labath added a comment.
The running aspect is somewhat unfortunate, as it forces the input to be
compatible with the host system. I wouldn't be surprised if the test broke
somewhere (in fact, I would be surprised if it didn't), because ir already
hardcodes a lot of assumptions about the host system.
Given that the fix really is about how types are parsed, it should be
sufficient to run something like "type lookup" on these types and verifying
that. Right now I get
struct string {
unsigned long length;
void *ptr;
}
for `type lookup string`. I'm guessing that after your patch the `void *` will
change into something else. Alternatively, you could just check for the
existing formatting of "target variable utf8", but leave a note that this does
not check the precise formatting of the variable, just its utf8-ness. Then we
can just update the expectation when the pretty printer bug is fixed.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D79559/new/
https://reviews.llvm.org/D79559
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
