Re: [PATCH RFC v3 13/21] ACPICA: Add new MADT GICC flags fields

On Mon, Dec 18, 2023 at 02:14:30PM +0100, Rafael J. Wysocki wrote:
> On Mon, Dec 18, 2023 at 10:23 AM Lorenzo Pieralisi
> <[email protected]> wrote:
> >
> > On Fri, Dec 15, 2023 at 04:53:28PM +0000, Russell King (Oracle) wrote:
> > > On Fri, Dec 15, 2023 at 04:23:22PM +0000, Jonathan Cameron wrote:
> > > > On Wed, 13 Dec 2023 12:50:18 +0000
> > > > Russell King (Oracle) <[email protected]> wrote:
> > > >
> > > > > From: James Morse <[email protected]>
> > > > >
> > > > > Add the new flag field to the MADT's GICC structure.
> > > > >
> > > > > 'Online Capable' indicates a disabled CPU can be enabled later. See
> > > > > ACPI specification 6.5 Tabel 5.37: GICC CPU Interface Flags.
> > > > >
> > > > > Signed-off-by: James Morse <[email protected]>
> > > > > Tested-by: Miguel Luis <[email protected]>
> > > > > Tested-by: Vishnu Pajjuri <[email protected]>
> > > > > Tested-by: Jianyong Wu <[email protected]>
> > > > > Signed-off-by: Russell King (Oracle) <[email protected]>
> > > >
> > > > I see there is an acpica pull request including this bit but with a different name
> > > > For reference.
> > > > https://github.com/acpica/acpica/pull/914/commits/453a5f67567786522021d5f6913f561f8b3cabf6
> > > >
> > > > +CC Lorenzo who submitted that.
> > >
> > > > > +#define ACPI_MADT_GICC_CPU_CAPABLE      (1<<3)   /* 03: CPU is online capable */
> > > >
> > > > ACPI_MADT_GICC_ONLINE_CAPABLE
> > >
> > > It's somewhat disappointing, but no big deal. It's easy enough to change
> > > "irqchip/gic-v3: Add support for ACPI's disabled but 'online capable' CPUs"
> > > to use Lorenzo's name when that patch hits - and it becomes one less
> > > patch in this patch set when Lorenzo's change eventually hits mainline.
> > >
> > > Does anyone know how long it may take for Lorenzo's change to get into
> > > mainline? Would it be by the 6.8 merge window or the following one?
> >
> > I wish I knew. I submitted ACPICA changes for the online capable bit
> > since I had to add additional flags on top (ie DMA coherent) and it
> > would not make sense to submit the latter without the former.
> >
> > I'd be great if the ACPICA headers can make it into Linux for the upcoming
> > merge window, not sure what I can do to fasttrack the process though
> > (I shall ping the maintainers).
> 
> If your upstream pull request has been merged, I can pick up Linux
> patches carrying Link: tags pointing to the upstream ACPICA commits in
> that pull request.

ACPICA PR was merged, sent the Linuxized version along with the GIC changes
here:

https://lore.kernel.org/lkml/[email protected]

Thanks,
Lorenzo