Re: [loop] e3f9387aea: stress-ng.loop.ops_per_sec 78.9% improvement

[Tetsuo Handa]

On 2021/12/15 22:58, kernel test robot wrote:
> 
> 
> Greeting,
> 
> FYI, we noticed a 78.9% improvement of stress-ng.loop.ops_per_sec due to commit:
> 
> 
> commit: e3f9387aea67742b9d1f4de8e5bb2fd08a8a4584 ("loop: Use pr_warn_once() for loop_control_remove() warning")
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> 

Avoiding printk() flooding will reduce latency, but I'm curious
why the testing program is trying ioctl(fd, LOOP_CTL_REMOVE, -1) ?
Since such ioctl() is pointless, measuring ops_per_sec is also pointless...