Minor problem/fixes with 1.3.96
Gonzalo Tornaria (gtornari@varela.reu.edu.uy)
Sun, 28 Apr 96 8:16:38 GMT
1) net/ipv4/Config.in
bash doesn't seems to like "if [...];then;fi". Menuconfig gets
confused too. Just comment it out:
--- net/ipv4/Config.in.old Sat Apr 27 16:10:37 1996
+++ net/ipv4/Config.in Sat Apr 27 16:32:57 1996
@@ -21,9 +21,9 @@
if [ "$CONFIG_NET_ALIAS" = "y" ]; then
tristate 'IP: aliasing support' CONFIG_IP_ALIAS
fi
-if [ "$CONFIG_KERNELD" = "y" ]; then
+#if [ "$CONFIG_KERNELD" = "y" ]; then
# bool 'IP: ARP daemon support (experimental)' CONFIG_ARPD
-fi
+#fi
comment '(it is safe to leave these untouched)'
bool 'IP: PC/TCP compatibility mode' CONFIG_INET_PCTCP
tristate 'IP: Reverse ARP' CONFIG_INET_RARP
2) drivers/block/rd.c
int rd_size was added to make ramdisk size boot time configurable.
But the variable would not be defined when rd is compiled as module,
but used anyway. Take it out of `#ifndef MODULE'. As a side-effect, you
can now `insmod rd rd_size=<size>'.
--- drivers/block/rd.c.old Sat Apr 27 15:44:12 1996
+++ drivers/block/rd.c Sat Apr 27 20:11:42 1996
@@ -90,11 +90,11 @@
* architecture-specific setup routine (from the stored bootsector
* information).
*/
+int rd_size = 4096; /* Size of the ramdisks */
#ifndef MODULE
int rd_doload = 0; /* 1 = load ramdisk, 0 = don't load */
int rd_prompt = 1; /* 1 = prompt for ramdisk, 0 = don't prompt */
int rd_image_start = 0; /* starting block # of image */
-int rd_size = 4096; /* Size of the ramdisks */
#ifdef CONFIG_BLK_DEV_INITRD
unsigned long initrd_start,initrd_end;
int mount_initrd = 1; /* zero if initrd should not be mounted */
3) nfs module
exit_mm is not exported. It should be added to ksyms... I don't
know if this is the rigth place. I don't know if this is right, I haven't
compiled this yet. I don't use nfs, so I couldn't even try it)
--- kernel/ksyms.c.old Sat Apr 27 01:03:54 1996
+++ kernel/ksyms.c Sat Apr 27 21:45:16 1996
@@ -120,6 +120,7 @@
X(do_munmap),
X(insert_vm_struct),
X(merge_segments),
+ X(exit_mm),
/* internal kernel memory management */
X(__get_free_pages),
--
Everything else looks fine... Also, the problem I had with 1.3.9[13],
that gcc would sleep forever if every process in the system is sleeping
seems to have gone.. At least I couldn't reproduce it. However, the other
problem still happens, namely "insmod sbpcd&sleep 1;rmmod sbpcd" oops'es..
(see my mail: "Reproducible Oops in 1.3.93").
Gonzalo