[Bug go/59408] New: [4.9 regression] Many Go tests FAIL with notesleep not on g0

[Bug go/59408] New: [4.9 regression] Many Go tests FAIL with notesleep not on g0

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59408

            Bug ID: 59408
           Summary: [4.9 regression] Many Go tests FAIL with notesleep not
                    on g0
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: go
          Assignee: ian at airs dot com
          Reporter: ro at gcc dot gnu.org
              Host: *-*-solaris2.*
            Target: *-*-solaris2.*
             Build: *-*-solaris2.*

Between 20131104 and 20131114, many Go tests (both in gcc/testsuite/go.* and
libgo) started to FAIL like

fatal error: notesleep not on g0

goroutine 35 [running]:

        :0

        :0

        :0

        :0

        :0

        :0

        :0
net.TestMutexStress
       
/var/gcc/regression/trunk/10-gcc-gas/build/i386-pc-solaris2.10/libgo/gotest24280/test/fd_mutex_test.go:186

        :0

        :0

        :0

goroutine 1 [chan receive]:
main.main
       
/var/gcc/regression/trunk/10-gcc-gas/build/i386-pc-solaris2.10/libgo/gotest24280/test/_testmain.go:167

goroutine 38 [runnable]:
net.$nested23
       
/var/gcc/regression/trunk/10-gcc-gas/build/i386-pc-solaris2.10/libgo/gotest24280/test/fd_mutex_test.go:174
created by net.TestMutexStress
       
/var/gcc/regression/trunk/10-gcc-gas/build/i386-pc-solaris2.10/libgo/gotest24280/test/fd_mutex_test.go:135
FAIL: net

This happens both for 32-bit and 64-bit tests, SPARC and x86, and constitues
a regression from 4.8 where at least the 32-bit Go tests almost all PASSed
on Solaris.

  Rainer

[Bug go/59408] [4.9 regression] Many Go tests FAIL with notesleep not on g0

[ro at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59408

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.9.0

[Bug go/59408] [4.9 regression] Many Go tests FAIL with notesleep not on g0

[ian at airs dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59408

Ian Lance Taylor <ian at airs dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |FIXED

--- Comment #2 from Ian Lance Taylor <ian at airs dot com> ---
This problem should be fixed now.  Sorry about that.

[Bug go/59408] [4.9 regression] Many Go tests FAIL with notesleep not on g0

[ian at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59408

--- Comment #1 from ian at gcc dot gnu.org <ian at gcc dot gnu.org> ---
Author: ian
Date: Fri Dec  6 18:26:27 2013
New Revision: 205756

URL: http://gcc.gnu.org/viewcvs?rev=205756&root=gcc&view=rev
Log:
    PR go/59408
runtime: Don't require g != m->g0 in sema notesleep.

Modified:
    trunk/libgo/runtime/lock_sema.c

[Bug go/59408] [4.9 regression] Many Go tests FAIL with notesleep not on g0

[ro at CeBiTec dot Uni-Bielefeld.DE]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59408

--- Comment #3 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #2 from Ian Lance Taylor <ian at airs dot com> ---
> This problem should be fixed now.  Sorry about that.

It does for the vast majority of the 32-bit tests.  I'll file separate
bugs for the few remaining 32-bit failures and the 64-bit tests.

Thanks.
        Rainer