Re: [PATCH v6 2/4] s390/pci: Reuse FMB buffer and preserve state in device re-enablement

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] s390/pci: zpci_fmb_reenable_device() unconditionally enables FMB
--

commit 7c558501997618a3d6ef7ed7b44346ed809628a3
Author: Omar Elghoul <oelghoul@linux.ibm.com>

s390/pci: Reuse FMB buffer and preserve state in device re-enablement

This commit introduces zpci_fmb_reenable_device() to check and ensure the
function measurement is enabled. It avoids freeing the FMB buffer during
disabling, allowing it to be reused when re-enabling measurement. To track
this, it adds an fmb_enabled flag to the zpci_dev structure.

> diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
> index c055a9ad09724..5d1f5b75bff5e 100644
> --- a/arch/s390/pci/pci.c
> +++ b/arch/s390/pci/pci.c
[ ... ]
> @@ -224,13 +229,39 @@ int zpci_fmb_disable_device(struct zpci_dev *zdev)
[ ... ]
> +int zpci_fmb_reenable_device(struct zpci_dev *zdev)
> +{
> +	u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_SET_MEASURE);
> +	struct zpci_fib fib = {0};
> +	u8 cc, status;
> +
> +	lockdep_assert_held(&zdev->fmb_lock);
> +
> +	if (!zdev->fmb_enabled)
> +		return zpci_fmb_enable_device(zdev);

[Severity: Medium]
If FMB is explicitly disabled by a user via debugfs or by a guest VM via the
VFIO FMB feature, will a subsequent device hot reset silently force FMB back
on?

During a hot reset, a sequence like zpci_hot_reset_device() ->
zpci_reenable_device() -> zpci_fmb_reenable_device() is triggered. If a guest
VM or user previously disabled FMB, zdev->fmb_enabled will be 0.

Does returning zpci_fmb_enable_device() here unconditionally enable it again,
overriding the user or guest VM configuration and going against the commit's
stated goal to preserve state?

> +
> +	fib.gd = zdev->gisa;
> +	cc = zpci_mod_fc(req, &fib, &status); /* Disable function measurement */
[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916213222.233-1-oelghoul@linux.ibm.com?part=2