The following code:
#include
#include
int main()
{
int32_t ival = 1;
uint32_t uval = 2;
printf("int = %"PRId32", uint = %"PRIu32".\n", ival, uval);
return 0;
}
when compiled with either gcc or clang on a 32-bit system and with the -Wall
flag produces the foll
>On Aug 22 16:45, Cary R. wrote:
>> I upgraded to the latest cygwin earlier today (it has probably been
>> a month or more since I last updated) and now a program that has been
>> working is crashing (segfault) when calling dlclose(). dlopen() and
>> calling a func
I upgraded to the latest cygwin earlier today (it has probably been
a month or more since I last updated) and now a program that has been
working is crashing (segfault) when calling dlclose(). dlopen() and
calling a function in the DLL still works just fine. An older version
of the program still wo
The following code demonstrates a subtle bug in the Cygwin version of strtod().
The value it generates is slightly different than the value when using the math
header files. This used to work correctly some time ago (months). I just took
the time to track the problem down in our regression suite
Actually compiling (with i686-w64-mingw32-gcc) the following simple C program
is also crashing.
int main()
{
return 0;
}
Cary
- Original Message -
From: Cary R.
Jon,
I just tried the new version of MinGW-w64 without the header update and the
compiler is crashing while
Jon,
I just tried the new version of MinGW-w64 without the header update and the
compiler is crashing while linking one of our C++ executables.
Here's the link command and the error message:
i686-w64-mingw32-g++ -o ivl.exe main.o async.o design_dump.o discipline.o
dup_expr.o elaborate.o elab_
I originally sent this a bit over a month ago and
have not received a rely. I just verified that the
problem still exists. I also verified that the same
tcsh version under Linux (Ubuntu 10.4) does not
have the problem, so this looks like a Cygwin
specific problem.
With the latest tcsh and cygwin o
With the latest tcsh and cygwin on a windows XP SP3 machine I'm seeing
a problem with file globing.
If you have a directory with three files tmp.c, vcd_priv.c and vcd_priv2.cc
(just create them with touch).
If you run ls *.cc it will display both vcd_priv.c and vcd_priv2.cc.
The bash shell does
Cygwin user working on smaller projects
where the speed difference is not a concern. The Cygwin version
also produces results that exactly match what is produced under
Linux.They probably belong with ngspice in the Science category.
I'm willing to provide help as needed.
Regards,
Cary R.
On my Linux machine I'm using the latest cppcheck from git
since it has bug fixes I need to test my code. Now that it's
mostly working I tried to compile it on Cygwin so I would
have it available there as well. I ran into a few problems
and after some debug it appears to be a problem with the
-D_GL
The /etc/fonts/local.conf file is not created by default, so I think the
outer most "if" in the post install script should have an else clause that
creates a minimal file if one does not exist. Something like:
/usr/share/lyx/fonts
I'm not certain this is exactly correct, but it should be
GTG from my point of view. The results from 1.5.25-6 match my patched
version.
Cary
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs
--
Unsubscribe info: http://cygwin.com/
Attached are the two ChangeLog entries from newlib and a patch of the math
directory differences between my modified 1.5.24 and 1.5.25. I did a quick
visual check and this single patch appears to match the two I originally
sent. Please let me know if I can be of further assistance.
Cary
_
In April/May 2007 I submitted a number of patches to fix math boundary
problems in newlib. These changes appear to have not made it into the
latest cygwin release. Though they are certainly in the newlib repository.
Is there a reason for this omission? Can the current cygwin release be
updated/fixe
On my Linux machine the rpc/xdr.h header file includes rpc/types.h and the
cygwin version does not. This results in configure scripts not being able
to verify that rpc/xdr.h is a valid file. Applying the following patch
makes this work as expected.
--- xdr.h.orig 2005-03-10 13:32:52.00100 -08
The latest update to git (1.5.3.1) changed how the manual pages are
displayed. Specifically references to other git manual pages are no longer
shown by name. Instead a reference number with a cross reference at the
end of the file is given. From a usability standpoint this is a real
inconvenience.
> > newlib and my version of glibc return +0 for all these cases.
> Hmm *my* version of glibc gets all 4 cases right:
We upgraded to a new version of RHEL and guess what. Well you know!
Patches for atan2() have been applied to newlib CVS.
Cary
__
One thing to keep in mind is that I was encouraged to submit patches to
make the newlib functions match the glibc implementation not the standard.
Which I don't currently have access to.
> Other potentially controversial special cases (may or may not be
> handled correctly by newlib -- I didn't c
Patch generated and applied to newlib CVS. As an added bonus I fixed a
few other inconsistencies in acos(), asin(), log() and log10().
Cary
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-
> gcc -fno-builtin ./pow-error.c
Yes -fno-builtin certainly makes the results
consistent, but I would consider this result slightly
less correct.
> Unless the function call is optimized away.
Hmm, I guess I didn't think something as complicated
as pow would get optimized away. Thanks for the
in
I had some more time to look into this and when the
simple C program I mentioned earlier uses variables
like the other program, incorrect results are
produced. I have attached this C/C++ program. I
certainly don't understand what is going on. I would
have expected pow to be pass-by value which shou
Hmmm, something weird is going on at my end. I just
wrote a standalone C program to test just these cases
and some of them now work as expected! I do agree that
-1.0 to either +-infinity should be a nan, but then
using that same logic why is -1.1 to infinity and 0.9
to minus infinity equal to infin
I'm not certain if this is a newlib compilation flag
problem, something that has been fixed in the 1.15
version of newlib or just that newlib does things
differently. Looking at the 1.15 code implies that
some of these should work and the rest appear to be a
deviation from other systems. I tested t
It appears that Jan would rather have LyX resolve this
problem. The basics are that fontconfig no longer
creates a default local.conf file, so LyX will need to
correctly handle this case.
It looks like adding an else clause to the existing -f
check that generates a default file will fix things
up.
Sorry for the tardy reply. Between too many hours
spent finishing a project and some much needed
rest/recovery after finishing it I have not been
monitoring the list as I would like to.
"Jan Nieuwenhuizen"
writes:
> "Cary R." <[EMAIL PROTECTED]> writes:
>
>
LyX 1.4.4-1 postinstall script has fatal error in sed
script.
The sed script is split across two lines which is
causing the script to fail while trying to add the lyx
font directory. This leaves a blank local.conf file. I
fixed the script by using a \n after the <\dir>
directive. I also added
The previous version 2.2.2-1 does include this file.
It also appears that the postinstall script needs some
updating or it may no longer be needed.
Yahoo! Music Unlimited
Access over 1 million songs.
http://mus
27 matches
Mail list logo
