Re: [PATCH] hw/nvme: Add iothread support

[Jinhao Fan]

Sure. I’ve already reworked this iothread patch upon the new irqfd patch. I 
think I can post a v2 patch today. Do you mean I include irqfd v3 in the new 
iothread patch series?

发自我的iPhone

> 在 2022年8月26日，15:12，Klaus Jensen <its@irrelevant.dk> 写道：
> 
> On Jul 20 17:00, Jinhao Fan wrote:
>> Add an option "iothread=x" to do emulation in a seperate iothread.
>> This improves the performance because QEMU's main loop is responsible
>> for a lot of other work while iothread is dedicated to NVMe emulation.
>> Moreover, emulating in iothread brings the potential of polling on
>> SQ/CQ doorbells, which I will bring up in a following patch.
>> 
>> Iothread can be enabled by:
>> -object iothread,id=nvme0 \
>> -device nvme,iothread=nvme0 \
>> 
>> Performance comparisons (KIOPS):
>> 
>> QD         1   4  16  64
>> QEMU      41 136 242 338
>> iothread  53 155 245 309
>> 
>> Signed-off-by: Jinhao Fan <fanjinhao21s@ict.ac.cn>
>> ---
>> hw/nvme/ctrl.c | 80 ++++++++++++++++++++++++++++++++++++++++++++++----
>> hw/nvme/ns.c   | 19 +++++++++---
>> hw/nvme/nvme.h |  6 +++-
>> 3 files changed, 95 insertions(+), 10 deletions(-)
>> 
> 
> Jinhao,
> 
> Are you gonna respin this based on the irqfd patches? I suggest you just
> add this work on top and post a series that is irqfd+iothread. Then, if
> we find the irqfd ready for merge, we can pick that up for the next
> release cycle early and continue on iothread work.

signature.asc
Description: Binary data