Bug#392592: a simpler patch
On Wed, Jan 03, 2007 at 11:51:32PM +0100, Bastian Blank wrote:
> On Tue, Jan 02, 2007 at 03:34:07PM -0700, dann frazier wrote:
> > Here's a patch that simply uses hardcoded definitions instead of
> > doing the dynamic tests that require architecture-specific scripts.
>
> We hardcode the compiler version also.
Yes, but we do not hardcoce specific versions of objdump or readelf, which
are also used by these scripts to generate their output.
My primary concern is that, in the future, the output of these scripts
will change and we (or our successors) will either not notice or
forget to update the hardcoded values.
Including the scripts in linux-kbuild will avoid this manual step
altogether, and allow for the possibility of other archs to provide
their own scripts in the future.
> Patches never comment out things, they always remove them.
ok
> > ifeq ($(GAS_STATUS),buggy)
> > $(error Sorry, you need a newer version of the assember, one that is built from \
>
> Just remove that part completely.
ok
> > @@ -77,7 +81,8 @@
> > $(Q)$(MAKE) $(build)=$(boot) $@
> >
> > unwcheck: vmlinux
> > - -$(Q)READELF=$(READELF) $(srctree)/arch/ia64/scripts/unwcheck.py $<
> > +# -$(Q)READELF=$(READELF) $(srctree)/arch/ia64/scripts/unwcheck.py $<
> > + @echo Warning: unwcheck test skipped to workaround missing arch scripts, see \#392592
>
> Warning not neccesary, just drop the whole target.
ok. I will incorporate your feedback on this version of the package
and revert my changes to linux-kbuild in svn.
However, I would like to continue discussing the idea of
including the scripts in linux-kbuild-2.6. If you are ok with the
idea but unhappy with my implementation, I'm certainly willing to make
modifications. If the concern is the timing, I'm certainly willing to
delay this till after etch - the freeze makes it very unlikely that
these harcoded values will become incorrect before the etch release.
--
dann frazier
Reply to: