linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Randy Dunlap <rdunlap@infradead.org>
To: Joe Perches <joe@perches.com>, Mihai Moldovan <ionic@ionic.de>,
	Masahiro Yamada <masahiroy@kernel.org>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] kconfig: nconf: stop endless search-up loops
Date: Sun, 28 Mar 2021 09:16:27 -0700	[thread overview]
Message-ID: <027ce5a1-cc48-3f79-fc05-bb587a9e86b1@infradead.org> (raw)
In-Reply-To: <6caae542c6f295a47b1012738b3b35055cf17cc4.camel@perches.com>

On 3/28/21 3:32 AM, Joe Perches wrote:
> On Sat, 2021-03-27 at 15:26 -0700, Randy Dunlap wrote:
>> On 3/27/21 3:12 PM, Mihai Moldovan wrote:
>>> * On 3/27/21 4:58 PM, Randy Dunlap wrote:
>>>> On 3/27/21 5:01 AM, Mihai Moldovan wrote:
>>>>> +		if ((-1 == index) && (index == match_start))
>>>>
>>>> checkpatch doesn't complain about this (and I wonder how it's missed), but
>>>> kernel style is (mostly) "constant goes on right hand side of comparison",
>>>> so
>>>> 		if ((index == -1) &&
>>>
>>> I can naturally send a V2 with that swapped.
>>>
>>> To my rationale: I made sure to use checkpatch, saw that it was accepted and
>>> even went for a quick git grep -- '-1 ==', which likewise returned enough
>>> results for me to call this consistent with the current code style.
>>>
>>> Maybe those matches were just frowned-upon, but forgotten-to-be-critized
>>> examples of this pattern being used.
>>
>> There is a test for it in checkpatch.pl but I also used checkpatch.pl
>> without it complaining, so I don't know what it takes to make the script
>> complain.
>>
>> 			if ($lead !~ /(?:$Operators|\.)\s*$/ &&
>> 			    $to !~ /^(?:Constant|[A-Z_][A-Z0-9_]*)$/ &&
>> 			    WARN("CONSTANT_COMPARISON",
>> 				 "Comparisons should place the constant on the right side of the test\n" . $herecurr) &&
> 
> Negative values aren't parsed well by the silly script as checkpatch
> isn't a real parser.

Yes, I get that.

> Basically, checkpatch only recognizes positive ints as constants.

OK, good to know.

> So it doesn't recognize uses like:
> 
> 	a * -5 > b
> 
> It doesn't parse -5 as a negative constant.
> 
> Though here it does seem the line with
>  			    $to !~ /^(?:Constant|[A-Z_][A-Z0-9_]*)$/ &&
> should be
>  			    $to !~ /^(?:$Constant|[A-Z_][A-Z0-9_]*)$/ &&
> 
> You are welcome to try to improve checkpatch, but it seems non-trivial
> and a relatively uncommon use in the kernel, so I won't.

I get that too.

> Most all of the existing uses seem to be in drivers/scsi/pm8001/pm8001_hwi.c

[snip]


thanks.
-- 
~Randy


  reply	other threads:[~2021-03-28 16:17 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-27 12:01 [PATCH] kconfig: nconf: stop endless search-up loops Mihai Moldovan
2021-03-27 15:58 ` Randy Dunlap
2021-03-27 22:12   ` Mihai Moldovan
2021-03-27 22:26     ` Randy Dunlap
2021-03-28  9:27       ` Mihai Moldovan
2021-03-28 10:37         ` Joe Perches
2021-03-28 10:32       ` Joe Perches
2021-03-28 16:16         ` Randy Dunlap [this message]
2021-03-28  9:52 ` [PATCH v2] " Mihai Moldovan
2021-04-10  5:47   ` Masahiro Yamada
2021-04-10  7:00     ` Mihai Moldovan
2021-04-10  9:12       ` Masahiro Yamada
2021-04-15  7:28 ` [PATCH v3] kconfig: nconf: stop endless search loops Mihai Moldovan
2021-04-16  5:40   ` Masahiro Yamada
2021-04-16 10:39     ` Mihai Moldovan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=027ce5a1-cc48-3f79-fc05-bb587a9e86b1@infradead.org \
    --to=rdunlap@infradead.org \
    --cc=ionic@ionic.de \
    --cc=joe@perches.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=masahiroy@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).