lkml.org 
[lkml]   [1998]   [Sep]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: showstopper race condition in sync() ??? [2.1.119|120]
On Thu, 17 Sep 1998, Cyrille Chepelov wrote:

>On Thu, 17 Sep 1998, Andrea Arcangeli wrote:
>
>> If you have a K6 as you said putting a spinlock can' t help you to avoid
>> the dquota deadlock.
>
>???? Isn't the K6 supposed to behave like a Pentium, at least
>interface-wise ?

As a Pentium _UP_. At least if something is not changed recently (usually
I am not in sync with harware news).

>(I may have employed the wrong terminology, if by 'spinlock' you (and the
>others) mean a hardware helper device for SMP machines. Sorry, I'm a
>rather recent convert. I meant bracketing the dquota [and the bdflush]
>with whatever Win32's EnterCriticalSection()/LeaveCriticalSection()
>translates to in Linux kernel code. Something to guarantee that only one
>thread can enter a particular piece of code at a time).

The point is that you are getting troubles with dquota on an UP hardware.
Yes? If so you can' t fix _your_ problem spinlokking some piece of code.
The kernel is not preemptive so you don' t need to care about nothig other
irq handlers in UP.

Andrea[s] Arcangeli


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

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.094 / U:0.300 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site