[PATCH] AArch64: Add optimized Q-register memcpy

Szabolcs Nagy szabolcs.nagy@arm.com
Wed Jul 15 08:16:08 GMT 2020


The 07/14/2020 16:17, Carlos O'Donell wrote:
> On 7/14/20 12:33 PM, Wilco Dijkstra wrote:
> > Add a new memcpy using 128-bit Q registers - this is faster on modern
> > cores and reduces codesize.  Similar to the generic memcpy, small cases
> > include copies up to 32 bytes.  64-128 byte copies are split into two
> > cases to improve performance of 64-96 byte copies.  Large copies align
> > the source rather than the destination.
> > 
> > bench-memcpy-random is ~9% faster than memcpy_falkor on Neoverse N1,
> > so make this memcpy the default on N1 (on Centriq it is 15% faster than
> > memcpy_falkor).
> > 
> > Passes GLIBC regression tests. OK for commit?
> 
> As release manager this is OK for 2.32 if Szabolcs says it's OK.

thanks.
this is ok to commit.

> > ---
> > 
> > diff --git a/sysdeps/aarch64/multiarch/Makefile b/sysdeps/aarch64/multiarch/Makefile
> > index 4377df0735287c210efd661188f9e6e3923c8003..e93c21e764a8d02b9f07f5030c31836a3f03f3e1 100644
> > --- a/sysdeps/aarch64/multiarch/Makefile
> > +++ b/sysdeps/aarch64/multiarch/Makefile
> > @@ -1,5 +1,5 @@
> >  ifeq ($(subdir),string)
> > -sysdep_routines += memcpy_generic memcpy_thunderx memcpy_thunderx2 \
> > +sysdep_routines += memcpy_generic memcpy_advsimd memcpy_thunderx memcpy_thunderx2 \
> >  		   memcpy_falkor \
> >  		   memcpy_new \
> >  		   memset_generic memset_falkor memset_emag memset_kunpeng \
> > diff --git a/sysdeps/aarch64/multiarch/ifunc-impl-list.c b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
> > index 0ccaf53e555e410569eb2be76ec7d5b4d7bc64a5..09feea97ea37ab923cf4a8557197d46adcd49204 100644
> > --- a/sysdeps/aarch64/multiarch/ifunc-impl-list.c
> > +++ b/sysdeps/aarch64/multiarch/ifunc-impl-list.c
> > @@ -42,11 +42,13 @@ __libc_ifunc_impl_list (const char *name, struct libc_ifunc_impl *array,
> >  	      IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_thunderx)
> >  	      IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_thunderx2)
> >  	      IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_falkor)
> > +	      IFUNC_IMPL_ADD (array, i, memcpy, 1, __memcpy_simd)
> >  	      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_thunderx2)
> >  	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_falkor)
> > +	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_simd)
> >  	      IFUNC_IMPL_ADD (array, i, memmove, 1, __memmove_generic))
> >    IFUNC_IMPL (i, name, memset,
> >  	      /* Enable this on non-falkor processors too so that other cores
> > diff --git a/sysdeps/aarch64/multiarch/memcpy.c b/sysdeps/aarch64/multiarch/memcpy.c
> > index 2fafefd5d23fc1528031b5fe52098218ed603b89..e6f3ae116701097d71a02e2a1f6bfdadc1eec34a 100644
> > --- a/sysdeps/aarch64/multiarch/memcpy.c
> > +++ b/sysdeps/aarch64/multiarch/memcpy.c
> > @@ -29,6 +29,7 @@
> >  extern __typeof (__redirect_memcpy) __libc_memcpy;
> >  
> >  extern __typeof (__redirect_memcpy) __memcpy_generic attribute_hidden;
> > +extern __typeof (__redirect_memcpy) __memcpy_simd attribute_hidden;
> >  extern __typeof (__redirect_memcpy) __memcpy_thunderx attribute_hidden;
> >  extern __typeof (__redirect_memcpy) __memcpy_thunderx2 attribute_hidden;
> >  extern __typeof (__redirect_memcpy) __memcpy_falkor attribute_hidden;
> > @@ -36,11 +37,11 @@ extern __typeof (__redirect_memcpy) __memcpy_falkor attribute_hidden;
> >  libc_ifunc (__libc_memcpy,
> >              (IS_THUNDERX (midr)
> >  	     ? __memcpy_thunderx
> > -	     : (IS_FALKOR (midr) || IS_PHECDA (midr) || IS_ARES (midr) || IS_KUNPENG920 (midr)
> > +	     : (IS_FALKOR (midr) || IS_PHECDA (midr) || IS_KUNPENG920 (midr)
> >  		? __memcpy_falkor
> >  		: (IS_THUNDERX2 (midr) || IS_THUNDERX2PA (midr)
> >  		  ? __memcpy_thunderx2
> > -		  : __memcpy_generic))));
> > +		  : (IS_ARES (midr) ? __memcpy_simd : __memcpy_generic)))));
> >  
> >  # undef memcpy
> >  strong_alias (__libc_memcpy, memcpy);
> > diff --git a/sysdeps/aarch64/multiarch/memcpy_advsimd.S b/sysdeps/aarch64/multiarch/memcpy_advsimd.S
> > new file mode 100644
> > index 0000000000000000000000000000000000000000..d4ba74777744c8bb5a83e43ab2d63ad8dab35203
> > --- /dev/null
> > +++ b/sysdeps/aarch64/multiarch/memcpy_advsimd.S
> > @@ -0,0 +1,247 @@
> > +/* Generic optimized memcpy using SIMD.
> > +   Copyright (C) 2020 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
> > +   <https://www.gnu.org/licenses/>.  */
> > +
> > +#include <sysdep.h>
> > +
> > +/* Assumptions:
> > + *
> > + * ARMv8-a, AArch64, Advanced SIMD, unaligned accesses.
> > + *
> > + */
> > +
> > +#define dstin	x0
> > +#define src	x1
> > +#define count	x2
> > +#define dst	x3
> > +#define srcend	x4
> > +#define dstend	x5
> > +#define A_l	x6
> > +#define A_lw	w6
> > +#define A_h	x7
> > +#define B_l	x8
> > +#define B_lw	w8
> > +#define B_h	x9
> > +#define C_lw	w10
> > +#define tmp1	x14
> > +
> > +#define A_q	q0
> > +#define B_q	q1
> > +#define C_q	q2
> > +#define D_q	q3
> > +#define E_q	q4
> > +#define F_q	q5
> > +#define G_q	q6
> > +#define H_q	q7
> > +
> > +
> > +/* This implementation supports both memcpy and memmove and shares most code.
> > +   It uses unaligned accesses and branchless sequences to keep the code small,
> > +   simple and improve performance.
> > +
> > +   Copies are split into 3 main cases: small copies of up to 32 bytes, medium
> > +   copies of up to 128 bytes, and large copies.  The overhead of the overlap
> > +   check in memmove is negligible since it is only required for large copies.
> > +
> > +   Large copies use a software pipelined loop processing 64 bytes per
> > +   iteration.  The destination pointer is 16-byte aligned to minimize
> > +   unaligned accesses.  The loop tail is handled by always copying 64 bytes
> > +   from the end.  */
> > +
> > +ENTRY (__memcpy_simd)
> > +	DELOUSE (0)
> > +	DELOUSE (1)
> > +	DELOUSE (2)
> > +
> > +	add	srcend, src, count
> > +	add	dstend, dstin, count
> > +	cmp	count, 128
> > +	b.hi	L(copy_long)
> > +	cmp	count, 32
> > +	b.hi	L(copy32_128)
> > +
> > +	/* Small copies: 0..32 bytes.  */
> > +	cmp	count, 16
> > +	b.lo	L(copy16)
> > +	ldr	A_q, [src]
> > +	ldr	B_q, [srcend, -16]
> > +	str	A_q, [dstin]
> > +	str	B_q, [dstend, -16]
> > +	ret
> > +
> > +	/* Copy 8-15 bytes.  */
> > +L(copy16):
> > +	tbz	count, 3, L(copy8)
> > +	ldr	A_l, [src]
> > +	ldr	A_h, [srcend, -8]
> > +	str	A_l, [dstin]
> > +	str	A_h, [dstend, -8]
> > +	ret
> > +
> > +	/* Copy 4-7 bytes.  */
> > +L(copy8):
> > +	tbz	count, 2, L(copy4)
> > +	ldr	A_lw, [src]
> > +	ldr	B_lw, [srcend, -4]
> > +	str	A_lw, [dstin]
> > +	str	B_lw, [dstend, -4]
> > +	ret
> > +
> > +	/* Copy 0..3 bytes using a branchless sequence.  */
> > +L(copy4):
> > +	cbz	count, L(copy0)
> > +	lsr	tmp1, count, 1
> > +	ldrb	A_lw, [src]
> > +	ldrb	C_lw, [srcend, -1]
> > +	ldrb	B_lw, [src, tmp1]
> > +	strb	A_lw, [dstin]
> > +	strb	B_lw, [dstin, tmp1]
> > +	strb	C_lw, [dstend, -1]
> > +L(copy0):
> > +	ret
> > +
> > +	.p2align 4
> > +	/* Medium copies: 33..128 bytes.  */
> > +L(copy32_128):
> > +	ldp	A_q, B_q, [src]
> > +	ldp	C_q, D_q, [srcend, -32]
> > +	cmp	count, 64
> > +	b.hi	L(copy128)
> > +	stp	A_q, B_q, [dstin]
> > +	stp	C_q, D_q, [dstend, -32]
> > +	ret
> > +
> > +	.p2align 4
> > +	/* Copy 65..128 bytes.  */
> > +L(copy128):
> > +	ldp	E_q, F_q, [src, 32]
> > +	cmp	count, 96
> > +	b.ls	L(copy96)
> > +	ldp	G_q, H_q, [srcend, -64]
> > +	stp	G_q, H_q, [dstend, -64]
> > +L(copy96):
> > +	stp	A_q, B_q, [dstin]
> > +	stp	E_q, F_q, [dstin, 32]
> > +	stp	C_q, D_q, [dstend, -32]
> > +	ret
> > +
> > +	/* Align loop64 below to 16 bytes.  */
> > +	nop
> > +
> > +	/* Copy more than 128 bytes.  */
> > +L(copy_long):
> > +	/* Copy 16 bytes and then align src to 16-byte alignment.  */
> > +	ldr	D_q, [src]
> > +	and	tmp1, src, 15
> > +	bic	src, src, 15
> > +	sub	dst, dstin, tmp1
> > +	add	count, count, tmp1	/* Count is now 16 too large.  */
> > +	ldp	A_q, B_q, [src, 16]
> > +	str	D_q, [dstin]
> > +	ldp	C_q, D_q, [src, 48]
> > +	subs	count, count, 128 + 16	/* Test and readjust count.  */
> > +	b.ls	L(copy64_from_end)
> > +L(loop64):
> > +	stp	A_q, B_q, [dst, 16]
> > +	ldp	A_q, B_q, [src, 80]
> > +	stp	C_q, D_q, [dst, 48]
> > +	ldp	C_q, D_q, [src, 112]
> > +	add	src, src, 64
> > +	add	dst, dst, 64
> > +	subs	count, count, 64
> > +	b.hi	L(loop64)
> > +
> > +	/* Write the last iteration and copy 64 bytes from the end.  */
> > +L(copy64_from_end):
> > +	ldp	E_q, F_q, [srcend, -64]
> > +	stp	A_q, B_q, [dst, 16]
> > +	ldp	A_q, B_q, [srcend, -32]
> > +	stp	C_q, D_q, [dst, 48]
> > +	stp	E_q, F_q, [dstend, -64]
> > +	stp	A_q, B_q, [dstend, -32]
> > +	ret
> > +
> > +END (__memcpy_simd)
> > +libc_hidden_builtin_def (__memcpy_simd)
> > +
> > +
> > +ENTRY (__memmove_simd)
> > +	DELOUSE (0)
> > +	DELOUSE (1)
> > +	DELOUSE (2)
> > +
> > +	add	srcend, src, count
> > +	add	dstend, dstin, count
> > +	cmp	count, 128
> > +	b.hi	L(move_long)
> > +	cmp	count, 32
> > +	b.hi	L(copy32_128)
> > +
> > +	/* Small moves: 0..32 bytes.  */
> > +	cmp	count, 16
> > +	b.lo	L(copy16)
> > +	ldr	A_q, [src]
> > +	ldr	B_q, [srcend, -16]
> > +	str	A_q, [dstin]
> > +	str	B_q, [dstend, -16]
> > +	ret
> > +
> > +L(move_long):
> > +	/* Only use backward copy if there is an overlap.  */
> > +	sub	tmp1, dstin, src
> > +	cbz	tmp1, L(move0)
> > +	cmp	tmp1, count
> > +	b.hs	L(copy_long)
> > +
> > +	/* Large backwards copy for overlapping copies.
> > +	   Copy 16 bytes and then align srcend to 16-byte alignment.  */
> > +L(copy_long_backwards):
> > +	ldr	D_q, [srcend, -16]
> > +	and	tmp1, srcend, 15
> > +	bic	srcend, srcend, 15
> > +	sub	count, count, tmp1
> > +	ldp	A_q, B_q, [srcend, -32]
> > +	str	D_q, [dstend, -16]
> > +	ldp	C_q, D_q, [srcend, -64]
> > +	sub	dstend, dstend, tmp1
> > +	subs	count, count, 128
> > +	b.ls	L(copy64_from_start)
> > +
> > +L(loop64_backwards):
> > +	stp	A_q, B_q, [dstend, -32]
> > +	ldp	A_q, B_q, [srcend, -96]
> > +	stp	C_q, D_q, [dstend, -64]
> > +	ldp	C_q, D_q, [srcend, -128]
> > +	sub	srcend, srcend, 64
> > +	sub	dstend, dstend, 64
> > +	subs	count, count, 64
> > +	b.hi	L(loop64_backwards)
> > +
> > +	/* Write the last iteration and copy 64 bytes from the start.  */
> > +L(copy64_from_start):
> > +	ldp	E_q, F_q, [src, 32]
> > +	stp	A_q, B_q, [dstend, -32]
> > +	ldp	A_q, B_q, [src]
> > +	stp	C_q, D_q, [dstend, -64]
> > +	stp	E_q, F_q, [dstin, 32]
> > +	stp	A_q, B_q, [dstin]
> > +L(move0):
> > +	ret
> > +
> > +END (__memmove_simd)
> > +libc_hidden_builtin_def (__memmove_simd)
> > diff --git a/sysdeps/aarch64/multiarch/memmove.c b/sysdeps/aarch64/multiarch/memmove.c
> > index ed5a47f6f83e7b0afcec60cb9fa0f09999eaacae..1229f8b89296eddd2e711490bb7fc0b35726b6f5 100644
> > --- a/sysdeps/aarch64/multiarch/memmove.c
> > +++ b/sysdeps/aarch64/multiarch/memmove.c
> > @@ -29,6 +29,7 @@
> >  extern __typeof (__redirect_memmove) __libc_memmove;
> >  
> >  extern __typeof (__redirect_memmove) __memmove_generic attribute_hidden;
> > +extern __typeof (__redirect_memmove) __memmove_simd attribute_hidden;
> >  extern __typeof (__redirect_memmove) __memmove_thunderx attribute_hidden;
> >  extern __typeof (__redirect_memmove) __memmove_thunderx2 attribute_hidden;
> >  extern __typeof (__redirect_memmove) __memmove_falkor attribute_hidden;
> > @@ -40,7 +41,7 @@ libc_ifunc (__libc_memmove,
> >  		? __memmove_falkor
> >  		: (IS_THUNDERX2 (midr) || IS_THUNDERX2PA (midr)
> >  		  ? __memmove_thunderx2
> > -		  : __memmove_generic))));
> > +		  : (IS_ARES (midr) ? __memmove_simd : __memmove_generic)))));
> >  
> >  # undef memmove
> >  strong_alias (__libc_memmove, memmove);
> > 
> > 
> 
> 
> -- 
> Cheers,
> Carlos.
> 


More information about the Libc-alpha mailing list