On Mon, Feb 04 2019, Mark Brown wrote:
> On Mon, Feb 04, 2019 at 09:34:56AM +1100, NeilBrown wrote:
>
>> It is extremely common in the kernel for a file to start
>> // SPDX-License-Identifier.....
>
>> and to have that immediately followed by a comment lile:
>
>> /*
>> * .....
>> * ....
>
> Yes, there was a lot of automated conversion AFAICT (and a lot of
> confusion with all this stuff only being documented in random mailing
> list posts for a long time).
>
>> This patch makes this file match much of the rest of the kernel. Why
>> do you want something different?
>
> Like I said because it makes the comments look more like someone
> actually meant to add a C++ comment - it's what the rest of the
> subsystem is doing too.
Ahh.. the argument "what the rest of the subsystem is doing" makes a lot
more sense to me than "look more like someone actually mean to add a C++
comment", because I really don't understand how
// SPDX-License-Identifier.....
/*
* old comment
*/
doesn't look like it was meant to be added.
Looking around the kernel, the pattern of "// SPDX-..." followed by a
"//" happens a lot in
99 arch/csky/
122 arch/nds32/
152 sound/soc/
201 drivers/media/
273 arch/arm/
(numbers are how many instances I found), and less so in
10 drivers/soc/
11 drivers/soundwire/
12 drivers/misc/
12 drivers/pci/
13 drivers/mfd/
13 drivers/power/
14 drivers/dma/
14 drivers/input/
14 scripts/coccinelle/
16 drivers/leds/
18 drivers/spi/
25 arch/arm64/
26 drivers/regulator/
28 drivers/clk/
29 drivers/net/
36 tools/testing/
52 drivers/pinctrl/
and much less else where.
Not at all in fs/ or mm/ or net/
It would be nice to have some consistency.
I would be in favour of avoiding // as much as possible.
The "// SPDX-" lines were (presumably) added nearly automatically.
I would prefer that if it were don't manually, it would always be
/*
* SPDX- ....
* etc
*/
but I guess that horse has left the gate, though we do have that pattern
in about 129 files.
What a mess.
Any way, if you make a case based on "that is what most other files in
drivers/spi/ do", then I won't have any disagreement with that.
Thanks,
NeilBrown
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
