lkml.org 
[lkml]   [2012]   [Dec]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectHow about changing block size to fast huge memory init of sysfs
From
Hi Hanse,

I am reading the drive/base/memory.c that you the
maintainer of it. I met a

problem, could you help me with it?

My panic is following: BUG: soft lockup - CPU#0 stuck
for 22s! [swapper:1]

. This is a warning from watch dog . My platform is 4tb memory ,
64bit x86, SLES sp2

3.0.13 kernel.

In the beginning, I found that a unnecessary search method named
“find_ memory_block”

then removed( similar with one commit I found later). It reduced time
from 50s to30s.

Unluckily ,still show the panic. I think the most expensive cost is
register process.

Because in 4TB 64bit, there are about 320000 block to register. If
we can reduce the

block number, then the time cost will be smaller. So my another
propose is to make block

size bigger, then sections_per_block will bigger, then more sections
belong to one block ,

then less block to register ,then less time will cost.

Is my propose suitable? If yes, how to change the block size.





Best regards,

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

\
 
 \ /
  Last update: 2012-12-10 05:01    [W:0.020 / U:2.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site