This is the mail archive of the newlib@sourceware.org mailing list for the newlib project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] sys/types.h: Avoid bit-manipulation of signed values


On Dec  7 08:11, Sebastian Huber wrote:
> 
> 
> On 03/12/15 10:09, Corinna Vinschen wrote:
> >Hi Nick,
> >
> >On Dec  3 18:45, Nick Withers wrote:
> >>>Hi all,
> >>>
> >>>On Fri, 2015-11-27 at 14:04 +0100, Corinna Vinschen wrote:
> >>>> >On Nov 27 19:08, Nick Withers wrote:
> >>>> >[...]
> >>>> >The first thing would be to check if sys/select.h from Cygwin is
> >>>> >sufficent for RTEMS as well, provided you apply my patch from
> >>>> >https://sourceware.org/ml/newlib/2015/msg00832.html  and move the
> >>>> >resulting winsup/cygwin/include/sys/select.h to
> >>>> >newlib/libc/include/sys.
> >>>> >
> >>>> >If that works fine, I apply my patch.
> >>>
> >>>I haven't yet tackled the rtems-libbsd part Sebastian mentioned, but
> >>>have made some progress.
> >>>
> >>>With the attached newlib-sys_select.patch* and rtems-sys_select.patch**
> >>>patches applied to Newlib and RTEMS Git masters respectively, the RTEMS
> >>>tests "syscall01" and "psxreaddir" seem to pass through PSim (psim
> >>>-syscall01.out and psxreaddir.out).
> >>>
> >>>
> >>>How do folk feel about this so far?
> >The additon for arm is fine so from the newlib side it looks good to go.
> 
> The difficult part is the compatibility with the libbsd which is based on
> FreeBSD version 9 code, but we can fix this later.

Ok, thanks.  I pushed Nick's and my patch.


Corinna

-- 
Corinna Vinschen
Cygwin Maintainer
Red Hat

Attachment: pgpnMFJsSyzmK.pgp
Description: PGP signature


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]