Le 19/10/2021 à 12:35, Kieran Bingham a écrit :
Quoting Christophe JAILLET (2021-10-16 09:40:29)
Some resources are allocated with 'pci_request_regions()', so use
'pci_release_regions()' to free them, instead of a verbose
'release_mem_region()'.
And the driver was even already using pci_release_regions() in
tw5864_initdev(), so indeed this makes it more consistent too.
I'm curious that tw5864_initdev() calls pci_enable_device() (and
pci_disable_device in it's error path), while tw5864_finidev() doesn't.
I agree with you. I should have spotted it when I've looked at the probe
and remove functions :(
Would you like to submit a patch to fix that on top of this one? or should I?
Either way is fine for me.
If it's fine for you to fix it, I leave it to you.
A more invasive fix could be to switch to 'pcim_enable_device()' and
remove both 'pci_release_regions()' and 'pci_disable_device()' calls.
CJ
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@xxxxxxxxxxxxxxxx>
There is no point in calling 'devm_kfree()'. The corresponding resource is
managed, so it will be fried automatically.
Indeed.
Signed-off-by: Christophe JAILLET <christophe.jaillet@xxxxxxxxxx>
---
drivers/media/pci/tw5864/tw5864-core.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/media/pci/tw5864/tw5864-core.c b/drivers/media/pci/tw5864/tw5864-core.c
index 23d3cae54a5d..fee3b7711901 100644
--- a/drivers/media/pci/tw5864/tw5864-core.c
+++ b/drivers/media/pci/tw5864/tw5864-core.c
@@ -333,11 +333,9 @@ static void tw5864_finidev(struct pci_dev *pci_dev)
/* release resources */
iounmap(dev->mmio);
- release_mem_region(pci_resource_start(pci_dev, 0),
- pci_resource_len(pci_dev, 0));
+ pci_release_regions(pci_dev);
v4l2_device_unregister(&dev->v4l2_dev);
- devm_kfree(&pci_dev->dev, dev);
}
static struct pci_driver tw5864_pci_driver = {
--
2.30.2