Yes a command line option is a possibility. Another
is for mempool alloc to check that the memory used is marked
noaccess initially and free to check that it is not
noaccess which feels as if it should cover basically
the same sort of problems as check_mempool_sane
does rather slowly (and indeed it seems to have checks on
itself in that it verifies the sort it does sorts!). But I
don't know if that would cause issues with other peoples
use of mempools.
From: Bart Van Assche [mailto:bart.vanassche@...]
Sent: Tuesday, March 25, 2008 12:26 AM
To: Solomon, Bernard
Cc: valgrind-users@...; Julian Seward
Subject: Re: [Valgrind-users] Couple of questions
On Tue, Mar 25, 2008 at 12:42 AM, Solomon, Bernard <bernard@...>
> I am not quite sure where I am supposed to post patches.
> This is the one removing the excessive checks though of
> course other options are possible (check every so often,
> have a command line switch etc.)
> I was using 3.3 but SVN seems the same.
> The 64 bit change I'll send separately if this is the
> right place to send it.
The check_mempool_sane() calls might be very useful to people who are
debugging their added memory allocation client requests. Maybe it's
better to modify the check_mempool_sane() function itself and to keep
the check_mempool_sane() calls. E.g. a command line option could be
added to Valgrind that lets the user specify whether or not to let
checking happen when check_mempool_sane() is called.