[Mesa-dev] [PATCH] genxml: Handle failure of Python codegen scripts.

[Dylan Baker]

Quoting Matt Turner (2016-10-26 14:06:02)
> On Wed, Oct 26, 2016 at 2:00 PM, Jason Ekstrand <jason at jlekstrand.net> wrote:
> > On Oct 26, 2016 12:51 PM, "Matt Turner" <mattst88 at gmail.com> wrote:
> >>
> >> ---
> >>  src/intel/Makefile.genxml.am | 2 +-
> >>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/src/intel/Makefile.genxml.am b/src/intel/Makefile.genxml.am
> >> index 2df223a..2ed2741 100644
> >> --- a/src/intel/Makefile.genxml.am
> >> +++ b/src/intel/Makefile.genxml.am
> >> @@ -33,7 +33,7 @@ $(GENXML_GENERATED_FILES): genxml/gen_pack_header.py
> >>
> >>  .xml_pack.h:
> >>         $(MKDIR_GEN)
> >> -       $(PYTHON_GEN) $(srcdir)/genxml/gen_pack_header.py $< > $@
> >> +       $(PYTHON_GEN) $(srcdir)/genxml/gen_pack_header.py $< > $@ ||
> >> ($(RM) $@; false)
> >
> > Thanks for fixing this!  I think the way we do this most other places is
> > with "> $@.tmp && mv $@.tmp $&” not that I care that much...
> 
> In fact, the opposite is true. We do the .tmp trick with the git sha1
> file, but use what's done in this patch in multiple places in
> Makefile.glsl.am and Makefile.nir.am, and it's even spread to
> freedreno and i965.

I like this approach better, it doesn't allow a previously building commit to
shadow a failing build, since if the generator is triggered and fails there
won't be a file at all, and the build will fail.

> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 455 bytes
Desc: signature
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20161026/63701a18/attachment.sig>