* [PATCH 1/3] Add support for rx flow hash configuration in a network device - [ethtool]
@ 2008-06-28 0:06 Santwona Behera
2008-09-14 1:23 ` Jeff Garzik
0 siblings, 1 reply; 2+ messages in thread
From: Santwona Behera @ 2008-06-28 0:06 UTC (permalink / raw)
To: netdev, davem, jeff
Cc: Matheos Worku, Michael Speer, gkernel-commit, Santwona Behera
[-- Attachment #1: Type: text/plain, Size: 1 bytes --]
[-- Attachment #2: ethtool_rx_flow_hash_cfg.patch --]
[-- Type: text/x-patch, Size: 12869 bytes --]
From: Santwona Behera <santwona.behera@sun.com>
Added new interfaces to ethtool to configure receive network flow distribution across multiple rx rings using hashing.
Patch is against ethtool version 6.
Signed-off-by: Santwona Behera <santwona.behera@sun.com>
---
ethtool-copy.h | 33 +++++++
ethtool.8 | 80 ++++++++++++++++++
ethtool.c | 254 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 367 insertions(+), 0 deletions(-)
diff --git a/ethtool-copy.h b/ethtool-copy.h
index 3a63224..e2a72f1 100644
--- a/ethtool-copy.h
+++ b/ethtool-copy.h
@@ -256,6 +256,12 @@ struct ethtool_perm_addr {
__u8 data[0];
};
+struct ethtool_rxnfc {
+ __u32 cmd;
+ __u32 flow_type;
+ __u64 data;
+};
+
#ifdef __KERNEL__
struct net_device;
@@ -373,6 +379,8 @@ struct ethtool_ops {
void (*complete)(struct net_device *);
u32 (*get_ufo)(struct net_device *);
int (*set_ufo)(struct net_device *, u32);
+ int (*get_rxhash)(struct net_device *, struct ethtool_rxnfc *);
+ int (*set_rxhash)(struct net_device *, struct ethtool_rxnfc *);
};
#endif /* __KERNEL__ */
@@ -415,6 +423,9 @@ struct ethtool_ops {
#define ETHTOOL_GGSO 0x00000023 /* Get GSO enable (ethtool_value) */
#define ETHTOOL_SGSO 0x00000024 /* Set GSO enable (ethtool_value) */
+#define ETHTOOL_GRXFH 0x00000029 /* Get RX flow hash configuration */
+#define ETHTOOL_SRXFH 0x0000002a /* Set RX flow hash configuration */
+
/* compatibility with older code */
#define SPARC_ETH_GSET ETHTOOL_GSET
#define SPARC_ETH_SSET ETHTOOL_SSET
@@ -501,4 +512,26 @@ struct ethtool_ops {
#define WAKE_MAGIC (1 << 5)
#define WAKE_MAGICSECURE (1 << 6) /* only meaningful if WAKE_MAGIC */
+/* L3-L4 network traffic flow types */
+#define TCP_V4_FLOW 0x01
+#define UDP_V4_FLOW 0x02
+#define SCTP_V4_FLOW 0x03
+#define AH_ESP_V4_FLOW 0x04
+#define TCP_V6_FLOW 0x05
+#define UDP_V6_FLOW 0x06
+#define SCTP_V6_FLOW 0x07
+#define AH_ESP_V6_FLOW 0x08
+
+/* L3-L4 network traffic flow hash options */
+#define RXH_DEV_PORT (1 << 0)
+#define RXH_L2DA (1 << 1)
+#define RXH_VLAN (1 << 2)
+#define RXH_L3_PROTO (1 << 3)
+#define RXH_IP_SRC (1 << 4)
+#define RXH_IP_DST (1 << 5)
+#define RXH_L4_BYTES_0_1 (1 << 6) /* src port in case of TCP/UDP/SCTP */
+#define RXH_L4_BYTES_2_3 (1 << 7) /* dst port in case of TCP/UDP/SCTP */
+#define RXH_DISCARD (1 << 31)
+
+
#endif /* _LINUX_ETHTOOL_H */
diff --git a/ethtool.8 b/ethtool.8
index cc6a46e..1beb387 100644
--- a/ethtool.8
+++ b/ethtool.8
@@ -40,6 +40,14 @@
.\" \(*WO - wol flags
.\"
.ds WO \fBp\fP|\fBu\fP|\fBm\fP|\fBb\fP|\fBa\fP|\fBg\fP|\fBs\fP|\fBd\fP...
+.\"
+.\" \(*FL - flow type values
+.\"
+.ds FL \fBtcp4\fP|\fBudp4\fP|\fBah4\fP|\fBsctp4\fP|\fBtcp6\fP|\fBudp6\fP|\fBah6\fP|\fBsctp6\fP
+.\"
+.\" \(*HO - hash options
+.\"
+.ds HO \fBp\fP|\fm\fP|\fBv\fP|\fBt\fP|\fBs\fP|\fBd\fP|\fBf\fP|\fBn\fP|\fBr\fP...
.TH ETHTOOL 8 "July 2007" "Ethtool version 6"
.SH NAME
ethtool \- Display or change ethernet card settings
@@ -188,6 +196,15 @@ ethtool \- Display or change ethernet card settings
.RB [ sopass \ \*(MA]
.RB [ msglvl
.IR N ]
+
+.B ethtool \-n
+.I ethX
+.RB [ rx-flow-hash \ \*(FL]
+
+.B ethtool \-N
+.I ethX
+.RB [ rx-flow-hash \ \*(FL
+.RB \ \*(HO]
.SH DESCRIPTION
.BI ethtool
is used for querying settings of an ethernet device and changing them.
@@ -413,6 +430,69 @@ bytes in ethernet MAC hex format (\*(MA).
.TP
.BI msglvl \ N
Sets the driver message level. Meanings differ per driver.
+.TP
+.B \-n \-\-show-nfc
+Retrieves the receive network flow classification configurations.
+.TP
+.BR rx-flow-hash \ \*(FL
+Retrieves the hash options for the specified network traffic type.
+.RS
+.PD 0
+.TP 3
+.BR "tcp4" " TCP over IPv4"
+.TP 3
+.BR "udp4" " UDP over IPv4"
+.TP 3
+.BR "ah4" " IPSEC AH/ESP over IPv4"
+.TP 3
+.BR "sctp4" " SCTP over IPv4"
+.TP 3
+.BR "tcp6" " TCP over IPv6"
+.TP 3
+.BR "udp6" " UDP over IPv6"
+.TP 3
+.BR "ah6" " IPSEC AH/ESP over IPv6"
+.TP 3
+.BR "sctp6" " SCTP over IPv6"
+.PD
+.RE
+.TP
+.B \-N \-\-config-nfc
+Configures the receive network flow classification.
+.TP
+.BR rx-flow-hash \ \*(FL \ \*(HO
+Configures the hash options for the specified network traffic type.
+.RS
+.PD 0
+.TP 3
+.B p
+Hash on the device port number on which the packet was received.
+.TP 3
+.B m
+Hash on the Layer 2 destination address of the rx packet.
+.TP 3
+.B v
+Hash on the VLAN tag of the rx packet.
+.TP 3
+.B t
+Hash on the Layer 3 protocol field of the rx packet.
+.TP 3
+.B s
+Hash on the IP source address of the rx packet.
+.TP 3
+.B d
+Hash on the IP destination address of the rx packet.
+.TP 3
+.B f
+Hash on bytes 0 and 1 of the Layer 4 header of the rx packet.
+.TP 3
+.B n
+Hash on bytes 2 and 3 of the Layer 4 header of the rx packet.
+.TP 3
+.B r
+Discard all packets of this flow type. When this option is set, all other options are ignored.
+.PD
+.RE
.SH BUGS
Not supported (in part or whole) on all ethernet drivers.
.SH AUTHOR
diff --git a/ethtool.c b/ethtool.c
index a668b49..46a63bd 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -69,6 +69,12 @@ static int do_scoalesce(int fd, struct ifreq *ifr);
static int do_goffload(int fd, struct ifreq *ifr);
static int do_soffload(int fd, struct ifreq *ifr);
static int do_gstats(int fd, struct ifreq *ifr);
+static int rxflow_str_to_type(const char *str);
+static int parse_rxfhashopts(char *optstr, u32 *data);
+static char *unparse_rxfhashopts(u64 opts);
+static int dump_rxfhash(int fhash, u64 val);
+static int do_srxclass(int fd, struct ifreq *ifr);
+static int do_grxclass(int fd, struct ifreq *ifr);
static enum {
MODE_HELP = -1,
@@ -90,6 +96,8 @@ static enum {
MODE_GOFFLOAD,
MODE_SOFFLOAD,
MODE_GSTATS,
+ MODE_GNFC,
+ MODE_SNFC,
} mode = MODE_GSET;
static struct option {
@@ -170,6 +178,14 @@ static struct option {
{ "-t", "--test", MODE_TEST, "Execute adapter self test",
" [ online | offline ]\n" },
{ "-S", "--statistics", MODE_GSTATS, "Show adapter statistics" },
+ { "-n", "--show-nfc", MODE_GNFC, "Show Rx network flow classification"
+ "options",
+ " [ rx-flow-hash tcp4|udp4|ah4|sctp4|"
+ "tcp6|udp6|ah6|sctp6 ]\n" },
+ { "-N", "--config-nfc", MODE_SNFC, "Configure Rx network flow "
+ "classification options",
+ " [ rx-flow-hash tcp4|udp4|ah4|sctp4|"
+ "tcp6|udp6|ah6|sctp6 p|m|v|t|s|d|f|n|r... ]\n" },
{ "-h", "--help", MODE_HELP, "Show this help" },
{}
};
@@ -266,6 +282,10 @@ static int seeprom_changed = 0;
static int seeprom_magic = 0;
static int seeprom_offset = -1;
static int seeprom_value = 0;
+static int rx_fhash_get = 0;
+static int rx_fhash_set = 0;
+static u32 rx_fhash_val = 0;
+static int rx_fhash_changed = 0;
static enum {
ONLINE=0,
OFFLINE,
@@ -394,6 +414,30 @@ static void parse_generic_cmdline(int argc, char **argp,
}
}
+static int rxflow_str_to_type(const char *str)
+{
+ int flow_type = 0;
+
+ if (!strcmp(str, "tcp4"))
+ flow_type = TCP_V4_FLOW;
+ else if (!strcmp(str, "udp4"))
+ flow_type = UDP_V4_FLOW;
+ else if (!strcmp(str, "ah4"))
+ flow_type = AH_ESP_V4_FLOW;
+ else if (!strcmp(str, "sctp4"))
+ flow_type = SCTP_V4_FLOW;
+ else if (!strcmp(str, "tcp6"))
+ flow_type = TCP_V6_FLOW;
+ else if (!strcmp(str, "udp6"))
+ flow_type = UDP_V6_FLOW;
+ else if (!strcmp(str, "ah6"))
+ flow_type = AH_ESP_V6_FLOW;
+ else if (!strcmp(str, "sctp6"))
+ flow_type = SCTP_V6_FLOW;
+
+ return flow_type;
+}
+
static void parse_cmdline(int argc, char **argp)
{
int i, k;
@@ -430,6 +474,8 @@ static void parse_cmdline(int argc, char **argp)
(mode == MODE_GOFFLOAD) ||
(mode == MODE_SOFFLOAD) ||
(mode == MODE_GSTATS) ||
+ (mode == MODE_GNFC) ||
+ (mode == MODE_SNFC) ||
(mode == MODE_PHYS_ID)) {
devname = argp[i];
break;
@@ -509,6 +555,48 @@ static void parse_cmdline(int argc, char **argp)
i = argc;
break;
}
+ if (mode == MODE_GNFC) {
+ if (!strcmp(argp[i], "rx-flow-hash")) {
+ i += 1;
+ if (i >= argc) {
+ show_usage(1);
+ break;
+ }
+ rx_fhash_get =
+ rxflow_str_to_type(argp[i]);
+ if (!rx_fhash_get)
+ show_usage(1);
+ } else
+ show_usage(1);
+ break;
+ }
+ if (mode == MODE_SNFC) {
+ if (!strcmp(argp[i], "rx-flow-hash")) {
+ i += 1;
+ if (i >= argc) {
+ show_usage(1);
+ break;
+ }
+ rx_fhash_set =
+ rxflow_str_to_type(argp[i]);
+ if (!rx_fhash_set) {
+ show_usage(1);
+ break;
+ }
+ i += 1;
+ if (i >= argc) {
+ show_usage(1);
+ break;
+ }
+ if (parse_rxfhashopts(argp[i],
+ &rx_fhash_val) < 0)
+ show_usage(1);
+ else
+ rx_fhash_changed = 1;
+ } else
+ show_usage(1);
+ break;
+ }
if (mode != MODE_SSET)
show_usage(1);
if (!strcmp(argp[i], "speed")) {
@@ -1000,6 +1088,84 @@ static int parse_sopass(char *src, unsigned char *dest)
return 0;
}
+static int parse_rxfhashopts(char *optstr, u32 *data)
+{
+ *data = 0;
+ while (*optstr) {
+ switch (*optstr) {
+ case 'p':
+ *data |= RXH_DEV_PORT;
+ break;
+ case 'm':
+ *data |= RXH_L2DA;
+ break;
+ case 'v':
+ *data |= RXH_VLAN;
+ break;
+ case 't':
+ *data |= RXH_L3_PROTO;
+ break;
+ case 's':
+ *data |= RXH_IP_SRC;
+ break;
+ case 'd':
+ *data |= RXH_IP_DST;
+ break;
+ case 'f':
+ *data |= RXH_L4_BYTES_0_1;
+ break;
+ case 'n':
+ *data |= RXH_L4_BYTES_2_3;
+ break;
+ case 'r':
+ *data |= RXH_DISCARD;
+ break;
+ default:
+ return -1;
+ }
+ optstr++;
+ }
+ return 0;
+}
+
+static char *unparse_rxfhashopts(u64 opts)
+{
+ static char buf[300];
+
+ memset(buf, 0, sizeof(buf));
+
+ if (opts) {
+ if (opts & RXH_DEV_PORT) {
+ strcat(buf, "Dev port\n");
+ }
+ if (opts & RXH_L2DA) {
+ strcat(buf, "L2DA\n");
+ }
+ if (opts & RXH_VLAN) {
+ strcat(buf, "VLAN tag\n");
+ }
+ if (opts & RXH_L3_PROTO) {
+ strcat(buf, "L3 proto\n");
+ }
+ if (opts & RXH_IP_SRC) {
+ strcat(buf, "IP SA\n");
+ }
+ if (opts & RXH_IP_DST) {
+ strcat(buf, "IP DA\n");
+ }
+ if (opts & RXH_L4_BYTES_0_1) {
+ strcat(buf, "L4 bytes 0 & 1 [TCP/UDP src port]\n");
+ }
+ if (opts & RXH_L4_BYTES_2_3) {
+ strcat(buf, "L4 bytes 2 & 3 [TCP/UDP dst port]\n");
+ }
+ } else {
+ sprintf(buf, "None");
+ }
+
+ return buf;
+}
+
static struct {
const char *name;
int (*func)(struct ethtool_drvinfo *info, struct ethtool_regs *regs);
@@ -1236,6 +1402,48 @@ static int dump_offload (int rx, int tx, int sg, int tso, int ufo, int gso)
return 0;
}
+static int dump_rxfhash(int fhash, u64 val)
+{
+ switch (fhash) {
+ case TCP_V4_FLOW:
+ fprintf(stdout, "TCP over IPV4 flows");
+ break;
+ case UDP_V4_FLOW:
+ fprintf(stdout, "UDP over IPV4 flows");
+ break;
+ case SCTP_V4_FLOW:
+ fprintf(stdout, "SCTP over IPV4 flows");
+ break;
+ case AH_ESP_V4_FLOW:
+ fprintf(stdout, "IPSEC AH over IPV4 flows");
+ break;
+ case TCP_V6_FLOW:
+ fprintf(stdout, "TCP over IPV6 flows");
+ break;
+ case UDP_V6_FLOW:
+ fprintf(stdout, "UDP over IPV6 flows");
+ break;
+ case SCTP_V6_FLOW:
+ fprintf(stdout, "SCTP over IPV6 flows");
+ break;
+ case AH_ESP_V6_FLOW:
+ fprintf(stdout, "IPSEC AH over IPV6 flows");
+ break;
+ default:
+ break;
+ }
+
+ if (val & RXH_DISCARD) {
+ fprintf(stdout, " - All matching flows discarded on RX\n");
+ return 0;
+ }
+ fprintf(stdout, " use these fields for computing Hash flow key:\n");
+
+ fprintf(stdout, "%s\n", unparse_rxfhashopts(val));
+
+ return 0;
+}
+
static int doit(void)
{
struct ifreq ifr;
@@ -1289,6 +1497,10 @@ static int doit(void)
return do_soffload(fd, &ifr);
} else if (mode == MODE_GSTATS) {
return do_gstats(fd, &ifr);
+ } else if (mode == MODE_GNFC) {
+ return do_grxclass(fd, &ifr);
+ } else if (mode == MODE_SNFC) {
+ return do_srxclass(fd, &ifr);
}
return 69;
@@ -2085,6 +2297,48 @@ static int do_gstats(int fd, struct ifreq *ifr)
return 0;
}
+
+static int do_srxclass(int fd, struct ifreq *ifr)
+{
+ int err;
+
+ if (rx_fhash_changed) {
+ struct ethtool_rxnfc nfccmd;
+
+ nfccmd.cmd = ETHTOOL_SRXFH;
+ nfccmd.flow_type = rx_fhash_set;
+ nfccmd.data = rx_fhash_val;
+
+ ifr->ifr_data = (caddr_t)&nfccmd;
+ err = ioctl(fd, SIOCETHTOOL, ifr);
+ if (err < 0)
+ perror("Cannot change RX network flow hashing options");
+
+ }
+
+ return 0;
+}
+
+static int do_grxclass(int fd, struct ifreq *ifr)
+{
+ int err;
+
+ if (rx_fhash_get) {
+ struct ethtool_rxnfc nfccmd;
+
+ nfccmd.cmd = ETHTOOL_GRXFH;
+ nfccmd.flow_type = rx_fhash_get;
+ ifr->ifr_data = (caddr_t)&nfccmd;
+ err = ioctl(fd, SIOCETHTOOL, ifr);
+ if (err < 0)
+ perror("Cannot get RX network flow hashing options");
+ else
+ dump_rxfhash(rx_fhash_get, nfccmd.data);
+ }
+
+ return 0;
+}
+
int main(int argc, char **argp, char **envp)
{
parse_cmdline(argc, argp);
--
1.5.2
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH 1/3] Add support for rx flow hash configuration in a network device - [ethtool]
2008-06-28 0:06 [PATCH 1/3] Add support for rx flow hash configuration in a network device - [ethtool] Santwona Behera
@ 2008-09-14 1:23 ` Jeff Garzik
0 siblings, 0 replies; 2+ messages in thread
From: Jeff Garzik @ 2008-09-14 1:23 UTC (permalink / raw)
To: Santwona Behera
Cc: netdev, davem, Matheos Worku, Michael Speer, gkernel-commit
applied
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2008-09-14 1:23 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-06-28 0:06 [PATCH 1/3] Add support for rx flow hash configuration in a network device - [ethtool] Santwona Behera
2008-09-14 1:23 ` Jeff Garzik
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).