All of lore.kernel.org
 help / color / mirror / Atom feed
* Re: New Reiser4 snapshot slower?
  2004-03-26 21:00 New Reiser4 snapshot slower? Jonathan Briggs
@ 2004-03-26 19:38 ` Hans Reiser
  2004-03-27  0:34 ` Jonathan Briggs
  1 sibling, 0 replies; 4+ messages in thread
From: Hans Reiser @ 2004-03-26 19:38 UTC (permalink / raw)
  To: Jonathan Briggs; +Cc: Reiserfs mail-list, zam

Jonathan Briggs wrote:

>This command:
># dd if=/dev/zero of=testfile bs=4k count=256k
>seems to take longer than it did before.  The difference seems to be all
>system time.
>
>I will have to reboot to an old version later, to get accurate numbers.
>
>I don't have debugging turned on, so that's not it.  Has something
>changed that would have made this slower?
>  
>
zam, please look into this report.

-- 
Hans


^ permalink raw reply	[flat|nested] 4+ messages in thread

* New Reiser4 snapshot slower?
@ 2004-03-26 21:00 Jonathan Briggs
  2004-03-26 19:38 ` Hans Reiser
  2004-03-27  0:34 ` Jonathan Briggs
  0 siblings, 2 replies; 4+ messages in thread
From: Jonathan Briggs @ 2004-03-26 21:00 UTC (permalink / raw)
  To: Reiserfs mail-list

[-- Attachment #1: Type: text/plain, Size: 380 bytes --]

This command:
# dd if=/dev/zero of=testfile bs=4k count=256k
seems to take longer than it did before.  The difference seems to be all
system time.

I will have to reboot to an old version later, to get accurate numbers.

I don't have debugging turned on, so that's not it.  Has something
changed that would have made this slower?
-- 
Jonathan Briggs
jbriggs@esoft.com

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: New Reiser4 snapshot slower?
  2004-03-26 21:00 New Reiser4 snapshot slower? Jonathan Briggs
  2004-03-26 19:38 ` Hans Reiser
@ 2004-03-27  0:34 ` Jonathan Briggs
  2004-03-28 16:37   ` Hans Reiser
  1 sibling, 1 reply; 4+ messages in thread
From: Jonathan Briggs @ 2004-03-27  0:34 UTC (permalink / raw)
  To: Reiserfs mail-list

[-- Attachment #1: Type: text/plain, Size: 1873 bytes --]

On Fri, 2004-03-26 at 14:00, Jonathan Briggs wrote:
> This command:
> # dd if=/dev/zero of=testfile bs=4k count=256k
> seems to take longer than it did before.  The difference seems to be all
> system time.
> 
> I will have to reboot to an old version later, to get accurate numbers.
> 
> I don't have debugging turned on, so that's not it.  Has something
> changed that would have made this slower?

I have some numbers for the previous snapshot vs the current
snapshot.  It doesn't look like there was much of a difference really. 
The real difference seems to be between a empty filesystem and one
that's been used as a home directory for several months.  That's where
my impression of slowness comes from: the last time I speed tested it,
my home directory was empty.

In /home:
With new R4 snapshot:
1 GB from /dev/zero into testfile
real: 38.301 s
system: 6.508 s
 
With old snapshot:
1 GB
real: 33 s
sys: 6.9 s
 
In fresh built R4 filesystem on loopback
Old snapshot
real: 24.2
system: 6.7
 
New snapshot
real: 22.7
system: 6.5

I also booted the new snapshot with profile=2 and created a 1 GB
testfile in my home directory and sorted the output of readprofile. 
Here are the interesting bits of that:
   132 handle_IRQ_event                           1.1786
    51 jnode_set_type                             1.5938
   175 clear_user                                 2.1875
  2311 protect_extent_nodes                       2.5340
  1060 do_softirq                                 6.6250
  1114 protect_reloc_node                        13.9250
  3503 __copy_from_user_ll                       27.3672
  8710 unprotect_extent_nodes                    45.3646
 12285 jlookup                                   54.8438
 33297 default_idle                             693.6875

-- 
Jonathan Briggs
jbriggs@esoft.com

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: New Reiser4 snapshot slower?
  2004-03-27  0:34 ` Jonathan Briggs
@ 2004-03-28 16:37   ` Hans Reiser
  0 siblings, 0 replies; 4+ messages in thread
From: Hans Reiser @ 2004-03-28 16:37 UTC (permalink / raw)
  To: Jonathan Briggs; +Cc: Reiserfs mail-list

The repacker should help when it is ready.

-- 
Hans


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2004-03-28 16:37 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-03-26 21:00 New Reiser4 snapshot slower? Jonathan Briggs
2004-03-26 19:38 ` Hans Reiser
2004-03-27  0:34 ` Jonathan Briggs
2004-03-28 16:37   ` Hans Reiser

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.