check patch: allow C99 comments

[marc-hb]

This is a logical revert of eb45907

Based on recently merged #4941, C99 comments are now OK. I never found
any rationale or even written down coding style for excluding them in
the first place.

Signed-off-by: Marc Herbert marc.herbert@intel.com

[paulstelian97]

I don't see how that PR hints at C99 comments being OK. They're only used for SPDX.

I do agree that avoiding C99 comments isn't exactly the best thing, especially given that we don't have compilers that can't handle them I think. Holding off on reviewing for now (no reason to reject, just not sure if this is fully to be accepted)

[marc-hb]

I don't see how that PR hints at C99 comments being OK. They're only used for SPDX.

https://github.com/thesofproject/sof/pull/4941/files uses // for some entire, 7 lines-long file headers like src/math/base2log.c and a couple others. SPDX is just one line.

In other files, C99 comments are not used at all, not even for SPDX.

Anyway, 4941 is not that important. The question this PR is asking is: does everyone agree to allow C99 comments? I suspect they've already been used for ages.

[marc-hb]

https://github.com/thesofproject/sof/pull/4941/files uses // for some entire, 7 lines-long file headers like src/math/base2log.c and a couple others. SPDX is just one line.

I think I got it: some people seem to assume SPDX magically includes and interprets the free form copyright notice following the SPDX-License-Identifier line. It does not, it's not magical. It takes a bit more effort to make copyright information known to SPDX, example from the https://spdx.github.io/spdx-spec/file-tags/ Annex (Informative)

SPDX-FileCopyrightText: 2019 Jane Doe <jane@example.com>
SPDX-FileCopyrightText: Copyright 2008-2010 John Smith
SPDX-FileCopyrightText: Copyright Example Company
SPDX-FileCopyrightText: Copyright contributors to the Foo project.

Off-topic sorry.

[lgirdwood]

@marc-hb have you told this to skip CI ? cant see any skip request in the PR, but the CI has stalled ?

[marc-hb]

I didn't ask to SKIP CI and I don't know why it stalled this time. It stalled after running checkpatch and this is a pure checkpatch configuration change which... does not even apply to sof-ci/jenkins as it always accepted C99 comments, see example in #4941.

sof-ci/jenkins has been running fine in other PRs so I didn't care.

[marc-hb]

SOFCI TEST

EDIT: many unavailable devices in https://sof-ci.01.org/sofpr/PR5046/build11320/devicetest/ but everything else is green.

[gkbldcig]

Can one of the admins verify this patch?

[marc-hb]

So no objection from anyone about C99 // comments in any SOF code review anymore? Speak now or forever hold your peace...

PS: while very similar, the Linux kernel has its own codestyle and policies not defined by SOF.

[marc-hb]

just not sure if this is fully to be accepted

I thought it was, but apparently it's still not: #5110 (comment)

Need to practice my tea leaves reading?

[marc-hb]

For even more fun: SPDX requires different comment style for .h files versus .c files, see 82220c8

[dbaluta]

@marc-hb everyone knows that, the real question here is why 😨

[marc-hb]

Apparently because .h files are often enough included in linnker scripts or assembly. Please help review this minor SPDX-README.md clarification:

• [SKIP CI] .github/SPDX-README.md: clarify .c/.h difference and applicability #6973