[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1579614873-21907-1-git-send-email-jing2.liu@linux.intel.com>
Date: Tue, 21 Jan 2020 21:54:28 +0800
From: Jing Liu <jing2.liu@...ux.intel.com>
To: virtio-dev@...ts.oasis-open.org
Cc: linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
qemu-devel@...gnu.org, Jing Liu <jing2.liu@...ux.intel.com>
Subject: [virtio-dev] [PATCH v2 0/5] virtio-mmio enhancement
The current virtio over MMIO has some limitations that impact the performance.
It only supports single legacy, dedicated interrupt and one virtqueue
notification register for all virtqueues which cause performance penalties.
To address such limitations, we proposed to update virtio-mmio spec with
two new feature bits to support MSI interrupt and enhancing notification mechanism.
For keeping virtio-mmio simple as it was, and considering practical usages, this
provides two kinds of mapping mode for device: MSI non-sharing mode and MSI sharing mode.
MSI non-sharng mode indicates a fixed static vector and event relationship specified
in spec, which can simplify the setup process and reduce vmexit, fitting for
a high interrupt rate request.
MSI sharing mode indicates a dynamic mapping, which is more like PCI does, fitting
for a non-high interrupt rate request.
Change Log:
v1->v2:
* Change version update to feature bit
* Add mask/unmask support
* Add two MSI sharing/non-sharing modes
* Change MSI registers layout and bits
Jing Liu (5):
virtio-mmio: Add feature bit for MMIO notification
virtio-mmio: Enhance queue notification support
virtio-mmio: Add feature bit for MMIO MSI
virtio-mmio: Introduce MSI details
virtio-mmio: MSI vector and event mapping
content.tex | 286 ++++++++++++++++++++++++++++++++++++++++++++++++++++++------
msi-state.c | 5 ++
2 files changed, 262 insertions(+), 29 deletions(-)
create mode 100644 msi-state.c
--
2.7.4
Powered by blists - more mailing lists