RE: [PATCH v5 13/27] vfio/pci: Bracket PCI resets with the CXL reset hooks

[Manish Honap]

> -----Original Message-----
> From: Alex Williamson <alex@xxxxxxxxxxx>
> Sent: Tuesday, September 22, 2026 7:45 AM
> To: Manish Honap <mhonap@xxxxxxxxxx>
> Cc: jgg@xxxxxxxx; Ankit Agrawal <ankita@xxxxxxxxxx>; jic23@xxxxxxxxxx;
> dave.jiang@xxxxxxxxx; alejandro.lucero-palau@xxxxxxx; Srirangan
> Madhavan <smadhavan@xxxxxxxxxx>; corbet@xxxxxxx;
> skhan@xxxxxxxxxxxxxxxxxxx; dave@xxxxxxxxxxxx; alison.schofield@xxxxxxxxx;
> vishal.l.verma@xxxxxxxxx; iweiny@xxxxxxxxxx; ming.li@xxxxxxxxxxxx; Yishai
> Hadas <yishaih@xxxxxxxxxx>; Shameer Kolothum Thodi
> <skolothumtho@xxxxxxxxxx>; kevin.tian@xxxxxxxxx; bhelgaas@xxxxxxxxxx;
> dmatlack@xxxxxxxxxx; kees@xxxxxxxxxx; gustavoars@xxxxxxxxxx; Neo Jia
> <cjia@xxxxxxxxxx>; Krishnakant Jaju <kjaju@xxxxxxxxxx>; Vikram Sethi
> <vsethi@xxxxxxxxxx>; Zhi Wang <zhiw@xxxxxxxxxx>; linux-
> doc@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx;
> linux-cxl@xxxxxxxxxxxxxxx; linux-pci@xxxxxxxxxxxxxxx; linux-
> kselftest@xxxxxxxxxxxxxxx; linux-hardening@xxxxxxxxxxxxxxx;
> alex@xxxxxxxxxxx; Simon Song (SW-GPU) <xinmengs@xxxxxxxxxx>
> Subject: Re: [PATCH v5 13/27] vfio/pci: Bracket PCI resets with the CXL reset
> hooks
> 
> External email: Use caution opening links or attachments
> 
> 
> On Thu, 17 Sep 2026 00:05:26 +0530
> <mhonap@xxxxxxxxxx> wrote:
> 
> > From: Manish Honap <mhonap@xxxxxxxxxx>
> >
> > A CXL Type-2 device must quiesce its coherent link before any reset
> > and restore its HDM decoder state afterward, or a reset can leave the
> > device mastering the bus over decoders that are no longer valid. This
> > has to happen around every reset the PCI core drives for the device,
> > not only the ones VFIO issues directly.
> >
> > Add reset_prepare and reset_done to the CXL ops and register the
> > matching pci_error_handlers callbacks.
> >
> > Assisted-by: LLM
> > Signed-off-by: Manish Honap <mhonap@xxxxxxxxxx>
> > ---
> >  drivers/vfio/pci/vfio_pci_core.c | 18 ++++++++++++++++++
> >  include/linux/vfio_pci_core.h    |  2 ++
> >  2 files changed, 20 insertions(+)
> >
> > diff --git a/drivers/vfio/pci/vfio_pci_core.c
> > b/drivers/vfio/pci/vfio_pci_core.c
> > index 2f593b2721a1..a91e07181847 100644
> > --- a/drivers/vfio/pci/vfio_pci_core.c
> > +++ b/drivers/vfio/pci/vfio_pci_core.c
> > @@ -2794,8 +2794,26 @@ int vfio_pci_core_sriov_configure(struct
> > vfio_pci_core_device *vdev,  }
> > EXPORT_SYMBOL_GPL(vfio_pci_core_sriov_configure);
> >
> > +static void vfio_pci_core_reset_prepare(struct pci_dev *pdev) {
> > +     struct vfio_pci_core_device *vdev = dev_get_drvdata(&pdev->dev);
> > +
> > +     if (vdev->cxl_ops)
> > +             vdev->cxl_ops->reset_prepare(vdev);
> > +}
> > +
> > +static void vfio_pci_core_reset_done(struct pci_dev *pdev) {
> > +     struct vfio_pci_core_device *vdev = dev_get_drvdata(&pdev->dev);
> > +
> > +     if (vdev->cxl_ops)
> > +             vdev->cxl_ops->reset_done(vdev); }
> 
> Flagging this because it's introducing one more vfio-pci-core use of the
> convention of storing the vfio_pci_core_device in drvdata that's a barrier for a
> Rust-based variant driver.  Likely this needs to be added to the per-variant
> boilerplate or macro so variants provide the vfio_pci_core_device pointer
> rather than expecting it in drvdata.
> Thanks,
> 
> Alex

Understood. In v6 the hooks will take the device:
vfio_pci_core_reset_prepare() and vfio_pci_core_reset_done() are
exported with a struct vfio_pci_core_device argument, and vfio-pci
calls them from its own pci_error_handlers in vfio_pci.c, which already
owns the drvdata.
vfio-pci-core will not add dev_get_drvdata() call, and a
variant that binds CXL devices will call the same two above mentioned helpers
from its own handlers. I am leaving vfio_pci_core_aer_err_detected() as it is in
this series.

> 
> > +
> >  const struct pci_error_handlers vfio_pci_core_err_handlers = {
> >       .error_detected = vfio_pci_core_aer_err_detected,
> > +     .reset_prepare = vfio_pci_core_reset_prepare,
> > +     .reset_done = vfio_pci_core_reset_done,
> >  };
> >  EXPORT_SYMBOL_GPL(vfio_pci_core_err_handlers);
> >
> > diff --git a/include/linux/vfio_pci_core.h
> > b/include/linux/vfio_pci_core.h index 643f32c2141f..6ebf9a26275e
> > 100644
> > --- a/include/linux/vfio_pci_core.h
> > +++ b/include/linux/vfio_pci_core.h
> > @@ -72,6 +72,8 @@ struct vfio_cxl_ops {
> >       void    (*release)(struct vfio_pci_core_device *vdev);
> >       int     (*open_device)(struct vfio_pci_core_device *vdev);
> >       void    (*close_device)(struct vfio_pci_core_device *vdev);
> > +     void    (*reset_prepare)(struct vfio_pci_core_device *vdev);
> > +     void    (*reset_done)(struct vfio_pci_core_device *vdev);
> >       /* Pinned per bound CXL device so vfio-cxl cannot unload under usage */
> >       struct module *owner;
> >  };