[go] Plan9 arm64 port

[Gerrit Bot (Gerrit)]

Gerrit Bot has uploaded the change for review

Commit message

Plan9 arm64 port

As 32-bit are staying in the past, the plan9 arm64 port is back. I'm running it on bare metal, on a Pi 4. I adapted previous work from @psilva261 (https://github.com/psilva261/go-arm64.plan9). Please advice.

Previous try of including the plan9 arm64 port is here: https://github.com/golang/go/issues/57540

Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

GitHub-Last-Rev: 09830347320ed27c7c8cc5b23acd81d8e200acaa

GitHub-Pull-Request: golang/go#76271

Change diff

diff --git a/src/cmd/dist/build.go b/src/cmd/dist/build.go
index 2b382a1..2b4818c 100644
--- a/src/cmd/dist/build.go
+++ b/src/cmd/dist/build.go
@@ -1811,6 +1811,7 @@
 	"plan9/386":       false,
 	"plan9/amd64":     false,
 	"plan9/arm":       false,
+	"plan9/arm64":     false,
 	"solaris/amd64":   true,
 	"windows/386":     true,
 	"windows/amd64":   true,
diff --git a/src/cmd/go/internal/work/exec.go b/src/cmd/go/internal/work/exec.go
index 0c9e96a..ce078ff 100644
--- a/src/cmd/go/internal/work/exec.go
+++ b/src/cmd/go/internal/work/exec.go
@@ -2075,6 +2075,7 @@
 	{0x00, 0x00, 0x01, 0xEB},                  // Plan 9 i386
 	{0x00, 0x00, 0x8a, 0x97},                  // Plan 9 amd64
 	{0x00, 0x00, 0x06, 0x47},                  // Plan 9 arm
+	{0x00, 0x00, 0x8c, 0x47},                  // Plan 9 arm64
 	{0x00, 0x61, 0x73, 0x6D},                  // WASM
 	{0x01, 0xDF},                              // XCOFF 32bit
 	{0x01, 0xF7},                              // XCOFF 64bit
diff --git a/src/cmd/go/testdata/script/work_env.txt b/src/cmd/go/testdata/script/work_env.txt
index 8b1779e..54e51da 100644
--- a/src/cmd/go/testdata/script/work_env.txt
+++ b/src/cmd/go/testdata/script/work_env.txt
@@ -1,7 +1,7 @@
 go env GOWORK
 stdout '^'$GOPATH'[\\/]src[\\/]go.work$'
 go env
-stdout '^(set )?GOWORK=''?'$GOPATH'[\\/]src[\\/]go.work''?$'
+stdout '^(set )?GOWORK=["'']?'$GOPATH'[\\/]src[\\/]go.work["'']?$'
 
 cd ..
 go env GOWORK
diff --git a/src/cmd/internal/objfile/plan9obj.go b/src/cmd/internal/objfile/plan9obj.go
index edd4023..f9801c0 100644
--- a/src/cmd/internal/objfile/plan9obj.go
+++ b/src/cmd/internal/objfile/plan9obj.go
@@ -137,6 +137,8 @@
 		return "amd64"
 	case plan9obj.MagicARM:
 		return "arm"
+	case plan9obj.MagicARM64:
+		return "arm64"
 	}
 	return ""
 }
diff --git a/src/cmd/link/internal/arm64/obj.go b/src/cmd/link/internal/arm64/obj.go
index 3d35815..cc3b113 100644
--- a/src/cmd/link/internal/arm64/obj.go
+++ b/src/cmd/link/internal/arm64/obj.go
@@ -47,6 +47,9 @@
 		Dwarfreglr: dwarfRegLR,
 		TrampLimit: 0x7c00000, // 26-bit signed offset * 4, leave room for PLT etc.
 
+		Plan9Magic:  0x8c47,
+		Plan9_64Bit: true,
+
 		Adddynrel:        adddynrel,
 		Archinit:         archinit,
 		Archreloc:        archreloc,
@@ -85,12 +88,12 @@
 		ld.Exitf("unknown -H option: %v", ctxt.HeadType)
 
 	case objabi.Hplan9: /* plan 9 */
-		ld.HEADR = 32
+		ld.HEADR = 32 + 8
 		if *ld.FlagRound == -1 {
-			*ld.FlagRound = 4096
+			*ld.FlagRound = 0x10000
 		}
 		if *ld.FlagTextAddr == -1 {
-			*ld.FlagTextAddr = ld.Rnd(4096, *ld.FlagRound) + int64(ld.HEADR)
+			*ld.FlagTextAddr = ld.Rnd(0x10000, *ld.FlagRound) + int64(ld.HEADR)
 		}
 
 	case objabi.Hlinux, /* arm64 elf */
diff --git a/src/cmd/link/internal/ld/main.go b/src/cmd/link/internal/ld/main.go
index d913953..8fee981 100644
--- a/src/cmd/link/internal/ld/main.go
+++ b/src/cmd/link/internal/ld/main.go
@@ -193,7 +193,7 @@
 	addstrdata1(ctxt, "runtime.buildVersion="+buildVersion)
 
 	// TODO(matloob): define these above and then check flag values here
-	if ctxt.Arch.Family == sys.AMD64 && buildcfg.GOOS == "plan9" {
+	if (ctxt.Arch.Family == sys.AMD64 || ctxt.Arch.Family == sys.ARM64) && buildcfg.GOOS == "plan9" {
 		flag.BoolVar(&flag8, "8", false, "use 64-bit addresses in symbol table")
 	}
 	flagHeadType := flag.String("H", "", "set header `type`")
diff --git a/src/cmd/link/internal/ld/symtab.go b/src/cmd/link/internal/ld/symtab.go
index a0345ca..dd2a16b 100644
--- a/src/cmd/link/internal/ld/symtab.go
+++ b/src/cmd/link/internal/ld/symtab.go
@@ -294,7 +294,7 @@
 	}
 	l := 4
 	addr := ldr.SymValue(s)
-	if ctxt.IsAMD64() && !flag8 {
+	if (ctxt.IsAMD64() || ctxt.IsARM64()) && !flag8 {
 		ctxt.Out.Write32b(uint32(addr >> 32))
 		l = 8
 	}
diff --git a/src/cmd/vendor/golang.org/x/sys/plan9/asm_plan9_arm64.s b/src/cmd/vendor/golang.org/x/sys/plan9/asm_plan9_arm64.s
new file mode 100644
index 0000000..9014318
--- /dev/null
+++ b/src/cmd/vendor/golang.org/x/sys/plan9/asm_plan9_arm64.s
@@ -0,0 +1,28 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+#include "textflag.h"
+
+// System call support for plan9 on arm64
+
+// Just jump to package syscall's implementation for all these functions.
+// The runtime may know about them.
+
+TEXT ·Syscall(SB),NOSPLIT,$0-64
+	JMP	syscall·Syscall(SB)
+
+TEXT ·Syscall6(SB),NOSPLIT,$0-88
+	JMP	syscall·Syscall6(SB)
+
+TEXT ·RawSyscall(SB),NOSPLIT,$0-56
+	JMP	syscall·RawSyscall(SB)
+
+TEXT ·RawSyscall6(SB),NOSPLIT,$0-80
+	JMP	syscall·RawSyscall6(SB)
+
+TEXT ·seek(SB),NOSPLIT,$0-56
+	JMP	syscall·seek(SB)
+
+TEXT ·exit(SB),NOSPLIT,$8-8
+	JMP	syscall·exit(SB)
diff --git a/src/cmd/vendor/golang.org/x/sys/plan9/zsyscall_plan9_arm64.go b/src/cmd/vendor/golang.org/x/sys/plan9/zsyscall_plan9_arm64.go
new file mode 100644
index 0000000..dd0265a
--- /dev/null
+++ b/src/cmd/vendor/golang.org/x/sys/plan9/zsyscall_plan9_arm64.go
@@ -0,0 +1,286 @@
+// go run mksyscall.go -l32 -plan9 -tags plan9,arm64 syscall_plan9.go
+// Code generated by the command above; DO NOT EDIT.
+
+//go:build plan9 && arm64
+// +build plan9,arm64
+
+package plan9
+
+import "unsafe"
+
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func fd2path(fd int, buf []byte) (err error) {
+	var _p0 unsafe.Pointer
+	if len(buf) > 0 {
+		_p0 = unsafe.Pointer(&buf[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FD2PATH, uintptr(fd), uintptr(_p0), uintptr(len(buf)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func pipe(p *[2]int32) (err error) {
+	r0, _, e1 := Syscall(SYS_PIPE, uintptr(unsafe.Pointer(p)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func await(s []byte) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(s) > 0 {
+		_p0 = unsafe.Pointer(&s[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_AWAIT, uintptr(_p0), uintptr(len(s)), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func open(path string, mode int) (fd int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_OPEN, uintptr(unsafe.Pointer(_p0)), uintptr(mode), 0)
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func create(path string, mode int, perm uint32) (fd int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_CREATE, uintptr(unsafe.Pointer(_p0)), uintptr(mode), uintptr(perm))
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func remove(path string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_REMOVE, uintptr(unsafe.Pointer(_p0)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func stat(path string, edir []byte) (n int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	var _p1 unsafe.Pointer
+	if len(edir) > 0 {
+		_p1 = unsafe.Pointer(&edir[0])
+	} else {
+		_p1 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_STAT, uintptr(unsafe.Pointer(_p0)), uintptr(_p1), uintptr(len(edir)))
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func bind(name string, old string, flag int) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(name)
+	if err != nil {
+		return
+	}
+	var _p1 *byte
+	_p1, err = BytePtrFromString(old)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_BIND, uintptr(unsafe.Pointer(_p0)), uintptr(unsafe.Pointer(_p1)), uintptr(flag))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func mount(fd int, afd int, old string, flag int, aname string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(old)
+	if err != nil {
+		return
+	}
+	var _p1 *byte
+	_p1, err = BytePtrFromString(aname)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall6(SYS_MOUNT, uintptr(fd), uintptr(afd), uintptr(unsafe.Pointer(_p0)), uintptr(flag), uintptr(unsafe.Pointer(_p1)), 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func wstat(path string, edir []byte) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	var _p1 unsafe.Pointer
+	if len(edir) > 0 {
+		_p1 = unsafe.Pointer(&edir[0])
+	} else {
+		_p1 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_WSTAT, uintptr(unsafe.Pointer(_p0)), uintptr(_p1), uintptr(len(edir)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func chdir(path string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_CHDIR, uintptr(unsafe.Pointer(_p0)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Dup(oldfd int, newfd int) (fd int, err error) {
+	r0, _, e1 := Syscall(SYS_DUP, uintptr(oldfd), uintptr(newfd), 0)
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Pread(fd int, p []byte, offset int64) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(p) > 0 {
+		_p0 = unsafe.Pointer(&p[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall6(SYS_PREAD, uintptr(fd), uintptr(_p0), uintptr(len(p)), uintptr(offset), uintptr(offset>>32), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Pwrite(fd int, p []byte, offset int64) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(p) > 0 {
+		_p0 = unsafe.Pointer(&p[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall6(SYS_PWRITE, uintptr(fd), uintptr(_p0), uintptr(len(p)), uintptr(offset), uintptr(offset>>32), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Close(fd int) (err error) {
+	r0, _, e1 := Syscall(SYS_CLOSE, uintptr(fd), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Fstat(fd int, edir []byte) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(edir) > 0 {
+		_p0 = unsafe.Pointer(&edir[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FSTAT, uintptr(fd), uintptr(_p0), uintptr(len(edir)))
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Fwstat(fd int, edir []byte) (err error) {
+	var _p0 unsafe.Pointer
+	if len(edir) > 0 {
+		_p0 = unsafe.Pointer(&edir[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FWSTAT, uintptr(fd), uintptr(_p0), uintptr(len(edir)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
diff --git a/src/debug/buildinfo/buildinfo.go b/src/debug/buildinfo/buildinfo.go
index d202d50..3ceb23c 100644
--- a/src/debug/buildinfo/buildinfo.go
+++ b/src/debug/buildinfo/buildinfo.go
@@ -274,7 +274,7 @@
 	if len(magic) >= 4 {
 		m := binary.BigEndian.Uint32(magic)
 		switch m {
-		case plan9obj.Magic386, plan9obj.MagicAMD64, plan9obj.MagicARM:
+		case plan9obj.Magic386, plan9obj.MagicAMD64, plan9obj.MagicARM, plan9obj.MagicARM64:
 			return true
 		}
 	}
diff --git a/src/debug/plan9obj/file.go b/src/debug/plan9obj/file.go
index 0880c3c..d89c6cd 100644
--- a/src/debug/plan9obj/file.go
+++ b/src/debug/plan9obj/file.go
@@ -130,7 +130,7 @@
 func parseMagic(magic []byte) (uint32, error) {
 	m := binary.BigEndian.Uint32(magic)
 	switch m {
-	case Magic386, MagicAMD64, MagicARM:
+	case Magic386, MagicAMD64, MagicARM, MagicARM64:
 		return m, nil
 	}
 	return 0, &formatError{0, "bad magic number", magic}
@@ -145,7 +145,7 @@
 	if _, err := r.ReadAt(magic[:], 0); err != nil {
 		return nil, err
 	}
-	_, err := parseMagic(magic[:])
+	m, err := parseMagic(magic[:])
 	if err != nil {
 		return nil, err
 	}
@@ -169,7 +169,11 @@
 			return nil, err
 		}
 		f.PtrSize = 8
-		f.LoadAddress = 0x200000
+		if m == MagicARM64 {
+			f.LoadAddress = 0x10000
+		} else {
+			f.LoadAddress = 0x200000
+		}
 		f.HdrSize += 8
 	}
 
diff --git a/src/debug/plan9obj/file_test.go b/src/debug/plan9obj/file_test.go
index 7e107bc..75498fc 100644
--- a/src/debug/plan9obj/file_test.go
+++ b/src/debug/plan9obj/file_test.go
@@ -38,6 +38,17 @@
 			{"pcsz", 0xca0, 0x7947},
 		},
 	},
+	{
+		"testdata/arm64-plan9-exec",
+		FileHeader{MagicARM64, 0x408, 0x1003c, 8, 0x10000, 40},
+		[]*SectionHeader{
+			{"text", 0x4a00, 0x28},
+			{"data", 0xaa0, 0x4a28},
+			{"syms", 0x294a, 0x54c8},
+			{"spsz", 0x0, 0x7e12},
+			{"pcsz", 0xa6c, 0x7e12},
+		},
+	},
 }
 
 func TestOpen(t *testing.T) {
diff --git a/src/debug/plan9obj/plan9obj.go b/src/debug/plan9obj/plan9obj.go
index 7a19451..98e97a2 100644
--- a/src/debug/plan9obj/plan9obj.go
+++ b/src/debug/plan9obj/plan9obj.go
@@ -33,4 +33,5 @@
 	Magic386   = (4*11+0)*11 + 7
 	MagicAMD64 = (4*26+0)*26 + 7 + Magic64
 	MagicARM   = (4*20+0)*20 + 7
+	MagicARM64 = (4*28+0)*28 + 7 + Magic64
 )
diff --git a/src/debug/plan9obj/testdata/arm64-plan9-exec b/src/debug/plan9obj/testdata/arm64-plan9-exec
new file mode 100755
index 0000000..3fa2c42
--- /dev/null
+++ b/src/debug/plan9obj/testdata/arm64-plan9-exec
Binary files differ
diff --git a/src/internal/platform/zosarch.go b/src/internal/platform/zosarch.go
index a2f5b22..a81735c 100644
--- a/src/internal/platform/zosarch.go
+++ b/src/internal/platform/zosarch.go
@@ -53,6 +53,7 @@
 	{"plan9", "386"},
 	{"plan9", "amd64"},
 	{"plan9", "arm"},
+	{"plan9", "arm64"},
 	{"solaris", "amd64"},
 	{"wasip1", "wasm"},
 	{"windows", "386"},
@@ -106,6 +107,7 @@
 	{"plan9", "386"}:       {},
 	{"plan9", "amd64"}:     {},
 	{"plan9", "arm"}:       {},
+	{"plan9", "arm64"}:     {},
 	{"solaris", "amd64"}:   {CgoSupported: true},
 	{"wasip1", "wasm"}:     {},
 	{"windows", "386"}:     {CgoSupported: true, FirstClass: true},
diff --git a/src/net/http/serve_test.go b/src/net/http/serve_test.go
index 4a16ba0..361beb1 100644
--- a/src/net/http/serve_test.go
+++ b/src/net/http/serve_test.go
@@ -4562,7 +4562,11 @@
 			if err != nil {
 				// Try to deflake spurious "connection reset by peer" under load.
 				// See golang.org/issue/22540.
-				time.Sleep(10 * time.Millisecond)
+				deflake := 10 * time.Millisecond
+				if runtime.GOOS == "plan9" {
+					deflake = 100 * time.Millisecond
+				}
+				time.Sleep(deflake)
 				res, err = cst.c.Get(cst.ts.URL)
 				if err != nil {
 					t.Error(err)
diff --git a/src/net/udpsock_test.go b/src/net/udpsock_test.go
index a79e9f8..a0cedb7 100644
--- a/src/net/udpsock_test.go
+++ b/src/net/udpsock_test.go
@@ -679,7 +679,7 @@
 	}
 
 	switch runtime.GOOS {
-	case "dragonfly", "openbsd":
+	case "dragonfly", "openbsd", "plan9":
 		// DragonflyBSD's IPv6 sockets are always IPv6-only, according to the man page:
 		// https://www.dragonflybsd.org/cgi/web-man?command=ip6 (search for IPV6_V6ONLY).
 		// OpenBSD's IPv6 sockets are always IPv6-only, according to the man page:
diff --git a/src/runtime/defs_plan9_arm64.go b/src/runtime/defs_plan9_arm64.go
new file mode 100644
index 0000000..e6803c9
--- /dev/null
+++ b/src/runtime/defs_plan9_arm64.go
@@ -0,0 +1,103 @@
+// Copyright 2015 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package runtime
+
+const _PAGESHIFT = 16
+const _PAGESIZE = 1 << _PAGESHIFT
+
+type ureg struct {
+	/* AArch64 registers */
+	r0  uint64 /* general registers */
+	r1  uint64 /* ... */
+	r2  uint64 /* ... */
+	r3  uint64 /* ... */
+	r4  uint64 /* ... */
+	r5  uint64 /* ... */
+	r6  uint64 /* ... */
+	r7  uint64 /* ... */
+	r8  uint64 /* ... */
+	r9  uint64 /* ... */
+	r10 uint64 /* ... */
+	r11 uint64 /* ... */
+	r12 uint64 /* ... */
+	r13 uint64 /* ... */
+	r14 uint64 /* ... */
+	r15 uint64 /* ... */
+	r16 uint64 /* ... */
+	r17 uint64 /* ... */
+	r18 uint64 /* ... */
+	r19 uint64 /* ... */
+	r20 uint64 /* ... */
+	r21 uint64 /* ... */
+	r22 uint64 /* ... */
+	r23 uint64 /* ... */
+	r24 uint64 /* ... */
+	r25 uint64 /* ... */
+	r26 uint64 /* ... */
+	r27 uint64 /* ... */
+	r28 uint64 /* ... */
+	r29 uint64 /* ... */
+	r30 uint64 /* link (lr) */
+	sp  uint64
+	pc  uint64 /* interrupted addr */
+	psr uint64
+	typ uint64 /* of exception */
+}
+
+type sigctxt struct {
+	u *ureg
+}
+
+//go:nosplit
+//go:nowritebarrierrec
+func (c *sigctxt) pc() uintptr { return uintptr(c.u.pc) }
+
+func (c *sigctxt) sp() uintptr { return uintptr(c.u.sp) }
+func (c *sigctxt) lr() uintptr { return uintptr(c.u.r30) }
+
+func (c *sigctxt) setpc(x uintptr)  { c.u.pc = uint64(x) }
+func (c *sigctxt) setsp(x uintptr)  { c.u.sp = uint64(x) }
+func (c *sigctxt) setlr(x uintptr)  { c.u.r30 = uint64(x) }
+func (c *sigctxt) savelr(x uintptr) { c.u.r0 = uint64(x) }
+
+func dumpregs(u *ureg) {
+	print("r0    ", hex(u.r0), "\n")
+	print("r1    ", hex(u.r1), "\n")
+	print("r2    ", hex(u.r2), "\n")
+	print("r3    ", hex(u.r3), "\n")
+	print("r4    ", hex(u.r4), "\n")
+	print("r5    ", hex(u.r5), "\n")
+	print("r6    ", hex(u.r6), "\n")
+	print("r7    ", hex(u.r7), "\n")
+	print("r8    ", hex(u.r8), "\n")
+	print("r9    ", hex(u.r9), "\n")
+	print("r10   ", hex(u.r10), "\n")
+	print("r11   ", hex(u.r11), "\n")
+	print("r12   ", hex(u.r12), "\n")
+	print("r13   ", hex(u.r13), "\n")
+	print("r14   ", hex(u.r14), "\n")
+	print("r15   ", hex(u.r15), "\n")
+	print("r16   ", hex(u.r16), "\n")
+	print("r17   ", hex(u.r17), "\n")
+	print("r18   ", hex(u.r18), "\n")
+	print("r19   ", hex(u.r19), "\n")
+	print("r20   ", hex(u.r20), "\n")
+	print("r21   ", hex(u.r21), "\n")
+	print("r22   ", hex(u.r22), "\n")
+	print("r23   ", hex(u.r23), "\n")
+	print("r24   ", hex(u.r24), "\n")
+	print("r25   ", hex(u.r25), "\n")
+	print("r26   ", hex(u.r26), "\n")
+	print("r27   ", hex(u.r27), "\n")
+	print("r28   ", hex(u.r28), "\n")
+	print("r29   ", hex(u.r29), "\n")
+	print("r30   ", hex(u.r30), "\n")
+	print("sp    ", hex(u.sp), "\n")
+	print("pc    ", hex(u.pc), "\n")
+	print("psr   ", hex(u.psr), "\n")
+	print("type  ", hex(u.typ), "\n")
+}
+
+func sigpanictramp()
diff --git a/src/runtime/os_plan9.go b/src/runtime/os_plan9.go
index 80c101f..f738101 100644
--- a/src/runtime/os_plan9.go
+++ b/src/runtime/os_plan9.go
@@ -587,10 +587,3 @@
 	// fall back to unix time
 	return int64(frombe(t[0]))
 }
-
-//go:nosplit
-func walltime() (sec int64, nsec int32) {
-	var t [1]uint64
-	readtime(&t[0], 1, 1)
-	return timesplit(frombe(t[0]))
-}
diff --git a/src/runtime/os_plan9_386.go b/src/runtime/os_plan9_386.go
new file mode 100644
index 0000000..66f09f4
--- /dev/null
+++ b/src/runtime/os_plan9_386.go
@@ -0,0 +1,12 @@
+// Copyright 2025 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package runtime
+
+//go:nosplit
+func walltime() (sec int64, nsec int32) {
+	var t [1]uint64
+	readtime(&t[0], 1, 1)
+	return timesplit(frombe(t[0]))
+}
diff --git a/src/runtime/os_plan9_amd64.go b/src/runtime/os_plan9_amd64.go
new file mode 100644
index 0000000..66f09f4
--- /dev/null
+++ b/src/runtime/os_plan9_amd64.go
@@ -0,0 +1,12 @@
+// Copyright 2025 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package runtime
+
+//go:nosplit
+func walltime() (sec int64, nsec int32) {
+	var t [1]uint64
+	readtime(&t[0], 1, 1)
+	return timesplit(frombe(t[0]))
+}
diff --git a/src/runtime/os_plan9_arm.go b/src/runtime/os_plan9_arm.go
index cce6229..2e9530e 100644
--- a/src/runtime/os_plan9_arm.go
+++ b/src/runtime/os_plan9_arm.go
@@ -13,3 +13,10 @@
 	// runtime·nanotime() is a poor approximation of CPU ticks that is enough for the profiler.
 	return nanotime()
 }
+
+//go:nosplit
+func walltime() (sec int64, nsec int32) {
+	var t [1]uint64
+	readtime(&t[0], 1, 1)
+	return timesplit(frombe(t[0]))
+}
diff --git a/src/runtime/os_plan9_arm64.go b/src/runtime/os_plan9_arm64.go
new file mode 100644
index 0000000..3d3dd7e
--- /dev/null
+++ b/src/runtime/os_plan9_arm64.go
@@ -0,0 +1,12 @@
+// Copyright 2015 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package runtime
+
+//go:nosplit
+func cputicks() int64 {
+	// Currently cputicks() is used in blocking profiler and to seed runtime·fastrand().
+	// runtime·nanotime() is a poor approximation of CPU ticks that is enough for the profiler.
+	return nanotime()
+}
diff --git a/src/runtime/rt0_plan9_arm64.s b/src/runtime/rt0_plan9_arm64.s
new file mode 100644
index 0000000..192ff38
--- /dev/null
+++ b/src/runtime/rt0_plan9_arm64.s
@@ -0,0 +1,16 @@
+// Copyright 2015 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.
+
+#include "textflag.h"
+
+//in plan 9 argc is at top of stack followed by ptrs to arguments
+
+TEXT _rt0_arm64_plan9(SB),NOSPLIT|NOFRAME,$0
+	MOVD	R0, _tos(SB)
+	MOVD	0(RSP), R0
+	MOVD	$8(RSP), R1
+	MOVD	$runtime·rt0_go(SB), R2
+	BL	(R2)
+
+GLOBL _tos(SB), NOPTR, $8
diff --git a/src/runtime/sys_plan9_arm64.s b/src/runtime/sys_plan9_arm64.s
new file mode 100644
index 0000000..f4d3a25
--- /dev/null
+++ b/src/runtime/sys_plan9_arm64.s
@@ -0,0 +1,298 @@
+// Copyright 2015 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.
+
+#include "go_asm.h"
+#include "go_tls.h"
+#include "textflag.h"
+
+// from ../syscall/zsysnum_plan9.go
+
+#define SYS_SYSR1       0
+#define SYS_BIND        2
+#define SYS_CHDIR       3
+#define SYS_CLOSE       4
+#define SYS_DUP         5
+#define SYS_ALARM       6
+#define SYS_EXEC        7
+#define SYS_EXITS       8
+#define SYS_FAUTH       10
+#define SYS_SEGBRK      12
+#define SYS_OPEN        14
+#define SYS_OSEEK       16
+#define SYS_SLEEP       17
+#define SYS_RFORK       19
+#define SYS_PIPE        21
+#define SYS_CREATE      22
+#define SYS_FD2PATH     23
+#define SYS_BRK_        24
+#define SYS_REMOVE      25
+#define SYS_NOTIFY      28
+#define SYS_NOTED       29
+#define SYS_SEGATTACH   30
+#define SYS_SEGDETACH   31
+#define SYS_SEGFREE     32
+#define SYS_SEGFLUSH    33
+#define SYS_RENDEZVOUS  34
+#define SYS_UNMOUNT     35
+#define SYS_SEMACQUIRE  37
+#define SYS_SEMRELEASE  38
+#define SYS_SEEK        39
+#define SYS_FVERSION    40
+#define SYS_ERRSTR      41
+#define SYS_STAT        42
+#define SYS_FSTAT       43
+#define SYS_WSTAT       44
+#define SYS_FWSTAT      45
+#define SYS_MOUNT       46
+#define SYS_AWAIT       47
+#define SYS_PREAD       50
+#define SYS_PWRITE      51
+#define SYS_TSEMACQUIRE 52
+#define SYS_NSEC        53
+
+//func open(name *byte, mode int, perm int32) int
+TEXT runtime·open(SB),NOSPLIT,$0-20
+	MOVD    $SYS_OPEN, R0
+	SVC	$0
+	MOVWU	R0, ret+16(FP)
+	RET
+
+//func pread(fd int32, buf unsafe.Pointer, nbytes int32, offset int64) int32
+TEXT runtime·pread(SB),NOSPLIT,$0-36
+	MOVD    $SYS_PREAD, R0
+	SVC	$0
+	MOVWU	R0, ret+32(FP)
+	RET
+
+//func pwrite(fd int32, buf unsafe.Pointer, nbytes int32, offset int64) int32
+TEXT runtime·pwrite(SB),NOSPLIT,$0-36
+	MOVD    $SYS_PWRITE, R0
+	SVC	$0
+	MOVWU	R0, ret+32(FP)
+	RET
+
+//func seek(fd int32, offset int64, whence int32) int64
+TEXT runtime·seek(SB),NOSPLIT,$0-32
+	MOVD	$ret+24(FP), R0
+	MOVWU	fd+0(FP), R2
+	MOVD	offset+8(FP), R3
+	MOVWU	whence+16(FP), R4
+
+	MOVD    $sysargs-0(SP), R1
+
+	MOVD	R0, 8(R1)
+	MOVWU	R2, 16(R1)
+	MOVD	R3, 24(R1)
+	MOVWU	R4, 32(R1)
+
+	MOVD	$SYS_SEEK, R0
+	SVC	$0
+
+	CMP	$-1, R0
+	BNE	2(PC)
+	MOVD	R0, ret+24(FP)
+	RET
+
+//func closefd(fd int32) int32
+TEXT runtime·closefd(SB),NOSPLIT,$0-12
+	MOVD	$SYS_CLOSE, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func dupfd(old, new int32) int32
+TEXT runtime·dupfd(SB),NOSPLIT,$0-12
+	MOVD	$SYS_DUP, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func exits(msg *byte)
+TEXT runtime·exits(SB),NOSPLIT,$0-8
+	MOVD    $SYS_EXITS, R0
+	SVC	$0
+	RET
+
+//func brk_(addr unsafe.Pointer) int32
+TEXT runtime·brk_(SB),NOSPLIT,$0-12
+	MOVD    $SYS_BRK_, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func sleep(ms int32) int32
+TEXT runtime·sleep(SB),NOSPLIT,$0-12
+	MOVD    $SYS_SLEEP, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func plan9_semacquire(addr *uint32, block int32) int32
+TEXT runtime·plan9_semacquire(SB),NOSPLIT,$0-20
+	MOVD	$SYS_SEMACQUIRE, R0
+	SVC	$0
+	MOVWU	R0, ret+16(FP)
+	RET
+
+//func plan9_tsemacquire(addr *uint32, ms int32) int32
+TEXT runtime·plan9_tsemacquire(SB),NOSPLIT,$0-20
+	MOVD	$SYS_TSEMACQUIRE, R0
+	SVC	$0
+	MOVWU	R0, ret+16(FP)
+	RET
+
+// func timesplit(u uint64) (sec int64, nsec int32)
+TEXT runtime·timesplit(SB),NOSPLIT,$0-16
+	// load u (nanoseconds)
+	MOVD	u+0(FP), R0
+
+	// compute sec = u / 1e9
+	MOVD	R0, R1
+	MOVD	$1000000000, R2
+	UDIV	R2, R1		// R1 = R1 / R2  -> seconds
+
+	// compute rem = u - sec * 1e9
+	MOVD	R1, R3
+	MUL	R3, R2		// R2 = sec * 1e9
+	SUB	R2, R0		// R0 = u - (sec*1e9) -> remainder (nsec)
+
+	// store results
+	MOVD	R1, sec+0(FP)
+	MOVWU	R0, nsec+8(FP)
+	RET
+
+//func nsec(*int64) int64
+TEXT runtime·nsec(SB),NOSPLIT|NOFRAME,$0-16
+	MOVD	$SYS_NSEC, R0
+	SVC	$0
+	MOVD	R0, ret+8(FP)
+	RET
+
+// func walltime() (sec int64, nsec int32)
+TEXT runtime·walltime(SB),NOSPLIT,$16-12
+	// use nsec system call to get current time in nanoseconds
+
+	MOVD	$SYS_NSEC, R0
+	SVC	$0
+
+	MOVD	R0, R1
+	MOVD	$1000000000, R2
+	UDIV	R2, R1
+
+	MOVD	R1, R3
+	MUL	R3, R2
+	SUB	R2, R0
+
+	MOVD	R1,sec+0(FP)
+	MOVWU	R0,nsec+8(FP)
+	RET
+
+//func notify(fn unsafe.Pointer) int32
+TEXT runtime·notify(SB),NOSPLIT,$0-12
+	MOVD	$SYS_NOTIFY, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func noted(mode int32) int32
+TEXT runtime·noted(SB),NOSPLIT,$0-12
+	MOVD	$SYS_NOTED, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func plan9_semrelease(addr *uint32, count int32) int32
+TEXT runtime·plan9_semrelease(SB),NOSPLIT,$0-20
+	MOVD	$SYS_SEMRELEASE, R0
+	SVC	$0
+	MOVWU	R0, ret+16(FP)
+	RET
+
+//func rfork(flags int32) int32
+TEXT runtime·rfork(SB),NOSPLIT,$0-12
+	MOVD	$SYS_RFORK, R0
+	SVC	$0
+	MOVWU	R0, ret+8(FP)
+	RET
+
+//func tstart_plan9(newm *m)
+TEXT runtime·tstart_plan9(SB),NOSPLIT,$8-8
+	MOVD	newm+0(FP), R1
+	MOVD	m_g0(R1), g
+
+	// Layout new m scheduler stack on os stack.
+	MOVD	RSP, R0
+	MOVD	R0, g_stack+stack_hi(g)
+	SUB	$(64*1024), R0
+	MOVD	R0, (g_stack+stack_lo)(g)
+	MOVD	R0, g_stackguard0(g)
+	MOVD	R0, g_stackguard1(g)
+
+	// Initialize procid from TOS struct.
+	MOVD	_tos(SB), R0
+	MOVWU	64(R0), R0
+	MOVD	R0, m_procid(R1)	// save pid as m->procid
+
+	BL	runtime·mstart(SB)
+
+	// Exit the thread.
+	MOVD	$0, R0
+	MOVD	R0, 8(RSP)
+	CALL	runtime·exits(SB)
+	JMP	0(PC)
+
+//func sigtramp(ureg, note unsafe.Pointer)
+TEXT runtime·sigtramp(SB),NOSPLIT,$0-16
+	// check that g and m exist
+	CMP	$0, g
+	BEQ	4(PC)
+	MOVD	g_m(g), R0
+	CMP 	$0, R0
+	BNE	2(PC)
+	BL	runtime·badsignal2(SB)	// will exit
+
+	// save args
+	MOVD	ureg+0(FP), R1
+	MOVD	note+8(FP), R2
+
+	// change stack
+	MOVD	m_gsignal(R0), R3
+	MOVD	(g_stack+stack_hi)(R3), R4
+	MOVD	R4, RSP
+
+	// make room for args, retval and g
+	SUB	$48, RSP
+
+	// save g
+	MOVD	g, R3
+	MOVD	R3, 40(RSP)
+
+	// g = m->gsignal
+	MOVD	m_gsignal(R0), g
+
+	// load args and call sighandler
+	MOVD R1, 8(RSP)
+	MOVD R2, 16(RSP)
+	MOVD R3, 24(RSP)
+
+	BL	runtime·sighandler(SB)
+	MOVWU	32(RSP), R0			// retval
+
+	// restore g
+	MOVD	40(RSP), g
+
+	// call noted(R0)
+	MOVD	R0, 8(RSP)
+	BL	runtime·noted(SB)
+	RET
+
+//func sigpanictramp()
+TEXT  runtime·sigpanictramp(SB),NOSPLIT,$0-0
+	MOVD.W	R0, -16(RSP)
+	B	runtime·sigpanic(SB)
+
+//func setfpmasks()
+// Mask all SSE floating-point exceptions (only amd64?)
+TEXT runtime·setfpmasks(SB),NOSPLIT,$0
+	RET
diff --git a/src/runtime/tls_arm64.h b/src/runtime/tls_arm64.h
index 3aa8c63..afa6cf6 100644
--- a/src/runtime/tls_arm64.h
+++ b/src/runtime/tls_arm64.h
@@ -36,6 +36,10 @@
 #define MRS_TPIDR_R0 WORD $0xd53bd040 // MRS TPIDR_EL0, R0
 #endif
 
+#ifdef GOOS_plan9
+#define MRS_TPIDR_R0 WORD $0xd53bd040 // MRS TPIDR_EL0, R0
+#endif
+
 #ifdef GOOS_windows
 #define TLS_windows
 #endif
diff --git a/src/syscall/asm_plan9_arm64.s b/src/syscall/asm_plan9_arm64.s
new file mode 100644
index 0000000..6605fb4
--- /dev/null
+++ b/src/syscall/asm_plan9_arm64.s
@@ -0,0 +1,216 @@
+// Copyright 2009 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+#include "textflag.h"
+#include "funcdata.h"
+
+#define SYS_ERRSTR 41	/* from zsysnum_plan9.go */
+#define SYS_SEEK 39	/* from zsysnum_plan9.go */
+
+// System call support for plan9 on arm64
+
+//func Syscall(trap, a1, a2, a3 uintptr) (r1, r2 uintptr, err ErrorString)
+TEXT	·Syscall(SB),NOSPLIT,$168-64
+	NO_LOCAL_POINTERS
+	BL	runtime·entersyscall(SB)
+
+	MOVD	trap+0(FP), R0
+	MOVD	a1+8(FP), R2
+	MOVD	a2+16(FP), R3
+	MOVD	a3+24(FP), R4
+
+	// move to syscall args
+	MOVD	R2, sysargs-192(FP)
+	MOVD	R3, sysargs-184(FP)
+	MOVD	R4, sysargs-176(FP)
+
+	SVC	$0
+
+	// put return values into r1, r2, err
+	MOVD	R0, r1+32(FP)
+	MOVD	R1, r2+40(FP)
+	MOVD	ZR, err+48(FP)
+
+	// put error if needed
+	CMP	$-1, R0
+	BEQ	syscallerr
+	BL	runtime·exitsyscall(SB)
+	MOVD	$·emptystring+0(SB), R2
+	B	syscallok
+syscallerr:
+	MOVD	$errbuf-128(SP), R2
+	MOVD	$128, R3
+
+	MOVD	$SYS_ERRSTR, R0
+	MOVD	R2, err-192(FP)
+	MOVD	R3, nerr-184(FP)
+	SVC	$0
+
+	BL	runtime·exitsyscall(SB)
+	BL	runtime·gostring(SB)
+	MOVD	$str-160(SP), R2
+syscallok:
+	MOVD	$err+48(FP), R1
+	MOVD	0(R2), R3
+	MOVD	8(R2), R4
+	MOVD	R3, 0(R1)
+	MOVD	R4, 8(R1)
+	RET
+
+
+//func Syscall6(trap, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2 uintptr, err ErrorString)
+// Actually Syscall5 but the rest of the code expects it to be named Syscall6.
+TEXT	·Syscall6(SB),NOSPLIT,$168-88
+	NO_LOCAL_POINTERS
+	BL	runtime·entersyscall(SB)
+
+	MOVD	trap+0(FP), R1
+	MOVD	a1+8(FP), R2
+	MOVD	a2+16(FP), R3
+	MOVD	a3+24(FP), R4
+	MOVD	a4+32(FP), R5
+	MOVD	a5+40(FP), R6
+	MOVD	a6+48(FP), R7
+
+	// dereference pointers
+	MOVD	R1, R0
+	MOVD	R2, sysargs-192(FP)
+	MOVD	R3, sysargs-184(FP)
+	MOVD	R4, sysargs-176(FP)
+	MOVD	R5, sysargs-168(FP)
+	MOVD	R6, sysargs-160(FP)
+	MOVD	R7, sysargs-152(FP)
+
+	SVC	$0
+
+	// put return value into r1, r2, err
+	MOVD	R0, r1+56(FP)
+	MOVD	R1, r2+64(FP)
+	MOVD	ZR, err+72(FP)
+
+	// put error if needed
+	CMP	$-1, R0
+	BEQ	syscall6err
+	BL	runtime·exitsyscall(SB)
+	MOVD	$·emptystring+0(SB), R2
+	B	syscall6ok
+syscall6err:
+	MOVD	$errbuf-128(SP), R2
+	MOVD	$128, R3
+
+	MOVD	$SYS_ERRSTR, R0
+	MOVD	R2, err-192(FP)
+	MOVD	R3, nerr-184(FP)
+	SVC	$0
+
+	BL	runtime·exitsyscall(SB)
+	BL	runtime·gostring(SB)
+	MOVD	$str-160(SP), R2
+syscall6ok:
+	MOVD	$err+72(FP), R1
+	MOVD	0(R2), R3
+	MOVD	8(R2), R4
+	MOVD	R3, 0(R1)
+	MOVD	R4, 8(R1)
+	RET
+
+//func RawSyscall(trap, a1, a2, a3 uintptr) (r1, r2, err uintptr)
+TEXT ·RawSyscall(SB),NOSPLIT,$24-56
+	MOVD	trap+0(FP), R1
+	MOVD	a1+8(FP), R2
+	MOVD	a2+16(FP), R3
+	MOVD	a3+24(FP), R4
+
+	// move to syscall args
+	MOVD	R1, R0
+	MOVD	R2, sysargs-48(FP)
+	MOVD	R3, sysargs-40(FP)
+	MOVD	R4, sysargs-32(FP)
+
+	SVC	$0
+
+	// put return values into r1, r2, err
+	MOVD	R0, r1+32(FP)
+	MOVD	R0, r2+40(FP)
+	MOVD	R0, err+48(FP)
+
+	RET
+
+//func RawSyscall6(trap, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2, err uintptr)
+// Actually RawSyscall5 but the rest of the code expects it to be named RawSyscall6.
+TEXT	·RawSyscall6(SB),NOSPLIT,$48-80
+	MOVD	trap+0(FP), R1
+	MOVD	a1+8(FP), R2
+	MOVD	a2+16(FP), R3
+	MOVD	a3+24(FP), R4
+	MOVD	a4+32(FP), R5
+	MOVD	a5+40(FP), R6
+	MOVD	a6+48(FP), R7
+
+	// move to syscall args
+	MOVD	R1, R0
+	MOVD	R2, sysargs-64(FP)
+	MOVD	R3, sysargs-56(FP)
+	MOVD	R4, sysargs-48(FP)
+	MOVD	R5, sysargs-40(FP)
+	MOVD	R6, sysargs-32(FP)
+	MOVD	R7, sysargs-24(FP)
+
+	SVC	$0
+
+	// put return values into r1, r2, err
+	MOVD	R0, r1+56(FP)
+	MOVD	R1, r2+64(FP)
+	MOVD	ZR, err+72(FP)
+
+	RET
+
+//func seek(placeholder uintptr, fd int, offset int64, whence int) (newoffset int64, err string)
+TEXT ·seek(SB),NOSPLIT,$168-56
+	NO_LOCAL_POINTERS
+
+	MOVD	$newoffset+32(FP), R0
+	MOVWU	fd+8(FP), R2
+	MOVD	offset+16(FP), R3
+	MOVWU	whence+24(FP), R4
+
+	// move to syscall args
+	MOVD	R0, sysargs-192(FP)
+	MOVWU	R2, sysargs-184(FP)
+	MOVD	R3, sysargs-176(FP)
+	MOVWU	R4, sysargs-168(FP)
+
+	MOVD	$SYS_SEEK, R0
+	SVC	$0
+
+	// put err
+	MOVD	ZR, err+40(FP)
+
+	// put error if needed
+	CMP	$-1, R0
+	BEQ	syscallerr
+	MOVD	$·emptystring+0(SB), R2
+	B	syscallok
+syscallerr:
+	MOVD	R0, newoffset+32(FP)
+
+	MOVD	$errbuf-128(SP), R2
+	MOVD	$128, R3
+
+	MOVD	$SYS_ERRSTR, R0
+	MOVD	R2, err-192(FP)
+	MOVD	R3, nerr-184(FP)
+	SVC	$0
+
+	BL	runtime·gostring(SB)
+	MOVD	$str-160(SP), R2
+syscallok:
+	MOVD	$err+40(FP), R1
+	MOVD	0(R2), R3
+	MOVD	8(R2), R4
+	MOVD	R3, 0(R1)
+	MOVD	R4, 8(R1)
+	RET
+
+
diff --git a/src/syscall/zsyscall_plan9_arm64.go b/src/syscall/zsyscall_plan9_arm64.go
new file mode 100644
index 0000000..2be5bd9
--- /dev/null
+++ b/src/syscall/zsyscall_plan9_arm64.go
@@ -0,0 +1,284 @@
+// mksyscall.pl -l32 -plan9 -tags plan9,arm64 syscall_plan9.go
+// Code generated by the command above; DO NOT EDIT.
+
+//go:build plan9 && arm64
+
+package syscall
+
+import "unsafe"
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func fd2path(fd int, buf []byte) (err error) {
+	var _p0 unsafe.Pointer
+	if len(buf) > 0 {
+		_p0 = unsafe.Pointer(&buf[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FD2PATH, uintptr(fd), uintptr(_p0), uintptr(len(buf)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func pipe(p *[2]int32) (err error) {
+	r0, _, e1 := Syscall(SYS_PIPE, uintptr(unsafe.Pointer(p)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func await(s []byte) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(s) > 0 {
+		_p0 = unsafe.Pointer(&s[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_AWAIT, uintptr(_p0), uintptr(len(s)), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func open(path string, mode int) (fd int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_OPEN, uintptr(unsafe.Pointer(_p0)), uintptr(mode), 0)
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func create(path string, mode int, perm uint32) (fd int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_CREATE, uintptr(unsafe.Pointer(_p0)), uintptr(mode), uintptr(perm))
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func remove(path string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_REMOVE, uintptr(unsafe.Pointer(_p0)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func stat(path string, edir []byte) (n int, err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	var _p1 unsafe.Pointer
+	if len(edir) > 0 {
+		_p1 = unsafe.Pointer(&edir[0])
+	} else {
+		_p1 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_STAT, uintptr(unsafe.Pointer(_p0)), uintptr(_p1), uintptr(len(edir)))
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func bind(name string, old string, flag int) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(name)
+	if err != nil {
+		return
+	}
+	var _p1 *byte
+	_p1, err = BytePtrFromString(old)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_BIND, uintptr(unsafe.Pointer(_p0)), uintptr(unsafe.Pointer(_p1)), uintptr(flag))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func mount(fd int, afd int, old string, flag int, aname string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(old)
+	if err != nil {
+		return
+	}
+	var _p1 *byte
+	_p1, err = BytePtrFromString(aname)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall6(SYS_MOUNT, uintptr(fd), uintptr(afd), uintptr(unsafe.Pointer(_p0)), uintptr(flag), uintptr(unsafe.Pointer(_p1)), 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func wstat(path string, edir []byte) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	var _p1 unsafe.Pointer
+	if len(edir) > 0 {
+		_p1 = unsafe.Pointer(&edir[0])
+	} else {
+		_p1 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_WSTAT, uintptr(unsafe.Pointer(_p0)), uintptr(_p1), uintptr(len(edir)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func chdir(path string) (err error) {
+	var _p0 *byte
+	_p0, err = BytePtrFromString(path)
+	if err != nil {
+		return
+	}
+	r0, _, e1 := Syscall(SYS_CHDIR, uintptr(unsafe.Pointer(_p0)), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Dup(oldfd int, newfd int) (fd int, err error) {
+	r0, _, e1 := Syscall(SYS_DUP, uintptr(oldfd), uintptr(newfd), 0)
+	fd = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Pread(fd int, p []byte, offset int64) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(p) > 0 {
+		_p0 = unsafe.Pointer(&p[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall6(SYS_PREAD, uintptr(fd), uintptr(_p0), uintptr(len(p)), uintptr(offset), uintptr(offset>>32), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Pwrite(fd int, p []byte, offset int64) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(p) > 0 {
+		_p0 = unsafe.Pointer(&p[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall6(SYS_PWRITE, uintptr(fd), uintptr(_p0), uintptr(len(p)), uintptr(offset), uintptr(offset>>32), 0)
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Close(fd int) (err error) {
+	r0, _, e1 := Syscall(SYS_CLOSE, uintptr(fd), 0, 0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Fstat(fd int, edir []byte) (n int, err error) {
+	var _p0 unsafe.Pointer
+	if len(edir) > 0 {
+		_p0 = unsafe.Pointer(&edir[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FSTAT, uintptr(fd), uintptr(_p0), uintptr(len(edir)))
+	n = int(r0)
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}
+
+// THIS FILE IS GENERATED BY THE COMMAND AT THE TOP; DO NOT EDIT
+
+func Fwstat(fd int, edir []byte) (err error) {
+	var _p0 unsafe.Pointer
+	if len(edir) > 0 {
+		_p0 = unsafe.Pointer(&edir[0])
+	} else {
+		_p0 = unsafe.Pointer(&_zero)
+	}
+	r0, _, e1 := Syscall(SYS_FWSTAT, uintptr(fd), uintptr(_p0), uintptr(len(edir)))
+	if int32(r0) == -1 {
+		err = e1
+	}
+	return
+}

Change information

Files:

• M src/cmd/dist/build.go
• M src/cmd/go/internal/work/exec.go
• M src/cmd/go/testdata/script/work_env.txt
• M src/cmd/internal/objfile/plan9obj.go
• M src/cmd/link/internal/arm64/obj.go
• M src/cmd/link/internal/ld/main.go
• M src/cmd/link/internal/ld/symtab.go
• A src/cmd/vendor/golang.org/x/sys/plan9/asm_plan9_arm64.s
• A src/cmd/vendor/golang.org/x/sys/plan9/zsyscall_plan9_arm64.go
• M src/debug/buildinfo/buildinfo.go
• M src/debug/plan9obj/file.go
• M src/debug/plan9obj/file_test.go
• M src/debug/plan9obj/plan9obj.go
• A src/debug/plan9obj/testdata/arm64-plan9-exec
• M src/internal/platform/zosarch.go
• M src/net/http/serve_test.go
• M src/net/udpsock_test.go
• A src/runtime/defs_plan9_arm64.go
• M src/runtime/os_plan9.go
• A src/runtime/os_plan9_386.go
• A src/runtime/os_plan9_amd64.go
• M src/runtime/os_plan9_arm.go
• A src/runtime/os_plan9_arm64.go
• A src/runtime/rt0_plan9_arm64.s
• A src/runtime/sys_plan9_arm64.s
• M src/runtime/tls_arm64.h
• A src/syscall/asm_plan9_arm64.s
• A src/syscall/zsyscall_plan9_arm64.go

Change size: XL

Delta: 28 files changed, 1319 insertions(+), 19 deletions(-)

Open in Gerrit

Related details

Attention set is empty

Submit Requirements:

• Code-Review
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newchange

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 1

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

[Gopher Robot (Gerrit)]

Gopher Robot added 1 comment

Patchset-level comments

File-level comment, Patchset 1 (Latest):

Gopher Robot . unresolved

I spotted some possible problems with your PR:

  1. The commit title should start with the primary affected package name followed by a colon, like "net/http: improve [...]".
  2. The first word in the commit title after the package should be a lowercase English word (usually a verb).
  3. You usually need to reference a bug number for all but trivial or cosmetic fixes. For this repo, the format is usually 'Fixes #12345' or 'Updates #12345' at the end of the commit message. Should you have a bug reference?

Please address any problems by updating the GitHub PR.

When complete, mark this comment as 'Done' and click the [blue 'Reply' button](https://go.dev/wiki/GerritBot#i-left-a-reply-to-a-comment-in-gerrit-but-no-one-but-me-can-see-it) above. These findings are based on heuristics; if a finding does not apply, briefly reply here saying so.

To update the commit title or commit message body shown here in Gerrit, you must edit the GitHub PR title and PR description (the first comment) in the GitHub web interface using the 'Edit' button or 'Edit' menu entry there. Note: pushing a new commit to the PR will not automatically update the commit message used by Gerrit.

For more details, see:

• [how to update commit messages](https://go.dev/wiki/GerritBot/#how-does-gerritbot-determine-the-final-commit-message) for PRs imported into Gerrit.
• the Go project's [conventions for commit messages](https://go.dev/doc/contribute#commit_messages) that you should follow.

(In general for Gerrit code reviews, the change author is expected to [log in to Gerrit](https://go-review.googlesource.com/login/) with a Gmail or other Google account and then close out each piece of feedback by marking it as 'Done' if implemented as suggested or otherwise reply to each review comment. See the [Review](https://go.dev/doc/contribute#review) section of the Contributing Guide for details.)

Open in Gerrit

Related details

Attention set is empty

Submit Requirements:

• Code-Review

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 1

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-Comment-Date: Wed, 12 Nov 2025 01:19:29 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Gopher Robot (Gerrit)]

Message from Gopher Robot

Congratulations on opening your first change. Thank you for your contribution!

Next steps:
A maintainer will review your change and provide feedback. See
https://go.dev/doc/contribute#review for more info and tips to get your
patch through code review.

Most changes in the Go project go through a few rounds of revision. This can be
surprising to people new to the project. The careful, iterative review process
is our way of helping mentor contributors and ensuring that their contributions
have a lasting impact.

During May-July and Nov-Jan the Go project is in a code freeze, during which
little code gets reviewed or merged. If a reviewer responds with a comment like
R=go1.11 or adds a tag like "wait-release", it means that this CL will be
reviewed as part of the next development cycle. See https://go.dev/s/release
for more details.

Open in Gerrit

Related details

Attention set is empty

Submit Requirements:

• Code-Review
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 1

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-Comment-Date: Wed, 12 Nov 2025 01:22:08 +0000

Gerrit-HasComments: No

Gerrit-Has-Labels: No

[Michael Pratt (Gerrit)]

Attention needed from Ian Lance Taylor and Keith Randall

Michael Pratt voted and added 1 comment

Votes added by Michael Pratt

Hold

+1

1 comment

Commit Message

Line 11, Patchset 1 (Latest):Previous try of including the plan9 arm64 port is here: https://github.com/golang/go/issues/57540

Michael Pratt . unresolved

Note that this proposal was declined. You will need to open a new proposal if you believe things have changed. It is nice that the total diff is quite small.

If you haven't seen it already, also take a look through https://go.dev/wiki/PortingPolicy for the general port requirements.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall

Submit Requirements:

• Code-Review

• No-Holds

• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 1

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Comment-Date: Wed, 12 Nov 2025 15:53:30 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: Yes

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #2 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall

• Michael Pratt

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 2

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Rafael Diniz added 1 comment

Commit Message

Line 11, Patchset 1:Previous try of including the plan9 arm64 port is here: https://github.com/golang/go/issues/57540

Michael Pratt . unresolved

Note that this proposal was declined. You will need to open a new proposal if you believe things have changed. It is nice that the total diff is quite small.

If you haven't seen it already, also take a look through https://go.dev/wiki/PortingPolicy for the general port requirements.

Rafael Diniz

Thanks Michael. I'll read.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 2

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Wed, 12 Nov 2025 20:50:51 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Michael Pratt <mpr...@google.com>

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Rafael Diniz added 1 comment

Commit Message

Line 11, Patchset 1:Previous try of including the plan9 arm64 port is here: https://github.com/golang/go/issues/57540

Michael Pratt . unresolved

Note that this proposal was declined. You will need to open a new proposal if you believe things have changed. It is nice that the total diff is quite small.

If you haven't seen it already, also take a look through https://go.dev/wiki/PortingPolicy for the general port requirements.

Rafael Diniz

Thanks Michael. I'll read.

Rafael Diniz

Just to clarify, I opened a proposal, which was promptly rejected:

https://github.com/golang/go/issues/76272

Should I open a new proposal or ask to re-open the one I opened, in order a discussion can actually happen?

Gerrit-Comment-Date: Wed, 12 Nov 2025 20:57:13 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Michael Pratt <mpr...@google.com>

Comment-In-Reply-To: Rafael Diniz <rafae...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #3 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 3

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 3 (Latest):

Richard Miller . unresolved

I think you have an ABI clash between the go arm64 compiler and the 9front syscall interface regarding alignment of arguments. For example `dupfd` has two int32 arguments. The go compiler packs these together onto the stack with 4-byte alignment, but the 9front syscall entry expects them each aligned to 8 bytes. Therefore the `dupfd(bintimeFD, 18)` call from `osinit` creates the wrong file descriptor, causing the `readtime` function to fail and prevent the go runtime from initialising.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 3

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Fri, 09 Jan 2026 13:23:30 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 3 (Latest):

Richard Miller . unresolved

I think you have an ABI clash between the go arm64 compiler and the 9front syscall interface regarding alignment of arguments. For example `dupfd` has two int32 arguments. The go compiler packs these together onto the stack with 4-byte alignment, but the 9front syscall entry expects them each aligned to 8 bytes. Therefore the `dupfd(bintimeFD, 18)` call from `osinit` creates the wrong file descriptor, causing the `readtime` function to fail and prevent the go runtime from initialising.

Richard Miller

Note that this comment applies specifically to the functions in `runtime/sys_plan9_arm64.s`.

Gerrit-Comment-Date: Fri, 09 Jan 2026 20:25:31 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall and Michael Pratt

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #4 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 4

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Rafael Diniz added 3 comments

Patchset-level comments

File-level comment, Patchset 1:

Gopher Robot . unresolved

I spotted some possible problems with your PR:

  1. The commit title should start with the primary affected package name followed by a colon, like "net/http: improve [...]".
  2. The first word in the commit title after the package should be a lowercase English word (usually a verb).
  3. You usually need to reference a bug number for all but trivial or cosmetic fixes. For this repo, the format is usually 'Fixes #12345' or 'Updates #12345' at the end of the commit message. Should you have a bug reference?

Please address any problems by updating the GitHub PR.

When complete, mark this comment as 'Done' and click the [blue 'Reply' button](https://go.dev/wiki/GerritBot#i-left-a-reply-to-a-comment-in-gerrit-but-no-one-but-me-can-see-it) above. These findings are based on heuristics; if a finding does not apply, briefly reply here saying so.

To update the commit title or commit message body shown here in Gerrit, you must edit the GitHub PR title and PR description (the first comment) in the GitHub web interface using the 'Edit' button or 'Edit' menu entry there. Note: pushing a new commit to the PR will not automatically update the commit message used by Gerrit.

For more details, see:

• [how to update commit messages](https://go.dev/wiki/GerritBot/#how-does-gerritbot-determine-the-final-commit-message) for PRs imported into Gerrit.
• the Go project's [conventions for commit messages](https://go.dev/doc/contribute#commit_messages) that you should follow.

(In general for Gerrit code reviews, the change author is expected to [log in to Gerrit](https://go-review.googlesource.com/login/) with a Gmail or other Google account and then close out each piece of feedback by marking it as 'Done' if implemented as suggested or otherwise reply to each review comment. See the [Review](https://go.dev/doc/contribute#review) section of the Contributing Guide for details.)

Rafael Diniz

Should I squash the commits to a single commit and or keep this MR with fine-grained commits, one commit per package (more or less how it is right now)? I'll edit the commit messages in any case.

File-level comment, Patchset 3:

Richard Miller . resolved

I think you have an ABI clash between the go arm64 compiler and the 9front syscall interface regarding alignment of arguments. For example `dupfd` has two int32 arguments. The go compiler packs these together onto the stack with 4-byte alignment, but the 9front syscall entry expects them each aligned to 8 bytes. Therefore the `dupfd(bintimeFD, 18)` call from `osinit` creates the wrong file descriptor, causing the `readtime` function to fail and prevent the go runtime from initialising.

Richard Miller

Note that this comment applies specifically to the functions in `runtime/sys_plan9_arm64.s`.

Rafael Diniz

Thanks for finding looking at this Richard! I'm working on it.

Commit Message

Line 11, Patchset 1:Previous try of including the plan9 arm64 port is here: https://github.com/golang/go/issues/57540

Michael Pratt . resolved

Note that this proposal was declined. You will need to open a new proposal if you believe things have changed. It is nice that the total diff is quite small.

If you haven't seen it already, also take a look through https://go.dev/wiki/PortingPolicy for the general port requirements.

Rafael Diniz

Thanks Michael. I'll read.

Rafael Diniz

Just to clarify, I opened a proposal, which was promptly rejected:

https://github.com/golang/go/issues/76272

Should I open a new proposal or ask to re-open the one I opened, in order a discussion can actually happen?

Rafael Diniz

Acknowledged

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 4

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Tue, 13 Jan 2026 19:07:33 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Gopher Robot <go...@golang.org>

Comment-In-Reply-To: Michael Pratt <mpr...@google.com>

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

Comment-In-Reply-To: Rafael Diniz <rafae...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #5 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 5

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Rafael Diniz added 1 comment

Patchset-level comments

File-level comment, Patchset 1:

Gopher Robot . resolved

I spotted some possible problems with your PR:

  1. The commit title should start with the primary affected package name followed by a colon, like "net/http: improve [...]".
  2. The first word in the commit title after the package should be a lowercase English word (usually a verb).
  3. You usually need to reference a bug number for all but trivial or cosmetic fixes. For this repo, the format is usually 'Fixes #12345' or 'Updates #12345' at the end of the commit message. Should you have a bug reference?

Please address any problems by updating the GitHub PR.

When complete, mark this comment as 'Done' and click the [blue 'Reply' button](https://go.dev/wiki/GerritBot#i-left-a-reply-to-a-comment-in-gerrit-but-no-one-but-me-can-see-it) above. These findings are based on heuristics; if a finding does not apply, briefly reply here saying so.

To update the commit title or commit message body shown here in Gerrit, you must edit the GitHub PR title and PR description (the first comment) in the GitHub web interface using the 'Edit' button or 'Edit' menu entry there. Note: pushing a new commit to the PR will not automatically update the commit message used by Gerrit.

For more details, see:

• [how to update commit messages](https://go.dev/wiki/GerritBot/#how-does-gerritbot-determine-the-final-commit-message) for PRs imported into Gerrit.
• the Go project's [conventions for commit messages](https://go.dev/doc/contribute#commit_messages) that you should follow.

(In general for Gerrit code reviews, the change author is expected to [log in to Gerrit](https://go-review.googlesource.com/login/) with a Gmail or other Google account and then close out each piece of feedback by marking it as 'Done' if implemented as suggested or otherwise reply to each review comment. See the [Review](https://go.dev/doc/contribute#review) section of the Contributing Guide for details.)

Rafael Diniz

Should I squash the commits to a single commit and or keep this MR with fine-grained commits, one commit per package (more or less how it is right now)? I'll edit the commit messages in any case.

Rafael Diniz

Acknowledged

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 5

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Mon, 26 Jan 2026 19:34:21 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Gopher Robot <go...@golang.org>

Comment-In-Reply-To: Rafael Diniz <rafae...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #6 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 6

[t hepudds (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

t hepudds added 1 comment

Patchset-level comments

File-level comment, Patchset 1:

Gopher Robot . unresolved

I spotted some possible problems with your PR:

  1. The commit title should start with the primary affected package name followed by a colon, like "net/http: improve [...]".
  2. The first word in the commit title after the package should be a lowercase English word (usually a verb).
  3. You usually need to reference a bug number for all but trivial or cosmetic fixes. For this repo, the format is usually 'Fixes #12345' or 'Updates #12345' at the end of the commit message. Should you have a bug reference?

Please address any problems by updating the GitHub PR.

When complete, mark this comment as 'Done' and click the [blue 'Reply' button](https://go.dev/wiki/GerritBot#i-left-a-reply-to-a-comment-in-gerrit-but-no-one-but-me-can-see-it) above. These findings are based on heuristics; if a finding does not apply, briefly reply here saying so.

To update the commit title or commit message body shown here in Gerrit, you must edit the GitHub PR title and PR description (the first comment) in the GitHub web interface using the 'Edit' button or 'Edit' menu entry there. Note: pushing a new commit to the PR will not automatically update the commit message used by Gerrit.

For more details, see:

• [how to update commit messages](https://go.dev/wiki/GerritBot/#how-does-gerritbot-determine-the-final-commit-message) for PRs imported into Gerrit.
• the Go project's [conventions for commit messages](https://go.dev/doc/contribute#commit_messages) that you should follow.

(In general for Gerrit code reviews, the change author is expected to [log in to Gerrit](https://go-review.googlesource.com/login/) with a Gmail or other Google account and then close out each piece of feedback by marking it as 'Done' if implemented as suggested or otherwise reply to each review comment. See the [Review](https://go.dev/doc/contribute#review) section of the Contributing Guide for details.)

Rafael Diniz

Should I squash the commits to a single commit and or keep this MR with fine-grained commits, one commit per package (more or less how it is right now)? I'll edit the commit messages in any case.

Rafael Diniz

Acknowledged

t hepudds

I happen to notice it looked like you pushed a new commit with a cleaned-up commit message to the GitHub PR. It looks like the commit message here in Gerrit is not yet updated.

In case it helps, here is a link to a reminder on how to update the Gerrit commit message for PRs imported into Gerrit:
https://go.dev/wiki/GerritBot/#how-does-gerritbot-determine-the-final-commit-message

Note that just pushing a new commit to the GitHub PR does not update the Gerrit commit message here.

Sorry, it's not quite a standard GitHub PR workflow.

After making changes in the GitHub web UI, be sure to then double-check the plain text shown in the Gerrit commit message (above) after the GitHub PR is imported into Gerrit to confirm you see your expected changes here.

Should I squash the commits to a single commit and or keep this MR with fine-grained commits [...]

You can upload as many commits to the GitHub PR as you like. GerritBot will handle squashing your commits into one change that Gerrit can handle.

There are some more details here:

• https://go.dev/doc/contribute#sending_a_change_github
• https://go.dev/wiki/GerritBot#frequently-asked-questions

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 6

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Tue, 27 Jan 2026 17:50:27 +0000

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #7 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 7

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #8 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 8

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and Richard Miller

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #9 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 9

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Richard Miller and t hepudds

Rafael Diniz added 1 comment

Patchset-level comments

File-level comment, Patchset 1:

Gopher Robot . resolved

Rafael Diniz

Thanks a lot for all the references. Already updated the Gerrit commit message.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller

• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 8

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Sat, 31 Jan 2026 22:28:44 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: t hepudds <thepud...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 9 (Latest):

Richard Miller . unresolved

There's still a problem with `runtime/sys_plan9_arm64.s` with rather amusing consequences:
```
cpu% ls -l lib/cache/go-build/f?
--rw-rw-rw- M 66 miller miller 10 Sep 28  1906 lib/cache/go-build/f0/f0ee9d7b8fbb0afac809cb6c46cee7fd3a68befdf643ccdd00533820a8926177-d
--rw-rw-rw- M 66 miller miller 175 Jan 21  1969 lib/cache/go-build/f1/f14b3be5d7964adcd8b70771002d2c7131550f47ff5b3acc36394514ecef55b8-a
--rw-rw-rw- M 66 miller miller  21 Aug 19  1920 lib/cache/go-build/f5/f502da81ad708823cb5ea6c61b7f07f3b49cd9d264dff170551d11c9c7f2f4bb-d
--rw-rw-rw- M 66 miller miller 175 Sep 19  1935 lib/cache/go-build/f7/f78fa6e66dfe2f70dd6f903d19bd2cd8d67611422a009f1290830eefb4c9cec9-a
--rw-rw-rw- M 66 miller miller 8874542 Sep  5  1906 lib/cache/go-build/fb/fbbd15b647dbc50179a6bea2abd6aa099a209a7a57ba38c812cb43eea2dc2477-d
```
The return argument offsets in `runtime/timesplit` are wrong, as you will see if you run `go vet`:
```
term%  go vet runtime
# runtime
# [runtime]
/home/miller/go-plan9-arm64/src/runtime/sys_plan9_arm64.s:186:1: [arm64] timesplit: invalid offset sec+0(FP); expected sec+8(FP)
/home/miller/go-plan9-arm64/src/runtime/sys_plan9_arm64.s:187:1: [arm64] timesplit: invalid offset nsec+8(FP); expected nsec+16(FP)
/home/miller/go-plan9-arm64/src/runtime/sys_plan9_arm64.s:198:1: [arm64] walltime: function walltime missing Go declaration
```
But fixing that will not be sufficient to make `time.Now` for plan9-arm64 correct, because the logic of the `timesplit` function is also not right. Anyway on 64-bit ARM, there's no need for the clever gyrations used on 32-bit ARM to simulate divide and remainder. I suggest this simpler version for arm64:
```

// func timesplit(u uint64) (sec int64, nsec int32)

TEXT runtime·timesplit(SB), NOSPLIT, $0-20
	MOVD	u+0(FP), R0
	MOVD	R0, R1
	MOVD	$1000000000, R2
	UDIV	R2, R1
	MUL	R1, R2
	SUB	R2, R0
	MOVD	R1,sec+8(FP)
	MOVWU	R0,nsec+16(FP)
	RET
```

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 9

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Sun, 01 Feb 2026 13:45:27 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 9 (Latest):

Richard Miller . unresolved

There seem to be other intermittent problems, as I found on another attempt to run `go vet`:
```
term% go vet -asmdecl runtime
go-plan9-arm64/src/crypto/internal/sysrand/rand.go:10:2: open /home/miller/go-plan9-arm64/src/os/types_windows.go: bad arg in system call
term% go vet -asmdecl runtime
# net [runtime.test]
fatal error: unexpected signal during runtime execution
[signal sys: trap: fault read code=0x0 addr=0x0 pc=0x84e5c]

runtime stack:
runtime.throw({0xa1c8f8?, 0x7f514?})
	go-plan9-arm64/src/runtime/panic.go:1229 +0x38 fp=0x1fffeeab8 sp=0x1fffeea88 pc=0x95ef8
runtime.sigpanic()
	go-plan9-arm64/src/runtime/os_plan9.go:78 +0x3a8 fp=0x1fffeeb08 sp=0x1fffeeab8 pc=0x5dba8
runtime.(*unwinder).resolveInternal(0x9d120?, 0x7c?, 0xe6?)
	go-plan9-arm64/src/runtime/traceback.go:258 +0x1c fp=0x1fffeeb98 sp=0x1fffeeb18 pc=0x84e5c
runtime.(*unwinder).initAt(0x1fffeecc0?, 0x2cbbc?, 0x1fffeecc0?, 0x6fc5c?, 0x48058808?, 0xa0?)
	go-plan9-arm64/src/runtime/traceback.go:225 +0x1cc fp=0x1fffeec88 sp=0x1fffeeb98 pc=0x84c1c
runtime.(*unwinder).init(...)
	go-plan9-arm64/src/runtime/traceback.go:130
runtime.copystack(0x48488960, 0x2000)
	go-plan9-arm64/src/runtime/stack.go:976 +0x46c fp=0x1fffeee48 sp=0x1fffeec88 pc=0x7a18c
runtime.newstack()
	go-plan9-arm64/src/runtime/stack.go:1187 +0x62c fp=0x1fffeef68 sp=0x1fffeee48 pc=0x7afec
runtime.morestack()
	go-plan9-arm64/src/runtime/asm_arm64.s:507 +0x70 fp=0x1fffeef68 sp=0x1fffeef68 pc=0x9b250

...

# go/build
# [go/build]
panic: assertion failed [recovered, repanicked]

goroutine 1 [running]:
go/types.(*Checker).handleBailout(0x4829f000, 0x486c9ad8)
	go-plan9-arm64/src/go/types/check.go:404 +0x94
panic({0x68b760?, 0x341f00?})
	go-plan9-arm64/src/runtime/panic.go:860 +0x12c
internal/pkgbits.assert(...)
	go-plan9-arm64/src/internal/pkgbits/support.go:11
internal/pkgbits.(*Decoder).Bool(0x486c5680)
	go-plan9-arm64/src/internal/pkgbits/decoder.go:365 +0xdc
go/internal/gcimporter.(*reader).typInfo(0x486c5680)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:254 +0x30
go/internal/gcimporter.(*pkgReader).objDictIdx(0x480d4c60, 0xb)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:574 +0x148
go/internal/gcimporter.(*pkgReader).objIdx(0x480d4c60, 0xb)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:477 +0x23c
go/internal/gcimporter.(*reader).obj(0x486c5bf8)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:440 +0x80
go/internal/gcimporter.(*reader).doTyp(0x486c5bf8)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:302 +0x80
go/internal/gcimporter.(*pkgReader).typIdx(0x480d4c60, {0x1f?, 0x0?}, 0x48246de0)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:279 +0x10c
go/internal/gcimporter.(*reader).typ(0x486c5ee8)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:249 +0x40
go/internal/gcimporter.(*reader).structType(0x486c5ee8)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:342 +0xbc
go/internal/gcimporter.(*reader).doTyp(0x486c5ee8)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:328 +0x33c
go/internal/gcimporter.(*pkgReader).typIdx(0x480d4c60, {0x5?, 0x0?}, 0x48246de0)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:279 +0x10c
go/internal/gcimporter.(*reader).typ(0x486c61d0)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:249 +0x40
go/internal/gcimporter.(*pkgReader).objIdx(0x480d4c60, 0x7)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:520 +0x3c8
go/internal/gcimporter.(*reader).obj(0x486c6488)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:440 +0x80
go/internal/gcimporter.(*reader).doTyp(0x486c6488)
	go-plan9-arm64/src/go/internal/gcimporter/ureader.go:302 +0x80

... etc

```
I wonder how thoroughly you have tested this port? Getting it to compile cleanly is only the first step.

Gerrit-Comment-Date: Sun, 01 Feb 2026 13:56:20 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Rafael Diniz added 2 comments

Patchset-level comments

Rafael Diniz

I'll investigate, my bad.

File-level comment, Patchset 9 (Latest):

Richard Miller . unresolved

There's still a problem with `runtime/sys_plan9_arm64.s` with rather amusing consequences:

Rafael Diniz

Changed. Thanks Dr. Miller!

Gerrit-Comment-Date: Mon, 02 Feb 2026 16:15:50 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #10 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 10

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #11 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 11

[SLSA Policy Verification Service (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

SLSA Policy Verification Service voted SLSA-Policy-Verified+1

SLSA-Policy-Verified

+1

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 11

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Fri, 15 May 2026 09:26:57 +0000

Gerrit-HasComments: No

Gerrit-Has-Labels: Yes

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Richard Miller and t hepudds

Rafael Diniz added 3 comments

Patchset-level comments

File-level comment, Patchset 9:

Richard Miller . resolved

Rafael Diniz

I should be fixed in Patchset 11.

File-level comment, Patchset 9:

Richard Miller . resolved

Rafael Diniz

go vet runtime now runs.

File-level comment, Patchset 11 (Latest):

Rafael Diniz . resolved

Some improvements of the arm64 Plan9 port.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• Richard Miller
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 11

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Fri, 15 May 2026 09:32:51 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

Comment-In-Reply-To: Rafael Diniz <rafae...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 11 (Latest):

Richard Miller . unresolved

Are the tests all running successfully for you? I'm seeing quite a few failures running `go tool dist test` on 9legacy:
```
FAIL compress/flate 9.490s
FAIL crypto/internal/fips140test 22.370s
FAIL go/internal/gcimporter 54.334s
FAIL internal/abi 2.932s
FAIL internal/godebugs 24.807s
FAIL internal/trace 58.438s
FAIL net/http 905.455s
FAIL net/http/httputil 900.798s
FAIL net/http/internal/http2 36.653s
FAIL runtime 149.852s
FAIL strconv 0.476s
FAIL testing 10.889s
FAIL cmd/addr2line 9.112s
FAIL cmd/compile/internal/importer 66.427s
FAIL cmd/compile/internal/inline/inlheur 11.493s
FAIL cmd/compile/internal/ssa 8.931s
FAIL cmd/compile/internal/test 115.702s
FAIL cmd/compile/internal/types2 179.038s
FAIL cmd/cover 39.653s
FAIL cmd/internal/archive 5.731s
FAIL cmd/internal/moddeps 76.040s
FAIL cmd/internal/obj 5.400s
FAIL cmd/internal/testdir 17.440s
FAIL cmd/link 67.860s
FAIL cmd/link/internal/ld 33.456s
FAIL cmd/pack 13.090s
```

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 11

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Fri, 15 May 2026 18:15:41 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #12 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 12

[SLSA Policy Verification Service (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

SLSA Policy Verification Service voted SLSA-Policy-Verified+1

SLSA-Policy-Verified

+1

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 12

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Fri, 15 May 2026 21:11:59 +0000

Gerrit-HasComments: No

Gerrit-Has-Labels: Yes

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Rafael Diniz added 1 comment

Patchset-level comments

File-level comment, Patchset 11:

Rafael Diniz

Thanks Dr. Miller. I just did a commit (Patchset 12) which fixes some issues (and add some workarounds for some tests so they pass at this point). I finally have a setup with qemu-aarch64 so I can test the changes faster.

Gerrit-Comment-Date: Sat, 16 May 2026 08:39:50 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 11:

Richard Miller

Still lots of errors. I suggest you submit a patch when you have ALL tests passing on real hardware. (Qemu is unlikely to be duplicating multicore behaviour accurately enough to pick up concurrency bugs.)

Gerrit-Comment-Date: Sat, 16 May 2026 09:36:56 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

Comment-In-Reply-To: Rafael Diniz <rafae...@gmail.com>

[SLSA Policy Verification Service (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

SLSA Policy Verification Service voted SLSA-Policy-Verified+1

SLSA-Policy-Verified

+1

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 13

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Sun, 17 May 2026 19:46:32 +0000

Gerrit-HasComments: No

Gerrit-Has-Labels: Yes

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #13 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Rafael Diniz added 2 comments

Patchset-level comments

File-level comment, Patchset 11:

Rafael Diniz

I managed to pass now many of the tests (go test dist), but there are some errors with networking, and some related to memory:

https://github.com/rafael2knokia/go/blob/plan9-arm64-dev/misc/plan9/arm64/dist-test-stdout-20260518.txt

If it is ok, I might submit a couple more patchsets, as I want to address the remaining issues separately. Then I'll fire up the Pi 4 to run the tests after the non-multicore-related errors are fixed.

File-level comment, Patchset 13 (Latest):

Rafael Diniz . unresolved

go tool dist test -run=^archive/tar$
ok  	archive/tar	4.658s
ok  	archive/zip	1.322s
ok  	bufio	0.666s
ok  	bytes	1.092s
ok  	cmp	0.217s
ok  	compress/bzip2	0.539s
ok  	compress/flate	3.946s
ok  	compress/gzip	8.524s
ok  	compress/lzw	0.401s
ok  	compress/zlib	0.508s
ok  	container/heap	0.359s
ok  	container/list	0.325s
ok  	container/ring	0.315s
ok  	context	1.789s
ok  	crypto	31.565s
ok  	crypto/aes	0.588s
ok  	crypto/cipher	16.763s
ok  	crypto/des	0.381s
ok  	crypto/dsa	0.279s
ok  	crypto/ecdh	0.601s
ok  	crypto/ecdsa	0.793s
ok  	crypto/ed25519	0.709s
ok  	crypto/elliptic	0.405s
ok  	crypto/fips140	0.492s
ok  	crypto/hkdf	0.345s
ok  	crypto/hmac	0.538s
ok  	crypto/hpke	8.548s
ok  	crypto/internal/boring	0.300s
?   	crypto/internal/boring/bbig	[no test files]
ok  	crypto/internal/boring/bcache	0.886s
?   	crypto/internal/boring/sig	[no test files]
?   	crypto/internal/constanttime	[no test files]
?   	crypto/internal/cryptotest	[no test files]
?   	crypto/internal/entropy	[no test files]
?   	crypto/internal/entropy/v1.0.0	[no test files]
ok  	crypto/internal/fips140	0.284s
ok  	crypto/internal/fips140/aes	0.368s
ok  	crypto/internal/fips140/aes/gcm	0.249s [no tests to run]
?   	crypto/internal/fips140/alias	[no test files]
ok  	crypto/internal/fips140/bigmod	0.542s
?   	crypto/internal/fips140/check	[no test files]
?   	crypto/internal/fips140/check/checktest	[no test files]
ok  	crypto/internal/fips140/drbg	0.235s [no tests to run]
ok  	crypto/internal/fips140/ecdh	0.302s
ok  	crypto/internal/fips140/ecdsa	0.440s
?   	crypto/internal/fips140/ed25519	[no test files]
ok  	crypto/internal/fips140/edwards25519	0.644s
ok  	crypto/internal/fips140/edwards25519/field	0.370s
?   	crypto/internal/fips140/hkdf	[no test files]
?   	crypto/internal/fips140/hmac	[no test files]
ok  	crypto/internal/fips140/mldsa	3.237s
ok  	crypto/internal/fips140/mlkem	0.804s
ok  	crypto/internal/fips140/nistec	0.291s [no tests to run]
ok  	crypto/internal/fips140/nistec/fiat	0.258s [no tests to run]
?   	crypto/internal/fips140/pbkdf2	[no test files]
ok  	crypto/internal/fips140/rsa	0.632s
?   	crypto/internal/fips140/sha256	[no test files]
?   	crypto/internal/fips140/sha3	[no test files]
?   	crypto/internal/fips140/sha512	[no test files]
?   	crypto/internal/fips140/ssh	[no test files]
ok  	crypto/internal/fips140/subtle	0.301s
?   	crypto/internal/fips140/tls12	[no test files]
?   	crypto/internal/fips140/tls13	[no test files]
ok  	crypto/internal/fips140cache	0.440s
ok  	crypto/internal/fips140deps	3.870s
?   	crypto/internal/fips140deps/byteorder	[no test files]
?   	crypto/internal/fips140deps/cpu	[no test files]
?   	crypto/internal/fips140deps/godebug	[no test files]
?   	crypto/internal/fips140deps/time	[no test files]
?   	crypto/internal/fips140hash	[no test files]
ok  	crypto/internal/fips140only	3.510s
ok  	crypto/internal/fips140test	24.638s
?   	crypto/internal/impl	[no test files]
?   	crypto/internal/rand	[no test files]
?   	crypto/internal/randutil	[no test files]
ok  	crypto/internal/sysrand	0.540s
?   	crypto/internal/sysrand/internal/seccomp	[no test files]
ok  	crypto/md5	0.550s
ok  	crypto/mlkem	0.427s
?   	crypto/mlkem/mlkemtest	[no test files]
ok  	crypto/pbkdf2	0.330s
ok  	crypto/rand	1.090s
ok  	crypto/rc4	0.401s
ok  	crypto/rsa	1.855s
ok  	crypto/sha1	0.472s
ok  	crypto/sha256	0.492s
ok  	crypto/sha3	3.976s
ok  	crypto/sha512	0.502s
ok  	crypto/subtle	0.671s
--- FAIL: TestGetClientCertificate (0.18s)
    --- FAIL: TestGetClientCertificate/TLSv12 (0.09s)
        handshake_client_test.go:2472: #1: client error: EOF
    --- FAIL: TestGetClientCertificate/TLSv13 (0.09s)
        handshake_client_test.go:2474: #2: expected client error "GetClientCertificate", but got "EOF"
--- FAIL: TestTLSUniqueMatches (0.02s)
    tls_test.go:504: write tcp 127.0.0.1:53628->127.0.0.1:36547: write /net/tcp/4/data: i/o on hungup channel
    tls_test.go:519: EOF
FAIL
FAIL	crypto/tls	11.553s
?   	crypto/tls/internal/fips140tls	[no test files]
ok  	crypto/x509	26.803s
?   	crypto/x509/pkix	[no test files]
ok  	database/sql	1.414s
ok  	database/sql/driver	0.273s
ok  	debug/buildinfo	1.033s
ok  	debug/dwarf	1.222s
ok  	debug/elf	1.307s
ok  	debug/gosym	0.603s
ok  	debug/macho	0.398s
ok  	debug/pe	1.612s
ok  	debug/plan9obj	0.302s
ok  	embed	0.375s [no tests to run]
ok  	embed/internal/embedtest	0.237s
?   	encoding	[no test files]
ok  	encoding/ascii85	0.197s
ok  	encoding/asn1	0.448s
ok  	encoding/base32	0.421s
ok  	encoding/base64	0.365s
ok  	encoding/binary	0.388s
ok  	encoding/csv	0.439s
ok  	encoding/gob	20.558s
ok  	encoding/hex	0.300s
ok  	encoding/json	2.197s
ok  	encoding/pem	1.318s
ok  	encoding/xml	0.626s
ok  	errors	0.252s
ok  	expvar	0.517s
ok  	flag	0.599s
ok  	fmt	0.893s
ok  	go/ast	0.557s
ok  	go/build	77.374s
ok  	go/build/constraint	0.280s
ok  	go/constant	0.406s
ok  	go/doc	1.368s
ok  	go/doc/comment	23.407s
ok  	go/format	0.330s
ok  	go/importer	10.624s
ok  	go/internal/gccgoimporter	0.758s
ok  	go/internal/gcimporter	75.176s
ok  	go/internal/srcimporter	30.868s
ok  	go/parser	1.475s
ok  	go/printer	1.914s
ok  	go/scanner	0.388s
ok  	go/token	1.147s
ok  	go/types	204.694s
ok  	go/version	0.318s
ok  	hash	0.357s
ok  	hash/adler32	0.273s
ok  	hash/crc32	0.356s
ok  	hash/crc64	0.333s
ok  	hash/fnv	0.312s
ok  	hash/maphash	1.186s
ok  	html	0.415s
ok  	html/template	1.491s
ok  	image	0.795s
ok  	image/color	0.260s
?   	image/color/palette	[no test files]
ok  	image/draw	0.855s
ok  	image/gif	0.996s
?   	image/internal/imageutil	[no test files]
ok  	image/jpeg	8.162s
ok  	image/png	1.920s
ok  	index/suffixarray	1.091s
ok  	internal/abi	1.671s
?   	internal/asan	[no test files]
?   	internal/bisect	[no test files]
ok  	internal/buildcfg	0.245s
?   	internal/bytealg	[no test files]
?   	internal/byteorder	[no test files]
?   	internal/cfg	[no test files]
ok  	internal/chacha8rand	0.396s
ok  	internal/copyright	17.321s
?   	internal/coverage	[no test files]
?   	internal/coverage/calloc	[no test files]
ok  	internal/coverage/cfile	25.225s
ok  	internal/coverage/cformat	0.326s
ok  	internal/coverage/cmerge	0.211s
?   	internal/coverage/decodecounter	[no test files]
?   	internal/coverage/decodemeta	[no test files]
?   	internal/coverage/encodecounter	[no test files]
?   	internal/coverage/encodemeta	[no test files]
ok  	internal/coverage/pods	0.438s
?   	internal/coverage/rtcov	[no test files]
ok  	internal/coverage/slicereader	0.297s
ok  	internal/coverage/slicewriter	0.308s
?   	internal/coverage/stringtab	[no test files]
ok  	internal/coverage/test	0.477s
?   	internal/coverage/uleb128	[no test files]
ok  	internal/cpu	0.466s
ok  	internal/dag	0.365s
ok  	internal/diff	0.376s
?   	internal/exportdata	[no test files]
?   	internal/filepathlite	[no test files]
ok  	internal/fmtsort	0.476s
ok  	internal/fuzz	0.454s
ok  	internal/gate	0.360s
?   	internal/goarch	[no test files]
ok  	internal/godebug	9.549s
ok  	internal/godebugs	0.333s
?   	internal/goexperiment	[no test files]
?   	internal/goos	[no test files]
?   	internal/goroot	[no test files]
ok  	internal/gover	0.320s
?   	internal/goversion	[no test files]
?   	internal/lazyregexp	[no test files]
?   	internal/lazytemplate	[no test files]
?   	internal/msan	[no test files]
?   	internal/nettrace	[no test files]
?   	internal/obscuretestdata	[no test files]
?   	internal/oserror	[no test files]
ok  	internal/pkgbits	0.204s
ok  	internal/platform	14.850s
ok  	internal/poll	0.534s
ok  	internal/profile	0.371s
?   	internal/profilerecord	[no test files]
?   	internal/race	[no test files]
ok  	internal/reflectlite	0.550s
ok  	internal/runtime/atomic	0.446s
ok  	internal/runtime/cgroup	0.583s
?   	internal/runtime/exithook	[no test files]
?   	internal/runtime/gc	[no test files]
?   	internal/runtime/gc/internal/gen	[no test files]
ok  	internal/runtime/gc/scan	0.213s
ok  	internal/runtime/maps	2.606s
ok  	internal/runtime/math	0.260s
?   	internal/runtime/pprof/label	[no test files]
ok  	internal/runtime/sys	0.300s
ok  	internal/runtime/wasitest	0.405s
ok  	internal/saferio	0.497s
ok  	internal/singleflight	0.531s
ok  	internal/strconv	1.217s
?   	internal/stringslite	[no test files]
ok  	internal/sync	3.120s
ok  	internal/synctest	9.001s
?   	internal/syscall/execenv	[no test files]
ok  	internal/syscall/unix	0.251s
ok  	internal/sysinfo	0.409s
?   	internal/syslist	[no test files]
ok  	internal/testenv	5.627s
?   	internal/testhash	[no test files]
?   	internal/testlog	[no test files]
?   	internal/testpty	[no test files]
ok  	internal/trace	288.211s
?   	internal/trace/internal/testgen	[no test files]
ok  	internal/trace/internal/tracev1	1.065s
?   	internal/trace/raw	[no test files]
ok  	internal/trace/testtrace	0.621s
ok  	internal/trace/tracev2	0.676s
?   	internal/trace/traceviewer	[no test files]
?   	internal/trace/traceviewer/format	[no test files]
?   	internal/trace/version	[no test files]
?   	internal/txtar	[no test files]
ok  	internal/types/errors	25.771s
ok  	internal/unsafeheader	0.557s
ok  	internal/xcoff	1.337s
ok  	internal/zstd	1.314s
ok  	io	1.251s
ok  	io/fs	2.026s
ok  	io/ioutil	18.554s
ok  	iter	0.712s
ok  	log	0.474s
?   	log/internal	[no test files]
ok  	log/slog	1.218s
?   	log/slog/internal	[no test files]
ok  	log/slog/internal/benchmarks	0.693s
ok  	log/slog/internal/buffer	0.468s
?   	log/syslog	[no test files]
ok  	maps	0.610s
ok  	math	0.705s
ok  	math/big	45.741s
ok  	math/big/internal/asmgen	0.964s
ok  	math/bits	1.333s
ok  	math/cmplx	0.531s
ok  	math/rand	1.926s
ok  	math/rand/v2	2.166s
ok  	mime	0.631s
ok  	mime/multipart	10.931s
ok  	mime/quotedprintable	0.973s
ok  	net	9.976s
--- FAIL: TestAutomaticHTTP2_ListenAndServe_GetCertificate (0.04s)
(...)

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 13

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Rafael Diniz <rafae...@gmail.com>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Mon, 18 May 2026 07:59:35 +0000

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Richard Miller added 10 comments

Patchset-level comments

File-level comment, Patchset 13 (Latest):

Richard Miller . unresolved

Most of the changes in this patch set appear to be related to running a builder on 9front instead of Plan 9. The patches are disabling tests which have run successfully for years on Plan 9 and 9legacy builders. 9front is a fork of Plan 9 which seems to have diverged so far as to be unable to run normal go tests. The title of this CL is "add plan9/arm64 port". I suggest sticking the to goal of passing all tests on Plan 9 with arm64. Porting to 9front can be a separate project.

File src/crypto/x509/root_plan9.go

Line 38, Patchset 13 (Latest): if os.IsNotExist(bestErr) {

Richard Miller . unresolved

Returning an error was the correct thing to do. If the cert file doesn't exist, it's not right to pretend it does. If you apply 9legacy patch ca.diff on the builder, the file will be there as needed.

File src/net/lookup_test.go

Line 869, Patchset 13 (Latest): if !strings.HasSuffix(err.Error(), errNoSuchHost.Error()) &&

Richard Miller . unresolved

This defeats the purpose of the test. Plan 9 correctly returns a NoSuchHost error. If 9front is returning something else, it's the 9front implementation which needs to be fixed.

Line 873, Patchset 13 (Latest): if !err.(*DNSError).IsNotFound &&

Richard Miller . unresolved

As above.

File src/net/timeout_test.go

Line 956, Patchset 13 (Latest): if runtime.GOOS == "plan9" {

Richard Miller . unresolved

This test is supported on Plan 9. Considerable work has been done (eg CL 420715) to ensure that it passes. If it breaks on a 9front builder, fix the 9front problem, don't disable the test.

File src/net/writev_test.go

Line 186, Patchset 13 (Latest): if runtime.GOOS == "plan9" {

Richard Miller . unresolved

This test passes on Plan 9; it's a 9front problem.

File src/runtime/goroutineleakprofile_test.go

Line 19, Patchset 13 (Latest): if testing.Short() && goos.IsPlan9 != 0 && goarch.IsArm64 != 0 {

Richard Miller . unresolved

This test passes on plan9-arm. If it times out on plan9-arm64 this seems to be a regression. Better to find and fix the plan9-arm64 problem than fudge the test.

File src/runtime/mgcscavenge_test.go

Line 302, Patchset 13 (Latest): if goos.IsPlan9 != 0 && goarch.IsArm64 != 0 {

Richard Miller . unresolved

The solution to exhausting physical memory is to enable swap space on the builder.

File src/runtime/mpagealloc_test.go

Line 15, Patchset 13 (Latest):func skipPageAllocSimulationOnPlan9Arm64(t *testing.T) {

Richard Miller . unresolved

The solution to physical memory exhaustion is to enable swap space on the builder.

File src/runtime/mpagecache_test.go

Line 15, Patchset 13 (Latest):func skipPageCacheSimulationOnPlan9Arm64(t *testing.T) {

Richard Miller . unresolved

The solution to exhausting physical memory is to enable swap space on the builder.

Gerrit-Comment-Date: Mon, 18 May 2026 10:35:28 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Richard Miller and t hepudds

Rafael Diniz added 13 comments

Patchset-level comments

File-level comment, Patchset 11:

Richard Miller . resolved

Rafael Diniz

Most tests passing now. Hopefully no more arm64 port issues.

File-level comment, Patchset 13 (Latest):

Rafael Diniz . resolved

Rafael Diniz

Done

File-level comment, Patchset 13 (Latest):

Rafael Diniz . resolved

Much improved stability, tests passing now.

File-level comment, Patchset 13 (Latest):

Richard Miller . resolved

Most of the changes in this patch set appear to be related to running a builder on 9front instead of Plan 9. The patches are disabling tests which have run successfully for years on Plan 9 and 9legacy builders. 9front is a fork of Plan 9 which seems to have diverged so far as to be unable to run normal go tests. The title of this CL is "add plan9/arm64 port". I suggest sticking the to goal of passing all tests on Plan 9 with arm64. Porting to 9front can be a separate project.

Rafael Diniz

Thanks Dr. Miller.
Done, I reverted the "waivers", and now all tests are passing. 0 test failures:
https://github.com/rafael2knokia/go/blob/plan9-arm64-dev/misc/plan9/arm64/test-report.md

File src/crypto/x509/root_plan9.go

Line 38, Patchset 13 (Latest): if os.IsNotExist(bestErr) {

Richard Miller . unresolved

Returning an error was the correct thing to do. If the cert file doesn't exist, it's not right to pretend it does. If you apply 9legacy patch ca.diff on the builder, the file will be there as needed.

Rafael Diniz

Indeed. Thanks for the pointer.

File src/net/lookup_test.go

Line 869, Patchset 13 (Latest): if !strings.HasSuffix(err.Error(), errNoSuchHost.Error()) &&

Richard Miller . resolved

This defeats the purpose of the test. Plan 9 correctly returns a NoSuchHost error. If 9front is returning something else, it's the 9front implementation which needs to be fixed.

Rafael Diniz

Correct, reverted.

Line 873, Patchset 13 (Latest): if !err.(*DNSError).IsNotFound &&

Richard Miller . resolved

As above.

Rafael Diniz

Reverted.

File src/net/timeout_test.go

Line 956, Patchset 13 (Latest): if runtime.GOOS == "plan9" {

Richard Miller . unresolved

This test is supported on Plan 9. Considerable work has been done (eg CL 420715) to ensure that it passes. If it breaks on a 9front builder, fix the 9front problem, don't disable the test.

Rafael Diniz

Fixed. I needed some small patching in 9front to get it passed (branch with the fixes here): https://github.com/rafael2knokia/9front/tree/tcpsplice-close-fix

File src/net/writev_test.go

Line 186, Patchset 13 (Latest): if runtime.GOOS == "plan9" {

Richard Miller . resolved

This test passes on Plan 9; it's a 9front problem.

Rafael Diniz

Reverted.

File src/runtime/goroutineleakprofile_test.go

Line 19, Patchset 13 (Latest): if testing.Short() && goos.IsPlan9 != 0 && goarch.IsArm64 != 0 {

Richard Miller . resolved

This test passes on plan9-arm. If it times out on plan9-arm64 this seems to be a regression. Better to find and fix the plan9-arm64 problem than fudge the test.

Rafael Diniz

Fixed.

File src/runtime/mgcscavenge_test.go

Line 302, Patchset 13 (Latest): if goos.IsPlan9 != 0 && goarch.IsArm64 != 0 {

Richard Miller . resolved

The solution to exhausting physical memory is to enable swap space on the builder.

Rafael Diniz

Reverted. Tests are passing now with swap.

File src/runtime/mpagealloc_test.go

Line 15, Patchset 13 (Latest):func skipPageAllocSimulationOnPlan9Arm64(t *testing.T) {

Richard Miller . resolved

The solution to physical memory exhaustion is to enable swap space on the builder.

Rafael Diniz

Reverted. Test are passing now.

File src/runtime/mpagecache_test.go

Line 15, Patchset 13 (Latest):func skipPageCacheSimulationOnPlan9Arm64(t *testing.T) {

Richard Miller . resolved

The solution to exhausting physical memory is to enable swap space on the builder.

Rafael Diniz

Reverted. Tests passing now.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• Richard Miller
• t hepudds

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Fri, 22 May 2026 07:59:28 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Richard Miller and t hepudds

Rafael Diniz voted and added 3 comments

Votes added by Rafael Diniz

Code-Review

+1

3 comments

Patchset-level comments

File-level comment, Patchset 13 (Latest):

Rafael Diniz . resolved

Port is stable (as much as it can be on Plan 9).

File src/crypto/x509/root_plan9.go

Line 38, Patchset 13 (Latest): if os.IsNotExist(bestErr) {

Richard Miller . resolved

Returning an error was the correct thing to do. If the cert file doesn't exist, it's not right to pretend it does. If you apply 9legacy patch ca.diff on the builder, the file will be there as needed.

Rafael Diniz

Indeed. Thanks for the pointer.

Rafael Diniz

Done

File src/net/timeout_test.go

Line 956, Patchset 13 (Latest): if runtime.GOOS == "plan9" {

Richard Miller . resolved

This test is supported on Plan 9. Considerable work has been done (eg CL 420715) to ensure that it passes. If it breaks on a 9front builder, fix the 9front problem, don't disable the test.

Rafael Diniz

Fixed. I needed some small patching in 9front to get it passed (branch with the fixes here): https://github.com/rafael2knokia/9front/tree/tcpsplice-close-fix

Rafael Diniz

Done

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Richard Miller
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 13

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Comment-Date: Fri, 10 Jul 2026 23:45:10 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: Yes

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz, Richard Miller and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #14 to this change.

Following approvals got outdated and were removed:

• Code-Review: +1 by Rafael Diniz

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• Rafael Diniz
• Richard Miller
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 14

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Richard Miller <millerr...@gmail.com>

Gerrit-Attention: Rafael Diniz <rafae...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 14 (Latest):

Richard Miller . unresolved

I have built this by bootstrapping from plan9-arm. Then, when I try to populate the cache with arm64 packages using the plan9-arm64 compiler, I get errors like this:
```
# internal/byteorder
<unknown line number>: internal compiler error: panic: intrinsic already exists for internal/runtime/atomic.Loaduintptr on s390x

goroutine 1 [running]:
runtime/debug.Stack()
	goroot/src/runtime/debug/stack.go:26 +0x64
cmd/compile/internal/base.FatalfAt({0x147c7d0?, 0x0?}, {0xaed72e, 0x9}, {0x4c04c048, 0x1, 0x1})
	goroot/src/cmd/compile/internal/base/print.go:232 +0x210
cmd/compile/internal/base.Fatalf(...)
	goroot/src/cmd/compile/internal/base/print.go:197
cmd/compile/internal/gc.handlePanic()
	goroot/src/cmd/compile/internal/gc/main.go:58 +0x90
panic({0x147c7d0?, 0x4c392e80?})
	goroot/src/runtime/panic.go:859 +0x120
cmd/compile/internal/ssagen.intrinsicBuilders.add(0x4c3968a0, 0x15d58e8, {0xb182cd, 0x17}, {0xaf367b, 0xb}, 0x14f3168)
	goroot/src/cmd/compile/internal/ssagen/intrinsics.go:51 +0x160
cmd/compile/internal/ssagen.intrinsicBuilders.alias(0x4c3968a0?, {0xb182cd, 0x17}, {0xaf367b, 0xb}, {0xb182cd, 0x17}, {0xae3ef8, 0x6}, {0x4c3b8280, ...})
	goroot/src/cmd/compile/internal/ssagen/intrinsics.go:80 +0xf8
cmd/compile/internal/ssagen.initIntrinsics.func3(...)
	goroot/src/cmd/compile/internal/ssagen/intrinsics.go:132
cmd/compile/internal/ssagen.initIntrinsics(0x0?)
	goroot/src/cmd/compile/internal/ssagen/intrinsics.go:695 +0x2438
cmd/compile/internal/ssagen.InitTables(...)
	goroot/src/cmd/compile/internal/ssagen/ssa.go:217
cmd/compile/internal/gc.Main(0x14f2d00)
	goroot/src/cmd/compile/internal/gc/main.go:215 +0xbfc
main.main()
	goroot/src/cmd/compile/main.go:57 +0x100

```
and this
```
# crypto/subtle
panic: d.nx != 0

goroutine 17 [running]:
crypto/internal/fips140/sha256.(*Digest).checkSum(...)
	goroot/src/crypto/internal/fips140/sha256/sha256.go:230
crypto/internal/fips140/sha256.(*Digest).Sum(0x4c416300, {0x4c434160, 0x0, 0x20})
	goroot/src/crypto/internal/fips140/sha256/sha256.go:204 +0x258
crypto/sha256.Sum256(...)
	goroot/src/crypto/sha256/sha256.go:60
cmd/internal/hash.Sum32({0x4c42a178, 0x8, 0x8})
	goroot/src/cmd/internal/hash/hash.go:27 +0x98
cmd/internal/obj.(*Link).GCLocalsSym(...)
	goroot/src/cmd/internal/obj/sym.go:243
cmd/compile/internal/liveness.(*Liveness).emit(0x4c4222c0)
	goroot/src/cmd/compile/internal/liveness/plive.go:1387 +0x478
cmd/compile/internal/liveness.Compute(0x4c4f8640, 0x4c444000, 0x0, 0x4c428540, 0x0)
	goroot/src/cmd/compile/internal/liveness/plive.go:1433 +0x894
cmd/compile/internal/ssagen.genssa(0x4c444000, 0x4c428540)
	goroot/src/cmd/compile/internal/ssagen/ssa.go:6988 +0xbc
cmd/compile/internal/ssagen.Compile(0x4c4f8640, 0x0, 0x4c3d4000?)
	goroot/src/cmd/compile/internal/ssagen/pgen.go:314 +0x480
cmd/compile/internal/gc.compileFunctions.func2()
	goroot/src/cmd/compile/internal/gc/compile.go:177 +0x6c
created by cmd/compile/internal/gc.compileFunctions in goroutine 1
	goroot/src/cmd/compile/internal/gc/compile.go:163 +0x108
```
and this
```
# time
panic: runtime error: slice bounds out of range [1280375023:64]

goroutine 33 [running]:
crypto/internal/fips140/sha256.(*Digest).Write(0x4c6bf140, {0x4c6bf0b8?, 0x32, 0x82909c?})
	goroot/src/crypto/internal/fips140/sha256/sha256.go:176 +0x254
crypto/internal/fips140/sha256.(*Digest).checkSum(...)
	goroot/src/crypto/internal/fips140/sha256/sha256.go:227
crypto/internal/fips140/sha256.(*Digest).Sum(0x4e5dff00, {0x4e840920, 0x0, 0x20})
	goroot/src/crypto/internal/fips140/sha256/sha256.go:204 +0xec
crypto/sha256.Sum256(...)
	goroot/src/crypto/sha256/sha256.go:60
cmd/internal/hash.Sum32({0x4e5e18d0, 0xe, 0x10})
	goroot/src/cmd/internal/hash/hash.go:27 +0x98
cmd/internal/obj.(*Link).GCLocalsSym(...)
	goroot/src/cmd/internal/obj/sym.go:243
cmd/compile/internal/liveness.(*Liveness).emit(0x4c143760)
	goroot/src/cmd/compile/internal/liveness/plive.go:1387 +0x478
cmd/compile/internal/liveness.Compute(0x4d15dcc0, 0x4c6881c0, 0x30, 0x4e848240, 0x0)
	goroot/src/cmd/compile/internal/liveness/plive.go:1433 +0x894
cmd/compile/internal/ssagen.genssa(0x4c6881c0, 0x4e848240)
	goroot/src/cmd/compile/internal/ssagen/ssa.go:6988 +0xbc
cmd/compile/internal/ssagen.Compile(0x4d15dcc0, 0x0, 0x4c5060b8?)
	goroot/src/cmd/compile/internal/ssagen/pgen.go:314 +0x480
cmd/compile/internal/gc.compileFunctions.func2()
	goroot/src/cmd/compile/internal/gc/compile.go:177 +0x6c
created by cmd/compile/internal/gc.compileFunctions in goroutine 1
	goroot/src/cmd/compile/internal/gc/compile.go:163 +0x108
```
and this
```
# cmd/vendor/github.com/google/pprof/internal/graph
<unknown line number>: internal compiler error: panic: runtime error: invalid memory address or nil pointer dereference

goroutine 1 [running]:
runtime/debug.Stack()
	/usr/miller/goroot/src/runtime/debug/stack.go:26 +0x64
cmd/compile/internal/base.FatalfAt({0x4c04e128?, 0x0?}, {0xaed72e, 0x9}, {0x4c04e158, 0x1, 0x1})
	/usr/miller/goroot/src/cmd/compile/internal/base/print.go:232 +0x210
cmd/compile/internal/base.Fatalf(...)
	/usr/miller/goroot/src/cmd/compile/internal/base/print.go:197
cmd/compile/internal/gc.handlePanic()
	/usr/miller/goroot/src/cmd/compile/internal/gc/main.go:58 +0x90
panic({0x1489b20?, 0x15912d0?})
	/usr/miller/goroot/src/runtime/panic.go:859 +0x120
cmd/compile/internal/ssagen.initIntrinsics(0x50?)
	/usr/miller/goroot/src/cmd/compile/internal/ssagen/intrinsics.go:114 +0x214
cmd/compile/internal/ssagen.InitTables(...)
	/usr/miller/goroot/src/cmd/compile/internal/ssagen/ssa.go:217
cmd/compile/internal/gc.Main(0x14f2d00)
	/usr/miller/goroot/src/cmd/compile/internal/gc/main.go:215 +0xbfc
main.main()
	/usr/miller/goroot/src/cmd/compile/main.go:57 +0x100
```

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Rafael Diniz

• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 14

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Rafael Diniz <rafae...@gmail.com>

Gerrit-Comment-Date: Sun, 12 Jul 2026 14:06:56 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Richard Miller (Gerrit)]

Patchset-level comments

Richard Miller

I rebooted the plan9 system using `*ncpu=1` to run on a single core. This time the arm64 native rebuild (`go build std cmd`) succeeded with no errors.

Have you tested this release on true SMP hardware, or only on an emulation of multiple cores? There may be a subtle flaw somewhere which is only triggered by some detail of the hardware behaviour (MMU TLB, caches, memory reordering) that isn't being accurately emulated on your test platform.

Gerrit-Comment-Date: Sun, 12 Jul 2026 16:17:45 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 14 (Latest):

Richard Miller . unresolved

I've now run `go tool dist test` (on just one core) and it does seem more reliable now. Well done.

In that test, there was one panic in `ssagen.initIntrinsics`, which suggests that SMP is exacerbating the problem but not causing it.

The other notable failure was the `runtime` test running forever, with address space expanded to the full available 4GB. Is this what you would expect for memory usage? What's the maximum user address space on the 9front platform you are using for tests?

Gerrit-Comment-Date: Mon, 13 Jul 2026 12:42:25 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 14 (Latest):

Richard Miller . resolved

I've now run `go tool dist test` (on just one core) and it does seem more reliable now. Well done.

In that test, there was one panic in `ssagen.initIntrinsics`, which suggests that SMP is exacerbating the problem but not causing it.

The other notable failure was the `runtime` test running forever, with address space expanded to the full available 4GB. Is this what you would expect for memory usage? What's the maximum user address space on the 9front platform you are using for tests?

Richard Miller

The infinite loop in `runtime.test` was my fault -- a kernel bug which I've now fixed. Thanks for flushing it out!

Gerrit-Comment-Date: Tue, 14 Jul 2026 16:01:17 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Richard Miller <millerr...@gmail.com>

[Richard Miller (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Richard Miller added 1 comment

Patchset-level comments

File-level comment, Patchset 14 (Latest):

Richard Miller . resolved

I have built this by bootstrapping from plan9-arm. Then, when I try to populate the cache with arm64 packages using the plan9-arm64 compiler, I get errors like this:

Richard Miller

Looks like this was a problem in my kernel too, also now corrected.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Rafael Diniz
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds

• No-Unresolved-Comments

• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 14

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Rafael Diniz <rafae...@gmail.com>

Gerrit-Comment-Date: Tue, 21 Jul 2026 15:16:47 +0000

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #15 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Rafael Diniz
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 15

[Rafael Diniz (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt and t hepudds

Message from Rafael Diniz

Thanks again for the detailed review. Summary of where PS15 stands, point by point:

1. Compiler-heap corruption (initIntrinsics / sha256 garbage / 'intrinsic already exists on s390x'): I think I found the mechanism. plan9/arm64 was the only Plan 9 port returning heap pages to the OS during sysUnused, via segfree (mem_plan9_arm64.go). Dropping a page's physical backing that way is unsafe under SMP unless the kernel guarantees a cross-core TLB shootdown: another core can keep a stale mapping and read/write the page after it's recycled, corrupting unrelated live allocations. In PS15 sysUnusedOS is now a no-op on plan9/arm64, matching plan9/{386,arm,amd64}; the segment reservation is kept. I want to be upfront about my testing: I can only run under QEMU/TCG, and it does NOT reproduce this - the exact packages from your crash logs, go build -a -p 4, ran 5 clean fresh-cache iterations on 4 emulated cores even WITHOUT the fix. So TCG isn't exercising the TLB path, consistent with your point that emulation won't reproduce multicore behaviour. I therefore cannot claim from my side that the panics are gone - could you retest PS15 on real hardware?

2. Disabling tests that pass on Plan 9 / 9legacy (PS13): you were right, and I've reverted all of them. PS15 no longer touches crypto/x509/root_plan9.go, net/{lookup,timeout,writev}_test.go, or runtime/{goroutineleakprofile,mgcscavenge,mpagealloc,mpagecache}_test.go. Where I'd seen failures they traced to the 9front loopback tcpsplice kernel bug (fixed 9front-side) and the segfree corruption above, not to anything needing a waiver.

3. ABI/timesplit/walltime (PS3, PS9): fixed. open/dupfd now 8-byte-align the int32 syscall args as you described; timesplit uses the simpler arm64 UDIV/MUL/SUB form you suggested (sec+8/nsec+16); walltime is a Go function again. go vet -asmdecl runtime and syscall are both clean.

4. Real SMP hardware (PS14): honestly, so far only QEMU/TCG, which as you noted won't model MMU/TLB/cache/reordering - the segfree bug is a concrete instance of exactly that. I don't yet have physical multicore arm64 results; I'll arrange hardware testing and won't ask for another substantive review until I can report real-hardware results.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 15

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Comment-Date: Mon, 27 Jul 2026 08:36:50 +0000

Gerrit-HasComments: No

Gerrit-Has-Labels: No

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #16 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt

• Rafael Diniz
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 16

Gerrit-Owner: Gerrit Bot <letsus...@gmail.com>

Gerrit-Reviewer: Ian Lance Taylor <ia...@golang.org>

Gerrit-Reviewer: Keith Randall <k...@golang.org>

Gerrit-Reviewer: Michael Pratt <mpr...@google.com>

Gerrit-Reviewer: Rafael Diniz <rafae...@gmail.com>

Gerrit-Reviewer: SLSA Policy Verification Service <devtools-gerritco...@google.com>

Gerrit-CC: Gopher Robot <go...@golang.org>

Gerrit-CC: Richard Miller <millerr...@gmail.com>

Gerrit-CC: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Keith Randall <k...@golang.org>

Gerrit-Attention: Ian Lance Taylor <ia...@golang.org>

Gerrit-Attention: t hepudds <thepud...@gmail.com>

Gerrit-Attention: Michael Pratt <mpr...@google.com>

Gerrit-Attention: Rafael Diniz <rafae...@gmail.com>

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #17 to this change.

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Rafael Diniz
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement
• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 17

[Gerrit Bot (Gerrit)]

Attention needed from Ian Lance Taylor, Keith Randall, Michael Pratt, Rafael Diniz and t hepudds

Gerrit Bot uploaded new patchset

Gerrit Bot uploaded patch set #18 to this change.

Following approvals got outdated and were removed:

• SLSA-Policy-Verified: +1 by SLSA Policy Verification Service

Open in Gerrit

Related details

Attention is currently required from:

• Ian Lance Taylor
• Keith Randall
• Michael Pratt
• Rafael Diniz
• t hepudds

Submit Requirements:

• Code-Review
• No-Holds
• No-Unresolved-Comments
• Review-Enforcement

• SLSA-Policy-Verified

• TryBots-Pass

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: newpatchset

Gerrit-Project: go

Gerrit-Branch: master

Gerrit-Change-Id: Id58aaf71cc337651292cf4f7d1bad2e57cbef4e6

Gerrit-Change-Number: 719643

Gerrit-PatchSet: 18