2005-12-22 22:31:41

by Ravikiran G Thirumalai

Subject: [patch] mm: Patch to convert global dirty_exceeded flag to per-node node_dirty_exceeded

Patch to convert global dirty_exceeded flag to per-node
node_dirty_exceeded.

dirty_exceeded ping pongs between nodes in order to force all cpus in
the system to increase the frequency of calls to balance_dirty_pages.

Currently dirty_exceeded is used by balance_dirty_pages_ratelimited to
force all CPUs in the system call balance_dirty_pages often, in order to
reduce the amount of dirty pages in the entire system (based on
dirty_thresh and one CPU exceeding thee ratelimits). As dirty_exceeded
is a global variable, it will ping-pong between nodes of a NUMA system
which is not good.

We think it is OK to reduce the affect of dirty_exceeded to node basis.
That will give us on one hand having multiple CPUs (of that node)
calling balance_dirty_pages often (based on same logic as before) and on
the other hand not causing ping-pong of dirty_exceeded with the other
nodes.

Following patch changes the global dirty_exceeded flag to a per-node
flag. Hence, balance_dirty_pages is called more often only on the node
which exceeded the dirty_threshold. The SMP case is left as was before.

Comments?

Signed-off-by: Alok N Kataria <[email protected]>
Signed-off-by: Ravikiran Thirumalai <[email protected]>
Signed-off-by: Shai Fultheim <[email protected]>

Index: linux-2.6.15-rc6/mm/page-writeback.c
===================================================================
--- linux-2.6.15-rc6.orig/mm/page-writeback.c 2005-12-21 18:29:38.000000000 -0800
+++ linux-2.6.15-rc6/mm/page-writeback.c 2005-12-21 23:43:33.000000000 -0800
@@ -46,7 +46,20 @@
static long ratelimit_pages = 32;

static long total_pages; /* The total number of pages in the machine. */
-static int dirty_exceeded; /* Dirty mem may be over limit */
+
+/*
+ * node_dirty_exceeded is used to indicate that dirty mem in the current node
+ * maybe over the limit and we need to increase the frequency of calls to
+ * balance_dirty_pages
+ */
+#ifdef CONFIG_NUMA
+static DEFINE_PER_CPU(int, _dirty_exceeded) = { 0 };
+#define node_dirty_exceeded \
+ per_cpu(_dirty_exceeded, node_to_first_cpu(numa_node_id()))
+#else
+static int node_dirty_exceeded;
+#endif
+

/*
* When balance_dirty_pages decides that the caller needs to perform some
@@ -212,7 +225,7 @@
if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh)
break;

- dirty_exceeded = 1;
+ node_dirty_exceeded = 1;

/* Note: nr_reclaimable denotes nr_dirty + nr_unstable.
* Unstable writes are a feature of certain networked
@@ -235,7 +248,7 @@
}

if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh)
- dirty_exceeded = 0;
+ node_dirty_exceeded = 0;

if (writeback_in_progress(bdi))
return; /* pdflush is already working this queue */
@@ -272,7 +285,7 @@
long ratelimit;

ratelimit = ratelimit_pages;
- if (dirty_exceeded)
+ if (node_dirty_exceeded)
ratelimit = 8;

/*

2005-12-22 22:55:06

by Ravikiran G Thirumalai

Subject: Re: [patch] mm: Patch to convert global dirty_exceeded flag to per-node node_dirty_exceeded

On Thu, Dec 22, 2005 at 02:31:39PM -0800, Ravikiran G Thirumalai wrote:
> Patch to convert global dirty_exceeded flag to per-node
> node_dirty_exceeded.

This depends on the compile fix for node_to_fist_cpu I posted before
if this code has to run on x86_64/ia64

http://marc.theaimsgroup.com/?l=linux-kernel&m=113529019215045&w=2

Thanks,
Kiran