diff options
author | Jamie Liu <jamieliu@google.com> | 2020-04-09 11:02:05 -0700 |
---|---|---|
committer | gVisor bot <gvisor-bot@google.com> | 2020-04-09 11:03:43 -0700 |
commit | 8f68be74919751775cfb3d8162a51351822a6f5f (patch) | |
tree | ecc1aa0b4aa295dde781ee032eb9326e7ba913f8 /pkg | |
parent | 7928aa345e334f2c68f8f03b71d8cabe79e8db7e (diff) |
Don't use REP string instructions in safecopy.memcpy.
PiperOrigin-RevId: 305718392
Diffstat (limited to 'pkg')
-rw-r--r-- | pkg/safecopy/memcpy_amd64.s | 111 |
1 files changed, 40 insertions, 71 deletions
diff --git a/pkg/safecopy/memcpy_amd64.s b/pkg/safecopy/memcpy_amd64.s index 129691d68..00b46c18f 100644 --- a/pkg/safecopy/memcpy_amd64.s +++ b/pkg/safecopy/memcpy_amd64.s @@ -55,15 +55,9 @@ TEXT ·memcpy(SB), NOSPLIT, $0-36 MOVQ from+8(FP), SI MOVQ n+16(FP), BX - // REP instructions have a high startup cost, so we handle small sizes - // with some straightline code. The REP MOVSQ instruction is really fast - // for large sizes. The cutover is approximately 2K. tail: - // move_129through256 or smaller work whether or not the source and the - // destination memory regions overlap because they load all data into - // registers before writing it back. move_256through2048 on the other - // hand can be used only when the memory regions don't overlap or the copy - // direction is forward. + // BSR+branch table make almost all memmove/memclr benchmarks worse. Not + // worth doing. TESTQ BX, BX JEQ move_0 CMPQ BX, $2 @@ -83,31 +77,45 @@ tail: JBE move_65through128 CMPQ BX, $256 JBE move_129through256 - // TODO: use branch table and BSR to make this just a single dispatch -/* - * forward copy loop - */ - CMPQ BX, $2048 - JLS move_256through2048 - - // Check alignment - MOVL SI, AX - ORL DI, AX - TESTL $7, AX - JEQ fwdBy8 - - // Do 1 byte at a time - MOVQ BX, CX - REP; MOVSB - RET - -fwdBy8: - // Do 8 bytes at a time - MOVQ BX, CX - SHRQ $3, CX - ANDQ $7, BX - REP; MOVSQ +move_257plus: + SUBQ $256, BX + MOVOU (SI), X0 + MOVOU X0, (DI) + MOVOU 16(SI), X1 + MOVOU X1, 16(DI) + MOVOU 32(SI), X2 + MOVOU X2, 32(DI) + MOVOU 48(SI), X3 + MOVOU X3, 48(DI) + MOVOU 64(SI), X4 + MOVOU X4, 64(DI) + MOVOU 80(SI), X5 + MOVOU X5, 80(DI) + MOVOU 96(SI), X6 + MOVOU X6, 96(DI) + MOVOU 112(SI), X7 + MOVOU X7, 112(DI) + MOVOU 128(SI), X8 + MOVOU X8, 128(DI) + MOVOU 144(SI), X9 + MOVOU X9, 144(DI) + MOVOU 160(SI), X10 + MOVOU X10, 160(DI) + MOVOU 176(SI), X11 + MOVOU X11, 176(DI) + MOVOU 192(SI), X12 + MOVOU X12, 192(DI) + MOVOU 208(SI), X13 + MOVOU X13, 208(DI) + MOVOU 224(SI), X14 + MOVOU X14, 224(DI) + MOVOU 240(SI), X15 + MOVOU X15, 240(DI) + CMPQ BX, $256 + LEAQ 256(SI), SI + LEAQ 256(DI), DI + JGE move_257plus JMP tail move_1or2: @@ -209,42 +217,3 @@ move_129through256: MOVOU -16(SI)(BX*1), X15 MOVOU X15, -16(DI)(BX*1) RET -move_256through2048: - SUBQ $256, BX - MOVOU (SI), X0 - MOVOU X0, (DI) - MOVOU 16(SI), X1 - MOVOU X1, 16(DI) - MOVOU 32(SI), X2 - MOVOU X2, 32(DI) - MOVOU 48(SI), X3 - MOVOU X3, 48(DI) - MOVOU 64(SI), X4 - MOVOU X4, 64(DI) - MOVOU 80(SI), X5 - MOVOU X5, 80(DI) - MOVOU 96(SI), X6 - MOVOU X6, 96(DI) - MOVOU 112(SI), X7 - MOVOU X7, 112(DI) - MOVOU 128(SI), X8 - MOVOU X8, 128(DI) - MOVOU 144(SI), X9 - MOVOU X9, 144(DI) - MOVOU 160(SI), X10 - MOVOU X10, 160(DI) - MOVOU 176(SI), X11 - MOVOU X11, 176(DI) - MOVOU 192(SI), X12 - MOVOU X12, 192(DI) - MOVOU 208(SI), X13 - MOVOU X13, 208(DI) - MOVOU 224(SI), X14 - MOVOU X14, 224(DI) - MOVOU 240(SI), X15 - MOVOU X15, 240(DI) - CMPQ BX, $256 - LEAQ 256(SI), SI - LEAQ 256(DI), DI - JGE move_256through2048 - JMP tail |