public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as
@ 2010-11-30 19:09 ro at gcc dot gnu.org
2010-12-13 11:16 ` [Bug target/46729] " ebotcazou at gcc dot gnu.org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: ro at gcc dot gnu.org @ 2010-11-30 19:09 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
Summary: Many 32-bit 30_threads execution tests fail on Solaris
10+/SPARC with as
Product: gcc
Version: 4.6.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned@gcc.gnu.org
ReportedBy: ro@gcc.gnu.org
CC: ebotcazou@gcc.gnu.org
Host: sparc-sun-solaris2.1[01]
Target: sparc-sun-solaris2.1[01]
Build: sparc-sun-solaris2.1[01]
Many 32-bit 30_threads libstdc++ execution tests fail on Solaris 10 and
11/SPARC
with Sun as. An example is
FAIL: 30_threads/async/42819.cc execution test
The test SEGVs like this:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1 (LWP 1)]
0x00019f40 in __gthread_active_p (this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/sparc-sun-solaris2.10/bits/gthr-default.h:216
216 if (__builtin_expect (__gthread_active_latest_value < 0, 0))
(gdb) where
#0 0x00019f40 in __gthread_active_p (this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/sparc-sun-solaris2.10/bits/gthr-default.h:216
#1 __gthread_once (this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/sparc-sun-solaris2.10/bits/gthr-default.h:725
#2 call_once<std::_Mem_fn<void
(std::__future_base::_State::*)(std::function<std::unique_ptr<std::__future_base::_Result_base,
std::__future_base::_Result_base::_Deleter>()>&, bool&)>,
std::__future_base::_State* const,
std::reference_wrapper<std::function<std::unique_ptr<std::__future_base::_Result_base,
std::__future_base::_Result_base::_Deleter>()> >, std::reference_wrapper<bool>
> (
this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/mutex:733
#3
std::__future_base::_State::_M_set_result(std::function<std::unique_ptr<std::__future_base::_Result_base,
std::__future_base::_Result_base::_Deleter> ()()>, bool) (this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/future:313
#4 std::__future_base::_Deferred_state<int>::_M_run_deferred (this=0x2be28)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/future:1295
#5 0x000171fc in std::__future_base::_State::wait (value=1)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/future:281
#6 std::__basic_future<int>::_M_get_result (value=1)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/future:515
#7 std::future<int>::get (value=1)
at
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/include/future:594
#8 work1 (value=1)
at
/vol/gcc/src/hg/trunk/local/libstdc++-v3/testsuite/30_threads/async/42819.cc:36
#9 0x00019acc in test01 ()
at
/vol/gcc/src/hg/trunk/local/libstdc++-v3/testsuite/30_threads/async/42819.cc:51
#10 0x00019b3c in main ()
at
/vol/gcc/src/hg/trunk/local/libstdc++-v3/testsuite/30_threads/async/42819.cc:57
(gdb) display/i $pc
1: x/i $pc
=> 0x19f40 <_ZNSt13__future_base15_Deferred_stateIiE15_M_run_deferredEv+340>:
st %g2, [ %g3 ]
(gdb) p/x $g2
$4 = 0x154c8
(gdb) p/x $g3
$5 = 0x10b52a40
As can be seen with pmap, %g3 isn't mapped:
ro@yoda 2243 > pmap 5431
5431: /var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++
00010000 48K r-x--
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/testsuite/42819.exe
0002A000 8K rwx--
/var/gcc/regression/trunk/10-gcc/build/sparc-sun-solaris2.10/libstdc++-v3/testsuite/42819.exe
0002C000 16K rwx-- [ heap ]
FE800000 1216K r-x-- /lib/libc.so.1
There is no problem with GNU as.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/46729] Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
@ 2010-12-13 11:16 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:19 ` ebotcazou at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu.org @ 2010-12-13 11:16 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed| |2010.12.13 11:16:10
AssignedTo|unassigned at gcc dot |ebotcazou at gcc dot
|gnu.org |gnu.org
Ever Confirmed|0 |1
--- Comment #1 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2010-12-13 11:16:10 UTC ---
Investigating.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/46729] Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
2010-12-13 11:16 ` [Bug target/46729] " ebotcazou at gcc dot gnu.org
@ 2010-12-19 12:19 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:20 ` ebotcazou at gcc dot gnu.org
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu.org @ 2010-12-19 12:19 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
--- Comment #2 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2010-12-19 12:19:16 UTC ---
Author: ebotcazou
Date: Sun Dec 19 12:19:12 2010
New Revision: 168049
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=168049
Log:
PR target/46729
* config/sparc/sparc.h (GLOBAL_OFFSET_TABLE_REGNUM): New macro.
(PIC_OFFSET_TABLE_REGNUM): Rewrite in terms of above macro.
* config/sparc/sparc.c (pic_helper_needed): Delete.
(global_offset_table): Likewise.
(pic_helper_symbol): Rename to...
(got_helper_rtx): ...this.
(global_offset_table_rtx): New global variable.
(sparc_got_symbol): Likewise.
(sparc_got): New static function.
(check_pic): Use local variable and call sparc_got.
(sparc_tls_symbol): Initialize to NULL_RTX.
(sparc_tls_got): In non-PIC mode, reload the GOT register for Sun TLS
and 32-bit ABI and copy the GOT symbol to a new register otherwise.
(get_pc_thunk_name): Rename local variable.
(gen_load_pcrel_sym): New wrapper around load_pcrel_sym{si,di}.
(load_pic_register): Rename to...
(load_got_register): ...this. Adjust and call gen_load_pcrel_sym.
(sparc_expand_prologue): Do not test flag_pic.
(sparc_output_mi_thunk): Use pic_offset_table_rtx directly.
(sparc_file_end): Test got_helper_rtx instead of pic_helper_needed.
Rename local variable and do not call get_pc_thunk_name again.
* config/sparc/sparc.md (load_pcrel_sym): Add operand #3.
Modified:
trunk/gcc/ChangeLog
trunk/gcc/config/sparc/sparc.c
trunk/gcc/config/sparc/sparc.h
trunk/gcc/config/sparc/sparc.md
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/46729] Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
2010-12-13 11:16 ` [Bug target/46729] " ebotcazou at gcc dot gnu.org
2010-12-19 12:19 ` ebotcazou at gcc dot gnu.org
@ 2010-12-19 12:20 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:25 ` [Bug target/46729] 32-bit 30_threads execution tests fail on Solaris 10/SPARC with Sun as ebotcazou at gcc dot gnu.org
2010-12-19 12:30 ` ebotcazou at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu.org @ 2010-12-19 12:20 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
--- Comment #3 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2010-12-19 12:20:12 UTC ---
Author: ebotcazou
Date: Sun Dec 19 12:20:08 2010
New Revision: 168050
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=168050
Log:
PR target/46729
* config/sparc/sparc.h (GLOBAL_OFFSET_TABLE_REGNUM): New macro.
(PIC_OFFSET_TABLE_REGNUM): Rewrite in terms of above macro.
* config/sparc/sparc.c (pic_helper_needed): Delete.
(global_offset_table): Likewise.
(pic_helper_symbol): Rename to...
(got_helper_rtx): ...this.
(global_offset_table_rtx): New global variable.
(sparc_got_symbol): Likewise.
(sparc_got): New static function.
(check_pic): Use local variable and call sparc_got.
(sparc_tls_symbol): Initialize to NULL_RTX.
(sparc_tls_got): In non-PIC mode, reload the GOT register for Sun TLS
and 32-bit ABI and copy the GOT symbol to a new register otherwise.
(get_pc_thunk_name): Rename local variable.
(gen_load_pcrel_sym): New wrapper around load_pcrel_sym{si,di}.
(load_pic_register): Rename to...
(load_got_register): ...this. Adjust and call gen_load_pcrel_sym.
(sparc_expand_prologue): Do not test flag_pic.
(sparc_output_mi_thunk): Use pic_offset_table_rtx directly.
(sparc_file_end): Test got_helper_rtx instead of pic_helper_needed.
Rename local variable and do not call get_pc_thunk_name again.
* config/sparc/sparc.md (load_pcrel_sym): Add operand #3.
Modified:
branches/gcc-4_5-branch/gcc/ChangeLog
branches/gcc-4_5-branch/gcc/config/sparc/sparc.c
branches/gcc-4_5-branch/gcc/config/sparc/sparc.h
branches/gcc-4_5-branch/gcc/config/sparc/sparc.md
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/46729] 32-bit 30_threads execution tests fail on Solaris 10/SPARC with Sun as
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
` (2 preceding siblings ...)
2010-12-19 12:20 ` ebotcazou at gcc dot gnu.org
@ 2010-12-19 12:25 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:30 ` ebotcazou at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu.org @ 2010-12-19 12:25 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
--- Comment #4 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2010-12-19 12:24:53 UTC ---
They should pass now.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/46729] 32-bit 30_threads execution tests fail on Solaris 10/SPARC with Sun as
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
` (3 preceding siblings ...)
2010-12-19 12:25 ` [Bug target/46729] 32-bit 30_threads execution tests fail on Solaris 10/SPARC with Sun as ebotcazou at gcc dot gnu.org
@ 2010-12-19 12:30 ` ebotcazou at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu.org @ 2010-12-19 12:30 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46729
Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Version|4.6.0 |4.5.2
Target Milestone|--- |4.5.3
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2010-12-19 12:30 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-11-30 19:09 [Bug target/46729] New: Many 32-bit 30_threads execution tests fail on Solaris 10+/SPARC with as ro at gcc dot gnu.org
2010-12-13 11:16 ` [Bug target/46729] " ebotcazou at gcc dot gnu.org
2010-12-19 12:19 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:20 ` ebotcazou at gcc dot gnu.org
2010-12-19 12:25 ` [Bug target/46729] 32-bit 30_threads execution tests fail on Solaris 10/SPARC with Sun as ebotcazou at gcc dot gnu.org
2010-12-19 12:30 ` ebotcazou at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).