Re: [PATCH] mini-os: enable compiler check for printk format types

[Ian Campbell <Ian.Campbell@citrix.com>]

On Fri, 2014-10-03 at 09:44 +0100, Thomas Leonard wrote:
> On 8 August 2014 15:32, Samuel Thibault <samuel.thibault@ens-lyon.org> wrote:
> > Thomas Leonard, le Fri 08 Aug 2014 15:28:22 +0100, a écrit :
> >> On 7 August 2014 15:55, Samuel Thibault <samuel.thibault@ens-lyon.org> wrote:
> >> > Thomas Leonard, le Wed 06 Aug 2014 10:44:00 +0100, a écrit :
> >> [...]
> >> >> diff --git a/extras/mini-os/netfront.c b/extras/mini-os/netfront.c
> >> >> index 44c3995..6f335fe 100644
> >> >> --- a/extras/mini-os/netfront.c
> >> >> +++ b/extras/mini-os/netfront.c
> >> >> @@ -327,8 +327,8 @@ struct netfront_dev *init_netfront(char *_nodename, void (*thenetif_rx)(unsigned
> >> >>      dev->fd = -1;
> >> >>  #endif
> >> >>
> >> >> -    printk("net TX ring size %d\n", NET_TX_RING_SIZE);
> >> >> -    printk("net RX ring size %d\n", NET_RX_RING_SIZE);
> >> >> +    printk("net TX ring size %llu\n", (unsigned long long) NET_TX_RING_SIZE);
> >> >> +    printk("net RX ring size %llu\n", (unsigned long long) NET_RX_RING_SIZE);
> >> >
> >> > lib/printf.c does not actually support %ll yet, it uses %L instead.
> >>
> >> Actually, it looks like it does:
> >>
> >> if (qualifier == 'l' && *fmt == 'l') {
> >>   qualifier = 'L';
> >>   ++fmt;
> >> }
> >
> > Oh, I missed this part of the code, thanks. For this exerpt of code,
> > long long is way too much, but for things like disk offsets and such,
> > it'd be preferrable to go for long long.
> 
> I see this patch never go applied. To be clear: the above comment was
> just for information. I agree with using %lu in the patch (as the
> latest version of the patch does).
> 
> So, I think this is ready to be applied:
> 
> http://lists.xenproject.org/archives/html/xen-devel/2014-08/msg00998.html

Samuel, are you ok with the patch after those clarifications?

Ian.




_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel