On Fri, Jan 07, 2011 at 09:04:55AM -0800, Tony Lindgren wrote: > * Felipe Balbi <balbi@xxxxxx> [110106 23:47]: > > > > The main point of using Sparse IRQ numbering is exactly avoiding > > pre-processor branches. Instead of defining ranges only when a device is > > compile, we can always keep the range allocated no matter if the device > > probes or not. So, I suggest dropping the ifdeffery on <plat/irqs.h> and > > move that to something like below: > > This is fine as long as we we don't run out of allocated IRQs before > switching over to sparse IRQ. The original reason for the ifdeffery > was to keep NR_IRQS below 256 when possible as it optimizes the interrupt > handling in that case. See NR_IRQS in arch/arm/include/asm/hardirq.h > for more info. I've sent patches (you're in Cc) enabling SPARSE_IRQ on OMAP. I boot tested on panda and overo and it was working fine. -- balbi -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
