Messages in this thread | | | Date | Thu, 1 Apr 2010 00:57:06 +0530 (IST) | From | Pavan Savoy <> | Subject | Re: [PATCH] drivers:staging: sources for ST core |
| |
--- On Wed, 31/3/10, Greg KH <gregkh@suse.de> wrote:
> From: Greg KH <gregkh@suse.de> > Subject: Re: [PATCH] drivers:staging: sources for ST core > To: "Pavan Savoy" <pavan_savoy@ti.com> > Cc: "Marcel Holtmann" <marcel@holtmann.org>, alan@lxorguk.ukuu.org.uk, linux-kernel@vger.kernel.org > Date: Wednesday, 31 March, 2010, 11:49 PM > On Wed, Mar 31, 2010 at 11:32:39PM > +0530, Pavan Savoy wrote: > > --- On Wed, 31/3/10, Greg KH <gregkh@suse.de> > wrote: > > > > > From: Greg KH <gregkh@suse.de> > > > Subject: Re: [PATCH] drivers:staging: sources for > ST core > > > To: "Pavan Savoy" <pavan_savoy@ti.com> > > > Cc: "Marcel Holtmann" <marcel@holtmann.org>, > alan@lxorguk.ukuu.org.uk, > linux-kernel@vger.kernel.org > > > Date: Wednesday, 31 March, 2010, 11:00 PM > > > On Wed, Mar 31, 2010 at 04:20:22AM > > > +0530, Pavan Savoy wrote: > > > > > > > > --- On Wed, 31/3/10, Pavan Savoy <pavan_savoy@ti.com> > > > wrote: > > > > > > > > > From: Pavan Savoy <pavan_savoy@ti.com> > > > > > Subject: Re: [PATCH] drivers:staging: > sources for > > > ST core > > > > > To: gregkh@suse.de > > > > > Cc: "Marcel Holtmann" <marcel@holtmann.org>, > > > alan@lxorguk.ukuu.org.uk > > > > > Date: Wednesday, 31 March, 2010, 4:11 > AM > > > > > --- On Wed, 31/3/10, Pavan Savoy > > > > > <pavan_savoy@yahoo.co.in> > > > > > wrote: > > > > > > > > > > > From: Pavan Savoy <pavan_savoy@yahoo.co.in> > > > > > > Subject: Re: [PATCH] > drivers:staging: > > > sources for ST > > > > > core > > > > > > To: "pavan_savoy@yahoo.co.in" > > > > > <pavan_savoy@yahoo.co.in> > > > > > > Date: Wednesday, 31 March, 2010, > 4:06 AM > > > > > > > From: Greg KH [gregkh@suse.de] > > > > > > > Sent: Wednesday, March 31, > 2010 3:17 > > > AM > > > > > > > To: Savoy, Pavan > > > > > > > Cc: Alan Cox; marcel@holtmann.org; > > > > > > > linux-kernel@vger.kernel.org > > > > > > > Subject: Re: [PATCH] > drivers:staging: > > > sources for > > > > > ST > > > > > > core > > > > > > > > > > > > > > On Wed, Mar 31, 2010 at > 02:35:55AM > > > +0530, Pavan > > > > > Savoy > > > > > > > wrote: > > > > > > > > So, something like the > below is > > > ok, I have > > > > > > defined my > > > > > > > own pr_fmt, > > > > > > > > however default log > level on my > > > board is 7, > > > > > and > > > > > > hence > > > > > > > pr_info is a bit > > > > > > > > more annoying than > usual. > > > > > > > > > > > > > > No, a driver should use > dev_dbg() and > > > other > > > > > > dev_printk() > > > > > > > calls, not > > > > > > > pr_debug() or anything like > that. > > > > > > > > > > > > > > Please don't roll your own > formats, use > > > the ones > > > > > that > > > > > > are > > > > > > > well defined > > > > > > > and uniquely describe your > driver and > > > device in > > > > > a > > > > > > format > > > > > > > that the whole > > > > > > > kernel uses. > > > > > > > > > > > > > > > > > > > forgot lkml the last time.. > > > > > > > > Nope, I couldn't find any instance of struct > device at > > > all, > > > > I need that to use dev_dbg right ? - None of > the > > > tty_* > > > > structure accessible by ldiscs seems to have > a > > > reference to > > > > it. > > > > Also I happened to look at other line > discipline > > > driver, if > > > > they have a smarter way of doing this, Nope > - n_tty, > > > n_hdlc, > > > > n_slip all seem to use plain old printks. > > > >? > > > > Any clues ?? > > > > > > Sorry, you are correct, we only have a struct > kref right > > > now for tty > > > core objects, not a struct device.? So nevermind, > this > > > should be fine. > > > > Oh cool. Thanks, So that leaves me with 1 pending item > from Alan which is to tie these 3 modules (KIM/Core/LL) up > onto a TTY device specific context, and avoid all global > ptrs. > > > > So without that is it good to go in ? > > Yes, care to do that and resubmit?
Oh, But that would take some major re-structuring with the driver, because with the existing driver being a platform_device/driver structure-I can't do that.
I have made it similar to TTY<->LDISCs, LDISCs register to TTY core, here BT/FM/GPS register to ST core. TTY has a array of ldisc registered which is global, locked resource, I have st_gdata which is global and locked resource.
Each ldisc is to be in context of tty, because multiple UART exist, but this is also a platform device and exports symbols for other drivers to register in, so essentially only 1 such device exist per platform.
I agree it's unlike other device drivers in kernel, But is this unacceptable this way ? Are there any specific architectural inputs ? is this supposed to be similar to other drivers ? Please suggest...
> thanks, > > greg k-h > -- > To unsubscribe from this list: send the line "unsubscribe > linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ >
The INTERNET now has a personality. YOURS! See your Yahoo! Homepage. http://in.yahoo.com/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |