From 290ba1089ecdca306cb2a0fbf3c7dc9cc29bc56f Mon Sep 17 00:00:00 2001
From: Siddhesh Poyarekar <siddhesh@sourceware.org>
Date: Thu, 5 Oct 2017 21:55:31 +0530
Subject: [PATCH 52/90] aarch64: Optimized implementation of memmove for
 Qualcomm Falkor

This is an optimized memmove implementation for the Qualcomm Falkor
processor core.  Due to the way the falkor memcpy needs to be written,
code cannot be easily shared between memmove and memcpy like in case
of other aarch64 memcpy implementations due to which this routine is
separate.  The underlying principle is the same as that of memcpy
where it tries to use registers with the same lower 4 bits for
fetching the same stream, thus optimizing hardware prefetcher
performance.

The memcpy copy loop copies 64 bytes at a time using the same register
pair since that's the way to train the hardware prefetcher on the
falkor core.  memmove cannot quite do that since it needs to avoid
overlaps, so it does the next best thing, i.e. has a 32 byte loop with
a 32 byte end (prefetch a loop ahead to account for overlapping
locations) with register pairs that alias so that they hit the same
prefetcher.  Due to this difference in loop size, they have to
currently be separate implementations but efforts are on to try and
get memmove to fall back into memcpy whenever it can without simply
duplicating all of the code.

Performance:

The routine fares around 20-25% better than the generic memmove for
most medium to large sizes (i.e. > 128 bytes) for the new walking
memmove benchmark (memmove-walk) with an unexplained regression
between 1K and 2K.  The minor regression is something worth looking
into for us, but the remaining gains are significant enough that we
would like this included upstream as we looking into the cause for the
regression.  Here is a snippet of the numbers as generated from the
microbenchmark by the compare_strings script.  Comparisons are against
__memmove_generic:

Function: memmove
Variant: walk
                                    __memmove_thunderx	__memmove_falkor	__memmove_generic
========================================================================================================================
<snip>
                        length=16384:  12508800.00 (  6.09%)	 11486800.00 ( 13.76%)	 13319600.00
                        length=16400:  13614200.00 ( -0.67%)	 11585000.00 ( 14.33%)	 13523600.00
                        length=16385:  13448400.00 (  0.10%)	 11732700.00 ( 12.84%)	 13461200.00
                        length=16399:  13594100.00 ( -0.22%)	 11859600.00 ( 12.57%)	 13564400.00
                        length=16386:  13211600.00 (  1.13%)	 11503800.00 ( 13.91%)	 13362400.00
                        length=16398:  13218600.00 (  2.12%)	 11573200.00 ( 14.30%)	 13504700.00
                        length=16387:  13510900.00 ( -0.37%)	 11744200.00 ( 12.76%)	 13461300.00
                        length=16397:  13603700.00 ( -0.15%)	 11878200.00 ( 12.55%)	 13583200.00
                        length=16388:  13461700.00 ( -0.13%)	 11558000.00 ( 14.03%)	 13444100.00
                        length=16396:  13517500.00 ( -0.03%)	 11561300.00 ( 14.45%)	 13513900.00
                        length=16389:  13534100.00 (  0.17%)	 11756800.00 ( 13.28%)	 13556900.00
                        length=16395:  13585600.00 (  0.11%)	 11791800.00 ( 13.30%)	 13601200.00
                        length=16390:  13480100.00 ( -0.13%)	 11685500.00 ( 13.20%)	 13462100.00
                        length=16394:  13529900.00 ( -0.23%)	 11549800.00 ( 14.43%)	 13498200.00
                        length=16391:  13595400.00 ( -0.26%)	 11768200.00 ( 13.22%)	 13560600.00
                        length=16393:  13567000.00 (  0.20%)	 11779700.00 ( 13.35%)	 13594700.00
                        length=32768:  71308800.00 ( -6.53%)	 50220800.00 ( 24.98%)	 66939200.00
                        length=32784:  72100800.00 (-11.55%)	 50114100.00 ( 22.47%)	 64636300.00
                        length=32769:  71767000.00 ( -7.10%)	 51238400.00 ( 23.54%)	 67010000.00
                        length=32783:  70113700.00 (-40.95%)	 51129000.00 ( -2.78%)	 49744400.00
                        length=32770:  71367600.00 ( -6.52%)	 50244700.00 ( 25.01%)	 67000900.00
                        length=32782:  64366700.00 (  4.71%)	 50101400.00 ( 25.83%)	 67545600.00
                        length=32771:  71440100.00 ( -6.51%)	 51263900.00 ( 23.57%)	 67074900.00
                        length=32781:  66993000.00 (  0.34%)	 51108300.00 ( 23.97%)	 67220300.00
                        length=32772:  71443900.00 (-60.50%)	 50062100.00 (-12.47%)	 44512600.00
                        length=32780:  71759100.00 ( -6.58%)	 50263200.00 ( 25.35%)	 67328600.00
                        length=32773:  71714900.00 (-33.21%)	 51076600.00 (  5.12%)	 53835400.00
                        length=32779:  71756900.00 ( -6.56%)	 51290800.00 ( 23.83%)	 67337800.00
                        length=32774:  59689300.00 (-34.55%)	 50068400.00 (-12.86%)	 44363300.00
                        length=32778:  71847500.00 (-18.20%)	 50084100.00 ( 17.61%)	 60786500.00
                        length=32775:  71599300.00 ( -6.54%)	 51278200.00 ( 23.70%)	 67204800.00
                        length=32777:  71862900.00 (-60.85%)	 51094000.00 (-14.36%)	 44677900.00
                        length=65536: 282848000.00 ( -6.60%)	199187000.00 ( 24.93%)	265325000.00
                        length=65552: 243285000.00 (-41.61%)	198512000.00 (-15.54%)	171805000.00
                        length=65537: 255415000.00 (-23.47%)	202499000.00 (  2.11%)	206858000.00
                        length=65551: 280122000.00 (-62.95%)	203349000.00 (-18.29%)	171911000.00
                        length=65538: 283676000.00 (-14.46%)	198368000.00 ( 19.96%)	247848000.00
                        length=65550: 275566000.00 (-51.76%)	198494000.00 ( -9.31%)	181581000.00
                        length=65539: 283699000.00 ( -6.58%)	203453000.00 ( 23.57%)	266195000.00
                        length=65549: 286572000.00 ( -6.65%)	202607000.00 ( 24.60%)	268712000.00
                        length=65540: 283710000.00 ( -6.59%)	199161000.00 ( 25.17%)	266160000.00
                        length=65548: 237573000.00 ( 11.48%)	198462000.00 ( 26.06%)	268395000.00
                        length=65541: 284150000.00 ( -6.58%)	203273000.00 ( 23.75%)	266600000.00
                        length=65547: 286250000.00 ( -6.70%)	202594000.00 ( 24.48%)	268263000.00
                        length=65542: 284167000.00 ( -6.60%)	199122000.00 ( 25.31%)	266584000.00
                        length=65546: 285656000.00 ( -6.59%)	198443000.00 ( 25.95%)	268002000.00
                        length=65543: 284600000.00 ( -6.58%)	203247000.00 ( 23.89%)	267030000.00
                        length=65545: 285665000.00 ( -6.40%)	202575000.00 ( 24.55%)	268472000.00
<snip>

	* sysdeps/aarch64/multiarch/Makefile (sysdep_routines): Add
	memmove_falkor.
	* sysdeps/aarch64/multiarch/ifunc-impl-list.c
	(__libc_ifunc_impl_list): Likewise.
	* sysdeps/aarch64/multiarch/memmove.c: Likewise.
	* sysdeps/aarch64/multiarch/memmove_falkor.S: New file.

diff --git a/ChangeLog b/ChangeLog
index 2aa1711834..d478f5d8e4 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,21 @@
 2017-10-10  Siddhesh Poyarekar  <siddhesh@sourceware.org>
 
+	* sysdeps/aarch64/multiarch/Makefile (sysdep_routines): Add
+	memmove_falkor.
+	* sysdeps/aarch64/multiarch/ifunc-impl-list.c
+	(__libc_ifunc_impl_list): Likewise.
+	* sysdeps/aarch64/multiarch/memmove.c: Likewise.
+	* sysdeps/aarch64/multiarch/memmove_falkor.S: New file.
+
+	* benchtests/bench-memmove-walk.c: New file.
+	* benchtests/Makefile (string-benchset): Add it.
+
+	* benchtests/bench-memset-walk.c: New file.
+	* benchtests/Makefile (string-benchset): Add it.
+
+	* benchtests/bench-memcpy-walk.c: New file.
+	* benchtests/Makefile (string-benchset): Add it.
+
 	* po/sv.po: Update translations.
 	* po/fr.po: Likewise.
 
diff --git a/sysdeps/aarch64/multiarch/Makefile b/sysdeps/aarch64/multiarch/Makefile
index 164ba1ae4b..9aa1e79a80 100644
--- a/sysdeps/aarch64/multiarch/Makefile
+++ b/sysdeps/aarch64/multiarch/Makefile
@@ -1,3 +1,4 @@
 ifeq ($(subdir),string)
-sysdep_routines += memcpy_generic memcpy_thunderx memcpy_falkor
+sysdep_routines += memcpy_generic memcpy_thunderx memcpy_falkor \
+		   memmove_falkor
 endif
diff --git a/sysdeps/aarch64/multiarch/ifunc-impl-list.c b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
index 8e873b3d9f..2cb74d5b43 100644
--- a/sysdeps/aarch64/multiarch/ifunc-impl-list.c
+++ b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
@@ -44,6 +44,7 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
 	      IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_generic))
   IFUNC_IMPL (i, name, memmove,
 	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_thunderx)
+	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_falkor)
 	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_generic))
 
   return i;
diff --git a/sysdeps/aarch64/multiarch/memmove.c b/sysdeps/aarch64/multiarch/memmove.c
index 34c6b29bd5..016f03ee50 100644
--- a/sysdeps/aarch64/multiarch/memmove.c
+++ b/sysdeps/aarch64/multiarch/memmove.c
@@ -30,9 +30,14 @@ extern __typeof (__redirect_memmove) __libc_memmove;
 
 extern __typeof (__redirect_memmove) __memmove_generic attribute_hidden;
 extern __typeof (__redirect_memmove) __memmove_thunderx attribute_hidden;
+extern __typeof (__redirect_memmove) __memmove_falkor attribute_hidden;
 
 libc_ifunc (__libc_memmove,
-            IS_THUNDERX (midr) ? __memmove_thunderx : __memmove_generic);
+            (IS_THUNDERX (midr)
+	     ? __memmove_thunderx
+	     : (IS_FALKOR (midr)
+		? __memmove_falkor
+		: __memmove_generic)));
 
 # undef memmove
 strong_alias (__libc_memmove, memmove);
diff --git a/sysdeps/aarch64/multiarch/memmove_falkor.S b/sysdeps/aarch64/multiarch/memmove_falkor.S
new file mode 100644
index 0000000000..3a4e6a2a8e
--- /dev/null
+++ b/sysdeps/aarch64/multiarch/memmove_falkor.S
@@ -0,0 +1,232 @@
+/* Copyright (C) 2017 Free Software Foundation, Inc.
+
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library.  If not, see
+   <http://www.gnu.org/licenses/>.  */
+
+#include <sysdep.h>
+
+/* Assumptions: ARMv8-a, AArch64, falkor, unaligned accesses.  */
+
+#define dstin	x0
+#define src	x1
+#define count	x2
+#define dstlen	x3
+#define dst	x3
+#define srcend	x4
+#define dstend	x5
+#define A_l	x6
+#define A_lw	w6
+#define A_h	x7
+#define A_hw	w7
+#define B_l	x8
+#define B_lw	w8
+#define B_h	x9
+#define C_l	x10
+#define C_h	x11
+#define D_l	x12
+#define D_h	x13
+#define E_l	src
+#define E_h	count
+#define F_l	srcend
+#define F_h	dst
+#define tmp1	x14
+
+/* Alias with A_l and A_h to train the prefetcher.  */
+#define Q_l	x22
+#define Q_h	x23
+
+/* RATIONALE:
+
+   The copy has 4 distinct parts:
+   * Small copies of 16 bytes and under
+   * Medium sized copies of 17-96 bytes
+   * Large copies where the source address is higher than the destination
+     (forward copies)
+   * Large copies where the destination address is higher than the source
+     (copy backward, or move).
+
+   We use only two registerpairs x6,x7 and x22,x23 for the copies and copy 32
+   bytes at a time to correctly train the hardware prefetcher for better
+   throughput.  */
+ENTRY_ALIGN (__memmove_falkor, 6)
+
+	sub	tmp1, dstin, src
+	add	srcend, src, count
+	add	dstend, dstin, count
+	cmp	count, 96
+	ccmp	tmp1, count, 2, hi
+	b.lo	L(move_long)
+
+	cmp	count, 16
+	b.ls	L(copy16)
+	cmp	count, 96
+	b.hi	L(copy_long)
+
+	/* Medium copies: 17..96 bytes.  */
+	sub	tmp1, count, 1
+	ldp	A_l, A_h, [src]
+	tbnz	tmp1, 6, L(copy96)
+	ldp	D_l, D_h, [srcend, -16]
+	tbz	tmp1, 5, 1f
+	ldp	B_l, B_h, [src, 16]
+	ldp	C_l, C_h, [srcend, -32]
+	stp	B_l, B_h, [dstin, 16]
+	stp	C_l, C_h, [dstend, -32]
+1:
+	stp	A_l, A_h, [dstin]
+	stp	D_l, D_h, [dstend, -16]
+	ret
+
+	.p2align 4
+	/* Small copies: 0..16 bytes.  */
+L(copy16):
+	cmp	count, 8
+	b.lo	1f
+	ldr	A_l, [src]
+	ldr	A_h, [srcend, -8]
+	str	A_l, [dstin]
+	str	A_h, [dstend, -8]
+	ret
+	.p2align 4
+1:
+	/* 4-7 */
+	tbz	count, 2, 1f
+	ldr	A_lw, [src]
+	ldr	A_hw, [srcend, -4]
+	str	A_lw, [dstin]
+	str	A_hw, [dstend, -4]
+	ret
+	.p2align 4
+1:
+	/* 2-3 */
+	tbz	count, 1, 1f
+	ldrh	A_lw, [src]
+	ldrh	A_hw, [srcend, -2]
+	strh	A_lw, [dstin]
+	strh	A_hw, [dstend, -2]
+	ret
+	.p2align 4
+1:
+	/* 0-1 */
+	tbz	count, 0, 1f
+	ldrb	A_lw, [src]
+	strb	A_lw, [dstin]
+1:	ret
+
+	.p2align 4
+	/* Copy 64..96 bytes.  Copy 64 bytes from the start and
+	   32 bytes from the end.  */
+L(copy96):
+	ldp	B_l, B_h, [src, 16]
+	ldp	C_l, C_h, [src, 32]
+	ldp	D_l, D_h, [src, 48]
+	ldp	E_l, E_h, [srcend, -32]
+	ldp	F_l, F_h, [srcend, -16]
+	stp	A_l, A_h, [dstin]
+	stp	B_l, B_h, [dstin, 16]
+	stp	C_l, C_h, [dstin, 32]
+	stp	D_l, D_h, [dstin, 48]
+	stp	E_l, E_h, [dstend, -32]
+	stp	F_l, F_h, [dstend, -16]
+	ret
+
+	/* Align SRC to 16 byte alignment so that we don't cross cache line
+	   boundaries on both loads and stores.  There are at least 96 bytes
+	   to copy, so copy 16 bytes unaligned and then align.  The loop
+	   copies 32 bytes per iteration and prefetches one iteration ahead.  */
+
+	.p2align 4
+L(copy_long):
+	sub	count, count, 64 + 16	/* Test and readjust count.  */
+	mov	B_l, Q_l
+	mov	B_h, Q_h
+	ldp	A_l, A_h, [src]
+	and	tmp1, src, 15
+	bic	src, src, 15
+	sub	dst, dstin, tmp1
+	add	count, count, tmp1	/* Count is now 16 too large.  */
+	ldp	Q_l, Q_h, [src, 16]!
+	stp	A_l, A_h, [dstin]
+	ldp	A_l, A_h, [src, 16]!
+
+L(loop64):
+	subs	count, count, 32
+	stp	Q_l, Q_h, [dst, 16]
+	ldp	Q_l, Q_h, [src, 16]!
+	stp	A_l, A_h, [dst, 32]!
+	ldp	A_l, A_h, [src, 16]!
+	b.hi	L(loop64)
+
+	/* Write the last full set of 32 bytes.  The remainder is at most 32
+	   bytes, so it is safe to always copy 32 bytes from the end even if
+	   there is just 1 byte left.  */
+L(last64):
+	ldp	C_l, C_h, [srcend, -32]
+	stp	Q_l, Q_h, [dst, 16]
+	ldp	Q_l, Q_h, [srcend, -16]
+	stp	A_l, A_h, [dst, 32]
+	stp	C_l, C_h, [dstend, -32]
+	stp	Q_l, Q_h, [dstend, -16]
+	mov	Q_l, B_l
+	mov	Q_h, B_h
+	ret
+
+	.p2align 4
+L(move_long):
+	cbz	tmp1, 3f
+
+	mov	B_l, Q_l
+	mov	B_h, Q_h
+
+	/* Align SRCEND to 16 byte alignment so that we don't cross cache line
+	   boundaries on both loads and stores.  There are at least 96 bytes
+	   to copy, so copy 16 bytes unaligned and then align.  The loop
+	   copies 32 bytes per iteration and prefetches one iteration ahead.  */
+
+	ldp	A_l, A_h, [srcend, -16]
+	and	tmp1, srcend, 15
+	sub	srcend, srcend, tmp1
+	ldp	Q_l, Q_h, [srcend, -16]!
+	stp	A_l, A_h, [dstend, -16]
+	sub	count, count, tmp1
+	ldp	A_l, A_h, [srcend, -16]!
+	sub	dstend, dstend, tmp1
+	sub	count, count, 64
+
+1:
+	subs	count, count, 32
+	stp	Q_l, Q_h, [dstend, -16]
+	ldp	Q_l, Q_h, [srcend, -16]!
+	stp	A_l, A_h, [dstend, -32]!
+	ldp	A_l, A_h, [srcend, -16]!
+	b.hi	1b
+
+	/* Write the last full set of 32 bytes.  The remainder is at most 32
+	   bytes, so it is safe to always copy 32 bytes from the start even if
+	   there is just 1 byte left.  */
+2:
+	ldp	C_l, C_h, [src, 16]
+	stp	Q_l, Q_h, [dstend, -16]
+	ldp	Q_l, Q_h, [src]
+	stp	A_l, A_h, [dstend, -32]
+	stp	C_l, C_h, [dstin, 16]
+	stp	Q_l, Q_h, [dstin]
+	mov	Q_l, B_l
+	mov	Q_h, B_h
+3:	ret
+
+END (__memmove_falkor)
+libc_hidden_builtin_def (__memmove_falkor)