From: Willy Tarreau <w@1wt.eu>
To: Ammar Faizi <ammarfaizi2@gnuweeb.org>
Cc: "Alviro Iskandar Setiawan" <alviro.iskandar@gnuweeb.org>,
"Thomas Weißschuh" <linux@weissschuh.net>,
"Nicholas Rosenberg" <inori@vnlx.org>,
"Michael William Jonathan" <moe@gnuweeb.org>,
"GNU/Weeb Mailing List" <gwml@vger.gnuweeb.org>,
"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>
Subject: Re: [RFC PATCH v1 2/5] tools/nolibc: x86-64: Use `rep stosb` for `memset()`
Date: Wed, 30 Aug 2023 17:23:22 +0200 [thread overview]
Message-ID: <ZO9e6h2jjVIMpBJP@1wt.eu> (raw)
In-Reply-To: <ZO9bv+GvgpphtGEi@biznet-home.integral.gnuweeb.org>
On Wed, Aug 30, 2023 at 10:09:51PM +0700, Ammar Faizi wrote:
> On Wed, Aug 30, 2023 at 09:24:45PM +0700, Alviro Iskandar Setiawan wrote:
> > Just a small idea to shrink this more, "mov %rdi, %rdx" and "mov %rdx,
> > %rax" can be replaced with "push %rdi" and "pop %rax" (they are just a
> > byte). So we can save 4 bytes more.
> >
> > 0000000000001500 <memset>:
> > 1500: 48 89 f0 mov %rsi,%rax
> > 1503: 48 89 d1 mov %rdx,%rcx
> > 1506: 57 push %rdi
> > 1507: f3 aa rep stos %al,%es:(%rdi)
> > 1509: 58 pop %rax
> > 150a: c3 ret
> >
> > But I know you don't like it because it costs extra memory access.
>
> Yes, that's an extra memory access. But I believe it doesn't hurt
> someone targetting -Os. In many cases, the compilers use push/pop to
> align the stack before a 'call' instruction. If they want to avoid extra
> memory access, they could have used "subq $8, %rsp" and "addq $8, %rsp".
Then "xchg %esi, %eax" is just one byte with no memory access ;-)
Willy
next prev parent reply other threads:[~2023-08-30 18:36 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-30 13:57 [RFC PATCH v1 0/5] nolibc x86-64 string functions Ammar Faizi
2023-08-30 13:57 ` [RFC PATCH v1 1/5] tools/nolibc: x86-64: Use `rep movsb` for `memcpy()` and `memmove()` Ammar Faizi
2023-08-30 13:57 ` [RFC PATCH v1 2/5] tools/nolibc: x86-64: Use `rep stosb` for `memset()` Ammar Faizi
2023-08-30 14:08 ` Alviro Iskandar Setiawan
2023-08-30 14:13 ` Ammar Faizi
2023-08-30 14:24 ` Alviro Iskandar Setiawan
2023-08-30 15:09 ` Ammar Faizi
2023-08-30 15:23 ` Willy Tarreau [this message]
2023-08-30 15:44 ` Ammar Faizi
2023-08-30 15:51 ` Willy Tarreau
2023-08-30 16:08 ` Ammar Faizi
2023-08-30 16:11 ` Alviro Iskandar Setiawan
2023-08-30 13:57 ` [RFC PATCH v1 3/5] tools/nolibc: x86-64: Use `rep cmpsb` for `memcmp()` Ammar Faizi
2023-08-30 21:26 ` Willy Tarreau
2023-09-01 3:24 ` Ammar Faizi
2023-09-01 3:35 ` Willy Tarreau
2023-09-01 7:27 ` Ammar Faizi
2023-09-04 8:26 ` David Laight
2023-08-30 13:57 ` [RFC PATCH v1 4/5] tools/nolibc: string: Remove the `_nolibc_memcpy_down()` function Ammar Faizi
2023-08-30 21:27 ` Willy Tarreau
2023-08-30 13:57 ` [RFC PATCH v1 5/5] tools/nolibc: string: Remove the `_nolibc_memcpy_up()` function Ammar Faizi
2023-08-30 21:32 ` [RFC PATCH v1 0/5] nolibc x86-64 string functions Willy Tarreau
2023-09-01 11:34 ` David Laight
2023-09-01 11:46 ` Willy Tarreau
2023-09-01 13:06 ` Ammar Faizi
2023-09-01 14:23 ` David Laight
2023-09-01 14:41 ` Ammar Faizi
2023-09-01 14:54 ` David Laight
2023-09-01 15:20 ` Ammar Faizi
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=ZO9e6h2jjVIMpBJP@1wt.eu \
--to=w@1wt.eu \
--cc=alviro.iskandar@gnuweeb.org \
--cc=ammarfaizi2@gnuweeb.org \
--cc=gwml@vger.gnuweeb.org \
--cc=inori@vnlx.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@weissschuh.net \
--cc=moe@gnuweeb.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 NNTP newsgroup(s).