Re: [MAINTAINERS SUMMIT] Other LLM-related topics - tags, newcomers, etc

On Sun, 19 Jul 2026 11:29:24 +0200,
Mauro Carvalho Chehab wrote:
> 
> On Sat, 18 Jul 2026 11:26:27 +0200
> Takashi Iwai <[email protected]> wrote:
> 
> > On Fri, 17 Jul 2026 02:58:12 +0200,
> > Mauro Carvalho Chehab wrote:
> > >   
> > > > So it's one say to say, we should figure out how to try to run Sashiko
> > > > on a local LLM, using open-weight models.  But it's going to be a lot
> > > > easier to propose such a thing than to actually do it.  
> > > 
> > > The main point is: do we really need 671B parameters? Those models
> > > speak a lot of different languages, have medical databases, and a lot
> > > of other random knowledge that are useless for kernel development.
> > > 
> > > I've been playing for a while with qwen 3.6 with 24KB context size,
> > > 36B parameters (3B activated), 4bits kv quantization and it does produce
> > > some decent results. The main limitation is the context size: it is
> > > probably not big enough to test big files (*)
> > > 
> > > (*) my GPU has 16GB and it is not dedicated to LLM - still, it does
> > >     present results on a reasonable time (a couple of minutes) and
> > >     with decent precision.  
> > 
> > I've been testing the kernel commit reviews with 16GB VRAM GPU, too.
> > It's an agent program based on Chris Mason's review prompts, and
> > targeted mainly for the verifications of our backport patches.  The
> > main models I tried were Qwen-3.6 35B, Gemma-4 26B and GPT-OSS 20B
> > (all 4-bit quantized).  And, my conclusion was that even such small
> > models can catch real bugs.
> > 
> > Yes, there are definitely many false-positives, and they don't always
> > follow the recent changes.  Also they cover much less changes than
> > Sashiko.  So, the results must be read with lots of grains of salt :)
> > 
> > The code review is different from the code generation, and the agent
> > divides the tasks so that the model can concentrate on each small
> > single task.  With that, small models can achieve in some level, too.
> > 
> > 
> > FWIW, the agent code is found at
> >   https://github.com/tiwai/kernel-review-agent.git
> > 
> > with HTML renderer and console viewer programs
> >   https://github.com/tiwai/review-table-gen.git
> >   https://github.com/tiwai/kreview-ui-rs.git
> 
> I just did a Sashiko review test using Ollama locally after applying 
> this PR:
> 
> 	https://github.com/sashiko-dev/sashiko/pull/338
> 
> The code was AI generated with qwen3.6 with Claude Opus reasoning
> (hf.co/rico03/Qwen3.6-27B-Claude-Opus-Reasoning-Distilled-GGUF:Q4_K_M).
> 
> I'll try to play with it a little bit to see how it works with
> real patch reviews.

Thanks!  Let me know if you can compare the review results from
different models qualitatively.  It'd be interesting.


Takashi