search for: netif_rx_ni

...CONFIG_4KSTACKS > - local_bh_disable(); > - netif_receive_skb(skb); > - local_bh_enable(); > + if (!rx_batched) { > + local_bh_disable(); > + netif_receive_skb(skb); > + local_bh_enable(); > + } else { > + tun_rx_batched(tfile, skb, more); > + } > #else > netif_rx_ni(skb); > #endif If rx_batched has been set, and we are talking to clients not using this new MSG_MORE facility (or such clients don't have multiple TX packets to send to you, thus MSG_MORE is often clear), you are doing a lot more work per-packet than the existing code. You take the queue...

...CONFIG_4KSTACKS > - local_bh_disable(); > - netif_receive_skb(skb); > - local_bh_enable(); > + if (!rx_batched) { > + local_bh_disable(); > + netif_receive_skb(skb); > + local_bh_enable(); > + } else { > + tun_rx_batched(tfile, skb, more); > + } > #else > netif_rx_ni(skb); > #endif If rx_batched has been set, and we are talking to clients not using this new MSG_MORE facility (or such clients don't have multiple TX packets to send to you, thus MSG_MORE is often clear), you are doing a lot more work per-packet than the existing code. You take the queue...

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

...t_header(skb, 0); rxhash = skb_get_hash(skb); + #ifndef CONFIG_4KSTACKS - local_bh_disable(); - netif_receive_skb(skb); - local_bh_enable(); + if (!rx_batched) { + local_bh_disable(); + netif_receive_skb(skb); + local_bh_enable(); + } else { + tun_rx_batched(tfile, skb, more); + } #else netif_rx_ni(skb); #endif @@ -1312,7 +1348,8 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from) if (!tun) return -EBADFD; - result = tun_get_user(tun, tfile, NULL, from, file->f_flags & O_NONBLOCK); + result = tun_get_user(tun, tfile, NULL, from, + file->f_...

...er(skb, 0); rxhash = skb_get_hash(skb); + #ifndef CONFIG_4KSTACKS - local_bh_disable(); - netif_receive_skb(skb); - local_bh_enable(); + if (!rx_batched) { + local_bh_disable(); + netif_receive_skb(skb); + local_bh_enable(); + } else { + err = tun_rx_batched(tfile, skb, more); + } #else netif_rx_ni(skb); #endif stats = get_cpu_ptr(tun->pcpu_stats); u64_stats_update_begin(&stats->syncp); - stats->rx_packets++; - stats->rx_bytes += len; + if (err) { + stats->rx_dropped++; + } else { + stats->rx_packets++; + stats->rx_bytes += len; + } u64_stats_update_end(&a...

...}; struct sk_buff *skb; @@ -1283,9 +1321,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, rxhash = skb_get_hash(skb); #ifndef CONFIG_4KSTACKS - local_bh_disable(); - netif_receive_skb(skb); - local_bh_enable(); + tun_rx_batched(tun, tfile, skb, more); #else netif_rx_ni(skb); #endif @@ -1311,7 +1347,8 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from) if (!tun) return -EBADFD; - result = tun_get_user(tun, tfile, NULL, from, file->f_flags & O_NONBLOCK); + result = tun_get_user(tun, tfile, NULL, from, + file->f_...

...20,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, skb_probe_transport_header(skb, 0); rxhash = skb_get_hash(skb); + #ifndef CONFIG_4KSTACKS - local_bh_disable(); - netif_receive_skb(skb); - local_bh_enable(); + tun_rx_batched(tun, tfile, skb, more); #else netif_rx_ni(skb); #endif @@ -1312,7 +1348,8 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from) if (!tun) return -EBADFD; - result = tun_get_user(tun, tfile, NULL, from, file->f_flags & O_NONBLOCK); + result = tun_get_user(tun, tfile, NULL, from, + file->f_...

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

vhost_net_ubuf_put_and_wait has a confusing name: it will actually also free it's argument. Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01 "vhost-net: flush outstanding DMAs on memory change" vhost_net_flush tries to use the argument after passing it to vhost_net_ubuf_put_and_wait, this results in use after free. To fix, don't free the argument in

vhost_net_ubuf_put_and_wait has a confusing name: it will actually also free it's argument. Thus since commit 1280c27f8e29acf4af2da914e80ec27c3dbd5c01 "vhost-net: flush outstanding DMAs on memory change" vhost_net_flush tries to use the argument after passing it to vhost_net_ubuf_put_and_wait, this results in use after free. To fix, don't free the argument in

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

...tun, struct tun_file *tfile, > skb_probe_transport_header(skb, 0); > > rxhash = skb_get_hash(skb); > + > #ifndef CONFIG_4KSTACKS > - local_bh_disable(); > - netif_receive_skb(skb); > - local_bh_enable(); > + tun_rx_batched(tun, tfile, skb, more); > #else > netif_rx_ni(skb); > #endif > @@ -1312,7 +1348,8 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from) > if (!tun) > return -EBADFD; > > - result = tun_get_user(tun, tfile, NULL, from, file->f_flags & O_NONBLOCK); > + result = tun_get_user(tun, tfile...

...tun, struct tun_file *tfile, > skb_probe_transport_header(skb, 0); > > rxhash = skb_get_hash(skb); > + > #ifndef CONFIG_4KSTACKS > - local_bh_disable(); > - netif_receive_skb(skb); > - local_bh_enable(); > + tun_rx_batched(tun, tfile, skb, more); > #else > netif_rx_ni(skb); > #endif > @@ -1312,7 +1348,8 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from) > if (!tun) > return -EBADFD; > > - result = tun_get_user(tun, tfile, NULL, from, file->f_flags & O_NONBLOCK); > + result = tun_get_user(tun, tfile...

...fffff814eb2f4>] net_rx_action+0x134/0x290 [ 5445.470445] [<ffffffff81065ea8>] __do_softirq+0xa8/0x210 [ 5445.478364] [<ffffffff8160831c>] call_softirq+0x1c/0x30 [ 5445.486145] <EOI> [<ffffffff8100c295>] ? do_softirq+0x65/0xa0 [ 5445.494076] [<ffffffff814eb7d8>] netif_rx_ni+0x28/0x30 [ 5445.501902] [<ffffffff8144c702>] tun_chr_aio_write+0x342/0x4f0 [ 5445.509771] [<ffffffff8144c3c0>] ? tun_chr_aio_read+0xd0/0xd0 [ 5445.517864] [<ffffffff81167db3>] do_sync_readv_writev+0xd3/0x110 [ 5445.526011] [<ffffffff81281ecc>] ? security_file_permission...

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

Hi: This series tries to implement tx batching support for vhost. This was done by using MSG_MORE as a hint for under layer socket. The backend (e.g tap) can then batch the packets temporarily in a list and submit it all once the number of bacthed exceeds a limitation. Tests shows obvious improvement on guest pktgen over over mlx4(noqueue) on host: Mpps -+%

Hi: Here are the patches I used for testing KVM with virtio-net using TSO. There are three patches for the tun device which are basically Rusty's patches with the mmap turned into copying (for correctness). Two patches are for the virtio-net frontend, one required to support receiving SG/TSO, and the other useful for testing SG per se. The other patch is to the KVM backend to make all this