[RFC,0/2] NMI IPI work in progress for Linux and OPAL

Message ID	20170912160553.13422-1-npiggin@gmail.com
Headers	show Return-Path: <skiboot-bounces+incoming=patchwork.ozlabs.org@lists.ozlabs.org> From: Nicholas Piggin <npiggin@gmail.com> To: linuxppc-dev@lists.ozlabs.org, skiboot@lists.ozlabs.org Date: Wed, 13 Sep 2017 02:05:51 +1000 Message-Id: <20170912160553.13422-1-npiggin@gmail.com> Subject: [Skiboot] [RFC PATCH 0/2] NMI IPI work in progress for Linux and OPAL Precedence: list Cc: Alistair Popple <alistair@popple.id.au> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Errors-To: skiboot-bounces+incoming=patchwork.ozlabs.org@lists.ozlabs.org Sender: "Skiboot" <skiboot-bounces+incoming=patchwork.ozlabs.org@lists.ozlabs.org>
Series	NMI IPI work in progress for Linux and OPAL \| expand [RFC,0/2] NMI IPI work in progress for Linux and OPAL [RFC,1/2] core: implement OPAL_SIGNAL_SYSTEM_RESET with POWER9 scoms [RFC,2/2] powerpc/powernv: implement NMI IPIs with OPAL_SIGNAL_SYSTEM_RESET

Message ID

20170912160553.13422-1-npiggin@gmail.com

Headers

From: Nicholas Piggin <npiggin@gmail.com>
To: linuxppc-dev@lists.ozlabs.org,
	skiboot@lists.ozlabs.org
Date: Wed, 13 Sep 2017 02:05:51 +1000
Message-Id: <20170912160553.13422-1-npiggin@gmail.com>
Subject: [Skiboot] [RFC PATCH 0/2] NMI IPI work in progress for Linux and
	OPAL
Precedence: list
Cc: Alistair Popple <alistair@popple.id.au>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
Errors-To: skiboot-bounces+incoming=patchwork.ozlabs.org@lists.ozlabs.org
Sender: "Skiboot"
	<skiboot-bounces+incoming=patchwork.ozlabs.org@lists.ozlabs.org>

Series

NMI IPI work in progress for Linux and OPAL | expand

Message

Nicholas Piggin Sept. 12, 2017, 4:05 p.m. UTC

Hi,

I have Linux and OPAL patches that make the NMI IPI facility work on
a POWER9 DD1 here, lightly tested. It works for threads that are running
and are in stop (at least the stop0_lite state enabled in DD1).

Comments on the OPAL patch in particular would be good. Next step will
be more testing with DD2 and different idle states.

This was taken mostly from pdbg but I have ended up changing a few
things (e.g., to make it work in idle state).

Thanks,
Nick