Re: [PATCH] Early init for security modules

[Chris Wright]

* Chris Wright ([e-mail]) wrote:
> As discussed before, here is a simple patch to allow for early
> initialization of security modules when compiled statically into the
> kernel.  The standard do_initcalls is too late for complete coverage of
> all filesystems and threads for example.  If this looks OK, I'd like to
> push it on to Linus.  Patch is against 2.5.69-bk.  It is tested on i386,
> and various arch maintainers are copied on relevant bits of patch.

This is just the arch specific linker bits for the early initialization
for security modules patch.  Does this look sane for these arches?

--- 1.7/arch/mips64/vmlinux.lds.S	Fri Feb 14 15:10:00 2003
+++ edited/arch/mips64/vmlinux.lds.S	Mon May 12 16:16:59 2003
@@ -53,6 +53,9 @@
   __con_initcall_start = .;
   .con_initcall.init : { *(.con_initcall.init) }
   __con_initcall_end = .;
+  __security_initcall_start = .;
+  .security_initcall.init : { *(.security_initcall.init) }
+  __security_initcall_end = .;
   . = ALIGN(4096);	/* Align double page for init_task_union */
   __init_end = .;
 
--- 1.8/arch/mips/vmlinux.lds.S	Fri Feb 14 15:09:55 2003
+++ edited/arch/mips/vmlinux.lds.S	Mon May 12 16:16:59 2003
@@ -54,6 +54,9 @@
   __con_initcall_start = .;
   .con_initcall.init : { *(.con_initcall.init) }
   __con_initcall_end = .;
+  __security_initcall_start = .;
+  .security_initcall.init : { *(.security_initcall.init) }
+  __security_initcall_end = .;
   . = ALIGN(4096);	/* Align double page for init_task_union */
   __init_end = .;