Re: kern/50375: layerfs (nullfs) locking problem leading to livelock

[Jeff Rizzo <riz%NetBSD.org@localhost>]

The following reply was made to PR kern/50375; it has been noted by GNATS.

From: Jeff Rizzo <riz%NetBSD.org@localhost>
To: gnats-bugs%NetBSD.org@localhost, hannken%NetBSD.org@localhost
Cc: 
Subject: Re: kern/50375: layerfs (nullfs) locking problem leading to livelock
Date: Tue, 10 Nov 2015 09:12:43 -0800

 On 11/10/15 02:20 AM, J. Hannken-Illjes wrote:
 > The following reply was made to PR kern/50375; it has been noted by GNATS.
 >
 > From: "J. Hannken-Illjes" <hannken%eis.cs.tu-bs.de@localhost>
 > To: gnats-bugs%NetBSD.org@localhost
 > Cc: Jeff Rizzo <riz%NetBSD.org@localhost>
 > Subject: Re: kern/50375: layerfs (nullfs) locking problem leading to livelock
 > Date: Tue, 10 Nov 2015 11:18:56 +0100
 >
 >   This deadlock looks like vmem exhaustion:
 >   
 >   1) Threads 19955, 26028, 25222 and 17766 wait for the vnode
 >      lock 0x92ffc4d4.
 >   
 >      This lock is held by thread 0x96ddc9e0 waiting in vmem_xalloc
 >      for vmem to become available.
 >   
 >   2) Thread 4849 waits for vnode lock 0x926cbb04.
 >   
 >      This lock is held by thread 0x96ddd4e0 also waiting
 >      in vmem_xalloc for vmem to become available.
 >   
 >   Has nothing to do with layerfs, no "VOP_LOCK > layer_lock > VOP_LOCK"
 >   sequences in the backtrace of the blocked threads.
 >   
 >   Could we conclude the patch fixes your issue with layerfs deadlock?
 >   
 >
 
 
 I agree, the issue in this PR, 50375, appears to be fixed.  Thank you 
 very much for your help.
 
 I do, however, have two hosts currently exhibiting the vmem exhaustion 
 problem, above.  I will troubleshoot that and open a new PR if appropriate.
 
 +j