> Hi Eric, Hi Richard,
>
> We recently ran across an ICE building a target libiberty for one of
> the multilibs of the mips64vrel-elf toolchain:
>
> .../libiberty/regex.c: In function 'byte_re_match_2_internal':
> .../libiberty/regex.c:7481: error: insn does not satisfy its constraints:
Hi Eric, Hi Richard,
We recently ran across an ICE building a target libiberty for one of
the multilibs of the mips64vrel-elf toolchain:
.../libiberty/regex.c: In function 'byte_re_match_2_internal':
.../libiberty/regex.c:7481: error: insn does not satisfy its constraints:
(insn 52
--- Comment #6 from pinskia at gcc dot gnu dot org 2007-03-29 07:30 ---
Fixed.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Status|NEW
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-03-29 07:15 ---
I am going to mark this as a rtl optimization issue because the tree level is
correct unless we get a PLUS_EXPR which is able to wrap without being
undefined.
--
pinskia at gcc dot gnu dot org changed:
--- Comment #5 from pinskia at gcc dot gnu dot org 2007-03-29 01:41 ---
Reduced self contained testcase:
Module NsLin
Real*8, Dimension(54,54) :: VXX, VXY,
& VYX, VYY, VXX2, VXY2, VYX2, VYY2
End Module NsLin
Subroutine SysNSL (N, NX, NY)
--- Comment #43 from janis at gcc dot gnu dot org 2007-03-29 01:12 ---
A couple of days ago in irc I agreed to come up with a version of the patch
that just handles the C tests. So far it works fine with C but breaks
everything else, but I haven't forgotten about it.
--
http://gcc.
--- Comment #1 from fxcoudert at gcc dot gnu dot org 2007-03-28 22:50
---
Same thing happens for local variables, character arguments and character
functions... Example of the first case:
subroutine foo(i)
integer(kind=8), intent(in) :: i
character(len=i) :: x
print *, len(x)
en
--- Comment #2 from pinskia at gcc dot gnu dot org 2007-03-28 22:49 ---
Actually I think the inline-asm should be:
#define CMOV_COMBO( val1, val2 ) \
asm volatile ( \
"cmpl %0, %3\n\t" \
"cmovl %3, %0\n\t" \
--- Comment #1 from ramiro at lisha dot ufsc dot br 2007-03-28 22:39
---
Created an attachment (id=13296)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13296&action=view)
.c, .i and .s files
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31386
This is reproduceable with gcc 4.0.0 and gcc SVN, but not with gcc 3.4.6.
I still don't really know what's the problem, but I think it may be that not
enough registers are available for output, and gcc is using the same register
for 2 different outputs.
gccs were built with:
--prefix=/usr --enabl
--- Comment #37 from anlauf at gmx dot de 2007-03-28 22:22 ---
(In reply to comment #36)
> Closing, will not backport to 4.2 unless someone feels strongly about it.
Jerry,
I hate to bother you, but it is not really fixed.
Add a comment line at the end of the namelist file gfcbug61a.nm
--- Comment #4 from simartin at gcc dot gnu dot org 2007-03-28 22:10
---
Fixed on the mainline.
--
simartin at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #3 from simartin at gcc dot gnu dot org 2007-03-28 22:06
---
Subject: Bug 29077
Author: simartin
Date: Wed Mar 28 22:06:01 2007
New Revision: 123312
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=123312
Log:
2007-03-28 Simon Martin <[EMAIL PROTECTED]>
PR
--- Comment #6 from tkoenig at gcc dot gnu dot org 2007-03-28 20:47 ---
Hi Jerry,
there is still the issue with kind=10 reals written with
CONVERT="swap".
Reopening until we've decided what to do with that :-)
Thomas
--
tkoenig at gcc dot gnu dot org changed:
What|
--- Comment #9 from dgregor at gcc dot gnu dot org 2007-03-28 19:46 ---
This problem has been solved by 16-bit tree codes:
http://gcc.gnu.org/ml/gcc-patches/2007-03/msg01721.html
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
---
--- Comment #3 from brooks at gcc dot gnu dot org 2007-03-28 19:17 ---
I had previously posted a patch to fix this, as well, but apparently the patch
tracker didn't pick up on it to attach it to the PR:
http://gcc.gnu.org/ml/gcc-patches/2007-03/msg01658.html
As mentioned on the patches
--- Comment #16 from pinskia at gcc dot gnu dot org 2007-03-28 19:24
---
(In reply to comment #15)
> maybe related: http://sourceware.org/bugzilla/show_bug.cgi?id=4267
It is unrelated really.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31364
--- Comment #15 from debian-gcc at lists dot debian dot org 2007-03-28
19:23 ---
maybe related: http://sourceware.org/bugzilla/show_bug.cgi?id=4267
building current 4.2 branch configured with --enable-secureplt
--enable-libstdcxx-debug, with current binutils trunk results in an ld segf
--- Comment #1 from brooks at gcc dot gnu dot org 2007-03-28 19:23 ---
Patch posted, but the patch tracker didn't see it:
http://gcc.gnu.org/ml/gcc-patches/2007-03/msg01657.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31356
--- Comment #3 from brooks at gcc dot gnu dot org 2007-03-28 19:22 ---
The patch tracker missed the patch I'd already posted for this one as well:
http://gcc.gnu.org/ml/gcc-patches/2007-03/msg01655.html
I think our fixes are fairly equivalent here; yours specifies the list of
language,
In compilation of a large library of code for decimal floating-point
arithmetic, I found that compilations with -g would succeed, but most
would fail with -O1, -O2, or -O3 on IA-32. I've taken a single example
and chopped away extraneous code to produce a standalone file
that exhibits the compiler
--- Comment #63 from bkoz at gcc dot gnu dot org 2007-03-28 18:18 ---
There is no fix a the moment.
However, I'm working on speculative fixes for newlib and linux, which are
predicated on the correct __cplusplus values. I may get to solaris too, if my
sanity stretches that far, or I m
--- Comment #62 from gdr at cs dot tamu dot edu 2007-03-28 17:54 ---
Subject: Re: __cplusplus defined to 1, should be 199711L
"bkoz at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| --- Comment #60 from bkoz at gcc dot gnu dot org 2007-03-28 17:48
---
|
| Mine.
|
| Curr
--- Comment #61 from bkoz at gcc dot gnu dot org 2007-03-28 17:49 ---
mine, try two
--
bkoz at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|gdr at
--- Comment #60 from bkoz at gcc dot gnu dot org 2007-03-28 17:48 ---
Mine.
Current libcpp patch only is:
Index: init.c
===
--- init.c (revision 123196)
+++ init.c (working copy)
@@ -376,7 +376,7 @@
}
i
--- Comment #59 from gdr at cs dot tamu dot edu 2007-03-28 17:43 ---
Subject: Re: __cplusplus defined to 1, should be 199711L
"bkoz at gcc dot gnu dot org" <[EMAIL PROTECTED]> writes:
| Request to re-assign to me.
Please, go ahead :-)
-- Gaby
--
http://gcc.gnu.org/bugzilla/show
--- Comment #10 from bangerth at dealii dot org 2007-03-28 17:37 ---
(In reply to comment #6)
> Always stopping is the right thing. Once Per finishes his fileline
> conversion, it will be possible to send cpplib's errors through
> diagnostic.c, and we will then be able to make this a fa
--- Comment #58 from bkoz at gcc dot gnu dot org 2007-03-28 17:32 ---
Request to re-assign to me.
-benjamin
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=1773
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dgregor at gcc dot gnu dot
|dot org
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dgregor at gcc dot gnu dot
|dot org
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dgregor at gcc dot gnu dot
|dot org
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |dgregor at gcc dot gnu dot
|dot org
--- Comment #16 from dgregor at gcc dot gnu dot org 2007-03-28 15:30
---
This functionality is part of the experimental C++0x mode, and is under control
of the -std=gnu++0x flag in the mainline (4.3.0) compiler.
--
dgregor at gcc dot gnu dot org changed:
What|Removed
--- Comment #6 from dgregor at gcc dot gnu dot org 2007-03-28 15:28 ---
Committed fix to mainline.
--
dgregor at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from hjl at gcc dot gnu dot org 2007-03-28 14:38 ---
Subject: Bug 31380
Author: hjl
Date: Wed Mar 28 14:38:24 2007
New Revision: 123299
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=123299
Log:
2007-03-28 Grigory Zagorodnev <[EMAIL PROTECTED]>
PR target
--- Comment #1 from nickc at redhat dot com 2007-03-28 13:50 ---
Hi Brooks,
I do not think that this is a bug, although it is possibly a documentation
issue. The --help=<> option restricts the output to a specified subset of the
full output obtained with just the --help option. Thus i
--- Comment #2 from nickc at redhat dot com 2007-03-28 13:40 ---
Subject: Re: --help= option isn't documented.
Hi Brooks,
The attached patch adds this missing documentation. It also takes
care of a small bug, whereby the name of the language whose options were
being displayed wa
--- Comment #5 from dgregor at gcc dot gnu dot org 2007-03-28 13:31 ---
Subject: Bug 29993
Author: dgregor
Date: Wed Mar 28 13:31:37 2007
New Revision: 123294
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=123294
Log:
2007-03-28 Douglas Gregor <[EMAIL PROTECTED]>
PR c++
--- Comment #1 from ubizjak at gmail dot com 2007-03-28 12:45 ---
Confirmed on i686-pc-linux-gnu with -O2 -msse2 -ftree-vectorize.
Backtrace:
#0 fancy_abort (file=0x871dab0 "../../gcc-svn/trunk/gcc/tree-data-ref.c",
line=2072, function=0x871eed6 "affine_function_equal_p") at
../../gcc
--- Comment #14 from tbm at cyrius dot com 2007-03-28 11:51 ---
Results are in, and it looks good. A bootstrap with the following languages
works fine: --enable-languages=c,c++,java,fortran,objc,obj-c++,ada,treelang
The test results with --enable-secureplt actually look better than
wit
--- Comment #2 from mano at roarinelk dot homelinux dot net 2007-03-28
11:50 ---
gcc-4.1.2 and 3.4.6 for linux; 3.3.5 and 2.95 for QNX also create
similar code without the dt instruction. How come your 3.4.4 is
so smart?
--
mano at roarinelk dot homelinux dot net changed:
--- Comment #17 from ebotcazou at gcc dot gnu dot org 2007-03-28 11:46
---
> Don't remember what fixed comment #13. Possibly, a change just
> made the issue latent.
OK, thanks. Here is what I needed on my 4.1 branch:
Index: cfgexpand.c
===
--- Comment #2 from nickc at redhat dot com 2007-03-28 11:38 ---
Subject: Re: gcc --help=target gives a linker error.
Hi Brooks,
I have a patch to fix this problem (attached). Unfortunately due to
an internal bugzilla error I cannot (currently) upload the patch to this PR.
Cheer
I found this with gfortran (gcc version 4.3.0 20070328), it seems to work with
the gfortran/gcc included in my Mandriva 2007 (gcc version 4.1.1 20060724), so
I've marked it as a regression.
$ cat test.f90
SUBROUTINE Test
IMPLICIT NONE
INTEGER :: i, j
REAL*8 :: C(3,100), rM(100), rMI
C
--- Comment #9 from satyaakam at yahoo dot co dot in 2007-03-28 09:32
---
Hi,
We are getting the same error on AIX 5.3.0.0 , with gcc 3.2.3.
any fix for this bug in near future.
regards
Satya
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=378
45 matches
Mail list logo
