linux-kernel - Re: [PATCH] clang-format: Update .clang-format with the latest for

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Mon, 21 Jan 2019 01:59:03 +0000
From:   Jason Gunthorpe <jgg@...lanox.com>
To:     Miguel Ojeda <miguel.ojeda.sandonis@...il.com>
CC:     Joe Perches <joe@...ches.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] clang-format: Update .clang-format with the latest
 for_each macro list

On Sat, Jan 19, 2019 at 10:29:50PM +0100, Miguel Ojeda wrote:
> Hi Joe,
> 
> On Sat, Jan 19, 2019 at 8:45 PM Joe Perches <joe@...ches.com> wrote:
> >
> > I suggest clang-format's ForEachMacros statement vector
> > be expanded to support regexes ala
> >
> > ForEachMacros:
> >   - Regex:           '\b(?:(?:[a-z0-9_]+_)?for_each(?:_[a-z_]+)|[a-z0-9_]+_for_each)\b'
> 
> Ah, yeah, of course. And it should be easy, because there is already
> support for regexes in the IncludeCategories option. However, the
> configuration file was intended for clang-format >= 4 (when I added
> the file, the intention was to make it usable as easily as possible,
> i.e. taking into account the availability of clang-format in distros).
> 
> Actually, if we were to change clang-format and then require in the
> kernel the latest release, we could go a bit further and add the
> missing features needed to match the kernel style almost completely. I
> have been trying to allocate some time to start that project in my
> backlog, but... :-)

It is unfortunate that clang format does not have a way for old
versions to parse new .clang-format files.

Perhaps the best course is to get that ability and then go ahead and
bump the min required version to there.. Otherwise this is a never
ending upgrade treadmill.

ie if clang-format would look for a .clang-format.VERSION first and
then try for the generic we could at least have a (somewhat painful)
path.. Or if it would run it through cpp first we could use
#ifdef/etc.

Jason