Skip to contentSkip to conversation

ktest: Updates for 4.11

PATCH New Steven Rostedt 1 participant 6 messages Feb 7, 2017

Subsystems (2)
lore
Share & tools
lore
author maintainer
#2 lore raw

PATCH 2/5 ktest: Fix child exit code processing

In reply to Steven Rostedt

From: "Steven Rostedt (VMware)" <[email protected]>

The child_exit errno needs to be shifted by 8 bits to compare against the return values for the bisect variables.

Fixes c5dacb88f0a64 ("ktest: Allow overriding bisect test results")
---
 tools/testing/ktest/ktest.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/ktest/ktest.pl +1 -1
index caa90d391a96..a64da242b824 100755
@@ -2633,7 +2633,7 @@ sub do_run_test {
     }
 
     waitpid $child_pid, 0;
-    $child_exit = $?;
+    $child_exit = $? >> 8;
 
     my $end_time = time;
     $test_time = $end_time - $start_time;
Signature · 1 line
2.10.2
author maintainer
#3 lore raw

PATCH 3/5 ktest: Add timeout to ssh command

In reply to Steven Rostedt

From: "Steven Rostedt (VMware)" <[email protected]>

Add a timeout to performing an ssh command. This will let testing if a machine is alive or not, or if something else may be amiss. A timeout can be passed to ssh, where ssh will fail if it does not complete within the given timeout.

---
 tools/testing/ktest/ktest.pl | 35 +++++++++++++++++++++++++++--------
 1 file changed, 27 insertions(+), 8 deletions(-)
tools/testing/ktest/ktest.pl +27 -8
index a64da242b824..d9bdd3d6dba6 100755
@@ -1668,19 +1668,18 @@ sub fail {
 }
 
 sub run_command {
-    my ($command, $redirect) = @_;
+    my ($command, $redirect, $timeout) = @_;
     my $start_time;
     my $end_time;
     my $dolog = 0;
     my $dord = 0;
     my $pid;
 
-    $start_time = time;
-
     $command =~ s/\$SSH_USER/$ssh_user/g;
     $command =~ s/\$MACHINE/$machine/g;
 
     doprint("$command ... ");
+    $start_time = time;
 
     $pid = open(CMD, "$command 2>&1 |") or
 	(fail "unable to exec $command" and return 0);
@@ -1697,14 +1696,34 @@ sub run_command {
 	$dord = 1;
     }
 
-    while (<CMD>) {
-	print LOG if ($dolog);
-	print RD  if ($dord);
+    my $hit_timeout = 0;
+
+    while (1) {
+	my $fp = \*CMD;
+	if (defined($timeout)) {
+	    doprint "timeout = $timeout\n";
+	}
+	my $line = wait_for_input($fp, $timeout);
+	if (!defined($line)) {
+	    my $now = time;
+	    if (defined($timeout) && (($now - $start_time) >= $timeout)) {
+		doprint "Hit timeout of $timeout, killing process\n";
+		$hit_timeout = 1;
+		kill 9, $pid;
+	    }
+	    last;
+	}
+	print LOG $line if ($dolog);
+	print RD $line if ($dord);
     }
 
     waitpid($pid, 0);
     my $failed = $?;
 
+    if ($hit_timeout) {
+	$failed = 1;
+    }
+
     close(CMD);
     close(LOG) if ($dolog);
     close(RD)  if ($dord);
@@ -1728,11 +1747,11 @@ sub run_command {
 }
 
 sub run_ssh {
-    my ($cmd) = @_;
+    my ($cmd, $timeout) = @_;
     my $cp_exec = $ssh_exec;
 
     $cp_exec =~ s/\$SSH_COMMAND/$cmd/g;
-    return run_command "$cp_exec";
+    return run_command "$cp_exec", undef , $timeout;
 }
 
 sub run_scp {
Signature · 1 line
2.10.2

Steven Rostedt

author maintainer
#0 lore raw

cover ktest: Updates for 4.11

git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest.git
for-next

Head SHA1: 5739438b725c09938bb6af9e4477c4283b8326cf

Steven Rostedt (VMware) (5):
ktest: Have POST_TEST run after the test has totally completed
ktest: Fix child exit code processing
ktest: Add timeout to ssh command
ktest.pl: Powercycle the box on reboot if no connection can be made
ktest: Add variable run_command_status to save status of commands executed

----

 tools/testing/ktest/ktest.pl | 111 +++++++++++++++++++++++++++++--------------
 1 file changed, 76 insertions(+), 35 deletions(-)
author maintainer
#1 lore raw

PATCH 1/5 ktest: Have POST_TEST run after the test has totally completed

From: "Steven Rostedt (VMware)" <[email protected]>

The POST_TEST config is to be executed after a test has fully compeleted, whether the test passed or failed. It currently is executed at the moment that the test has been decided if it failed or not. As the test does other clean ups, it isn't truly finished. Move the POST_TEST execution to after all the test cleanups have been done.

---
 tools/testing/ktest/ktest.pl | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)
tools/testing/ktest/ktest.pl +12 -8
index d08e214ec6e7..caa90d391a96 100755
@@ -1412,6 +1412,10 @@ sub dodie {
 	    system("stty $stty_orig");
     }
 
+    if (defined($post_test)) {
+	run_command $post_test;
+    }
+
     die @_, "\n";
 }
 
@@ -1624,10 +1628,6 @@ sub save_logs {
 
 sub fail {
 
-	if (defined($post_test)) {
-		run_command $post_test;
-	}
-
 	if ($die_on_failure) {
 		dodie @_;
 	}
@@ -1660,6 +1660,10 @@ sub fail {
 	    save_logs "fail", $store_failures;
         }
 
+	if (defined($post_test)) {
+		run_command $post_test;
+	}
+
 	return 1;
 }
 
@@ -2489,10 +2493,6 @@ sub halt {
 sub success {
     my ($i) = @_;
 
-    if (defined($post_test)) {
-	run_command $post_test;
-    }
-
     $successes++;
 
     my $name = "";
@@ -2517,6 +2517,10 @@ sub success {
 	doprint "Reboot and wait $sleep_time seconds\n";
 	reboot_to_good $sleep_time;
     }
+
+    if (defined($post_test)) {
+	run_command $post_test;
+    }
 }
 
 sub answer_bisect {
Signature · 1 line
2.10.2
author maintainer
#5 lore raw

PATCH 5/5 ktest: Add variable run_command_status to save status of commands executed

In reply to Steven Rostedt

From: "Steven Rostedt (VMware)" <[email protected]>

Create a variable called run_command_status that saves the status of the executed commands and can be used by other functions later to test for status.

---
 tools/testing/ktest/ktest.pl | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)
tools/testing/ktest/ktest.pl +11 -11
index 6a1484cc4436..29470b554711 100755
@@ -179,6 +179,7 @@ my $localversion;
 my $iteration = 0;
 my $successes = 0;
 my $stty_orig;
+my $run_command_status = 0;
 
 my $bisect_good;
 my $bisect_bad;
@@ -1736,11 +1737,8 @@ sub run_command {
     }
 
     waitpid($pid, 0);
-    my $failed = $?;
-
-    if ($hit_timeout) {
-	$failed = 1;
-    }
+    # shift 8 for real exit status
+    $run_command_status = $? >> 8;
 
     close(CMD);
     close(LOG) if ($dolog);
@@ -1755,13 +1753,17 @@ sub run_command {
 	doprint "[$delta seconds] ";
     }
 
-    if ($failed) {
+    if ($hit_timeout) {
+	$run_command_status = 1;
+    }
+
+    if ($run_command_status) {
 	doprint "FAILED!\n";
     } else {
 	doprint "SUCCESS\n";
     }
 
-    return !$failed;
+    return !$run_command_status;
 }
 
 sub run_ssh {
@@ -2578,16 +2580,15 @@ sub answer_bisect {
 }
 
 sub child_run_test {
-    my $failed = 0;
 
     # child should have no power
     $reboot_on_error = 0;
     $poweroff_on_error = 0;
     $die_on_failure = 1;
 
-    run_command $run_test, $testlog or $failed = 1;
+    run_command $run_test, $testlog;
 
-    exit $failed;
+    exit $run_command_status;
 }
 
 my $child_done;
@@ -3371,7 +3372,6 @@ sub config_bisect {
     save_config \%good_configs, $good_config;
     save_config \%bad_configs, $bad_config;
 
-
     if (defined($config_bisect_check) && $config_bisect_check ne "0") {
 	if ($config_bisect_check ne "good") {
 	    doprint "Testing bad config\n";
Signature · 1 line
2.10.2
author maintainer
#4 lore raw

PATCH 4/5 ktest.pl: Powercycle the box on reboot if no connection can be made

In reply to Steven Rostedt

From: "Steven Rostedt (VMware)" <[email protected]>

When performing a reboot of the test box, try to ssh to it. If it can't connect for 5 seconds, then powercycle the box. This is useful because the reboot is done via ssh, and if you can't ssh to the box because it is hung, the reboot fails to reboot.

---
 tools/testing/ktest/ktest.pl | 40 +++++++++++++++++++++++++++++-----------
 1 file changed, 29 insertions(+), 11 deletions(-)
tools/testing/ktest/ktest.pl +29 -11
index d9bdd3d6dba6..6a1484cc4436 100755
@@ -1325,26 +1325,44 @@ sub wait_for_monitor;
 
 sub reboot {
     my ($time) = @_;
+    my $powercycle = 0;
 
-    # Make sure everything has been written to disk
-    run_ssh("sync");
+    # test if the machine can be connected to within 5 seconds
+    my $stat = run_ssh("echo check machine status", 5);
+    if (!$stat) {
+	doprint("power cycle\n");
+	$powercycle = 1;
+    }
+
+    if ($powercycle) {
+	run_command "$power_cycle";
 
-    if (defined($time)) {
 	start_monitor;
 	# flush out current monitor
 	# May contain the reboot success line
 	wait_for_monitor 1;
-    }
 
-    # try to reboot normally
-    if (run_command $reboot) {
-	if (defined($powercycle_after_reboot)) {
-	    sleep $powercycle_after_reboot;
+    } else {
+	# Make sure everything has been written to disk
+	run_ssh("sync");
+
+	if (defined($time)) {
+	    start_monitor;
+	    # flush out current monitor
+	    # May contain the reboot success line
+	    wait_for_monitor 1;
+	}
+
+	# try to reboot normally
+	if (run_command $reboot) {
+	    if (defined($powercycle_after_reboot)) {
+		sleep $powercycle_after_reboot;
+		run_command "$power_cycle";
+	    }
+	} else {
+	    # nope? power cycle it.
 	    run_command "$power_cycle";
 	}
-    } else {
-	# nope? power cycle it.
-	run_command "$power_cycle";
     }
 
     if (defined($time)) {
Signature · 1 line
2.10.2