lkml.org 
[lkml]   [2009]   [Oct]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 6/9] swap_info: swap_map of chars not shorts
On Thu, 15 Oct 2009 23:17:20 +0100 (BST)
Hugh Dickins <hugh.dickins@tiscali.co.uk> wrote:

> On Thu, 15 Oct 2009, KAMEZAWA Hiroyuki wrote:
> > On Thu, 15 Oct 2009 01:53:52 +0100 (BST)
> > Hugh Dickins <hugh.dickins@tiscali.co.uk> wrote:
> > > @@ -1175,6 +1175,12 @@ static int try_to_unuse(unsigned int typ
> > > * If that's wrong, then we should worry more about
> > > * exit_mmap() and do_munmap() cases described above:
> > > * we might be resetting SWAP_MAP_MAX too early here.
> > > + *
> > > + * Yes, that's wrong: though very unlikely, swap count 0x7ffe
> > > + * could surely occur if pid_max raised from PID_MAX_DEFAULT;
> >
> > Just a nitpick.
> >
> > Hmm, logically, our MAX COUNT is 0x7e after this patch. Then, how about not
> > mentioning to 0x7ffe and PID_MAX ? as..
> >
> > Yes, that's wrong: we now use SWAP_MAP_MAX as 0x7e, very easy to overflow.
> > next patch will...
>
> Perhaps we're reading it differently: I was there inserting a comment
> on what was already said above (with no wish to change that existing
> comment), then going on (immediately below) to mention how this patch
> is now lowering SWAP_MAP_MAX to 0x7e, making the situation even worse,
> but no worries because the next patch fixes it.
>
yes.

> If you are seeing a nit there, I'm afraid it's one too small for my
> eye!

I don't think it's very troublesome, but in these days, people seems to love
"bisect", Then, comments for change and comments for code should be divided, IMHO.

> And the lifetime of this comment, in Linus's git history, will
> be (I'm guessing) a fraction of a second - becoming a non-issue, it
> rightly gets deleted in the next patch.

ya, thanks.

Regards,
-Kame



\
 
 \ /
  Last update: 2009-10-16 02:05    [W:0.093 / U:2.892 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site