Messages in this thread | | | Date | Wed, 05 Sep 2012 17:37:25 -0700 | From | Darren Hart <> | Subject | Re: [PATCH] pch_uart: Add eg20t_port lock field, avoid recursive spinlocks |
| |
On 09/05/2012 05:25 PM, Darren Hart wrote: > > > On 09/05/2012 05:18 PM, Greg Kroah-Hartman wrote: >> On Wed, Sep 05, 2012 at 05:14:48PM -0700, Greg Kroah-Hartman wrote: >>> On Wed, Sep 05, 2012 at 05:04:07PM -0700, Darren Hart wrote: >>>> The following patch has been included in linux-next >>>> (fe89def79c48e2149abdd1e816523e69a9067191) but has not yet landed in mainline >>>> nor been queued for stable so far as I can determine. This patch addresses a >>>> deadlock in mainline and is a prerequisite for an additional fix required by the >>>> PREEMPT_RT kernel. Can we get this pulled into 3.4.11 please? >>> >>> 3.4.11? It has to hit Linus's tree first. >>> >>>> Perhaps I am >>>> jumping the gun, but this patch was originally pulled on June 19, 2012. >>> >>> Remember, we missed a pull cycle for tty due to other problems, I >>> thought I picked all of the different pieces needed for 3.6, but I must >>> of missed this one. >> >> Nope, it made it, it is commit 2588aba002d14e938c2f56d299ecf3e7ce1302a5. > > Doh, I pulled master and stable, but only checked stable. Sigh. My > apologies Greg. > >> >> Now, do you want that patch in the -stable releases? If so, how far >> back? :) > > Yes, back to 3.0 would be ideal. It needs mangling for 3.2 and back > though. I will send patches for 3.4, 3.2 and possibly 3.0 following the > stable_kernel_rules.txt procedure.
On second thought, there are way too many changes to pch_uart that are required before this patch can really be applied prior to 3.4. I suspect these are not all appropriate for -stable. I'd be happy just getting this into 3.4.11. 2588aba002d14e938c2f56d299ecf3e7ce1302a5 cherry-picks cleanly to 3.4.
Thanks,
-- Darren Hart Intel Open Source Technology Center Yocto Project - Technical Lead - Linux Kernel
| |