0069455bcb
Patch series "Memory allocation profiling", v6. Overview: Low overhead [1] per-callsite memory allocation profiling. Not just for debug kernels, overhead low enough to be deployed in production. Example output: root@moria-kvm:~# sort -rn /proc/allocinfo 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node ... Usage: kconfig options: - CONFIG_MEM_ALLOC_PROFILING - CONFIG_MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT - CONFIG_MEM_ALLOC_PROFILING_DEBUG adds warnings for allocations that weren't accounted because of a missing annotation sysctl: /proc/sys/vm/mem_profiling Runtime info: /proc/allocinfo Notes: [1]: Overhead To measure the overhead we are comparing the following configurations: (1) Baseline with CONFIG_MEMCG_KMEM=n (2) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) (3) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) (4) Enabled at runtime (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n && /proc/sys/vm/mem_profiling=1) (5) Baseline with CONFIG_MEMCG_KMEM=y && allocating with __GFP_ACCOUNT (6) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) && CONFIG_MEMCG_KMEM=y (7) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) && CONFIG_MEMCG_KMEM=y Performance overhead: To evaluate performance we implemented an in-kernel test executing multiple get_free_page/free_page and kmalloc/kfree calls with allocation sizes growing from 8 to 240 bytes with CPU frequency set to max and CPU affinity set to a specific CPU to minimize the noise. Below are results from running the test on Ubuntu 22.04.2 LTS with 6.8.0-rc1 kernel on 56 core Intel Xeon: kmalloc pgalloc (1 baseline) 6.764s 16.902s (2 default disabled) 6.793s (+0.43%) 17.007s (+0.62%) (3 default enabled) 7.197s (+6.40%) 23.666s (+40.02%) (4 runtime enabled) 7.405s (+9.48%) 23.901s (+41.41%) (5 memcg) 13.388s (+97.94%) 48.460s (+186.71%) (6 def disabled+memcg) 13.332s (+97.10%) 48.105s (+184.61%) (7 def enabled+memcg) 13.446s (+98.78%) 54.963s (+225.18%) Memory overhead: Kernel size: text data bss dec diff (1) 26515311 18890222 17018880 62424413 (2) 26524728 19423818 16740352 62688898 264485 (3) 26524724 19423818 16740352 62688894 264481 (4) 26524728 19423818 16740352 62688898 264485 (5) 26541782 18964374 16957440 62463596 39183 Memory consumption on a 56 core Intel CPU with 125GB of memory: Code tags: 192 kB PageExts: 262144 kB (256MB) SlabExts: 9876 kB (9.6MB) PcpuExts: 512 kB (0.5MB) Total overhead is 0.2% of total memory. Benchmarks: Hackbench tests run 100 times: hackbench -s 512 -l 200 -g 15 -f 25 -P baseline disabled profiling enabled profiling avg 0.3543 0.3559 (+0.0016) 0.3566 (+0.0023) stdev 0.0137 0.0188 0.0077 hackbench -l 10000 baseline disabled profiling enabled profiling avg 6.4218 6.4306 (+0.0088) 6.5077 (+0.0859) stdev 0.0933 0.0286 0.0489 stress-ng tests: stress-ng --class memory --seq 4 -t 60 stress-ng --class cpu --seq 4 -t 60 Results posted at: https://evilpiepirate.org/~kent/memalloc_prof_v4_stress-ng/ [2] https://lore.kernel.org/all/20240306182440.2003814-1-surenb@google.com/ This patch (of 37): The next patch drops vmalloc.h from a system header in order to fix a circular dependency; this adds it to all the files that were pulling it in implicitly. [kent.overstreet@linux.dev: fix arch/alpha/lib/memcpy.c] Link: https://lkml.kernel.org/r/20240327002152.3339937-1-kent.overstreet@linux.dev [surenb@google.com: fix arch/x86/mm/numa_32.c] Link: https://lkml.kernel.org/r/20240402180933.1663992-1-surenb@google.com [kent.overstreet@linux.dev: a few places were depending on sizes.h] Link: https://lkml.kernel.org/r/20240404034744.1664840-1-kent.overstreet@linux.dev [arnd@arndb.de: fix mm/kasan/hw_tags.c] Link: https://lkml.kernel.org/r/20240404124435.3121534-1-arnd@kernel.org [surenb@google.com: fix arc build] Link: https://lkml.kernel.org/r/20240405225115.431056-1-surenb@google.com Link: https://lkml.kernel.org/r/20240321163705.3067592-1-surenb@google.com Link: https://lkml.kernel.org/r/20240321163705.3067592-2-surenb@google.com Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> Signed-off-by: Suren Baghdasaryan <surenb@google.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com> Tested-by: Kees Cook <keescook@chromium.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Alex Gaynor <alex.gaynor@gmail.com> Cc: Alice Ryhl <aliceryhl@google.com> Cc: Andreas Hindborg <a.hindborg@samsung.com> Cc: Benno Lossin <benno.lossin@proton.me> Cc: "Björn Roy Baron" <bjorn3_gh@protonmail.com> Cc: Boqun Feng <boqun.feng@gmail.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Gary Guo <gary@garyguo.net> Cc: Miguel Ojeda <ojeda@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Wedson Almeida Filho <wedsonaf@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
306 lines
7.5 KiB
C
306 lines
7.5 KiB
C
// SPDX-License-Identifier: GPL-2.0 OR MIT
|
|
|
|
/*
|
|
* Xen para-virtual DRM device
|
|
*
|
|
* Copyright (C) 2016-2018 EPAM Systems Inc.
|
|
*
|
|
* Author: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
|
|
*/
|
|
|
|
#include <linux/dma-buf.h>
|
|
#include <linux/scatterlist.h>
|
|
#include <linux/shmem_fs.h>
|
|
#include <linux/vmalloc.h>
|
|
|
|
#include <drm/drm_gem.h>
|
|
#include <drm/drm_prime.h>
|
|
#include <drm/drm_probe_helper.h>
|
|
|
|
#include <xen/balloon.h>
|
|
#include <xen/xen.h>
|
|
|
|
#include "xen_drm_front.h"
|
|
#include "xen_drm_front_gem.h"
|
|
|
|
struct xen_gem_object {
|
|
struct drm_gem_object base;
|
|
|
|
size_t num_pages;
|
|
struct page **pages;
|
|
|
|
/* set for buffers allocated by the backend */
|
|
bool be_alloc;
|
|
|
|
/* this is for imported PRIME buffer */
|
|
struct sg_table *sgt_imported;
|
|
};
|
|
|
|
static inline struct xen_gem_object *
|
|
to_xen_gem_obj(struct drm_gem_object *gem_obj)
|
|
{
|
|
return container_of(gem_obj, struct xen_gem_object, base);
|
|
}
|
|
|
|
static int gem_alloc_pages_array(struct xen_gem_object *xen_obj,
|
|
size_t buf_size)
|
|
{
|
|
xen_obj->num_pages = DIV_ROUND_UP(buf_size, PAGE_SIZE);
|
|
xen_obj->pages = kvmalloc_array(xen_obj->num_pages,
|
|
sizeof(struct page *), GFP_KERNEL);
|
|
return !xen_obj->pages ? -ENOMEM : 0;
|
|
}
|
|
|
|
static void gem_free_pages_array(struct xen_gem_object *xen_obj)
|
|
{
|
|
kvfree(xen_obj->pages);
|
|
xen_obj->pages = NULL;
|
|
}
|
|
|
|
static int xen_drm_front_gem_object_mmap(struct drm_gem_object *gem_obj,
|
|
struct vm_area_struct *vma)
|
|
{
|
|
struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
|
|
int ret;
|
|
|
|
vma->vm_ops = gem_obj->funcs->vm_ops;
|
|
|
|
/*
|
|
* Clear the VM_PFNMAP flag that was set by drm_gem_mmap(), and set the
|
|
* vm_pgoff (used as a fake buffer offset by DRM) to 0 as we want to map
|
|
* the whole buffer.
|
|
*/
|
|
vm_flags_mod(vma, VM_MIXEDMAP | VM_DONTEXPAND, VM_PFNMAP);
|
|
vma->vm_pgoff = 0;
|
|
|
|
/*
|
|
* According to Xen on ARM ABI (xen/include/public/arch-arm.h):
|
|
* all memory which is shared with other entities in the system
|
|
* (including the hypervisor and other guests) must reside in memory
|
|
* which is mapped as Normal Inner Write-Back Outer Write-Back
|
|
* Inner-Shareable.
|
|
*/
|
|
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
|
|
|
|
/*
|
|
* vm_operations_struct.fault handler will be called if CPU access
|
|
* to VM is here. For GPUs this isn't the case, because CPU doesn't
|
|
* touch the memory. Insert pages now, so both CPU and GPU are happy.
|
|
*
|
|
* FIXME: as we insert all the pages now then no .fault handler must
|
|
* be called, so don't provide one
|
|
*/
|
|
ret = vm_map_pages(vma, xen_obj->pages, xen_obj->num_pages);
|
|
if (ret < 0)
|
|
DRM_ERROR("Failed to map pages into vma: %d\n", ret);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static const struct vm_operations_struct xen_drm_drv_vm_ops = {
|
|
.open = drm_gem_vm_open,
|
|
.close = drm_gem_vm_close,
|
|
};
|
|
|
|
static const struct drm_gem_object_funcs xen_drm_front_gem_object_funcs = {
|
|
.free = xen_drm_front_gem_object_free,
|
|
.get_sg_table = xen_drm_front_gem_get_sg_table,
|
|
.vmap = xen_drm_front_gem_prime_vmap,
|
|
.vunmap = xen_drm_front_gem_prime_vunmap,
|
|
.mmap = xen_drm_front_gem_object_mmap,
|
|
.vm_ops = &xen_drm_drv_vm_ops,
|
|
};
|
|
|
|
static struct xen_gem_object *gem_create_obj(struct drm_device *dev,
|
|
size_t size)
|
|
{
|
|
struct xen_gem_object *xen_obj;
|
|
int ret;
|
|
|
|
xen_obj = kzalloc(sizeof(*xen_obj), GFP_KERNEL);
|
|
if (!xen_obj)
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
xen_obj->base.funcs = &xen_drm_front_gem_object_funcs;
|
|
|
|
ret = drm_gem_object_init(dev, &xen_obj->base, size);
|
|
if (ret < 0) {
|
|
kfree(xen_obj);
|
|
return ERR_PTR(ret);
|
|
}
|
|
|
|
return xen_obj;
|
|
}
|
|
|
|
static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
|
|
{
|
|
struct xen_drm_front_drm_info *drm_info = dev->dev_private;
|
|
struct xen_gem_object *xen_obj;
|
|
int ret;
|
|
|
|
size = round_up(size, PAGE_SIZE);
|
|
xen_obj = gem_create_obj(dev, size);
|
|
if (IS_ERR(xen_obj))
|
|
return xen_obj;
|
|
|
|
if (drm_info->front_info->cfg.be_alloc) {
|
|
/*
|
|
* backend will allocate space for this buffer, so
|
|
* only allocate array of pointers to pages
|
|
*/
|
|
ret = gem_alloc_pages_array(xen_obj, size);
|
|
if (ret < 0)
|
|
goto fail;
|
|
|
|
/*
|
|
* allocate ballooned pages which will be used to map
|
|
* grant references provided by the backend
|
|
*/
|
|
ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
|
|
xen_obj->pages);
|
|
if (ret < 0) {
|
|
DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
|
|
xen_obj->num_pages, ret);
|
|
gem_free_pages_array(xen_obj);
|
|
goto fail;
|
|
}
|
|
|
|
xen_obj->be_alloc = true;
|
|
return xen_obj;
|
|
}
|
|
/*
|
|
* need to allocate backing pages now, so we can share those
|
|
* with the backend
|
|
*/
|
|
xen_obj->num_pages = DIV_ROUND_UP(size, PAGE_SIZE);
|
|
xen_obj->pages = drm_gem_get_pages(&xen_obj->base);
|
|
if (IS_ERR(xen_obj->pages)) {
|
|
ret = PTR_ERR(xen_obj->pages);
|
|
xen_obj->pages = NULL;
|
|
goto fail;
|
|
}
|
|
|
|
return xen_obj;
|
|
|
|
fail:
|
|
DRM_ERROR("Failed to allocate buffer with size %zu\n", size);
|
|
return ERR_PTR(ret);
|
|
}
|
|
|
|
struct drm_gem_object *xen_drm_front_gem_create(struct drm_device *dev,
|
|
size_t size)
|
|
{
|
|
struct xen_gem_object *xen_obj;
|
|
|
|
xen_obj = gem_create(dev, size);
|
|
if (IS_ERR(xen_obj))
|
|
return ERR_CAST(xen_obj);
|
|
|
|
return &xen_obj->base;
|
|
}
|
|
|
|
void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
|
|
{
|
|
struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
|
|
|
|
if (xen_obj->base.import_attach) {
|
|
drm_prime_gem_destroy(&xen_obj->base, xen_obj->sgt_imported);
|
|
gem_free_pages_array(xen_obj);
|
|
} else {
|
|
if (xen_obj->pages) {
|
|
if (xen_obj->be_alloc) {
|
|
xen_free_unpopulated_pages(xen_obj->num_pages,
|
|
xen_obj->pages);
|
|
gem_free_pages_array(xen_obj);
|
|
} else {
|
|
drm_gem_put_pages(&xen_obj->base,
|
|
xen_obj->pages, true, false);
|
|
}
|
|
}
|
|
}
|
|
drm_gem_object_release(gem_obj);
|
|
kfree(xen_obj);
|
|
}
|
|
|
|
struct page **xen_drm_front_gem_get_pages(struct drm_gem_object *gem_obj)
|
|
{
|
|
struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
|
|
|
|
return xen_obj->pages;
|
|
}
|
|
|
|
struct sg_table *xen_drm_front_gem_get_sg_table(struct drm_gem_object *gem_obj)
|
|
{
|
|
struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
|
|
|
|
if (!xen_obj->pages)
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
return drm_prime_pages_to_sg(gem_obj->dev,
|
|
xen_obj->pages, xen_obj->num_pages);
|
|
}
|
|
|
|
struct drm_gem_object *
|
|
xen_drm_front_gem_import_sg_table(struct drm_device *dev,
|
|
struct dma_buf_attachment *attach,
|
|
struct sg_table *sgt)
|
|
{
|
|
struct xen_drm_front_drm_info *drm_info = dev->dev_private;
|
|
struct xen_gem_object *xen_obj;
|
|
size_t size;
|
|
int ret;
|
|
|
|
size = attach->dmabuf->size;
|
|
xen_obj = gem_create_obj(dev, size);
|
|
if (IS_ERR(xen_obj))
|
|
return ERR_CAST(xen_obj);
|
|
|
|
ret = gem_alloc_pages_array(xen_obj, size);
|
|
if (ret < 0)
|
|
return ERR_PTR(ret);
|
|
|
|
xen_obj->sgt_imported = sgt;
|
|
|
|
ret = drm_prime_sg_to_page_array(sgt, xen_obj->pages,
|
|
xen_obj->num_pages);
|
|
if (ret < 0)
|
|
return ERR_PTR(ret);
|
|
|
|
ret = xen_drm_front_dbuf_create(drm_info->front_info,
|
|
xen_drm_front_dbuf_to_cookie(&xen_obj->base),
|
|
0, 0, 0, size, sgt->sgl->offset,
|
|
xen_obj->pages);
|
|
if (ret < 0)
|
|
return ERR_PTR(ret);
|
|
|
|
DRM_DEBUG("Imported buffer of size %zu with nents %u\n",
|
|
size, sgt->orig_nents);
|
|
|
|
return &xen_obj->base;
|
|
}
|
|
|
|
int xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj,
|
|
struct iosys_map *map)
|
|
{
|
|
struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
|
|
void *vaddr;
|
|
|
|
if (!xen_obj->pages)
|
|
return -ENOMEM;
|
|
|
|
/* Please see comment in gem_mmap_obj on mapping and attributes. */
|
|
vaddr = vmap(xen_obj->pages, xen_obj->num_pages,
|
|
VM_MAP, PAGE_KERNEL);
|
|
if (!vaddr)
|
|
return -ENOMEM;
|
|
iosys_map_set_vaddr(map, vaddr);
|
|
|
|
return 0;
|
|
}
|
|
|
|
void xen_drm_front_gem_prime_vunmap(struct drm_gem_object *gem_obj,
|
|
struct iosys_map *map)
|
|
{
|
|
vunmap(map->vaddr);
|
|
}
|