public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/15420] New: g++ produce bad assembler code when using -O1
@ 2004-05-14 2:50 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 3:47 ` [Bug target/15420] " pinskia at gcc dot gnu dot org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com @ 2004-05-14 2:50 UTC (permalink / raw)
To: gcc-bugs
g++ manage to compile a file with -O0 whereas -O1 makes it fail.
I didn't try to isolate the problem. I hope you may excuse my laziness...
The problem appears when compiling xalan 1.8.0 which can be found at :
http://apache.fastorama.com/dist/xml/xalan-c/Xalan-C_1_8_0-src.tar.gz
(it may be useless I attach all the files ?)
When I try to compile src/xalanc/XalanEXSLT/XalanEXSLTString.cpp I get that
strange error :
lhuard ~/src/xml-xalan/c/src/xalanc % g++ -m64 -O1 -DNDEBUG -m64
-DXML_BITSTOBUILD_64 -mcpu=v9 -fno-elide-constructors -Wall -fPIC -DSOLARIS
-D_REENTRANT -DXALAN_INMEM_MSG_LOADER -c -I/users/lhuard/src/xml-xalan/c/src
-I/users/lhuard/src/xml-xalan/c/include -I../../nls/include
-I/users/lhuard/src/xerces-c-src_2_5_0/src/
-I/users/lhuard/src/xerces-c-src_2_5_0/include/xercesc
-I/users/lhuard/src/xerces-c-src_2_5_0/include/ -o ../../obj/XalanEXSLTString.o
/users/lhuard/src/xml-xalan/c/src/xalanc/XalanEXSLT/XalanEXSLTString.cpp
/usr/ccs/bin/as: "/tmp/lhuard/ccB9twdG.s", line 887: error: invalid (misaligned)
register
whereas it works with -O0.
I hope those few informations will permit anybody to reproduce the problem. If
not, tell me, I might send the .s output of cc1 or give any helpfull information.
Let me know which information may be usefull.
--
Summary: g++ produce bad assembler code when using -O1
Product: gcc
Version: 3.2.1
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: nospam dot gccsolarix dot 20 dot lenex at spamgourmet
dot com
CC: gcc-bugs at gcc dot gnu dot org
GCC build triplet: sparc-sun-solaris2.8
GCC host triplet: sparc-sun-solaris2.8
GCC target triplet: sparc-sun-solaris2.8
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/15420] g++ produce bad assembler code when using -O1
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
@ 2004-05-14 3:47 ` pinskia at gcc dot gnu dot org
2004-05-14 13:34 ` ebotcazou at gcc dot gnu dot org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-05-14 3:47 UTC (permalink / raw)
To: gcc-bugs
--
What |Removed |Added
----------------------------------------------------------------------------
Component|c++ |target
Keywords| |wrong-code
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/15420] g++ produce bad assembler code when using -O1
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 3:47 ` [Bug target/15420] " pinskia at gcc dot gnu dot org
@ 2004-05-14 13:34 ` ebotcazou at gcc dot gnu dot org
2004-05-14 18:38 ` nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-05-14 13:34 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-05-13 19:19 -------
We do not support GCC 3.2.x anymore, only GCC 3.3.x and later.
However, the error message sounds familiar: could you post the fragment of
assembly code around line 887?
--
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |WAITING
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/15420] g++ produce bad assembler code when using -O1
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 3:47 ` [Bug target/15420] " pinskia at gcc dot gnu dot org
2004-05-14 13:34 ` ebotcazou at gcc dot gnu dot org
@ 2004-05-14 18:38 ` nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 19:13 ` bangerth at dealii dot org
2004-07-10 0:59 ` pinskia at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com @ 2004-05-14 18:38 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com 2004-05-14 13:49 -------
I didn't notice a newer version exist since my administrator set this old
version as default.
The bug is corrected since 3.3:x versions : I tried with 3.2.3. It is buggy. I
tried with 3.3.1. It works.
No further investigation is needed since the bug is fixed in up-to-date versions.
However, here is an excerpt of the buggy assembler code :
871 call
_ZN10xalanc_1_827XalanReferenceCountedObject15removeReferenceEPS0_, 0
872 ldx [%fp+1999], %o0
873 ba,pt %xcc, .LL836
874 nop
875 .LL713:
876 ldx [%i2], %o0
877 ldx [%o0+208], %o1
878 .LLEHB27:
879 call %o1, 0
880 mov %i2, %o0
881 .LLEHE27:
882 stx %o0, [%fp+2007]
883 cmp %l0, 1
884 bne,pt %icc, .LL729
885 mov %o0, %i5
886 ldd [%fp+1991], %f2
887 fdtox %f2, %f3 <--- The problem was here.
888 st %f3, [%fp+1979]
889 ldx [%i1], %o1
890 lduh [%o1], %i1
891 mov 0, %o1
892 mov -1, %o2
Thks
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/15420] g++ produce bad assembler code when using -O1
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
` (2 preceding siblings ...)
2004-05-14 18:38 ` nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
@ 2004-05-14 19:13 ` bangerth at dealii dot org
2004-07-10 0:59 ` pinskia at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: bangerth at dealii dot org @ 2004-05-14 19:13 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From bangerth at dealii dot org 2004-05-14 14:05 -------
Apparently already fixed. So closing.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |RESOLVED
Resolution| |FIXED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug target/15420] g++ produce bad assembler code when using -O1
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
` (3 preceding siblings ...)
2004-05-14 19:13 ` bangerth at dealii dot org
@ 2004-07-10 0:59 ` pinskia at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-07-10 0:59 UTC (permalink / raw)
To: gcc-bugs
--
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |3.3.1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15420
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2004-07-10 0:59 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-05-14 2:50 [Bug c++/15420] New: g++ produce bad assembler code when using -O1 nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 3:47 ` [Bug target/15420] " pinskia at gcc dot gnu dot org
2004-05-14 13:34 ` ebotcazou at gcc dot gnu dot org
2004-05-14 18:38 ` nospam dot gccsolarix dot 20 dot lenex at spamgourmet dot com
2004-05-14 19:13 ` bangerth at dealii dot org
2004-07-10 0:59 ` pinskia at gcc dot gnu dot 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).