> -----Original Message----- > From: Jonathan Cameron <[email protected]> > Sent: Saturday, September 26, 2026 2:11 AM > To: Manish Honap <[email protected]> > Cc: [email protected]; [email protected]; Ankit Agrawal <[email protected]>; > [email protected]; [email protected]; Srirangan > Madhavan <[email protected]>; [email protected]; > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; [email protected]; Yishai > Hadas <[email protected]>; Shameer Kolothum Thodi > <[email protected]>; [email protected]; [email protected]; > [email protected]; [email protected]; [email protected]; Neo Jia > <[email protected]>; Krishnakant Jaju <[email protected]>; Vikram Sethi > <[email protected]>; Zhi Wang <[email protected]>; linux- > [email protected]; [email protected]; [email protected]; > [email protected]; [email protected]; linux- > [email protected]; [email protected] > Subject: Re: [PATCH v5 04/27] cxl: Add cxl_reset_dvsec_sequence() for vfio-pci > > External email: Use caution opening links or attachments > > > On Thu, 17 Sep 2026 00:05:17 +0530 > <[email protected]> wrote: > > > From: Manish Honap <[email protected]> > > > > vfio-pci needs to run a CXL reset on a passed-through Type-2 device > > without the host-memory handling in cxl_reset_function(): the memory > > behind the decoder belongs to the guest, so the HDM range collection > > and host CPU cache flush that the host path performs must be skipped. > > Hi Manish, > > This cache flush being skipped bit concerns me. Who is issuing the necessary > flushes in the vfio case? That is there to ensure we don't end up with random > lines cached by PA (as well as VA if relevant), so they affect a guest just as > much as the host. > I'm not sure why it makes any difference who the memory belongs to.
Yes, you are right. I feel the reasoning in the commit message is wrong. cxl_reset v14 already makes the invalidation mandatory on the host path: it invalidates before the reset and again after the restore, and refuses the reset when cpu_cache_has_invalidate_memregion() is false. For v6, I will run the same sequence in vfio reset entry. On a host where no cache maintenance driver has registered, invalidate memregion check fails and CXL reset can't be run. vfio-cxl checks the same condition at bind and continues without CXL support, so it never gives a guest HDM memory it could not clear and invalidate on the next reset. > > > > > Add cxl_reset_dvsec_sequence(), which runs the DVSEC reset under the > > device lock and restores HDM state through > > cxl_restore_state_after_pci_reset(), and cxl_reset_capable() to gate > > it on a function-scoped reset. Both are exported to the vfio-cxl > > module so vfio-pci core does not import the CXL namespace. > > > > CXL Reset always clears memory, so there is no caller Memory Clear choice. > > > > Assisted-by: LLM > > Signed-off-by: Manish Honap <[email protected]>
