Add Niagara-4 optimized memcpy and mempcpy.

	* sysdeps/sparc/sparc64/multiarch/memcpy-niagara4.S: New file.
	* sysdeps/sparc/sparc32/sparcv9/multiarch/memcpy-niagara4.S: New
	file.
	* sysdeps/sparc/sparc64/multiarch/Makefile: Add to
	sysdep_routines.
	* sysdeps/sparc/sparc32/sparcv9/multiarch/Makefile: Likewise.
	* sysdeps/sparc/sparc64/multiarch/memcpy.S: Use Niagara-4 memcpy
	and mempcpy when HWCAP_SPARC_CRYPTO is set.

3 files changed, 354 insertions, 3 deletions

diff --git a/sysdeps/sparc/sparc64/multiarch/Makefile b/sysdeps/sparc/sparc64/multiarch/Makefile
index 4d45042a95..377dfc7330 100644
--- a/sysdeps/sparc/sparc64/multiarch/Makefile
+++ b/sysdeps/sparc/sparc64/multiarch/Makefile
@@ -1,4 +1,4 @@
 ifeq ($(subdir),string)
 sysdep_routines += memcpy-ultra3 memcpy-niagara1 memcpy-niagara2 \
-		   memset-niagara1
+		   memset-niagara1 memcpy-niagara4
 endif
diff --git a/sysdeps/sparc/sparc64/multiarch/memcpy-niagara4.S b/sysdeps/sparc/sparc64/multiarch/memcpy-niagara4.S
new file mode 100644
index 0000000000..7bb05b9818
--- /dev/null
+++ b/sysdeps/sparc/sparc64/multiarch/memcpy-niagara4.S
@@ -0,0 +1,329 @@
+/* Copy SIZE bytes from SRC to DEST.  For SUN4V Niagara-4.
+   Copyright (C) 2012 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by David S. Miller (davem@davemloft.net)
+
+   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>
+
+#define ASI_BLK_INIT_QUAD_LDD_P	0xe2
+
+#define FPRS_FEF		0x04
+
+/* On T4 it is very expensive to access ASRs like %fprs and
+ * %asi, avoiding a read or a write can save ~50 cycles.
+ */
+#define FPU_ENTER			\
+	rd	%fprs, %o5;		\
+	andcc	%o5, FPRS_FEF, %g0;	\
+	be,a,pn	%icc, 999f;		\
+	 wr	%g0, FPRS_FEF, %fprs;	\
+	999:
+
+#define VISEntryHalf FPU_ENTER
+#define VISExitHalf and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs
+
+#define GLOBAL_SPARE	%g5
+
+#define STORE_ASI	ASI_BLK_INIT_QUAD_LDD_P
+#define EX_LD(x)	x
+#define EX_ST(x)	x
+#define EX_RETVAL(x)	x
+#define LOAD(type,addr,dest)	type [addr], dest
+#define STORE(type,src,addr)	type src, [addr]
+#define STORE_INIT(src,addr)	stxa src, [addr] STORE_ASI
+
+#if !defined NOT_IN_libc
+
+	.register	%g2,#scratch
+	.register	%g3,#scratch
+	.register	%g6,#scratch
+
+	.text
+
+ENTRY(__mempcpy_niagara4)
+	ba,pt		%icc, 101f
+	 add		%o0, %o2, %o3
+END(__mempcpy_niagara4)
+
+	.align		32
+ENTRY(__memcpy_niagara4)
+100:	/* %o0=dst, %o1=src, %o2=len */
+	mov		%o0, %o3
+101:
+	brz,pn		%o2, .Lexit
+	 cmp		%o2, 3
+	ble,pn		%icc, .Ltiny
+	 cmp		%o2, 19
+	ble,pn		%icc, .Lsmall
+	 or		%o0, %o1, %g2
+	cmp		%o2, 128
+	bl,pn		%icc, .Lmedium
+	 nop
+
+.Llarge:/* len >= 0x80 */
+	/* First get dest 8 byte aligned.  */
+	sub		%g0, %o0, %g1
+	and		%g1, 0x7, %g1
+	brz,pt		%g1, 51f
+	 sub		%o2, %g1, %o2
+
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+	add		%o1, 1, %o1
+	subcc		%g1, 1, %g1
+	add		%o0, 1, %o0
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stb, %g2, %o0 - 0x01))
+
+51:	LOAD(prefetch, %o1 + 0x040, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x080, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x0c0, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x100, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x140, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x180, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x1c0, #n_reads_strong)
+	LOAD(prefetch, %o1 + 0x200, #n_reads_strong)
+
+	/* Check if we can use the straight fully aligned
+	 * loop, or we require the alignaddr/faligndata variant.
+	 */
+	andcc		%o1, 0x7, %o5
+	bne,pn		%icc, .Llarge_src_unaligned
+	 sub		%g0, %o0, %g1
+
+	/* Legitimize the use of initializing stores by getting dest
+	 * to be 64-byte aligned.
+	 */
+	and		%g1, 0x3f, %g1
+	brz,pt		%g1, .Llarge_aligned
+	 sub		%o2, %g1, %o2
+
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g2))
+	add		%o1, 8, %o1
+	subcc		%g1, 8, %g1
+	add		%o0, 8, %o0
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stx, %g2, %o0 - 0x08))
+
+.Llarge_aligned:
+	/* len >= 0x80 && src 8-byte aligned && dest 8-byte aligned */
+	andn		%o2, 0x3f, %o4
+	sub		%o2, %o4, %o2
+
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+	add		%o1, 0x40, %o1
+	EX_LD(LOAD(ldx, %o1 - 0x38, %g2))
+	subcc		%o4, 0x40, %o4
+	EX_LD(LOAD(ldx, %o1 - 0x30, %g3))
+	EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE))
+	EX_LD(LOAD(ldx, %o1 - 0x20, %o5))
+	EX_ST(STORE_INIT(%g1, %o0))
+	add		%o0, 0x08, %o0
+	EX_ST(STORE_INIT(%g2, %o0))
+	add		%o0, 0x08, %o0
+	EX_LD(LOAD(ldx, %o1 - 0x18, %g2))
+	EX_ST(STORE_INIT(%g3, %o0))
+	add		%o0, 0x08, %o0
+	EX_LD(LOAD(ldx, %o1 - 0x10, %g3))
+	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+	add		%o0, 0x08, %o0
+	EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE))
+	EX_ST(STORE_INIT(%o5, %o0))
+	add		%o0, 0x08, %o0
+	EX_ST(STORE_INIT(%g2, %o0))
+	add		%o0, 0x08, %o0
+	EX_ST(STORE_INIT(%g3, %o0))
+	add		%o0, 0x08, %o0
+	EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+	add		%o0, 0x08, %o0
+	bne,pt		%icc, 1b
+	 LOAD(prefetch, %o1 + 0x200, #n_reads_strong)
+
+	membar		#StoreLoad | #StoreStore
+
+	brz,pn		%o2, .Lexit
+	 cmp		%o2, 19
+	ble,pn		%icc, .Lsmall_unaligned
+	 nop
+	ba,a,pt		%icc, .Lmedium_noprefetch
+
+.Lexit:	retl
+	 mov		EX_RETVAL(%o3), %o0
+
+.Llarge_src_unaligned:
+	andn		%o2, 0x3f, %o4
+	sub		%o2, %o4, %o2
+	VISEntryHalf
+	alignaddr	%o1, %g0, %g1
+	add		%o1, %o4, %o1
+	EX_LD(LOAD(ldd, %g1 + 0x00, %f0))
+1:	EX_LD(LOAD(ldd, %g1 + 0x08, %f2))
+	subcc		%o4, 0x40, %o4
+	EX_LD(LOAD(ldd, %g1 + 0x10, %f4))
+	EX_LD(LOAD(ldd, %g1 + 0x18, %f6))
+	EX_LD(LOAD(ldd, %g1 + 0x20, %f8))
+	EX_LD(LOAD(ldd, %g1 + 0x28, %f10))
+	EX_LD(LOAD(ldd, %g1 + 0x30, %f12))
+	EX_LD(LOAD(ldd, %g1 + 0x38, %f14))
+	faligndata	%f0, %f2, %f16
+	EX_LD(LOAD(ldd, %g1 + 0x40, %f0))
+	faligndata	%f2, %f4, %f18
+	add		%g1, 0x40, %g1
+	faligndata	%f4, %f6, %f20
+	faligndata	%f6, %f8, %f22
+	faligndata	%f8, %f10, %f24
+	faligndata	%f10, %f12, %f26
+	faligndata	%f12, %f14, %f28
+	faligndata	%f14, %f0, %f30
+	EX_ST(STORE(std, %f16, %o0 + 0x00))
+	EX_ST(STORE(std, %f18, %o0 + 0x08))
+	EX_ST(STORE(std, %f20, %o0 + 0x10))
+	EX_ST(STORE(std, %f22, %o0 + 0x18))
+	EX_ST(STORE(std, %f24, %o0 + 0x20))
+	EX_ST(STORE(std, %f26, %o0 + 0x28))
+	EX_ST(STORE(std, %f28, %o0 + 0x30))
+	EX_ST(STORE(std, %f30, %o0 + 0x38))
+	add		%o0, 0x40, %o0
+	bne,pt		%icc, 1b
+	 LOAD(prefetch, %g1 + 0x200, #n_reads_strong)
+	VISExitHalf
+
+	brz,pn		%o2, .Lexit
+	 cmp		%o2, 19
+	ble,pn		%icc, .Lsmall_unaligned
+	 nop
+	ba,a,pt		%icc, .Lmedium_unaligned
+
+.Lmedium:
+	LOAD(prefetch, %o1 + 0x40, #n_reads_strong)
+	andcc		%g2, 0x7, %g0
+	bne,pn		%icc, .Lmedium_unaligned
+	 nop
+.Lmedium_noprefetch:
+	andncc		%o2, 0x20 - 1, %o5
+	be,pn		%icc, 2f
+	 sub		%o2, %o5, %o2
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+	EX_LD(LOAD(ldx, %o1 + 0x08, %g2))
+	EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE))
+	EX_LD(LOAD(ldx, %o1 + 0x18, %o4))
+	add		%o1, 0x20, %o1
+	subcc		%o5, 0x20, %o5
+	EX_ST(STORE(stx, %g1, %o0 + 0x00))
+	EX_ST(STORE(stx, %g2, %o0 + 0x08))
+	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10))
+	EX_ST(STORE(stx, %o4, %o0 + 0x18))
+	bne,pt		%icc, 1b
+	 add		%o0, 0x20, %o0
+2:	andcc		%o2, 0x18, %o5
+	be,pt		%icc, 3f
+	 sub		%o2, %o5, %o2
+1:	EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+	add		%o1, 0x08, %o1
+	add		%o0, 0x08, %o0
+	subcc		%o5, 0x08, %o5
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stx, %g1, %o0 - 0x08))
+3:	brz,pt		%o2, .Lexit
+	 cmp		%o2, 0x04
+	bl,pn		%icc, .Ltiny
+	 nop
+	EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+	add		%o1, 0x04, %o1
+	add		%o0, 0x04, %o0
+	subcc		%o2, 0x04, %o2
+	bne,pn		%icc, .Ltiny
+	 EX_ST(STORE(stw, %g1, %o0 - 0x04))
+	ba,a,pt		%icc, .Lexit
+.Lmedium_unaligned:
+	/* First get dest 8 byte aligned.  */
+	sub		%g0, %o0, %g1
+	and		%g1, 0x7, %g1
+	brz,pt		%g1, 2f
+	 sub		%o2, %g1, %o2
+
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+	add		%o1, 1, %o1
+	subcc		%g1, 1, %g1
+	add		%o0, 1, %o0
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stb, %g2, %o0 - 0x01))
+2:
+	and		%o1, 0x7, %g1
+	brz,pn		%g1, .Lmedium_noprefetch
+	 sll		%g1, 3, %g1
+	mov		64, %g2
+	sub		%g2, %g1, %g2
+	andn		%o1, 0x7, %o1
+	EX_LD(LOAD(ldx, %o1 + 0x00, %o4))
+	sllx		%o4, %g1, %o4
+	andn		%o2, 0x08 - 1, %o5
+	sub		%o2, %o5, %o2
+1:	EX_LD(LOAD(ldx, %o1 + 0x08, %g3))
+	add		%o1, 0x08, %o1
+	subcc		%o5, 0x08, %o5
+	srlx		%g3, %g2, GLOBAL_SPARE
+	or		GLOBAL_SPARE, %o4, GLOBAL_SPARE
+	EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00))
+	add		%o0, 0x08, %o0
+	bne,pt		%icc, 1b
+	 sllx		%g3, %g1, %o4
+	srl		%g1, 3, %g1
+	add		%o1, %g1, %o1
+	brz,pn		%o2, .Lexit
+	 nop
+	ba,pt		%icc, .Lsmall_unaligned
+
+.Ltiny:
+	EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+	subcc		%o2, 1, %o2
+	be,pn		%icc, .Lexit
+	 EX_ST(STORE(stb, %g1, %o0 + 0x00))
+	EX_LD(LOAD(ldub, %o1 + 0x01, %g1))
+	subcc		%o2, 1, %o2
+	be,pn		%icc, .Lexit
+	 EX_ST(STORE(stb, %g1, %o0 + 0x01))
+	EX_LD(LOAD(ldub, %o1 + 0x02, %g1))
+	ba,pt		%icc, .Lexit
+	 EX_ST(STORE(stb, %g1, %o0 + 0x02))
+
+.Lsmall:
+	andcc		%g2, 0x3, %g0
+	bne,pn		%icc, .Lsmall_unaligned
+	 andn		%o2, 0x4 - 1, %o5
+	sub		%o2, %o5, %o2
+1:
+	EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+	add		%o1, 0x04, %o1
+	subcc		%o5, 0x04, %o5
+	add		%o0, 0x04, %o0
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stw, %g1, %o0 - 0x04))
+	brz,pt		%o2, .Lexit
+	 nop
+	ba,a,pt		%icc, .Ltiny
+
+.Lsmall_unaligned:
+1:	EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+	add		%o1, 1, %o1
+	add		%o0, 1, %o0
+	subcc		%o2, 1, %o2
+	bne,pt		%icc, 1b
+	 EX_ST(STORE(stb, %g1, %o0 - 0x01))
+	ba,a,pt		%icc, .Lexit
+END(__memcpy_niagara4)
+
+#endif
diff --git a/sysdeps/sparc/sparc64/multiarch/memcpy.S b/sysdeps/sparc/sparc64/multiarch/memcpy.S
index 20c72d9bbd..50a40883c0 100644
--- a/sysdeps/sparc/sparc64/multiarch/memcpy.S
+++ b/sysdeps/sparc/sparc64/multiarch/memcpy.S
@@ -26,8 +26,19 @@ ENTRY(memcpy)
 # ifdef SHARED
 	SETUP_PIC_REG_LEAF(o3, o5)
 # endif
-	andcc	%o0, HWCAP_SPARC_N2, %g0
+	set	HWCAP_SPARC_CRYPTO, %o1
+	andcc	%o0, %o1, %g0
 	be	1f
+	 andcc	%o0, HWCAP_SPARC_N2, %g0
+# ifdef SHARED
+	sethi	%gdop_hix22(__memcpy_niagara4), %o1
+	xor	%o1, %gdop_lox10(__memcpy_niagara4), %o1
+# else
+	set	__memcpy_niagara4, %o1
+# endif
+	ba	10f
+	 nop
+1:	be	1f
 	 andcc	%o0, HWCAP_SPARC_BLKINIT, %g0
 # ifdef SHARED
 	sethi	%gdop_hix22(__memcpy_niagara2), %o1
@@ -77,8 +88,19 @@ ENTRY(__mempcpy)
 # ifdef SHARED
 	SETUP_PIC_REG_LEAF(o3, o5)
 # endif
-	andcc	%o0, HWCAP_SPARC_N2, %g0
+	set	HWCAP_SPARC_CRYPTO, %o1
+	andcc	%o0, %o1, %g0
 	be	1f
+	 andcc	%o0, HWCAP_SPARC_N2, %g0
+# ifdef SHARED
+	sethi	%gdop_hix22(__mempcpy_niagara4), %o1
+	xor	%o1, %gdop_lox10(__mempcpy_niagara4), %o1
+# else
+	set	__mempcpy_niagara4, %o1
+# endif
+	ba	10f
+	 nop
+1:	be	1f
 	 andcc	%o0, HWCAP_SPARC_BLKINIT, %g0
 # ifdef SHARED
 	sethi	%gdop_hix22(__mempcpy_niagara2), %o1