[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gluster-devel] posix-locks bugs
From: |
Anand Avati |
Subject: |
Re: [Gluster-devel] posix-locks bugs |
Date: |
Sat, 16 Feb 2008 16:12:11 +0530 |
Timo,
patch-666 fixes this. Thanks for the pointer!
avati
2007/12/18, Timo Sirainen <address@hidden>:
>
> new_posix_lock (struct flock *flock, transport_t *transport, pid_t
> client_pid)
> ..
> if (flock->l_len == 0)
> lock->fl_end = ULONG_MAX;
>
> fl_end is "signed long long" type pretty much everywhere. The main bug
> here is that with 64bit systems ULONG_MAX gets converted to -1, and
> l_len=0 locks break completely.
>
> With 32bit systems it also breaks if doing lock ranges above 4GB in a
> file, but this is probably less of a problem.
>
> So, this should work better:
>
> lock->fl_end = LLONG_MAX;
>
> (BTW. Why aren't PGP signatures allowed here? I got "The message's
> content type was not explicitly allowed")
>
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
--
If I traveled to the end of the rainbow
As Dame Fortune did intend,
Murphy would be there to tell me
The pot's at the other end.
- Re: [Gluster-devel] posix-locks bugs,
Anand Avati <=