Something seems to be: breaking alignment, changing the allocator used, or trashing that memory.

In my case, the address of the mutex is: 0x5603a3666630  (*1) below
and the surrounding memory:

0x5603a3666610: n\0\0\03\0\0\0\0\0\0\0(*2)\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd1\0\0\0\0\0\0\0
0x5603a3666630: (*1)\x01\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd\xdd
0x5603a3666650: \xdd\xdd\xdd\xdd\xdd\xdd\xdd\xddQ\0\0\0\0\0\0\0\x90ds\xa3\x03V\0\0\xf0\xc4\xfbq\a\x7f\0\0

The fun thing is that the nbytes field lines up at (*2) which is full of DEADBYTEs, and thus tries to find the tail of the allocation at p + 15987178197214944733  (which happily segfaults)