Messages in this thread | | | Date | Thu, 15 Jul 2021 12:31:02 +0100 | From | Sudeep Holla <> | Subject | Re: [PATCH 12/13] mailbox: pcc: Add support for PCCT extended PCC subspaces(type 3/4) |
| |
On Wed, Jul 14, 2021 at 07:52:16PM +0100, Cristian Marussi wrote: > On Thu, Jul 08, 2021 at 07:08:50PM +0100, Sudeep Holla wrote: > > With all the plumbing in place to avoid accessing PCCT type and other > > fields directly from the PCCT table all the time, let us now add the > > support for extended PCC subspaces(type 3 and 4). > > > > Signed-off-by: Sudeep Holla <sudeep.holla@arm.com> > > --- > > Hi Sudeep, > > just a few general observation on Type 3/4 subspaces from the spec > Table 14.7: > > - "If a slave-subspace is present in the PCCT, then the platform interrupt flag must be set to 1." table 14.7 > > Maybe is worth to WARN on this if this assumption is violated by the > ACPI table we found. > > - "Note that if interrupts are edge triggered, then each subspace must have its own unique interrupt." > > Same here, maybe it's worth also to check this, so that after all the > pchan->db_irq has been obtained no edge triggered irqs are duplicated > before requesting them. >
Both valid points and in my TODO list, just forgot to address this as I didn't have a setup to check/flag these up. I will address them in next version.
Thanks for reviewing these patches.
-- Regards, Sudeep
| |