]> Pileus Git - ~andy/linux/commit
powerpc: Make hard_irq_disable() do the right thing vs. irq tracing
authorBenjamin Herrenschmidt <benh@kernel.crashing.org>
Mon, 6 May 2013 21:04:02 +0000 (21:04 +0000)
committerBenjamin Herrenschmidt <benh@kernel.crashing.org>
Tue, 7 May 2013 07:13:57 +0000 (17:13 +1000)
commit5737789c8340620d7b542d1d4e9b197de8eb2801
treebd441fd76c05008b482c190d219978f9df8bb972
parentd5dae721308ac5d5db03bf3aebaa15fcdcfb4862
powerpc: Make hard_irq_disable() do the right thing vs. irq tracing

If hard_irq_disable() is called while interrupts are already soft-disabled
(which is the most common case) all is already well.

However you can (and in some cases want) to call it while everything is
enabled (to make sure you don't get a lazy even, for example before entry
into KVM guests) and in this case we need to inform the irq tracer that
the irqs are going off.

We have to change the inline into a macro to avoid an include circular
dependency hell hole.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
arch/powerpc/include/asm/hw_irq.h