[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190201153726.e3ce44c4a3fb22f5f5d6bcf3@suse.de>
Date: Fri, 1 Feb 2019 15:37:26 +0100
From: Thomas Bogendoerfer <tbogendoerfer@...e.de>
To: Christoph Hellwig <hch@...radead.org>
Cc: Bjorn Helgaas <bhelgaas@...gle.com>,
James Hogan <jhogan@...nel.org>, linux-kernel@...r.kernel.org,
linux-mips@...r.kernel.org, linux-pci@...r.kernel.org,
Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
Paul Burton <paul.burton@...s.com>,
Ralf Baechle <ralf@...ux-mips.org>
Subject: Re: [PATCH 7/7] MIPS: SGI-IP27: abstract chipset irq from bridge
On Mon, 28 Jan 2019 08:27:21 -0800
Christoph Hellwig <hch@...radead.org> wrote:
> On Mon, Jan 28, 2019 at 03:01:35PM +0100, Thomas Bogendoerfer wrote:
> > On Mon, 28 Jan 2019 05:33:17 -0800
> > Christoph Hellwig <hch@...radead.org> wrote:
> >
> > > Shouldnt this just use chained irqchip drivers instead?
> >
> > you mean using irq_set_chained_handler() ? If yes, this IMHO doesn't look usefull
> > because it's used for adding a secondary interrupt controller. But what I need
> > is telling bridge ASIC to direct the xtalk IRQ packet to a specific HUB/HEART/BEDROCK
> > from the HUB/HEART/BEDROCK specific code. And want to avoid dragging in bridge details
> > to that specific code.
>
> Yes, but don't we have nested interrupt controllers here? Even if they
> don't really do much in the fast path the setup does look chained to me.
> Then again I'm not really an expert in the irq handling code nor in this
> hardware, so maybe Thomas or Marc might have a better idea.
found a presentation from Marc and what I need here is irqdomain hierarchy.
Which meana converting the whole hub irq code to irqdomain first. Let's see
how easy this will be:-)
Thomas.
--
SUSE Linux GmbH
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Powered by blists - more mailing lists