Bruno Haible <[EMAIL PROTECTED]> writes:
> * lib/stdint.in.h (_STDINT_MAX): Subtract 1 from an unused signed
> integer shift in the signed case. Fixes warnings with OSF/1 5.1 cc.
Thanks for that fix.
Bruno Haible <[EMAIL PROTECTED]> writes:
> I propose to change the round* macros as follows, so as to
> 1. fix the Solaris problem with roundf, roundl,
> 2. use the gl_FUNC_FLOOR*_LIBS, gl_FUNC_CEIL*_LIBS macros introduced
> earlier today.
> And the corresponding changes to the 'math' mod
> Yes, please apply.
Done. After some merging, I applied this:
2007-10-28 Bruno Haible <[EMAIL PROTECTED]>
* lib/math.in.h: Test REPLACE_ISFINITE instead of HAVE_DECL_ISFINITE.
* m4/isfinite.m4 (gl_ISFINITE): Initialize REPLACE_ISFINITE instead of
HAVE_DECL_ISFINITE.
Hi Ben,
On 2007-10-22 you wrote:
> >> I am not sure about this. gl_FUNC_FLOORF checks for floorf and
> >> provides a substitute if it is not available. But for roundf I
> >> was planning to use the system floorf if it was available and, if
> >> not, use the roundf implementation that does not ne
Bruno Haible <[EMAIL PROTECTED]> writes:
> But I don't much like that the symbol HAVE_DECL_ISFINITE denotes something
> different than whether the isfinite macro is declared. This will be confusing
> in the future. OK to replace it by the inversion of REPLACE_ISFINITE, like
> this?
Yes, please ap
Hi Paul,
The gnulib-substituted stdint.h causes a compiler warning from the
OSF/1 5.1 cc compiler, at every use of PTRDIFF_MAX or SIG_ATOMIC_MAX.
cc: Warning: test-stdint.c, line 260: In the declaration of
"verify_error_if_negative_size__", integer overflow occurs in evaluating the
expression
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Tim Rice on 10/28/2007 7:15 PM:
>
> It looks like there are some regressions in 1.4.10 on some of my platforms.
Thanks for the report. Many of these issues have already been fixed in
gnulib, and will thus be in m4 1.4.11 when I release
Hi,
On Solaris 10, with Sun C 5.0, in 32-bit mode, the functions floorf, ceilf,
floorl, ceill are declared, and are present in /lib/libm.so.2, but are not
defined in the libm that the compiler uses (libm.so.1 I guess). As a
consequence, the corresponding tests don't link.
So we need to check not
Hi Ben,
A week ago I promised this:
> OK, I can provide a macro that just tests whether floorf() needs libm.
Here it is (committed):
2007-10-28 Bruno Haible <[EMAIL PROTECTED]>
* m4/floorl.m4 (gl_FUNC_FLOORL_LIBS): New macro, extracted from
gl_FUNC_FLOORL. Cache the result.
Ben,
But I don't much like that the symbol HAVE_DECL_ISFINITE denotes something
different than whether the isfinite macro is declared. This will be confusing
in the future. OK to replace it by the inversion of REPLACE_ISFINITE, like
this?
2007-10-28 Bruno Haible <[EMAIL PROTECTED]>
*
Hi Ben,
> +2007-10-27 Ben Pfaff <[EMAIL PROTECTED]>
> +
> + Ralf Wildenhues reported that Tru64 4.0D declares the round
> + functions but does not have definitions.
> + * m4/check-math-lib.m4 (gl_CHECK_MATH_LIB): If the target function
> + cannot be found in any library, set the
Ben Pfaff wrote:
> Here is the module as I committed it:
>
> 2007-10-27 Ben Pfaff <[EMAIL PROTECTED]>
>
> + * lib/math.in.h: Define isfinite macro and prototypes for
> + gl_isfinitef, gl_isfinited, gl_isfinitel if we are providing
> + implementations.
> + * m4/math_h.m4: New s
Sam Steingold wrote:
> > There is an alternative modules 'canonicalize-lgpl'; it has fewer
> > dependencies.
>
> how is it functionally different?
> if it is not, why two versions?
It is functionally the same. The 'canonicalize' module also has a mode
CAN_ALL_BUT_LAST that correspondings to clisp
Simon Josefsson wrote:
> > The complexity of distributing a library under one license and the
> > corresponding tools under another license is very manageable. I do this
> > for libiconv and gettext. You distribute both COPYINGv2 and COPYINGv3, and
> > add a statement to the README, saying that the
Hi Bruno,
openbsd$ ./seq 4
1
nan
nan
4
I tracked this to printf-posix's use of isnanl.
That function from gnulib returns true for the long-double
values of 2 and 3 (exponent is 0 in those cases, yet this
code returns 1:
if (exponent == 0)
return (m.word[1] >> 31);
I don't have t
Hello Ralf,
Ralf Wildenhues wrote:
> The Intel compiler 9.0 gives these warnings on GNU/Linux:
>
> | ../../gllib/unistr/u16-chr.c(39): warning #266: function declared implicitly
> | switch (u16_uctomb_aux (c, uc, 2))
> |
> | ../../gllib/unistr/u8-chr.c(39): warning #266: function declared
> 2007-10-20 Bruno Haible <[EMAIL PROTECTED]>
>
> * modules/error (Notice): Mention how to augment the XGETTEXT_OPTIONS.
> * modules/vasprintf (Notice): Likewise.
> * modules/xprintf (Notice): Likewise.
> * modules/xvasprintf (Notice): Likewise.
I now applied this. Augme
Jim Meyering wrote on 2007-10-20:
> Here's what I want :-)
>
> Simply add an attribute to each affected module indicating that a
> function takes a printf-style format string, and then let tools do
> the rest, possibly even including deriving the arg offset based on
> attribute dir
Jim Meyering wrote on 2007-10-20:
> ... just saying "out of memory" provides
> no clue as to which output file may have just been corrupted
> as a result of this failure. Nor does it mention that the error
> arose during an attempt at formatted output.
>
> I'll bet we have enough of the pieces to
> * Bruno Haible <[EMAIL PROTECTED]> [2007-10-28 05:43:54 +0200]:
>
> Sam Steingold wrote:
>> > But it's better to use canonicalize_file_name()
>>
>> it comes with a huge dependecy set...
>
> There is an alternative modules 'canonicalize-lgpl'; it has fewer
> dependencies.
how is it functionally
The Intel compiler 9.0 gives these warnings on GNU/Linux:
| ../../gllib/unistr/u16-chr.c(39): warning #266: function declared implicitly
| switch (u16_uctomb_aux (c, uc, 2))
|
| ../../gllib/unistr/u8-chr.c(39): warning #266: function declared implicitly
| switch (u8_uctomb_aux (c, uc,
21 matches
Mail list logo
