This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Unreliable BFD caching heuristic
- From: Pedro Alves <palves at redhat dot com>
- To: lgustavo at codesourcery dot com
- Cc: "gdb at sourceware dot org" <gdb at sourceware dot org>, Tom Tromey <tromey at redhat dot com>, "Maciej W. Rozycki" <macro at codesourcery dot com>
- Date: Mon, 25 Nov 2013 18:21:04 +0000
- Subject: Re: Unreliable BFD caching heuristic
- Authentication-results: sourceware.org; auth=none
- References: <528E454F dot 6060003 at codesourcery dot com> <528E49B8 dot 8090409 at redhat dot com> <52938CB2 dot 8080700 at codesourcery dot com>
On 11/25/2013 05:45 PM, Luis Machado wrote:
> We can't rely on the timestamp due to some filesystems having 1 second
> or 2 seconds resolution. That doesn't seem enough.
We can't rely on timestamps to be sure the files are the same, but
we can assume that if the files have different timestamps, they're
not the same. IOW, if we have some other fallback file comparison
method that is more expensive, we only need to apply it if the
timestamps match; if they don't match, we can assume the file
is not the same. Sure, if the same file is accessed through
different filesystems, we can end up thinking we had two handles
for different files, but then all that's lost is we don't apply the
bfd sharing optimization in that rare scenario.
--
Pedro Alves