qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-11-23 10:00:03 -07:00
Code Releases Activity
c71acbfe83
go/test/codegen/arithmetic.go
erifan01 600259b099 cmd/compile: use depth first topological sort algorithm for layout 
The current layout algorithm tries to put consecutive blocks together,
so the priority of the successor block is higher than the priority of
the zero indegree block. This algorithm is beneficial for subsequent
register allocation, but will result in more branch instructions.
The depth-first topological sorting algorithm is a well-known layout
algorithm, which has applications in many languages, and it helps to
reduce branch instructions. This CL applies it to the layout pass.
The test results show that it helps to reduce the code size.

This CL also includes the following changes:
1, Removed the primary predecessor mechanism. The new layout algorithm is
  not very friendly to register allocator in some cases, in order to adapt
  to the new layout algorithm, a new primary predecessor selection strategy
  is introduced.
2, Since the new layout implementation may place non-loop blocks between
  loop blocks, some adaptive modifications have also been made to looprotate
  pass.
3, The layout also affects the results of codegen, so this CL also adjusted
  several codegen tests accordingly.

It is inevitable that this CL will cause the code size or performance of a
few functions to decrease, but the number of cases it improves is much larger
than the number of cases it drops.

Statistical data from compilecmp on linux/amd64 is as follow:
name                      old time/op       new time/op       delta
Template                        382ms ± 4%        382ms ± 4%    ~     (p=0.497 n=49+50)
Unicode                         170ms ± 9%        169ms ± 8%    ~     (p=0.344 n=48+50)
GoTypes                         2.01s ± 4%        2.01s ± 4%    ~     (p=0.628 n=50+48)
Compiler                        190ms ±10%        189ms ± 9%    ~     (p=0.734 n=50+50)
SSA                             11.8s ± 2%        11.8s ± 3%    ~     (p=0.877 n=50+50)
Flate                           241ms ± 9%        241ms ± 8%    ~     (p=0.897 n=50+49)
GoParser                        366ms ± 3%        361ms ± 4%  -1.21%  (p=0.004 n=47+50)
Reflect                         835ms ± 3%        838ms ± 3%    ~     (p=0.275 n=50+49)
Tar                             336ms ± 4%        335ms ± 3%    ~     (p=0.454 n=48+48)
XML                             433ms ± 4%        431ms ± 3%    ~     (p=0.071 n=49+48)
LinkCompiler                    706ms ± 4%        705ms ± 4%    ~     (p=0.608 n=50+49)
ExternalLinkCompiler            1.85s ± 3%        1.83s ± 2%  -1.47%  (p=0.000 n=49+48)
LinkWithoutDebugCompiler        437ms ± 5%        437ms ± 6%    ~     (p=0.953 n=49+50)
[Geo mean]                      615ms             613ms       -0.37%

name                      old alloc/op      new alloc/op      delta
Template                       38.7MB ± 1%       38.7MB ± 1%    ~     (p=0.834 n=50+50)
Unicode                        28.1MB ± 0%       28.1MB ± 0%  -0.22%  (p=0.000 n=49+50)
GoTypes                         168MB ± 1%        168MB ± 1%    ~     (p=0.054 n=47+47)
Compiler                       23.0MB ± 1%       23.0MB ± 1%    ~     (p=0.432 n=50+50)
SSA                            1.54GB ± 0%       1.54GB ± 0%  +0.21%  (p=0.000 n=50+50)
Flate                          23.6MB ± 1%       23.6MB ± 1%    ~     (p=0.153 n=43+46)
GoParser                       35.1MB ± 1%       35.1MB ± 2%    ~     (p=0.202 n=50+50)
Reflect                        84.7MB ± 1%       84.7MB ± 1%    ~     (p=0.333 n=48+49)
Tar                            34.5MB ± 1%       34.5MB ± 1%    ~     (p=0.406 n=46+49)
XML                            44.3MB ± 2%       44.2MB ± 3%    ~     (p=0.981 n=50+50)
LinkCompiler                    131MB ± 0%        128MB ± 0%  -2.74%  (p=0.000 n=50+50)
ExternalLinkCompiler            120MB ± 0%        120MB ± 0%  +0.01%  (p=0.007 n=50+50)
LinkWithoutDebugCompiler       77.3MB ± 0%       77.3MB ± 0%  -0.02%  (p=0.000 n=50+50)
[Geo mean]                     69.3MB            69.1MB       -0.22%

file      before    after     Δ        %
addr2line 4104220   4043684   -60536   -1.475%
api       5342502   5249678   -92824   -1.737%
asm       4973785   4858257   -115528  -2.323%
buildid   2667844   2625660   -42184   -1.581%
cgo       4686849   4616313   -70536   -1.505%
compile   23667431  23268406  -399025  -1.686%
cover     4959676   4874108   -85568   -1.725%
dist      3515934   3450422   -65512   -1.863%
doc       3995581   3925469   -70112   -1.755%
fix       3379202   3318522   -60680   -1.796%
link      6743249   6629913   -113336  -1.681%
nm        4047529   3991777   -55752   -1.377%
objdump   4456151   4388151   -68000   -1.526%
pack      2435040   2398072   -36968   -1.518%
pprof     13804080  13565808  -238272  -1.726%
test2json 2690043   2645987   -44056   -1.638%
trace     10418492  10232716  -185776  -1.783%
vet       7258259   7121259   -137000  -1.888%
total     113145867 111204202 -1941665 -1.716%

The situation on linux/arm64 is as follow:
name                      old time/op       new time/op       delta
Template                        280ms ± 1%        282ms ± 1%  +0.75%  (p=0.000 n=46+48)
Unicode                         124ms ± 2%        124ms ± 2%  +0.37%  (p=0.045 n=50+50)
GoTypes                         1.69s ± 1%        1.70s ± 1%  +0.56%  (p=0.000 n=49+50)
Compiler                        122ms ± 1%        123ms ± 1%  +0.93%  (p=0.000 n=50+50)
SSA                             12.6s ± 1%        12.7s ± 0%  +0.72%  (p=0.000 n=50+50)
Flate                           170ms ± 1%        172ms ± 1%  +0.97%  (p=0.000 n=49+49)
GoParser                        262ms ± 1%        263ms ± 1%  +0.39%  (p=0.000 n=49+48)
Reflect                         639ms ± 1%        650ms ± 1%  +1.63%  (p=0.000 n=49+49)
Tar                             243ms ± 1%        245ms ± 1%  +0.82%  (p=0.000 n=50+50)
XML                             324ms ± 1%        327ms ± 1%  +0.72%  (p=0.000 n=50+49)
LinkCompiler                    597ms ± 1%        596ms ± 1%  -0.27%  (p=0.001 n=48+47)
ExternalLinkCompiler            1.90s ± 1%        1.88s ± 1%  -1.00%  (p=0.000 n=50+50)
LinkWithoutDebugCompiler        364ms ± 1%        363ms ± 1%    ~     (p=0.220 n=49+50)
[Geo mean]                      485ms             488ms       +0.49%

name                      old alloc/op      new alloc/op      delta
Template                       38.7MB ± 0%       38.8MB ± 1%    ~     (p=0.093 n=43+49)
Unicode                        28.4MB ± 0%       28.4MB ± 0%  +0.03%  (p=0.000 n=49+45)
GoTypes                         169MB ± 1%        169MB ± 1%  +0.23%  (p=0.010 n=50+50)
Compiler                       23.2MB ± 1%       23.2MB ± 1%  +0.11%  (p=0.000 n=40+44)
SSA                            1.54GB ± 0%       1.55GB ± 0%  +0.45%  (p=0.000 n=47+49)
Flate                          23.8MB ± 2%       23.8MB ± 1%    ~     (p=0.543 n=50+50)
GoParser                       35.3MB ± 1%       35.4MB ± 1%    ~     (p=0.792 n=50+50)
Reflect                        85.2MB ± 1%       85.2MB ± 0%    ~     (p=0.055 n=50+47)
Tar                            34.5MB ± 1%       34.5MB ± 1%  +0.06%  (p=0.015 n=50+50)
XML                            43.8MB ± 2%       43.9MB ± 2%  +0.19%  (p=0.000 n=48+48)
LinkCompiler                    137MB ± 0%        136MB ± 0%  -0.92%  (p=0.000 n=50+50)
ExternalLinkCompiler            127MB ± 0%        127MB ± 0%    ~     (p=0.516 n=50+50)
LinkWithoutDebugCompiler       84.0MB ± 0%       84.0MB ± 0%    ~     (p=0.057 n=50+50)
[Geo mean]                     70.4MB            70.4MB       +0.01%

file      before    after     Δ        %
addr2line 4021557   4002933   -18624   -0.463%
api       5127847   5028503   -99344   -1.937%
asm       5034716   4936836   -97880   -1.944%
buildid   2608118   2594094   -14024   -0.538%
cgo       4488592   4398320   -90272   -2.011%
compile   22501129  22213592  -287537  -1.278%
cover     4742301   4713573   -28728   -0.606%
dist      3388071   3365311   -22760   -0.672%
doc       3802250   3776082   -26168   -0.688%
fix       3306147   3216939   -89208   -2.698%
link      6404483   6363699   -40784   -0.637%
nm        3941026   3921930   -19096   -0.485%
objdump   4383330   4295122   -88208   -2.012%
pack      2404547   2389515   -15032   -0.625%
pprof     12996234  12856818  -139416  -1.073%
test2json 2668500   2586788   -81712   -3.062%
trace     9816276   9609580   -206696  -2.106%
vet       6900682   6787338   -113344  -1.643%
total     108535806 107056973 -1478833 -1.363%

Change-Id: Iaec1cdcaacca8025e9babb0fb8a532fddb70c87d
Reviewed-on: https://go-review.googlesource.com/c/go/+/255239
Reviewed-by: eric fang <eric.fang@arm.com>
Reviewed-by: Keith Randall <khr@golang.org>
Trust: eric fang <eric.fang@arm.com>
2021-03-16 02:44:54 +00:00

554 lines
13 KiB
Go
Raw Blame History

// asmcheck
// Copyright 2018 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package codegen
// This file contains codegen tests related to arithmetic
// simplifications and optimizations on integer types.
// For codegen tests on float types, see floats.go.
// ----------------- //
// Subtraction //
// ----------------- //
var ef int
func SubMem(arr []int, b, c, d int) int {
// 386:`SUBL\s[A-Z]+,\s8\([A-Z]+\)`
// amd64:`SUBQ\s[A-Z]+,\s16\([A-Z]+\)`
arr[2] -= b
// 386:`SUBL\s[A-Z]+,\s12\([A-Z]+\)`
// amd64:`SUBQ\s[A-Z]+,\s24\([A-Z]+\)`
arr[3] -= b
// 386:`DECL\s16\([A-Z]+\)`
arr[4]--
// 386:`ADDL\s[$]-20,\s20\([A-Z]+\)`
arr[5] -= 20
// 386:`SUBL\s\([A-Z]+\)\([A-Z]+\*4\),\s[A-Z]+`
ef -= arr[b]
// 386:`SUBL\s[A-Z]+,\s\([A-Z]+\)\([A-Z]+\*4\)`
arr[c] -= b
// 386:`ADDL\s[$]-15,\s\([A-Z]+\)\([A-Z]+\*4\)`
arr[d] -= 15
// 386:`DECL\s\([A-Z]+\)\([A-Z]+\*4\)`
arr[b]--
// amd64:`DECQ\s64\([A-Z]+\)`
arr[8]--
// 386:"SUBL\t4"
// amd64:"SUBQ\t8"
return arr[0] - arr[1]
}
func SubFromConst(a int) int {
// ppc64le: `SUBC\tR[0-9]+,\s[$]40,\sR`
// ppc64: `SUBC\tR[0-9]+,\s[$]40,\sR`
b := 40 - a
return b
}
func SubFromConstNeg(a int) int {
// ppc64le: `ADD\t[$]40,\sR[0-9]+,\sR`
// ppc64: `ADD\t[$]40,\sR[0-9]+,\sR`
c := 40 - (-a)
return c
}
func SubSubFromConst(a int) int {
// ppc64le: `ADD\t[$]20,\sR[0-9]+,\sR`
// ppc64: `ADD\t[$]20,\sR[0-9]+,\sR`
c := 40 - (20 - a)
return c
}
func AddSubFromConst(a int) int {
// ppc64le: `SUBC\tR[0-9]+,\s[$]60,\sR`
// ppc64: `SUBC\tR[0-9]+,\s[$]60,\sR`
c := 40 + (20 - a)
return c
}
func NegSubFromConst(a int) int {
// ppc64le: `ADD\t[$]-20,\sR[0-9]+,\sR`
// ppc64: `ADD\t[$]-20,\sR[0-9]+,\sR`
c := -(20 - a)
return c
}
func NegAddFromConstNeg(a int) int {
// ppc64le: `SUBC\tR[0-9]+,\s[$]40,\sR`
// ppc64: `SUBC\tR[0-9]+,\s[$]40,\sR`
c := -(-40 + a)
return c
}
// -------------------- //
// Multiplication //
// -------------------- //
func Pow2Muls(n1, n2 int) (int, int) {
// amd64:"SHLQ\t[$]5",-"IMULQ"
// 386:"SHLL\t[$]5",-"IMULL"
// arm:"SLL\t[$]5",-"MUL"
// arm64:"LSL\t[$]5",-"MUL"
// ppc64:"SLD\t[$]5",-"MUL"
// ppc64le:"SLD\t[$]5",-"MUL"
a := n1 * 32
// amd64:"SHLQ\t[$]6",-"IMULQ"
// 386:"SHLL\t[$]6",-"IMULL"
// arm:"SLL\t[$]6",-"MUL"
// arm64:`NEG\sR[0-9]+<<6,\sR[0-9]+`,-`LSL`,-`MUL`
// ppc64:"SLD\t[$]6","NEG\\sR[0-9]+,\\sR[0-9]+",-"MUL"
// ppc64le:"SLD\t[$]6","NEG\\sR[0-9]+,\\sR[0-9]+",-"MUL"
b := -64 * n2
return a, b
}
func Mul_96(n int) int {
// amd64:`SHLQ\t[$]5`,`LEAQ\t\(.*\)\(.*\*2\),`,-`IMULQ`
// 386:`SHLL\t[$]5`,`LEAL\t\(.*\)\(.*\*2\),`,-`IMULL`
// arm64:`LSL\t[$]5`,`ADD\sR[0-9]+<<1,\sR[0-9]+`,-`MUL`
// arm:`SLL\t[$]5`,`ADD\sR[0-9]+<<1,\sR[0-9]+`,-`MUL`
// s390x:`SLD\t[$]5`,`SLD\t[$]6`,-`MULLD`
return n * 96
}
func Mul_n120(n int) int {
// s390x:`SLD\t[$]3`,`SLD\t[$]7`,-`MULLD`
return n * -120
}
func MulMemSrc(a []uint32, b []float32) {
// 386:`IMULL\s4\([A-Z]+\),\s[A-Z]+`
a[0] *= a[1]
// 386/sse2:`MULSS\s4\([A-Z]+\),\sX[0-9]+`
// amd64:`MULSS\s4\([A-Z]+\),\sX[0-9]+`
b[0] *= b[1]
}
// Multiplications merging tests
func MergeMuls1(n int) int {
// amd64:"IMUL3Q\t[$]46"
// 386:"IMUL3L\t[$]46"
return 15*n + 31*n // 46n
}
func MergeMuls2(n int) int {
// amd64:"IMUL3Q\t[$]23","ADDQ\t[$]29"
// 386:"IMUL3L\t[$]23","ADDL\t[$]29"
return 5*n + 7*(n+1) + 11*(n+2) // 23n + 29
}
func MergeMuls3(a, n int) int {
// amd64:"ADDQ\t[$]19",-"IMULQ\t[$]19"
// 386:"ADDL\t[$]19",-"IMULL\t[$]19"
return a*n + 19*n // (a+19)n
}
func MergeMuls4(n int) int {
// amd64:"IMUL3Q\t[$]14"
// 386:"IMUL3L\t[$]14"
return 23*n - 9*n // 14n
}
func MergeMuls5(a, n int) int {
// amd64:"ADDQ\t[$]-19",-"IMULQ\t[$]19"
// 386:"ADDL\t[$]-19",-"IMULL\t[$]19"
return a*n - 19*n // (a-19)n
}
// -------------- //
// Division //
// -------------- //
func DivMemSrc(a []float64) {
// 386/sse2:`DIVSD\s8\([A-Z]+\),\sX[0-9]+`
// amd64:`DIVSD\s8\([A-Z]+\),\sX[0-9]+`
a[0] /= a[1]
}
func Pow2Divs(n1 uint, n2 int) (uint, int) {
// 386:"SHRL\t[$]5",-"DIVL"
// amd64:"SHRQ\t[$]5",-"DIVQ"
// arm:"SRL\t[$]5",-".*udiv"
// arm64:"LSR\t[$]5",-"UDIV"
// ppc64:"SRD"
// ppc64le:"SRD"
a := n1 / 32 // unsigned
// amd64:"SARQ\t[$]6",-"IDIVQ"
// 386:"SARL\t[$]6",-"IDIVL"
// arm:"SRA\t[$]6",-".*udiv"
// arm64:"ASR\t[$]6",-"SDIV"
// ppc64:"SRAD"
// ppc64le:"SRAD"
b := n2 / 64 // signed
return a, b
}
// Check that constant divisions get turned into MULs
func ConstDivs(n1 uint, n2 int) (uint, int) {
// amd64:"MOVQ\t[$]-1085102592571150095","MULQ",-"DIVQ"
// 386:"MOVL\t[$]-252645135","MULL",-"DIVL"
// arm64:`MOVD`,`UMULH`,-`DIV`
// arm:`MOVW`,`MUL`,-`.*udiv`
a := n1 / 17 // unsigned
// amd64:"MOVQ\t[$]-1085102592571150095","IMULQ",-"IDIVQ"
// 386:"MOVL\t[$]-252645135","IMULL",-"IDIVL"
// arm64:`MOVD`,`SMULH`,-`DIV`
// arm:`MOVW`,`MUL`,-`.*udiv`
b := n2 / 17 // signed
return a, b
}
func FloatDivs(a []float32) float32 {
// amd64:`DIVSS\s8\([A-Z]+\),\sX[0-9]+`
// 386/sse2:`DIVSS\s8\([A-Z]+\),\sX[0-9]+`
return a[1] / a[2]
}
func Pow2Mods(n1 uint, n2 int) (uint, int) {
// 386:"ANDL\t[$]31",-"DIVL"
// amd64:"ANDQ\t[$]31",-"DIVQ"
// arm:"AND\t[$]31",-".*udiv"
// arm64:"AND\t[$]31",-"UDIV"
// ppc64:"ANDCC\t[$]31"
// ppc64le:"ANDCC\t[$]31"
a := n1 % 32 // unsigned
// 386:"SHRL",-"IDIVL"
// amd64:"SHRQ",-"IDIVQ"
// arm:"SRA",-".*udiv"
// arm64:"ASR",-"REM"
// ppc64:"SRAD"
// ppc64le:"SRAD"
b := n2 % 64 // signed
return a, b
}
// Check that signed divisibility checks get converted to AND on low bits
func Pow2DivisibleSigned(n1, n2 int) (bool, bool) {
// 386:"TESTL\t[$]63",-"DIVL",-"SHRL"
// amd64:"TESTQ\t[$]63",-"DIVQ",-"SHRQ"
// arm:"AND\t[$]63",-".*udiv",-"SRA"
// arm64:"AND\t[$]63",-"UDIV",-"ASR"
// ppc64:"ANDCC\t[$]63",-"SRAD"
// ppc64le:"ANDCC\t[$]63",-"SRAD"
a := n1%64 == 0 // signed divisible
// 386:"TESTL\t[$]63",-"DIVL",-"SHRL"
// amd64:"TESTQ\t[$]63",-"DIVQ",-"SHRQ"
// arm:"AND\t[$]63",-".*udiv",-"SRA"
// arm64:"AND\t[$]63",-"UDIV",-"ASR"
// ppc64:"ANDCC\t[$]63",-"SRAD"
// ppc64le:"ANDCC\t[$]63",-"SRAD"
b := n2%64 != 0 // signed indivisible
return a, b
}
// Check that constant modulo divs get turned into MULs
func ConstMods(n1 uint, n2 int) (uint, int) {
// amd64:"MOVQ\t[$]-1085102592571150095","MULQ",-"DIVQ"
// 386:"MOVL\t[$]-252645135","MULL",-"DIVL"
// arm64:`MOVD`,`UMULH`,-`DIV`
// arm:`MOVW`,`MUL`,-`.*udiv`
a := n1 % 17 // unsigned
// amd64:"MOVQ\t[$]-1085102592571150095","IMULQ",-"IDIVQ"
// 386:"MOVL\t[$]-252645135","IMULL",-"IDIVL"
// arm64:`MOVD`,`SMULH`,-`DIV`
// arm:`MOVW`,`MUL`,-`.*udiv`
b := n2 % 17 // signed
return a, b
}
// Check that divisibility checks x%c==0 are converted to MULs and rotates
func Divisible(n1 uint, n2 int) (bool, bool, bool, bool) {
// amd64:"MOVQ\t[$]-6148914691236517205","IMULQ","ROLQ\t[$]63",-"DIVQ"
// 386:"IMUL3L\t[$]-1431655765","ROLL\t[$]31",-"DIVQ"
// arm64:"MOVD\t[$]-6148914691236517205","MUL","ROR",-"DIV"
// arm:"MUL","CMP\t[$]715827882",-".*udiv"
// ppc64:"MULLD","ROTL\t[$]63"
// ppc64le:"MULLD","ROTL\t[$]63"
evenU := n1%6 == 0
// amd64:"MOVQ\t[$]-8737931403336103397","IMULQ",-"ROLQ",-"DIVQ"
// 386:"IMUL3L\t[$]678152731",-"ROLL",-"DIVQ"
// arm64:"MOVD\t[$]-8737931403336103397","MUL",-"ROR",-"DIV"
// arm:"MUL","CMP\t[$]226050910",-".*udiv"
// ppc64:"MULLD",-"ROTL"
// ppc64le:"MULLD",-"ROTL"
oddU := n1%19 == 0
// amd64:"IMULQ","ADD","ROLQ\t[$]63",-"DIVQ"
// 386:"IMUL3L\t[$]-1431655765","ADDL\t[$]715827882","ROLL\t[$]31",-"DIVQ"
// arm64:"MUL","ADD\t[$]3074457345618258602","ROR",-"DIV"
// arm:"MUL","ADD\t[$]715827882",-".*udiv"
// ppc64/power8:"MULLD","ADD","ROTL\t[$]63"
// ppc64le/power8:"MULLD","ADD","ROTL\t[$]63"
// ppc64/power9:"MADDLD","ROTL\t[$]63"
// ppc64le/power9:"MADDLD","ROTL\t[$]63"
evenS := n2%6 == 0
// amd64:"IMULQ","ADD",-"ROLQ",-"DIVQ"
// 386:"IMUL3L\t[$]678152731","ADDL\t[$]113025455",-"ROLL",-"DIVQ"
// arm64:"MUL","ADD\t[$]485440633518672410",-"ROR",-"DIV"
// arm:"MUL","ADD\t[$]113025455",-".*udiv"
// ppc64/power8:"MULLD","ADD",-"ROTL"
// ppc64/power9:"MADDLD",-"ROTL"
// ppc64le/power8:"MULLD","ADD",-"ROTL"
// ppc64le/power9:"MADDLD",-"ROTL"
oddS := n2%19 == 0
return evenU, oddU, evenS, oddS
}
// Check that fix-up code is not generated for divisions where it has been proven that
// that the divisor is not -1 or that the dividend is > MinIntNN.
func NoFix64A(divr int64) (int64, int64) {
var d int64 = 42
var e int64 = 84
if divr > 5 {
d /= divr // amd64:-"JMP"
e %= divr // amd64:-"JMP"
// The following statement is to avoid conflict between the above check
// and the normal JMP generated at the end of the block.
d += e
}
return d, e
}
func NoFix64B(divd int64) (int64, int64) {
var d int64
var e int64
var divr int64 = -1
if divd > -9223372036854775808 {
d = divd / divr // amd64:-"JMP"
e = divd % divr // amd64:-"JMP"
d += e
}
return d, e
}
func NoFix32A(divr int32) (int32, int32) {
var d int32 = 42
var e int32 = 84
if divr > 5 {
// amd64:-"JMP"
// 386:-"JMP"
d /= divr
// amd64:-"JMP"
// 386:-"JMP"
e %= divr
d += e
}
return d, e
}
func NoFix32B(divd int32) (int32, int32) {
var d int32
var e int32
var divr int32 = -1
if divd > -2147483648 {
// amd64:-"JMP"
// 386:-"JMP"
d = divd / divr
// amd64:-"JMP"
// 386:-"JMP"
e = divd % divr
d += e
}
return d, e
}
func NoFix16A(divr int16) (int16, int16) {
var d int16 = 42
var e int16 = 84
if divr > 5 {
// amd64:-"JMP"
// 386:-"JMP"
d /= divr
// amd64:-"JMP"
// 386:-"JMP"
e %= divr
d += e
}
return d, e
}
func NoFix16B(divd int16) (int16, int16) {
var d int16
var e int16
var divr int16 = -1
if divd > -32768 {
// amd64:-"JMP"
// 386:-"JMP"
d = divd / divr
// amd64:-"JMP"
// 386:-"JMP"
e = divd % divr
d += e
}
return d, e
}
// Check that len() and cap() calls divided by powers of two are
// optimized into shifts and ands
func LenDiv1(a []int) int {
// 386:"SHRL\t[$]10"
// amd64:"SHRQ\t[$]10"
// arm64:"LSR\t[$]10",-"SDIV"
// arm:"SRL\t[$]10",-".*udiv"
// ppc64:"SRD"\t[$]10"
// ppc64le:"SRD"\t[$]10"
return len(a) / 1024
}
func LenDiv2(s string) int {
// 386:"SHRL\t[$]11"
// amd64:"SHRQ\t[$]11"
// arm64:"LSR\t[$]11",-"SDIV"
// arm:"SRL\t[$]11",-".*udiv"
// ppc64:"SRD\t[$]11"
// ppc64le:"SRD\t[$]11"
return len(s) / (4097 >> 1)
}
func LenMod1(a []int) int {
// 386:"ANDL\t[$]1023"
// amd64:"ANDQ\t[$]1023"
// arm64:"AND\t[$]1023",-"SDIV"
// arm/6:"AND",-".*udiv"
// arm/7:"BFC",-".*udiv",-"AND"
// ppc64:"ANDCC\t[$]1023"
// ppc64le:"ANDCC\t[$]1023"
return len(a) % 1024
}
func LenMod2(s string) int {
// 386:"ANDL\t[$]2047"
// amd64:"ANDQ\t[$]2047"
// arm64:"AND\t[$]2047",-"SDIV"
// arm/6:"AND",-".*udiv"
// arm/7:"BFC",-".*udiv",-"AND"
// ppc64:"ANDCC\t[$]2047"
// ppc64le:"ANDCC\t[$]2047"
return len(s) % (4097 >> 1)
}
func CapDiv(a []int) int {
// 386:"SHRL\t[$]12"
// amd64:"SHRQ\t[$]12"
// arm64:"LSR\t[$]12",-"SDIV"
// arm:"SRL\t[$]12",-".*udiv"
// ppc64:"SRD\t[$]12"
// ppc64le:"SRD\t[$]12"
return cap(a) / ((1 << 11) + 2048)
}
func CapMod(a []int) int {
// 386:"ANDL\t[$]4095"
// amd64:"ANDQ\t[$]4095"
// arm64:"AND\t[$]4095",-"SDIV"
// arm/6:"AND",-".*udiv"
// arm/7:"BFC",-".*udiv",-"AND"
// ppc64:"ANDCC\t[$]4095"
// ppc64le:"ANDCC\t[$]4095"
return cap(a) % ((1 << 11) + 2048)
}
func AddMul(x int) int {
// amd64:"LEAQ\t1"
return 2*x + 1
}
func MULA(a, b, c uint32) (uint32, uint32, uint32) {
// arm:`MULA`,-`MUL\s`
// arm64:`MADDW`,-`MULW`
r0 := a*b + c
// arm:`MULA`,-`MUL\s`
// arm64:`MADDW`,-`MULW`
r1 := c*79 + a
// arm:`ADD`,-`MULA`,-`MUL\s`
// arm64:`ADD`,-`MADD`,-`MULW`
r2 := b*64 + c
return r0, r1, r2
}
func MULS(a, b, c uint32) (uint32, uint32, uint32) {
// arm/7:`MULS`,-`MUL\s`
// arm/6:`SUB`,`MUL\s`,-`MULS`
// arm64:`MSUBW`,-`MULW`
r0 := c - a*b
// arm/7:`MULS`,-`MUL\s`
// arm/6:`SUB`,`MUL\s`,-`MULS`
// arm64:`MSUBW`,-`MULW`
r1 := a - c*79
// arm/7:`SUB`,-`MULS`,-`MUL\s`
// arm64:`SUB`,-`MSUBW`,-`MULW`
r2 := c - b*64
return r0, r1, r2
}
func addSpecial(a, b, c uint32) (uint32, uint32, uint32) {
// amd64:`INCL`
a++
// amd64:`DECL`
b--
// amd64:`SUBL.*-128`
c += 128
return a, b, c
}
// Divide -> shift rules usually require fixup for negative inputs.
// If the input is non-negative, make sure the fixup is eliminated.
func divInt(v int64) int64 {
if v < 0 {
return 0
}
// amd64:-`.*SARQ.*63,`, -".*SHRQ", ".*SARQ.*[$]9,"
return v / 512
}
// The reassociate rules "x - (z + C) -> (x - z) - C" and
// "(z + C) -x -> C + (z - x)" can optimize the following cases.
func constantFold1(i0, j0, i1, j1, i2, j2, i3, j3 int) (int, int, int, int) {
// arm64:"SUB","ADD\t[$]2"
r0 := (i0 + 3) - (j0 + 1)
// arm64:"SUB","SUB\t[$]4"
r1 := (i1 - 3) - (j1 + 1)
// arm64:"SUB","ADD\t[$]4"
r2 := (i2 + 3) - (j2 - 1)
// arm64:"SUB","SUB\t[$]2"
r3 := (i3 - 3) - (j3 - 1)
return r0, r1, r2, r3
}
// The reassociate rules "x - (z + C) -> (x - z) - C" and
// "(C - z) - x -> C - (z + x)" can optimize the following cases.
func constantFold2(i0, j0, i1, j1 int) (int, int) {
// arm64:"ADD","MOVD\t[$]2","SUB"
r0 := (3 - i0) - (j0 + 1)
// arm64:"ADD","MOVD\t[$]4","SUB"
r1 := (3 - i1) - (j1 - 1)
return r0, r1
}
func constantFold3(i, j int) int {
// arm64: "MOVD\t[$]30","MUL",-"ADD",-"LSL"
r := (5 * i) * (6 * j)
return r
}
View Git Blame Copy Permalink