lkml.org 
[lkml]   [2015]   [Apr]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 0/13] Parallel struct page initialisation v4
On 04/28/2015 10:36 AM, Mel Gorman wrote:
> The bulk of the changes here are related to Andrew's feedback. Functionally
> there is almost no difference.
>
> Changelog since v3
> o Fix section-related warning
> o Comments, clarifications, checkpatch
> o Report the number of pages initialised
>
> Changelog since v2
> o Reduce overhead of topology_init
> o Remove boot-time kernel parameter to enable/disable
> o Enable on UMA
>
> Changelog since v1
> o Always initialise low zones
> o Typo corrections
> o Rename parallel mem init to parallel struct page init
> o Rebase to 4.0
>
> Struct page initialisation had been identified as one of the reasons why
> large machines take a long time to boot. Patches were posted a long time ago
> to defer initialisation until they were first used. This was rejected on
> the grounds it should not be necessary to hurt the fast paths. This series
> reuses much of the work from that time but defers the initialisation of
> memory to kswapd so that one thread per node initialises memory local to
> that node.
>
> After applying the series and setting the appropriate Kconfig variable I
> see this in the boot log on a 64G machine
>
> [ 7.383764] kswapd 0 initialised deferred memory in 188ms
> [ 7.404253] kswapd 1 initialised deferred memory in 208ms
> [ 7.411044] kswapd 3 initialised deferred memory in 216ms
> [ 7.411551] kswapd 2 initialised deferred memory in 216ms
>
> On a 1TB machine, I see
>
> [ 8.406511] kswapd 3 initialised deferred memory in 1116ms
> [ 8.428518] kswapd 1 initialised deferred memory in 1140ms
> [ 8.435977] kswapd 0 initialised deferred memory in 1148ms
> [ 8.437416] kswapd 2 initialised deferred memory in 1148ms
>
> Once booted the machine appears to work as normal. Boot times were measured
> from the time shutdown was called until ssh was available again. In the
> 64G case, the boot time savings are negligible. On the 1TB machine, the
> savings were 16 seconds.
>
> It would be nice if the people that have access to really large machines
> would test this series and report how much boot time is reduced.
>
>

I ran a bootup timing test on a 12-TB 16-socket IvyBridge-EX system.
From grub menu to ssh login, the bootup time was 453s before the patch
and 265s after the patch - a saving of 188s (42%). I used a different OS
environment and config file with this test and so the timing data
weren't comparable with my previous testing data. The kswapd log entries
were

[ 45.973967] kswapd 4 initialised 197655470 pages in 4390ms
[ 45.974214] kswapd 7 initialised 197655470 pages in 4390ms
[ 45.976692] kswapd 15 initialised 197654299 pages in 4390ms
[ 45.993284] kswapd 0 initialised 197131131 pages in 4410ms
[ 46.032735] kswapd 9 initialised 197655470 pages in 4447ms
[ 46.065856] kswapd 8 initialised 197655470 pages in 4481ms
[ 46.066615] kswapd 1 initialised 197622702 pages in 4483ms
[ 46.077995] kswapd 2 initialised 197655470 pages in 4495ms
[ 46.219508] kswapd 13 initialised 197655470 pages in 4633ms
[ 46.224358] kswapd 3 initialised 197655470 pages in 4641ms
[ 46.228441] kswapd 11 initialised 197655470 pages in 4643ms
[ 46.232258] kswapd 12 initialised 197655470 pages in 4647ms
[ 46.239659] kswapd 10 initialised 197655470 pages in 4654ms
[ 46.243402] kswapd 14 initialised 197655470 pages in 4657ms
[ 46.250368] kswapd 5 initialised 197655470 pages in 4666ms
[ 46.254659] kswapd 6 initialised 197655470 pages in 4670ms

Cheers,
Longman


\
 
 \ /
  Last update: 2015-04-29 03:21    [W:1.079 / U:0.088 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site