diff options
Diffstat (limited to 'arch/s390/kernel/vdso64/vgetrandom-chacha.S')
| -rw-r--r-- | arch/s390/kernel/vdso64/vgetrandom-chacha.S | 76 |
1 files changed, 36 insertions, 40 deletions
diff --git a/arch/s390/kernel/vdso64/vgetrandom-chacha.S b/arch/s390/kernel/vdso64/vgetrandom-chacha.S index d802b0a96f41..09c034c2f853 100644 --- a/arch/s390/kernel/vdso64/vgetrandom-chacha.S +++ b/arch/s390/kernel/vdso64/vgetrandom-chacha.S @@ -1,7 +1,9 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/stringify.h> #include <linux/linkage.h> #include <asm/alternative.h> +#include <asm/dwarf.h> #include <asm/fpu-insn.h> #define STATE0 %v0 @@ -12,9 +14,6 @@ #define COPY1 %v5 #define COPY2 %v6 #define COPY3 %v7 -#define PERM4 %v16 -#define PERM8 %v17 -#define PERM12 %v18 #define BEPERM %v19 #define TMP0 %v20 #define TMP1 %v21 @@ -23,13 +22,11 @@ .section .rodata - .balign 128 -.Lconstants: + .balign 32 +SYM_DATA_START_LOCAL(chacha20_constants) .long 0x61707865,0x3320646e,0x79622d32,0x6b206574 # endian-neutral - .long 0x04050607,0x08090a0b,0x0c0d0e0f,0x00010203 # rotl 4 bytes - .long 0x08090a0b,0x0c0d0e0f,0x00010203,0x04050607 # rotl 8 bytes - .long 0x0c0d0e0f,0x00010203,0x04050607,0x08090a0b # rotl 12 bytes .long 0x03020100,0x07060504,0x0b0a0908,0x0f0e0d0c # byte swap +SYM_DATA_END(chacha20_constants) .text /* @@ -43,13 +40,14 @@ * size_t nblocks) */ SYM_FUNC_START(__arch_chacha20_blocks_nostack) - larl %r1,.Lconstants + CFI_STARTPROC + larl %r1,chacha20_constants /* COPY0 = "expand 32-byte k" */ VL COPY0,0,,%r1 - /* PERM4-PERM12,BEPERM = byte selectors for VPERM */ - VLM PERM4,BEPERM,16,%r1 + /* BEPERM = byte selectors for VPERM */ + ALTERNATIVE __stringify(VL BEPERM,16,,%r1), "brcl 0,0", ALT_FACILITY(148) /* COPY1,COPY2 = key */ VLM COPY1,COPY2,0,%r3 @@ -89,11 +87,11 @@ SYM_FUNC_START(__arch_chacha20_blocks_nostack) VERLLF STATE1,STATE1,7 /* STATE1[0,1,2,3] = STATE1[1,2,3,0] */ - VPERM STATE1,STATE1,STATE1,PERM4 + VSLDB STATE1,STATE1,STATE1,4 /* STATE2[0,1,2,3] = STATE2[2,3,0,1] */ - VPERM STATE2,STATE2,STATE2,PERM8 + VSLDB STATE2,STATE2,STATE2,8 /* STATE3[0,1,2,3] = STATE3[3,0,1,2] */ - VPERM STATE3,STATE3,STATE3,PERM12 + VSLDB STATE3,STATE3,STATE3,12 /* STATE0 += STATE1, STATE3 = rotl32(STATE3 ^ STATE0, 16) */ VAF STATE0,STATE0,STATE1 @@ -116,32 +114,38 @@ SYM_FUNC_START(__arch_chacha20_blocks_nostack) VERLLF STATE1,STATE1,7 /* STATE1[0,1,2,3] = STATE1[3,0,1,2] */ - VPERM STATE1,STATE1,STATE1,PERM12 + VSLDB STATE1,STATE1,STATE1,12 /* STATE2[0,1,2,3] = STATE2[2,3,0,1] */ - VPERM STATE2,STATE2,STATE2,PERM8 + VSLDB STATE2,STATE2,STATE2,8 /* STATE3[0,1,2,3] = STATE3[1,2,3,0] */ - VPERM STATE3,STATE3,STATE3,PERM4 + VSLDB STATE3,STATE3,STATE3,4 brctg %r0,.Ldoubleround - /* OUTPUT0 = STATE0 + STATE0 */ + /* OUTPUT0 = STATE0 + COPY0 */ VAF STATE0,STATE0,COPY0 - /* OUTPUT1 = STATE1 + STATE1 */ + /* OUTPUT1 = STATE1 + COPY1 */ VAF STATE1,STATE1,COPY1 - /* OUTPUT2 = STATE2 + STATE2 */ + /* OUTPUT2 = STATE2 + COPY2 */ VAF STATE2,STATE2,COPY2 - /* OUTPUT2 = STATE3 + STATE3 */ + /* OUTPUT3 = STATE3 + COPY3 */ VAF STATE3,STATE3,COPY3 - /* - * 32 bit wise little endian store to OUTPUT. If the vector - * enhancement facility 2 is not installed use the slow path. - */ - ALTERNATIVE "brc 0xf,.Lstoreslow", "nop", ALT_FACILITY(148) - VSTBRF STATE0,0,,%r2 - VSTBRF STATE1,16,,%r2 - VSTBRF STATE2,32,,%r2 - VSTBRF STATE3,48,,%r2 -.Lstoredone: + ALTERNATIVE \ + __stringify( \ + /* Convert STATE to little endian and store to OUTPUT */\ + VPERM TMP0,STATE0,STATE0,BEPERM; \ + VPERM TMP1,STATE1,STATE1,BEPERM; \ + VPERM TMP2,STATE2,STATE2,BEPERM; \ + VPERM TMP3,STATE3,STATE3,BEPERM; \ + VSTM TMP0,TMP3,0,%r2), \ + __stringify( \ + /* 32 bit wise little endian store to OUTPUT */ \ + VSTBRF STATE0,0,,%r2; \ + VSTBRF STATE1,16,,%r2; \ + VSTBRF STATE2,32,,%r2; \ + VSTBRF STATE3,48,,%r2; \ + brcl 0,0), \ + ALT_FACILITY(148) /* ++COPY3.COUNTER */ /* alsih %r3,1 */ @@ -173,13 +177,5 @@ SYM_FUNC_START(__arch_chacha20_blocks_nostack) VZERO TMP3 br %r14 - -.Lstoreslow: - /* Convert STATE to little endian format and store to OUTPUT */ - VPERM TMP0,STATE0,STATE0,BEPERM - VPERM TMP1,STATE1,STATE1,BEPERM - VPERM TMP2,STATE2,STATE2,BEPERM - VPERM TMP3,STATE3,STATE3,BEPERM - VSTM TMP0,TMP3,0,%r2 - j .Lstoredone + CFI_ENDPROC SYM_FUNC_END(__arch_chacha20_blocks_nostack) |