diff options
Diffstat (limited to 'arch/powerpc/lib')
| -rw-r--r-- | arch/powerpc/lib/Makefile | 5 | ||||
| -rw-r--r-- | arch/powerpc/lib/checksum_64.S | 27 | ||||
| -rw-r--r-- | arch/powerpc/lib/code-patching.c | 14 | ||||
| -rw-r--r-- | arch/powerpc/lib/copypage_64.S | 4 | ||||
| -rw-r--r-- | arch/powerpc/lib/copypage_power7.S | 165 | ||||
| -rw-r--r-- | arch/powerpc/lib/copyuser_power7.S | 157 | ||||
| -rw-r--r-- | arch/powerpc/lib/crtsavres.S | 5 | ||||
| -rw-r--r-- | arch/powerpc/lib/hweight_64.S | 14 | ||||
| -rw-r--r-- | arch/powerpc/lib/ldstfp.S | 12 | ||||
| -rw-r--r-- | arch/powerpc/lib/memcpy_64.S | 4 | ||||
| -rw-r--r-- | arch/powerpc/lib/memcpy_power7.S | 647 | ||||
| -rw-r--r-- | arch/powerpc/lib/string.S | 2 | ||||
| -rw-r--r-- | arch/powerpc/lib/string_64.S | 202 | ||||
| -rw-r--r-- | arch/powerpc/lib/vmx-helper.c (renamed from arch/powerpc/lib/copyuser_power7_vmx.c) | 27 |
14 files changed, 1199 insertions, 86 deletions
diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile index 7735a2c2e6d..746e0c895cd 100644 --- a/arch/powerpc/lib/Makefile +++ b/arch/powerpc/lib/Makefile @@ -17,14 +17,15 @@ obj-$(CONFIG_HAS_IOMEM) += devres.o obj-$(CONFIG_PPC64) += copypage_64.o copyuser_64.o \ memcpy_64.o usercopy_64.o mem_64.o string.o \ checksum_wrappers_64.o hweight_64.o \ - copyuser_power7.o + copyuser_power7.o string_64.o copypage_power7.o \ + memcpy_power7.o obj-$(CONFIG_XMON) += sstep.o ldstfp.o obj-$(CONFIG_KPROBES) += sstep.o ldstfp.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += sstep.o ldstfp.o ifeq ($(CONFIG_PPC64),y) obj-$(CONFIG_SMP) += locks.o -obj-$(CONFIG_ALTIVEC) += copyuser_power7_vmx.o +obj-$(CONFIG_ALTIVEC) += vmx-helper.o endif obj-$(CONFIG_PPC_LIB_RHEAP) += rheap.o diff --git a/arch/powerpc/lib/checksum_64.S b/arch/powerpc/lib/checksum_64.S index 18245af38ae..167f72555d6 100644 --- a/arch/powerpc/lib/checksum_64.S +++ b/arch/powerpc/lib/checksum_64.S @@ -65,9 +65,6 @@ _GLOBAL(csum_tcpudp_magic) srwi r3,r3,16 blr -#define STACKFRAMESIZE 256 -#define STK_REG(i) (112 + ((i)-14)*8) - /* * Computes the checksum of a memory block at buff, length len, * and adds in "sum" (32-bit). @@ -114,9 +111,9 @@ _GLOBAL(csum_partial) mtctr r6 stdu r1,-STACKFRAMESIZE(r1) - std r14,STK_REG(r14)(r1) - std r15,STK_REG(r15)(r1) - std r16,STK_REG(r16)(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) ld r6,0(r3) ld r9,8(r3) @@ -175,9 +172,9 @@ _GLOBAL(csum_partial) adde r0,r0,r15 adde r0,r0,r16 - ld r14,STK_REG(r14)(r1) - ld r15,STK_REG(r15)(r1) - ld r16,STK_REG(r16)(r1) + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) addi r1,r1,STACKFRAMESIZE andi. r4,r4,63 @@ -299,9 +296,9 @@ dest; sth r6,0(r4) mtctr r6 stdu r1,-STACKFRAMESIZE(r1) - std r14,STK_REG(r14)(r1) - std r15,STK_REG(r15)(r1) - std r16,STK_REG(r16)(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) source; ld r6,0(r3) source; ld r9,8(r3) @@ -382,9 +379,9 @@ dest; std r16,56(r4) adde r0,r0,r15 adde r0,r0,r16 - ld r14,STK_REG(r14)(r1) - ld r15,STK_REG(r15)(r1) - ld r16,STK_REG(r16)(r1) + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) addi r1,r1,STACKFRAMESIZE andi. r5,r5,63 diff --git a/arch/powerpc/lib/code-patching.c b/arch/powerpc/lib/code-patching.c index 7c975d43e3f..dd223b3eb33 100644 --- a/arch/powerpc/lib/code-patching.c +++ b/arch/powerpc/lib/code-patching.c @@ -13,17 +13,23 @@ #include <linux/mm.h> #include <asm/page.h> #include <asm/code-patching.h> +#include <asm/uaccess.h> -void patch_instruction(unsigned int *addr, unsigned int instr) +int patch_instruction(unsigned int *addr, unsigned int instr) { - *addr = instr; + int err; + + err = __put_user(instr, addr); + if (err) + return err; asm ("dcbst 0, %0; sync; icbi 0,%0; sync; isync" : : "r" (addr)); + return 0; } -void patch_branch(unsigned int *addr, unsigned long target, int flags) +int patch_branch(unsigned int *addr, unsigned long target, int flags) { - patch_instruction(addr, create_branch(addr, target, flags)); + return patch_instruction(addr, create_branch(addr, target, flags)); } unsigned int create_branch(const unsigned int *addr, diff --git a/arch/powerpc/lib/copypage_64.S b/arch/powerpc/lib/copypage_64.S index 53dcb6b1b70..9f9434a8526 100644 --- a/arch/powerpc/lib/copypage_64.S +++ b/arch/powerpc/lib/copypage_64.S @@ -17,7 +17,11 @@ PPC64_CACHES: .section ".text" _GLOBAL(copy_page) +BEGIN_FTR_SECTION lis r5,PAGE_SIZE@h +FTR_SECTION_ELSE + b .copypage_power7 +ALT_FTR_SECTION_END_IFCLR(CPU_FTR_VMX_COPY) ori r5,r5,PAGE_SIZE@l BEGIN_FTR_SECTION ld r10,PPC64_CACHES@toc(r2) diff --git a/arch/powerpc/lib/copypage_power7.S b/arch/powerpc/lib/copypage_power7.S new file mode 100644 index 00000000000..0ef75bf0695 --- /dev/null +++ b/arch/powerpc/lib/copypage_power7.S @@ -0,0 +1,165 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2012 + * + * Author: Anton Blanchard <anton@au.ibm.com> + */ +#include <asm/page.h> +#include <asm/ppc_asm.h> + +_GLOBAL(copypage_power7) + /* + * We prefetch both the source and destination using enhanced touch + * instructions. We use a stream ID of 0 for the load side and + * 1 for the store side. Since source and destination are page + * aligned we don't need to clear the bottom 7 bits of either + * address. + */ + ori r9,r3,1 /* stream=1 */ + +#ifdef CONFIG_PPC_64K_PAGES + lis r7,0x0E01 /* depth=7, units=512 */ +#else + lis r7,0x0E00 /* depth=7 */ + ori r7,r7,0x1000 /* units=32 */ +#endif + ori r10,r7,1 /* stream=1 */ + + lis r8,0x8000 /* GO=1 */ + clrldi r8,r8,32 + +.machine push +.machine "power4" + dcbt r0,r4,0b01000 + dcbt r0,r7,0b01010 + dcbtst r0,r9,0b01000 + dcbtst r0,r10,0b01010 + eieio + dcbt r0,r8,0b01010 /* GO */ +.machine pop + +#ifdef CONFIG_ALTIVEC + mflr r0 + std r3,48(r1) + std r4,56(r1) + std r0,16(r1) + stdu r1,-STACKFRAMESIZE(r1) + bl .enter_vmx_copy + cmpwi r3,0 + ld r0,STACKFRAMESIZE+16(r1) + ld r3,STACKFRAMESIZE+48(r1) + ld r4,STACKFRAMESIZE+56(r1) + mtlr r0 + + li r0,(PAGE_SIZE/128) + mtctr r0 + + beq .Lnonvmx_copy + + addi r1,r1,STACKFRAMESIZE + + li r6,16 + li r7,32 + li r8,48 + li r9,64 + li r10,80 + li r11,96 + li r12,112 + + .align 5 +1: lvx vr7,r0,r4 + lvx vr6,r4,r6 + lvx vr5,r4,r7 + lvx vr4,r4,r8 + lvx vr3,r4,r9 + lvx vr2,r4,r10 + lvx vr1,r4,r11 + lvx vr0,r4,r12 + addi r4,r4,128 + stvx vr7,r0,r3 + stvx vr6,r3,r6 + stvx vr5,r3,r7 + stvx vr4,r3,r8 + stvx vr3,r3,r9 + stvx vr2,r3,r10 + stvx vr1,r3,r11 + stvx vr0,r3,r12 + addi r3,r3,128 + bdnz 1b + + b .exit_vmx_copy /* tail call optimise */ + +#else + li r0,(PAGE_SIZE/128) + mtctr r0 + + stdu r1,-STACKFRAMESIZE(r1) +#endif + +.Lnonvmx_copy: + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) + std r17,STK_REG(R17)(r1) + std r18,STK_REG(R18)(r1) + std r19,STK_REG(R19)(r1) + std r20,STK_REG(R20)(r1) + +1: ld r0,0(r4) + ld r5,8(r4) + ld r6,16(r4) + ld r7,24(r4) + ld r8,32(r4) + ld r9,40(r4) + ld r10,48(r4) + ld r11,56(r4) + ld r12,64(r4) + ld r14,72(r4) + ld r15,80(r4) + ld r16,88(r4) + ld r17,96(r4) + ld r18,104(r4) + ld r19,112(r4) + ld r20,120(r4) + addi r4,r4,128 + std r0,0(r3) + std r5,8(r3) + std r6,16(r3) + std r7,24(r3) + std r8,32(r3) + std r9,40(r3) + std r10,48(r3) + std r11,56(r3) + std r12,64(r3) + std r14,72(r3) + std r15,80(r3) + std r16,88(r3) + std r17,96(r3) + std r18,104(r3) + std r19,112(r3) + std r20,120(r3) + addi r3,r3,128 + bdnz 1b + + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) + ld r17,STK_REG(R17)(r1) + ld r18,STK_REG(R18)(r1) + ld r19,STK_REG(R19)(r1) + ld r20,STK_REG(R20)(r1) + addi r1,r1,STACKFRAMESIZE + blr diff --git a/arch/powerpc/lib/copyuser_power7.S b/arch/powerpc/lib/copyuser_power7.S index 497db7b23bb..f9ede7c6606 100644 --- a/arch/powerpc/lib/copyuser_power7.S +++ b/arch/powerpc/lib/copyuser_power7.S @@ -19,9 +19,6 @@ */ #include <asm/ppc_asm.h> -#define STACKFRAMESIZE 256 -#define STK_REG(i) (112 + ((i)-14)*8) - .macro err1 100: .section __ex_table,"a" @@ -57,26 +54,26 @@ .Ldo_err4: - ld r16,STK_REG(r16)(r1) - ld r15,STK_REG(r15)(r1) - ld r14,STK_REG(r14)(r1) + ld r16,STK_REG(R16)(r1) + ld r15,STK_REG(R15)(r1) + ld r14,STK_REG(R14)(r1) .Ldo_err3: - bl .exit_vmx_copy + bl .exit_vmx_usercopy ld r0,STACKFRAMESIZE+16(r1) mtlr r0 b .Lexit #endif /* CONFIG_ALTIVEC */ .Ldo_err2: - ld r22,STK_REG(r22)(r1) - ld r21,STK_REG(r21)(r1) - ld r20,STK_REG(r20)(r1) - ld r19,STK_REG(r19)(r1) - ld r18,STK_REG(r18)(r1) - ld r17,STK_REG(r17)(r1) - ld r16,STK_REG(r16)(r1) - ld r15,STK_REG(r15)(r1) - ld r14,STK_REG(r14)(r1) + ld r22,STK_REG(R22)(r1) + ld r21,STK_REG(R21)(r1) + ld r20,STK_REG(R20)(r1) + ld r19,STK_REG(R19)(r1) + ld r18,STK_REG(R18)(r1) + ld r17,STK_REG(R17)(r1) + ld r16,STK_REG(R16)(r1) + ld r15,STK_REG(R15)(r1) + ld r14,STK_REG(R14)(r1) .Lexit: addi r1,r1,STACKFRAMESIZE .Ldo_err1: @@ -137,15 +134,15 @@ err1; stw r0,0(r3) mflr r0 stdu r1,-STACKFRAMESIZE(r1) - std r14,STK_REG(r14)(r1) - std r15,STK_REG(r15)(r1) - std r16,STK_REG(r16)(r1) - std r17,STK_REG(r17)(r1) - std r18,STK_REG(r18)(r1) - std r19,STK_REG(r19)(r1) - std r20,STK_REG(r20)(r1) - std r21,STK_REG(r21)(r1) - std r22,STK_REG(r22)(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) + std r17,STK_REG(R17)(r1) + std r18,STK_REG(R18)(r1) + std r19,STK_REG(R19)(r1) + std r20,STK_REG(R20)(r1) + std r21,STK_REG(R21)(r1) + std r22,STK_REG(R22)(r1) std r0,STACKFRAMESIZE+16(r1) srdi r6,r5,7 @@ -192,15 +189,15 @@ err2; std r21,120(r3) clrldi r5,r5,(64-7) - ld r14,STK_REG(r14)(r1) - ld r15,STK_REG(r15)(r1) - ld r16,STK_REG(r16)(r1) - ld r17,STK_REG(r17)(r1) - ld r18,STK_REG(r18)(r1) - ld r19,STK_REG(r19)(r1) - ld r20,STK_REG(r20)(r1) - ld r21,STK_REG(r21)(r1) - ld r22,STK_REG(r22)(r1) + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) + ld r17,STK_REG(R17)(r1) + ld r18,STK_REG(R18)(r1) + ld r19,STK_REG(R19)(r1) + ld r20,STK_REG(R20)(r1) + ld r21,STK_REG(R21)(r1) + ld r22,STK_REG(R22)(r1) addi r1,r1,STACKFRAMESIZE /* Up to 127B to go */ @@ -290,7 +287,7 @@ err1; stb r0,0(r3) mflr r0 std r0,16(r1) stdu r1,-STACKFRAMESIZE(r1) - bl .enter_vmx_copy + bl .enter_vmx_usercopy cmpwi r3,0 ld r0,STACKFRAMESIZE+16(r1) ld r3,STACKFRAMESIZE+48(r1) @@ -298,6 +295,68 @@ err1; stb r0,0(r3) ld r5,STACKFRAMESIZE+64(r1) mtlr r0 + /* + * We prefetch both the source and destination using enhanced touch + * instructions. We use a stream ID of 0 for the load side and + * 1 for the store side. + */ + clrrdi r6,r4,7 + clrrdi r9,r3,7 + ori r9,r9,1 /* stream=1 */ + + srdi r7,r5,7 /* length in cachelines, capped at 0x3FF */ + cmpldi r7,0x3FF + ble 1f + li r7,0x3FF +1: lis r0,0x0E00 /* depth=7 */ + sldi r7,r7,7 + or r7,r7,r0 + ori r10,r7,1 /* stream=1 */ + + lis r8,0x8000 /* GO=1 */ + clrldi r8,r8,32 + +.machine push +.machine "power4" + dcbt r0,r6,0b01000 + dcbt r0,r7,0b01010 + dcbtst r0,r9,0b01000 + dcbtst r0,r10,0b01010 + eieio + dcbt r0,r8,0b01010 /* GO */ +.machine pop + + /* + * We prefetch both the source and destination using enhanced touch + * instructions. We use a stream ID of 0 for the load side and + * 1 for the store side. + */ + clrrdi r6,r4,7 + clrrdi r9,r3,7 + ori r9,r9,1 /* stream=1 */ + + srdi r7,r5,7 /* length in cachelines, capped at 0x3FF */ + cmpldi cr1,r7,0x3FF + ble cr1,1f + li r7,0x3FF +1: lis r0,0x0E00 /* depth=7 */ + sldi r7,r7,7 + or r7,r7,r0 + ori r10,r7,1 /* stream=1 */ + + lis r8,0x8000 /* GO=1 */ + clrldi r8,r8,32 + +.machine push +.machine "power4" + dcbt r0,r6,0b01000 + dcbt r0,r7,0b01010 + dcbtst r0,r9,0b01000 + dcbtst r0,r10,0b01010 + eieio + dcbt r0,r8,0b01010 /* GO */ +.machine pop + beq .Lunwind_stack_nonvmx_copy /* @@ -378,9 +437,9 @@ err3; stvx vr0,r3,r11 7: sub r5,r5,r6 srdi r6,r5,7 - std r14,STK_REG(r14)(r1) - std r15,STK_REG(r15)(r1) - std r16,STK_REG(r16)(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) li r12,64 li r14,80 @@ -415,9 +474,9 @@ err4; stvx vr0,r3,r16 addi r3,r3,128 bdnz 8b - ld r14,STK_REG(r14)(r1) - ld r15,STK_REG(r15)(r1) - ld r16,STK_REG(r16)(r1) + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) /* Up to 127B to go */ clrldi r5,r5,(64-7) @@ -476,7 +535,7 @@ err3; lbz r0,0(r4) err3; stb r0,0(r3) 15: addi r1,r1,STACKFRAMESIZE - b .exit_vmx_copy /* tail call optimise */ + b .exit_vmx_usercopy /* tail call optimise */ .Lvmx_unaligned_copy: /* Get the destination 16B aligned */ @@ -563,9 +622,9 @@ err3; stvx vr11,r3,r11 7: sub r5,r5,r6 srdi r6,r5,7 - std r14,STK_REG(r14)(r1) - std r15,STK_REG(r15)(r1) - std r16,STK_REG(r16)(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) li r12,64 li r14,80 @@ -608,9 +667,9 @@ err4; stvx vr15,r3,r16 addi r3,r3,128 bdnz 8b - ld r14,STK_REG(r14)(r1) - ld r15,STK_REG(r15)(r1) - ld r16,STK_REG(r16)(r1) + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) /* Up to 127B to go */ clrldi r5,r5,(64-7) @@ -679,5 +738,5 @@ err3; lbz r0,0(r4) err3; stb r0,0(r3) 15: addi r1,r1,STACKFRAMESIZE - b .exit_vmx_copy /* tail call optimise */ + b .exit_vmx_usercopy /* tail call optimise */ #endif /* CONFiG_ALTIVEC */ diff --git a/arch/powerpc/lib/crtsavres.S b/arch/powerpc/lib/crtsavres.S index 1c893f05d22..b2c68ce139a 100644 --- a/arch/powerpc/lib/crtsavres.S +++ b/arch/powerpc/lib/crtsavres.S @@ -41,12 +41,13 @@ #include <asm/ppc_asm.h> .file "crtsavres.S" - .section ".text" #ifdef CONFIG_CC_OPTIMIZE_FOR_SIZE #ifndef CONFIG_PPC64 + .section ".text" + /* Routines for saving integer registers, called by the compiler. */ /* Called with r11 pointing to the stack header word of the caller of the */ /* function, just beyond the end of the integer save area. */ @@ -232,6 +233,8 @@ _GLOBAL(_rest32gpr_31_x) #else /* CONFIG_PPC64 */ + .section ".text.save.restore","ax",@progbits + .globl _savegpr0_14 _savegpr0_14: std r14,-144(r1) diff --git a/arch/powerpc/lib/hweight_64.S b/arch/powerpc/lib/hweight_64.S index fda27868cf8..9b96ff2ecd4 100644 --- a/arch/powerpc/lib/hweight_64.S +++ b/arch/powerpc/lib/hweight_64.S @@ -28,7 +28,7 @@ BEGIN_FTR_SECTION nop nop FTR_SECTION_ELSE - PPC_POPCNTB(r3,r3) + PPC_POPCNTB(R3,R3) clrldi r3,r3,64-8 blr ALT_FTR_SECTION_END_IFCLR(CPU_FTR_POPCNTB) @@ -42,14 +42,14 @@ BEGIN_FTR_SECTION nop FTR_SECTION_ELSE BEGIN_FTR_SECTION_NESTED(50) - PPC_POPCNTB(r3,r3) + PPC_POPCNTB(R3,R3) srdi r4,r3,8 add r3,r4,r3 clrldi r3,r3,64-8 blr FTR_SECTION_ELSE_NESTED(50) clrlwi r3,r3,16 - PPC_POPCNTW(r3,r3) + PPC_POPCNTW(R3,R3) clrldi r3,r3,64-8 blr ALT_FTR_SECTION_END_NESTED_IFCLR(CPU_FTR_POPCNTD, 50) @@ -66,7 +66,7 @@ BEGIN_FTR_SECTION nop FTR_SECTION_ELSE BEGIN_FTR_SECTION_NESTED(51) - PPC_POPCNTB(r3,r3) + PPC_POPCNTB(R3,R3) srdi r4,r3,16 add r3,r4,r3 srdi r4,r3,8 @@ -74,7 +74,7 @@ FTR_SECTION_ELSE clrldi r3,r3,64-8 blr FTR_SECTION_ELSE_NESTED(51) - PPC_POPCNTW(r3,r3) + PPC_POPCNTW(R3,R3) clrldi r3,r3,64-8 blr ALT_FTR_SECTION_END_NESTED_IFCLR(CPU_FTR_POPCNTD, 51) @@ -93,7 +93,7 @@ BEGIN_FTR_SECTION nop FTR_SECTION_ELSE BEGIN_FTR_SECTION_NESTED(52) - PPC_POPCNTB(r3,r3) + PPC_POPCNTB(R3,R3) srdi r4,r3,32 add r3,r4,r3 srdi r4,r3,16 @@ -103,7 +103,7 @@ FTR_SECTION_ELSE clrldi r3,r3,64-8 blr FTR_SECTION_ELSE_NESTED(52) - PPC_POPCNTD(r3,r3) + PPC_POPCNTD(R3,R3) clrldi r3,r3,64-8 blr ALT_FTR_SECTION_END_NESTED_IFCLR(CPU_FTR_POPCNTD, 52) diff --git a/arch/powerpc/lib/ldstfp.S b/arch/powerpc/lib/ldstfp.S index 6a85380520b..85aec08ab23 100644 --- a/arch/powerpc/lib/ldstfp.S +++ b/arch/powerpc/lib/ldstfp.S @@ -330,13 +330,13 @@ _GLOBAL(do_lxvd2x) MTMSRD(r7) isync beq cr7,1f - STXVD2X(0,r1,r8) + STXVD2X(0,R1,R8) 1: li r9,-EFAULT -2: LXVD2X(0,0,r4) +2: LXVD2X(0,R0,R4) li r9,0 3: beq cr7,4f bl put_vsr - LXVD2X(0,r1,r8) + LXVD2X(0,R1,R8) 4: PPC_LL r0,STKFRM+PPC_LR_STKOFF(r1) mtlr r0 MTMSRD(r6) @@ -358,13 +358,13 @@ _GLOBAL(do_stxvd2x) MTMSRD(r7) isync beq cr7,1f - STXVD2X(0,r1,r8) + STXVD2X(0,R1,R8) bl get_vsr 1: li r9,-EFAULT -2: STXVD2X(0,0,r4) +2: STXVD2X(0,R0,R4) li r9,0 3: beq cr7,4f - LXVD2X(0,r1,r8) + LXVD2X(0,R1,R8) 4: PPC_LL r0,STKFRM+PPC_LR_STKOFF(r1) mtlr r0 MTMSRD(r6) diff --git a/arch/powerpc/lib/memcpy_64.S b/arch/powerpc/lib/memcpy_64.S index 82fea3963e1..d2bbbc8d7dc 100644 --- a/arch/powerpc/lib/memcpy_64.S +++ b/arch/powerpc/lib/memcpy_64.S @@ -11,7 +11,11 @@ .align 7 _GLOBAL(memcpy) +BEGIN_FTR_SECTION std r3,48(r1) /* save destination pointer for return value */ +FTR_SECTION_ELSE + b memcpy_power7 +ALT_FTR_SECTION_END_IFCLR(CPU_FTR_VMX_COPY) PPC_MTOCRF(0x01,r5) cmpldi cr1,r5,16 neg r6,r3 # LS 3 bits = # bytes to 8-byte dest bdry diff --git a/arch/powerpc/lib/memcpy_power7.S b/arch/powerpc/lib/memcpy_power7.S new file mode 100644 index 00000000000..0efdc51bc71 --- /dev/null +++ b/arch/powerpc/lib/memcpy_power7.S @@ -0,0 +1,647 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2012 + * + * Author: Anton Blanchard <anton@au.ibm.com> + */ +#include <asm/ppc_asm.h> + +_GLOBAL(memcpy_power7) +#ifdef CONFIG_ALTIVEC + cmpldi r5,16 + cmpldi cr1,r5,4096 + + std r3,48(r1) + + blt .Lshort_copy + bgt cr1,.Lvmx_copy +#else + cmpldi r5,16 + + std r3,48(r1) + + blt .Lshort_copy +#endif + +.Lnonvmx_copy: + /* Get the source 8B aligned */ + neg r6,r4 + mtocrf 0x01,r6 + clrldi r6,r6,(64-3) + + bf cr7*4+3,1f + lbz r0,0(r4) + addi r4,r4,1 + stb r0,0(r3) + addi r3,r3,1 + +1: bf cr7*4+2,2f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +2: bf cr7*4+1,3f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +3: sub r5,r5,r6 + cmpldi r5,128 + blt 5f + + mflr r0 + stdu r1,-STACKFRAMESIZE(r1) + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) + std r17,STK_REG(R17)(r1) + std r18,STK_REG(R18)(r1) + std r19,STK_REG(R19)(r1) + std r20,STK_REG(R20)(r1) + std r21,STK_REG(R21)(r1) + std r22,STK_REG(R22)(r1) + std r0,STACKFRAMESIZE+16(r1) + + srdi r6,r5,7 + mtctr r6 + + /* Now do cacheline (128B) sized loads and stores. */ + .align 5 +4: + ld r0,0(r4) + ld r6,8(r4) + ld r7,16(r4) + ld r8,24(r4) + ld r9,32(r4) + ld r10,40(r4) + ld r11,48(r4) + ld r12,56(r4) + ld r14,64(r4) + ld r15,72(r4) + ld r16,80(r4) + ld r17,88(r4) + ld r18,96(r4) + ld r19,104(r4) + ld r20,112(r4) + ld r21,120(r4) + addi r4,r4,128 + std r0,0(r3) + std r6,8(r3) + std r7,16(r3) + std r8,24(r3) + std r9,32(r3) + std r10,40(r3) + std r11,48(r3) + std r12,56(r3) + std r14,64(r3) + std r15,72(r3) + std r16,80(r3) + std r17,88(r3) + std r18,96(r3) + std r19,104(r3) + std r20,112(r3) + std r21,120(r3) + addi r3,r3,128 + bdnz 4b + + clrldi r5,r5,(64-7) + + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) + ld r17,STK_REG(R17)(r1) + ld r18,STK_REG(R18)(r1) + ld r19,STK_REG(R19)(r1) + ld r20,STK_REG(R20)(r1) + ld r21,STK_REG(R21)(r1) + ld r22,STK_REG(R22)(r1) + addi r1,r1,STACKFRAMESIZE + + /* Up to 127B to go */ +5: srdi r6,r5,4 + mtocrf 0x01,r6 + +6: bf cr7*4+1,7f + ld r0,0(r4) + ld r6,8(r4) + ld r7,16(r4) + ld r8,24(r4) + ld r9,32(r4) + ld r10,40(r4) + ld r11,48(r4) + ld r12,56(r4) + addi r4,r4,64 + std r0,0(r3) + std r6,8(r3) + std r7,16(r3) + std r8,24(r3) + std r9,32(r3) + std r10,40(r3) + std r11,48(r3) + std r12,56(r3) + addi r3,r3,64 + + /* Up to 63B to go */ +7: bf cr7*4+2,8f + ld r0,0(r4) + ld r6,8(r4) + ld r7,16(r4) + ld r8,24(r4) + addi r4,r4,32 + std r0,0(r3) + std r6,8(r3) + std r7,16(r3) + std r8,24(r3) + addi r3,r3,32 + + /* Up to 31B to go */ +8: bf cr7*4+3,9f + ld r0,0(r4) + ld r6,8(r4) + addi r4,r4,16 + std r0,0(r3) + std r6,8(r3) + addi r3,r3,16 + +9: clrldi r5,r5,(64-4) + + /* Up to 15B to go */ +.Lshort_copy: + mtocrf 0x01,r5 + bf cr7*4+0,12f + lwz r0,0(r4) /* Less chance of a reject with word ops */ + lwz r6,4(r4) + addi r4,r4,8 + stw r0,0(r3) + stw r6,4(r3) + addi r3,r3,8 + +12: bf cr7*4+1,13f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +13: bf cr7*4+2,14f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +14: bf cr7*4+3,15f + lbz r0,0(r4) + stb r0,0(r3) + +15: ld r3,48(r1) + blr + +.Lunwind_stack_nonvmx_copy: + addi r1,r1,STACKFRAMESIZE + b .Lnonvmx_copy + +#ifdef CONFIG_ALTIVEC +.Lvmx_copy: + mflr r0 + std r4,56(r1) + std r5,64(r1) + std r0,16(r1) + stdu r1,-STACKFRAMESIZE(r1) + bl .enter_vmx_copy + cmpwi r3,0 + ld r0,STACKFRAMESIZE+16(r1) + ld r3,STACKFRAMESIZE+48(r1) + ld r4,STACKFRAMESIZE+56(r1) + ld r5,STACKFRAMESIZE+64(r1) + mtlr r0 + + /* + * We prefetch both the source and destination using enhanced touch + * instructions. We use a stream ID of 0 for the load side and + * 1 for the store side. + */ + clrrdi r6,r4,7 + clrrdi r9,r3,7 + ori r9,r9,1 /* stream=1 */ + + srdi r7,r5,7 /* length in cachelines, capped at 0x3FF */ + cmpldi cr1,r7,0x3FF + ble cr1,1f + li r7,0x3FF +1: lis r0,0x0E00 /* depth=7 */ + sldi r7,r7,7 + or r7,r7,r0 + ori r10,r7,1 /* stream=1 */ + + lis r8,0x8000 /* GO=1 */ + clrldi r8,r8,32 + +.machine push +.machine "power4" + dcbt r0,r6,0b01000 + dcbt r0,r7,0b01010 + dcbtst r0,r9,0b01000 + dcbtst r0,r10,0b01010 + eieio + dcbt r0,r8,0b01010 /* GO */ +.machine pop + + beq .Lunwind_stack_nonvmx_copy + + /* + * If source and destination are not relatively aligned we use a + * slower permute loop. + */ + xor r6,r4,r3 + rldicl. r6,r6,0,(64-4) + bne .Lvmx_unaligned_copy + + /* Get the destination 16B aligned */ + neg r6,r3 + mtocrf 0x01,r6 + clrldi r6,r6,(64-4) + + bf cr7*4+3,1f + lbz r0,0(r4) + addi r4,r4,1 + stb r0,0(r3) + addi r3,r3,1 + +1: bf cr7*4+2,2f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +2: bf cr7*4+1,3f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +3: bf cr7*4+0,4f + ld r0,0(r4) + addi r4,r4,8 + std r0,0(r3) + addi r3,r3,8 + +4: sub r5,r5,r6 + + /* Get the desination 128B aligned */ + neg r6,r3 + srdi r7,r6,4 + mtocrf 0x01,r7 + clrldi r6,r6,(64-7) + + li r9,16 + li r10,32 + li r11,48 + + bf cr7*4+3,5f + lvx vr1,r0,r4 + addi r4,r4,16 + stvx vr1,r0,r3 + addi r3,r3,16 + +5: bf cr7*4+2,6f + lvx vr1,r0,r4 + lvx vr0,r4,r9 + addi r4,r4,32 + stvx vr1,r0,r3 + stvx vr0,r3,r9 + addi r3,r3,32 + +6: bf cr7*4+1,7f + lvx vr3,r0,r4 + lvx vr2,r4,r9 + lvx vr1,r4,r10 + lvx vr0,r4,r11 + addi r4,r4,64 + stvx vr3,r0,r3 + stvx vr2,r3,r9 + stvx vr1,r3,r10 + stvx vr0,r3,r11 + addi r3,r3,64 + +7: sub r5,r5,r6 + srdi r6,r5,7 + + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) + + li r12,64 + li r14,80 + li r15,96 + li r16,112 + + mtctr r6 + + /* + * Now do cacheline sized loads and stores. By this stage the + * cacheline stores are also cacheline aligned. + */ + .align 5 +8: + lvx vr7,r0,r4 + lvx vr6,r4,r9 + lvx vr5,r4,r10 + lvx vr4,r4,r11 + lvx vr3,r4,r12 + lvx vr2,r4,r14 + lvx vr1,r4,r15 + lvx vr0,r4,r16 + addi r4,r4,128 + stvx vr7,r0,r3 + stvx vr6,r3,r9 + stvx vr5,r3,r10 + stvx vr4,r3,r11 + stvx vr3,r3,r12 + stvx vr2,r3,r14 + stvx vr1,r3,r15 + stvx vr0,r3,r16 + addi r3,r3,128 + bdnz 8b + + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) + + /* Up to 127B to go */ + clrldi r5,r5,(64-7) + srdi r6,r5,4 + mtocrf 0x01,r6 + + bf cr7*4+1,9f + lvx vr3,r0,r4 + lvx vr2,r4,r9 + lvx vr1,r4,r10 + lvx vr0,r4,r11 + addi r4,r4,64 + stvx vr3,r0,r3 + stvx vr2,r3,r9 + stvx vr1,r3,r10 + stvx vr0,r3,r11 + addi r3,r3,64 + +9: bf cr7*4+2,10f + lvx vr1,r0,r4 + lvx vr0,r4,r9 + addi r4,r4,32 + stvx vr1,r0,r3 + stvx vr0,r3,r9 + addi r3,r3,32 + +10: bf cr7*4+3,11f + lvx vr1,r0,r4 + addi r4,r4,16 + stvx vr1,r0,r3 + addi r3,r3,16 + + /* Up to 15B to go */ +11: clrldi r5,r5,(64-4) + mtocrf 0x01,r5 + bf cr7*4+0,12f + ld r0,0(r4) + addi r4,r4,8 + std r0,0(r3) + addi r3,r3,8 + +12: bf cr7*4+1,13f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +13: bf cr7*4+2,14f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +14: bf cr7*4+3,15f + lbz r0,0(r4) + stb r0,0(r3) + +15: addi r1,r1,STACKFRAMESIZE + ld r3,48(r1) + b .exit_vmx_copy /* tail call optimise */ + +.Lvmx_unaligned_copy: + /* Get the destination 16B aligned */ + neg r6,r3 + mtocrf 0x01,r6 + clrldi r6,r6,(64-4) + + bf cr7*4+3,1f + lbz r0,0(r4) + addi r4,r4,1 + stb r0,0(r3) + addi r3,r3,1 + +1: bf cr7*4+2,2f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +2: bf cr7*4+1,3f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +3: bf cr7*4+0,4f + lwz r0,0(r4) /* Less chance of a reject with word ops */ + lwz r7,4(r4) + addi r4,r4,8 + stw r0,0(r3) + stw r7,4(r3) + addi r3,r3,8 + +4: sub r5,r5,r6 + + /* Get the desination 128B aligned */ + neg r6,r3 + srdi r7,r6,4 + mtocrf 0x01,r7 + clrldi r6,r6,(64-7) + + li r9,16 + li r10,32 + li r11,48 + + lvsl vr16,0,r4 /* Setup permute control vector */ + lvx vr0,0,r4 + addi r4,r4,16 + + bf cr7*4+3,5f + lvx vr1,r0,r4 + vperm vr8,vr0,vr1,vr16 + addi r4,r4,16 + stvx vr8,r0,r3 + addi r3,r3,16 + vor vr0,vr1,vr1 + +5: bf cr7*4+2,6f + lvx vr1,r0,r4 + vperm vr8,vr0,vr1,vr16 + lvx vr0,r4,r9 + vperm vr9,vr1,vr0,vr16 + addi r4,r4,32 + stvx vr8,r0,r3 + stvx vr9,r3,r9 + addi r3,r3,32 + +6: bf cr7*4+1,7f + lvx vr3,r0,r4 + vperm vr8,vr0,vr3,vr16 + lvx vr2,r4,r9 + vperm vr9,vr3,vr2,vr16 + lvx vr1,r4,r10 + vperm vr10,vr2,vr1,vr16 + lvx vr0,r4,r11 + vperm vr11,vr1,vr0,vr16 + addi r4,r4,64 + stvx vr8,r0,r3 + stvx vr9,r3,r9 + stvx vr10,r3,r10 + stvx vr11,r3,r11 + addi r3,r3,64 + +7: sub r5,r5,r6 + srdi r6,r5,7 + + std r14,STK_REG(R14)(r1) + std r15,STK_REG(R15)(r1) + std r16,STK_REG(R16)(r1) + + li r12,64 + li r14,80 + li r15,96 + li r16,112 + + mtctr r6 + + /* + * Now do cacheline sized loads and stores. By this stage the + * cacheline stores are also cacheline aligned. + */ + .align 5 +8: + lvx vr7,r0,r4 + vperm vr8,vr0,vr7,vr16 + lvx vr6,r4,r9 + vperm vr9,vr7,vr6,vr16 + lvx vr5,r4,r10 + vperm vr10,vr6,vr5,vr16 + lvx vr4,r4,r11 + vperm vr11,vr5,vr4,vr16 + lvx vr3,r4,r12 + vperm vr12,vr4,vr3,vr16 + lvx vr2,r4,r14 + vperm vr13,vr3,vr2,vr16 + lvx vr1,r4,r15 + vperm vr14,vr2,vr1,vr16 + lvx vr0,r4,r16 + vperm vr15,vr1,vr0,vr16 + addi r4,r4,128 + stvx vr8,r0,r3 + stvx vr9,r3,r9 + stvx vr10,r3,r10 + stvx vr11,r3,r11 + stvx vr12,r3,r12 + stvx vr13,r3,r14 + stvx vr14,r3,r15 + stvx vr15,r3,r16 + addi r3,r3,128 + bdnz 8b + + ld r14,STK_REG(R14)(r1) + ld r15,STK_REG(R15)(r1) + ld r16,STK_REG(R16)(r1) + + /* Up to 127B to go */ + clrldi r5,r5,(64-7) + srdi r6,r5,4 + mtocrf 0x01,r6 + + bf cr7*4+1,9f + lvx vr3,r0,r4 + vperm vr8,vr0,vr3,vr16 + lvx vr2,r4,r9 + vperm vr9,vr3,vr2,vr16 + lvx vr1,r4,r10 + vperm vr10,vr2,vr1,vr16 + lvx vr0,r4,r11 + vperm vr11,vr1,vr0,vr16 + addi r4,r4,64 + stvx vr8,r0,r3 + stvx vr9,r3,r9 + stvx vr10,r3,r10 + stvx vr11,r3,r11 + addi r3,r3,64 + +9: bf cr7*4+2,10f + lvx vr1,r0,r4 + vperm vr8,vr0,vr1,vr16 + lvx vr0,r4,r9 + vperm vr9,vr1,vr0,vr16 + addi r4,r4,32 + stvx vr8,r0,r3 + stvx vr9,r3,r9 + addi r3,r3,32 + +10: bf cr7*4+3,11f + lvx vr1,r0,r4 + vperm vr8,vr0,vr1,vr16 + addi r4,r4,16 + stvx vr8,r0,r3 + addi r3,r3,16 + + /* Up to 15B to go */ +11: clrldi r5,r5,(64-4) + addi r4,r4,-16 /* Unwind the +16 load offset */ + mtocrf 0x01,r5 + bf cr7*4+0,12f + lwz r0,0(r4) /* Less chance of a reject with word ops */ + lwz r6,4(r4) + addi r4,r4,8 + stw r0,0(r3) + stw r6,4(r3) + addi r3,r3,8 + +12: bf cr7*4+1,13f + lwz r0,0(r4) + addi r4,r4,4 + stw r0,0(r3) + addi r3,r3,4 + +13: bf cr7*4+2,14f + lhz r0,0(r4) + addi r4,r4,2 + sth r0,0(r3) + addi r3,r3,2 + +14: bf cr7*4+3,15f + lbz r0,0(r4) + stb r0,0(r3) + +15: addi r1,r1,STACKFRAMESIZE + ld r3,48(r1) + b .exit_vmx_copy /* tail call optimise */ +#endif /* CONFiG_ALTIVEC */ diff --git a/arch/powerpc/lib/string.S b/arch/powerpc/lib/string.S index 093d6316435..1b5a0a09d60 100644 --- a/arch/powerpc/lib/string.S +++ b/arch/powerpc/lib/string.S @@ -119,6 +119,7 @@ _GLOBAL(memchr) 2: li r3,0 blr +#ifdef CONFIG_PPC32 _GLOBAL(__clear_user) addi r6,r3,-4 li r3,0 @@ -160,3 +161,4 @@ _GLOBAL(__clear_user) PPC_LONG 1b,91b PPC_LONG 8b,92b .text +#endif diff --git a/arch/powerpc/lib/string_64.S b/arch/powerpc/lib/string_64.S new file mode 100644 index 00000000000..3b1e48049fa --- /dev/null +++ b/arch/powerpc/lib/string_64.S @@ -0,0 +1,202 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2012 + * + * Author: Anton Blanchard <anton@au.ibm.com> + */ + +#include <asm/ppc_asm.h> +#include <asm/asm-offsets.h> + + .section ".toc","aw" +PPC64_CACHES: + .tc ppc64_caches[TC],ppc64_caches + .section ".text" + +/** + * __clear_user: - Zero a block of memory in user space, with less checking. + * @to: Destination address, in user space. + * @n: Number of bytes to zero. + * + * Zero a block of memory in user space. Caller must check + * the specified block with access_ok() before calling this function. + * + * Returns number of bytes that could not be cleared. + * On success, this will be zero. + */ + + .macro err1 +100: + .section __ex_table,"a" + .align 3 + .llong 100b,.Ldo_err1 + .previous + .endm + + .macro err2 +200: + .section __ex_table,"a" + .align 3 + .llong 200b,.Ldo_err2 + .previous + .endm + + .macro err3 +300: + .section __ex_table,"a" + .align 3 + .llong 300b,.Ldo_err3 + .previous + .endm + +.Ldo_err1: + mr r3,r8 + +.Ldo_err2: + mtctr r4 +1: +err3; stb r0,0(r3) + addi r3,r3,1 + addi r4,r4,-1 + bdnz 1b + +.Ldo_err3: + mr r3,r4 + blr + +_GLOBAL(__clear_user) + cmpdi r4,32 + neg r6,r3 + li r0,0 + blt .Lshort_clear + mr r8,r3 + mtocrf 0x01,r6 + clrldi r6,r6,(64-3) + + /* Get the destination 8 byte aligned */ + bf cr7*4+3,1f +err1; stb r0,0(r3) + addi r3,r3,1 + +1: bf cr7*4+2,2f +err1; sth r0,0(r3) + addi r3,r3,2 + +2: bf cr7*4+1,3f +err1; stw r0,0(r3) + addi r3,r3,4 + +3: sub r4,r4,r6 + + cmpdi r4,32 + cmpdi cr1,r4,512 + blt .Lshort_clear + bgt cr1,.Llong_clear + +.Lmedium_clear: + srdi r6,r4,5 + mtctr r6 + + /* Do 32 byte chunks */ +4: +err2; std r0,0(r3) +err2; std r0,8(r3) +err2; std r0,16(r3) +err2; std r0,24(r3) + addi r3,r3,32 + addi r4,r4,-32 + bdnz 4b + +.Lshort_clear: + /* up to 31 bytes to go */ + cmpdi r4,16 + blt 6f +err2; std r0,0(r3) +err2; std r0,8(r3) + addi r3,r3,16 + addi r4,r4,-16 + + /* Up to 15 bytes to go */ +6: mr r8,r3 + clrldi r4,r4,(64-4) + mtocrf 0x01,r4 + bf cr7*4+0,7f +err1; std r0,0(r3) + addi r3,r3,8 + +7: bf cr7*4+1,8f +err1; stw r0,0(r3) + addi r3,r3,4 + +8: bf cr7*4+2,9f +err1; sth r0,0(r3) + addi r3,r3,2 + +9: bf cr7*4+3,10f +err1; stb r0,0(r3) + +10: li r3,0 + blr + +.Llong_clear: + ld r5,PPC64_CACHES@toc(r2) + + bf cr7*4+0,11f +err2; std r0,0(r3) + addi r3,r3,8 + addi r4,r4,-8 + + /* Destination is 16 byte aligned, need to get it cacheline aligned */ +11: lwz r7,DCACHEL1LOGLINESIZE(r5) + lwz r9,DCACHEL1LINESIZE(r5) + + /* + * With worst case alignment the long clear loop takes a minimum + * of 1 byte less than 2 cachelines. + */ + sldi r10,r9,2 + cmpd r4,r10 + blt .Lmedium_clear + + neg r6,r3 + addi r10,r9,-1 + and. r5,r6,r10 + beq 13f + + srdi r6,r5,4 + mtctr r6 + mr r8,r3 +12: +err1; std r0,0(r3) +err1; std r0,8(r3) + addi r3,r3,16 + bdnz 12b + + sub r4,r4,r5 + +13: srd r6,r4,r7 + mtctr r6 + mr r8,r3 +14: +err1; dcbz r0,r3 + add r3,r3,r9 + bdnz 14b + + and r4,r4,r10 + + cmpdi r4,32 + blt .Lshort_clear + b .Lmedium_clear diff --git a/arch/powerpc/lib/copyuser_power7_vmx.c b/arch/powerpc/lib/vmx-helper.c index bf2654f2b68..3cf529ceec5 100644 --- a/arch/powerpc/lib/copyuser_power7_vmx.c +++ b/arch/powerpc/lib/vmx-helper.c @@ -22,7 +22,7 @@ #include <linux/hardirq.h> #include <asm/switch_to.h> -int enter_vmx_copy(void) +int enter_vmx_usercopy(void) { if (in_interrupt()) return 0; @@ -44,8 +44,31 @@ int enter_vmx_copy(void) * This function must return 0 because we tail call optimise when calling * from __copy_tofrom_user_power7 which returns 0 on success. */ -int exit_vmx_copy(void) +int exit_vmx_usercopy(void) { pagefault_enable(); return 0; } + +int enter_vmx_copy(void) +{ + if (in_interrupt()) + return 0; + + preempt_disable(); + + enable_kernel_altivec(); + + return 1; +} + +/* + * All calls to this function will be optimised into tail calls. We are + * passed a pointer to the destination which we return as required by a + * memcpy implementation. + */ +void *exit_vmx_copy(void *dest) +{ + preempt_enable(); + return dest; +} |