Aneesh Kumar K.V wrote:
> Dan Williams <dan.j.williams@xxxxxxxxx> writes:
>
>
> ....
>
> > +
> > +static int pci_tsm_lock(struct pci_dev *pdev, struct tsm_dev *tsm_dev)
> > +{
> > + const struct pci_tsm_ops *ops = tsm_pci_ops(tsm_dev);
> > + struct pci_tsm *tsm;
> > + int rc;
> > +
> > + ACQUIRE(device_intr, lock)(&pdev->dev);
> > + if ((rc = ACQUIRE_ERR(device_intr, &lock)))
> > + return rc;
> > +
> > + if (pdev->dev.driver)
> > + return -EBUSY;
> > +
> > + tsm = ops->lock(pdev);
> > + if (IS_ERR(tsm))
> > + return PTR_ERR(tsm);
> > +
> > + pdev->tsm = tsm;
> > + return 0;
> > +}
> >
>
> This is slightly different from connect() callback in that we don't have
> pdev->tsm initialized when calling ->lock() callback. Should we do
> something like below? (I also included the arch changes to show how
> destructor is being used.)
Do you need to walk pdev->tsm when you are creating the tsm context?
For example, pass @pdev and the lock context structure to
rsi_device_lock()?
