Re: [PATCH v4 13/14] xen: ifdef inclusion of <asm/grant_table.h> in <xen/grant_table.h>

[Oleksii <oleksii.kurochko@xxxxxxxxx>]

On Mon, 2023-11-27 at 15:41 +0100, Jan Beulich wrote:
> On 27.11.2023 15:13, Oleksii Kurochko wrote:
> > --- a/xen/arch/ppc/include/asm/grant_table.h
> > +++ /dev/null
> > @@ -1,5 +0,0 @@
> > -/* SPDX-License-Identifier: GPL-2.0-only */
> > -#ifndef __ASM_PPC_GRANT_TABLE_H__
> > -#define __ASM_PPC_GRANT_TABLE_H__
> > -
> > -#endif /* __ASM_PPC_GRANT_TABLE_H__ */
> 
> Removing this header would be correct only if GRANT_TABLE had a
> "depends on
> !PPC", I'm afraid. Recall that the earlier randconfig adjustment in
> CI was
> actually requested to be undone, at which point what an arch's
> defconfig
> says isn't necessarily what a randconfig should use.
We can do depends on !PPC && !RISCV but shouldn't it be enough only to
turn CONFIG_GRANT_TABLE off in defconfig and set CONFIG_GRANT_TABLE=n
in EXTRA_XEN_CONFIG?

Some time ago I also tried to redefine "Config GRANT_TABLE" in arch-
specific Kconfig + defconfig + EXTRA_XEN_CONFIG and it works for me.
Could it be solution instead of "depends on..." ?

One more question I have do we really need this randconfig? On RISC-V
side, I launched several time this patch series ( from v1 to v4 + runs
during test of patch series ) and I haven't faced case
when CONFIG_GRANT_TABLE=n. ( but I turned the config off in defconfig +
EXTRA_XEN_CONFIG ). Also when it "Config GRANT_TABLE" was re-defined in
arch-specific KConfig I haven't face an issue with CONFIG_GRANT_TABLE
too.

~ Oleksii