* [PATCH 01/21] Create vmstat.c/.h from page_alloc.c/.h
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
@ 2006-06-12 21:12 ` Christoph Lameter
2006-06-12 21:12 ` [PATCH 02/21] Basic ZVC (zoned vm counter) implementation Christoph Lameter
` (19 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:12 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Move counter code from page_alloc.c/page-flags.h to vmstat.c/h.
Create vmstat.c/vmstat.h by separating the counter code and the proc functions.
Move the vm_stat_text array before zoneinfo_show.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Index: linux-2.6.17-rc6-cl/include/linux/page-flags.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/page-flags.h 2006-06-12 12:42:50.853428498 -0700
+++ linux-2.6.17-rc6-cl/include/linux/page-flags.h 2006-06-12 12:43:12.081607020 -0700
@@ -5,11 +5,7 @@
#ifndef PAGE_FLAGS_H
#define PAGE_FLAGS_H
-#include <linux/percpu.h>
-#include <linux/cache.h>
-#include <linux/types.h>
-
-#include <asm/pgtable.h>
+#include <linux/vmstat.h>
/*
* Various page->flags bits:
@@ -105,134 +101,6 @@
#endif
/*
- * Global page accounting. One instance per CPU. Only unsigned longs are
- * allowed.
- *
- * - Fields can be modified with xxx_page_state and xxx_page_state_zone at
- * any time safely (which protects the instance from modification by
- * interrupt.
- * - The __xxx_page_state variants can be used safely when interrupts are
- * disabled.
- * - The __xxx_page_state variants can be used if the field is only
- * modified from process context and protected from preemption, or only
- * modified from interrupt context. In this case, the field should be
- * commented here.
- */
-struct page_state {
- unsigned long nr_dirty; /* Dirty writeable pages */
- unsigned long nr_writeback; /* Pages under writeback */
- unsigned long nr_unstable; /* NFS unstable pages */
- unsigned long nr_page_table_pages;/* Pages used for pagetables */
- unsigned long nr_mapped; /* mapped into pagetables.
- * only modified from process context */
- unsigned long nr_slab; /* In slab */
-#define GET_PAGE_STATE_LAST nr_slab
-
- /*
- * The below are zeroed by get_page_state(). Use get_full_page_state()
- * to add up all these.
- */
- unsigned long pgpgin; /* Disk reads */
- unsigned long pgpgout; /* Disk writes */
- unsigned long pswpin; /* swap reads */
- unsigned long pswpout; /* swap writes */
-
- unsigned long pgalloc_high; /* page allocations */
- unsigned long pgalloc_normal;
- unsigned long pgalloc_dma32;
- unsigned long pgalloc_dma;
-
- unsigned long pgfree; /* page freeings */
- unsigned long pgactivate; /* pages moved inactive->active */
- unsigned long pgdeactivate; /* pages moved active->inactive */
-
- unsigned long pgfault; /* faults (major+minor) */
- unsigned long pgmajfault; /* faults (major only) */
-
- unsigned long pgrefill_high; /* inspected in refill_inactive_zone */
- unsigned long pgrefill_normal;
- unsigned long pgrefill_dma32;
- unsigned long pgrefill_dma;
-
- unsigned long pgsteal_high; /* total highmem pages reclaimed */
- unsigned long pgsteal_normal;
- unsigned long pgsteal_dma32;
- unsigned long pgsteal_dma;
-
- unsigned long pgscan_kswapd_high;/* total highmem pages scanned */
- unsigned long pgscan_kswapd_normal;
- unsigned long pgscan_kswapd_dma32;
- unsigned long pgscan_kswapd_dma;
-
- unsigned long pgscan_direct_high;/* total highmem pages scanned */
- unsigned long pgscan_direct_normal;
- unsigned long pgscan_direct_dma32;
- unsigned long pgscan_direct_dma;
-
- unsigned long pginodesteal; /* pages reclaimed via inode freeing */
- unsigned long slabs_scanned; /* slab objects scanned */
- unsigned long kswapd_steal; /* pages reclaimed by kswapd */
- unsigned long kswapd_inodesteal;/* reclaimed via kswapd inode freeing */
- unsigned long pageoutrun; /* kswapd's calls to page reclaim */
- unsigned long allocstall; /* direct reclaim calls */
-
- unsigned long pgrotated; /* pages rotated to tail of the LRU */
- unsigned long nr_bounce; /* pages for bounce buffers */
-};
-
-extern void get_page_state(struct page_state *ret);
-extern void get_page_state_node(struct page_state *ret, int node);
-extern void get_full_page_state(struct page_state *ret);
-extern unsigned long read_page_state_offset(unsigned long offset);
-extern void mod_page_state_offset(unsigned long offset, unsigned long delta);
-extern void __mod_page_state_offset(unsigned long offset, unsigned long delta);
-
-#define read_page_state(member) \
- read_page_state_offset(offsetof(struct page_state, member))
-
-#define mod_page_state(member, delta) \
- mod_page_state_offset(offsetof(struct page_state, member), (delta))
-
-#define __mod_page_state(member, delta) \
- __mod_page_state_offset(offsetof(struct page_state, member), (delta))
-
-#define inc_page_state(member) mod_page_state(member, 1UL)
-#define dec_page_state(member) mod_page_state(member, 0UL - 1)
-#define add_page_state(member,delta) mod_page_state(member, (delta))
-#define sub_page_state(member,delta) mod_page_state(member, 0UL - (delta))
-
-#define __inc_page_state(member) __mod_page_state(member, 1UL)
-#define __dec_page_state(member) __mod_page_state(member, 0UL - 1)
-#define __add_page_state(member,delta) __mod_page_state(member, (delta))
-#define __sub_page_state(member,delta) __mod_page_state(member, 0UL - (delta))
-
-#define page_state(member) (*__page_state(offsetof(struct page_state, member)))
-
-#define state_zone_offset(zone, member) \
-({ \
- unsigned offset; \
- if (is_highmem(zone)) \
- offset = offsetof(struct page_state, member##_high); \
- else if (is_normal(zone)) \
- offset = offsetof(struct page_state, member##_normal); \
- else if (is_dma32(zone)) \
- offset = offsetof(struct page_state, member##_dma32); \
- else \
- offset = offsetof(struct page_state, member##_dma); \
- offset; \
-})
-
-#define __mod_page_state_zone(zone, member, delta) \
- do { \
- __mod_page_state_offset(state_zone_offset(zone, member), (delta)); \
- } while (0)
-
-#define mod_page_state_zone(zone, member, delta) \
- do { \
- mod_page_state_offset(state_zone_offset(zone, member), (delta)); \
- } while (0)
-
-/*
* Manipulation of page state flags
*/
#define PageLocked(page) \
Index: linux-2.6.17-rc6-cl/mm/Makefile
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/Makefile 2006-06-12 12:42:52.028160490 -0700
+++ linux-2.6.17-rc6-cl/mm/Makefile 2006-06-12 12:43:12.082583522 -0700
@@ -10,7 +10,7 @@ mmu-$(CONFIG_MMU) := fremap.o highmem.o
obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
page_alloc.o page-writeback.o pdflush.o \
readahead.o swap.o truncate.o vmscan.o \
- prio_tree.o util.o mmzone.o $(mmu-y)
+ prio_tree.o util.o mmzone.o vmstat.o $(mmu-y)
obj-$(CONFIG_SWAP) += page_io.o swap_state.o swapfile.o thrash.o
obj-$(CONFIG_SWAP_PREFETCH) += swap_prefetch.o
Index: linux-2.6.17-rc6-cl/include/linux/mm.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mm.h 2006-06-12 12:42:50.747966275 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mm.h 2006-06-12 12:43:12.083560024 -0700
@@ -4,6 +4,7 @@
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/capability.h>
+#include <linux/vmstat.h>
#ifdef __KERNEL__
@@ -37,7 +38,6 @@ extern int sysctl_legacy_va_layout;
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/processor.h>
-#include <asm/atomic.h>
#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
Index: linux-2.6.17-rc6-cl/include/linux/vmstat.h
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ linux-2.6.17-rc6-cl/include/linux/vmstat.h 2006-06-12 12:43:12.084536526 -0700
@@ -0,0 +1,135 @@
+#ifndef _LINUX_VMSTAT_H
+#define _LINUX_VMSTAT_H
+
+#include <linux/types.h>
+
+/*
+ * Global page accounting. One instance per CPU. Only unsigned longs are
+ * allowed.
+ *
+ * - Fields can be modified with xxx_page_state and xxx_page_state_zone at
+ * any time safely (which protects the instance from modification by
+ * interrupt.
+ * - The __xxx_page_state variants can be used safely when interrupts are
+ * disabled.
+ * - The __xxx_page_state variants can be used if the field is only
+ * modified from process context and protected from preemption, or only
+ * modified from interrupt context. In this case, the field should be
+ * commented here.
+ */
+struct page_state {
+ unsigned long nr_dirty; /* Dirty writeable pages */
+ unsigned long nr_writeback; /* Pages under writeback */
+ unsigned long nr_unstable; /* NFS unstable pages */
+ unsigned long nr_page_table_pages;/* Pages used for pagetables */
+ unsigned long nr_mapped; /* mapped into pagetables.
+ * only modified from process context */
+ unsigned long nr_slab; /* In slab */
+#define GET_PAGE_STATE_LAST nr_slab
+
+ /*
+ * The below are zeroed by get_page_state(). Use get_full_page_state()
+ * to add up all these.
+ */
+ unsigned long pgpgin; /* Disk reads */
+ unsigned long pgpgout; /* Disk writes */
+ unsigned long pswpin; /* swap reads */
+ unsigned long pswpout; /* swap writes */
+
+ unsigned long pgalloc_high; /* page allocations */
+ unsigned long pgalloc_normal;
+ unsigned long pgalloc_dma32;
+ unsigned long pgalloc_dma;
+
+ unsigned long pgfree; /* page freeings */
+ unsigned long pgactivate; /* pages moved inactive->active */
+ unsigned long pgdeactivate; /* pages moved active->inactive */
+
+ unsigned long pgfault; /* faults (major+minor) */
+ unsigned long pgmajfault; /* faults (major only) */
+
+ unsigned long pgrefill_high; /* inspected in refill_inactive_zone */
+ unsigned long pgrefill_normal;
+ unsigned long pgrefill_dma32;
+ unsigned long pgrefill_dma;
+
+ unsigned long pgsteal_high; /* total highmem pages reclaimed */
+ unsigned long pgsteal_normal;
+ unsigned long pgsteal_dma32;
+ unsigned long pgsteal_dma;
+
+ unsigned long pgscan_kswapd_high;/* total highmem pages scanned */
+ unsigned long pgscan_kswapd_normal;
+ unsigned long pgscan_kswapd_dma32;
+ unsigned long pgscan_kswapd_dma;
+
+ unsigned long pgscan_direct_high;/* total highmem pages scanned */
+ unsigned long pgscan_direct_normal;
+ unsigned long pgscan_direct_dma32;
+ unsigned long pgscan_direct_dma;
+
+ unsigned long pginodesteal; /* pages reclaimed via inode freeing */
+ unsigned long slabs_scanned; /* slab objects scanned */
+ unsigned long kswapd_steal; /* pages reclaimed by kswapd */
+ unsigned long kswapd_inodesteal;/* reclaimed via kswapd inode freeing */
+ unsigned long pageoutrun; /* kswapd's calls to page reclaim */
+ unsigned long allocstall; /* direct reclaim calls */
+
+ unsigned long pgrotated; /* pages rotated to tail of the LRU */
+ unsigned long nr_bounce; /* pages for bounce buffers */
+};
+
+extern void get_page_state(struct page_state *ret);
+extern void get_page_state_node(struct page_state *ret, int node);
+extern void get_full_page_state(struct page_state *ret);
+extern unsigned long read_page_state_offset(unsigned long offset);
+extern void mod_page_state_offset(unsigned long offset, unsigned long delta);
+extern void __mod_page_state_offset(unsigned long offset, unsigned long delta);
+
+#define read_page_state(member) \
+ read_page_state_offset(offsetof(struct page_state, member))
+
+#define mod_page_state(member, delta) \
+ mod_page_state_offset(offsetof(struct page_state, member), (delta))
+
+#define __mod_page_state(member, delta) \
+ __mod_page_state_offset(offsetof(struct page_state, member), (delta))
+
+#define inc_page_state(member) mod_page_state(member, 1UL)
+#define dec_page_state(member) mod_page_state(member, 0UL - 1)
+#define add_page_state(member,delta) mod_page_state(member, (delta))
+#define sub_page_state(member,delta) mod_page_state(member, 0UL - (delta))
+
+#define __inc_page_state(member) __mod_page_state(member, 1UL)
+#define __dec_page_state(member) __mod_page_state(member, 0UL - 1)
+#define __add_page_state(member,delta) __mod_page_state(member, (delta))
+#define __sub_page_state(member,delta) __mod_page_state(member, 0UL - (delta))
+
+#define page_state(member) (*__page_state(offsetof(struct page_state, member)))
+
+#define state_zone_offset(zone, member) \
+({ \
+ unsigned offset; \
+ if (is_highmem(zone)) \
+ offset = offsetof(struct page_state, member##_high); \
+ else if (is_normal(zone)) \
+ offset = offsetof(struct page_state, member##_normal); \
+ else if (is_dma32(zone)) \
+ offset = offsetof(struct page_state, member##_dma32); \
+ else \
+ offset = offsetof(struct page_state, member##_dma); \
+ offset; \
+})
+
+#define __mod_page_state_zone(zone, member, delta) \
+ do { \
+ __mod_page_state_offset(state_zone_offset(zone, member), (delta)); \
+ } while (0)
+
+#define mod_page_state_zone(zone, member, delta) \
+ do { \
+ mod_page_state_offset(state_zone_offset(zone, member), (delta)); \
+ } while (0)
+
+#endif /* _LINUX_VMSTAT_H */
+
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 12:51:44.438570245 -0700
@@ -0,0 +1,417 @@
+/*
+ * linux/mm/vmstat.c
+ *
+ * Manages VM statistics
+ * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds
+ */
+
+#include <linux/config.h>
+#include <linux/mm.h>
+
+/*
+ * Accumulate the page_state information across all CPUs.
+ * The result is unavoidably approximate - it can change
+ * during and after execution of this function.
+ */
+static DEFINE_PER_CPU(struct page_state, page_states) = {0};
+
+atomic_t nr_pagecache = ATOMIC_INIT(0);
+EXPORT_SYMBOL(nr_pagecache);
+#ifdef CONFIG_SMP
+DEFINE_PER_CPU(long, nr_pagecache_local) = 0;
+#endif
+
+static void __get_page_state(struct page_state *ret, int nr, cpumask_t *cpumask)
+{
+ unsigned cpu;
+
+ memset(ret, 0, nr * sizeof(unsigned long));
+ cpus_and(*cpumask, *cpumask, cpu_online_map);
+
+ for_each_cpu_mask(cpu, *cpumask) {
+ unsigned long *in;
+ unsigned long *out;
+ unsigned off;
+ unsigned next_cpu;
+
+ in = (unsigned long *)&per_cpu(page_states, cpu);
+
+ next_cpu = next_cpu(cpu, *cpumask);
+ if (likely(next_cpu < NR_CPUS))
+ prefetch(&per_cpu(page_states, next_cpu));
+
+ out = (unsigned long *)ret;
+ for (off = 0; off < nr; off++)
+ *out++ += *in++;
+ }
+}
+
+void get_page_state_node(struct page_state *ret, int node)
+{
+ int nr;
+ cpumask_t mask = node_to_cpumask(node);
+
+ nr = offsetof(struct page_state, GET_PAGE_STATE_LAST);
+ nr /= sizeof(unsigned long);
+
+ __get_page_state(ret, nr+1, &mask);
+}
+
+void get_page_state(struct page_state *ret)
+{
+ int nr;
+ cpumask_t mask = CPU_MASK_ALL;
+
+ nr = offsetof(struct page_state, GET_PAGE_STATE_LAST);
+ nr /= sizeof(unsigned long);
+
+ __get_page_state(ret, nr + 1, &mask);
+}
+
+void get_full_page_state(struct page_state *ret)
+{
+ cpumask_t mask = CPU_MASK_ALL;
+
+ __get_page_state(ret, sizeof(*ret) / sizeof(unsigned long), &mask);
+}
+
+unsigned long read_page_state_offset(unsigned long offset)
+{
+ unsigned long ret = 0;
+ int cpu;
+
+ for_each_online_cpu(cpu) {
+ unsigned long in;
+
+ in = (unsigned long)&per_cpu(page_states, cpu) + offset;
+ ret += *((unsigned long *)in);
+ }
+ return ret;
+}
+
+void __mod_page_state_offset(unsigned long offset, unsigned long delta)
+{
+ void *ptr;
+
+ ptr = &__get_cpu_var(page_states);
+ *(unsigned long *)(ptr + offset) += delta;
+}
+EXPORT_SYMBOL(__mod_page_state_offset);
+
+void mod_page_state_offset(unsigned long offset, unsigned long delta)
+{
+ unsigned long flags;
+ void *ptr;
+
+ local_irq_save(flags);
+ ptr = &__get_cpu_var(page_states);
+ *(unsigned long *)(ptr + offset) += delta;
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(mod_page_state_offset);
+
+void __get_zone_counts(unsigned long *active, unsigned long *inactive,
+ unsigned long *free, struct pglist_data *pgdat)
+{
+ struct zone *zones = pgdat->node_zones;
+ int i;
+
+ *active = 0;
+ *inactive = 0;
+ *free = 0;
+ for (i = 0; i < MAX_NR_ZONES; i++) {
+ *active += zones[i].nr_active;
+ *inactive += zones[i].nr_inactive;
+ *free += zones[i].free_pages;
+ }
+}
+
+void get_zone_counts(unsigned long *active,
+ unsigned long *inactive, unsigned long *free)
+{
+ struct pglist_data *pgdat;
+
+ *active = 0;
+ *inactive = 0;
+ *free = 0;
+ for_each_online_pgdat(pgdat) {
+ unsigned long l, m, n;
+ __get_zone_counts(&l, &m, &n, pgdat);
+ *active += l;
+ *inactive += m;
+ *free += n;
+ }
+}
+
+#ifdef CONFIG_PROC_FS
+
+#include <linux/seq_file.h>
+
+static void *frag_start(struct seq_file *m, loff_t *pos)
+{
+ pg_data_t *pgdat;
+ loff_t node = *pos;
+ for (pgdat = first_online_pgdat();
+ pgdat && node;
+ pgdat = next_online_pgdat(pgdat))
+ --node;
+
+ return pgdat;
+}
+
+static void *frag_next(struct seq_file *m, void *arg, loff_t *pos)
+{
+ pg_data_t *pgdat = (pg_data_t *)arg;
+
+ (*pos)++;
+ return next_online_pgdat(pgdat);
+}
+
+static void frag_stop(struct seq_file *m, void *arg)
+{
+}
+
+/*
+ * This walks the free areas for each zone.
+ */
+static int frag_show(struct seq_file *m, void *arg)
+{
+ pg_data_t *pgdat = (pg_data_t *)arg;
+ struct zone *zone;
+ struct zone *node_zones = pgdat->node_zones;
+ unsigned long flags;
+ int order;
+
+ for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
+ if (!populated_zone(zone))
+ continue;
+
+ spin_lock_irqsave(&zone->lock, flags);
+ seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name);
+ for (order = 0; order < MAX_ORDER; ++order)
+ seq_printf(m, "%6lu ", zone->free_area[order].nr_free);
+ spin_unlock_irqrestore(&zone->lock, flags);
+ seq_putc(m, '\n');
+ }
+ return 0;
+}
+
+struct seq_operations fragmentation_op = {
+ .start = frag_start,
+ .next = frag_next,
+ .stop = frag_stop,
+ .show = frag_show,
+};
+
+static char *vmstat_text[] = {
+ "nr_dirty",
+ "nr_writeback",
+ "nr_unstable",
+ "nr_page_table_pages",
+ "nr_mapped",
+ "nr_slab",
+
+ "pgpgin",
+ "pgpgout",
+ "pswpin",
+ "pswpout",
+
+ "pgalloc_high",
+ "pgalloc_normal",
+ "pgalloc_dma32",
+ "pgalloc_dma",
+
+ "pgfree",
+ "pgactivate",
+ "pgdeactivate",
+
+ "pgfault",
+ "pgmajfault",
+
+ "pgrefill_high",
+ "pgrefill_normal",
+ "pgrefill_dma32",
+ "pgrefill_dma",
+
+ "pgsteal_high",
+ "pgsteal_normal",
+ "pgsteal_dma32",
+ "pgsteal_dma",
+
+ "pgscan_kswapd_high",
+ "pgscan_kswapd_normal",
+ "pgscan_kswapd_dma32",
+ "pgscan_kswapd_dma",
+
+ "pgscan_direct_high",
+ "pgscan_direct_normal",
+ "pgscan_direct_dma32",
+ "pgscan_direct_dma",
+
+ "pginodesteal",
+ "slabs_scanned",
+ "kswapd_steal",
+ "kswapd_inodesteal",
+ "pageoutrun",
+ "allocstall",
+
+ "pgrotated",
+ "nr_bounce",
+};
+
+/*
+ * Output information about zones in @pgdat.
+ */
+static int zoneinfo_show(struct seq_file *m, void *arg)
+{
+ pg_data_t *pgdat = arg;
+ struct zone *zone;
+ struct zone *node_zones = pgdat->node_zones;
+ unsigned long flags;
+
+ for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; zone++) {
+ int i;
+
+ if (!populated_zone(zone))
+ continue;
+
+ spin_lock_irqsave(&zone->lock, flags);
+ seq_printf(m, "Node %d, zone %8s", pgdat->node_id, zone->name);
+ seq_printf(m,
+ "\n pages free %lu"
+ "\n min %lu"
+ "\n low %lu"
+ "\n high %lu"
+ "\n active %lu"
+ "\n inactive %lu"
+ "\n scanned %lu (a: %lu i: %lu)"
+ "\n spanned %lu"
+ "\n present %lu",
+ zone->free_pages,
+ zone->pages_min,
+ zone->pages_low,
+ zone->pages_high,
+ zone->nr_active,
+ zone->nr_inactive,
+ zone->pages_scanned,
+ zone->nr_scan_active, zone->nr_scan_inactive,
+ zone->spanned_pages,
+ zone->present_pages);
+ seq_printf(m,
+ "\n protection: (%lu",
+ zone->lowmem_reserve[0]);
+ for (i = 1; i < ARRAY_SIZE(zone->lowmem_reserve); i++)
+ seq_printf(m, ", %lu", zone->lowmem_reserve[i]);
+ seq_printf(m,
+ ")"
+ "\n pagesets");
+ for_each_online_cpu(i) {
+ struct per_cpu_pageset *pageset;
+ int j;
+
+ pageset = zone_pcp(zone, i);
+ for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) {
+ if (pageset->pcp[j].count)
+ break;
+ }
+ if (j == ARRAY_SIZE(pageset->pcp))
+ continue;
+ for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) {
+ seq_printf(m,
+ "\n cpu: %i pcp: %i"
+ "\n count: %i"
+ "\n high: %i"
+ "\n batch: %i",
+ i, j,
+ pageset->pcp[j].count,
+ pageset->pcp[j].high,
+ pageset->pcp[j].batch);
+ }
+#ifdef CONFIG_NUMA
+ seq_printf(m,
+ "\n numa_hit: %lu"
+ "\n numa_miss: %lu"
+ "\n numa_foreign: %lu"
+ "\n interleave_hit: %lu"
+ "\n local_node: %lu"
+ "\n other_node: %lu",
+ pageset->numa_hit,
+ pageset->numa_miss,
+ pageset->numa_foreign,
+ pageset->interleave_hit,
+ pageset->local_node,
+ pageset->other_node);
+#endif
+ }
+ seq_printf(m,
+ "\n all_unreclaimable: %u"
+ "\n prev_priority: %i"
+ "\n temp_priority: %i"
+ "\n start_pfn: %lu",
+ zone->all_unreclaimable,
+ zone->prev_priority,
+ zone->temp_priority,
+ zone->zone_start_pfn);
+ spin_unlock_irqrestore(&zone->lock, flags);
+ seq_putc(m, '\n');
+ }
+ return 0;
+}
+
+struct seq_operations zoneinfo_op = {
+ .start = frag_start, /* iterate over all zones. The same as in
+ * fragmentation. */
+ .next = frag_next,
+ .stop = frag_stop,
+ .show = zoneinfo_show,
+};
+
+static void *vmstat_start(struct seq_file *m, loff_t *pos)
+{
+ struct page_state *ps;
+
+ if (*pos >= ARRAY_SIZE(vmstat_text))
+ return NULL;
+
+ ps = kmalloc(sizeof(*ps), GFP_KERNEL);
+ m->private = ps;
+ if (!ps)
+ return ERR_PTR(-ENOMEM);
+ get_full_page_state(ps);
+ ps->pgpgin /= 2; /* sectors -> kbytes */
+ ps->pgpgout /= 2;
+ return (unsigned long *)ps + *pos;
+}
+
+static void *vmstat_next(struct seq_file *m, void *arg, loff_t *pos)
+{
+ (*pos)++;
+ if (*pos >= ARRAY_SIZE(vmstat_text))
+ return NULL;
+ return (unsigned long *)m->private + *pos;
+}
+
+static int vmstat_show(struct seq_file *m, void *arg)
+{
+ unsigned long *l = arg;
+ unsigned long off = l - (unsigned long *)m->private;
+
+ seq_printf(m, "%s %lu\n", vmstat_text[off], *l);
+ return 0;
+}
+
+static void vmstat_stop(struct seq_file *m, void *arg)
+{
+ kfree(m->private);
+ m->private = NULL;
+}
+
+struct seq_operations vmstat_op = {
+ .start = vmstat_start,
+ .next = vmstat_next,
+ .stop = vmstat_stop,
+ .show = vmstat_show,
+};
+
+#endif /* CONFIG_PROC_FS */
+
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 12:42:52.042808021 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 12:43:12.087466032 -0700
@@ -1306,141 +1306,6 @@ static void show_node(struct zone *zone)
#endif
/*
- * Accumulate the page_state information across all CPUs.
- * The result is unavoidably approximate - it can change
- * during and after execution of this function.
- */
-static DEFINE_PER_CPU(struct page_state, page_states) = {0};
-
-atomic_t nr_pagecache = ATOMIC_INIT(0);
-EXPORT_SYMBOL(nr_pagecache);
-#ifdef CONFIG_SMP
-DEFINE_PER_CPU(long, nr_pagecache_local) = 0;
-#endif
-
-static void __get_page_state(struct page_state *ret, int nr, cpumask_t *cpumask)
-{
- unsigned cpu;
-
- memset(ret, 0, nr * sizeof(unsigned long));
- cpus_and(*cpumask, *cpumask, cpu_online_map);
-
- for_each_cpu_mask(cpu, *cpumask) {
- unsigned long *in;
- unsigned long *out;
- unsigned off;
- unsigned next_cpu;
-
- in = (unsigned long *)&per_cpu(page_states, cpu);
-
- next_cpu = next_cpu(cpu, *cpumask);
- if (likely(next_cpu < NR_CPUS))
- prefetch(&per_cpu(page_states, next_cpu));
-
- out = (unsigned long *)ret;
- for (off = 0; off < nr; off++)
- *out++ += *in++;
- }
-}
-
-void get_page_state_node(struct page_state *ret, int node)
-{
- int nr;
- cpumask_t mask = node_to_cpumask(node);
-
- nr = offsetof(struct page_state, GET_PAGE_STATE_LAST);
- nr /= sizeof(unsigned long);
-
- __get_page_state(ret, nr+1, &mask);
-}
-
-void get_page_state(struct page_state *ret)
-{
- int nr;
- cpumask_t mask = CPU_MASK_ALL;
-
- nr = offsetof(struct page_state, GET_PAGE_STATE_LAST);
- nr /= sizeof(unsigned long);
-
- __get_page_state(ret, nr + 1, &mask);
-}
-
-void get_full_page_state(struct page_state *ret)
-{
- cpumask_t mask = CPU_MASK_ALL;
-
- __get_page_state(ret, sizeof(*ret) / sizeof(unsigned long), &mask);
-}
-
-unsigned long read_page_state_offset(unsigned long offset)
-{
- unsigned long ret = 0;
- int cpu;
-
- for_each_online_cpu(cpu) {
- unsigned long in;
-
- in = (unsigned long)&per_cpu(page_states, cpu) + offset;
- ret += *((unsigned long *)in);
- }
- return ret;
-}
-
-void __mod_page_state_offset(unsigned long offset, unsigned long delta)
-{
- void *ptr;
-
- ptr = &__get_cpu_var(page_states);
- *(unsigned long *)(ptr + offset) += delta;
-}
-EXPORT_SYMBOL(__mod_page_state_offset);
-
-void mod_page_state_offset(unsigned long offset, unsigned long delta)
-{
- unsigned long flags;
- void *ptr;
-
- local_irq_save(flags);
- ptr = &__get_cpu_var(page_states);
- *(unsigned long *)(ptr + offset) += delta;
- local_irq_restore(flags);
-}
-EXPORT_SYMBOL(mod_page_state_offset);
-
-void __get_zone_counts(unsigned long *active, unsigned long *inactive,
- unsigned long *free, struct pglist_data *pgdat)
-{
- struct zone *zones = pgdat->node_zones;
- int i;
-
- *active = 0;
- *inactive = 0;
- *free = 0;
- for (i = 0; i < MAX_NR_ZONES; i++) {
- *active += zones[i].nr_active;
- *inactive += zones[i].nr_inactive;
- *free += zones[i].free_pages;
- }
-}
-
-void get_zone_counts(unsigned long *active,
- unsigned long *inactive, unsigned long *free)
-{
- struct pglist_data *pgdat;
-
- *active = 0;
- *inactive = 0;
- *free = 0;
- for_each_online_pgdat(pgdat) {
- unsigned long l, m, n;
- __get_zone_counts(&l, &m, &n, pgdat);
- *active += l;
- *inactive += m;
- *free += n;
- }
-}
-
-/*
* The node's effective length of inactive_list(s).
*/
unsigned long nr_free_inactive_pages_node(int nid)
@@ -2360,278 +2225,6 @@ void __init free_area_init(unsigned long
__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL);
}
-#ifdef CONFIG_PROC_FS
-
-#include <linux/seq_file.h>
-
-static void *frag_start(struct seq_file *m, loff_t *pos)
-{
- pg_data_t *pgdat;
- loff_t node = *pos;
- for (pgdat = first_online_pgdat();
- pgdat && node;
- pgdat = next_online_pgdat(pgdat))
- --node;
-
- return pgdat;
-}
-
-static void *frag_next(struct seq_file *m, void *arg, loff_t *pos)
-{
- pg_data_t *pgdat = (pg_data_t *)arg;
-
- (*pos)++;
- return next_online_pgdat(pgdat);
-}
-
-static void frag_stop(struct seq_file *m, void *arg)
-{
-}
-
-/*
- * This walks the free areas for each zone.
- */
-static int frag_show(struct seq_file *m, void *arg)
-{
- pg_data_t *pgdat = (pg_data_t *)arg;
- struct zone *zone;
- struct zone *node_zones = pgdat->node_zones;
- unsigned long flags;
- int order;
-
- for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
- if (!populated_zone(zone))
- continue;
-
- spin_lock_irqsave(&zone->lock, flags);
- seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name);
- for (order = 0; order < MAX_ORDER; ++order)
- seq_printf(m, "%6lu ", zone->free_area[order].nr_free);
- spin_unlock_irqrestore(&zone->lock, flags);
- seq_putc(m, '\n');
- }
- return 0;
-}
-
-struct seq_operations fragmentation_op = {
- .start = frag_start,
- .next = frag_next,
- .stop = frag_stop,
- .show = frag_show,
-};
-
-/*
- * Output information about zones in @pgdat.
- */
-static int zoneinfo_show(struct seq_file *m, void *arg)
-{
- pg_data_t *pgdat = arg;
- struct zone *zone;
- struct zone *node_zones = pgdat->node_zones;
- unsigned long flags;
-
- for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; zone++) {
- int i;
-
- if (!populated_zone(zone))
- continue;
-
- spin_lock_irqsave(&zone->lock, flags);
- seq_printf(m, "Node %d, zone %8s", pgdat->node_id, zone->name);
- seq_printf(m,
- "\n pages free %lu"
- "\n min %lu"
- "\n low %lu"
- "\n high %lu"
- "\n active %lu"
- "\n inactive %lu"
- "\n scanned %lu (a: %lu i: %lu)"
- "\n spanned %lu"
- "\n present %lu",
- zone->free_pages,
- zone->pages_min,
- zone->pages_low,
- zone->pages_high,
- zone->nr_active,
- zone->nr_inactive,
- zone->pages_scanned,
- zone->nr_scan_active, zone->nr_scan_inactive,
- zone->spanned_pages,
- zone->present_pages);
- seq_printf(m,
- "\n protection: (%lu",
- zone->lowmem_reserve[0]);
- for (i = 1; i < ARRAY_SIZE(zone->lowmem_reserve); i++)
- seq_printf(m, ", %lu", zone->lowmem_reserve[i]);
- seq_printf(m,
- ")"
- "\n pagesets");
- for_each_online_cpu(i) {
- struct per_cpu_pageset *pageset;
- int j;
-
- pageset = zone_pcp(zone, i);
- for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) {
- if (pageset->pcp[j].count)
- break;
- }
- if (j == ARRAY_SIZE(pageset->pcp))
- continue;
- for (j = 0; j < ARRAY_SIZE(pageset->pcp); j++) {
- seq_printf(m,
- "\n cpu: %i pcp: %i"
- "\n count: %i"
- "\n high: %i"
- "\n batch: %i",
- i, j,
- pageset->pcp[j].count,
- pageset->pcp[j].high,
- pageset->pcp[j].batch);
- }
-#ifdef CONFIG_NUMA
- seq_printf(m,
- "\n numa_hit: %lu"
- "\n numa_miss: %lu"
- "\n numa_foreign: %lu"
- "\n interleave_hit: %lu"
- "\n local_node: %lu"
- "\n other_node: %lu",
- pageset->numa_hit,
- pageset->numa_miss,
- pageset->numa_foreign,
- pageset->interleave_hit,
- pageset->local_node,
- pageset->other_node);
-#endif
- }
- seq_printf(m,
- "\n all_unreclaimable: %u"
- "\n prev_priority: %i"
- "\n temp_priority: %i"
- "\n start_pfn: %lu",
- zone->all_unreclaimable,
- zone->prev_priority,
- zone->temp_priority,
- zone->zone_start_pfn);
- spin_unlock_irqrestore(&zone->lock, flags);
- seq_putc(m, '\n');
- }
- return 0;
-}
-
-struct seq_operations zoneinfo_op = {
- .start = frag_start, /* iterate over all zones. The same as in
- * fragmentation. */
- .next = frag_next,
- .stop = frag_stop,
- .show = zoneinfo_show,
-};
-
-static char *vmstat_text[] = {
- "nr_dirty",
- "nr_writeback",
- "nr_unstable",
- "nr_page_table_pages",
- "nr_mapped",
- "nr_slab",
-
- "pgpgin",
- "pgpgout",
- "pswpin",
- "pswpout",
-
- "pgalloc_high",
- "pgalloc_normal",
- "pgalloc_dma32",
- "pgalloc_dma",
-
- "pgfree",
- "pgactivate",
- "pgdeactivate",
-
- "pgfault",
- "pgmajfault",
-
- "pgrefill_high",
- "pgrefill_normal",
- "pgrefill_dma32",
- "pgrefill_dma",
-
- "pgsteal_high",
- "pgsteal_normal",
- "pgsteal_dma32",
- "pgsteal_dma",
-
- "pgscan_kswapd_high",
- "pgscan_kswapd_normal",
- "pgscan_kswapd_dma32",
- "pgscan_kswapd_dma",
-
- "pgscan_direct_high",
- "pgscan_direct_normal",
- "pgscan_direct_dma32",
- "pgscan_direct_dma",
-
- "pginodesteal",
- "slabs_scanned",
- "kswapd_steal",
- "kswapd_inodesteal",
- "pageoutrun",
- "allocstall",
-
- "pgrotated",
- "nr_bounce",
-};
-
-static void *vmstat_start(struct seq_file *m, loff_t *pos)
-{
- struct page_state *ps;
-
- if (*pos >= ARRAY_SIZE(vmstat_text))
- return NULL;
-
- ps = kmalloc(sizeof(*ps), GFP_KERNEL);
- m->private = ps;
- if (!ps)
- return ERR_PTR(-ENOMEM);
- get_full_page_state(ps);
- ps->pgpgin /= 2; /* sectors -> kbytes */
- ps->pgpgout /= 2;
- return (unsigned long *)ps + *pos;
-}
-
-static void *vmstat_next(struct seq_file *m, void *arg, loff_t *pos)
-{
- (*pos)++;
- if (*pos >= ARRAY_SIZE(vmstat_text))
- return NULL;
- return (unsigned long *)m->private + *pos;
-}
-
-static int vmstat_show(struct seq_file *m, void *arg)
-{
- unsigned long *l = arg;
- unsigned long off = l - (unsigned long *)m->private;
-
- seq_printf(m, "%s %lu\n", vmstat_text[off], *l);
- return 0;
-}
-
-static void vmstat_stop(struct seq_file *m, void *arg)
-{
- kfree(m->private);
- m->private = NULL;
-}
-
-struct seq_operations vmstat_op = {
- .start = vmstat_start,
- .next = vmstat_next,
- .stop = vmstat_stop,
- .show = vmstat_show,
-};
-
-#endif /* CONFIG_PROC_FS */
-
#ifdef CONFIG_HOTPLUG_CPU
static int page_alloc_cpu_notify(struct notifier_block *self,
unsigned long action, void *hcpu)
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 02/21] Basic ZVC (zoned vm counter) implementation
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
2006-06-12 21:12 ` [PATCH 01/21] Create vmstat.c/.h from page_alloc.c/.h Christoph Lameter
@ 2006-06-12 21:12 ` Christoph Lameter
2006-06-13 5:37 ` zoned vm counters: per zone counter functionality Nick Piggin
2006-06-12 21:13 ` [PATCH 03/21] Convert nr_mapped to per zone counter Christoph Lameter
` (18 subsequent siblings)
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:12 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: per zone counter functionality
From: Christoph Lameter <clameter@sgi.com>
Per zone counter infrastructure
The counters that we currently have for the VM are split per processor.
The processor however has not much to do with the zone these pages belong
to. We cannot tell f.e. how many ZONE_DMA pages are dirty.
So we are blind to potentially inbalances in the usage of memory in various
zones. F.e. in a NUMA system we cannot tell how many pages are dirty on
a particular node. If we knew then we could put measures into the VM to balance
the use of memory between different zones and different nodes in a NUMA
system. For example it would be possible to limit the dirty pages per node
so that fast local memory is kept available even if a process is dirtying
huge amounts of pages.
Another example is zone reclaim. We do not know how many unmapped pages exist
per zone. So we just have to try to reclaim. If it is not working then we
pause and try again later. It would be better if we knew when it makes sense
to reclaim unmapped pages from a zone. This patchset allows the determination
of the number of unmapped pages per zone. We can remove the zone reclaim
interval with the counters introduced here.
Futhermore the ability to have various usage statistics available will allow
the development of new NUMA balancing algorithms that may be able to improve
the decision making in the scheduler of when to move a process to another node
and hopefully will also enable automatic page migration through a user space
program that can analyse the memory load distribution and then rebalance
memory use in order to increase performance.
The counter framework here implements differential counters for each processor
in struct zone. The differential counters are consolidated when a threshold
is exceeded (like done in the current implementation for nr_pageache), when
slab reaping occurs or when a consolidation function is called.
Consolidation uses atomic operations and accumulates counters per zone in the
zone structure and also globally in the vm_stat array. VM functions can
access the counts by simply indexing a global or zone specific array.
The arrangement of counters in an array also simplifies processing when output
has to be generated for /proc/*.
Counters can be updated by calling *_zone_page_state or __*_zone_page_state
analogous to *_page_state. The second group of functions can be called if it
is known that interrupts are disabled.
Special optimized increment and decrement functions are provided. These can
avoid certain checks and use increment or decrement instructions that an
architecture may provide.
We also add a new CONFIG_DMA_IS_NORMAL that signifies that an architecture
can do DMA to all memory and therefore ZONE_NORMAL will not be populated.
This is only currently set for ia64 and only affects node_page_state(). In
the best case node_page_state can be reduced to retrieving a single counter
for the one zone on the node.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 12:42:50.753825287 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 12:54:33.238664762 -0700
@@ -46,6 +46,9 @@ struct zone_padding {
#define ZONE_PADDING(name)
#endif
+enum zone_stat_item {
+ NR_STAT_ITEMS };
+
struct per_cpu_pages {
int count; /* number of pages in the list */
int high; /* high watermark, emptying needed */
@@ -55,6 +58,10 @@ struct per_cpu_pages {
struct per_cpu_pageset {
struct per_cpu_pages pcp[2]; /* 0: hot. 1: cold */
+#ifdef CONFIG_SMP
+ s8 vm_stat_diff[NR_STAT_ITEMS];
+#endif
+
#ifdef CONFIG_NUMA
unsigned long numa_hit; /* allocated in intended node */
unsigned long numa_miss; /* allocated in non intended node */
@@ -170,6 +177,8 @@ struct zone {
/* A count of how many reclaimers are scanning this zone */
atomic_t reclaim_in_progress;
+ /* Zone statistics */
+ atomic_long_t vm_stat[NR_STAT_ITEMS];
/*
* timestamp (in jiffies) of the last zone reclaim that did not
* result in freeing of pages. This is used to avoid repeated scans
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 12:43:12.087466032 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 12:54:33.240617767 -0700
@@ -2143,6 +2143,7 @@ static void __meminit free_area_init_cor
zone->nr_scan_inactive = 0;
zone->nr_active = 0;
zone->nr_inactive = 0;
+ zap_zone_vm_stats(zone);
atomic_set(&zone->reclaim_in_progress, 0);
if (!size)
continue;
@@ -2254,6 +2255,7 @@ static int page_alloc_cpu_notify(struct
}
local_irq_enable();
+ refresh_cpu_vm_stats(cpu);
}
return NOTIFY_OK;
}
Index: linux-2.6.17-rc6-cl/mm/slab.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/slab.c 2006-06-12 12:42:52.056479050 -0700
+++ linux-2.6.17-rc6-cl/mm/slab.c 2006-06-12 12:54:33.242570771 -0700
@@ -3826,6 +3826,7 @@ next:
check_irq_on();
mutex_unlock(&cache_chain_mutex);
next_reap_node();
+ refresh_cpu_vm_stats(smp_processor_id());
/* Set up the next iteration */
schedule_delayed_work(&__get_cpu_var(reap_work), REAPTIMEOUT_CPUC);
}
Index: linux-2.6.17-rc6-cl/include/linux/vmstat.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/vmstat.h 2006-06-12 12:43:12.084536526 -0700
+++ linux-2.6.17-rc6-cl/include/linux/vmstat.h 2006-06-12 12:54:33.243547273 -0700
@@ -2,6 +2,9 @@
#define _LINUX_VMSTAT_H
#include <linux/types.h>
+#include <linux/config.h>
+#include <linux/mmzone.h>
+#include <asm/atomic.h>
/*
* Global page accounting. One instance per CPU. Only unsigned longs are
@@ -131,5 +134,84 @@ extern void __mod_page_state_offset(unsi
mod_page_state_offset(state_zone_offset(zone, member), (delta)); \
} while (0)
+/*
+ * Zone based page accounting with per cpu differentials.
+ */
+extern atomic_long_t vm_stat[NR_STAT_ITEMS];
+
+static inline unsigned long global_page_state(enum zone_stat_item item)
+{
+ long x = atomic_long_read(&vm_stat[item]);
+#ifdef CONFIG_SMP
+ if (x < 0)
+ x = 0;
+#endif
+ return x;
+}
+
+static inline unsigned long zone_page_state(struct zone *zone,
+ enum zone_stat_item item)
+{
+ long x = atomic_long_read(&zone->vm_stat[item]);
+#ifdef CONFIG_SMP
+ if (x < 0)
+ x = 0;
+#endif
+ return x;
+}
+
+#ifdef CONFIG_NUMA
+/*
+ * Determine the per node value of a stat item. This function
+ * is called frequently in a NUMA machine, so try to be as
+ * frugal as possible.
+ */
+static inline unsigned long node_page_state(int node,
+ enum zone_stat_item item)
+{
+ struct zone *zones = NODE_DATA(node)->node_zones;
+
+ return
+#ifndef CONFIG_DMA_IS_NORMAL
+#if !defined(CONFIG_DMA_IS_DMA32) && BITS_PER_LONG >= 64
+ zone_page_state(&zones[ZONE_DMA32], item) +
+#endif
+ zone_page_state(&zones[ZONE_NORMAL], item) +
+#endif
+#ifdef CONFIG_HIGHMEM
+ zone_page_state(&zones[ZONE_HIGHMEM], item) +
+#endif
+ zone_page_state(&zones[ZONE_DMA], item);
+}
+#else
+#define node_page_state(node, item) global_page_state(item)
+#endif
+
+void __mod_zone_page_state(struct zone *, enum zone_stat_item item, int);
+void __inc_zone_page_state(struct page *, enum zone_stat_item);
+void __dec_zone_page_state(struct page *, enum zone_stat_item);
+
+#define __add_zone_page_state(__z, __i, __d) __mod_zone_page_state(__z, __i, __d)
+#define __sub_zone_page_state(__z, __i, __d) __mod_zone_page_state(__z, __i,-(__d))
+
+void mod_zone_page_state(struct zone *, enum zone_stat_item, int);
+void inc_zone_page_state(struct page *, enum zone_stat_item);
+void dec_zone_page_state(struct page *, enum zone_stat_item);
+
+#define add_zone_page_state(__z, __i, __d) mod_zone_page_state(__z, __i, __d)
+#define sub_zone_page_state(__z, __i, __d) mod_zone_page_state(__z, __i, -(__d))
+
+static inline void zap_zone_vm_stats(struct zone *zone) {
+ memset(zone->vm_stat, 0, sizeof(zone->vm_stat));
+}
+
+#ifdef CONFIG_SMP
+void refresh_cpu_vm_stats(int);
+void refresh_vm_stats(void);
+#else
+static inline void refresh_cpu_vm_stats(int cpu) { }
+static inline void refresh_vm_stats(void) { }
+#endif
+
#endif /* _LINUX_VMSTAT_H */
Index: linux-2.6.17-rc6-cl/arch/ia64/Kconfig
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/ia64/Kconfig 2006-06-12 12:42:41.828596355 -0700
+++ linux-2.6.17-rc6-cl/arch/ia64/Kconfig 2006-06-12 13:31:07.490528917 -0700
@@ -70,6 +70,10 @@ config DMA_IS_DMA32
bool
default y
+config DMA_IS_NORMAL
+ bool
+ default y
+
choice
prompt "System type"
default IA64_GENERIC
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 12:51:44.438570245 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:07:49.853393476 -0700
@@ -3,10 +3,15 @@
*
* Manages VM statistics
* Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds
+ *
+ * zoned VM statistics
+ * Copyright (C) 2006 Silicon Graphics, Inc.,
+ * Christoph Lameter <christoph@lameter.com>
*/
#include <linux/config.h>
#include <linux/mm.h>
+#include <linux/module.h>
/*
* Accumulate the page_state information across all CPUs.
@@ -143,6 +148,259 @@ void get_zone_counts(unsigned long *acti
}
}
+/*
+ * Manage combined zone based / global counters
+ *
+ * vm_stat contains the global counters
+ */
+atomic_long_t vm_stat[NR_STAT_ITEMS];
+
+static inline void zone_page_state_add(long x, struct zone *zone,
+ enum zone_stat_item item)
+{
+ atomic_long_add(x, &zone->vm_stat[item]);
+ atomic_long_add(x, &vm_stat[item]);
+}
+
+#ifdef CONFIG_SMP
+
+#define STAT_THRESHOLD 32
+
+/*
+ * Determine pointer to currently valid differential byte given a zone and
+ * the item number.
+ *
+ * Preemption must be off
+ */
+static inline s8 *diff_pointer(struct zone *zone, enum zone_stat_item item)
+{
+ return &zone_pcp(zone, smp_processor_id())->vm_stat_diff[item];
+}
+
+/*
+ * For use when we know that interrupts are disabled.
+ */
+void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
+ int delta)
+{
+ s8 *p;
+ long x;
+
+ p = diff_pointer(zone, item);
+ x = delta + *p;
+
+ if (unlikely(x > STAT_THRESHOLD || x < -STAT_THRESHOLD)) {
+ zone_page_state_add(x, zone, item);
+ x = 0;
+ }
+
+ *p = x;
+}
+EXPORT_SYMBOL(__mod_zone_page_state);
+
+/*
+ * For an unknown interrupt state
+ */
+void mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
+ int delta)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ __mod_zone_page_state(zone, item, delta);
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(mod_zone_page_state);
+
+/*
+ * Optimized increment and decrement functions.
+ *
+ * These are only for a single page and therefore can take a struct page *
+ * argument instead of struct zone *. This allows the inclusion of the code
+ * generated for page_zone(page) into the optimized functions.
+ *
+ * No overflow check is necessary and therefore the differential can be
+ * incremented or decremented in place which may allow the compilers to
+ * generate better code.
+ *
+ * The increment or decrement is known and therefore one boundary check can
+ * be omitted.
+ *
+ * Some processors have inc/dec instructions that are atomic vs an interrupt.
+ * However, the code must first determine the differential location in a zone
+ * based on the processor number and then inc/dec the counter. There is no
+ * guarantee without disabling preemption that the processor will not change
+ * in between and therefore the atomicity vs. interrupt cannot be exploited
+ * in a useful way here.
+ */
+void __inc_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ struct zone *zone = page_zone(page);
+ s8 *p = diff_pointer(zone, item);
+
+ (*p)++;
+
+ if (unlikely(*p > STAT_THRESHOLD)) {
+ zone_page_state_add(*p, zone, item);
+ *p = 0;
+ }
+}
+EXPORT_SYMBOL(__inc_zone_page_state);
+
+void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ struct zone *zone = page_zone(page);
+ s8 *p = diff_pointer(zone, item);
+
+ (*p)--;
+
+ if (unlikely(*p < -STAT_THRESHOLD)) {
+ zone_page_state_add(*p, zone, item);
+ *p = 0;
+ }
+}
+EXPORT_SYMBOL(__dec_zone_page_state);
+
+void inc_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ unsigned long flags;
+ struct zone *zone;
+ s8 *p;
+
+ zone = page_zone(page);
+ local_irq_save(flags);
+ p = diff_pointer(zone, item);
+
+ (*p)++;
+
+ if (unlikely(*p > STAT_THRESHOLD)) {
+ zone_page_state_add(*p, zone, item);
+ *p = 0;
+ }
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(inc_zone_page_state);
+
+void dec_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ unsigned long flags;
+ struct zone *zone;
+ s8 *p;
+
+ zone = page_zone(page);
+ local_irq_save(flags);
+ p = diff_pointer(zone, item);
+
+ (*p)--;
+
+ if (unlikely(*p < -STAT_THRESHOLD)) {
+ zone_page_state_add(*p, zone, item);
+ *p = 0;
+ }
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(dec_zone_page_state);
+
+/*
+ * Update the zone counters for one cpu.
+ */
+void refresh_cpu_vm_stats(int cpu)
+{
+ struct zone *zone;
+ int i;
+ unsigned long flags;
+
+ for_each_zone(zone) {
+ struct per_cpu_pageset *pcp;
+
+ pcp = zone_pcp(zone, cpu);
+
+ for (i = 0; i < NR_STAT_ITEMS; i++)
+ if (pcp->vm_stat_diff[i]) {
+ local_irq_save(flags);
+ zone_page_state_add(pcp->vm_stat_diff[i],
+ zone, i);
+ pcp->vm_stat_diff[i] = 0;
+ local_irq_restore(flags);
+ }
+ }
+}
+
+static void __refresh_cpu_vm_stats(void *dummy)
+{
+ refresh_cpu_vm_stats(smp_processor_id());
+}
+
+/*
+ * Consolidate all counters.
+ *
+ * Note that the result is less inaccurate but still inaccurate
+ * if concurrent processes are allowed to run.
+ */
+void refresh_vm_stats(void)
+{
+ on_each_cpu(__refresh_cpu_vm_stats, NULL, 0, 1);
+}
+EXPORT_SYMBOL(refresh_vm_stats);
+
+#else /* CONFIG_SMP */
+
+/*
+ * We do not maintain differentials in a single processor configuration.
+ * The functions directly modify the zone and global counters.
+ */
+
+void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
+ int delta)
+{
+ zone_page_state_add(delta, zone, item);
+}
+EXPORT_SYMBOL(__mod_zone_page_state);
+
+void mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
+ int delta)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ zone_page_state_add(delta, zone, item);
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(mod_zone_page_state);
+
+void __inc_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ zone_page_state_add(1, page_zone(page), item);
+}
+EXPORT_SYMBOL(__inc_zone_page_state);
+
+void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ zone_page_state_add(-1, page_zone(page), item);
+}
+EXPORT_SYMBOL(__dec_zone_page_state);
+
+void inc_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ zone_page_state_add(1, page_zone(page), item);
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(inc_zone_page_state);
+
+void dec_zone_page_state(struct page *page, enum zone_stat_item item)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ zone_page_state_add( -1, page_zone(page), item);
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(dec_zone_page_state);
+#endif
+
#ifdef CONFIG_PROC_FS
#include <linux/seq_file.h>
@@ -204,6 +462,9 @@ struct seq_operations fragmentation_op =
};
static char *vmstat_text[] = {
+ /* Zoned VM counters */
+
+ /* Page state */
"nr_dirty",
"nr_writeback",
"nr_unstable",
@@ -297,6 +558,11 @@ static int zoneinfo_show(struct seq_file
zone->nr_scan_active, zone->nr_scan_inactive,
zone->spanned_pages,
zone->present_pages);
+
+ for (i = 0; i < NR_STAT_ITEMS; i++)
+ seq_printf(m, "\n %-12s %lu", vmstat_text[i],
+ zone_page_state(zone, i));
+
seq_printf(m,
"\n protection: (%lu",
zone->lowmem_reserve[0]);
@@ -368,19 +634,25 @@ struct seq_operations zoneinfo_op = {
static void *vmstat_start(struct seq_file *m, loff_t *pos)
{
+ unsigned long *v;
struct page_state *ps;
+ int i;
if (*pos >= ARRAY_SIZE(vmstat_text))
return NULL;
- ps = kmalloc(sizeof(*ps), GFP_KERNEL);
- m->private = ps;
- if (!ps)
+ v = kmalloc(NR_STAT_ITEMS *sizeof(unsigned long)
+ + sizeof(*ps), GFP_KERNEL);
+ m->private = v;
+ if (!v)
return ERR_PTR(-ENOMEM);
+ for (i = 0; i < NR_STAT_ITEMS; i++)
+ v[i] = global_page_state(i);
+ ps = (struct page_state *)(v + NR_STAT_ITEMS);
get_full_page_state(ps);
ps->pgpgin /= 2; /* sectors -> kbytes */
ps->pgpgout /= 2;
- return (unsigned long *)ps + *pos;
+ return v + *pos;
}
static void *vmstat_next(struct seq_file *m, void *arg, loff_t *pos)
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: zoned vm counters: per zone counter functionality
2006-06-12 21:12 ` [PATCH 02/21] Basic ZVC (zoned vm counter) implementation Christoph Lameter
@ 2006-06-13 5:37 ` Nick Piggin
2006-06-13 15:55 ` Christoph Lameter
0 siblings, 1 reply; 39+ messages in thread
From: Nick Piggin @ 2006-06-13 5:37 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti,
linux-mm, Andi Kleen, Dave Chinner
Hi Christoph,
Looks like a nice patchset. I really like that you've moved the counters
out of page alloc.
Is there any point in using a more meaningful namespace prefix than NR_
for the zone_stat_items?
> +enum zone_stat_item {
> + NR_STAT_ITEMS };
> +
--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: zoned vm counters: per zone counter functionality
2006-06-13 5:37 ` zoned vm counters: per zone counter functionality Nick Piggin
@ 2006-06-13 15:55 ` Christoph Lameter
2006-06-14 1:21 ` Nick Piggin
0 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-13 15:55 UTC (permalink / raw)
To: Nick Piggin
Cc: linux-kernel, akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti,
linux-mm, Andi Kleen, Dave Chinner
On Tue, 13 Jun 2006, Nick Piggin wrote:
> Is there any point in using a more meaningful namespace prefix than NR_
> for the zone_stat_items?
>
>
> > +enum zone_stat_item {
> > + NR_STAT_ITEMS };
> > +
How about
NR_VM_ZONE_STAT_ITEMS ?
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: zoned vm counters: per zone counter functionality
2006-06-13 15:55 ` Christoph Lameter
@ 2006-06-14 1:21 ` Nick Piggin
2006-06-14 5:37 ` Hugh Dickins
2006-06-14 16:14 ` Christoph Lameter
0 siblings, 2 replies; 39+ messages in thread
From: Nick Piggin @ 2006-06-14 1:21 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti,
linux-mm, Andi Kleen, Dave Chinner
Christoph Lameter wrote:
>On Tue, 13 Jun 2006, Nick Piggin wrote:
>
>
>>Is there any point in using a more meaningful namespace prefix than NR_
>>for the zone_stat_items?
>>
>>
>>
>>>+enum zone_stat_item {
>>>+ NR_STAT_ITEMS };
>>>+
>>>
>
>How about
>
>NR_VM_ZONE_STAT_ITEMS ?
>
>
>
I guess that's OK.
Hmm, then NR_ANON would become VM_ZONE_STAT_NR_ANON? That might be a bit
long for your tastes, maybe the prefix could be hidden by "clever" macros?
--
Send instant messages to your online friends http://au.messenger.yahoo.com
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: zoned vm counters: per zone counter functionality
2006-06-14 1:21 ` Nick Piggin
@ 2006-06-14 5:37 ` Hugh Dickins
2006-06-14 5:53 ` Andi Kleen
2006-06-14 16:14 ` Christoph Lameter
1 sibling, 1 reply; 39+ messages in thread
From: Hugh Dickins @ 2006-06-14 5:37 UTC (permalink / raw)
To: Nick Piggin
Cc: Christoph Lameter, linux-kernel, akpm, Con Kolivas,
Marcelo Tosatti, linux-mm, Andi Kleen, Dave Chinner
On Wed, 14 Jun 2006, Nick Piggin wrote:
>
> Hmm, then NR_ANON would become VM_ZONE_STAT_NR_ANON? That might be a bit
> long for your tastes, maybe the prefix could be hidden by "clever" macros?
Don't even begin to think of "clever" macros.
Hugh
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: zoned vm counters: per zone counter functionality
2006-06-14 5:37 ` Hugh Dickins
@ 2006-06-14 5:53 ` Andi Kleen
0 siblings, 0 replies; 39+ messages in thread
From: Andi Kleen @ 2006-06-14 5:53 UTC (permalink / raw)
To: Hugh Dickins
Cc: Nick Piggin, Christoph Lameter, linux-kernel, akpm, Con Kolivas,
Marcelo Tosatti, linux-mm, Dave Chinner
On Wednesday 14 June 2006 07:37, Hugh Dickins wrote:
> On Wed, 14 Jun 2006, Nick Piggin wrote:
> >
> > Hmm, then NR_ANON would become VM_ZONE_STAT_NR_ANON? That might be a bit
> > long for your tastes, maybe the prefix could be hidden by "clever" macros?
>
> Don't even begin to think of "clever" macros.
Yes they cause cancer of the grep. SCNR.
-Andi
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: zoned vm counters: per zone counter functionality
2006-06-14 1:21 ` Nick Piggin
2006-06-14 5:37 ` Hugh Dickins
@ 2006-06-14 16:14 ` Christoph Lameter
1 sibling, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-14 16:14 UTC (permalink / raw)
To: Nick Piggin
Cc: linux-kernel, akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti,
linux-mm, Andi Kleen, Dave Chinner
On Wed, 14 Jun 2006, Nick Piggin wrote:
> I guess that's OK.
>
> Hmm, then NR_ANON would become VM_ZONE_STAT_NR_ANON? That might be a bit
> long for your tastes, maybe the prefix could be hidden by "clever" macros?
I only changed the NR_STAT_ITEMS but kept the rest since these symbols are
used frequently.
^ permalink raw reply [flat|nested] 39+ messages in thread
* [PATCH 03/21] Convert nr_mapped to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
2006-06-12 21:12 ` [PATCH 01/21] Create vmstat.c/.h from page_alloc.c/.h Christoph Lameter
2006-06-12 21:12 ` [PATCH 02/21] Basic ZVC (zoned vm counter) implementation Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 04/21] swap_prefetch: Convert nr_mapped to ZVC Christoph Lameter
` (17 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_mapped to per zone counter
From: Christoph Lameter <clameter@sgi.com>
nr_mapped is important because it allows a determination how many pages of a
zone are not mapped, which would allow a more efficient means of determining
when we need to reclaim memory in a zone.
We take the nr_mapped field out of the page state structure and define a new
per zone counter named NR_MAPPED.
We replace the use of nr_mapped in various kernel locations. This avoids the
looping over all processors in try_to_free_pages(), writeback, reclaim (swap +
zone reclaim).
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/i386/mm/pgtable.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c 2006-06-12 12:57:23.381448638 -0700
@@ -61,7 +61,7 @@ void show_mem(void)
get_page_state(&ps);
printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
- printk(KERN_INFO "%lu pages mapped\n", ps.nr_mapped);
+ printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_MAPPED));
printk(KERN_INFO "%lu pages slab\n", ps.nr_slab);
printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
}
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 12:42:42.797286410 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 12:57:23.382425140 -0700
@@ -54,8 +54,6 @@ static ssize_t node_read_meminfo(struct
ps.nr_dirty = 0;
if ((long)ps.nr_writeback < 0)
ps.nr_writeback = 0;
- if ((long)ps.nr_mapped < 0)
- ps.nr_mapped = 0;
if ((long)ps.nr_slab < 0)
ps.nr_slab = 0;
@@ -84,7 +82,7 @@ static ssize_t node_read_meminfo(struct
nid, K(i.freeram - i.freehigh),
nid, K(ps.nr_dirty),
nid, K(ps.nr_writeback),
- nid, K(ps.nr_mapped),
+ nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(ps.nr_slab));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 12:42:48.170000806 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 12:57:23.383401642 -0700
@@ -190,7 +190,7 @@ static int meminfo_read_proc(char *page,
K(i.freeswap),
K(ps.nr_dirty),
K(ps.nr_writeback),
- K(ps.nr_mapped),
+ K(global_page_state(NR_MAPPED)),
K(ps.nr_slab),
K(allowed),
K(committed),
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 12:54:33.238664762 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 12:57:23.383401642 -0700
@@ -47,6 +47,9 @@ struct zone_padding {
#endif
enum zone_stat_item {
+ NR_MAPPED, /* mapped into pagetables.
+ only modified from process context */
+
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 12:54:33.240617767 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 12:57:23.385354646 -0700
@@ -1409,7 +1409,7 @@ void show_free_areas(void)
ps.nr_unstable,
nr_free_pages(),
ps.nr_slab,
- ps.nr_mapped,
+ global_page_state(NR_MAPPED),
ps.nr_page_table_pages);
for_each_zone(zone) {
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-12 12:42:52.043784523 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-12 12:57:23.386331148 -0700
@@ -111,7 +111,7 @@ static void get_writeback_state(struct w
{
wbs->nr_dirty = read_page_state(nr_dirty);
wbs->nr_unstable = read_page_state(nr_unstable);
- wbs->nr_mapped = read_page_state(nr_mapped);
+ wbs->nr_mapped = global_page_state(NR_MAPPED);
wbs->nr_writeback = read_page_state(nr_writeback);
}
Index: linux-2.6.17-rc6-cl/mm/rmap.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/rmap.c 2006-06-12 12:42:52.051596540 -0700
+++ linux-2.6.17-rc6-cl/mm/rmap.c 2006-06-12 12:57:23.386331148 -0700
@@ -455,7 +455,7 @@ static void __page_set_anon_rmap(struct
* nr_mapped state can be updated without turning off
* interrupts because it is not modified via interrupt.
*/
- __inc_page_state(nr_mapped);
+ __inc_zone_page_state(page, NR_MAPPED);
}
/**
@@ -499,7 +499,7 @@ void page_add_new_anon_rmap(struct page
void page_add_file_rmap(struct page *page)
{
if (atomic_inc_and_test(&page->_mapcount))
- __inc_page_state(nr_mapped);
+ __inc_zone_page_state(page, NR_MAPPED);
}
/**
@@ -531,7 +531,7 @@ void page_remove_rmap(struct page *page)
*/
if (page_test_and_clear_dirty(page))
set_page_dirty(page);
- __dec_page_state(nr_mapped);
+ __dec_zone_page_state(page, NR_MAPPED);
}
}
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-12 12:42:52.067220573 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-12 12:57:23.388284152 -0700
@@ -996,7 +996,7 @@ unsigned long try_to_free_pages(struct z
}
for (priority = DEF_PRIORITY; priority >= 0; priority--) {
- sc.nr_mapped = read_page_state(nr_mapped);
+ sc.nr_mapped = global_page_state(NR_MAPPED);
sc.nr_scanned = 0;
if (!priority)
disable_swap_token();
@@ -1081,7 +1081,7 @@ loop_again:
total_scanned = 0;
nr_reclaimed = 0;
sc.may_writepage = !laptop_mode,
- sc.nr_mapped = read_page_state(nr_mapped);
+ sc.nr_mapped = global_page_state(NR_MAPPED);
inc_page_state(pageoutrun);
@@ -1416,7 +1416,7 @@ unsigned long shrink_all_memory(unsigned
for (prio = DEF_PRIORITY; prio >= 0; prio--) {
unsigned long nr_to_scan = nr_pages - ret;
- sc.nr_mapped = read_page_state(nr_mapped);
+ sc.nr_mapped = global_page_state(NR_MAPPED);
sc.nr_scanned = 0;
ret += shrink_all_zones(nr_to_scan, prio, pass, &sc);
@@ -1558,7 +1558,7 @@ static int __zone_reclaim(struct zone *z
struct scan_control sc = {
.may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE),
.may_swap = !!(zone_reclaim_mode & RECLAIM_SWAP),
- .nr_mapped = read_page_state(nr_mapped),
+ .nr_mapped = global_page_state(NR_MAPPED),
.swap_cluster_max = max_t(unsigned long, nr_pages,
SWAP_CLUSTER_MAX),
.gfp_mask = gfp_mask,
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 12:55:05.481785246 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:00:45.074036260 -0700
@@ -463,13 +463,13 @@ struct seq_operations fragmentation_op =
static char *vmstat_text[] = {
/* Zoned VM counters */
+ "nr_mapped",
/* Page state */
"nr_dirty",
"nr_writeback",
"nr_unstable",
"nr_page_table_pages",
- "nr_mapped",
"nr_slab",
"pgpgin",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 04/21] swap_prefetch: Convert nr_mapped to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (2 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 03/21] Convert nr_mapped to per zone counter Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 23:36 ` Con Kolivas
2006-06-12 21:13 ` [PATCH 05/21] Conversion of nr_pagecache to per zone counter Christoph Lameter
` (16 subsequent siblings)
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
swap_prefetch: nr_mapped conversion
This is separate on request by Andrew.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-10 22:04:26.763047220 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 11:29:47.664697453 -0700
@@ -394,7 +394,8 @@ static int prefetch_suitable(void)
* even if the slab is being allocated on a remote node. This
* would be expensive to fix and not of great significance.
*/
- limit = ps.nr_mapped + ps.nr_slab + ps.nr_dirty +
+ limit = node_page_state(node, NR_MAPPED) +
+ ps.nr_slab + ps.nr_dirty +
ps.nr_unstable + total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 05/21] Conversion of nr_pagecache to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (3 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 04/21] swap_prefetch: Convert nr_mapped to ZVC Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 06/21] Remove nr_mapped from scan controls structure Christoph Lameter
` (15 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_pagecache to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Currently a single atomic variable is used to establish the size of the page
cache in the whole machine. The zoned VM counters have the same method of
implementation as the nr_pagecache code but also allow the determination of
the pagecache size per zone.
Remove the special implementation for nr_pagecache and make it a zoned
counter.
Updates of the page cache counters are always performed with interrupts off.
We can therefore use the __ variant here.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/sparc64/kernel/sys_sunos32.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/sparc64/kernel/sys_sunos32.c 2006-06-12 12:42:42.240680230 -0700
+++ linux-2.6.17-rc6-cl/arch/sparc64/kernel/sys_sunos32.c 2006-06-12 13:00:50.575648675 -0700
@@ -155,7 +155,7 @@ asmlinkage int sunos_brk(u32 baddr)
* simple, it hopefully works in most obvious cases.. Easy to
* fool it, but this should catch most mistakes.
*/
- freepages = get_page_cache_size();
+ freepages = global_page_state(NR_PAGECACHE);
freepages >>= 1;
freepages += nr_free_pages();
freepages += nr_swap_pages;
Index: linux-2.6.17-rc6-cl/arch/sparc/kernel/sys_sunos.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/sparc/kernel/sys_sunos.c 2006-06-12 12:42:42.249468748 -0700
+++ linux-2.6.17-rc6-cl/arch/sparc/kernel/sys_sunos.c 2006-06-12 13:00:50.577601679 -0700
@@ -196,7 +196,7 @@ asmlinkage int sunos_brk(unsigned long b
* simple, it hopefully works in most obvious cases.. Easy to
* fool it, but this should catch most mistakes.
*/
- freepages = get_page_cache_size();
+ freepages = global_page_state(NR_PAGECACHE);
freepages >>= 1;
freepages += nr_free_pages();
freepages += nr_swap_pages;
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 12:57:23.383401642 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:00:50.578578181 -0700
@@ -142,7 +142,8 @@ static int meminfo_read_proc(char *page,
allowed = ((totalram_pages - hugetlb_total_pages())
* sysctl_overcommit_ratio / 100) + total_swap_pages;
- cached = get_page_cache_size() - total_swapcache_pages - i.bufferram;
+ cached = global_page_state(NR_PAGECACHE) -
+ total_swapcache_pages - i.bufferram;
if (cached < 0)
cached = 0;
Index: linux-2.6.17-rc6-cl/include/linux/pagemap.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/pagemap.h 2006-06-12 12:42:50.853428498 -0700
+++ linux-2.6.17-rc6-cl/include/linux/pagemap.h 2006-06-12 13:00:50.579554683 -0700
@@ -115,51 +115,6 @@ int add_to_page_cache_lru(struct page *p
extern void remove_from_page_cache(struct page *page);
extern void __remove_from_page_cache(struct page *page);
-extern atomic_t nr_pagecache;
-
-#ifdef CONFIG_SMP
-
-#define PAGECACHE_ACCT_THRESHOLD max(16, NR_CPUS * 2)
-DECLARE_PER_CPU(long, nr_pagecache_local);
-
-/*
- * pagecache_acct implements approximate accounting for pagecache.
- * vm_enough_memory() do not need high accuracy. Writers will keep
- * an offset in their per-cpu arena and will spill that into the
- * global count whenever the absolute value of the local count
- * exceeds the counter's threshold.
- *
- * MUST be protected from preemption.
- * current protection is mapping->page_lock.
- */
-static inline void pagecache_acct(int count)
-{
- long *local;
-
- local = &__get_cpu_var(nr_pagecache_local);
- *local += count;
- if (*local > PAGECACHE_ACCT_THRESHOLD || *local < -PAGECACHE_ACCT_THRESHOLD) {
- atomic_add(*local, &nr_pagecache);
- *local = 0;
- }
-}
-
-#else
-
-static inline void pagecache_acct(int count)
-{
- atomic_add(count, &nr_pagecache);
-}
-#endif
-
-static inline unsigned long get_page_cache_size(void)
-{
- int ret = atomic_read(&nr_pagecache);
- if (unlikely(ret < 0))
- ret = 0;
- return ret;
-}
-
/*
* Return byte-offset into filesystem object for page.
*/
Index: linux-2.6.17-rc6-cl/mm/filemap.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/filemap.c 2006-06-12 12:42:52.024254482 -0700
+++ linux-2.6.17-rc6-cl/mm/filemap.c 2006-06-12 13:00:50.581507687 -0700
@@ -126,7 +126,7 @@ void __remove_from_page_cache(struct pag
radix_tree_delete(&mapping->page_tree, page->index);
page->mapping = NULL;
mapping->nrpages--;
- pagecache_acct(-1);
+ __dec_zone_page_state(page, NR_PAGECACHE);
}
EXPORT_SYMBOL(__remove_from_page_cache);
@@ -424,7 +424,7 @@ int add_to_page_cache(struct page *page,
page->mapping = mapping;
page->index = offset;
mapping->nrpages++;
- pagecache_acct(1);
+ __inc_zone_page_state(page, NR_PAGECACHE);
}
write_unlock_irq(&mapping->tree_lock);
radix_tree_preload_end();
Index: linux-2.6.17-rc6-cl/mm/mmap.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/mmap.c 2006-06-12 12:42:52.037925511 -0700
+++ linux-2.6.17-rc6-cl/mm/mmap.c 2006-06-12 13:00:50.582484189 -0700
@@ -96,7 +96,7 @@ int __vm_enough_memory(long pages, int c
if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) {
unsigned long n;
- free = get_page_cache_size();
+ free = global_page_state(NR_PAGECACHE);
free += nr_swap_pages;
/*
Index: linux-2.6.17-rc6-cl/mm/nommu.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/nommu.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/mm/nommu.c 2006-06-12 13:00:50.583460691 -0700
@@ -1122,7 +1122,7 @@ int __vm_enough_memory(long pages, int c
if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) {
unsigned long n;
- free = get_page_cache_size();
+ free = global_page_state(NR_PAGECACHE);
free += nr_swap_pages;
/*
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 12:57:23.385354646 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:00:50.584437193 -0700
@@ -2231,16 +2231,11 @@ static int page_alloc_cpu_notify(struct
unsigned long action, void *hcpu)
{
int cpu = (unsigned long)hcpu;
- long *count;
unsigned long *src, *dest;
if (action == CPU_DEAD) {
int i;
- /* Drain local pagecache count. */
- count = &per_cpu(nr_pagecache_local, cpu);
- atomic_add(*count, &nr_pagecache);
- *count = 0;
local_irq_disable();
__drain_pages(cpu);
Index: linux-2.6.17-rc6-cl/mm/swap_state.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_state.c 2006-06-12 12:42:52.062338063 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_state.c 2006-06-12 13:00:50.585413695 -0700
@@ -89,7 +89,7 @@ static int __add_to_swap_cache(struct pa
SetPageSwapCache(page);
set_page_private(page, entry.val);
total_swapcache_pages++;
- pagecache_acct(1);
+ __inc_zone_page_state(page, NR_PAGECACHE);
}
write_unlock_irq(&swapper_space.tree_lock);
radix_tree_preload_end();
@@ -135,7 +135,7 @@ void __delete_from_swap_cache(struct pag
set_page_private(page, 0);
ClearPageSwapCache(page);
total_swapcache_pages--;
- pagecache_acct(-1);
+ __dec_zone_page_state(page, NR_PAGECACHE);
INC_CACHE_INFO(del_total);
}
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 12:57:23.383401642 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:00:50.585413695 -0700
@@ -49,7 +49,7 @@ struct zone_padding {
enum zone_stat_item {
NR_MAPPED, /* mapped into pagetables.
only modified from process context */
-
+ NR_PAGECACHE,
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/arch/s390/appldata/appldata_mem.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/s390/appldata/appldata_mem.c 2006-06-12 12:42:42.183066607 -0700
+++ linux-2.6.17-rc6-cl/arch/s390/appldata/appldata_mem.c 2006-06-12 13:00:50.586390197 -0700
@@ -130,7 +130,8 @@ static void appldata_get_mem_data(void *
mem_data->totalhigh = P2K(val.totalhigh);
mem_data->freehigh = P2K(val.freehigh);
mem_data->bufferram = P2K(val.bufferram);
- mem_data->cached = P2K(atomic_read(&nr_pagecache) - val.bufferram);
+ mem_data->cached = P2K(global_page_state(NR_PAGECACHE)
+ - val.bufferram);
si_swapinfo(&val);
mem_data->totalswap = P2K(val.totalswap);
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 12:57:23.382425140 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:00:50.587366699 -0700
@@ -69,6 +69,7 @@ static ssize_t node_read_meminfo(struct
"Node %d LowFree: %8lu kB\n"
"Node %d Dirty: %8lu kB\n"
"Node %d Writeback: %8lu kB\n"
+ "Node %d PageCache: %8lu kB\n"
"Node %d Mapped: %8lu kB\n"
"Node %d Slab: %8lu kB\n",
nid, K(i.totalram),
@@ -82,6 +83,7 @@ static ssize_t node_read_meminfo(struct
nid, K(i.freeram - i.freehigh),
nid, K(ps.nr_dirty),
nid, K(ps.nr_writeback),
+ nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(ps.nr_slab));
n += hugetlb_report_node_meminfo(nid, buf + n);
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:00:45.074036260 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:01:24.304028650 -0700
@@ -464,6 +464,7 @@ struct seq_operations fragmentation_op =
static char *vmstat_text[] = {
/* Zoned VM counters */
"nr_mapped",
+ "nr_pagecache",
/* Page state */
"nr_dirty",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 06/21] Remove nr_mapped from scan controls structure
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (4 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 05/21] Conversion of nr_pagecache to per zone counter Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 07/21] Split NR_ANON off from NR_MAPPED Christoph Lameter
` (14 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned VM stats: Remove nr_mapped from scan control
From: Christoph Lameter <clameter@sgi.com>
We can now access the number of pages in a mapped state in an inexpensive
way in shrink_active_list. So drop the nr_mapped field from scan_control.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-10 17:25:04.636700267 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-10 19:50:44.835034840 -0700
@@ -48,8 +48,6 @@ struct scan_control {
/* Incremented by the number of inactive pages that were scanned */
unsigned long nr_scanned;
- unsigned long nr_mapped; /* From page_state */
-
/* This context's GFP mask */
gfp_t gfp_mask;
@@ -749,7 +747,8 @@ static void shrink_active_list(unsigned
* how much memory
* is mapped.
*/
- mapped_ratio = (sc->nr_mapped * 100) / vm_total_pages;
+ mapped_ratio = (global_page_state(NR_MAPPED) * 100) /
+ vm_total_pages;
/*
* Now decide how much we really want to unmap some pages. The
@@ -996,7 +995,6 @@ unsigned long try_to_free_pages(struct z
}
for (priority = DEF_PRIORITY; priority >= 0; priority--) {
- sc.nr_mapped = global_page_state(NR_MAPPED);
sc.nr_scanned = 0;
if (!priority)
disable_swap_token();
@@ -1081,7 +1079,6 @@ loop_again:
total_scanned = 0;
nr_reclaimed = 0;
sc.may_writepage = !laptop_mode,
- sc.nr_mapped = global_page_state(NR_MAPPED);
inc_page_state(pageoutrun);
@@ -1416,7 +1413,6 @@ unsigned long shrink_all_memory(unsigned
for (prio = DEF_PRIORITY; prio >= 0; prio--) {
unsigned long nr_to_scan = nr_pages - ret;
- sc.nr_mapped = global_page_state(NR_MAPPED);
sc.nr_scanned = 0;
ret += shrink_all_zones(nr_to_scan, prio, pass, &sc);
@@ -1558,7 +1554,6 @@ static int __zone_reclaim(struct zone *z
struct scan_control sc = {
.may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE),
.may_swap = !!(zone_reclaim_mode & RECLAIM_SWAP),
- .nr_mapped = global_page_state(NR_MAPPED),
.swap_cluster_max = max_t(unsigned long, nr_pages,
SWAP_CLUSTER_MAX),
.gfp_mask = gfp_mask,
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 07/21] Split NR_ANON off from NR_MAPPED
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (5 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 06/21] Remove nr_mapped from scan controls structure Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 08/21] swap_prefetch: Split NR_ANON off NR_MAPPED Christoph Lameter
` (13 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned VM stats: Add NR_ANON
From: Christoph Lameter <clameter@sgi.com>
The current NR_MAPPED is used by zone reclaim and the dirty load
calculation as the number of mapped pagecache pages. However, that is not
true. NR_MAPPED includes the mapped anonymous pages. This patch
separates those and therefore allows an accurate tracking of the anonymous
pages per zone.
It then becomes possible to determine the number of unmapped pages
per zone and we can avoid scanning for unmapped pages if there
are none.
Also it may now be possible to determine the mapped/unmapped ratio in
get_dirty_limit. Isnt the number of anonymous pages irrelevant in that
calculation?
Note that this will change the meaning of the number of mapped pages
reported in /proc/vmstat /proc/meminfo and in the per node statistics.
This may affect user space tools that monitor these counters!
NR_MAPPED works like NR_DIRTY. It is only valid for pagecache pages.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:00:50.578578181 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:01:30.621020255 -0700
@@ -168,6 +168,7 @@ static int meminfo_read_proc(char *page,
"SwapFree: %8lu kB\n"
"Dirty: %8lu kB\n"
"Writeback: %8lu kB\n"
+ "Anonymous: %8lu kB\n"
"Mapped: %8lu kB\n"
"Slab: %8lu kB\n"
"CommitLimit: %8lu kB\n"
@@ -191,6 +192,7 @@ static int meminfo_read_proc(char *page,
K(i.freeswap),
K(ps.nr_dirty),
K(ps.nr_writeback),
+ K(global_page_state(NR_ANON)),
K(global_page_state(NR_MAPPED)),
K(ps.nr_slab),
K(allowed),
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:00:50.585413695 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:01:30.622973259 -0700
@@ -47,7 +47,8 @@ struct zone_padding {
#endif
enum zone_stat_item {
- NR_MAPPED, /* mapped into pagetables.
+ NR_ANON, /* Mapped anonymous pages */
+ NR_MAPPED, /* pagecache pages mapped into pagetables.
only modified from process context */
NR_PAGECACHE,
NR_STAT_ITEMS };
Index: linux-2.6.17-rc6-cl/mm/rmap.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/rmap.c 2006-06-12 12:57:23.386331148 -0700
+++ linux-2.6.17-rc6-cl/mm/rmap.c 2006-06-12 13:01:30.623949761 -0700
@@ -455,7 +455,7 @@ static void __page_set_anon_rmap(struct
* nr_mapped state can be updated without turning off
* interrupts because it is not modified via interrupt.
*/
- __inc_zone_page_state(page, NR_MAPPED);
+ __inc_zone_page_state(page, NR_ANON);
}
/**
@@ -531,7 +531,8 @@ void page_remove_rmap(struct page *page)
*/
if (page_test_and_clear_dirty(page))
set_page_dirty(page);
- __dec_zone_page_state(page, NR_MAPPED);
+ __dec_zone_page_state(page,
+ PageAnon(page) ? NR_ANON : NR_MAPPED);
}
}
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-12 13:01:27.580192947 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-12 13:01:30.624926263 -0700
@@ -747,7 +747,8 @@ static void shrink_active_list(unsigned
* how much memory
* is mapped.
*/
- mapped_ratio = (global_page_state(NR_MAPPED) * 100) /
+ mapped_ratio = ((global_page_state(NR_MAPPED) +
+ global_page_state(NR_ANON)) * 100) /
vm_total_pages;
/*
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:00:50.587366699 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:01:30.625902765 -0700
@@ -71,6 +71,7 @@ static ssize_t node_read_meminfo(struct
"Node %d Writeback: %8lu kB\n"
"Node %d PageCache: %8lu kB\n"
"Node %d Mapped: %8lu kB\n"
+ "Node %d Anonymous: %8lu kB\n"
"Node %d Slab: %8lu kB\n",
nid, K(i.totalram),
nid, K(i.freeram),
@@ -85,6 +86,7 @@ static ssize_t node_read_meminfo(struct
nid, K(ps.nr_writeback),
nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
+ nid, K(node_page_state(nid, NR_ANON)),
nid, K(ps.nr_slab));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-12 12:57:23.386331148 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-12 13:01:30.625902765 -0700
@@ -111,7 +111,8 @@ static void get_writeback_state(struct w
{
wbs->nr_dirty = read_page_state(nr_dirty);
wbs->nr_unstable = read_page_state(nr_unstable);
- wbs->nr_mapped = global_page_state(NR_MAPPED);
+ wbs->nr_mapped = global_page_state(NR_MAPPED) +
+ global_page_state(NR_ANON);
wbs->nr_writeback = read_page_state(nr_writeback);
}
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:01:24.304028650 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:02:08.289585892 -0700
@@ -465,6 +465,7 @@ static char *vmstat_text[] = {
/* Zoned VM counters */
"nr_mapped",
"nr_pagecache",
+ "nr_anon",
/* Page state */
"nr_dirty",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 08/21] swap_prefetch: Split NR_ANON off NR_MAPPED
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (6 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 07/21] Split NR_ANON off from NR_MAPPED Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 23:36 ` Con Kolivas
2006-06-12 21:13 ` [PATCH 09/21] zone_reclaim: remove /proc/sys/vm/zone_reclaim_interval Christoph Lameter
` (12 subsequent siblings)
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
swap_prefetch: add NR_ANON
Separated out by request from Andrew.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-10 14:57:06.644921771 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-10 15:39:44.582122436 -0700
@@ -395,6 +395,7 @@ static int prefetch_suitable(void)
* would be expensive to fix and not of great significance.
*/
limit = node_page_state(node, NR_MAPPED) +
+ node_page_state(node, NR_ANON) +
ps.nr_slab + ps.nr_dirty +
ps.nr_unstable + total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 09/21] zone_reclaim: remove /proc/sys/vm/zone_reclaim_interval
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (7 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 08/21] swap_prefetch: Split NR_ANON off NR_MAPPED Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 10/21] Conversion of nr_slab to per zone counter Christoph Lameter
` (11 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: use per zone counters to remove zone_reclaim_interval
From: Christoph Lameter <clameter@sgi.com>
The zone_reclaim_interval was necessary because we were not able to determine
how many unmapped pages exist in a zone. Therefore we had to scan in
intervals to figure out if any pages were unmapped.
With the zoned counters and NR_ANON we now know the number of pagecache pages
and the number of mapped pages in a zone. So we can simply skip the reclaim
if there is an insufficient number of unmapped pages. We use SWAP_CLUSTER_MAX
as the boundary.
Drop all support for zone_reclaim_interval.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:33:11.045379511 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:37:33.840632482 -0700
@@ -183,12 +183,6 @@ struct zone {
/* Zone statistics */
atomic_long_t vm_stat[NR_STAT_ITEMS];
- /*
- * timestamp (in jiffies) of the last zone reclaim that did not
- * result in freeing of pages. This is used to avoid repeated scans
- * if all memory in the zone is in use.
- */
- unsigned long last_unsuccessful_zone_reclaim;
/*
* prev_priority holds the scanning priority for this zone. It is
Index: linux-2.6.17-rc6-cl/include/linux/swap.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/swap.h 2006-06-12 12:42:51.046775908 -0700
+++ linux-2.6.17-rc6-cl/include/linux/swap.h 2006-06-12 13:33:11.120570168 -0700
@@ -190,7 +190,6 @@ extern long vm_total_pages;
#ifdef CONFIG_NUMA
extern int zone_reclaim_mode;
-extern int zone_reclaim_interval;
extern int zone_reclaim(struct zone *, gfp_t, unsigned int);
#else
#define zone_reclaim_mode 0
Index: linux-2.6.17-rc6-cl/include/linux/sysctl.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/sysctl.h 2006-06-12 12:42:51.074117966 -0700
+++ linux-2.6.17-rc6-cl/include/linux/sysctl.h 2006-06-12 13:33:11.121546670 -0700
@@ -191,7 +191,6 @@ enum
VM_DROP_PAGECACHE=29, /* int: nuke lots of pagecache */
VM_PERCPU_PAGELIST_FRACTION=30,/* int: fraction of pages in each percpu_pagelist */
VM_ZONE_RECLAIM_MODE=31, /* reclaim local zone memory before going off node */
- VM_ZONE_RECLAIM_INTERVAL=32, /* time period to wait after reclaim failure */
VM_PANIC_ON_OOM=33, /* panic at out-of-memory */
VM_VDSO_ENABLED=34, /* map VDSO into new processes? */
VM_SWAP_PREFETCH=35, /* swap prefetch */
Index: linux-2.6.17-rc6-cl/kernel/sysctl.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/kernel/sysctl.c 2006-06-12 12:42:51.948087320 -0700
+++ linux-2.6.17-rc6-cl/kernel/sysctl.c 2006-06-12 13:33:11.123499674 -0700
@@ -1015,15 +1015,6 @@ static ctl_table vm_table[] = {
.strategy = &sysctl_intvec,
.extra1 = &zero,
},
- {
- .ctl_name = VM_ZONE_RECLAIM_INTERVAL,
- .procname = "zone_reclaim_interval",
- .data = &zone_reclaim_interval,
- .maxlen = sizeof(zone_reclaim_interval),
- .mode = 0644,
- .proc_handler = &proc_dointvec_jiffies,
- .strategy = &sysctl_jiffies,
- },
#endif
#ifdef CONFIG_X86_32
{
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-12 13:33:11.047332515 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-12 13:33:11.124476176 -0700
@@ -1530,11 +1530,6 @@ int zone_reclaim_mode __read_mostly;
#define RECLAIM_SLAB (1<<3) /* Do a global slab shrink if the zone is out of memory */
/*
- * Mininum time between zone reclaim scans
- */
-int zone_reclaim_interval __read_mostly = 30*HZ;
-
-/*
* Priority for ZONE_RECLAIM. This determines the fraction of pages
* of a node considered for each zone_reclaim. 4 scans 1/16th of
* a zone.
@@ -1599,16 +1594,6 @@ static int __zone_reclaim(struct zone *z
p->reclaim_state = NULL;
current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
-
- if (nr_reclaimed == 0) {
- /*
- * We were unable to reclaim enough pages to stay on node. We
- * now allow off node accesses for a certain time period before
- * trying again to reclaim pages from the local zone.
- */
- zone->last_unsuccessful_zone_reclaim = jiffies;
- }
-
return nr_reclaimed >= nr_pages;
}
@@ -1618,13 +1603,17 @@ int zone_reclaim(struct zone *zone, gfp_
int node_id;
/*
- * Do not reclaim if there was a recent unsuccessful attempt at zone
- * reclaim. In that case we let allocations go off node for the
- * zone_reclaim_interval. Otherwise we would scan for each off-node
- * page allocation.
+ * Do not reclaim if there are not enough reclaimable pages in this
+ * zone that would satify this allocations.
+ *
+ * All unmapped pagecache pages are reclaimable.
+ *
+ * Both counters may be temporarily off a bit so we use
+ * SWAP_CLUSTER_MAX as the boundary. It may also be good to
+ * leave a few frequently used unmapped pagecache pages around.
*/
- if (time_before(jiffies,
- zone->last_unsuccessful_zone_reclaim + zone_reclaim_interval))
+ if (zone_page_state(zone, NR_PAGECACHE) -
+ zone_page_state(zone, NR_MAPPED) < SWAP_CLUSTER_MAX)
return 0;
/*
Index: linux-2.6.17-rc6-cl/Documentation/sysctl/vm.txt
===================================================================
--- linux-2.6.17-rc6-cl.orig/Documentation/sysctl/vm.txt 2006-06-12 12:42:42.519959822 -0700
+++ linux-2.6.17-rc6-cl/Documentation/sysctl/vm.txt 2006-06-12 13:36:53.817719845 -0700
@@ -28,7 +28,6 @@ Currently, these files are in /proc/sys/
- block_dump
- drop-caches
- zone_reclaim_mode
-- zone_reclaim_interval
- panic_on_oom
- swap_prefetch
- readahead_ratio
@@ -172,18 +171,6 @@ in all nodes of the system.
================================================================
-zone_reclaim_interval:
-
-The time allowed for off node allocations after zone reclaim
-has failed to reclaim enough pages to allow a local allocation.
-
-Time is set in seconds and set by default to 30 seconds.
-
-Reduce the interval if undesired off node allocations occur. However, too
-frequent scans will have a negative impact onoff node allocation performance.
-
-=============================================================
-
panic_on_oom
This enables or disables panic on out-of-memory feature. If this is set to 1,
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 10/21] Conversion of nr_slab to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (8 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 09/21] zone_reclaim: remove /proc/sys/vm/zone_reclaim_interval Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 11/21] swap_prefetch: Conversion of nr_slab to ZVC Christoph Lameter
` (10 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_slab to per zone counter
From: Christoph Lameter <clameter@sgi.com>
- Allows reclaim to access counter without looping over processor counts.
- Allows accurate statistics on how many pages are used in a zone by
the slab. This may become useful to balance slab allocations over
various zones.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/i386/mm/pgtable.c 2006-06-12 12:57:23.381448638 -0700
+++ linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c 2006-06-12 13:02:16.916004785 -0700
@@ -62,7 +62,7 @@ void show_mem(void)
printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_MAPPED));
- printk(KERN_INFO "%lu pages slab\n", ps.nr_slab);
+ printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
}
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:01:30.625902765 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:02:16.916981287 -0700
@@ -54,8 +54,6 @@ static ssize_t node_read_meminfo(struct
ps.nr_dirty = 0;
if ((long)ps.nr_writeback < 0)
ps.nr_writeback = 0;
- if ((long)ps.nr_slab < 0)
- ps.nr_slab = 0;
n = sprintf(buf, "\n"
"Node %d MemTotal: %8lu kB\n"
@@ -87,7 +85,7 @@ static ssize_t node_read_meminfo(struct
nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(node_page_state(nid, NR_ANON)),
- nid, K(ps.nr_slab));
+ nid, K(node_page_state(nid, NR_SLAB)));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
}
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:01:30.621020255 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:02:16.917957789 -0700
@@ -194,7 +194,7 @@ static int meminfo_read_proc(char *page,
K(ps.nr_writeback),
K(global_page_state(NR_ANON)),
K(global_page_state(NR_MAPPED)),
- K(ps.nr_slab),
+ K(global_page_state(NR_SLAB)),
K(allowed),
K(committed),
K(ps.nr_page_table_pages),
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:02:14.488420750 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:02:16.918934291 -0700
@@ -51,6 +51,7 @@ enum zone_stat_item {
NR_MAPPED, /* pagecache pages mapped into pagetables.
only modified from process context */
NR_PAGECACHE,
+ NR_SLAB, /* Pages used by slab allocator */
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 13:00:50.584437193 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:02:16.920887295 -0700
@@ -1408,7 +1408,7 @@ void show_free_areas(void)
ps.nr_writeback,
ps.nr_unstable,
nr_free_pages(),
- ps.nr_slab,
+ global_page_state(NR_SLAB),
global_page_state(NR_MAPPED),
ps.nr_page_table_pages);
Index: linux-2.6.17-rc6-cl/mm/slab.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/slab.c 2006-06-12 12:54:33.242570771 -0700
+++ linux-2.6.17-rc6-cl/mm/slab.c 2006-06-12 13:02:16.923816801 -0700
@@ -1525,7 +1525,7 @@ static void *kmem_getpages(struct kmem_c
nr_pages = (1 << cachep->gfporder);
if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
atomic_add(nr_pages, &slab_reclaim_pages);
- add_page_state(nr_slab, nr_pages);
+ add_zone_page_state(page_zone(page), NR_SLAB, nr_pages);
for (i = 0; i < nr_pages; i++)
__SetPageSlab(page + i);
return page_address(page);
@@ -1545,7 +1545,7 @@ static void kmem_freepages(struct kmem_c
__ClearPageSlab(page);
page++;
}
- sub_page_state(nr_slab, nr_freed);
+ sub_zone_page_state(page_zone(page), NR_SLAB, nr_freed);
if (current->reclaim_state)
current->reclaim_state->reclaimed_slab += nr_freed;
free_pages((unsigned long)addr, cachep->gfporder);
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-12 13:02:14.493303260 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-12 13:02:16.924793303 -0700
@@ -1372,7 +1372,7 @@ unsigned long shrink_all_memory(unsigned
for_each_zone(zone)
lru_pages += zone->nr_active + zone->nr_inactive;
- nr_slab = read_page_state(nr_slab);
+ nr_slab = global_page_state(NR_SLAB);
/* If slab caches are huge, it's better to hit them first */
while (nr_slab >= lru_pages) {
reclaim_state.reclaimed_slab = 0;
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:02:08.289585892 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:02:26.602904877 -0700
@@ -466,13 +466,13 @@ static char *vmstat_text[] = {
"nr_mapped",
"nr_pagecache",
"nr_anon",
+ "nr_slab",
/* Page state */
"nr_dirty",
"nr_writeback",
"nr_unstable",
"nr_page_table_pages",
- "nr_slab",
"pgpgin",
"pgpgout",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 11/21] swap_prefetch: Conversion of nr_slab to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (9 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 10/21] Conversion of nr_slab to per zone counter Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 12/21] Conversion of nr_pagetables to per zone counter Christoph Lameter
` (9 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
swap_prefetch: Use NR_SLAB
This removes a potential problem for swap_prefetch. Use NR_SLAB
and remove the comment stating the problem.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12 11:30:59.995187077 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 11:55:06.458392224 -0700
@@ -389,14 +389,11 @@ static int prefetch_suitable(void)
/*
* >2/3 of the ram on this node is mapped, slab, swapcache or
* dirty, we need to leave some free for pagecache.
- * Note that currently nr_slab is innacurate on numa because
- * nr_slab is incremented on the node doing the accounting
- * even if the slab is being allocated on a remote node. This
- * would be expensive to fix and not of great significance.
*/
limit = node_page_state(node, NR_MAPPED) +
node_page_state(node, NR_ANON) +
- ps.nr_slab + ps.nr_dirty +
+ node_page_state(node, NR_SLAB) +
+ ps.nr_dirty +
ps.nr_unstable + total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 12/21] Conversion of nr_pagetables to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (10 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 11/21] swap_prefetch: Conversion of nr_slab to ZVC Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 13/21] Conversion of nr_dirty " Christoph Lameter
` (8 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_pagetable to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_page_table_pages to a per zone counter
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/i386/mm/pgtable.c 2006-06-12 13:02:16.916004785 -0700
+++ linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c 2006-06-12 13:02:33.254836674 -0700
@@ -63,7 +63,7 @@ void show_mem(void)
printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_MAPPED));
printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
- printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
+ printk(KERN_INFO "%lu pages pagetables\n", global_page_state(NR_PAGETABLE));
}
/*
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:02:16.917957789 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:02:33.256789678 -0700
@@ -197,7 +197,7 @@ static int meminfo_read_proc(char *page,
K(global_page_state(NR_SLAB)),
K(allowed),
K(committed),
- K(ps.nr_page_table_pages),
+ K(global_page_state(NR_PAGETABLE)),
(unsigned long)VMALLOC_TOTAL >> 10,
vmi.used >> 10,
vmi.largest_chunk >> 10
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:02:16.918934291 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:02:33.257766180 -0700
@@ -52,6 +52,7 @@ enum zone_stat_item {
only modified from process context */
NR_PAGECACHE,
NR_SLAB, /* Pages used by slab allocator */
+ NR_PAGETABLE, /* used for pagetables */
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/memory.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/memory.c 2006-06-12 12:42:52.030113494 -0700
+++ linux-2.6.17-rc6-cl/mm/memory.c 2006-06-12 13:02:33.259719184 -0700
@@ -127,7 +127,7 @@ static void free_pte_range(struct mmu_ga
pmd_clear(pmd);
pte_lock_deinit(page);
pte_free_tlb(tlb, page);
- dec_page_state(nr_page_table_pages);
+ dec_zone_page_state(page, NR_PAGETABLE);
tlb->mm->nr_ptes--;
}
@@ -312,7 +312,7 @@ int __pte_alloc(struct mm_struct *mm, pm
pte_free(new);
} else {
mm->nr_ptes++;
- inc_page_state(nr_page_table_pages);
+ inc_zone_page_state(new, NR_PAGETABLE);
pmd_populate(mm, pmd, new);
}
spin_unlock(&mm->page_table_lock);
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 13:02:16.920887295 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:02:33.260695686 -0700
@@ -1410,7 +1410,7 @@ void show_free_areas(void)
nr_free_pages(),
global_page_state(NR_SLAB),
global_page_state(NR_MAPPED),
- ps.nr_page_table_pages);
+ global_page_state(NR_PAGETABLE));
for_each_zone(zone) {
int i;
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:02:16.916981287 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:02:33.261672188 -0700
@@ -70,6 +70,7 @@ static ssize_t node_read_meminfo(struct
"Node %d PageCache: %8lu kB\n"
"Node %d Mapped: %8lu kB\n"
"Node %d Anonymous: %8lu kB\n"
+ "Node %d PageTables: %8lu kB\n"
"Node %d Slab: %8lu kB\n",
nid, K(i.totalram),
nid, K(i.freeram),
@@ -85,6 +86,7 @@ static ssize_t node_read_meminfo(struct
nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(node_page_state(nid, NR_ANON)),
+ nid, K(node_page_state(nid, NR_PAGETABLE)),
nid, K(node_page_state(nid, NR_SLAB)));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
Index: linux-2.6.17-rc6-cl/arch/um/kernel/skas/mmu.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/um/kernel/skas/mmu.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/arch/um/kernel/skas/mmu.c 2006-06-12 13:02:33.262648690 -0700
@@ -152,7 +152,7 @@ void destroy_context_skas(struct mm_stru
free_page(mmu->id.stack);
pte_lock_deinit(virt_to_page(mmu->last_page_table));
pte_free_kernel((pte_t *) mmu->last_page_table);
- dec_page_state(nr_page_table_pages);
+ dec_zone_page_state(virt_to_page(mmu->last_page_table), NR_PAGETABLE);
#ifdef CONFIG_3_LEVEL_PGTABLES
pmd_free((pmd_t *) mmu->last_pmd);
#endif
Index: linux-2.6.17-rc6-cl/arch/arm/mm/mm-armv.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/arm/mm/mm-armv.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/arch/arm/mm/mm-armv.c 2006-06-12 13:02:33.263625192 -0700
@@ -227,7 +227,7 @@ void free_pgd_slow(pgd_t *pgd)
pte = pmd_page(*pmd);
pmd_clear(pmd);
- dec_page_state(nr_page_table_pages);
+ dec_zone_page_state(virt_to_page((unsigned long *)pgd), NR_PAGETABLE);
pte_lock_deinit(pte);
pte_free(pte);
pmd_free(pmd);
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:02:26.602904877 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:02:46.391718420 -0700
@@ -467,12 +467,12 @@ static char *vmstat_text[] = {
"nr_pagecache",
"nr_anon",
"nr_slab",
+ "nr_page_table_pages",
/* Page state */
"nr_dirty",
"nr_writeback",
"nr_unstable",
- "nr_page_table_pages",
"pgpgin",
"pgpgout",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 13/21] Conversion of nr_dirty to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (11 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 12/21] Conversion of nr_pagetables to per zone counter Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 14/21] swap_prefetch: Conversion of nr_dirty to ZVC Christoph Lameter
` (7 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_dirty to per zone counter
From: Christoph Lameter <clameter@sgi.com>
This makes nr_dirty a per zone counter. Looping over all processors is
avoided during writeback state determination.
The counter aggregation for nr_dirty had to be undone in the NFS layer since
we summed up the page counts from multiple zones.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/i386/mm/pgtable.c 2006-06-12 13:02:33.254836674 -0700
+++ linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c 2006-06-12 13:02:52.201905469 -0700
@@ -59,7 +59,7 @@ void show_mem(void)
printk(KERN_INFO "%d pages swap cached\n", cached);
get_page_state(&ps);
- printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
+ printk(KERN_INFO "%lu pages dirty\n", global_page_state(NR_DIRTY));
printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_MAPPED));
printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:02:33.261672188 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:02:52.202881971 -0700
@@ -50,8 +50,6 @@ static ssize_t node_read_meminfo(struct
__get_zone_counts(&active, &inactive, &free, NODE_DATA(nid));
/* Check for negative values in these approximate counters */
- if ((long)ps.nr_dirty < 0)
- ps.nr_dirty = 0;
if ((long)ps.nr_writeback < 0)
ps.nr_writeback = 0;
@@ -81,7 +79,7 @@ static ssize_t node_read_meminfo(struct
nid, K(i.freehigh),
nid, K(i.totalram - i.totalhigh),
nid, K(i.freeram - i.freehigh),
- nid, K(ps.nr_dirty),
+ nid, K(node_page_state(nid, NR_DIRTY)),
nid, K(ps.nr_writeback),
nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
Index: linux-2.6.17-rc6-cl/fs/buffer.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/buffer.c 2006-06-12 12:42:47.348762564 -0700
+++ linux-2.6.17-rc6-cl/fs/buffer.c 2006-06-12 13:02:52.204834975 -0700
@@ -854,7 +854,7 @@ int __set_page_dirty_buffers(struct page
write_lock_irq(&mapping->tree_lock);
if (page->mapping) { /* Race with truncate? */
if (mapping_cap_account_dirty(mapping))
- inc_page_state(nr_dirty);
+ __inc_zone_page_state(page, NR_DIRTY);
radix_tree_tag_set(&mapping->page_tree,
page_index(page),
PAGECACHE_TAG_DIRTY);
Index: linux-2.6.17-rc6-cl/fs/fs-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/fs-writeback.c 2006-06-12 12:42:47.502073390 -0700
+++ linux-2.6.17-rc6-cl/fs/fs-writeback.c 2006-06-12 13:02:52.204834975 -0700
@@ -472,7 +472,7 @@ void sync_inodes_sb(struct super_block *
.range_start = 0,
.range_end = LLONG_MAX,
};
- unsigned long nr_dirty = read_page_state(nr_dirty);
+ unsigned long nr_dirty = global_page_state(NR_DIRTY);
unsigned long nr_unstable = read_page_state(nr_unstable);
wbc.nr_to_write = nr_dirty + nr_unstable +
Index: linux-2.6.17-rc6-cl/fs/nfs/pagelist.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/nfs/pagelist.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/fs/nfs/pagelist.c 2006-06-12 13:02:52.205811477 -0700
@@ -315,6 +315,7 @@ nfs_scan_lock_dirty(struct nfs_inode *nf
req->wb_index, NFS_PAGE_TAG_DIRTY);
nfs_list_remove_request(req);
nfs_list_add_request(req, dst);
+ inc_zone_page_state(req->wb_page, NR_DIRTY);
res++;
}
}
Index: linux-2.6.17-rc6-cl/fs/nfs/write.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/nfs/write.c 2006-06-12 12:42:48.111410682 -0700
+++ linux-2.6.17-rc6-cl/fs/nfs/write.c 2006-06-12 13:02:52.206787979 -0700
@@ -497,7 +497,7 @@ nfs_mark_request_dirty(struct nfs_page *
nfs_list_add_request(req, &nfsi->dirty);
nfsi->ndirty++;
spin_unlock(&nfsi->req_lock);
- inc_page_state(nr_dirty);
+ inc_zone_page_state(req->wb_page, NR_DIRTY);
mark_inode_dirty(inode);
}
@@ -598,7 +598,6 @@ nfs_scan_dirty(struct inode *inode, stru
if (nfsi->ndirty != 0) {
res = nfs_scan_lock_dirty(nfsi, dst, idx_start, npages);
nfsi->ndirty -= res;
- sub_page_state(nr_dirty,res);
if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty))
printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n");
}
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:02:33.256789678 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:02:52.207764481 -0700
@@ -190,7 +190,7 @@ static int meminfo_read_proc(char *page,
K(i.freeram-i.freehigh),
K(i.totalswap),
K(i.freeswap),
- K(ps.nr_dirty),
+ K(global_page_state(NR_DIRTY)),
K(ps.nr_writeback),
K(global_page_state(NR_ANON)),
K(global_page_state(NR_MAPPED)),
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:02:33.257766180 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:02:52.209717485 -0700
@@ -53,6 +53,7 @@ enum zone_stat_item {
NR_PAGECACHE,
NR_SLAB, /* Pages used by slab allocator */
NR_PAGETABLE, /* used for pagetables */
+ NR_DIRTY,
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 13:02:33.260695686 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:02:52.210693987 -0700
@@ -1404,7 +1404,7 @@ void show_free_areas(void)
"unstable:%lu free:%u slab:%lu mapped:%lu pagetables:%lu\n",
active,
inactive,
- ps.nr_dirty,
+ global_page_state(NR_DIRTY),
ps.nr_writeback,
ps.nr_unstable,
nr_free_pages(),
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-12 13:01:30.625902765 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-12 13:02:52.211670489 -0700
@@ -109,7 +109,7 @@ struct writeback_state
static void get_writeback_state(struct writeback_state *wbs)
{
- wbs->nr_dirty = read_page_state(nr_dirty);
+ wbs->nr_dirty = global_page_state(NR_DIRTY);
wbs->nr_unstable = read_page_state(nr_unstable);
wbs->nr_mapped = global_page_state(NR_MAPPED) +
global_page_state(NR_ANON);
@@ -641,7 +641,7 @@ int __set_page_dirty_nobuffers(struct pa
if (mapping2) { /* Race with truncate? */
BUG_ON(mapping2 != mapping);
if (mapping_cap_account_dirty(mapping))
- inc_page_state(nr_dirty);
+ __inc_zone_page_state(page, NR_DIRTY);
radix_tree_tag_set(&mapping->page_tree,
page_index(page), PAGECACHE_TAG_DIRTY);
}
@@ -728,9 +728,9 @@ int test_clear_page_dirty(struct page *p
radix_tree_tag_clear(&mapping->page_tree,
page_index(page),
PAGECACHE_TAG_DIRTY);
- write_unlock_irqrestore(&mapping->tree_lock, flags);
if (mapping_cap_account_dirty(mapping))
- dec_page_state(nr_dirty);
+ __dec_zone_page_state(page, NR_DIRTY);
+ write_unlock_irqrestore(&mapping->tree_lock, flags);
return 1;
}
write_unlock_irqrestore(&mapping->tree_lock, flags);
@@ -761,7 +761,7 @@ int clear_page_dirty_for_io(struct page
if (mapping) {
if (TestClearPageDirty(page)) {
if (mapping_cap_account_dirty(mapping))
- dec_page_state(nr_dirty);
+ dec_zone_page_state(page, NR_DIRTY);
return 1;
}
return 0;
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:02:46.391718420 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:02:58.922192406 -0700
@@ -468,9 +468,9 @@ static char *vmstat_text[] = {
"nr_anon",
"nr_slab",
"nr_page_table_pages",
+ "nr_dirty",
/* Page state */
- "nr_dirty",
"nr_writeback",
"nr_unstable",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 14/21] swap_prefetch: Conversion of nr_dirty to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (12 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 13/21] Conversion of nr_dirty " Christoph Lameter
@ 2006-06-12 21:13 ` Christoph Lameter
2006-06-12 23:38 ` Con Kolivas
2006-06-12 21:14 ` [PATCH 15/21] reiser4: Conversiion " Christoph Lameter
` (6 subsequent siblings)
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:13 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
swap_prefetch: Convert nr_dirty to zone count
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12 12:55:29.984174144 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 12:55:33.862840200 -0700
@@ -393,7 +393,7 @@ static int prefetch_suitable(void)
limit = node_page_state(node, NR_MAPPED) +
node_page_state(node, NR_ANON) +
node_page_state(node, NR_SLAB) +
- ps.nr_dirty +
+ node_page_state(node, NR_DIRTY) +
ps.nr_unstable + total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 15/21] reiser4: Conversiion of nr_dirty to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (13 preceding siblings ...)
2006-06-12 21:13 ` [PATCH 14/21] swap_prefetch: Conversion of nr_dirty to ZVC Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 21:14 ` [PATCH 16/21] Conversion of nr_writeback to per zone counter Christoph Lameter
` (5 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: reiser4: conversion of nr_dirty to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_dirty to a per zone counter
Requested by Andrew.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/fs/reiser4/as_ops.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/reiser4/as_ops.c 2006-06-10 11:44:14.557965359 -0700
+++ linux-2.6.17-rc6-cl/fs/reiser4/as_ops.c 2006-06-10 15:05:55.473509770 -0700
@@ -83,7 +83,7 @@ int reiser4_set_page_dirty(struct page *
if (page->mapping) {
assert("vs-1652", page->mapping == mapping);
if (mapping_cap_account_dirty(mapping))
- inc_page_state(nr_dirty);
+ __inc_zone_page_state(page, NR_DIRTY);
radix_tree_tag_set(&mapping->page_tree,
page->index,
PAGECACHE_TAG_REISER4_MOVED);
Index: linux-2.6.17-rc6-cl/fs/reiser4/page_cache.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/reiser4/page_cache.c 2006-06-10 11:44:14.641944528 -0700
+++ linux-2.6.17-rc6-cl/fs/reiser4/page_cache.c 2006-06-10 15:05:17.209277244 -0700
@@ -464,7 +464,7 @@ int set_page_dirty_internal(struct page
if (!TestSetPageDirty(page)) {
if (mapping_cap_account_dirty(mapping))
- inc_page_state(nr_dirty);
+ inc_zone_page_state(page, NR_DIRTY);
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
}
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 16/21] Conversion of nr_writeback to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (14 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 15/21] reiser4: Conversiion " Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 21:14 ` [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC Christoph Lameter
` (4 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_writeback to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_writeback to per zone counter.
This removes the last page_state counter from arch/i386/mm/pgtable.c
so we drop the page_state from there.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/arch/i386/mm/pgtable.c 2006-06-12 13:02:52.201905469 -0700
+++ linux-2.6.17-rc6-cl/arch/i386/mm/pgtable.c 2006-06-12 13:03:04.914008833 -0700
@@ -30,7 +30,6 @@ void show_mem(void)
struct page *page;
pg_data_t *pgdat;
unsigned long i;
- struct page_state ps;
unsigned long flags;
printk(KERN_INFO "Mem-info:\n");
@@ -58,9 +57,8 @@ void show_mem(void)
printk(KERN_INFO "%d pages shared\n", shared);
printk(KERN_INFO "%d pages swap cached\n", cached);
- get_page_state(&ps);
printk(KERN_INFO "%lu pages dirty\n", global_page_state(NR_DIRTY));
- printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
+ printk(KERN_INFO "%lu pages writeback\n", global_page_state(NR_WRITEBACK));
printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_MAPPED));
printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
printk(KERN_INFO "%lu pages pagetables\n", global_page_state(NR_PAGETABLE));
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:02:52.202881971 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:03:04.914985335 -0700
@@ -49,9 +49,6 @@ static ssize_t node_read_meminfo(struct
get_page_state_node(&ps, nid);
__get_zone_counts(&active, &inactive, &free, NODE_DATA(nid));
- /* Check for negative values in these approximate counters */
- if ((long)ps.nr_writeback < 0)
- ps.nr_writeback = 0;
n = sprintf(buf, "\n"
"Node %d MemTotal: %8lu kB\n"
@@ -80,7 +77,7 @@ static ssize_t node_read_meminfo(struct
nid, K(i.totalram - i.totalhigh),
nid, K(i.freeram - i.freehigh),
nid, K(node_page_state(nid, NR_DIRTY)),
- nid, K(ps.nr_writeback),
+ nid, K(node_page_state(nid, NR_WRITEBACK)),
nid, K(node_page_state(nid, NR_PAGECACHE)),
nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(node_page_state(nid, NR_ANON)),
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:02:52.207764481 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:03:04.915961837 -0700
@@ -191,7 +191,7 @@ static int meminfo_read_proc(char *page,
K(i.totalswap),
K(i.freeswap),
K(global_page_state(NR_DIRTY)),
- K(ps.nr_writeback),
+ K(global_page_state(NR_WRITEBACK)),
K(global_page_state(NR_ANON)),
K(global_page_state(NR_MAPPED)),
K(global_page_state(NR_SLAB)),
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:02:52.209717485 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:03:04.916938339 -0700
@@ -54,6 +54,7 @@ enum zone_stat_item {
NR_SLAB, /* Pages used by slab allocator */
NR_PAGETABLE, /* used for pagetables */
NR_DIRTY,
+ NR_WRITEBACK,
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/include/linux/page-flags.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/page-flags.h 2006-06-12 12:43:12.081607020 -0700
+++ linux-2.6.17-rc6-cl/include/linux/page-flags.h 2006-06-12 13:03:04.917914841 -0700
@@ -176,7 +176,7 @@
do { \
if (!test_and_set_bit(PG_writeback, \
&(page)->flags)) \
- inc_page_state(nr_writeback); \
+ inc_zone_page_state(page, NR_WRITEBACK); \
} while (0)
#define TestSetPageWriteback(page) \
({ \
@@ -184,14 +184,14 @@
ret = test_and_set_bit(PG_writeback, \
&(page)->flags); \
if (!ret) \
- inc_page_state(nr_writeback); \
+ inc_zone_page_state(page, NR_WRITEBACK); \
ret; \
})
#define ClearPageWriteback(page) \
do { \
if (test_and_clear_bit(PG_writeback, \
&(page)->flags)) \
- dec_page_state(nr_writeback); \
+ dec_zone_page_state(page, NR_WRITEBACK); \
} while (0)
#define TestClearPageWriteback(page) \
({ \
@@ -199,7 +199,7 @@
ret = test_and_clear_bit(PG_writeback, \
&(page)->flags); \
if (ret) \
- dec_page_state(nr_writeback); \
+ dec_zone_page_state(page, NR_WRITEBACK); \
ret; \
})
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 13:02:52.210693987 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:03:04.919867845 -0700
@@ -1405,7 +1405,7 @@ void show_free_areas(void)
active,
inactive,
global_page_state(NR_DIRTY),
- ps.nr_writeback,
+ global_page_state(NR_WRITEBACK),
ps.nr_unstable,
nr_free_pages(),
global_page_state(NR_SLAB),
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-12 13:02:52.211670489 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-12 13:03:04.919867845 -0700
@@ -113,7 +113,7 @@ static void get_writeback_state(struct w
wbs->nr_unstable = read_page_state(nr_unstable);
wbs->nr_mapped = global_page_state(NR_MAPPED) +
global_page_state(NR_ANON);
- wbs->nr_writeback = read_page_state(nr_writeback);
+ wbs->nr_writeback = global_page_state(NR_WRITEBACK);
}
/*
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:02:58.922192406 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:03:12.036614604 -0700
@@ -469,9 +469,9 @@ static char *vmstat_text[] = {
"nr_slab",
"nr_page_table_pages",
"nr_dirty",
+ "nr_writeback",
/* Page state */
- "nr_writeback",
"nr_unstable",
"pgpgin",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (15 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 16/21] Conversion of nr_writeback to per zone counter Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 23:38 ` Con Kolivas
2006-06-12 21:14 ` [PATCH 18/21] Conversion of nr_unstable to per zone counter Christoph Lameter
` (3 subsequent siblings)
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: swap_prefetch: conversion of nr_writeback to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_writeback to per zone counter
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-10 15:09:12.691768066 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-10 15:18:17.095560797 -0700
@@ -381,7 +381,7 @@ static int prefetch_suitable(void)
get_page_state_node(&ps, node);
/* We shouldn't prefetch when we are doing writeback */
- if (ps.nr_writeback) {
+ if (node_page_state(node, NR_WRITEBACK)) {
node_clear(node, sp_stat.prefetch_nodes);
continue;
}
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC
2006-06-12 21:14 ` [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC Christoph Lameter
@ 2006-06-12 23:38 ` Con Kolivas
0 siblings, 0 replies; 39+ messages in thread
From: Con Kolivas @ 2006-06-12 23:38 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tuesday 13 June 2006 07:14, Christoph Lameter wrote:
> Subject: swap_prefetch: conversion of nr_writeback to per zone counter
> From: Christoph Lameter <clameter@sgi.com>
>
> Conversion of nr_writeback to per zone counter
>
> Signed-off-by: Christoph Lameter <clameter@sgi.com>
> Signed-off-by: Andrew Morton <akpm@osdl.org>
Acked-by: Con Kolivas <kernel@kolivas.org>
--
-ck
^ permalink raw reply [flat|nested] 39+ messages in thread
* [PATCH 18/21] Conversion of nr_unstable to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (16 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 21:14 ` [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC Christoph Lameter
` (2 subsequent siblings)
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_unstable to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_unstable to a per zone counter
This converts the last critical page state of the VM and therefore
we need to remove several functions that were depending on
GET_PAGE_STATE_LAST in order to make the kernel compile again.
We are only left with event type counters in page state.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:03:04.914985335 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:03:17.226722868 -0700
@@ -40,13 +40,11 @@ static ssize_t node_read_meminfo(struct
int n;
int nid = dev->id;
struct sysinfo i;
- struct page_state ps;
unsigned long inactive;
unsigned long active;
unsigned long free;
si_meminfo_node(&i, nid);
- get_page_state_node(&ps, nid);
__get_zone_counts(&active, &inactive, &free, NODE_DATA(nid));
@@ -66,6 +64,7 @@ static ssize_t node_read_meminfo(struct
"Node %d Mapped: %8lu kB\n"
"Node %d Anonymous: %8lu kB\n"
"Node %d PageTables: %8lu kB\n"
+ "Node %d Unstable: %8lu kB\n"
"Node %d Slab: %8lu kB\n",
nid, K(i.totalram),
nid, K(i.freeram),
@@ -82,6 +81,7 @@ static ssize_t node_read_meminfo(struct
nid, K(node_page_state(nid, NR_MAPPED)),
nid, K(node_page_state(nid, NR_ANON)),
nid, K(node_page_state(nid, NR_PAGETABLE)),
+ nid, K(node_page_state(nid, NR_UNSTABLE)),
nid, K(node_page_state(nid, NR_SLAB)));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
Index: linux-2.6.17-rc6-cl/fs/fs-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/fs-writeback.c 2006-06-12 13:02:52.204834975 -0700
+++ linux-2.6.17-rc6-cl/fs/fs-writeback.c 2006-06-12 13:03:17.227699370 -0700
@@ -473,7 +473,7 @@ void sync_inodes_sb(struct super_block *
.range_end = LLONG_MAX,
};
unsigned long nr_dirty = global_page_state(NR_DIRTY);
- unsigned long nr_unstable = read_page_state(nr_unstable);
+ unsigned long nr_unstable = global_page_state(NR_UNSTABLE);
wbc.nr_to_write = nr_dirty + nr_unstable +
(inodes_stat.nr_inodes - inodes_stat.nr_unused) +
Index: linux-2.6.17-rc6-cl/fs/nfs/write.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/nfs/write.c 2006-06-12 13:02:52.206787979 -0700
+++ linux-2.6.17-rc6-cl/fs/nfs/write.c 2006-06-12 13:03:17.228675872 -0700
@@ -525,7 +525,7 @@ nfs_mark_request_commit(struct nfs_page
nfs_list_add_request(req, &nfsi->commit);
nfsi->ncommit++;
spin_unlock(&nfsi->req_lock);
- inc_page_state(nr_unstable);
+ inc_zone_page_state(req->wb_page, NR_UNSTABLE);
mark_inode_dirty(inode);
}
#endif
@@ -1382,7 +1382,6 @@ static void nfs_commit_done(struct rpc_t
{
struct nfs_write_data *data = calldata;
struct nfs_page *req;
- int res = 0;
dprintk("NFS: %4d nfs_commit_done (status %d)\n",
task->tk_pid, task->tk_status);
@@ -1420,9 +1419,8 @@ static void nfs_commit_done(struct rpc_t
nfs_mark_request_dirty(req);
next:
nfs_clear_page_writeback(req);
- res++;
+ dec_zone_page_state(req->wb_page, NR_UNSTABLE);
}
- sub_page_state(nr_unstable,res);
}
static const struct rpc_call_ops nfs_commit_ops = {
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:03:04.916938339 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:03:17.229652374 -0700
@@ -55,6 +55,7 @@ enum zone_stat_item {
NR_PAGETABLE, /* used for pagetables */
NR_DIRTY,
NR_WRITEBACK,
+ NR_UNSTABLE, /* NFS unstable pages */
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/page_alloc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page_alloc.c 2006-06-12 13:03:04.919867845 -0700
+++ linux-2.6.17-rc6-cl/mm/page_alloc.c 2006-06-12 13:03:17.231605378 -0700
@@ -1361,7 +1361,6 @@ void si_meminfo_node(struct sysinfo *val
*/
void show_free_areas(void)
{
- struct page_state ps;
int cpu, temperature;
unsigned long active;
unsigned long inactive;
@@ -1393,7 +1392,6 @@ void show_free_areas(void)
}
}
- get_page_state(&ps);
get_zone_counts(&active, &inactive, &free);
printk("Free pages: %11ukB (%ukB HighMem)\n",
@@ -1406,7 +1404,7 @@ void show_free_areas(void)
inactive,
global_page_state(NR_DIRTY),
global_page_state(NR_WRITEBACK),
- ps.nr_unstable,
+ global_page_state(NR_UNSTABLE),
nr_free_pages(),
global_page_state(NR_SLAB),
global_page_state(NR_MAPPED),
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-12 13:03:04.919867845 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-12 13:03:17.232581880 -0700
@@ -110,7 +110,7 @@ struct writeback_state
static void get_writeback_state(struct writeback_state *wbs)
{
wbs->nr_dirty = global_page_state(NR_DIRTY);
- wbs->nr_unstable = read_page_state(nr_unstable);
+ wbs->nr_unstable = global_page_state(NR_UNSTABLE);
wbs->nr_mapped = global_page_state(NR_MAPPED) +
global_page_state(NR_ANON);
wbs->nr_writeback = global_page_state(NR_WRITEBACK);
Index: linux-2.6.17-rc6-cl/fs/proc/proc_misc.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/fs/proc/proc_misc.c 2006-06-12 13:03:04.915961837 -0700
+++ linux-2.6.17-rc6-cl/fs/proc/proc_misc.c 2006-06-12 13:03:17.233558382 -0700
@@ -120,7 +120,6 @@ static int meminfo_read_proc(char *page,
{
struct sysinfo i;
int len;
- struct page_state ps;
unsigned long inactive;
unsigned long active;
unsigned long free;
@@ -129,7 +128,6 @@ static int meminfo_read_proc(char *page,
struct vmalloc_info vmi;
long cached;
- get_page_state(&ps);
get_zone_counts(&active, &inactive, &free);
/*
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:03:12.036614604 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:03:33.967873583 -0700
@@ -470,10 +470,9 @@ static char *vmstat_text[] = {
"nr_page_table_pages",
"nr_dirty",
"nr_writeback",
-
- /* Page state */
"nr_unstable",
+ /* Event counters */
"pgpgin",
"pgpgout",
"pswpin",
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (17 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 18/21] Conversion of nr_unstable to per zone counter Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 23:40 ` Con Kolivas
2006-06-12 21:14 ` [PATCH 20/21] Conversion of nr_bounce to per zone counter Christoph Lameter
2006-06-12 21:14 ` [PATCH 21/21] Remove useless struct wbs Christoph Lameter
20 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: swap_prefetch: conversion of nr_unstable to per zone counter
From: Christoph Lameter <clameter@sgi.com>
The determination of the vm state is now not that expensive
anymore after we remove the use of the page state.
Remove the logic to avoid the expensive checks.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12 12:55:39.368358636 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 12:55:41.321362692 -0700
@@ -298,7 +298,7 @@ static int prefetch_suitable(void)
{
unsigned long limit;
struct zone *z;
- int node, ret = 0, test_pagestate = 0;
+ int node, ret = 0;
/* Purposefully racy */
if (test_bit(0, &swapped.busy)) {
@@ -306,19 +306,6 @@ static int prefetch_suitable(void)
goto out;
}
- /*
- * get_page_state and above_background_load are expensive so we only
- * perform them every SWAP_CLUSTER_MAX prefetched_pages.
- * We test to see if we're above_background_load as disk activity
- * even at low priority can cause interrupt induced scheduling
- * latencies.
- */
- if (!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX)) {
- if (above_background_load())
- goto out;
- test_pagestate = 1;
- }
-
clear_current_prefetch_free();
/*
@@ -357,7 +344,6 @@ static int prefetch_suitable(void)
*/
for_each_node_mask(node, sp_stat.prefetch_nodes) {
struct node_stats *ns = &sp_stat.node[node];
- struct page_state ps;
/*
* We check to see that pages are not being allocated
@@ -375,11 +361,6 @@ static int prefetch_suitable(void)
} else
ns->last_free = ns->current_free;
- if (!test_pagestate)
- continue;
-
- get_page_state_node(&ps, node);
-
/* We shouldn't prefetch when we are doing writeback */
if (node_page_state(node, NR_WRITEBACK)) {
node_clear(node, sp_stat.prefetch_nodes);
@@ -394,7 +375,8 @@ static int prefetch_suitable(void)
node_page_state(node, NR_ANON) +
node_page_state(node, NR_SLAB) +
node_page_state(node, NR_DIRTY) +
- ps.nr_unstable + total_swapcache_pages;
+ node_page_state(node, NR_UNSTABLE) +
+ total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
continue;
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 21:14 ` [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC Christoph Lameter
@ 2006-06-12 23:40 ` Con Kolivas
2006-06-12 23:48 ` Christoph Lameter
0 siblings, 1 reply; 39+ messages in thread
From: Con Kolivas @ 2006-06-12 23:40 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tuesday 13 June 2006 07:14, Christoph Lameter wrote:
> Subject: swap_prefetch: conversion of nr_unstable to per zone counter
> From: Christoph Lameter <clameter@sgi.com>
>
> The determination of the vm state is now not that expensive
> anymore after we remove the use of the page state.
> Remove the logic to avoid the expensive checks.
>
> Signed-off-by: Christoph Lameter <clameter@sgi.com>
> Signed-off-by: Andrew Morton <akpm@osdl.org>
Nack. You're changing some other code unintentionally.
>
> Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
> ===================================================================
> --- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12
> 12:55:39.368358636 -0700 +++
> linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 12:55:41.321362692 -0700
> @@ -298,7 +298,7 @@ static int prefetch_suitable(void)
> {
> unsigned long limit;
> struct zone *z;
> - int node, ret = 0, test_pagestate = 0;
> + int node, ret = 0;
>
> /* Purposefully racy */
> if (test_bit(0, &swapped.busy)) {
> @@ -306,19 +306,6 @@ static int prefetch_suitable(void)
> goto out;
> }
>
> - /*
> - * get_page_state and above_background_load are expensive so we only
> - * perform them every SWAP_CLUSTER_MAX prefetched_pages.
> - * We test to see if we're above_background_load as disk activity
> - * even at low priority can cause interrupt induced scheduling
> - * latencies.
> - */
> - if (!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX)) {
> - if (above_background_load())
> - goto out;
We still need the above_background_load check, just remove the test_pagestate
check and it can be checked every time. Drop the comment about get_page_state
only.
> - test_pagestate = 1;
> - }
> -
> clear_current_prefetch_free();
>
> /*
> @@ -357,7 +344,6 @@ static int prefetch_suitable(void)
> */
> for_each_node_mask(node, sp_stat.prefetch_nodes) {
> struct node_stats *ns = &sp_stat.node[node];
> - struct page_state ps;
>
> /*
> * We check to see that pages are not being allocated
> @@ -375,11 +361,6 @@ static int prefetch_suitable(void)
> } else
> ns->last_free = ns->current_free;
>
> - if (!test_pagestate)
> - continue;
> -
> - get_page_state_node(&ps, node);
> -
> /* We shouldn't prefetch when we are doing writeback */
> if (node_page_state(node, NR_WRITEBACK)) {
> node_clear(node, sp_stat.prefetch_nodes);
> @@ -394,7 +375,8 @@ static int prefetch_suitable(void)
> node_page_state(node, NR_ANON) +
> node_page_state(node, NR_SLAB) +
> node_page_state(node, NR_DIRTY) +
> - ps.nr_unstable + total_swapcache_pages;
> + node_page_state(node, NR_UNSTABLE) +
> + total_swapcache_pages;
> if (limit > ns->prefetch_watermark) {
> node_clear(node, sp_stat.prefetch_nodes);
> continue;
--
-ck
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 23:40 ` Con Kolivas
@ 2006-06-12 23:48 ` Christoph Lameter
2006-06-12 23:57 ` Con Kolivas
0 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 23:48 UTC (permalink / raw)
To: Con Kolivas
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tue, 13 Jun 2006, Con Kolivas wrote:
> Nack. You're changing some other code unintentionally.
Is this okay?
Subject: swap_prefetch: conversion of nr_unstable to per zone counter
From: Christoph Lameter <clameter@sgi.com>
The determination of the vm state is now not that expensive
anymore after we remove the use of the page state.
Remove the logic to avoid the expensive checks.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12 13:37:47.283159568 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 16:46:48.504626417 -0700
@@ -357,7 +357,6 @@ static int prefetch_suitable(void)
*/
for_each_node_mask(node, sp_stat.prefetch_nodes) {
struct node_stats *ns = &sp_stat.node[node];
- struct page_state ps;
/*
* We check to see that pages are not being allocated
@@ -375,11 +374,6 @@ static int prefetch_suitable(void)
} else
ns->last_free = ns->current_free;
- if (!test_pagestate)
- continue;
-
- get_page_state_node(&ps, node);
-
/* We shouldn't prefetch when we are doing writeback */
if (node_page_state(node, NR_WRITEBACK)) {
node_clear(node, sp_stat.prefetch_nodes);
@@ -394,7 +388,8 @@ static int prefetch_suitable(void)
node_page_state(node, NR_ANON) +
node_page_state(node, NR_SLAB) +
node_page_state(node, NR_DIRTY) +
- ps.nr_unstable + total_swapcache_pages;
+ node_page_state(node, NR_UNSTABLE) +
+ total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
continue;
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 23:48 ` Christoph Lameter
@ 2006-06-12 23:57 ` Con Kolivas
2006-06-12 23:59 ` Con Kolivas
0 siblings, 1 reply; 39+ messages in thread
From: Con Kolivas @ 2006-06-12 23:57 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tuesday 13 June 2006 09:48, Christoph Lameter wrote:
> On Tue, 13 Jun 2006, Con Kolivas wrote:
> > Nack. You're changing some other code unintentionally.
>
> Is this okay?
Not quite.
Remove the test_pagestate variable entirely and leave the check something
like:
if (!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX) &&
above_background_load())
goto out;
Thanks
>
> Subject: swap_prefetch: conversion of nr_unstable to per zone counter
> From: Christoph Lameter <clameter@sgi.com>
>
> The determination of the vm state is now not that expensive
> anymore after we remove the use of the page state.
> Remove the logic to avoid the expensive checks.
>
> Signed-off-by: Christoph Lameter <clameter@sgi.com>
> Signed-off-by: Andrew Morton <akpm@osdl.org>
>
> Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
> ===================================================================
> --- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12
> 13:37:47.283159568 -0700 +++
> linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 16:46:48.504626417 -0700
> @@ -357,7 +357,6 @@ static int prefetch_suitable(void)
> */
> for_each_node_mask(node, sp_stat.prefetch_nodes) {
> struct node_stats *ns = &sp_stat.node[node];
> - struct page_state ps;
>
> /*
> * We check to see that pages are not being allocated
> @@ -375,11 +374,6 @@ static int prefetch_suitable(void)
> } else
> ns->last_free = ns->current_free;
>
> - if (!test_pagestate)
> - continue;
> -
> - get_page_state_node(&ps, node);
> -
> /* We shouldn't prefetch when we are doing writeback */
> if (node_page_state(node, NR_WRITEBACK)) {
> node_clear(node, sp_stat.prefetch_nodes);
> @@ -394,7 +388,8 @@ static int prefetch_suitable(void)
> node_page_state(node, NR_ANON) +
> node_page_state(node, NR_SLAB) +
> node_page_state(node, NR_DIRTY) +
> - ps.nr_unstable + total_swapcache_pages;
> + node_page_state(node, NR_UNSTABLE) +
> + total_swapcache_pages;
> if (limit > ns->prefetch_watermark) {
> node_clear(node, sp_stat.prefetch_nodes);
> continue;
--
-ck
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 23:57 ` Con Kolivas
@ 2006-06-12 23:59 ` Con Kolivas
2006-06-13 0:08 ` Christoph Lameter
0 siblings, 1 reply; 39+ messages in thread
From: Con Kolivas @ 2006-06-12 23:59 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tuesday 13 June 2006 09:57, Con Kolivas wrote:
> On Tuesday 13 June 2006 09:48, Christoph Lameter wrote:
> > On Tue, 13 Jun 2006, Con Kolivas wrote:
> > > Nack. You're changing some other code unintentionally.
> >
> > Is this okay?
>
> Not quite.
>
> Remove the test_pagestate variable entirely and leave the check something
> like:
>
> if (!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX) &&
> above_background_load())
> goto out;
>
> Thanks
Oh and
The comment should read something like:
/*
* above_background_load() is expensive so we only perform
* it every SWAP_CLUSTER_MAX prefetched_pages.
* We test to see if we're above_background_load as disk activity
* even at low priority can cause interrupt induced scheduling
* latencies.
*/
--
-ck
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-12 23:59 ` Con Kolivas
@ 2006-06-13 0:08 ` Christoph Lameter
2006-06-13 0:19 ` Con Kolivas
0 siblings, 1 reply; 39+ messages in thread
From: Christoph Lameter @ 2006-06-13 0:08 UTC (permalink / raw)
To: Con Kolivas
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tue, 13 Jun 2006, Con Kolivas wrote:
> The comment should read something like:
If we need another round then maybe it would be best if you would do that
patch.
This?
Subject: swap_prefetch: conversion of nr_unstable to per zone counter
From: Christoph Lameter <clameter@sgi.com>
The determination of the vm state is now not that expensive
anymore after we remove the use of the page state.
Change the logic to avoid the expensive checks.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Con Kolivas <kernel@kolivas.org>
Index: linux-2.6.17-rc6-cl/mm/swap_prefetch.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/swap_prefetch.c 2006-06-12 13:37:47.283159568 -0700
+++ linux-2.6.17-rc6-cl/mm/swap_prefetch.c 2006-06-12 17:06:44.875945042 -0700
@@ -298,7 +298,7 @@ static int prefetch_suitable(void)
{
unsigned long limit;
struct zone *z;
- int node, ret = 0, test_pagestate = 0;
+ int node, ret = 0;
/* Purposefully racy */
if (test_bit(0, &swapped.busy)) {
@@ -307,17 +307,15 @@ static int prefetch_suitable(void)
}
/*
- * get_page_state and above_background_load are expensive so we only
- * perform them every SWAP_CLUSTER_MAX prefetched_pages.
+ * above_background_load() is expensive so we only perform
+ * it every SWAP_CLUSTER_MAX prefetched_pages.
* We test to see if we're above_background_load as disk activity
* even at low priority can cause interrupt induced scheduling
* latencies.
*/
- if (!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX)) {
- if (above_background_load())
+ if ((!(sp_stat.prefetched_pages % SWAP_CLUSTER_MAX)) &&
+ above_background_load())
goto out;
- test_pagestate = 1;
- }
clear_current_prefetch_free();
@@ -357,7 +355,6 @@ static int prefetch_suitable(void)
*/
for_each_node_mask(node, sp_stat.prefetch_nodes) {
struct node_stats *ns = &sp_stat.node[node];
- struct page_state ps;
/*
* We check to see that pages are not being allocated
@@ -375,11 +372,6 @@ static int prefetch_suitable(void)
} else
ns->last_free = ns->current_free;
- if (!test_pagestate)
- continue;
-
- get_page_state_node(&ps, node);
-
/* We shouldn't prefetch when we are doing writeback */
if (node_page_state(node, NR_WRITEBACK)) {
node_clear(node, sp_stat.prefetch_nodes);
@@ -394,7 +386,8 @@ static int prefetch_suitable(void)
node_page_state(node, NR_ANON) +
node_page_state(node, NR_SLAB) +
node_page_state(node, NR_DIRTY) +
- ps.nr_unstable + total_swapcache_pages;
+ node_page_state(node, NR_UNSTABLE) +
+ total_swapcache_pages;
if (limit > ns->prefetch_watermark) {
node_clear(node, sp_stat.prefetch_nodes);
continue;
^ permalink raw reply [flat|nested] 39+ messages in thread* Re: [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC
2006-06-13 0:08 ` Christoph Lameter
@ 2006-06-13 0:19 ` Con Kolivas
0 siblings, 0 replies; 39+ messages in thread
From: Con Kolivas @ 2006-06-13 0:19 UTC (permalink / raw)
To: Christoph Lameter
Cc: linux-kernel, akpm, Hugh Dickins, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner
On Tuesday 13 June 2006 10:08, Christoph Lameter wrote:
> On Tue, 13 Jun 2006, Con Kolivas wrote:
> > The comment should read something like:
>
> If we need another round then maybe it would be best if you would do that
> patch.
Sorry about that I wasn't trying to pester you.
> This?
Looks good :)
> Subject: swap_prefetch: conversion of nr_unstable to per zone counter
> From: Christoph Lameter <clameter@sgi.com>
>
> The determination of the vm state is now not that expensive
> anymore after we remove the use of the page state.
> Change the logic to avoid the expensive checks.
>
> Signed-off-by: Christoph Lameter <clameter@sgi.com>
> Signed-off-by: Con Kolivas <kernel@kolivas.org>
Thanks for your work on this.
--
-ck
^ permalink raw reply [flat|nested] 39+ messages in thread
* [PATCH 20/21] Conversion of nr_bounce to per zone counter
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (18 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
2006-06-12 21:14 ` [PATCH 21/21] Remove useless struct wbs Christoph Lameter
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: conversion of nr_bounce to per zone counter
From: Christoph Lameter <clameter@sgi.com>
Conversion of nr_bounce to a per zone counter
nr_bounce is only used for proc output. So it could be left as an
event counter. However, the event counters are not accurate and nr_bounce is
categorizing types of pages in a zone. So we really need this to also be a
per zone counter.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/drivers/base/node.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/drivers/base/node.c 2006-06-12 13:03:17.226722868 -0700
+++ linux-2.6.17-rc6-cl/drivers/base/node.c 2006-06-12 13:03:40.552426736 -0700
@@ -65,6 +65,7 @@ static ssize_t node_read_meminfo(struct
"Node %d Anonymous: %8lu kB\n"
"Node %d PageTables: %8lu kB\n"
"Node %d Unstable: %8lu kB\n"
+ "Node %d Bounce: %8lu kB\n"
"Node %d Slab: %8lu kB\n",
nid, K(i.totalram),
nid, K(i.freeram),
@@ -82,6 +83,7 @@ static ssize_t node_read_meminfo(struct
nid, K(node_page_state(nid, NR_ANON)),
nid, K(node_page_state(nid, NR_PAGETABLE)),
nid, K(node_page_state(nid, NR_UNSTABLE)),
+ nid, K(node_page_state(nid, NR_BOUNCE)),
nid, K(node_page_state(nid, NR_SLAB)));
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:03:17.229652374 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:03:40.553403238 -0700
@@ -56,6 +56,7 @@ enum zone_stat_item {
NR_DIRTY,
NR_WRITEBACK,
NR_UNSTABLE, /* NFS unstable pages */
+ NR_BOUNCE,
NR_STAT_ITEMS };
struct per_cpu_pages {
Index: linux-2.6.17-rc6-cl/mm/highmem.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/highmem.c 2006-06-05 17:57:02.000000000 -0700
+++ linux-2.6.17-rc6-cl/mm/highmem.c 2006-06-12 13:03:40.553403238 -0700
@@ -316,7 +316,7 @@ static void bounce_end_io(struct bio *bi
continue;
mempool_free(bvec->bv_page, pool);
- dec_page_state(nr_bounce);
+ dec_zone_page_state(bvec->bv_page, NR_BOUNCE);
}
bio_endio(bio_orig, bio_orig->bi_size, err);
@@ -397,7 +397,7 @@ static void __blk_queue_bounce(request_q
to->bv_page = mempool_alloc(pool, q->bounce_gfp);
to->bv_len = from->bv_len;
to->bv_offset = from->bv_offset;
- inc_page_state(nr_bounce);
+ inc_zone_page_state(to->bv_page, NR_BOUNCE);
if (rw == WRITE) {
char *vto, *vfrom;
Index: linux-2.6.17-rc6-cl/mm/vmstat.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmstat.c 2006-06-12 13:03:33.967873583 -0700
+++ linux-2.6.17-rc6-cl/mm/vmstat.c 2006-06-12 13:03:54.972432138 -0700
@@ -471,6 +471,7 @@ static char *vmstat_text[] = {
"nr_dirty",
"nr_writeback",
"nr_unstable",
+ "nr_bounce",
/* Event counters */
"pgpgin",
@@ -518,7 +519,6 @@ static char *vmstat_text[] = {
"allocstall",
"pgrotated",
- "nr_bounce",
};
/*
^ permalink raw reply [flat|nested] 39+ messages in thread* [PATCH 21/21] Remove useless struct wbs
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
` (19 preceding siblings ...)
2006-06-12 21:14 ` [PATCH 20/21] Conversion of nr_bounce to per zone counter Christoph Lameter
@ 2006-06-12 21:14 ` Christoph Lameter
20 siblings, 0 replies; 39+ messages in thread
From: Christoph Lameter @ 2006-06-12 21:14 UTC (permalink / raw)
To: linux-kernel
Cc: akpm, Hugh Dickins, Con Kolivas, Marcelo Tosatti, Nick Piggin,
linux-mm, Andi Kleen, Dave Chinner, Christoph Lameter
Subject: zoned vm counters: remove useless writeback structure
From: Christoph Lameter <clameter@sgi.com>
Remove writeback state
We can remove some functions now that were needed to calculate the page state
for writeback control since these statistics are now directly available.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/mm/page-writeback.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/page-writeback.c 2006-06-10 17:25:05.237248907 -0700
+++ linux-2.6.17-rc6-cl/mm/page-writeback.c 2006-06-10 19:27:44.138352158 -0700
@@ -99,23 +99,6 @@ EXPORT_SYMBOL(laptop_mode);
static void background_writeout(unsigned long _min_pages);
-struct writeback_state
-{
- unsigned long nr_dirty;
- unsigned long nr_unstable;
- unsigned long nr_mapped;
- unsigned long nr_writeback;
-};
-
-static void get_writeback_state(struct writeback_state *wbs)
-{
- wbs->nr_dirty = global_page_state(NR_DIRTY);
- wbs->nr_unstable = global_page_state(NR_UNSTABLE);
- wbs->nr_mapped = global_page_state(NR_MAPPED) +
- global_page_state(NR_ANON);
- wbs->nr_writeback = global_page_state(NR_WRITEBACK);
-}
-
/*
* Work out the current dirty-memory clamping and background writeout
* thresholds.
@@ -134,8 +117,7 @@ static void get_writeback_state(struct w
* clamping level.
*/
static void
-get_dirty_limits(struct writeback_state *wbs, long *pbackground, long *pdirty,
- struct address_space *mapping)
+get_dirty_limits(long *pbackground, long *pdirty, struct address_space *mapping)
{
int background_ratio; /* Percentages */
int dirty_ratio;
@@ -145,8 +127,6 @@ get_dirty_limits(struct writeback_state
unsigned long available_memory = total_pages;
struct task_struct *tsk;
- get_writeback_state(wbs);
-
#ifdef CONFIG_HIGHMEM
/*
* If this mapping can only allocate from low memory,
@@ -157,7 +137,9 @@ get_dirty_limits(struct writeback_state
#endif
- unmapped_ratio = 100 - (wbs->nr_mapped * 100) / total_pages;
+ unmapped_ratio = 100 - ((global_page_state(NR_MAPPED) +
+ global_page_state(NR_ANON)) * 100) /
+ total_pages;
dirty_ratio = vm_dirty_ratio;
if (dirty_ratio > unmapped_ratio / 2)
@@ -190,7 +172,6 @@ get_dirty_limits(struct writeback_state
*/
static void balance_dirty_pages(struct address_space *mapping)
{
- struct writeback_state wbs;
long nr_reclaimable;
long background_thresh;
long dirty_thresh;
@@ -208,11 +189,12 @@ static void balance_dirty_pages(struct a
.range_cyclic = 1,
};
- get_dirty_limits(&wbs, &background_thresh,
- &dirty_thresh, mapping);
- nr_reclaimable = wbs.nr_dirty + wbs.nr_unstable;
- if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh)
- break;
+ get_dirty_limits(&background_thresh, &dirty_thresh, mapping);
+ nr_reclaimable = global_page_state(NR_DIRTY) +
+ global_page_state(NR_UNSTABLE);
+ if (nr_reclaimable + global_page_state(NR_WRITEBACK) <=
+ dirty_thresh)
+ break;
if (!dirty_exceeded)
dirty_exceeded = 1;
@@ -225,11 +207,14 @@ static void balance_dirty_pages(struct a
*/
if (nr_reclaimable) {
writeback_inodes(&wbc);
- get_dirty_limits(&wbs, &background_thresh,
- &dirty_thresh, mapping);
- nr_reclaimable = wbs.nr_dirty + wbs.nr_unstable;
- if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh)
- break;
+ get_dirty_limits(&background_thresh,
+ &dirty_thresh, mapping);
+ nr_reclaimable = global_page_state(NR_DIRTY) +
+ global_page_state(NR_UNSTABLE);
+ if (nr_reclaimable +
+ global_page_state(NR_WRITEBACK)
+ <= dirty_thresh)
+ break;
pages_written += write_chunk - wbc.nr_to_write;
if (pages_written >= write_chunk)
break; /* We've done our duty */
@@ -237,8 +222,9 @@ static void balance_dirty_pages(struct a
blk_congestion_wait(WRITE, HZ/10);
}
- if (nr_reclaimable + wbs.nr_writeback <= dirty_thresh && dirty_exceeded)
- dirty_exceeded = 0;
+ if (nr_reclaimable + global_page_state(NR_WRITEBACK)
+ <= dirty_thresh && dirty_exceeded)
+ dirty_exceeded = 0;
if (writeback_in_progress(bdi))
return; /* pdflush is already working this queue */
@@ -300,12 +286,11 @@ EXPORT_SYMBOL(balance_dirty_pages_rateli
void throttle_vm_writeout(void)
{
- struct writeback_state wbs;
long background_thresh;
long dirty_thresh;
for ( ; ; ) {
- get_dirty_limits(&wbs, &background_thresh, &dirty_thresh, NULL);
+ get_dirty_limits(&background_thresh, &dirty_thresh, NULL);
/*
* Boost the allowable dirty threshold a bit for page
@@ -313,8 +298,9 @@ void throttle_vm_writeout(void)
*/
dirty_thresh += dirty_thresh / 10; /* wheeee... */
- if (wbs.nr_unstable + wbs.nr_writeback <= dirty_thresh)
- break;
+ if (global_page_state(NR_UNSTABLE) +
+ global_page_state(NR_WRITEBACK) <= dirty_thresh)
+ break;
blk_congestion_wait(WRITE, HZ/10);
}
}
@@ -337,12 +323,12 @@ static void background_writeout(unsigned
};
for ( ; ; ) {
- struct writeback_state wbs;
long background_thresh;
long dirty_thresh;
- get_dirty_limits(&wbs, &background_thresh, &dirty_thresh, NULL);
- if (wbs.nr_dirty + wbs.nr_unstable < background_thresh
+ get_dirty_limits(&background_thresh, &dirty_thresh, NULL);
+ if (global_page_state(NR_DIRTY) +
+ global_page_state(NR_UNSTABLE) < background_thresh
&& min_pages <= 0)
break;
wbc.encountered_congestion = 0;
@@ -366,12 +352,9 @@ static void background_writeout(unsigned
*/
int wakeup_pdflush(long nr_pages)
{
- if (nr_pages == 0) {
- struct writeback_state wbs;
-
- get_writeback_state(&wbs);
- nr_pages = wbs.nr_dirty + wbs.nr_unstable;
- }
+ if (nr_pages == 0)
+ nr_pages = global_page_state(NR_DIRTY) +
+ global_page_state(NR_UNSTABLE);
return pdflush_operation(background_writeout, nr_pages);
}
@@ -402,7 +385,6 @@ static void wb_kupdate(unsigned long arg
unsigned long start_jif;
unsigned long next_jif;
long nr_to_write;
- struct writeback_state wbs;
struct writeback_control wbc = {
.bdi = NULL,
.sync_mode = WB_SYNC_NONE,
@@ -415,11 +397,11 @@ static void wb_kupdate(unsigned long arg
sync_supers();
- get_writeback_state(&wbs);
oldest_jif = jiffies - dirty_expire_interval;
start_jif = jiffies;
next_jif = start_jif + dirty_writeback_interval;
- nr_to_write = wbs.nr_dirty + wbs.nr_unstable +
+ nr_to_write = global_page_state(NR_DIRTY) +
+ global_page_state(NR_UNSTABLE) +
(inodes_stat.nr_inodes - inodes_stat.nr_unused);
while (nr_to_write > 0) {
wbc.encountered_congestion = 0;
^ permalink raw reply [flat|nested] 39+ messages in thread