qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-11-06 11:26:12 -07:00
Code Releases Activity
bf2ef26be3
go/src/runtime/sys_linux_ppc64x.s

950 lines
22 KiB
ArmAsm
Raw Normal View History

[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// Copyright 2014 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.
all: add //go:build lines to assembly files Don't add them to files in vendor and cmd/vendor though. These will be pulled in by updating the respective dependencies. For #41184 Change-Id: Icc57458c9b3033c347124323f33084c85b224c70 Reviewed-on: https://go-review.googlesource.com/c/go/+/319389 Trust: Tobias Klauser <tobias.klauser@gmail.com> Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
2021-05-12 09:55:42 -06:00
//go:build linux && (ppc64 || ppc64le)
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
//
all: power64 is now ppc64 Fixes #8654. LGTM=austin R=austin CC=golang-codereviews https://golang.org/cl/180600043
2014-12-05 17:13:20 -07:00
// System calls and other sys.stuff for ppc64, Linux
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
//
[dev.cc] runtime: convert power64 assembly files for C to Go transition The power64 equivalent of CL 168510043 LGTM=rsc R=rsc CC=golang-codereviews https://golang.org/cl/178940043
2014-11-18 13:19:26 -07:00
#include "go_asm.h"
#include "go_tls.h"
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
#include "textflag.h"
reflect, runtime, runtime/cgo: use ppc64 asm constant for fixed frame size Shared libraries on ppc64le will require a larger minimum stack frame (because the ABI mandates that the TOC pointer is available at 24(R1)). Part 3 of that is using a #define in the ppc64 assembly to refer to the size of the fixed part of the stack (finding all these took me about a week!). Change-Id: I50f22fe1c47af1ec59da1bd7ea8f84a4750df9b7 Reviewed-on: https://go-review.googlesource.com/15525 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-10-08 03:34:29 -06:00
#include "asm_ppc64x.h"
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#define SYS_exit 1
#define SYS_read 3
#define SYS_write 4
#define SYS_open 5
#define SYS_close 6
runtime: log all thread stack traces during GODEBUG=crash on Linux and OS X Normally, a panic/throw only shows the thread stack for the current thread and all paused goroutines. Goroutines running on other threads, or other threads running on their system stacks, are opaque. Change that when GODEBUG=crash, by passing a SIGQUIT around to all the threads when GODEBUG=crash. If this works out reasonably well, we might make the SIGQUIT relay part of the standard panic/throw death, perhaps eliding idle m's. Change-Id: If7dd354f7f3a6e326d17c254afcf4f7681af2f8b Reviewed-on: https://go-review.googlesource.com/2811 Reviewed-by: Rick Hudson <rlh@golang.org>
2015-01-14 09:18:24 -07:00
#define SYS_getpid 20
#define SYS_kill 37
runtime: inform arena placement using sbrk(0) On 32-bit architectures (or if we fail to map a 64-bit-style arena), we try to map the heap arena just above the end of the process image. While we can accept any address, using lower addresses is preferable because lower addresses cause us to map less of the heap bitmap. However, if a program is linked against C code that has global constructors, those constructors may call brk/sbrk to allocate memory (e.g., many C malloc implementations do this for small allocations). The brk also starts just above the process image, so this may adjust the brk past the beginning of where we want to put the heap arena. In this case, the kernel will pick a different address for the arena and it will usually be very high (at least, as these things go in a 32-bit address space). Fix this by consulting the current value of the brk and using this in addition to the end of the process image to compute the initial arena placement. This is implemented only on Linux currently, since we have no evidence that it's an issue on any other OSes. Fixes #19831. Change-Id: Id64b45d08d8c91e4f50d92d0339146250b04f2f8 Reviewed-on: https://go-review.googlesource.com/39810 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-04-06 12:32:37 -06:00
#define SYS_brk 45
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#define SYS_fcntl 55
#define SYS_mmap 90
#define SYS_munmap 91
#define SYS_setitimer 104
#define SYS_clone 120
#define SYS_sched_yield 158
runtime: change GNU/Linux usleep to use nanosleep Ever since we added sleep to the runtime back in 2008, we've implemented it on GNU/Linux with the select (or pselect or pselect6) system call. But the Linux kernel has a nanosleep system call, which should be a tiny bit more efficient since it doesn't have to check to see whether there are any file descriptors. So use it. Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a Reviewed-on: https://go-review.googlesource.com/108538 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2018-04-20 16:30:52 -06:00
#define SYS_nanosleep 162
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#define SYS_rt_sigreturn 172
#define SYS_rt_sigaction 173
#define SYS_rt_sigprocmask 174
#define SYS_sigaltstack 185
#define SYS_madvise 205
#define SYS_mincore 206
#define SYS_gettid 207
#define SYS_futex 221
#define SYS_sched_getaffinity 223
#define SYS_exit_group 234
#define SYS_epoll_create 236
#define SYS_epoll_ctl 237
#define SYS_epoll_wait 238
runtime: add timer_create syscalls for Linux Updates #35057 Change-Id: Id702b502fa4e4005ba1e450a945bc4420a8a8b8c Reviewed-on: https://go-review.googlesource.com/c/go/+/342052 Run-TryBot: Rhys Hiltner <rhys@justin.tv> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com> Trust: Than McIntosh <thanm@google.com>
2021-08-13 09:34:25 -06:00
#define SYS_timer_create 240
#define SYS_timer_settime 241
#define SYS_timer_delete 244
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#define SYS_clock_gettime 246
runtime: use tgkill for raise raise uses tkill to send a signal to the current thread. For this use, tgkill is functionally equivalent to tkill expect that it also takes the pid as the first argument. Using tgkill makes it simpler to run a Go program in a strict sandbox. With kill and tgkill, the sandbox policy (e.g., seccomp) can prevent the program from sending signals to other processes by checking that the first argument == getpid(). With tkill, the policy must whitelist all tids in the process, which is effectively impossible given Go's dynamic thread creation. Fixes #27548 Change-Id: I8ed282ef1f7215b02ef46de144493e36454029ea Reviewed-on: https://go-review.googlesource.com/133975 Run-TryBot: Michael Pratt <mpratt@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-09-06 18:21:59 -06:00
#define SYS_tgkill 250
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#define SYS_epoll_create1 315
runtime: define nonblockingPipe This requires defining pipe, pipe2, and setNonblock for various platforms. The new function is currently only used on AIX. It will be used by later CLs in this series. Updates #27707 Change-Id: Id2f987b66b4c66a3ef40c22484ff1d14f58e9b31 Reviewed-on: https://go-review.googlesource.com/c/go/+/171822 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Michael Knyszek <mknyszek@google.com> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2019-04-03 17:31:13 -06:00
#define SYS_pipe2 317
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·exit(SB),NOSPLIT|NOFRAME,$0-4
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW code+0(FP), R3
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_exit_group
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: make it possible to exit Go-created threads Currently, threads created by the runtime exist until the whole program exits. For #14592 and #20395, we want to be able to exit and clean up threads created by the runtime. This commit implements that mechanism. The main difficulty is how to clean up the g0 stack. In cgo mode and on Solaris and Windows where the OS manages thread stacks, we simply arrange to return from mstart and let the system clean up the thread. If the runtime allocated the g0 stack, then we use a new exitThread syscall wrapper that arranges to clear a flag in the M once the stack can safely be reaped and call the thread termination syscall. exitThread is based on the existing exit1 wrapper, which was always meant to terminate the calling thread. However, exit1 has never been used since it was introduced 9 years ago, so it was broken on several platforms. exitThread also has the additional complication of having to flag that the stack is unused, which requires some tricks on platforms that use the stack for syscalls. This still leaves the problem of how to reap the unused g0 stacks. For this, we move the M from allm to a new freem list as part of the M exiting. Later, allocm scans the freem list, finds Ms that are marked as done with their stack, removes these from the list and frees their g0 stacks. This also allows these Ms to be garbage collected. This CL does not yet use any of this functionality. Follow-up CLs will. Likewise, there are no new tests in this CL because we'll need follow-up functionality to test it. Change-Id: Ic851ee74227b6d39c6fc1219fc71b45d3004bc63 Reviewed-on: https://go-review.googlesource.com/46037 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2017-06-16 13:54:21 -06:00
// func exitThread(wait *uint32)
TEXT runtime·exitThread(SB),NOSPLIT|NOFRAME,$0-8
MOVD wait+0(FP), R1
// We're done using the stack.
MOVW $0, R2
SYNC
MOVW R2, (R1)
MOVW $0, R3 // exit code
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_exit
runtime: make it possible to exit Go-created threads Currently, threads created by the runtime exist until the whole program exits. For #14592 and #20395, we want to be able to exit and clean up threads created by the runtime. This commit implements that mechanism. The main difficulty is how to clean up the g0 stack. In cgo mode and on Solaris and Windows where the OS manages thread stacks, we simply arrange to return from mstart and let the system clean up the thread. If the runtime allocated the g0 stack, then we use a new exitThread syscall wrapper that arranges to clear a flag in the M once the stack can safely be reaped and call the thread termination syscall. exitThread is based on the existing exit1 wrapper, which was always meant to terminate the calling thread. However, exit1 has never been used since it was introduced 9 years ago, so it was broken on several platforms. exitThread also has the additional complication of having to flag that the stack is unused, which requires some tricks on platforms that use the stack for syscalls. This still leaves the problem of how to reap the unused g0 stacks. For this, we move the M from allm to a new freem list as part of the M exiting. Later, allocm scans the freem list, finds Ms that are marked as done with their stack, removes these from the list and frees their g0 stacks. This also allows these Ms to be garbage collected. This CL does not yet use any of this functionality. Follow-up CLs will. Likewise, there are no new tests in this CL because we'll need follow-up functionality to test it. Change-Id: Ic851ee74227b6d39c6fc1219fc71b45d3004bc63 Reviewed-on: https://go-review.googlesource.com/46037 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
2017-06-16 13:54:21 -06:00
JMP 0(PC)
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·open(SB),NOSPLIT|NOFRAME,$0-20
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD name+0(FP), R3
MOVW mode+8(FP), R4
MOVW perm+12(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_open
runtime: Update open/close/read/write to return -1 on error. Error detection code copied from syscall, where presumably we actually do it right. Note that we throw the errno away. The runtime doesn't use it. Fixes #10052 Change-Id: I8de77dda6bf287276b137646c26b84fa61554ec8 Reviewed-on: https://go-review.googlesource.com/6571 Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
2015-03-02 21:16:48 -07:00
BVC 2(PC)
MOVW $-1, R3
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+16(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·closefd(SB),NOSPLIT|NOFRAME,$0-12
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW fd+0(FP), R3
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_close
runtime: Update open/close/read/write to return -1 on error. Error detection code copied from syscall, where presumably we actually do it right. Note that we throw the errno away. The runtime doesn't use it. Fixes #10052 Change-Id: I8de77dda6bf287276b137646c26b84fa61554ec8 Reviewed-on: https://go-review.googlesource.com/6571 Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
2015-03-02 21:16:48 -07:00
BVC 2(PC)
MOVW $-1, R3
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+8(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: wrap nanotime, walltime, and write In preparation for general faketime support, this renames the existing nanotime, walltime, and write functions to nanotime1, walltime1, and write1 and wraps them with trivial Go functions. This will let us inject different implementations on all platforms when faketime is enabled. Updates #30439. Change-Id: Ice5ccc513a32a6d89ea051638676d3ee05b00418 Reviewed-on: https://go-review.googlesource.com/c/go/+/192738 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2019-09-01 08:37:44 -06:00
TEXT runtime·write1(SB),NOSPLIT|NOFRAME,$0-28
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD fd+0(FP), R3
MOVD p+8(FP), R4
MOVW n+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_write
runtime: Update open/close/read/write to return -1 on error. Error detection code copied from syscall, where presumably we actually do it right. Note that we throw the errno away. The runtime doesn't use it. Fixes #10052 Change-Id: I8de77dda6bf287276b137646c26b84fa61554ec8 Reviewed-on: https://go-review.googlesource.com/6571 Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
2015-03-02 21:16:48 -07:00
BVC 2(PC)
runtime: change read and write to return negative errno value The internal read and write functions used to return -1 on error; change them to return a negative errno value instead. This will be used by later CLs in this series. For most targets this is a simplification, although for ones that call into libc it is a complication. Updates #27707 Change-Id: Id02bf9487f03e7e88e4f2b85e899e986738697ad Reviewed-on: https://go-review.googlesource.com/c/go/+/171823 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Michael Knyszek <mknyszek@google.com>
2019-04-05 12:42:37 -06:00
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·read(SB),NOSPLIT|NOFRAME,$0-28
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW fd+0(FP), R3
MOVD p+8(FP), R4
MOVW n+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_read
runtime: Update open/close/read/write to return -1 on error. Error detection code copied from syscall, where presumably we actually do it right. Note that we throw the errno away. The runtime doesn't use it. Fixes #10052 Change-Id: I8de77dda6bf287276b137646c26b84fa61554ec8 Reviewed-on: https://go-review.googlesource.com/6571 Run-TryBot: Keith Randall <khr@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
2015-03-02 21:16:48 -07:00
BVC 2(PC)
runtime: change read and write to return negative errno value The internal read and write functions used to return -1 on error; change them to return a negative errno value instead. This will be used by later CLs in this series. For most targets this is a simplification, although for ones that call into libc it is a complication. Updates #27707 Change-Id: Id02bf9487f03e7e88e4f2b85e899e986738697ad Reviewed-on: https://go-review.googlesource.com/c/go/+/171823 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Michael Knyszek <mknyszek@google.com>
2019-04-05 12:42:37 -06:00
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: define nonblockingPipe This requires defining pipe, pipe2, and setNonblock for various platforms. The new function is currently only used on AIX. It will be used by later CLs in this series. Updates #27707 Change-Id: Id2f987b66b4c66a3ef40c22484ff1d14f58e9b31 Reviewed-on: https://go-review.googlesource.com/c/go/+/171822 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Michael Knyszek <mknyszek@google.com> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2019-04-03 17:31:13 -06:00
// func pipe2(flags int32) (r, w int32, errno int32)
TEXT runtime·pipe2(SB),NOSPLIT|NOFRAME,$0-20
ADD $FIXED_FRAME+8, R1, R3
MOVW flags+0(FP), R4
SYSCALL $SYS_pipe2
MOVW R3, errno+16(FP)
RET
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
TEXT runtime·usleep(SB),NOSPLIT,$16-4
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVW usec+0(FP), R3
MOVD R3, R5
MOVW $1000000, R4
DIVD R4, R3
[dev.power64] runtime: fix usleep not to smash saved LR LGTM=minux R=minux CC=golang-codereviews https://golang.org/cl/130760043
2014-08-15 13:28:08 -06:00
MOVD R3, 8(R1)
runtime: change GNU/Linux usleep to use nanosleep Ever since we added sleep to the runtime back in 2008, we've implemented it on GNU/Linux with the select (or pselect or pselect6) system call. But the Linux kernel has a nanosleep system call, which should be a tiny bit more efficient since it doesn't have to check to see whether there are any file descriptors. So use it. Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a Reviewed-on: https://go-review.googlesource.com/108538 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2018-04-20 16:30:52 -06:00
MOVW $1000, R4
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MULLD R3, R4
SUB R4, R5
[dev.power64] runtime: fix usleep not to smash saved LR LGTM=minux R=minux CC=golang-codereviews https://golang.org/cl/130760043
2014-08-15 13:28:08 -06:00
MOVD R5, 16(R1)
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: change GNU/Linux usleep to use nanosleep Ever since we added sleep to the runtime back in 2008, we've implemented it on GNU/Linux with the select (or pselect or pselect6) system call. But the Linux kernel has a nanosleep system call, which should be a tiny bit more efficient since it doesn't have to check to see whether there are any file descriptors. So use it. Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a Reviewed-on: https://go-review.googlesource.com/108538 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2018-04-20 16:30:52 -06:00
// nanosleep(&ts, 0)
ADD $8, R1, R3
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVW $0, R4
runtime: change GNU/Linux usleep to use nanosleep Ever since we added sleep to the runtime back in 2008, we've implemented it on GNU/Linux with the select (or pselect or pselect6) system call. But the Linux kernel has a nanosleep system call, which should be a tiny bit more efficient since it doesn't have to check to see whether there are any file descriptors. So use it. Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a Reviewed-on: https://go-review.googlesource.com/108538 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2018-04-20 16:30:52 -06:00
SYSCALL $SYS_nanosleep
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: set m.procid always on Linux For debuggers and other program inspectors. Fixes #9914. Change-Id: I670728cea28c045e6eaba1808c550ee2f34d16ff Reviewed-on: https://go-review.googlesource.com/11341 Reviewed-by: Austin Clements <austin@google.com>
2015-06-22 10:32:05 -06:00
TEXT runtime·gettid(SB),NOSPLIT,$0-4
SYSCALL $SYS_gettid
MOVW R3, ret+0(FP)
RET
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·raise(SB),NOSPLIT|NOFRAME,$0
runtime: use tgkill for raise raise uses tkill to send a signal to the current thread. For this use, tgkill is functionally equivalent to tkill expect that it also takes the pid as the first argument. Using tgkill makes it simpler to run a Go program in a strict sandbox. With kill and tgkill, the sandbox policy (e.g., seccomp) can prevent the program from sending signals to other processes by checking that the first argument == getpid(). With tkill, the policy must whitelist all tids in the process, which is effectively impossible given Go's dynamic thread creation. Fixes #27548 Change-Id: I8ed282ef1f7215b02ef46de144493e36454029ea Reviewed-on: https://go-review.googlesource.com/133975 Run-TryBot: Michael Pratt <mpratt@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-09-06 18:21:59 -06:00
SYSCALL $SYS_getpid
MOVW R3, R14
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_gettid
runtime: use tgkill for raise raise uses tkill to send a signal to the current thread. For this use, tgkill is functionally equivalent to tkill expect that it also takes the pid as the first argument. Using tgkill makes it simpler to run a Go program in a strict sandbox. With kill and tgkill, the sandbox policy (e.g., seccomp) can prevent the program from sending signals to other processes by checking that the first argument == getpid(). With tkill, the policy must whitelist all tids in the process, which is effectively impossible given Go's dynamic thread creation. Fixes #27548 Change-Id: I8ed282ef1f7215b02ef46de144493e36454029ea Reviewed-on: https://go-review.googlesource.com/133975 Run-TryBot: Michael Pratt <mpratt@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-09-06 18:21:59 -06:00
MOVW R3, R4 // arg 2 tid
MOVW R14, R3 // arg 1 pid
MOVW sig+0(FP), R5 // arg 3
SYSCALL $SYS_tgkill
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·raiseproc(SB),NOSPLIT|NOFRAME,$0
runtime: log all thread stack traces during GODEBUG=crash on Linux and OS X Normally, a panic/throw only shows the thread stack for the current thread and all paused goroutines. Goroutines running on other threads, or other threads running on their system stacks, are opaque. Change that when GODEBUG=crash, by passing a SIGQUIT around to all the threads when GODEBUG=crash. If this works out reasonably well, we might make the SIGQUIT relay part of the standard panic/throw death, perhaps eliding idle m's. Change-Id: If7dd354f7f3a6e326d17c254afcf4f7681af2f8b Reviewed-on: https://go-review.googlesource.com/2811 Reviewed-by: Rick Hudson <rlh@golang.org>
2015-01-14 09:18:24 -07:00
SYSCALL $SYS_getpid
MOVW R3, R3 // arg 1 pid
MOVW sig+0(FP), R4 // arg 2
SYSCALL $SYS_kill
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
runtime: log all thread stack traces during GODEBUG=crash on Linux and OS X Normally, a panic/throw only shows the thread stack for the current thread and all paused goroutines. Goroutines running on other threads, or other threads running on their system stacks, are opaque. Change that when GODEBUG=crash, by passing a SIGQUIT around to all the threads when GODEBUG=crash. If this works out reasonably well, we might make the SIGQUIT relay part of the standard panic/throw death, perhaps eliding idle m's. Change-Id: If7dd354f7f3a6e326d17c254afcf4f7681af2f8b Reviewed-on: https://go-review.googlesource.com/2811 Reviewed-by: Rick Hudson <rlh@golang.org>
2015-01-14 09:18:24 -07:00
runtime: M-targeted signals for Linux We'll add a test once all of the POSIX platforms are done. For #10958, #24543. Change-Id: If7e3f14e8391791364877629bf415d9f8e788b0a Reviewed-on: https://go-review.googlesource.com/c/go/+/201401 Run-TryBot: Austin Clements <austin@google.com> Reviewed-by: Cherry Zhang <cherryyz@google.com>
2019-10-03 08:19:38 -06:00
TEXT ·getpid(SB),NOSPLIT|NOFRAME,$0-8
SYSCALL $SYS_getpid
MOVD R3, ret+0(FP)
RET
TEXT ·tgkill(SB),NOSPLIT|NOFRAME,$0-24
MOVD tgid+0(FP), R3
MOVD tid+8(FP), R4
MOVD sig+16(FP), R5
SYSCALL $SYS_tgkill
RET
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·setitimer(SB),NOSPLIT|NOFRAME,$0-24
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW mode+0(FP), R3
MOVD new+8(FP), R4
MOVD old+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_setitimer
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: add timer_create syscalls for Linux Updates #35057 Change-Id: Id702b502fa4e4005ba1e450a945bc4420a8a8b8c Reviewed-on: https://go-review.googlesource.com/c/go/+/342052 Run-TryBot: Rhys Hiltner <rhys@justin.tv> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com> Trust: Than McIntosh <thanm@google.com>
2021-08-13 09:34:25 -06:00
TEXT runtime·timer_create(SB),NOSPLIT,$0-28
MOVW clockid+0(FP), R3
MOVD sevp+8(FP), R4
MOVD timerid+16(FP), R5
SYSCALL $SYS_timer_create
MOVW R3, ret+24(FP)
RET
TEXT runtime·timer_settime(SB),NOSPLIT,$0-28
MOVW timerid+0(FP), R3
MOVW flags+4(FP), R4
MOVD new+8(FP), R5
MOVD old+16(FP), R6
SYSCALL $SYS_timer_settime
MOVW R3, ret+24(FP)
RET
TEXT runtime·timer_delete(SB),NOSPLIT,$0-12
MOVW timerid+0(FP), R3
SYSCALL $SYS_timer_delete
MOVW R3, ret+8(FP)
RET
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·mincore(SB),NOSPLIT|NOFRAME,$0-28
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD addr+0(FP), R3
MOVD n+8(FP), R4
MOVD dst+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_mincore
runtime: fix errno sign for some mmap and mincore cases The caller of mmap expects it to return a positive errno value, but the linux-arm64 and nacl-386 system calls returned a negative errno value. Correct them to negate the errno value. The caller of mincore expects it to return a negative errno value (yes, this is inconsistent), but the linux-mips64x and linux-ppc64x system call returned a positive errno value. Correct them to negate the errno value. Add a test that mmap returns errno with the correct sign. Brad added a test for mincore's errno value in https://golang.org/cl/19457. Fixes #14297. Change-Id: I2b93f32e679bd1eae1c9aef9ae7bcf0ba39521b5 Reviewed-on: https://go-review.googlesource.com/19455 Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: Minux Ma <minux@golang.org> Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-02-10 22:46:51 -07:00
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: rename walltime1 to walltime Change-Id: Iec9de5ca56eb68d524bbaa0668515dbd09ad38a1 Reviewed-on: https://go-review.googlesource.com/c/go/+/314770 Trust: Ian Lance Taylor <iant@golang.org> Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Michael Knyszek <mknyszek@google.com>
2021-04-28 12:42:34 -06:00
// func walltime() (sec int64, nsec int32)
TEXT runtime·walltime(SB),NOSPLIT,$16-12
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD R1, R15 // R15 is unchanged by C code
MOVD g_m(g), R21 // R21 = m
MOVD $0, R3 // CLOCK_REALTIME
MOVD runtime·vdsoClockgettimeSym(SB), R12 // Check for VDSO availability
CMP R12, R0
BEQ fallback
// Set vdsoPC and vdsoSP for SIGPROF traceback.
runtime: make nanotime1 reentrant Currently, nanotime1 (and walltime1) is not reentrant, in that it sets m.vdsoSP at entry and clears it at exit. If a signal lands in between, and nanotime1 is called from the signal handler, it will clear m.vdsoSP while we are still in nanotime1. If (in the unlikely event) it is signaled again, m.vdsoSP will be wrong, which may cause the stack unwinding code to crash. This CL makes it reentrant, by saving/restoring the previous vdsoPC and vdsoSP, instead of setting it to 0 at exit. TODO: have some way to test? Change-Id: I9ee53b251f1d8a5a489c71d4b4c0df1dee70c3e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/246763 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2020-08-04 18:25:10 -06:00
// Save the old values on stack and restore them on exit,
// so this function is reentrant.
MOVD m_vdsoPC(R21), R4
MOVD m_vdsoSP(R21), R5
MOVD R4, 32(R1)
MOVD R5, 40(R1)
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD LR, R14
runtime: set vdsoSP to caller's SP consistently m.vdsoSP should be set to the SP of the caller of nanotime1, instead of the SP of nanotime1 itself, which matches m.vdsoPC. Otherwise the unmatched vdsoPC and vdsoSP would make the stack trace look like recursive. We already do it correctly on AMD64, 386, and RISCV64. This CL fixes the rest. Fixes #47324. Change-Id: I98b6fcfbe9fc6bdd28b8fe2a1299b7c505371dd4 Reviewed-on: https://go-review.googlesource.com/c/go/+/337590 Trust: Cherry Mui <cherryyz@google.com> Trust: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
2021-09-22 16:32:45 -06:00
MOVD $ret-FIXED_FRAME(FP), R5 // caller's SP
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD R14, m_vdsoPC(R21)
runtime: set vdsoSP to caller's SP consistently m.vdsoSP should be set to the SP of the caller of nanotime1, instead of the SP of nanotime1 itself, which matches m.vdsoPC. Otherwise the unmatched vdsoPC and vdsoSP would make the stack trace look like recursive. We already do it correctly on AMD64, 386, and RISCV64. This CL fixes the rest. Fixes #47324. Change-Id: I98b6fcfbe9fc6bdd28b8fe2a1299b7c505371dd4 Reviewed-on: https://go-review.googlesource.com/c/go/+/337590 Trust: Cherry Mui <cherryyz@google.com> Trust: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
2021-09-22 16:32:45 -06:00
MOVD R5, m_vdsoSP(R21)
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD m_curg(R21), R6
CMP g, R6
BNE noswitch
MOVD m_g0(R21), R7
MOVD (g_sched+gobuf_sp)(R7), R1 // Set SP to g0 stack
noswitch:
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
SUB $16, R1 // Space for results
RLDICR $0, R1, $59, R1 // Align for C code
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD R12, CTR
MOVD R1, R4
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
// Store g on gsignal's stack, so if we receive a signal
// during VDSO code we can find the g.
// If we don't have a signal stack, we won't receive signal,
// so don't bother saving g.
// When using cgo, we already saved g on TLS, also don't save
// g here.
// Also don't save g if we are already on the signal stack.
// We won't get a nested signal.
MOVBZ runtime·iscgo(SB), R22
CMP R22, $0
BNE nosaveg
MOVD m_gsignal(R21), R22 // g.m.gsignal
CMP R22, $0
BEQ nosaveg
CMP g, R22
BEQ nosaveg
MOVD (g_stack+stack_lo)(R22), R22 // g.m.gsignal.stack.lo
MOVD g, (R22)
BL (CTR) // Call from VDSO
MOVD $0, (R22) // clear g slot, R22 is unchanged by C code
JMP finish
nosaveg:
BL (CTR) // Call from VDSO
finish:
MOVD $0, R0 // Restore R0
MOVD 0(R1), R3 // sec
MOVD 8(R1), R5 // nsec
MOVD R15, R1 // Restore SP
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
runtime: make nanotime1 reentrant Currently, nanotime1 (and walltime1) is not reentrant, in that it sets m.vdsoSP at entry and clears it at exit. If a signal lands in between, and nanotime1 is called from the signal handler, it will clear m.vdsoSP while we are still in nanotime1. If (in the unlikely event) it is signaled again, m.vdsoSP will be wrong, which may cause the stack unwinding code to crash. This CL makes it reentrant, by saving/restoring the previous vdsoPC and vdsoSP, instead of setting it to 0 at exit. TODO: have some way to test? Change-Id: I9ee53b251f1d8a5a489c71d4b4c0df1dee70c3e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/246763 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2020-08-04 18:25:10 -06:00
// Restore vdsoPC, vdsoSP
// We don't worry about being signaled between the two stores.
// If we are not in a signal handler, we'll restore vdsoSP to 0,
// and no one will care about vdsoPC. If we are in a signal handler,
// we cannot receive another signal.
MOVD 40(R1), R6
MOVD R6, m_vdsoSP(R21)
MOVD 32(R1), R6
MOVD R6, m_vdsoPC(R21)
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
return:
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVD R3, sec+0(FP)
MOVW R5, nsec+8(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
// Syscall fallback
fallback:
ADD $32, R1, R4
SYSCALL $SYS_clock_gettime
MOVD 32(R1), R3
MOVD 40(R1), R5
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
JMP return
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
runtime: make nanotime1 reentrant Currently, nanotime1 (and walltime1) is not reentrant, in that it sets m.vdsoSP at entry and clears it at exit. If a signal lands in between, and nanotime1 is called from the signal handler, it will clear m.vdsoSP while we are still in nanotime1. If (in the unlikely event) it is signaled again, m.vdsoSP will be wrong, which may cause the stack unwinding code to crash. This CL makes it reentrant, by saving/restoring the previous vdsoPC and vdsoSP, instead of setting it to 0 at exit. TODO: have some way to test? Change-Id: I9ee53b251f1d8a5a489c71d4b4c0df1dee70c3e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/246763 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2020-08-04 18:25:10 -06:00
TEXT runtime·nanotime1(SB),NOSPLIT,$16-8
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD $1, R3 // CLOCK_MONOTONIC
MOVD R1, R15 // R15 is unchanged by C code
MOVD g_m(g), R21 // R21 = m
MOVD runtime·vdsoClockgettimeSym(SB), R12 // Check for VDSO availability
CMP R12, R0
BEQ fallback
// Set vdsoPC and vdsoSP for SIGPROF traceback.
runtime: make nanotime1 reentrant Currently, nanotime1 (and walltime1) is not reentrant, in that it sets m.vdsoSP at entry and clears it at exit. If a signal lands in between, and nanotime1 is called from the signal handler, it will clear m.vdsoSP while we are still in nanotime1. If (in the unlikely event) it is signaled again, m.vdsoSP will be wrong, which may cause the stack unwinding code to crash. This CL makes it reentrant, by saving/restoring the previous vdsoPC and vdsoSP, instead of setting it to 0 at exit. TODO: have some way to test? Change-Id: I9ee53b251f1d8a5a489c71d4b4c0df1dee70c3e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/246763 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2020-08-04 18:25:10 -06:00
// Save the old values on stack and restore them on exit,
// so this function is reentrant.
MOVD m_vdsoPC(R21), R4
MOVD m_vdsoSP(R21), R5
MOVD R4, 32(R1)
MOVD R5, 40(R1)
runtime: set vdsoSP to caller's SP consistently m.vdsoSP should be set to the SP of the caller of nanotime1, instead of the SP of nanotime1 itself, which matches m.vdsoPC. Otherwise the unmatched vdsoPC and vdsoSP would make the stack trace look like recursive. We already do it correctly on AMD64, 386, and RISCV64. This CL fixes the rest. Fixes #47324. Change-Id: I98b6fcfbe9fc6bdd28b8fe2a1299b7c505371dd4 Reviewed-on: https://go-review.googlesource.com/c/go/+/337590 Trust: Cherry Mui <cherryyz@google.com> Trust: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
2021-09-22 16:32:45 -06:00
MOVD LR, R14 // R14 is unchanged by C code
MOVD $ret-FIXED_FRAME(FP), R5 // caller's SP
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD R14, m_vdsoPC(R21)
runtime: set vdsoSP to caller's SP consistently m.vdsoSP should be set to the SP of the caller of nanotime1, instead of the SP of nanotime1 itself, which matches m.vdsoPC. Otherwise the unmatched vdsoPC and vdsoSP would make the stack trace look like recursive. We already do it correctly on AMD64, 386, and RISCV64. This CL fixes the rest. Fixes #47324. Change-Id: I98b6fcfbe9fc6bdd28b8fe2a1299b7c505371dd4 Reviewed-on: https://go-review.googlesource.com/c/go/+/337590 Trust: Cherry Mui <cherryyz@google.com> Trust: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
2021-09-22 16:32:45 -06:00
MOVD R5, m_vdsoSP(R21)
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD m_curg(R21), R6
CMP g, R6
BNE noswitch
MOVD m_g0(R21), R7
MOVD (g_sched+gobuf_sp)(R7), R1 // Set SP to g0 stack
noswitch:
SUB $16, R1 // Space for results
RLDICR $0, R1, $59, R1 // Align for C code
MOVD R12, CTR
MOVD R1, R4
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
// Store g on gsignal's stack, so if we receive a signal
// during VDSO code we can find the g.
// If we don't have a signal stack, we won't receive signal,
// so don't bother saving g.
// When using cgo, we already saved g on TLS, also don't save
// g here.
// Also don't save g if we are already on the signal stack.
// We won't get a nested signal.
MOVBZ runtime·iscgo(SB), R22
CMP R22, $0
BNE nosaveg
MOVD m_gsignal(R21), R22 // g.m.gsignal
CMP R22, $0
BEQ nosaveg
CMP g, R22
BEQ nosaveg
MOVD (g_stack+stack_lo)(R22), R22 // g.m.gsignal.stack.lo
MOVD g, (R22)
BL (CTR) // Call from VDSO
MOVD $0, (R22) // clear g slot, R22 is unchanged by C code
JMP finish
nosaveg:
BL (CTR) // Call from VDSO
finish:
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
MOVD $0, R0 // Restore R0
MOVD 0(R1), R3 // sec
MOVD 8(R1), R5 // nsec
MOVD R15, R1 // Restore SP
runtime: make nanotime1 reentrant Currently, nanotime1 (and walltime1) is not reentrant, in that it sets m.vdsoSP at entry and clears it at exit. If a signal lands in between, and nanotime1 is called from the signal handler, it will clear m.vdsoSP while we are still in nanotime1. If (in the unlikely event) it is signaled again, m.vdsoSP will be wrong, which may cause the stack unwinding code to crash. This CL makes it reentrant, by saving/restoring the previous vdsoPC and vdsoSP, instead of setting it to 0 at exit. TODO: have some way to test? Change-Id: I9ee53b251f1d8a5a489c71d4b4c0df1dee70c3e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/246763 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2020-08-04 18:25:10 -06:00
// Restore vdsoPC, vdsoSP
// We don't worry about being signaled between the two stores.
// If we are not in a signal handler, we'll restore vdsoSP to 0,
// and no one will care about vdsoPC. If we are in a signal handler,
// we cannot receive another signal.
MOVD 40(R1), R6
MOVD R6, m_vdsoSP(R21)
MOVD 32(R1), R6
MOVD R6, m_vdsoPC(R21)
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
return:
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// sec is in R3, nsec in R5
// return nsec in R3
MOVD $1000000000, R4
MULLD R4, R3
ADD R5, R3
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD R3, ret+0(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
// Syscall fallback
fallback:
ADD $32, R1, R4
SYSCALL $SYS_clock_gettime
MOVD 32(R1), R3
runtime: fix wrong offset when calling ppc64x nanotime syscall There is a wrong offset when getting the results of a clock_gettime syscall. Although the syscall will never be called in native ppc64x, QEMU doesn't implement VDSO, so it will return wrong values. Fixes #36592 Change-Id: Icf838075228dcdd62cf2c1279aa983e5993d66ee Reviewed-on: https://go-review.googlesource.com/c/go/+/215397 Reviewed-by: Tobias Klauser <tobias.klauser@gmail.com>
2020-01-17 13:59:59 -07:00
MOVD 40(R1), R5
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
JMP return
runtime: add support for VDSO on ppc64x for use in walltime/nanotime This change adds support for VDSO on ppc64x, making it possible to avoid a syscall in walltime and nanotime. BenchmarkClockVDSOAndFallbackPaths/vDSO-192 20000000 66.0 ns/op BenchmarkClockVDSOAndFallbackPaths/Fallback-192 1000000 1456 ns/op Change-Id: I3373bd804b6f122961de3ae9d034e6ccf35748e6 Reviewed-on: https://go-review.googlesource.com/131135 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
2018-08-23 17:16:19 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·rtsigprocmask(SB),NOSPLIT|NOFRAME,$0-28
runtime: unify some signal handling functions Unify the OS-specific versions of msigsave, msigrestore, sigblock, updatesigmask, and unblocksig into single versions in signal_unix.go. To do this, make sigprocmask work the same way on all systems, which required adding a definition of sigprocmask for linux and openbsd. Also add a single OS-specific function sigmaskToSigset. Change-Id: I7cbf75131dddb57eeefe648ef845b0791404f785 Reviewed-on: https://go-review.googlesource.com/29689 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-23 18:54:51 -06:00
MOVW how+0(FP), R3
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD new+8(FP), R4
MOVD old+16(FP), R5
MOVW size+24(FP), R6
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_rt_sigprocmask
BVC 2(PC)
cmd/asm: detect invalid DS form offsets for ppc64x While debugging a recent regression it was discovered that the assembler for ppc64x was not always generating the correct instruction for DS form loads and stores. When an instruction is DS form then the offset must be a multiple of 4, and if it isn't then bits outside the offset field were being incorrectly set resulting in unexpected and incorrect instructions. This change adds a check to determine when the opcode is DS form and then verifies that the offset is a multiple of 4 before generating the instruction, otherwise logs an error. This also changes a few asm files that were using unaligned offsets for DS form loads and stores. In the runtime package these were instructions intended to cause a crash so using aligned or unaligned offsets doesn't change that behavior. Change-Id: Ie3a7e1e65dcc9933b54de7a46a054da8459cb56f Reviewed-on: https://go-review.googlesource.com/40476 Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
2017-04-12 12:22:16 -06:00
MOVD R0, 0xf0(R0) // crash
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·rt_sigaction(SB),NOSPLIT|NOFRAME,$0-36
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD sig+0(FP), R3
MOVD new+8(FP), R4
MOVD old+16(FP), R5
MOVD size+24(FP), R6
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_rt_sigaction
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+32(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime/cgo,cmd/internal/obj/ppc64: fix signals with cgo Recently some tsan tests were enabled on ppc64le which had not been enabled before. This resulted in failures on systems with tsan available, and while debugging it was determined that there were other issues related to the use of signals with cgo. Signals were not being forwarded within programs linked against libtsan because the nocgo sigaction was being called for ppc64le with or without cgo. Adding callCgoSigaction and calling that allows signals to be registered so that signal forwarding works. For linux-ppc64 and aix-ppc64, this won't change. On linux-ppc64 there is no cgo. I can't test aix-ppc64 so those owners can enable it if they want. In reviewing comments about sigtramp in sys_linux_arm64 it was noted that a previous issue in arm64 due to missing callee save registers could also be a problem on ppc64x, so code was added to save and restore those. Also, the use of R31 as a temp register in some cases caused an issue since it is a nonvolatile register in C and was being clobbered in cases where the C code expected it to be valid. The code sequences to load these addresses were changed to avoid the use of R31 when loading such an address. To get around a vet error, the stubs_ppc64x.go file in runtime was split into stubs_ppc64.go and stubs_ppc64le.go. Updates #45040 Change-Id: Ia4ecff950613cbe1b89471790b1d3819d5b5cfb9 Reviewed-on: https://go-review.googlesource.com/c/go/+/306369 Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
2021-03-31 11:28:47 -06:00
#ifdef GOARCH_ppc64le
// Call the function stored in _cgo_sigaction using the GCC calling convention.
TEXT runtime·callCgoSigaction(SB),NOSPLIT,$0
MOVD sig+0(FP), R3
MOVD new+8(FP), R4
MOVD old+16(FP), R5
MOVD _cgo_sigaction(SB), R12
MOVD R12, CTR // R12 should contain the function address
MOVD R1, R15 // Save R1
MOVD R2, 24(R1) // Save R2
SUB $48, R1 // reserve 32 (frame) + 16 bytes for sp-8 where fp may be saved.
RLDICR $0, R1, $59, R1 // Align to 16 bytes for C code
BL (CTR)
XOR R0, R0, R0 // Clear R0 as Go expects
MOVD R15, R1 // Restore R1
MOVD 24(R1), R2 // Restore R2
MOVW R3, ret+24(FP) // Return result
RET
#endif
runtime: signal forwarding Forward signals to signal handlers installed before Go installs its own, under certain circumstances. In particular, as iant@ suggests, signals are forwarded iff: (1) a non-SIG_DFL signal handler existed before Go, and (2) signal is synchronous (i.e., one of SIGSEGV, SIGBUS, SIGFPE), and (3a) signal occured on a non-Go thread, or (3b) signal occurred on a Go thread but in CGo code. Supported only on Linux, for now. Change-Id: I403219ee47b26cf65da819fb86cf1ec04d3e25f5 Reviewed-on: https://go-review.googlesource.com/8712 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-04-09 12:12:12 -06:00
TEXT runtime·sigfwd(SB),NOSPLIT,$0-32
MOVW sig+8(FP), R3
MOVD info+16(FP), R4
MOVD ctx+24(FP), R5
runtime, runtime/cgo: conform to PIC register use rules in ppc64 asm PIC code on ppc64le uses R2 as a TOC pointer and when calling a function through a function pointer must ensure the function pointer is in R12. These rules are easy enough to follow unconditionally in our assembly, so do that. Change-Id: Icfc4e47ae5dfbe15f581cbdd785cdeed6e40bc32 Reviewed-on: https://go-review.googlesource.com/15526 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-10-08 03:49:39 -06:00
MOVD fn+0(FP), R12
MOVD R12, CTR
runtime: signal forwarding Forward signals to signal handlers installed before Go installs its own, under certain circumstances. In particular, as iant@ suggests, signals are forwarded iff: (1) a non-SIG_DFL signal handler existed before Go, and (2) signal is synchronous (i.e., one of SIGSEGV, SIGBUS, SIGFPE), and (3a) signal occured on a non-Go thread, or (3b) signal occurred on a Go thread but in CGo code. Supported only on Linux, for now. Change-Id: I403219ee47b26cf65da819fb86cf1ec04d3e25f5 Reviewed-on: https://go-review.googlesource.com/8712 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-04-09 12:12:12 -06:00
BL (CTR)
cmd/compile, cmd/link, runtime: on ppc64x, maintain the TOC pointer in R2 when compiling PIC The PowerPC ISA does not have a PC-relative load instruction, which poses obvious challenges when generating position-independent code. The way the ELFv2 ABI addresses this is to specify that r2 points to a per "module" (shared library or executable) TOC pointer. Maintaining this pointer requires cooperation between codegen and the system linker: * Non-leaf functions leave space on the stack at r1+24 to save the TOC pointer. * A call to a function that *might* have to go via a PLT stub must be followed by a nop instruction that the system linker can replace with "ld r1, 24(r1)" to restore the TOC pointer (only when dynamically linking Go code). * When calling a function via a function pointer, the address of the function must be in r12, and the first couple of instructions (the "global entry point") of the called function use this to derive the address of the TOC for the module it is in. * When calling a function that is implemented in the same module, the system linker adjusts the call to skip over the instructions mentioned above (the "local entry point"), assuming that r2 is already correctly set. So this changeset adds the global entry point instructions, sets the metadata so the system linker knows where the local entry point is, inserts code to save the TOC pointer at 24(r1), adds a nop after any call not known to be local and copes with the odd non-local code transfer in the runtime (e.g. the stuff around jmpdefer). It does not actually compile PIC yet. Change-Id: I7522e22bdfd2f891745a900c60254fe9e372c854 Reviewed-on: https://go-review.googlesource.com/15967 Reviewed-by: Russ Cox <rsc@golang.org>
2015-10-15 20:42:09 -06:00
MOVD 24(R1), R2
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
runtime: signal forwarding Forward signals to signal handlers installed before Go installs its own, under certain circumstances. In particular, as iant@ suggests, signals are forwarded iff: (1) a non-SIG_DFL signal handler existed before Go, and (2) signal is synchronous (i.e., one of SIGSEGV, SIGBUS, SIGFPE), and (3a) signal occured on a non-Go thread, or (3b) signal occurred on a Go thread but in CGo code. Supported only on Linux, for now. Change-Id: I403219ee47b26cf65da819fb86cf1ec04d3e25f5 Reviewed-on: https://go-review.googlesource.com/8712 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-04-09 12:12:12 -06:00
runtime: fix vet complaints for linux/arm64, linux/mips*, linux/ppc64*, linux/s390x Working toward making the tree vet-safe instead of having so many exceptions in cmd/vet/all/whitelist. This CL makes "go vet -unsafeptr=false runtime" happy for these GOOS/GOARCHes, except for an unresolved complaint on mips/mipsle that is a bug in vet, while keeping "GO_BUILDER_NAME=misc-vetall go tool dist test" happy too. For #31916. Change-Id: I6ef7e982a2fdbbfbc22cee876ca37ac54d8109e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/176102 Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2019-05-08 11:59:48 -06:00
TEXT runtime·sigreturn(SB),NOSPLIT,$0-0
RET
all: power64 is now ppc64 Fixes #8654. LGTM=austin R=austin CC=golang-codereviews https://golang.org/cl/180600043
2014-12-05 17:13:20 -07:00
#ifdef GOARCH_ppc64le
// ppc64le doesn't need function descriptors
runtime/cgo,cmd/internal/obj/ppc64: fix signals with cgo Recently some tsan tests were enabled on ppc64le which had not been enabled before. This resulted in failures on systems with tsan available, and while debugging it was determined that there were other issues related to the use of signals with cgo. Signals were not being forwarded within programs linked against libtsan because the nocgo sigaction was being called for ppc64le with or without cgo. Adding callCgoSigaction and calling that allows signals to be registered so that signal forwarding works. For linux-ppc64 and aix-ppc64, this won't change. On linux-ppc64 there is no cgo. I can't test aix-ppc64 so those owners can enable it if they want. In reviewing comments about sigtramp in sys_linux_arm64 it was noted that a previous issue in arm64 due to missing callee save registers could also be a problem on ppc64x, so code was added to save and restore those. Also, the use of R31 as a temp register in some cases caused an issue since it is a nonvolatile register in C and was being clobbered in cases where the C code expected it to be valid. The code sequences to load these addresses were changed to avoid the use of R31 when loading such an address. To get around a vet error, the stubs_ppc64x.go file in runtime was split into stubs_ppc64.go and stubs_ppc64le.go. Updates #45040 Change-Id: Ia4ecff950613cbe1b89471790b1d3819d5b5cfb9 Reviewed-on: https://go-review.googlesource.com/c/go/+/306369 Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
2021-03-31 11:28:47 -06:00
// Save callee-save registers in the case of signal forwarding.
// Same as on ARM64 https://golang.org/issue/31827 .
TEXT runtime·sigtramp(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#else
// function descriptor for the real sigtramp
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·sigtramp(SB),NOSPLIT|NOFRAME,$0
runtime: fix vet complaints for linux/arm64, linux/mips*, linux/ppc64*, linux/s390x Working toward making the tree vet-safe instead of having so many exceptions in cmd/vet/all/whitelist. This CL makes "go vet -unsafeptr=false runtime" happy for these GOOS/GOARCHes, except for an unresolved complaint on mips/mipsle that is a bug in vet, while keeping "GO_BUILDER_NAME=misc-vetall go tool dist test" happy too. For #31916. Change-Id: I6ef7e982a2fdbbfbc22cee876ca37ac54d8109e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/176102 Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2019-05-08 11:59:48 -06:00
DWORD $sigtramp<>(SB)
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
DWORD $0
DWORD $0
runtime: mark sigtramp as TOPFRAME Currently throw() in the signal handler results in "fatal error: unknown return pc from runtime.sigreturn ...". Marking sigtramp as TOPFRAME allows gentraceback to stop tracebacks at the end of a signal handler, since there is not much beyond sigtramp. This is just done on Linux for now, but may apply to other Unix systems as well. Change-Id: I96edcb945283f417a5bfe00ce2fb2b1a0d578692 Reviewed-on: https://go-review.googlesource.com/c/go/+/402190 Reviewed-by: Austin Clements <austin@google.com> Run-TryBot: Michael Pratt <mpratt@google.com> Auto-Submit: Michael Pratt <mpratt@google.com> TryBot-Result: Gopher Robot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
2022-04-26 12:42:34 -06:00
TEXT sigtramp<>(SB),NOSPLIT|NOFRAME|TOPFRAME,$0
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
#endif
runtime/cgo,cmd/internal/obj/ppc64: fix signals with cgo Recently some tsan tests were enabled on ppc64le which had not been enabled before. This resulted in failures on systems with tsan available, and while debugging it was determined that there were other issues related to the use of signals with cgo. Signals were not being forwarded within programs linked against libtsan because the nocgo sigaction was being called for ppc64le with or without cgo. Adding callCgoSigaction and calling that allows signals to be registered so that signal forwarding works. For linux-ppc64 and aix-ppc64, this won't change. On linux-ppc64 there is no cgo. I can't test aix-ppc64 so those owners can enable it if they want. In reviewing comments about sigtramp in sys_linux_arm64 it was noted that a previous issue in arm64 due to missing callee save registers could also be a problem on ppc64x, so code was added to save and restore those. Also, the use of R31 as a temp register in some cases caused an issue since it is a nonvolatile register in C and was being clobbered in cases where the C code expected it to be valid. The code sequences to load these addresses were changed to avoid the use of R31 when loading such an address. To get around a vet error, the stubs_ppc64x.go file in runtime was split into stubs_ppc64.go and stubs_ppc64le.go. Updates #45040 Change-Id: Ia4ecff950613cbe1b89471790b1d3819d5b5cfb9 Reviewed-on: https://go-review.googlesource.com/c/go/+/306369 Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
2021-03-31 11:28:47 -06:00
// Start with standard C stack frame layout and linkage.
MOVD LR, R0
MOVD R0, 16(R1) // Save LR in caller's frame.
MOVW CR, R0 // Save CR in caller's frame
MOVD R0, 8(R1)
// The stack must be acquired here and not
// in the automatic way based on stack size
// since that sequence clobbers R31 before it
// gets saved.
// We are being ultra safe here in saving the
// Vregs. The case where they might need to
// be saved is very unlikely.
MOVDU R1, -544(R1)
MOVD R14, 64(R1)
MOVD R15, 72(R1)
MOVD R16, 80(R1)
MOVD R17, 88(R1)
MOVD R18, 96(R1)
MOVD R19, 104(R1)
MOVD R20, 112(R1)
MOVD R21, 120(R1)
MOVD R22, 128(R1)
MOVD R23, 136(R1)
MOVD R24, 144(R1)
MOVD R25, 152(R1)
MOVD R26, 160(R1)
MOVD R27, 168(R1)
MOVD R28, 176(R1)
MOVD R29, 184(R1)
MOVD g, 192(R1) // R30
MOVD R31, 200(R1)
FMOVD F14, 208(R1)
FMOVD F15, 216(R1)
FMOVD F16, 224(R1)
FMOVD F17, 232(R1)
FMOVD F18, 240(R1)
FMOVD F19, 248(R1)
FMOVD F20, 256(R1)
FMOVD F21, 264(R1)
FMOVD F22, 272(R1)
FMOVD F23, 280(R1)
FMOVD F24, 288(R1)
FMOVD F25, 296(R1)
FMOVD F26, 304(R1)
FMOVD F27, 312(R1)
FMOVD F28, 320(R1)
FMOVD F29, 328(R1)
FMOVD F30, 336(R1)
FMOVD F31, 344(R1)
// Save V regs
// STXVD2X and LXVD2X used since
// we aren't sure of alignment.
// Endianness doesn't matter
// if we are just loading and
// storing values.
MOVD $352, R7 // V20
STXVD2X VS52, (R7)(R1)
ADD $16, R7 // V21 368
STXVD2X VS53, (R7)(R1)
ADD $16, R7 // V22 384
STXVD2X VS54, (R7)(R1)
ADD $16, R7 // V23 400
STXVD2X VS55, (R7)(R1)
ADD $16, R7 // V24 416
STXVD2X VS56, (R7)(R1)
ADD $16, R7 // V25 432
STXVD2X VS57, (R7)(R1)
ADD $16, R7 // V26 448
STXVD2X VS58, (R7)(R1)
ADD $16, R7 // V27 464
STXVD2X VS59, (R7)(R1)
ADD $16, R7 // V28 480
STXVD2X VS60, (R7)(R1)
ADD $16, R7 // V29 496
STXVD2X VS61, (R7)(R1)
ADD $16, R7 // V30 512
STXVD2X VS62, (R7)(R1)
ADD $16, R7 // V31 528
STXVD2X VS63, (R7)(R1)
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// initialize essential registers (just in case)
BL runtime·reginit(SB)
cmd/cgo, runtime/cgo: support ppc64 This implements support for calls to and from C in the ppc64 C ABI, as well as supporting functionality such as an entry point from the dynamic linker. Change-Id: I68da6df50d5638cb1a3d3fef773fb412d7bf631a Reviewed-on: https://go-review.googlesource.com/2009 Reviewed-by: Russ Cox <rsc@golang.org>
2014-12-16 16:34:55 -07:00
// this might be called in external code context,
// where g is not set.
runtime: use unsigned load for iscgo variable in ppc64x runtime asm This changes the runtime asm code that loads iscgo to use MOVBZ instead of MOVB, avoiding an unnecessary sign extension. This is most significant in runtime.save_g, reducing the size from 8 to 7 instructions. Change-Id: Iaa2121464b5309e1f27fd91b19b5603c7aaf619d Reviewed-on: https://go-review.googlesource.com/c/144217 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Chase <drchase@google.com>
2018-10-22 14:02:43 -06:00
MOVBZ runtime·iscgo(SB), R6
runtime: fix crash during VDSO calls on PowerPC This patch reinstates a fix for PowerPC with regard to making VDSO calls while receiving a signal, and subsequently crashing. The crash happens because certain VDSO calls can modify the r30 register, which is where g is stored. This change was reverted for PowerPC because r30 is supposed to be a non-volatile register. This is true, but that only makes a guarantee across function calls, but not "within" a function call. This patch was seemingly fine before because the Linux kernel still had hand rolled assembly VDSO function calls, however with a recent change to C function calls it seems the compiler used can generate instructions which temporarily clobber r30. This means that when we receive a signal during one of these calls the value of r30 will not be the g as the runtime expects, causing a segfault. You can see from this assembly dump how the register is clobbered during the call: (the following is from a 5.13rc2 kernel) ``` Dump of assembler code for function __cvdso_clock_gettime_data: 0x00007ffff7ff0700 <+0>: cmplwi r4,15 0x00007ffff7ff0704 <+4>: bgt 0x7ffff7ff07f0 <__cvdso_clock_gettime_data+240> 0x00007ffff7ff0708 <+8>: li r9,1 0x00007ffff7ff070c <+12>: slw r9,r9,r4 0x00007ffff7ff0710 <+16>: andi. r10,r9,2179 0x00007ffff7ff0714 <+20>: beq 0x7ffff7ff0810 <__cvdso_clock_gettime_data+272> 0x00007ffff7ff0718 <+24>: rldicr r10,r4,4,59 0x00007ffff7ff071c <+28>: lis r9,32767 0x00007ffff7ff0720 <+32>: std r30,-16(r1) 0x00007ffff7ff0724 <+36>: std r31,-8(r1) 0x00007ffff7ff0728 <+40>: add r6,r3,r10 0x00007ffff7ff072c <+44>: ori r4,r9,65535 0x00007ffff7ff0730 <+48>: lwz r8,0(r3) 0x00007ffff7ff0734 <+52>: andi. r9,r8,1 0x00007ffff7ff0738 <+56>: bne 0x7ffff7ff07d0 <__cvdso_clock_gettime_data+208> 0x00007ffff7ff073c <+60>: lwsync 0x00007ffff7ff0740 <+64>: mftb r30 <---- RIGHT HERE => 0x00007ffff7ff0744 <+68>: ld r12,40(r6) ``` What I believe is happening is that the kernel changed the PowerPC VDSO calls to use standard C calls instead of using hand rolled assembly. The hand rolled assembly calls never touched r30, so this change was safe to roll back. That does not seem to be the case anymore as on the 5.13rc2 kernel the compiler *is* generating assembly which modifies r30, making this change again unsafe and causing a crash when the program receives a signal during these calls (which will happen often due to async preempt). This change happened here: https://lwn.net/ml/linux-kernel/235e5571959cfa89ced081d7e838ed5ff38447d2.1601365870.git.christophe.leroy@csgroup.eu/. I realize this was reverted due to unexplained hangs in PowerPC builders, but I think we should reinstate this change and investigate those issues separately: https://github.com/golang/go/commit/f4ca3c1e0a2066ca4f7bd6203866d282ed34acf2 Fixes #46803 Change-Id: Ib18d7bbfc80a1a9cb558f0098878d41081324b52 GitHub-Last-Rev: c3002bcfca3ef58b27485e31328e6297b7a9dfe7 GitHub-Pull-Request: golang/go#46767 Reviewed-on: https://go-review.googlesource.com/c/go/+/328110 Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Trust: Lynn Boger <laboger@linux.vnet.ibm.com>
2021-06-17 14:22:40 -06:00
CMP R6, $0
cmd/cgo, runtime/cgo: support ppc64 This implements support for calls to and from C in the ppc64 C ABI, as well as supporting functionality such as an entry point from the dynamic linker. Change-Id: I68da6df50d5638cb1a3d3fef773fb412d7bf631a Reviewed-on: https://go-review.googlesource.com/2009 Reviewed-by: Russ Cox <rsc@golang.org>
2014-12-16 16:34:55 -07:00
BEQ 2(PC)
BL runtime·load_g(SB)
reflect, runtime, runtime/cgo: use ppc64 asm constant for fixed frame size Shared libraries on ppc64le will require a larger minimum stack frame (because the ABI mandates that the TOC pointer is available at 24(R1)). Part 3 of that is using a #define in the ppc64 assembly to refer to the size of the fixed part of the stack (finding all these took me about a week!). Change-Id: I50f22fe1c47af1ec59da1bd7ea8f84a4750df9b7 Reviewed-on: https://go-review.googlesource.com/15525 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-10-08 03:34:29 -06:00
MOVW R3, FIXED_FRAME+0(R1)
MOVD R4, FIXED_FRAME+8(R1)
MOVD R5, FIXED_FRAME+16(R1)
runtime, runtime/cgo: conform to PIC register use rules in ppc64 asm PIC code on ppc64le uses R2 as a TOC pointer and when calling a function through a function pointer must ensure the function pointer is in R12. These rules are easy enough to follow unconditionally in our assembly, so do that. Change-Id: Icfc4e47ae5dfbe15f581cbdd785cdeed6e40bc32 Reviewed-on: https://go-review.googlesource.com/15526 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-10-08 03:49:39 -06:00
MOVD $runtime·sigtrampgo(SB), R12
MOVD R12, CTR
runtime: signal forwarding Forward signals to signal handlers installed before Go installs its own, under certain circumstances. In particular, as iant@ suggests, signals are forwarded iff: (1) a non-SIG_DFL signal handler existed before Go, and (2) signal is synchronous (i.e., one of SIGSEGV, SIGBUS, SIGFPE), and (3a) signal occured on a non-Go thread, or (3b) signal occurred on a Go thread but in CGo code. Supported only on Linux, for now. Change-Id: I403219ee47b26cf65da819fb86cf1ec04d3e25f5 Reviewed-on: https://go-review.googlesource.com/8712 Reviewed-by: Ian Lance Taylor <iant@golang.org>
2015-04-09 12:12:12 -06:00
BL (CTR)
runtime/cgo,cmd/internal/obj/ppc64: fix signals with cgo Recently some tsan tests were enabled on ppc64le which had not been enabled before. This resulted in failures on systems with tsan available, and while debugging it was determined that there were other issues related to the use of signals with cgo. Signals were not being forwarded within programs linked against libtsan because the nocgo sigaction was being called for ppc64le with or without cgo. Adding callCgoSigaction and calling that allows signals to be registered so that signal forwarding works. For linux-ppc64 and aix-ppc64, this won't change. On linux-ppc64 there is no cgo. I can't test aix-ppc64 so those owners can enable it if they want. In reviewing comments about sigtramp in sys_linux_arm64 it was noted that a previous issue in arm64 due to missing callee save registers could also be a problem on ppc64x, so code was added to save and restore those. Also, the use of R31 as a temp register in some cases caused an issue since it is a nonvolatile register in C and was being clobbered in cases where the C code expected it to be valid. The code sequences to load these addresses were changed to avoid the use of R31 when loading such an address. To get around a vet error, the stubs_ppc64x.go file in runtime was split into stubs_ppc64.go and stubs_ppc64le.go. Updates #45040 Change-Id: Ia4ecff950613cbe1b89471790b1d3819d5b5cfb9 Reviewed-on: https://go-review.googlesource.com/c/go/+/306369 Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
2021-03-31 11:28:47 -06:00
MOVD 24(R1), R2 // Should this be here? Where is it saved?
// Starts at 64; FIXED_FRAME is 32
MOVD 64(R1), R14
MOVD 72(R1), R15
MOVD 80(R1), R16
MOVD 88(R1), R17
MOVD 96(R1), R18
MOVD 104(R1), R19
MOVD 112(R1), R20
MOVD 120(R1), R21
MOVD 128(R1), R22
MOVD 136(R1), R23
MOVD 144(R1), R24
MOVD 152(R1), R25
MOVD 160(R1), R26
MOVD 168(R1), R27
MOVD 176(R1), R28
MOVD 184(R1), R29
MOVD 192(R1), g // R30
MOVD 200(R1), R31
FMOVD 208(R1), F14
FMOVD 216(R1), F15
FMOVD 224(R1), F16
FMOVD 232(R1), F17
FMOVD 240(R1), F18
FMOVD 248(R1), F19
FMOVD 256(R1), F20
FMOVD 264(R1), F21
FMOVD 272(R1), F22
FMOVD 280(R1), F23
FMOVD 288(R1), F24
FMOVD 292(R1), F25
FMOVD 300(R1), F26
FMOVD 308(R1), F27
FMOVD 316(R1), F28
FMOVD 328(R1), F29
FMOVD 336(R1), F30
FMOVD 344(R1), F31
MOVD $352, R7
LXVD2X (R7)(R1), VS52
ADD $16, R7 // 368 V21
LXVD2X (R7)(R1), VS53
ADD $16, R7 // 384 V22
LXVD2X (R7)(R1), VS54
ADD $16, R7 // 400 V23
LXVD2X (R7)(R1), VS55
ADD $16, R7 // 416 V24
LXVD2X (R7)(R1), VS56
ADD $16, R7 // 432 V25
LXVD2X (R7)(R1), VS57
ADD $16, R7 // 448 V26
LXVD2X (R7)(R1), VS58
ADD $16, R8 // 464 V27
LXVD2X (R7)(R1), VS59
ADD $16, R7 // 480 V28
LXVD2X (R7)(R1), VS60
ADD $16, R7 // 496 V29
LXVD2X (R7)(R1), VS61
ADD $16, R7 // 512 V30
LXVD2X (R7)(R1), VS62
ADD $16, R7 // 528 V31
LXVD2X (R7)(R1), VS63
ADD $544, R1
MOVD 8(R1), R0
MOVFL R0, $0xff
MOVD 16(R1), R0
MOVD R0, LR
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
runtime: support symbolic backtrace of C code in a cgo crash The new function runtime.SetCgoTraceback may be used to register stack traceback and symbolizer functions, written in C, to do a stack traceback from cgo code. There is a sample implementation of runtime.SetCgoSymbolizer at github.com/ianlancetaylor/cgosymbolizer. Just importing that package is sufficient to get symbolic C backtraces. Currently only supported on linux/amd64. Change-Id: If96ee2eb41c6c7379d407b9561b87557bfe47341 Reviewed-on: https://go-review.googlesource.com/17761 Reviewed-by: Austin Clements <austin@google.com>
2015-12-11 18:16:48 -07:00
#ifdef GOARCH_ppc64le
// ppc64le doesn't need function descriptors
runtime: support cgo traceback on PPC64LE Code essentially mirrors AMD64 implementation. Change-Id: I39f7f099ce11fdc3772df039998cc11947bb22a2 Reviewed-on: https://go-review.googlesource.com/72270 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-20 12:26:28 -06:00
TEXT runtime·cgoSigtramp(SB),NOSPLIT|NOFRAME,$0
// The stack unwinder, presumably written in C, may not be able to
// handle Go frame correctly. So, this function is NOFRAME, and we
all: fix a bunch of misspellings Change-Id: I94cebca86706e072fbe3be782d3edbe0e22b9432 GitHub-Last-Rev: 8e15a40545704fb21b41a8768079f2da19341ef3 GitHub-Pull-Request: golang/go#28067 Reviewed-on: https://go-review.googlesource.com/c/140437 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-10-07 19:19:51 -06:00
// save/restore LR manually.
runtime: support cgo traceback on PPC64LE Code essentially mirrors AMD64 implementation. Change-Id: I39f7f099ce11fdc3772df039998cc11947bb22a2 Reviewed-on: https://go-review.googlesource.com/72270 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-20 12:26:28 -06:00
MOVD LR, R10
// We're coming from C code, initialize essential registers.
CALL runtime·reginit(SB)
// If no traceback function, do usual sigtramp.
MOVD runtime·cgoTraceback(SB), R6
CMP $0, R6
BEQ sigtramp
// If no traceback support function, which means that
// runtime/cgo was not linked in, do usual sigtramp.
MOVD _cgo_callers(SB), R6
CMP $0, R6
BEQ sigtramp
// Set up g register.
CALL runtime·load_g(SB)
// Figure out if we are currently in a cgo call.
// If not, just do usual sigtramp.
runtime/cgo,cmd/internal/obj/ppc64: fix signals with cgo Recently some tsan tests were enabled on ppc64le which had not been enabled before. This resulted in failures on systems with tsan available, and while debugging it was determined that there were other issues related to the use of signals with cgo. Signals were not being forwarded within programs linked against libtsan because the nocgo sigaction was being called for ppc64le with or without cgo. Adding callCgoSigaction and calling that allows signals to be registered so that signal forwarding works. For linux-ppc64 and aix-ppc64, this won't change. On linux-ppc64 there is no cgo. I can't test aix-ppc64 so those owners can enable it if they want. In reviewing comments about sigtramp in sys_linux_arm64 it was noted that a previous issue in arm64 due to missing callee save registers could also be a problem on ppc64x, so code was added to save and restore those. Also, the use of R31 as a temp register in some cases caused an issue since it is a nonvolatile register in C and was being clobbered in cases where the C code expected it to be valid. The code sequences to load these addresses were changed to avoid the use of R31 when loading such an address. To get around a vet error, the stubs_ppc64x.go file in runtime was split into stubs_ppc64.go and stubs_ppc64le.go. Updates #45040 Change-Id: Ia4ecff950613cbe1b89471790b1d3819d5b5cfb9 Reviewed-on: https://go-review.googlesource.com/c/go/+/306369 Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <carlos.seo@linaro.org>
2021-03-31 11:28:47 -06:00
// compared to ARM64 and others.
runtime: support cgo traceback on PPC64LE Code essentially mirrors AMD64 implementation. Change-Id: I39f7f099ce11fdc3772df039998cc11947bb22a2 Reviewed-on: https://go-review.googlesource.com/72270 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-20 12:26:28 -06:00
CMP $0, g
BEQ sigtrampnog // g == nil
MOVD g_m(g), R6
CMP $0, R6
BEQ sigtramp // g.m == nil
MOVW m_ncgo(R6), R7
CMPW $0, R7
BEQ sigtramp // g.m.ncgo = 0
MOVD m_curg(R6), R7
CMP $0, R7
BEQ sigtramp // g.m.curg == nil
MOVD g_syscallsp(R7), R7
CMP $0, R7
BEQ sigtramp // g.m.curg.syscallsp == 0
MOVD m_cgoCallers(R6), R7 // R7 is the fifth arg in C calling convention.
CMP $0, R7
BEQ sigtramp // g.m.cgoCallers == nil
MOVW m_cgoCallersUse(R6), R8
CMPW $0, R8
BNE sigtramp // g.m.cgoCallersUse != 0
// Jump to a function in runtime/cgo.
// That function, written in C, will call the user's traceback
// function with proper unwind info, and will then call back here.
// The first three arguments, and the fifth, are already in registers.
// Set the two remaining arguments now.
MOVD runtime·cgoTraceback(SB), R6
MOVD $runtime·sigtramp(SB), R8
MOVD _cgo_callers(SB), R12
MOVD R12, CTR
MOVD R10, LR // restore LR
JMP (CTR)
sigtramp:
MOVD R10, LR // restore LR
JMP runtime·sigtramp(SB)
sigtrampnog:
// Signal arrived on a non-Go thread. If this is SIGPROF, get a
// stack trace.
CMPW R3, $27 // 27 == SIGPROF
BNE sigtramp
// Lock sigprofCallersUse (cas from 0 to 1).
MOVW $1, R7
MOVD $runtime·sigprofCallersUse(SB), R8
SYNC
LWAR (R8), R6
CMPW $0, R6
BNE sigtramp
STWCCC R7, (R8)
BNE -4(PC)
ISYNC
// Jump to the traceback function in runtime/cgo.
// It will call back to sigprofNonGo, which will ignore the
// arguments passed in registers.
// First three arguments to traceback function are in registers already.
MOVD runtime·cgoTraceback(SB), R6
MOVD $runtime·sigprofCallers(SB), R7
MOVD $runtime·sigprofNonGoWrapper<>(SB), R8
MOVD _cgo_callers(SB), R12
MOVD R12, CTR
MOVD R10, LR // restore LR
JMP (CTR)
runtime: support symbolic backtrace of C code in a cgo crash The new function runtime.SetCgoTraceback may be used to register stack traceback and symbolizer functions, written in C, to do a stack traceback from cgo code. There is a sample implementation of runtime.SetCgoSymbolizer at github.com/ianlancetaylor/cgosymbolizer. Just importing that package is sufficient to get symbolic C backtraces. Currently only supported on linux/amd64. Change-Id: If96ee2eb41c6c7379d407b9561b87557bfe47341 Reviewed-on: https://go-review.googlesource.com/17761 Reviewed-by: Austin Clements <austin@google.com>
2015-12-11 18:16:48 -07:00
#else
// function descriptor for the real sigtramp
TEXT runtime·cgoSigtramp(SB),NOSPLIT|NOFRAME,$0
runtime: fix vet complaints for linux/arm64, linux/mips*, linux/ppc64*, linux/s390x Working toward making the tree vet-safe instead of having so many exceptions in cmd/vet/all/whitelist. This CL makes "go vet -unsafeptr=false runtime" happy for these GOOS/GOARCHes, except for an unresolved complaint on mips/mipsle that is a bug in vet, while keeping "GO_BUILDER_NAME=misc-vetall go tool dist test" happy too. For #31916. Change-Id: I6ef7e982a2fdbbfbc22cee876ca37ac54d8109e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/176102 Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2019-05-08 11:59:48 -06:00
DWORD $cgoSigtramp<>(SB)
runtime: support symbolic backtrace of C code in a cgo crash The new function runtime.SetCgoTraceback may be used to register stack traceback and symbolizer functions, written in C, to do a stack traceback from cgo code. There is a sample implementation of runtime.SetCgoSymbolizer at github.com/ianlancetaylor/cgosymbolizer. Just importing that package is sufficient to get symbolic C backtraces. Currently only supported on linux/amd64. Change-Id: If96ee2eb41c6c7379d407b9561b87557bfe47341 Reviewed-on: https://go-review.googlesource.com/17761 Reviewed-by: Austin Clements <austin@google.com>
2015-12-11 18:16:48 -07:00
DWORD $0
DWORD $0
runtime: fix vet complaints for linux/arm64, linux/mips*, linux/ppc64*, linux/s390x Working toward making the tree vet-safe instead of having so many exceptions in cmd/vet/all/whitelist. This CL makes "go vet -unsafeptr=false runtime" happy for these GOOS/GOARCHes, except for an unresolved complaint on mips/mipsle that is a bug in vet, while keeping "GO_BUILDER_NAME=misc-vetall go tool dist test" happy too. For #31916. Change-Id: I6ef7e982a2fdbbfbc22cee876ca37ac54d8109e5 Reviewed-on: https://go-review.googlesource.com/c/go/+/176102 Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2019-05-08 11:59:48 -06:00
TEXT cgoSigtramp<>(SB),NOSPLIT,$0
JMP sigtramp<>(SB)
runtime: support symbolic backtrace of C code in a cgo crash The new function runtime.SetCgoTraceback may be used to register stack traceback and symbolizer functions, written in C, to do a stack traceback from cgo code. There is a sample implementation of runtime.SetCgoSymbolizer at github.com/ianlancetaylor/cgosymbolizer. Just importing that package is sufficient to get symbolic C backtraces. Currently only supported on linux/amd64. Change-Id: If96ee2eb41c6c7379d407b9561b87557bfe47341 Reviewed-on: https://go-review.googlesource.com/17761 Reviewed-by: Austin Clements <austin@google.com>
2015-12-11 18:16:48 -07:00
#endif
runtime: support cgo traceback on PPC64LE Code essentially mirrors AMD64 implementation. Change-Id: I39f7f099ce11fdc3772df039998cc11947bb22a2 Reviewed-on: https://go-review.googlesource.com/72270 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-20 12:26:28 -06:00
TEXT runtime·sigprofNonGoWrapper<>(SB),NOSPLIT,$0
// We're coming from C code, set up essential register, then call sigprofNonGo.
CALL runtime·reginit(SB)
runtime: use per-thread profiler for SetCgoTraceback platforms Updates #35057 Change-Id: I61d772a2cbfb27540fb70c14676c68593076ca94 Reviewed-on: https://go-review.googlesource.com/c/go/+/342054 Run-TryBot: Rhys Hiltner <rhys@justin.tv> TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com> Trust: Michael Knyszek <mknyszek@google.com>
2021-08-13 10:01:13 -06:00
MOVW R3, FIXED_FRAME+0(R1) // sig
MOVD R4, FIXED_FRAME+8(R1) // info
MOVD R5, FIXED_FRAME+16(R1) // ctx
runtime: support cgo traceback on PPC64LE Code essentially mirrors AMD64 implementation. Change-Id: I39f7f099ce11fdc3772df039998cc11947bb22a2 Reviewed-on: https://go-review.googlesource.com/72270 Run-TryBot: Cherry Zhang <cherryyz@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-20 12:26:28 -06:00
CALL runtime·sigprofNonGo(SB)
RET
runtime: support symbolic backtrace of C code in a cgo crash The new function runtime.SetCgoTraceback may be used to register stack traceback and symbolizer functions, written in C, to do a stack traceback from cgo code. There is a sample implementation of runtime.SetCgoSymbolizer at github.com/ianlancetaylor/cgosymbolizer. Just importing that package is sufficient to get symbolic C backtraces. Currently only supported on linux/amd64. Change-Id: If96ee2eb41c6c7379d407b9561b87557bfe47341 Reviewed-on: https://go-review.googlesource.com/17761 Reviewed-by: Austin Clements <austin@google.com>
2015-12-11 18:16:48 -07:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·mmap(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD addr+0(FP), R3
MOVD n+8(FP), R4
MOVW prot+16(FP), R5
MOVW flags+20(FP), R6
MOVW fd+24(FP), R7
MOVW off+28(FP), R8
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_mmap
runtime: separate error result for mmap Currently mmap returns an unsafe.Pointer that encodes OS errors as values less than 4096. In practice this is okay, but it borders on being really unsafe: for example, the value has to be checked immediately after return and if stack copying were ever to observe such a value, it would panic. It's also not remotely idiomatic. Fix this by making mmap return a separate pointer value and error, like a normal Go function. Updates #22218. Change-Id: Iefd965095ffc82cc91118872753a5d39d785c3a6 Reviewed-on: https://go-review.googlesource.com/71270 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-10-16 18:28:29 -06:00
BVC ok
MOVD $0, p+32(FP)
MOVD R3, err+40(FP)
RET
ok:
MOVD R3, p+32(FP)
MOVD $0, err+40(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·munmap(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD addr+0(FP), R3
MOVD n+8(FP), R4
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_munmap
BVC 2(PC)
cmd/asm: detect invalid DS form offsets for ppc64x While debugging a recent regression it was discovered that the assembler for ppc64x was not always generating the correct instruction for DS form loads and stores. When an instruction is DS form then the offset must be a multiple of 4, and if it isn't then bits outside the offset field were being incorrectly set resulting in unexpected and incorrect instructions. This change adds a check to determine when the opcode is DS form and then verifies that the offset is a multiple of 4 before generating the instruction, otherwise logs an error. This also changes a few asm files that were using unaligned offsets for DS form loads and stores. In the runtime package these were instructions intended to cause a crash so using aligned or unaligned offsets doesn't change that behavior. Change-Id: Ie3a7e1e65dcc9933b54de7a46a054da8459cb56f Reviewed-on: https://go-review.googlesource.com/40476 Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
2017-04-12 12:22:16 -06:00
MOVD R0, 0xf0(R0)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·madvise(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD addr+0(FP), R3
MOVD n+8(FP), R4
MOVW flags+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_madvise
runtime: use MADV_FREE on Linux if available On Linux, sysUnused currently uses madvise(MADV_DONTNEED) to signal the kernel that a range of allocated memory contains unneeded data. After a successful call, the range (but not the data it contained before the call to madvise) is still available but the first access to that range will unconditionally incur a page fault (needed to 0-fill the range). A faster alternative is MADV_FREE, available since Linux 4.5. The mechanism is very similar, but the page fault will only be incurred if the kernel, between the call to madvise and the first access, decides to reuse that memory for something else. In sysUnused, test whether MADV_FREE is supported and fall back to MADV_DONTNEED in case it isn't. This requires making the return value of the madvise syscall available to the caller, so change runtime.madvise to return it. Fixes #23687 Change-Id: I962c3429000dd9f4a00846461ad128b71201bb04 Reviewed-on: https://go-review.googlesource.com/135395 Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-09-14 01:57:06 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// int64 futex(int32 *uaddr, int32 op, int32 val,
// struct timespec *timeout, int32 *uaddr2, int32 val2);
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·futex(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD addr+0(FP), R3
MOVW op+8(FP), R4
MOVW val+12(FP), R5
MOVD ts+16(FP), R6
MOVD addr2+24(FP), R7
MOVW val3+32(FP), R8
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_futex
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+40(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
// int64 clone(int32 flags, void *stk, M *mp, G *gp, void (*fn)(void));
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·clone(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVW flags+0(FP), R3
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD stk+8(FP), R4
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// Copy mp, gp, fn off parent stack for use by child.
// Careful: Linux system call clobbers ???.
all: fix assembly vet issues Add missing function prototypes. Fix function prototypes. Use FP references instead of SP references. Fix variable names. Update comments. Clean up whitespace. (Not for vet.) All fairly minor fixes to make vet happy. Updates #11041 Change-Id: Ifab2cdf235ff61cdc226ab1d84b8467b5ac9446c Reviewed-on: https://go-review.googlesource.com/27713 Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
2016-07-11 17:05:57 -06:00
MOVD mp+16(FP), R7
MOVD gp+24(FP), R8
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVD fn+32(FP), R12
MOVD R7, -8(R4)
MOVD R8, -16(R4)
MOVD R12, -24(R4)
MOVD $1234, R7
MOVD R7, -32(R4)
SYSCALL $SYS_clone
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// In parent, return.
CMP R3, $0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
BEQ 3(PC)
MOVW R3, ret+40(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// In child, on new stack.
// initialize essential registers
BL runtime·reginit(SB)
MOVD -32(R1), R7
CMP R7, $1234
BEQ 2(PC)
MOVD R0, 0(R0)
// Initialize m->procid to Linux tid
SYSCALL $SYS_gettid
runtime: merge clone0 and clone We initially added clone0 to handle the case when G or M don't exist, but it turns out that we could have just modified clone. (It also helps that the function we're invoking in clone0 no longer needs arguments.) As a side-effect, newosproc0 is now supported on all linux archs. Change-Id: Ie603af75d8f164310fc16446052d83743961f3ca Reviewed-on: https://go-review.googlesource.com/9164 Reviewed-by: David Crawshaw <crawshaw@golang.org>
2015-04-17 18:27:07 -06:00
MOVD -24(R1), R12 // fn
MOVD -16(R1), R8 // g
MOVD -8(R1), R7 // m
CMP R7, $0
BEQ nog
CMP R8, $0
BEQ nog
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVD R3, m_procid(R7)
// TODO: setup TLS.
// In child, set up new stack
MOVD R7, g_m(R8)
MOVD R8, g
//CALL runtime·stackcheck(SB)
runtime: merge clone0 and clone We initially added clone0 to handle the case when G or M don't exist, but it turns out that we could have just modified clone. (It also helps that the function we're invoking in clone0 no longer needs arguments.) As a side-effect, newosproc0 is now supported on all linux archs. Change-Id: Ie603af75d8f164310fc16446052d83743961f3ca Reviewed-on: https://go-review.googlesource.com/9164 Reviewed-by: David Crawshaw <crawshaw@golang.org>
2015-04-17 18:27:07 -06:00
nog:
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// Call fn
MOVD R12, CTR
BL (CTR)
runtime: fix build after CL 9164 on Linux There is an assumption that the function executed in child thread created by runtime.close should not return. And different systems enforce that differently: some exit that thread, some exit the whole process. The test TestNewOSProc0 introduced in CL 9161 breaks that assumption, so we need to adjust the code to only exit the thread should the called function return. Change-Id: Id631cb2f02ec6fbd765508377a79f3f96c6a2ed6 Reviewed-on: https://go-review.googlesource.com/9246 Reviewed-by: Dave Cheney <dave@cheney.net>
2015-04-22 16:03:18 -06:00
// It shouldn't return. If it does, exit that thread.
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVW $111, R3
runtime: fix build after CL 9164 on Linux There is an assumption that the function executed in child thread created by runtime.close should not return. And different systems enforce that differently: some exit that thread, some exit the whole process. The test TestNewOSProc0 introduced in CL 9161 breaks that assumption, so we need to adjust the code to only exit the thread should the called function return. Change-Id: Id631cb2f02ec6fbd765508377a79f3f96c6a2ed6 Reviewed-on: https://go-review.googlesource.com/9246 Reviewed-by: Dave Cheney <dave@cheney.net>
2015-04-22 16:03:18 -06:00
SYSCALL $SYS_exit
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
BR -2(PC) // keep exiting
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·sigaltstack(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVD new+0(FP), R3
MOVD old+8(FP), R4
SYSCALL $SYS_sigaltstack
BVC 2(PC)
cmd/asm: detect invalid DS form offsets for ppc64x While debugging a recent regression it was discovered that the assembler for ppc64x was not always generating the correct instruction for DS form loads and stores. When an instruction is DS form then the offset must be a multiple of 4, and if it isn't then bits outside the offset field were being incorrectly set resulting in unexpected and incorrect instructions. This change adds a check to determine when the opcode is DS form and then verifies that the offset is a multiple of 4 before generating the instruction, otherwise logs an error. This also changes a few asm files that were using unaligned offsets for DS form loads and stores. In the runtime package these were instructions intended to cause a crash so using aligned or unaligned offsets doesn't change that behavior. Change-Id: Ie3a7e1e65dcc9933b54de7a46a054da8459cb56f Reviewed-on: https://go-review.googlesource.com/40476 Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
2017-04-12 12:22:16 -06:00
MOVD R0, 0xf0(R0) // crash
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·osyield(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_sched_yield
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·sched_getaffinity(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVD pid+0(FP), R3
MOVD len+8(FP), R4
MOVD buf+16(FP), R5
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_sched_getaffinity
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// int32 runtime·epollcreate(int32 size);
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·epollcreate(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW size+0(FP), R3
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_epoll_create
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+8(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// int32 runtime·epollcreate1(int32 flags);
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·epollcreate1(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW flags+0(FP), R3
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_epoll_create1
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+8(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
// func epollctl(epfd, op, fd int32, ev *epollEvent) int
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·epollctl(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW epfd+0(FP), R3
MOVW op+4(FP), R4
MOVW fd+8(FP), R5
MOVD ev+16(FP), R6
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_epoll_ctl
runtime: fix errno sign for epollctl on mips, mips64 and ppc64 The caller of epollctl expects it to return a negative errno value, but it returns a positive errno value on mips, mips64 and ppc64. The change fixes this. Updates #23446 Change-Id: Ie6372eca6c23de21964caaaa433c9a45ef93531e Reviewed-on: https://go-review.googlesource.com/89235 Reviewed-by: Carlos Eduardo Seo <cseo@linux.vnet.ibm.com> Reviewed-by: Ian Lance Taylor <iant@golang.org> Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org>
2018-01-23 00:56:24 -07:00
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// int32 runtime·epollwait(int32 epfd, EpollEvent *ev, int32 nev, int32 timeout);
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·epollwait(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW epfd+0(FP), R3
MOVD ev+8(FP), R4
MOVW nev+16(FP), R5
MOVW timeout+20(FP), R6
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
SYSCALL $SYS_epoll_wait
runtime: fix syscall error returns on mips/mips64/ppc64 The linux syscall functions used in runtime are designed around the calling convention of returning errors as negative numbers. On some other systems (like mips and ppc) the actual syscalls signal errors in other ways. This means that the assembly implementations of the syscall functions on these platforms need to transform the return values in the error cases to match the expected negative errno values. This was addressed for certain syscalls in https://golang.org/cl/19455 and https://golang.org/cl/89235. This patch handles the rest of the syscall functions in sys_linux_*.s that return any value for mips/mips64/ppc64. Fixes #23446 Change-Id: I302100261231f76d5850ab2c2ea080170d7dba72 GitHub-Last-Rev: e358e2b08c76897b13f917cfa12b5085e20337f9 GitHub-Pull-Request: golang/go#26606 Reviewed-on: https://go-review.googlesource.com/125895 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-07-25 12:44:07 -06:00
BVC 2(PC)
NEG R3 // caller expects negative errno
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW R3, ret+24(FP)
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
// void runtime·closeonexec(int32 fd);
cmd/internal/obj, runtime: add NOFRAME flag to suppress stack frame set up on ppc64x Replace the confusing game where a frame size of $-8 would suppress the implicit setting up of a stack frame with a nice explicit flag. The code to set up the function prologue is still a little confusing but better than it was. Change-Id: I1d49278ff42c6bc734ebfb079998b32bc53f8d9a Reviewed-on: https://go-review.googlesource.com/15670 Reviewed-by: Minux Ma <minux@golang.org>
2015-10-08 17:44:27 -06:00
TEXT runtime·closeonexec(SB),NOSPLIT|NOFRAME,$0
[dev.power64] runtime: power64 fixes and ports of changes Fix include paths that got moved in the great pkg/ rename. Add missing runtime/arch_* files for power64. Port changes that happened on default since branching to runtime/{asm,atomic,sys_linux}_power64x.s (precise stacks, calling convention change, various new and deleted functions. Port struct renaming and fix some bugs in runtime/defs_linux_power64.h. LGTM=rsc R=rsc, dave CC=golang-codereviews https://golang.org/cl/161450043
2014-10-27 15:27:03 -06:00
MOVW fd+0(FP), R3 // fd
[dev.power64] runtime: manually written parts for linux/power64 and linux/power64le, checkpoint LGTM=rsc R=rsc, iant CC=golang-codereviews https://golang.org/cl/122430043
2014-08-12 17:48:49 -06:00
MOVD $2, R4 // F_SETFD
MOVD $1, R5 // FD_CLOEXEC
SYSCALL $SYS_fcntl
all: use RET instead of RETURN on ppc64 All of the architectures except ppc64 have only "RET" for the return mnemonic. ppc64 used to have only "RETURN", but commit cf06ea6 introduced RET as a synonym for RETURN to make ppc64 consistent with the other architectures. However, that commit was never followed up to make the code itself consistent by eliminating uses of RETURN. This commit replaces all uses of RETURN in the ppc64 assembly with RET. This was done with sed -i 's/\<RETURN\>/RET/' **/*_ppc64x.s plus one manual change to syscall/asm.s. Change-Id: I3f6c8d2be157df8841d48de988ee43f3e3087995 Reviewed-on: https://go-review.googlesource.com/10672 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Russ Cox <rsc@golang.org> Reviewed-by: Minux Ma <minux@golang.org>
2015-06-03 12:59:27 -06:00
RET
runtime: inform arena placement using sbrk(0) On 32-bit architectures (or if we fail to map a 64-bit-style arena), we try to map the heap arena just above the end of the process image. While we can accept any address, using lower addresses is preferable because lower addresses cause us to map less of the heap bitmap. However, if a program is linked against C code that has global constructors, those constructors may call brk/sbrk to allocate memory (e.g., many C malloc implementations do this for small allocations). The brk also starts just above the process image, so this may adjust the brk past the beginning of where we want to put the heap arena. In this case, the kernel will pick a different address for the arena and it will usually be very high (at least, as these things go in a 32-bit address space). Fix this by consulting the current value of the brk and using this in addition to the end of the process image to compute the initial arena placement. This is implemented only on Linux currently, since we have no evidence that it's an issue on any other OSes. Fixes #19831. Change-Id: Id64b45d08d8c91e4f50d92d0339146250b04f2f8 Reviewed-on: https://go-review.googlesource.com/39810 Run-TryBot: Austin Clements <austin@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
2017-04-06 12:32:37 -06:00
// func sbrk0() uintptr
TEXT runtime·sbrk0(SB),NOSPLIT|NOFRAME,$0
// Implemented as brk(NULL).
MOVD $0, R3
SYSCALL $SYS_brk
MOVD R3, ret+0(FP)
RET
runtime: fix vet complaints for linux/386 Working toward making the tree vet-safe instead of having so many exceptions in cmd/vet/all/whitelist. This CL makes "GOOS=linux GOARCH=386 go vet -unsafeptr=false runtime" happy, while keeping "GO_BUILDER_NAME=misc-vetall go tool dist test" happy too. For #31916. Change-Id: I3e5586a7ff6e359357350d0602c2259493280ded Reviewed-on: https://go-review.googlesource.com/c/go/+/176099 Run-TryBot: Russ Cox <rsc@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Austin Clements <austin@google.com>
2019-05-08 09:44:15 -06:00
TEXT runtime·access(SB),$0-20
MOVD R0, 0(R0) // unimplemented, only needed for android; declared in stubs_linux.go
MOVW R0, ret+16(FP) // for vet
RET
TEXT runtime·connect(SB),$0-28
MOVD R0, 0(R0) // unimplemented, only needed for android; declared in stubs_linux.go
MOVW R0, ret+24(FP) // for vet
RET
TEXT runtime·socket(SB),$0-20
MOVD R0, 0(R0) // unimplemented, only needed for android; declared in stubs_linux.go
MOVW R0, ret+16(FP) // for vet
RET
Copy Permalink