On 25 April 2012 16:42, Khem Raj <[email protected]> wrote: > On Wed, Apr 25, 2012 at 4:53 AM, Bernhard Reutner-Fischer > <[email protected]> wrote: >> On 24 April 2012 23:54, Peter Mazinger <[email protected]> wrote: >>>> Peter, >>>> >>> >>>> I get alot of errors when compiling for x86_64 (i.e. 64bit), e.g.: >>>> >>>> libc/misc/dirent/versionsort.c:16:1: warning: return type defaults to >>>> 'int' [enabled by default] >>>> libc/misc/dirent/versionsort.c:16:1: warning: function declaration >>>> isn't a prototype [-Wstrict-prototypes] >>> >>> means, your headers do not provide a prototype for versionsort() >> >> No, it was a typo in the macro name (see surplus 's' in other mail in >> this thread). >> fixed. >> >>> fcntl is a special case, not comparable with the other X64() functions (at >>> least in uClibc), this case can be discussed on ml if you want to. >> >> Yes, i know. I think the correct guard is >> #if defined __UCLIBC_HAS_LFS__ && __WORDSIZE == 32 && !defined __NR_fcntl64 >> fixed. >> >> missing comma for ppc32 in posix_fadvise.c -- fixed. >> >> Another question, on powerpc, i see: >> >> Makefile.in:139: warning: overriding commands for target >> `include/bits/local_lim.h' >> libpthread/nptl/sysdeps/unix/sysv/linux/Makefile.commonarch:202: >> warning: ignoring old commands for target `include/bits/local_lim.h' >> >> Can you advise on this on, please? > > Bernhard > > Be aware that the patches that are eliminated only make it work > on x86 other architectures are still regressing so do first find > the offending patches that cause this.
I am of course aware of this, yes. See uClibc.org/~aldot/uClibc/psm-future.tar.gz for my current set (plus separately proposed fix for PR3853 in bugzilla) that is in testing on a couple of arches. cheers, _______________________________________________ uClibc mailing list [email protected] http://lists.busybox.net/mailman/listinfo/uclibc
