mem += -((uintptr_t)mem + sizeof(struct pthread)) & (libc.tls_align-1);
td = (pthread_t)mem;
mem += sizeof(struct pthread);
for (i=1, p=libc.tls_head; p; i++, p=p->next) {
mem += -((uintptr_t)mem + sizeof(struct pthread)) & (libc.tls_align-1);
td = (pthread_t)mem;
mem += sizeof(struct pthread);
for (i=1, p=libc.tls_head; p; i++, p=p->next) {