Re: [PATCH 431/606] mfd: khadas-mcu: Convert to i2c's .probe_new()

[Lee Jones]

On Wed, 07 Dec 2022, Lee Jones wrote:

> On Wed, 07 Dec 2022, Uwe Kleine-König wrote:
> 
> > Hello Lee,
> > 
> > On Tue, Dec 06, 2022 at 04:56:59PM +0000, Lee Jones wrote:
> > > > On Tue, Dec 06, 2022 at 04:10:51PM +0000, Lee Jones wrote:
> > > > > On Tue, 06 Dec 2022, Uwe Kleine-König wrote:
> > > > > 
> > > > > > Hey Lee,
> > > > > > 
> > > > > > On Mon, Nov 21, 2022 at 04:51:21PM +0000, Lee Jones wrote:
> > > > > > > On Mon, 21 Nov 2022, Uwe Kleine-König wrote:
> > > > > > > 
> > > > > > > > Hello Lee,
> > > > > > > > 
> > > > > > > > On Mon, Nov 21, 2022 at 12:32:10PM +0000, Lee Jones wrote:
> > > > > > > > > On Fri, 18 Nov 2022, Uwe Kleine-König wrote:
> > > > > > > > > 
> > > > > > > > > > From: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> > > > > > > > > > 
> > > > > > > > > > The probe function doesn't make use of the i2c_device_id * parameter so it
> > > > > > > > > > can be trivially converted.
> > > > > > > > > > 
> > > > > > > > > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
> > > > > > > > > > ---
> > > > > > > > > >  drivers/mfd/khadas-mcu.c | 5 ++---
> > > > > > > > > >  1 file changed, 2 insertions(+), 3 deletions(-)
> > > > > > > > > 
> > > > > > > > > After a week or so, please collect-up all the tags you have received
> > > > > > > > > and submit a per-subsystem set for me to hoover up, thanks.
> > > > > > > > 
> > > > > > > > For mfd I'd do:
> > > > > > > > 
> > > > > > > > 	git checkout mfd/for-next
> > > > > > > > 	b4 am -P 413-481 20221118224540.619276-1-uwe@xxxxxxxxxxxxxxxxx
> > > > > > > > 	git am ./20221118_uwe_i2c_complete_conversion_to_i2c_probe_new.mbx
> > > > > > > > 	git send-email --to .... --cc .... mfd/for-next
> > > > > > > 
> > > > > > > That's just crazy enough to work.
> > > > > > > 
> > > > > > > Thanks for the tip.
> > > > > > 
> > > > > > On irc you said you'd care for application of these patches ("I plan to
> > > > > > attempt the b4 solution"), they didn't land in next yet. Do you need a
> > > > > > reminder? Something else?
> > > > > 
> > > > > I applied them, but they fail to build and I haven't had time to
> > > > > investigate.  I guess they depend on some patches that have been
> > > > > accepted into another (input?) and are now in -next.  Any idea if they
> > > > > are available on some immutable branch that I can pull from?
> > > > 
> > > > If in your tree i2c_client_get_device_id() is missing, you want to pull
> > > > 
> > > > 	https://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git i2c/client_device_id_helper-immutable
> > > 
> > > Ideal, thanks.
> > 
> > I see you added my commits to your tree now, however there is a problem:
> > 
> >   $ git log --oneline --graph --boundary linus/master..FETCH_HEAD
> >   ...
> >   *   7281458f4396 Merge branches 'ib-i2c-mfd-client_dev_id_helper-6.2' and 'ib-mfd-uwes-i2c-probe_new-6.2' into ibs-for-mfd-merged
> >   |\  
> >   | * c066a1632bc3 mfd: wm8994-core: Convert to i2c's .probe_new()
> >   | * 0f22cf00762d mfd: wm8400-core: Convert to i2c's .probe_new()
> >   | * 090f49b250ee mfd: wm8350-i2c: Convert to i2c's .probe_new()
> >   | * 7174af1be41d mfd: wm831x-i2c: Convert to i2c's .probe_new()
> >   | ...
> >   | * 63909fec136e mfd: adp5520: Convert to i2c's .probe_new()
> >   | * 549d4f3207f8 mfd: act8945a: Convert to i2c's .probe_new()
> >   | * 623f79babaf1 mfd: aat2870-core: Convert to i2c's .probe_new()
> >   | * ce41e4ae7cac mfd: 88pm805: Convert to i2c's .probe_new()
> >   | * 02aa483f1f98 mfd: 88pm800: Convert to i2c's .probe_new()
> >   * | 662233731d66 i2c: core: Introduce i2c_client_get_device_id helper function
> >   |/
> >   o 9abf2313adc1 (tag: v6.1-rc1) Linux 6.1-rc1
> > 
> > That means that the commit introducing i2c_client_get_device_id() isn't
> > an ancestor of the commits that make use of the new function. So
> > 63909fec136e (which is the first commit making use of the new function)
> > likely won't compile:
> > 
> > 	$ git grep i2c_client_get_device_id 63909fec136e
> > 	63909fec136e:drivers/mfd/adp5520.c:     const struct i2c_device_id *id = i2c_client_get_device_id(client);
> > 
> > Starting with 7281458f4396 everything is fine again, but still this
> > hurts a bisection.
> 
> Fair point - I'll rebase the my topic branch on top of Wolfram's.

Okay, try now.

-- 
Lee Jones [李琼斯]