> Hi, > I am right assuming that if we get a page fault for a user mode virtual > address, curlwp can't be NULL (e.g. on x86, in the > !KERNELMODE(frame->tf_cs, frame->tf_eflags) or frame->tf_trapno contains > T_USER) ? i think you are right. page fault (in x86 and MI sense, at least) with curlwp == NULL is a bug. YAMAMOTO Takashi