NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: kern/41374: getnewvnode deadlock

To: gnats-bugs%NetBSD.org@localhost
Subject: Re: kern/41374: getnewvnode deadlock
From: Andrew Doran <ad%netbsd.org@localhost>
Date: Sun, 10 May 2009 08:31:19 +0000

On Thu, May 07, 2009 at 04:45:01AM +0000, yamt%mwd.biglobe.ne.jp@localhost 
wrote:

>       in the case of getcleanvnode,
>       set VI_XLOCK before incrementing v_usecount?

I think insufficient because vtryget() is lockless. How about making XLOCK
a flag bit in v_useconut?

Follow-Ups:
- Re: kern/41374: getnewvnode deadlock
  - From: YAMAMOTO Takashi

References:
- kern/41374: getnewvnode deadlock
  - From: yamt

Prev by Date: kern/41400: uhci_poll_hub "c->c_magic == CALLOUT_MAGIC" assertion failure
Next by Date: Re: kern/41374: getnewvnode deadlock
Previous by Thread: kern/41374: getnewvnode deadlock
Next by Thread: Re: kern/41374: getnewvnode deadlock
Indexes:

Home | Main Index | Thread Index | Old Index