From ff7e299914a2dfa1c3fd7abd267535f7586ac413 Mon Sep 17 00:00:00 2001
From: Wangyang Guo <wangyang.guo@intel.com>
Date: Mon, 15 Nov 2021 06:12:16 +0000
Subject: [PATCH 1/2] Avoid short distance rep movsb

---
 .../multiarch/memmove-vec-unaligned-erms.S    | 22 +++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
index c952576c..07299dc0 100644
--- a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
+++ b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S
@@ -56,6 +56,14 @@
 # endif
 #endif
 
+/* Avoid short distance rep movsb only with non-SSE vector.  */
+#ifndef AVOID_SHORT_DISTANCE_REP_MOVSB
+# define AVOID_SHORT_DISTANCE_REP_MOVSB 1
+#else
+# define AVOID_SHORT_DISTANCE_REP_MOVSB 0
+#endif
+
+
 #ifndef PREFETCH
 # define PREFETCH(addr) prefetcht0 addr
 #endif
@@ -235,7 +243,21 @@ L(movsb):
 	cmpq	%r9, %rdi
 	/* Avoid slow backward REP MOVSB.  */
 	jb	L(more_8x_vec_backward)
+# if AVOID_SHORT_DISTANCE_REP_MOVSB
+	movq	%rdi, %rcx
+	subq	%rsi, %rcx
+	jmp	2f
+# endif
 1:
+# if AVOID_SHORT_DISTANCE_REP_MOVSB
+	movq	%rsi, %rcx
+	subq	%rdi, %rcx
+2:
+/* Avoid "rep movsb" if RCX, the distance between source and destination,
+   is N*4GB + [1..63] with N >= 0.  */
+	cmpl	$63, %ecx
+	jbe	L(more_2x_vec)	/* Avoid "rep movsb" if ECX <= 63.  */
+# endif
 	movq	%rdx, %rcx
 	rep movsb
 L(nop):
-- 
2.27.0