
*
* Note that DMA may be going on right up to the point where the page-
* vector is unpopulated in delayed destroy. Hence keep the
@@ -46,6 +48,9 @@ struct i915_ttm_tt {
struct ttm_tt ttm;
struct device *dev;
struct sg_table *cached_st;
+
+ bool is_shmem;
+ struct file *filp;
};
static const struct ttm_place sys_placement_flags = {
@@ -179,12 +184,88 @@ i915_ttm_placement_from_obj(const struct drm_i915_gem_object *obj,
placement->busy_placement = busy;
}
+static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev,
+ struct ttm_tt *ttm,
+ struct ttm_operation_ctx *ctx)
+{
+ struct drm_i915_private *i915 = container_of(bdev, typeof(*i915), bdev);
+ struct intel_memory_region *mr = i915->mm.regions[INTEL_MEMORY_SYSTEM];
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ const unsigned int max_segment = i915_sg_segment_size();
+ const size_t size = ttm->num_pages << PAGE_SHIFT;
+ struct file *filp = i915_tt->filp;
+ struct sgt_iter sgt_iter;
+ struct sg_table *st;
+ struct page *page;
+ unsigned long i;
+ int err;
+
+ if (!filp) {
+ struct address_space *mapping;
+ gfp_t mask;
+
+ filp = shmem_file_setup("i915-shmem-tt", size, VM_NORESERVE);
+ if (IS_ERR(filp))
+ return PTR_ERR(filp);
+
+ mask = GFP_HIGHUSER | __GFP_RECLAIMABLE;
+
+ mapping = filp->f_mapping;
+ mapping_set_gfp_mask(mapping, mask);
+ GEM_BUG_ON(!(mapping_gfp_mask(mapping) & __GFP_RECLAIM));
+
+ i915_tt->filp = filp;
+ }
+
+ st = shmem_alloc_st(i915, size, mr, filp->f_mapping, max_segment);
+ if (IS_ERR(st))
+ return PTR_ERR(st);
+
+ err = dma_map_sg_attrs(i915_tt->dev,
+ st->sgl, st->nents,
+ DMA_BIDIRECTIONAL,
+ DMA_ATTR_SKIP_CPU_SYNC);
+ if (err <= 0) {
+ err = -EINVAL;
+ goto err_free_st;
+ }
+
+ i = 0;
+ for_each_sgt_page(page, sgt_iter, st)
+ ttm->pages[i++] = page;
+
+ if (ttm->page_flags & TTM_TT_FLAG_SWAPPED)
+ ttm->page_flags &= ~TTM_TT_FLAG_SWAPPED;
+
+ i915_tt->cached_st = st;
+ return 0;
+
+err_free_st:
+ shmem_free_st(st, filp->f_mapping, false, false);
+ return err;
+}
+
+static void i915_ttm_tt_shmem_unpopulate(struct ttm_tt *ttm)
+{
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ bool backup = ttm->page_flags & TTM_TT_FLAG_SWAPPED;
+
+ dma_unmap_sg(i915_tt->dev, i915_tt->cached_st->sgl,
+ i915_tt->cached_st->nents,
+ DMA_BIDIRECTIONAL);
+
+ shmem_free_st(fetch_and_zero(&i915_tt->cached_st),
+ file_inode(i915_tt->filp)->i_mapping,
+ backup, backup);
+}
+
static struct ttm_tt *i915_ttm_tt_create(struct ttm_buffer_object *bo,
uint32_t page_flags)
{
struct ttm_resource_manager *man =
ttm_manager_type(bo->bdev, bo->resource->mem_type);
struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo);
+ enum ttm_caching caching = i915_ttm_select_tt_caching(obj);
struct i915_ttm_tt *i915_tt;
int ret;
@@ -196,36 +277,62 @@ static struct ttm_tt *i915_ttm_tt_create(struct ttm_buffer_object *bo,
man->use_tt)
page_flags |= TTM_TT_FLAG_ZERO_ALLOC;
- ret = ttm_tt_init(&i915_tt->ttm, bo, page_flags,
- i915_ttm_select_tt_caching(obj));
- if (ret) {
- kfree(i915_tt);
- return NULL;
+ if (i915_gem_object_is_shrinkable(obj) && caching == ttm_cached) {
+ page_flags |= TTM_TT_FLAG_EXTERNAL |
+ TTM_TT_FLAG_EXTERNAL_MAPPABLE;
+ i915_tt->is_shmem = true;
}
+ ret = ttm_tt_init(&i915_tt->ttm, bo, page_flags, caching);
+ if (ret)
+ goto err_free;
+
i915_tt->dev = obj->base.dev->dev;
return &i915_tt->ttm;
+
+err_free:
+ kfree(i915_tt);
+ return NULL;
+}
+
+static int i915_ttm_tt_populate(struct ttm_device *bdev,
+ struct ttm_tt *ttm,
+ struct ttm_operation_ctx *ctx)
+{
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+
+ if (i915_tt->is_shmem)
+ return i915_ttm_tt_shmem_populate(bdev, ttm, ctx);
+
+ return ttm_pool_alloc(&bdev->pool, ttm, ctx);
}
static void i915_ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm)
{
struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
- if (i915_tt->cached_st) {
- dma_unmap_sgtable(i915_tt->dev, i915_tt->cached_st,
- DMA_BIDIRECTIONAL, 0);
- sg_free_table(i915_tt->cached_st);
- kfree(i915_tt->cached_st);
- i915_tt->cached_st = NULL;
+ if (i915_tt->is_shmem) {
+ i915_ttm_tt_shmem_unpopulate(ttm);
+ } else {
+ if (i915_tt->cached_st) {
+ dma_unmap_sgtable(i915_tt->dev, i915_tt->cached_st,
+ DMA_BIDIRECTIONAL, 0);
+ sg_free_table(i915_tt->cached_st);
+ kfree(i915_tt->cached_st);
+ i915_tt->cached_st = NULL;
+ }
+ ttm_pool_free(&bdev->pool, ttm);
}
- ttm_pool_free(&bdev->pool, ttm);
}
static void i915_ttm_tt_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
{
struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ if (i915_tt->filp)
+ fput(i915_tt->filp);
+
ttm_tt_fini(ttm);
kfree(i915_tt);
}
@@ -235,6 +342,14 @@ static bool i915_ttm_eviction_valuable(struct ttm_buffer_object *bo,
{
struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo);
+ /*
+ * EXTERNAL objects should never be swapped out by TTM, instead we need
+ * to handle that ourselves. TTM will already skip such objects for us,
+ * but we would like to avoid grabbing locks for no good reason.
+ */
+ if (bo->ttm && bo->ttm->page_flags & TTM_TT_FLAG_EXTERNAL)
+ return -EBUSY;
+
/* Will do for now. Our pinned objects are still on TTM's LRU lists */
return i915_gem_object_evictable(obj);
}
@@ -328,9 +443,11 @@ static void i915_ttm_adjust_gem_after_move(struct drm_i915_gem_object *obj)
i915_gem_object_set_cache_coherency(obj, cache_level);
}
-static void i915_ttm_purge(struct drm_i915_gem_object *obj)
+static int i915_ttm_purge(struct drm_i915_gem_object *obj)
{
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
struct ttm_operation_ctx ctx = {
.interruptible = true,
.no_wait_gpu = false,
@@ -339,17 +456,74 @@ static void i915_ttm_purge(struct drm_i915_gem_object *obj)
int ret;
if (obj->mm.madv == __I915_MADV_PURGED)
- return;
+ return 0;
- /* TTM's purge interface. Note that we might be reentering. */
ret = ttm_bo_validate(bo, &place, &ctx);
- if (!ret) {
- obj->write_domain = 0;
- obj->read_domains = 0;
- i915_ttm_adjust_gem_after_move(obj);
- i915_ttm_free_cached_io_st(obj);
- obj->mm.madv = __I915_MADV_PURGED;
+ if (ret)
+ return ret;
+
+ if (bo->ttm && i915_tt->filp) {
+ /*
+ * The below fput(which eventually calls shmem_truncate) might
+ * be delayed by worker, so when directly called to purge the
+ * pages(like by the shrinker) we should try to be more
+ * aggressive and release the pages immediately.
+ */
+ shmem_truncate_range(file_inode(i915_tt->filp),
+ 0, (loff_t)-1);
+ fput(fetch_and_zero(&i915_tt->filp));
}
+
+ obj->write_domain = 0;
+ obj->read_domains = 0;
+ i915_ttm_adjust_gem_after_move(obj);
+ i915_ttm_free_cached_io_st(obj);
+ obj->mm.madv = __I915_MADV_PURGED;
+ return 0;
+}
+
+static int i915_ttm_shrinker_release_pages(struct drm_i915_gem_object *obj,
+ bool should_writeback)
+{
+ struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
+ struct ttm_operation_ctx ctx = {
+ .interruptible = true,
+ .no_wait_gpu = false,
+ };
+ struct ttm_placement place = {};
+ int ret;
+
+ if (!bo->ttm || bo->resource->mem_type != TTM_PL_SYSTEM)
+ return 0;
+
+ GEM_BUG_ON(!i915_tt->is_shmem);
+
+ if (!i915_tt->filp)
+ return 0;
+
+ switch (obj->mm.madv) {
+ case I915_MADV_DONTNEED:
+ return i915_ttm_purge(obj);
+ case __I915_MADV_PURGED:
+ return 0;
+ }
+
+ if (bo->ttm->page_flags & TTM_TT_FLAG_SWAPPED)
+ return 0;
+
+ bo->ttm->page_flags |= TTM_TT_FLAG_SWAPPED;
+ ret = ttm_bo_validate(bo, &place, &ctx);
+ if (ret) {
+ bo->ttm->page_flags &= ~TTM_TT_FLAG_SWAPPED;
+ return ret;
+ }
+
+ if (should_writeback)
+ __shmem_writeback(obj->base.size, i915_tt->filp->f_mapping);
+
+ return 0;
}
static void i915_ttm_swap_notify(struct ttm_buffer_object *bo)
@@ -620,6 +794,7 @@ static unsigned long i915_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
static struct ttm_device_funcs i915_ttm_bo_driver = {
.ttm_tt_create = i915_ttm_tt_create,
+ .ttm_tt_populate = i915_ttm_tt_populate,
.ttm_tt_unpopulate = i915_ttm_tt_unpopulate,
.ttm_tt_destroy = i915_ttm_tt_destroy,
.eviction_valuable = i915_ttm_eviction_valuable,
@@ -687,12 +862,17 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
}
if (!i915_gem_object_has_pages(obj)) {
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
+
/* Object either has a page vector or is an iomem object */
st = bo->ttm ? i915_ttm_tt_get_st(bo->ttm) : obj->ttm.cached_io_st;
if (IS_ERR(st))
return PTR_ERR(st);
__i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
+ if (!bo->ttm || !i915_tt->is_shmem)
+ i915_gem_object_make_unshrinkable(obj);
}
return ret;
@@ -772,6 +952,8 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
{
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
/*
* Don't manipulate the TTM LRUs while in TTM bo destruction.
@@ -784,7 +966,10 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
* Put on the correct LRU list depending on the MADV status
*/
spin_lock(&bo->bdev->lru_lock);
- if (obj->mm.madv != I915_MADV_WILLNEED) {
+ if (bo->ttm && i915_tt->filp) {
+ /* Try to keep shmem_tt from being considered for shrinking. */
+ bo->priority = TTM_MAX_BO_PRIORITY - 1;
+ } else if (obj->mm.madv != I915_MADV_WILLNEED) {
bo->priority = I915_TTM_PRIO_PURGE;
} else if (!i915_gem_object_has_pages(obj)) {
if (bo->priority < I915_TTM_PRIO_HAS_PAGES)
@@ -886,9 +1071,12 @@ static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
.truncate = i915_ttm_purge,
+ .shrinker_release_pages = i915_ttm_shrinker_release_pages,
+
.adjust_lru = i915_ttm_adjust_lru,
.delayed_free = i915_ttm_delayed_free,
.migrate = i915_ttm_migrate,
+
.mmap_offset = i915_ttm_mmap_offset,
.mmap_ops = &vm_ops_ttm,
};
@@ -943,7 +1131,6 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
obj->mm.region = intel_memory_region_get(mem);
INIT_LIST_HEAD(&obj->mm.region_link);
- i915_gem_object_make_unshrinkable(obj);
INIT_RADIX_TREE(&obj->ttm.get_io_page.radix, GFP_KERNEL | __GFP_NOWARN);
mutex_init(&obj->ttm.get_io_page.lock);
bo_type = (obj->flags & I915_BO_ALLOC_USER) ? ttm_bo_type_device :
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 2/8] drm/i915/ttm: add tt shmem backend
@ 2021-10-05 18:23 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:23 UTC (permalink / raw)
To: intel-gfx
Cc: dri-devel, Thomas Hellström, Christian König, Oak Zeng
For cached objects we can allocate our pages directly in shmem. This
should make it possible(in a later patch) to utilise the existing
i915-gem shrinker code for such objects. For now this is still disabled.
v2(Thomas):
- Add optional try_to_writeback hook for objects. Importantly we need
to check if the object is even still shrinkable; in between us
dropping the shrinker LRU lock and acquiring the object lock it could for
example have been moved. Also we need to differentiate between
"lazy" shrinking and the immediate writeback mode. Also later we need to
handle objects which don't even have mm.pages, so bundling this into
put_pages() would require somehow handling that edge case, hence
just letting the ttm backend handle everything in try_to_writeback
doesn't seem too bad.
v3(Thomas):
- Likely a bad idea to touch the object from the unpopulate hook,
since it's not possible to hold a reference, without also creating
circular dependency, so likely this is too fragile. For now just
ensure we at least mark the pages as dirty/accessed when called from the
shrinker on WILLNEED objects.
- s/try_to_writeback/shrinker_release_pages, since this can do more
than just writeback.
- Get rid of do_backup boolean and just set the SWAPPED flag prior to
calling unpopulate.
- Keep shmem_tt as lowest priority for the TTM LRU bo_swapout walk, since
these just get skipped anyway. We can try to come up with something
better later.
v4(Thomas):
- s/PCI_DMA/DMA/. Also drop NO_KERNEL_MAPPING and NO_WARN, which
apparently doesn't do anything with streaming mappings.
- Just pass along the error for ->truncate, and assume nothing.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Oak Zeng <oak.zeng@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.h | 11 +-
.../gpu/drm/i915/gem/i915_gem_object_types.h | 4 +-
drivers/gpu/drm/i915/gem/i915_gem_pages.c | 6 +-
drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 18 +-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 17 +-
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 233 ++++++++++++++++--
6 files changed, 247 insertions(+), 42 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 9df3ee60604e..e641db297e0e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -93,7 +93,6 @@ void i915_gem_flush_free_objects(struct drm_i915_private *i915);
struct sg_table *
__i915_gem_object_unset_pages(struct drm_i915_gem_object *obj);
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
/**
* i915_gem_object_lookup_rcu - look up a temporary GEM object from its handle
@@ -449,7 +448,7 @@ i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
}
int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
+int i915_gem_object_truncate(struct drm_i915_gem_object *obj);
void i915_gem_object_writeback(struct drm_i915_gem_object *obj);
/**
@@ -612,6 +611,14 @@ int i915_gem_object_wait_migration(struct drm_i915_gem_object *obj,
bool i915_gem_object_placement_possible(struct drm_i915_gem_object *obj,
enum intel_memory_type type);
+struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment);
+void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup);
+void __shmem_writeback(size_t size, struct address_space *mapping);
+
#ifdef CONFIG_MMU_NOTIFIER
static inline bool
i915_gem_object_is_userptr(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 7c3da4e3e737..7dd5f804aab3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -54,8 +54,10 @@ struct drm_i915_gem_object_ops {
int (*get_pages)(struct drm_i915_gem_object *obj);
void (*put_pages)(struct drm_i915_gem_object *obj,
struct sg_table *pages);
- void (*truncate)(struct drm_i915_gem_object *obj);
+ int (*truncate)(struct drm_i915_gem_object *obj);
void (*writeback)(struct drm_i915_gem_object *obj);
+ int (*shrinker_release_pages)(struct drm_i915_gem_object *obj,
+ bool should_writeback);
int (*pread)(struct drm_i915_gem_object *obj,
const struct drm_i915_gem_pread *arg);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index 8eb1c3a6fc9c..ea6d9b3d2d6b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -158,11 +158,13 @@ int i915_gem_object_pin_pages_unlocked(struct drm_i915_gem_object *obj)
}
/* Immediately discard the backing storage */
-void i915_gem_object_truncate(struct drm_i915_gem_object *obj)
+int i915_gem_object_truncate(struct drm_i915_gem_object *obj)
{
drm_gem_free_mmap_offset(&obj->base);
if (obj->ops->truncate)
- obj->ops->truncate(obj);
+ return obj->ops->truncate(obj);
+
+ return 0;
}
/* Try to discard unwanted pages */
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 36b711ae9e28..8375dc23f233 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -25,8 +25,8 @@ static void check_release_pagevec(struct pagevec *pvec)
cond_resched();
}
-static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
- bool dirty, bool backup)
+void shmem_free_st(struct sg_table *st, struct address_space *mapping,
+ bool dirty, bool backup)
{
struct sgt_iter sgt_iter;
struct pagevec pvec;
@@ -52,10 +52,10 @@ static void shmem_free_st(struct sg_table *st, struct address_space *mapping,
kfree(st);
}
-static struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
- size_t size, struct intel_memory_region *mr,
- struct address_space *mapping,
- unsigned int max_segment)
+struct sg_table *shmem_alloc_st(struct drm_i915_private *i915,
+ size_t size, struct intel_memory_region *mr,
+ struct address_space *mapping,
+ unsigned int max_segment)
{
const unsigned long page_count = size / PAGE_SIZE;
unsigned long i;
@@ -286,7 +286,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj)
return ret;
}
-static void
+static int
shmem_truncate(struct drm_i915_gem_object *obj)
{
/*
@@ -298,9 +298,11 @@ shmem_truncate(struct drm_i915_gem_object *obj)
shmem_truncate_range(file_inode(obj->base.filp), 0, (loff_t)-1);
obj->mm.madv = __I915_MADV_PURGED;
obj->mm.pages = ERR_PTR(-EFAULT);
+
+ return 0;
}
-static void __shmem_writeback(size_t size, struct address_space *mapping)
+void __shmem_writeback(size_t size, struct address_space *mapping)
{
struct writeback_control wbc = {
.sync_mode = WB_SYNC_NONE,
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index 5ab136ffdeb2..ae2a8d54b7a4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -56,19 +56,24 @@ static bool unsafe_drop_pages(struct drm_i915_gem_object *obj,
return false;
}
-static void try_to_writeback(struct drm_i915_gem_object *obj,
- unsigned int flags)
+static int try_to_writeback(struct drm_i915_gem_object *obj, unsigned int flags)
{
+ if (obj->ops->shrinker_release_pages)
+ return obj->ops->shrinker_release_pages(obj,
+ flags & I915_SHRINK_WRITEBACK);
+
switch (obj->mm.madv) {
case I915_MADV_DONTNEED:
i915_gem_object_truncate(obj);
- return;
+ return 0;
case __I915_MADV_PURGED:
- return;
+ return 0;
}
if (flags & I915_SHRINK_WRITEBACK)
i915_gem_object_writeback(obj);
+
+ return 0;
}
/**
@@ -222,8 +227,8 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww,
}
if (!__i915_gem_object_put_pages(obj)) {
- try_to_writeback(obj, shrink);
- count += obj->base.size >> PAGE_SHIFT;
+ if (!try_to_writeback(obj, shrink))
+ count += obj->base.size >> PAGE_SHIFT;
}
if (!ww)
i915_gem_object_unlock(obj);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 74a1ffd0d7dd..0fe1eb8f38e9 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -35,6 +35,8 @@
* @ttm: The base TTM page vector.
* @dev: The struct device used for dma mapping and unmapping.
* @cached_st: The cached scatter-gather table.
+ * @is_shmem: Set if using shmem.
+ * @filp: The shmem file, if using shmem backend.
*
* Note that DMA may be going on right up to the point where the page-
* vector is unpopulated in delayed destroy. Hence keep the
@@ -46,6 +48,9 @@ struct i915_ttm_tt {
struct ttm_tt ttm;
struct device *dev;
struct sg_table *cached_st;
+
+ bool is_shmem;
+ struct file *filp;
};
static const struct ttm_place sys_placement_flags = {
@@ -179,12 +184,88 @@ i915_ttm_placement_from_obj(const struct drm_i915_gem_object *obj,
placement->busy_placement = busy;
}
+static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev,
+ struct ttm_tt *ttm,
+ struct ttm_operation_ctx *ctx)
+{
+ struct drm_i915_private *i915 = container_of(bdev, typeof(*i915), bdev);
+ struct intel_memory_region *mr = i915->mm.regions[INTEL_MEMORY_SYSTEM];
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ const unsigned int max_segment = i915_sg_segment_size();
+ const size_t size = ttm->num_pages << PAGE_SHIFT;
+ struct file *filp = i915_tt->filp;
+ struct sgt_iter sgt_iter;
+ struct sg_table *st;
+ struct page *page;
+ unsigned long i;
+ int err;
+
+ if (!filp) {
+ struct address_space *mapping;
+ gfp_t mask;
+
+ filp = shmem_file_setup("i915-shmem-tt", size, VM_NORESERVE);
+ if (IS_ERR(filp))
+ return PTR_ERR(filp);
+
+ mask = GFP_HIGHUSER | __GFP_RECLAIMABLE;
+
+ mapping = filp->f_mapping;
+ mapping_set_gfp_mask(mapping, mask);
+ GEM_BUG_ON(!(mapping_gfp_mask(mapping) & __GFP_RECLAIM));
+
+ i915_tt->filp = filp;
+ }
+
+ st = shmem_alloc_st(i915, size, mr, filp->f_mapping, max_segment);
+ if (IS_ERR(st))
+ return PTR_ERR(st);
+
+ err = dma_map_sg_attrs(i915_tt->dev,
+ st->sgl, st->nents,
+ DMA_BIDIRECTIONAL,
+ DMA_ATTR_SKIP_CPU_SYNC);
+ if (err <= 0) {
+ err = -EINVAL;
+ goto err_free_st;
+ }
+
+ i = 0;
+ for_each_sgt_page(page, sgt_iter, st)
+ ttm->pages[i++] = page;
+
+ if (ttm->page_flags & TTM_TT_FLAG_SWAPPED)
+ ttm->page_flags &= ~TTM_TT_FLAG_SWAPPED;
+
+ i915_tt->cached_st = st;
+ return 0;
+
+err_free_st:
+ shmem_free_st(st, filp->f_mapping, false, false);
+ return err;
+}
+
+static void i915_ttm_tt_shmem_unpopulate(struct ttm_tt *ttm)
+{
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ bool backup = ttm->page_flags & TTM_TT_FLAG_SWAPPED;
+
+ dma_unmap_sg(i915_tt->dev, i915_tt->cached_st->sgl,
+ i915_tt->cached_st->nents,
+ DMA_BIDIRECTIONAL);
+
+ shmem_free_st(fetch_and_zero(&i915_tt->cached_st),
+ file_inode(i915_tt->filp)->i_mapping,
+ backup, backup);
+}
+
static struct ttm_tt *i915_ttm_tt_create(struct ttm_buffer_object *bo,
uint32_t page_flags)
{
struct ttm_resource_manager *man =
ttm_manager_type(bo->bdev, bo->resource->mem_type);
struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo);
+ enum ttm_caching caching = i915_ttm_select_tt_caching(obj);
struct i915_ttm_tt *i915_tt;
int ret;
@@ -196,36 +277,62 @@ static struct ttm_tt *i915_ttm_tt_create(struct ttm_buffer_object *bo,
man->use_tt)
page_flags |= TTM_TT_FLAG_ZERO_ALLOC;
- ret = ttm_tt_init(&i915_tt->ttm, bo, page_flags,
- i915_ttm_select_tt_caching(obj));
- if (ret) {
- kfree(i915_tt);
- return NULL;
+ if (i915_gem_object_is_shrinkable(obj) && caching == ttm_cached) {
+ page_flags |= TTM_TT_FLAG_EXTERNAL |
+ TTM_TT_FLAG_EXTERNAL_MAPPABLE;
+ i915_tt->is_shmem = true;
}
+ ret = ttm_tt_init(&i915_tt->ttm, bo, page_flags, caching);
+ if (ret)
+ goto err_free;
+
i915_tt->dev = obj->base.dev->dev;
return &i915_tt->ttm;
+
+err_free:
+ kfree(i915_tt);
+ return NULL;
+}
+
+static int i915_ttm_tt_populate(struct ttm_device *bdev,
+ struct ttm_tt *ttm,
+ struct ttm_operation_ctx *ctx)
+{
+ struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+
+ if (i915_tt->is_shmem)
+ return i915_ttm_tt_shmem_populate(bdev, ttm, ctx);
+
+ return ttm_pool_alloc(&bdev->pool, ttm, ctx);
}
static void i915_ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm)
{
struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
- if (i915_tt->cached_st) {
- dma_unmap_sgtable(i915_tt->dev, i915_tt->cached_st,
- DMA_BIDIRECTIONAL, 0);
- sg_free_table(i915_tt->cached_st);
- kfree(i915_tt->cached_st);
- i915_tt->cached_st = NULL;
+ if (i915_tt->is_shmem) {
+ i915_ttm_tt_shmem_unpopulate(ttm);
+ } else {
+ if (i915_tt->cached_st) {
+ dma_unmap_sgtable(i915_tt->dev, i915_tt->cached_st,
+ DMA_BIDIRECTIONAL, 0);
+ sg_free_table(i915_tt->cached_st);
+ kfree(i915_tt->cached_st);
+ i915_tt->cached_st = NULL;
+ }
+ ttm_pool_free(&bdev->pool, ttm);
}
- ttm_pool_free(&bdev->pool, ttm);
}
static void i915_ttm_tt_destroy(struct ttm_device *bdev, struct ttm_tt *ttm)
{
struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm);
+ if (i915_tt->filp)
+ fput(i915_tt->filp);
+
ttm_tt_fini(ttm);
kfree(i915_tt);
}
@@ -235,6 +342,14 @@ static bool i915_ttm_eviction_valuable(struct ttm_buffer_object *bo,
{
struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo);
+ /*
+ * EXTERNAL objects should never be swapped out by TTM, instead we need
+ * to handle that ourselves. TTM will already skip such objects for us,
+ * but we would like to avoid grabbing locks for no good reason.
+ */
+ if (bo->ttm && bo->ttm->page_flags & TTM_TT_FLAG_EXTERNAL)
+ return -EBUSY;
+
/* Will do for now. Our pinned objects are still on TTM's LRU lists */
return i915_gem_object_evictable(obj);
}
@@ -328,9 +443,11 @@ static void i915_ttm_adjust_gem_after_move(struct drm_i915_gem_object *obj)
i915_gem_object_set_cache_coherency(obj, cache_level);
}
-static void i915_ttm_purge(struct drm_i915_gem_object *obj)
+static int i915_ttm_purge(struct drm_i915_gem_object *obj)
{
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
struct ttm_operation_ctx ctx = {
.interruptible = true,
.no_wait_gpu = false,
@@ -339,17 +456,74 @@ static void i915_ttm_purge(struct drm_i915_gem_object *obj)
int ret;
if (obj->mm.madv == __I915_MADV_PURGED)
- return;
+ return 0;
- /* TTM's purge interface. Note that we might be reentering. */
ret = ttm_bo_validate(bo, &place, &ctx);
- if (!ret) {
- obj->write_domain = 0;
- obj->read_domains = 0;
- i915_ttm_adjust_gem_after_move(obj);
- i915_ttm_free_cached_io_st(obj);
- obj->mm.madv = __I915_MADV_PURGED;
+ if (ret)
+ return ret;
+
+ if (bo->ttm && i915_tt->filp) {
+ /*
+ * The below fput(which eventually calls shmem_truncate) might
+ * be delayed by worker, so when directly called to purge the
+ * pages(like by the shrinker) we should try to be more
+ * aggressive and release the pages immediately.
+ */
+ shmem_truncate_range(file_inode(i915_tt->filp),
+ 0, (loff_t)-1);
+ fput(fetch_and_zero(&i915_tt->filp));
}
+
+ obj->write_domain = 0;
+ obj->read_domains = 0;
+ i915_ttm_adjust_gem_after_move(obj);
+ i915_ttm_free_cached_io_st(obj);
+ obj->mm.madv = __I915_MADV_PURGED;
+ return 0;
+}
+
+static int i915_ttm_shrinker_release_pages(struct drm_i915_gem_object *obj,
+ bool should_writeback)
+{
+ struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
+ struct ttm_operation_ctx ctx = {
+ .interruptible = true,
+ .no_wait_gpu = false,
+ };
+ struct ttm_placement place = {};
+ int ret;
+
+ if (!bo->ttm || bo->resource->mem_type != TTM_PL_SYSTEM)
+ return 0;
+
+ GEM_BUG_ON(!i915_tt->is_shmem);
+
+ if (!i915_tt->filp)
+ return 0;
+
+ switch (obj->mm.madv) {
+ case I915_MADV_DONTNEED:
+ return i915_ttm_purge(obj);
+ case __I915_MADV_PURGED:
+ return 0;
+ }
+
+ if (bo->ttm->page_flags & TTM_TT_FLAG_SWAPPED)
+ return 0;
+
+ bo->ttm->page_flags |= TTM_TT_FLAG_SWAPPED;
+ ret = ttm_bo_validate(bo, &place, &ctx);
+ if (ret) {
+ bo->ttm->page_flags &= ~TTM_TT_FLAG_SWAPPED;
+ return ret;
+ }
+
+ if (should_writeback)
+ __shmem_writeback(obj->base.size, i915_tt->filp->f_mapping);
+
+ return 0;
}
static void i915_ttm_swap_notify(struct ttm_buffer_object *bo)
@@ -620,6 +794,7 @@ static unsigned long i915_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
static struct ttm_device_funcs i915_ttm_bo_driver = {
.ttm_tt_create = i915_ttm_tt_create,
+ .ttm_tt_populate = i915_ttm_tt_populate,
.ttm_tt_unpopulate = i915_ttm_tt_unpopulate,
.ttm_tt_destroy = i915_ttm_tt_destroy,
.eviction_valuable = i915_ttm_eviction_valuable,
@@ -687,12 +862,17 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
}
if (!i915_gem_object_has_pages(obj)) {
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
+
/* Object either has a page vector or is an iomem object */
st = bo->ttm ? i915_ttm_tt_get_st(bo->ttm) : obj->ttm.cached_io_st;
if (IS_ERR(st))
return PTR_ERR(st);
__i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
+ if (!bo->ttm || !i915_tt->is_shmem)
+ i915_gem_object_make_unshrinkable(obj);
}
return ret;
@@ -772,6 +952,8 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
{
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
+ struct i915_ttm_tt *i915_tt =
+ container_of(bo->ttm, typeof(*i915_tt), ttm);
/*
* Don't manipulate the TTM LRUs while in TTM bo destruction.
@@ -784,7 +966,10 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
* Put on the correct LRU list depending on the MADV status
*/
spin_lock(&bo->bdev->lru_lock);
- if (obj->mm.madv != I915_MADV_WILLNEED) {
+ if (bo->ttm && i915_tt->filp) {
+ /* Try to keep shmem_tt from being considered for shrinking. */
+ bo->priority = TTM_MAX_BO_PRIORITY - 1;
+ } else if (obj->mm.madv != I915_MADV_WILLNEED) {
bo->priority = I915_TTM_PRIO_PURGE;
} else if (!i915_gem_object_has_pages(obj)) {
if (bo->priority < I915_TTM_PRIO_HAS_PAGES)
@@ -886,9 +1071,12 @@ static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
.truncate = i915_ttm_purge,
+ .shrinker_release_pages = i915_ttm_shrinker_release_pages,
+
.adjust_lru = i915_ttm_adjust_lru,
.delayed_free = i915_ttm_delayed_free,
.migrate = i915_ttm_migrate,
+
.mmap_offset = i915_ttm_mmap_offset,
.mmap_ops = &vm_ops_ttm,
};
@@ -943,7 +1131,6 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
obj->mm.region = intel_memory_region_get(mem);
INIT_LIST_HEAD(&obj->mm.region_link);
- i915_gem_object_make_unshrinkable(obj);
INIT_RADIX_TREE(&obj->ttm.get_io_page.radix, GFP_KERNEL | __GFP_NOWARN);
mutex_init(&obj->ttm.get_io_page.lock);
bo_type = (obj->flags & I915_BO_ALLOC_USER) ? ttm_bo_type_device :
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 3/8] drm/i915/gtt: drop unneeded make_unshrinkable
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
We already do this when mapping the pages.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gt/gen6_ppgtt.c | 1 -
drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
index 890191f286e3..baea9770200a 100644
--- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
@@ -185,7 +185,6 @@ static void gen6_alloc_va_range(struct i915_address_space *vm,
pt = stash->pt[0];
__i915_gem_object_pin_pages(pt->base);
- i915_gem_object_make_unshrinkable(pt->base);
fill32_px(pt, vm->scratch[0]->encode);
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 037a9a6e4889..8af2f709571c 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -301,7 +301,6 @@ static void __gen8_ppgtt_alloc(struct i915_address_space * const vm,
pt = stash->pt[!!lvl];
__i915_gem_object_pin_pages(pt->base);
- i915_gem_object_make_unshrinkable(pt->base);
fill_px(pt, vm->scratch[lvl]->encode);
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 3/8] drm/i915/gtt: drop unneeded make_unshrinkable
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
We already do this when mapping the pages.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gt/gen6_ppgtt.c | 1 -
drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
index 890191f286e3..baea9770200a 100644
--- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
@@ -185,7 +185,6 @@ static void gen6_alloc_va_range(struct i915_address_space *vm,
pt = stash->pt[0];
__i915_gem_object_pin_pages(pt->base);
- i915_gem_object_make_unshrinkable(pt->base);
fill32_px(pt, vm->scratch[0]->encode);
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 037a9a6e4889..8af2f709571c 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -301,7 +301,6 @@ static void __gen8_ppgtt_alloc(struct i915_address_space * const vm,
pt = stash->pt[!!lvl];
__i915_gem_object_pin_pages(pt->base);
- i915_gem_object_make_unshrinkable(pt->base);
fill_px(pt, vm->scratch[lvl]->encode);
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 4/8] drm/i915: drop unneeded make_unshrinkable in free_object
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
The comment here is no longer accurate, since the current shrinker code
requires a full ref before touching any objects. Also unset_pages()
should already do the required make_unshrinkable() for us, if needed,
which is also nicely balanced with set_pages().
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.c | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 76ce6a1500bc..1dc3c1940c32 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -337,15 +337,6 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj)
*/
atomic_inc(&i915->mm.free_count);
- /*
- * This serializes freeing with the shrinker. Since the free
- * is delayed, first by RCU then by the workqueue, we want the
- * shrinker to be able to free pages of unreferenced objects,
- * or else we may oom whilst there are plenty of deferred
- * freed objects.
- */
- i915_gem_object_make_unshrinkable(obj);
-
/*
* Since we require blocking on struct_mutex to unbind the freed
* object from the GPU before releasing resources back to the
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 4/8] drm/i915: drop unneeded make_unshrinkable in free_object
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
The comment here is no longer accurate, since the current shrinker code
requires a full ref before touching any objects. Also unset_pages()
should already do the required make_unshrinkable() for us, if needed,
which is also nicely balanced with set_pages().
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.c | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 76ce6a1500bc..1dc3c1940c32 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -337,15 +337,6 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj)
*/
atomic_inc(&i915->mm.free_count);
- /*
- * This serializes freeing with the shrinker. Since the free
- * is delayed, first by RCU then by the workqueue, we want the
- * shrinker to be able to free pages of unreferenced objects,
- * or else we may oom whilst there are plenty of deferred
- * freed objects.
- */
- i915_gem_object_make_unshrinkable(obj);
-
/*
* Since we require blocking on struct_mutex to unbind the freed
* object from the GPU before releasing resources back to the
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 5/8] drm/i915: add some kernel-doc for shrink_pin and friends
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
Attempt to document shrink_pin and the other relevant interfaces that
interact with it, before we start messing with it.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
.../gpu/drm/i915/gem/i915_gem_object_types.h | 24 +++++++++++++-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 31 +++++++++++++++++++
2 files changed, 54 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 7dd5f804aab3..f4233c4e8d2e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -461,6 +461,28 @@ struct drm_i915_gem_object {
* instead go through the pin/unpin interfaces.
*/
atomic_t pages_pin_count;
+
+ /**
+ * @shrink_pin: Prevents the pages from being made visible to
+ * the shrinker, while the shrink_pin is non-zero. Most users
+ * should pretty much never have to care about this, outside of
+ * some special use cases.
+ *
+ * By default most objects will start out as visible to the
+ * shrinker(if I915_GEM_OBJECT_IS_SHRINKABLE) as soon as the
+ * backing pages are attached to the object, like in
+ * __i915_gem_object_set_pages(). They will then be removed the
+ * shrinker list once the pages are released.
+ *
+ * The @shrink_pin is incremented by calling
+ * i915_gem_object_make_unshrinkable(), which will also remove
+ * the object from the shrinker list, if the pin count was zero.
+ *
+ * Callers will then typically call
+ * i915_gem_object_make_shrinkable() or
+ * i915_gem_object_make_purgeable() to decrement the pin count,
+ * and make the pages visible again.
+ */
atomic_t shrink_pin;
/**
@@ -522,7 +544,7 @@ struct drm_i915_gem_object {
struct i915_gem_object_page_iter get_dma_page;
/**
- * Element within i915->mm.unbound_list or i915->mm.bound_list,
+ * Element within i915->mm.shrink_list or i915->mm.purge_list,
* locked by i915->mm.obj_lock.
*/
struct list_head link;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index ae2a8d54b7a4..66121fedc655 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -463,6 +463,16 @@ void i915_gem_shrinker_taints_mutex(struct drm_i915_private *i915,
#define obj_to_i915(obj__) to_i915((obj__)->base.dev)
+/**
+ * i915_gem_object_make_unshrinkable - Hide the object from the shrinker. By
+ * default all object types that support shrinking(see IS_SHRINKABLE), will also
+ * make the object visible to the shrinker after allocating the system memory
+ * pages.
+ * @obj: The GEM object.
+ *
+ * This is typically used for special kernel internal objects that can't be
+ * easily processed by the shrinker, like if they are perma-pinned.
+ */
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
{
struct drm_i915_private *i915 = obj_to_i915(obj);
@@ -513,12 +523,33 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
+/**
+ * i915_gem_object_make_shrinkable - Move the object to the tail of the
+ * shrinkable list. Objects on this list might be swapped out. Used with
+ * WILLNEED objects.
+ * @obj: The GEM object.
+ *
+ * MUST only be called on objects which have backing pages.
+ *
+ * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
+ */
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
{
__i915_gem_object_make_shrinkable(obj,
&obj_to_i915(obj)->mm.shrink_list);
}
+/**
+ * i915_gem_object_make_purgeable - Move the object to the tail of the purgeable
+ * list. Used with DONTNEED objects. Unlike with shrinkable objects, the
+ * shrinker will attempt to discard the backing pages, instead of trying to swap
+ * them out.
+ * @obj: The GEM object.
+ *
+ * MUST only be called on objects which have backing pages.
+ *
+ * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
+ */
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
{
__i915_gem_object_make_shrinkable(obj,
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 5/8] drm/i915: add some kernel-doc for shrink_pin and friends
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
Attempt to document shrink_pin and the other relevant interfaces that
interact with it, before we start messing with it.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
.../gpu/drm/i915/gem/i915_gem_object_types.h | 24 +++++++++++++-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 31 +++++++++++++++++++
2 files changed, 54 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 7dd5f804aab3..f4233c4e8d2e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -461,6 +461,28 @@ struct drm_i915_gem_object {
* instead go through the pin/unpin interfaces.
*/
atomic_t pages_pin_count;
+
+ /**
+ * @shrink_pin: Prevents the pages from being made visible to
+ * the shrinker, while the shrink_pin is non-zero. Most users
+ * should pretty much never have to care about this, outside of
+ * some special use cases.
+ *
+ * By default most objects will start out as visible to the
+ * shrinker(if I915_GEM_OBJECT_IS_SHRINKABLE) as soon as the
+ * backing pages are attached to the object, like in
+ * __i915_gem_object_set_pages(). They will then be removed the
+ * shrinker list once the pages are released.
+ *
+ * The @shrink_pin is incremented by calling
+ * i915_gem_object_make_unshrinkable(), which will also remove
+ * the object from the shrinker list, if the pin count was zero.
+ *
+ * Callers will then typically call
+ * i915_gem_object_make_shrinkable() or
+ * i915_gem_object_make_purgeable() to decrement the pin count,
+ * and make the pages visible again.
+ */
atomic_t shrink_pin;
/**
@@ -522,7 +544,7 @@ struct drm_i915_gem_object {
struct i915_gem_object_page_iter get_dma_page;
/**
- * Element within i915->mm.unbound_list or i915->mm.bound_list,
+ * Element within i915->mm.shrink_list or i915->mm.purge_list,
* locked by i915->mm.obj_lock.
*/
struct list_head link;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index ae2a8d54b7a4..66121fedc655 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -463,6 +463,16 @@ void i915_gem_shrinker_taints_mutex(struct drm_i915_private *i915,
#define obj_to_i915(obj__) to_i915((obj__)->base.dev)
+/**
+ * i915_gem_object_make_unshrinkable - Hide the object from the shrinker. By
+ * default all object types that support shrinking(see IS_SHRINKABLE), will also
+ * make the object visible to the shrinker after allocating the system memory
+ * pages.
+ * @obj: The GEM object.
+ *
+ * This is typically used for special kernel internal objects that can't be
+ * easily processed by the shrinker, like if they are perma-pinned.
+ */
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
{
struct drm_i915_private *i915 = obj_to_i915(obj);
@@ -513,12 +523,33 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
+/**
+ * i915_gem_object_make_shrinkable - Move the object to the tail of the
+ * shrinkable list. Objects on this list might be swapped out. Used with
+ * WILLNEED objects.
+ * @obj: The GEM object.
+ *
+ * MUST only be called on objects which have backing pages.
+ *
+ * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
+ */
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
{
__i915_gem_object_make_shrinkable(obj,
&obj_to_i915(obj)->mm.shrink_list);
}
+/**
+ * i915_gem_object_make_purgeable - Move the object to the tail of the purgeable
+ * list. Used with DONTNEED objects. Unlike with shrinkable objects, the
+ * shrinker will attempt to discard the backing pages, instead of trying to swap
+ * them out.
+ * @obj: The GEM object.
+ *
+ * MUST only be called on objects which have backing pages.
+ *
+ * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
+ */
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
{
__i915_gem_object_make_shrinkable(obj,
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 6/8] drm/i915/ttm: move shrinker management into adjust_lru
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
We currently just evict lmem objects to system memory when under memory
pressure. For this case we might lack the usual object mm.pages, which
effectively hides the pages from the i915-gem shrinker, until we
actually "attach" the TT to the object, or in the case of lmem-only
objects it just gets migrated back to lmem when touched again.
For all cases we can just adjust the i915 shrinker LRU each time we also
adjust the TTM LRU. The two cases we care about are:
1) When something is moved by TTM, including when initially populating
an object. Importantly this covers the case where TTM moves something from
lmem <-> smem, outside of the normal get_pages() interface, which
should still ensure the shmem pages underneath are reclaimable.
2) When calling into i915_gem_object_unlock(). The unlock should
ensure the object is removed from the shinker LRU, if it was indeed
swapped out, or just purged, when the shrinker drops the object lock.
v2(Thomas):
- Handle managing the shrinker LRU in adjust_lru, where it is always
safe to touch the object.
v3(Thomas):
- Pretty much a re-write. This time piggy back off the shrink_pin
stuff, which actually seems to fit quite well for what we want here.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.h | 8 +++
.../gpu/drm/i915/gem/i915_gem_object_types.h | 25 ++++++-
drivers/gpu/drm/i915/gem/i915_gem_pages.c | 5 +-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 45 +++++++++++--
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 65 +++++++++++++++++--
5 files changed, 130 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index e641db297e0e..3eac8cf2ae10 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -294,6 +294,12 @@ i915_gem_object_is_shrinkable(const struct drm_i915_gem_object *obj)
return i915_gem_object_type_has(obj, I915_GEM_OBJECT_IS_SHRINKABLE);
}
+static inline bool
+i915_gem_object_has_self_managed_shrink_list(const struct drm_i915_gem_object *obj)
+{
+ return i915_gem_object_type_has(obj, I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST);
+}
+
static inline bool
i915_gem_object_is_proxy(const struct drm_i915_gem_object *obj)
{
@@ -531,6 +537,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
+void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
+void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
static inline bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index f4233c4e8d2e..9dbbca682a77 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -34,9 +34,11 @@ struct i915_lut_handle {
struct drm_i915_gem_object_ops {
unsigned int flags;
-#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
-#define I915_GEM_OBJECT_IS_PROXY BIT(2)
-#define I915_GEM_OBJECT_NO_MMAP BIT(3)
+#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
+/* Skip the shrinker management in set_pages/unset_pages */
+#define I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST BIT(2)
+#define I915_GEM_OBJECT_IS_PROXY BIT(3)
+#define I915_GEM_OBJECT_NO_MMAP BIT(4)
/* Interface between the GEM object and its backing storage.
* get_pages() is called once prior to the use of the associated set
@@ -485,6 +487,23 @@ struct drm_i915_gem_object {
*/
atomic_t shrink_pin;
+ /**
+ * @ttm_shrinker_status: Whether TTM is currently holding a
+ * shrink_pin for this object. Protected by the object lock.
+ *
+ * I915_TTM_SHRINKER_UNPINNED: We don't have an extra shrink_pin
+ * for this object. The underlying pages or ttm_tt is using
+ * shmem pages underneath, and as such this object might be
+ * currently visible to the shrinker.
+ *
+ * I915_TTM_SHRINKER_PINNED: We are holding shrink_pin for this
+ * object, which prevents the shrinker from seeing this object.
+ * The object is not currently using shmem pages undearneath.
+ */
+#define I915_TTM_SHRINKER_UNPINNED 1
+#define I915_TTM_SHRINKER_PINNED 2
+ int ttm_shrinker_status;
+
/**
* Priority list of potential placements for this object.
*/
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index ea6d9b3d2d6b..308e22a80af4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -68,7 +68,7 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
shrinkable = false;
}
- if (shrinkable) {
+ if (shrinkable && !i915_gem_object_has_self_managed_shrink_list(obj)) {
struct list_head *list;
unsigned long flags;
@@ -210,7 +210,8 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj)
if (i915_gem_object_is_volatile(obj))
obj->mm.madv = I915_MADV_WILLNEED;
- i915_gem_object_make_unshrinkable(obj);
+ if (!i915_gem_object_has_self_managed_shrink_list(obj))
+ i915_gem_object_make_unshrinkable(obj);
if (obj->mm.mapping) {
unmap_object(obj, page_mask_bits(obj->mm.mapping));
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index 66121fedc655..dde0a5c232f8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -497,13 +497,12 @@ void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
-static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
- struct list_head *head)
+static void ___i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
+ struct list_head *head)
{
struct drm_i915_private *i915 = obj_to_i915(obj);
unsigned long flags;
- GEM_BUG_ON(!i915_gem_object_has_pages(obj));
if (!i915_gem_object_is_shrinkable(obj))
return;
@@ -523,6 +522,38 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
+/**
+ * __i915_gem_object_make_shrinkable - Move the object to the tail of the
+ * shrinkable list. Objects on this list might be swapped out. Used with
+ * WILLNEED objects.
+ * @obj: The GEM object.
+ *
+ * DO NOT USE. This is intended to be called on very special objects that don't
+ * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
+ * underneath.
+ */
+void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
+{
+ ___i915_gem_object_make_shrinkable(obj,
+ &obj_to_i915(obj)->mm.shrink_list);
+}
+
+/**
+ * __i915_gem_object_make_purgeable - Move the object to the tail of the
+ * purgeable list. Objects on this list might be swapped out. Used with
+ * DONTNEED objects.
+ * @obj: The GEM object.
+ *
+ * DO NOT USE. This is intended to be called on very special objects that don't
+ * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
+ * underneath.
+ */
+void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
+{
+ ___i915_gem_object_make_shrinkable(obj,
+ &obj_to_i915(obj)->mm.purge_list);
+}
+
/**
* i915_gem_object_make_shrinkable - Move the object to the tail of the
* shrinkable list. Objects on this list might be swapped out. Used with
@@ -535,8 +566,8 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
*/
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
{
- __i915_gem_object_make_shrinkable(obj,
- &obj_to_i915(obj)->mm.shrink_list);
+ GEM_BUG_ON(!i915_gem_object_has_pages(obj));
+ __i915_gem_object_make_shrinkable(obj);
}
/**
@@ -552,6 +583,6 @@ void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
*/
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
{
- __i915_gem_object_make_shrinkable(obj,
- &obj_to_i915(obj)->mm.purge_list);
+ GEM_BUG_ON(!i915_gem_object_has_pages(obj));
+ __i915_gem_object_make_purgeable(obj);
}
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 0fe1eb8f38e9..98b7ead1a008 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -851,7 +851,6 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
return i915_ttm_err_to_gem(ret);
}
- i915_ttm_adjust_lru(obj);
if (bo->ttm && !ttm_tt_is_populated(bo->ttm)) {
ret = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
if (ret)
@@ -871,10 +870,9 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
return PTR_ERR(st);
__i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
- if (!bo->ttm || !i915_tt->is_shmem)
- i915_gem_object_make_unshrinkable(obj);
}
+ i915_ttm_adjust_lru(obj);
return ret;
}
@@ -945,8 +943,6 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
* If the object is not destroyed next, The TTM eviction logic
* and shrinkers will move it out if needed.
*/
-
- i915_ttm_adjust_lru(obj);
}
static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
@@ -954,6 +950,9 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
struct i915_ttm_tt *i915_tt =
container_of(bo->ttm, typeof(*i915_tt), ttm);
+ bool shrinkable =
+ bo->ttm && i915_tt->filp && ttm_tt_is_populated(bo->ttm);
+ int status;
/*
* Don't manipulate the TTM LRUs while in TTM bo destruction.
@@ -962,11 +961,42 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
if (!kref_read(&bo->kref))
return;
+ /*
+ * We skip managing the shrinker LRU in set_pages() and just manage
+ * everything here. This does at least solve the issue with having
+ * temporary shmem mappings(like with evicted lmem) not being visible to
+ * the shrinker. Only our shmem objects are shrinkable, everything else
+ * we keep as unshrinkable.
+ *
+ * To make sure everything plays nice we keep an extra shrink pin in TTM
+ * if the underlying pages are not currently shrinkable. Once we release
+ * our pin, like when the pages are moved to shmem, the pages will then
+ * be added to the shrinker LRU, assuming the caller isn't also holding
+ * a pin.
+ *
+ * TODO: consider maybe also bumping the shrinker list here when we have
+ * already unpinned it, which should give us something more like an LRU.
+ */
+ status = obj->mm.ttm_shrinker_status;
+ if (shrinkable) {
+ if (status != I915_TTM_SHRINKER_UNPINNED) {
+ if (obj->mm.madv == I915_MADV_WILLNEED)
+ __i915_gem_object_make_shrinkable(obj);
+ else
+ __i915_gem_object_make_purgeable(obj);
+ status = I915_TTM_SHRINKER_UNPINNED;
+ }
+ } else if (status != I915_TTM_SHRINKER_PINNED) {
+ i915_gem_object_make_unshrinkable(obj);
+ status = I915_TTM_SHRINKER_PINNED;
+ }
+ obj->mm.ttm_shrinker_status = status;
+
/*
* Put on the correct LRU list depending on the MADV status
*/
spin_lock(&bo->bdev->lru_lock);
- if (bo->ttm && i915_tt->filp) {
+ if (shrinkable) {
/* Try to keep shmem_tt from being considered for shrinking. */
bo->priority = TTM_MAX_BO_PRIORITY - 1;
} else if (obj->mm.madv != I915_MADV_WILLNEED) {
@@ -1067,6 +1097,7 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.name = "i915_gem_object_ttm",
+ .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
@@ -1089,6 +1120,19 @@ void i915_ttm_bo_destroy(struct ttm_buffer_object *bo)
mutex_destroy(&obj->ttm.get_io_page.lock);
if (obj->ttm.created) {
+ /*
+ * We freely manage the shrinker LRU outide of the mm.pages life
+ * cycle. As a result when destroying the object we should be
+ * extra paranoid and ensure we remove it from the LRU, before
+ * we free the object.
+ *
+ * Touching the ttm_shrinker_status outside of the object lock
+ * here should be safe now that the last GEM object ref was
+ * dropped.
+ */
+ if (obj->mm.ttm_shrinker_status == I915_TTM_SHRINKER_UNPINNED)
+ i915_gem_object_make_unshrinkable(obj);
+
i915_ttm_backup_free(obj);
/* This releases all gem object bindings to the backend. */
@@ -1141,6 +1185,15 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
/* Forcing the page size is kernel internal only */
GEM_BUG_ON(page_size && obj->mm.n_placements);
+ /*
+ * Keep an extra shrink pin to prevent the object from being made
+ * shrinkable too early. If the ttm_tt is ever allocated in shmem, we
+ * drop the pin. The TTM backend manages the shrinker LRU itself,
+ * outside of the normal mm.pages life cycle.
+ */
+ i915_gem_object_make_unshrinkable(obj);
+ obj->mm.ttm_shrinker_status = I915_TTM_SHRINKER_PINNED;
+
/*
* If this function fails, it will call the destructor, but
* our caller still owns the object. So no freeing in the
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 6/8] drm/i915/ttm: move shrinker management into adjust_lru
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
We currently just evict lmem objects to system memory when under memory
pressure. For this case we might lack the usual object mm.pages, which
effectively hides the pages from the i915-gem shrinker, until we
actually "attach" the TT to the object, or in the case of lmem-only
objects it just gets migrated back to lmem when touched again.
For all cases we can just adjust the i915 shrinker LRU each time we also
adjust the TTM LRU. The two cases we care about are:
1) When something is moved by TTM, including when initially populating
an object. Importantly this covers the case where TTM moves something from
lmem <-> smem, outside of the normal get_pages() interface, which
should still ensure the shmem pages underneath are reclaimable.
2) When calling into i915_gem_object_unlock(). The unlock should
ensure the object is removed from the shinker LRU, if it was indeed
swapped out, or just purged, when the shrinker drops the object lock.
v2(Thomas):
- Handle managing the shrinker LRU in adjust_lru, where it is always
safe to touch the object.
v3(Thomas):
- Pretty much a re-write. This time piggy back off the shrink_pin
stuff, which actually seems to fit quite well for what we want here.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_object.h | 8 +++
.../gpu/drm/i915/gem/i915_gem_object_types.h | 25 ++++++-
drivers/gpu/drm/i915/gem/i915_gem_pages.c | 5 +-
drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 45 +++++++++++--
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 65 +++++++++++++++++--
5 files changed, 130 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index e641db297e0e..3eac8cf2ae10 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -294,6 +294,12 @@ i915_gem_object_is_shrinkable(const struct drm_i915_gem_object *obj)
return i915_gem_object_type_has(obj, I915_GEM_OBJECT_IS_SHRINKABLE);
}
+static inline bool
+i915_gem_object_has_self_managed_shrink_list(const struct drm_i915_gem_object *obj)
+{
+ return i915_gem_object_type_has(obj, I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST);
+}
+
static inline bool
i915_gem_object_is_proxy(const struct drm_i915_gem_object *obj)
{
@@ -531,6 +537,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
+void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
+void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
static inline bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index f4233c4e8d2e..9dbbca682a77 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -34,9 +34,11 @@ struct i915_lut_handle {
struct drm_i915_gem_object_ops {
unsigned int flags;
-#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
-#define I915_GEM_OBJECT_IS_PROXY BIT(2)
-#define I915_GEM_OBJECT_NO_MMAP BIT(3)
+#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
+/* Skip the shrinker management in set_pages/unset_pages */
+#define I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST BIT(2)
+#define I915_GEM_OBJECT_IS_PROXY BIT(3)
+#define I915_GEM_OBJECT_NO_MMAP BIT(4)
/* Interface between the GEM object and its backing storage.
* get_pages() is called once prior to the use of the associated set
@@ -485,6 +487,23 @@ struct drm_i915_gem_object {
*/
atomic_t shrink_pin;
+ /**
+ * @ttm_shrinker_status: Whether TTM is currently holding a
+ * shrink_pin for this object. Protected by the object lock.
+ *
+ * I915_TTM_SHRINKER_UNPINNED: We don't have an extra shrink_pin
+ * for this object. The underlying pages or ttm_tt is using
+ * shmem pages underneath, and as such this object might be
+ * currently visible to the shrinker.
+ *
+ * I915_TTM_SHRINKER_PINNED: We are holding shrink_pin for this
+ * object, which prevents the shrinker from seeing this object.
+ * The object is not currently using shmem pages undearneath.
+ */
+#define I915_TTM_SHRINKER_UNPINNED 1
+#define I915_TTM_SHRINKER_PINNED 2
+ int ttm_shrinker_status;
+
/**
* Priority list of potential placements for this object.
*/
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index ea6d9b3d2d6b..308e22a80af4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -68,7 +68,7 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
shrinkable = false;
}
- if (shrinkable) {
+ if (shrinkable && !i915_gem_object_has_self_managed_shrink_list(obj)) {
struct list_head *list;
unsigned long flags;
@@ -210,7 +210,8 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj)
if (i915_gem_object_is_volatile(obj))
obj->mm.madv = I915_MADV_WILLNEED;
- i915_gem_object_make_unshrinkable(obj);
+ if (!i915_gem_object_has_self_managed_shrink_list(obj))
+ i915_gem_object_make_unshrinkable(obj);
if (obj->mm.mapping) {
unmap_object(obj, page_mask_bits(obj->mm.mapping));
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index 66121fedc655..dde0a5c232f8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -497,13 +497,12 @@ void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
-static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
- struct list_head *head)
+static void ___i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
+ struct list_head *head)
{
struct drm_i915_private *i915 = obj_to_i915(obj);
unsigned long flags;
- GEM_BUG_ON(!i915_gem_object_has_pages(obj));
if (!i915_gem_object_is_shrinkable(obj))
return;
@@ -523,6 +522,38 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
}
+/**
+ * __i915_gem_object_make_shrinkable - Move the object to the tail of the
+ * shrinkable list. Objects on this list might be swapped out. Used with
+ * WILLNEED objects.
+ * @obj: The GEM object.
+ *
+ * DO NOT USE. This is intended to be called on very special objects that don't
+ * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
+ * underneath.
+ */
+void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
+{
+ ___i915_gem_object_make_shrinkable(obj,
+ &obj_to_i915(obj)->mm.shrink_list);
+}
+
+/**
+ * __i915_gem_object_make_purgeable - Move the object to the tail of the
+ * purgeable list. Objects on this list might be swapped out. Used with
+ * DONTNEED objects.
+ * @obj: The GEM object.
+ *
+ * DO NOT USE. This is intended to be called on very special objects that don't
+ * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
+ * underneath.
+ */
+void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
+{
+ ___i915_gem_object_make_shrinkable(obj,
+ &obj_to_i915(obj)->mm.purge_list);
+}
+
/**
* i915_gem_object_make_shrinkable - Move the object to the tail of the
* shrinkable list. Objects on this list might be swapped out. Used with
@@ -535,8 +566,8 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
*/
void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
{
- __i915_gem_object_make_shrinkable(obj,
- &obj_to_i915(obj)->mm.shrink_list);
+ GEM_BUG_ON(!i915_gem_object_has_pages(obj));
+ __i915_gem_object_make_shrinkable(obj);
}
/**
@@ -552,6 +583,6 @@ void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
*/
void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
{
- __i915_gem_object_make_shrinkable(obj,
- &obj_to_i915(obj)->mm.purge_list);
+ GEM_BUG_ON(!i915_gem_object_has_pages(obj));
+ __i915_gem_object_make_purgeable(obj);
}
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 0fe1eb8f38e9..98b7ead1a008 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -851,7 +851,6 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
return i915_ttm_err_to_gem(ret);
}
- i915_ttm_adjust_lru(obj);
if (bo->ttm && !ttm_tt_is_populated(bo->ttm)) {
ret = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
if (ret)
@@ -871,10 +870,9 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
return PTR_ERR(st);
__i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
- if (!bo->ttm || !i915_tt->is_shmem)
- i915_gem_object_make_unshrinkable(obj);
}
+ i915_ttm_adjust_lru(obj);
return ret;
}
@@ -945,8 +943,6 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
* If the object is not destroyed next, The TTM eviction logic
* and shrinkers will move it out if needed.
*/
-
- i915_ttm_adjust_lru(obj);
}
static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
@@ -954,6 +950,9 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
struct i915_ttm_tt *i915_tt =
container_of(bo->ttm, typeof(*i915_tt), ttm);
+ bool shrinkable =
+ bo->ttm && i915_tt->filp && ttm_tt_is_populated(bo->ttm);
+ int status;
/*
* Don't manipulate the TTM LRUs while in TTM bo destruction.
@@ -962,11 +961,42 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
if (!kref_read(&bo->kref))
return;
+ /*
+ * We skip managing the shrinker LRU in set_pages() and just manage
+ * everything here. This does at least solve the issue with having
+ * temporary shmem mappings(like with evicted lmem) not being visible to
+ * the shrinker. Only our shmem objects are shrinkable, everything else
+ * we keep as unshrinkable.
+ *
+ * To make sure everything plays nice we keep an extra shrink pin in TTM
+ * if the underlying pages are not currently shrinkable. Once we release
+ * our pin, like when the pages are moved to shmem, the pages will then
+ * be added to the shrinker LRU, assuming the caller isn't also holding
+ * a pin.
+ *
+ * TODO: consider maybe also bumping the shrinker list here when we have
+ * already unpinned it, which should give us something more like an LRU.
+ */
+ status = obj->mm.ttm_shrinker_status;
+ if (shrinkable) {
+ if (status != I915_TTM_SHRINKER_UNPINNED) {
+ if (obj->mm.madv == I915_MADV_WILLNEED)
+ __i915_gem_object_make_shrinkable(obj);
+ else
+ __i915_gem_object_make_purgeable(obj);
+ status = I915_TTM_SHRINKER_UNPINNED;
+ }
+ } else if (status != I915_TTM_SHRINKER_PINNED) {
+ i915_gem_object_make_unshrinkable(obj);
+ status = I915_TTM_SHRINKER_PINNED;
+ }
+ obj->mm.ttm_shrinker_status = status;
+
/*
* Put on the correct LRU list depending on the MADV status
*/
spin_lock(&bo->bdev->lru_lock);
- if (bo->ttm && i915_tt->filp) {
+ if (shrinkable) {
/* Try to keep shmem_tt from being considered for shrinking. */
bo->priority = TTM_MAX_BO_PRIORITY - 1;
} else if (obj->mm.madv != I915_MADV_WILLNEED) {
@@ -1067,6 +1097,7 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.name = "i915_gem_object_ttm",
+ .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
@@ -1089,6 +1120,19 @@ void i915_ttm_bo_destroy(struct ttm_buffer_object *bo)
mutex_destroy(&obj->ttm.get_io_page.lock);
if (obj->ttm.created) {
+ /*
+ * We freely manage the shrinker LRU outide of the mm.pages life
+ * cycle. As a result when destroying the object we should be
+ * extra paranoid and ensure we remove it from the LRU, before
+ * we free the object.
+ *
+ * Touching the ttm_shrinker_status outside of the object lock
+ * here should be safe now that the last GEM object ref was
+ * dropped.
+ */
+ if (obj->mm.ttm_shrinker_status == I915_TTM_SHRINKER_UNPINNED)
+ i915_gem_object_make_unshrinkable(obj);
+
i915_ttm_backup_free(obj);
/* This releases all gem object bindings to the backend. */
@@ -1141,6 +1185,15 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
/* Forcing the page size is kernel internal only */
GEM_BUG_ON(page_size && obj->mm.n_placements);
+ /*
+ * Keep an extra shrink pin to prevent the object from being made
+ * shrinkable too early. If the ttm_tt is ever allocated in shmem, we
+ * drop the pin. The TTM backend manages the shrinker LRU itself,
+ * outside of the normal mm.pages life cycle.
+ */
+ i915_gem_object_make_unshrinkable(obj);
+ obj->mm.ttm_shrinker_status = I915_TTM_SHRINKER_PINNED;
+
/*
* If this function fails, it will call the destructor, but
* our caller still owns the object. So no freeing in the
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 7/8] drm/i915/ttm: use cached system pages when evicting lmem
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
This should let us do an accelerated copy directly to the shmem pages
when temporarily moving lmem-only objects, where the i915-gem shrinker
can later kick in to swap out the pages, if needed.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 98b7ead1a008..48069b5a664e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -134,11 +134,11 @@ static enum ttm_caching
i915_ttm_select_tt_caching(const struct drm_i915_gem_object *obj)
{
/*
- * Objects only allowed in system get cached cpu-mappings.
- * Other objects get WC mapping for now. Even if in system.
+ * Objects only allowed in system get cached cpu-mappings, or when
+ * evicting lmem-only buffers to system for swapping. Other objects get
+ * WC mapping for now. Even if in system.
*/
- if (obj->mm.region->type == INTEL_MEMORY_SYSTEM &&
- obj->mm.n_placements <= 1)
+ if (obj->mm.n_placements <= 1)
return ttm_cached;
return ttm_write_combined;
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 7/8] drm/i915/ttm: use cached system pages when evicting lmem
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
This should let us do an accelerated copy directly to the shmem pages
when temporarily moving lmem-only objects, where the i915-gem shrinker
can later kick in to swap out the pages, if needed.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 98b7ead1a008..48069b5a664e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -134,11 +134,11 @@ static enum ttm_caching
i915_ttm_select_tt_caching(const struct drm_i915_gem_object *obj)
{
/*
- * Objects only allowed in system get cached cpu-mappings.
- * Other objects get WC mapping for now. Even if in system.
+ * Objects only allowed in system get cached cpu-mappings, or when
+ * evicting lmem-only buffers to system for swapping. Other objects get
+ * WC mapping for now. Even if in system.
*/
- if (obj->mm.region->type == INTEL_MEMORY_SYSTEM &&
- obj->mm.n_placements <= 1)
+ if (obj->mm.n_placements <= 1)
return ttm_cached;
return ttm_write_combined;
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [PATCH v6 8/8] drm/i915/ttm: enable shmem tt backend
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
@ 2021-10-05 18:24 ` Matthew Auld
-1 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
Turn on the shmem tt backend, and enable shrinking.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 48069b5a664e..66caee29a884 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -1097,7 +1097,8 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.name = "i915_gem_object_ttm",
- .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
+ .flags = I915_GEM_OBJECT_IS_SHRINKABLE |
+ I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] [PATCH v6 8/8] drm/i915/ttm: enable shmem tt backend
@ 2021-10-05 18:24 ` Matthew Auld
0 siblings, 0 replies; 28+ messages in thread
From: Matthew Auld @ 2021-10-05 18:24 UTC (permalink / raw)
To: intel-gfx; +Cc: dri-devel, Thomas Hellström
Turn on the shmem tt backend, and enable shrinking.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
index 48069b5a664e..66caee29a884 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
@@ -1097,7 +1097,8 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
.name = "i915_gem_object_ttm",
- .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
+ .flags = I915_GEM_OBJECT_IS_SHRINKABLE |
+ I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
.get_pages = i915_ttm_get_pages,
.put_pages = i915_ttm_put_pages,
--
2.26.3
^ permalink raw reply related [flat|nested] 28+ messages in thread
* [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
` (7 preceding siblings ...)
(?)
@ 2021-10-05 23:55 ` Patchwork
-1 siblings, 0 replies; 28+ messages in thread
From: Patchwork @ 2021-10-05 23:55 UTC (permalink / raw)
To: Matthew Auld; +Cc: intel-gfx
== Series Details ==
Series: series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
URL : https://patchwork.freedesktop.org/series/95476/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
605eef61ebf9 drm/i915/gem: Break out some shmem backend utils
b116a99cc04e drm/i915/ttm: add tt shmem backend
-:16: WARNING:COMMIT_LOG_LONG_LINE: Possible unwrapped commit description (prefer a maximum 75 chars per line)
#16:
dropping the shrinker LRU lock and acquiring the object lock it could for
total: 0 errors, 1 warnings, 0 checks, 486 lines checked
14f612b9093f drm/i915/gtt: drop unneeded make_unshrinkable
32c05a4de159 drm/i915: drop unneeded make_unshrinkable in free_object
0999a655bf2a drm/i915: add some kernel-doc for shrink_pin and friends
c6e18e36eb58 drm/i915/ttm: move shrinker management into adjust_lru
-:19: WARNING:COMMIT_LOG_LONG_LINE: Possible unwrapped commit description (prefer a maximum 75 chars per line)
#19:
an object. Importantly this covers the case where TTM moves something from
total: 0 errors, 1 warnings, 0 checks, 264 lines checked
df1f7fa12f7b drm/i915/ttm: use cached system pages when evicting lmem
fc764587e539 drm/i915/ttm: enable shmem tt backend
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Intel-gfx] ✗ Fi.CI.SPARSE: warning for series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
` (8 preceding siblings ...)
(?)
@ 2021-10-05 23:57 ` Patchwork
-1 siblings, 0 replies; 28+ messages in thread
From: Patchwork @ 2021-10-05 23:57 UTC (permalink / raw)
To: Matthew Auld; +Cc: intel-gfx
== Series Details ==
Series: series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
URL : https://patchwork.freedesktop.org/series/95476/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.2
Fast mode used, each commit won't be checked separately.
-
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:27:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:27:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:27:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:32:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:32:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:49:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:49:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:49:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:56:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_engine_stats.h:56:9: warning: trying to copy expression type 31
+drivers/gpu/drm/i915/gt/intel_reset.c:1392:5: warning: context imbalance in 'intel_gt_reset_trylock' - different lock contexts for basic block
+drivers/gpu/drm/i915/i915_perf.c:1442:15: warning: memset with byte count of 16777216
+drivers/gpu/drm/i915/i915_perf.c:1496:15: warning: memset with byte count of 16777216
+./include/asm-generic/bitops/find.h:112:45: warning: shift count is negative (-262080)
+./include/asm-generic/bitops/find.h:32:31: warning: shift count is negative (-262080)
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_read16' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_read32' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_read64' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_read8' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_write16' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_write32' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'fwtable_write8' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'gen6_write16' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'gen6_write32' - different lock contexts for basic block
+./include/linux/spinlock.h:418:9: warning: context imbalance in 'gen6_write8' - different lock contexts for basic block
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
` (9 preceding siblings ...)
(?)
@ 2021-10-06 0:27 ` Patchwork
-1 siblings, 0 replies; 28+ messages in thread
From: Patchwork @ 2021-10-06 0:27 UTC (permalink / raw)
To: Matthew Auld; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 6644 bytes --]
== Series Details ==
Series: series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
URL : https://patchwork.freedesktop.org/series/95476/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10685 -> Patchwork_21257
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/index.html
Known issues
------------
Here are the changes found in Patchwork_21257 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@cs-gfx:
- fi-kbl-soraka: NOTRUN -> [SKIP][1] ([fdo#109271]) +21 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@amdgpu/amd_basic@cs-gfx.html
* igt@debugfs_test@read_all_entries:
- fi-kbl-soraka: [PASS][2] -> [DMESG-WARN][3] ([i915#1982] / [i915#262])
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-kbl-soraka/igt@debugfs_test@read_all_entries.html
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@debugfs_test@read_all_entries.html
* igt@gem_exec_suspend@basic-s0:
- fi-tgl-u2: [PASS][4] -> [FAIL][5] ([i915#1888])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-tgl-u2/igt@gem_exec_suspend@basic-s0.html
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-tgl-u2/igt@gem_exec_suspend@basic-s0.html
* igt@gem_huc_copy@huc-copy:
- fi-kbl-soraka: NOTRUN -> [SKIP][6] ([fdo#109271] / [i915#2190])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@gem_huc_copy@huc-copy.html
* igt@i915_selftest@live@gt_pm:
- fi-kbl-soraka: NOTRUN -> [DMESG-FAIL][7] ([i915#1886] / [i915#2291])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@i915_selftest@live@gt_pm.html
* igt@kms_chamelium@common-hpd-after-suspend:
- fi-kbl-soraka: NOTRUN -> [SKIP][8] ([fdo#109271] / [fdo#111827]) +8 similar issues
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@kms_chamelium@common-hpd-after-suspend.html
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-b:
- fi-cfl-8109u: [PASS][9] -> [DMESG-WARN][10] ([i915#295]) +17 similar issues
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-cfl-8109u/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-b.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-cfl-8109u/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-b.html
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d:
- fi-kbl-soraka: NOTRUN -> [SKIP][11] ([fdo#109271] / [i915#533])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d.html
* igt@runner@aborted:
- fi-bdw-5557u: NOTRUN -> [FAIL][12] ([i915#1602] / [i915#2029])
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-bdw-5557u/igt@runner@aborted.html
#### Possible fixes ####
* igt@core_hotunplug@unbind-rebind:
- fi-tgl-u2: [INCOMPLETE][13] -> [PASS][14]
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-tgl-u2/igt@core_hotunplug@unbind-rebind.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-tgl-u2/igt@core_hotunplug@unbind-rebind.html
* igt@gem_exec_suspend@basic-s0:
- fi-kbl-soraka: [INCOMPLETE][15] ([i915#155]) -> [PASS][16]
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-kbl-soraka/igt@gem_exec_suspend@basic-s0.html
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-kbl-soraka/igt@gem_exec_suspend@basic-s0.html
* igt@gem_exec_suspend@basic-s3:
- fi-tgl-u2: [FAIL][17] ([i915#1888]) -> [PASS][18]
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-tgl-u2/igt@gem_exec_suspend@basic-s3.html
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-tgl-u2/igt@gem_exec_suspend@basic-s3.html
#### Warnings ####
* igt@kms_frontbuffer_tracking@basic:
- fi-cfl-8109u: [FAIL][19] ([i915#2546]) -> [DMESG-WARN][20] ([i915#295])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/fi-cfl-8109u/igt@kms_frontbuffer_tracking@basic.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/fi-cfl-8109u/igt@kms_frontbuffer_tracking@basic.html
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#155]: https://gitlab.freedesktop.org/drm/intel/issues/155
[i915#1602]: https://gitlab.freedesktop.org/drm/intel/issues/1602
[i915#1886]: https://gitlab.freedesktop.org/drm/intel/issues/1886
[i915#1888]: https://gitlab.freedesktop.org/drm/intel/issues/1888
[i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
[i915#2029]: https://gitlab.freedesktop.org/drm/intel/issues/2029
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#2291]: https://gitlab.freedesktop.org/drm/intel/issues/2291
[i915#2546]: https://gitlab.freedesktop.org/drm/intel/issues/2546
[i915#262]: https://gitlab.freedesktop.org/drm/intel/issues/262
[i915#295]: https://gitlab.freedesktop.org/drm/intel/issues/295
[i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
Participating hosts (41 -> 34)
------------------------------
Missing (7): fi-ilk-m540 bat-dg1-6 fi-hsw-4200u fi-bsw-cyan bat-adlp-4 fi-ctg-p8600 bat-jsl-1
Build changes
-------------
* Linux: CI_DRM_10685 -> Patchwork_21257
CI-20190529: 20190529
CI_DRM_10685: 36c3656c997b07f326d6b967efb1b75e01713773 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6232: effad6af5678be711a2c3e58e182319de784de54 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_21257: fc764587e539a0720d5501e4b3ab5fe4bcdf3bed @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
fc764587e539 drm/i915/ttm: enable shmem tt backend
df1f7fa12f7b drm/i915/ttm: use cached system pages when evicting lmem
c6e18e36eb58 drm/i915/ttm: move shrinker management into adjust_lru
0999a655bf2a drm/i915: add some kernel-doc for shrink_pin and friends
32c05a4de159 drm/i915: drop unneeded make_unshrinkable in free_object
14f612b9093f drm/i915/gtt: drop unneeded make_unshrinkable
b116a99cc04e drm/i915/ttm: add tt shmem backend
605eef61ebf9 drm/i915/gem: Break out some shmem backend utils
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/index.html
[-- Attachment #2: Type: text/html, Size: 7930 bytes --]
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
` (10 preceding siblings ...)
(?)
@ 2021-10-06 5:17 ` Patchwork
-1 siblings, 0 replies; 28+ messages in thread
From: Patchwork @ 2021-10-06 5:17 UTC (permalink / raw)
To: Matthew Auld; +Cc: intel-gfx
[-- Attachment #1: Type: text/plain, Size: 30301 bytes --]
== Series Details ==
Series: series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils
URL : https://patchwork.freedesktop.org/series/95476/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10685_full -> Patchwork_21257_full
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Known issues
------------
Here are the changes found in Patchwork_21257_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@feature_discovery@display-2x:
- shard-tglb: NOTRUN -> [SKIP][1] ([i915#1839])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@feature_discovery@display-2x.html
* igt@gem_ctx_persistence@legacy-engines-queued:
- shard-snb: NOTRUN -> [SKIP][2] ([fdo#109271] / [i915#1099]) +2 similar issues
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-snb5/igt@gem_ctx_persistence@legacy-engines-queued.html
* igt@gem_eio@unwedge-stress:
- shard-iclb: [PASS][3] -> [TIMEOUT][4] ([i915#2369] / [i915#2481] / [i915#3070])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb6/igt@gem_eio@unwedge-stress.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb1/igt@gem_eio@unwedge-stress.html
* igt@gem_exec_fair@basic-deadline:
- shard-apl: NOTRUN -> [FAIL][5] ([i915#2846])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl1/igt@gem_exec_fair@basic-deadline.html
* igt@gem_exec_fair@basic-none-share@rcs0:
- shard-iclb: [PASS][6] -> [FAIL][7] ([i915#2842])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb6/igt@gem_exec_fair@basic-none-share@rcs0.html
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb8/igt@gem_exec_fair@basic-none-share@rcs0.html
* igt@gem_exec_fair@basic-none@vcs0:
- shard-kbl: [PASS][8] -> [FAIL][9] ([i915#2842]) +2 similar issues
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl3/igt@gem_exec_fair@basic-none@vcs0.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl3/igt@gem_exec_fair@basic-none@vcs0.html
* igt@gem_exec_fair@basic-pace@vecs0:
- shard-kbl: NOTRUN -> [FAIL][10] ([i915#2842])
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl2/igt@gem_exec_fair@basic-pace@vecs0.html
* igt@gem_exec_schedule@semaphore-power:
- shard-skl: [PASS][11] -> [DMESG-WARN][12] ([i915#1982])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl6/igt@gem_exec_schedule@semaphore-power.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl2/igt@gem_exec_schedule@semaphore-power.html
* igt@gem_exec_suspend@basic-s0:
- shard-kbl: [PASS][13] -> [DMESG-WARN][14] ([i915#165] / [i915#180] / [i915#262] / [i915#62] / [i915#92])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@gem_exec_suspend@basic-s0.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@gem_exec_suspend@basic-s0.html
* igt@gem_huc_copy@huc-copy:
- shard-apl: NOTRUN -> [SKIP][15] ([fdo#109271] / [i915#2190])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl2/igt@gem_huc_copy@huc-copy.html
* igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-mc-ccs:
- shard-iclb: NOTRUN -> [SKIP][16] ([i915#768])
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-mc-ccs.html
* igt@gem_userptr_blits@readonly-pwrite-unsync:
- shard-iclb: NOTRUN -> [SKIP][17] ([i915#3297])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@gem_userptr_blits@readonly-pwrite-unsync.html
* igt@gem_userptr_blits@vma-merge:
- shard-snb: NOTRUN -> [FAIL][18] ([i915#2724])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-snb7/igt@gem_userptr_blits@vma-merge.html
* igt@gem_workarounds@suspend-resume-context:
- shard-apl: [PASS][19] -> [DMESG-WARN][20] ([i915#180]) +1 similar issue
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-apl3/igt@gem_workarounds@suspend-resume-context.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl6/igt@gem_workarounds@suspend-resume-context.html
* igt@gen3_render_tiledy_blits:
- shard-tglb: NOTRUN -> [SKIP][21] ([fdo#109289]) +1 similar issue
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@gen3_render_tiledy_blits.html
* igt@gen9_exec_parse@bb-secure:
- shard-tglb: NOTRUN -> [SKIP][22] ([i915#2856]) +1 similar issue
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@gen9_exec_parse@bb-secure.html
* igt@i915_pm_lpsp@screens-disabled:
- shard-tglb: NOTRUN -> [SKIP][23] ([i915#1902])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@i915_pm_lpsp@screens-disabled.html
* igt@i915_pm_rc6_residency@media-rc6-accuracy:
- shard-tglb: NOTRUN -> [SKIP][24] ([fdo#109289] / [fdo#111719])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@i915_pm_rc6_residency@media-rc6-accuracy.html
* igt@i915_pm_rpm@pc8-residency:
- shard-tglb: NOTRUN -> [SKIP][25] ([fdo#109506] / [i915#2411])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@i915_pm_rpm@pc8-residency.html
* igt@i915_pm_rpm@system-suspend-modeset:
- shard-skl: [PASS][26] -> [INCOMPLETE][27] ([i915#151])
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl1/igt@i915_pm_rpm@system-suspend-modeset.html
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl1/igt@i915_pm_rpm@system-suspend-modeset.html
* igt@kms_big_fb@y-tiled-32bpp-rotate-0:
- shard-glk: [PASS][28] -> [DMESG-WARN][29] ([i915#118] / [i915#95]) +1 similar issue
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-glk7/igt@kms_big_fb@y-tiled-32bpp-rotate-0.html
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-glk7/igt@kms_big_fb@y-tiled-32bpp-rotate-0.html
* igt@kms_big_fb@yf-tiled-addfb:
- shard-tglb: NOTRUN -> [SKIP][30] ([fdo#111615]) +2 similar issues
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_big_fb@yf-tiled-addfb.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip:
- shard-apl: NOTRUN -> [SKIP][31] ([fdo#109271] / [i915#3777]) +5 similar issues
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl8/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip.html
* igt@kms_ccs@pipe-a-bad-rotation-90-y_tiled_gen12_mc_ccs:
- shard-kbl: NOTRUN -> [SKIP][32] ([fdo#109271] / [i915#3886]) +5 similar issues
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl1/igt@kms_ccs@pipe-a-bad-rotation-90-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
- shard-apl: NOTRUN -> [SKIP][33] ([fdo#109271] / [i915#3886]) +13 similar issues
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl1/igt@kms_ccs@pipe-a-ccs-on-another-bo-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-b-bad-rotation-90-y_tiled_gen12_mc_ccs:
- shard-iclb: NOTRUN -> [SKIP][34] ([fdo#109278] / [i915#3886])
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@kms_ccs@pipe-b-bad-rotation-90-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
- shard-tglb: NOTRUN -> [SKIP][35] ([i915#3689] / [i915#3886]) +4 similar issues
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs.html
* igt@kms_ccs@pipe-d-bad-pixel-format-y_tiled_ccs:
- shard-snb: NOTRUN -> [SKIP][36] ([fdo#109271]) +296 similar issues
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-snb7/igt@kms_ccs@pipe-d-bad-pixel-format-y_tiled_ccs.html
* igt@kms_ccs@pipe-d-missing-ccs-buffer-yf_tiled_ccs:
- shard-tglb: NOTRUN -> [SKIP][37] ([i915#3689]) +3 similar issues
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_ccs@pipe-d-missing-ccs-buffer-yf_tiled_ccs.html
* igt@kms_cdclk@plane-scaling:
- shard-tglb: NOTRUN -> [SKIP][38] ([i915#3742])
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_cdclk@plane-scaling.html
* igt@kms_chamelium@dp-mode-timings:
- shard-apl: NOTRUN -> [SKIP][39] ([fdo#109271] / [fdo#111827]) +17 similar issues
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl2/igt@kms_chamelium@dp-mode-timings.html
* igt@kms_chamelium@vga-hpd-without-ddc:
- shard-snb: NOTRUN -> [SKIP][40] ([fdo#109271] / [fdo#111827]) +16 similar issues
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-snb7/igt@kms_chamelium@vga-hpd-without-ddc.html
* igt@kms_color_chamelium@pipe-d-ctm-max:
- shard-kbl: NOTRUN -> [SKIP][41] ([fdo#109271] / [fdo#111827]) +10 similar issues
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl7/igt@kms_color_chamelium@pipe-d-ctm-max.html
* igt@kms_color_chamelium@pipe-d-ctm-red-to-blue:
- shard-tglb: NOTRUN -> [SKIP][42] ([fdo#109284] / [fdo#111827]) +6 similar issues
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_color_chamelium@pipe-d-ctm-red-to-blue.html
* igt@kms_content_protection@uevent:
- shard-kbl: NOTRUN -> [FAIL][43] ([i915#2105])
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl1/igt@kms_content_protection@uevent.html
* igt@kms_cursor_crc@pipe-a-cursor-alpha-transparent:
- shard-kbl: [PASS][44] -> [DMESG-WARN][45] ([i915#180] / [i915#203] / [i915#62] / [i915#92]) +2 similar issues
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl7/igt@kms_cursor_crc@pipe-a-cursor-alpha-transparent.html
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@kms_cursor_crc@pipe-a-cursor-alpha-transparent.html
* igt@kms_cursor_crc@pipe-c-cursor-256x256-offscreen:
- shard-kbl: [PASS][46] -> [DMESG-WARN][47] ([i915#62] / [i915#92]) +3 similar issues
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@kms_cursor_crc@pipe-c-cursor-256x256-offscreen.html
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@kms_cursor_crc@pipe-c-cursor-256x256-offscreen.html
* igt@kms_cursor_crc@pipe-d-cursor-32x10-offscreen:
- shard-tglb: NOTRUN -> [SKIP][48] ([i915#3359]) +3 similar issues
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb1/igt@kms_cursor_crc@pipe-d-cursor-32x10-offscreen.html
* igt@kms_cursor_crc@pipe-d-cursor-32x32-rapid-movement:
- shard-tglb: NOTRUN -> [SKIP][49] ([i915#3319])
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@kms_cursor_crc@pipe-d-cursor-32x32-rapid-movement.html
* igt@kms_cursor_crc@pipe-d-cursor-512x170-offscreen:
- shard-tglb: NOTRUN -> [SKIP][50] ([fdo#109279] / [i915#3359]) +1 similar issue
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_cursor_crc@pipe-d-cursor-512x170-offscreen.html
* igt@kms_cursor_edge_walk@pipe-d-64x64-right-edge:
- shard-skl: NOTRUN -> [SKIP][51] ([fdo#109271]) +4 similar issues
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl2/igt@kms_cursor_edge_walk@pipe-d-64x64-right-edge.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic:
- shard-tglb: NOTRUN -> [SKIP][52] ([i915#4103])
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-atomic.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
- shard-skl: [PASS][53] -> [FAIL][54] ([i915#2346])
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl1/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
- shard-kbl: [PASS][55] -> [DMESG-WARN][56] ([i915#165] / [i915#180] / [i915#533] / [i915#62] / [i915#92])
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
* igt@kms_cursor_legacy@flip-vs-cursor-busy-crc-legacy:
- shard-kbl: [PASS][57] -> [DMESG-WARN][58] ([i915#165] / [i915#62] / [i915#92]) +10 similar issues
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@kms_cursor_legacy@flip-vs-cursor-busy-crc-legacy.html
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@kms_cursor_legacy@flip-vs-cursor-busy-crc-legacy.html
* igt@kms_cursor_legacy@pipe-d-torture-move:
- shard-iclb: NOTRUN -> [SKIP][59] ([fdo#109278]) +1 similar issue
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@kms_cursor_legacy@pipe-d-torture-move.html
* igt@kms_flip@2x-single-buffer-flip-vs-dpms-off-vs-modeset-interruptible:
- shard-iclb: NOTRUN -> [SKIP][60] ([fdo#109274])
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@kms_flip@2x-single-buffer-flip-vs-dpms-off-vs-modeset-interruptible.html
* igt@kms_flip@plain-flip-fb-recreate@a-edp1:
- shard-skl: [PASS][61] -> [FAIL][62] ([i915#2122])
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl9/igt@kms_flip@plain-flip-fb-recreate@a-edp1.html
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl9/igt@kms_flip@plain-flip-fb-recreate@a-edp1.html
* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs:
- shard-apl: NOTRUN -> [SKIP][63] ([fdo#109271] / [i915#2672])
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl1/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile:
- shard-iclb: [PASS][64] -> [SKIP][65] ([i915#3701]) +1 similar issue
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb7/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile.html
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb2/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-blt:
- shard-iclb: NOTRUN -> [SKIP][66] ([fdo#109280]) +2 similar issues
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-move:
- shard-tglb: NOTRUN -> [SKIP][67] ([fdo#111825]) +21 similar issues
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-move.html
* igt@kms_frontbuffer_tracking@fbcpsr-suspend:
- shard-tglb: [PASS][68] -> [INCOMPLETE][69] ([i915#2411] / [i915#456])
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb6/igt@kms_frontbuffer_tracking@fbcpsr-suspend.html
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb7/igt@kms_frontbuffer_tracking@fbcpsr-suspend.html
* igt@kms_frontbuffer_tracking@psr-rgb101010-draw-mmap-wc:
- shard-kbl: NOTRUN -> [SKIP][70] ([fdo#109271]) +117 similar issues
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl2/igt@kms_frontbuffer_tracking@psr-rgb101010-draw-mmap-wc.html
* igt@kms_hdr@bpc-switch-dpms:
- shard-skl: [PASS][71] -> [FAIL][72] ([i915#1188]) +1 similar issue
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl9/igt@kms_hdr@bpc-switch-dpms.html
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl7/igt@kms_hdr@bpc-switch-dpms.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-kbl: [PASS][73] -> [DMESG-WARN][74] ([i915#180]) +6 similar issues
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl1/igt@kms_hdr@bpc-switch-suspend.html
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl1/igt@kms_hdr@bpc-switch-suspend.html
* igt@kms_pipe_crc_basic@suspend-read-crc-pipe-d:
- shard-kbl: NOTRUN -> [SKIP][75] ([fdo#109271] / [i915#533])
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl7/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-d.html
- shard-apl: NOTRUN -> [SKIP][76] ([fdo#109271] / [i915#533]) +2 similar issues
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl3/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-d.html
* igt@kms_plane@plane-position-hole@pipe-b-planes:
- shard-kbl: [PASS][77] -> [DMESG-WARN][78] ([i915#165] / [i915#180] / [i915#62] / [i915#92]) +12 similar issues
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@kms_plane@plane-position-hole@pipe-b-planes.html
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@kms_plane@plane-position-hole@pipe-b-planes.html
* igt@kms_plane_alpha_blend@pipe-b-alpha-basic:
- shard-kbl: NOTRUN -> [FAIL][79] ([fdo#108145] / [i915#265])
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl7/igt@kms_plane_alpha_blend@pipe-b-alpha-basic.html
* igt@kms_plane_alpha_blend@pipe-b-coverage-7efc:
- shard-skl: [PASS][80] -> [FAIL][81] ([fdo#108145] / [i915#265])
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl1/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl9/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html
* igt@kms_plane_alpha_blend@pipe-c-alpha-basic:
- shard-apl: NOTRUN -> [FAIL][82] ([fdo#108145] / [i915#265]) +2 similar issues
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl1/igt@kms_plane_alpha_blend@pipe-c-alpha-basic.html
* igt@kms_plane_alpha_blend@pipe-c-alpha-transparent-fb:
- shard-apl: NOTRUN -> [FAIL][83] ([i915#265])
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl6/igt@kms_plane_alpha_blend@pipe-c-alpha-transparent-fb.html
* igt@kms_plane_lowres@pipe-a-tiling-none:
- shard-tglb: NOTRUN -> [SKIP][84] ([i915#3536])
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@kms_plane_lowres@pipe-a-tiling-none.html
* igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping:
- shard-apl: NOTRUN -> [SKIP][85] ([fdo#109271] / [i915#2733])
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl3/igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping.html
* igt@kms_psr2_sf@cursor-plane-update-sf:
- shard-tglb: NOTRUN -> [SKIP][86] ([i915#2920]) +1 similar issue
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_psr2_sf@cursor-plane-update-sf.html
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-5:
- shard-apl: NOTRUN -> [SKIP][87] ([fdo#109271] / [i915#658]) +4 similar issues
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl2/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-5.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-2:
- shard-kbl: NOTRUN -> [SKIP][88] ([fdo#109271] / [i915#658]) +3 similar issues
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl1/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-2.html
* igt@kms_psr2_su@frontbuffer:
- shard-iclb: [PASS][89] -> [SKIP][90] ([fdo#109642] / [fdo#111068] / [i915#658])
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb2/igt@kms_psr2_su@frontbuffer.html
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb3/igt@kms_psr2_su@frontbuffer.html
* igt@kms_psr@psr2_cursor_mmap_gtt:
- shard-tglb: NOTRUN -> [FAIL][91] ([i915#132] / [i915#3467])
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@kms_psr@psr2_cursor_mmap_gtt.html
* igt@kms_psr@psr2_sprite_blt:
- shard-iclb: [PASS][92] -> [SKIP][93] ([fdo#109441]) +1 similar issue
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb2/igt@kms_psr@psr2_sprite_blt.html
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb3/igt@kms_psr@psr2_sprite_blt.html
* igt@kms_vblank@pipe-a-ts-continuation-suspend:
- shard-kbl: [PASS][94] -> [DMESG-WARN][95] ([i915#180] / [i915#295])
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl2/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl1/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
* igt@nouveau_crc@pipe-b-ctx-flip-detection:
- shard-tglb: NOTRUN -> [SKIP][96] ([i915#2530]) +2 similar issues
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@nouveau_crc@pipe-b-ctx-flip-detection.html
* igt@perf@blocking:
- shard-skl: [PASS][97] -> [FAIL][98] ([i915#1542])
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl2/igt@perf@blocking.html
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl8/igt@perf@blocking.html
* igt@prime_nv_api@i915_nv_reimport_twice_check_flink_name:
- shard-apl: NOTRUN -> [SKIP][99] ([fdo#109271]) +244 similar issues
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl1/igt@prime_nv_api@i915_nv_reimport_twice_check_flink_name.html
* igt@prime_nv_pcopy@test1_macro:
- shard-tglb: NOTRUN -> [SKIP][100] ([fdo#109291])
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@prime_nv_pcopy@test1_macro.html
* igt@prime_nv_test@i915_blt_fill_nv_read:
- shard-iclb: NOTRUN -> [SKIP][101] ([fdo#109291])
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb7/igt@prime_nv_test@i915_blt_fill_nv_read.html
* igt@prime_vgem@coherency-gtt:
- shard-tglb: NOTRUN -> [SKIP][102] ([fdo#111656])
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@prime_vgem@coherency-gtt.html
* igt@prime_vgem@fence-flip-hang:
- shard-tglb: NOTRUN -> [SKIP][103] ([fdo#109295])
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@prime_vgem@fence-flip-hang.html
* igt@sysfs_clients@fair-0:
- shard-tglb: NOTRUN -> [SKIP][104] ([i915#2994]) +1 similar issue
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@sysfs_clients@fair-0.html
* igt@sysfs_clients@fair-1:
- shard-apl: NOTRUN -> [SKIP][105] ([fdo#109271] / [i915#2994]) +3 similar issues
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl3/igt@sysfs_clients@fair-1.html
* igt@sysfs_clients@sema-50:
- shard-kbl: NOTRUN -> [SKIP][106] ([fdo#109271] / [i915#2994]) +1 similar issue
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl2/igt@sysfs_clients@sema-50.html
#### Possible fixes ####
* igt@gem_eio@in-flight-contexts-1us:
- shard-tglb: [TIMEOUT][107] ([i915#3063]) -> [PASS][108]
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb5/igt@gem_eio@in-flight-contexts-1us.html
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@gem_eio@in-flight-contexts-1us.html
* igt@gem_eio@unwedge-stress:
- shard-tglb: [TIMEOUT][109] ([i915#2369] / [i915#3063] / [i915#3648]) -> [PASS][110]
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb7/igt@gem_eio@unwedge-stress.html
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb6/igt@gem_eio@unwedge-stress.html
* igt@gem_exec_fair@basic-deadline:
- shard-kbl: [FAIL][111] ([i915#2846]) -> [PASS][112]
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl4/igt@gem_exec_fair@basic-deadline.html
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl6/igt@gem_exec_fair@basic-deadline.html
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-glk: [FAIL][113] ([i915#2842]) -> [PASS][114]
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-glk9/igt@gem_exec_fair@basic-pace-share@rcs0.html
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-glk6/igt@gem_exec_fair@basic-pace-share@rcs0.html
* igt@gem_exec_fair@basic-pace@bcs0:
- shard-tglb: [FAIL][115] ([i915#2842]) -> [PASS][116] +1 similar issue
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb2/igt@gem_exec_fair@basic-pace@bcs0.html
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb6/igt@gem_exec_fair@basic-pace@bcs0.html
* igt@i915_pm_rpm@system-suspend-execbuf:
- shard-tglb: [INCOMPLETE][117] ([i915#2411] / [i915#456] / [i915#750]) -> [PASS][118]
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb7/igt@i915_pm_rpm@system-suspend-execbuf.html
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb2/igt@i915_pm_rpm@system-suspend-execbuf.html
* igt@kms_cursor_crc@pipe-c-cursor-suspend:
- shard-kbl: [DMESG-WARN][119] ([i915#180]) -> [PASS][120] +3 similar issues
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl4/igt@kms_cursor_crc@pipe-c-cursor-suspend.html
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl2/igt@kms_cursor_crc@pipe-c-cursor-suspend.html
* igt@kms_fbcon_fbt@fbc-suspend:
- shard-kbl: [INCOMPLETE][121] ([i915#155] / [i915#180] / [i915#636]) -> [PASS][122]
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-kbl6/igt@kms_fbcon_fbt@fbc-suspend.html
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-kbl2/igt@kms_fbcon_fbt@fbc-suspend.html
* igt@kms_flip@flip-vs-expired-vblank@c-edp1:
- shard-skl: [FAIL][123] ([i915#79]) -> [PASS][124]
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl7/igt@kms_flip@flip-vs-expired-vblank@c-edp1.html
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl4/igt@kms_flip@flip-vs-expired-vblank@c-edp1.html
* igt@kms_flip@flip-vs-suspend-interruptible@a-edp1:
- shard-tglb: [INCOMPLETE][125] ([i915#2411] / [i915#456]) -> [PASS][126]
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-tglb7/igt@kms_flip@flip-vs-suspend-interruptible@a-edp1.html
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-tglb5/igt@kms_flip@flip-vs-suspend-interruptible@a-edp1.html
* igt@kms_flip@flip-vs-suspend-interruptible@c-dp1:
- shard-apl: [DMESG-WARN][127] ([i915#180]) -> [PASS][128] +1 similar issue
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-apl8/igt@kms_flip@flip-vs-suspend-interruptible@c-dp1.html
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-apl8/igt@kms_flip@flip-vs-suspend-interruptible@c-dp1.html
* igt@kms_plane_alpha_blend@pipe-a-constant-alpha-min:
- shard-skl: [FAIL][129] ([fdo#108145] / [i915#265]) -> [PASS][130]
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl5/igt@kms_plane_alpha_blend@pipe-a-constant-alpha-min.html
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl7/igt@kms_plane_alpha_blend@pipe-a-constant-alpha-min.html
* igt@kms_prop_blob@blob-multiple:
- shard-skl: [DMESG-WARN][131] ([i915#1982]) -> [PASS][132]
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl5/igt@kms_prop_blob@blob-multiple.html
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl7/igt@kms_prop_blob@blob-multiple.html
* igt@kms_psr@psr2_primary_mmap_cpu:
- shard-iclb: [SKIP][133] ([fdo#109441]) -> [PASS][134] +1 similar issue
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb4/igt@kms_psr@psr2_primary_mmap_cpu.html
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb2/igt@kms_psr@psr2_primary_mmap_cpu.html
* igt@perf@short-reads:
- shard-skl: [FAIL][135] ([i915#51]) -> [PASS][136]
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-skl1/igt@perf@short-reads.html
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-skl9/igt@perf@short-reads.html
#### Warnings ####
* igt@i915_pm_rc6_residency@rc6-fence:
- shard-iclb: [WARN][137] ([i915#2684]) -> [WARN][138] ([i915#1804] / [i915#2684])
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10685/shard-iclb2/igt@i915_pm_rc6_residency@rc6-fence.html
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/shard-iclb6/i
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_21257/index.html
[-- Attachment #2: Type: text/html, Size: 33581 bytes --]
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [PATCH v6 3/8] drm/i915/gtt: drop unneeded make_unshrinkable
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
@ 2021-10-06 8:02 ` Thomas Hellström
-1 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:02 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> We already do this when mapping the pages.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gt/gen6_ppgtt.c | 1 -
> drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 1 -
> 2 files changed, 2 deletions(-)
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
>
> diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> index 890191f286e3..baea9770200a 100644
> --- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> +++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> @@ -185,7 +185,6 @@ static void gen6_alloc_va_range(struct i915_address_space *vm,
>
> pt = stash->pt[0];
> __i915_gem_object_pin_pages(pt->base);
> - i915_gem_object_make_unshrinkable(pt->base);
>
> fill32_px(pt, vm->scratch[0]->encode);
>
> diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> index 037a9a6e4889..8af2f709571c 100644
> --- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> +++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> @@ -301,7 +301,6 @@ static void __gen8_ppgtt_alloc(struct i915_address_space * const vm,
>
> pt = stash->pt[!!lvl];
> __i915_gem_object_pin_pages(pt->base);
> - i915_gem_object_make_unshrinkable(pt->base);
>
> fill_px(pt, vm->scratch[lvl]->encode);
>
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [Intel-gfx] [PATCH v6 3/8] drm/i915/gtt: drop unneeded make_unshrinkable
@ 2021-10-06 8:02 ` Thomas Hellström
0 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:02 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> We already do this when mapping the pages.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gt/gen6_ppgtt.c | 1 -
> drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 1 -
> 2 files changed, 2 deletions(-)
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
>
> diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> index 890191f286e3..baea9770200a 100644
> --- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> +++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
> @@ -185,7 +185,6 @@ static void gen6_alloc_va_range(struct i915_address_space *vm,
>
> pt = stash->pt[0];
> __i915_gem_object_pin_pages(pt->base);
> - i915_gem_object_make_unshrinkable(pt->base);
>
> fill32_px(pt, vm->scratch[0]->encode);
>
> diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> index 037a9a6e4889..8af2f709571c 100644
> --- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> +++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> @@ -301,7 +301,6 @@ static void __gen8_ppgtt_alloc(struct i915_address_space * const vm,
>
> pt = stash->pt[!!lvl];
> __i915_gem_object_pin_pages(pt->base);
> - i915_gem_object_make_unshrinkable(pt->base);
>
> fill_px(pt, vm->scratch[lvl]->encode);
>
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [PATCH v6 4/8] drm/i915: drop unneeded make_unshrinkable in free_object
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
@ 2021-10-06 8:03 ` Thomas Hellström
-1 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:03 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> The comment here is no longer accurate, since the current shrinker code
> requires a full ref before touching any objects. Also unset_pages()
> should already do the required make_unshrinkable() for us, if needed,
> which is also nicely balanced with set_pages().
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gem/i915_gem_object.c | 9 ---------
> 1 file changed, 9 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
> index 76ce6a1500bc..1dc3c1940c32 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
> @@ -337,15 +337,6 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj)
> */
> atomic_inc(&i915->mm.free_count);
>
> - /*
> - * This serializes freeing with the shrinker. Since the free
> - * is delayed, first by RCU then by the workqueue, we want the
> - * shrinker to be able to free pages of unreferenced objects,
> - * or else we may oom whilst there are plenty of deferred
> - * freed objects.
> - */
> - i915_gem_object_make_unshrinkable(obj);
> -
> /*
> * Since we require blocking on struct_mutex to unbind the freed
> * object from the GPU before releasing resources back to the
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [Intel-gfx] [PATCH v6 4/8] drm/i915: drop unneeded make_unshrinkable in free_object
@ 2021-10-06 8:03 ` Thomas Hellström
0 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:03 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> The comment here is no longer accurate, since the current shrinker code
> requires a full ref before touching any objects. Also unset_pages()
> should already do the required make_unshrinkable() for us, if needed,
> which is also nicely balanced with set_pages().
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gem/i915_gem_object.c | 9 ---------
> 1 file changed, 9 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
> index 76ce6a1500bc..1dc3c1940c32 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
> @@ -337,15 +337,6 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj)
> */
> atomic_inc(&i915->mm.free_count);
>
> - /*
> - * This serializes freeing with the shrinker. Since the free
> - * is delayed, first by RCU then by the workqueue, we want the
> - * shrinker to be able to free pages of unreferenced objects,
> - * or else we may oom whilst there are plenty of deferred
> - * freed objects.
> - */
> - i915_gem_object_make_unshrinkable(obj);
> -
> /*
> * Since we require blocking on struct_mutex to unbind the freed
> * object from the GPU before releasing resources back to the
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [PATCH v6 5/8] drm/i915: add some kernel-doc for shrink_pin and friends
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
@ 2021-10-06 8:03 ` Thomas Hellström
-1 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:03 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> Attempt to document shrink_pin and the other relevant interfaces that
> interact with it, before we start messing with it.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> .../gpu/drm/i915/gem/i915_gem_object_types.h | 24 +++++++++++++-
> drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 31 +++++++++++++++++++
> 2 files changed, 54 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> index 7dd5f804aab3..f4233c4e8d2e 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> @@ -461,6 +461,28 @@ struct drm_i915_gem_object {
> * instead go through the pin/unpin interfaces.
> */
> atomic_t pages_pin_count;
> +
> + /**
> + * @shrink_pin: Prevents the pages from being made visible to
> + * the shrinker, while the shrink_pin is non-zero. Most users
> + * should pretty much never have to care about this, outside of
> + * some special use cases.
> + *
> + * By default most objects will start out as visible to the
> + * shrinker(if I915_GEM_OBJECT_IS_SHRINKABLE) as soon as the
> + * backing pages are attached to the object, like in
> + * __i915_gem_object_set_pages(). They will then be removed the
> + * shrinker list once the pages are released.
> + *
> + * The @shrink_pin is incremented by calling
> + * i915_gem_object_make_unshrinkable(), which will also remove
> + * the object from the shrinker list, if the pin count was zero.
> + *
> + * Callers will then typically call
> + * i915_gem_object_make_shrinkable() or
> + * i915_gem_object_make_purgeable() to decrement the pin count,
> + * and make the pages visible again.
> + */
> atomic_t shrink_pin;
>
> /**
> @@ -522,7 +544,7 @@ struct drm_i915_gem_object {
> struct i915_gem_object_page_iter get_dma_page;
>
> /**
> - * Element within i915->mm.unbound_list or i915->mm.bound_list,
> + * Element within i915->mm.shrink_list or i915->mm.purge_list,
> * locked by i915->mm.obj_lock.
> */
> struct list_head link;
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> index ae2a8d54b7a4..66121fedc655 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> @@ -463,6 +463,16 @@ void i915_gem_shrinker_taints_mutex(struct drm_i915_private *i915,
>
> #define obj_to_i915(obj__) to_i915((obj__)->base.dev)
>
> +/**
> + * i915_gem_object_make_unshrinkable - Hide the object from the shrinker. By
> + * default all object types that support shrinking(see IS_SHRINKABLE), will also
> + * make the object visible to the shrinker after allocating the system memory
> + * pages.
> + * @obj: The GEM object.
> + *
> + * This is typically used for special kernel internal objects that can't be
> + * easily processed by the shrinker, like if they are perma-pinned.
> + */
> void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
> {
> struct drm_i915_private *i915 = obj_to_i915(obj);
> @@ -513,12 +523,33 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> +/**
> + * i915_gem_object_make_shrinkable - Move the object to the tail of the
> + * shrinkable list. Objects on this list might be swapped out. Used with
> + * WILLNEED objects.
> + * @obj: The GEM object.
> + *
> + * MUST only be called on objects which have backing pages.
> + *
> + * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
> + */
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> {
> __i915_gem_object_make_shrinkable(obj,
> &obj_to_i915(obj)->mm.shrink_list);
> }
>
> +/**
> + * i915_gem_object_make_purgeable - Move the object to the tail of the purgeable
> + * list. Used with DONTNEED objects. Unlike with shrinkable objects, the
> + * shrinker will attempt to discard the backing pages, instead of trying to swap
> + * them out.
> + * @obj: The GEM object.
> + *
> + * MUST only be called on objects which have backing pages.
> + *
> + * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
> + */
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> {
> __i915_gem_object_make_shrinkable(obj,
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [Intel-gfx] [PATCH v6 5/8] drm/i915: add some kernel-doc for shrink_pin and friends
@ 2021-10-06 8:03 ` Thomas Hellström
0 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:03 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> Attempt to document shrink_pin and the other relevant interfaces that
> interact with it, before we start messing with it.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> .../gpu/drm/i915/gem/i915_gem_object_types.h | 24 +++++++++++++-
> drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 31 +++++++++++++++++++
> 2 files changed, 54 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> index 7dd5f804aab3..f4233c4e8d2e 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> @@ -461,6 +461,28 @@ struct drm_i915_gem_object {
> * instead go through the pin/unpin interfaces.
> */
> atomic_t pages_pin_count;
> +
> + /**
> + * @shrink_pin: Prevents the pages from being made visible to
> + * the shrinker, while the shrink_pin is non-zero. Most users
> + * should pretty much never have to care about this, outside of
> + * some special use cases.
> + *
> + * By default most objects will start out as visible to the
> + * shrinker(if I915_GEM_OBJECT_IS_SHRINKABLE) as soon as the
> + * backing pages are attached to the object, like in
> + * __i915_gem_object_set_pages(). They will then be removed the
> + * shrinker list once the pages are released.
> + *
> + * The @shrink_pin is incremented by calling
> + * i915_gem_object_make_unshrinkable(), which will also remove
> + * the object from the shrinker list, if the pin count was zero.
> + *
> + * Callers will then typically call
> + * i915_gem_object_make_shrinkable() or
> + * i915_gem_object_make_purgeable() to decrement the pin count,
> + * and make the pages visible again.
> + */
> atomic_t shrink_pin;
>
> /**
> @@ -522,7 +544,7 @@ struct drm_i915_gem_object {
> struct i915_gem_object_page_iter get_dma_page;
>
> /**
> - * Element within i915->mm.unbound_list or i915->mm.bound_list,
> + * Element within i915->mm.shrink_list or i915->mm.purge_list,
> * locked by i915->mm.obj_lock.
> */
> struct list_head link;
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> index ae2a8d54b7a4..66121fedc655 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> @@ -463,6 +463,16 @@ void i915_gem_shrinker_taints_mutex(struct drm_i915_private *i915,
>
> #define obj_to_i915(obj__) to_i915((obj__)->base.dev)
>
> +/**
> + * i915_gem_object_make_unshrinkable - Hide the object from the shrinker. By
> + * default all object types that support shrinking(see IS_SHRINKABLE), will also
> + * make the object visible to the shrinker after allocating the system memory
> + * pages.
> + * @obj: The GEM object.
> + *
> + * This is typically used for special kernel internal objects that can't be
> + * easily processed by the shrinker, like if they are perma-pinned.
> + */
> void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
> {
> struct drm_i915_private *i915 = obj_to_i915(obj);
> @@ -513,12 +523,33 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> +/**
> + * i915_gem_object_make_shrinkable - Move the object to the tail of the
> + * shrinkable list. Objects on this list might be swapped out. Used with
> + * WILLNEED objects.
> + * @obj: The GEM object.
> + *
> + * MUST only be called on objects which have backing pages.
> + *
> + * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
> + */
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> {
> __i915_gem_object_make_shrinkable(obj,
> &obj_to_i915(obj)->mm.shrink_list);
> }
>
> +/**
> + * i915_gem_object_make_purgeable - Move the object to the tail of the purgeable
> + * list. Used with DONTNEED objects. Unlike with shrinkable objects, the
> + * shrinker will attempt to discard the backing pages, instead of trying to swap
> + * them out.
> + * @obj: The GEM object.
> + *
> + * MUST only be called on objects which have backing pages.
> + *
> + * MUST be balanced with previous call to i915_gem_object_make_unshrinkable().
> + */
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> {
> __i915_gem_object_make_shrinkable(obj,
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [PATCH v6 6/8] drm/i915/ttm: move shrinker management into adjust_lru
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
@ 2021-10-06 8:10 ` Thomas Hellström
-1 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:10 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> We currently just evict lmem objects to system memory when under memory
> pressure. For this case we might lack the usual object mm.pages, which
> effectively hides the pages from the i915-gem shrinker, until we
> actually "attach" the TT to the object, or in the case of lmem-only
> objects it just gets migrated back to lmem when touched again.
>
> For all cases we can just adjust the i915 shrinker LRU each time we also
> adjust the TTM LRU. The two cases we care about are:
>
> 1) When something is moved by TTM, including when initially populating
> an object. Importantly this covers the case where TTM moves something from
> lmem <-> smem, outside of the normal get_pages() interface, which
> should still ensure the shmem pages underneath are reclaimable.
>
> 2) When calling into i915_gem_object_unlock(). The unlock should
> ensure the object is removed from the shinker LRU, if it was indeed
> swapped out, or just purged, when the shrinker drops the object lock.
>
> v2(Thomas):
> - Handle managing the shrinker LRU in adjust_lru, where it is always
> safe to touch the object.
> v3(Thomas):
> - Pretty much a re-write. This time piggy back off the shrink_pin
> stuff, which actually seems to fit quite well for what we want here.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gem/i915_gem_object.h | 8 +++
> .../gpu/drm/i915/gem/i915_gem_object_types.h | 25 ++++++-
> drivers/gpu/drm/i915/gem/i915_gem_pages.c | 5 +-
> drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 45 +++++++++++--
> drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 65 +++++++++++++++++--
> 5 files changed, 130 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> index e641db297e0e..3eac8cf2ae10 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> @@ -294,6 +294,12 @@ i915_gem_object_is_shrinkable(const struct drm_i915_gem_object *obj)
> return i915_gem_object_type_has(obj, I915_GEM_OBJECT_IS_SHRINKABLE);
> }
>
> +static inline bool
> +i915_gem_object_has_self_managed_shrink_list(const struct drm_i915_gem_object *obj)
> +{
> + return i915_gem_object_type_has(obj, I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST);
> +}
> +
> static inline bool
> i915_gem_object_is_proxy(const struct drm_i915_gem_object *obj)
> {
> @@ -531,6 +537,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
>
> void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> +void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> +void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
>
> static inline bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> index f4233c4e8d2e..9dbbca682a77 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> @@ -34,9 +34,11 @@ struct i915_lut_handle {
>
> struct drm_i915_gem_object_ops {
> unsigned int flags;
> -#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
> -#define I915_GEM_OBJECT_IS_PROXY BIT(2)
> -#define I915_GEM_OBJECT_NO_MMAP BIT(3)
> +#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
> +/* Skip the shrinker management in set_pages/unset_pages */
> +#define I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST BIT(2)
> +#define I915_GEM_OBJECT_IS_PROXY BIT(3)
> +#define I915_GEM_OBJECT_NO_MMAP BIT(4)
>
> /* Interface between the GEM object and its backing storage.
> * get_pages() is called once prior to the use of the associated set
> @@ -485,6 +487,23 @@ struct drm_i915_gem_object {
> */
> atomic_t shrink_pin;
>
> + /**
> + * @ttm_shrinker_status: Whether TTM is currently holding a
> + * shrink_pin for this object. Protected by the object lock.
> + *
> + * I915_TTM_SHRINKER_UNPINNED: We don't have an extra shrink_pin
> + * for this object. The underlying pages or ttm_tt is using
> + * shmem pages underneath, and as such this object might be
> + * currently visible to the shrinker.
> + *
> + * I915_TTM_SHRINKER_PINNED: We are holding shrink_pin for this
> + * object, which prevents the shrinker from seeing this object.
> + * The object is not currently using shmem pages undearneath.
> + */
> +#define I915_TTM_SHRINKER_UNPINNED 1
> +#define I915_TTM_SHRINKER_PINNED 2
> + int ttm_shrinker_status;
Would it be possible to use a single bool ttm_shrinkable here? Also see
below.
...
> +
> /**
> * Priority list of potential placements for this object.
> */
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> index ea6d9b3d2d6b..308e22a80af4 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> @@ -68,7 +68,7 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
> shrinkable = false;
> }
>
> - if (shrinkable) {
> + if (shrinkable && !i915_gem_object_has_self_managed_shrink_list(obj)) {
> struct list_head *list;
> unsigned long flags;
>
> @@ -210,7 +210,8 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj)
> if (i915_gem_object_is_volatile(obj))
> obj->mm.madv = I915_MADV_WILLNEED;
>
> - i915_gem_object_make_unshrinkable(obj);
> + if (!i915_gem_object_has_self_managed_shrink_list(obj))
> + i915_gem_object_make_unshrinkable(obj);
>
> if (obj->mm.mapping) {
> unmap_object(obj, page_mask_bits(obj->mm.mapping));
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> index 66121fedc655..dde0a5c232f8 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> @@ -497,13 +497,12 @@ void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> -static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> - struct list_head *head)
> +static void ___i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> + struct list_head *head)
> {
> struct drm_i915_private *i915 = obj_to_i915(obj);
> unsigned long flags;
>
> - GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> if (!i915_gem_object_is_shrinkable(obj))
> return;
>
> @@ -523,6 +522,38 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> +/**
> + * __i915_gem_object_make_shrinkable - Move the object to the tail of the
> + * shrinkable list. Objects on this list might be swapped out. Used with
> + * WILLNEED objects.
> + * @obj: The GEM object.
> + *
> + * DO NOT USE. This is intended to be called on very special objects that don't
> + * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
> + * underneath.
> + */
> +void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> +{
> + ___i915_gem_object_make_shrinkable(obj,
> + &obj_to_i915(obj)->mm.shrink_list);
> +}
> +
> +/**
> + * __i915_gem_object_make_purgeable - Move the object to the tail of the
> + * purgeable list. Objects on this list might be swapped out. Used with
> + * DONTNEED objects.
> + * @obj: The GEM object.
> + *
> + * DO NOT USE. This is intended to be called on very special objects that don't
> + * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
> + * underneath.
> + */
> +void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> +{
> + ___i915_gem_object_make_shrinkable(obj,
> + &obj_to_i915(obj)->mm.purge_list);
> +}
> +
> /**
> * i915_gem_object_make_shrinkable - Move the object to the tail of the
> * shrinkable list. Objects on this list might be swapped out. Used with
> @@ -535,8 +566,8 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> */
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> {
> - __i915_gem_object_make_shrinkable(obj,
> - &obj_to_i915(obj)->mm.shrink_list);
> + GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> + __i915_gem_object_make_shrinkable(obj);
> }
>
> /**
> @@ -552,6 +583,6 @@ void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> */
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> {
> - __i915_gem_object_make_shrinkable(obj,
> - &obj_to_i915(obj)->mm.purge_list);
> + GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> + __i915_gem_object_make_purgeable(obj);
> }
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> index 0fe1eb8f38e9..98b7ead1a008 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> @@ -851,7 +851,6 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
> return i915_ttm_err_to_gem(ret);
> }
>
> - i915_ttm_adjust_lru(obj);
> if (bo->ttm && !ttm_tt_is_populated(bo->ttm)) {
> ret = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
> if (ret)
> @@ -871,10 +870,9 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
> return PTR_ERR(st);
>
> __i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
> - if (!bo->ttm || !i915_tt->is_shmem)
> - i915_gem_object_make_unshrinkable(obj);
> }
>
> + i915_ttm_adjust_lru(obj);
> return ret;
> }
>
> @@ -945,8 +943,6 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
> * If the object is not destroyed next, The TTM eviction logic
> * and shrinkers will move it out if needed.
> */
> -
> - i915_ttm_adjust_lru(obj);
> }
>
> static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> @@ -954,6 +950,9 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
> struct i915_ttm_tt *i915_tt =
> container_of(bo->ttm, typeof(*i915_tt), ttm);
> + bool shrinkable =
> + bo->ttm && i915_tt->filp && ttm_tt_is_populated(bo->ttm);
> + int status;
>
> /*
> * Don't manipulate the TTM LRUs while in TTM bo destruction.
> @@ -962,11 +961,42 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> if (!kref_read(&bo->kref))
> return;
>
> + /*
> + * We skip managing the shrinker LRU in set_pages() and just manage
> + * everything here. This does at least solve the issue with having
> + * temporary shmem mappings(like with evicted lmem) not being visible to
> + * the shrinker. Only our shmem objects are shrinkable, everything else
> + * we keep as unshrinkable.
> + *
> + * To make sure everything plays nice we keep an extra shrink pin in TTM
> + * if the underlying pages are not currently shrinkable. Once we release
> + * our pin, like when the pages are moved to shmem, the pages will then
> + * be added to the shrinker LRU, assuming the caller isn't also holding
> + * a pin.
> + *
> + * TODO: consider maybe also bumping the shrinker list here when we have
> + * already unpinned it, which should give us something more like an LRU.
> + */
> + status = obj->mm.ttm_shrinker_status;
Then perhaps
if (shrinkable != obj->mm.ttm_shrinkable) {
...
obj->mm.ttm_shrinkable = shrinkable;
}
Otherwise LGTM,
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> + if (shrinkable) {
> + if (status != I915_TTM_SHRINKER_UNPINNED) {
> + if (obj->mm.madv == I915_MADV_WILLNEED)
> + __i915_gem_object_make_shrinkable(obj);
> + else
> + __i915_gem_object_make_purgeable(obj);
> + status = I915_TTM_SHRINKER_UNPINNED;
> + }
> + } else if (status != I915_TTM_SHRINKER_PINNED) {
> + i915_gem_object_make_unshrinkable(obj);
> + status = I915_TTM_SHRINKER_PINNED;
> + }
> + obj->mm.ttm_shrinker_status = status;
> +
> /*
> * Put on the correct LRU list depending on the MADV status
> */
> spin_lock(&bo->bdev->lru_lock);
> - if (bo->ttm && i915_tt->filp) {
> + if (shrinkable) {
> /* Try to keep shmem_tt from being considered for shrinking. */
> bo->priority = TTM_MAX_BO_PRIORITY - 1;
> } else if (obj->mm.madv != I915_MADV_WILLNEED) {
> @@ -1067,6 +1097,7 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
>
> static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
> .name = "i915_gem_object_ttm",
> + .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
>
> .get_pages = i915_ttm_get_pages,
> .put_pages = i915_ttm_put_pages,
> @@ -1089,6 +1120,19 @@ void i915_ttm_bo_destroy(struct ttm_buffer_object *bo)
> mutex_destroy(&obj->ttm.get_io_page.lock);
>
> if (obj->ttm.created) {
> + /*
> + * We freely manage the shrinker LRU outide of the mm.pages life
> + * cycle. As a result when destroying the object we should be
> + * extra paranoid and ensure we remove it from the LRU, before
> + * we free the object.
> + *
> + * Touching the ttm_shrinker_status outside of the object lock
> + * here should be safe now that the last GEM object ref was
> + * dropped.
> + */
> + if (obj->mm.ttm_shrinker_status == I915_TTM_SHRINKER_UNPINNED)
> + i915_gem_object_make_unshrinkable(obj);
> +
> i915_ttm_backup_free(obj);
>
> /* This releases all gem object bindings to the backend. */
> @@ -1141,6 +1185,15 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
> /* Forcing the page size is kernel internal only */
> GEM_BUG_ON(page_size && obj->mm.n_placements);
>
> + /*
> + * Keep an extra shrink pin to prevent the object from being made
> + * shrinkable too early. If the ttm_tt is ever allocated in shmem, we
> + * drop the pin. The TTM backend manages the shrinker LRU itself,
> + * outside of the normal mm.pages life cycle.
> + */
> + i915_gem_object_make_unshrinkable(obj);
> + obj->mm.ttm_shrinker_status = I915_TTM_SHRINKER_PINNED;
> +
> /*
> * If this function fails, it will call the destructor, but
> * our caller still owns the object. So no freeing in the
^ permalink raw reply [flat|nested] 28+ messages in thread
* Re: [Intel-gfx] [PATCH v6 6/8] drm/i915/ttm: move shrinker management into adjust_lru
@ 2021-10-06 8:10 ` Thomas Hellström
0 siblings, 0 replies; 28+ messages in thread
From: Thomas Hellström @ 2021-10-06 8:10 UTC (permalink / raw)
To: Matthew Auld, intel-gfx; +Cc: dri-devel
On 10/5/21 20:24, Matthew Auld wrote:
> We currently just evict lmem objects to system memory when under memory
> pressure. For this case we might lack the usual object mm.pages, which
> effectively hides the pages from the i915-gem shrinker, until we
> actually "attach" the TT to the object, or in the case of lmem-only
> objects it just gets migrated back to lmem when touched again.
>
> For all cases we can just adjust the i915 shrinker LRU each time we also
> adjust the TTM LRU. The two cases we care about are:
>
> 1) When something is moved by TTM, including when initially populating
> an object. Importantly this covers the case where TTM moves something from
> lmem <-> smem, outside of the normal get_pages() interface, which
> should still ensure the shmem pages underneath are reclaimable.
>
> 2) When calling into i915_gem_object_unlock(). The unlock should
> ensure the object is removed from the shinker LRU, if it was indeed
> swapped out, or just purged, when the shrinker drops the object lock.
>
> v2(Thomas):
> - Handle managing the shrinker LRU in adjust_lru, where it is always
> safe to touch the object.
> v3(Thomas):
> - Pretty much a re-write. This time piggy back off the shrink_pin
> stuff, which actually seems to fit quite well for what we want here.
>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> ---
> drivers/gpu/drm/i915/gem/i915_gem_object.h | 8 +++
> .../gpu/drm/i915/gem/i915_gem_object_types.h | 25 ++++++-
> drivers/gpu/drm/i915/gem/i915_gem_pages.c | 5 +-
> drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 45 +++++++++++--
> drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 65 +++++++++++++++++--
> 5 files changed, 130 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> index e641db297e0e..3eac8cf2ae10 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
> @@ -294,6 +294,12 @@ i915_gem_object_is_shrinkable(const struct drm_i915_gem_object *obj)
> return i915_gem_object_type_has(obj, I915_GEM_OBJECT_IS_SHRINKABLE);
> }
>
> +static inline bool
> +i915_gem_object_has_self_managed_shrink_list(const struct drm_i915_gem_object *obj)
> +{
> + return i915_gem_object_type_has(obj, I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST);
> +}
> +
> static inline bool
> i915_gem_object_is_proxy(const struct drm_i915_gem_object *obj)
> {
> @@ -531,6 +537,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
>
> void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj);
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> +void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj);
> +void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj);
>
> static inline bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> index f4233c4e8d2e..9dbbca682a77 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
> @@ -34,9 +34,11 @@ struct i915_lut_handle {
>
> struct drm_i915_gem_object_ops {
> unsigned int flags;
> -#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
> -#define I915_GEM_OBJECT_IS_PROXY BIT(2)
> -#define I915_GEM_OBJECT_NO_MMAP BIT(3)
> +#define I915_GEM_OBJECT_IS_SHRINKABLE BIT(1)
> +/* Skip the shrinker management in set_pages/unset_pages */
> +#define I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST BIT(2)
> +#define I915_GEM_OBJECT_IS_PROXY BIT(3)
> +#define I915_GEM_OBJECT_NO_MMAP BIT(4)
>
> /* Interface between the GEM object and its backing storage.
> * get_pages() is called once prior to the use of the associated set
> @@ -485,6 +487,23 @@ struct drm_i915_gem_object {
> */
> atomic_t shrink_pin;
>
> + /**
> + * @ttm_shrinker_status: Whether TTM is currently holding a
> + * shrink_pin for this object. Protected by the object lock.
> + *
> + * I915_TTM_SHRINKER_UNPINNED: We don't have an extra shrink_pin
> + * for this object. The underlying pages or ttm_tt is using
> + * shmem pages underneath, and as such this object might be
> + * currently visible to the shrinker.
> + *
> + * I915_TTM_SHRINKER_PINNED: We are holding shrink_pin for this
> + * object, which prevents the shrinker from seeing this object.
> + * The object is not currently using shmem pages undearneath.
> + */
> +#define I915_TTM_SHRINKER_UNPINNED 1
> +#define I915_TTM_SHRINKER_PINNED 2
> + int ttm_shrinker_status;
Would it be possible to use a single bool ttm_shrinkable here? Also see
below.
...
> +
> /**
> * Priority list of potential placements for this object.
> */
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> index ea6d9b3d2d6b..308e22a80af4 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> @@ -68,7 +68,7 @@ void __i915_gem_object_set_pages(struct drm_i915_gem_object *obj,
> shrinkable = false;
> }
>
> - if (shrinkable) {
> + if (shrinkable && !i915_gem_object_has_self_managed_shrink_list(obj)) {
> struct list_head *list;
> unsigned long flags;
>
> @@ -210,7 +210,8 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj)
> if (i915_gem_object_is_volatile(obj))
> obj->mm.madv = I915_MADV_WILLNEED;
>
> - i915_gem_object_make_unshrinkable(obj);
> + if (!i915_gem_object_has_self_managed_shrink_list(obj))
> + i915_gem_object_make_unshrinkable(obj);
>
> if (obj->mm.mapping) {
> unmap_object(obj, page_mask_bits(obj->mm.mapping));
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> index 66121fedc655..dde0a5c232f8 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
> @@ -497,13 +497,12 @@ void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj)
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> -static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> - struct list_head *head)
> +static void ___i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> + struct list_head *head)
> {
> struct drm_i915_private *i915 = obj_to_i915(obj);
> unsigned long flags;
>
> - GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> if (!i915_gem_object_is_shrinkable(obj))
> return;
>
> @@ -523,6 +522,38 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> spin_unlock_irqrestore(&i915->mm.obj_lock, flags);
> }
>
> +/**
> + * __i915_gem_object_make_shrinkable - Move the object to the tail of the
> + * shrinkable list. Objects on this list might be swapped out. Used with
> + * WILLNEED objects.
> + * @obj: The GEM object.
> + *
> + * DO NOT USE. This is intended to be called on very special objects that don't
> + * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
> + * underneath.
> + */
> +void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> +{
> + ___i915_gem_object_make_shrinkable(obj,
> + &obj_to_i915(obj)->mm.shrink_list);
> +}
> +
> +/**
> + * __i915_gem_object_make_purgeable - Move the object to the tail of the
> + * purgeable list. Objects on this list might be swapped out. Used with
> + * DONTNEED objects.
> + * @obj: The GEM object.
> + *
> + * DO NOT USE. This is intended to be called on very special objects that don't
> + * yet have mm.pages, but are guaranteed to have potentially reclaimable pages
> + * underneath.
> + */
> +void __i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> +{
> + ___i915_gem_object_make_shrinkable(obj,
> + &obj_to_i915(obj)->mm.purge_list);
> +}
> +
> /**
> * i915_gem_object_make_shrinkable - Move the object to the tail of the
> * shrinkable list. Objects on this list might be swapped out. Used with
> @@ -535,8 +566,8 @@ static void __i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj,
> */
> void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> {
> - __i915_gem_object_make_shrinkable(obj,
> - &obj_to_i915(obj)->mm.shrink_list);
> + GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> + __i915_gem_object_make_shrinkable(obj);
> }
>
> /**
> @@ -552,6 +583,6 @@ void i915_gem_object_make_shrinkable(struct drm_i915_gem_object *obj)
> */
> void i915_gem_object_make_purgeable(struct drm_i915_gem_object *obj)
> {
> - __i915_gem_object_make_shrinkable(obj,
> - &obj_to_i915(obj)->mm.purge_list);
> + GEM_BUG_ON(!i915_gem_object_has_pages(obj));
> + __i915_gem_object_make_purgeable(obj);
> }
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> index 0fe1eb8f38e9..98b7ead1a008 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c
> @@ -851,7 +851,6 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
> return i915_ttm_err_to_gem(ret);
> }
>
> - i915_ttm_adjust_lru(obj);
> if (bo->ttm && !ttm_tt_is_populated(bo->ttm)) {
> ret = ttm_tt_populate(bo->bdev, bo->ttm, &ctx);
> if (ret)
> @@ -871,10 +870,9 @@ static int __i915_ttm_get_pages(struct drm_i915_gem_object *obj,
> return PTR_ERR(st);
>
> __i915_gem_object_set_pages(obj, st, i915_sg_dma_sizes(st->sgl));
> - if (!bo->ttm || !i915_tt->is_shmem)
> - i915_gem_object_make_unshrinkable(obj);
> }
>
> + i915_ttm_adjust_lru(obj);
> return ret;
> }
>
> @@ -945,8 +943,6 @@ static void i915_ttm_put_pages(struct drm_i915_gem_object *obj,
> * If the object is not destroyed next, The TTM eviction logic
> * and shrinkers will move it out if needed.
> */
> -
> - i915_ttm_adjust_lru(obj);
> }
>
> static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> @@ -954,6 +950,9 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> struct ttm_buffer_object *bo = i915_gem_to_ttm(obj);
> struct i915_ttm_tt *i915_tt =
> container_of(bo->ttm, typeof(*i915_tt), ttm);
> + bool shrinkable =
> + bo->ttm && i915_tt->filp && ttm_tt_is_populated(bo->ttm);
> + int status;
>
> /*
> * Don't manipulate the TTM LRUs while in TTM bo destruction.
> @@ -962,11 +961,42 @@ static void i915_ttm_adjust_lru(struct drm_i915_gem_object *obj)
> if (!kref_read(&bo->kref))
> return;
>
> + /*
> + * We skip managing the shrinker LRU in set_pages() and just manage
> + * everything here. This does at least solve the issue with having
> + * temporary shmem mappings(like with evicted lmem) not being visible to
> + * the shrinker. Only our shmem objects are shrinkable, everything else
> + * we keep as unshrinkable.
> + *
> + * To make sure everything plays nice we keep an extra shrink pin in TTM
> + * if the underlying pages are not currently shrinkable. Once we release
> + * our pin, like when the pages are moved to shmem, the pages will then
> + * be added to the shrinker LRU, assuming the caller isn't also holding
> + * a pin.
> + *
> + * TODO: consider maybe also bumping the shrinker list here when we have
> + * already unpinned it, which should give us something more like an LRU.
> + */
> + status = obj->mm.ttm_shrinker_status;
Then perhaps
if (shrinkable != obj->mm.ttm_shrinkable) {
...
obj->mm.ttm_shrinkable = shrinkable;
}
Otherwise LGTM,
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> + if (shrinkable) {
> + if (status != I915_TTM_SHRINKER_UNPINNED) {
> + if (obj->mm.madv == I915_MADV_WILLNEED)
> + __i915_gem_object_make_shrinkable(obj);
> + else
> + __i915_gem_object_make_purgeable(obj);
> + status = I915_TTM_SHRINKER_UNPINNED;
> + }
> + } else if (status != I915_TTM_SHRINKER_PINNED) {
> + i915_gem_object_make_unshrinkable(obj);
> + status = I915_TTM_SHRINKER_PINNED;
> + }
> + obj->mm.ttm_shrinker_status = status;
> +
> /*
> * Put on the correct LRU list depending on the MADV status
> */
> spin_lock(&bo->bdev->lru_lock);
> - if (bo->ttm && i915_tt->filp) {
> + if (shrinkable) {
> /* Try to keep shmem_tt from being considered for shrinking. */
> bo->priority = TTM_MAX_BO_PRIORITY - 1;
> } else if (obj->mm.madv != I915_MADV_WILLNEED) {
> @@ -1067,6 +1097,7 @@ static u64 i915_ttm_mmap_offset(struct drm_i915_gem_object *obj)
>
> static const struct drm_i915_gem_object_ops i915_gem_ttm_obj_ops = {
> .name = "i915_gem_object_ttm",
> + .flags = I915_GEM_OBJECT_SELF_MANAGED_SHRINK_LIST,
>
> .get_pages = i915_ttm_get_pages,
> .put_pages = i915_ttm_put_pages,
> @@ -1089,6 +1120,19 @@ void i915_ttm_bo_destroy(struct ttm_buffer_object *bo)
> mutex_destroy(&obj->ttm.get_io_page.lock);
>
> if (obj->ttm.created) {
> + /*
> + * We freely manage the shrinker LRU outide of the mm.pages life
> + * cycle. As a result when destroying the object we should be
> + * extra paranoid and ensure we remove it from the LRU, before
> + * we free the object.
> + *
> + * Touching the ttm_shrinker_status outside of the object lock
> + * here should be safe now that the last GEM object ref was
> + * dropped.
> + */
> + if (obj->mm.ttm_shrinker_status == I915_TTM_SHRINKER_UNPINNED)
> + i915_gem_object_make_unshrinkable(obj);
> +
> i915_ttm_backup_free(obj);
>
> /* This releases all gem object bindings to the backend. */
> @@ -1141,6 +1185,15 @@ int __i915_gem_ttm_object_init(struct intel_memory_region *mem,
> /* Forcing the page size is kernel internal only */
> GEM_BUG_ON(page_size && obj->mm.n_placements);
>
> + /*
> + * Keep an extra shrink pin to prevent the object from being made
> + * shrinkable too early. If the ttm_tt is ever allocated in shmem, we
> + * drop the pin. The TTM backend manages the shrinker LRU itself,
> + * outside of the normal mm.pages life cycle.
> + */
> + i915_gem_object_make_unshrinkable(obj);
> + obj->mm.ttm_shrinker_status = I915_TTM_SHRINKER_PINNED;
> +
> /*
> * If this function fails, it will call the destructor, but
> * our caller still owns the object. So no freeing in the
^ permalink raw reply [flat|nested] 28+ messages in thread
end of thread, other threads:[~2021-10-06 8:10 UTC | newest]
Thread overview: 28+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-05 18:23 [PATCH v6 1/8] drm/i915/gem: Break out some shmem backend utils Matthew Auld
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
2021-10-05 18:23 ` [PATCH v6 2/8] drm/i915/ttm: add tt shmem backend Matthew Auld
2021-10-05 18:23 ` [Intel-gfx] " Matthew Auld
2021-10-05 18:24 ` [PATCH v6 3/8] drm/i915/gtt: drop unneeded make_unshrinkable Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-06 8:02 ` Thomas Hellström
2021-10-06 8:02 ` [Intel-gfx] " Thomas Hellström
2021-10-05 18:24 ` [PATCH v6 4/8] drm/i915: drop unneeded make_unshrinkable in free_object Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-06 8:03 ` Thomas Hellström
2021-10-06 8:03 ` [Intel-gfx] " Thomas Hellström
2021-10-05 18:24 ` [PATCH v6 5/8] drm/i915: add some kernel-doc for shrink_pin and friends Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-06 8:03 ` Thomas Hellström
2021-10-06 8:03 ` [Intel-gfx] " Thomas Hellström
2021-10-05 18:24 ` [PATCH v6 6/8] drm/i915/ttm: move shrinker management into adjust_lru Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-06 8:10 ` Thomas Hellström
2021-10-06 8:10 ` [Intel-gfx] " Thomas Hellström
2021-10-05 18:24 ` [PATCH v6 7/8] drm/i915/ttm: use cached system pages when evicting lmem Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-05 18:24 ` [PATCH v6 8/8] drm/i915/ttm: enable shmem tt backend Matthew Auld
2021-10-05 18:24 ` [Intel-gfx] " Matthew Auld
2021-10-05 23:55 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [v6,1/8] drm/i915/gem: Break out some shmem backend utils Patchwork
2021-10-05 23:57 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-10-06 0:27 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-10-06 5:17 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.