Re: [PATCH v14 10/16] cxl: Cache endpoint HDM state during PCI enumeration

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] cxl: Misplaced error logging suppresses PCI_COMMAND restore failure logs
--

commit 62840404e5420d7cde318d8a3f21873bf84154cf
Author: Srirangan Madhavan <smadhavan@nvidia.com>

cxl: Cache endpoint HDM state during PCI enumeration

This commit introduces a cache for CXL HDM state during PCI enumeration
to support driver-independent resets. It temporarily modifies PCI_COMMAND
to access MMIO, reads the cache state, and then restores the PCI_COMMAND
register.

> diff --git a/drivers/cxl/core/resource.c b/drivers/cxl/core/resource.c
> index 8d2fbc49a06fb..c3f53c1b93fd8 100644
> --- a/drivers/cxl/core/resource.c
> +++ b/drivers/cxl/core/resource.c

[ ... ]

> +static struct cxl_hdm_info *cxl_pci_hdm_read_info(struct pci_dev *pdev,
> +						  struct cxl_register_map *map,
> +						  int dvsec)
> +{

[ ... ]

> +	info = __cxl_pci_hdm_read_info(pdev, map, dvsec);
> +	rc = pci_write_config_word(pdev, PCI_COMMAND, command);
> +	if (!rc)
> +		return info;
> +
> +	if (IS_ERR(info)) {
> +		pci_err(pdev,
> +			"failed to restore PCI_COMMAND after CXL HDM cache init: %d\n",
> +			pcibios_err_to_errno(rc));
> +		return info;
> +	}
> +
> +	kfree(info);
> +	return ERR_PTR(pcibios_err_to_errno(rc));
> +}

[Severity: Medium]
Is the error logging for the PCI_COMMAND restore correctly scoped here?

If __cxl_pci_hdm_read_info() succeeds (info is a valid pointer) but the
subsequent pci_write_config_word() to restore PCI_COMMAND fails, this skips
the if (IS_ERR(info)) block entirely.

As a result, it correctly frees info and returns an error pointer, but
silently suppresses the pci_err() log about the restore failure. Conversely,
if both the read and the restore fail, it logs the restore failure but
returns the read failure. 

Does this code unintentionally hide the restore failure logs when the cache
read is successful?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261001092227.3004747-1-smadhavan@nvidia.com?part=10