| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <c553c26f-1f9c-e5ab-b71b-b2b5eeb59ad2@suse.cz>
Date: Mon, 14 Jun 2021 13:07:14 +0200
From: Vlastimil Babka <vbabka@...e.cz>
To: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Christoph Lameter <cl@...ux.com>,
David Rientjes <rientjes@...gle.com>,
Pekka Enberg <penberg@...nel.org>,
Joonsoo Kim <iamjoonsoo.kim@....com>
Cc: Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
Thomas Gleixner <tglx@...utronix.de>,
Mel Gorman <mgorman@...hsingularity.net>,
Jesper Dangaard Brouer <brouer@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Jann Horn <jannh@...gle.com>
Subject: Re: [RFC v2 33/34] mm, slub: use migrate_disable() on PREEMPT_RT
On 6/9/21 1:39 PM, Vlastimil Babka wrote:
> We currently use preempt_disable() (directly or via get_cpu_ptr()) to stabilize
> the pointer to kmem_cache_cpu. On PREEMPT_RT this would be incompatible with
> the list_lock spinlock. We can use migrate_disable() instead, but that
> increases overhead on !PREEMPT_RT as it's an unconditional function call even
> though it's ultimately a migrate_disable() there.
>
> In order to get the best available mechanism on both PREEMPT_RT and
> !PREEMPT_RT, introduce private slub_get_cpu_ptr() and slub_put_cpu_ptr()
> wrappers and use them.
>
> Signed-off-by: Vlastimil Babka <vbabka@...e.cz>
> ---
> mm/slub.c | 41 +++++++++++++++++++++++++++++++----------
> 1 file changed, 31 insertions(+), 10 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 12e966f07f19..caa206213e72 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -115,6 +115,26 @@
> * the fast path and disables lockless freelists.
> */
>
> +/*
> + * We could simply use migrate_disable()/enable() but as long as it's a
> + * function call even on !PREEMPT_RT, use inline preempt_disable() there.
> + */
> +#ifdef CONFIG_PREEMPT_RT
> +#define slub_get_cpu_ptr(var) get_cpu_ptr(var)
> +#define slub_put_cpu_ptr(var) put_cpu_ptr(var)
After Mel's report and bisect pointing to this patch, I realized I got the
#ifdef wrong and it should be #ifnded
Fixed patch follows:
----8<----
>From e79e166414b8c22e3bcf73a9383bafe802fa64d2 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@...e.cz>
Date: Fri, 21 May 2021 14:03:23 +0200
Subject: [PATCH] mm, slub: use migrate_disable() on PREEMPT_RT
We currently use preempt_disable() (directly or via get_cpu_ptr()) to stabilize
the pointer to kmem_cache_cpu. On PREEMPT_RT this would be incompatible with
the list_lock spinlock. We can use migrate_disable() instead, but that
increases overhead on !PREEMPT_RT as it's an unconditional function call even
though it's ultimately a migrate_disable() there.
In order to get the best available mechanism on both PREEMPT_RT and
!PREEMPT_RT, introduce private slub_get_cpu_ptr() and slub_put_cpu_ptr()
wrappers and use them.
Signed-off-by: Vlastimil Babka <vbabka@...e.cz>
---
mm/slub.c | 41 +++++++++++++++++++++++++++++++----------
1 file changed, 31 insertions(+), 10 deletions(-)
diff --git a/mm/slub.c b/mm/slub.c
index 12e966f07f19..f0359b0c8154 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -115,6 +115,26 @@
* the fast path and disables lockless freelists.
*/
+/*
+ * We could simply use migrate_disable()/enable() but as long as it's a
+ * function call even on !PREEMPT_RT, use inline preempt_disable() there.
+ */
+#ifndef CONFIG_PREEMPT_RT
+#define slub_get_cpu_ptr(var) get_cpu_ptr(var)
+#define slub_put_cpu_ptr(var) put_cpu_ptr(var)
+#else
+#define slub_get_cpu_ptr(var) \
+({ \
+ migrate_disable(); \
+ this_cpu_ptr(var); \
+})
+#define slub_put_cpu_ptr(var) \
+do { \
+ (void)(var); \
+ migrate_enable(); \
+} while (0)
+#endif
+
#ifdef CONFIG_SLUB_DEBUG
#ifdef CONFIG_SLUB_DEBUG_ON
DEFINE_STATIC_KEY_TRUE(slub_debug_enabled);
@@ -2419,7 +2439,7 @@ static void put_cpu_partial(struct kmem_cache *s, struct page *page, int drain)
int pages;
int pobjects;
- preempt_disable();
+ slub_get_cpu_ptr(s->cpu_slab);
do {
pages = 0;
pobjects = 0;
@@ -2450,7 +2470,7 @@ static void put_cpu_partial(struct kmem_cache *s, struct page *page, int drain)
} while (this_cpu_cmpxchg(s->cpu_slab->partial, oldpage, page)
!= oldpage);
- preempt_enable();
+ slub_put_cpu_ptr(s->cpu_slab);
#endif /* CONFIG_SLUB_CPU_PARTIAL */
}
@@ -2759,7 +2779,7 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
if (unlikely(!pfmemalloc_match(page, gfpflags)))
goto deactivate_slab;
- /* must check again c->page in case IRQ handler changed it */
+ /* must check again c->page in case we got preempted and it changed */
local_irq_save(flags);
if (unlikely(page != c->page)) {
local_irq_restore(flags);
@@ -2818,7 +2838,8 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
}
if (unlikely(!slub_percpu_partial(c))) {
local_irq_restore(flags);
- goto new_objects; /* stolen by an IRQ handler */
+ /* we were preempted and partial list got empty */
+ goto new_objects;
}
page = c->page = slub_percpu_partial(c);
@@ -2834,9 +2855,9 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
if (freelist)
goto check_new_page;
- put_cpu_ptr(s->cpu_slab);
+ slub_put_cpu_ptr(s->cpu_slab);
page = new_slab(s, gfpflags, node);
- c = get_cpu_ptr(s->cpu_slab);
+ c = slub_get_cpu_ptr(s->cpu_slab);
if (unlikely(!page)) {
slab_out_of_memory(s, gfpflags, node);
@@ -2919,12 +2940,12 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
* cpu before disabling preemption. Need to reload cpu area
* pointer.
*/
- c = get_cpu_ptr(s->cpu_slab);
+ c = slub_get_cpu_ptr(s->cpu_slab);
#endif
p = ___slab_alloc(s, gfpflags, node, addr, c);
#ifdef CONFIG_PREEMPTION
- put_cpu_ptr(s->cpu_slab);
+ slub_put_cpu_ptr(s->cpu_slab);
#endif
return p;
}
@@ -3445,7 +3466,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
* IRQs, which protects against PREEMPT and interrupts
* handlers invoking normal fastpath.
*/
- c = get_cpu_ptr(s->cpu_slab);
+ c = slub_get_cpu_ptr(s->cpu_slab);
local_irq_disable();
for (i = 0; i < size; i++) {
@@ -3491,7 +3512,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
}
c->tid = next_tid(c->tid);
local_irq_enable();
- put_cpu_ptr(s->cpu_slab);
+ slub_put_cpu_ptr(s->cpu_slab);
/*
* memcg and kmem_cache debug support and memory initialization.
--
2.32.0
Powered by blists - more mailing lists