public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: Richard Henderson <richard.henderson@linaro.org>,
	libc-alpha@sourceware.org, Jeff Law <jeffreyalaw@gmail.com>,
	Xi Ruoyao <xry111@xry111.site>,
	Noah Goldstein <goldstein.w.n@gmail.com>
Subject: Re: [PATCH v12 14/31] string: Improve generic strrchr with memrchr and strlen
Date: Mon, 6 Feb 2023 11:01:04 -0300	[thread overview]
Message-ID: <a248b824-6d81-c4ae-52ed-8deaa7836e15@linaro.org> (raw)
In-Reply-To: <f9005160-0a35-8a2c-1307-ee86aedf7071@linaro.org>



On 04/02/23 00:06, Richard Henderson wrote:
> On 2/2/23 08:11, Adhemerval Zanella wrote:
>> Now that both strlen and memrchr have word vectorized implementation,
>> it should be faster to implement strrchr based on memrchr over the
>> string length instead of calling strchr on a loop.
>>
>> 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 multi-arch (it covers both LE and BE
>> for 64 and 32 bits).
>> ---
>>   string/strrchr.c | 18 +-----------------
>>   1 file changed, 1 insertion(+), 17 deletions(-)
> 
> memrchr needs libc_hidden_builtin_proto.  On riscv64:
> 
>    c:   00000097                auipc   ra,0x0
>                         c: R_RISCV_CALL __GI_strlen
>                         c: R_RISCV_RELAX        *ABS*
>   10:   000080e7                jalr    ra # c <__GI_strrchr+0xc>
> ...
>   24:   00000317                auipc   t1,0x0
>                         24: R_RISCV_CALL_PLT    __memrchr
>                         24: R_RISCV_RELAX       *ABS*
>   28:   00030067                jr      t1 # 24 <.LVL2+0x8>
> 
> 
> r~

Similar to strchr [1], static linker ends up creating the expected local
call:

$ riscv64-glibc-linux-gnu-objdump -dr libc.so
[...]
0000000000088f5c <strrchr>:
   88f5c:       7179                    addi    sp,sp,-48
   88f5e:       e84a                    sd      s2,16(sp)
   88f60:       000e7917                auipc   s2,0xe7
   88f64:       73893903                ld      s2,1848(s2) # 170698 <__stack_chk_guard@GLIBC_2.27>
   88f68:       00093783                ld      a5,0(s2)
[...]
   88f9a:       6145                    addi    sp,sp,48
   88f9c:       c70fd06f                j       8640c <memrchr>
   88fa0:       193500ef                jal     ra,d9932 <__stack_chk_fail>
[...]

But I agree that we should change to have the libc_hidden_builin_proto
in this case.  I will send a patch to fix it.

[1] https://sourceware.org/pipermail/libc-alpha/2023-February/145322.html

  reply	other threads:[~2023-02-06 14:01 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-02 18:11 [PATCH v12 00/31] Improve generic string routines Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 01/31] Parameterize op_t from memcopy.h Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 02/31] Parameterize OP_T_THRES " Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 03/31] Add string vectorized find and detection functions Adhemerval Zanella
2023-02-03  0:24   ` Richard Henderson
2023-02-03 12:39     ` Adhemerval Zanella Netto
2023-02-03 13:40       ` Adhemerval Zanella Netto
2023-02-04  2:54         ` Richard Henderson
2023-02-02 18:11 ` [PATCH v12 04/31] string: Improve generic strlen Adhemerval Zanella
2023-02-03 23:23   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 05/31] string: Improve generic strchrnul Adhemerval Zanella
2023-02-03 23:23   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 06/31] string: Improve generic strchr Adhemerval Zanella
2023-02-03 23:24   ` Noah Goldstein
2023-02-04  2:58   ` Richard Henderson
2023-02-06 13:07     ` Adhemerval Zanella Netto
2023-02-02 18:11 ` [PATCH v12 07/31] string: Improve generic strcmp Adhemerval Zanella
2023-02-03 23:25   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 08/31] string: Improve generic strncmp Adhemerval Zanella
2023-02-03  0:34   ` Richard Henderson
2023-02-21  9:28   ` Szabolcs Nagy
2023-02-21 12:24     ` Adhemerval Zanella Netto
2023-02-02 18:11 ` [PATCH v12 09/31] string: Improve generic stpcpy Adhemerval Zanella
2023-02-03 23:26   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 10/31] string: Improve generic strcpy Adhemerval Zanella
2023-02-03 23:26   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 11/31] string: Improve generic memchr Adhemerval Zanella
2023-02-03 23:26   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 12/31] string: Improve generic strnlen with memchr Adhemerval Zanella
2023-02-03 23:27   ` Noah Goldstein
2023-02-02 18:11 ` [PATCH v12 13/31] string: Improve generic memrchr Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 14/31] string: Improve generic strrchr with memrchr and strlen Adhemerval Zanella
2023-02-04  3:06   ` Richard Henderson
2023-02-06 14:01     ` Adhemerval Zanella Netto [this message]
2023-02-02 18:11 ` [PATCH v12 15/31] hppa: Add memcopy.h Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 16/31] hppa: Add string-fza.h, string-fzc.h, and string-fzi.h Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 17/31] alpha: Add string-fza, string-fzb.h, string-fzi.h, and string-shift.h Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 18/31] arm: Add string-fza.h Adhemerval Zanella
2023-02-20 13:24   ` Szabolcs Nagy
2023-02-20 13:45     ` Szabolcs Nagy
2023-02-20 14:01       ` Adhemerval Zanella Netto
2023-02-20 16:12         ` Szabolcs Nagy
2023-02-02 18:11 ` [PATCH v12 19/31] powerpc: " Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 20/31] sh: Add string-fzb.h Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 21/31] riscv: Add string-fza.h and string-fzi.h Adhemerval Zanella
2023-02-03 19:47   ` Noah Goldstein
2023-02-06 16:14     ` Adhemerval Zanella Netto
2023-02-02 18:11 ` [PATCH v12 22/31] string: Hook up the default implementation on test-strlen Adhemerval Zanella
2023-02-03 23:30   ` Noah Goldstein
2023-02-06 17:36     ` Adhemerval Zanella Netto
2023-02-02 18:11 ` [PATCH v12 23/31] string: Hook up the default implementation on test-strnlen Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 24/31] string: Hook up the default implementation on test-strchr Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 25/31] string: Hook up the default implementation on test-strcmp Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 26/31] string: Hook up the default implementation on test-strncmp Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 27/31] string: Hook up the default implementation on test-stpcpy Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 28/31] string: Hook up the default implementation on test-strcpy Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 29/31] string: Hook up the default implementation on test-memchr Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 30/31] string: Hook up the default implementation on test-memrchr Adhemerval Zanella
2023-02-02 18:11 ` [PATCH v12 31/31] string: Hook up the default implementation on test-strrchr Adhemerval Zanella
2023-02-03  1:55   ` Richard Henderson
2023-02-02 18:17 ` [PATCH v12 00/31] Improve generic string routines Adhemerval Zanella Netto
2023-02-02 19:54 [PATCH v12 14/31] string: Improve generic strrchr with memrchr and strlen Wilco Dijkstra
2023-02-03 12:15 ` 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=a248b824-6d81-c4ae-52ed-8deaa7836e15@linaro.org \
    --to=adhemerval.zanella@linaro.org \
    --cc=goldstein.w.n@gmail.com \
    --cc=jeffreyalaw@gmail.com \
    --cc=libc-alpha@sourceware.org \
    --cc=richard.henderson@linaro.org \
    --cc=xry111@xry111.site \
    /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).