It seems that Al Viro's patch to support large /proc/mounts, etc., has a
small mistake that broke (at least some) compiles where CONFIG_MODULES is
not defined... (That's the bug I reported to the list earlier tonight
FWIW.)
Here's a tested (on two systems, one with modules and one without) patch
against 2.4.15-pre3. The patch also seems to apply to 2.4.13-ac8, although
I have not tested it with that kernel version.
-Barry K. Nathan <[email protected]>
diff -ruN linux-2.4.15-pre3/fs/proc/proc_misc.c linux-2.4.15-pre3-bkn1/fs/proc/proc_misc.c
--- linux-2.4.15-pre3/fs/proc/proc_misc.c Sun Nov 11 16:43:57 2001
+++ linux-2.4.15-pre3-bkn1/fs/proc/proc_misc.c Sun Nov 11 20:32:11 2001
@@ -568,9 +568,11 @@
entry = create_proc_entry("mounts", 0, NULL);
if (entry)
entry->proc_fops = &proc_mounts_operations;
+#ifdef CONFIG_MODULES
entry = create_proc_entry("ksyms", 0, NULL);
if (entry)
entry->proc_fops = &proc_ksyms_operations;
+#endif
proc_root_kcore = create_proc_entry("kcore", S_IRUSR, NULL);
if (proc_root_kcore) {
proc_root_kcore->proc_fops = &proc_kcore_operations;