[Bug c/33877] Request for __VA_ARGC__

--- Comment #1 from igodard at pacbell dot net 2008-07-23 06:59 --- Second the motion - I too have had to tie myself in knots (generating source code that is then compiled...) to get the effect of this extension. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33877

Re: [Bug preprocessor/36906] New: #pragma GNU poison index()

Try the attribute deprecated. Sent from my iPhone On Jul 22, 2008, at 23:50, "Dave at Yost dot com" <[EMAIL PROTECTED] > wrote: There should be a way to poison a symbol only when it is used as a function or macro call. I suggest that following a symbol with a pair of parentheses might be

[Bug preprocessor/36906] #pragma GNU poison index()

--- Comment #1 from pinskia at gmail dot com 2008-07-23 07:09 --- Subject: Re: New: #pragma GNU poison index() Try the attribute deprecated. Sent from my iPhone On Jul 22, 2008, at 23:50, "Dave at Yost dot com" <[EMAIL PROTECTED] > wrote: > There should be a way to poison a symbo

[Bug tree-optimization/36861] boost's compressed avl confuses GCC

--- Comment #17 from igaztanaga at gmail dot com 2008-07-23 08:03 --- Andrew, I think I do write some crappy code, but the slowdown is also present in the non-optimized version. FWIW, there are STL implementations that embed the color bit in rbtrees in the parent pointer so the optimiza

[Bug middle-end/36885] [TUPLES] tuples branch bootstrap failure in libstdc++-v3

--- Comment #9 from rguenth at gcc dot gnu dot org 2008-07-23 08:33 --- Created an attachment (id=15944) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15944&action=view) somewhat reduced testcase Fails at -O2 -g. Two errors this time ;) ./cc1plus -quiet -O2 locale_init.3.ii -g l

[Bug fortran/36894] question on gfortran character*(*)

--- Comment #4 from burnus at gcc dot gnu dot org 2008-07-23 09:25 --- First of all: If you want to see in C-like syntax which interface gfortran generates, compile "gfortran -fdump-tree-original -c myopenbf.f", which generates a myopenbf.f.003t.original file. Secondly: myopenbf_(bf

[Bug tree-optimization/36830] [4.4 Regression] STORAGE_ERROR raised compiling s-os_lib.adb

--- Comment #7 from ebotcazou at gcc dot gnu dot org 2008-07-23 09:35 --- I'm seeing this as well since the PRE rewrite. -- ebotcazou at gcc dot gnu dot org changed: What|Removed |Added -

[Bug target/36840] Fortran complex array multiply missed optimization

--- Comment #3 from rguenth at gcc dot gnu dot org 2008-07-23 09:37 --- The problem is that SLP and the reduction operation do not mix and that the vectorizer doesn't understand the complex component accessors. : # S.4_40 = PHI D.1045_26 = S.4_40 + -1; CR.27_44 = REALPART_EXPR <(

[Bug fortran/36841] Eliminate gfortran_sum_r8 call for calculation involving multidimensional array multiplication followed by a sum along first dimension

--- Comment #1 from rguenth at gcc dot gnu dot org 2008-07-23 09:40 --- Confirmed. The middle-end array work will address this in a generic way. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added ---

[Bug fortran/36842] Fortran: Minimize heap allocation of temporary arrays.

--- Comment #1 from rguenth at gcc dot gnu dot org 2008-07-23 09:43 --- Confirmed. Dependency analysis should see that no temporary is required here. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added --

[Bug target/36609] AVR wrong code using incorrect RTL for test reversal pattern

--- Comment #1 from hutchinsonandy at gcc dot gnu dot org 2008-07-23 10:43 --- Created an attachment (id=15945) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15945&action=view) Patch to target files http://gcc.gnu.org/ml/gcc-patches/2008-07/msg00391.html -- http://gcc.gnu.or

[Bug fortran/36894] question on gfortran character*(*)

--- Comment #5 from jhuddleston at hughes dot net 2008-07-23 10:49 --- thank you burnus for your time and analysis -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36894

[Bug bootstrap/36907] New: [4.4 Regression]: Gcc failed to bootstrap

Gcc 4.4 revision 138076 failed to boot strap on Linux/ia64: /export/gnu/import/svn/gcc-test/bld/./prev-gcc/xgcc -B/export/gnu/import/svn/gcc-test/bld/./prev-gcc/ -B/usr/local/ia64-unknown-linux-gnu/bin/ -c -DUSE_LIBUNWIND_EXCEPTIONS -g -O2 -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #1 from hjl dot tools at gmail dot com 2008-07-23 12:41 --- It also fails to boostrap on Linux/x86-64: libtool: compile: /export/gnu/import/svn/gcc-test/bld/./gcc/xgcc -B/export/gnu/import/svn/gcc-test/bld/./gcc/ -B/usr/local/x86_64-unknown-linux-gnu/bin/ -B/usr/local/x86_6

[Bug bootstrap/36908] New: bootstrap forever with BOOT_CFLAGS="-O2 -ftree-loop-distribution"

d with: ../configure --prefix=/abuild/tbily/install/ --enable-bootstrap --disable-multilib --enable-checks --enable-languages=c,c++ Thread model: posix gcc version 4.4.0 20080723 (experimental) (GCC) -- Summary: bootstrap forever with BOOT_CFLAGS="-O2 -ftree-loop-

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #2 from hjl dot tools at gmail dot com 2008-07-23 13:28 --- On ia64: Program received signal SIGSEGV, Segmentation fault. 0x40e16760 in ia64_variable_issue (dump=0x203f9e50, sched_verbose=0, insn=0x23fec6e0, can_issue_more=1) at ../../src/gcc/

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #3 from hjl dot tools at gmail dot com 2008-07-23 13:39 --- The problem on Fedora 9/x86-64 may be due to FILE is undeclared in extern void cl_optimization_print (FILE *, int, struct cl_optimization *); extern void cl_target_option_print (FILE *, int, struct cl_target_option

[Bug c/33877] Request for __VA_ARGC__

--- Comment #2 from aldot at gcc dot gnu dot org 2008-07-23 13:41 --- I'm curious as to why you think that would be useful? What would you do with it? -- aldot at gcc dot gnu dot org changed: What|Removed |Added

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #4 from hjl dot tools at gmail dot com 2008-07-23 13:46 --- (In reply to comment #3) > The problem on Fedora 9/x86-64 may be due to FILE is undeclared in > > extern void cl_optimization_print (FILE *, int, struct cl_optimization *); > extern void cl_target_option_print (FILE

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #5 from hjl dot tools at gmail dot com 2008-07-23 14:31 --- decode_options doesn't allow backend to override options, which breaks ia64 and maybe other targets. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36907

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #6 from gnu at the-meissners dot org 2008-07-23 15:22 --- Subject: Re: [4.4 Regression]: Gcc failed to bootstrap On Wed, Jul 23, 2008 at 01:46:30PM -, hjl dot tools at gmail dot com wrote: > > > --- Comment #4 from hjl dot tools at gmail dot com 2008-07-23 13:46

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #7 from hjl dot tools at gmail dot com 2008-07-23 15:33 --- (In reply to comment #6) > > I didn't have objective c, java, or ADA on the system. I'm going to kick off > a > build shortly with the following patch to see if it allows objc to build: > I don't believe you nee

[Bug fortran/36909] New: INTENT(IN) formal -> No need to call _gfortran_internal_unpack

Currently, even for a INTENT(IN) variable _gfortran_internal_unpack is called. D.1036 = _gfortran_internal_pack (&parm.1); foo (D.1036); if ((integer(kind=4)[0:] *) parm.1.data != (integer(kind=4)[0:] *) D.1036) { _gfortran_internal_unpack (&parm.1, D.1036); if (D

[Bug c/35058] -Werror= works only with some warnings

--- Comment #3 from manu at gcc dot gnu dot org 2008-07-23 15:58 --- Subject: Bug 35058 Author: manu Date: Wed Jul 23 15:57:49 2008 New Revision: 138089 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=138089 Log: 2008-07-23 Manuel Lopez-Ibanez <[EMAIL PROTECTED]> PR 35

[Bug c/35058] -Werror= works only with some warnings

--- Comment #4 from manu at gcc dot gnu dot org 2008-07-23 16:00 --- This should be fixed in GCC 4.4. If you find any new cases please REOPEN this with a testcase. -- manu at gcc dot gnu dot org changed: What|Removed |Added

[Bug fortran/29952] Flag to give runtime information " array temporary was created for argument"

--- Comment #3 from burnus at gcc dot gnu dot org 2008-07-23 16:12 --- Test case for the run-time check. Only for the second call a warning is needed. One way to add it is the following. One simply inserts if (parm.1.data != D.1036) printf("test.f90:5: In call to FOO, an array tem

[Bug c++/36910] New: missing diagnostic for initializing function pointer with incompatible exception specification

Gcc 4.3.1 fails to diagnose the invalid initialization of pf0 (nothrow) from f0 (may throw) in the following program. $ cat t.cpp && gcc -W -Wall -Wno-unused -c -pedantic t.cpp void f0 () { } void f1 () throw () { } void f2 () __attribute__ ((nothrow)); int main () { void (*pf0)() throw () =

[Bug fortran/35937] Wrong type for charlength of function

--- Comment #5 from pault at gcc dot gnu dot org 2008-07-23 16:28 --- (In reply to comment #4) Hi FX Your patch looks fine to me. In particular, you have tackled it at the right place. However, I have a question: why is this not sufficient? se.expr = gfc_evaluate_now (fold_convert (g

[Bug fortran/36909] INTENT(IN) formal -> No need to call _gfortran_internal_unpack

--- Comment #1 from burnus at gcc dot gnu dot org 2008-07-23 17:13 --- Can probably combined with fix of the run-time part of PR 29952. In gfc_conv_function_call one calls gfc_conv_array_parameter; currently the formal symbol is not passed and thus one cannot check it in gfc_conv_array_

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #8 from pinskia at gcc dot gnu dot org 2008-07-23 17:41 --- (In reply to comment #4) > It happens when building libobjc. The real underlying issue with libobjc was recorded as PR 24775. I have not time to work on it lately though. -- http://gcc.gnu.org/bugzilla/show_bu

[Bug c++/36911] New: Virtual function problem

When a virtual function is called from a base class constructor, it always calls the base class's version of the function, even if a derived class overrides it. Example code: #include using namespace std; class Base { public: Base(); virtual void overridable(); }; class Derived

[Bug c++/36911] Virtual function problem

--- Comment #1 from brian at dessent dot net 2008-07-23 18:10 --- Subject: Re: New: Virtual function problem The behavior is correct, see . -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36911

[Bug c++/36911] Virtual function problem

--- Comment #2 from pinskia at gcc dot gnu dot org 2008-07-23 18:14 --- Yes this is correct as you are calling the virtual function inside a constructor which means the virtual of that class is going to be called always ... This is the only way you can call a pure virtual function in fa

[Bug c/33877] Request for __VA_ARGC__

--- Comment #3 from Dave at Yost dot com 2008-07-23 18:33 --- This feature enables the declaration (via macro) of a variadic function that requires neither an argument containing a count of the number of variadic arguments (which is busywork clutter prone to unchecked error) nor an out-o

[Bug c++/36910] missing diagnostic for initializing function pointer with incompatible exception specification

--- Comment #1 from pinskia at gcc dot gnu dot org 2008-07-23 18:33 --- *** This bug has been marked as a duplicate of 12255 *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added --

[Bug c++/12255] exception-specifications unchecked during assignment of pointer to function

--- Comment #4 from pinskia at gcc dot gnu dot org 2008-07-23 18:33 --- *** Bug 36910 has been marked as a duplicate of this bug. *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added ---

[Bug debug/33740] gcc 4.1.1 on AIX - gdb and TV fail to see static value due to likely gcc compiler bug

--- Comment #3 from seppo at totalviewtech dot com 2008-07-23 18:46 --- Hi Andrew, Sorry about the delay in response. We had some issues w/ installation and this fell through the cracks. Anyways, I just tested gcc 4.2.3 and it fails the same way as 4.1.1. Thanks, Seppo XXX

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #9 from meissner at gcc dot gnu dot org 2008-07-23 19:04 --- Subject: Bug 36907 Author: meissner Date: Wed Jul 23 19:03:40 2008 New Revision: 138091 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=138091 Log: Fix PR 36907, breakage in building libobj with function sp

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #10 from gnu at the-meissners dot org 2008-07-23 19:06 --- Subject: Re: [4.4 Regression]: Gcc failed to bootstrap On Wed, Jul 23, 2008 at 01:46:30PM -, hjl dot tools at gmail dot com wrote: > > > --- Comment #4 from hjl dot tools at gmail dot com 2008-07-23 13:46

[Bug bootstrap/36907] [4.4 Regression]: Gcc failed to bootstrap

--- Comment #11 from hjl dot tools at gmail dot com 2008-07-23 20:16 --- If you don't want to check/change every target, you should make your changes to c-common.c optional. You only enable your changes for those targets you know they work correctly. -- http://gcc.gnu.org/bugzilla/

[Bug c++/17729] [4.2/4.3/4.4 Regression] Duplicate __attribute__((deprecated)) warning

--- Comment #12 from mrs at apple dot com 2008-07-23 21:12 --- radr://4232969 -- mrs at apple dot com changed: What|Removed |Added CC|

[Bug c++/36912] New: ICE when building with "-frounding-math -g"

static const double Eps = .001; static const double Tol = 1.0+Eps; int main(int, char *[]) { return 0; } Fedora 9. Works fine on MacOs 10.5 with gcc 4.0. setting Eps to be an integral value (but keeping its type to be double) makes the ICE go away. -- Summary: ICE when buildin

[Bug c++/36912] ICE when building with "-frounding-math -g"

--- Comment #1 from pinskia at gcc dot gnu dot org 2008-07-23 21:21 --- t.c:5: error: initializer for floating value is not a floating constant t.c:5: internal compiler error: tree check: expected real_cst, have plus_expr in output_constant, at varasm.c:4429 Please submit a full bug repo

[Bug c/36913] New: rewind does not work for files over 8GB

rewind does not work for files over 8GB (in my case). Closing the file and opening it again does work. -- Summary: rewind does not work for files over 8GB Product: gcc Version: 3.4.4 Status: UNCONFIRMED Severity: normal Priority: P

[Bug c/36913] rewind does not work for files over 8GB

--- Comment #1 from paolo dot carlini at oracle dot com 2008-07-24 01:14 --- rewind is a C library function, this is the Bugzilla for the GNU compiler collection, the wrong place to report about its possible misbehaviors. -- paolo dot carlini at oracle dot com changed: W

[Bug libgcj/27823] Found a problem with the JNI methods declared and implemented

--- Comment #24 from Riot dot Nrrrd dot Bugzilla at isolar dot DynDNS dot ORG 2008-07-24 01:58 --- (In reply to comment #0) > When I do make a gmake bootstrap of GCC 4.1.1 I have the following output: > > gmake[9]: Entering directory > `/opt/local/src/gcc-obj/sparc-sun-solaris2.9/sparc

[Bug c/36913] rewind does not work for files over 8GB

--- Comment #2 from m_olshansky at yahoo dot com 2008-07-24 02:04 --- Subject: Re: rewind does not work for files over 8GB Do you imply that it's source is universal? If so, can it be true that the GNU implementation of some lower level function called by 'rewind' is incorrect? --- O

[Bug c/36914] New: va_list is treatead as an array when complied in 64bits

Hi, I did a fair enough search to see if it an already reported bug, or non bug and I didn't find anything. Bellow is a simple program that illustrate the problem. The issue only happen when compiling in 64bits. With -m32 flag it works as expected. The issue seems to be when va_list in returned in

[Bug c/36914] va_list is treatead as an array when complied in 64bits

--- Comment #1 from fede dot naum at gmail dot com 2008-07-24 05:04 --- Created an attachment (id=15946) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15946&action=view) source file The source file containing the piece of code that is not able to compile in 64bits -- http://g

[Bug c/36914] va_list is treatead as an array when complied in 64bits

--- Comment #2 from fede dot naum at gmail dot com 2008-07-24 05:06 --- Created an attachment (id=15947) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15947&action=view) Preprocessed file test1.i The preprocessed file after running gcc -v -save-temps -m64 -Wall test1.c -o test1.o