Re: [PATCHSET v3 0/5] Support for RWF_UNCACHED

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 12/11/19 6:08 PM, Linus Torvalds wrote:
> On Wed, Dec 11, 2019 at 3:41 PM Jens Axboe <axboe@xxxxxxxxx> wrote:
>>
>> Here's what I did to reproduce:
> 
> Gaah. I have a fairly fast ssd (but it's "consumer fast" - Samsung 970
> Pro - I'm assuming yours is a different class), but I encrypt my disk,
> so I only get about 15k iops and I see kcyrptd in my profiles, not
> kswapd.
> 
> I didn't even try to worry about RWF_UNCACHED or RWF_NOACCESS, since I
> wanted to see the problem. But I think that with my setup, I can't
> really see it just due to my IO being slow ;(
> 
> I do see xas_create and kswapd0, but it's 0.42%. I'm assuming it's the
> very first signs of this problem, but it's certainly not all that
> noticeable.

15K is likely too slow to really show an issue, I'm afraid. The 970
is no slouch, but your crypt setup will likely hamper it a lot. You
don't have a non-encrypted partition on it? I always set aside a
decently sized empty partition on my laptop to be able to run these
kinds of things without having to resort to a test box.

Latency is where it's down the most, you'll likely do better with
more threads to get the IOPS up.

-- 
Jens Axboe




[Index of Archives]     [Linux RAID]     [Linux SCSI]     [Linux ATA RAID]     [IDE]     [Linux Wireless]     [Linux Kernel]     [ATH6KL]     [Linux Bluetooth]     [Linux Netdev]     [Kernel Newbies]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Device Mapper]

  Powered by Linux