On 3/22/23 9:05 AM, Mathieu Poirier wrote:
Hi Tanmay,
On Tue, Mar 21, 2023 at 09:09:36PM -0700, Tanmay Shah wrote:
This patch enhances rproc_put() to support remoteproc clusters
with multiple child nodes as in rproc_get_by_phandle().
Signed-off-by: Tarak Reddy <tarak.reddy@xxxxxxx>
Signed-off-by: Tanmay Shah <tanmay.shah@xxxxxxx>
---
drivers/remoteproc/remoteproc_core.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c
index a3e7c8798381..e7e451012615 100644
--- a/drivers/remoteproc/remoteproc_core.c
+++ b/drivers/remoteproc/remoteproc_core.c
@@ -2560,6 +2560,19 @@ EXPORT_SYMBOL(rproc_free);
void rproc_put(struct rproc *rproc)
{
module_put(rproc->dev.parent->driver->owner);
There is something wrong here - this should have been removed.
Thanks Mathieu. Sure this needs to be fixed.
This is result of manually picking up patch from my side.
I will try to find better automated way to pick-up patches not available
on mailing list.
+ struct platform_device *cluster_pdev;
+
+ if (rproc->dev.parent) {
This condition is not needed, please remove.
Ack.
+ if (rproc->dev.parent->driver) {
+ module_put(rproc->dev.parent->driver->owner);
+ } else {
+ cluster_pdev = of_find_device_by_node(rproc->dev.parent->of_node->parent);
+ if (cluster_pdev) {
+ module_put(cluster_pdev->dev.driver->owner);
+ put_device(&cluster_pdev->dev);
I am not sure if cluster_pdev->dev should be dropped here.
Should we drop it in platform driver after rproc_free() ?
+ }
+ }
+ }
Some in-lined documentation, the way I did in patch 1/2 would be appreciated.
Otherwize I think the above enhancement make sense.
Ack I will document in next revision.
Thanks,
Mathieu
put_device(&rproc->dev);
Also, if we decide to drop cluster->dev here then,
should we drop reference of rproc->dev before cluster->dev ?
}
EXPORT_SYMBOL(rproc_put);
--
2.25.1
