|
From: Jeremy F. <je...@go...> - 2005-02-14 17:57:47
|
On Mon, 2005-02-14 at 11:54 +0000, Julian Seward wrote: > Frankly I'm unsympathetic to the view that even after a program > does a bad write which is reported by Memcheck, it should not > actually trash Valgrind. My view is, once a bad write is > reported, you can just forget the rest of your run on the basis > that V _might_ have gotten trashed. I think that's a bit simplistic. I tend to overlook reports coming from other people's code in my program if I'm fairly sure they're not relevant to the particular problem I'm trying to track down; if I know that they're not trashing Valgrind, then that helps. I would always run with pointercheck enabled; the only time I disable it is with programs which use ioctls which create mmaps outside the client address space. > I said as much in the valgrind-1.0 release notes, and I can't > remember getting complaints about this behaviour. Yeah, I'm pretty sure that because people ignored the advice. We regularly get "bug" reports from people saying that Valgrind crashed after hundreds of client warnings. J |