Skip to main content

Browse the archive

Revision 251b97f552b1ad414cc5a9ccc8e4e94503edd5fc authored by Peter Zijlstra on 04 July 2008, 16:59:24 UTC, committed by Linus Torvalds on 04 July 2008, 17:40:04 UTC 
mm: dirty page accounting vs VM_MIXEDMAP
 
Dirty page accounting accurately measures the amound of dirty pages in
writable shared mappings by mapping the pages RO (as indicated by
vma_wants_writenotify).  We then trap on first write and call
set_page_dirty() on the page, after which we map the page RW and
continue execution.

When we launder dirty pages, we call clear_page_dirty_for_io() which
clears both the dirty flag, and maps the page RO again before we start
writeout so that the story can repeat itself.

vma_wants_writenotify() excludes VM_PFNMAP on the basis that we cannot
do the regular dirty page stuff on raw PFNs and the memory isn't going
anywhere anyway.

The recently introduced VM_MIXEDMAP mixes both !pfn_valid() and
pfn_valid() pages in a single mapping.

We can't do dirty page accounting on !pfn_valid() pages as stated
above, and mapping them RO causes them to be COW'ed on write, which
breaks VM_SHARED semantics.

Excluding VM_MIXEDMAP in vma_wants_writenotify() would mean we don't do
the regular dirty page accounting for the pfn_valid() pages, which
would bring back all the head-aches from inaccurate dirty page
accounting.

So instead, we let the !pfn_valid() pages get mapped RO, but fix them
up unconditionally in the fault path.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: "Jared Hulbert" <jaredeh@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent cde5353
Raw File
ratelimit.c 
/*
 * ratelimit.c - Do something with rate limit.
 *
 * Isolated from kernel/printk.c by Dave Young <hidave.darkstar@gmail.com>
 *
 * This file is released under the GPLv2.
 *
 */

#include <linux/kernel.h>
#include <linux/jiffies.h>
#include <linux/module.h>

/*
 * __ratelimit - rate limiting
 * @ratelimit_jiffies: minimum time in jiffies between two callbacks
 * @ratelimit_burst: number of callbacks we do before ratelimiting
 *
 * This enforces a rate limit: not more than @ratelimit_burst callbacks
 * in every ratelimit_jiffies
 */
int __ratelimit(int ratelimit_jiffies, int ratelimit_burst)
{
	static DEFINE_SPINLOCK(ratelimit_lock);
	static unsigned toks = 10 * 5 * HZ;
	static unsigned long last_msg;
	static int missed;
	unsigned long flags;
	unsigned long now = jiffies;

	spin_lock_irqsave(&ratelimit_lock, flags);
	toks += now - last_msg;
	last_msg = now;
	if (toks > (ratelimit_burst * ratelimit_jiffies))
		toks = ratelimit_burst * ratelimit_jiffies;
	if (toks >= ratelimit_jiffies) {
		int lost = missed;

		missed = 0;
		toks -= ratelimit_jiffies;
		spin_unlock_irqrestore(&ratelimit_lock, flags);
		if (lost)
			printk(KERN_WARNING "%s: %d messages suppressed\n",
				__func__, lost);
		return 1;
	}
	missed++;
	spin_unlock_irqrestore(&ratelimit_lock, flags);
	return 0;
}
EXPORT_SYMBOL(__ratelimit);
back to top