--- Comment #16 from ghazi at gcc dot gnu dot org 2008-05-19 06:19 ---
Fixed in 4.4.0 by adding a new builtin for systems that care about the sign of
isinf's return value. E.g. do this:
#define isinf(x) __builtin_isinf_sign(x)
--
ghazi at gcc dot gnu dot org changed:
W
--
ghazi at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |ghazi at gcc dot gnu dot org
|dot org
--- Comment #15 from ghazi at gcc dot gnu dot org 2008-05-19 06:16 ---
Addressed this by adding __builtin_isinf_sign()
--
ghazi at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #68 from vmakarov at redhat dot com 2008-05-19 02:08 ---
The patch solving IRA problem is described in
http://gcc.gnu.org/ml/gcc-patches/2008-05/msg01093.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26854
--- Comment #67 from vmakarov at gcc dot gnu dot org 2008-05-19 02:03
---
Subject: Bug 26854
Author: vmakarov
Date: Mon May 19 02:02:52 2008
New Revision: 135523
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=135523
Log:
2008-05-18 Vladimir Makarov <[EMAIL PROTECTED]>
--- Comment #66 from vmakarov at redhat dot com 2008-05-19 02:00 ---
The problem with IRA was in too many allocnos to be chosen for spilling. The
most tome was spent in choosing the best allocno for spilling. The patch
solving the problem is coming.
--
http://gcc.gnu.org/bugzilla/
--- Comment #6 from Bernd dot Donner at gmx dot net 2008-05-19 00:18
---
Sorry, this is my fault your comments are correct.
--
Bernd dot Donner at gmx dot net changed:
What|Removed |Added
---
--- Comment #5 from paolo dot carlini at oracle dot com 2008-05-19 00:16
---
And you are certainly wrong that the Standard allows to omit the definition, in
this case. This is 9.4.2/4 oc C++03, note the last sentence:
If a static data member is of const integral or const enumeration ty
--- Comment #6 from paolo dot carlini at oracle dot com 2008-05-19 00:10
---
*** Bug 36261 has been marked as a duplicate of this bug. ***
--
paolo dot carlini at oracle dot com changed:
What|Removed |Added
--- Comment #4 from paolo dot carlini at oracle dot com 2008-05-19 00:10
---
*** This bug has been marked as a duplicate of 30745 ***
--
paolo dot carlini at oracle dot com changed:
What|Removed |Added
---
--- Comment #3 from Bernd dot Donner at gmx dot net 2008-05-19 00:00
---
(In reply to comment #2)
> In other terms, just add, after the declaration of struct A:
>
> const int A::N;
>
According to the C++ standard "static const" member variables can be declarend
_and_ defined immediat
--- Comment #14 from ghazi at gcc dot gnu dot org 2008-05-18 23:20 ---
Subject: Bug 35509
Author: ghazi
Date: Sun May 18 23:19:38 2008
New Revision: 135517
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=135517
Log:
PR middle-end/35509
* builtins.c (mathfn_built_
--- Comment #2 from fxcoudert at gcc dot gnu dot org 2008-05-18 23:07
---
Fixed on 4.4 and 4.3.
--
fxcoudert at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #1 from hakonrk at fys dot uio dot no 2008-05-18 23:07 ---
Created an attachment (id=15653)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15653&action=view)
Uses gigabytes of memory when compiled with optimizations on 4.3.0
--
http://gcc.gnu.org/bugzilla/show_bug.c
While compiling the NES emulator FCE Ultra for my new Fedora 9 system (x86-64),
I noticed that gcc 4.3.0 (or at least Fedora's version of it) used so much
memory on some of the files that my system almost crashed. The preprocessed
source code that trigger this bug is attached here. This seems to
--- Comment #2 from paolo dot carlini at oracle dot com 2008-05-18 22:58
---
In other terms, just add, after the declaration of struct A:
const int A::N;
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36261
--- Comment #1 from pinskia at gcc dot gnu dot org 2008-05-18 22:52 ---
std::min(1, A::N)
std::min takes a reference to a const type which means you need the defintion
of A::N including the definition.
--
pinskia at gcc dot gnu dot org changed:
What|Removed
This small program cannot be compiled. It failes with an "undefined reference
to `A::N'" error. Replacing the main function line with:
std::cout << "Number: " << std::min(1, (int)A::N) << "\n";
results in a successful compilation.
This is the small program to reproduce the bug:
#include
#includ
--- Comment #2 from reichelt at gcc dot gnu dot org 2008-05-18 21:57
---
Fixed on mainline - probably by the patch for PR 35331.
--
reichelt at gcc dot gnu dot org changed:
What|Removed |Added
--
--- Comment #3 from pault at gcc dot gnu dot org 2008-05-18 21:02 ---
Frank,
I am downgrading this to 'normal' severity - 'major' is reserved for fairly
cataclysmic bugs for the whole of gcc. 'ice-on-valid-code' will catch the
attention of gfortran developers.
I'll try to come back to
--- Comment #44 from jakub at gcc dot gnu dot org 2008-05-18 20:24 ---
The bug I've reported is fixed both on the trunk and on the branch, although
with different patches.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #43 from jakub at gcc dot gnu dot org 2008-05-18 20:20 ---
Subject: Bug 36090
Author: jakub
Date: Sun May 18 20:19:55 2008
New Revision: 135509
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=135509
Log:
PR target/36090
* gcc.c-torture/execute/20080502
--- Comment #42 from jakub at gcc dot gnu dot org 2008-05-18 20:19 ---
Subject: Bug 36090
Author: jakub
Date: Sun May 18 20:19:03 2008
New Revision: 135508
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=135508
Log:
PR target/36090
* gcc.c-torture/execute/20080502
See http://gcc.gnu.org/ml/fortran/2008-05/msg00207.html
Currently, the keywords used in intrinsic.texi don't match intrinsic.c.
(So far for all non-vendor extensions, intrinsic.c was correct.)
TODO:
- FSTAT: Fix intrinsic.{texi,c} argument-name mismatch
- FREE: Fix intrinsic.{texi,c}
--- Comment #41 from jakub at gcc dot gnu dot org 2008-05-18 18:31 ---
The bug is fixed there (though with the safer rs6000 specific fix), just the
testcase hasn't been committed. I'll do that later tonight.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36090
On Sun, May 18, 2008 at 11:24 AM, John Kacur <[EMAIL PROTECTED]> wrote:
> int main(void)
> {
> unsigned j = ~0 >> 8;
The above is signed right shift which will keep the sign bit and replicate it.
> unsigned i = ~0;
> i >>= 8;
While this is an unsigned right shift.
-- P
cat rshift.c
#include
#include
int main(void)
{
unsigned j = ~0 >> 8;
unsigned i = ~0;
i >>= 8;
printf("i = %x\n", i);
printf("j = %x\n", j);
exit(0);
}
gcc -o rshift rshift.c -Wall -g
[EMAIL PROTECTED]:~/b1systems/ckurs> ./rshift
i = ff
j
--- Comment #4 from jv244 at cam dot ac dot uk 2008-05-18 17:59 ---
the end of the backtrace (notice the depth) might be useful as well:
#78681 0x00486421 in gt_ggc_mx_lang_tree_node (x_p=) at ./gt-fortran-f95-lang.h:290
#78682 0x00632f85 in gt_ggc_mx_cgraph_node (x_p=)
/tmp/ccJOx9ry.s
GNU Fortran (GCC) version 4.4.0 20080518 (experimental) [trunk revision 135494]
(x86_64-unknown-linux-gnu)
compiled by GNU C version 4.4.0 20080518 (experimental) [trunk revision
135494], GMP version 4.2.1, MPFR version 2.2.1.
GGC heuristics: --param ggc-min-expand=30
--- Comment #2 from dominiq at lps dot ens dot fr 2008-05-18 17:51 ---
I got similar errors for an incremental update:
...
/opt/gcc/i686-darwin/./prev-gcc/xgcc -B/opt/gcc/i686-darwin/./prev-gcc/
-B/opt/gcc/gcc4.4w/i686-apple-darwin9/bin/ -c -g -O2 -fomit-frame-pointer
-DIN_GCC -W -Wa
--- Comment #1 from ubizjak at gmail dot com 2008-05-18 17:02 ---
Hm, I have reverted my TARGET_SECONDARY_RELOAD patch while this issue is being
investigated. For some reason, copying %esi to Q_REGS class doesn't happen...
Fixed by reverting the problematic patch for now.
--
ubizjak
--- Comment #1 from rguenth at gcc dot gnu dot org 2008-05-18 16:59 ---
Zdenek probably remembers why this was done. In any case, posting a patch
to gcc-patches would be appropriate.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #9 from rguenth at gcc dot gnu dot org 2008-05-18 16:58 ---
Did you investigate whether IRA fixes this issue?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36222
--- Comment #4 from pault at gcc dot gnu dot org 2008-05-18 16:00 ---
(In reply to comment #3)
The patch that I have been working on, which is held up by memory leak
problems, actually cures this fellow too.
I am active again on all the allocatable component bugs.
Watch this space.
Pa
--- Comment #8 from uweigand at gcc dot gnu dot org 2008-05-18 15:58
---
That special case in find_reloads is really about a different situation.
We do not have a simple move here.
The problem also is not really related to vector instruction in particular;
reload doesn't at all care wh
--- Comment #13 from ghazi at gcc dot gnu dot org 2008-05-18 15:51 ---
Patch to implement a separate builtin isinf_sign posted here:
http://gcc.gnu.org/ml/gcc-patches/2008-05/msg01067.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35509
--- Comment #3 from pault at gcc dot gnu dot org 2008-05-18 15:37 ---
> I have to admit, I did not quickly see in the dump (-fdump-tree-original) why
> it is failing; at a glance, both calling "neg_at" in MAIN__ and the assignment
> in "neg_at" itself look ok; and "t2 = -t1" also works.
win9
--host=i686-apple-darwin9 --target=i686-apple-darwin9 --with-gnu-as
--with-tune=generic --prefix=/opt/gnu/gcc/gcc-4.4.0 --enable-debug=no
--disable-nls --enable-languages=c,c++,objc,fortran,ada,obj-c++,java
Thread model: posix
gcc version 4.4.0 20080518 (experimental) [trunk revision 135503]
--- Comment #3 from fxcoudert at gcc dot gnu dot org 2008-05-18 14:55
---
I have some more information, but it still doesn't make sense. I'm comparing
the codepaths taken in the front-end by:
implicit none
character(len=5), dimension(3,3), parameter :: &
p = reshape(["a", "b",
--- Comment #1 from charlet at gcc dot gnu dot org 2008-05-18 11:15 ---
--enable-threads=gnat is obsolete and will be removed as soon as someone
submits a patch to that effect. Your report simply shows that nobody is
using it these days (and use the default instead).
Arno
--
charlet
--- Comment #7 from burnus at gcc dot gnu dot org 2008-05-18 11:11 ---
Fixed on the trunk (4.4.0). Thanks for the patch.
--
burnus at gcc dot gnu dot org changed:
What|Removed |Added
-
--- Comment #6 from burnus at gcc dot gnu dot org 2008-05-18 11:10 ---
Subject: Bug 36251
Author: burnus
Date: Sun May 18 11:10:11 2008
New Revision: 135495
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=135495
Log:
2008-05-18 Steven G. Kargl <[EMAIL PROTECTED]>
PR fo
--- Comment #2 from tkoenig at gcc dot gnu dot org 2008-05-18 10:42 ---
The problem is with the array parameter.
$ cat spread.f90
implicit none
character(len=5), dimension(3,3), parameter :: &
p = reshape(["a", "b", "c", "d", "e", "f", "g", "h", "i"], [3,3])
character(len=5),
--- Comment #1 from pault at gcc dot gnu dot org 2008-05-18 10:03 ---
Confirmed.
This is quite revealing...
implicit none
character(len=5), dimension(3,3), parameter :: &
p = reshape(["a", "b", "c", "d", "e", "f", "g", "h", "i"], [3,3])
character(len=5), dimension(3,3) :: m1
The function copy_loop_headers after duplicate the region for a loop,
237 /* Ensure that the latch and the preheader is simple (we know that
they
238 are not now, since there was the loop exit condition. */
239 split_edge (loop_preheader_edge (loop));
240 split_edge (lo
45 matches
Mail list logo
