On 10/3/19 4:57 PM, Jakub Jelinek wrote:
On Thu, Oct 03, 2019 at 03:57:17PM -0400, Jason Merrill wrote:
That sounds better.
Ok.
And, second comment, not 100% sure about the unions, e.g. whether we
shouldn't somehow try to figure what is the active union member and only use
the active one rather than trying all.
Hmm, good question. Can we get away with not recursing for unions? Trying
all of them might end up choosing an inactive member that shares a common
initial sequence with the active member. Which is OK for reads, but not for
writes.
Nothing in check-c++-all cares about unions there, so like this?
Worst case if we have a testcase that will break because of that
we'll need to figure out what to do. But in unions it could very well just
find an inactive member without common initial sequence that just happens to
have the right type at the same offset.
So, is the following ok for trunk if it passes full bootstrap/regtest?
In attachment is the interdiff from the previous patch.
OK, thanks.
Jason
