| To: | Guy Streeter <guy.streeter@gmail.com> |
|---|---|
| Subject: | Re: NR_IRQS and possible irq values on malta |
| From: | "Maciej W. Rozycki" <macro@linux-mips.org> |
| Date: | Thu, 3 Feb 2005 22:45:06 +0000 (GMT) |
| Cc: | linux-mips <linux-mips@linux-mips.org> |
| In-reply-to: | <52dd17640502031426660c7196@mail.gmail.com> |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <52dd17640502031426660c7196@mail.gmail.com> |
| Sender: | linux-mips-bounce@linux-mips.org |
On Thu, 3 Feb 2005, Guy Streeter wrote: > In mips_pcibios_iack() the irq number is obtained like this: > > irq = GT_READ(GT_PCI0_IACK_OFS); > irq &= 0xff; > > (for coreLV, but similarly for others). This value is used to index > into an array of size NR_IRQS, which is set in mach-generic/irq.h to > 128. > I realize that when things work right, the value of irq isn't likely > to get that big, but it seems to me it should be masked down to 127 or > the NR_IRQS value should be larger. I have no opinion which is the > right thing to do. It looks like a problem in __do_IRQ() (in kernel/irq/handle.c) -- it should probably call "BUG_ON(irq >= NR_IRQS)". MIPS-specific code doesn't care. Maciej |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | NR_IRQS and possible irq values on malta, Guy Streeter |
|---|---|
| Next by Date: | Re: [PATCH] Fix Kconfig for Broadcom SWARM, Ralf Baechle |
| Previous by Thread: | NR_IRQS and possible irq values on malta, Guy Streeter |
| Next by Thread: | dcache aliasing problem on fork, Atsushi Nemoto |
| Indexes: | [Date] [Thread] [Top] [All Lists] |