From: Marco Elver <elver@google.com>
To: Andi Kleen <ak@linux.intel.com>
Cc: Oscar Salvador <osalvador@suse.de>,
andrey.konovalov@linux.dev,
Andrew Morton <akpm@linux-foundation.org>,
Andrey Konovalov <andreyknvl@gmail.com>,
Alexander Potapenko <glider@google.com>,
Dmitry Vyukov <dvyukov@google.com>,
Vlastimil Babka <vbabka@suse.cz>,
kasan-dev@googlegroups.com,
Evgenii Stepanov <eugenis@google.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Andrey Konovalov <andreyknvl@google.com>
Subject: Re: [PATCH v4 12/22] lib/stackdepot: use read/write lock
Date: Thu, 11 Jan 2024 10:48:19 +0100 [thread overview]
Message-ID: <CANpmjNNdWwGsD3JRcEqpq_ywwDFoxsBjz6n=6vL5YksNsPyqHw@mail.gmail.com> (raw)
In-Reply-To: <87sf34lrn3.fsf@linux.intel.com>
On Thu, 11 Jan 2024 at 00:01, Andi Kleen <ak@linux.intel.com> wrote:
>
> Oscar Salvador <osalvador@suse.de> writes:
> >>
> >> With this change, multiple users can still look up records in parallel.
>
> That's a severe misunderstanding -- rwlocks always bounce a cache line,
> so the parallelism is significantly reduced.
>
> Normally rwlocks are only worth it if your critical region is quite long.
>
> >>
> >> This is preparatory patch for implementing the eviction of stack records
> >> from the stack depot.
> >>
> >> Reviewed-by: Alexander Potapenko <glider@google.com>
> >> Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
> >
> > Reviewed-by: Oscar Salvador <osalvador@suse.de>
>
>
> Has anyone benchmarked this on a high core count machine? It sounds
> pretty bad if every lock aquisition starts bouncing a single cache line.
>
> Consider using RCU or similar.
stackdepot is severely limited in what kernel facilities it may use
due to being used by such low level facilities as the allocator
itself.
I've been suggesting percpu-rwsem here, but looking at it in more
detail that doesn't work because percpu-rwsem wants to sleep, but
stackdepot must work in non-sleepable contexts. :-/
next prev parent reply other threads:[~2024-01-11 9:49 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-20 17:46 [PATCH v4 00/22] stackdepot: allow evicting stack traces andrey.konovalov
2023-11-20 17:46 ` [PATCH v4 01/22] lib/stackdepot: print disabled message only if truly disabled andrey.konovalov
2024-01-03 8:14 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 02/22] lib/stackdepot: check disabled flag when fetching andrey.konovalov
2024-01-03 8:18 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 03/22] lib/stackdepot: simplify __stack_depot_save andrey.konovalov
2024-01-03 8:19 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 04/22] lib/stackdepot: drop valid bit from handles andrey.konovalov
2024-01-03 8:21 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 05/22] lib/stackdepot: add depot_fetch_stack helper andrey.konovalov
2024-01-03 8:24 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 06/22] lib/stackdepot: use fixed-sized slots for stack records andrey.konovalov
2024-01-03 8:30 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 07/22] lib/stackdepot: fix and clean-up atomic annotations andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 08/22] lib/stackdepot: rework helpers for depot_alloc_stack andrey.konovalov
2024-01-03 8:42 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 09/22] lib/stackdepot: rename next_pool_required to new_pool_required andrey.konovalov
2024-01-03 8:44 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 10/22] lib/stackdepot: store next pool pointer in new_pool andrey.konovalov
2024-01-03 9:06 ` Oscar Salvador
2023-11-20 17:47 ` [PATCH v4 11/22] lib/stackdepot: store free stack records in a freelist andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 12/22] lib/stackdepot: use read/write lock andrey.konovalov
2024-01-03 9:14 ` Oscar Salvador
2024-01-10 23:01 ` Andi Kleen
2024-01-11 9:48 ` Marco Elver [this message]
2024-01-11 12:36 ` Andi Kleen
2024-01-11 19:08 ` Marco Elver
2024-01-12 2:38 ` Andrey Konovalov
2024-01-12 8:24 ` Marco Elver
2024-01-12 22:15 ` Marco Elver
2024-01-13 1:24 ` Andi Kleen
2024-01-13 9:12 ` Marco Elver
2024-01-13 9:19 ` Andi Kleen
2024-01-13 9:23 ` Marco Elver
2024-01-13 9:30 ` Marco Elver
2024-01-13 9:31 ` Andi Kleen
2024-01-24 14:15 ` Breno Leitao
2024-01-24 14:21 ` Marco Elver
2024-01-24 16:24 ` Breno Leitao
2023-11-20 17:47 ` [PATCH v4 13/22] lib/stackdepot: use list_head for stack record links andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 14/22] kmsan: use stack_depot_save instead of __stack_depot_save andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 15/22] lib/stackdepot, kasan: add flags to __stack_depot_save and rename andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 16/22] lib/stackdepot: add refcount for records andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 17/22] lib/stackdepot: allow users to evict stack traces andrey.konovalov
2024-01-04 8:52 ` Oscar Salvador
2024-01-04 9:25 ` Marco Elver
2024-01-04 10:19 ` Oscar Salvador
2024-01-04 10:42 ` Marco Elver
2023-11-20 17:47 ` [PATCH v4 18/22] kasan: remove atomic accesses to stack ring entries andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 19/22] kasan: check object_size in kasan_complete_mode_report_info andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 20/22] kasan: use stack_depot_put for tag-based modes andrey.konovalov
2023-11-20 17:47 ` [PATCH v4 21/22] kasan: use stack_depot_put for Generic mode andrey.konovalov
2023-11-22 3:17 ` [BISECTED] Boot hangs when SLUB_DEBUG_ON=y Hyeonggon Yoo
2023-11-22 12:37 ` [REGRESSION] Boot hangs when SLUB_DEBUG_ON=y and KASAN_GENERIC=y Hyeonggon Yoo
2023-11-22 23:13 ` [BISECTED] Boot hangs when SLUB_DEBUG_ON=y Andrey Konovalov
2023-11-20 17:47 ` [PATCH v4 22/22] lib/stackdepot: adjust DEPOT_POOLS_CAP for KMSAN andrey.konovalov
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='CANpmjNNdWwGsD3JRcEqpq_ywwDFoxsBjz6n=6vL5YksNsPyqHw@mail.gmail.com' \
--to=elver@google.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=andrey.konovalov@linux.dev \
--cc=andreyknvl@gmail.com \
--cc=andreyknvl@google.com \
--cc=dvyukov@google.com \
--cc=eugenis@google.com \
--cc=glider@google.com \
--cc=kasan-dev@googlegroups.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=osalvador@suse.de \
--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