From: Harry Yoo <harry.yoo@oracle.com>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: Suren Baghdasaryan <surenb@google.com>,
"Liam R. Howlett" <Liam.Howlett@oracle.com>,
Christoph Lameter <cl@linux.com>,
David Rientjes <rientjes@google.com>,
Roman Gushchin <roman.gushchin@linux.dev>,
Hyeonggon Yoo <42.hyeyoo@gmail.com>,
Uladzislau Rezki <urezki@gmail.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
rcu@vger.kernel.org, maple-tree@lists.infradead.org
Subject: Re: [PATCH RFC v2 07/10] slab: determine barn status racily outside of lock
Date: Tue, 25 Feb 2025 17:54:22 +0900 [thread overview]
Message-ID: <Z72FPj_FzhCZpRsk@harry> (raw)
In-Reply-To: <20250214-slub-percpu-caches-v2-7-88592ee0966a@suse.cz>
On Fri, Feb 14, 2025 at 05:27:43PM +0100, Vlastimil Babka wrote:
> The possibility of many barn operations is determined by the current
> number of full or empty sheaves. Taking the barn->lock just to find out
> that e.g. there are no empty sheaves results in unnecessary overhead and
> lock contention. Thus perform these checks outside of the lock with a
> data_race() annotated variable read and fail quickly without taking the
> lock.
>
> Checks for sheaf availability that racily succeed have to be obviously
> repeated under the lock for correctness, but we can skip repeating
> checks if there are too many sheaves on the given list as the limits
> don't need to be strict.
>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Looks good to me,
Reviewed-by: Harry Yoo <harry.yoo@oracle.com>
in kmem_cache_return_sheaf:
> if (!pcs->spare) {
> pcs->spare = sheaf;
> sheaf = NULL;
> } else if (pcs->barn->nr_full >= MAX_FULL_SHEAVES) {
> /* racy check */
> barn = pcs->barn;
> keep = true;
> }
By the way this code also needs data_race()?
--
Cheers,
Harry
> ---
> mm/slub.c | 57 ++++++++++++++++++++++++++++++++++-----------------------
> 1 file changed, 34 insertions(+), 23 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index c1df7cf22267f28f743404531bef921e25fac086..72e6437f1d74bfacbb1cd7642af42929c48cc66a 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2685,9 +2685,12 @@ static struct slab_sheaf *barn_get_empty_sheaf(struct node_barn *barn)
> struct slab_sheaf *empty = NULL;
> unsigned long flags;
>
> + if (!data_race(barn->nr_empty))
> + return NULL;
> +
> spin_lock_irqsave(&barn->lock, flags);
>
> - if (barn->nr_empty) {
> + if (likely(barn->nr_empty)) {
> empty = list_first_entry(&barn->sheaves_empty,
> struct slab_sheaf, barn_list);
> list_del(&empty->barn_list);
> @@ -2703,38 +2706,36 @@ static int barn_put_empty_sheaf(struct node_barn *barn,
> struct slab_sheaf *sheaf, bool ignore_limit)
> {
> unsigned long flags;
> - int ret = 0;
> +
> + /* we don't repeat the check under barn->lock as it's not critical */
> + if (!ignore_limit && data_race(barn->nr_empty) >= MAX_EMPTY_SHEAVES)
> + return -E2BIG;
>
> spin_lock_irqsave(&barn->lock, flags);
>
> - if (!ignore_limit && barn->nr_empty >= MAX_EMPTY_SHEAVES) {
> - ret = -E2BIG;
> - } else {
> - list_add(&sheaf->barn_list, &barn->sheaves_empty);
> - barn->nr_empty++;
> - }
> + list_add(&sheaf->barn_list, &barn->sheaves_empty);
> + barn->nr_empty++;
>
> spin_unlock_irqrestore(&barn->lock, flags);
> - return ret;
> + return 0;
> }
>
> static int barn_put_full_sheaf(struct node_barn *barn, struct slab_sheaf *sheaf,
> bool ignore_limit)
> {
> unsigned long flags;
> - int ret = 0;
> +
> + /* we don't repeat the check under barn->lock as it's not critical */
> + if (!ignore_limit && data_race(barn->nr_full) >= MAX_FULL_SHEAVES)
> + return -E2BIG;
>
> spin_lock_irqsave(&barn->lock, flags);
>
> - if (!ignore_limit && barn->nr_full >= MAX_FULL_SHEAVES) {
> - ret = -E2BIG;
> - } else {
> - list_add(&sheaf->barn_list, &barn->sheaves_full);
> - barn->nr_full++;
> - }
> + list_add(&sheaf->barn_list, &barn->sheaves_full);
> + barn->nr_full++;
>
> spin_unlock_irqrestore(&barn->lock, flags);
> - return ret;
> + return 0;
> }
>
> static struct slab_sheaf *barn_get_full_or_empty_sheaf(struct node_barn *barn)
> @@ -2742,6 +2743,9 @@ static struct slab_sheaf *barn_get_full_or_empty_sheaf(struct node_barn *barn)
> struct slab_sheaf *sheaf = NULL;
> unsigned long flags;
>
> + if (!data_race(barn->nr_full) && !data_race(barn->nr_empty))
> + return NULL;
> +
> spin_lock_irqsave(&barn->lock, flags);
>
> if (barn->nr_full) {
> @@ -2772,9 +2776,12 @@ barn_replace_empty_sheaf(struct node_barn *barn, struct slab_sheaf *empty)
> struct slab_sheaf *full = NULL;
> unsigned long flags;
>
> + if (!data_race(barn->nr_full))
> + return NULL;
> +
> spin_lock_irqsave(&barn->lock, flags);
>
> - if (barn->nr_full) {
> + if (likely(barn->nr_full)) {
> full = list_first_entry(&barn->sheaves_full, struct slab_sheaf,
> barn_list);
> list_del(&full->barn_list);
> @@ -2797,19 +2804,23 @@ barn_replace_full_sheaf(struct node_barn *barn, struct slab_sheaf *full)
> struct slab_sheaf *empty;
> unsigned long flags;
>
> + /* we don't repeat this check under barn->lock as it's not critical */
> + if (data_race(barn->nr_full) >= MAX_FULL_SHEAVES)
> + return ERR_PTR(-E2BIG);
> + if (!data_race(barn->nr_empty))
> + return ERR_PTR(-ENOMEM);
> +
> spin_lock_irqsave(&barn->lock, flags);
>
> - if (barn->nr_full >= MAX_FULL_SHEAVES) {
> - empty = ERR_PTR(-E2BIG);
> - } else if (!barn->nr_empty) {
> - empty = ERR_PTR(-ENOMEM);
> - } else {
> + if (likely(barn->nr_empty)) {
> empty = list_first_entry(&barn->sheaves_empty, struct slab_sheaf,
> barn_list);
> list_del(&empty->barn_list);
> list_add(&full->barn_list, &barn->sheaves_full);
> barn->nr_empty--;
> barn->nr_full++;
> + } else {
> + empty = ERR_PTR(-ENOMEM);
> }
>
> spin_unlock_irqrestore(&barn->lock, flags);
>
> --
> 2.48.1
>
next prev parent reply other threads:[~2025-02-25 8:54 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-14 16:27 [PATCH RFC v2 00/10] SLUB percpu sheaves Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 01/10] slab: add opt-in caching layer of " Vlastimil Babka
2025-02-22 22:46 ` Suren Baghdasaryan
2025-02-22 22:56 ` Suren Baghdasaryan
2025-03-12 14:57 ` Vlastimil Babka
2025-03-12 15:14 ` Suren Baghdasaryan
2025-03-17 10:09 ` Vlastimil Babka
2025-02-24 8:04 ` Harry Yoo
2025-03-12 14:59 ` Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 02/10] slab: add sheaf support for batching kfree_rcu() operations Vlastimil Babka
2025-02-22 23:08 ` Suren Baghdasaryan
2025-03-12 16:19 ` Vlastimil Babka
2025-02-24 8:40 ` Harry Yoo
2025-03-12 16:16 ` Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 03/10] locking/local_lock: Introduce localtry_lock_t Vlastimil Babka
2025-02-17 14:19 ` Sebastian Andrzej Siewior
2025-02-17 14:35 ` Vlastimil Babka
2025-02-17 15:07 ` Sebastian Andrzej Siewior
2025-02-18 18:41 ` Alexei Starovoitov
2025-02-26 17:00 ` Davidlohr Bueso
2025-02-26 17:15 ` Alexei Starovoitov
2025-02-26 19:28 ` Davidlohr Bueso
2025-02-14 16:27 ` [PATCH RFC v2 04/10] locking/local_lock: add localtry_trylock() Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 05/10] slab: switch percpu sheaves locking to localtry_lock Vlastimil Babka
2025-02-23 2:33 ` Suren Baghdasaryan
2025-02-24 13:08 ` Harry Yoo
2025-02-14 16:27 ` [PATCH RFC v2 06/10] slab: sheaf prefilling for guaranteed allocations Vlastimil Babka
2025-02-23 3:54 ` Suren Baghdasaryan
2025-02-25 7:30 ` Harry Yoo
2025-03-12 17:09 ` Vlastimil Babka
2025-02-25 8:00 ` Harry Yoo
2025-03-12 18:16 ` Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 07/10] slab: determine barn status racily outside of lock Vlastimil Babka
2025-02-23 4:00 ` Suren Baghdasaryan
2025-02-25 8:54 ` Harry Yoo [this message]
2025-03-12 18:23 ` Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 08/10] tools: Add testing support for changes to rcu and slab for sheaves Vlastimil Babka
2025-02-23 4:24 ` Suren Baghdasaryan
2025-02-14 16:27 ` [PATCH RFC v2 09/10] tools: Add sheafs support to testing infrastructure Vlastimil Babka
2025-02-14 16:27 ` [PATCH RFC v2 10/10] maple_tree: use percpu sheaves for maple_node_cache Vlastimil Babka
2025-02-23 4:27 ` Suren Baghdasaryan
2025-02-14 18:28 ` [PATCH RFC v2 00/10] SLUB percpu sheaves Christoph Lameter (Ampere)
2025-02-23 0:19 ` Kent Overstreet
2025-02-23 4:44 ` Suren Baghdasaryan
2025-02-24 1:36 ` Suren Baghdasaryan
2025-02-24 1:43 ` Suren Baghdasaryan
2025-02-24 20:53 ` Vlastimil Babka
2025-02-24 21:12 ` Suren Baghdasaryan
2025-02-25 20:26 ` Suren Baghdasaryan
2025-03-04 10:54 ` Vlastimil Babka
2025-03-04 18:35 ` Suren Baghdasaryan
2025-03-04 19:08 ` Liam R. Howlett
2025-03-14 17:10 ` Suren Baghdasaryan
2025-03-17 11:08 ` Vlastimil Babka
2025-03-17 18:56 ` Suren Baghdasaryan
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=Z72FPj_FzhCZpRsk@harry \
--to=harry.yoo@oracle.com \
--cc=42.hyeyoo@gmail.com \
--cc=Liam.Howlett@oracle.com \
--cc=cl@linux.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=maple-tree@lists.infradead.org \
--cc=rcu@vger.kernel.org \
--cc=rientjes@google.com \
--cc=roman.gushchin@linux.dev \
--cc=surenb@google.com \
--cc=urezki@gmail.com \
--cc=vbabka@suse.cz \
/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