lkml.org 
[lkml]   [2009]   [May]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [Patch 06/12] Use the new wrapper routines to access debug registers in process/thread code
On Fri, May 29, 2009 at 12:49:03PM +0200, Frederic Weisbecker wrote:
> On Fri, May 29, 2009 at 02:31:46PM +0530, K.Prasad wrote:
> > On Thu, May 28, 2009 at 04:42:38PM +1000, David Gibson wrote:
> > > On Mon, May 11, 2009 at 05:23:44PM +0530, K.Prasad wrote:
> > > > From: Alan Stern <stern@rowland.harvard.edu>
> > > >
> > > > This patch enables the use of abstract debug registers in
> > > > process-handling routines.
> > >
> > > [snip]
> > > >
> > > > + p->thread.io_bitmap_ptr = NULL;
> > >
> > > Why is manipulating the io_bitmap_ptr relevant to debug register
> > > handling?
> >
> > I *re-read* the patch but was unable to find how this change had sneaked
> > in. It shouldn't be there although it is harmless.
>
>
> When I reviewed this patch, I also ended stucked on it.
> But actually I guess I found the sense, this is only for
> convenience.
>
> Look at the current copy_thread() in arch/x86/kernel/process32.c
>
> If p->thread.io_bitmap_ptr fails to be duplicated, we set
> p->thread.io_bitmap_max = 0 and return -ENOMEM
>
> Now look at the patch.
> If we fail to copy the hardware thread virtual registers we
> want to exit with io_bitmap_ptr = NULL
> If we fail to copy the io_bitmap, we want to free the breakpoint
> and exit.
> If we fail further, we want to free breakpoints and io_bitmap_ptr
>
> The out section then tries to:
>
> -free the breakpoints
> -free p->thread.io_bitmap_ptr
>
>
> So it's important to set io_bitmap_ptr to NULL so that
> we know whether we have to release it or not.
>
>

aah...yes. It tricked me! It is needed to bring the desired
error-return behaviour of copy_thread(). Please ignore this patch (the
updation of the comments can be brought in through a separate
enhancement patch...see below).

> > Hi Frederic,
> > I am attaching a new version of this patch 06/12 that:
> >
> > - removes the line that assigns NULL to "p->thread.io_bitmap_ptr"
>
>
> Dangerous. Unless p->thread.io_bitmap_ptr is already zeroed out
> at this stage?
>
>
> > - Updates the comment in __switch_to() function which was stale (was
> > relevant when 'last_debugged_task' was used to detect lazy debug
> > register switching).
> >
> > Kindly integrate this version in lieu of the older version sent here:
> > http://lkml.org/lkml/2009/5/21/149.
>
>
> Ok. Well it would be nice if you resend the whole series actually :)
> Do you have another fix pending?
>
> Thanks!
>

In the process of responding to David Gibson's comments I agreed to
a few minor/cosmetic changes - say like updation of comments, renaming
functions or variables, etc.

Given that the patchset is on the verge of integration into -tip tree, I
would prefer them to be done through a separate patch (on -tip) for
enhancement. Kindly let me know what you think about proceeding this
way.

Thanks,
K.Prasad






\
 
 \ /
  Last update: 2009-05-29 15:55    [W:0.042 / U:0.296 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site