2013/10/22 Urias McCullough <umccullough@xxxxxxxxx>: >> It's not silly. With the previous version of the code, if we failed to >> map the rsdt, it would short circuit at the rsdt != NULL, and >> consequently fall through and crash trying to dereference >> rsdt->length. > > > FWIW, this was also CID 610890, which I'll close now :) Isn't it leaking rsdtArea now? How about rsdt == null || strncmp... ? /Fredrik Holmqvist, TQH