Re: PATCH powerpc: Merge asm-ppc*/hardirq.h
From: Jon Loeliger <hidden>
Date: 2005-09-14 13:41:34
On Wed, 2005-09-14 at 06:05, Christoph Hellwig wrote:
On Tue, Sep 13, 2005 at 03:38:40PM -0500, Jon Loeliger wrote:quoted
+#ifndef _ASM_POWERPC_HARDIRQ_H +#define _ASM_POWERPC_HARDIRQ_H +#ifdef __KERNEL__the __KERNEL__ ifdefs is not needed. <asm/hardirq.h> is only included from <linux/hardirq.h> which doesn't have anything user-visible.
Will do.
quoted
+ +#include <linux/config.h>not needed.quoted
+#include <linux/cache.h> + +#ifdef __powerpc64__ +#include <linux/preempt.h> +#else +#include <linux/smp_lock.h> +#include <asm/irq.h> +#endifWe shouldn't need either of these include blocks at all.
OK. I'll do some out-ripping and verify that it all compiles still.
quoted
+ +/* The __last_jiffy_stamp field is needed to ensure that no decrementer + * interrupt is lost on SMP machines. Since on most CPUs it is in the same + * cache line as local_irq_count, it is cheap to access and is also used on UP + * for uniformity. + */ +typedef struct { + unsigned int __softirq_pending; /* set_bit is used on this */ + unsigned int __last_jiffy_stamp; +} ____cacheline_aligned irq_cpustat_t;I'd suggest just using a DECLARE_PER_CPU variable for last_jiffy_stamp. In facct I plan to get rid of irq_cpustat_t completely at some point.
Hmmm... So, I don't want to mess that up. How about I leave it as I've suggested and leave that step up to you as a follow on? Thanks, jdl