Re: [PATCH v2 01/15] sched: Add task enqueue/dequeue trace points

[Peter Zijlstra]

On Fri, Oct 02, 2026 at 09:09:01AM +0200, Gabriele Monaco wrote:
> On Thu, 2026-10-01 at 17:49 +0200, Peter Zijlstra wrote:
> > On Thu, Oct 01, 2026 at 05:20:28PM +0200, Gabriele Monaco wrote:
> > > From: Nam Cao <namcao@xxxxxxxxxxxxx>
> > > 
> > > Add trace points into enqueue_task() and dequeue_task().
> > > 
> > > Signed-off-by: Nam Cao <namcao@xxxxxxxxxxxxx>
> > > Suggested-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > 
> > Obviously I don't remember anything at all. But also, you 'forgot' to
> > send me the rest of the patches which might hold a clue. As is I'm
> > clueless as to why we want this, since the Changelog also offers none.
> 
> Right..
> 
> This commit is part of [1], but borrowed from [2], which is actually more
> interesting to understand why I need those tracepoints.
> 
> [2] contains 3 other RV deadline monitors, I should have at least CC'd you in
> [3] and [4].. Essentially the enqueue and dequeue tracepoints are used to:
> 
>  * model the event from dl_server idle to ready (a new fair task is available)
>  * understand when a task is moved to and from another scheduler runqueue
> 
> Thanks,
> Gabriele
> 
> [1] - https://lore.kernel.org/lkml/20261001152042.124445-1-gmonaco@xxxxxxxxxx
> [2] - https://lore.kernel.org/lkml/20260929124908.177676-1-gmonaco@xxxxxxxxxx
> [3] - https://lore.kernel.org/lkml/20260929124908.177676-8-gmonaco@xxxxxxxxxx
> [4] - https://lore.kernel.org/lkml/20260929124908.177676-9-gmonaco@xxxxxxxxxx

So mostly I flat out out ignore partial series. It means I have to go
dig around to figure out the whole picture, and anything that requires
extra effort goes on the 'later' pile, and you know where that lives.

Now, as to 3,4; I'm confused. If I search 'enqueue' I see there is a
tracepoint attach thingy in the .c file, but I don't see the
corresponding bits in the .dot file. And per my understanding the .c
thing is an artifact of the .dot file.

So what gives?

Also, IMO it would be *much* preferred if the .rst file would not exist.
Just have the graph-easy and the description as a comment in the .dot
file. Then you have the whole thing in one *readable* place.

...insert rant on what a piece of shit rst is here...