Re: [PATCH 00/21] kGraft

[One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk>]

On Wed, 25 Jun 2014 13:05:29 +0200
Jiri Slaby <jslaby@suse.cz> wrote:

> Hi,
> 
> this is a repost of the second round of RFC on kGraft, the linux
> kernel online patching developed at SUSE. This repost only widened the
> target audience for broader review, no code change happened.

-		wait_event_freezable(khubd_wait,
+		wait_event_freezable(khubd_wait,
  ({ kgr_task_safe(current);

The changes are somewhat ugly with all the kgraft crap leaking into plces
like jbd and freezer and usb. That says to me its not well isolated and
there must be a better way of hiding that kgr stuff so we don't have to
kepe putting more code into all the kthreads, and inevitably missing them
or have people getting it wrong.

You also wake up all the kthreads when some of them might not expect to
be woken and may not check for the case of a bogus wake.

You add instructions to various hotpaths (despite the CONFIG comment).
Have you done timing analysis of their impact when turned on ?

Can you explain a bit more about why the path you've chosen was used as
opposed to using the power management freeze/resume path. Would that not
be a lot less intrusive ?

Don't get me wrong - I'd like a good working ksplice/graft !

Alan