Hi
On Thu, Sep 29, 2022 at 9:53 PM Richard Henderson <
richard.hender...@linaro.org> wrote:
> On 9/29/22 06:41, marcandre.lur...@redhat.com wrote:
> > void qemu_thread_naming(bool enable)
> > {
> > /* But note we don't actually name them on Windows yet */
> > name_threads = enable;
> >
> > - fprintf(stderr, "qemu: thread naming not supported on this host\n");
> > + if (enable && !load_set_thread_description()) {
> > + fprintf(stderr, "qemu: thread naming not supported on this
> host\n");
> > + }
> > }
>
> Comment is out of date, and I think it would be better to *not* set
> name_threads if not
> supported, rather than...
>
Comment removed.
>
>
> > +static bool
> > +set_thread_description(HANDLE h, const char *name)
> > +{
> > + HRESULT hr;
> > + g_autofree wchar_t *namew = NULL;
> > +
> > + if (!load_set_thread_description() || !name) {
> > + return false;
> > + }
>
> ... have to re-query load_set_thread_description later.
>
The load_set_thread_description() function is actually a "one-time"
function, it doesn't re-load.
>
> Also, unused return value; might as well be void.
>
Right, maybe it should warn if it failed to set the name?
--
Marc-André Lureau
