torvalds/linux
3
0
Fork 0
mirror of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2026-03-22 07:27:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

drm/i915: Use huge tmpfs mountpoint helpers

Browse Source 
Make use of the new drm_gem_huge_mnt_create() and
drm_gem_get_huge_mnt() helpers to avoid code duplication. Now that
it's just a few lines long, the single function in i915_gemfs.c is
moved into i915_gem_shmem.c.

v3:
- use huge tmpfs mountpoint in drm_device
- move i915_gemfs.c into i915_gem_shmem.c

v4:
- clean up mountpoint creation error handling

v5:
- use drm_gem_has_huge_mnt() helper

v7:
- include <drm/drm_print.h> in i915_gem_shmem.c

v8:
- keep logging notice message with CONFIG_TRANSPARENT_HUGEPAGE=n
- don't access huge_mnt field with CONFIG_TRANSPARENT_HUGEPAGE=n

v9:
- replace drm_gem_has_huge_mnt() by drm_gem_get_huge_mnt()
- remove useless ternary op test in selftests/huge_pages.c

v12:
- fix layering violation in selftests (Tvrtko)
- fix incorrect filename in commit message

v13:
- add Tvrtko A-b

Signed-off-by: Loïc Molinari <loic.molinari@collabora.com>
Acked-by: Tvrtko Ursulin <tursulin@ursulin.net>
Link: https://patch.msgid.link/20251205182231.194072-6-loic.molinari@collabora.com
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
This commit is contained in:
Loïc Molinari
2025-12-05 19:22:26 +01:00
committed by Boris Brezillon
parent 6e0b1b8201
commit a8a9a59022
7 changed files with 50 additions and 125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
drivers/gpu/drm/i915/Makefile
View File

@@ -169,8 +169,7 @@ gem-y += \
gem/i915_gem_ttm_move.o \
gem/i915_gem_ttm_pm.o \
gem/i915_gem_userptr.o \
gem/i915_gem_wait.o \
gem/i915_gemfs.o
gem/i915_gem_wait.o
i915-y += \
$(gem-y) \
i915_active.o \

9
drivers/gpu/drm/i915/gem/i915_gem_object_types.h
View File

@@ -348,12 +348,13 @@ struct drm_i915_gem_object {
*/
#define I915_BO_ALLOC_GPU_ONLY BIT(6)
#define I915_BO_ALLOC_CCS_AUX BIT(7)
#define I915_BO_ALLOC_NOTHP BIT(8)
/*
* Object is allowed to retain its initial data and will not be cleared on first
* access if used along with I915_BO_ALLOC_USER. This is mainly to keep
* preallocated framebuffer data intact while transitioning it to i915drmfb.
*/
#define I915_BO_PREALLOC BIT(8)
#define I915_BO_PREALLOC BIT(9)
#define I915_BO_ALLOC_FLAGS (I915_BO_ALLOC_CONTIGUOUS | \
I915_BO_ALLOC_VOLATILE | \
I915_BO_ALLOC_CPU_CLEAR | \
@@ -363,9 +364,9 @@ struct drm_i915_gem_object {
I915_BO_ALLOC_GPU_ONLY | \
I915_BO_ALLOC_CCS_AUX | \
I915_BO_PREALLOC)
#define I915_BO_READONLY BIT(9)
#define I915_TILING_QUIRK_BIT 10 /* unknown swizzling; do not release! */
#define I915_BO_PROTECTED BIT(11)
#define I915_BO_READONLY BIT(10)
#define I915_TILING_QUIRK_BIT 11 /* unknown swizzling; do not release! */
#define I915_BO_PROTECTED BIT(12)
/**
* @mem_flags - Mutable placement-related flags
*

58
drivers/gpu/drm/i915/gem/i915_gem_shmem.c
View File

@@ -9,14 +9,16 @@
#include <linux/uio.h>
#include <drm/drm_cache.h>
#include <drm/drm_gem.h>
#include <drm/drm_print.h>
#include "gem/i915_gem_region.h"
#include "i915_drv.h"
#include "i915_gem_object.h"
#include "i915_gem_tiling.h"
#include "i915_gemfs.h"
#include "i915_scatterlist.h"
#include "i915_trace.h"
#include "i915_utils.h"
/*
* Move folios to appropriate lru and release the batch, decrementing the
@@ -494,9 +496,11 @@ const struct drm_i915_gem_object_ops i915_gem_shmem_ops = {
static int __create_shmem(struct drm_i915_private *i915,
struct drm_gem_object *obj,
resource_size_t size)
resource_size_t size,
unsigned int flags)
{
unsigned long flags = VM_NORESERVE;
unsigned long shmem_flags = VM_NORESERVE;
struct vfsmount *huge_mnt;
struct file *filp;
drm_gem_private_object_init(&i915->drm, obj, size);
@@ -515,11 +519,12 @@ static int __create_shmem(struct drm_i915_private *i915,
if (BITS_PER_LONG == 64 && size > MAX_LFS_FILESIZE)
return -E2BIG;
if (i915->mm.gemfs)
filp = shmem_file_setup_with_mnt(i915->mm.gemfs, "i915", size,
flags);
huge_mnt = drm_gem_get_huge_mnt(&i915->drm);
if (!(flags & I915_BO_ALLOC_NOTHP) && huge_mnt)
filp = shmem_file_setup_with_mnt(huge_mnt, "i915", size,
shmem_flags);
else
filp = shmem_file_setup("i915", size, flags);
filp = shmem_file_setup("i915", size, shmem_flags);
if (IS_ERR(filp))
return PTR_ERR(filp);
@@ -548,7 +553,7 @@ static int shmem_object_init(struct intel_memory_region *mem,
gfp_t mask;
int ret;
ret = __create_shmem(i915, &obj->base, size);
ret = __create_shmem(i915, &obj->base, size, flags);
if (ret)
return ret;
@@ -644,21 +649,40 @@ fail:
static int init_shmem(struct intel_memory_region *mem)
{
i915_gemfs_init(mem->i915);
struct drm_i915_private *i915 = mem->i915;
/*
* By creating our own shmemfs mountpoint, we can pass in
* mount flags that better match our usecase.
*
* One example, although it is probably better with a per-file
* control, is selecting huge page allocations ("huge=within_size").
* However, we only do so on platforms which benefit from it, or to
* offset the overhead of iommu lookups, where with latter it is a net
* win even on platforms which would otherwise see some performance
* regressions such a slow reads issue on Broadwell and Skylake.
*/
if (GRAPHICS_VER(i915) < 11 && !i915_vtd_active(i915))
goto no_thp;
drm_gem_huge_mnt_create(&i915->drm, "within_size");
if (drm_gem_get_huge_mnt(&i915->drm))
drm_info(&i915->drm, "Using Transparent Hugepages\n");
else
drm_notice(&i915->drm,
"Transparent Hugepage support is recommended for optimal performance%s\n",
GRAPHICS_VER(i915) >= 11 ? " on this platform!" :
" when IOMMU is enabled!");
no_thp:
intel_memory_region_set_name(mem, "system");
return 0; /* We have fallback to the kernel mnt if gemfs init failed. */
}
static int release_shmem(struct intel_memory_region *mem)
{
i915_gemfs_fini(mem->i915);
return 0;
return 0; /* We have fallback to the kernel mnt if huge mnt failed. */
}
static const struct intel_memory_region_ops shmem_region_ops = {
.init = init_shmem,
.release = release_shmem,
.init_object = shmem_object_init,
};

71
drivers/gpu/drm/i915/gem/i915_gemfs.c
View File

@@ -1,71 +0,0 @@
// SPDX-License-Identifier: MIT
/*
* Copyright © 2017 Intel Corporation
*/
#include <linux/fs.h>
#include <linux/mount.h>
#include <linux/fs_context.h>
#include <drm/drm_print.h>
#include "i915_drv.h"
#include "i915_gemfs.h"
#include "i915_utils.h"
void i915_gemfs_init(struct drm_i915_private *i915)
{
struct file_system_type *type;
struct fs_context *fc;
struct vfsmount *gemfs;
int ret;
/*
* By creating our own shmemfs mountpoint, we can pass in
* mount flags that better match our usecase.
*
* One example, although it is probably better with a per-file
* control, is selecting huge page allocations ("huge=within_size").
* However, we only do so on platforms which benefit from it, or to
* offset the overhead of iommu lookups, where with latter it is a net
* win even on platforms which would otherwise see some performance
* regressions such a slow reads issue on Broadwell and Skylake.
*/
if (GRAPHICS_VER(i915) < 11 && !i915_vtd_active(i915))
return;
if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE))
goto err;
type = get_fs_type("tmpfs");
if (!type)
goto err;
fc = fs_context_for_mount(type, SB_KERNMOUNT);
if (IS_ERR(fc))
goto err;
ret = vfs_parse_fs_string(fc, "source", "tmpfs");
if (!ret)
ret = vfs_parse_fs_string(fc, "huge", "within_size");
if (!ret)
gemfs = fc_mount_longterm(fc);
put_fs_context(fc);
if (ret)
goto err;
i915->mm.gemfs = gemfs;
drm_info(&i915->drm, "Using Transparent Hugepages\n");
return;
err:
drm_notice(&i915->drm,
"Transparent Hugepage support is recommended for optimal performance%s\n",
GRAPHICS_VER(i915) >= 11 ? " on this platform!" :
" when IOMMU is enabled!");
}
void i915_gemfs_fini(struct drm_i915_private *i915)
{
kern_unmount(i915->mm.gemfs);
}

14
drivers/gpu/drm/i915/gem/i915_gemfs.h
View File

@@ -1,14 +0,0 @@
/* SPDX-License-Identifier: MIT */
/*
* Copyright © 2017 Intel Corporation
*/
#ifndef __I915_GEMFS_H__
#define __I915_GEMFS_H__
struct drm_i915_private;
void i915_gemfs_init(struct drm_i915_private *i915);
void i915_gemfs_fini(struct drm_i915_private *i915);
#endif

15
drivers/gpu/drm/i915/gem/selftests/huge_pages.c
View File

@@ -1316,7 +1316,7 @@ typedef struct drm_i915_gem_object *
static inline bool igt_can_allocate_thp(struct drm_i915_private *i915)
{
return i915->mm.gemfs && has_transparent_hugepage();
return !!drm_gem_get_huge_mnt(&i915->drm);
}
static struct drm_i915_gem_object *
@@ -1761,7 +1761,6 @@ static int igt_tmpfs_fallback(void *arg)
struct drm_i915_private *i915 = arg;
struct i915_address_space *vm;
struct i915_gem_context *ctx;
struct vfsmount *gemfs = i915->mm.gemfs;
struct drm_i915_gem_object *obj;
struct i915_vma *vma;
struct file *file;
@@ -1779,15 +1778,8 @@ static int igt_tmpfs_fallback(void *arg)
}
vm = i915_gem_context_get_eb_vm(ctx);
/*
* Make sure that we don't burst into a ball of flames upon falling back
* to tmpfs, which we rely on if on the off-chance we encounter a failure
* when setting up gemfs.
*/
i915->mm.gemfs = NULL;
obj = i915_gem_object_create_shmem(i915, PAGE_SIZE);
obj = i915_gem_object_create_region(i915->mm.regions[INTEL_REGION_SMEM],
PAGE_SIZE, 0, I915_BO_ALLOC_NOTHP);
if (IS_ERR(obj)) {
err = PTR_ERR(obj);
goto out_restore;
@@ -1819,7 +1811,6 @@ static int igt_tmpfs_fallback(void *arg)
out_put:
i915_gem_object_put(obj);
out_restore:
i915->mm.gemfs = gemfs;
i915_vm_put(vm);
out:

5
drivers/gpu/drm/i915/i915_drv.h
View File

@@ -141,11 +141,6 @@ struct i915_gem_mm {
*/
atomic_t free_count;
/**
* tmpfs instance used for shmem backed objects
*/
struct vfsmount *gemfs;
struct intel_memory_region *regions[INTEL_REGION_UNKNOWN];
struct notifier_block oom_notifier;