Fwd: Re: [Valgrind-users] valgrind's internal instruction cache (translation table)

[Michael E. L. <me...@co...>]

doh, forgot to cc the list.

----------  Forwarded Message  ----------

Subject: Re: [Valgrind-users] valgrind's internal instruction cache=20
(translation table)
Date: Fri, 7 May 2004 12:46:19 -0400
From: "Michael E. Locasto" <me...@co...>
To: Nicholas Nethercote <nj...@ca...>

On Friday 07 May 2004 12:37 pm, Nicholas Nethercote wrote:
> On Fri, 7 May 2004, Michael E. Locasto wrote:
> > I was wondering (before I dig into the source and fiddle about) if th=
e
> > size of Valgrind's instruction cache was adjustable
>
> I think it's a build-time constant, it may be governed by the VG_TT_SIZ=
E
> constant in coregrind/vg_transtab.c.

Yep, as I found out. There are also a bunch of assert()'s that deal with =
the
bounds on the table size and number of free entries.

> Why do you want to change this?

I was looking for a way to disable the translation table entirely; that i=
s,
get VG to instrument just one instruction (or maybe just one basic block)=
 at
a time and then forget about it. I'm trying to compare VG to a system tha=
t
translates one instruction at a time and then does not cache the translat=
ion.

I could try to mimic this by forcing VG to flush frequently (via a low
VG_TT_LIMIT_PERCENT value) but that introduces the overhead of doing the
LRU into the measurements.

Thanks for the pointer and your help so far...

Cheers,
Michael

-------------------------------------------------------