This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
[PATCH] x86-64: Skip zero length in __mem[pcpy|move|set]_erms
- From: "H.J. Lu" <hongjiu dot lu at intel dot com>
- To: GNU C Library <libc-alpha at sourceware dot org>
- Date: Wed, 23 May 2018 10:28:53 -0700
- Subject: [PATCH] x86-64: Skip zero length in __mem[pcpy|move|set]_erms
- Reply-to: "H.J. Lu" <hjl dot tools at gmail dot com>
This patch skips zero length in __mempcpy_erms, __memmove_erms and
__memset_erms.
Tested on x86-64. I will check it in shortly.
H.J.
---
* sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
(__mempcpy_erms): Skip zero length.
(__memmove_erms): Likewise.
* sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
(__memset_erms): Likewise.
---
sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S | 6 ++++++
sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S | 5 +++++
2 files changed, 11 insertions(+)
diff --git a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
index 1404131bb7..e2ede45e9f 100644
--- a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
+++ b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
@@ -156,6 +156,9 @@ END (__mempcpy_chk_erms)
/* Only used to measure performance of REP MOVSB. */
ENTRY (__mempcpy_erms)
movq %rdi, %rax
+ /* Skip zero length. */
+ testq %rdx, %rdx
+ jz 2f
addq %rdx, %rax
jmp L(start_movsb)
END (__mempcpy_erms)
@@ -167,6 +170,9 @@ END (__memmove_chk_erms)
ENTRY (__memmove_erms)
movq %rdi, %rax
+ /* Skip zero length. */
+ testq %rdx, %rdx
+ jz 2f
L(start_movsb):
movq %rdx, %rcx
cmpq %rsi, %rdi
diff --git a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
index 0a827f3e47..dc9cb88b37 100644
--- a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
+++ b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S
@@ -128,6 +128,11 @@ END (__memset_chk_erms)
/* Only used to measure performance of REP STOSB. */
ENTRY (__memset_erms)
+ /* Skip zero length. */
+ testq %rdx, %rdx
+ jnz L(stosb)
+ movq %rdi, %rax
+ ret
# else
/* Provide a hidden symbol to debugger. */
.hidden MEMSET_SYMBOL (__memset, erms)
--
2.17.0