From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: libc-alpha@sourceware.org,
Richard Henderson <richard.henderson@linaro.org>,
Noah Goldstein <goldstein.w.n@gmail.com>
Subject: [PATCH v9 08/22] string: Improve generic strcmp
Date: Tue, 17 Jan 2023 17:00:00 -0300 [thread overview]
Message-ID: <20230117200014.1299923-9-adhemerval.zanella@linaro.org> (raw)
In-Reply-To: <20230117200014.1299923-1-adhemerval.zanella@linaro.org>
New generic implementation tries to use word operations along with
the new string-fz{b,i} functions even for inputs with different
alignments (with still uses aligned access plus merge operation
to get a correct word by word comparison).
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 <richard.henderson@linaro.org>
---
string/strcmp.c | 119 +++++++++++++++++++++++++++++++++++++++++-------
1 file changed, 103 insertions(+), 16 deletions(-)
diff --git a/string/strcmp.c b/string/strcmp.c
index 053f5a8d2b..c7fcd61f5c 100644
--- a/string/strcmp.c
+++ b/string/strcmp.c
@@ -15,33 +15,120 @@
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
+#include <stdint.h>
+#include <string-extbyte.h>
+#include <string-fzb.h>
+#include <string-fzi.h>
#include <string.h>
+#include <memcopy.h>
-#undef strcmp
-
-#ifndef STRCMP
-# define STRCMP strcmp
+#ifdef STRCMP
+# define strcmp STRCMP
#endif
+static inline int
+final_cmp (const op_t w1, const op_t w2)
+{
+ /* It can not use index_first_zero_ne because it must not compare past the
+ final '\0' is present (and final_cmp is called before has_zero check).
+ */
+ for (size_t i = 0; i < sizeof (op_t); i++)
+ {
+ unsigned char c1 = extractbyte (w1, i);
+ unsigned char c2 = extractbyte (w2, i);
+ if (c1 == '\0' || c1 != c2)
+ return c1 - c2;
+ }
+ return 0;
+}
+
+/* Aligned loop: if a difference is found, exit to compare the bytes. Else
+ if a zero is found we have equal strings. */
+static inline int
+strcmp_aligned_loop (const op_t *x1, const op_t *x2, op_t w1)
+{
+ op_t w2 = *x2++;
+
+ while (w1 == w2)
+ {
+ if (has_zero (w1))
+ return 0;
+ w1 = *x1++;
+ w2 = *x2++;
+ }
+
+ return final_cmp (w1, w2);
+}
+
+/* Unaligned loop: align the first partial of P2, with 0xff for the rest of
+ the bytes so that we can also apply the has_zero test to see if we have
+ already reached EOS. If we have, then we can simply fall through to the
+ final comparison. */
+static inline int
+strcmp_unaligned_loop (const op_t *x1, const op_t *x2, op_t w1, uintptr_t ofs)
+{
+ op_t w2a = *x2++;
+ uintptr_t sh_1 = ofs * CHAR_BIT;
+ uintptr_t sh_2 = sizeof(op_t) * CHAR_BIT - sh_1;
+
+ op_t w2 = MERGE (w2a, sh_1, (op_t)-1, sh_2);
+ if (!has_zero (w2))
+ {
+ op_t w2b;
+
+ /* Unaligned loop. The invariant is that W2B, which is "ahead" of W1,
+ does not contain end-of-string. Therefore it is safe (and necessary)
+ to read another word from each while we do not have a difference. */
+ while (1)
+ {
+ w2b = *x2++;
+ w2 = MERGE (w2a, sh_1, w2b, sh_2);
+ if (w1 != w2)
+ return final_cmp (w1, w2);
+ if (has_zero (w2b))
+ break;
+ w1 = *x1++;
+ w2a = w2b;
+ }
+
+ /* Zero found in the second partial of P2. If we had EOS in the aligned
+ word, we have equality. */
+ if (has_zero (w1))
+ return 0;
+
+ /* Load the final word of P1 and align the final partial of P2. */
+ w1 = *x1++;
+ w2 = MERGE (w2b, sh_1, 0, sh_2);
+ }
+
+ return final_cmp (w1, w2);
+}
+
/* Compare S1 and S2, returning less than, equal to or
greater than zero if S1 is lexicographically less than,
equal to or greater than S2. */
int
-STRCMP (const char *p1, const char *p2)
+strcmp (const char *p1, const char *p2)
{
- const unsigned char *s1 = (const unsigned char *) p1;
- const unsigned char *s2 = (const unsigned char *) p2;
- unsigned char c1, c2;
-
- do
+ /* Handle the unaligned bytes of p1 first. */
+ uintptr_t n = -(uintptr_t)p1 % sizeof(op_t);
+ for (int i = 0; i < n; ++i)
{
- c1 = (unsigned char) *s1++;
- c2 = (unsigned char) *s2++;
- if (c1 == '\0')
- return c1 - c2;
+ unsigned char c1 = *p1++;
+ unsigned char c2 = *p2++;
+ int diff = c1 - c2;
+ if (c1 == '\0' || diff != 0)
+ return diff;
}
- while (c1 == c2);
- return c1 - c2;
+ /* P1 is now aligned to op_t. P2 may or may not be. */
+ const op_t *x1 = (const op_t *) p1;
+ op_t w1 = *x1++;
+ uintptr_t ofs = (uintptr_t) p2 % sizeof(op_t);
+ return ofs == 0
+ ? strcmp_aligned_loop (x1, (const op_t *)p2, w1)
+ : strcmp_unaligned_loop (x1, (const op_t *)(p2 - ofs), w1, ofs);
}
+#ifndef STRCMP
libc_hidden_builtin_def (strcmp)
+#endif
--
2.34.1
next prev parent reply other threads:[~2023-01-17 20:00 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-17 19:59 [PATCH v9 00/22] Improve generic string routines Adhemerval Zanella
2023-01-17 19:59 ` [PATCH v9 01/22] Parameterize op_t from memcopy.h Adhemerval Zanella
2023-01-17 19:59 ` [PATCH v9 02/22] Parameterize OP_T_THRES " Adhemerval Zanella
2023-01-17 19:59 ` [PATCH v9 03/22] Add string vectorized find and detection functions Adhemerval Zanella
2023-01-18 1:35 ` Richard Henderson
2023-01-18 1:54 ` Richard Henderson
2023-01-18 12:42 ` Adhemerval Zanella Netto
2023-01-18 19:24 ` Richard Henderson
2023-01-18 19:38 ` Adhemerval Zanella Netto
2023-01-17 19:59 ` [PATCH v9 04/22] string: Improve generic strlen Adhemerval Zanella
2023-01-17 19:59 ` [PATCH v9 05/22] string: Improve generic strnlen Adhemerval Zanella
2023-01-17 19:59 ` [PATCH v9 06/22] string: Improve generic strchr Adhemerval Zanella
2023-01-18 1:42 ` Richard Henderson
2023-01-17 19:59 ` [PATCH v9 07/22] string: Improve generic strchrnul Adhemerval Zanella
2023-01-17 20:00 ` Adhemerval Zanella [this message]
2023-01-17 20:00 ` [PATCH v9 09/22] string: Improve generic memchr Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 10/22] string: Improve generic memrchr Adhemerval Zanella
2023-01-18 2:10 ` Richard Henderson
2023-01-17 20:00 ` [PATCH v9 11/22] hppa: Add memcopy.h Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 12/22] hppa: Add string-fzb.h and string-fzi.h Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 13/22] alpha: " Adhemerval Zanella
2023-01-18 1:58 ` Richard Henderson
2023-01-17 20:00 ` [PATCH v9 14/22] arm: Add string-fza.h Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 15/22] powerpc: " Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 16/22] sh: Add string-fzb.h Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 17/22] string: Hook up the default implementation on test-strlen Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 18/22] string: Hook up the default implementation on test-strnlen Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 19/22] string: Hook up the default implementation on test-strchr Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 20/22] string: Hook up the default implementation on test-strcmp Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 21/22] string: Hook up the default implementation on test-memchr Adhemerval Zanella
2023-01-17 20:00 ` [PATCH v9 22/22] string: Hook up the default implementation on test-memrchr Adhemerval Zanella
2023-01-18 18:51 ` Carlos O'Donell
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=20230117200014.1299923-9-adhemerval.zanella@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=goldstein.w.n@gmail.com \
--cc=libc-alpha@sourceware.org \
--cc=richard.henderson@linaro.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).