Re: [Stable-review] [48/70] vmscan: all_unreclaimable() use zone->all_unreclaimable as a name

[KOSAKI Motohiro]

> On Thu, 2011-04-21 at 13:24 +0900, KOSAKI Motohiro wrote:
> > > On Tue, 2011-04-19 at 13:08 -0700, Greg KH wrote:
> > > > 2.6.38-stable review patch.  If anyone has any objections, please let us know.
> > > > 
> > > > ------------------
> > > > 
> > > > From: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
> > > > 
> > > > commit 929bea7c714220fc76ce3f75bef9056477c28e74 upstream.
> > > > 
> > > > all_unreclaimable check in direct reclaim has been introduced at 2.6.19
> > > > by following commit.
> > > > 
> > > > 	2006 Sep 25; commit 408d8544; oom: use unreclaimable info
> > > > 
> > > > And it went through strange history. firstly, following commit broke
> > > > the logic unintentionally.
> > > > 
> > > > 	2008 Apr 29; commit a41f24ea; page allocator: smarter retry of
> > > > 				      costly-order allocations
> > > [...]
> > > 
> > > So presumably this needs to be fixed in 2.6.32.y and other longterm
> > > series as well.  Though there seems to be a whole series of fixes
> > > required in 2.6.32.y!
> > > 
> > > Are you going to look after that, or should someone else prepare
> > > backports?  (I'm certainly not volunteering - I don't have the VM
> > > knowledge to work out what needs doing.)
> > 
> > Hi Ben
> > 
> > Honestly, I didn't prepare. If my remember is correct, you are debian 
> > guy. So, Can I think the backport 2.6.32.y help debian people? If so,
> > it's good thing to increase my priority to do this.
> 
> Most of the 'enterprise' and long-term supported distributions (Debian,
> Ubuntu, SLE, OEL and RHEL) have kernels based on 2.6.32.  RH seem to be
> doing their own thing but the rest of us are using 2.6.32.y as a basis.

I've finished see current head of longterm-2.6.32. It has

        2006 Sep 25; commit 408d8544; oom: use unreclaimable info
        2008 Apr 29; commit a41f24ea; page allocator: smarter retry of
                                      costly-order allocations

and, doesn't have

        2010 Jun 04; commit bb21c7ce; vmscan: fix do_try_to_free_pages()
                                      return value when priority==0
        2010 Sep 22: commit d1908362: vmscan: check all_unreclaimable
                                      in direct reclaim path
        2011 Apr 14 commit 929bea7c7: vmscan: all_unreclaimable() use 
                                      zone->all_unreclaimable as a name

Then, the code mean
 1) commit 408d8544 doesn't works as well.
 2) But, there is no hangup risk as commit 929bea7c7 described.

So, I think there is no worth to backport. Two years no bug report mean
it's no big matter. And we can't make a patch which include the above
three patch and its dependencies smaller than 100 lines.

Thanks.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/