From: "Cristian Rodríguez" <crrodriguez@opensuse.org>
To: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Cc: GNU C Library <libc-alpha@sourceware.org>
Subject: Re: [PATCH] Improve performance of IO locks
Date: Sun, 7 Aug 2022 08:51:08 -0400 [thread overview]
Message-ID: <CAPBLoAee8JZCnGxbtd+9VuU7Fyi6iQf29a0rkGMCfGFpGKMMag@mail.gmail.com> (raw)
In-Reply-To: <AM5PR0801MB1668FF146572E2069648B5F1839A9@AM5PR0801MB1668.eurprd08.prod.outlook.com>
Sounds good, tested on x86_64.
On Mon, Aug 1, 2022 at 7:06 AM Wilco Dijkstra via Libc-alpha
<libc-alpha@sourceware.org> wrote:
>
>
> ping
>
>
> Improve performance of recursive IO locks by adding a fast path for
> the single-threaded case. To reduce the number of memory accesses for
> locking/unlocking, only increment the recursion counter if the lock
> is already taken.
>
> On Neoverse V1, a microbenchmark with many small freads improved by
> 2.9 times. Multithreaded performance improved by 2%.
>
> Passes GLIBC testsuite, OK for commit?
>
> ---
>
> diff --git a/sysdeps/nptl/stdio-lock.h b/sysdeps/nptl/stdio-lock.h
> index afa0b779c81d7dd915f8edb6c0974e4f231d4e0a..45823cd1629d3e3efecc64a7d07706a6e6de9af1 100644
> --- a/sysdeps/nptl/stdio-lock.h
> +++ b/sysdeps/nptl/stdio-lock.h
> @@ -37,12 +37,18 @@ typedef struct { int lock; int cnt; void *owner; } _IO_lock_t;
> #define _IO_lock_lock(_name) \
> do { \
> void *__self = THREAD_SELF; \
> - if ((_name).owner != __self) \
> + if (SINGLE_THREAD_P && (_name).owner == NULL) \
> + { \
> + (_name).lock = LLL_LOCK_INITIALIZER_LOCKED; \
> + (_name).owner = __self; \
> + } \
> + else if ((_name).owner != __self) \
> { \
> lll_lock ((_name).lock, LLL_PRIVATE); \
> - (_name).owner = __self; \
> + (_name).owner = __self; \
> } \
> - ++(_name).cnt; \
> + else \
> + ++(_name).cnt; \
> } while (0)
>
> #define _IO_lock_trylock(_name) \
> @@ -52,10 +58,7 @@ typedef struct { int lock; int cnt; void *owner; } _IO_lock_t;
> if ((_name).owner != __self) \
> { \
> if (lll_trylock ((_name).lock) == 0) \
> - { \
> - (_name).owner = __self; \
> - (_name).cnt = 1; \
> - } \
> + (_name).owner = __self; \
> else \
> __result = EBUSY; \
> } \
> @@ -66,11 +69,18 @@ typedef struct { int lock; int cnt; void *owner; } _IO_lock_t;
>
> #define _IO_lock_unlock(_name) \
> do { \
> - if (--(_name).cnt == 0) \
> + if (SINGLE_THREAD_P && (_name).cnt == 0) \
> + { \
> + (_name).owner = NULL; \
> + (_name).lock = 0; \
> + } \
> + else if ((_name).cnt == 0) \
> { \
> - (_name).owner = NULL; \
> + (_name).owner = NULL; \
> lll_unlock ((_name).lock, LLL_PRIVATE); \
> } \
> + else \
> + --(_name).cnt; \
> } while (0)
>
>
next prev parent reply other threads:[~2022-08-07 12:51 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-08 16:32 Wilco Dijkstra
2022-08-01 11:06 ` Wilco Dijkstra
2022-08-07 12:51 ` Cristian Rodríguez [this message]
2022-08-15 22:27 ` Mark Wielaard
2022-08-16 3:07 ` Thomas Fitzsimmons
2022-08-16 7:31 ` Florian Weimer
2022-08-16 10:00 ` Mark Wielaard
2022-08-16 10:08 ` Florian Weimer
2022-08-17 13:45 ` Mark Wielaard
2022-08-22 10:25 ` Florian Weimer
2022-08-22 14:58 ` Frank Ch. Eigler
2022-08-16 11:19 ` Andreas Schwab
2022-08-16 10:24 ` Wilco Dijkstra
2022-08-16 13:18 ` Thomas Fitzsimmons
2022-08-16 14:31 ` Thomas Fitzsimmons
2022-08-17 11:53 ` Mark Wielaard
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=CAPBLoAee8JZCnGxbtd+9VuU7Fyi6iQf29a0rkGMCfGFpGKMMag@mail.gmail.com \
--to=crrodriguez@opensuse.org \
--cc=Wilco.Dijkstra@arm.com \
--cc=libc-alpha@sourceware.org \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).