Put a new module at the end of the list intead of making new modules at the top of the list. find_module_all() start the hunt using the first entry on the list, if we assume that the modules which are first loaded are the most frequently looked for modules this should provide a tiny optimization. With a 8vcpu 8 GiB RAM kvm guest with kdevops with kdevops this saves about 4 MiB of RAM max use during the kmod tests 0008 which loops and runs loading a module in a loop through kthreads while not affecting the time to test. There could be some minor savings for systems that have repeated module requests for subsystems that are not yet optimized (ACPI frequency drivers come to mind) so in theory this could help there but that is just pure speculation. Signed-off-by: Luis Chamberlain <mcgrof@xxxxxxxxxx> --- kernel/module/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/module/main.c b/kernel/module/main.c index 0ad26455def2..32955b7819b3 100644 --- a/kernel/module/main.c +++ b/kernel/module/main.c @@ -2682,7 +2682,7 @@ static int add_unformed_module(struct module *mod) goto out; mod_update_bounds(mod); - list_add_rcu(&mod->list, &modules); + list_add_tail_rcu(&mod->list, &modules); mod_tree_insert(mod); err = 0; -- 2.39.1
