linux-kernel - Re: Commit Links [was: Linux 6.17-rc5]

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]

Message-ID: <81e33495-ad29-49b1-bb45-ee44f9f2a5ce@kernel.org>
Date: Mon, 13 Oct 2025 10:53:36 +0200
From: Jiri Slaby <jirislaby@...nel.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: Commit Links [was: Linux 6.17-rc5]

On 13. 09. 25, 0:22, Linus Torvalds wrote:
> On Thu, 11 Sept 2025 at 23:24, Jiri Slaby <jirislaby@...nel.org> wrote:
>> OK, can we have Submitted-at: or something for the above which you could
>> completely ignore?
> 
> No.
> 
> We're not adding garbage that I'm "supposed to ignore".
> 
> Stop this idiotic thread. Stop adding crap automatically.

Pardon me? I understand you are tired of this (so am I), but can we stay 
constructive instead of constantly resistive? This is not helping.
> I have made it very clear that you can add "Link" to your submissions
> as long as IT IS NOT SOME AUTOMATIC MEANINGLESS GARBAGE.
Yes, that's why I proposed something different than Link:.

> I have also pointed out that you can find the original submission in
> various much better ways than the meaningless link think you argue
> for.

Obvious question: how? Do you have some power tool which we all are 
missing?

One example for all: for 23743ba64709, I want to get the applied v2 
20250825125607.2478-3-calixte.pernot@...noble-inp.org and not mistakenly 
resent v3 20250909202629.9386-2-calixte.pernot@...noble-inp.org. How do 
I do that (if there was not a Link)? I can look up for tons of these 
false pos/negs.

Seeming solutions: people are adding git-notes for some trees (tip, 
Vasant's tree). People are implementing b4 dig. People are using lore 
search. All those approaches are hopeless as they simply do NOT work as 
Link does. [Some maintainers I talked to inside SUSE are going to keep 
_this kind_ of Link (but I guess you will start refusing their PRs 
eventually).]

My stance: if I bisect to a commit which does not yield me the right 
message-id in a second, I won't report it and stop right there. I am not 
going to burn my time by dubious searching here and there. The same as 
you don't want to burn your time clicking only-for-you-useless Links -- 
only viewed from the opposite side. And well, provided I alone upgrade 
the Tumbleweed kernel when a new release comes, I have to find bugs in 
final releases quite often.

thanks,
-- 
js
suse labs