On Fri, 2015-03-13 at 15:40 -0400, Konrad Rzeszutek Wilk wrote:
> On Fri, Mar 13, 2015 at 03:31:12PM -0400, Konrad Rzeszutek Wilk wrote:
> > Instead put the -Exx in errno.
> >
> > Signed-off-by: Konrad Rzeszutek Wilk <konrad.w...@oracle.com>
> > ---
> > tools/libxc/xc_domain.c | 6 ++++--
> > 1 file changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/tools/libxc/xc_domain.c b/tools/libxc/xc_domain.c
> > index 845d1d7..eb3a5f2 100644
> > --- a/tools/libxc/xc_domain.c
> > +++ b/tools/libxc/xc_domain.c
> > @@ -771,8 +771,10 @@ int xc_domain_get_tsc_info(xc_interface *xch,
> >
> > info = xc_hypercall_buffer_alloc(xch, info, sizeof(*info));
> > if ( info == NULL )
> > - return -ENOMEM;
> > -
> > + {
> > + errno = ENOMEM;
> > + return -1;
> > + }
>
> While I was doing that I saw a lot of other code doing:
>
>
> if ( xc_hypercall_bounce_pre (..) )
> return -1;
>
> Should they all be fixed up to do:
> 'errno = ENOMEM'
>
> before returning -1?
I think xc_hypercall_bounce_pre should probably set an appropriate errno
on failure rather than making all the callers have to think about it. It
may indirectly already do so via mmap? I didn't actually go look...
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
