converting cap_set_pg() to for_each_task_pid()

I have a pending patch that converts cap_set_pg() to the for_each_task_pid() API. Could you review this, and if it pass, include it in your tree?

Thanks,
Bill

cap_set_pg() wants to find all processes in a given process group. This converts it to use for_each_task_pid().

 capability.c |   16 +++++++++-------
 1 files changed, 9 insertions(+), 7 deletions(-)

diff -urpN wli-2.5.51-bk1-4/kernel/capability.c wli-2.5.51-bk1-5/kernel/capability.c
@@ -84,13 +84,15 @@ static inline void cap_set_pg(int pgrp, 
 			      kernel_cap_t *inheritable,
 			      kernel_cap_t *permitted)
 {
-     task_t *g, *target;
-
-     do_each_thread(g, target) {
-             if (target->pgrp != pgrp)
-                     continue;
-	     security_capset_set(target, effective, inheritable, permitted);
-     } while_each_thread(g, target);
+	task_t *g, *target;
+	struct list_head *l;
+	struct pid *pid;
+
+	for_each_task_pid(pgrp, PIDTYPE_PGID, g, l, pid) {
+		target = g;
+		while_each_thread(g, target)
+			security_capset_set(target, effective, inheritable, permitted);
+	}
 }
 
 /*

This is functionally the same, right?
And if so, why do the change, is this now faster somehow?

thanks,

greg k-h

It is functionally equivalent. The motivation is not so much performance as API conversion. I am personally taking it upon myself to deprecate for_each_process() and do_each_thread() and update the mergeably correctible callers. The deprecation motives are both efficiency and correctness related.

It should improve performance in whatever sense performance is interesting from the point of view of this function; however, it is not motivated by that fact (and performance is not of interest with respect to most security functions). It is API conversion.

Thanks,
Bill

I have no problem with this change.
-chris

Linux Security Modules     http://lsm.immunix.org     http://lsm.bkbits.net

Applied to my tree, I'll send it to Linus in a bit.

thanks,

greg k-h