On Mon, Nov 20, 2023 at 07:34:29PM +0100, Vlastimil Babka wrote:
> This will eliminate a call between compilation units through
> __kmem_cache_alloc_node() and allow better inlining of the allocation
> fast path.
>
> Reviewed-by: Kees Cook <keescook@xxxxxxxxxxxx>
> Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx>
> ---
> mm/slab.h | 3 --
> mm/slab_common.c | 119 ----------------------------------------------------
> mm/slub.c | 126 +++++++++++++++++++++++++++++++++++++++++++++++++++----
> 3 files changed, 118 insertions(+), 130 deletions(-)
>
> diff --git a/mm/slab.h b/mm/slab.h
> index 7d7cc7af614e..54deeb0428c6 100644
> --- a/mm/slab.h
> +++ b/mm/slab.h
> @@ -416,9 +416,6 @@ kmalloc_slab(size_t size, gfp_t flags, unsigned long caller)
> return kmalloc_caches[kmalloc_type(flags, caller)][index];
> }
>
> -void *__kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags,
> - int node, size_t orig_size,
> - unsigned long caller);
> gfp_t kmalloc_fix_flags(gfp_t flags);
>
> /* Functions provided by the slab allocators */
> diff --git a/mm/slab_common.c b/mm/slab_common.c
> index 31ade17a7ad9..238293b1dbe1 100644
> --- a/mm/slab_common.c
> +++ b/mm/slab_common.c
> @@ -936,50 +936,6 @@ void __init create_kmalloc_caches(slab_flags_t flags)
> slab_state = UP;
> }
>
> -static void *__kmalloc_large_node(size_t size, gfp_t flags, int node);
> -static __always_inline
> -void *__do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller)
> -{
> - struct kmem_cache *s;
> - void *ret;
> -
> - if (unlikely(size > KMALLOC_MAX_CACHE_SIZE)) {
> - ret = __kmalloc_large_node(size, flags, node);
> - trace_kmalloc(caller, ret, size,
> - PAGE_SIZE << get_order(size), flags, node);
> - return ret;
> - }
> -
> - if (unlikely(!size))
> - return ZERO_SIZE_PTR;
> -
> - s = kmalloc_slab(size, flags, caller);
> -
> - ret = __kmem_cache_alloc_node(s, flags, node, size, caller);
> - ret = kasan_kmalloc(s, ret, size, flags);
> - trace_kmalloc(caller, ret, size, s->size, flags, node);
> - return ret;
> -}
> -
> -void *__kmalloc_node(size_t size, gfp_t flags, int node)
> -{
> - return __do_kmalloc_node(size, flags, node, _RET_IP_);
> -}
> -EXPORT_SYMBOL(__kmalloc_node);
> -
> -void *__kmalloc(size_t size, gfp_t flags)
> -{
> - return __do_kmalloc_node(size, flags, NUMA_NO_NODE, _RET_IP_);
> -}
> -EXPORT_SYMBOL(__kmalloc);
> -
> -void *__kmalloc_node_track_caller(size_t size, gfp_t flags,
> - int node, unsigned long caller)
> -{
> - return __do_kmalloc_node(size, flags, node, caller);
> -}
> -EXPORT_SYMBOL(__kmalloc_node_track_caller);
> -
> /**
> * __ksize -- Report full size of underlying allocation
> * @object: pointer to the object
> @@ -1016,30 +972,6 @@ size_t __ksize(const void *object)
> return slab_ksize(folio_slab(folio)->slab_cache);
> }
>
> -void *kmalloc_trace(struct kmem_cache *s, gfp_t gfpflags, size_t size)
> -{
> - void *ret = __kmem_cache_alloc_node(s, gfpflags, NUMA_NO_NODE,
> - size, _RET_IP_);
> -
> - trace_kmalloc(_RET_IP_, ret, size, s->size, gfpflags, NUMA_NO_NODE);
> -
> - ret = kasan_kmalloc(s, ret, size, gfpflags);
> - return ret;
> -}
> -EXPORT_SYMBOL(kmalloc_trace);
> -
> -void *kmalloc_node_trace(struct kmem_cache *s, gfp_t gfpflags,
> - int node, size_t size)
> -{
> - void *ret = __kmem_cache_alloc_node(s, gfpflags, node, size, _RET_IP_);
> -
> - trace_kmalloc(_RET_IP_, ret, size, s->size, gfpflags, node);
> -
> - ret = kasan_kmalloc(s, ret, size, gfpflags);
> - return ret;
> -}
> -EXPORT_SYMBOL(kmalloc_node_trace);
> -
> gfp_t kmalloc_fix_flags(gfp_t flags)
> {
> gfp_t invalid_mask = flags & GFP_SLAB_BUG_MASK;
> @@ -1052,57 +984,6 @@ gfp_t kmalloc_fix_flags(gfp_t flags)
> return flags;
> }
>
> -/*
> - * To avoid unnecessary overhead, we pass through large allocation requests
> - * directly to the page allocator. We use __GFP_COMP, because we will need to
> - * know the allocation order to free the pages properly in kfree.
> - */
> -
> -static void *__kmalloc_large_node(size_t size, gfp_t flags, int node)
> -{
> - struct page *page;
> - void *ptr = NULL;
> - unsigned int order = get_order(size);
> -
> - if (unlikely(flags & GFP_SLAB_BUG_MASK))
> - flags = kmalloc_fix_flags(flags);
> -
> - flags |= __GFP_COMP;
> - page = alloc_pages_node(node, flags, order);
> - if (page) {
> - ptr = page_address(page);
> - mod_lruvec_page_state(page, NR_SLAB_UNRECLAIMABLE_B,
> - PAGE_SIZE << order);
> - }
> -
> - ptr = kasan_kmalloc_large(ptr, size, flags);
> - /* As ptr might get tagged, call kmemleak hook after KASAN. */
> - kmemleak_alloc(ptr, size, 1, flags);
> - kmsan_kmalloc_large(ptr, size, flags);
> -
> - return ptr;
> -}
> -
> -void *kmalloc_large(size_t size, gfp_t flags)
> -{
> - void *ret = __kmalloc_large_node(size, flags, NUMA_NO_NODE);
> -
> - trace_kmalloc(_RET_IP_, ret, size, PAGE_SIZE << get_order(size),
> - flags, NUMA_NO_NODE);
> - return ret;
> -}
> -EXPORT_SYMBOL(kmalloc_large);
> -
> -void *kmalloc_large_node(size_t size, gfp_t flags, int node)
> -{
> - void *ret = __kmalloc_large_node(size, flags, node);
> -
> - trace_kmalloc(_RET_IP_, ret, size, PAGE_SIZE << get_order(size),
> - flags, node);
> - return ret;
> -}
> -EXPORT_SYMBOL(kmalloc_large_node);
> -
> #ifdef CONFIG_SLAB_FREELIST_RANDOM
> /* Randomize a generic freelist */
> static void freelist_randomize(unsigned int *list,
> diff --git a/mm/slub.c b/mm/slub.c
> index 2baa9e94d9df..d6bc15929d22 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -3851,14 +3851,6 @@ void *kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
> }
> EXPORT_SYMBOL(kmem_cache_alloc_lru);
>
> -void *__kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags,
> - int node, size_t orig_size,
> - unsigned long caller)
> -{
> - return slab_alloc_node(s, NULL, gfpflags, node,
> - caller, orig_size);
> -}
> -
> /**
> * kmem_cache_alloc_node - Allocate an object on the specified node
> * @s: The cache to allocate from.
> @@ -3882,6 +3874,124 @@ void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t gfpflags, int node)
> }
> EXPORT_SYMBOL(kmem_cache_alloc_node);
>
> +/*
> + * To avoid unnecessary overhead, we pass through large allocation requests
> + * directly to the page allocator. We use __GFP_COMP, because we will need to
> + * know the allocation order to free the pages properly in kfree.
> + */
> +static void *__kmalloc_large_node(size_t size, gfp_t flags, int node)
> +{
> + struct page *page;
> + void *ptr = NULL;
> + unsigned int order = get_order(size);
> +
> + if (unlikely(flags & GFP_SLAB_BUG_MASK))
> + flags = kmalloc_fix_flags(flags);
> +
> + flags |= __GFP_COMP;
> + page = alloc_pages_node(node, flags, order);
> + if (page) {
> + ptr = page_address(page);
> + mod_lruvec_page_state(page, NR_SLAB_UNRECLAIMABLE_B,
> + PAGE_SIZE << order);
> + }
> +
> + ptr = kasan_kmalloc_large(ptr, size, flags);
> + /* As ptr might get tagged, call kmemleak hook after KASAN. */
> + kmemleak_alloc(ptr, size, 1, flags);
> + kmsan_kmalloc_large(ptr, size, flags);
> +
> + return ptr;
> +}
> +
> +void *kmalloc_large(size_t size, gfp_t flags)
> +{
> + void *ret = __kmalloc_large_node(size, flags, NUMA_NO_NODE);
> +
> + trace_kmalloc(_RET_IP_, ret, size, PAGE_SIZE << get_order(size),
> + flags, NUMA_NO_NODE);
> + return ret;
> +}
> +EXPORT_SYMBOL(kmalloc_large);
> +
> +void *kmalloc_large_node(size_t size, gfp_t flags, int node)
> +{
> + void *ret = __kmalloc_large_node(size, flags, node);
> +
> + trace_kmalloc(_RET_IP_, ret, size, PAGE_SIZE << get_order(size),
> + flags, node);
> + return ret;
> +}
> +EXPORT_SYMBOL(kmalloc_large_node);
> +
> +static __always_inline
> +void *__do_kmalloc_node(size_t size, gfp_t flags, int node,
> + unsigned long caller)
> +{
> + struct kmem_cache *s;
> + void *ret;
> +
> + if (unlikely(size > KMALLOC_MAX_CACHE_SIZE)) {
> + ret = __kmalloc_large_node(size, flags, node);
> + trace_kmalloc(caller, ret, size,
> + PAGE_SIZE << get_order(size), flags, node);
> + return ret;
> + }
> +
> + if (unlikely(!size))
> + return ZERO_SIZE_PTR;
> +
> + s = kmalloc_slab(size, flags, caller);
> +
> + ret = slab_alloc_node(s, NULL, flags, node, caller, size);
> + ret = kasan_kmalloc(s, ret, size, flags);
> + trace_kmalloc(caller, ret, size, s->size, flags, node);
> + return ret;
> +}
> +
> +void *__kmalloc_node(size_t size, gfp_t flags, int node)
> +{
> + return __do_kmalloc_node(size, flags, node, _RET_IP_);
> +}
> +EXPORT_SYMBOL(__kmalloc_node);
> +
> +void *__kmalloc(size_t size, gfp_t flags)
> +{
> + return __do_kmalloc_node(size, flags, NUMA_NO_NODE, _RET_IP_);
> +}
> +EXPORT_SYMBOL(__kmalloc);
> +
> +void *__kmalloc_node_track_caller(size_t size, gfp_t flags,
> + int node, unsigned long caller)
> +{
> + return __do_kmalloc_node(size, flags, node, caller);
> +}
> +EXPORT_SYMBOL(__kmalloc_node_track_caller);
> +
> +void *kmalloc_trace(struct kmem_cache *s, gfp_t gfpflags, size_t size)
> +{
> + void *ret = slab_alloc_node(s, NULL, gfpflags, NUMA_NO_NODE,
> + _RET_IP_, size);
> +
> + trace_kmalloc(_RET_IP_, ret, size, s->size, gfpflags, NUMA_NO_NODE);
> +
> + ret = kasan_kmalloc(s, ret, size, gfpflags);
> + return ret;
> +}
> +EXPORT_SYMBOL(kmalloc_trace);
> +
> +void *kmalloc_node_trace(struct kmem_cache *s, gfp_t gfpflags,
> + int node, size_t size)
> +{
> + void *ret = slab_alloc_node(s, NULL, gfpflags, node, _RET_IP_, size);
> +
> + trace_kmalloc(_RET_IP_, ret, size, s->size, gfpflags, node);
> +
> + ret = kasan_kmalloc(s, ret, size, gfpflags);
> + return ret;
> +}
> +EXPORT_SYMBOL(kmalloc_node_trace);
> +
> static noinline void free_to_partial_list(
> struct kmem_cache *s, struct slab *slab,
> void *head, void *tail, int bulk_cnt,
>
> --
Looks good to me,
Reviewed-by: Hyeonggon Yoo <42.hyeyoo@xxxxxxxxx>
> 2.42.1
>
>
