similar to: [PATCH 1/2] vdpa: ifcvf: return err when fail to request config irq

On 2020/4/24 ??6:04, Zhu Lingshan wrote: > This commit implements config interrupt support > in IFC VF > > Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> > --- > drivers/vdpa/ifcvf/ifcvf_base.c | 3 +++ > drivers/vdpa/ifcvf/ifcvf_base.h | 2 ++ > drivers/vdpa/ifcvf/ifcvf_main.c | 22 +++++++++++++++++++++- > 3 files changed, 26 insertions(+), 1

Formerly, ifcvf driver has implemented a lazy-initialization mechanism for the virtqueues and other config space contents, it would store all configurations that passed down from the userspace, then load them to the device config space upon DRIVER_OK. This can not serve live migration, so this series implement an immediate initialization mechanism, which means rather than the former store-load

On 2020/7/13 ??6:22, Zhu, Lingshan wrote: > > > On 7/13/2020 4:33 PM, Jason Wang wrote: >> >> On 2020/7/12 ??10:49, Zhu Lingshan wrote: >>> This commit replaced irq_request/free() with helpers in vDPA >>> core, so that it can request/free irq and setup irq offloading >>> on order. >>> >>> Signed-off-by: Zhu Lingshan <lingshan.zhu

On 2020/7/13 ??6:22, Zhu, Lingshan wrote: > > > On 7/13/2020 4:33 PM, Jason Wang wrote: >> >> On 2020/7/12 ??10:49, Zhu Lingshan wrote: >>> This commit replaced irq_request/free() with helpers in vDPA >>> core, so that it can request/free irq and setup irq offloading >>> on order. >>> >>> Signed-off-by: Zhu Lingshan <lingshan.zhu

Formerly, ifcvf driver has implemented a lazy-initialization mechanism for the virtqueues and other config space contents, it would store all configurations that passed down from the userspace, then load them to the device config space upon DRIVER_OK. This can not serve live migration, so this series implement an immediate initialization mechanism, which means rather than the former store-load

This commit synchronize irqs of the virtqueues and config space in the reset routine. Thus ifcvf_stop_hw() and reset() are refactored as well. Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> --- drivers/vdpa/ifcvf/ifcvf_base.c | 41 +++++++++++++++++++++-------- drivers/vdpa/ifcvf/ifcvf_base.h | 1 + drivers/vdpa/ifcvf/ifcvf_main.c | 46 +++++---------------------------- 3 files

On Mon, May 8, 2023 at 6:05?PM Zhu Lingshan <lingshan.zhu at intel.com> wrote: > > This commit synchronize irqs of the virtqueues > and config space in the reset routine. > Thus ifcvf_stop_hw() and reset() are refactored as well. > > Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> > --- > drivers/vdpa/ifcvf/ifcvf_base.c | 41

On 2020/7/23 ??5:12, Jason Wang wrote: > We ignore the err of requesting config interrupt, fix this. > > Fixes: e7991f376a4d ("ifcvf: implement config interrupt in IFCVF") > Cc: Zhu Lingshan <lingshan.zhu at intel.com> > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vdpa/ifcvf/ifcvf_main.c | 4 ++++ > 1 file changed, 4

In this commit, virtqueue operations including: set_vq_num(), set_vq_address(), set_vq_ready() and get_vq_ready() access PCI registers directly to take immediate actions. Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> --- drivers/vdpa/ifcvf/ifcvf_base.c | 58 ++++++++++++++++++++------------- drivers/vdpa/ifcvf/ifcvf_base.h | 10 +++--- drivers/vdpa/ifcvf/ifcvf_main.c | 16

This commit implements a new function ifcvf_get_driver_feature() which read driver_features from virtio registers. To be less ambiguous, ifcvf_set_features() is renamed to ifcvf_set_driver_features(), and ifcvf_get_features() is renamed to ifcvf_get_dev_features() which returns the provisioned vDPA device features. Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> ---

On Thu, Mar 26, 2020 at 3:08 PM Jason Wang <jasowang at redhat.com> wrote: > > From: Zhu Lingshan <lingshan.zhu at intel.com> > > This commit introduced two layers to drive IFC VF: > > (1) ifcvf_base layer, which handles IFC VF NIC hardware operations and > configurations. > > (2) ifcvf_main layer, which complies to VDPA bus framework, >

On 2020/4/26 ??4:42, Zhu Lingshan wrote: > This commit implements config interrupt support > in IFC VF > > Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> > --- > drivers/vdpa/ifcvf/ifcvf_base.c | 3 +++ > drivers/vdpa/ifcvf/ifcvf_base.h | 3 +++ > drivers/vdpa/ifcvf/ifcvf_main.c | 22 +++++++++++++++++++++- > 3 files changed, 27 insertions(+), 1

On 2020/7/12 ??10:49, Zhu Lingshan wrote: > This commit replaced irq_request/free() with helpers in vDPA > core, so that it can request/free irq and setup irq offloading > on order. > > Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> > --- > drivers/vdpa/ifcvf/ifcvf_main.c | 11 ++++++----- > 1 file changed, 6 insertions(+), 5 deletions(-) > > diff

On 2020/7/16 ??7:23, Zhu Lingshan wrote: > This commit replaced irq_request/free() with helpers in vDPA > core, so that it can request/free irq and setup irq offloading > on order. > > Signed-off-by: Zhu Lingshan <lingshan.zhu at intel.com> > Suggested-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vdpa/ifcvf/ifcvf_main.c | 14 +++++++++----- > 1

Live migration of vdpa would typically require re-instate vdpa device with an idential set of configs on the destination node, same way as how source node created the device in the first place. In order to allow live migration orchestration software to export the initial set of vdpa attributes with which the device was created, it will be useful if the vdpa tool can report the config on demand

Now that many more architectures build vhost, a couple of these (um, and arm with deprecated oabi) have reported build failures with randconfig, however fixes for that need a bit more discussion/testing and will be merged separately. Not a regression - these previously simply didn't have vhost at all. Also, there's some DMA API code in the vdpa simulator is hacky - if no solution surfaces

Now that many more architectures build vhost, a couple of these (um, and arm with deprecated oabi) have reported build failures with randconfig, however fixes for that need a bit more discussion/testing and will be merged separately. Not a regression - these previously simply didn't have vhost at all. Also, there's some DMA API code in the vdpa simulator is hacky - if no solution surfaces

get_vq_align returns u16 now, but that's not enough for systems/devices with 64K pages. All callers assign it to a u32 variable anyway, so let's just change the return value type to u32. Cc: "Zhu, Lingshan" <lingshan.zhu at intel.com> Reported-by: Arnd Bergmann <arnd at arndb.de> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

The following changes since commit ceaa837f96adb69c0df0397937cd74991d5d821a: Linux 6.2-rc8 (2023-02-12 14:10:17 -0800) are available in the Git repository at: https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git tags/for_linus for you to fetch changes up to deeacf35c922da579637f5db625af20baafc66ed: vdpa/mlx5: support device features provisioning (2023-02-20 19:27:00 -0500)

On 2020/5/11 ??6:11, Zhu, Lingshan wrote: > > > On 5/11/2020 5:26 PM, Jason Wang wrote: >> >> On 2020/5/11 ??3:19, Zhu Lingshan wrote: >>> This commit move IRQ request and free operations from probe() >>> to VIRTIO status change handler to comply with VIRTIO spec. >>> >>> VIRTIO spec 1.1, section 2.1.2 Device Requirements: Device Status