ext2: buffer_head usage with and without O_DIRECT

2005-07-24 Thread li nux
This is with respect to 2.4.28 on
http://lxr.linux.no/source/?v=2.4.18 

When i do read/write on ext2 without opening files
with O_DIRECT, i can see buffer_head constantly
increasing in /proc/slabinfo.

But when I open files with O_DIRECT, the i/o is done
without using buffer_head, /proc/slabinfo shows this
as constant throughout the i/o.
There is no other i/o activity on the box.

Stacks below show that both of them creates
buffer_head's. Any idea why this is happening ?

As per the code, Without O_DIRECT, stack is:
(See fs/buffer.c)
submit_bh_rsector
submit_bh
block_read_full_page (This calls create_buffers to
create buffer_head's)
ext2_readpage
do_generic_file_read
generic_file_new_read
generic_file_read

With O_DIRECT:
brw_kiovec (This creates buffer_head's)
generic_direct_sector_IO (cals
prepare_direct_IO_iobuf)
ext2_direct_IO
generic_file_direct_IO 
generic_file_new_read
generic_file_read

-Thanks.




Start your day with Yahoo! - make it your home page 
http://www.yahoo.com/r/hs 
 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/


Oprofile: patch for Call-graph on AMD64

2005-07-13 Thread li nux
Is oprofile patch available for call-graph support on
AMD64 ?
http://oprofile.sourceforge.net/patches/ has only for
IA64 and x86.
What is the status of progress on AMD64 with respect
to this.

Thanks.



__ 
Yahoo! Mail 
Stay connected, organized, and protected. Take the tour: 
http://tour.mail.yahoo.com/mailtour.html 

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/