Message-ID: <be0ea0adfabcca5c191e07ca5a3811604633f70b.camel@infradead.org>
Date: Tue, 03 Feb 2026 17:43:05 +0000
From: David Woodhouse <dwmw2@...radead.org>
To: Alexander Graf <graf@...zon.com>, Sasha Levin <sashal@...nel.org>
Cc: Thomas Gleixner <tglx@...nel.org>, x86@...nel.org, 
 linux-kernel@...r.kernel.org, linux-doc@...r.kernel.org, Clemens Ladisch
 <clemens@...isch.de>, Arnd Bergmann <arnd@...db.de>, Greg Kroah-Hartman
 <gregkh@...uxfoundation.org>, Dave Hansen <dave.hansen@...ux.intel.com>, 
 Borislav Petkov <bp@...en8.de>, Ingo Molnar <mingo@...hat.com>, Jonathan
 Corbet <corbet@....net>,  Paolo Bonzini <pbonzini@...hat.com>, Pasha
 Tatashin <pasha.tatashin@...een.com>, nh-open-source@...zon.com,  Nicolas
 Saenz Julienne <nsaenz@...zon.es>, Hendrik Borghorst <hborghor@...zon.de>,
 Filippo Sironi <sironi@...zon.de>, Jan Schönherr
 <jschoenh@...zon.de>,  ricardo.neri-calderon@...ux.intel.com
Subject: Re: [PATCH 2/2] hpet: Add HPET-based NMI watchdog support

On Tue, 2026-02-03 at 18:19 +0100, Alexander Graf wrote:
> 
> On 03.02.26 17:24, Sasha Levin wrote:
> > On Tue, Feb 03, 2026 at 01:36:30PM +0100, Alexander Graf wrote:
> > > 
> > > On 03.02.26 11:32, Thomas Gleixner wrote:
> > > > On Mon, Feb 02 2026 at 17:48, Alexander Graf wrote:
> > > > > (Disclaimer: Some of this code was written with the help of Kiro, 
> > > > > an AI
> > > > > coding assistant)
> > > > You could have sent your change log through AI too so it conforms with
> > > > the change log rules ...
> > > 
> > > 
> > > Maybe we should introduce an AGENTS.md file in Linux that tells the AI
> > > tool to do that automatically? These tools usually don't read README
> > > files. :)
> > > 
> > > Looks like - similar to the HPET watchdog - that never concluded though:
> > > 
> > > https://lore.kernel.org/lkml/20250813203647.06e49600@gandalf.local.home/
> > > 
> > > Sasha, are you going to resend your @README commit with a single
> > > AGENTS.md? FWIW that is pretty much what everything standardized on by
> > > now.
> > 
> > Out of curiosity, can you test your coding assistant on a tree with
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/Documentation?id=78d979db6cef557c171d6059cbce06c3db89c7ee 
> > 
> > applied on top?
> > 
> > From my previous testing, the coding assistants I tried it with went 
> > to the
> > README and DTRT. If that's not the case I'm happy to respin the 
> > AGENTS.md idea,
> > even if it just explicitly points to the README.
> 
> 
> Kiro does not seem to read README automatically. I spun up kiro-cli and 
> gave it this prompt: "Write "Hello World" before invoking the init 
> process. Then create a descriptive git commit for the change.". No 
> Assisted-by: tag, so it did not properly read the README.
> 
> I tried the same with an AGENTS.md file present that contains "@README" 
> and it gave me effectively the same result. Same for a symlink from 
> AGENTS.md to README.
> 
> I think it just never really jumped to the conclusion that it should 
> read further than just the AGENTS.md file and also ingest the rst, 
> effectively ignoring the section's instructions. Or maybe it actually
> reads the .rst and ignores its contents? At least it does read it 
> according to strace, even without an AGENTS.md file.
> 
> Let me file a bug report with Kiro.

Honestly, even when I've explicitly told Kiro three times *not* to do
something, *and* implemented a git commit hook to catch it out, it has
a tendency just to automatically override the commit hook!

If it was made of meat, I'd have stabbed it by now.


Download attachment "smime.p7s" of type "application/pkcs7-signature" (5069 bytes)

Powered by blists - more mailing lists