Re: [PATCH] HID: amd_sfh: Validate PCI BAR size before mapping

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] High Severity Finding: API Usage Violation. The transport driver uses hid_input_report() instead of hid_safe_input_report(), risking out-of-bounds reads.
--

commit daa6f21fdcd011d62323dd6e5390e89dbb4f38d2
Author: Slawomir Stepien <sst@poczta.fm>

    HID: amd_sfh: Validate PCI BAR size before mapping

This commit adds a check to verify that PCI BAR 2 is at least 128KB before
mapping it with pcim_iomap_regions(). This prevents an out-of-bounds memory
access during the probe function if the driver is bound to a device with a
smaller BAR 2.

Link: https://syzkaller.appspot.com/bug?extid=4eadd4dfe9e66522bae8
Link: https://syzkaller.appspot.com/ai_job?id=3bc1c45c-548f-4ab5-8243-d2c8ec321d6c

> diff --git a/drivers/hid/amd-sfh-hid/amd_sfh_common.h b/drivers/hid/amd-sfh-hid/amd_sfh_common.h
> index 78f830c133e5c..c0c324565bee4 100644
> --- a/drivers/hid/amd-sfh-hid/amd_sfh_common.h
> +++ b/drivers/hid/amd-sfh-hid/amd_sfh_common.h
[ ... ]
> @@ -17,6 +17,8 @@
>  #define PCI_DEVICE_ID_AMD_MP2		0x15E4
>  #define PCI_DEVICE_ID_AMD_MP2_1_1	0x164A
>  
> +#define AMD_SFH_MIN_BAR_SIZE		(128 * 1024)
> +
>  #define AMD_C2P_MSG(regno) (0x10500 + ((regno) * 4))
>  #define AMD_P2C_MSG(regno) (0x10680 + ((regno) * 4))
>  
> diff --git a/drivers/hid/amd-sfh-hid/amd_sfh_pcie.c b/drivers/hid/amd-sfh-hid/amd_sfh_pcie.c
> index 4b81cebdc3359..039b6ac327d3d 100644
> --- a/drivers/hid/amd-sfh-hid/amd_sfh_pcie.c
> +++ b/drivers/hid/amd-sfh-hid/amd_sfh_pcie.c
[ ... ]
> @@ -451,6 +451,16 @@ static int amd_mp2_pci_probe(struct pci_dev *pdev, const struct pci_device_id *i
>  	if (rc)
>  		return rc;
>  
> +	if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
> +		dev_err(&pdev->dev, "BAR 2 is not IORESOURCE_MEM\n");
> +		return -ENODEV;
> +	}
> +
> +	if (pci_resource_len(pdev, 2) < AMD_SFH_MIN_BAR_SIZE) {
> +		dev_err(&pdev->dev, "BAR 2 is too small\n");
> +		return -EINVAL;
> +	}
> +
>  	rc = pcim_iomap_regions(pdev, BIT(2), DRIVER_NAME);
>  	if (rc)
>  		return rc;

[Severity: High]
This is a pre-existing issue, but does this driver use the correct input
reporting API?

In drivers/hid/amd-sfh-hid/amd_sfh_client.c:amd_sfh_work():

	if (node_type == HID_FEATURE_REPORT) {
		report_size = mp2_ops->get_feat_rep(sensor_index, report_id,
						    cli_data->feature_report[current_index]);
		if (report_size)
			hid_input_report(cli_data->hid_sensor_hubs[current_index],
					 cli_data->report_type[current_index],
					 cli_data->feature_report[current_index], report_size, 0);

And later for HID_INPUT_REPORT:

		if (report_size)
			hid_input_report(cli_data->hid_sensor_hubs[current_index],
					 cli_data->report_type[current_index],
					 in_data->input_report[current_index], report_size, 0);

Also in amd_sfh_work_buffer():

			hid_input_report(cli_data->hid_sensor_hubs[i], HID_INPUT_REPORT,
					 in_data->input_report[i], report_size, 0);

Calling hid_input_report() instead of hid_safe_input_report() in a transport
driver can lead to out-of-bounds reads in the HID core if the hardware
supplies unvalidated report sizes. Should these be changed to use
hid_safe_input_report() to provide mandatory buffer boundary checks?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/94704655-d654-4500-8122-b80bea9bd908@mail.kernel.org?part=1