[PATCH] remove page_launder() from bdflush

[PATCH] remove page_launder() from bdflush

[Marcelo Tosatti]

Hi Linus, 

There is no reason why bdflush should call page_launder().

Its pretty obvious that bdflush's job is to only write out _buffers_. 

Under my tests this patch makes things faster.

Guess why? Because bdflush is writing out buffers when it should instead
blocking inside try_to_free_pages().

Please apply. 

--- fs/buffer.c.orig    Tue May 15 03:13:05 2001
+++ fs/buffer.c Tue May 15 03:13:22 2001
@@ -2703,8 +2703,6 @@
                CHECK_EMERGENCY_SYNC
 
                flushed = flush_dirty_buffers(0);
-               if (free_shortage())
-                       flushed += page_launder(GFP_KERNEL, 0);
 
                /*
                 * If there are still a lot of dirty buffers around,


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/

Re: [PATCH] remove page_launder() from bdflush

[David Woodhouse]

marcelo@conectiva.com.br said:
>  There is no reason why bdflush should call page_launder().
> Its pretty obvious that bdflush's job is to only write out _buffers_. 
> Under my tests this patch makes things faster. 

Oh good. ISTR last time I looked at implementing CONFIG_BLK_DEV I got as far
as trying to remove bdflush() before getting confused at finding
page_launder() in it, and going on to more important things.

--
dwmw2


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/