wangpc-pp wrote: > > Based on my rough understanding, this is expected? > > What do you mean? > > Isn't this test needs to be updated or disabled?
I think this ASAN failure is not caused by this PR because these memorys are allocated/deallocated by `memory_resource` directly, **there is nothing about `new`/`delete`**. We allocate 50 bytes via `void* ret = r1.allocate(50);` and deallocate 1 byte via `r1.deallocate(ret, 1);` and that's the reason why ASAN failed, because the size of the allocated type (50 bytes) and the size of the deallocated type (1 byte) are really not matched. I don't know if this is what we want to test, if so, we may add `// UNSUPPORTED: asan` to skip this test in ASAN builds. cc @ldionne @Quuxplusone https://github.com/llvm/llvm-project/pull/90373 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
