On 1/16/25 7:58 AM, Bao D. Nguyen wrote:
On 1/15/2025 2:46 PM, Bart Van Assche wrote:
This patch improves performance as follows on a test setup with UFSHCI
3.0 controller:
- With the mq-deadline scheduler: 2.5x more IOPS for small writes.
- When not using an I/O scheduler compared to using mq-deadline with
zone locking: 4x more IOPS for small writes.
Wondering if the change has been tried on 4.x hosts and using different
IO schedulers? Any performance improvements?
Hi Bao,
Agreed that it would be great to have results for UFSHCI 4.0 host
controllers. Something that's unfortunate is that my UFSHCI 4.0
test setup runs kernel 6.6 and that I have not yet succeeded at
backporting the zone write plugging patches (a requirement for this
patch series) to the 6.6 kernel. A test setup with UFSHCI 4.0
controller and kernel 6.12 should become available to me around the
end of March (two months from now).
Thanks,
Bart.
