Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1717905 > unrolled thread

[PATCH v2 0/2] enable hires timer to timeout datagram socket

Back to article view | Back to linux.kernel

Contents

  [PATCH v2 0/2] enable hires timer to timeout datagram socket Vallish Vaidyeshwara <vallish@amazon.com> - 2017-08-23 02:20 +0200
    [PATCH v2 2/2] selftests/net: add test to verify datagram socket timeout Vallish Vaidyeshwara <vallish@amazon.com> - 2017-08-23 02:20 +0200
    [PATCH v2 1/2] net: enable high resolution timer mode to timeout datagram sockets Vallish Vaidyeshwara <vallish@amazon.com> - 2017-08-23 02:20 +0200
    Re: [PATCH v2 0/2] enable hires timer to timeout datagram socket David Miller <davem@davemloft.net> - 2017-08-23 06:40 +0200

#1717905 — [PATCH v2 0/2] enable hires timer to timeout datagram socket

v1->v2:
 - Cong Wang pointed out MAX_SCHEDULE_TIMEOUT wraparound, fixed the
   patch 1/2 to accommodate MAX_SCHEDULE_TIMEOUT wait time
 - Changed format of printing total time from float to long in
   selftests patch 2/2

Hello Dave,

Resending the patch series to include netdev mailing list with a
cover letter.

I am submitting 2 patch series to enable hires timer to timeout
datagram sockets (AF_UNIX & AF_INET domain) and test code to test
timeout accuracy on these sockets.

There has been a behavior change in 4.9 kernel with refactoring of Kernel
timer wheel in 4.8. We have a use case wherein our datagram socket
application is sensitive to socket timeout including long timeouts.

One of the test runs with a timeout value of 180 seconds timed out at
190 seconds.
[root@]# ./datagram_sock_timeout 180000
datagram_sock_timeout failed: took 190.00 seconds
[root@]#
The same program when run on a 4.4 kernel would timeout more accurately and
the kernel added slack was not noticeable to user application.

Patch 1: Has core code change of enabling hires timer to timeout datagram
	 socket on AF_UNIX and AF_INET domain
	 Patch 2: Test code to report regression in timeout behavior related to
		 patch 1

Vallish Vaidyeshwara (2):
  net: enable high resolution timer mode to timeout datagram sockets
  selftests/net: add test to verify datagram socket timeout

 net/core/datagram.c                                |  14 ++-
 tools/testing/selftests/net/Makefile               |   3 +-
 .../testing/selftests/net/datagram_sock_timeout.c  | 119 +++++++++++++++++++++
 .../selftests/net/run_datagram_sock_timeout.sh     |  12 +++
 4 files changed, 146 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/net/datagram_sock_timeout.c
 create mode 100755 tools/testing/selftests/net/run_datagram_sock_timeout.sh

-- 
2.7.3.AMZN

[toc] | [next] | [standalone]

#1717906 — [PATCH v2 2/2] selftests/net: add test to verify datagram socket timeout

AF_UNIX and AF_INET datagram sockets use high resolution timer to time
SO_RCVTIMEO value used with setsockopt(2). This test checks for the
accuracy of kernel notifying these sockets timeout to application. Test
program has code to check AF_UNIX socket, however the kernel function used
to timeout AF_INET socket is the same kernel function used by AF_UNIX as
well which is __skb_wait_for_more_packets().

Reported-by: Manjula Peiris <thelgep@amazon.com>
Reviewed-by: Eduardo Valentin <eduval@amazon.com>
Reviewed-by: Anchal Agarwal <anchalag@amazon.com>
Signed-off-by: Vallish Vaidyeshwara <vallish@amazon.com>
---
 tools/testing/selftests/net/Makefile               |   3 +-
 .../testing/selftests/net/datagram_sock_timeout.c  | 119 +++++++++++++++++++++
 .../selftests/net/run_datagram_sock_timeout.sh     |  12 +++
 3 files changed, 133 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/net/datagram_sock_timeout.c
 create mode 100755 tools/testing/selftests/net/run_datagram_sock_timeout.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index f6c9dbf..eb5a8c7 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -3,11 +3,12 @@
 CFLAGS =  -Wall -Wl,--no-as-needed -O2 -g
 CFLAGS += -I../../../../usr/include/
 
-TEST_PROGS := run_netsocktests run_afpackettests test_bpf.sh netdevice.sh
+TEST_PROGS := run_netsocktests run_afpackettests test_bpf.sh netdevice.sh run_datagram_sock_timeout.sh
 TEST_GEN_FILES =  socket
 TEST_GEN_FILES += psock_fanout psock_tpacket
 TEST_GEN_FILES += reuseport_bpf reuseport_bpf_cpu reuseport_bpf_numa
 TEST_GEN_FILES += reuseport_dualstack
+TEST_GEN_FILES += datagram_sock_timeout
 
 include ../lib.mk
 
diff --git a/tools/testing/selftests/net/datagram_sock_timeout.c b/tools/testing/selftests/net/datagram_sock_timeout.c
new file mode 100644
index 0000000..362becf
--- /dev/null
+++ b/tools/testing/selftests/net/datagram_sock_timeout.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright 2017 Amazon.com, Inc.
+ * Author: Manjula Peiris <thelgep@amazon.com>
+ *         Vallish Vaidyeshwara <vallish@amazon.com>
+ *
+ * selftests/net: test to verify datagram socket timeout
+ *
+ * AF_UNIX and AF_INET datagram sockets use high resolution timer to time
+ * SO_RCVTIMEO value used with setsockopt(2). This test checks for the accuracy
+ * of kernel notifying these sockets timeout to application. Test program has
+ * code to check AF_UNIX socket, however the kernel function used to timeout
+ * AF_INET socket is the same kernel function used by AF_UNIX as well which is
+ * __skb_wait_for_more_packets().
+ *
+ * License (GPLv2):
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. * See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.
+ */
+
+#include <stdlib.h>
+#include <sys/socket.h>
+#include <unistd.h>
+#include <signal.h>
+#include <stdio.h>
+#include <fcntl.h>
+#include <err.h>
+#include <errno.h>
+#include <sys/un.h>
+#include <time.h>
+#include <assert.h>
+
+#define BUF_SIZE 128
+#define KB 1024
+#define NUM_FD 2
+
+static int set_socket_timeout(int sockfd, unsigned int ms)
+{
+	int ret;
+	struct timeval timeout;
+	socklen_t cb = sizeof(timeout);
+
+	timeout.tv_sec = ms / 1000;
+	timeout.tv_usec = (ms % 1000) * 1000;
+	ret = setsockopt(sockfd, SOL_SOCKET, SO_RCVTIMEO, &timeout, cb);
+	return ret;
+}
+
+int main(int argc, char **argv)
+{
+	char err[BUF_SIZE];
+	int ret;
+	int fds[NUM_FD];
+	struct msghdr message;
+	char buffer[KB];
+	struct sockaddr_storage src_addr;
+	struct iovec iov[1];
+	time_t start, end;
+	unsigned int timeout;
+
+	iov[0].iov_base = buffer;
+	iov[0].iov_len = sizeof(buffer);
+	message.msg_name = &src_addr;
+	message.msg_namelen = sizeof(src_addr);
+	message.msg_iov = iov;
+	message.msg_iovlen = 1;
+	message.msg_control = 0;
+	message.msg_controllen = 0;
+
+	if (argc != 2) {
+		fprintf(stderr,
+			"datagram_sock_timeout failed: no timeout specified\n");
+		return -1;
+	}
+	timeout = (unsigned int)(atoi(argv[1]));
+
+	if (socketpair(AF_UNIX, SOCK_DGRAM, 0, fds) != 0) {
+		strerror_r(errno, err, BUF_SIZE);
+		fprintf(stderr, "socketpair() call failed with %s\n", err);
+		return -1;
+	}
+
+	if (set_socket_timeout(fds[0], timeout) != 0) {
+		strerror_r(errno, err, BUF_SIZE);
+		fprintf(stderr, "setsockopt() call failed with %s\n", err);
+		return -1;
+	}
+
+	start = time(NULL);
+	ret = (int)recvmsg(fds[0], &message, 0);
+	end = time(NULL);
+	if (!(ret == -1 && errno == 11)) {
+		fprintf(stderr,
+			"datagram_sock_timeout failed: test was interrupted\n");
+		return -1;
+	}
+
+	if ((end - start) != (timeout / 1000)) {
+		fprintf(stderr,
+			"datagram_sock_timeout failed: took %lu seconds\n",
+			(long)(end - start));
+		return -1;
+	}
+
+	close(fds[0]);
+	close(fds[1]);
+
+	fprintf(stderr, "datagram_sock_timeout passed\n");
+	return 0;
+}
diff --git a/tools/testing/selftests/net/run_datagram_sock_timeout.sh b/tools/testing/selftests/net/run_datagram_sock_timeout.sh
new file mode 100755
index 0000000..d5f4f82
--- /dev/null
+++ b/tools/testing/selftests/net/run_datagram_sock_timeout.sh
@@ -0,0 +1,12 @@
+#!/bin/sh
+# Runs datagram socket timeout test
+
+echo "--------------------"
+echo "running run_datagram_sock_timeout test"
+echo "--------------------"
+./datagram_sock_timeout 180000
+if [ $? -ne 0 ]; then
+	echo "[FAIL]"
+else
+	echo "[PASS]"
+fi
-- 
2.7.3.AMZN

[toc] | [prev] | [next] | [standalone]

#1717907 — [PATCH v2 1/2] net: enable high resolution timer mode to timeout datagram sockets

Enable high resolution timer mode to time SO_RCVTIMEO value used with
setsockopt(2) on AF_UNIX and AF_INET datagram sockets. By default,
SO_RCVTIMEO uses low resolution timer which is good for most of socket
use cases.

Background:
Kernel timer wheel was refactored in 4.8 to avoid drawbacks with previous
implementation:
https://lwn.net/Articles/691064/
Unlike the previous "kernel timer wheel" implementation in 4.4 which aimed
for accuracy by paying cost for cascading tracked timers at the boundary of
256 jiffies, the new timer wheel implementation gets rid of cascading
latency by paying a price for being less accurate for far off timers.

Use Case:
New implementation is good for most of socket use cases. However we have a
use case where our application is sensitive to socket timeout including
long timeouts.  Please refer to test code as part of this patch series.
One of the test runs with a timeout value of 180 seconds timed out at
190 seconds.
[root@]# ./datagram_sock_timeout 180000
datagram_sock_timeout failed: took 190.00 seconds
[root@]#
The same program when run on a 4.4 kernel would timeout more acurately and
the kernel added slack was not noticeable to user application.

Interesting text:
a) Standards for setsockopt:
http://pubs.opengroup.org/onlinepubs/009695399/functions/setsockopt.html
<snip>
SO_RCVTIMEO
Sets the timeout value that specifies the maximum amount of time an input
function waits until it completes. It accepts a timeval structure with the
number of seconds and microseconds specifying the limit on how long to wait
for an input operation to complete. If a receive operation has blocked for
this much time without receiving additional data, it shall return with a
partial count or errno set to [EAGAIN] or [EWOULDBLOCK] if no data is
received. The default for this option is zero, which indicates that a
receive operation shall not time out. This option takes a timeval
structure. Note that not all implementations allow this option to be set.
<end snip>
This only talks about the maximum time and the current behavior indeed
follows this standard. System call does not return before the time
specified and it does return EAGAIN.
b) Man page for SETSOCKOPT(3P):
<snip>
The  option_name  argument  specifies  a  single  option to set. It can be
one of the socket-level options defined in <sys_socket.h> and described in
Section 2.10.16, Use of Options.  If option_name is equal to SO_RCVTIMEO
or SO_SNDTIMEO and the implementation supports setting the option, it is
unspecified whether the struct timeval  pointed  to by  option_value  is
stored  as  provided by this function or is rounded up to align with the
resolution of the clock being used. If setsockopt() is called with
option_name equal to SO_ACCEPTCONN, SO_ERROR, or SO_TYPE, the behavior is
unspecified.
<end snip>
Behavior is unspecified.
3) Man page for SELECT(2):
<snip>
Note  that  the  timeout  interval  will  be  rounded up to the system
clock granularity, and kernel scheduling delays mean that the blocking
interval may overrun by a small amount.  If both fields of the timeval
structure are zero, then select() returns immediately.  (This is useful
for polling.)  If timeout is NULL (no timeout),  select()  can block
indefinitely.
<end snip>
Select system call guarantees timeout interval and inturn uses highres
timer.

Reported-by: Manjula Peiris <thelgep@amazon.com>
Reviewed-by: Eduardo Valentin <eduval@amazon.com>
Reviewed-by: Anchal Agarwal <anchalag@amazon.com>
Signed-off-by: Vallish Vaidyeshwara <vallish@amazon.com>
---
 net/core/datagram.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index ee5647b..832c147 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -80,6 +80,7 @@ static int receiver_wake_function(wait_queue_entry_t *wait, unsigned int mode, i
 		return 0;
 	return autoremove_wake_function(wait, mode, sync, key);
 }
+
 /*
  * Wait for the last received packet to be different from skb
  */
@@ -87,6 +88,9 @@ int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p,
 				const struct sk_buff *skb)
 {
 	int error;
+	ktime_t expires;
+	struct timespec64 time;
+	unsigned long pre_sched_time;
 	DEFINE_WAIT_FUNC(wait, receiver_wake_function);
 
 	prepare_to_wait_exclusive(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
@@ -116,7 +120,15 @@ int __skb_wait_for_more_packets(struct sock *sk, int *err, long *timeo_p,
 		goto interrupted;
 
 	error = 0;
-	*timeo_p = schedule_timeout(*timeo_p);
+	/* Wait using highres timer */
+	time.tv_sec = *timeo_p / HZ;
+	time.tv_nsec = jiffies_to_nsecs(*timeo_p % HZ);
+	expires = ktime_add_safe(ktime_get(), timespec64_to_ktime(time));
+	pre_sched_time = jiffies;
+	if (schedule_hrtimeout(&expires, HRTIMER_MODE_ABS))
+		*timeo_p = jiffies - pre_sched_time;
+	else
+		*timeo_p = 0;
 out:
 	finish_wait(sk_sleep(sk), &wait);
 	return error;
-- 
2.7.3.AMZN

[toc] | [prev] | [next] | [standalone]

#1718027

From: Vallish Vaidyeshwara <vallish@amazon.com>
Date: Wed, 23 Aug 2017 00:10:25 +0000

> I am submitting 2 patch series to enable hires timer to timeout
> datagram sockets (AF_UNIX & AF_INET domain) and test code to test
> timeout accuracy on these sockets.

This is not reasonable.

If you want high resolution events with real guarantees, please use
the kernel interfaces which provide this as explained to you as
feedback by other reviewers.

I'm not applying this, sorry.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web