|
From: Michael E. L. <me...@co...> - 2004-05-07 16:48:05
|
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 ------------------------------------------------------- |