Thread (7 messages) flat view 7 messages, 2 authors, 2006-08-24

Re: [PATCH] powerpc: update mpc7448hpc2 board irq supportusingdevice tree

From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: 2006-08-24 08:53:39

On Thu, 2006-08-24 at 16:36 +0800, Zang Roy-r61911 wrote:
On Thu, 2006-08-24 at 13:50, Benjamin Herrenschmidt wrote:
quoted
quoted
-struct hw_interrupt_type tsi108_pci_irq = {
+static struct irq_chip tsi108_pci_irq = {
      .typename = "tsi108_PCI_int",
-     .enable = tsi108_pci_irq_enable,
-     .disable = tsi108_pci_irq_disable,
+     .mask = tsi108_pci_irq_disable,
      .ack = tsi108_pci_irq_ack,
      .end = tsi108_pci_irq_end,
+     .unmask = tsi108_pci_irq_enable,
 };
 
 /*
While the patch as-is looks ok, it also looks like you could take
advantage of the new genirq code to clean up your TSI irq handling a
bit. You probably don't need a end() handler anymore provided that you
properly set your main handler to be either level, edge or fasteoi (in
which case, you need an eoi handler).
I need the end() handler. The main handler can not proper process the
case.
None of the 3 main handlers in kernel/irq/chip.c (You can set them at
host_map time with set_irq_chip_and_handler, if you don't do it, you end
up with the old-style main handler, but one of the 3 new ones might be
more suitable to your flow) ? Strange... I would need to look in more
details, but it's not urgent, the current stuff should work.

Ben
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help