-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 3/24/2007 7:04 PM:
> Hi,
>
> How can one distinguish +0.0 and -0.0. printf() needs to be able to do it.
> How can this be done portably? I can see two ways:
> a) By knowing the bit position of the sign bit.
> b) By doi
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 3/24/2007 9:17 PM:
> James Youngman wrote:
>> 2007-03-24 James Youngman <[EMAIL PROTECTED]>
>>
>> * lib/stat-time.h (get_stat_birthtime): New function for
>> retrieving st_birthtime as provided by UFS2 (hen
James Youngman wrote:
> 2007-03-24 James Youngman <[EMAIL PROTECTED]>
>
> * lib/stat-time.h (get_stat_birthtime): New function for
> retrieving st_birthtime as provided by UFS2 (hence *BSD).
> * m4/stat-time.m4 (gl_STAT_BIRTHTIME): Probe for st_birthtime
> and its v
tsearch() should be declared in , not "tsearch.h". I'm removing
tsearch.h, making gnulib more POSIX aligned.
2007-03-24 Bruno Haible <[EMAIL PROTECTED]>
* modules/search: New file.
* lib/search_.h: New file, incorporating lib/tsearch.h.
* m4/search_h.m4: New file.
NetBSD.
This platform not only lacks all 'long double' math functions, starting with
frexpl() and ldexpl().
It also initializes the x86 FPU control word to a value that causes all
floating-point operations to round to 'double' precision (53 mantissa bits).
Nearly all 'long double' operations ther
Hi,
How can one distinguish +0.0 and -0.0. printf() needs to be able to do it.
How can this be done portably? I can see two ways:
a) By knowing the bit position of the sign bit.
b) By doing a bit-for-bit comparison against +0.0.
The second approach appears to be implementable with less portab
This fixes a problem that would only be noticeable on platform without
'long double' support.
2007-03-24 Bruno Haible <[EMAIL PROTECTED]>
* lib/vasnprintf.c [!HAVE_LONG_DOUBLE]: Include printf-frexp.h. Don't
include isnanl-nolibm.h.
*** lib/vasnprintf.c22 Mar 2007 02:04:01
Ian Lance Taylor wrote:
I believe there is a comprehensible distinction between "compiler will
not assume that signed overflow is undefined behaviour" and "compiler
will cause all arithmetic to wrap around."
In any case, I have no plans to continue working on this. I described
my work in consi
Robert Dewar <[EMAIL PROTECTED]> writes:
> > You're right, I shouldn't have said "implementation defined."
> > What will happen with -fno-strict-overflow is whatever the processor
> > ISA happens to do when a signed arithmetic operation overflows. For
> > ordinary machines it will just wrap.
>
>
Apologies if I have something signficant wrong; it's my first
nontrivial gnulib patch.
2007-03-24 James Youngman <[EMAIL PROTECTED]>
* lib/stat-time.h (get_stat_birthtime): New function for
retrieving st_birthtime as provided by UFS2 (hence *BSD).
* m4/stat-time.m4 (g
Ian Lance Taylor wrote:
You're right, I shouldn't have said "implementation defined."
What will happen with -fno-strict-overflow is whatever the processor
ISA happens to do when a signed arithmetic operation overflows. For
ordinary machines it will just wrap.
Given that all ordinary machines
Robert Dewar <[EMAIL PROTECTED]> writes:
> Ian Lance Taylor wrote:
>
> > The new option -fstrict-overflow tells gcc that it can assume the
> > strict signed overflow semantics prescribed by the language standard.
> > This option is enabled by default at -O2 and higher. Using
> > -fno-strict-over
Bruno Haible <[EMAIL PROTECTED]> writes:
> The result of these considerations is this. OK to commit?
Looks fine, please do!
/Simon
> 2007-03-17 Bruno Haible <[EMAIL PROTECTED]>
>
> * m4/locale-fr.m4 (gt_LOCALE_FR, gt_LOCALE_FR_UTF8): Check also the
> locale's decimal-point character.
This was buggy. Fixing it:
2007-03-24 Bruno Haible <[EMAIL PROTECTED]>
* m4/locale-fr.m4 (gt_LOCALE_FR, gt_LOCALE_FR_
Hi Simon,
A "gnulib-tool --create-testdir --with-tests" directory is giving me this
output during "make check" on NetBSD 3.0:
Could not read file: No such file or directory
BAD: out[len] not zero: No such file or directory
Read -1077940928 from /etc/resolv.conf...
Read 0 from /dev/null...
Part 3 is a cleanup of the module structure.
2007-03-24 Bruno Haible <[EMAIL PROTECTED]>
* modules/unistr/base (Depends-on): Remove utf8-ucs4-unsafe,
utf16-ucs4-unsafe, utf8-ucs4, utf16-ucs4, ucs4-utf8, ucs4-utf16.
* modules/unistr/u8-mbtouc: Add source files from modul
Here's the second part: making the source file naming more consistent.
2007-03-24 Bruno Haible <[EMAIL PROTECTED]>
* lib/unistr/u8-mbtouc-aux.c: Renamed from lib/unistr/utf8-ucs4.c.
Enable the function only if HAVE_INLINE.
* lib/unistr/u8-mbtouc-unsafe-aux.c: Renamed fro
Hi,
Compiling with -Wall gave me these warnings:
ucs4-utf16.h:33: warning: static declaration for `u16_uctomb' follows non-static
ucs4-utf16.h:33: warning: static declaration for `u16_uctomb' follows non-static
ucs4-utf16.h:33: warning: static declaration for `u16_uctomb' follows non-static
ucs4-
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Let's take this question to the Austin group, for a definitive word from
the POSIX folks. The question originally arose on the bash mailing lists
(http://lists.gnu.org/archive/html/bug-bash/2007-03/msg00067.html and
following, see also
http://lists.gn
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Eric Blake wrote:
>> Rather than use lots of casts, coreutils does this in a common header:
>>
>> /* Convert a possibly-signed character to an unsigned character. This is
>>a bit safer than casting to unsigned char, since it catches some type
>>er
Eric Blake wrote:
> Rather than use lots of casts, coreutils does this in a common header:
>
> /* Convert a possibly-signed character to an unsigned character. This is
>a bit safer than casting to unsigned char, since it catches some type
>errors that the cast doesn't. */
> static inline
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 3/24/2007 5:22 AM:
>
> When you look at the code, you see that the user only has to set the variable
> ARGP_HELP_FMT to a value containing non-ASCII characters, to make the program
> crash. Note that the code in glibc does
Hi,
Compiling the 'argp' module on NetBSD with "-Wall" yields these warnings:
argp-help.c:172: warning: subscript has type `char'
argp-help.c:174: warning: subscript has type `char'
argp-help.c:181: warning: subscript has type `char'
argp-help.c:185: warning: subscript has type `char'
argp-help.c
Robert Dewar wrote:
Ian Lance Taylor wrote:
The new option -fstrict-overflow tells gcc that it can assume the
strict signed overflow semantics prescribed by the language standard.
This option is enabled by default at -O2 and higher. Using
-fno-strict-overflow will tell gcc that it can not assum
24 matches
Mail list logo
