Hi,

> Compared to other SPEC benchmarks, the fraction of overhead due to checks
> is
> > very low for gcc. Much overhead seems to come from heap poisoning and
> from
> > the quarantine queue, but there are also some 30% of overhead that hide
> > elsewhere… I suspect that the allocator might be to blame. This suspicion
> > arose because perf showed a high number of samples in the allocator.
>


> Have you compared ASan perf stats to those with glibc allocator or
> tcmalloc?
>

That comparison was against glibc's allocator. I haven't tried tcmalloc.
Also note that gcc is a single-threaded benchmark AFAIK.

I'd like to try other benchmarks and other allocators... do you have a good
idea of what else I should try?

In ASan allocator objects belonging to different size classes cannot
> be adjacent in memory.
>

OK, that makes sense. In this case the redzone optimizations should indeed
be sound.

A slightly unrelated thing that we're interested in is heap
> randomization under ASan.
>

I see. Not directly related to what I'm trying to achieve, but an
interesting idea in general! I don't have many thoughts regarding that
right now, except that the need for deterministic replay is very important.
Users should have the possibility to manually set a seed, so they can
reproduce an issue once it occurred.

Also, this might be very interesting for the stack, too. I think even a
slightly randomized stack layout (through random redzone sizes or variable
reordering) could make it harder to write exploits. I know ASan is a
debugging tool more than a security hardening tool, but I'd still find this
interesting.

Cheers,
Jonas

-- 
You received this message because you are subscribed to the Google Groups 
"address-sanitizer" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to