Re: [PATCH v28 4/5] sfc: obtain and map cxl range using devm_cxl_probe_mem

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Resource Leak in Error Path during driver probe.
--

commit 1149d7e604f29e1a5ccc2f2d5632fca5c521f213
Author: Alejandro Lucero <alucerop@amd.com>

sfc: obtain and map cxl range using devm_cxl_probe_mem

Use core API for safely obtain the CXL range linked to an HDM committed
by the BIOS. Map such a range for being used as the ctpio buffer.

> diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
> index da008462096d0..abfa0ce2b4d1e 100644
> --- a/drivers/net/ethernet/sfc/efx.c
> +++ b/drivers/net/ethernet/sfc/efx.c

[ ... ]

> @@ -1244,6 +1245,7 @@ static int efx_pci_probe(struct pci_dev *pci_dev,
>   fail3:
>  	efx_fini_io(efx);
>   fail2:
> +	efx_cxl_exit(probe_data);

[Severity: Medium]
In efx_pci_probe(), if efx_cxl_init() fails, the code jumps to fail2:

	rc = efx_cxl_init(probe_data);
	if (rc) {
		pci_err(pci_dev, "CXL initialization failed with error %d\n", rc);
		goto fail2;
	}

Does skipping fail3 here cause a resource leak by bypassing efx_fini_io()?

efx_init_io() allocates memory and reserves PCI regions that are not
devres-managed, so bypassing efx_fini_io() would leave the memory mapped
and the device enabled.

>  	efx_fini_struct(efx);
>   fail1:
>  	WARN_ON(rc > 0);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260618181806.118745-1-alejandro.lucero-palau@amd.com?part=4