On Mon, Mar 09, 2026 at 07:38:28AM -0700, Erni Sri Satya Vennela wrote:
...
> diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> b/drivers/net/ethernet/microsoft/mana/gdma_main.c
...
> @@ -2128,6 +2140,9 @@ int mana_gd_suspend(struct pci_dev *pdev, pm_message_t
> state)
>
> mana_gd_cleanup(pdev);
>
> + debugfs_remove_recursive(gc->mana_pci_debugfs);
> + gc->mana_pci_debugfs = NULL;
Hi Erni,
The same cleanup of mana_pci_debugfs already appears in a couple of other
places. It seems that all such cleanup is now paired with a call to
mana_gd_cleanup().
So could you consider performing the mana_pci_debugfs cleanup in
mana_gd_cleanup()? Possibly also renaming that function?
> +
> return 0;
> }
>
> @@ -2140,6 +2155,12 @@ int mana_gd_resume(struct pci_dev *pdev)
> struct gdma_context *gc = pci_get_drvdata(pdev);
> int err;
>
> + if (gc->is_pf)
> + gc->mana_pci_debugfs = debugfs_create_dir("0",
> mana_debugfs_root);
> + else
> + gc->mana_pci_debugfs =
> debugfs_create_dir(pci_slot_name(pdev->slot),
> + mana_debugfs_root);
Likewise the setup of mana_pci_debugfs seems to now always be paired
with a call to mana_gd_setup().
> +
> err = mana_gd_setup(pdev);
> if (err)
> return err;
...
