qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-10-04 22:31:22 -06:00
Code Releases Activity
d91219e458
go/src/pkg/runtime/sys_linux_arm.s

443 lines
8.3 KiB
ArmAsm
Raw Normal View History

Added enough arm related scaffolding to create a simple hello.c program linking against the runtime. R=rsc APPROVED=rsc DELTA=178 (175 added, 0 deleted, 3 changed) OCL=29283 CL=29380
2009-05-26 12:18:42 -06:00
// Copyright 2009 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.
//
// System calls and other sys.stuff for arm, Linux
//
runtime: separate out auto-generated files, take 2 This is like the ill-fated CL 5493063 except that I have written a shell script (autogen.sh) instead of thinking I could possibly write a correct Makefile. R=golang-dev, r CC=golang-dev https://golang.org/cl/5496075
2011-12-19 13:51:13 -07:00
#include "zasm_GOOS_GOARCH.h"
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
runtime: check for ARM syscall failures While we are at it, also update some comments. Tested on Linux/ARM builder. R=rsc, golang-dev CC=golang-dev https://golang.org/cl/5696047
2012-02-23 13:43:14 -07:00
// for EABI, as we don't support OABI
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
#define SYS_BASE 0x0
Runtime is now starting up with a dummy c program as target: - morestack and gosave/gogo/gocall support - memclr and memset from inferno - bugfixes in _rt0_arm R=rsc APPROVED=rsc DELTA=304 (174 added, 36 deleted, 94 changed) OCL=30636 CL=30642
2009-06-23 12:54:23 -06:00
#define SYS_exit (SYS_BASE + 1)
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
#define SYS_read (SYS_BASE + 3)
Runtime is now starting up with a dummy c program as target: - morestack and gosave/gogo/gocall support - memclr and memset from inferno - bugfixes in _rt0_arm R=rsc APPROVED=rsc DELTA=304 (174 added, 36 deleted, 94 changed) OCL=30636 CL=30642
2009-06-23 12:54:23 -06:00
#define SYS_write (SYS_BASE + 4)
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
#define SYS_open (SYS_BASE + 5)
#define SYS_close (SYS_BASE + 6)
arm: fix build on android R=kaib CC=golang-dev https://golang.org/cl/206059
2010-02-10 16:01:02 -07:00
#define SYS_gettimeofday (SYS_BASE + 78)
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
#define SYS_clone (SYS_BASE + 120)
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
#define SYS_rt_sigreturn (SYS_BASE + 173)
#define SYS_rt_sigaction (SYS_BASE + 174)
runtime: linux signal masking Fixes #3101 (Linux). R=golang-dev, bradfitz, minux.ma CC=golang-dev https://golang.org/cl/5696043
2012-02-23 12:43:58 -07:00
#define SYS_rt_sigprocmask (SYS_BASE + 175)
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
#define SYS_sigaltstack (SYS_BASE + 186)
Runtime is now starting up with a dummy c program as target: - morestack and gosave/gogo/gocall support - memclr and memset from inferno - bugfixes in _rt0_arm R=rsc APPROVED=rsc DELTA=304 (174 added, 36 deleted, 94 changed) OCL=30636 CL=30642
2009-06-23 12:54:23 -06:00
#define SYS_mmap2 (SYS_BASE + 192)
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
#define SYS_futex (SYS_BASE + 240)
#define SYS_exit_group (SYS_BASE + 248)
runtime: use manual stack for garbage collection Old code was using recursion to traverse object graph. New code uses an explicit stack, cutting the per-pointer footprint to two words during the recursion and avoiding the standard allocator and stack splitting code. in test/garbage: Reduces parser runtime by 2-3% Reduces Peano runtime by 40% Increases tree runtime by 4-5% R=r CC=golang-dev https://golang.org/cl/2150042
2010-09-07 07:57:22 -06:00
#define SYS_munmap (SYS_BASE + 91)
runtime: madvise and SysUnused for Linux SysUnused being a direct call to madvise MADV_DONTNEED. R=golang-dev, dave CC=golang-dev https://golang.org/cl/5477057
2011-12-12 14:33:13 -07:00
#define SYS_madvise (SYS_BASE + 220)
runtime: os-specific types and code for setitimer R=r CC=golang-dev https://golang.org/cl/4273097
2011-03-23 09:31:42 -06:00
#define SYS_setitimer (SYS_BASE + 104)
runtime: SysMap uses MAP_FIXED if needed on 64-bit Linux This change was adapted from gccgo's libgo/runtime/mem.c at Ian Taylor's suggestion. It fixes all.bash failing with "address space conflict: map() =" on amd64 Linux with kernel version 2.6.32.8-grsec-2.1.14-modsign-xeon-64. With this change, SysMap will use MAP_FIXED to allocate its desired address space, after first calling mincore to check that there is nothing else mapped there. R=iant, dave, n13m3y3r, rsc CC=golang-dev https://golang.org/cl/4438091
2011-06-07 22:50:10 -06:00
#define SYS_mincore (SYS_BASE + 219)
runtime: turn "too many EPIPE" into real SIGPIPE Tested on Linux and OS X, amd64 and 386. R=r, iant CC=golang-dev https://golang.org/cl/4452046
2011-04-25 14:58:00 -06:00
#define SYS_gettid (SYS_BASE + 224)
#define SYS_tkill (SYS_BASE + 238)
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
#define SYS_sched_yield (SYS_BASE + 158)
runtime: runtime.usleep() bugfix on darwin/amd64 and linux/arm pkg/runtime/sys_darwin_amd64.s: fixes syscall select nr pkg/runtime/sys_linux_arm.s: uses newselect instead of the now unimplemented (old) select, also fixes the wrong div/mod statements in runtime.usleep. Fixes #2633 R=golang-dev, dave, rsc CC=golang-dev https://golang.org/cl/5504096
2012-01-10 21:48:02 -07:00
#define SYS_select (SYS_BASE + 142) // newselect
runtime: size arena to fit in virtual address space limit For Brad. Now FreeBSD/386 binaries run on nearlyfreespeech.net. Fixes #2302. R=golang-dev, r CC=golang-dev https://golang.org/cl/5700060
2012-02-24 13:28:51 -07:00
#define SYS_ugetrlimit (SYS_BASE + 191)
runtime: use sched_getaffinity for runtime.NumCPU() on Linux Fixes #3921. R=iant CC=golang-dev https://golang.org/cl/6448132
2012-08-09 20:05:26 -06:00
#define SYS_sched_getaffinity (SYS_BASE + 242)
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
#define SYS_clock_gettime (SYS_BASE + 263)
net, runtime: enable runtime-integrated pollster on Linux/ARM. Results from linux/arm on a Samsung Chromebook (from dfc): localhost(~/go/src/pkg/net) % ~/go/misc/benchcmp {old,new}.txt benchmark old ns/op new ns/op delta BenchmarkTCP4OneShot 568840 350526 -38.38% BenchmarkTCP4OneShot-2 359054 206708 -42.43% BenchmarkTCP4OneShotTimeout 637464 363550 -42.97% BenchmarkTCP4OneShotTimeout-2 374255 216695 -42.10% BenchmarkTCP4Persistent 184974 64984 -64.87% BenchmarkTCP4Persistent-2 109902 47195 -57.06% BenchmarkTCP4PersistentTimeout 210039 64789 -69.15% BenchmarkTCP4PersistentTimeout-2 124284 43374 -65.10% BenchmarkTCP6OneShot 672278 362116 -46.14% BenchmarkTCP6OneShot-2 383631 216400 -43.59% BenchmarkTCP6OneShotTimeout 680740 378306 -44.43% BenchmarkTCP6OneShotTimeout-2 397524 230152 -42.10% BenchmarkTCP6Persistent 172346 65292 -62.12% BenchmarkTCP6Persistent-2 106229 42096 -60.37% BenchmarkTCP6PersistentTimeout 161149 65138 -59.58% BenchmarkTCP6PersistentTimeout-2 152276 44548 -70.75% R=golang-dev, dave, bradfitz, dvyukov, rsc CC=golang-dev https://golang.org/cl/7820045
2013-03-15 14:01:56 -06:00
#define SYS_epoll_create (SYS_BASE + 250)
#define SYS_epoll_ctl (SYS_BASE + 251)
#define SYS_epoll_wait (SYS_BASE + 252)
#define SYS_epoll_create1 (SYS_BASE + 357)
#define SYS_fcntl (SYS_BASE + 55)
Runtime is now starting up with a dummy c program as target: - morestack and gosave/gogo/gocall support - memclr and memset from inferno - bugfixes in _rt0_arm R=rsc APPROVED=rsc DELTA=304 (174 added, 36 deleted, 94 changed) OCL=30636 CL=30642
2009-06-23 12:54:23 -06:00
add cache flushing call after generating closure. go/test: passes 88% (306/347) R=rsc https://golang.org/cl/152089
2009-11-12 00:23:11 -07:00
#define ARM_BASE (SYS_BASE + 0x0f0000)
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
TEXT runtime·open(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW $SYS_open, R7
SWI $0
RET
TEXT runtime·close(SB),7,$0
MOVW 0(FP), R0
MOVW $SYS_close, R7
SWI $0
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·write(SB),7,$0
changed 5c calling convention to use stack exclusively for in params. a number of fixes to assembly routines that assumed R0 had the first arg. one stack offset fix, arm pushes the link register on stack top. go/test: passes 65% (235/364) tests R=rsc APPROVED=rsc DELTA=20 (11 added, 0 deleted, 9 changed) OCL=34809 CL=34812
2009-09-18 17:45:41 -06:00
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_write, R7
SWI $0
Added enough arm related scaffolding to create a simple hello.c program linking against the runtime. R=rsc APPROVED=rsc DELTA=178 (175 added, 0 deleted, 3 changed) OCL=29283 CL=29380
2009-05-26 12:18:42 -06:00
RET
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
TEXT runtime·read(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW $SYS_read, R7
SWI $0
RET
runtime: size arena to fit in virtual address space limit For Brad. Now FreeBSD/386 binaries run on nearlyfreespeech.net. Fixes #2302. R=golang-dev, r CC=golang-dev https://golang.org/cl/5700060
2012-02-24 13:28:51 -07:00
TEXT runtime·getrlimit(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW $SYS_ugetrlimit, R7
SWI $0
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·exit(SB),7,$-4
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW 0(FP), R0
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_exit_group, R7
SWI $0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW $1234, R0
MOVW $1002, R1
MOVW R0, (R1) // fail hard
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·exit1(SB),7,$-4
proper syscall convention for arm R=rsc http://go/go-review/1015009
2009-10-25 12:51:16 -06:00
MOVW 0(FP), R0
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_exit, R7
SWI $0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW $1234, R0
MOVW $1003, R1
MOVW R0, (R1) // fail hard
Adding a batch of missing system calls. R=rsc APPROVED=rsc DELTA=1329 (1264 added, 1 deleted, 64 changed) OCL=30040 CL=30158
2009-06-10 12:53:07 -06:00
runtime: accept GOTRACEBACK=crash to mean 'crash after panic' This provides a way to generate core dumps when people need them. The settings are: GOTRACEBACK=0 no traceback on panic, just exit GOTRACEBACK=1 default - traceback on panic, then exit GOTRACEBACK=2 traceback including runtime frames on panic, then exit GOTRACEBACK=crash traceback including runtime frames on panic, then crash Fixes #3257. R=golang-dev, devon.odell, r, daniel.morsing, ality CC=golang-dev https://golang.org/cl/7666044
2013-03-14 23:11:03 -06:00
TEXT runtime·raise(SB),7,$-4
runtime: turn "too many EPIPE" into real SIGPIPE Tested on Linux and OS X, amd64 and 386. R=r, iant CC=golang-dev https://golang.org/cl/4452046
2011-04-25 14:58:00 -06:00
MOVW $SYS_gettid, R7
SWI $0
// arg 1 tid already in R0 from gettid
runtime: accept GOTRACEBACK=crash to mean 'crash after panic' This provides a way to generate core dumps when people need them. The settings are: GOTRACEBACK=0 no traceback on panic, just exit GOTRACEBACK=1 default - traceback on panic, then exit GOTRACEBACK=2 traceback including runtime frames on panic, then exit GOTRACEBACK=crash traceback including runtime frames on panic, then crash Fixes #3257. R=golang-dev, devon.odell, r, daniel.morsing, ality CC=golang-dev https://golang.org/cl/7666044
2013-03-14 23:11:03 -06:00
MOVW sig+0(FP), R1 // arg 2 - signal
runtime: turn "too many EPIPE" into real SIGPIPE Tested on Linux and OS X, amd64 and 386. R=r, iant CC=golang-dev https://golang.org/cl/4452046
2011-04-25 14:58:00 -06:00
MOVW $SYS_tkill, R7
SWI $0
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·mmap(SB),7,$0
changed 5c calling convention to use stack exclusively for in params. a number of fixes to assembly routines that assumed R0 had the first arg. one stack offset fix, arm pushes the link register on stack top. go/test: passes 65% (235/364) tests R=rsc APPROVED=rsc DELTA=20 (11 added, 0 deleted, 9 changed) OCL=34809 CL=34812
2009-09-18 17:45:41 -06:00
MOVW 0(FP), R0
Runtime is now starting up with a dummy c program as target: - morestack and gosave/gogo/gocall support - memclr and memset from inferno - bugfixes in _rt0_arm R=rsc APPROVED=rsc DELTA=304 (174 added, 36 deleted, 94 changed) OCL=30636 CL=30642
2009-06-23 12:54:23 -06:00
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW 12(FP), R3
MOVW 16(FP), R4
MOVW 20(FP), R5
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_mmap2, R7
SWI $0
runtime: check for ARM syscall failures While we are at it, also update some comments. Tested on Linux/ARM builder. R=rsc, golang-dev CC=golang-dev https://golang.org/cl/5696047
2012-02-23 13:43:14 -07:00
MOVW $0xfffff001, R6
CMP R6, R0
RSB.HI $0, R0
Adding a batch of missing system calls. R=rsc APPROVED=rsc DELTA=1329 (1264 added, 1 deleted, 64 changed) OCL=30040 CL=30158
2009-06-10 12:53:07 -06:00
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·munmap(SB),7,$0
runtime: use manual stack for garbage collection Old code was using recursion to traverse object graph. New code uses an explicit stack, cutting the per-pointer footprint to two words during the recursion and avoiding the standard allocator and stack splitting code. in test/garbage: Reduces parser runtime by 2-3% Reduces Peano runtime by 40% Increases tree runtime by 4-5% R=r CC=golang-dev https://golang.org/cl/2150042
2010-09-07 07:57:22 -06:00
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW $SYS_munmap, R7
SWI $0
runtime: check for ARM syscall failures While we are at it, also update some comments. Tested on Linux/ARM builder. R=rsc, golang-dev CC=golang-dev https://golang.org/cl/5696047
2012-02-23 13:43:14 -07:00
MOVW $0xfffff001, R6
CMP R6, R0
runtime: use m and g, instead of R9 and R10, in ARM assembly files also don't clobber R9 if it is about to crash. In response to https://golang.org/cl/9251043/#msg2. R=golang-dev, khr, khr, dave CC=golang-dev https://golang.org/cl/9778046
2013-05-28 06:13:02 -06:00
MOVW.HI $0, R8 // crash on syscall failure
MOVW.HI R8, (R8)
runtime: use manual stack for garbage collection Old code was using recursion to traverse object graph. New code uses an explicit stack, cutting the per-pointer footprint to two words during the recursion and avoiding the standard allocator and stack splitting code. in test/garbage: Reduces parser runtime by 2-3% Reduces Peano runtime by 40% Increases tree runtime by 4-5% R=r CC=golang-dev https://golang.org/cl/2150042
2010-09-07 07:57:22 -06:00
RET
runtime: madvise and SysUnused for Linux SysUnused being a direct call to madvise MADV_DONTNEED. R=golang-dev, dave CC=golang-dev https://golang.org/cl/5477057
2011-12-12 14:33:13 -07:00
TEXT runtime·madvise(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW $SYS_madvise, R7
SWI $0
runtime: ignore failure from madvise When we release memory to the OS, if the OS doesn't want us to release it (for example, because the program executed mlockall(MCL_FUTURE)), madvise will fail. Ignore the failure instead of crashing. Fixes #3435. R=ken2 CC=golang-dev https://golang.org/cl/6998052
2012-12-22 13:06:28 -07:00
// ignore failure - maybe pages are locked
runtime: madvise and SysUnused for Linux SysUnused being a direct call to madvise MADV_DONTNEED. R=golang-dev, dave CC=golang-dev https://golang.org/cl/5477057
2011-12-12 14:33:13 -07:00
RET
runtime: fix arm build R=adg, dfc, r CC=golang-dev https://golang.org/cl/4296042
2011-03-25 10:30:49 -06:00
TEXT runtime·setitimer(SB),7,$0
runtime: os-specific types and code for setitimer R=r CC=golang-dev https://golang.org/cl/4273097
2011-03-23 09:31:42 -06:00
MOVW 0(FP), R0
MOVW 4(FP), R1
runtime: fix arm build R=adg, dfc, r CC=golang-dev https://golang.org/cl/4296042
2011-03-25 10:30:49 -06:00
MOVW 8(FP), R2
runtime: os-specific types and code for setitimer R=r CC=golang-dev https://golang.org/cl/4273097
2011-03-23 09:31:42 -06:00
MOVW $SYS_setitimer, R7
SWI $0
RET
runtime: SysMap uses MAP_FIXED if needed on 64-bit Linux This change was adapted from gccgo's libgo/runtime/mem.c at Ian Taylor's suggestion. It fixes all.bash failing with "address space conflict: map() =" on amd64 Linux with kernel version 2.6.32.8-grsec-2.1.14-modsign-xeon-64. With this change, SysMap will use MAP_FIXED to allocate its desired address space, after first calling mincore to check that there is nothing else mapped there. R=iant, dave, n13m3y3r, rsc CC=golang-dev https://golang.org/cl/4438091
2011-06-07 22:50:10 -06:00
TEXT runtime·mincore(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW $SYS_mincore, R7
SWI $0
RET
time: new Time, Duration, ZoneInfo types R=r, bradfitz, gri, dsymonds, iant CC=golang-dev https://golang.org/cl/5392041
2011-11-30 09:59:44 -07:00
TEXT time·now(SB), 7, $32
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
MOVW $0, R0 // CLOCK_REALTIME
MOVW $8(R13), R1 // timespec
MOVW $SYS_clock_gettime, R7
time: new Time, Duration, ZoneInfo types R=r, bradfitz, gri, dsymonds, iant CC=golang-dev https://golang.org/cl/5392041
2011-11-30 09:59:44 -07:00
SWI $0
MOVW 8(R13), R0 // sec
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
MOVW 12(R13), R2 // nsec
time: new Time, Duration, ZoneInfo types R=r, bradfitz, gri, dsymonds, iant CC=golang-dev https://golang.org/cl/5392041
2011-11-30 09:59:44 -07:00
MOVW R0, 0(FP)
MOVW $0, R1
MOVW R1, 4(FP)
MOVW R2, 8(FP)
RET
time: faster Nanoseconds call runtime knows how to get the time of day without allocating memory. R=golang-dev, dsymonds, dave, hectorchu, r, cw CC=golang-dev https://golang.org/cl/5297078
2011-11-03 15:35:28 -06:00
// int64 nanotime(void) so really
// void nanotime(int64 *nsec)
TEXT runtime·nanotime(SB),7,$32
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
MOVW $0, R0 // CLOCK_REALTIME
MOVW $8(R13), R1 // timespec
MOVW $SYS_clock_gettime, R7
arm: fix build on android R=kaib CC=golang-dev https://golang.org/cl/206059
2010-02-10 16:01:02 -07:00
SWI $0
runtime: fix linux/arm nanotime? TBR=r CC=golang-dev https://golang.org/cl/5354041
2011-11-04 06:38:10 -06:00
MOVW 8(R13), R0 // sec
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
MOVW 12(R13), R2 // nsec
runtime: fix linux/arm nanotime? TBR=r CC=golang-dev https://golang.org/cl/5354041
2011-11-04 06:38:10 -06:00
MOVW $1000000000, R3
MULLU R0, R3, (R1, R0)
MOVW $0, R4
ADD.S R2, R0
ADC R4, R1
runtime: use clock_gettime to get ns resolution for time.now & runtime.nanotime For Linux/{386,arm}, FreeBSD/{386,amd64,arm}, NetBSD/{386,amd64}, OpenBSD/{386,amd64}. Note: our Darwin implementation already has ns resolution. Linux/386 (Core i7-2600 @ 3.40GHz, kernel 3.5.2-gentoo) benchmark old ns/op new ns/op delta BenchmarkNow 110 118 +7.27% Linux/ARM (ARM Cortex-A8 @ 800MHz, kernel 2.6.32.28 android) benchmark old ns/op new ns/op delta BenchmarkNow 625 542 -13.28% Linux/ARM (ARM Cortex-A9 @ 1GHz, Pandaboard) benchmark old ns/op new ns/op delta BenchmarkNow 992 909 -8.37% FreeBSD 9-REL-p1/amd64 (Dell R610 Server with Xeon X5650 @ 2.67GHz) benchmark old ns/op new ns/op delta BenchmarkNow 699 695 -0.57% FreeBSD 9-REL-p1/amd64 (Atom D525 @ 1.80GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1553 1658 +6.76% OpenBSD/amd64 (Dell E6410 with i5 CPU M 540 @ 2.53GHz) benchmark old ns/op new ns/op delta BenchmarkNow 1262 1236 -2.06% OpenBSD/i386 (Asus eeePC 701 with Intel Celeron M 900MHz - locked to 631MHz) benchmark old ns/op new ns/op delta BenchmarkNow 5089 5043 -0.90% NetBSD/i386 (VMware VM with Core i5 CPU @ 2.7GHz) benchmark old ns/op new ns/op delta BenchmarkNow 277 278 +0.36% NetBSD/amd64 (VMware VM with Core i5 CPU @ 2.7Ghz) benchmark old ns/op new ns/op delta BenchmarkNow 103 105 +1.94% Thanks Maxim Khitrov, Joel Sing, and Dave Cheney for providing benchmark data. R=jsing, dave, rsc CC=golang-dev https://golang.org/cl/6820120
2012-12-18 07:57:25 -07:00
runtime: fix linux/arm nanotime? TBR=r CC=golang-dev https://golang.org/cl/5354041
2011-11-04 06:38:10 -06:00
MOVW 0(FP), R3
MOVW R0, 0(R3)
MOVW R1, 4(R3)
arm: fix build on android R=kaib CC=golang-dev https://golang.org/cl/206059
2010-02-10 16:01:02 -07:00
RET
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
// int32 futex(int32 *uaddr, int32 op, int32 val,
Adding a batch of missing system calls. R=rsc APPROVED=rsc DELTA=1329 (1264 added, 1 deleted, 64 changed) OCL=30040 CL=30158
2009-06-10 12:53:07 -06:00
// struct timespec *timeout, int32 *uaddr2, int32 val2);
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·futex(SB),7,$0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW 4(SP), R0
MOVW 8(SP), R1
MOVW 12(SP), R2
MOVW 16(SP), R3
MOVW 20(SP), R4
MOVW 24(SP), R5
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_futex, R7
SWI $0
Adding a batch of missing system calls. R=rsc APPROVED=rsc DELTA=1329 (1264 added, 1 deleted, 64 changed) OCL=30040 CL=30158
2009-06-10 12:53:07 -06:00
RET
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
runtime: use "mp" and "gp" instead of "m" and "g" for local variable name to avoid confusion with the global "m" and "g". R=golang-dev, minux.ma, rsc CC=bradfitz, golang-dev https://golang.org/cl/6939064
2012-12-18 09:30:29 -07:00
// int32 clone(int32 flags, void *stack, M *mp, G *gp, void (*fn)(void));
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·clone(SB),7,$0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW flags+0(FP), R0
MOVW stack+4(FP), R1
MOVW $0, R2 // parent tid ptr
MOVW $0, R3 // tls_val
MOVW $0, R4 // child tid ptr
MOVW $0, R5
runtime: use "mp" and "gp" instead of "m" and "g" for local variable name to avoid confusion with the global "m" and "g". R=golang-dev, minux.ma, rsc CC=bradfitz, golang-dev https://golang.org/cl/6939064
2012-12-18 09:30:29 -07:00
// Copy mp, gp, fn off parent stack for use by child.
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
// TODO(kaib): figure out which registers are clobbered by clone and avoid stack copying
MOVW $-16(R1), R1
MOVW mm+8(FP), R6
MOVW R6, 0(R1)
MOVW gg+12(FP), R6
MOVW R6, 4(R1)
MOVW fn+16(FP), R6
MOVW R6, 8(R1)
MOVW $1234, R6
MOVW R6, 12(R1)
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_clone, R7
SWI $0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
// In parent, return.
CMP $0, R0
BEQ 2(PC)
RET
// Paranoia: check that SP is as we expect. Use R13 to avoid linker 'fixup'
MOVW 12(R13), R0
MOVW $1234, R1
CMP R0, R1
BEQ 2(PC)
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
BL runtime·abort(SB)
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW 0(R13), m
MOVW 4(R13), g
// paranoia; check they are not nil
MOVW 0(m), R0
MOVW 0(g), R0
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
BL runtime·emptyfunc(SB) // fault if stack check is wrong
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
// Initialize m->procid to Linux tid
make 5g executables run on android/arm hardware. change OABI usage to EABI. go/test: passes 85% (296/347) on random android phone. R=rsc http://go/go-review/1024003
2009-11-05 23:53:08 -07:00
MOVW $SYS_gettid, R7
SWI $0
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
MOVW R0, m_procid(m)
// Call fn
MOVW 8(R13), R0
MOVW $16(R13), R13
BL (R0)
MOVW $0, R0
MOVW R0, 4(R13)
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
BL runtime·exit1(SB)
clone and futex go/test: passes 99% (343/347) R=rsc http://go/go-review/1016004
2009-10-29 22:21:14 -06:00
// It shouldn't return
MOVW $1234, R0
MOVW $1005, R1
MOVW R0, (R1)
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·sigaltstack(SB),7,$0
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW $SYS_sigaltstack, R7
SWI $0
runtime: check for ARM syscall failures While we are at it, also update some comments. Tested on Linux/ARM builder. R=rsc, golang-dev CC=golang-dev https://golang.org/cl/5696047
2012-02-23 13:43:14 -07:00
MOVW $0xfffff001, R6
CMP R6, R0
runtime: use m and g, instead of R9 and R10, in ARM assembly files also don't clobber R9 if it is about to crash. In response to https://golang.org/cl/9251043/#msg2. R=golang-dev, khr, khr, dave CC=golang-dev https://golang.org/cl/9778046
2013-05-28 06:13:02 -06:00
MOVW.HI $0, R8 // crash on syscall failure
MOVW.HI R8, (R8)
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·sigtramp(SB),7,$24
runtime: cgo support for Linux/ARM Part 2 of CL 5601044 (cgo: Linux/ARM support) R=dave, rsc CC=golang-dev https://golang.org/cl/5989057
2012-05-04 04:20:09 -06:00
// this might be called in external code context,
// where g and m are not set.
runtime/cgo: make symbol naming consistent The naming in this package is a disaster. Make it all consistent. Remove some 'static' from functions that will be referred to from other files soon. This CL is purely renames using global search and replace. Submitting separately so that real changes will not be drowned out by these renames in future CLs. TBR=iant CC=golang-dev https://golang.org/cl/7416046
2013-02-28 14:24:38 -07:00
// first save R0, because _cgo_load_gm will clobber it
runtime: cgo support for Linux/ARM Part 2 of CL 5601044 (cgo: Linux/ARM support) R=dave, rsc CC=golang-dev https://golang.org/cl/5989057
2012-05-04 04:20:09 -06:00
MOVW R0, 4(R13)
runtime/cgo: make symbol naming consistent The naming in this package is a disaster. Make it all consistent. Remove some 'static' from functions that will be referred to from other files soon. This CL is purely renames using global search and replace. Submitting separately so that real changes will not be drowned out by these renames in future CLs. TBR=iant CC=golang-dev https://golang.org/cl/7416046
2013-02-28 14:24:38 -07:00
MOVW _cgo_load_gm(SB), R0
runtime: cgo support for Linux/ARM Part 2 of CL 5601044 (cgo: Linux/ARM support) R=dave, rsc CC=golang-dev https://golang.org/cl/5989057
2012-05-04 04:20:09 -06:00
CMP $0, R0
BL.NE (R0)
runtime: make arm signal handler call runtime.badsignal In preparation for CL 9249043 (already submitted). Fixes #5553. R=golang-dev, iant, capnm9, dave CC=golang-dev https://golang.org/cl/9251043
2013-05-27 06:46:53 -06:00
CMP $0, m
runtime: correctly handle signals received on foreign threads Fixes #3250. R=rsc CC=golang-dev https://golang.org/cl/10757044
2013-07-11 14:39:39 -06:00
BNE 4(PC)
runtime: make arm signal handler call runtime.badsignal In preparation for CL 9249043 (already submitted). Fixes #5553. R=golang-dev, iant, capnm9, dave CC=golang-dev https://golang.org/cl/9251043
2013-05-27 06:46:53 -06:00
// signal number is already prepared in 4(R13)
runtime: correctly handle signals received on foreign threads Fixes #3250. R=rsc CC=golang-dev https://golang.org/cl/10757044
2013-07-11 14:39:39 -06:00
MOVW $runtime·badsignal(SB), R11
BL (R11)
runtime: make arm signal handler call runtime.badsignal In preparation for CL 9249043 (already submitted). Fixes #5553. R=golang-dev, iant, capnm9, dave CC=golang-dev https://golang.org/cl/9251043
2013-05-27 06:46:53 -06:00
RET
runtime: pass to signal handler value of g at time of signal The existing code assumed that signals only arrived while executing on the goroutine stack (g == m->curg), not while executing on the scheduler stack (g == m->g0). Most of the signal handling trampolines correctly saved and restored g already, but the sighandler C code did not have access to it. Some rewriting of assembly to make the various implementations as similar as possible. Will need to change Windows too but I don't understand how sigtramp gets called there. R=r CC=golang-dev https://golang.org/cl/4203042
2011-02-23 12:47:42 -07:00
// save g
MOVW g, R3
MOVW g, 20(R13)
runtime: parallelize garbage collector mark + sweep Running test/garbage/parser.out. On a 4-core Lenovo X201s (Linux): 31.12u 0.60s 31.74r 1 cpu, no atomics 32.27u 0.58s 32.86r 1 cpu, atomic instructions 33.04u 0.83s 27.47r 2 cpu On a 16-core Xeon (Linux): 33.08u 0.65s 33.80r 1 cpu, no atomics 34.87u 1.12s 29.60r 2 cpu 36.00u 1.87s 28.43r 3 cpu 36.46u 2.34s 27.10r 4 cpu 38.28u 3.85s 26.92r 5 cpu 37.72u 5.25s 26.73r 6 cpu 39.63u 7.11s 26.95r 7 cpu 39.67u 8.10s 26.68r 8 cpu On a 2-core MacBook Pro Core 2 Duo 2.26 (circa 2009, MacBookPro5,5): 39.43u 1.45s 41.27r 1 cpu, no atomics 43.98u 2.95s 38.69r 2 cpu On a 2-core Mac Mini Core 2 Duo 1.83 (circa 2008; Macmini2,1): 48.81u 2.12s 51.76r 1 cpu, no atomics 57.15u 4.72s 51.54r 2 cpu The handoff algorithm is really only good for two cores. Beyond that we will need to so something more sophisticated, like have each core hand off to the next one, around a circle. Even so, the code is a good checkpoint; for now we'll limit the number of gc procs to at most 2. R=dvyukov CC=golang-dev https://golang.org/cl/4641082
2011-09-30 07:40:01 -06:00
runtime: pass to signal handler value of g at time of signal The existing code assumed that signals only arrived while executing on the goroutine stack (g == m->curg), not while executing on the scheduler stack (g == m->g0). Most of the signal handling trampolines correctly saved and restored g already, but the sighandler C code did not have access to it. Some rewriting of assembly to make the various implementations as similar as possible. Will need to change Windows too but I don't understand how sigtramp gets called there. R=r CC=golang-dev https://golang.org/cl/4203042
2011-02-23 12:47:42 -07:00
// g = m->gsignal
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
MOVW m_gsignal(m), g
runtime: pass to signal handler value of g at time of signal The existing code assumed that signals only arrived while executing on the goroutine stack (g == m->curg), not while executing on the scheduler stack (g == m->g0). Most of the signal handling trampolines correctly saved and restored g already, but the sighandler C code did not have access to it. Some rewriting of assembly to make the various implementations as similar as possible. Will need to change Windows too but I don't understand how sigtramp gets called there. R=r CC=golang-dev https://golang.org/cl/4203042
2011-02-23 12:47:42 -07:00
// copy arguments for call to sighandler
runtime: cgo support for Linux/ARM Part 2 of CL 5601044 (cgo: Linux/ARM support) R=dave, rsc CC=golang-dev https://golang.org/cl/5989057
2012-05-04 04:20:09 -06:00
// R0 is already saved above
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
MOVW R1, 8(R13)
MOVW R2, 12(R13)
runtime: pass to signal handler value of g at time of signal The existing code assumed that signals only arrived while executing on the goroutine stack (g == m->curg), not while executing on the scheduler stack (g == m->g0). Most of the signal handling trampolines correctly saved and restored g already, but the sighandler C code did not have access to it. Some rewriting of assembly to make the various implementations as similar as possible. Will need to change Windows too but I don't understand how sigtramp gets called there. R=r CC=golang-dev https://golang.org/cl/4203042
2011-02-23 12:47:42 -07:00
MOVW R3, 16(R13)
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
BL runtime·sighandler(SB)
runtime: parallelize garbage collector mark + sweep Running test/garbage/parser.out. On a 4-core Lenovo X201s (Linux): 31.12u 0.60s 31.74r 1 cpu, no atomics 32.27u 0.58s 32.86r 1 cpu, atomic instructions 33.04u 0.83s 27.47r 2 cpu On a 16-core Xeon (Linux): 33.08u 0.65s 33.80r 1 cpu, no atomics 34.87u 1.12s 29.60r 2 cpu 36.00u 1.87s 28.43r 3 cpu 36.46u 2.34s 27.10r 4 cpu 38.28u 3.85s 26.92r 5 cpu 37.72u 5.25s 26.73r 6 cpu 39.63u 7.11s 26.95r 7 cpu 39.67u 8.10s 26.68r 8 cpu On a 2-core MacBook Pro Core 2 Duo 2.26 (circa 2009, MacBookPro5,5): 39.43u 1.45s 41.27r 1 cpu, no atomics 43.98u 2.95s 38.69r 2 cpu On a 2-core Mac Mini Core 2 Duo 1.83 (circa 2008; Macmini2,1): 48.81u 2.12s 51.76r 1 cpu, no atomics 57.15u 4.72s 51.54r 2 cpu The handoff algorithm is really only good for two cores. Beyond that we will need to so something more sophisticated, like have each core hand off to the next one, around a circle. Even so, the code is a good checkpoint; for now we'll limit the number of gc procs to at most 2. R=dvyukov CC=golang-dev https://golang.org/cl/4641082
2011-09-30 07:40:01 -06:00
runtime: pass to signal handler value of g at time of signal The existing code assumed that signals only arrived while executing on the goroutine stack (g == m->curg), not while executing on the scheduler stack (g == m->g0). Most of the signal handling trampolines correctly saved and restored g already, but the sighandler C code did not have access to it. Some rewriting of assembly to make the various implementations as similar as possible. Will need to change Windows too but I don't understand how sigtramp gets called there. R=r CC=golang-dev https://golang.org/cl/4203042
2011-02-23 12:47:42 -07:00
// restore g
MOVW 20(R13), g
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
RET
runtime: linux signal masking Fixes #3101 (Linux). R=golang-dev, bradfitz, minux.ma CC=golang-dev https://golang.org/cl/5696043
2012-02-23 12:43:58 -07:00
TEXT runtime·rtsigprocmask(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW 12(FP), R3
MOVW $SYS_rt_sigprocmask, R7
SWI $0
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·rt_sigaction(SB),7,$0
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW 12(FP), R3
MOVW $SYS_rt_sigaction, R7
SWI $0
RET
runtime: ,s/[a-zA-Z0-9_]+/runtime·&/g, almost Prefix all external symbols in runtime by runtime·, to avoid conflicts with possible symbols of the same name in linked-in C libraries. The obvious conflicts are printf, malloc, and free, but hide everything to avoid future pain. The symbols left alone are: ** known to cgo ** _cgo_free _cgo_malloc libcgo_thread_start initcgo ncgocall ** known to linker ** _rt0_$GOARCH _rt0_$GOARCH_$GOOS text etext data end pclntab epclntab symtab esymtab ** known to C compiler ** _divv _modv _div64by32 etc (arch specific) Tested on darwin/386, darwin/amd64, linux/386, linux/amd64. Built (but not tested) for freebsd/386, freebsd/amd64, linux/arm, windows/386. R=r, PeterGo CC=golang-dev https://golang.org/cl/2899041
2010-11-04 12:00:19 -06:00
TEXT runtime·sigreturn(SB),7,$0
runtime: various arm fixes * correct symbol table size * do not reorder functions in output * traceback * signal handling * use same code for go + defer * handle leaf functions in symbol table R=kaib, dpx CC=golang-dev https://golang.org/cl/884041
2010-04-05 13:51:09 -06:00
MOVW $SYS_rt_sigreturn, R7
SWI $0
RET
runtime: use kernel-supplied cas on linux/arm Using the kernel-supplied compare-and-swap code on linux/arm means that runtime doesn't have to care whether this is GOARM=5 or GOARM=6 anymore. Fixes #1494. R=r, r2 CC=golang-dev https://golang.org/cl/4245043
2011-02-25 12:29:55 -07:00
runtime: parallelize garbage collector mark + sweep Running test/garbage/parser.out. On a 4-core Lenovo X201s (Linux): 31.12u 0.60s 31.74r 1 cpu, no atomics 32.27u 0.58s 32.86r 1 cpu, atomic instructions 33.04u 0.83s 27.47r 2 cpu On a 16-core Xeon (Linux): 33.08u 0.65s 33.80r 1 cpu, no atomics 34.87u 1.12s 29.60r 2 cpu 36.00u 1.87s 28.43r 3 cpu 36.46u 2.34s 27.10r 4 cpu 38.28u 3.85s 26.92r 5 cpu 37.72u 5.25s 26.73r 6 cpu 39.63u 7.11s 26.95r 7 cpu 39.67u 8.10s 26.68r 8 cpu On a 2-core MacBook Pro Core 2 Duo 2.26 (circa 2009, MacBookPro5,5): 39.43u 1.45s 41.27r 1 cpu, no atomics 43.98u 2.95s 38.69r 2 cpu On a 2-core Mac Mini Core 2 Duo 1.83 (circa 2008; Macmini2,1): 48.81u 2.12s 51.76r 1 cpu, no atomics 57.15u 4.72s 51.54r 2 cpu The handoff algorithm is really only good for two cores. Beyond that we will need to so something more sophisticated, like have each core hand off to the next one, around a circle. Even so, the code is a good checkpoint; for now we'll limit the number of gc procs to at most 2. R=dvyukov CC=golang-dev https://golang.org/cl/4641082
2011-09-30 07:40:01 -06:00
TEXT runtime·usleep(SB),7,$12
MOVW usec+0(FP), R0
MOVW R0, R1
MOVW $1000000, R2
runtime: runtime.usleep() bugfix on darwin/amd64 and linux/arm pkg/runtime/sys_darwin_amd64.s: fixes syscall select nr pkg/runtime/sys_linux_arm.s: uses newselect instead of the now unimplemented (old) select, also fixes the wrong div/mod statements in runtime.usleep. Fixes #2633 R=golang-dev, dave, rsc CC=golang-dev https://golang.org/cl/5504096
2012-01-10 21:48:02 -07:00
DIV R2, R0
MOD R2, R1
MOVW R0, 4(SP)
MOVW R1, 8(SP)
runtime: parallelize garbage collector mark + sweep Running test/garbage/parser.out. On a 4-core Lenovo X201s (Linux): 31.12u 0.60s 31.74r 1 cpu, no atomics 32.27u 0.58s 32.86r 1 cpu, atomic instructions 33.04u 0.83s 27.47r 2 cpu On a 16-core Xeon (Linux): 33.08u 0.65s 33.80r 1 cpu, no atomics 34.87u 1.12s 29.60r 2 cpu 36.00u 1.87s 28.43r 3 cpu 36.46u 2.34s 27.10r 4 cpu 38.28u 3.85s 26.92r 5 cpu 37.72u 5.25s 26.73r 6 cpu 39.63u 7.11s 26.95r 7 cpu 39.67u 8.10s 26.68r 8 cpu On a 2-core MacBook Pro Core 2 Duo 2.26 (circa 2009, MacBookPro5,5): 39.43u 1.45s 41.27r 1 cpu, no atomics 43.98u 2.95s 38.69r 2 cpu On a 2-core Mac Mini Core 2 Duo 1.83 (circa 2008; Macmini2,1): 48.81u 2.12s 51.76r 1 cpu, no atomics 57.15u 4.72s 51.54r 2 cpu The handoff algorithm is really only good for two cores. Beyond that we will need to so something more sophisticated, like have each core hand off to the next one, around a circle. Even so, the code is a good checkpoint; for now we'll limit the number of gc procs to at most 2. R=dvyukov CC=golang-dev https://golang.org/cl/4641082
2011-09-30 07:40:01 -06:00
MOVW $0, R0
MOVW $0, R1
MOVW $0, R2
MOVW $0, R3
MOVW $4(SP), R4
MOVW $SYS_select, R7
SWI $0
RET
runtime: check for ARM syscall failures While we are at it, also update some comments. Tested on Linux/ARM builder. R=rsc, golang-dev CC=golang-dev https://golang.org/cl/5696047
2012-02-23 13:43:14 -07:00
// Use kernel version instead of native armcas in asm_arm.s.
// See ../sync/atomic/asm_linux_arm.s for details.
runtime: use kernel-supplied cas on linux/arm Using the kernel-supplied compare-and-swap code on linux/arm means that runtime doesn't have to care whether this is GOARM=5 or GOARM=6 anymore. Fixes #1494. R=r, r2 CC=golang-dev https://golang.org/cl/4245043
2011-02-25 12:29:55 -07:00
TEXT cas<>(SB),7,$0
MOVW $0xffff0fc0, PC
TEXT runtime·cas(SB),7,$0
MOVW valptr+0(FP), R2
MOVW old+4(FP), R0
runtime, sync/atomic: fix arm cas Works around bug in kernel implementation on old ARM5 kernels. Bug was fixed on 26 Nov 2007 (between 2.6.23 and 2.6.24) but old kernels persist. Fixes #1750. R=dfc, golang-dev CC=golang-dev https://golang.org/cl/4436072
2011-05-02 08:49:19 -06:00
casagain:
runtime: use kernel-supplied cas on linux/arm Using the kernel-supplied compare-and-swap code on linux/arm means that runtime doesn't have to care whether this is GOARM=5 or GOARM=6 anymore. Fixes #1494. R=r, r2 CC=golang-dev https://golang.org/cl/4245043
2011-02-25 12:29:55 -07:00
MOVW new+8(FP), R1
BL cas<>(SB)
runtime, sync/atomic: fix arm cas Works around bug in kernel implementation on old ARM5 kernels. Bug was fixed on 26 Nov 2007 (between 2.6.23 and 2.6.24) but old kernels persist. Fixes #1750. R=dfc, golang-dev CC=golang-dev https://golang.org/cl/4436072
2011-05-02 08:49:19 -06:00
BCC cascheck
MOVW $1, R0
runtime: use kernel-supplied cas on linux/arm Using the kernel-supplied compare-and-swap code on linux/arm means that runtime doesn't have to care whether this is GOARM=5 or GOARM=6 anymore. Fixes #1494. R=r, r2 CC=golang-dev https://golang.org/cl/4245043
2011-02-25 12:29:55 -07:00
RET
runtime, sync/atomic: fix arm cas Works around bug in kernel implementation on old ARM5 kernels. Bug was fixed on 26 Nov 2007 (between 2.6.23 and 2.6.24) but old kernels persist. Fixes #1750. R=dfc, golang-dev CC=golang-dev https://golang.org/cl/4436072
2011-05-02 08:49:19 -06:00
cascheck:
// Kernel lies; double-check.
MOVW valptr+0(FP), R2
MOVW old+4(FP), R0
MOVW 0(R2), R3
CMP R0, R3
BEQ casagain
MOVW $0, R0
RET
runtime: use kernel-supplied cas on linux/arm Using the kernel-supplied compare-and-swap code on linux/arm means that runtime doesn't have to care whether this is GOARM=5 or GOARM=6 anymore. Fixes #1494. R=r, r2 CC=golang-dev https://golang.org/cl/4245043
2011-02-25 12:29:55 -07:00
TEXT runtime·casp(SB),7,$0
B runtime·cas(SB)
runtime: improve Linux mutex The implementation is hybrid active/passive spin/blocking mutex. The design minimizes amount of context switches and futex calls. The idea is that all critical sections in runtime are intentially small, so pure blocking mutex behaves badly causing a lot of context switches, thread parking/unparking and kernel calls. Note that some synthetic benchmarks become somewhat slower, that's due to increased contention on other data structures, it should not affect programs that do any real work. On 2 x Intel E5620, 8 HT cores, 2.4GHz benchmark old ns/op new ns/op delta BenchmarkSelectContended 521.00 503.00 -3.45% BenchmarkSelectContended-2 661.00 320.00 -51.59% BenchmarkSelectContended-4 1139.00 629.00 -44.78% BenchmarkSelectContended-8 2870.00 878.00 -69.41% BenchmarkSelectContended-16 5276.00 818.00 -84.50% BenchmarkChanContended 112.00 103.00 -8.04% BenchmarkChanContended-2 631.00 174.00 -72.42% BenchmarkChanContended-4 682.00 272.00 -60.12% BenchmarkChanContended-8 1601.00 520.00 -67.52% BenchmarkChanContended-16 3100.00 372.00 -88.00% BenchmarkChanSync 253.00 239.00 -5.53% BenchmarkChanSync-2 5030.00 4648.00 -7.59% BenchmarkChanSync-4 4826.00 4694.00 -2.74% BenchmarkChanSync-8 4778.00 4713.00 -1.36% BenchmarkChanSync-16 5289.00 4710.00 -10.95% BenchmarkChanProdCons0 273.00 254.00 -6.96% BenchmarkChanProdCons0-2 599.00 400.00 -33.22% BenchmarkChanProdCons0-4 1168.00 659.00 -43.58% BenchmarkChanProdCons0-8 2831.00 1057.00 -62.66% BenchmarkChanProdCons0-16 4197.00 1037.00 -75.29% BenchmarkChanProdCons10 150.00 140.00 -6.67% BenchmarkChanProdCons10-2 607.00 268.00 -55.85% BenchmarkChanProdCons10-4 1137.00 404.00 -64.47% BenchmarkChanProdCons10-8 2115.00 828.00 -60.85% BenchmarkChanProdCons10-16 4283.00 855.00 -80.04% BenchmarkChanProdCons100 117.00 110.00 -5.98% BenchmarkChanProdCons100-2 558.00 218.00 -60.93% BenchmarkChanProdCons100-4 722.00 287.00 -60.25% BenchmarkChanProdCons100-8 1840.00 431.00 -76.58% BenchmarkChanProdCons100-16 3394.00 448.00 -86.80% BenchmarkChanProdConsWork0 2014.00 1996.00 -0.89% BenchmarkChanProdConsWork0-2 1207.00 1127.00 -6.63% BenchmarkChanProdConsWork0-4 1913.00 611.00 -68.06% BenchmarkChanProdConsWork0-8 3016.00 949.00 -68.53% BenchmarkChanProdConsWork0-16 4320.00 1154.00 -73.29% BenchmarkChanProdConsWork10 1906.00 1897.00 -0.47% BenchmarkChanProdConsWork10-2 1123.00 1033.00 -8.01% BenchmarkChanProdConsWork10-4 1076.00 571.00 -46.93% BenchmarkChanProdConsWork10-8 2748.00 1096.00 -60.12% BenchmarkChanProdConsWork10-16 4600.00 1105.00 -75.98% BenchmarkChanProdConsWork100 1884.00 1852.00 -1.70% BenchmarkChanProdConsWork100-2 1235.00 1146.00 -7.21% BenchmarkChanProdConsWork100-4 1217.00 619.00 -49.14% BenchmarkChanProdConsWork100-8 1534.00 509.00 -66.82% BenchmarkChanProdConsWork100-16 4126.00 918.00 -77.75% BenchmarkSyscall 34.40 33.30 -3.20% BenchmarkSyscall-2 160.00 121.00 -24.38% BenchmarkSyscall-4 131.00 136.00 +3.82% BenchmarkSyscall-8 139.00 131.00 -5.76% BenchmarkSyscall-16 161.00 168.00 +4.35% BenchmarkSyscallWork 950.00 950.00 +0.00% BenchmarkSyscallWork-2 481.00 480.00 -0.21% BenchmarkSyscallWork-4 268.00 270.00 +0.75% BenchmarkSyscallWork-8 156.00 169.00 +8.33% BenchmarkSyscallWork-16 188.00 184.00 -2.13% BenchmarkSemaSyntNonblock 36.40 35.60 -2.20% BenchmarkSemaSyntNonblock-2 81.40 45.10 -44.59% BenchmarkSemaSyntNonblock-4 126.00 108.00 -14.29% BenchmarkSemaSyntNonblock-8 112.00 112.00 +0.00% BenchmarkSemaSyntNonblock-16 110.00 112.00 +1.82% BenchmarkSemaSyntBlock 35.30 35.30 +0.00% BenchmarkSemaSyntBlock-2 118.00 124.00 +5.08% BenchmarkSemaSyntBlock-4 105.00 108.00 +2.86% BenchmarkSemaSyntBlock-8 101.00 111.00 +9.90% BenchmarkSemaSyntBlock-16 112.00 118.00 +5.36% BenchmarkSemaWorkNonblock 810.00 811.00 +0.12% BenchmarkSemaWorkNonblock-2 476.00 414.00 -13.03% BenchmarkSemaWorkNonblock-4 238.00 228.00 -4.20% BenchmarkSemaWorkNonblock-8 140.00 126.00 -10.00% BenchmarkSemaWorkNonblock-16 117.00 116.00 -0.85% BenchmarkSemaWorkBlock 810.00 811.00 +0.12% BenchmarkSemaWorkBlock-2 454.00 466.00 +2.64% BenchmarkSemaWorkBlock-4 243.00 241.00 -0.82% BenchmarkSemaWorkBlock-8 145.00 137.00 -5.52% BenchmarkSemaWorkBlock-16 132.00 123.00 -6.82% BenchmarkContendedSemaphore 123.00 102.00 -17.07% BenchmarkContendedSemaphore-2 34.80 34.90 +0.29% BenchmarkContendedSemaphore-4 34.70 34.80 +0.29% BenchmarkContendedSemaphore-8 34.70 34.70 +0.00% BenchmarkContendedSemaphore-16 34.80 34.70 -0.29% BenchmarkMutex 26.80 26.00 -2.99% BenchmarkMutex-2 108.00 45.20 -58.15% BenchmarkMutex-4 103.00 127.00 +23.30% BenchmarkMutex-8 109.00 147.00 +34.86% BenchmarkMutex-16 102.00 152.00 +49.02% BenchmarkMutexSlack 27.00 26.90 -0.37% BenchmarkMutexSlack-2 149.00 165.00 +10.74% BenchmarkMutexSlack-4 121.00 209.00 +72.73% BenchmarkMutexSlack-8 101.00 158.00 +56.44% BenchmarkMutexSlack-16 97.00 129.00 +32.99% BenchmarkMutexWork 792.00 794.00 +0.25% BenchmarkMutexWork-2 407.00 409.00 +0.49% BenchmarkMutexWork-4 220.00 209.00 -5.00% BenchmarkMutexWork-8 267.00 160.00 -40.07% BenchmarkMutexWork-16 315.00 300.00 -4.76% BenchmarkMutexWorkSlack 792.00 793.00 +0.13% BenchmarkMutexWorkSlack-2 406.00 404.00 -0.49% BenchmarkMutexWorkSlack-4 225.00 212.00 -5.78% BenchmarkMutexWorkSlack-8 268.00 136.00 -49.25% BenchmarkMutexWorkSlack-16 300.00 300.00 +0.00% BenchmarkRWMutexWrite100 27.10 27.00 -0.37% BenchmarkRWMutexWrite100-2 33.10 40.80 +23.26% BenchmarkRWMutexWrite100-4 113.00 88.10 -22.04% BenchmarkRWMutexWrite100-8 119.00 95.30 -19.92% BenchmarkRWMutexWrite100-16 148.00 109.00 -26.35% BenchmarkRWMutexWrite10 29.60 29.40 -0.68% BenchmarkRWMutexWrite10-2 111.00 61.40 -44.68% BenchmarkRWMutexWrite10-4 270.00 208.00 -22.96% BenchmarkRWMutexWrite10-8 204.00 185.00 -9.31% BenchmarkRWMutexWrite10-16 261.00 190.00 -27.20% BenchmarkRWMutexWorkWrite100 1040.00 1036.00 -0.38% BenchmarkRWMutexWorkWrite100-2 593.00 580.00 -2.19% BenchmarkRWMutexWorkWrite100-4 470.00 365.00 -22.34% BenchmarkRWMutexWorkWrite100-8 468.00 289.00 -38.25% BenchmarkRWMutexWorkWrite100-16 604.00 374.00 -38.08% BenchmarkRWMutexWorkWrite10 951.00 951.00 +0.00% BenchmarkRWMutexWorkWrite10-2 1001.00 928.00 -7.29% BenchmarkRWMutexWorkWrite10-4 1555.00 1006.00 -35.31% BenchmarkRWMutexWorkWrite10-8 2085.00 1171.00 -43.84% BenchmarkRWMutexWorkWrite10-16 2082.00 1614.00 -22.48% R=rsc, iant, msolo, fw, iant CC=golang-dev https://golang.org/cl/4711045
2011-07-29 10:44:06 -06:00
TEXT runtime·osyield(SB),7,$0
MOVW $SYS_sched_yield, R7
SWI $0
RET
runtime: use sched_getaffinity for runtime.NumCPU() on Linux Fixes #3921. R=iant CC=golang-dev https://golang.org/cl/6448132
2012-08-09 20:05:26 -06:00
TEXT runtime·sched_getaffinity(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW $SYS_sched_getaffinity, R7
SWI $0
RET
net, runtime: enable runtime-integrated pollster on Linux/ARM. Results from linux/arm on a Samsung Chromebook (from dfc): localhost(~/go/src/pkg/net) % ~/go/misc/benchcmp {old,new}.txt benchmark old ns/op new ns/op delta BenchmarkTCP4OneShot 568840 350526 -38.38% BenchmarkTCP4OneShot-2 359054 206708 -42.43% BenchmarkTCP4OneShotTimeout 637464 363550 -42.97% BenchmarkTCP4OneShotTimeout-2 374255 216695 -42.10% BenchmarkTCP4Persistent 184974 64984 -64.87% BenchmarkTCP4Persistent-2 109902 47195 -57.06% BenchmarkTCP4PersistentTimeout 210039 64789 -69.15% BenchmarkTCP4PersistentTimeout-2 124284 43374 -65.10% BenchmarkTCP6OneShot 672278 362116 -46.14% BenchmarkTCP6OneShot-2 383631 216400 -43.59% BenchmarkTCP6OneShotTimeout 680740 378306 -44.43% BenchmarkTCP6OneShotTimeout-2 397524 230152 -42.10% BenchmarkTCP6Persistent 172346 65292 -62.12% BenchmarkTCP6Persistent-2 106229 42096 -60.37% BenchmarkTCP6PersistentTimeout 161149 65138 -59.58% BenchmarkTCP6PersistentTimeout-2 152276 44548 -70.75% R=golang-dev, dave, bradfitz, dvyukov, rsc CC=golang-dev https://golang.org/cl/7820045
2013-03-15 14:01:56 -06:00
// int32 runtime·epollcreate(int32 size)
TEXT runtime·epollcreate(SB),7,$0
MOVW 0(FP), R0
MOVW $SYS_epoll_create, R7
SWI $0
RET
// int32 runtime·epollcreate1(int32 flags)
TEXT runtime·epollcreate1(SB),7,$0
MOVW 0(FP), R0
MOVW $SYS_epoll_create1, R7
SWI $0
RET
// int32 runtime·epollctl(int32 epfd, int32 op, int32 fd, EpollEvent *ev)
TEXT runtime·epollctl(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW 12(FP), R3
MOVW $SYS_epoll_ctl, R7
SWI $0
RET
// int32 runtime·epollwait(int32 epfd, EpollEvent *ev, int32 nev, int32 timeout)
TEXT runtime·epollwait(SB),7,$0
MOVW 0(FP), R0
MOVW 4(FP), R1
MOVW 8(FP), R2
MOVW 12(FP), R3
MOVW $SYS_epoll_wait, R7
SWI $0
RET
// void runtime·closeonexec(int32 fd)
TEXT runtime·closeonexec(SB),7,$0
MOVW 0(FP), R0 // fd
MOVW $2, R1 // F_SETFD
MOVW $1, R2 // FD_CLOEXEC
MOVW $SYS_fcntl, R7
SWI $0
RET
Copy Permalink