search for: config_tun_vnet_cross_l

...p.c b/drivers/net/macvtap.c index 0327d9d..dc0a47c 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -48,11 +48,60 @@ struct macvtap_queue { #define MACVTAP_FEATURES (IFF_VNET_HDR | IFF_MULTI_QUEUE) #define MACVTAP_VNET_LE 0x80000000 +#define MACVTAP_VNET_BE 0x40000000 + +#ifdef CONFIG_TUN_VNET_CROSS_LE +static inline bool macvtap_legacy_is_little_endian(struct macvtap_queue *q) +{ + return q->flags & MACVTAP_VNET_BE ? false : + virtio_legacy_is_little_endian(); +} + +static long macvtap_get_vnet_be(struct macvtap_queue *q, int __user *sp) +{ + int s = !!(q->flags & MACVTAP_VNET_BE...

...distinct implementation for cross-endian that > don't reuse the existing code and defines then ? I think implementation and interface are fine, just the documentation can be improved a bit. How about: "Support for cross-endian vnet headers on little-endian kernels". Accordingly CONFIG_TUN_VNET_CROSS_LE ? -- MST

...distinct implementation for cross-endian that > don't reuse the existing code and defines then ? I think implementation and interface are fine, just the documentation can be improved a bit. How about: "Support for cross-endian vnet headers on little-endian kernels". Accordingly CONFIG_TUN_VNET_CROSS_LE ? -- MST

...> > don't reuse the existing code and defines then ? > > I think implementation and interface are fine, just the documentation > can be improved a bit. > > How about: > "Support for cross-endian vnet headers on little-endian kernels". > > Accordingly CONFIG_TUN_VNET_CROSS_LE > > ? > Sure. And what about also renaming the ioctl to TUNSETVNETCROSSLE then ? -- Greg

Only cosmetic and documentation changes since v5. --- Greg Kurz (8): virtio: introduce virtio_is_little_endian() helper tun: add tun_is_little_endian() helper macvtap: introduce macvtap_is_little_endian() helper vringh: introduce vringh_is_little_endian() helper vhost: introduce vhost_is_little_endian() helper virtio: add explicit big-endian support to memory

Only cosmetic and documentation changes since v5. --- Greg Kurz (8): virtio: introduce virtio_is_little_endian() helper tun: add tun_is_little_endian() helper macvtap: introduce macvtap_is_little_endian() helper vringh: introduce vringh_is_little_endian() helper vhost: introduce vhost_is_little_endian() helper virtio: add explicit big-endian support to memory

Hi, This patchset allows vhost to be used with legacy virtio when guest and host have a different endianness. It is compatible with modern virtio and can be fully compiled out through kernel config. FWIW, I could flawlessly kexec/reboot guests from ppc64 to ppc64le and back. I could also migrate from a ppc64 to a ppc64le host and back. No regressions on x86 as expected. My experimental QEMU tree

Hi, This patchset allows vhost to be used with legacy virtio when guest and host have a different endianness. It is compatible with modern virtio and can be fully compiled out through kernel config. FWIW, I could flawlessly kexec/reboot guests from ppc64 to ppc64le and back. I could also migrate from a ppc64 to a ppc64le host and back. No regressions on x86 as expected. My experimental QEMU tree

On Fri, Apr 10, 2015 at 12:20:21PM +0200, Greg Kurz wrote: > The VNET_LE flag was introduced to fix accesses to virtio 1.0 headers > that are always little-endian. It can also be used to handle the special > case of a legacy little-endian device implemented by a big-endian host. > > Let's add a flag and ioctls for big-endian devices as well. If both flags > are set,

On Fri, Apr 10, 2015 at 12:20:21PM +0200, Greg Kurz wrote: > The VNET_LE flag was introduced to fix accesses to virtio 1.0 headers > that are always little-endian. It can also be used to handle the special > case of a legacy little-endian device implemented by a big-endian host. > > Let's add a flag and ioctls for big-endian devices as well. If both flags > are set,

Setting the IF flag can cause an VM exit. So we should avoid touching the IF flag until absolutely necessary. This patch change the way the paravirt arch_local_irq_restore() works by checking the previous flags value and call arch_local_irq_enable() only if the IF flag was set previously. On a 32 vCPUs KVM guest running the AIM7 five-sec workload, the performance increased slightly from 302136.32

Setting the IF flag can cause an VM exit. So we should avoid touching the IF flag until absolutely necessary. This patch change the way the paravirt arch_local_irq_restore() works by checking the previous flags value and call arch_local_irq_enable() only if the IF flag was set previously. On a 32 vCPUs KVM guest running the AIM7 five-sec workload, the performance increased slightly from 302136.32

...=y CONFIG_MACVLAN=m # CONFIG_MACVTAP is not set # CONFIG_VXLAN is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y # CONFIG_NETCONSOLE_DYNAMIC is not set CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_RIONET=m CONFIG_RIONET_TX_SIZE=128 CONFIG_RIONET_RX_SIZE=128 # CONFIG_TUN is not set # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y # CONFIG_VIRTIO_NET is not set # CONFIG_NLMON is not set CONFIG_ARCNET=m CONFIG_ARCNET_1201=m CONFIG_ARCNET_1051=m CONFIG_ARCNET_RAW=m # CONFIG_ARCNET_CAP is not set CONFIG_ARCNET_COM90xx=m CONFIG_ARCNET_COM90xxIO=m CONFIG_ARCNET_RIM_I=m # CONFIG_ARCNET_COM20020 is not set...

...ot set # CONFIG_NET_FC is not set # CONFIG_IFB is not set # CONFIG_NET_TEAM is not set # CONFIG_MACVLAN is not set # CONFIG_VXLAN is not set # CONFIG_GENEVE is not set # CONFIG_GTP is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_TUN=y # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y CONFIG_VIRTIO_NET=y # CONFIG_NLMON is not set # CONFIG_ARCNET is not set CONFIG_ATM_DRIVERS=y # CONFIG_ATM_DUMMY is not set CONFIG_ATM_TCP=y # CONFIG_ATM_LANAI is not set # CONFIG_ATM_ENI is not set # CONFIG_ATM_FIRESTREAM is not set # CONFIG_ATM_ZATM is not set # CONFIG_A...

...ot set # CONFIG_NET_FC is not set # CONFIG_IFB is not set # CONFIG_NET_TEAM is not set # CONFIG_MACVLAN is not set # CONFIG_VXLAN is not set # CONFIG_GENEVE is not set # CONFIG_GTP is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_TUN=y # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y CONFIG_VIRTIO_NET=y # CONFIG_NLMON is not set # CONFIG_ARCNET is not set CONFIG_ATM_DRIVERS=y # CONFIG_ATM_DUMMY is not set CONFIG_ATM_TCP=y # CONFIG_ATM_LANAI is not set # CONFIG_ATM_ENI is not set # CONFIG_ATM_FIRESTREAM is not set # CONFIG_ATM_ZATM is not set # CONFIG_A...

...et > # CONFIG_NET_TEAM is not set > # CONFIG_MACVLAN is not set > # CONFIG_VXLAN is not set > # CONFIG_GENEVE is not set > # CONFIG_GTP is not set > # CONFIG_MACSEC is not set > CONFIG_NETCONSOLE=y > CONFIG_NETPOLL=y > CONFIG_NET_POLL_CONTROLLER=y > CONFIG_TUN=y > # CONFIG_TUN_VNET_CROSS_LE is not set > CONFIG_VETH=y > CONFIG_VIRTIO_NET=y > # CONFIG_NLMON is not set > # CONFIG_ARCNET is not set > CONFIG_ATM_DRIVERS=y > # CONFIG_ATM_DUMMY is not set > CONFIG_ATM_TCP=y > # CONFIG_ATM_LANAI is not set > # CONFIG_ATM_ENI is not set > # CONFIG_ATM_FIRESTREAM...

...FIG_IFB is not set # CONFIG_NET_TEAM is not set # CONFIG_MACVLAN is not set # CONFIG_VXLAN is not set # CONFIG_GENEVE is not set # CONFIG_GTP is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y # CONFIG_NETCONSOLE_DYNAMIC is not set CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_TUN=y # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y CONFIG_VIRTIO_NET=y # CONFIG_NLMON is not set # CONFIG_ARCNET is not set CONFIG_ATM_DRIVERS=y # CONFIG_ATM_DUMMY is not set CONFIG_ATM_TCP=y # CONFIG_ATM_LANAI is not set # CONFIG_ATM_ENI is not set # CONFIG_ATM_FIRESTREAM is not set # CONFIG_ATM_ZATM is not set # CONFIG_A...

...NFIG_IFB=y # CONFIG_NET_TEAM is not set CONFIG_MACVLAN=y CONFIG_MACVTAP=y # CONFIG_VXLAN is not set # CONFIG_GENEVE is not set # CONFIG_GTP is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y CONFIG_NETCONSOLE_DYNAMIC=y CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_TUN=y CONFIG_TAP=y # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y CONFIG_VIRTIO_NET=y # CONFIG_NLMON is not set CONFIG_SUNGEM_PHY=y CONFIG_ARCNET=y CONFIG_ARCNET_1201=y CONFIG_ARCNET_1051=y CONFIG_ARCNET_RAW=y CONFIG_ARCNET_CAP=y CONFIG_ARCNET_COM90xx=y CONFIG_ARCNET_COM90xxIO=y CONFIG_ARCNET_RIM_I=y CONFIG_ARCNET_COM20020=y CONFIG_ARCNE...

...NFIG_IFB=y # CONFIG_NET_TEAM is not set CONFIG_MACVLAN=y CONFIG_MACVTAP=y # CONFIG_VXLAN is not set # CONFIG_GENEVE is not set # CONFIG_GTP is not set # CONFIG_MACSEC is not set CONFIG_NETCONSOLE=y CONFIG_NETCONSOLE_DYNAMIC=y CONFIG_NETPOLL=y CONFIG_NET_POLL_CONTROLLER=y CONFIG_TUN=y CONFIG_TAP=y # CONFIG_TUN_VNET_CROSS_LE is not set CONFIG_VETH=y CONFIG_VIRTIO_NET=y # CONFIG_NLMON is not set CONFIG_SUNGEM_PHY=y CONFIG_ARCNET=y CONFIG_ARCNET_1201=y CONFIG_ARCNET_1051=y CONFIG_ARCNET_RAW=y CONFIG_ARCNET_CAP=y CONFIG_ARCNET_COM90xx=y CONFIG_ARCNET_COM90xxIO=y CONFIG_ARCNET_RIM_I=y CONFIG_ARCNET_COM20020=y CONFIG_ARCNE...

This patch series aims to move the logical block size checking to the block code. This was inspired by missing check for valid logical block size in virtio-blk which causes the kernel to crash in a weird way later on when it is invalid. I added blk_is_valid_logical_block_size which returns true iff the block size is one of supported sizes. I added this check to virtio-blk, and also converted