netdev - Re: [patch net-next 01/10] genetlink: don't merge dumpit split op for different cmds into single iter

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

Message-ID: <ZSaGiSKL5/ocFYOE@nanopsycho>
Date: Wed, 11 Oct 2023 13:27:05 +0200
From: Jiri Pirko <jiri@...nulli.us>
To: Jakub Kicinski <kuba@...nel.org>
Cc: netdev@...r.kernel.org, pabeni@...hat.com, davem@...emloft.net,
	edumazet@...gle.com, jacob.e.keller@...el.com,
	johannes@...solutions.net
Subject: Re: [patch net-next 01/10] genetlink: don't merge dumpit split op
 for different cmds into single iter

Wed, Oct 11, 2023 at 08:08:57AM CEST, jiri@...nulli.us wrote:
>Tue, Oct 10, 2023 at 08:48:45PM CEST, kuba@...nel.org wrote:
>>On Tue, 10 Oct 2023 13:08:20 +0200 Jiri Pirko wrote:
>>> Fixes: b8fd60c36a44 ("genetlink: allow families to use split ops directly")
>>
>>Drop Fixes, add "currently no family declares ops which could trigger
>>this issue".
>
>Yeah, we need fixes semantics written down somewhere.
>I can do it, sure.

I found 2 mentions that relate to netdev regarging Fixes:

Quoting Documentation/process/submitting-patches.rst:
If your patch fixes a bug in a specific commit, e.g. you found an issue using
``git bisect``, please use the 'Fixes:' tag with the first 12 characters of
the SHA-1 ID, and the one line summary. 

Quoting Documentation/process/maintainer-netdev.rst:
 - for fixes the ``Fixes:`` tag is required, regardless of the tree

This patch fixes a bug, sure, bug is not hit by existing code, but still
it is present.

Why it is wrong to put "Fixes" in this case?
Could you please document this?

>
>
>>
>>>  	if (i + cnt < family->n_split_ops &&
>>> -	    family->split_ops[i + cnt].flags & GENL_CMD_CAP_DUMP) {
>>> +	    family->split_ops[i + cnt].flags & GENL_CMD_CAP_DUMP &&
>>> +	    (!cnt ||
>>> +	     (cnt && family->split_ops[i + cnt].cmd == iter->doit.cmd))) {
>>
>>Why are you checking cnt? if do was not found cmd will be 0, which
>>cannot mis-match.
>
>Correct. Will remove cnt check.