Re: [PATCH V2 mlx5-next 14/14] vfio/mlx5: Use its own PCI reset_done error handler
From: Yishai Hadas <yishaih@nvidia.com>
Date: 2021-10-21 10:39:21
Also in:
kvm, linux-pci
From: Yishai Hadas <yishaih@nvidia.com>
Date: 2021-10-21 10:39:21
Also in:
kvm, linux-pci
On 10/20/2021 8:45 PM, Alex Williamson wrote:
However, regardless of the migration recovery strategy, userspace needs a means to get the device back into an initial state in a deterministic way without closing and re-opening the device (or polling for an arbitrary length of time). That's the minimum viable product here. Thanks, Alex
OK, in V3 I'll use the algorithm that Jason has suggested in this mail thread to let things be done in a deterministic way without using a WQ. Thanks, Yishai