Re: [PATCH RFC v2 18/18] tools/testing/cxl: Add Dynamic Capacity events

[Ira Weiny]

Jonathan Cameron wrote:
> On Mon, 28 Aug 2023 22:21:09 -0700
> Ira Weiny <ira.weiny@xxxxxxxxx> wrote:
> 
> > OS software needs to be alerted when new extents arrive on a Dynamic
> > Capacity Device (DCD).  On test DCDs extents are added through sysfs.
> > 
> > Add events on DCD extent injection.  Directly call the event irq
> > callback to simulate irqs to process the test extents.
> > 
> > Signed-off-by: Ira Weiny <ira.weiny@xxxxxxxxx>
> Trivial comments inline.
> 
> Reviewed-by: Jonathan.Cameron@xxxxxxxxxx>
> 
> > ---
> >  tools/testing/cxl/test/mem.c | 57 ++++++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 57 insertions(+)
> > 
> > diff --git a/tools/testing/cxl/test/mem.c b/tools/testing/cxl/test/mem.c
> > index d6041a2145c5..20364fee9df9 100644
> > --- a/tools/testing/cxl/test/mem.c
> > +++ b/tools/testing/cxl/test/mem.c
> > @@ -2008,6 +2008,41 @@ static bool new_extent_valid(struct device *dev, size_t new_start,
> >  	return false;
> >  }
> >  
> > +struct dcd_event_dyn_cap dcd_event_rec_template = {
> > +	.hdr = {
> > +		.id = UUID_INIT(0xca95afa7, 0xf183, 0x4018,
> > +				0x8c, 0x2f, 0x95, 0x26, 0x8e, 0x10, 0x1a, 0x2a),
> > +		.length = sizeof(struct dcd_event_dyn_cap),
> > +	},
> > +};
> > +
> > +static int send_dc_event(struct mock_event_store *mes, enum dc_event type,
> > +			 u64 start, u64 length, const char *tag_str)
> 
> Arguably it's not sending the event, but rather adding it to the event log and
> flicking the irq. So maybe naming needs some thought?

I spent all my naming energy on what to call extents at each layer...  ;-)

Yea I'll rename to add_dc_event() or something like that.

> 
> > +{
> > +	struct device *dev = mes->mds->cxlds.dev;
> > +	struct dcd_event_dyn_cap *dcd_event_rec;
> > +
> > +	dcd_event_rec = devm_kzalloc(dev, sizeof(*dcd_event_rec), GFP_KERNEL);
> > +	if (!dcd_event_rec)
> > +		return -ENOMEM;
> > +
> > +	memcpy(dcd_event_rec, &dcd_event_rec_template, sizeof(*dcd_event_rec));
> 
> devm_kmemdup?

Yep would work well.

Thanks again for all the review,
Ira