From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: Jeff Law <jeffreyalaw@gmail.com>, libc-alpha@sourceware.org
Subject: Re: [PATCH v10 00/24] Improve generic string routines
Date: Fri, 27 Jan 2023 13:52:43 -0300 [thread overview]
Message-ID: <ae7663ec-7480-e2cb-95ee-1fa9b9dcb356@linaro.org> (raw)
In-Reply-To: <cc345726-bfbb-405b-603e-6659d8a82afd@gmail.com>
On 20/01/23 20:00, Jeff Law via Libc-alpha wrote:
>
>
> On 1/20/23 14:15, Adhemerval Zanella via Libc-alpha wrote:
>> It is done by:
>>
>> 1. Parametrizing the internal routines (for instance the find zero
>> in a word) so each architecture can reimplement without the need
>> to reimplement the whole routine.
>>
>> 2. Vectorizing more string implementations (for instance strcpy
>> and strcmp).
>>
>> 3. Change some implementations to use already possible optimized
>> ones (strnlen and strchr). It makes new ports to focus on
>> only provide optimized implementation of a hardful symbols
>> (for instance memchr) and make its improvement to be used in
>> a larger set of routines.
>>
>> I checked on x86_64-linux-gnu, i686-linux-gnu, powerpc-linux-gnu,
>> and powerpc64-linux-gnu by removing the arch-specific assembly
>> implementation and disabling multiarch (it covers both LE and BE
>> for 64 and 32 bits). I also checked the string routines on alpha, hppa,
>> and sh.
> [ ... ]
> You guys are making work for me! :)
>
> We've got hand-written RISC-V implementations of various str* and mem* routines that I nearly asked to get merged for glibc-2.37. I held off thinking that there wasn't much benefit to including it in 2.37 vs 2.38. Bad call on my part!
>
Does this patchset work for the RISCV routine are aiming to optimize?
Do you have any strategy you think it would be profitable to add in
the generic framework?
Hand-optimize routines might squeeze some more cycles on some scenarios,
but at least the default framework should give a better performance.
I see that most of the arch-specific optimization adds alignment
consideration (specially if architecture provides fast unaligned access)
and loop unrolling for main loop. For former it would require a
different implementation, but for latter I think we can add per-arch
compiler flags to force unrolling (as we do for powerpc memmove/worcopy).
next prev parent reply other threads:[~2023-01-27 16:52 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-20 21:15 Adhemerval Zanella
2023-01-20 21:15 ` [PATCH v10 01/24] Parameterize op_t from memcopy.h Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 02/24] Parameterize OP_T_THRES " Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 03/24] Add string vectorized find and detection functions Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 04/24] string: Improve generic strlen Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 05/24] string: Improve generic strnlen Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 06/24] string: Improve generic strchr Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 07/24] string: Improve generic strchrnul Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 08/24] string: Improve generic strcmp Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 09/24] string: Improve generic strncmp Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 10/24] string: Improve generic memchr Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 11/24] string: Improve generic memrchr Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 12/24] hppa: Add memcopy.h Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 13/24] hppa: Add string-fzb.h and string-fzi.h Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 14/24] alpha: " Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 15/24] arm: Add string-fza.h Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 16/24] powerpc: " Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 17/24] sh: Add string-fzb.h Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 18/24] string: Hook up the default implementation on test-strlen Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 19/24] string: Hook up the default implementation on test-strnlen Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 20/24] string: Hook up the default implementation on test-strchr Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 21/24] string: Hook up the default implementation on test-strcmp Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 22/24] string: Hook up the default implementation on test-strncmp Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 23/24] string: Hook up the default implementation on test-memchr Adhemerval Zanella
2023-01-20 21:16 ` [PATCH v10 24/24] string: Hook up the default implementation on test-memrchr Adhemerval Zanella
2023-01-20 23:00 ` [PATCH v10 00/24] Improve generic string routines Jeff Law
2023-01-27 16:52 ` Adhemerval Zanella Netto [this message]
2023-01-27 19:36 ` Richard Henderson
2023-01-30 13:26 ` Adhemerval Zanella Netto
2023-01-30 18:37 ` Jeff Law
2023-01-28 14:30 ` Xi Ruoyao
2023-01-30 13:10 ` Adhemerval Zanella Netto
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=ae7663ec-7480-e2cb-95ee-1fa9b9dcb356@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=jeffreyalaw@gmail.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).