From: Szabolcs Nagy <Szabolcs.Nagy@arm.com>
To: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Cc: 'GNU C Library' <libc-alpha@sourceware.org>
Subject: Re: [PATCH] AArch64: Improve SVE memcpy and memmove
Date: Mon, 6 Feb 2023 14:29:24 +0000 [thread overview]
Message-ID: <Y+EOxNmCAEiW4LQ6@arm.com> (raw)
In-Reply-To: <PAWPR08MB898248BE9DB331AC6DBEF4B183D79@PAWPR08MB8982.eurprd08.prod.outlook.com>
The 02/03/2023 13:05, Wilco Dijkstra wrote:
> Improve SVE memcpy/memmove by copying 2 vectors if the size is small enough.
> This improves performance of random memcpy by ~9% on Neoverse V1, and
> memcpy/memmove of 33-64 bytes become ~16% faster.
>
> Passes regress, OK for commit?
This is ok to commit, thanks.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
>
> ---
>
> diff --git a/sysdeps/aarch64/multiarch/memcpy_sve.S b/sysdeps/aarch64/multiarch/memcpy_sve.S
> index f4dc214f60bf25e818eb6b8de2d4093ad0c886e1..d11be6a44301af4bfd7fa4900555b769dc58d34d 100644
> --- a/sysdeps/aarch64/multiarch/memcpy_sve.S
> +++ b/sysdeps/aarch64/multiarch/memcpy_sve.S
> @@ -67,14 +67,15 @@ ENTRY (__memcpy_sve)
>
> cmp count, 128
> b.hi L(copy_long)
> - cmp count, 32
> + cntb vlen
> + cmp count, vlen, lsl 1
> b.hi L(copy32_128)
> -
> whilelo p0.b, xzr, count
> - cntb vlen
> - tbnz vlen, 4, L(vlen128)
> - ld1b z0.b, p0/z, [src]
> - st1b z0.b, p0, [dstin]
> + whilelo p1.b, vlen, count
> + ld1b z0.b, p0/z, [src, 0, mul vl]
> + ld1b z1.b, p1/z, [src, 1, mul vl]
> + st1b z0.b, p0, [dstin, 0, mul vl]
> + st1b z1.b, p1, [dstin, 1, mul vl]
> ret
>
> /* Medium copies: 33..128 bytes. */
> @@ -102,14 +103,6 @@ L(copy96):
> stp C_q, D_q, [dstend, -32]
> ret
>
> -L(vlen128):
> - whilelo p1.b, vlen, count
> - ld1b z0.b, p0/z, [src, 0, mul vl]
> - ld1b z1.b, p1/z, [src, 1, mul vl]
> - st1b z0.b, p0, [dstin, 0, mul vl]
> - st1b z1.b, p1, [dstin, 1, mul vl]
> - ret
> -
> .p2align 4
> /* Copy more than 128 bytes. */
> L(copy_long):
> @@ -158,14 +151,15 @@ ENTRY (__memmove_sve)
>
> cmp count, 128
> b.hi L(move_long)
> - cmp count, 32
> + cntb vlen
> + cmp count, vlen, lsl 1
> b.hi L(copy32_128)
> -
> whilelo p0.b, xzr, count
> - cntb vlen
> - tbnz vlen, 4, L(vlen128)
> - ld1b z0.b, p0/z, [src]
> - st1b z0.b, p0, [dstin]
> + whilelo p1.b, vlen, count
> + ld1b z0.b, p0/z, [src, 0, mul vl]
> + ld1b z1.b, p1/z, [src, 1, mul vl]
> + st1b z0.b, p0, [dstin, 0, mul vl]
> + st1b z1.b, p1, [dstin, 1, mul vl]
> ret
>
> .p2align 4
>
prev parent reply other threads:[~2023-02-06 14:29 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-03 13:05 Wilco Dijkstra
2023-02-06 14:29 ` Szabolcs Nagy [this message]
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=Y+EOxNmCAEiW4LQ6@arm.com \
--to=szabolcs.nagy@arm.com \
--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).