There was a bug in this patch from 2007-05-06: when padding was requested
and the snprintf result did not fit into the initially allocated buffer,
the function could return a truncated result instead of calling snprintf
once again with a larger buffer. This fixes it.
2007-05-18 Bruno Haible <[EM
Daniel Jacobowitz wrote:
> When is it O(N*2) - only with high precision?
O(N^2) where N = max(abs(exponent of x), precision).
When N = 3000, and the elementary operation is a division instruction which
takes a few nanoseconds, the result is in the range of several milliseconds.
Bruno
On Fri, May 18, 2007 at 10:23:45PM +0200, Bruno Haible wrote:
> > Shouldn't you at least write in the documentation that the conversion
> > may be unusably slow? I know I wouldn't use this in GDB with user
> > input if it might take all year; that's hardly more useful.
>
> I wouldn't say that is
Daniel Jacobowitz wrote:
> Gosh, that's a huge amount of scary code
You are welcome to proofread it. I would be delighted.
> (actually from GMP?)
Only the naming of the base type comes from GMP. The 'divide' function is
taken from CLN. The other functions are straightforward coding.
> Shouldn't
On Fri, May 18, 2007 at 08:24:53PM +0200, Bruno Haible wrote:
> On mingw and BeOS, the native *printf functions don't support 'long double'
> arguments. This works around it.
>
> Conversion of floating-point numbers to decimal requires intermediate numbers
> arbitrary length, either multiprecision
On Fri, May 18, 2007 at 07:34:55PM +0200, Bruno Haible wrote:
> On mingw, the native printf does not support directives like "%lld". This
> works around it.
>
> 2007-05-18 Bruno Haible <[EMAIL PROTECTED]>
>
> * lib/vasnprintf.m4 (VASNPRINTF) [WIN32]: Use %I64d instead of %lld
> for
The snprintf function on BeOS produces no output if the specified size is
>= 0x300. Since gnulib's sprintf.c passed a size of 0x7fff, it
fully hits this bug.
The workaround is to use the native sprintf instead.
2007-05-18 Bruno Haible <[EMAIL PROTECTED]>
* lib/vasnprintf.c (US
On mingw, the native printf does not support directives like "%lld". This
works around it.
2007-05-18 Bruno Haible <[EMAIL PROTECTED]>
* lib/vasnprintf.m4 (VASNPRINTF) [WIN32]: Use %I64d instead of %lld
for printing 64-bit integers. Needed for mingw.
*** lib/vasnprintf.c6 M
On AIX 5.1, frexpl is in libc (not libm) and doesn't work (see frexpl.m4).
The printf-frexpl module needs to know that it cannot use this function.
2007-05-18 Bruno Haible <[EMAIL PROTECTED]>
* m4/printf-frexpl.m4 (gl_FUNC_PRINTF_FREXPL): Invoke
gl_FUNC_FREXPL_WORKS.
*
For purposes of *printf correction, one needs an implementation of frexpl()
without access to libm. This new module does it.
2007-05-18 Bruno Haible <[EMAIL PROTECTED]>
* modules/frexpl-nolibm: New file.
* m4/frexpl.m4 (gl_FUNC_FREXPL_NO_LIBM): New macro.
==
On 5/18/07, Greg Schafer <[EMAIL PROTECTED]> wrote:
Alexander Kahl wrote:
> gcc -std=gnu99 -I. -march=pentium4 -O2 -pipe -fomit-frame-pointer -s
-fgnu89-inline -c utimecmp.c
> In file included from utimecmp.c:33:
> utimens.h:2: error: conflicting types for 'futimens'
> /tools/include/sys/
Alexander Kahl <[EMAIL PROTECTED]> wrote:
> gcc -std=gnu99 -I. -march=pentium4 -O2 -pipe -fomit-frame-pointer -s
> -fgnu89-inline -c utimecmp.c
> In file included from utimecmp.c:33:
> utimens.h:2: error: conflicting types for 'futimens'
> /tools/include/sys/stat.h:370: error: previous decla
12 matches
Mail list logo
