Re: [RFC PATCH v1 1/1] xen/Makefile: introduce ARCH_FIXED_CONFIG for randconfig

[Oleksii <oleksii.kurochko@xxxxxxxxx>]

On Thu, 2023-12-07 at 17:01 -0800, Stefano Stabellini wrote:
> On Thu, 7 Dec 2023, Oleksii wrote:
> > On Thu, 2023-12-07 at 20:17 +0000, Andrew Cooper wrote:
> > > On 07/12/2023 5:03 pm, Oleksii Kurochko wrote:
> > > > ARCH_FIXED_CONFIG is required in the case of randconfig
> > > > and CI for configs that aren't ready or are not
> > > > supposed to be implemented for specific architecture.
> > > > These configs should always be disabled to prevent randconfig
> > > > related tests from failing.
> > > > 
> > > > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>
> > > > ---
> > > >  xen/Makefile | 5 ++++-
> > > >  1 file changed, 4 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/xen/Makefile b/xen/Makefile
> > > > index ca571103c8..8ae8fe1480 100644
> > > > --- a/xen/Makefile
> > > > +++ b/xen/Makefile
> > > > @@ -336,11 +336,14 @@ ifeq ($(config-build),y)
> > > >  # *config targets only - make sure prerequisites are updated,
> > > > and
> > > > descend
> > > >  # in tools/kconfig to make the *config target
> > > >  
> > > > +ARCH_FORCED_CONFIG :=
> > > > $(srctree)/arch/$(SRCARCH)/configs/randomforced.config
> > > > +
> > > >  # Create a file for KCONFIG_ALLCONFIG which depends on the
> > > > environment.
> > > >  # This will be use by kconfig targets
> > > > allyesconfig/allmodconfig/allnoconfig/randconfig
> > > >  filechk_kconfig_allconfig = \
> > > >      $(if $(findstring n,$(XEN_HAS_CHECKPOLICY)), echo
> > > > 'CONFIG_XSM_FLASK_POLICY=n';) \
> > > > -    $(if $(KCONFIG_ALLCONFIG), cat $(KCONFIG_ALLCONFIG);) \
> > > > +    $(if $(KCONFIG_ALLCONFIG), cat $(KCONFIG_ALLCONFIG); \
> > > > +    $(if $(wildcard $(ARCH_FORCED_CONFIG)), cat
> > > > $(ARCH_FORCED_CONFIG);) ) \
> > > >      :
> > > >  
> > > >  .allconfig.tmp: FORCE
> > > 
> > > We already have infrastructure for this.  What's wrong with
> > > EXTRA_FIXED_RANDCONFIG?
> > Everything is fine; I don't know why there was only an issue with
> > CONFIG_GRANT_TABLE on PPC. On the RISC-V side, there were more
> > configs
> > issues, prompting me to include all the configurations not
> > implemented
> > for RISC-V in EXTRA_FIXED_RANDCONFIG. You can find the added
> > configurations in this commit:
> > https://lore.kernel.org/xen-devel/b4e85f8f58787b4d179022973ce25673d6b56e36.1700761381.git.oleksii.kurochko@xxxxxxxxx/#Z31automation:gitlab-ci:build.yaml
> > 
> > One challenge is that the same configurations need to be added
> > multiple
> > times for each build test using randconfig.
> 
> That's a lot of extra configs to add. Could you use a yaml anchor or
> a
> .something to include? So that you define the full list only once at
> the
> top of the file and then reuse it everywhere as needed.
I didn't know that it work. If there is such options, I think it can be
a solution.

Thanks. I'll try.

> 
> 
> > Another reason for this approach is a suggestion from Jan (probably
> > I
> > misunderstood it), who proposed using a template to instruct
> > randconfig
> > not to modify currently unnecessary configurations. You can find
> > the
> > suggestion and discussion here:
> > https://lore.kernel.org/xen-devel/008d0c66-6816-4d12-9e1f-1878e982f9fc@xxxxxxxx/
> > 
> > Perhaps we could enhance the build script to fetch "fixed" configs
> > from
> > the architecture-specific fixed-defconfig instead of modifying the
> > Makefile directly.
> 
> Sorry I missed the original thread somehow. Please use "automation"
> as
> subject line tag for automation patches.
Sure. I'll do next time.

~ Oleksii