On Fri, 2024-08-02 at 12:49 +0200, Thomas Gleixner wrote:

On Fri, Aug 02 2024 at 09:07, David Woodhouse wrote:

quoted

On Thu, 2024-08-01 at 20:57 +0200, Thomas Gleixner wrote:

quoted

It's not counting right out of reset. But once it started counting it's
tedious to stop :)

My reading of the data sheet definitely suggests that it *shouldn't*
be.

Mode 0 says: "The output will be initially low after the mode set
operation. After the count is loaded into the selected count
register... the counter will count."

Hmm. Indeed. That does not stop the counter, but it prevents the
interrupt from firing over and over.

So fine, we can go with the patch from Li, but the changelog needs a
rewrite and the code want's a big fat comment.

Nah, it wants to be MODE, COUNT, COUNT, MODE to handle all known
implementations.

Already posted as [PATCH 2/1] (with big fat comments and a version of
your test) at

https://lore.kernel.org/kvm/3bc237678ade809cc685fedb8c1a3d435e590639.camel@infradead.org/ (local)

Although I just realised that I edited the first patch (to *remove* the
now-bogus comments about the stop sequence) before posting that one, so
they don't follow cleanly from one another; there's a trivial conflict.
I also forgot to remove the pre-1999 typedefs from the test program,
despite fixing it to use <stdint.h> like it's the 21st century :)

Top two commits of
https://git.infradead.org/users/dwmw2/linux.git/shortlog/refs/heads/clocks

I'll repost properly if you're happy with them?