Package: glibc-doc-reference
Version: 2.10.1-1
Severity: normal
File: /usr/share/info/libc.info.gz
error_print_progname doesn't appear in the indexes, except perhaps as a
dodgy ( entry at the start of the Variable Index node.
It looks like a {} grouping typo the @deftypevar per below.
(Or is it
Package: libc6
Version: 2.10.1-3
Severity: normal
The program nomallocinfo.c below prints a bit
malloc version=1
heap nr=0
sizes
then gets a segfault. malloc_info() won't have much to say when there's
been no mallocs yet, but I think it shouldn't segfault in that case.
#include
Package: libc6
Version: 2.10.1-3
Severity: normal
The program mallocinfoquotes.c prints
system type=current size=135168
/
system type=max size=135168
/
which looks like two \n meant to be \, to close the quoted values.
#include stdio.h
#include stdlib.h
#include malloc.h
int
Package: libc6
Version: 2.10.1-3
Severity: normal
With a ja_JP.UTF8 locale defined, the program foo.c below prints
E3 80 87 00 E4 B8 80 00 E4 BA 8C 00
where I expected the 00 null bytes to be semi-colons, separating the
symbols for the alt digits, per POSIX
Package: libc6
Version: 2.3.6.ds1-4
Severity: normal
Tags: patch
When given a negative imaginary number, csqrt() returns a negative
real part, whereas the principal root should be the one with positive
real part. Sample program below.
I believe this was fixed upstream last year. It'd be nice
Package: glibc-doc
Version: 2.3.6-5
Severity: normal
The Attribute Meanings and File Times nodes describe microsecond
st_atime_usec etc fields for the stat structure, but it seems they're
not provided on a gnu/linux system (judging by the little program
below failing to compile).
It'd be good if
Package: glibc-doc
Version: 2.3.6-5
Severity: normal
The File Times node shows
-- Function: int futimes (int *FD, struct timeval TVP[2])
but from looking at /usr/include/sys/time.h I think the fd parameter
should be just int FD (not a pointer).
-- System Information:
Debian
Package: glibc-doc
Version: 2.3.6-5
Severity: minor
In the Function Index node, sbrk appears at the start as *sbrk.
It probably wants a grouping like
{void *}
in the .texi, so the * isn't taken as part of the function name.
-- System Information:
Debian Release: testing/unstable
APT
Package: glibc-doc
Version: 2.3.6-5
Severity: wishlist
It'd be nice if int8_t and friends, which are described in the
Integers node, could be in the Type Index so that in info an i
index search can hit them, and also Emacs info-look symbol lookup.
-- System Information:
Debian Release:
Package: glibc-doc
Version: 2.3.5-6
Severity: normal
A program
#include langinfo.h
#include stdio.h
int
main (void)
{
printf (%s\n, nl_langinfo (DECIMAL_POINT));
return 0;
}
compiled with gcc foo.c gives an error
foo.c:7: error: 'DECIMAL_POINT'
Package: locales
Version: 2.3.2.ds1-22
Severity: minor
/usr/share/man/man5/locale.gen.5.gz refers to
/usr/share/i18n/charsets
whereas the directory seems to be
/usr/share/i18n/charmaps
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT
Package: glibc-doc
Version: 2.3.2.ds1-19
Severity: normal
In the TZ Variable node, the TZ=:chars form is described,
If CHARACTERS begins with a slash, it is an absolute file
name; otherwise the library looks for the file
`/share/lib/zoneinfo/CHARACTERS'.
But there seems
Package: glibc-doc
Version: 2.3.2.ds1-19
Severity: normal
In the Processor Time node, times() is described with
The return value is the calling process' CPU time (the same value
you get from `clock()'.
But I think that's not the case on a Debian system. For instance,
#include
Package: glibc-doc
Version: 2.3.2.ds1-18
Severity: normal
In the Constants for Sysconf node, _SC_CLK_TCK is described as being
the parameter corresponding to CLOCKS_PER_SEC, but I believe that
should be CLK_TCK. Eg.
#include time.h
#include unistd.h
main () {
printf (%d, %d,
Package: glibc-doc
Version: 2.3.2.ds1-18
Severity: normal
In the CPU Time node, CLK_TCK is described as an obsolete name for
CLOCKS_PER_SEC. But in Debian (i386 at least) the two are not the
same, eg.
#include time.h
main () { printf (%d %d\n, CLK_TCK, CLOCKS_PER_SEC); }
shows 100 and
Jeff Bailey [EMAIL PROTECTED] writes:
What arch is this for, please?
Sorry, yes, this is i386.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Package: libc6
Version: 2.3.1-5
A program foo.c,
#include stdio.h
#include stdlib.h
int
main (void)
{
printf (%f\n, strtod(0x10p-1,NULL));
}
prints 1.0, where I expected it to print 8.0, since the exponent for a
hex float is a power of
Jeff Bailey [EMAIL PROTECTED] writes:
What arch is this for, please?
Sorry, yes, this is i386.
Package: libc6
Version: 2.3.1-3
I noticed scanf and friends seem to demand an exponent on hex format
float inputs, whereas my reading of the doco suggests it should be
optional.
The program
int
main ()
{
double d = -999;
int ret;
ret =
Package: glibc-doc
Version: 2.3.1-3
Severity: minor
System Parameters in the manual refers to sysctl.h for the
prototypes and defines for sysctl(), but there doesn't seem to be a
/usr/include/sysctl.h in libc6-dev (2.3.1-3). I wonder if that doco
should say sys/sysctl.h, which exists.
--
To
Package: glibc-doc
Version: 2.3.1-3
Severity: minor
System Parameters in the manual refers to sysctl.h for the
prototypes and defines for sysctl(), but there doesn't seem to be a
/usr/include/sysctl.h in libc6-dev (2.3.1-3). I wonder if that doco
should say sys/sysctl.h, which exists.
Austin Donnelly [EMAIL PROTECTED] writes:
Package: libc6
Version: 2.1.1
The definition of strstr(3) in ./sysdeps/i386/i486/bits/string.h
includes a fragment of assembly:
[snip]
Note that it clears the direction flag (cld), but never restores it to
its previous value. This could be a
22 matches
Mail list logo
