From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: Richard Henderson <rth@twiddle.net>,
Noah Goldstein <goldstein.w.n@gmail.com>
Cc: libc-alpha@sourceware.org
Subject: Re: [PATCH v5 08/17] string: Improve generic strchrnul
Date: Tue, 10 Jan 2023 11:18:42 -0300 [thread overview]
Message-ID: <ab08e6d3-4005-6364-d566-0a909093b84e@linaro.org> (raw)
In-Reply-To: <f708a08f-33fe-c881-529f-e6241b7d6c45@twiddle.net>
On 09/01/23 20:33, Richard Henderson wrote:
> On 1/9/23 12:35, Adhemerval Zanella Netto wrote:
>>
>>
>> On 05/01/23 20:17, Noah Goldstein wrote:
>>> On Mon, Sep 19, 2022 at 1:04 PM Adhemerval Zanella via Libc-alpha
>>> <libc-alpha@sourceware.org> wrote:
>>>>
>>>> New algorithm have the following key differences:
>>>>
>>>> - Reads first word unaligned and use string-maskoff function to
>>>> remove unwanted data. This strategy follow arch-specific
>>>> optimization used on aarch64 and powerpc.
>>>>
>>>> - Use string-fz{b,i} functions.
>>>>
>>>> Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64-linux-gnu,
>>>> and powerpc-linux-gnu by removing the arch-specific assembly
>>>> implementation and disabling multi-arch (it covers both LE and BE
>>>> for 64 and 32 bits).
>>>>
>>>> Co-authored-by: Richard Henderson <rth@twiddle.net>
>>>> ---
>>>> string/strchrnul.c | 156 +++---------------
>>>> .../power4/multiarch/strchrnul-ppc32.c | 4 -
>>>> sysdeps/s390/strchrnul-c.c | 2 -
>>>> 3 files changed, 24 insertions(+), 138 deletions(-)
>>>>
>>>> diff --git a/string/strchrnul.c b/string/strchrnul.c
>>>> index 0cc1fc6bb0..67defa3dab 100644
>>>> --- a/string/strchrnul.c
>>>> +++ b/string/strchrnul.c
>>>> @@ -1,10 +1,5 @@
>>>> /* Copyright (C) 1991-2022 Free Software Foundation, Inc.
>>>> This file is part of the GNU C Library.
>>>> - Based on strlen implementation by Torbjorn Granlund (tege@sics.se),
>>>> - with help from Dan Sahlin (dan@sics.se) and
>>>> - bug fix and commentary by Jim Blandy (jimb@ai.mit.edu);
>>>> - adaptation to strchr suggested by Dick Karpinski (dick@cca.ucsf.edu),
>>>> - and implemented by Roland McGrath (roland@ai.mit.edu).
>>>>
>>>> The GNU C Library is free software; you can redistribute it and/or
>>>> modify it under the terms of the GNU Lesser General Public
>>>> @@ -21,146 +16,43 @@
>>>> <https://www.gnu.org/licenses/>. */
>>>>
>>>> #include <string.h>
>>>> -#include <memcopy.h>
>>>> #include <stdlib.h>
>>>> +#include <stdint.h>
>>>> +#include <string-fza.h>
>>>> +#include <string-fzb.h>
>>>> +#include <string-fzi.h>
>>>> +#include <string-maskoff.h>
>>>>
>>>> #undef __strchrnul
>>>> #undef strchrnul
>>>>
>>>> -#ifndef STRCHRNUL
>>>> -# define STRCHRNUL __strchrnul
>>>> +#ifdef STRCHRNUL
>>>> +# define __strchrnul STRCHRNUL
>>>> #endif
>>>>
>>>> /* Find the first occurrence of C in S or the final NUL byte. */
>>>> char *
>>>> -STRCHRNUL (const char *s, int c_in)
>>>> +__strchrnul (const char *str, int c_in)
>>>> {
>>>> - const unsigned char *char_ptr;
>>>> - const unsigned long int *longword_ptr;
>>>> - unsigned long int longword, magic_bits, charmask;
>>>> - unsigned char c;
>>>> -
>>>> - c = (unsigned char) c_in;
>>>> -
>>>> - /* Handle the first few characters by reading one character at a time.
>>>> - Do this until CHAR_PTR is aligned on a longword boundary. */
>>>> - for (char_ptr = (const unsigned char *) s;
>>>> - ((unsigned long int) char_ptr & (sizeof (longword) - 1)) != 0;
>>>> - ++char_ptr)
>>>> - if (*char_ptr == c || *char_ptr == '\0')
>>>> - return (void *) char_ptr;
>>>> -
>>>> - /* All these elucidatory comments refer to 4-byte longwords,
>>>> - but the theory applies equally well to 8-byte longwords. */
>>>> -
>>>> - longword_ptr = (unsigned long int *) char_ptr;
>>>> -
>>>> - /* Bits 31, 24, 16, and 8 of this number are zero. Call these bits
>>>> - the "holes." Note that there is a hole just to the left of
>>>> - each byte, with an extra at the end:
>>>> -
>>>> - bits: 01111110 11111110 11111110 11111111
>>>> - bytes: AAAAAAAA BBBBBBBB CCCCCCCC DDDDDDDD
>>>> -
>>>> - The 1-bits make sure that carries propagate to the next 0-bit.
>>>> - The 0-bits provide holes for carries to fall into. */
>>>> - magic_bits = -1;
>>>> - magic_bits = magic_bits / 0xff * 0xfe << 1 >> 1 | 1;
>>>> -
>>>> - /* Set up a longword, each of whose bytes is C. */
>>>> - charmask = c | (c << 8);
>>>> - charmask |= charmask << 16;
>>>> - if (sizeof (longword) > 4)
>>>> - /* Do the shift in two steps to avoid a warning if long has 32 bits. */
>>>> - charmask |= (charmask << 16) << 16;
>>>> - if (sizeof (longword) > 8)
>>>> - abort ();
>>>> -
>>>> - /* Instead of the traditional loop which tests each character,
>>>> - we will test a longword at a time. The tricky part is testing
>>>> - if *any of the four* bytes in the longword in question are zero. */
>>>> - for (;;)
>>>> - {
>>>> - /* We tentatively exit the loop if adding MAGIC_BITS to
>>>> - LONGWORD fails to change any of the hole bits of LONGWORD.
>>>> -
>>>> - 1) Is this safe? Will it catch all the zero bytes?
>>>> - Suppose there is a byte with all zeros. Any carry bits
>>>> - propagating from its left will fall into the hole at its
>>>> - least significant bit and stop. Since there will be no
>>>> - carry from its most significant bit, the LSB of the
>>>> - byte to the left will be unchanged, and the zero will be
>>>> - detected.
>>>> + /* Set up a word, each of whose bytes is C. */
>>>> + op_t repeated_c = repeat_bytes (c_in);
>>>>
>>>> - 2) Is this worthwhile? Will it ignore everything except
>>>> - zero bytes? Suppose every byte of LONGWORD has a bit set
>>>> - somewhere. There will be a carry into bit 8. If bit 8
>>>> - is set, this will carry into bit 16. If bit 8 is clear,
>>>> - one of bits 9-15 must be set, so there will be a carry
>>>> - into bit 16. Similarly, there will be a carry into bit
>>>> - 24. If one of bits 24-30 is set, there will be a carry
>>>> - into bit 31, so all of the hole bits will be changed.
>>>> + /* Align the input address to op_t. */
>>>> + uintptr_t s_int = (uintptr_t) str;
>>>> + const op_t *word_ptr = word_containing (str);
>>>>
>>>> - The one misfire occurs when bits 24-30 are clear and bit
>>>> - 31 is set; in this case, the hole at bit 31 is not
>>>> - changed. If we had access to the processor carry flag,
>>>> - we could close this loophole by putting the fourth hole
>>>> - at bit 32!
>>>> + /* Read the first aligned word, but force bytes before the string to
>>>> + match neither zero nor goal (we make sure the high bit of each byte
>>>> + is 1, and the low 7 bits are all the opposite of the goal byte). */
>>>> + op_t bmask = create_mask (s_int);
>>>> + op_t word = (*word_ptr | bmask) ^ (repeated_c & highbit_mask (bmask));
>>>
>>> Think much clearer (and probably better codegen) is:
>>> find_zero_eq_low/all(word, repeated) >> (s_int * CHAR_BIT)
>>
>> It does not seem to work, at least not replacing the two lines with:
>>
>> op_t word = find_zero_eq_all/low (*word_ptr, repeated_c) >> (s_int * CHAR_BIT);
>
> Oh, two fine points:
>
> (1) big-endian would want shifting left,
> (2) alpha would want shifting by bits not bytes,
> because the cmpbge insn produces an 8-bit mask.
>
> so you'd need to hide this shift in the headers like create_mask().
Alright, the following works:
static __always_inline op_t
check_mask (op_t word, uintptr_t s_int)
{
if (__BYTE_ORDER == __LITTLE_ENDIAN)
return word >> (CHAR_BIT * (s_int % sizeof (s_int)));
else
return word << (CHAR_BIT * (s_int % sizeof (s_int)));
}
char *
__strchrnul (const char *str, int c_in)
{
op_t repeated_c = repeat_bytes (c_in);
uintptr_t s_int = (uintptr_t) str;
const op_t *word_ptr = word_containing (str);
op_t word = *word_ptr;
op_t mask = check_mask (find_zero_eq_all (word, repeated_c), s_int);
if (mask != 0)
return (char *) str + index_first_(mask);
do
word = *++word_ptr;
while (! has_zero_eq (word, repeated_c));
op_t found = index_first_zero_eq (word, repeated_c);
return (char *) word_ptr + found;
}
I had to use find_zero_eq_all to avoid uninitialized bytes, that triggered
some regression on tests that use strchr (for instance test-strpbrk).
I will update the patch based on this version.
next prev parent reply other threads:[~2023-01-10 14:18 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-19 19:59 [PATCH v5 00/17] Improve generic string routines Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 01/17] Parameterize op_t from memcopy.h Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 02/17] Parameterize OP_T_THRES " Adhemerval Zanella
2022-09-20 10:49 ` Carlos O'Donell
2022-09-19 19:59 ` [PATCH v5 03/17] Add string-maskoff.h generic header Adhemerval Zanella
2022-09-20 11:43 ` Carlos O'Donell
2022-09-22 17:31 ` Adhemerval Zanella Netto
2023-01-05 22:49 ` Noah Goldstein
2023-01-05 23:26 ` Alejandro Colomar
2023-01-09 18:19 ` Adhemerval Zanella Netto
2023-01-09 18:02 ` Adhemerval Zanella Netto
2022-09-19 19:59 ` [PATCH v5 04/17] Add string vectorized find and detection functions Adhemerval Zanella
2023-01-05 22:53 ` Noah Goldstein
2023-01-09 18:51 ` Adhemerval Zanella Netto
2023-01-05 23:04 ` Noah Goldstein
2023-01-09 19:34 ` Adhemerval Zanella Netto
2022-09-19 19:59 ` [PATCH v5 05/17] string: Improve generic strlen Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 06/17] string: Improve generic strnlen Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 07/17] string: Improve generic strchr Adhemerval Zanella
2023-01-05 23:09 ` Noah Goldstein
2023-01-05 23:19 ` Noah Goldstein
2023-01-09 19:39 ` Adhemerval Zanella Netto
2022-09-19 19:59 ` [PATCH v5 08/17] string: Improve generic strchrnul Adhemerval Zanella
2023-01-05 23:17 ` Noah Goldstein
2023-01-09 20:35 ` Adhemerval Zanella Netto
2023-01-09 20:49 ` Richard Henderson
2023-01-09 20:59 ` Noah Goldstein
2023-01-09 21:01 ` Noah Goldstein
2023-01-09 23:33 ` Richard Henderson
2023-01-10 14:18 ` Adhemerval Zanella Netto [this message]
2023-01-10 16:24 ` Richard Henderson
2023-01-10 17:16 ` Noah Goldstein
2023-01-10 18:19 ` Adhemerval Zanella Netto
2023-01-10 17:17 ` Noah Goldstein
2023-01-10 18:16 ` Adhemerval Zanella Netto
2022-09-19 19:59 ` [PATCH v5 09/17] string: Improve generic strcmp Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 10/17] string: Improve generic memchr Adhemerval Zanella
2023-01-05 23:47 ` Noah Goldstein
2023-01-09 20:50 ` Adhemerval Zanella Netto
2023-01-05 23:49 ` Noah Goldstein
2023-01-09 20:51 ` Adhemerval Zanella Netto
2023-01-09 21:26 ` Noah Goldstein
2023-01-10 14:33 ` Adhemerval Zanella Netto
2022-09-19 19:59 ` [PATCH v5 11/17] string: Improve generic memrchr Adhemerval Zanella
2023-01-05 23:51 ` Noah Goldstein
2022-09-19 19:59 ` [PATCH v5 12/17] hppa: Add memcopy.h Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 13/17] hppa: Add string-fzb.h and string-fzi.h Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 14/17] alpha: " Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 15/17] arm: Add string-fza.h Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 16/17] powerpc: " Adhemerval Zanella
2022-09-19 19:59 ` [PATCH v5 17/17] sh: Add string-fzb.h Adhemerval Zanella
2022-12-05 17:07 ` [PATCH v5 00/17] Improve generic string routines Xi Ruoyao
2023-01-05 21:56 ` Adhemerval Zanella Netto
2023-01-05 23:52 ` Noah Goldstein
2023-01-06 13:43 ` 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=ab08e6d3-4005-6364-d566-0a909093b84e@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=goldstein.w.n@gmail.com \
--cc=libc-alpha@sourceware.org \
--cc=rth@twiddle.net \
/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).