From: Dmitry Vyukov <dvyukov@google.com>
To: Marco Elver <elver@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Alexander Potapenko <glider@google.com>,
Jann Horn <jannh@google.com>,
Aleksandr Nogikh <nogikh@google.com>,
Taras Madan <tarasmadan@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
kasan-dev@googlegroups.com
Subject: Re: [PATCH v2 2/5] kfence: count unexpectedly skipped allocations
Date: Tue, 21 Sep 2021 12:48:33 +0200 [thread overview]
Message-ID: <CACT4Y+bJ=i=i5eecERcAWMcZuYtU1587WEpRa=SH+bGvmoCayA@mail.gmail.com> (raw)
In-Reply-To: <20210921101014.1938382-2-elver@google.com>
On Tue, 21 Sept 2021 at 12:10, Marco Elver <elver@google.com> wrote:
>
> Maintain a counter to count allocations that are skipped due to being
> incompatible (oversized, incompatible gfp flags) or no capacity.
>
> This is to compute the fraction of allocations that could not be
> serviced by KFENCE, which we expect to be rare.
>
> Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
> ---
> v2:
> * Do not count deadlock-avoidance skips.
> ---
> mm/kfence/core.c | 16 +++++++++++++---
> 1 file changed, 13 insertions(+), 3 deletions(-)
>
> diff --git a/mm/kfence/core.c b/mm/kfence/core.c
> index 7a97db8bc8e7..249d75b7e5ee 100644
> --- a/mm/kfence/core.c
> +++ b/mm/kfence/core.c
> @@ -112,6 +112,8 @@ enum kfence_counter_id {
> KFENCE_COUNTER_FREES,
> KFENCE_COUNTER_ZOMBIES,
> KFENCE_COUNTER_BUGS,
> + KFENCE_COUNTER_SKIP_INCOMPAT,
> + KFENCE_COUNTER_SKIP_CAPACITY,
> KFENCE_COUNTER_COUNT,
> };
> static atomic_long_t counters[KFENCE_COUNTER_COUNT];
> @@ -121,6 +123,8 @@ static const char *const counter_names[] = {
> [KFENCE_COUNTER_FREES] = "total frees",
> [KFENCE_COUNTER_ZOMBIES] = "zombie allocations",
> [KFENCE_COUNTER_BUGS] = "total bugs",
> + [KFENCE_COUNTER_SKIP_INCOMPAT] = "skipped allocations (incompatible)",
> + [KFENCE_COUNTER_SKIP_CAPACITY] = "skipped allocations (capacity)",
> };
> static_assert(ARRAY_SIZE(counter_names) == KFENCE_COUNTER_COUNT);
>
> @@ -271,8 +275,10 @@ static void *kfence_guarded_alloc(struct kmem_cache *cache, size_t size, gfp_t g
> list_del_init(&meta->list);
> }
> raw_spin_unlock_irqrestore(&kfence_freelist_lock, flags);
> - if (!meta)
> + if (!meta) {
> + atomic_long_inc(&counters[KFENCE_COUNTER_SKIP_CAPACITY]);
> return NULL;
> + }
>
> if (unlikely(!raw_spin_trylock_irqsave(&meta->lock, flags))) {
> /*
> @@ -740,8 +746,10 @@ void *__kfence_alloc(struct kmem_cache *s, size_t size, gfp_t flags)
> * Perform size check before switching kfence_allocation_gate, so that
> * we don't disable KFENCE without making an allocation.
> */
> - if (size > PAGE_SIZE)
> + if (size > PAGE_SIZE) {
> + atomic_long_inc(&counters[KFENCE_COUNTER_SKIP_INCOMPAT]);
> return NULL;
> + }
>
> /*
> * Skip allocations from non-default zones, including DMA. We cannot
> @@ -749,8 +757,10 @@ void *__kfence_alloc(struct kmem_cache *s, size_t size, gfp_t flags)
> * properties (e.g. reside in DMAable memory).
> */
> if ((flags & GFP_ZONEMASK) ||
> - (s->flags & (SLAB_CACHE_DMA | SLAB_CACHE_DMA32)))
> + (s->flags & (SLAB_CACHE_DMA | SLAB_CACHE_DMA32))) {
> + atomic_long_inc(&counters[KFENCE_COUNTER_SKIP_INCOMPAT]);
> return NULL;
> + }
>
> /*
> * allocation_gate only needs to become non-zero, so it doesn't make
> --
> 2.33.0.464.g1972c5931b-goog
>
next prev parent reply other threads:[~2021-09-21 10:48 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-21 10:10 [PATCH v2 1/5] stacktrace: move filter_irq_stacks() to kernel/stacktrace.c Marco Elver
2021-09-21 10:10 ` [PATCH v2 2/5] kfence: count unexpectedly skipped allocations Marco Elver
2021-09-21 10:48 ` Dmitry Vyukov [this message]
2021-09-21 10:10 ` [PATCH v2 3/5] kfence: move saving stack trace of allocations into __kfence_alloc() Marco Elver
2021-09-21 10:50 ` Dmitry Vyukov
2021-09-21 10:10 ` [PATCH v2 4/5] kfence: limit currently covered allocations when pool nearly full Marco Elver
2021-09-21 11:05 ` Dmitry Vyukov
2021-09-21 11:35 ` Marco Elver
2021-09-21 10:10 ` [PATCH v2 5/5] kfence: add note to documentation about skipping covered allocations Marco Elver
2021-09-21 10:51 ` Dmitry Vyukov
2021-09-21 10:47 ` [PATCH v2 1/5] stacktrace: move filter_irq_stacks() to kernel/stacktrace.c Dmitry Vyukov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CACT4Y+bJ=i=i5eecERcAWMcZuYtU1587WEpRa=SH+bGvmoCayA@mail.gmail.com' \
--to=dvyukov@google.com \
--cc=akpm@linux-foundation.org \
--cc=elver@google.com \
--cc=glider@google.com \
--cc=jannh@google.com \
--cc=kasan-dev@googlegroups.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nogikh@google.com \
--cc=tarasmadan@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox