[Bug testsuite/33082] [4.3 Regression] Revision 127491 causes FAIL: gcc.dg/dfp/convert-bfp-fold.c (test for excess errors)

--- Comment #3 from ubizjak at gmail dot com 2007-08-16 07:12 --- Patch at http://gcc.gnu.org/ml/gcc-patches/2007-08/msg00982.html -- ubizjak at gmail dot com changed: What|Removed |Added

[Bug libstdc++/33084] New: Small typo in valarray header

There appears to be a simple typo in header valarray. Loction is Line 1012 in file libstdc++-v3/include/std/valarray in the current svn trunc in the third of three operators defined by the macro _DEFINE_BINARY_OPERATOR I think this line should be changed from return _Expr<_Closure, _Tp>(_Closu

[Bug c++/31132] [4.1/4.2/4.3 regression] ICE on inconsistent friend declaration

--- Comment #3 from paolo at gcc dot gnu dot org 2007-08-16 09:05 --- Subject: Bug 31132 Author: paolo Date: Thu Aug 16 09:05:17 2007 New Revision: 127535 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127535 Log: /cp 2007-08-16 Paolo Carlini <[EMAIL PROTECTED]> PR c+

[Bug c++/31132] [4.1/4.2 regression] ICE on inconsistent friend declaration

--- Comment #4 from pcarlini at suse dot de 2007-08-16 09:08 --- Fixing the problem on 4_1-branch and 4_2-branch depends on fixing 27211... -- pcarlini at suse dot de changed: What|Removed |Added

[Bug libstdc++/33084] Small typo in valarray header

--- Comment #1 from pcarlini at suse dot de 2007-08-16 09:41 --- Thanks. This isn't a regression, I'm going to apply your patch to mainline and 4_2-branch only. -- pcarlini at suse dot de changed: What|Removed |Added ---

[Bug c++/33012] ICE on throwing copy of object returned by reference from method with traits-deduced return-type

--- Comment #3 from raymond at corvil dot com 2007-08-16 09:50 --- Yes, it does not happen with recent releases. I logged it because I couldn't find any hint of this problem having been seen and noted anywhere. I just wanted to let you know about it in case the problem has simply been

[Bug middle-end/32897] Invalid rematerialisation of subregs

--- Comment #1 from rsandifo at gcc dot gnu dot org 2007-08-16 10:16 --- Subject: Bug 32897 Author: rsandifo Date: Thu Aug 16 10:16:15 2007 New Revision: 127536 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127536 Log: gcc/ PR middle-end/32897 * reload.c (find_

[Bug middle-end/179] gcc -O2 -Wuninitialized missing warning with &var under 2.95.x and 3.x

--- Comment #10 from manu at gcc dot gnu dot org 2007-08-16 10:19 --- Some analysis http://gcc.gnu.org/ml/gcc/2007-08/msg00271.html -- manu at gcc dot gnu dot org changed: What|Removed |Added

[Bug middle-end/32897] Invalid rematerialisation of subregs

--- Comment #2 from rsandifo at gcc dot gnu dot org 2007-08-16 10:21 --- Patch applied. -- rsandifo at gcc dot gnu dot org changed: What|Removed |Added Statu

[Bug libgcj/33085] New: liblt_prog_compiler_pic_GCJ='-DDLL_EXPORT' is wrong

Bug reported by Ross Ridge to mingw users list When building libgcj for windows targets, libtool puts -DDLL_EXPORT in compiler flag: mingw* | cygwin* | pw32* | os2*) # This hack is so that the source file can tell whether it is being # built for inclusion in a dll (and should exp

[Bug c/33083] noreturn function should be invoked via JMP

--- Comment #1 from pinskia at gcc dot gnu dot org 2007-08-16 10:38 --- *** This bug has been marked as a duplicate of 10837 *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added --

[Bug rtl-optimization/10837] noreturn attribute causes no sibling calling optimization

--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-16 10:38 --- *** Bug 33083 has been marked as a duplicate of this bug. *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added ---

[Bug middle-end/33086] New: warn for read-only uninitialized variables passed as arguments

void use(const int *); void foo(void) { int i; use(&i); } At least for languages where 'const' is actually enforced, we should warn for this. For languages where the 'const' can be cast away and 'i' can be initialized by 'use' the situation is less clear (although personally I think we should

[Bug other/33087] New: Add dependency information to object files and make use of it

It would be very nice if GCC could save the dependency information for an object file in the file itself (ideally including the checksums of the source files used to build it) and optionally check this information when compiling, so that it would skip files which are up-to-date. I think that this

[Bug other/33087] Add dependency information to object files and make use of it

--- Comment #1 from raselmsh at hotmail dot com 2007-08-16 10:55 --- Not related to the bug, but is there a way of hiding e-mail addresses in this bugzilla? I have switched from my real address to a more or less disposable one to get around this, but it would be nicer to be able to use

[Bug middle-end/10138] warn for uninitialized arrays passed as const* arguments

--- Comment #9 from manu at gcc dot gnu dot org 2007-08-16 10:57 --- Let's simplify this report. This one is now about int atoi(const char *); int foo() { char buf[10]; return atoi(buf); } As comment #3 mentions, this is a combination of 1) Report use of uninitialized array

[Bug middle-end/33086] warn for read-only uninitialized variables passed as arguments

--- Comment #1 from pinskia at gcc dot gnu dot org 2007-08-16 10:59 --- > At least for languages where 'const' is actually enforced There is none, unless you are taking about fortran "in" arguments. So we need to mark such argument as special. Now if you have the full program (or at l

[Bug middle-end/33086] warn for read-only uninitialized variables passed as arguments

-- pinskia at gcc dot gnu dot org changed: What|Removed |Added CC||pinskia at gcc dot gnu dot |

[Bug other/33087] Add dependency information to object files and make use of it

--- Comment #2 from pinskia at gcc dot gnu dot org 2007-08-16 11:02 --- You can use ccache to get this same behavior. You can also use -MD to get automatic dependency checking. >ideally including the checksums of the source files used to build it That will not work for some stuff wher

[Bug other/33087] Add dependency information to object files and make use of it

--- Comment #3 from raselmsh at hotmail dot com 2007-08-16 11:14 --- How reliable is ccache? I have tried it (or was it a different C Cache?), but uninstalled after I found that it caused some compile runs to fail, presumably after misselecting the object file from the cache. To clarif

[Bug middle-end/33086] warn for read-only uninitialized variables passed as arguments

--- Comment #2 from manu at gcc dot gnu dot org 2007-08-16 11:19 --- (In reply to comment #1) > > At least for languages where 'const' is actually enforced > > There is none, void use(const int *a) { a[0] = 5; } void foo(void) { int i; use(&i); } new.c:3: error: assignment of

[Bug fortran/20896] [4.2 and 4.1 only] ambiguous interface not detected

--- Comment #9 from fxcoudert at gcc dot gnu dot org 2007-08-16 11:50 --- (In reply to comment #8) > Paul, what is the status on this one? ping? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20896

[Bug fortran/19292] [meta-bug] g77 features lacking in gfortran

--- Comment #31 from fxcoudert at gcc dot gnu dot org 2007-08-16 12:10 --- Marking this as FIXED, see http://gcc.gnu.org/ml/fortran/2007-08/msg00327.html for details. -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added ---

[Bug middle-end/33088] New: [4.1/4.2/4.3 Regression] spurious exceptions with -ffloat-store

Since RTH's patch , stores to the real or imaginary part of a complex variable are represented using a load of the other part and a COMPLEX_EXPR. If the variable is not yet initialized, and the load of the other part does not get optimized a

[Bug libfortran/33079] Optional empty strings do not appear to be 'PRESENT'

-- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added URL||http://gcc.gnu.org/ml/gcc- |

[Bug libfortran/32989] GETARG intrinsic

--- Comment #2 from fxcoudert at gcc dot gnu dot org 2007-08-16 13:28 --- Well, the gfortran doc says "Shall be of type INTEGER(4), N >= 0". This is in contradiction with the g77 doc, which says: "Pos: INTEGER not wider than the default kind; scalar; INTENT(IN)." Note that the argument

[Bug libfortran/33079] Optional empty strings do not appear to be 'PRESENT'

--- Comment #4 from patchapp at dberlin dot org 2007-08-16 13:25 --- Subject: Bug number PR33079 A patch for this bug has been added to the patch tracker. The mailing list url for the patch is http://gcc.gnu.org/ml/gcc-patches/2007-08/msg01012.html -- http://gcc.gnu.org/bugzilla/sh

[Bug rtl-optimization/33089] New: combine not combining add + test in if (a + b) > 0

Consider this simple test case: extern void foo (void); extern void bar (void); void bgtutest (unsigned int a, unsigned int b) { if (a + b > 0) foo (); else bar (); } gcc mainline revision 127532 -O2 -S -dp: bgtutest: .LFB2: addl%edi, %esi # 7 *addsi_1/1 [l

[Bug middle-end/33029] [4.3 Regression] libgcc2.c:1890: internal compiler error: in local_cprop_pass, at gcse.c:3236

--- Comment #5 from steven at gcc dot gnu dot org 2007-08-16 14:35 --- Starting program: /home/steven/devel/build-test/gcc/cc1 -g -O2 -fPIC -Dpa64=1 -mlong-calls -fkeep-inline-functions libgcc2.i __multc3 Analyzing compilation unit Performing interprocedural optimizations Assembli

[Bug c/23872] .t02.original dump weirdness

--- Comment #4 from manu at gcc dot gnu dot org 2007-08-16 14:39 --- I think this is confirmed. Patches welcome! -- manu at gcc dot gnu dot org changed: What|Removed |Added --

[Bug middle-end/33029] [4.3 Regression] libgcc2.c:1890: internal compiler error: in local_cprop_pass, at gcse.c:3236

--- Comment #6 from steven at gcc dot gnu dot org 2007-08-16 14:54 --- Created an attachment (id=14064) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14064&action=view) remove nested libcall handling from gcse.c This is the *only* place in the entire compiler where GCC handles nes

[Bug middle-end/33086] warn for read-only uninitialized variables passed as arguments

--- Comment #3 from pinskia at gcc dot gnu dot org 2007-08-16 14:58 --- void use(const int *a) { int *b = (int*)a; b[0] = 5; } void foo(void) { int i; use(&i); } -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33086

[Bug middle-end/33029] [4.3 Regression] libgcc2.c:1890: internal compiler error: in local_cprop_pass, at gcse.c:3236

--- Comment #7 from steven at gcc dot gnu dot org 2007-08-16 15:05 --- Fix for the bug: Index: lower-subreg.c === --- lower-subreg.c (revision 127558) +++ lower-subreg.c (working copy) @@ -897,7 +897,7 @@ resolve_

[Bug fortran/33090] New: Unable to build on AIX 4.3.3

gfortran won't build on AIX 4.3.3 as it doesn't have weak support: ... /opt/build/gcc-4.1.2-objdir/./gcc/xgcc -B/opt/build/gcc-4.1.2-objdir/./gcc/ -B/o pt/TWWfsw/gcc41/powerpc-ibm-aix4.3.3.0/bin/ -B/opt/TWWfsw/gcc41/powerpc-ibm-aix4 .3.3.0/lib/ -isystem /opt/TWWfsw/gcc41/powerpc-ibm-aix4.3.3.0/incl

[Bug libfortran/32989] GETARG intrinsic

--- Comment #3 from burnus at gcc dot gnu dot org 2007-08-16 15:19 --- > Well, the gfortran doc says "Shall be of type INTEGER(4), N >= 0". This is in > contradiction with the g77 doc, which says: "Pos: INTEGER not wider than the > default kind; scalar; INTENT(IN)." Well, N >= 0 vs. pos

[Bug middle-end/22456] [4.1/4.2/4.3 regression] missing "is used uninitialized" warning

--- Comment #15 from dnovillo at gcc dot gnu dot org 2007-08-16 15:22 --- If not an exact duplicate, it's strongly related to 18501. The code pattern is slightly different so it may be worth keeping around. Adding a dependency on 18501. -- dnovillo at gcc dot gnu dot org changed:

[Bug c++/33091] New: [c++0x] ICE using remove_reference on variadic param pack

Environment: System: Linux coffeebuzz.miovision.corp 2.6.18-8.1.8.el5 #1 SMP Tue Jul 10 06:39:17 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux Architecture: x86_64 host: x86_64-unknown-linux-gnu build: x86_64-unknown-linux-gnu target: x86_64-unknown-linux-gnu configured with: ../gcc/configure CC=gcc43 C

[Bug c++/33091] [c++0x] ICE using remove_reference on variadic param pack

--- Comment #1 from chris dot fairles at gmail dot com 2007-08-16 15:57 --- Created an attachment (id=14065) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14065&action=view) source file that produces ICE attached source file that causes ICE -- http://gcc.gnu.org/bugzilla/sho

[Bug c++/33091] [c++0x] ICE using remove_reference on variadic param pack

--- Comment #2 from chris dot fairles at gmail dot com 2007-08-16 15:58 --- Created an attachment (id=14066) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14066&action=view) -save-temps compiler output -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33091

[Bug c++/33091] [c++0x] ICE using remove_reference on variadic param pack

--- Comment #3 from chris dot fairles at gmail dot com 2007-08-16 15:59 --- Created an attachment (id=14067) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14067&action=view) -save-temps compiler output -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33091

[Bug middle-end/33092] New: Using -O1 -fno-tree-salias results in ICE

While turning off optimizations to try and track down a compiler bug I noticed that if I used -fno-tree-salias with -O1 the compiler would ICE: x.c: In function 'main': x.c:1: internal compiler error: in verify_curr_properties, at passes.c:1035 Please submit a full bug report, with preprocessed so

[Bug c++/33025] [4.3 Regression] Wrong calling of placement new with conditionals

-- jakub at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |jakub at gcc dot gnu dot org |dot org

[Bug c++/33091] [c++0x] ICE using remove_reference on variadic param pack

--- Comment #4 from chris dot fairles at gmail dot com 2007-08-16 16:16 --- Adding: template tuple(const tuple<_UElements...>& __in, typename std::enable_if< std::is_same< std::integral_constant, std::integral_constant>::value>::type* = 0) : _Inherited(__in) { } (whether its

[Bug middle-end/33092] [4.3 Regrsssion] Using -O1 -fno-tree-salias results in ICE

-- pinskia at gcc dot gnu dot org changed: What|Removed |Added CC||dberlin at gcc dot gnu dot |

[Bug fortran/33072] "use mod, only: operator(.sub.)" matches any procedure "sub"

--- Comment #2 from patchapp at dberlin dot org 2007-08-16 16:50 --- Subject: Bug number PR33072 A patch for this bug has been added to the patch tracker. The mailing list url for the patch is http://gcc.gnu.org/ml/gcc-patches/2007-08/msg01039.html -- http://gcc.gnu.org/bugzilla/sh

[Bug middle-end/33092] [4.3 Regrsssion] Using -O1 -fno-tree-salias results in ICE

--- Comment #1 from dberlin at gcc dot gnu dot org 2007-08-16 17:37 --- Subject: Re: [4.3 Regrsssion] Using -O1 -fno-tree-salias results in ICE Yeah, we either need to remove salias, or force create an empty may_alias pass that returns TODO_may_alias but does nothing else. I'm not sur

[Bug fortran/33072] "use mod, only: operator(.sub.)" matches any procedure "sub"

--- Comment #3 from burnus at gcc dot gnu dot org 2007-08-16 18:17 --- Subject: Bug 33072 Author: burnus Date: Thu Aug 16 18:17:46 2007 New Revision: 127564 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127564 Log: 2007-08-16 Tobias Burnus <[EMAIL PROTECTED]> PR fort

[Bug fortran/33072] "use mod, only: operator(.sub.)" matches any procedure "sub"

--- Comment #4 from burnus at gcc dot gnu dot org 2007-08-16 18:20 --- FIXED - but only the initially reported bug. The other I push back to PR31298. -- burnus at gcc dot gnu dot org changed: What|Removed |Added

[Bug libfortran/32972] performance of pack/unpack

--- Comment #4 from tkoenig at gcc dot gnu dot org 2007-08-16 18:16 --- Hi Dominique, would you be willing to give the patch at http://gcc.gnu.org/ml/fortran/2007-08/msg00295.html a spin on your Mac to check whether this is OK? That would be great, because I don't have a big-endian s

[Bug testsuite/33082] [4.3 Regression] Revision 127491 causes FAIL: gcc.dg/dfp/convert-bfp-fold.c (test for excess errors)

--- Comment #4 from uros at gcc dot gnu dot org 2007-08-16 18:30 --- Subject: Bug 33082 Author: uros Date: Thu Aug 16 18:30:14 2007 New Revision: 127565 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127565 Log: PR testsuite/33082 * gcc.dg/dfp/convert-dfp-fold.c:

[Bug testsuite/33082] [4.3 Regression] Revision 127491 causes FAIL: gcc.dg/dfp/convert-bfp-fold.c (test for excess errors)

--- Comment #5 from ubizjak at gmail dot com 2007-08-16 18:33 --- Fixed. -- ubizjak at gmail dot com changed: What|Removed |Added Status|UNCONFIRMED

[Bug bootstrap/33065] warning in Comparing stages 2 and 3: cd: stage3-gcc: No such file or directory

--- Comment #7 from mbo dot massimo at tiscali dot it 2007-08-16 19:27 --- (From update of attachment 14063) make logs generated building gcc in a directory different from source directory -- mbo dot massimo at tiscali dot it changed: What|Removed

[Bug target/33011] [4.3 Regression] frv: error: bad output_move_single operand

--- Comment #3 from rask at gcc dot gnu dot org 2007-08-16 19:49 --- Revision 127532 works. -- rask at gcc dot gnu dot org changed: What|Removed |Added Status

[Bug c++/33093] New: ICE in C++ frontend with constant defined in template specialization

When I compile this test case with gcc 4.1, 4.2, or mainline, I get an ICE: foo.cc: In instantiation of ‘const unsigned char B::c’: foo.cc:20: instantiated from ‘const unsigned char C >::c’ foo.cc:26: instantiated from here foo.cc:20: internal compiler error: in instantiate_decl, at cp/pt.c:14

[Bug c++/33093] ICE in C++ frontend with constant defined in template specialization

--- Comment #1 from pinskia at gcc dot gnu dot org 2007-08-16 21:01 --- >As far as I can see this is valid C++ code. It is not but the way to fix it is doing: template<> const unsigned char B::c = 1; - cut -- This is a dup of bug 24791 *** This bug has been marked as a duplicat

[Bug c++/24791] ICE on invalid instantiation of template's static member

--- Comment #11 from pinskia at gcc dot gnu dot org 2007-08-16 21:01 --- *** Bug 33093 has been marked as a duplicate of this bug. *** -- pinskia at gcc dot gnu dot org changed: What|Removed |Added -

[Bug c++/31751] [4.1/4.2 regression] ICE with forgotten member declaration

--- Comment #2 from reichelt at gcc dot gnu dot org 2007-08-16 22:42 --- Hi Paolo, this was apparently fixed by your patch for PR27211 (and not PR33035). A backport would be nice to fix the regression here. -- reichelt at gcc dot gnu dot org changed: What|Removed

[Bug fortran/33090] Unable to build on AIX 4.3.3

--- Comment #1 from pinskia at gcc dot gnu dot org 2007-08-17 00:32 --- Huh? The warnings should be ok. What exact error are you getting because I don' t see -Werror on the command line so the warnings should not be stoping the build? -- pinskia at gcc dot gnu dot org changed:

[Bug target/33075] (mingw) Dwarf2-EH causes stack leak with VLA

--- Comment #4 from pinskia at gcc dot gnu dot org 2007-08-17 00:36 --- >I think this may be related to the deallocation problem reported in PR 19771 It is unrelated as the BIND_EXPR is there for this testcase and so is the builtin to dealloc the memory. -- pinskia at gcc dot gnu do

[Bug c/30013] Multiple flaws in decimal floating-point arithmetic conversions fixed

--- Comment #4 from janis at gcc dot gnu dot org 2007-08-17 00:40 --- Ben just mentioned this PR to me, I hadn't seen it before. I'm working on a patch to support TFmode for powerpc*-linux, and I'll talk to HJ about proper support for XFmode. Initially we didn't support long double bec

[Bug c++/33094] New: ICE on valid C++ virtual template static member in namespace

This C++ code gets an ICE with gcc 4.2 and with mainline. I believe this is valid C++ code. namespace { template class A { virtual T f1() { return c; } static const T c = 0; }; A v; } foo.cc: In instantiation of ‘const int ::A::c’: foo.cc:7: instantiated from ‘T::A::f1() [with T = int]

Jeremie added you as a friend

Jeremie sent you a friend request on MySpace. Click this link to join and Jeremie will be added as your friend: http://www1.myspace.com/reloc.cfm?c=2&id=A8516BAB-FB7B-4C88-9CFA-C0ABB2253419 Thanks, your friends at MySpace - This email was sent by someone who kno

[Bug middle-end/32970] [4.3 Regression] C++ frontend can not handle vector pointer constant parameter

--- Comment #8 from bje at gcc dot gnu dot org 2007-08-17 05:24 --- Subject: Bug 32970 Author: bje Date: Fri Aug 17 05:24:24 2007 New Revision: 127578 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=127578 Log: PR middle-end/32970 gcc/ * tree.c (reconstruct_comple

[Bug fortran/33095] New: MAX with optional arguments gives run-time error

This is a fallout from PR31198. Using a = MAX(1,2, a1, a2) where a1 and a2 are optional arguments: If both a1 and a2 are not present, gfortran gives a bogus run-time error message: Fortran runtime error: Second argument of 'max' intrinsic should be present Looking at the dump, gfortran optimi

[Bug middle-end/32970] [4.3 Regression] C++ frontend can not handle vector pointer constant parameter

--- Comment #9 from bje at gcc dot gnu dot org 2007-08-17 06:49 --- Fixed in r127578. -- bje at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONF