From mboxrd@z Thu Jan 1 00:00:00 1970 From: hollis@austin.ibm.com Date: Mon, 3 Dec 2001 16:46:16 -0600 To: Michael Sokolov Cc: linuxppc-dev@lists.linuxppc.org, davidm@amberdata.demon.co.uk Subject: Re: RFC: i8259.c & PCI int ack Message-ID: <20011203164616.I421@austin.ibm.com> References: <0112032237.AA11733@ivan.Harhan.ORG> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <0112032237.AA11733@ivan.Harhan.ORG> Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: On Mon, Dec 03, 2001 at 02:37:56PM -0800, Michael Sokolov wrote: > hollis@austin.ibm.com wrote: > > > - i8259_init now takes a 'long intack_addr', which on PReP is 0xbffffff0. > > For platforms that only poll, this argument is irrelevant and can be 0x0 > > > > [...] > > > > If this is the right idea, I intend to change all 8259 users: > > - ppc_md.get_irq = i8259_irq -> ppc_md.get_irq = i8259_poll > > - i8259_init() -> i8259_init(0x0) > > - i8259_irq(smp_processor_id()) -> i8259_poll() > > With this patch on platforms that only poll, your code will still do an > ioremap on i8259_init's argument, even if it's 0. I'm not sure that is good. Good point; fixed. -Hollis ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/