[bpf-next PATCH v3 0/7] sockmap sample update

[bpf-next PATCH v3 0/7] sockmap sample update

[John Fastabend]

The sockmap sample is pretty simple at the moment. All it does is open
a few sockets attach BPF programs/sockmaps and sends a few packets.

However, for testing and debugging I wanted to have more control over
the sendmsg format and data than provided by tools like iperf3/netperf,
etc. The reason is for testing BPF programs and stream parser it is
helpful to be able submit multiple sendmsg calls with different msg
layouts. For example lots of 1B iovs or a single large MB of data, etc.

Additionally, my current test setup requires an entire orchestration
layer (cilium) to run. As well as lighttpd and http traffic generators
or for kafka testing brokers and clients. This makes it a bit more
difficult when doing performance optimizations to incrementally test
small changes and come up with performance delta's and perf numbers.

By adding a few more options and an additional few tests the sockmap
sample program can show a more complete example and do some of the
above. Because the sample program is self contained it doesn't require
additional infrastructure to run either.

This series, although still fairly crude, does provide some nice
additions. They are

  - a new sendmsg tests with a sender and recv threads
  - a new base tests so we can get metrics/data without BPF
  - multiple GBps of throughput on base and sendmsg tests
  - automatically set rlimit and common variables

That said the UI is still primitive, more features could be added,
more tests might be useful, the reporting is bare bones, etc. But,
IMO lets push this now rather than sit on it for weeks until I get
time to do the above improvements. Additional patches can address
the other limitations/issues.

v2: removed bogus file added by patch 3/7
v3: 1/7 replace goto out with returns, remove sighandler update,
    2/7 free iov in error cases
    3/7 fix bogus makefile change, bail out early on errors

Thanks Daniel and Martin for the reviews!
---

John Fastabend (7):
      bpf: refactor sockmap sample program update for arg parsing
      bpf: add sendmsg option for testing BPF programs
      bpf: sockmap sample, use fork() for send and recv
      bpf: sockmap sample, report bytes/sec
      bpf: sockmap sample add base test without any BPF for comparison
      bpf: sockmap put client sockets in blocking mode
      bpf: sockmap set rlimit


 samples/sockmap/sockmap_user.c |  383 +++++++++++++++++++++++++++++++++++-----
 1 file changed, 331 insertions(+), 52 deletions(-)

[bpf-next PATCH v3 1/7] bpf: refactor sockmap sample program update for arg parsing

[John Fastabend]

sockmap sample program takes arguments from cmd line but it reads them
in using offsets into the array. Because we want to add more arguments
in the future lets do proper argument handling.

Also refactor code to pull apart sock init and ping/pong test. This
allows us to add new tests in the future.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |  164 ++++++++++++++++++++++++++++------------
 1 file changed, 113 insertions(+), 51 deletions(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index 7cc9d22..17400d4 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -35,6 +35,8 @@
 #include <assert.h>
 #include <libgen.h>
 
+#include <getopt.h>
+
 #include "../bpf/bpf_load.h"
 #include "../bpf/bpf_util.h"
 #include "../bpf/libbpf.h"
@@ -46,15 +48,39 @@
 #define S1_PORT 10000
 #define S2_PORT 10001
 
-static int sockmap_test_sockets(int rate, int dot)
+/* global sockets */
+int s1, s2, c1, c2, p1, p2;
+
+static const struct option long_options[] = {
+	{"help",	no_argument,		NULL, 'h' },
+	{"cgroup",	required_argument,	NULL, 'c' },
+	{"rate",	required_argument,	NULL, 'r' },
+	{"verbose",	no_argument,		NULL, 'v' },
+	{0, 0, NULL, 0 }
+};
+
+static void usage(char *argv[])
 {
-	int i, sc, err, max_fd, one = 1;
-	int s1, s2, c1, c2, p1, p2;
+	int i;
+
+	printf(" Usage: %s --cgroup <cgroup_path>\n", argv[0]);
+	printf(" options:\n");
+	for (i = 0; long_options[i].name != 0; i++) {
+		printf(" --%-12s", long_options[i].name);
+		if (long_options[i].flag != NULL)
+			printf(" flag (internal value:%d)\n",
+				*long_options[i].flag);
+		else
+			printf(" -%c\n", long_options[i].val);
+	}
+	printf("\n");
+}
+
+static int sockmap_init_sockets(void)
+{
+	int i, err, one = 1;
 	struct sockaddr_in addr;
-	struct timeval timeout;
-	char buf[1024] = {0};
 	int *fds[4] = {&s1, &s2, &c1, &c2};
-	fd_set w;
 
 	s1 = s2 = p1 = p2 = c1 = c2 = 0;
 
@@ -63,8 +89,7 @@ static int sockmap_test_sockets(int rate, int dot)
 		*fds[i] = socket(AF_INET, SOCK_STREAM, 0);
 		if (*fds[i] < 0) {
 			perror("socket s1 failed()");
-			err = *fds[i];
-			goto out;
+			return errno;
 		}
 	}
 
@@ -74,7 +99,7 @@ static int sockmap_test_sockets(int rate, int dot)
 				 (char *)&one, sizeof(one));
 		if (err) {
 			perror("setsockopt failed()");
-			goto out;
+			return errno;
 		}
 	}
 
@@ -83,7 +108,7 @@ static int sockmap_test_sockets(int rate, int dot)
 		err = ioctl(*fds[i], FIONBIO, (char *)&one);
 		if (err < 0) {
 			perror("ioctl s1 failed()");
-			goto out;
+			return errno;
 		}
 	}
 
@@ -96,14 +121,14 @@ static int sockmap_test_sockets(int rate, int dot)
 	err = bind(s1, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0) {
 		perror("bind s1 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = bind(s2, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0) {
 		perror("bind s2 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	/* Listen server sockets */
@@ -111,14 +136,14 @@ static int sockmap_test_sockets(int rate, int dot)
 	err = listen(s1, 32);
 	if (err < 0) {
 		perror("listen s1 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = listen(s2, 32);
 	if (err < 0) {
 		perror("listen s1 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	/* Initiate Connect */
@@ -126,46 +151,56 @@ static int sockmap_test_sockets(int rate, int dot)
 	err = connect(c1, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0 && errno != EINPROGRESS) {
 		perror("connect c1 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = connect(c2, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0 && errno != EINPROGRESS) {
 		perror("connect c2 failed()\n");
-		goto out;
+		return errno;
+	} else if (err < 0) {
+		err = 0;
 	}
 
 	/* Accept Connecrtions */
 	p1 = accept(s1, NULL, NULL);
 	if (p1 < 0) {
 		perror("accept s1 failed()\n");
-		goto out;
+		return errno;
 	}
 
 	p2 = accept(s2, NULL, NULL);
 	if (p2 < 0) {
 		perror("accept s1 failed()\n");
-		goto out;
+		return errno;
 	}
 
-	max_fd = p2;
-	timeout.tv_sec = 10;
-	timeout.tv_usec = 0;
-
 	printf("connected sockets: c1 <-> p1, c2 <-> p2\n");
 	printf("cgroups binding: c1(%i) <-> s1(%i) - - - c2(%i) <-> s2(%i)\n",
 		c1, s1, c2, s2);
+	return 0;
+}
+
+static int forever_ping_pong(int rate, int verbose)
+{
+	struct timeval timeout;
+	char buf[1024] = {0};
+	int sc;
+
+	timeout.tv_sec = 10;
+	timeout.tv_usec = 0;
 
 	/* Ping/Pong data from client to server */
 	sc = send(c1, buf, sizeof(buf), 0);
 	if (sc < 0) {
 		perror("send failed()\n");
-		goto out;
+		return sc;
 	}
 
 	do {
-		int s, rc, i;
+		int s, rc, i, max_fd = p2;
+		fd_set w;
 
 		/* FD sets */
 		FD_ZERO(&w);
@@ -193,7 +228,7 @@ static int sockmap_test_sockets(int rate, int dot)
 			if (rc < 0) {
 				if (errno != EWOULDBLOCK) {
 					perror("recv failed()\n");
-					break;
+					return rc;
 				}
 			}
 
@@ -205,35 +240,61 @@ static int sockmap_test_sockets(int rate, int dot)
 			sc = send(i, buf, rc, 0);
 			if (sc < 0) {
 				perror("send failed()\n");
-				break;
+				return sc;
 			}
 		}
-		sleep(rate);
-		if (dot) {
+
+		if (rate)
+			sleep(rate);
+
+		if (verbose) {
 			printf(".");
 			fflush(stdout);
 
 		}
 	} while (running);
 
-out:
-	close(s1);
-	close(s2);
-	close(p1);
-	close(p2);
-	close(c1);
-	close(c2);
-	return err;
+	return 0;
 }
 
 int main(int argc, char **argv)
 {
-	int rate = 1, dot = 1;
+	int rate = 1, verbose = 0;
+	int opt, longindex, err, cg_fd = 0;
 	char filename[256];
-	int err, cg_fd;
-	char *cg_path;
 
-	cg_path = argv[argc - 1];
+	while ((opt = getopt_long(argc, argv, "hvc:r:",
+				  long_options, &longindex)) != -1) {
+		switch (opt) {
+		/* Cgroup configuration */
+		case 'c':
+			cg_fd = open(optarg, O_DIRECTORY, O_RDONLY);
+			if (cg_fd < 0) {
+				fprintf(stderr,
+					"ERROR: (%i) open cg path failed: %s\n",
+					cg_fd, optarg);
+				return cg_fd;
+			}
+			break;
+		case 'r':
+			rate = atoi(optarg);
+			break;
+		case 'v':
+			verbose = 1;
+			break;
+		case 'h':
+		default:
+			usage(argv);
+			return -1;
+		}
+	}
+
+	if (!cg_fd) {
+		fprintf(stderr, "%s requires cgroup option: --cgroup <path>\n",
+			argv[0]);
+		return -1;
+	}
+
 	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
 
 	running = 1;
@@ -247,14 +308,6 @@ int main(int argc, char **argv)
 		return 1;
 	}
 
-	/* Cgroup configuration */
-	cg_fd = open(cg_path, O_DIRECTORY, O_RDONLY);
-	if (cg_fd < 0) {
-		fprintf(stderr, "ERROR: (%i) open cg path failed: %s\n",
-			cg_fd, cg_path);
-		return cg_fd;
-	}
-
 	/* Attach programs to sockmap */
 	err = bpf_prog_attach(prog_fd[0], map_fd[0],
 				BPF_SK_SKB_STREAM_PARSER, 0);
@@ -280,12 +333,21 @@ int main(int argc, char **argv)
 		return err;
 	}
 
-	err = sockmap_test_sockets(rate, dot);
+	err = sockmap_init_sockets();
 	if (err) {
 		fprintf(stderr, "ERROR: test socket failed: %d\n", err);
-		return err;
+		goto out;
 	}
-	return 0;
+
+	err = forever_ping_pong(rate, verbose);
+out:
+	close(s1);
+	close(s2);
+	close(p1);
+	close(p2);
+	close(c1);
+	close(c2);
+	return err;
 }
 
 void running_handler(int a)

Re: [bpf-next PATCH v3 1/7] bpf: refactor sockmap sample program update for arg parsing

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:06:17PM -0800, John Fastabend wrote:
> sockmap sample program takes arguments from cmd line but it reads them
> in using offsets into the array. Because we want to add more arguments
> in the future lets do proper argument handling.
> 
> Also refactor code to pull apart sock init and ping/pong test. This
> allows us to add new tests in the future.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
One nit below.

Acked-by: Martin KaFai Lau <kafai@fb.com>

> ---
[ ... ]
> @@ -280,12 +333,21 @@ int main(int argc, char **argv)
>  		return err;
>  	}
>  
> -	err = sockmap_test_sockets(rate, dot);
> +	err = sockmap_init_sockets();
>  	if (err) {
>  		fprintf(stderr, "ERROR: test socket failed: %d\n", err);
> -		return err;
> +		goto out;
>  	}
> -	return 0;
> +
> +	err = forever_ping_pong(rate, verbose);
> +out:
> +	close(s1);
> +	close(s2);
> +	close(p1);
> +	close(p2);
> +	close(c1);
> +	close(c2);
close(cg_fd);

> +	return err;
>  }
>  
>  void running_handler(int a)
> 

[bpf-next PATCH v3 2/7] bpf: add sendmsg option for testing BPF programs

[John Fastabend]

When testing BPF programs using sockmap I often want to have more
control over how sendmsg is exercised. This becomes even more useful
as new sockmap program types are added.

This adds a test type option to select type of test to run. Currently,
only "ping" and "sendmsg" are supported, but more can be added as
needed.

The new help argument gives the following,

 Usage: ./sockmap --cgroup <cgroup_path>
 options:
 --help         -h
 --cgroup       -c
 --rate         -r
 --verbose      -v
 --iov_count    -i
 --length       -l
 --test         -t

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |  147 +++++++++++++++++++++++++++++++++++++++-
 1 file changed, 144 insertions(+), 3 deletions(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index 17400d4..8ec7dbf 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -56,6 +56,9 @@
 	{"cgroup",	required_argument,	NULL, 'c' },
 	{"rate",	required_argument,	NULL, 'r' },
 	{"verbose",	no_argument,		NULL, 'v' },
+	{"iov_count",	required_argument,	NULL, 'i' },
+	{"length",	required_argument,	NULL, 'l' },
+	{"test",	required_argument,	NULL, 't' },
 	{0, 0, NULL, 0 }
 };
 
@@ -182,6 +185,117 @@ static int sockmap_init_sockets(void)
 	return 0;
 }
 
+struct msg_stats {
+	size_t bytes_sent;
+	size_t bytes_recvd;
+};
+
+static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
+		    struct msg_stats *s, bool tx)
+{
+	struct msghdr msg = {0};
+	struct iovec *iov;
+	int i, flags = 0;
+
+	iov = calloc(iov_count, sizeof(struct iovec));
+	if (!iov)
+		return -ENOMEM;
+
+	for (i = 0; i < iov_count; i++) {
+		char *d = calloc(iov_length, sizeof(char));
+
+		if (!d) {
+			fprintf(stderr, "iov_count %i/%i OOM\n", i, iov_count);
+			free(iov);
+			return -ENOMEM;
+		}
+		iov[i].iov_base = d;
+		iov[i].iov_len = iov_length;
+	}
+
+	msg.msg_iov = iov;
+	msg.msg_iovlen = iov_count;
+
+	if (tx) {
+		for (i = 0; i < cnt; i++) {
+			int sent = sendmsg(fd, &msg, flags);
+
+			if (sent < 0) {
+				perror("send loop error:");
+				free(iov);
+				return sent;
+			}
+			s->bytes_sent += sent;
+		}
+	} else {
+		int slct, recv, max_fd = fd;
+		struct timeval timeout;
+		float total_bytes;
+		fd_set w;
+
+		total_bytes = (float)iov_count * (float)iov_length * (float)cnt;
+		while (s->bytes_recvd < total_bytes) {
+			timeout.tv_sec = 1;
+			timeout.tv_usec = 0;
+
+			/* FD sets */
+			FD_ZERO(&w);
+			FD_SET(fd, &w);
+
+			slct = select(max_fd + 1, &w, NULL, NULL, &timeout);
+			if (slct == -1) {
+				perror("select()");
+				goto out_errno;
+			} else if (!slct) {
+				fprintf(stderr, "unexpected timeout\n");
+				goto out_errno;
+			}
+
+			recv = recvmsg(fd, &msg, flags);
+			if (recv < 0) {
+				if (errno != EWOULDBLOCK) {
+					perror("recv failed()\n");
+					goto out_errno;
+				}
+			}
+
+			s->bytes_recvd += recv;
+		}
+	}
+
+	for (i = 0; i < iov_count; i++)
+		free(iov[i].iov_base);
+	free(iov);
+	return 0;
+out_errno:
+	free(iov);
+	return errno;
+}
+
+static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
+{
+	struct msg_stats s = {0};
+	int err;
+
+	err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
+	if (err) {
+		fprintf(stderr,
+			"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
+			iov_count, iov_buf, cnt, err);
+		return err;
+	}
+
+	msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
+	if (err)
+		fprintf(stderr,
+			"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
+			iov_count, iov_buf, cnt, err);
+
+	fprintf(stdout, "sendmsg: TX_bytes %zu RX_bytes %zu\n",
+		s.bytes_sent, s.bytes_recvd);
+	return err;
+}
+
 static int forever_ping_pong(int rate, int verbose)
 {
 	struct timeval timeout;
@@ -257,13 +371,19 @@ static int forever_ping_pong(int rate, int verbose)
 	return 0;
 }
 
+enum {
+	PING_PONG,
+	SENDMSG,
+};
+
 int main(int argc, char **argv)
 {
-	int rate = 1, verbose = 0;
+	int iov_count = 1, length = 1024, rate = 1, verbose = 0;
 	int opt, longindex, err, cg_fd = 0;
+	int test = PING_PONG;
 	char filename[256];
 
-	while ((opt = getopt_long(argc, argv, "hvc:r:",
+	while ((opt = getopt_long(argc, argv, "hvc:r:i:l:t:",
 				  long_options, &longindex)) != -1) {
 		switch (opt) {
 		/* Cgroup configuration */
@@ -282,6 +402,22 @@ int main(int argc, char **argv)
 		case 'v':
 			verbose = 1;
 			break;
+		case 'i':
+			iov_count = atoi(optarg);
+			break;
+		case 'l':
+			length = atoi(optarg);
+			break;
+		case 't':
+			if (memcmp(optarg, "ping", 4) == 0) {
+				test = PING_PONG;
+			} else if (memcmp(optarg, "sendmsg", 7) == 0) {
+				test = SENDMSG;
+			} else {
+				usage(argv);
+				return -1;
+			}
+			break;
 		case 'h':
 		default:
 			usage(argv);
@@ -339,7 +475,12 @@ int main(int argc, char **argv)
 		goto out;
 	}
 
-	err = forever_ping_pong(rate, verbose);
+	if (test == PING_PONG)
+		err = forever_ping_pong(rate, verbose);
+	else if (test == SENDMSG)
+		err = sendmsg_test(iov_count, length, rate, verbose);
+	else
+		fprintf(stderr, "unknown test\n");
 out:
 	close(s1);
 	close(s2);

Re: [bpf-next PATCH v3 2/7] bpf: add sendmsg option for testing BPF programs

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:06:34PM -0800, John Fastabend wrote:
> When testing BPF programs using sockmap I often want to have more
> control over how sendmsg is exercised. This becomes even more useful
> as new sockmap program types are added.
> 
> This adds a test type option to select type of test to run. Currently,
> only "ping" and "sendmsg" are supported, but more can be added as
> needed.
> 
> The new help argument gives the following,
> 
>  Usage: ./sockmap --cgroup <cgroup_path>
>  options:
>  --help         -h
>  --cgroup       -c
>  --rate         -r
>  --verbose      -v
>  --iov_count    -i
>  --length       -l
>  --test         -t
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
> ---
>  samples/sockmap/sockmap_user.c |  147 +++++++++++++++++++++++++++++++++++++++-
>  1 file changed, 144 insertions(+), 3 deletions(-)
> 
> diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
> index 17400d4..8ec7dbf 100644
> --- a/samples/sockmap/sockmap_user.c
> +++ b/samples/sockmap/sockmap_user.c
> @@ -56,6 +56,9 @@
>  	{"cgroup",	required_argument,	NULL, 'c' },
>  	{"rate",	required_argument,	NULL, 'r' },
>  	{"verbose",	no_argument,		NULL, 'v' },
> +	{"iov_count",	required_argument,	NULL, 'i' },
> +	{"length",	required_argument,	NULL, 'l' },
> +	{"test",	required_argument,	NULL, 't' },
>  	{0, 0, NULL, 0 }
>  };
>  
> @@ -182,6 +185,117 @@ static int sockmap_init_sockets(void)
>  	return 0;
>  }
>  
> +struct msg_stats {
> +	size_t bytes_sent;
> +	size_t bytes_recvd;
> +};
> +
> +static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
> +		    struct msg_stats *s, bool tx)
> +{
> +	struct msghdr msg = {0};
> +	struct iovec *iov;
> +	int i, flags = 0;
> +
> +	iov = calloc(iov_count, sizeof(struct iovec));
> +	if (!iov)
> +		return -ENOMEM;
I think errno has already been set to ENOMEM (instead of
-ENOMEM), so may directly use it instead.

> +
> +	for (i = 0; i < iov_count; i++) {
> +		char *d = calloc(iov_length, sizeof(char));
> +
> +		if (!d) {
> +			fprintf(stderr, "iov_count %i/%i OOM\n", i, iov_count);
> +			free(iov);
> +			return -ENOMEM;
The new "out_errno" label below should include freeing
all iov[i].iov_base.

Also, instead of a return here, I think you meant a "goto out_errno;"
such that the earlier "calloc(iov_length, sizeof(char));"
can also be freed during error.

Same for the following error return/goto cases

> +		}
> +		iov[i].iov_base = d;
> +		iov[i].iov_len = iov_length;
> +	}
> +
> +	msg.msg_iov = iov;
> +	msg.msg_iovlen = iov_count;
> +
> +	if (tx) {
> +		for (i = 0; i < cnt; i++) {
> +			int sent = sendmsg(fd, &msg, flags);
> +
> +			if (sent < 0) {
> +				perror("send loop error:");
> +				free(iov);
> +				return sent;
> +			}
> +			s->bytes_sent += sent;
> +		}
> +	} else {
> +		int slct, recv, max_fd = fd;
> +		struct timeval timeout;
> +		float total_bytes;
> +		fd_set w;
> +
> +		total_bytes = (float)iov_count * (float)iov_length * (float)cnt;
> +		while (s->bytes_recvd < total_bytes) {
> +			timeout.tv_sec = 1;
> +			timeout.tv_usec = 0;
> +
> +			/* FD sets */
> +			FD_ZERO(&w);
> +			FD_SET(fd, &w);
> +
> +			slct = select(max_fd + 1, &w, NULL, NULL, &timeout);
> +			if (slct == -1) {
> +				perror("select()");
> +				goto out_errno;
> +			} else if (!slct) {
> +				fprintf(stderr, "unexpected timeout\n");
> +				goto out_errno;
Just in case, I think errno == 0 here here.

> +			}
> +
> +			recv = recvmsg(fd, &msg, flags);
> +			if (recv < 0) {
> +				if (errno != EWOULDBLOCK) {
> +					perror("recv failed()\n");
> +					goto out_errno;
> +				}
> +			}
> +
> +			s->bytes_recvd += recv;
> +		}
> +	}
> +
> +	for (i = 0; i < iov_count; i++)
> +		free(iov[i].iov_base);
> +	free(iov);
> +	return 0;
> +out_errno:
> +	free(iov);
> +	return errno;
> +}
> +
> +static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
> +{
> +	struct msg_stats s = {0};
> +	int err;
> +
> +	err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
> +	if (err) {
> +		fprintf(stderr,
> +			"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
> +			iov_count, iov_buf, cnt, err);
> +		return err;
> +	}
> +
> +	msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
err = msg_loop(...);

> +	if (err)
> +		fprintf(stderr,
> +			"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
> +			iov_count, iov_buf, cnt, err);
> +
> +	fprintf(stdout, "sendmsg: TX_bytes %zu RX_bytes %zu\n",
> +		s.bytes_sent, s.bytes_recvd);
> +	return err;
> +}
> +
>  static int forever_ping_pong(int rate, int verbose)
>  {
>  	struct timeval timeout;
> @@ -257,13 +371,19 @@ static int forever_ping_pong(int rate, int verbose)
>  	return 0;
>  }
>  
> +enum {
> +	PING_PONG,
> +	SENDMSG,
> +};
> +
>  int main(int argc, char **argv)
>  {
> -	int rate = 1, verbose = 0;
> +	int iov_count = 1, length = 1024, rate = 1, verbose = 0;
>  	int opt, longindex, err, cg_fd = 0;
> +	int test = PING_PONG;
>  	char filename[256];
>  
> -	while ((opt = getopt_long(argc, argv, "hvc:r:",
> +	while ((opt = getopt_long(argc, argv, "hvc:r:i:l:t:",
>  				  long_options, &longindex)) != -1) {
>  		switch (opt) {
>  		/* Cgroup configuration */
> @@ -282,6 +402,22 @@ int main(int argc, char **argv)
>  		case 'v':
>  			verbose = 1;
>  			break;
> +		case 'i':
> +			iov_count = atoi(optarg);
> +			break;
> +		case 'l':
> +			length = atoi(optarg);
> +			break;
> +		case 't':
> +			if (memcmp(optarg, "ping", 4) == 0) {
strcmp() may feel safer here just in case optarg could be shorter.

> +				test = PING_PONG;
> +			} else if (memcmp(optarg, "sendmsg", 7) == 0) {
Same here.

> +				test = SENDMSG;
> +			} else {
> +				usage(argv);
> +				return -1;
> +			}
> +			break;
>  		case 'h':
>  		default:
>  			usage(argv);
> @@ -339,7 +475,12 @@ int main(int argc, char **argv)
>  		goto out;
>  	}
>  
> -	err = forever_ping_pong(rate, verbose);
> +	if (test == PING_PONG)
> +		err = forever_ping_pong(rate, verbose);
> +	else if (test == SENDMSG)
> +		err = sendmsg_test(iov_count, length, rate, verbose);
> +	else
> +		fprintf(stderr, "unknown test\n");
>  out:
>  	close(s1);
>  	close(s2);
> 

[bpf-next PATCH v3 3/7] bpf: sockmap sample, use fork() for send and recv

[John Fastabend]

Currently for SENDMSG tests first send completes then recv runs. This
does not work well for large data sizes and/or many iterations. So
fork the recv and send handler so that we run both send and recv. In
the future we can add a parameter to do more than a single fork of
tx/rx.

With this we can get many GBps of data which helps exercise the
sockmap code.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |   58 +++++++++++++++++++++++++++++-----------
 1 file changed, 42 insertions(+), 16 deletions(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index 8ec7dbf..bbe9587 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -23,6 +23,7 @@
 #include <stdbool.h>
 #include <signal.h>
 #include <fcntl.h>
+#include <sys/wait.h>
 
 #include <sys/time.h>
 #include <sys/types.h>
@@ -195,7 +196,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 {
 	struct msghdr msg = {0};
 	struct iovec *iov;
-	int i, flags = 0;
+	int i, flags = MSG_NOSIGNAL;
 
 	iov = calloc(iov_count, sizeof(struct iovec));
 	if (!iov)
@@ -274,25 +275,50 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 
 static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
 {
+	int txpid, rxpid, err = 0;
 	struct msg_stats s = {0};
-	int err;
-
-	err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
-	if (err) {
-		fprintf(stderr,
-			"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
-			iov_count, iov_buf, cnt, err);
-		return err;
+	int status;
+
+	errno = 0;
+
+	rxpid = fork();
+	if (rxpid == 0) {
+		err = msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
+		if (err)
+			fprintf(stderr,
+				"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
+				iov_count, iov_buf, cnt, err);
+		fprintf(stdout, "rx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
+			s.bytes_sent, s.bytes_recvd);
+		shutdown(p2, SHUT_RDWR);
+		shutdown(p1, SHUT_RDWR);
+		exit(1);
+	} else if (rxpid == -1) {
+		perror("msg_loop_rx: ");
+		return errno;
 	}
 
-	msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
-	if (err)
-		fprintf(stderr,
-			"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
-			iov_count, iov_buf, cnt, err);
+	txpid = fork();
+	if (txpid == 0) {
+		err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
+		if (err)
+			fprintf(stderr,
+				"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
+				iov_count, iov_buf, cnt, err);
+		fprintf(stdout, "tx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
+			s.bytes_sent, s.bytes_recvd);
+		shutdown(c1, SHUT_RDWR);
+		exit(1);
+	} else if (txpid == -1) {
+		perror("msg_loop_tx: ");
+		return errno;
+	}
 
-	fprintf(stdout, "sendmsg: TX_bytes %zu RX_bytes %zu\n",
-		s.bytes_sent, s.bytes_recvd);
+	assert(waitpid(rxpid, &status, 0) == rxpid);
+	if (!txpid)
+		goto out;
+	assert(waitpid(txpid, &status, 0) == txpid);
+out:
 	return err;
 }
 

Re: [bpf-next PATCH v3 3/7] bpf: sockmap sample, use fork() for send and recv

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:06:54PM -0800, John Fastabend wrote:
> Currently for SENDMSG tests first send completes then recv runs. This
> does not work well for large data sizes and/or many iterations. So
> fork the recv and send handler so that we run both send and recv. In
> the future we can add a parameter to do more than a single fork of
> tx/rx.
> 
> With this we can get many GBps of data which helps exercise the
> sockmap code.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Some nits.

Acked-by: Martin KaFai Lau <kafai@fb.com>

> ---
[ ... ]
> @@ -274,25 +275,50 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
>  
>  static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
>  {
> +	int txpid, rxpid, err = 0;
>  	struct msg_stats s = {0};
> -	int err;
> -
> -	err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
> -	if (err) {
> -		fprintf(stderr,
> -			"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
> -			iov_count, iov_buf, cnt, err);
> -		return err;
> +	int status;
> +
> +	errno = 0;
> +
> +	rxpid = fork();
> +	if (rxpid == 0) {
> +		err = msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
> +		if (err)
> +			fprintf(stderr,
> +				"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
> +				iov_count, iov_buf, cnt, err);
> +		fprintf(stdout, "rx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
> +			s.bytes_sent, s.bytes_recvd);
> +		shutdown(p2, SHUT_RDWR);
> +		shutdown(p1, SHUT_RDWR);
> +		exit(1);
> +	} else if (rxpid == -1) {
> +		perror("msg_loop_rx: ");
> +		return errno;
>  	}
>  
> -	msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
> -	if (err)
> -		fprintf(stderr,
> -			"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
> -			iov_count, iov_buf, cnt, err);
> +	txpid = fork();
> +	if (txpid == 0) {
> +		err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
> +		if (err)
> +			fprintf(stderr,
> +				"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
> +				iov_count, iov_buf, cnt, err);
> +		fprintf(stdout, "tx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
> +			s.bytes_sent, s.bytes_recvd);
> +		shutdown(c1, SHUT_RDWR);
> +		exit(1);
> +	} else if (txpid == -1) {
> +		perror("msg_loop_tx: ");
> +		return errno;
> +	}
>  
> -	fprintf(stdout, "sendmsg: TX_bytes %zu RX_bytes %zu\n",
> -		s.bytes_sent, s.bytes_recvd);
> +	assert(waitpid(rxpid, &status, 0) == rxpid);
> +	if (!txpid)
This case won't be hit?

> +		goto out;
> +	assert(waitpid(txpid, &status, 0) == txpid);
> +out:
>  	return err;
>  }
>  
> 

Re: [bpf-next PATCH v3 3/7] bpf: sockmap sample, use fork() for send and recv

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:06:54PM -0800, John Fastabend wrote:
> Currently for SENDMSG tests first send completes then recv runs. This
> does not work well for large data sizes and/or many iterations. So
> fork the recv and send handler so that we run both send and recv. In
> the future we can add a parameter to do more than a single fork of
> tx/rx.
> 
> With this we can get many GBps of data which helps exercise the
> sockmap code.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
One nit.

Acked-by: Martin KaFai Lau <kafai@fb.com>

> ---
[ ... ]
> @@ -274,25 +275,50 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
>  
>  static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
>  {
> +	int txpid, rxpid, err = 0;
>  	struct msg_stats s = {0};
> -	int err;
> -
> -	err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
> -	if (err) {
> -		fprintf(stderr,
> -			"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
> -			iov_count, iov_buf, cnt, err);
> -		return err;
> +	int status;
> +
> +	errno = 0;
> +
> +	rxpid = fork();
> +	if (rxpid == 0) {
> +		err = msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
> +		if (err)
> +			fprintf(stderr,
> +				"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
> +				iov_count, iov_buf, cnt, err);
> +		fprintf(stdout, "rx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
> +			s.bytes_sent, s.bytes_recvd);
> +		shutdown(p2, SHUT_RDWR);
> +		shutdown(p1, SHUT_RDWR);
> +		exit(1);
> +	} else if (rxpid == -1) {
> +		perror("msg_loop_rx: ");
> +		return errno;
>  	}
>  
> -	msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
> -	if (err)
> -		fprintf(stderr,
> -			"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
> -			iov_count, iov_buf, cnt, err);
> +	txpid = fork();
> +	if (txpid == 0) {
> +		err = msg_loop(c1, iov_count, iov_buf, cnt, &s, true);
> +		if (err)
> +			fprintf(stderr,
> +				"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
> +				iov_count, iov_buf, cnt, err);
> +		fprintf(stdout, "tx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
> +			s.bytes_sent, s.bytes_recvd);
> +		shutdown(c1, SHUT_RDWR);
> +		exit(1);
> +	} else if (txpid == -1) {
> +		perror("msg_loop_tx: ");
> +		return errno;
> +	}
>  
> -	fprintf(stdout, "sendmsg: TX_bytes %zu RX_bytes %zu\n",
> -		s.bytes_sent, s.bytes_recvd);
> +	assert(waitpid(rxpid, &status, 0) == rxpid);
> +	if (!txpid)
This case won't be hit?

> +		goto out;
> +	assert(waitpid(txpid, &status, 0) == txpid);
> +out:
>  	return err;
>  }
>  
> 

[bpf-next PATCH v3 4/7] bpf: sockmap sample, report bytes/sec

[John Fastabend]

Report bytes/sec sent as well as total bytes. Useful to get rough
idea how different configurations and usage patterns perform with
sockmap.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |   37 ++++++++++++++++++++++++++++++++-----
 1 file changed, 32 insertions(+), 5 deletions(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index bbe9587..442fc00 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -24,6 +24,7 @@
 #include <signal.h>
 #include <fcntl.h>
 #include <sys/wait.h>
+#include <time.h>
 
 #include <sys/time.h>
 #include <sys/types.h>
@@ -189,14 +190,16 @@ static int sockmap_init_sockets(void)
 struct msg_stats {
 	size_t bytes_sent;
 	size_t bytes_recvd;
+	struct timespec start;
+	struct timespec end;
 };
 
 static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 		    struct msg_stats *s, bool tx)
 {
 	struct msghdr msg = {0};
+	int err, i, flags = MSG_NOSIGNAL;
 	struct iovec *iov;
-	int i, flags = MSG_NOSIGNAL;
 
 	iov = calloc(iov_count, sizeof(struct iovec));
 	if (!iov)
@@ -218,6 +221,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 	msg.msg_iovlen = iov_count;
 
 	if (tx) {
+		clock_gettime(CLOCK_MONOTONIC, &s->start);
 		for (i = 0; i < cnt; i++) {
 			int sent = sendmsg(fd, &msg, flags);
 
@@ -228,6 +232,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 			}
 			s->bytes_sent += sent;
 		}
+		clock_gettime(CLOCK_MONOTONIC, &s->end);
 	} else {
 		int slct, recv, max_fd = fd;
 		struct timeval timeout;
@@ -235,6 +240,9 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 		fd_set w;
 
 		total_bytes = (float)iov_count * (float)iov_length * (float)cnt;
+		err = clock_gettime(CLOCK_MONOTONIC, &s->start);
+		if (err < 0)
+			perror("recv start time: ");
 		while (s->bytes_recvd < total_bytes) {
 			timeout.tv_sec = 1;
 			timeout.tv_usec = 0;
@@ -246,15 +254,18 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 			slct = select(max_fd + 1, &w, NULL, NULL, &timeout);
 			if (slct == -1) {
 				perror("select()");
+				clock_gettime(CLOCK_MONOTONIC, &s->end);
 				goto out_errno;
 			} else if (!slct) {
 				fprintf(stderr, "unexpected timeout\n");
+				clock_gettime(CLOCK_MONOTONIC, &s->end);
 				goto out_errno;
 			}
 
 			recv = recvmsg(fd, &msg, flags);
 			if (recv < 0) {
 				if (errno != EWOULDBLOCK) {
+					clock_gettime(CLOCK_MONOTONIC, &s->end);
 					perror("recv failed()\n");
 					goto out_errno;
 				}
@@ -262,6 +273,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 
 			s->bytes_recvd += recv;
 		}
+		clock_gettime(CLOCK_MONOTONIC, &s->end);
 	}
 
 	for (i = 0; i < iov_count; i++)
@@ -273,11 +285,14 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 	return errno;
 }
 
+static float giga = 1000000000;
+
 static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
 {
 	int txpid, rxpid, err = 0;
 	struct msg_stats s = {0};
 	int status;
+	float sent_Bps = 0, recvd_Bps = 0;
 
 	errno = 0;
 
@@ -288,10 +303,16 @@ static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
 			fprintf(stderr,
 				"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
 				iov_count, iov_buf, cnt, err);
-		fprintf(stdout, "rx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
-			s.bytes_sent, s.bytes_recvd);
 		shutdown(p2, SHUT_RDWR);
 		shutdown(p1, SHUT_RDWR);
+		if (s.end.tv_sec - s.start.tv_sec) {
+			sent_Bps = s.bytes_sent / (s.end.tv_sec - s.start.tv_sec);
+			recvd_Bps = s.bytes_recvd / (s.end.tv_sec - s.start.tv_sec);
+		}
+		fprintf(stdout,
+			"rx_sendmsg: TX: %zuB %fB/s %fGB/s RX: %zuB %fB/s %fGB/s\n",
+			s.bytes_sent, sent_Bps, sent_Bps/giga,
+			s.bytes_recvd, recvd_Bps, recvd_Bps/giga);
 		exit(1);
 	} else if (rxpid == -1) {
 		perror("msg_loop_rx: ");
@@ -305,9 +326,15 @@ static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
 			fprintf(stderr,
 				"msg_loop_tx: iov_count %i iov_buf %i cnt %i err %i\n",
 				iov_count, iov_buf, cnt, err);
-		fprintf(stdout, "tx_sendmsg: TX_bytes %zu RX_bytes %zu\n",
-			s.bytes_sent, s.bytes_recvd);
 		shutdown(c1, SHUT_RDWR);
+		if (s.end.tv_sec - s.start.tv_sec) {
+			sent_Bps = s.bytes_sent / (s.end.tv_sec - s.start.tv_sec);
+			recvd_Bps = s.bytes_recvd / (s.end.tv_sec - s.start.tv_sec);
+		}
+		fprintf(stdout,
+			"tx_sendmsg: TX: %zuB %fB/s %f GB/s RX: %zuB %fB/s %fGB/s\n",
+			s.bytes_sent, sent_Bps, sent_Bps/giga,
+			s.bytes_recvd, recvd_Bps, recvd_Bps/giga);
 		exit(1);
 	} else if (txpid == -1) {
 		perror("msg_loop_tx: ");

Re: [bpf-next PATCH v3 4/7] bpf: sockmap sample, report bytes/sec

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:07:10PM -0800, John Fastabend wrote:
> Report bytes/sec sent as well as total bytes. Useful to get rough
> idea how different configurations and usage patterns perform with
> sockmap.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>

[bpf-next PATCH v3 5/7] bpf: sockmap sample add base test without any BPF for comparison

[John Fastabend]

Add a base test that does not use BPF hooks to test baseline case.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |   26 +++++++++++++++++++++-----
 1 file changed, 21 insertions(+), 5 deletions(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index 442fc00..c3295a7 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -287,18 +287,24 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 
 static float giga = 1000000000;
 
-static int sendmsg_test(int iov_count, int iov_buf, int cnt, int verbose)
+static int sendmsg_test(int iov_count, int iov_buf, int cnt,
+			int verbose, bool base)
 {
-	int txpid, rxpid, err = 0;
+	float sent_Bps = 0, recvd_Bps = 0;
+	int rx_fd, txpid, rxpid, err = 0;
 	struct msg_stats s = {0};
 	int status;
-	float sent_Bps = 0, recvd_Bps = 0;
 
 	errno = 0;
 
+	if (base)
+		rx_fd = p1;
+	else
+		rx_fd = p2;
+
 	rxpid = fork();
 	if (rxpid == 0) {
-		err = msg_loop(p2, iov_count, iov_buf, cnt, &s, false);
+		err = msg_loop(rx_fd, iov_count, iov_buf, cnt, &s, false);
 		if (err)
 			fprintf(stderr,
 				"msg_loop_rx: iov_count %i iov_buf %i cnt %i err %i\n",
@@ -427,6 +433,7 @@ static int forever_ping_pong(int rate, int verbose)
 enum {
 	PING_PONG,
 	SENDMSG,
+	BASE,
 };
 
 int main(int argc, char **argv)
@@ -466,6 +473,8 @@ int main(int argc, char **argv)
 				test = PING_PONG;
 			} else if (memcmp(optarg, "sendmsg", 7) == 0) {
 				test = SENDMSG;
+			} else if (memcmp(optarg, "base", 4) == 0) {
+				test = BASE;
 			} else {
 				usage(argv);
 				return -1;
@@ -491,6 +500,10 @@ int main(int argc, char **argv)
 	/* catch SIGINT */
 	signal(SIGINT, running_handler);
 
+	/* If base test skip BPF setup */
+	if (test == BASE)
+		goto run;
+
 	if (load_bpf_file(filename)) {
 		fprintf(stderr, "load_bpf_file: (%s) %s\n",
 			filename, strerror(errno));
@@ -522,6 +535,7 @@ int main(int argc, char **argv)
 		return err;
 	}
 
+run:
 	err = sockmap_init_sockets();
 	if (err) {
 		fprintf(stderr, "ERROR: test socket failed: %d\n", err);
@@ -531,7 +545,9 @@ int main(int argc, char **argv)
 	if (test == PING_PONG)
 		err = forever_ping_pong(rate, verbose);
 	else if (test == SENDMSG)
-		err = sendmsg_test(iov_count, length, rate, verbose);
+		err = sendmsg_test(iov_count, length, rate, verbose, false);
+	else if (test == BASE)
+		err = sendmsg_test(iov_count, length, rate, verbose, true);
 	else
 		fprintf(stderr, "unknown test\n");
 out:

Re: [bpf-next PATCH v3 5/7] bpf: sockmap sample add base test without any BPF for comparison

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:07:29PM -0800, John Fastabend wrote:
> Add a base test that does not use BPF hooks to test baseline case.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>

[bpf-next PATCH v3 6/7] bpf: sockmap put client sockets in blocking mode

[John Fastabend]

Put client sockets in blocking mode otherwise with sendmsg tests
its easy to overrun the socket buffers which results in the test
being aborted.

The original non-blocking was added to handle listen/accept with
a single thread the client/accepted sockets do not need to be
non-blocking.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index c3295a7..818766b 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -109,7 +109,7 @@ static int sockmap_init_sockets(void)
 	}
 
 	/* Non-blocking sockets */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < 2; i++) {
 		err = ioctl(*fds[i], FIONBIO, (char *)&one);
 		if (err < 0) {
 			perror("ioctl s1 failed()");

Re: [bpf-next PATCH v3 6/7] bpf: sockmap put client sockets in blocking mode

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:07:45PM -0800, John Fastabend wrote:
> Put client sockets in blocking mode otherwise with sendmsg tests
> its easy to overrun the socket buffers which results in the test
> being aborted.
> 
> The original non-blocking was added to handle listen/accept with
> a single thread the client/accepted sockets do not need to be
> non-blocking.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>

[bpf-next PATCH v3 7/7] bpf: sockmap set rlimit

[John Fastabend]

Avoid extra step of setting limit from cmdline and do it directly in
the program.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
---
 samples/sockmap/sockmap_user.c |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/samples/sockmap/sockmap_user.c b/samples/sockmap/sockmap_user.c
index 818766b..a6dab97 100644
--- a/samples/sockmap/sockmap_user.c
+++ b/samples/sockmap/sockmap_user.c
@@ -27,6 +27,7 @@
 #include <time.h>
 
 #include <sys/time.h>
+#include <sys/resource.h>
 #include <sys/types.h>
 
 #include <linux/netlink.h>
@@ -439,6 +440,7 @@ enum {
 int main(int argc, char **argv)
 {
 	int iov_count = 1, length = 1024, rate = 1, verbose = 0;
+	struct rlimit r = {10 * 1024 * 1024, RLIM_INFINITY};
 	int opt, longindex, err, cg_fd = 0;
 	int test = PING_PONG;
 	char filename[256];
@@ -493,6 +495,11 @@ int main(int argc, char **argv)
 		return -1;
 	}
 
+	if (setrlimit(RLIMIT_MEMLOCK, &r)) {
+		perror("setrlimit(RLIMIT_MEMLOCK)");
+		return 1;
+	}
+
 	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
 
 	running = 1;

Re: [bpf-next PATCH v3 7/7] bpf: sockmap set rlimit

[Martin KaFai Lau]

On Thu, Jan 11, 2018 at 09:08:02PM -0800, John Fastabend wrote:
> Avoid extra step of setting limit from cmdline and do it directly in
> the program.
> 
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>