Hi Vivek!

On 11/29/06, Vivek Goyal <[EMAIL PROTECTED]> wrote:
> On Mon, Nov 27, 2006 at 05:34:33PM +0900, Magnus Damm wrote:
> > On Mon, 2006-11-27 at 17:02 +0900, Horms wrote:
> > > On Mon, Nov 27, 2006 at 03:51:09PM +0900, Magnus Damm wrote:
> > > > kexec-tools: add nr_matched argument to kexec_iomem_for_each_line()
> > > >
> > > > This patch passes the number of matched lines using a pointer argument
> > > > instead of using the return value as suggested by Vivek.
> > > >
> > > > Signed-off-by: Magnus Damm <[EMAIL PROTECTED]>
> > >
> > > I have to say that I don't really see the merit of this change. I think
> > > its quite ok for functions that process a number of items, (like
> > > sprintf()), return the number of items processed. Though I am prepared
> > > to put it in anyway. Some more specific coments are below.
> >
> > I totally agree with you - I don't see the merit either. I think the
> > code just becomes messier with this change.
> >
> > Vivek, what do you think? Do you have a better implementation?
> >
> Hi Magnus,
>
> Rethinking about it I think horms's sprintf() example makes sense. So
> I am fine with the previous patch. Success 0, Error negative code, and
> number of lines matched is positive value returned.

Good, we'll keep the code as-is then.

Thanks!

/ magnus
_______________________________________________
fastboot mailing list
fastboot@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/fastboot

Reply via email to