| Subject | Re: [PATCH v2 17/43] microblaze: Use set_current_blocked() and block_sigmask() | From | Matt Fleming <> | Date | Wed, 24 Aug 2011 11:21:27 +0100 |
| |
On Wed, 2011-08-24 at 10:49 +0200, Michal Simek wrote: > > v2 of this patch depends on "[PATCH 01/43] signal: Add block_sigmask() > > for adding sigmask to current->blocked" so they need to go through the > > same tree but this patch would benefit from some maintainer ACK's. > > Please add there my ACK to it too.
OK will do, thanks.
> > > > Michal, I dropped your Acked-by because I felt this patch changed > > quite dramatically since v1 and warrants another review. I hope that's > > OK. > > That's fine. I have run LTP tests and results look good.
Excellent!
> I will add patches 14-16/43 v2 to microblaze next branch keep them > for a while. I will propose them to Linus tree for v3.2.
Actually, I just realised this is where it gets a little complicated. If you take those 3 patches and they're not in Oleg's tree, [PATCH 17/43] doesn't apply cleanly. Fixing 17/43 up so it applies without patches 14-16 would result in something like this in handle_signal(),
if (!(ka->sa.sa_flags & SA_NODEFER)) { block_sigmask(ka, sig); } return 1;
Which Linus would have to fixup to its original form when he merges the two trees. I think all the microblaze patches should go through Oleg's tree, but it's really up to you two and Linus.
Oleg, if Michal does take patches 14-16 I can send you an updated patch to apply.
-- Matt Fleming, Intel Open Source Technology Center
|