Virtio-net supports "hardware" RSS with toeplitz key.
Also, it allows receiving calculated hash in vheader
that may be used with RPS.
Added ethtools callbacks to manipulate RSS.
Technically hash calculation may be set only for
SRC+DST and SRC+DST+PORTSRC+PORTDST hashflows.
The completely disabling hash calculation for TCP or UDP
would disable hash calculation for IP.
RSS/RXHASH is disabled by default.
Changes since v4:
* Refactored code.
* Fixed sparse warnings, there would be warnings for using
restricted fields in vnet header structure.
Changes since v3:
* Moved hunks a bit.
* Added indirection table zero size check
for hash report only feature.
* Added virtio_skb_set_hash() helper instead of in-place routine.
Changes since v2:
* Fixed issue with calculating padded header length.
During review/tests, there was found an issue that
will crash the kernel if VIRTIO_NET_F_MRG_RXBUF
was not set. (thx to Jason Wang <[email protected]>)
* Refactored the code according to review.
Changes since v1:
* Refactored virtnet_set_hashflow.
* Refactored virtio_net_ctrl_rss.
* Moved hunks between patches a bit.
Changes since rfc:
* Code refactored.
* Patches reformatted.
* Added feature validation.
Andrew (1):
drivers/net/virtio_net: Added RSS hash report control.
Andrew Melnychenko (3):
drivers/net/virtio_net: Fixed padded vheader to use v1 with hash.
drivers/net/virtio_net: Added basic RSS support.
drivers/net/virtio_net: Added RSS hash report.
drivers/net/virtio_net.c | 389 +++++++++++++++++++++++++++++++++++++--
1 file changed, 376 insertions(+), 13 deletions(-)
--
2.35.1
Added features for RSS.
Added initialization, RXHASH feature and ethtool ops.
By default RSS/RXHASH is disabled.
Virtio RSS "IPv6 extensions" hashes disabled.
Added ethtools ops to set key and indirection table.
Signed-off-by: Andrew Melnychenko <[email protected]>
---
drivers/net/virtio_net.c | 192 +++++++++++++++++++++++++++++++++++++--
1 file changed, 186 insertions(+), 6 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b9ed7c55d9a0..b5f2bb426a7b 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -169,6 +169,24 @@ struct receive_queue {
struct xdp_rxq_info xdp_rxq;
};
+/* This structure can contain rss message with maximum settings for indirection table and keysize
+ * Note, that default structure that describes RSS configuration virtio_net_rss_config
+ * contains same info but can't handle table values.
+ * In any case, structure would be passed to virtio hw through sg_buf split by parts
+ * because table sizes may be differ according to the device configuration.
+ */
+#define VIRTIO_NET_RSS_MAX_KEY_SIZE 40
+#define VIRTIO_NET_RSS_MAX_TABLE_LEN 128
+struct virtio_net_ctrl_rss {
+ u32 hash_types;
+ u16 indirection_table_mask;
+ u16 unclassified_queue;
+ u16 indirection_table[VIRTIO_NET_RSS_MAX_TABLE_LEN];
+ u16 max_tx_vq;
+ u8 hash_key_length;
+ u8 key[VIRTIO_NET_RSS_MAX_KEY_SIZE];
+};
+
/* Control VQ buffers: protected by the rtnl lock */
struct control_buf {
struct virtio_net_ctrl_hdr hdr;
@@ -178,6 +196,7 @@ struct control_buf {
u8 allmulti;
__virtio16 vid;
__virtio64 offloads;
+ struct virtio_net_ctrl_rss rss;
};
struct virtnet_info {
@@ -206,6 +225,12 @@ struct virtnet_info {
/* Host will merge rx buffers for big packets (shake it! shake it!) */
bool mergeable_rx_bufs;
+ /* Host supports rss and/or hash report */
+ bool has_rss;
+ u8 rss_key_size;
+ u16 rss_indir_table_size;
+ u32 rss_hash_types_supported;
+
/* Has control virtqueue */
bool has_cvq;
@@ -2184,6 +2209,57 @@ static void virtnet_get_ringparam(struct net_device *dev,
ring->tx_pending = ring->tx_max_pending;
}
+static bool virtnet_commit_rss_command(struct virtnet_info *vi)
+{
+ struct net_device *dev = vi->dev;
+ struct scatterlist sgs[4];
+ unsigned int sg_buf_size;
+
+ /* prepare sgs */
+ sg_init_table(sgs, 4);
+
+ sg_buf_size = offsetof(struct virtio_net_ctrl_rss, indirection_table);
+ sg_set_buf(&sgs[0], &vi->ctrl->rss, sg_buf_size);
+
+ sg_buf_size = sizeof(uint16_t) * (vi->ctrl->rss.indirection_table_mask + 1);
+ sg_set_buf(&sgs[1], vi->ctrl->rss.indirection_table, sg_buf_size);
+
+ sg_buf_size = offsetof(struct virtio_net_ctrl_rss, key)
+ - offsetof(struct virtio_net_ctrl_rss, max_tx_vq);
+ sg_set_buf(&sgs[2], &vi->ctrl->rss.max_tx_vq, sg_buf_size);
+
+ sg_buf_size = vi->rss_key_size;
+ sg_set_buf(&sgs[3], vi->ctrl->rss.key, sg_buf_size);
+
+ if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_MQ,
+ VIRTIO_NET_CTRL_MQ_RSS_CONFIG, sgs)) {
+ dev_warn(&dev->dev, "VIRTIONET issue with committing RSS sgs\n");
+ return false;
+ }
+ return true;
+}
+
+static void virtnet_init_default_rss(struct virtnet_info *vi)
+{
+ u32 indir_val = 0;
+ int i = 0;
+
+ vi->ctrl->rss.hash_types = vi->rss_hash_types_supported;
+ vi->ctrl->rss.indirection_table_mask = vi->rss_indir_table_size
+ ? vi->rss_indir_table_size - 1 : 0;
+ vi->ctrl->rss.unclassified_queue = 0;
+
+ for (; i < vi->rss_indir_table_size; ++i) {
+ indir_val = ethtool_rxfh_indir_default(i, vi->curr_queue_pairs);
+ vi->ctrl->rss.indirection_table[i] = indir_val;
+ }
+
+ vi->ctrl->rss.max_tx_vq = vi->curr_queue_pairs;
+ vi->ctrl->rss.hash_key_length = vi->rss_key_size;
+
+ netdev_rss_key_fill(vi->ctrl->rss.key, vi->rss_key_size);
+}
+
static void virtnet_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
@@ -2412,6 +2488,71 @@ static void virtnet_update_settings(struct virtnet_info *vi)
vi->duplex = duplex;
}
+static u32 virtnet_get_rxfh_key_size(struct net_device *dev)
+{
+ return ((struct virtnet_info *)netdev_priv(dev))->rss_key_size;
+}
+
+static u32 virtnet_get_rxfh_indir_size(struct net_device *dev)
+{
+ return ((struct virtnet_info *)netdev_priv(dev))->rss_indir_table_size;
+}
+
+static int virtnet_get_rxfh(struct net_device *dev, u32 *indir, u8 *key, u8 *hfunc)
+{
+ struct virtnet_info *vi = netdev_priv(dev);
+ int i;
+
+ if (indir) {
+ for (i = 0; i < vi->rss_indir_table_size; ++i)
+ indir[i] = vi->ctrl->rss.indirection_table[i];
+ }
+
+ if (key)
+ memcpy(key, vi->ctrl->rss.key, vi->rss_key_size);
+
+ if (hfunc)
+ *hfunc = ETH_RSS_HASH_TOP;
+
+ return 0;
+}
+
+static int virtnet_set_rxfh(struct net_device *dev, const u32 *indir, const u8 *key, const u8 hfunc)
+{
+ struct virtnet_info *vi = netdev_priv(dev);
+ int i;
+
+ if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+ return -EOPNOTSUPP;
+
+ if (indir) {
+ for (i = 0; i < vi->rss_indir_table_size; ++i)
+ vi->ctrl->rss.indirection_table[i] = indir[i];
+ }
+ if (key)
+ memcpy(vi->ctrl->rss.key, key, vi->rss_key_size);
+
+ virtnet_commit_rss_command(vi);
+
+ return 0;
+}
+
+static int virtnet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, u32 *rule_locs)
+{
+ struct virtnet_info *vi = netdev_priv(dev);
+ int rc = 0;
+
+ switch (info->cmd) {
+ case ETHTOOL_GRXRINGS:
+ info->data = vi->curr_queue_pairs;
+ break;
+ default:
+ rc = -EOPNOTSUPP;
+ }
+
+ return rc;
+}
+
static const struct ethtool_ops virtnet_ethtool_ops = {
.supported_coalesce_params = ETHTOOL_COALESCE_MAX_FRAMES,
.get_drvinfo = virtnet_get_drvinfo,
@@ -2427,6 +2568,11 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
.set_link_ksettings = virtnet_set_link_ksettings,
.set_coalesce = virtnet_set_coalesce,
.get_coalesce = virtnet_get_coalesce,
+ .get_rxfh_key_size = virtnet_get_rxfh_key_size,
+ .get_rxfh_indir_size = virtnet_get_rxfh_indir_size,
+ .get_rxfh = virtnet_get_rxfh,
+ .set_rxfh = virtnet_set_rxfh,
+ .get_rxnfc = virtnet_get_rxnfc,
};
static void virtnet_freeze_down(struct virtio_device *vdev)
@@ -2679,6 +2825,16 @@ static int virtnet_set_features(struct net_device *dev,
vi->guest_offloads = offloads;
}
+ if ((dev->features ^ features) & NETIF_F_RXHASH) {
+ if (features & NETIF_F_RXHASH)
+ vi->ctrl->rss.hash_types = vi->rss_hash_types_supported;
+ else
+ vi->ctrl->rss.hash_types = VIRTIO_NET_HASH_REPORT_NONE;
+
+ if (!virtnet_commit_rss_command(vi))
+ return -EINVAL;
+ }
+
return 0;
}
@@ -3073,6 +3229,8 @@ static bool virtnet_validate_features(struct virtio_device *vdev)
"VIRTIO_NET_F_CTRL_VQ") ||
VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_MQ, "VIRTIO_NET_F_CTRL_VQ") ||
VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_CTRL_MAC_ADDR,
+ "VIRTIO_NET_F_CTRL_VQ") ||
+ VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_RSS,
"VIRTIO_NET_F_CTRL_VQ"))) {
return false;
}
@@ -3113,13 +3271,14 @@ static int virtnet_probe(struct virtio_device *vdev)
u16 max_queue_pairs;
int mtu;
- /* Find if host supports multiqueue virtio_net device */
- err = virtio_cread_feature(vdev, VIRTIO_NET_F_MQ,
- struct virtio_net_config,
- max_virtqueue_pairs, &max_queue_pairs);
+ /* Find if host supports multiqueue/rss virtio_net device */
+ max_queue_pairs = 1;
+ if (virtio_has_feature(vdev, VIRTIO_NET_F_MQ) || virtio_has_feature(vdev, VIRTIO_NET_F_RSS))
+ max_queue_pairs =
+ virtio_cread16(vdev, offsetof(struct virtio_net_config, max_virtqueue_pairs));
/* We need at least 2 queue's */
- if (err || max_queue_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN ||
+ if (max_queue_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN ||
max_queue_pairs > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX ||
!virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ))
max_queue_pairs = 1;
@@ -3207,6 +3366,23 @@ static int virtnet_probe(struct virtio_device *vdev)
if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF))
vi->mergeable_rx_bufs = true;
+ if (virtio_has_feature(vdev, VIRTIO_NET_F_RSS)) {
+ vi->has_rss = true;
+ vi->rss_indir_table_size =
+ virtio_cread16(vdev, offsetof(struct virtio_net_config,
+ rss_max_indirection_table_length));
+ vi->rss_key_size =
+ virtio_cread8(vdev, offsetof(struct virtio_net_config, rss_max_key_size));
+
+ vi->rss_hash_types_supported =
+ virtio_cread32(vdev, offsetof(struct virtio_net_config, supported_hash_types));
+ vi->rss_hash_types_supported &=
+ ~(VIRTIO_NET_RSS_HASH_TYPE_IP_EX |
+ VIRTIO_NET_RSS_HASH_TYPE_TCP_EX |
+ VIRTIO_NET_RSS_HASH_TYPE_UDP_EX);
+
+ dev->hw_features |= NETIF_F_RXHASH;
+ }
if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF) ||
virtio_has_feature(vdev, VIRTIO_F_VERSION_1))
vi->hdr_len = sizeof(struct virtio_net_hdr_mrg_rxbuf);
@@ -3275,6 +3451,9 @@ static int virtnet_probe(struct virtio_device *vdev)
}
}
+ if (vi->has_rss)
+ virtnet_init_default_rss(vi);
+
err = register_netdev(dev);
if (err) {
pr_debug("virtio_net: registering device failed\n");
@@ -3406,7 +3585,8 @@ static struct virtio_device_id id_table[] = {
VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ, \
VIRTIO_NET_F_CTRL_MAC_ADDR, \
VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, \
- VIRTIO_NET_F_SPEED_DUPLEX, VIRTIO_NET_F_STANDBY
+ VIRTIO_NET_F_SPEED_DUPLEX, VIRTIO_NET_F_STANDBY, \
+ VIRTIO_NET_F_RSS
static unsigned int features[] = {
VIRTNET_FEATURES,
--
2.35.1
Added features for RSS hash report.
If hash is provided - it sets to skb.
Added checks if rss and/or hash are enabled together.
Signed-off-by: Andrew Melnychenko <[email protected]>
---
drivers/net/virtio_net.c | 55 +++++++++++++++++++++++++++++++++++-----
1 file changed, 49 insertions(+), 6 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b5f2bb426a7b..c9472c30e8a2 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -227,6 +227,7 @@ struct virtnet_info {
/* Host supports rss and/or hash report */
bool has_rss;
+ bool has_rss_hash_report;
u8 rss_key_size;
u16 rss_indir_table_size;
u32 rss_hash_types_supported;
@@ -1148,6 +1149,35 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
return NULL;
}
+static void virtio_skb_set_hash(const struct virtio_net_hdr_v1_hash *hdr_hash,
+ struct sk_buff *skb)
+{
+ enum pkt_hash_types rss_hash_type;
+
+ if (!hdr_hash || !skb)
+ return;
+
+ switch ((int)hdr_hash->hash_report) {
+ case VIRTIO_NET_HASH_REPORT_TCPv4:
+ case VIRTIO_NET_HASH_REPORT_UDPv4:
+ case VIRTIO_NET_HASH_REPORT_TCPv6:
+ case VIRTIO_NET_HASH_REPORT_UDPv6:
+ case VIRTIO_NET_HASH_REPORT_TCPv6_EX:
+ case VIRTIO_NET_HASH_REPORT_UDPv6_EX:
+ rss_hash_type = PKT_HASH_TYPE_L4;
+ break;
+ case VIRTIO_NET_HASH_REPORT_IPv4:
+ case VIRTIO_NET_HASH_REPORT_IPv6:
+ case VIRTIO_NET_HASH_REPORT_IPv6_EX:
+ rss_hash_type = PKT_HASH_TYPE_L3;
+ break;
+ case VIRTIO_NET_HASH_REPORT_NONE:
+ default:
+ rss_hash_type = PKT_HASH_TYPE_NONE;
+ }
+ skb_set_hash(skb, (unsigned int)hdr_hash->hash_value, rss_hash_type);
+}
+
static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
void *buf, unsigned int len, void **ctx,
unsigned int *xdp_xmit,
@@ -1182,6 +1212,8 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
return;
hdr = skb_vnet_hdr(skb);
+ if (dev->features & NETIF_F_RXHASH && vi->has_rss_hash_report)
+ virtio_skb_set_hash((const struct virtio_net_hdr_v1_hash *)hdr, skb);
if (hdr->hdr.flags & VIRTIO_NET_HDR_F_DATA_VALID)
skb->ip_summed = CHECKSUM_UNNECESSARY;
@@ -2232,7 +2264,8 @@ static bool virtnet_commit_rss_command(struct virtnet_info *vi)
sg_set_buf(&sgs[3], vi->ctrl->rss.key, sg_buf_size);
if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_MQ,
- VIRTIO_NET_CTRL_MQ_RSS_CONFIG, sgs)) {
+ vi->has_rss ? VIRTIO_NET_CTRL_MQ_RSS_CONFIG
+ : VIRTIO_NET_CTRL_MQ_HASH_CONFIG, sgs)) {
dev_warn(&dev->dev, "VIRTIONET issue with committing RSS sgs\n");
return false;
}
@@ -3231,6 +3264,8 @@ static bool virtnet_validate_features(struct virtio_device *vdev)
VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_CTRL_MAC_ADDR,
"VIRTIO_NET_F_CTRL_VQ") ||
VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_RSS,
+ "VIRTIO_NET_F_CTRL_VQ") ||
+ VIRTNET_FAIL_ON(vdev, VIRTIO_NET_F_HASH_REPORT,
"VIRTIO_NET_F_CTRL_VQ"))) {
return false;
}
@@ -3366,8 +3401,13 @@ static int virtnet_probe(struct virtio_device *vdev)
if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF))
vi->mergeable_rx_bufs = true;
- if (virtio_has_feature(vdev, VIRTIO_NET_F_RSS)) {
+ if (virtio_has_feature(vdev, VIRTIO_NET_F_HASH_REPORT))
+ vi->has_rss_hash_report = true;
+
+ if (virtio_has_feature(vdev, VIRTIO_NET_F_RSS))
vi->has_rss = true;
+
+ if (vi->has_rss || vi->has_rss_hash_report) {
vi->rss_indir_table_size =
virtio_cread16(vdev, offsetof(struct virtio_net_config,
rss_max_indirection_table_length));
@@ -3383,8 +3423,11 @@ static int virtnet_probe(struct virtio_device *vdev)
dev->hw_features |= NETIF_F_RXHASH;
}
- if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF) ||
- virtio_has_feature(vdev, VIRTIO_F_VERSION_1))
+
+ if (vi->has_rss_hash_report)
+ vi->hdr_len = sizeof(struct virtio_net_hdr_v1_hash);
+ else if (virtio_has_feature(vdev, VIRTIO_NET_F_MRG_RXBUF) ||
+ virtio_has_feature(vdev, VIRTIO_F_VERSION_1))
vi->hdr_len = sizeof(struct virtio_net_hdr_mrg_rxbuf);
else
vi->hdr_len = sizeof(struct virtio_net_hdr);
@@ -3451,7 +3494,7 @@ static int virtnet_probe(struct virtio_device *vdev)
}
}
- if (vi->has_rss)
+ if (vi->has_rss || vi->has_rss_hash_report)
virtnet_init_default_rss(vi);
err = register_netdev(dev);
@@ -3586,7 +3629,7 @@ static struct virtio_device_id id_table[] = {
VIRTIO_NET_F_CTRL_MAC_ADDR, \
VIRTIO_NET_F_MTU, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS, \
VIRTIO_NET_F_SPEED_DUPLEX, VIRTIO_NET_F_STANDBY, \
- VIRTIO_NET_F_RSS
+ VIRTIO_NET_F_RSS, VIRTIO_NET_F_HASH_REPORT
static unsigned int features[] = {
VIRTNET_FEATURES,
--
2.35.1
Now it's possible to control supported hashflows.
Added hashflow set/get callbacks.
Also, disabling RXH_IP_SRC/DST for TCP would disable then for UDP.
TCP and UDP supports only:
ethtool -U eth0 rx-flow-hash tcp4 sd
RXH_IP_SRC + RXH_IP_DST
ethtool -U eth0 rx-flow-hash tcp4 sdfn
RXH_IP_SRC + RXH_IP_DST + RXH_L4_B_0_1 + RXH_L4_B_2_3
Disabling happens because VirtioNET hashtype for IP doesn't check L4 proto,
it works for all IP packets(TCP, UDP, ICMP, etc.).
For TCP and UDP, it's possible to set IP+PORT hashes.
But disabling IP hashes will disable them for TCP and UDP simultaneously.
It's possible to set IP+PORT for TCP/UDP and disable/enable IP
for everything else(UDP, ICMP, etc.).
Signed-off-by: Andrew Melnychenko <[email protected]>
---
drivers/net/virtio_net.c | 141 ++++++++++++++++++++++++++++++++++++++-
1 file changed, 140 insertions(+), 1 deletion(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c9472c30e8a2..17eeb4f807e3 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -231,6 +231,7 @@ struct virtnet_info {
u8 rss_key_size;
u16 rss_indir_table_size;
u32 rss_hash_types_supported;
+ u32 rss_hash_types_saved;
/* Has control virtqueue */
bool has_cvq;
@@ -2278,6 +2279,7 @@ static void virtnet_init_default_rss(struct virtnet_info *vi)
int i = 0;
vi->ctrl->rss.hash_types = vi->rss_hash_types_supported;
+ vi->rss_hash_types_saved = vi->rss_hash_types_supported;
vi->ctrl->rss.indirection_table_mask = vi->rss_indir_table_size
? vi->rss_indir_table_size - 1 : 0;
vi->ctrl->rss.unclassified_queue = 0;
@@ -2293,6 +2295,121 @@ static void virtnet_init_default_rss(struct virtnet_info *vi)
netdev_rss_key_fill(vi->ctrl->rss.key, vi->rss_key_size);
}
+static void virtnet_get_hashflow(const struct virtnet_info *vi, struct ethtool_rxnfc *info)
+{
+ info->data = 0;
+ switch (info->flow_type) {
+ case TCP_V4_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_TCPv4) {
+ info->data = RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ } else if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv4) {
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+ }
+ break;
+ case TCP_V6_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_TCPv6) {
+ info->data = RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ } else if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv6) {
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+ }
+ break;
+ case UDP_V4_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_UDPv4) {
+ info->data = RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ } else if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv4) {
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+ }
+ break;
+ case UDP_V6_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_UDPv6) {
+ info->data = RXH_IP_SRC | RXH_IP_DST |
+ RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ } else if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv6) {
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+ }
+ break;
+ case IPV4_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv4)
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+
+ break;
+ case IPV6_FLOW:
+ if (vi->rss_hash_types_saved & VIRTIO_NET_RSS_HASH_TYPE_IPv6)
+ info->data = RXH_IP_SRC | RXH_IP_DST;
+
+ break;
+ default:
+ info->data = 0;
+ break;
+ }
+}
+
+static bool virtnet_set_hashflow(struct virtnet_info *vi, struct ethtool_rxnfc *info)
+{
+ u32 new_hashtypes = vi->rss_hash_types_saved;
+ bool is_disable = info->data & RXH_DISCARD;
+ bool is_l4 = info->data == (RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 | RXH_L4_B_2_3);
+
+ /* supports only 'sd', 'sdfn' and 'r' */
+ if (!((info->data == (RXH_IP_SRC | RXH_IP_DST)) | is_l4 | is_disable))
+ return false;
+
+ switch (info->flow_type) {
+ case TCP_V4_FLOW:
+ new_hashtypes &= ~(VIRTIO_NET_RSS_HASH_TYPE_IPv4 | VIRTIO_NET_RSS_HASH_TYPE_TCPv4);
+ if (!is_disable)
+ new_hashtypes |= VIRTIO_NET_RSS_HASH_TYPE_IPv4
+ | (is_l4 ? VIRTIO_NET_RSS_HASH_TYPE_TCPv4 : 0);
+ break;
+ case UDP_V4_FLOW:
+ new_hashtypes &= ~(VIRTIO_NET_RSS_HASH_TYPE_IPv4 | VIRTIO_NET_RSS_HASH_TYPE_UDPv4);
+ if (!is_disable)
+ new_hashtypes |= VIRTIO_NET_RSS_HASH_TYPE_IPv4
+ | (is_l4 ? VIRTIO_NET_RSS_HASH_TYPE_UDPv4 : 0);
+ break;
+ case IPV4_FLOW:
+ new_hashtypes &= ~VIRTIO_NET_RSS_HASH_TYPE_IPv4;
+ if (!is_disable)
+ new_hashtypes = VIRTIO_NET_RSS_HASH_TYPE_IPv4;
+ break;
+ case TCP_V6_FLOW:
+ new_hashtypes &= ~(VIRTIO_NET_RSS_HASH_TYPE_IPv6 | VIRTIO_NET_RSS_HASH_TYPE_TCPv6);
+ if (!is_disable)
+ new_hashtypes |= VIRTIO_NET_RSS_HASH_TYPE_IPv6
+ | (is_l4 ? VIRTIO_NET_RSS_HASH_TYPE_TCPv6 : 0);
+ break;
+ case UDP_V6_FLOW:
+ new_hashtypes &= ~(VIRTIO_NET_RSS_HASH_TYPE_IPv6 | VIRTIO_NET_RSS_HASH_TYPE_UDPv6);
+ if (!is_disable)
+ new_hashtypes |= VIRTIO_NET_RSS_HASH_TYPE_IPv6
+ | (is_l4 ? VIRTIO_NET_RSS_HASH_TYPE_UDPv6 : 0);
+ break;
+ case IPV6_FLOW:
+ new_hashtypes &= ~VIRTIO_NET_RSS_HASH_TYPE_IPv6;
+ if (!is_disable)
+ new_hashtypes = VIRTIO_NET_RSS_HASH_TYPE_IPv6;
+ break;
+ default:
+ /* unsupported flow */
+ return false;
+ }
+
+ /* if unsupported hashtype was set */
+ if (new_hashtypes != (new_hashtypes & vi->rss_hash_types_supported))
+ return false;
+
+ if (new_hashtypes != vi->rss_hash_types_saved) {
+ vi->rss_hash_types_saved = new_hashtypes;
+ vi->ctrl->rss.hash_types = vi->rss_hash_types_saved;
+ if (vi->dev->features & NETIF_F_RXHASH)
+ return virtnet_commit_rss_command(vi);
+ }
+
+ return true;
+}
static void virtnet_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
@@ -2578,6 +2695,27 @@ static int virtnet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info,
switch (info->cmd) {
case ETHTOOL_GRXRINGS:
info->data = vi->curr_queue_pairs;
+ break;
+ case ETHTOOL_GRXFH:
+ virtnet_get_hashflow(vi, info);
+ break;
+ default:
+ rc = -EOPNOTSUPP;
+ }
+
+ return rc;
+}
+
+static int virtnet_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info)
+{
+ struct virtnet_info *vi = netdev_priv(dev);
+ int rc = 0;
+
+ switch (info->cmd) {
+ case ETHTOOL_SRXFH:
+ if (!virtnet_set_hashflow(vi, info))
+ rc = -EINVAL;
+
break;
default:
rc = -EOPNOTSUPP;
@@ -2606,6 +2744,7 @@ static const struct ethtool_ops virtnet_ethtool_ops = {
.get_rxfh = virtnet_get_rxfh,
.set_rxfh = virtnet_set_rxfh,
.get_rxnfc = virtnet_get_rxnfc,
+ .set_rxnfc = virtnet_set_rxnfc,
};
static void virtnet_freeze_down(struct virtio_device *vdev)
@@ -2860,7 +2999,7 @@ static int virtnet_set_features(struct net_device *dev,
if ((dev->features ^ features) & NETIF_F_RXHASH) {
if (features & NETIF_F_RXHASH)
- vi->ctrl->rss.hash_types = vi->rss_hash_types_supported;
+ vi->ctrl->rss.hash_types = vi->rss_hash_types_saved;
else
vi->ctrl->rss.hash_types = VIRTIO_NET_HASH_REPORT_NONE;
--
2.35.1
On Mon, 28 Mar 2022 20:53:32 +0300 Andrew Melnychenko wrote:
> Virtio-net supports "hardware" RSS with toeplitz key.
> Also, it allows receiving calculated hash in vheader
> that may be used with RPS.
> Added ethtools callbacks to manipulate RSS.
>
> Technically hash calculation may be set only for
> SRC+DST and SRC+DST+PORTSRC+PORTDST hashflows.
> The completely disabling hash calculation for TCP or UDP
> would disable hash calculation for IP.
>
> RSS/RXHASH is disabled by default.
# Form letter - net-next is closed
We have already sent the networking pull request for 5.18
and therefore net-next is closed for new drivers, features,
code refactoring and optimizations. We are currently accepting
bug fixes only.
Please repost when net-next reopens after 5.18-rc1 is cut.
RFC patches sent for review only are obviously welcome at any time.
On Mon, Mar 28, 2022 at 08:53:35PM +0300, Andrew Melnychenko wrote:
> +static void virtio_skb_set_hash(const struct virtio_net_hdr_v1_hash *hdr_hash,
> + struct sk_buff *skb)
> +{
> + enum pkt_hash_types rss_hash_type;
> +
> + if (!hdr_hash || !skb)
> + return;
> +
> + switch ((int)hdr_hash->hash_report) {
> + case VIRTIO_NET_HASH_REPORT_TCPv4:
> + case VIRTIO_NET_HASH_REPORT_UDPv4:
> + case VIRTIO_NET_HASH_REPORT_TCPv6:
> + case VIRTIO_NET_HASH_REPORT_UDPv6:
> + case VIRTIO_NET_HASH_REPORT_TCPv6_EX:
> + case VIRTIO_NET_HASH_REPORT_UDPv6_EX:
> + rss_hash_type = PKT_HASH_TYPE_L4;
> + break;
> + case VIRTIO_NET_HASH_REPORT_IPv4:
> + case VIRTIO_NET_HASH_REPORT_IPv6:
> + case VIRTIO_NET_HASH_REPORT_IPv6_EX:
> + rss_hash_type = PKT_HASH_TYPE_L3;
> + break;
> + case VIRTIO_NET_HASH_REPORT_NONE:
> + default:
> + rss_hash_type = PKT_HASH_TYPE_NONE;
> + }
> + skb_set_hash(skb, (unsigned int)hdr_hash->hash_value, rss_hash_type);
> +}
> +
> static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
> void *buf, unsigned int len, void **ctx,
> unsigned int *xdp_xmit,
I just noticed this- endian-ness broken here, you need _le16_to_cpu for
hash_report and __le_32_to_cpu for hash_value.
--
MST