Re: Any known issues with this combination?

2009-11-04 Thread Michael S. Zick
On Wed November 4 2009, sf...@users.sourceforge.net wrote:
 
 Hello Kyle,
 
 Kyle Bader:
  I plan on using 2.6.31.5 vanilla from git, merging it with aufs2-31 and
  applying the grsecurity patch:
   :::
  Are you aware of any incompatibilities?
 
 As far as I know, there is no such problem reported.
 But it depends upon these things,
 - grsecurity patch or other patches
 - changes made in mainline 2.6.31 -- 2.6.31.5
   Basically the aufs2 trees are based upon the first version of every
   mainline release, 2.6.31(.0) for example.
 
 If nobody has ever tried 2.6.31.5, you will be the first one. :-)
 

Yes, let the mailing list know - good or bad.
I am planning to try building for 2.6.31.x + auFS for MIPSel.

Mike
 
 J. R. Okajima
 
 --
 Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
 trial. Simplify your report design, integration and deployment - and focus on 
 what you do best, core application coding. Discover what's new with
 Crystal Reports now.  http://p.sf.net/sfu/bobj-july
 
 



--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july


Re: possible deadlock

2009-11-04 Thread sfjro

michael rodriguez:
 Here is an example of some blocked processes using the following kernel 
 combo with frame pointers enabled:

I don't think these stacktraces reliable. While I don't know how you
collect them, I am afraid there is something wrong on your system.
All these stacktraces contain bogus info, but I can see a kind of trend
which is,
lookup
aufs_d_revalidate
aufs_read_lock
di_write_lock
down_write

If there was a process which still holds down_read(), other processes
trying down_write() would stop. But I could not find such process.

If there really exists a deadlock or something, the kernel debugging
feature LOCKDEP will help you. Try enabling PROVE_LOCKING or something.


J. R. Okajima

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july