On Tue, Jul 14, 2015 at 11:23:27AM +0100, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> > On Mon, 2015-07-13 at 18:12 +0100, Ian Jackson wrote:
> > > There are no call sites in libxl which can pass a NULL.
> > >
> > > I think that if we are to retain th
On Tue, 2015-07-14 at 14:58 +0100, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> > On Tue, 2015-07-14 at 11:23 +0100, Ian Jackson wrote:
> > > If we don't want to support it we can just remove the check and let it
> > > crash.
> >
> > An assert at least
Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> On Tue, 2015-07-14 at 11:23 +0100, Ian Jackson wrote:
> > If we don't want to support it we can just remove the check and let it
> > crash.
>
> An assert at least gets us a slightly improved message when it happens.
If that log
On Tue, 2015-07-14 at 11:23 +0100, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> > On Mon, 2015-07-13 at 18:12 +0100, Ian Jackson wrote:
> > > There are no call sites in libxl which can pass a NULL.
> > >
> > > I think that if we are to retain this feat
Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> On Mon, 2015-07-13 at 18:12 +0100, Ian Jackson wrote:
> > There are no call sites in libxl which can pass a NULL.
> >
> > I think that if we are to retain this feature, it ought to be
> > documented, at least.
>
> Or add an ass
On Mon, 2015-07-13 at 18:12 +0100, Ian Jackson wrote:
> Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> > I rather dislike subjects of the form "fix $function", since it gives
> > very little clue to someone reading the shortlog what is going on.
>
> Yes.
>
> > In this case
Ian Campbell writes ("Re: [PATCH 1/9] libxl: fix libxl__abs_path"):
> I rather dislike subjects of the form "fix $function", since it gives
> very little clue to someone reading the shortlog what is going on.
Yes.
> In this case I think "libxl: make libxl__abs_path correctly handle a
> NULL argum
On Mon, Jul 13, 2015 at 10:57:32AM +0100, Ian Campbell wrote:
> On Fri, 2015-07-10 at 19:00 +0100, Wei Liu wrote:
>
> I rather dislike subjects of the form "fix $function", since it gives
> very little clue to someone reading the shortlog what is going on.
>
> In this case I think "libxl: make li
On Fri, 2015-07-10 at 19:00 +0100, Wei Liu wrote:
I rather dislike subjects of the form "fix $function", since it gives
very little clue to someone reading the shortlog what is going on.
In this case I think "libxl: make libxl__abs_path correctly handle a
NULL argument" would be an accurate descr
