cmd/trace: unrecognized failures

[gopherbot]

#!watchflakes
post <- pkg == "cmd/trace" && test == "" && date < "2023-03-01"

Issue created automatically to collect these failures.

Example (log):

/tmp/workdir/go/pkg/tool/dragonfly_amd64/link: running gcc failed: exit status 1
/usr/libexec/binutils234/elf/ld.gold: error: cannot find -lresolv
collect2: error: ld returned 1 exit status

— watchflakes

[gopherbot]

Found new dashboard test flakes for:

#!watchflakes
post <- pkg == "cmd/trace" && test == ""

2022-11-07 17:58 dragonfly-amd64-622 website@0bdc4717 go@a6642e67 cmd/trace [build] (log)

/tmp/workdir/go/pkg/tool/dragonfly_amd64/link: running gcc failed: exit status 1
/usr/libexec/binutils234/elf/ld.gold: error: cannot find -lresolv
collect2: error: ld returned 1 exit status

2022-11-07 17:58 dragonfly-amd64-622 website@0bdc4717 go@41be62e1 cmd/trace [build] (log)

/tmp/workdir/go/pkg/tool/dragonfly_amd64/link: running gcc failed: exit status 1
/usr/libexec/binutils234/elf/ld.gold: error: cannot find -lresolv
collect2: error: ld returned 1 exit status

2022-11-07 21:25 dragonfly-amd64-622 tools@003fde14 go@6939659a cmd/trace [build] (log)

/tmp/workdir/go/pkg/tool/dragonfly_amd64/link: running gcc failed: exit status 1
/usr/libexec/binutils234/elf/ld.gold: error: cannot find -lresolv
collect2: error: ld returned 1 exit status

— watchflakes

[bcmills]

Duplicate of #56787

[gopherbot]

Found new dashboard test flakes for:

#!watchflakes
post <- pkg == "cmd/trace" && test == ""

2023-01-24 19:47 freebsd-arm-paulzhol go@ffbd194f cmd/trace (log)

PASS
exec: WaitDelay expired before I/O complete

— watchflakes

[prattmic]

#56785 (comment) is #58230

[gopherbot]

Found new dashboard test flakes for:

#!watchflakes
post <- pkg == "cmd/trace" && test == ""

2023-08-01 14:26 linux-riscv64-unmatched go@d4b46b09 cmd/trace [build] (log)

runtime: marked free object in span 0x3fb03caa58, elemsize=16 freeindex=0 (bad use of unsafe.Pointer? try -d=checkptr)
0x3fac2c6000 free  unmarked
0x3fac2c6010 alloc marked  
0x3fac2c6020 alloc marked  
0x3fac2c6030 free  unmarked
0x3fac2c6040 free  unmarked
0x3fac2c6050 alloc marked  
0x3fac2c6060 free  unmarked
0x3fac2c6070 alloc marked  
0x3fac2c6080 alloc marked  
...
fatal error: found pointer to free object

goroutine 3 [running]:
runtime.throw({0x301a46?, 0x3fac2c75d0?})
	runtime/panic.go:837 +0x44 fp=0x3fac035598 sp=0x3fac035570 pc=0x452b4
runtime.(*mspan).reportZombies(0x3fb03caa58)
	runtime/mgcsweep.go:861 +0x2f0 fp=0x3fac035618 sp=0x3fac035598 pc=0x35c88
runtime.(*sweepLocked).sweep(0x5486c0?, 0x58?)
	runtime/mgcsweep.go:644 +0x9c4 fp=0x3fac035728 sp=0x3fac035618 pc=0x35464
runtime.sweepone()
	runtime/mgcsweep.go:393 +0x140 fp=0x3fac035780 sp=0x3fac035728 pc=0x347c0
runtime.bgsweep(0x0?)
	runtime/mgcsweep.go:301 +0x14c fp=0x3fac0357c8 sp=0x3fac035780 pc=0x3451c
runtime.gcenable.func1()
	runtime/mgc.go:203 +0x2c fp=0x3fac0357d8 sp=0x3fac0357c8 pc=0x28d4c
runtime.goexit()
	runtime/asm_riscv64.s:512 +0x4 fp=0x3fac0357d8 sp=0x3fac0357d8 pc=0x750a4
created by runtime.gcenable in goroutine 1
	runtime/mgc.go:203 +0x70

— watchflakes

[mknyszek]

This looks like memory corruption on RISC-V. Is there an open issue for that?

[mknyszek]

The latest failure is definitely not the same as the previous failures here. I'll close this and open a new issue.

[bcmills]

I'll close this and open a new issue.

For the moment, to stop watchflakes from reopening issues you need to add a date predicate.
(Longer term I'd like that to become unnecessary; see #57632 (comment).)