|
From: Jeremy F. <je...@go...> - 2002-11-18 17:44:49
|
On Mon, 2002-11-18 at 08:03, Nicholas Nethercote wrote: > It's performance is better than Valgrind's; without instrumentation > programs under Linux run about 1.0--2.0 times slower, with Valgrind it's > more like 5 times slower. > > It's at www.cag.lcs.mit.edu/dynamorio/. The slides from the ASPLOS > tutorial are very interesting. The whole thing was only released in June. Looks interesting. It seems their biggest performance win was using BB chaining. They also implemented a trace cache, though it seems to be a fair bit of complexity for hardly any incremental improvement. > It clearly has some advantages over Valgrind (eg. performance, runs under > Windows). Now I'm trying to work out what advantages has over it... V is under the GPL, as opposed to: permission to copy and use (but not distribute to others) this software for the sole purpose of internal evaluation is hereby granted without fee. Oh, and they don't seem to have released any source... J |