I'm wondering if this is an endian issue with the UID translation? Do you know which endian mode your aarch64 boxes are booted it? IBM ships on PPC, but I'm not sure they've tested recent versions, so something may have crept in.
THANK YOU for replying to the thread. I checked on this and both x86_64 and aarch64 are little endian. I built this test program to confirm including this one that configure sometimes uses: #include <inttypes.h> int main(int argc, char ** argv){ volatile uint32_t i=0x01234567; // return 0 for big endian, 1 for little endian. return (*((uint8_t*)(&i))) == 0x67; } It is returning results for little endian. sh-4.2# uname -m aarch64 sh-4.2# cc i.c sh-4.2# ./a.out sh-4.2# echo $? 1 Let me know if we have more ideas to try. Best, Erik