Re: [PATCH v20 2/3] vfio-pci/zdev: Add a device feature for error information

[Alex Williamson]

On Tue, 30 Jun 2026 09:55:52 -0700
Farhan Ali <alifm@xxxxxxxxxxxxx> wrote:

> For zPCI devices, we have platform specific error information. The platform
> firmware provides this error information to the operating system in an
> architecture specific mechanism. To enable recovery from userspace for
> these devices, we want to expose this error information to userspace. Add a
> new device feature to expose this information.
> 
> Userspace needs to be provide a buffer of fixed size. This size is provided
> to userspace via the VFIO_DEVICE_INFO_CAP_ZPCI_BASE capability.
> 
> Reviewed-by: Matthew Rosato <mjrosato@xxxxxxxxxxxxx>
> Reviewed-by: Niklas Schnelle <schnelle@xxxxxxxxxxxxx>
> Signed-off-by: Farhan Ali <alifm@xxxxxxxxxxxxx>
> ---
>  arch/s390/include/asm/pci.h      |  2 ++
>  arch/s390/pci/pci_event.c        | 19 ++++++++++++++
>  drivers/vfio/pci/vfio_pci_core.c |  2 ++
>  drivers/vfio/pci/vfio_pci_priv.h |  9 +++++++
>  drivers/vfio/pci/vfio_pci_zdev.c | 45 ++++++++++++++++++++++++++++++--
>  include/uapi/linux/vfio.h        | 23 ++++++++++++++++
>  include/uapi/linux/vfio_zdev.h   |  2 ++
>  7 files changed, 100 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
> index 016386f7ef4a..88a125b92bdd 100644
> --- a/arch/s390/include/asm/pci.h
> +++ b/arch/s390/include/asm/pci.h
> @@ -364,6 +364,8 @@ int zpci_clear_error_state(struct zpci_dev *zdev);
>  int zpci_reset_load_store_blocked(struct zpci_dev *zdev);
>  void zpci_start_mediated_recovery(struct zpci_dev *zdev);
>  void zpci_stop_mediated_recovery(struct zpci_dev *zdev);
> +int zpci_get_pending_error(struct zpci_dev *zdev,
> +			   struct zpci_ccdf_err *ccdf);
>  
>  #ifdef CONFIG_NUMA
>  
> diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c
> index 8934b640a095..3ad16038e005 100644
> --- a/arch/s390/pci/pci_event.c
> +++ b/arch/s390/pci/pci_event.c
> @@ -77,6 +77,25 @@ static int zpci_store_pci_error(struct pci_dev *pdev,
>  	return 0;
>  }
>  
> +int zpci_get_pending_error(struct zpci_dev *zdev,
> +			   struct zpci_ccdf_err *ccdf)
> +{
> +	int head = 0;

Nit, unnecessary initialization.

Otherwise LGTM and I can drop the initialization on commit if there's
no other reason for respin.  I think we just need a branch from Bjorn
to base this on once he's happy with the PCI series.  Thanks,

Alex

> +
> +	guard(mutex)(&zdev->pending_errs_lock);
> +
> +	if (!zdev->pending_errs.count)
> +		return -ENOMSG;
> +
> +	head = zdev->pending_errs.head % ZPCI_ERR_PENDING_MAX;
> +	memcpy(ccdf, &zdev->pending_errs.err[head],
> +	       sizeof(struct zpci_ccdf_err));
> +	zdev->pending_errs.head++;
> +	zdev->pending_errs.count--;
> +	return 0;
> +}
> +EXPORT_SYMBOL_GPL(zpci_get_pending_error);
> +
>  void zpci_start_mediated_recovery(struct zpci_dev *zdev)
>  {
>  	guard(mutex)(&zdev->pending_errs_lock);