From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailserv2.iuinc.com (IDENT:qmailr@mailserv2.iuinc.com [206.245.164.55]) by puffin.external.hp.com (8.9.3/8.9.3) with SMTP id HAA15166 for ; Tue, 1 Feb 2000 07:01:23 -0700 Received: from udlkern.fc.hp.com (jsm@udlkern.fc.hp.com [15.1.52.48]) by hpfcla.fc.hp.com (8.9.1/8.9.1) with ESMTP id GAA19339 for ; Tue, 1 Feb 2000 06:02:20 -0700 (MST) Received: (from jsm@localhost) by udlkern.fc.hp.com (8.8.6 (PHNE_14041)/8.7.1) id GAA21104 for parisc-linux@thepuffingroup.com; Tue, 1 Feb 2000 06:02:19 -0700 (MST) Date: Tue, 1 Feb 2000 06:02:19 -0700 (MST) From: John Marvin Message-Id: <200002011302.GAA21104@udlkern.fc.hp.com> To: parisc-linux@thepuffingroup.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [parisc-linux] init_task.c dependency problem List-ID: With my most recent commit's, I made a change in a header file that arch/parisc/kernel/init_task.c depends on. However, due to the makefile games used to build this file (to properly align the task structure), the dependencies for init_task.c are not honored (the .depend file lists dependencies for init_task.o, but we build init_task0.o and init_task1.o instead). What this means is that init_task.c will not be rebuilt if you just cvs update your build area. In that case, the kernel will halt immediately after the "execve /sbin/init ...." message (i.e. it will print halted and then print a register dump). If you are building from a new area, or you make clean, you will not have this problem. A simple workaround (if you don't want to do a make clean) is to "touch arch/parisc/kernel/init_task.c" so that it will be rebuilt. You could also just remove init_task0.o and init_task1.o to force the rebuild. Yesterday's commits allowed fork() from userland to work. Today's commits allow exec() from userland to work. John Marvin jsm@fc.hp.com