[PATCH 07/10] drm/i915: internal buffers use ttm backend
Robert Beckett
bob.beckett at collabora.com
Mon Jun 6 18:29:25 UTC 2022
Create a kernel only internal memory region that uses ttm pool allocator to
allocate volatile system pages.
Refactor internal buffer backend to simply allocate from this new
region.
Signed-off-by: Robert Beckett <bob.beckett at collabora.com>
---
drivers/gpu/drm/i915/gem/i915_gem_internal.c | 187 +------------------
drivers/gpu/drm/i915/gem/i915_gem_internal.h | 5 -
drivers/gpu/drm/i915/gem/i915_gem_ttm.h | 1 +
drivers/gpu/drm/i915/i915_pci.c | 4 +-
drivers/gpu/drm/i915/intel_memory_region.c | 6 +-
drivers/gpu/drm/i915/intel_memory_region.h | 2 +
6 files changed, 15 insertions(+), 190 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.c b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
index c698f95af15f..a83751867ac7 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_internal.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.c
@@ -4,188 +4,9 @@
* Copyright © 2014-2016 Intel Corporation
*/
-#include <linux/scatterlist.h>
-#include <linux/slab.h>
-#include <linux/swiotlb.h>
-
+#include "gem/i915_gem_internal.h"
+#include "gem/i915_gem_region.h"
#include "i915_drv.h"
-#include "i915_gem.h"
-#include "i915_gem_internal.h"
-#include "i915_gem_object.h"
-#include "i915_scatterlist.h"
-#include "i915_utils.h"
-
-#define QUIET (__GFP_NORETRY | __GFP_NOWARN)
-#define MAYFAIL (__GFP_RETRY_MAYFAIL | __GFP_NOWARN)
-
-static void internal_free_pages(struct sg_table *st)
-{
- struct scatterlist *sg;
-
- for (sg = st->sgl; sg; sg = __sg_next(sg)) {
- if (sg_page(sg))
- __free_pages(sg_page(sg), get_order(sg->length));
- }
-
- sg_free_table(st);
- kfree(st);
-}
-
-static int i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj)
-{
- struct drm_i915_private *i915 = to_i915(obj->base.dev);
- struct sg_table *st;
- struct scatterlist *sg;
- unsigned int sg_page_sizes;
- unsigned int npages;
- int max_order;
- gfp_t gfp;
-
- max_order = MAX_ORDER;
-#ifdef CONFIG_SWIOTLB
- if (is_swiotlb_active(obj->base.dev->dev)) {
- unsigned int max_segment;
-
- max_segment = swiotlb_max_segment();
- if (max_segment) {
- max_segment = max_t(unsigned int, max_segment,
- PAGE_SIZE) >> PAGE_SHIFT;
- max_order = min(max_order, ilog2(max_segment));
- }
- }
-#endif
-
- gfp = GFP_KERNEL | __GFP_HIGHMEM | __GFP_RECLAIMABLE;
- if (IS_I965GM(i915) || IS_I965G(i915)) {
- /* 965gm cannot relocate objects above 4GiB. */
- gfp &= ~__GFP_HIGHMEM;
- gfp |= __GFP_DMA32;
- }
-
-create_st:
- st = kmalloc(sizeof(*st), GFP_KERNEL);
- if (!st)
- return -ENOMEM;
-
- npages = obj->base.size / PAGE_SIZE;
- if (sg_alloc_table(st, npages, GFP_KERNEL)) {
- kfree(st);
- return -ENOMEM;
- }
-
- sg = st->sgl;
- st->nents = 0;
- sg_page_sizes = 0;
-
- do {
- int order = min(fls(npages) - 1, max_order);
- struct page *page;
-
- do {
- page = alloc_pages(gfp | (order ? QUIET : MAYFAIL),
- order);
- if (page)
- break;
- if (!order--)
- goto err;
-
- /* Limit subsequent allocations as well */
- max_order = order;
- } while (1);
-
- sg_set_page(sg, page, PAGE_SIZE << order, 0);
- sg_page_sizes |= PAGE_SIZE << order;
- st->nents++;
-
- npages -= 1 << order;
- if (!npages) {
- sg_mark_end(sg);
- break;
- }
-
- sg = __sg_next(sg);
- } while (1);
-
- if (i915_gem_gtt_prepare_pages(obj, st)) {
- /* Failed to dma-map try again with single page sg segments */
- if (get_order(st->sgl->length)) {
- internal_free_pages(st);
- max_order = 0;
- goto create_st;
- }
- goto err;
- }
-
- __i915_gem_object_set_pages(obj, st, sg_page_sizes);
-
- return 0;
-
-err:
- sg_set_page(sg, NULL, 0, 0);
- sg_mark_end(sg);
- internal_free_pages(st);
-
- return -ENOMEM;
-}
-
-static void i915_gem_object_put_pages_internal(struct drm_i915_gem_object *obj,
- struct sg_table *pages)
-{
- i915_gem_gtt_finish_pages(obj, pages);
- internal_free_pages(pages);
-
- obj->mm.dirty = false;
-
- __start_cpu_write(obj);
-}
-
-static const struct drm_i915_gem_object_ops i915_gem_object_internal_ops = {
- .name = "i915_gem_object_internal",
- .flags = I915_GEM_OBJECT_IS_SHRINKABLE,
- .get_pages = i915_gem_object_get_pages_internal,
- .put_pages = i915_gem_object_put_pages_internal,
-};
-
-struct drm_i915_gem_object *
-__i915_gem_object_create_internal(struct drm_i915_private *i915,
- const struct drm_i915_gem_object_ops *ops,
- phys_addr_t size)
-{
- static struct lock_class_key lock_class;
- struct drm_i915_gem_object *obj;
- unsigned int cache_level;
-
- GEM_BUG_ON(!size);
- GEM_BUG_ON(!IS_ALIGNED(size, PAGE_SIZE));
-
- if (overflows_type(size, obj->base.size))
- return ERR_PTR(-E2BIG);
-
- obj = i915_gem_object_alloc();
- if (!obj)
- return ERR_PTR(-ENOMEM);
-
- drm_gem_private_object_init(&i915->drm, &obj->base, size);
- i915_gem_object_init(obj, ops, &lock_class, 0);
- obj->mem_flags |= I915_BO_FLAG_STRUCT_PAGE;
-
- /*
- * Mark the object as volatile, such that the pages are marked as
- * dontneed whilst they are still pinned. As soon as they are unpinned
- * they are allowed to be reaped by the shrinker, and the caller is
- * expected to repopulate - the contents of this object are only valid
- * whilst active and pinned.
- */
- i915_gem_object_set_volatile(obj);
-
- obj->read_domains = I915_GEM_DOMAIN_CPU;
- obj->write_domain = I915_GEM_DOMAIN_CPU;
-
- cache_level = HAS_LLC(i915) ? I915_CACHE_LLC : I915_CACHE_NONE;
- i915_gem_object_set_cache_coherency(obj, cache_level);
-
- return obj;
-}
/**
* i915_gem_object_create_internal: create an object with volatile pages
@@ -206,5 +27,7 @@ struct drm_i915_gem_object *
i915_gem_object_create_internal(struct drm_i915_private *i915,
phys_addr_t size)
{
- return __i915_gem_object_create_internal(i915, &i915_gem_object_internal_ops, size);
+ return i915_gem_object_create_region(i915->mm.regions[INTEL_REGION_INTERNAL],
+ size, 0, I915_BO_ALLOC_VOLATILE);
}
+
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_internal.h b/drivers/gpu/drm/i915/gem/i915_gem_internal.h
index 6664e06112fc..524e1042b20f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_internal.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_internal.h
@@ -15,9 +15,4 @@ struct drm_i915_private;
struct drm_i915_gem_object *
i915_gem_object_create_internal(struct drm_i915_private *i915,
phys_addr_t size);
-struct drm_i915_gem_object *
-__i915_gem_object_create_internal(struct drm_i915_private *i915,
- const struct drm_i915_gem_object_ops *ops,
- phys_addr_t size);
-
#endif /* __I915_GEM_INTERNAL_H__ */
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.h b/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
index 73e371aa3850..cb4cab8a13e1 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
@@ -92,4 +92,5 @@ static inline bool i915_ttm_cpu_maps_iomem(struct ttm_resource *mem)
/* Once / if we support GGTT, this is also false for cached ttm_tts */
return mem->mem_type != I915_PL_SYSTEM;
}
+
#endif
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 5ad9884874c2..58f2bb21d888 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -159,7 +159,7 @@
.page_sizes = I915_GTT_PAGE_SIZE_4K
#define GEN_DEFAULT_REGIONS \
- .memory_regions = REGION_SMEM | REGION_STOLEN_SMEM
+ .memory_regions = REGION_SMEM | REGION_STOLEN_SMEM | REGION_INTERNAL
#define I830_FEATURES \
GEN(2), \
@@ -897,7 +897,7 @@ static const struct intel_device_info rkl_info = {
};
#define DGFX_FEATURES \
- .memory_regions = REGION_SMEM | REGION_LMEM | REGION_STOLEN_LMEM, \
+ .memory_regions = REGION_SMEM | REGION_LMEM | REGION_STOLEN_LMEM | REGION_INTERNAL, \
.has_llc = 0, \
.has_pxp = 0, \
.has_snoop = 1, \
diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c
index e38d2db1c3e3..02b7e6dae78f 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/intel_memory_region.c
@@ -31,6 +31,10 @@ static const struct {
.class = INTEL_MEMORY_STOLEN_LOCAL,
.instance = 0,
},
+ [INTEL_REGION_INTERNAL] = {
+ .class = INTEL_MEMORY_SYSTEM,
+ .instance = 1,
+ }
};
static int __iopagetest(struct intel_memory_region *mem,
@@ -309,7 +313,7 @@ int intel_memory_regions_hw_probe(struct drm_i915_private *i915)
instance = intel_region_map[i].instance;
switch (type) {
case INTEL_MEMORY_SYSTEM:
- if (IS_DGFX(i915))
+ if (IS_DGFX(i915) || i == INTEL_REGION_INTERNAL)
mem = i915_gem_ttm_system_setup(i915, type,
instance);
else
diff --git a/drivers/gpu/drm/i915/intel_memory_region.h b/drivers/gpu/drm/i915/intel_memory_region.h
index 3d8378c1b447..4038913bc467 100644
--- a/drivers/gpu/drm/i915/intel_memory_region.h
+++ b/drivers/gpu/drm/i915/intel_memory_region.h
@@ -35,6 +35,7 @@ enum intel_region_id {
INTEL_REGION_LMEM_3,
INTEL_REGION_STOLEN_SMEM,
INTEL_REGION_STOLEN_LMEM,
+ INTEL_REGION_INTERNAL,
INTEL_REGION_UNKNOWN, /* Should be last */
};
@@ -42,6 +43,7 @@ enum intel_region_id {
#define REGION_LMEM BIT(INTEL_REGION_LMEM_0)
#define REGION_STOLEN_SMEM BIT(INTEL_REGION_STOLEN_SMEM)
#define REGION_STOLEN_LMEM BIT(INTEL_REGION_STOLEN_LMEM)
+#define REGION_INTERNAL BIT(INTEL_REGION_INTERNAL)
#define I915_ALLOC_CONTIGUOUS BIT(0)
--
2.25.1
More information about the Intel-gfx-trybot
mailing list