This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: valgrind support for glibc/kernel arm64 HWCAPS
On Thursday 06 July 2017 11:48 AM, Florian Weimer wrote:
> There is a page boundary check upfront, or this only happens after the
> pointer is aligned and the load cannot cross a page boundary.
>
>> It would be nice to get some more details of this one as well:
>>
>> (b) some of the normal versions are hyper-optimised, which fools
>> Memcheck and cause spurious value warnings. Our versions are
>> simpler.
>>
>> specifically whether it is x86-specific or true for other architectures
>> as well
>
> I'm not sure if this is what (b) is about, but for some instructions or
> instruction sequences it is hard to implement proper uninitialized value
> tracking, so that valgrind reports a dependency on uninitialized data
> which does not in fact exist because the bits are masked away in
> practice, but valgrind does not see this due to this problem.
OK, thanks for the explanation.
> But this isn't what I meant. There is an expectation that even after we
> add SSE5-based IFUNCs to glibc, you can still use current valgrind to
> run programs on SSE5-capable systems.
Sure, I don't disagree with that.
Siddhesh