Re: [rfc][patch] dynamic data structure switching

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Peter Zijlstra
Date: Monday, September 10, 2007 - 4:55 am

On Sun, 2007-09-02 at 20:27 +0200, Nick Piggin wrote:


With the seqlock protecting the whole xfer a lookup for a non-existing
item might spin here for a very long time.

Maybe we should create this sleeping seqlock we talked and provide both
a sleeping and non sleeping version of this lookup function.


So its up to the caller to take resize_mutex, but there is no mention of
this requirement. Why cannot it be taken inside this function?


<snip>



-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[rfc][patch] dynamic data structure switching, Nick Piggin, (Sun Sep 2, 11:27 am)
Re: [rfc][patch] dynamic data structure switching, Nick Piggin, (Sun Sep 2, 11:36 am)
Re: [rfc][patch] dynamic data structure switching, Peter Zijlstra, (Mon Sep 10, 4:55 am)
Re: [rfc][patch] dynamic data structure switching, Nick Piggin, (Mon Sep 10, 6:39 am)