linux-kernel - Re: Dynamic debug on by default?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 14 Aug 2013 11:02:00 -0700
From:	Greg KH <gregkh@...uxfoundation.org>
To:	Sarah Sharp <sarah.a.sharp@...ux.intel.com>
Cc:	Joe Perches <joe@...ches.com>,
	Xenia Ragiadakou <burzalodowa@...il.com>,
	Jason Baron <jbaron@...mai.com>, linux-usb@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: Dynamic debug on by default?

On Wed, Aug 14, 2013 at 10:20:43AM -0700, Sarah Sharp wrote:
> On Wed, Aug 14, 2013 at 10:04:10AM -0700, Greg KH wrote:
> > On Wed, Aug 14, 2013 at 09:51:54AM -0700, Joe Perches wrote:
> > > On Wed, 2013-08-14 at 09:40 -0700, Sarah Sharp wrote:
> > > > Hi Xenia,
> > > > 
> > > > I'm a bit confused.  I thought that debugging messages would be turned
> > > > off by default for a module if CONFIG_DYNAMIC_DEBUG was turned on.  When
> > > > I tested your patch to remove the CONFIG_USB_XHCI_HCD_DEBUGGING and just
> > > > use dev_dbg, the messages from the xHCI driver appeared in dmesg by
> > > > default.
> > > > 
> > > > That generates a lot of log spew.  We can have distros add a boot
> > > > parameter option to turn off debug messages, but that boot parameter is
> > > > limited to 1023 characters.  I'm concerned that if more drivers add
> > > > dynamic debugging, the distros will eventually run out of space in the
> > > > dynamic debugging boot parameter.  I know Greg was ripping out debugging
> > > > config options in other USB drivers, so this is a bit concerning.
> > > > 
> > > > Jason, is there a way within the xHCI driver to say that dynamic
> > > > debugging should be off by default?  I've looked through the
> > > > documentation, and I can't find anything like that documented.
> > > 
> > > #undefine DEBUG
> > > 
> > > > I've attached my .config file, in case I have something misconfigured.
> > > 
> > > Because of:
> > > 
> > > drivers/usb/host/Makefile:ccflags-$(CONFIG_USB_DEBUG) := -DDEBUG
> > 
> > Yes, that's the problem, I'm trying to get rid of the CONFIG_USB_DEBUG
> > flag, and I'm almost there.  There are some host controllers that still
> > use it for some things that I have not cleaned up fully yet.
> 
> Do you think those patches will land in 3.12?  If not, I'll add an
> #undefine DEBUG to xhci.h to avoid the log spew.

Don't undefine it, if a user sets CONFIG_USB_DEBUG they obviously want
to see that spew :)

As for when it can be removed, no, it's probably not going to make 3.12,
although a lot of the work is already done and in my usb-next tree.
There are still a number of places where this is still used in the host
controller drivers that I haven't cleaned up yet:

~/linux/work/usb/drivers/usb/host $ git grep -w "ifdef DEBUG" | wc -l
37

~/linux/work/usb/drivers/usb/host $ git grep -wl "ifdef DEBUG"
ehci-fsl.c
ehci-hcd.c
ehci-q.c
ehci.h
fotg210-hcd.c
fotg210.h
fusbh200-hcd.c
fusbh200.h
imx21-hcd.c
imx21-hcd.h
ohci-dbg.c
ohci-hcd.c
ohci-spear.c
ohci.h
oxu210hp-hcd.c
uhci-debug.c
uhci-hcd.c

The hardest ones will be ohci and ehci, as there are different code
paths for when debugging is enabled (the debug stats generation and
dumping).

I'm swamped with "real work" at the moment and will not get to these
anytime soon, but if someone else wants to pick up the work and run with
it, I would love to see it done.

thanks,

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