Re: [gentoo-kernel] 3.12 Kernels : Regression in the handling of usbatm debug traces

[Greg KH]

On Thu, Mar 06, 2014 at 09:38:52AM +0100, Eric F. GARIOUD wrote:
> On Tuesday 04 March 2014 19:16:35 Greg KH wrote:
> > On Mon, Mar 03, 2014 at 09:39:11AM +0100, Eric F. GARIOUD wrote:
> > > Having CONFIG_USB_ATM set and CONFIG_USB_DEBUG not set, syslogs are now
> > > overfed with kyriads of
> > >
> > > << usbatm_rx_process: xyzt callbacks suppressed >>
> > >
> > > messages per minute.
> > >
> > > - My understanding is that a commit from Greg K-H caused the trouble :
> > >
> > > quoted from https://lkml.org/lkml/2013/9/2/429 :
> > >
> > > Greg Kroah-Hartman (46):...
> > >
> > >       USB: usbatm: don't rely on CONFIG_USB_DEBUG      USB: usbatm: remove
> > >
> > > CONFIG_USB_DEBUG dependancy
> > >
> > > - My understanding is that Krzysztof Mazur reported that behaviour as part
> > > of
> > >
> > > https://lkml.org/lkml/2013/10/26/65
> > >
> > > Greg K-H is involved in the discussion and a patch seems to have been
> > > produced. I did not manage to find it commited in 3.12
> > >
> > > - Are there good reasons for this patch not to be commited ?
> >
> > What specific patch are you referring to?
>
> http://lkml.iu.edu//hypermail/linux/kernel/1310.3/01055.html
> That is : [PATCH V2] printk: pr_debug_ratelimited: check state first toreduce
> "callbacks suppressed" messages.
>
> And
>
> http://lkml.iu.edu//hypermail/linux/kernel/1310.3/01056.html
> That is : [PATCH V2] usbatm: Fix dynamic_debug / ratelimited atm_dbg
> andatm_rldbg macros
>
> > > - If not, as I understand 3.12 is longterm and maintained by Jiri Slaby,
> > > could someone forward him a request to commit Joe Perches' work ?
> >
> > Why can't you?
>
> Heu... I am nothing but an humble end-user you know. When, in some thread I
> read Greg K-H writing that he will commit the patch and AFAIK does not, then I
> presume there is something wrong in my own understanding.
> If, in addition to this, neither the reporter of the bug nor the author of the
> patch push for committing then, at the very least, I do not feel personally
> entitled to bother the longterm maintainer with this.

This is all fixed in 3.13, so don't even worry about 3.12 anymore, you
should have been upgraded already.

greg k-h