Re: [PATCH v1 1/1] lib/crypto: tests: Add KUnit tests for AES

[David Laight]

On Fri, 16 Jan 2026 21:55:04 +0100
Holger Dengler <dengler@xxxxxxxxxxxxx> wrote:

> On 16/01/2026 20:44, Eric Biggers wrote:
...
> > The warm-up loops in the existing benchmarks are both for cache warming
> > and to get the CPU frequency fast and fixed.  It's not anything
> > sophisticated, but rather just something that's simple and seems to
> > works well enough across CPUs without depending on any special APIs.  If
> > your CPU doesn't do much frequency scaling, you may not notice a
> > difference, but other CPUs may need it.  
> 
> Do you have a gut feeling how many iterations it takes to get the CPU speed
> up? If it takes less than 50 iterations, it would be sufficient with the new
> method.

It may not matter what you do to get the cpu speed fixed.
Looping calling ktime_get_ns() for 'long enough' should do it.
That would be test independent but the 'long enough' very
cpu dependent.
The benchmarks probably ought to have some common API - even if it
just in the kunit code.

The advantage of counting cpu clocks is the frequency then doesn't
matter as much - L1 cache miss timings might change.

The difficulty is finding a cpu clock counter. Architecture dependent
and may not exist (you don't want the fixed frequency 'sanitised' TSC).

	David