This is assertion is no longer necessary since we never record constants in the live set anyway. (They are never recorded in the initial live set, and constant bases are removed near line 2119)
Details
Details
Diff Detail
Diff Detail
Event Timeline
Comment Actions
Drop by comments.
test/Transforms/RewriteStatepointsForGC/constants.ll | ||
---|---|---|
107 | The bitcast seems unnecessary -- do we need it for reproducing the crash? |
Comment Actions
LGTM w/the test simplifications Sanjoy pointed out.
FTR, I've reviewed the code which handles constant base and derived pointers in the entire file and it appears to be correct. I had thought we'd already removed this assert and am more than a bit surprised to find we hadn't.
The bitcast seems unnecessary -- do we need it for reproducing the crash?