From: Larry Woodman <lwoodman@redhat.com>
Date: Mon, 4 May 2009 17:26:20 -0400
Subject: [mm] add tracepoints
Message-id: 1241472380.10978.29.camel@dhcp-100-19-198.bos.redhat.com
O-Subject: Re: [RHEL 5-U4 Patch] mm tracepoints
Bugzilla: 493444
RH-Acked-by: Dave Anderson <anderson@redhat.com>

This patch is basically a RHEL5-U4 backport of the mm tracepoints
submitted upstream:

diff --git a/include/trace/mm.h b/include/trace/mm.h
new file mode 100644
index 0000000..e18fa8b
--- /dev/null
+++ b/include/trace/mm.h
@@ -0,0 +1,111 @@
+#ifndef _TRACE_MM_H
+#define _TRACE_MM_H
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM mm
+
+#include <linux/tracepoint.h>
+#include <linux/mm.h>
+
+DEFINE_TRACE(mm_anon_fault,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_anon_pgin,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_anon_cow,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_anon_userfree,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_anon_unmap,
+	TPPROTO(unsigned long pfn, int success),
+	TPARGS(pfn, success));
+
+DEFINE_TRACE(mm_filemap_fault,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_filemap_cow,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_filemap_unmap,
+	TPPROTO(unsigned long pfn, int success),
+	TPARGS(pfn, success));
+
+DEFINE_TRACE(mm_filemap_userunmap,
+	TPPROTO(struct mm_struct *mm, unsigned long address, unsigned long pfn),
+	TPARGS(mm, address, pfn));
+
+DEFINE_TRACE(mm_pagereclaim_pgout,
+	TPPROTO(unsigned long pfn, int anon),
+	TPARGS(pfn, anon));
+
+DEFINE_TRACE(mm_pagereclaim_free,
+	TPPROTO(unsigned long pfn, int anon),
+	TPARGS(pfn, anon));
+
+DEFINE_TRACE(mm_pdflush_bgwriteout,
+	TPPROTO(unsigned long count),
+	TPARGS(count));
+
+DEFINE_TRACE(mm_pdflush_kupdate,
+	TPPROTO(unsigned long count),
+	TPARGS(count));
+
+DEFINE_TRACE(mm_page_allocation,
+	TPPROTO(unsigned long pfn, unsigned long free),
+	TPARGS(pfn, free));
+
+DEFINE_TRACE(mm_kswapd_runs,
+	TPPROTO(unsigned long reclaimed),
+	TPARGS(reclaimed));
+
+DEFINE_TRACE(mm_directreclaim_reclaimall,
+	TPPROTO(unsigned long priority),
+	TPARGS(priority));
+
+DEFINE_TRACE(mm_directreclaim_reclaimzone,
+	TPPROTO(unsigned long reclaimed),
+	TPARGS(reclaimed));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkzone,
+	TPPROTO(unsigned long reclaimed),
+	TPARGS(reclaimed));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkactive,
+	TPPROTO(unsigned long scanned),
+	TPARGS(scanned));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkactive_a2a,
+	TPPROTO(unsigned long pfn),
+	TPARGS(pfn));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkactive_a2i,
+	TPPROTO(unsigned long pfn),
+	TPARGS(pfn));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkinactive,
+	TPPROTO(unsigned long reclaimed),
+	TPARGS(reclaimed));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkinactive_i2a,
+	TPPROTO(unsigned long pfn),
+	TPARGS(pfn));
+
+DEFINE_TRACE(mm_pagereclaim_shrinkinactive_i2i,
+	TPPROTO(unsigned long pfn),
+	TPARGS(pfn));
+
+DEFINE_TRACE(mm_page_free,
+	TPPROTO(unsigned long pfn),
+	TPARGS(pfn));
+
+#undef TRACE_SYSTEM
+#endif
diff --git a/mm/filemap.c b/mm/filemap.c
index 6aa48e1..f899d86 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -31,6 +31,7 @@
 #include <linux/syscalls.h>
 #include <linux/cpuset.h>
 #include <linux/hardirq.h> /* for BUG_ON(!in_atomic()) only */
+#include <trace/mm.h>
 #include "internal.h"
 #include <trace/filemap.h>
 
@@ -1502,6 +1503,7 @@ success:
 	mark_page_accessed(page);
 	if (type)
 		*type = majmin;
+	trace_mm_filemap_fault(area->vm_mm, address, page_to_pfn(page));
 	return page;
 
 outside_data_content:
diff --git a/mm/memory.c b/mm/memory.c
index cffa52a..bbd2ba5 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -51,6 +51,7 @@
 #include <linux/init.h>
 #include <linux/writeback.h>
 #include <linux/mmu_notifier.h>
+#include <trace/mm.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -768,14 +769,18 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 						addr) != page->index)
 				set_pte_at(mm, addr, pte,
 					   pgoff_to_pte(page->index));
-			if (PageAnon(page))
+			if (PageAnon(page)) {
 				anon_rss--;
-			else {
+				trace_mm_anon_userfree(mm, addr,
+							page_to_pfn(page));
+			} else {
 				if (pte_dirty(ptent))
 					set_page_dirty(page);
 				if (pte_young(ptent))
 					SetPageReferenced(page);
 				file_rss--;
+				trace_mm_filemap_userunmap(mm, addr,
+							page_to_pfn(page));
 			}
 			page_remove_rmap(page);
 			tlb_remove_page(tlb, page);
@@ -1952,9 +1957,13 @@ gotten:
 			if (!PageAnon(old_page)) {
 				dec_mm_counter(mm, file_rss);
 				inc_mm_counter(mm, anon_rss);
+				trace_mm_filemap_cow(mm, address,
+						page_to_pfn(new_page));
 			}
-		} else
+		} else {
 			inc_mm_counter(mm, anon_rss);
+			trace_mm_anon_cow(mm, address, page_to_pfn(new_page));
+		}
 		flush_cache_page(vma, address, pte_pfn(orig_pte));
 		entry = mk_pte(new_page, vma->vm_page_prot);
 		entry = maybe_mkwrite(pte_mkdirty(entry), vma);
@@ -2343,7 +2352,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		int write_access, pte_t orig_pte)
 {
 	spinlock_t *ptl;
-	struct page *page;
+	struct page *page = NULL;
 	swp_entry_t entry;
 	pte_t pte;
 	int ret = VM_FAULT_MINOR;
@@ -2426,6 +2435,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 unlock:
 	pte_unmap_unlock(page_table, ptl);
 out:
+	trace_mm_anon_pgin(mm, address, page_to_pfn(page));
 	return ret;
 out_nomap:
 	pte_unmap_unlock(page_table, ptl);
@@ -2487,6 +2497,7 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	lazy_mmu_prot_update(entry);
 unlock:
 	pte_unmap_unlock(page_table, ptl);
+	trace_mm_anon_fault(mm, address, page_to_pfn(page));
 	return VM_FAULT_MINOR;
 release:
 	page_cache_release(page);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 6b3a818..d337e45 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -31,6 +31,7 @@
 #include <linux/cpu.h>
 #include <linux/syscalls.h>
 #include <linux/rmap.h>
+#include <trace/mm.h>
 
 /*
  * The maximum number of pages to writeout in a single bdflush/kupdate
@@ -359,6 +360,7 @@ static void background_writeout(unsigned long _min_pages)
 				break;
 		}
 	}
+	trace_mm_pdflush_bgwriteout(_min_pages);
 }
 
 /*
@@ -421,6 +423,7 @@ static void wb_kupdate(unsigned long arg)
 	nr_to_write = global_page_state(NR_FILE_DIRTY) +
 			global_page_state(NR_UNSTABLE_NFS) +
 			(inodes_stat.nr_inodes - inodes_stat.nr_unused);
+	trace_mm_pdflush_kupdate(nr_to_write);
 	while (nr_to_write > 0) {
 		wbc.encountered_congestion = 0;
 		wbc.nr_to_write = max_writeback_pages;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 11b1925..4bbf86b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -37,6 +37,7 @@
 #include <linux/vmalloc.h>
 #include <linux/mempolicy.h>
 #include <linux/stop_machine.h>
+#include <trace/mm.h>
 
 #include <asm/tlbflush.h>
 #include <asm/div64.h>
@@ -745,6 +746,7 @@ static void fastcall free_hot_cold_page(struct page *page, int cold)
 	if (free_pages_check(page))
 		return;
 
+	trace_mm_page_free(page_to_pfn(page));
 	kernel_map_pages(page, 1, 0);
 
 	pcp = &zone_pcp(zone, get_cpu())->pcp[cold];
@@ -834,6 +836,7 @@ again:
 	BUG_ON(bad_range(zone, page));
 	if (prep_new_page(page, order, gfp_flags))
 		goto again;
+	trace_mm_page_allocation(page_to_pfn(page), zone->free_pages);
 	return page;
 
 failed:
@@ -919,9 +922,8 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 		}
 
 		page = buffered_rmqueue(zonelist, *z, order, gfp_mask);
-		if (page) {
+		if (page) 
 			break;
-		}
 	} while (*(++z) != NULL);
 	return page;
 }
diff --git a/mm/rmap.c b/mm/rmap.c
index 0673a13..12bb716 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -54,6 +54,7 @@
 #include <linux/rcupdate.h>
 #include <linux/module.h>
 #include <linux/mmu_notifier.h>
+#include <trace/mm.h>
 
 #include <asm/tlbflush.h>
 
@@ -808,6 +809,7 @@ static int try_to_unmap_anon(struct page *page, int migration)
 			break;
 	}
 	spin_unlock(&anon_vma->lock);
+	trace_mm_anon_unmap(page_to_pfn(page), ret == SWAP_SUCCESS);
 	return ret;
 }
 
@@ -904,6 +906,7 @@ static int try_to_unmap_file(struct page *page, int migration)
 		vma->vm_private_data = NULL;
 out:
 	spin_unlock(&mapping->i_mmap_lock);
+	trace_mm_filemap_unmap(page_to_pfn(page), ret == SWAP_SUCCESS);
 	return ret;
 }
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 1f13601..8a11dd8 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -40,6 +40,7 @@
 #include <asm/div64.h>
 
 #include <linux/swapops.h>
+#include <trace/mm.h>
 
 #include "internal.h"
 
@@ -374,6 +375,7 @@ static pageout_t pageout(struct page *page, struct address_space *mapping)
 			ClearPageReclaim(page);
 		}
 
+		trace_mm_pagereclaim_pgout(page_to_pfn(page), PageAnon(page));
 		return PAGE_SUCCESS;
 	}
 
@@ -560,15 +562,18 @@ free_it:
 		nr_reclaimed++;
 		if (!pagevec_add(&freed_pvec, page))
 			__pagevec_release_nonlru(&freed_pvec);
+		trace_mm_pagereclaim_free(page_to_pfn(page), PageAnon(page));
 		continue;
 
 activate_locked:
 		SetPageActive(page);
 		pgactivate++;
+		trace_mm_pagereclaim_shrinkinactive_i2a(page_to_pfn(page));
 keep_locked:
 		unlock_page(page);
 keep:
 		list_add(&page->lru, &ret_pages);
+		trace_mm_pagereclaim_shrinkinactive_i2i(page_to_pfn(page));
 		BUG_ON(PageLRU(page));
 	}
 	list_splice(&ret_pages, page_list);
@@ -697,6 +702,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 done:
 	local_irq_enable();
 	pagevec_release(&pvec);
+	trace_mm_pagereclaim_shrinkinactive(nr_reclaimed);
 	return nr_reclaimed;
 }
 
@@ -813,6 +819,7 @@ force_reclaim_mapped:
 			    (total_swap_pages == 0 && PageAnon(page)) ||
 			    page_referenced(page, 0)) {
 				list_add(&page->lru, &l_active);
+				trace_mm_pagereclaim_shrinkactive_a2a(page_to_pfn(page));
 				continue;
 			}
 		}
@@ -832,6 +839,7 @@ force_reclaim_mapped:
 
 		list_move(&page->lru, &zone->inactive_list);
 		pgmoved++;
+		trace_mm_pagereclaim_shrinkactive_a2i(page_to_pfn(page));
 		if (!pagevec_add(&pvec, page)) {
 			zone->nr_inactive += pgmoved;
 			spin_unlock_irq(&zone->lru_lock);
@@ -874,6 +882,7 @@ force_reclaim_mapped:
 	__count_vm_events(PGDEACTIVATE, pgdeactivate);
 	spin_unlock_irq(&zone->lru_lock);
 
+	trace_mm_pagereclaim_shrinkactive(pgscanned);
 	pagevec_release(&pvec);
 }
 
@@ -945,6 +954,7 @@ static void shrink_zone(int priority, struct zone *zone,
 
 	throttle_vm_writeout();
 
+	trace_mm_pagereclaim_shrinkzone(nr_reclaimed);
 	atomic_dec(&zone->reclaim_in_progress);
 }
 
@@ -988,6 +998,7 @@ static void shrink_zones(int priority, struct zone **zones,
 
 		shrink_zone(priority, zone, sc);
 	}
+	trace_mm_directreclaim_reclaimall(priority);
 }
  
 /*
@@ -1255,6 +1266,7 @@ out:
 		goto loop_again;
 	}
 
+	trace_mm_kswapd_runs(sc.nr_reclaimed);
 	return sc.nr_reclaimed;
 }
 
@@ -1658,6 +1670,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 
 	p->reclaim_state = NULL;
 	current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
+	trace_mm_directreclaim_reclaimzone(sc.nr_reclaimed);
 	return sc.nr_reclaimed >= nr_pages;
 }