lkml.org 
[lkml]   [2011]   [May]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    SubjectRe: [PATCH 2/3] perf_events: fix validation of events using an extra reg (v2)
    From
    On Mon, May 23, 2011 at 9:47 AM, Peter Zijlstra <peterz@infradead.org> wrote:
    > On Fri, 2011-05-20 at 18:45 +0200, Stephane Eranian wrote:
    >> On Fri, May 20, 2011 at 5:05 PM, Peter Zijlstra <peterz@infradead.org> wrote:
    >> > On Fri, 2011-05-20 at 16:37 +0200, Stephane Eranian wrote:
    >> >> The is_fake field is necessary to avoid a lockdep issue having
    >> >> to do with interrupt masking and era->lock.
    >> >
    >> > Why not simply disable IRQs in both validate_* functions?
    >> >
    >> Could do that too if you think that's cleaner.
    >
    > Its saner than making that code path conditional, but an alternative
    > solution is to give the fake thing a different lock class.
    >
    Looks like using irqsave()/irqrestore() would be easier. Let me respin
    the patch series for this.


    \
     
     \ /
      Last update: 2011-05-23 11:17    [W:0.020 / U:11.832 seconds]
    ©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site