Re: [PATCH] module: avoid calling synchronize_rcu()

[Luis Chamberlain]

On Wed, Mar 02, 2022 at 10:14:00AM +0100, Miroslav Benes wrote:
> Hi,
> 
> On Wed, 2 Mar 2022, cgel.zte@xxxxxxxxx wrote:
> 
> > From: Lv Ruyi (CGEL ZTE) <lv.ruyi@xxxxxxxxxx>
> > 
> > Kfree_rcu() usually results in even simpler code than does 
> > synchronize_rcu() without synchronize_rcu()'s multi-millisecond
> > latency, so replace synchronize_rcu() with kfree_rcu().
> > 
> > Reported-by: Zeal Robot <zealci@xxxxxxxxxx>
> > Signed-off-by: Lv Ruyi (CGEL ZTE) <lv.ruyi@xxxxxxxxxx>
> > ---
> >  kernel/module.c | 3 +--
> >  1 file changed, 1 insertion(+), 2 deletions(-)
> > 
> > diff --git a/kernel/module.c b/kernel/module.c
> > index 6cea788fd965..767b5f9e5819 100644
> > --- a/kernel/module.c
> > +++ b/kernel/module.c
> > @@ -4138,8 +4138,7 @@ static int load_module(struct load_info *info, const char __user *uargs,
> >   ddebug_cleanup:
> >  	ftrace_release_mod(mod);
> >  	dynamic_debug_remove(mod, info->debug);
> > -	synchronize_rcu();
> > -	kfree(mod->args);
> > +	kfree_rcu(mod->args);
> 
> this has been proposed already. synchronize_rcu() and kfree() here are not 
> really tied together. See the discussion at 
> https://lore.kernel.org/all/alpine.LSU.2.21.2111301132220.3922@xxxxxxxxxxxxx/T/#u

Aaron, can you add a nice comment here to explain this while at it?
Otherwise this will be lost tribal knowledge.

Lv Ruyi, please open source your Zeal Robot. Thanks!

  Luis