Re: [PATCH 1/3] jump label: add enabled/disabled state to jumplabel key entries

[Mathieu Desnoyers]

* Steven Rostedt (rostedt@xxxxxxxxxxx) wrote:
> On Tue, 2010-11-23 at 18:43 -0500, Mathieu Desnoyers wrote:
> > * Jason Baron (jbaron@xxxxxxxxxx) wrote:
> > [...]
> > > +static void update_jump_label_module(struct module *mod)
> > > +{
> > > +	struct hlist_head *head;
> > > +	struct hlist_node *node, *node_next, *module_node, *module_node_next;
> > > +	struct jump_label_entry *e;
> > > +	struct jump_label_module_entry *e_module;
> > > +	struct jump_entry *iter;
> > > +	int i, count;
> > > +
> > > +	/* if the module doesn't have jump label entries, just return */
> > > +	if (!mod->num_jump_entries)
> > > +		return;
> > > +
> > > +	for (i = 0; i < JUMP_LABEL_TABLE_SIZE; i++) {
> > > +		head = &jump_label_table[i];
> > > +		hlist_for_each_entry_safe(e, node, node_next, head, hlist) {
> > > +			if (!e->enabled)
> > > +				continue;
> > > +			hlist_for_each_entry_safe(e_module, module_node,
> > > +						  module_node_next,
> > > +						  &(e->modules), hlist) {
> > > +				if (e_module->mod != mod)
> > > +					continue;
> > 
> > Ouch.
> > 
> > Could you iterate on the loaded/unloaded module jump labels and do hash
> > table lookups rather than doing this O(n^3) iteration ?
> 
> This does not look O(n^3) to me.
> 
> It's a hash table, thus the first two loops is just iterating O(n) items
> in the hash.

Good point.

> 
> And the third loop is all the modules that use a particular event.
> 
> So it is O(n*m)  where n is the number of events, and m is the number of
> modules attached to the events. And that's only if all events are used
> by those modules. The actual case is much smaller.

Still, I wonder if the O(n) iteration on all events could be shrinked to
an interation on only the events present in the loaded/unloaded module ?
I think I did something like that for immediate values already. It might
apply (or not) here, just a thought.

Thanks,

Mathieu


-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/