| Message ID | 20220929014555.112323-1-lingshan.zhu@intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Conditionally read fields in dev cfg space | expand |
Hi Michael, Jason starts his vacation this afternoon, and next week is our national holiday. He has acked 3 ~ 6 of this series before, and I have made improvements based on his comments. Do you have any comments on patches 1 and 2? Thanks, Zhu Lingshan On 9/29/2022 9:45 AM, Zhu Lingshan wrote: > This series intends to read the fields in virtio-net device > configuration space conditionally on the feature bits, > this means: > > MTU exists if VIRTIO_NET_F_MTU is set > MAC exists if VIRTIO_NET_F_NET is set > MQ exists if VIRTIO_NET_F_MQ or VIRTIO_NET_F_RSS is set. > > This series report device features to userspace and invokes > vdpa_config_ops.get_config() rather than > vdpa_get_config_unlocked() to read the device config spcae, > so no races in vdpa_set_features_unlocked() > > Thanks! > > Changes form V2: > remove unnacessary checking for vdev->config->get_status (Jason) > > Changes from V1: > 1)Better comments for VDPA_ATTR_VDPA_DEV_SUPPORTED_FEATURES, > only in the header file(Jason) > 2)Split original 3/4 into separate patches(Jason) > 3)Check FEATURES_OK for reporting driver features > in vdpa_dev_config_fill (Jason) > 4) Add iproute2 example for reporting device features > > Zhu Lingshan (6): > vDPA: allow userspace to query features of a vDPA device > vDPA: only report driver features if FEATURES_OK is set > vDPA: check VIRTIO_NET_F_RSS for max_virtqueue_paris's presence > vDPA: check virtio device features to detect MQ > vDPA: fix spars cast warning in vdpa_dev_net_mq_config_fill > vDPA: conditionally read MTU and MAC in dev cfg space > > drivers/vdpa/vdpa.c | 68 ++++++++++++++++++++++++++++++--------- > include/uapi/linux/vdpa.h | 4 +++ > 2 files changed, 56 insertions(+), 16 deletions(-) >
On Thu, Sep 29, 2022 at 03:23:46PM +0800, Zhu, Lingshan wrote: > Hi Michael, > > Jason starts his vacation this afternoon, and next week is our national > holiday. > He has acked 3 ~ 6 of this series before, and I have made improvements based > on his comments. > Do you have any comments on patches 1 and 2? No, I'll merge for next. > Thanks, > Zhu Lingshan > On 9/29/2022 9:45 AM, Zhu Lingshan wrote: > > This series intends to read the fields in virtio-net device > > configuration space conditionally on the feature bits, > > this means: > > > > MTU exists if VIRTIO_NET_F_MTU is set > > MAC exists if VIRTIO_NET_F_NET is set > > MQ exists if VIRTIO_NET_F_MQ or VIRTIO_NET_F_RSS is set. > > > > This series report device features to userspace and invokes > > vdpa_config_ops.get_config() rather than > > vdpa_get_config_unlocked() to read the device config spcae, > > so no races in vdpa_set_features_unlocked() > > > > Thanks! > > > > Changes form V2: > > remove unnacessary checking for vdev->config->get_status (Jason) > > > > Changes from V1: > > 1)Better comments for VDPA_ATTR_VDPA_DEV_SUPPORTED_FEATURES, > > only in the header file(Jason) > > 2)Split original 3/4 into separate patches(Jason) > > 3)Check FEATURES_OK for reporting driver features > > in vdpa_dev_config_fill (Jason) > > 4) Add iproute2 example for reporting device features > > > > Zhu Lingshan (6): > > vDPA: allow userspace to query features of a vDPA device > > vDPA: only report driver features if FEATURES_OK is set > > vDPA: check VIRTIO_NET_F_RSS for max_virtqueue_paris's presence > > vDPA: check virtio device features to detect MQ > > vDPA: fix spars cast warning in vdpa_dev_net_mq_config_fill > > vDPA: conditionally read MTU and MAC in dev cfg space > > > > drivers/vdpa/vdpa.c | 68 ++++++++++++++++++++++++++++++--------- > > include/uapi/linux/vdpa.h | 4 +++ > > 2 files changed, 56 insertions(+), 16 deletions(-) > >
On 9/29/2022 3:38 PM, Michael S. Tsirkin wrote: > On Thu, Sep 29, 2022 at 03:23:46PM +0800, Zhu, Lingshan wrote: >> Hi Michael, >> >> Jason starts his vacation this afternoon, and next week is our national >> holiday. >> He has acked 3 ~ 6 of this series before, and I have made improvements based >> on his comments. >> Do you have any comments on patches 1 and 2? > > No, I'll merge for next. Thanks! > >> Thanks, >> Zhu Lingshan >> On 9/29/2022 9:45 AM, Zhu Lingshan wrote: >>> This series intends to read the fields in virtio-net device >>> configuration space conditionally on the feature bits, >>> this means: >>> >>> MTU exists if VIRTIO_NET_F_MTU is set >>> MAC exists if VIRTIO_NET_F_NET is set >>> MQ exists if VIRTIO_NET_F_MQ or VIRTIO_NET_F_RSS is set. >>> >>> This series report device features to userspace and invokes >>> vdpa_config_ops.get_config() rather than >>> vdpa_get_config_unlocked() to read the device config spcae, >>> so no races in vdpa_set_features_unlocked() >>> >>> Thanks! >>> >>> Changes form V2: >>> remove unnacessary checking for vdev->config->get_status (Jason) >>> >>> Changes from V1: >>> 1)Better comments for VDPA_ATTR_VDPA_DEV_SUPPORTED_FEATURES, >>> only in the header file(Jason) >>> 2)Split original 3/4 into separate patches(Jason) >>> 3)Check FEATURES_OK for reporting driver features >>> in vdpa_dev_config_fill (Jason) >>> 4) Add iproute2 example for reporting device features >>> >>> Zhu Lingshan (6): >>> vDPA: allow userspace to query features of a vDPA device >>> vDPA: only report driver features if FEATURES_OK is set >>> vDPA: check VIRTIO_NET_F_RSS for max_virtqueue_paris's presence >>> vDPA: check virtio device features to detect MQ >>> vDPA: fix spars cast warning in vdpa_dev_net_mq_config_fill >>> vDPA: conditionally read MTU and MAC in dev cfg space >>> >>> drivers/vdpa/vdpa.c | 68 ++++++++++++++++++++++++++++++--------- >>> include/uapi/linux/vdpa.h | 4 +++ >>> 2 files changed, 56 insertions(+), 16 deletions(-) >>>