Re: [PATCH] lib: test module for find_*_bit() functions

From: Geert Uytterhoeven
Date: Fri Nov 24 2017 - 10:46:14 EST


Hi Yury,

On Fri, Nov 24, 2017 at 3:30 PM, Yury Norov <ynorov@xxxxxxxxxxxxxxxxxx> wrote:
> Below the updates proposed in this thread.

Thank you!

> From 959700bd7e7f586171c15a4130a9888acac02daf Mon Sep 17 00:00:00 2001
> From: Yury Norov <ynorov@xxxxxxxxxxxxxxxxxx>
> Date: Wed, 22 Nov 2017 17:21:40 +0300
> Subject: [PATCH] improve lib/test_find_bit
>
> As suggested in review comments:
> * printk: align numbers using whitespaces instead of tabs;
> * return error value from init() to avoid calling rmmod if testing again;
> * use ktime_get instead of get_cycles as some arches don't support it;
> * rename test_find_bit.c to find_bit_benchmark.c.
>
> The output in dmesg (on QEMU arm64):
> [ 38.823430] Start testing find_bit() with random-filled bitmap
> [ 38.845358] find_next_bit: 20138448 ns, 163968 iterations
> [ 38.856217] find_next_zero_bit: 10615328 ns, 163713 iterations
> [ 38.863564] find_last_bit: 7111888 ns, 163967 iterations
> [ 40.944796] find_first_bit: 2081007216 ns, 163968 iterations
> [ 40.944975]
> [ 40.944975] Start testing find_bit() with sparse bitmap
> [ 40.945268] find_next_bit: 73216 ns, 656 iterations
> [ 40.967858] find_next_zero_bit: 22461008 ns, 327025 iterations
> [ 40.968047] find_last_bit: 62320 ns, 656 iterations
> [ 40.978060] find_first_bit: 9889360 ns, 656 iterations
>
> Signed-off-by: Yury Norov <ynorov@xxxxxxxxxxxxxxxxxx>

Tested-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds