lkml.org 
[lkml]   [2008]   [Jan]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [RFC][PATCH] VFS: create /proc/<pid>/mountinfo
From
Date
> Pavel Machek wrote:
> > On Sun 2008-01-20 09:23:00, Miklos Szeredi wrote:
> >>> Miklos Szeredi wrote:
> >>>> - for mount ID's use IDA (from the IDR library) instead of a 32bit
> >>>> counter, which could overflow
> >>> IDAs tend to get reused quickly, which can cause race conditions. Any
> >>> reason not to just use a 64-bit counter?
> >> They tend to become hard to parse/compare for humans after a while.
> >> And all this is basically only for humans, so race conditions don't
> >> really matter. Also a changed mount with a reused ID is easily
> >> identified by comparing the other fields.
> >
> > Hmm, smart humans only compare last few digits if they don't care
> > about 100% reliability, and dumb software compares 64bits easily...
> > Pavel
>
> Indeed.
>
> And this is most certainly NOT only for humans, and race conditions most
> certainly matter.

Use case please? What will this info be used for, other than for
feedback for humans about the state of the propagation tree?

Face it, userspace is inherently racy. Inode numbers, device numbers,
whatever are being reused all the time, we live with it, even if it's
programs, and not just humans.

But it's not even an important design decision, the ID allocation can
be swapped at any time. If you insist, I'll change it to a 64bit
counter, and it'll just suck a little more, but no permanent damage
done ;)

Miklos


\
 
 \ /
  Last update: 2008-01-23 20:35    [W:0.065 / U:0.220 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site