Re: [PATCH RT 2/5] a few open coded completions

[Ben Hutchings]

On Mon, 2014-03-10 at 14:14 -0400, Steven Rostedt wrote:
> On Mon, 10 Mar 2014 17:52:42 +0000
> Ben Hutchings <ben@xxxxxxxxxxxxxxx> wrote:
> 
> > On Mon, 2014-03-10 at 10:29 -0400, Steven Rostedt wrote:
> > > On Sun, 09 Mar 2014 00:27:21 +0000
> > > Ben Hutchings <ben@xxxxxxxxxxxxxxx> wrote:
> > > 
> > > > On Wed, 2013-12-11 at 14:49 -0500, Steven Rostedt wrote:
> > > > > Here's the new patch:
> > > > 
> > > > Doesn't add #includes and doesn't change any types, so of course it
> > > > doesn't compile.  WTF?
> > > > 
> > > 
> > > Hi Ben,
> > > 
> > > This is an old thread. Which kernel did you apply it to. Before I send
> > > any patch set out, I run a small test (heavy stress test, and also
> > > compile several different configs). This patch was against 3.10-rt, and
> > > may have already had the includes needed. I know the original patch
> > > compiled, but I may not have tested the second patch.
> > > 
> > > Do you need this patch? Or did I miss something for 3.2-rt?
> > 
> > You applied it to 3.2-rt:
> > 
> > commit 916c8d2de3c4759f57426fa3cfee63b491e88939
> > Author: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
> > Date:   Mon Oct 28 11:50:06 2013 +0100
> > 
> >     swait: Add a few more users
> > 
> > and 3.4-rt, 3.8-rt, 3.10-rt.  So far as I can see, it is broken on all
> > of those branches, though I only actually tried building 3.2-rt.
> > 
> 
> Can you send me your config. I may just not be hitting it. I've booted
> and ran all these kernels I've posted.
> 
> I don't run a make allmodconfig as I do on my upstream work. Maybe I'll
> have to add that to the mix.

You can enable all the affected code with:

CONFIG_HERMES=m
CONFIG_ORINOCO_USB=m
CONFIG_USB_GADGET=m
CONFIG_USB_FUNCTIONFS=m
CONFIG_USB_GADGETFS=m

Ben.

-- 
Ben Hutchings
Any sufficiently advanced bug is indistinguishable from a feature.

Attachment: signature.asc
Description: This is a digitally signed message part