Re: [RFC PATCH 3/3] hw/block/nvme: add nvme_inject_state HMP command

[Minwoo Im <minwoo.im.dev@gmail.com>]

On 21-02-10 21:33:50, Klaus Jensen wrote:
> On Feb 11 04:52, Minwoo Im wrote:
> > nvme_inject_state command is to give a controller state to be.
> > Human Monitor Interface(HMP) supports users to make controller to a
> > specified state of:
> > 
> > 	normal:			Normal state (no injection)
> > 	cmd-interrupted:	Commands will be interrupted internally
> > 
> > This patch is just a start to give dynamic command from the HMP to the
> > QEMU NVMe device model.  If "cmd-interrupted" state is given, then the
> > controller will return all the CQ entries with Command Interrupts status
> > code.
> > 
> > Usage:
> > 	-device nvme,id=nvme0,....
> > 
> > 	(qemu) nvme_inject_state nvme0 cmd-interrupted
> > 
> > 	<All the commands will be interrupted internally>
> > 
> > 	(qemu) nvme_inject_state nvme0 normal
> > 
> > This feature is required to test Linux kernel NVMe driver for the
> > command retry feature.
> > 
> 
> This is super cool and commands like this feel much nicer than the
> qom-set approach that the SMART critical warning feature took.

This interface is super easy to inject some errors to the running
device with a function call-back.

> But... looking at the existing commands I don't think we can "bloat" it
> up with a device specific command like this, but I don't know the policy
> around this.

Me neither, but I've seen the PCI AER error injection feature from
the existing commands, so I suggested this command to control the
NVMe device itself like error injection.

> If an HMP command is out, then we should be able to make do with the
> qom-set approach just fine though.

Hope so.