[PATCH 2 of 2] util: drop statmtimesec

[Yuya Nishihara]

On Thu, 19 Nov 2015 23:00:00 -0800, Pierre-Yves David wrote:
> On 11/19/2015 03:04 PM, Yuya Nishihara wrote:
> > On Thu, 19 Nov 2015 16:14:48 -0500, Bryan O'Sullivan wrote:
> >> On Thu, Nov 19, 2015 at 3:59 PM, Laurent Charignon <lcharignon at fb.com>
> >> wrote:
> >>
> >>> That looks good to me, I will let Bryan, Augie or Pierre-Yves have another
> >>> look.
> >>>
> >>
> >> Ship it!
> >
> > Doesn't this disable the nsec filecache accuracy?
> > I'll check it later.
> 
> Can you elaborate on your concern here? I'll wait on your feedback to
> move forward with this patch.

Matt already pointed out it. (I didn't read the commit message this morning,
sorry.)

"This has some impact on our file cache validation code in that it lowers
timestamp resolution. But as we still have to deal with low-resolution
filesystems, we're not relying on this anyway."

Well, it's true our file cache has a problem on HFS+ or ext3, but float mtime
can avoid problems on modern filesystems now. And we haven't come up with an
alternative solution yet.

So, instead of disabling float timestamps globally, perhaps we can reimplement
osutil.stat() and lstat() and use them where integer timestamps are necessary.
I'll try it.