On Mon, Apr 3, 2017 at 10:12 AM, Horia Geantă <horia.geanta@xxxxxxx> wrote:
> The way Job Ring platform devices are created and released does not
> allow for multiple create-release cycles.
>
> JR0 Platform device creation error
> JR0 Platform device creation error
> caam 2100000.caam: no queues configured, terminating
> caam: probe of 2100000.caam failed with error -12
>
> The reason is that platform devices are created for each job ring:
>
> for_each_available_child_of_node(nprop, np)
> if (of_device_is_compatible(np, "fsl,sec-v4.0-job-ring") ||
> of_device_is_compatible(np, "fsl,sec4.0-job-ring")) {
> ctrlpriv->jrpdev[ring] =
> of_platform_device_create(np, NULL, dev);
>
> which sets OF_POPULATED on the device node, but then it cleans these up:
>
> /* Remove platform devices for JobRs */
> for (ring = 0; ring < ctrlpriv->total_jobrs; ring++) {
> if (ctrlpriv->jrpdev[ring])
> of_device_unregister(ctrlpriv->jrpdev[ring]);
> }
>
> which leaves OF_POPULATED set.
>
> Use of_platform_populate / of_platform_depopulate instead.
> This allows for a bit of driver clean-up, jrpdev is no longer needed.
>
> Logic changes a bit too:
> -exit in case of_platform_populate fails, since currently even QI backend
> depends on JR; true, we no longer support the case when "some" of the JR
> DT nodes are incorrect
> -when cleaning up, caam_remove() would also depopulate RTIC in case
> it would have been populated somewhere else - not the case for now
>
> Fixes: 313ea293e9c4d ("crypto: caam - Add Platform driver for Job Ring")
> Reported-by: Russell King <rmk+kernel@xxxxxxxxxxxxxxx>
> Suggested-by: Rob Herring <robh+dt@xxxxxxxxxx>
> Signed-off-by: Horia Geantă <horia.geanta@xxxxxxx>
Acked-by: Rob Herring <robh@xxxxxxxxxx>
