lkml.org 
[lkml]   [2010]   [Nov]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 14/14] ptrace: remove the extra wake_up_process() from ptrace_detach()
Hello,

On 11/28/2010 09:44 PM, Oleg Nesterov wrote:
> Today I lost the concentration at 13/14 ;)
>
> Will continue tomorrow. As for this patch,

Eh, well, you've already found enough holes in the patchset. It's
like a swiss cheese. I'll update and repost.

> On 11/26, Tejun Heo wrote:
>>
>> This wake_up_process() has a turbulent history. This is a remnant
>> from ancient ptrace implementation and patently wrong. Commit
>> 95a3540d (ptrace_detach: the wrong wakeup breaks the ERESTARTxxx
>> logic) removed it
>
> Yes. This obviously means I personally like this change. In fact,
> I never understood this wakeup, and I was glad to find the reason
> to send the patch.

Another thing is that the use of wake_up_process() in ptrace is rather
scary even in ptrace_resume(). It will wake up even uninterruptible
sleeps. Is there any reason it's not just waking up
TASK_INTERRUPTIBLE | STOPPED | TRACED?

Thanks.

--
tejun


\
 
 \ /
  Last update: 2010-11-29 14:57    [W:5.739 / U:0.060 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site