Re: [GIT PULL FOR v3.16] davinci updates

[Hans Verkuil <hverkuil@xs4all.nl>]

On 05/24/2014 12:45 AM, Mauro Carvalho Chehab wrote:
> Em Fri, 23 May 2014 11:07:25 +0200
> Hans Verkuil <hverkuil@xs4all.nl> escreveu:
> 
>> Hi Mauro,
>>
>> These are cleanup patches for the davinci drivers. A total of about 1200 lines
>> of code are removed. Not bad!
>>
>> Regards,
>>
>> 	Hans
>>
>>
>> The following changes since commit e899966f626f1f657a4a7bac736c0b9ae5a243ea:
>>
>>   Merge tag 'v3.15-rc6' into patchwork (2014-05-21 23:03:15 -0300)
>>
>> are available in the git repository at:
>>
>>
>>   git://linuxtv.org/hverkuil/media_tree.git davinci
>>
>> for you to fetch changes up to c1022cd59bb34dbb435cda9a2fc98bb6fb931f61:
>>
>>   media: davinci: vpif: add Copyright message (2014-05-23 10:12:34 +0200)
>>
>> ----------------------------------------------------------------
>> Lad, Prabhakar (49):
>>       media: davinci: vpif_display: initialize vb2 queue and DMA context during probe
>>       media: davinci: vpif_display: drop buf_init() callback
>>       media: davinci: vpif_display: use vb2_ops_wait_prepare/finish helper functions
>>       media: davinci: vpif_display: release buffers in case start_streaming() call back fails
>>       media: davinci: vpif_display: drop buf_cleanup() callback
>>       media: davinci: vpif_display: improve vpif_buffer_prepare() callback
>>       media: davinci: vpif_display: improve vpif_buffer_queue_setup() function
>>       media: davinci: vpif_display: improve start/stop_streaming callbacks
>>       media: davinci: vpif_display: use vb2_fop_mmap/poll
>>       media: davinci: vpif_display: use v4l2_fh_open and vb2_fop_release
>>       media: davinci: vpif_display: use vb2_ioctl_* helpers
>>       media: davinci: vpif_display: drop unused member fbuffers
>>       media: davinci: vpif_display: drop reserving memory for device
>>       media: davinci: vpif_display: drop unnecessary field memory
>>       media: davinci: vpif_display: drop numbuffers field from common_obj
>>       media: davinic: vpif_display: drop started member from struct common_obj
>>       media: davinci: vpif_display: initialize the video device in single place
>>       media: davinci: vpif_display: drop unneeded module params
>>       media: davinci: vpif_display: drop cropcap
>>       media: davinci: vpif_display: group v4l2_ioctl_ops
>>       media: davinci: vpif_display: use SIMPLE_DEV_PM_OPS
>>       media: davinci: vpif_display: return -ENODATA for *dv_timings calls
>>       media: davinci: vpif_display: return -ENODATA for *std calls
>>       media: davinci; vpif_display: fix checkpatch error
>>       media: davinci: vpif_display: fix v4l-complinace issues
>>       media: davinci: vpif_capture: initalize vb2 queue and DMA context during probe
>>       media: davinci: vpif_capture: drop buf_init() callback
>>       media: davinci: vpif_capture: use vb2_ops_wait_prepare/finish helper functions
>>       media: davinci: vpif_capture: release buffers in case start_streaming() call back fails
>>       media: davinci: vpif_capture: drop buf_cleanup() callback
>>       media: davinci: vpif_capture: improve vpif_buffer_prepare() callback
>>       media: davinci: vpif_capture: improve vpif_buffer_queue_setup() function
>>       media: davinci: vpif_capture: improve start/stop_streaming callbacks
>>       media: davinci: vpif_capture: use vb2_fop_mmap/poll
>>       media: davinci: vpif_capture: use v4l2_fh_open and vb2_fop_release
>>       media: davinci: vpif_capture: use vb2_ioctl_* helpers
>>       media: davinci: vpif_capture: drop reserving memory for device
>>       media: davinci: vpif_capture: drop unnecessary field memory
>>       media: davinic: vpif_capture: drop started member from struct common_obj
>>       media: davinci: vpif_capture: initialize the video device in single place
> 
>>       media: davinci: vpif_capture: drop unneeded module params
> 
> Enough!
> 
> I'm tired of guessing why those bad commented are needed and what them are
> actually doing.
> 
> In this particular case:
> 
> Why those module parameters were needed before, but aren't needed anymore?
> What changed? The removal of module parameters is a sort of API change.
> 
> So, I _DO_ expect them to be very well justified.
> 
> Please, properly describe _ALL_ patches, or I'll NACK the pull requests.
> 
> This time, I applied everything up to the patch before this one. On a next
> pull request without proper descriptions, I'll likely just stop on the first
> patch missing description (or with a crappy one).

Next time you see patches with insufficient commit log text just send them back
with 'Changes Requested'. I don't mind since I have a bit of a blind spot for
that myself. It's good training for me.

But in this case you accepted the patch ("drop unneeded module params") which
really needed a better description (again, blind spot on my side, I should
have caught that), and then stopped merging the remaining patches. But those
remaining patches all have proper commit logs (at least in my view), so I am
requesting that you pull in the remaining patches. 

If you think that the commit logs for the remaining patches isn't good enough,
just let me know and I will improve them.

Regards,

	Hans