On Tue, Aug 09, 2005 at 11:29:31AM -0700, Linus Torvalds wrote: > > > On Tue, 9 Aug 2005, Peter Hagervall wrote: > > > > Yes, assuming c8b9ce8f1ef27a0e77c87ea1581c6e2b33753e7a is the current > > HEAD. My arch is x86. > > Yup, that's what I tested too. > > > The straightforward way of using it is just prepending the command line > > with 'valgrind', so in this case I have: > > I just get "missing --tool option" from that. > > I did a "valgrind --tool=memcheck" thing, and yes, I see two invalid reads > in clean_up_insns too. > > I suspect that valgrind could find better errors if it was told about > "allocate()" and "free_one_entry()". As it is, it won't notice any normal > allocation mis-use, because it just sees them as one mmap. The > clean_up_insns one seems to be from some of the (rare) malloc/free users, > namely the ptr_list allocations. > > The fact that the pointer list that is warned about was free'd by > "simplify_switch()" is interesting - that sure as hell is some very > special code. And valgrind did _not_ complain about kernel/sched.c, for > example, so it's certainly not a normal error. > > > I'll try to track down the problem later tonight. > > Thanks. I'll see if I get around to looking closer later too. > > Linus
I think I have it narrowed down a bit now, it seems that ac8ce3636d83e25d8a316fdce221e36c04c02dc1 is OK, whereas its child, 972609f0d10e3bcb10a9132802b2edb8b01d305f is not. valgrind still complains about invalid reads in clean_up_insns, but at least it doesn't segfault. -- Peter Hagervall - To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
