[PATCH] page_uptodate locking scalability

Use a bit spin lock in the first buffer of the page to synchronise asynch
IO buffer completions, instead of the global page_uptodate_lock, which is
showing some scalabilty problems.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

fs/buffer.c

@@ -513,8 +513,8 @@ static void free_more_memory(void)
  */
 static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 {
-	static DEFINE_SPINLOCK(page_uptodate_lock);
 	unsigned long flags;
+	struct buffer_head *first;
 	struct buffer_head *tmp;
 	struct page *page;
 	int page_uptodate = 1;
@@ -536,7 +536,9 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	 * two buffer heads end IO at almost the same time and both
 	 * decide that the page is now completely done.
 	 */
-	spin_lock_irqsave(&page_uptodate_lock, flags);
+	first = page_buffers(page);
+	local_irq_save(flags);
+	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
 	clear_buffer_async_read(bh);
 	unlock_buffer(bh);
 	tmp = bh;
@@ -549,7 +551,8 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 		}
 		tmp = tmp->b_this_page;
 	} while (tmp != bh);
-	spin_unlock_irqrestore(&page_uptodate_lock, flags);
+	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
+	local_irq_restore(flags);
 
 	/*
 	 * If none of the buffers had errors and they are all
@@ -561,7 +564,8 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
 	return;
 
 still_busy:
-	spin_unlock_irqrestore(&page_uptodate_lock, flags);
+	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
+	local_irq_restore(flags);
 	return;
 }
 
@@ -572,8 +576,8 @@ still_busy:
 void end_buffer_async_write(struct buffer_head *bh, int uptodate)
 {
 	char b[BDEVNAME_SIZE];
-	static DEFINE_SPINLOCK(page_uptodate_lock);
 	unsigned long flags;
+	struct buffer_head *first;
 	struct buffer_head *tmp;
 	struct page *page;
 
@@ -594,7 +598,10 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
 		SetPageError(page);
 	}
 
-	spin_lock_irqsave(&page_uptodate_lock, flags);
+	first = page_buffers(page);
+	local_irq_save(flags);
+	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
+
 	clear_buffer_async_write(bh);
 	unlock_buffer(bh);
 	tmp = bh->b_this_page;
@@ -605,12 +612,14 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
 		}
 		tmp = tmp->b_this_page;
 	}
-	spin_unlock_irqrestore(&page_uptodate_lock, flags);
+	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
+	local_irq_restore(flags);
 	end_page_writeback(page);
 	return;
 
 still_busy:
-	spin_unlock_irqrestore(&page_uptodate_lock, flags);
+	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
+	local_irq_restore(flags);
 	return;
 }
 

include/linux/buffer_head.h

@@ -19,6 +19,9 @@ enum bh_state_bits {
 	BH_Dirty,	/* Is dirty */
 	BH_Lock,	/* Is locked */
 	BH_Req,		/* Has been submitted for I/O */
+	BH_Uptodate_Lock,/* Used by the first bh in a page, to serialise
+			  * IO completion of other buffers in the page
+			  */
 
 	BH_Mapped,	/* Has a disk mapping */
 	BH_New,		/* Disk mapping was newly created by get_block */