[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1988292.s3joEKjhKX@localhost.localdomain>
Date: Sat, 18 Sep 2021 13:13:48 +0200
From: "Fabio M. De Francesco" <fmdefrancesco@...il.com>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: Larry Finger <Larry.Finger@...inger.net>,
Phillip Potter <phil@...lpotter.co.uk>,
Pavel Skripkin <paskripkin@...il.com>,
linux-staging@...ts.linux.dev, linux-kernel@...r.kernel.org,
David Laight <david.Laight@...lab.com>,
Dan Carpenter <dan.carpenter@...cle.com>
Subject: Re: [PATCH v7 15/19] staging: r8188eu: clean up usbctrl_vendorreq()
On Friday, September 17, 2021 4:44:17 PM CEST Greg Kroah-Hartman wrote:
> On Fri, Sep 17, 2021 at 09:18:33AM +0200, Fabio M. De Francesco wrote:
> > Clean up usbctrl_vendoreq() in usb_ops_linux.c.
> >
> > List of changes:
> >
> > 1) Rename variables:
> > pdata => data
> > pio_priv => io_priv
> > pintfhdl => intfhdl
> > wvalue => address.
> > 2) Reorder variables declarations according to the "Reverse Xmas Tree"
> > style.
> > 3) Remove unnecessary test for "!pIo_buf".
> > 4) Move comments one line below code.
> > 5) Remove unnecessary excess parentheses.
> > 6) Remove unnecessary extra spaces.
> > 7) Remove unnecessary comments.
> > 8) Fix grammar errors (checksumed => checksummed).
>
> When you find yourself listing all of the different things you have done
> in a single commit, that is a HUGE hint that you need to break this up
> into smaller pieces.
>
> Please do so here, this should not be just one change, as it's almost
> impossible to look at this and "know" it's all still the same logic
> happening here. But if you had broken this down into 8 different
> changes, then it would have been obvious and I could easily have applied
> the changes.
Dear Greg,
My first thought when I read you message was to simply delete this patch
because usbctrl_vendorreq() is going to be deleted in 18/19. But then I
rethought of the original purpose behind this patch and (after talking with
Pavel) we decided to do the task you asked and split this patch into 8
smaller ones. The only reason is because, as you noticed, we "[]are moving
code around", so, although I'm not required to clean up code in
usbctrl_vendorreq(), I'm required to make the new usb_read() and usb_write()
the cleaner the possible. This preventive clean up helps me a lot.
Obviously I guess that I'm required to split also the next patch of this
series in 3 because there are also there 3 different kind of clean-ups.
So, we'll have a total of 11 clean-ups.
>
> I've taken the first 14 patches in this series, it's great work, thank
> you all for doing this.
Thanks to you for the "great work". We appreciated it.
> But this, and the remaining patches in here
> need to be split up more to make it obvious that the changes are correct
> and should be accepted. Please feel free to start the numbering of the
> patch series over now, given that the first 14 are now merged into my
> tree.
We're working on this.
Thanks,
Fabio
>
> thanks,
>
> greg k-h
>
Powered by blists - more mailing lists