Re: ARM gcc 4.1 optimization bug.
Hi Daniel, I have already reported this bug. The bug number is #27363. I also tried the gcc snapshot 4.1.1-20060421. The bug is not fixed in this version too. Thanks & Regards yfw On 5/1/06, Daniel Jacobowitz <[EMAIL PROTECTED]> wrote: On Sun, Apr 30, 2006 at 11:03:05AM +0800, Fengwei Yin wrote: > Hi, > I am using gcc4.1 for ARM to build Linux kernel. But there is a bug > related to the gcc > optimization. I assume this is correct mail list to report this bug. > If not, please let me know. No, if you have a bug report, please use the bug reporting system. Please see: http://gcc.gnu.org/bugs.html -- Daniel Jacobowitz CodeSourcery
[Ada] two regressions c64103c & cd5003g in 113391 vs. 113355
On Aurora SPARC Linux release 2.0 (Kashmir FC3)/TI UltraSparc IIi (Sabre) sun4u, I get two new failures... splitting /usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/tests/c6/c64103b.ada into: c64103b.adb BUILD c64103b.adb gnatmake --GCC="/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/" -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support c64103b.adb -largs --GCC="/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/" /usr/local/src/trunk/objdir/gcc/xgcc -c -B/usr/local/src/trunk/objdir/gcc/ -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support c64103b.adb gnatbind -aO./ -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support -I- -x c64103b.ali gnatlink c64103b.ali --GCC=/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/ RUN c64103b ,.,. C64103B ACATS 2.5 06-05-01 02:49:01 C64103B FOR IN-OUT PARAMETERS OF A SCALAR TYPE, CONSTRAINT_ERROR IS RAISED: BEFORE A SUBPROGRAM CALL WHEN THE CONVERTED ACTUAL PARAMETER IS OUTSIDE THE RANGE OF THE FORMAL PARAMETER'S SUBTYPE; AFTER A SUBPROGRAM CALL WHEN THE CONVERTED FORMAL PARAMETER IS OUTSIDE THE RANGE OF THE ACTUAL PARAMETER'S SUBTYPE. C64103B PASSED . PASS: c64103b BUILD FAIL: c64103c splitting /usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/tests/cd/cd5003g.ada into: cd5003g_pack2.ads cd5003g_pack2.adb cd5003g.adb BUILD cd5003g.adb gnatmake --GCC="/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/" -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support cd5003g.adb -largs --GCC="/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/" /usr/local/src/trunk/objdir/gcc/xgcc -c -B/usr/local/src/trunk/objdir/gcc/ -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support cd5003g.adb RUN cd5003g /usr/local/src/trunk/objdir/gcc/xgcc -c -B/usr/local/src/trunk/objdir/gcc/ -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support cd5003g_pack2.adb /usr/local/src/trunk/objdir/gcc/xgcc -c -I./ -B/usr/local/src/trunk/objdir/gcc/ -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support -I- /usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support/spprt13.ads /usr/local/src/trunk/objdir/gcc/xgcc -c -I./ -B/usr/local/src/trunk/objdir/gcc/ -gnatws -O2 -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support -I- /usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support/fcndecl.ads gnatbind -aO./ -I/usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/support -I- -x cd5003g.ali gnatlink cd5003g.ali --GCC=/usr/local/src/trunk/objdir/gcc/xgcc -B/usr/local/src/trunk/objdir/gcc/ /usr/local/src/trunk/gcc/gcc/testsuite/ada/acats/run_all.sh: line 16: /usr/local/src/trunk/objdir/gcc/testsuite/ada/acats/tests/cd/cd5003g/cd5003g: No such file or directory FAIL: cd5003g Any ideas? -- Cheers, /ChJ
Re: [Ada] two regressions c64103c & cd5003g in 113391 vs. 113355
> Any ideas? Re-run the testsuite, they most likely will disappear. -- Eric Botcazou
SPEC2000 x86 test machine down
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 The x86 box that runs SPEC2000 daily has hardware problems. It will be looked at sometime this week, but I'm not sure when it will be back online. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.3 (GNU/Linux) iD8DBQFEVjI/UTa2oAUaiwQRAlCAAKCcy/UeeawYAVAMXTKJ/5RBaN8RoACfffrf 1CUGqOTtjqXxKV3F2N1kqRc= =kAZf -END PGP SIGNATURE-
gcc-3.4.6 and pdp11
I would like to know who supports pdp11 for 3.4.6 or even the latest 4.x version of gcc? Is it still being maintained ? I'm trying to cross compile a pdp11 on my i686 and I'm having problems. If I can talk to the maintainer(s) I can get some help and maybe even get involved in pdp11 support. Bill
Re: Windows Unicode and GCC
Hello, As a quick reminder, the problem that I encountered arised when trying to compile source files that are NOT encoded with the same encoding as the system header files. My current Linux machine uses UTF-8, but I am trying to compile files that were created using Windows "unicode". To make a reproducible test, I created a simple hello world program that includes stdio.h. The file hi-utf16.c, created with Notepad and saved in "unicode", contains a BOM which is, in essence, a small header at the beginning of the file that indicates the encoding. nicolas:~> gcc -finput-charset=UTF-16 hi-utf16.c hi-utf16.c:1:19:failure to convert UTF-16 to UTF-8 It appears that CPP is telling libiconv to convert the source file from UTF-16 to UTF-8, which works, but as soon as it hits the include file, it fails. Of course, stdio.h is stored in UTF-8 on the system so trying to convert it from UTF-16 will fail right away. Now, it would be nice if every file used the same unicode encoding, but that's not always possible, especially when source control is involved. This issue touches interoperability between Windows and UNIX and also "legacy" (ie. pre-UTF-8) source files in general. My suggestion is to have CPP open a file, read the first up to 4 bytes to figure out if there is a BOM. If so, calculate the encoding and pass it libiconv. I believe that's what vim does, btw. In short, we would have the encoding detected in the following order: 1-BOM 2-finput-charset option 3-LC_CTYPE I am even thinking of how sweet it would be to be able to specify the encoding per file or directory to take care of the encodings that are not auto-detectable (Latin-1 for example), but that is probably another project of its own! Best regards, Nicolas
exposing SH's fpscr support
I'd like to extend the FPSCR support functions in SH's libgcc.a to
include a way for applications to change the bits that GCC doesn't
care about (exception handling, denormals, etc). I've come up with
this routine and (after much head-banging) managed to test it on real
hardware. The idea is that GCC gets to manage the SZ and PR bits, and
the application is free to manage the remainder of the bits, without
the two interfering with each other.
I chose the single change_fpscr function (I'm not committed to that
name) as it allows you to set the multi-bit fields in a single call
(mask the whole field, set the new value), but a two-function
set/clear pair would work too.
At this point, I would like some feedback about the API and how to
cleanly (namespace-wise) add it to libgcc.a.
static int
get_fpscr()
{
int ret;
asm volatile ("sts fpscr,%0" : "=r" (ret));
return ret;
}
static void
put_fpscr(int nv)
{
asm volatile ("lds %0,fpscr" : : "r" (nv));
}
#define FR 0x0020
#define SZ 0x0010
#define PR 0x0008
#define DN 0x0004
#define RN 0x0003
#define RN_N 0
#define RN_Z 1
extern int __fpscr_values[2];
void
change_fpscr(int off, int on)
{
int b = get_fpscr();
off = ~off;
off |= 0x0018;
on &= ~ 0x0018;
b &= off;
b |= on;
put_fpscr(b);
__fpscr_values[0] &= off;
__fpscr_values[0] |= on;
__fpscr_values[1] &= off;
__fpscr_values[1] |= on;
}
hex file
I have a hex file that i want to read and then write to another file and preserve the hex info. What is the best approach? thanks, ** David DeSimone Los Alamos National Laboratory Advanced Nuclear Technology N-2 Mail stop: B228 Los Alamos, NM 87545 phone: (505)606-0949
Re: gcc-3.4.6 and pdp11
On Mon, 2006-05-01 14:43:06 -0400, Bill Cunningham <[EMAIL PROTECTED]> wrote: > I would like to know who supports pdp11 for 3.4.6 or even the latest 4.x > version of gcc? Is it still being maintained ? I'm trying to cross compile a > pdp11 on my i686 and I'm having problems. If I can talk to the maintainer(s) > I can get some help and maybe even get involved in pdp11 support. I don't know about the state of pdp11 in GCC, but a pdp11 related bug was fixed in binutils just some weeks ago. It takes somebody to notice, so it's not 100% dead:) MfG, JBG -- Jan-Benedict Glaw [EMAIL PROTECTED]. +49-172-7608481 _ O _ "Eine Freie Meinung in einem Freien Kopf| Gegen Zensur | Gegen Krieg _ _ O für einen Freien Staat voll Freier Bürger" | im Internet! | im Irak! O O O ret = do_actions((curr | FREE_SPEECH) & ~(NEW_COPYRIGHT_LAW | DRM | TCPA)); signature.asc Description: Digital signature
Re: Suggestion for GCC (C & C++) enhancement - static variable initialisation ordering
On Apr 29, 2006, at 7:45 AM, Manfred von Willich wrote: Any interested GCC maintainers/contributors: I'd encourage you to work up a solid proposal for ISO/ANSI and propose it there. Some of the issues you bring up have already been discussed in that forum and decided, I'd doubt that they'd want to change those decisions. Specifically: (b) detect (unfortunately not at compile time) when correct initialisation is impossible due to a cyclic dependency. was thought about and rejected as was the idea about running an init order solver at runtime (beyond those aspects already present in the standard).
Re: hex file
On May 1, 2006, at 1:26 PM, David Desimone wrote: I have a hex file that i want to read and then write to another file and preserve the hex info. Wrong list.
Re: dwarf2 compiling problem
> gcc:version 4.02 > target: DLX CPU (port in progress) > > Does anyone have an idea what can cause this? > Is this a bug in gcc or in target cpu files? It's almost certainly a bug in your target files. The quickest course of action would be to run cc1 under gdb, feed it with the same source file (preprocessed first, of course) and see if you can infer the cause of the crash from the backtrack and variables in that stack frame. Ben
svn problems
svn is giving me grief, I'm merging and getting: mrs $ svn cp -r113703 svn+ssh://src.apple.com/svn/fsf-gcc/gcc/ branches/apple-local-200502-branch/gcc/testsuite/g++.dg/ bitreverse-10.C bitreverse-10.C svn: Entry for 'bitreverse-10.C' exists (though the working file is missing) mrs $ svn add bitreverse-10.C svn: warning: 'bitreverse-10.C' not found mrs $ touch bitreverse-10.C mrs $ svn add bitreverse-10.C A bitreverse-10.C mrs $ svn cp -r113703 svn+ssh://src.apple.com/svn/fsf-gcc/gcc/ branches/apple-local-200502-branch/gcc/testsuite/g++.dg/ bitreverse-10.C bitreverse-10.C svn: File 'bitreverse-10.C' already exists mrs $ svn rm --force bitreverse-10.C D bitreverse-10.C mrs $ svn cp -r113703 svn+ssh://src.apple.com/svn/fsf-gcc/gcc/ branches/apple-local-200502-branch/gcc/testsuite/g++.dg/ bitreverse-10.C bitreverse-10.C svn: Entry for 'bitreverse-10.C' exists (though the working file is missing) How do I `fix' this. I know how to fix it with multiple checkins, don't want to do that, if I can avoid that. I can't seem to find a spelling that works. :-( In cvs, this just works. :-(^2.
Re: svn problems
> mrs $ svn cp -r113703 svn+ssh://src.apple.com/svn/fsf-gcc/gcc/ > branches/apple-local-200502-branch/gcc/testsuite/g++.dg/ > bitreverse-10.C bitreverse-10.C > svn: Entry for 'bitreverse-10.C' exists (though the working file is > missing) > > How do I `fix' this. I know how to fix it with multiple checkins, > don't want to do that, if I can avoid that. I can't seem to find a > spelling that works. :-( What exactly are you trying to achieve? Why not just copy the file (with normal cp, for an existing checkout) then svn add it. > In cvs, this just works. :-(^2. I find that hard to believe. cvs doesn't even have the concept of a copy. Paul
Re: svn problems
On May 1, 2006, at 7:43 PM, Paul Brook wrote: What exactly are you trying to achieve? The first order operation is is a new fangled concept called a merge. The second order operation is to dig out from under svn bugs with additional commands to get the state of the working copy correct, so that when I do my ci, the repo will then have what I want in it, aka, the merge product. Why not just copy the file (with normal cp, for an existing checkout) then svn add it. Well, conceptually, I want the history for the file, hence the cp. I was trying to get around yet another svn bug that causes checkins of merge products to fail. In cvs, this just works. :-(^2. I find that hard to believe. cvs doesn't even have the concept of a copy. Trust me, merging works in cvs, I've used it before. It amazes me that svn can't do a merge.
Re: svn problems
> It amazes me that svn can't do a merge. That's patently inaccurate. Have you tried using the svnmerge script that uses SVN properties to remember which changesets you have (and haven't) merged? Ben
Re: svn problems
> Well, conceptually, I want the history for the file, hence the cp. I > was trying to get around yet another svn bug that causes checkins of > merge products to fail. What bugs? If you file them, and tell us how to reproduce them, maybe we could fix them?
Re: svn problems
On 5/2/06, Mike Stump <[EMAIL PROTECTED]> wrote: It amazes me that svn can't do a merge. Just because you can't do it, doesn't mean svn can't do it. Other people have been maintaining branches with svn for months now without trouble. Gr. Steven
Re: svn problems
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mike Stump wrote: > It amazes me that svn can't do a merge. > You obviously have not read the documentation nor browsed the GCC wiki. Doing merges with svn is amazingly simple. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.3 (GNU/Linux) iD8DBQFEVt1XUTa2oAUaiwQRAsHIAKCE2t/7WW2JgA3hXQqDSIxBtvyTKgCgjxE9 JN+HQuUQrjdPcbvLl50yY2Q= =M7ty -END PGP SIGNATURE-
Re: [Ada] two regressions c64103c & cd5003g in 113391 vs. 113355
On 5/1/06, Eric Botcazou <[EMAIL PROTECTED]> wrote: > Any ideas? Re-run the testsuite, they most likely will disappear. right... I somehow had memory kernel related issues... A new one now... c9a011b -- Cheers, /ChJ
