Merge branch 'for-5.14/amd-sfh' into for-linus

- support for Renoir and Cezanne SoCs
- support for Ambient Light Sensor
- support for Human Presence Detection sensor

all from Basavaraj Natikar

72 files changed, 8843 insertions, 887 deletions

diff --git a/tools/testing/ktest/examples/vmware.conf b/tools/testing/ktest/examples/vmware.conf
new file mode 100644
index 000000000000..61958163d242
--- /dev/null
+++ b/tools/testing/ktest/examples/vmware.conf
@@ -0,0 +1,137 @@
+#
+# This config is an example usage of ktest.pl with a vmware guest
+#
+# VMware Setup:
+# -------------
+# - Edit the Virtual Machine ("Edit virtual machine settings")
+# - Add a Serial Port
+#   - You almost certainly want it set "Connect at power on"
+#   - Select "Use socket (named pipe)"
+#   - Select a name that you'll recognize, like 'ktestserialpipe'
+#   - From: Server
+#   - To: A Virtual Machine
+#   - Save
+# - Make sure you note the name, it will be in the base directory of the
+#   virtual machine (where the "disks" are stored.  The default
+#   is /var/lib/vmware/<virtual machine name>/<the name you entered above>
+#
+# - Make note of the path to the VM
+# </End VMware setup>
+#
+# The guest is called 'Guest' and this would be something that
+# could be run on the host to test a virtual machine target.
+
+MACHINE = Guest
+
+# Name of the serial pipe you set in the VMware settings
+VMWARE_SERIAL_NAME = <the name you entered above>
+
+# Define a variable of the name of the VM
+# Noting this needs to be the name of the kmx file, and usually, the
+# name of the directory that it's in.  If the directory and name
+# differ change the VMWARE_VM_DIR accordingly.
+# Please ommit the .kmx extension
+VMWARE_VM_NAME = <virtual machine name>
+
+# VM dir name.  This is usually the same as the virtual machine's name,
+# but not always the case.  Change if they differ
+VMWARE_VM_DIR = ${VMWARE_VM_NAME}
+
+# Base directory that the Virtual machine is contained in
+# /var/lib/vmware is the default on Linux
+VMWARE_VM_BASE_DIR = /var/lib/vmware/${VMWARE_VM_DIR}
+
+# Use ncat to read the unix pipe.  Anything that can read the Unix Pipe
+# and output it's contents to stdout will work
+CONSOLE = /usr/bin/ncat -U ${VMWARE_VM_BASE_DIR}/${VMWARE_SERIAL_NAME}
+
+# Define what version of Workstation you are using
+# This is used by vmrun to use the appropriate appripriate pieces to 
+# test this.  In all likelihood you want 'ws' or 'player' 
+# Valid options:
+# 	ws - Workstation (Windows or Linux host)
+# 	fusion - Fusion (Mac host)
+# 	player - Using VMware Player (Windows or Linux host)
+# Note: vmrun has to run directly on the host machine
+VMWARE_HOST_TYPE = ws
+
+# VMware provides `vmrun` to allow you to do certain things to the virtual machine
+# This should hard reset the VM and force a boot
+VMWARE_POWER_CYCLE = /usr/bin/vmrun -T ${VMWARE_HOST_TYPE} reset ${VMWARE_VM_BASE_DIR}/${VMWARE_VM_NAME}.kmx nogui
+
+#*************************************#
+# This part is the same as test.conf  #
+#*************************************#
+
+# The include files will set up the type of test to run. Just set TEST to
+# which test you want to run.
+#
+# TESTS = patchcheck, randconfig, boot, test, config-bisect, bisect, min-config
+#
+# See the include/*.conf files that define these tests
+#
+TEST := patchcheck
+
+# Some tests may have more than one test to run. Define MULTI := 1 to run
+# the extra tests.
+MULTI := 0
+
+# In case you want to differentiate which type of system you are testing
+BITS := 64
+
+# REBOOT = none, error, fail, empty
+#  See include/defaults.conf
+REBOOT := empty
+
+
+# The defaults file will set up various settings that can be used by all
+# machine configs.
+INCLUDE include/defaults.conf
+
+
+#*************************************#
+# Now we are different from test.conf #
+#*************************************#
+
+
+# The example here assumes that Guest is running a Fedora release
+# that uses dracut for its initfs. The POST_INSTALL will be executed
+# after the install of the kernel and modules are complete.
+#
+POST_INSTALL = ${SSH} /sbin/dracut -f /boot/initramfs-test.img $KERNEL_VERSION
+
+# Guests sometimes get stuck on reboot. We wait 3 seconds after running
+# the reboot command and then  do a full power-cycle of the guest.
+# This forces the guest to restart.
+#
+POWERCYCLE_AFTER_REBOOT = 3
+
+# We do the same after the halt command, but this time we wait 20 seconds.
+POWEROFF_AFTER_HALT = 20
+
+
+# As the defaults.conf file has a POWER_CYCLE option already defined,
+# and options can not be defined in the same section more than once
+# (all DEFAULTS sections are considered the same). We use the
+# DEFAULTS OVERRIDE to tell ktest.pl to ignore the previous defined
+# options, for the options set in the OVERRIDE section.
+#
+DEFAULTS OVERRIDE
+
+# Instead of using the default POWER_CYCLE option defined in
+# defaults.conf, we use virsh to cycle it. To do so, we destroy
+# the guest, wait 5 seconds, and then start it up again.
+# Crude, but effective.
+#
+POWER_CYCLE = ${VMWARE_POWER_CYCLE}
+
+
+DEFAULTS
+
+# The following files each handle a different test case.
+# Having them included allows you to set up more than one machine and share
+# the same tests.
+INCLUDE include/patchcheck.conf
+INCLUDE include/tests.conf
+INCLUDE include/bisect.conf
+INCLUDE include/min-config.conf
diff --git a/tools/testing/ktest/ktest.pl b/tools/testing/ktest/ktest.pl
index 4e2450964517..09d1578f9d66 100755
--- a/tools/testing/ktest/ktest.pl
+++ b/tools/testing/ktest/ktest.pl
@@ -24,7 +24,7 @@ my %evals;
 
 #default opts
 my %default = (
-    "MAILER"			=> "sendmail",  # default mailer
+    "MAILER"			=> "sendmail",	# default mailer
     "EMAIL_ON_ERROR"		=> 1,
     "EMAIL_WHEN_FINISHED"	=> 1,
     "EMAIL_WHEN_CANCELED"	=> 0,
@@ -36,15 +36,15 @@ my %default = (
     "CLOSE_CONSOLE_SIGNAL"	=> "INT",
     "TIMEOUT"			=> 120,
     "TMP_DIR"			=> "/tmp/ktest/\${MACHINE}",
-    "SLEEP_TIME"		=> 60,	# sleep time between tests
+    "SLEEP_TIME"		=> 60,		# sleep time between tests
     "BUILD_NOCLEAN"		=> 0,
     "REBOOT_ON_ERROR"		=> 0,
     "POWEROFF_ON_ERROR"		=> 0,
     "REBOOT_ON_SUCCESS"		=> 1,
     "POWEROFF_ON_SUCCESS"	=> 0,
     "BUILD_OPTIONS"		=> "",
-    "BISECT_SLEEP_TIME"		=> 60,   # sleep time between bisects
-    "PATCHCHECK_SLEEP_TIME"	=> 60, # sleep time between patch checks
+    "BISECT_SLEEP_TIME"		=> 60,		# sleep time between bisects
+    "PATCHCHECK_SLEEP_TIME"	=> 60, 		# sleep time between patch checks
     "CLEAR_LOG"			=> 0,
     "BISECT_MANUAL"		=> 0,
     "BISECT_SKIP"		=> 1,
@@ -512,6 +512,69 @@ $config_help{"REBOOT_SCRIPT"} = << "EOF"
 EOF
     ;
 
+# used with process_expression()
+my $d = 0;
+
+# defined before get_test_name()
+my $in_die = 0;
+
+# defined before process_warning_line()
+my $check_build_re = ".*:.*(warning|error|Error):.*";
+my $utf8_quote = "\\x{e2}\\x{80}(\\x{98}|\\x{99})";
+
+# defined before child_finished()
+my $child_done;
+
+# config_ignore holds the configs that were set (or unset) for
+# a good config and we will ignore these configs for the rest
+# of a config bisect. These configs stay as they were.
+my %config_ignore;
+
+# config_set holds what all configs were set as.
+my %config_set;
+
+# config_off holds the set of configs that the bad config had disabled.
+# We need to record them and set them in the .config when running
+# olddefconfig, because olddefconfig keeps the defaults.
+my %config_off;
+
+# config_off_tmp holds a set of configs to turn off for now
+my @config_off_tmp;
+
+# config_list is the set of configs that are being tested
+my %config_list;
+my %null_config;
+
+my %dependency;
+
+# found above run_config_bisect()
+my $pass = 1;
+
+# found above add_dep()
+
+my %depends;
+my %depcount;
+my $iflevel = 0;
+my @ifdeps;
+
+# prevent recursion
+my %read_kconfigs;
+
+# found above test_this_config()
+my %min_configs;
+my %keep_configs;
+my %save_configs;
+my %processed_configs;
+my %nochange_config;
+
+#
+# These are first defined here, main function later on
+#
+sub run_command;
+sub start_monitor;
+sub end_monitor;
+sub wait_for_monitor;
+
 sub _logit {
     if (defined($opt{"LOG_FILE"})) {
 	print LOG @_;
@@ -537,7 +600,7 @@ sub read_prompt {
     my $ans;
 
     for (;;) {
-	if ($cancel) {
+        if ($cancel) {
 	    print "$prompt [y/n/C] ";
 	} else {
 	    print "$prompt [Y/n] ";
@@ -760,7 +823,7 @@ sub process_variables {
     # remove the space added in the beginning
     $retval =~ s/ //;
 
-    return "$retval"
+    return "$retval";
 }
 
 sub set_value {
@@ -863,7 +926,6 @@ sub value_defined {
 	defined($opt{$2});
 }
 
-my $d = 0;
 sub process_expression {
     my ($name, $val) = @_;
 
@@ -978,7 +1040,6 @@ sub __read_config {
 	    $override = 0;
 
 	    if ($type eq "TEST_START") {
-
 		if ($num_tests_set) {
 		    die "$name: $.: Can not specify both NUM_TESTS and TEST_START\n";
 		}
@@ -1048,7 +1109,6 @@ sub __read_config {
 		$test_num = $old_test_num;
 		$repeat = $old_repeat;
 	    }
-
 	} elsif (/^\s*ELSE\b(.*)$/) {
 	    if (!$if) {
 		die "$name: $.: ELSE found with out matching IF section\n$_";
@@ -1095,7 +1155,7 @@ sub __read_config {
 		    }
 		}
 	    }
-		
+
 	    if ( ! -r $file ) {
 		die "$name: $.: Can't read file $file\n$_";
 	    }
@@ -1186,13 +1246,13 @@ sub __read_config {
 }
 
 sub get_test_case {
-	print "What test case would you like to run?\n";
-	print " (build, install or boot)\n";
-	print " Other tests are available but require editing ktest.conf\n";
-	print " (see tools/testing/ktest/sample.conf)\n";
-	my $ans = <STDIN>;
-	chomp $ans;
-	$default{"TEST_TYPE"} = $ans;
+    print "What test case would you like to run?\n";
+    print " (build, install or boot)\n";
+    print " Other tests are available but require editing ktest.conf\n";
+    print " (see tools/testing/ktest/sample.conf)\n";
+    my $ans = <STDIN>;
+    chomp $ans;
+    $default{"TEST_TYPE"} = $ans;
 }
 
 sub read_config {
@@ -1368,11 +1428,6 @@ sub eval_option {
     return $option;
 }
 
-sub run_command;
-sub start_monitor;
-sub end_monitor;
-sub wait_for_monitor;
-
 sub reboot {
     my ($time) = @_;
     my $powercycle = 0;
@@ -1457,8 +1512,6 @@ sub do_not_reboot {
 	($test_type eq "config_bisect" && $opt{"CONFIG_BISECT_TYPE[$i]"} eq "build");
 }
 
-my $in_die = 0;
-
 sub get_test_name() {
     my $name;
 
@@ -1471,7 +1524,6 @@ sub get_test_name() {
 }
 
 sub dodie {
-
     # avoid recursion
     return if ($in_die);
     $in_die = 1;
@@ -1481,10 +1533,8 @@ sub dodie {
     doprint "CRITICAL FAILURE... [TEST $i] ", @_, "\n";
 
     if ($reboot_on_error && !do_not_reboot) {
-
 	doprint "REBOOTING\n";
 	reboot_to_good;
-
     } elsif ($poweroff_on_error && defined($power_off)) {
 	doprint "POWERING OFF\n";
 	`$power_off`;
@@ -1519,13 +1569,14 @@ sub dodie {
 	    close O;
 	    close L;
 	}
-        send_email("KTEST: critical failure for test $i [$name]",
-                "Your test started at $script_start_time has failed with:\n@_\n", $log_file);
+
+	send_email("KTEST: critical failure for test $i [$name]",
+		"Your test started at $script_start_time has failed with:\n@_\n", $log_file);
     }
 
     if ($monitor_cnt) {
-	    # restore terminal settings
-	    system("stty $stty_orig");
+	# restore terminal settings
+	system("stty $stty_orig");
     }
 
     if (defined($post_test)) {
@@ -1709,81 +1760,81 @@ sub wait_for_monitor {
 }
 
 sub save_logs {
-	my ($result, $basedir) = @_;
-	my @t = localtime;
-	my $date = sprintf "%04d%02d%02d%02d%02d%02d",
-		1900+$t[5],$t[4],$t[3],$t[2],$t[1],$t[0];
+    my ($result, $basedir) = @_;
+    my @t = localtime;
+    my $date = sprintf "%04d%02d%02d%02d%02d%02d",
+	1900+$t[5],$t[4],$t[3],$t[2],$t[1],$t[0];
 
-	my $type = $build_type;
-	if ($type =~ /useconfig/) {
-	    $type = "useconfig";
-	}
+    my $type = $build_type;
+    if ($type =~ /useconfig/) {
+	$type = "useconfig";
+    }
 
-	my $dir = "$machine-$test_type-$type-$result-$date";
+    my $dir = "$machine-$test_type-$type-$result-$date";
 
-	$dir = "$basedir/$dir";
+    $dir = "$basedir/$dir";
 
-	if (!-d $dir) {
-	    mkpath($dir) or
-		dodie "can't create $dir";
-	}
+    if (!-d $dir) {
+	mkpath($dir) or
+	    dodie "can't create $dir";
+    }
 
-	my %files = (
-		"config" => $output_config,
-		"buildlog" => $buildlog,
-		"dmesg" => $dmesg,
-		"testlog" => $testlog,
-	);
+    my %files = (
+	"config" => $output_config,
+	"buildlog" => $buildlog,
+	"dmesg" => $dmesg,
+	"testlog" => $testlog,
+    );
 
-	while (my ($name, $source) = each(%files)) {
-		if (-f "$source") {
-			cp "$source", "$dir/$name" or
-				dodie "failed to copy $source";
-		}
+    while (my ($name, $source) = each(%files)) {
+	if (-f "$source") {
+	    cp "$source", "$dir/$name" or
+		dodie "failed to copy $source";
 	}
+    }
 
-	doprint "*** Saved info to $dir ***\n";
+    doprint "*** Saved info to $dir ***\n";
 }
 
 sub fail {
 
-	if ($die_on_failure) {
-		dodie @_;
-	}
+    if ($die_on_failure) {
+	dodie @_;
+    }
 
-	doprint "FAILED\n";
+    doprint "FAILED\n";
 
-	my $i = $iteration;
+    my $i = $iteration;
 
-	# no need to reboot for just building.
-	if (!do_not_reboot) {
-	    doprint "REBOOTING\n";
-	    reboot_to_good $sleep_time;
-	}
+    # no need to reboot for just building.
+    if (!do_not_reboot) {
+	doprint "REBOOTING\n";
+	reboot_to_good $sleep_time;
+    }
 
-	my $name = "";
+    my $name = "";
 
-	if (defined($test_name)) {
-	    $name = " ($test_name)";
-	}
+    if (defined($test_name)) {
+	$name = " ($test_name)";
+    }
 
-	print_times;
+    print_times;
 
-	doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
-	doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
-	doprint "KTEST RESULT: TEST $i$name Failed: ", @_, "\n";
-	doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
-	doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
+    doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
+    doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
+    doprint "KTEST RESULT: TEST $i$name Failed: ", @_, "\n";
+    doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
+    doprint "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
 
-	if (defined($store_failures)) {
-	    save_logs "fail", $store_failures;
-        }
+    if (defined($store_failures)) {
+	save_logs "fail", $store_failures;
+    }
 
-	if (defined($post_test)) {
-		run_command $post_test;
-	}
+    if (defined($post_test)) {
+	run_command $post_test;
+    }
 
-	return 1;
+    return 1;
 }
 
 sub run_command {
@@ -1915,8 +1966,8 @@ sub _get_grub_index {
     my ($command, $target, $skip) = @_;
 
     return if (defined($grub_number) && defined($last_grub_menu) &&
-	       $last_grub_menu eq $grub_menu && defined($last_machine) &&
-	       $last_machine eq $machine);
+	$last_grub_menu eq $grub_menu && defined($last_machine) &&
+	$last_machine eq $machine);
 
     doprint "Find $reboot_type menu ... ";
     $grub_number = -1;
@@ -1924,8 +1975,8 @@ sub _get_grub_index {
     my $ssh_grub = $ssh_exec;
     $ssh_grub =~ s,\$SSH_COMMAND,$command,g;
 
-    open(IN, "$ssh_grub |")
-	or dodie "unable to execute $command";
+    open(IN, "$ssh_grub |") or
+	dodie "unable to execute $command";
 
     my $found = 0;
 
@@ -1969,9 +2020,9 @@ sub get_grub_index {
 	$target = '^menuentry.*' . $grub_menu_qt;
 	$skip = '^menuentry\s|^submenu\s';
     } elsif ($reboot_type eq "grub2bls") {
-        $command = $grub_bls_get;
-        $target = '^title=.*' . $grub_menu_qt;
-        $skip = '^title=';
+	$command = $grub_bls_get;
+	$target = '^title=.*' . $grub_menu_qt;
+	$skip = '^title=';
     } else {
 	return;
     }
@@ -1979,8 +2030,7 @@ sub get_grub_index {
     _get_grub_index($command, $target, $skip);
 }
 
-sub wait_for_input
-{
+sub wait_for_input {
     my ($fp, $time) = @_;
     my $start_time;
     my $rin;
@@ -2096,7 +2146,6 @@ sub monitor {
     my $version_found = 0;
 
     while (!$done) {
-
 	if ($bug && defined($stop_after_failure) &&
 	    $stop_after_failure >= 0) {
 	    my $time = $stop_after_failure - (time - $failure_start);
@@ -2349,9 +2398,6 @@ sub start_monitor_and_install {
     return monitor;
 }
 
-my $check_build_re = ".*:.*(warning|error|Error):.*";
-my $utf8_quote = "\\x{e2}\\x{80}(\\x{98}|\\x{99})";
-
 sub process_warning_line {
     my ($line) = @_;
 
@@ -2394,7 +2440,7 @@ sub check_buildlog {
 	while (<IN>) {
 	    if (/$check_build_re/) {
 		my $warning = process_warning_line $_;
-		
+
 		$warnings_list{$warning} = 1;
 	    }
 	}
@@ -2571,7 +2617,6 @@ sub build {
 	    run_command "mv $outputdir/config_temp $output_config" or
 		dodie "moving config_temp";
 	}
-
     } elsif (!$noclean) {
 	unlink "$output_config";
 	run_command "$make mrproper" or
@@ -2594,6 +2639,9 @@ sub build {
     # Run old config regardless, to enforce min configurations
     make_oldconfig;
 
+    if (not defined($build_options)){
+	$build_options = "";
+    }
     my $build_ret = run_command "$make $build_options", $buildlog;
 
     if (defined($post_build)) {
@@ -2649,14 +2697,15 @@ sub success {
 
     print_times;
 
-    doprint "\n\n*******************************************\n";
-    doprint     "*******************************************\n";
-    doprint     "KTEST RESULT: TEST $i$name SUCCESS!!!!         **\n";
-    doprint     "*******************************************\n";
-    doprint     "*******************************************\n";
+    doprint "\n\n";
+    doprint "*******************************************\n";
+    doprint "*******************************************\n";
+    doprint "KTEST RESULT: TEST $i$name SUCCESS!!!!   **\n";
+    doprint "*******************************************\n";
+    doprint "*******************************************\n";
 
     if (defined($store_successes)) {
-        save_logs "success", $store_successes;
+	save_logs "success", $store_successes;
     }
 
     if ($i != $opt{"NUM_TESTS"} && !do_not_reboot) {
@@ -2698,8 +2747,6 @@ sub child_run_test {
     exit $run_command_status;
 }
 
-my $child_done;
-
 sub child_finished {
     $child_done = 1;
 }
@@ -3031,7 +3078,6 @@ sub bisect {
     }
 
     if ($do_check) {
-
 	# get current HEAD
 	my $head = get_sha1("HEAD");
 
@@ -3071,13 +3117,11 @@ sub bisect {
 	run_command "git bisect replay $replay" or
 	    dodie "failed to run replay";
     } else {
-
 	run_command "git bisect good $good" or
 	    dodie "could not set bisect good to $good";
 
 	run_git_bisect "git bisect bad $bad" or
 	    dodie "could not set bisect bad to $bad";
-
     }
 
     if (defined($start)) {
@@ -3103,35 +3147,13 @@ sub bisect {
     success $i;
 }
 
-# config_ignore holds the configs that were set (or unset) for
-# a good config and we will ignore these configs for the rest
-# of a config bisect. These configs stay as they were.
-my %config_ignore;
-
-# config_set holds what all configs were set as.
-my %config_set;
-
-# config_off holds the set of configs that the bad config had disabled.
-# We need to record them and set them in the .config when running
-# olddefconfig, because olddefconfig keeps the defaults.
-my %config_off;
-
-# config_off_tmp holds a set of configs to turn off for now
-my @config_off_tmp;
-
-# config_list is the set of configs that are being tested
-my %config_list;
-my %null_config;
-
-my %dependency;
-
 sub assign_configs {
     my ($hash, $config) = @_;
 
     doprint "Reading configs from $config\n";
 
-    open (IN, $config)
-	or dodie "Failed to read $config";
+    open (IN, $config) or
+	dodie "Failed to read $config";
 
     while (<IN>) {
 	chomp;
@@ -3219,8 +3241,6 @@ sub config_bisect_end {
     doprint "***************************************\n\n";
 }
 
-my $pass = 1;
-
 sub run_config_bisect {
     my ($good, $bad, $last_result) = @_;
     my $reset = "";
@@ -3243,13 +3263,13 @@ sub run_config_bisect {
 
     $ret = run_config_bisect_test $config_bisect_type;
     if ($ret) {
-        doprint "NEW GOOD CONFIG ($pass)\n";
+	doprint "NEW GOOD CONFIG ($pass)\n";
 	system("cp $output_config $tmpdir/good_config.tmp.$pass");
 	$pass++;
 	# Return 3 for good config
 	return 3;
     } else {
-        doprint "NEW BAD CONFIG ($pass)\n";
+	doprint "NEW BAD CONFIG ($pass)\n";
 	system("cp $output_config $tmpdir/bad_config.tmp.$pass");
 	$pass++;
 	# Return 4 for bad config
@@ -3284,10 +3304,11 @@ sub config_bisect {
 
     if (!defined($config_bisect_exec)) {
 	# First check the location that ktest.pl ran
-	my @locations = ( "$pwd/config-bisect.pl",
-			  "$dirname/config-bisect.pl",
-			  "$builddir/tools/testing/ktest/config-bisect.pl",
-			  undef );
+	my @locations = (
+		"$pwd/config-bisect.pl",
+		"$dirname/config-bisect.pl",
+		"$builddir/tools/testing/ktest/config-bisect.pl",
+		undef );
 	foreach my $loc (@locations) {
 	    doprint "loc = $loc\n";
 	    $config_bisect_exec = $loc;
@@ -3368,7 +3389,7 @@ sub config_bisect {
     } while ($ret == 3 || $ret == 4);
 
     if ($ret == 2) {
-        config_bisect_end "$good_config.tmp", "$bad_config.tmp";
+	config_bisect_end "$good_config.tmp", "$bad_config.tmp";
     }
 
     return $ret if ($ret < 0);
@@ -3511,14 +3532,6 @@ sub patchcheck {
     return 1;
 }
 
-my %depends;
-my %depcount;
-my $iflevel = 0;
-my @ifdeps;
-
-# prevent recursion
-my %read_kconfigs;
-
 sub add_dep {
     # $config depends on $dep
     my ($config, $dep) = @_;
@@ -3548,7 +3561,6 @@ sub read_kconfig {
     my $cont = 0;
     my $line;
 
-
     if (! -f $kconfig) {
 	doprint "file $kconfig does not exist, skipping\n";
 	return;
@@ -3630,8 +3642,8 @@ sub read_kconfig {
 
 sub read_depends {
     # find out which arch this is by the kconfig file
-    open (IN, $output_config)
-	or dodie "Failed to read $output_config";
+    open (IN, $output_config) or
+	dodie "Failed to read $output_config";
     my $arch;
     while (<IN>) {
 	if (m,Linux/(\S+)\s+\S+\s+Kernel Configuration,) {
@@ -3657,7 +3669,7 @@ sub read_depends {
 
     if (! -f $kconfig && $arch =~ /\d$/) {
 	my $orig = $arch;
- 	# some subarchs have numbers, truncate them
+	# some subarchs have numbers, truncate them
 	$arch =~ s/\d*$//;
 	$kconfig = "$builddir/arch/$arch/Kconfig";
 	if (! -f $kconfig) {
@@ -3706,7 +3718,6 @@ sub get_depends {
     my @configs;
 
     while ($dep =~ /[$valid]/) {
-
 	if ($dep =~ /^[^$valid]*([$valid]+)/) {
 	    my $conf = "CONFIG_" . $1;
 
@@ -3721,12 +3732,6 @@ sub get_depends {
     return @configs;
 }
 
-my %min_configs;
-my %keep_configs;
-my %save_configs;
-my %processed_configs;
-my %nochange_config;
-
 sub test_this_config {
     my ($config) = @_;
 
@@ -3852,7 +3857,7 @@ sub make_min_config {
     foreach my $config (@config_keys) {
 	my $kconfig = chomp_config $config;
 	if (!defined $depcount{$kconfig}) {
-		$depcount{$kconfig} = 0;
+	    $depcount{$kconfig} = 0;
 	}
     }
 
@@ -3887,7 +3892,6 @@ sub make_min_config {
     my $take_two = 0;
 
     while (!$done) {
-
 	my $config;
 	my $found;
 
@@ -3898,7 +3902,7 @@ sub make_min_config {
 
 	# Sort keys by who is most dependent on
 	@test_configs = sort  { $depcount{chomp_config($b)} <=> $depcount{chomp_config($a)} }
-			  @test_configs ;
+	    @test_configs ;
 
 	# Put configs that did not modify the config at the end.
 	my $reset = 1;
@@ -3954,13 +3958,13 @@ sub make_min_config {
 	my $failed = 0;
 	build "oldconfig" or $failed = 1;
 	if (!$failed) {
-		start_monitor_and_install or $failed = 1;
+	    start_monitor_and_install or $failed = 1;
 
-		if ($type eq "test" && !$failed) {
-		    do_run_test or $failed = 1;
-		}
+	    if ($type eq "test" && !$failed) {
+		do_run_test or $failed = 1;
+	    }
 
-		end_monitor;
+	    end_monitor;
 	}
 
 	$in_bisect = 0;
@@ -3974,8 +3978,8 @@ sub make_min_config {
 
 	    # update new ignore configs
 	    if (defined($ignore_config)) {
-		open (OUT, ">$temp_config")
-		    or dodie "Can't write to $temp_config";
+		open (OUT, ">$temp_config") or
+		    dodie "Can't write to $temp_config";
 		foreach my $config (keys %save_configs) {
 		    print OUT "$save_configs{$config}\n";
 		}
@@ -4002,8 +4006,8 @@ sub make_min_config {
 	    }
 
 	    # Save off all the current mandatory configs
-	    open (OUT, ">$temp_config")
-		or dodie "Can't write to $temp_config";
+	    open (OUT, ">$temp_config") or
+		dodie "Can't write to $temp_config";
 	    foreach my $config (keys %keep_configs) {
 		print OUT "$keep_configs{$config}\n";
 	    }
@@ -4041,7 +4045,6 @@ sub make_warnings_file {
 
     open(IN, $buildlog) or dodie "Can't open $buildlog";
     while (<IN>) {
-
 	# Some compilers use UTF-8 extended for quotes
 	# for distcc heterogeneous systems, this causes issues
 	s/$utf8_quote/'/g;
@@ -4057,98 +4060,6 @@ sub make_warnings_file {
     success $i;
 }
 
-$#ARGV < 1 or die "ktest.pl version: $VERSION\n   usage: ktest.pl [config-file]\n";
-
-if ($#ARGV == 0) {
-    $ktest_config = $ARGV[0];
-    if (! -f $ktest_config) {
-	print "$ktest_config does not exist.\n";
-	if (!read_yn "Create it?") {
-	    exit 0;
-	}
-    }
-}
-
-if (! -f $ktest_config) {
-    $newconfig = 1;
-    get_test_case;
-    open(OUT, ">$ktest_config") or die "Can not create $ktest_config";
-    print OUT << "EOF"
-# Generated by ktest.pl
-#
-
-# PWD is a ktest.pl variable that will result in the process working
-# directory that ktest.pl is executed in.
-
-# THIS_DIR is automatically assigned the PWD of the path that generated
-# the config file. It is best to use this variable when assigning other
-# directory paths within this directory. This allows you to easily
-# move the test cases to other locations or to other machines.
-#
-THIS_DIR := $variable{"PWD"}
-
-# Define each test with TEST_START
-# The config options below it will override the defaults
-TEST_START
-TEST_TYPE = $default{"TEST_TYPE"}
-
-DEFAULTS
-EOF
-;
-    close(OUT);
-}
-read_config $ktest_config;
-
-if (defined($opt{"LOG_FILE"})) {
-    $opt{"LOG_FILE"} = eval_option("LOG_FILE", $opt{"LOG_FILE"}, -1);
-}
-
-# Append any configs entered in manually to the config file.
-my @new_configs = keys %entered_configs;
-if ($#new_configs >= 0) {
-    print "\nAppending entered in configs to $ktest_config\n";
-    open(OUT, ">>$ktest_config") or die "Can not append to $ktest_config";
-    foreach my $config (@new_configs) {
-	print OUT "$config = $entered_configs{$config}\n";
-	$opt{$config} = process_variables($entered_configs{$config});
-    }
-}
-
-if (defined($opt{"LOG_FILE"})) {
-    if ($opt{"CLEAR_LOG"}) {
-	unlink $opt{"LOG_FILE"};
-    }
-    open(LOG, ">> $opt{LOG_FILE}") or die "Can't write to $opt{LOG_FILE}";
-    LOG->autoflush(1);
-}
-
-doprint "\n\nSTARTING AUTOMATED TESTS\n\n";
-
-for (my $i = 0, my $repeat = 1; $i <= $opt{"NUM_TESTS"}; $i += $repeat) {
-
-    if (!$i) {
-	doprint "DEFAULT OPTIONS:\n";
-    } else {
-	doprint "\nTEST $i OPTIONS";
-	if (defined($repeat_tests{$i})) {
-	    $repeat = $repeat_tests{$i};
-	    doprint " ITERATE $repeat";
-	}
-	doprint "\n";
-    }
-
-    foreach my $option (sort keys %opt) {
-
-	if ($option =~ /\[(\d+)\]$/) {
-	    next if ($i != $1);
-	} else {
-	    next if ($i);
-	}
-
-	doprint "$option = $opt{$option}\n";
-    }
-}
-
 sub option_defined {
     my ($option) = @_;
 
@@ -4261,7 +4172,6 @@ sub do_send_mail {
 }
 
 sub send_email {
-
     if (defined($mailto)) {
 	if (!defined($mailer)) {
 	    doprint "No email sent: email or mailer not specified in config.\n";
@@ -4274,12 +4184,103 @@ sub send_email {
 sub cancel_test {
     if ($email_when_canceled) {
 	my $name = get_test_name;
-        send_email("KTEST: Your [$name] test was cancelled",
-                "Your test started at $script_start_time was cancelled: sig int");
+	send_email("KTEST: Your [$name] test was cancelled",
+	    "Your test started at $script_start_time was cancelled: sig int");
     }
     die "\nCaught Sig Int, test interrupted: $!\n"
 }
 
+$#ARGV < 1 or die "ktest.pl version: $VERSION\n   usage: ktest.pl [config-file]\n";
+
+if ($#ARGV == 0) {
+    $ktest_config = $ARGV[0];
+    if (! -f $ktest_config) {
+	print "$ktest_config does not exist.\n";
+	if (!read_yn "Create it?") {
+	    exit 0;
+	}
+    }
+}
+
+if (! -f $ktest_config) {
+    $newconfig = 1;
+    get_test_case;
+    open(OUT, ">$ktest_config") or die "Can not create $ktest_config";
+    print OUT << "EOF"
+# Generated by ktest.pl
+#
+
+# PWD is a ktest.pl variable that will result in the process working
+# directory that ktest.pl is executed in.
+
+# THIS_DIR is automatically assigned the PWD of the path that generated
+# the config file. It is best to use this variable when assigning other
+# directory paths within this directory. This allows you to easily
+# move the test cases to other locations or to other machines.
+#
+THIS_DIR := $variable{"PWD"}
+
+# Define each test with TEST_START
+# The config options below it will override the defaults
+TEST_START
+TEST_TYPE = $default{"TEST_TYPE"}
+
+DEFAULTS
+EOF
+;
+    close(OUT);
+}
+read_config $ktest_config;
+
+if (defined($opt{"LOG_FILE"})) {
+    $opt{"LOG_FILE"} = eval_option("LOG_FILE", $opt{"LOG_FILE"}, -1);
+}
+
+# Append any configs entered in manually to the config file.
+my @new_configs = keys %entered_configs;
+if ($#new_configs >= 0) {
+    print "\nAppending entered in configs to $ktest_config\n";
+    open(OUT, ">>$ktest_config") or die "Can not append to $ktest_config";
+    foreach my $config (@new_configs) {
+	print OUT "$config = $entered_configs{$config}\n";
+	$opt{$config} = process_variables($entered_configs{$config});
+    }
+}
+
+if (defined($opt{"LOG_FILE"})) {
+    if ($opt{"CLEAR_LOG"}) {
+	unlink $opt{"LOG_FILE"};
+    }
+    open(LOG, ">> $opt{LOG_FILE}") or die "Can't write to $opt{LOG_FILE}";
+    LOG->autoflush(1);
+}
+
+doprint "\n\nSTARTING AUTOMATED TESTS\n\n";
+
+for (my $i = 0, my $repeat = 1; $i <= $opt{"NUM_TESTS"}; $i += $repeat) {
+
+    if (!$i) {
+	doprint "DEFAULT OPTIONS:\n";
+    } else {
+	doprint "\nTEST $i OPTIONS";
+	if (defined($repeat_tests{$i})) {
+	    $repeat = $repeat_tests{$i};
+	    doprint " ITERATE $repeat";
+	}
+	doprint "\n";
+    }
+
+    foreach my $option (sort keys %opt) {
+	if ($option =~ /\[(\d+)\]$/) {
+	    next if ($i != $1);
+	} else {
+	    next if ($i);
+	}
+
+	doprint "$option = $opt{$option}\n";
+    }
+}
+
 $SIG{INT} = qw(cancel_test);
 
 # First we need to do is the builds
@@ -4323,15 +4324,15 @@ for (my $i = 1; $i <= $opt{"NUM_TESTS"}; $i++) {
 
     # The first test may override the PRE_KTEST option
     if ($i == 1) {
-        if (defined($pre_ktest)) {
-            doprint "\n";
-            run_command $pre_ktest;
-        }
-        if ($email_when_started) {
+	if (defined($pre_ktest)) {
+	    doprint "\n";
+	    run_command $pre_ktest;
+	}
+	if ($email_when_started) {
 	    my $name = get_test_name;
-            send_email("KTEST: Your [$name] test was started",
-                "Your test was started on $script_start_time");
-        }
+	    send_email("KTEST: Your [$name] test was started",
+		"Your test was started on $script_start_time");
+	}
     }
 
     # Any test can override the POST_KTEST option
@@ -4409,7 +4410,7 @@ for (my $i = 1; $i <= $opt{"NUM_TESTS"}; $i++) {
 	my $ret = run_command $pre_test;
 	if (!$ret && defined($pre_test_die) &&
 	    $pre_test_die) {
-	    dodie "failed to pre_test\n";
+		dodie "failed to pre_test\n";
 	}
     }
 
@@ -4503,12 +4504,11 @@ if ($opt{"POWEROFF_ON_SUCCESS"}) {
     run_command $switch_to_good;
 }
 
-
 doprint "\n    $successes of $opt{NUM_TESTS} tests were successful\n\n";
 
 if ($email_when_finished) {
     send_email("KTEST: Your test has finished!",
-            "$successes of $opt{NUM_TESTS} tests started at $script_start_time were successful!");
+	"$successes of $opt{NUM_TESTS} tests started at $script_start_time were successful!");
 }
 
 if (defined($opt{"LOG_FILE"})) {
@@ -4517,3 +4517,12 @@ if (defined($opt{"LOG_FILE"})) {
 }
 
 exit 0;
+
+##
+# The following are here to standardize tabs/spaces/etc across the most likely editors
+###
+
+# Local Variables:
+# mode: perl
+# End:
+# vim: softtabstop=4
diff --git a/tools/testing/nvdimm/test/iomap.c b/tools/testing/nvdimm/test/iomap.c
index c62d372d426f..ed563bdd88f3 100644
--- a/tools/testing/nvdimm/test/iomap.c
+++ b/tools/testing/nvdimm/test/iomap.c
@@ -62,7 +62,7 @@ struct nfit_test_resource *get_nfit_res(resource_size_t resource)
 }
 EXPORT_SYMBOL(get_nfit_res);
 
-void __iomem *__nfit_test_ioremap(resource_size_t offset, unsigned long size,
+static void __iomem *__nfit_test_ioremap(resource_size_t offset, unsigned long size,
 		void __iomem *(*fallback_fn)(resource_size_t, unsigned long))
 {
 	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 9b185bf82da8..54f367cbadae 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -1871,9 +1871,16 @@ static void smart_init(struct nfit_test *t)
 	}
 }
 
+static size_t sizeof_spa(struct acpi_nfit_system_address *spa)
+{
+	/* until spa location cookie support is added... */
+	return sizeof(*spa) - 8;
+}
+
 static int nfit_test0_alloc(struct nfit_test *t)
 {
-	size_t nfit_size = sizeof(struct acpi_nfit_system_address) * NUM_SPA
+	struct acpi_nfit_system_address *spa = NULL;
+	size_t nfit_size = sizeof_spa(spa) * NUM_SPA
 			+ sizeof(struct acpi_nfit_memory_map) * NUM_MEM
 			+ sizeof(struct acpi_nfit_control_region) * NUM_DCR
 			+ offsetof(struct acpi_nfit_control_region,
@@ -1937,7 +1944,8 @@ static int nfit_test0_alloc(struct nfit_test *t)
 
 static int nfit_test1_alloc(struct nfit_test *t)
 {
-	size_t nfit_size = sizeof(struct acpi_nfit_system_address) * 2
+	struct acpi_nfit_system_address *spa = NULL;
+	size_t nfit_size = sizeof_spa(spa) * 2
 		+ sizeof(struct acpi_nfit_memory_map) * 2
 		+ offsetof(struct acpi_nfit_control_region, window_size) * 2;
 	int i;
@@ -2000,7 +2008,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	 */
 	spa = nfit_buf;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 0+1;
 	spa->address = t->spa_set_dma[0];
@@ -2014,7 +2022,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 1+1;
 	spa->address = t->spa_set_dma[1];
@@ -2024,7 +2032,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa2 (dcr0) dimm0 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 2+1;
 	spa->address = t->dcr_dma[0];
@@ -2034,7 +2042,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa3 (dcr1) dimm1 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 3+1;
 	spa->address = t->dcr_dma[1];
@@ -2044,7 +2052,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa4 (dcr2) dimm2 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 4+1;
 	spa->address = t->dcr_dma[2];
@@ -2054,7 +2062,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa5 (dcr3) dimm3 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 	spa->range_index = 5+1;
 	spa->address = t->dcr_dma[3];
@@ -2064,7 +2072,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa6 (bdw for dcr0) dimm0 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 6+1;
 	spa->address = t->dimm_dma[0];
@@ -2074,7 +2082,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa7 (bdw for dcr1) dimm1 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 7+1;
 	spa->address = t->dimm_dma[1];
@@ -2084,7 +2092,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa8 (bdw for dcr2) dimm2 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 8+1;
 	spa->address = t->dimm_dma[2];
@@ -2094,7 +2102,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	/* spa9 (bdw for dcr3) dimm3 */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 	spa->range_index = 9+1;
 	spa->address = t->dimm_dma[3];
@@ -2581,7 +2589,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		/* spa10 (dcr4) dimm4 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_DCR), 16);
 		spa->range_index = 10+1;
 		spa->address = t->dcr_dma[4];
@@ -2595,7 +2603,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 		spa->range_index = 11+1;
 		spa->address = t->spa_set_dma[2];
@@ -2605,7 +2613,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		/* spa12 (bdw for dcr4) dimm4 */
 		spa = nfit_buf + offset;
 		spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-		spa->header.length = sizeof(*spa);
+		spa->header.length = sizeof_spa(spa);
 		memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_BDW), 16);
 		spa->range_index = 12+1;
 		spa->address = t->dimm_dma[4];
@@ -2739,7 +2747,7 @@ static void nfit_test1_setup(struct nfit_test *t)
 	/* spa0 (flat range with no bdw aliasing) */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_PM), 16);
 	spa->range_index = 0+1;
 	spa->address = t->spa_set_dma[0];
@@ -2749,7 +2757,7 @@ static void nfit_test1_setup(struct nfit_test *t)
 	/* virtual cd region */
 	spa = nfit_buf + offset;
 	spa->header.type = ACPI_NFIT_TYPE_SYSTEM_ADDRESS;
-	spa->header.length = sizeof(*spa);
+	spa->header.length = sizeof_spa(spa);
 	memcpy(spa->range_guid, to_nfit_uuid(NFIT_SPA_VCD), 16);
 	spa->range_index = 0;
 	spa->address = t->spa_set_dma[1];
diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index 6c575cf34a71..bc3299a20338 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -25,6 +25,7 @@ TARGETS += ir
 TARGETS += kcmp
 TARGETS += kexec
 TARGETS += kvm
+TARGETS += landlock
 TARGETS += lib
 TARGETS += livepatch
 TARGETS += lkdtm
diff --git a/tools/testing/selftests/arm64/bti/test.c b/tools/testing/selftests/arm64/bti/test.c
index 656b04976ccc..67b77ab83c20 100644
--- a/tools/testing/selftests/arm64/bti/test.c
+++ b/tools/testing/selftests/arm64/bti/test.c
@@ -6,6 +6,7 @@
 
 #include "system.h"
 
+#include <stddef.h>
 #include <linux/errno.h>
 #include <linux/auxvec.h>
 #include <linux/signal.h>
diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 12ee40284da0..2060bc122c53 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -40,7 +40,7 @@ struct ipv6_packet pkt_v6 = {
 	.tcp.doff = 5,
 };
 
-static int settimeo(int fd, int timeout_ms)
+int settimeo(int fd, int timeout_ms)
 {
 	struct timeval timeout = { .tv_sec = 3 };
 
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 7205f8afdba1..5e0d51c07b63 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -33,6 +33,7 @@ struct ipv6_packet {
 } __packed;
 extern struct ipv6_packet pkt_v6;
 
+int settimeo(int fd, int timeout_ms);
 int start_server(int family, int type, const char *addr, __u16 port,
 		 int timeout_ms);
 int connect_to_fd(int server_fd, int timeout_ms);
diff --git a/tools/testing/selftests/bpf/prog_tests/ringbuf.c b/tools/testing/selftests/bpf/prog_tests/ringbuf.c
index de78617f6550..f9a8ae331963 100644
--- a/tools/testing/selftests/bpf/prog_tests/ringbuf.c
+++ b/tools/testing/selftests/bpf/prog_tests/ringbuf.c
@@ -86,8 +86,9 @@ void test_ringbuf(void)
 	const size_t rec_sz = BPF_RINGBUF_HDR_SZ + sizeof(struct sample);
 	pthread_t thread;
 	long bg_ret = -1;
-	int err, cnt;
+	int err, cnt, rb_fd;
 	int page_size = getpagesize();
+	void *mmap_ptr, *tmp_ptr;
 
 	skel = test_ringbuf__open();
 	if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
@@ -101,6 +102,52 @@ void test_ringbuf(void)
 	if (CHECK(err != 0, "skel_load", "skeleton load failed\n"))
 		goto cleanup;
 
+	rb_fd = bpf_map__fd(skel->maps.ringbuf);
+	/* good read/write cons_pos */
+	mmap_ptr = mmap(NULL, page_size, PROT_READ | PROT_WRITE, MAP_SHARED, rb_fd, 0);
+	ASSERT_OK_PTR(mmap_ptr, "rw_cons_pos");
+	tmp_ptr = mremap(mmap_ptr, page_size, 2 * page_size, MREMAP_MAYMOVE);
+	if (!ASSERT_ERR_PTR(tmp_ptr, "rw_extend"))
+		goto cleanup;
+	ASSERT_ERR(mprotect(mmap_ptr, page_size, PROT_EXEC), "exec_cons_pos_protect");
+	ASSERT_OK(munmap(mmap_ptr, page_size), "unmap_rw");
+
+	/* bad writeable prod_pos */
+	mmap_ptr = mmap(NULL, page_size, PROT_WRITE, MAP_SHARED, rb_fd, page_size);
+	err = -errno;
+	ASSERT_ERR_PTR(mmap_ptr, "wr_prod_pos");
+	ASSERT_EQ(err, -EPERM, "wr_prod_pos_err");
+
+	/* bad writeable data pages */
+	mmap_ptr = mmap(NULL, page_size, PROT_WRITE, MAP_SHARED, rb_fd, 2 * page_size);
+	err = -errno;
+	ASSERT_ERR_PTR(mmap_ptr, "wr_data_page_one");
+	ASSERT_EQ(err, -EPERM, "wr_data_page_one_err");
+	mmap_ptr = mmap(NULL, page_size, PROT_WRITE, MAP_SHARED, rb_fd, 3 * page_size);
+	ASSERT_ERR_PTR(mmap_ptr, "wr_data_page_two");
+	mmap_ptr = mmap(NULL, 2 * page_size, PROT_WRITE, MAP_SHARED, rb_fd, 2 * page_size);
+	ASSERT_ERR_PTR(mmap_ptr, "wr_data_page_all");
+
+	/* good read-only pages */
+	mmap_ptr = mmap(NULL, 4 * page_size, PROT_READ, MAP_SHARED, rb_fd, 0);
+	if (!ASSERT_OK_PTR(mmap_ptr, "ro_prod_pos"))
+		goto cleanup;
+
+	ASSERT_ERR(mprotect(mmap_ptr, 4 * page_size, PROT_WRITE), "write_protect");
+	ASSERT_ERR(mprotect(mmap_ptr, 4 * page_size, PROT_EXEC), "exec_protect");
+	ASSERT_ERR_PTR(mremap(mmap_ptr, 0, 4 * page_size, MREMAP_MAYMOVE), "ro_remap");
+	ASSERT_OK(munmap(mmap_ptr, 4 * page_size), "unmap_ro");
+
+	/* good read-only pages with initial offset */
+	mmap_ptr = mmap(NULL, page_size, PROT_READ, MAP_SHARED, rb_fd, page_size);
+	if (!ASSERT_OK_PTR(mmap_ptr, "ro_prod_pos"))
+		goto cleanup;
+
+	ASSERT_ERR(mprotect(mmap_ptr, page_size, PROT_WRITE), "write_protect");
+	ASSERT_ERR(mprotect(mmap_ptr, page_size, PROT_EXEC), "exec_protect");
+	ASSERT_ERR_PTR(mremap(mmap_ptr, 0, 3 * page_size, MREMAP_MAYMOVE), "ro_remap");
+	ASSERT_OK(munmap(mmap_ptr, page_size), "unmap_ro");
+
 	/* only trigger BPF program for current process */
 	skel->bss->pid = getpid();
 
diff --git a/tools/testing/selftests/bpf/prog_tests/snprintf.c b/tools/testing/selftests/bpf/prog_tests/snprintf.c
index a958c22aec75..dffbcaa1ec98 100644
--- a/tools/testing/selftests/bpf/prog_tests/snprintf.c
+++ b/tools/testing/selftests/bpf/prog_tests/snprintf.c
@@ -43,6 +43,8 @@ void test_snprintf_positive(void)
 	if (!ASSERT_OK_PTR(skel, "skel_open"))
 		return;
 
+	skel->bss->pid = getpid();
+
 	if (!ASSERT_OK(test_snprintf__attach(skel), "skel_attach"))
 		goto cleanup;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/tc_redirect.c b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
new file mode 100644
index 000000000000..5703c918812b
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
@@ -0,0 +1,785 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+
+/*
+ * This test sets up 3 netns (src <-> fwd <-> dst). There is no direct veth link
+ * between src and dst. The netns fwd has veth links to each src and dst. The
+ * client is in src and server in dst. The test installs a TC BPF program to each
+ * host facing veth in fwd which calls into i) bpf_redirect_neigh() to perform the
+ * neigh addr population and redirect or ii) bpf_redirect_peer() for namespace
+ * switch from ingress side; it also installs a checker prog on the egress side
+ * to drop unexpected traffic.
+ */
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <linux/limits.h>
+#include <linux/sysctl.h>
+#include <linux/if_tun.h>
+#include <linux/if.h>
+#include <sched.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <sys/stat.h>
+#include <sys/mount.h>
+
+#include "test_progs.h"
+#include "network_helpers.h"
+#include "test_tc_neigh_fib.skel.h"
+#include "test_tc_neigh.skel.h"
+#include "test_tc_peer.skel.h"
+
+#define NS_SRC "ns_src"
+#define NS_FWD "ns_fwd"
+#define NS_DST "ns_dst"
+
+#define IP4_SRC "172.16.1.100"
+#define IP4_DST "172.16.2.100"
+#define IP4_TUN_SRC "172.17.1.100"
+#define IP4_TUN_FWD "172.17.1.200"
+#define IP4_PORT 9004
+
+#define IP6_SRC "0::1:dead:beef:cafe"
+#define IP6_DST "0::2:dead:beef:cafe"
+#define IP6_TUN_SRC "1::1:dead:beef:cafe"
+#define IP6_TUN_FWD "1::2:dead:beef:cafe"
+#define IP6_PORT 9006
+
+#define IP4_SLL "169.254.0.1"
+#define IP4_DLL "169.254.0.2"
+#define IP4_NET "169.254.0.0"
+
+#define MAC_DST_FWD "00:11:22:33:44:55"
+#define MAC_DST "00:22:33:44:55:66"
+
+#define IFADDR_STR_LEN 18
+#define PING_ARGS "-i 0.2 -c 3 -w 10 -q"
+
+#define SRC_PROG_PIN_FILE "/sys/fs/bpf/test_tc_src"
+#define DST_PROG_PIN_FILE "/sys/fs/bpf/test_tc_dst"
+#define CHK_PROG_PIN_FILE "/sys/fs/bpf/test_tc_chk"
+
+#define TIMEOUT_MILLIS 10000
+
+#define log_err(MSG, ...) \
+	fprintf(stderr, "(%s:%d: errno: %s) " MSG "\n", \
+		__FILE__, __LINE__, strerror(errno), ##__VA_ARGS__)
+
+static const char * const namespaces[] = {NS_SRC, NS_FWD, NS_DST, NULL};
+
+static int write_file(const char *path, const char *newval)
+{
+	FILE *f;
+
+	f = fopen(path, "r+");
+	if (!f)
+		return -1;
+	if (fwrite(newval, strlen(newval), 1, f) != 1) {
+		log_err("writing to %s failed", path);
+		fclose(f);
+		return -1;
+	}
+	fclose(f);
+	return 0;
+}
+
+struct nstoken {
+	int orig_netns_fd;
+};
+
+static int setns_by_fd(int nsfd)
+{
+	int err;
+
+	err = setns(nsfd, CLONE_NEWNET);
+	close(nsfd);
+
+	if (!ASSERT_OK(err, "setns"))
+		return err;
+
+	/* Switch /sys to the new namespace so that e.g. /sys/class/net
+	 * reflects the devices in the new namespace.
+	 */
+	err = unshare(CLONE_NEWNS);
+	if (!ASSERT_OK(err, "unshare"))
+		return err;
+
+	err = umount2("/sys", MNT_DETACH);
+	if (!ASSERT_OK(err, "umount2 /sys"))
+		return err;
+
+	err = mount("sysfs", "/sys", "sysfs", 0, NULL);
+	if (!ASSERT_OK(err, "mount /sys"))
+		return err;
+
+	err = mount("bpffs", "/sys/fs/bpf", "bpf", 0, NULL);
+	if (!ASSERT_OK(err, "mount /sys/fs/bpf"))
+		return err;
+
+	return 0;
+}
+
+/**
+ * open_netns() - Switch to specified network namespace by name.
+ *
+ * Returns token with which to restore the original namespace
+ * using close_netns().
+ */
+static struct nstoken *open_netns(const char *name)
+{
+	int nsfd;
+	char nspath[PATH_MAX];
+	int err;
+	struct nstoken *token;
+
+	token = malloc(sizeof(struct nstoken));
+	if (!ASSERT_OK_PTR(token, "malloc token"))
+		return NULL;
+
+	token->orig_netns_fd = open("/proc/self/ns/net", O_RDONLY);
+	if (!ASSERT_GE(token->orig_netns_fd, 0, "open /proc/self/ns/net"))
+		goto fail;
+
+	snprintf(nspath, sizeof(nspath), "%s/%s", "/var/run/netns", name);
+	nsfd = open(nspath, O_RDONLY | O_CLOEXEC);
+	if (!ASSERT_GE(nsfd, 0, "open netns fd"))
+		goto fail;
+
+	err = setns_by_fd(nsfd);
+	if (!ASSERT_OK(err, "setns_by_fd"))
+		goto fail;
+
+	return token;
+fail:
+	free(token);
+	return NULL;
+}
+
+static void close_netns(struct nstoken *token)
+{
+	ASSERT_OK(setns_by_fd(token->orig_netns_fd), "setns_by_fd");
+	free(token);
+}
+
+static int netns_setup_namespaces(const char *verb)
+{
+	const char * const *ns = namespaces;
+	char cmd[128];
+
+	while (*ns) {
+		snprintf(cmd, sizeof(cmd), "ip netns %s %s", verb, *ns);
+		if (!ASSERT_OK(system(cmd), cmd))
+			return -1;
+		ns++;
+	}
+	return 0;
+}
+
+struct netns_setup_result {
+	int ifindex_veth_src_fwd;
+	int ifindex_veth_dst_fwd;
+};
+
+static int get_ifaddr(const char *name, char *ifaddr)
+{
+	char path[PATH_MAX];
+	FILE *f;
+	int ret;
+
+	snprintf(path, PATH_MAX, "/sys/class/net/%s/address", name);
+	f = fopen(path, "r");
+	if (!ASSERT_OK_PTR(f, path))
+		return -1;
+
+	ret = fread(ifaddr, 1, IFADDR_STR_LEN, f);
+	if (!ASSERT_EQ(ret, IFADDR_STR_LEN, "fread ifaddr")) {
+		fclose(f);
+		return -1;
+	}
+	fclose(f);
+	return 0;
+}
+
+static int get_ifindex(const char *name)
+{
+	char path[PATH_MAX];
+	char buf[32];
+	FILE *f;
+	int ret;
+
+	snprintf(path, PATH_MAX, "/sys/class/net/%s/ifindex", name);
+	f = fopen(path, "r");
+	if (!ASSERT_OK_PTR(f, path))
+		return -1;
+
+	ret = fread(buf, 1, sizeof(buf), f);
+	if (!ASSERT_GT(ret, 0, "fread ifindex")) {
+		fclose(f);
+		return -1;
+	}
+	fclose(f);
+	return atoi(buf);
+}
+
+#define SYS(fmt, ...)						\
+	({							\
+		char cmd[1024];					\
+		snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__);	\
+		if (!ASSERT_OK(system(cmd), cmd))		\
+			goto fail;				\
+	})
+
+static int netns_setup_links_and_routes(struct netns_setup_result *result)
+{
+	struct nstoken *nstoken = NULL;
+	char veth_src_fwd_addr[IFADDR_STR_LEN+1] = {};
+
+	SYS("ip link add veth_src type veth peer name veth_src_fwd");
+	SYS("ip link add veth_dst type veth peer name veth_dst_fwd");
+
+	SYS("ip link set veth_dst_fwd address " MAC_DST_FWD);
+	SYS("ip link set veth_dst address " MAC_DST);
+
+	if (get_ifaddr("veth_src_fwd", veth_src_fwd_addr))
+		goto fail;
+
+	result->ifindex_veth_src_fwd = get_ifindex("veth_src_fwd");
+	if (result->ifindex_veth_src_fwd < 0)
+		goto fail;
+	result->ifindex_veth_dst_fwd = get_ifindex("veth_dst_fwd");
+	if (result->ifindex_veth_dst_fwd < 0)
+		goto fail;
+
+	SYS("ip link set veth_src netns " NS_SRC);
+	SYS("ip link set veth_src_fwd netns " NS_FWD);
+	SYS("ip link set veth_dst_fwd netns " NS_FWD);
+	SYS("ip link set veth_dst netns " NS_DST);
+
+	/** setup in 'src' namespace */
+	nstoken = open_netns(NS_SRC);
+	if (!ASSERT_OK_PTR(nstoken, "setns src"))
+		goto fail;
+
+	SYS("ip addr add " IP4_SRC "/32 dev veth_src");
+	SYS("ip addr add " IP6_SRC "/128 dev veth_src nodad");
+	SYS("ip link set dev veth_src up");
+
+	SYS("ip route add " IP4_DST "/32 dev veth_src scope global");
+	SYS("ip route add " IP4_NET "/16 dev veth_src scope global");
+	SYS("ip route add " IP6_DST "/128 dev veth_src scope global");
+
+	SYS("ip neigh add " IP4_DST " dev veth_src lladdr %s",
+	    veth_src_fwd_addr);
+	SYS("ip neigh add " IP6_DST " dev veth_src lladdr %s",
+	    veth_src_fwd_addr);
+
+	close_netns(nstoken);
+
+	/** setup in 'fwd' namespace */
+	nstoken = open_netns(NS_FWD);
+	if (!ASSERT_OK_PTR(nstoken, "setns fwd"))
+		goto fail;
+
+	/* The fwd netns automatically gets a v6 LL address / routes, but also
+	 * needs v4 one in order to start ARP probing. IP4_NET route is added
+	 * to the endpoints so that the ARP processing will reply.
+	 */
+	SYS("ip addr add " IP4_SLL "/32 dev veth_src_fwd");
+	SYS("ip addr add " IP4_DLL "/32 dev veth_dst_fwd");
+	SYS("ip link set dev veth_src_fwd up");
+	SYS("ip link set dev veth_dst_fwd up");
+
+	SYS("ip route add " IP4_SRC "/32 dev veth_src_fwd scope global");
+	SYS("ip route add " IP6_SRC "/128 dev veth_src_fwd scope global");
+	SYS("ip route add " IP4_DST "/32 dev veth_dst_fwd scope global");
+	SYS("ip route add " IP6_DST "/128 dev veth_dst_fwd scope global");
+
+	close_netns(nstoken);
+
+	/** setup in 'dst' namespace */
+	nstoken = open_netns(NS_DST);
+	if (!ASSERT_OK_PTR(nstoken, "setns dst"))
+		goto fail;
+
+	SYS("ip addr add " IP4_DST "/32 dev veth_dst");
+	SYS("ip addr add " IP6_DST "/128 dev veth_dst nodad");
+	SYS("ip link set dev veth_dst up");
+
+	SYS("ip route add " IP4_SRC "/32 dev veth_dst scope global");
+	SYS("ip route add " IP4_NET "/16 dev veth_dst scope global");
+	SYS("ip route add " IP6_SRC "/128 dev veth_dst scope global");
+
+	SYS("ip neigh add " IP4_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS("ip neigh add " IP6_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+
+	close_netns(nstoken);
+
+	return 0;
+fail:
+	if (nstoken)
+		close_netns(nstoken);
+	return -1;
+}
+
+static int netns_load_bpf(void)
+{
+	SYS("tc qdisc add dev veth_src_fwd clsact");
+	SYS("tc filter add dev veth_src_fwd ingress bpf da object-pinned "
+	    SRC_PROG_PIN_FILE);
+	SYS("tc filter add dev veth_src_fwd egress bpf da object-pinned "
+	    CHK_PROG_PIN_FILE);
+
+	SYS("tc qdisc add dev veth_dst_fwd clsact");
+	SYS("tc filter add dev veth_dst_fwd ingress bpf da object-pinned "
+	    DST_PROG_PIN_FILE);
+	SYS("tc filter add dev veth_dst_fwd egress bpf da object-pinned "
+	    CHK_PROG_PIN_FILE);
+
+	return 0;
+fail:
+	return -1;
+}
+
+static void test_tcp(int family, const char *addr, __u16 port)
+{
+	int listen_fd = -1, accept_fd = -1, client_fd = -1;
+	char buf[] = "testing testing";
+	int n;
+	struct nstoken *nstoken;
+
+	nstoken = open_netns(NS_DST);
+	if (!ASSERT_OK_PTR(nstoken, "setns dst"))
+		return;
+
+	listen_fd = start_server(family, SOCK_STREAM, addr, port, 0);
+	if (!ASSERT_GE(listen_fd, 0, "listen"))
+		goto done;
+
+	close_netns(nstoken);
+	nstoken = open_netns(NS_SRC);
+	if (!ASSERT_OK_PTR(nstoken, "setns src"))
+		goto done;
+
+	client_fd = connect_to_fd(listen_fd, TIMEOUT_MILLIS);
+	if (!ASSERT_GE(client_fd, 0, "connect_to_fd"))
+		goto done;
+
+	accept_fd = accept(listen_fd, NULL, NULL);
+	if (!ASSERT_GE(accept_fd, 0, "accept"))
+		goto done;
+
+	if (!ASSERT_OK(settimeo(accept_fd, TIMEOUT_MILLIS), "settimeo"))
+		goto done;
+
+	n = write(client_fd, buf, sizeof(buf));
+	if (!ASSERT_EQ(n, sizeof(buf), "send to server"))
+		goto done;
+
+	n = read(accept_fd, buf, sizeof(buf));
+	ASSERT_EQ(n, sizeof(buf), "recv from server");
+
+done:
+	if (nstoken)
+		close_netns(nstoken);
+	if (listen_fd >= 0)
+		close(listen_fd);
+	if (accept_fd >= 0)
+		close(accept_fd);
+	if (client_fd >= 0)
+		close(client_fd);
+}
+
+static int test_ping(int family, const char *addr)
+{
+	const char *ping = family == AF_INET6 ? "ping6" : "ping";
+
+	SYS("ip netns exec " NS_SRC " %s " PING_ARGS " %s > /dev/null", ping, addr);
+	return 0;
+fail:
+	return -1;
+}
+
+static void test_connectivity(void)
+{
+	test_tcp(AF_INET, IP4_DST, IP4_PORT);
+	test_ping(AF_INET, IP4_DST);
+	test_tcp(AF_INET6, IP6_DST, IP6_PORT);
+	test_ping(AF_INET6, IP6_DST);
+}
+
+static int set_forwarding(bool enable)
+{
+	int err;
+
+	err = write_file("/proc/sys/net/ipv4/ip_forward", enable ? "1" : "0");
+	if (!ASSERT_OK(err, "set ipv4.ip_forward=0"))
+		return err;
+
+	err = write_file("/proc/sys/net/ipv6/conf/all/forwarding", enable ? "1" : "0");
+	if (!ASSERT_OK(err, "set ipv6.forwarding=0"))
+		return err;
+
+	return 0;
+}
+
+static void test_tc_redirect_neigh_fib(struct netns_setup_result *setup_result)
+{
+	struct nstoken *nstoken = NULL;
+	struct test_tc_neigh_fib *skel = NULL;
+	int err;
+
+	nstoken = open_netns(NS_FWD);
+	if (!ASSERT_OK_PTR(nstoken, "setns fwd"))
+		return;
+
+	skel = test_tc_neigh_fib__open();
+	if (!ASSERT_OK_PTR(skel, "test_tc_neigh_fib__open"))
+		goto done;
+
+	if (!ASSERT_OK(test_tc_neigh_fib__load(skel), "test_tc_neigh_fib__load"))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_src, SRC_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " SRC_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_chk, CHK_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " CHK_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_dst, DST_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " DST_PROG_PIN_FILE))
+		goto done;
+
+	if (netns_load_bpf())
+		goto done;
+
+	/* bpf_fib_lookup() checks if forwarding is enabled */
+	if (!ASSERT_OK(set_forwarding(true), "enable forwarding"))
+		goto done;
+
+	test_connectivity();
+
+done:
+	if (skel)
+		test_tc_neigh_fib__destroy(skel);
+	close_netns(nstoken);
+}
+
+static void test_tc_redirect_neigh(struct netns_setup_result *setup_result)
+{
+	struct nstoken *nstoken = NULL;
+	struct test_tc_neigh *skel = NULL;
+	int err;
+
+	nstoken = open_netns(NS_FWD);
+	if (!ASSERT_OK_PTR(nstoken, "setns fwd"))
+		return;
+
+	skel = test_tc_neigh__open();
+	if (!ASSERT_OK_PTR(skel, "test_tc_neigh__open"))
+		goto done;
+
+	skel->rodata->IFINDEX_SRC = setup_result->ifindex_veth_src_fwd;
+	skel->rodata->IFINDEX_DST = setup_result->ifindex_veth_dst_fwd;
+
+	err = test_tc_neigh__load(skel);
+	if (!ASSERT_OK(err, "test_tc_neigh__load"))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_src, SRC_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " SRC_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_chk, CHK_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " CHK_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_dst, DST_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " DST_PROG_PIN_FILE))
+		goto done;
+
+	if (netns_load_bpf())
+		goto done;
+
+	if (!ASSERT_OK(set_forwarding(false), "disable forwarding"))
+		goto done;
+
+	test_connectivity();
+
+done:
+	if (skel)
+		test_tc_neigh__destroy(skel);
+	close_netns(nstoken);
+}
+
+static void test_tc_redirect_peer(struct netns_setup_result *setup_result)
+{
+	struct nstoken *nstoken;
+	struct test_tc_peer *skel;
+	int err;
+
+	nstoken = open_netns(NS_FWD);
+	if (!ASSERT_OK_PTR(nstoken, "setns fwd"))
+		return;
+
+	skel = test_tc_peer__open();
+	if (!ASSERT_OK_PTR(skel, "test_tc_peer__open"))
+		goto done;
+
+	skel->rodata->IFINDEX_SRC = setup_result->ifindex_veth_src_fwd;
+	skel->rodata->IFINDEX_DST = setup_result->ifindex_veth_dst_fwd;
+
+	err = test_tc_peer__load(skel);
+	if (!ASSERT_OK(err, "test_tc_peer__load"))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_src, SRC_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " SRC_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_chk, CHK_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " CHK_PROG_PIN_FILE))
+		goto done;
+
+	err = bpf_program__pin(skel->progs.tc_dst, DST_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " DST_PROG_PIN_FILE))
+		goto done;
+
+	if (netns_load_bpf())
+		goto done;
+
+	if (!ASSERT_OK(set_forwarding(false), "disable forwarding"))
+		goto done;
+
+	test_connectivity();
+
+done:
+	if (skel)
+		test_tc_peer__destroy(skel);
+	close_netns(nstoken);
+}
+
+static int tun_open(char *name)
+{
+	struct ifreq ifr;
+	int fd, err;
+
+	fd = open("/dev/net/tun", O_RDWR);
+	if (!ASSERT_GE(fd, 0, "open /dev/net/tun"))
+		return -1;
+
+	memset(&ifr, 0, sizeof(ifr));
+
+	ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
+	if (*name)
+		strncpy(ifr.ifr_name, name, IFNAMSIZ);
+
+	err = ioctl(fd, TUNSETIFF, &ifr);
+	if (!ASSERT_OK(err, "ioctl TUNSETIFF"))
+		goto fail;
+
+	SYS("ip link set dev %s up", name);
+
+	return fd;
+fail:
+	close(fd);
+	return -1;
+}
+
+#define MAX(a, b) ((a) > (b) ? (a) : (b))
+enum {
+	SRC_TO_TARGET = 0,
+	TARGET_TO_SRC = 1,
+};
+
+static int tun_relay_loop(int src_fd, int target_fd)
+{
+	fd_set rfds, wfds;
+
+	FD_ZERO(&rfds);
+	FD_ZERO(&wfds);
+
+	for (;;) {
+		char buf[1500];
+		int direction, nread, nwrite;
+
+		FD_SET(src_fd, &rfds);
+		FD_SET(target_fd, &rfds);
+
+		if (select(1 + MAX(src_fd, target_fd), &rfds, NULL, NULL, NULL) < 0) {
+			log_err("select failed");
+			return 1;
+		}
+
+		direction = FD_ISSET(src_fd, &rfds) ? SRC_TO_TARGET : TARGET_TO_SRC;
+
+		nread = read(direction == SRC_TO_TARGET ? src_fd : target_fd, buf, sizeof(buf));
+		if (nread < 0) {
+			log_err("read failed");
+			return 1;
+		}
+
+		nwrite = write(direction == SRC_TO_TARGET ? target_fd : src_fd, buf, nread);
+		if (nwrite != nread) {
+			log_err("write failed");
+			return 1;
+		}
+	}
+}
+
+static void test_tc_redirect_peer_l3(struct netns_setup_result *setup_result)
+{
+	struct test_tc_peer *skel = NULL;
+	struct nstoken *nstoken = NULL;
+	int err;
+	int tunnel_pid = -1;
+	int src_fd, target_fd;
+	int ifindex;
+
+	/* Start a L3 TUN/TAP tunnel between the src and dst namespaces.
+	 * This test is using TUN/TAP instead of e.g. IPIP or GRE tunnel as those
+	 * expose the L2 headers encapsulating the IP packet to BPF and hence
+	 * don't have skb in suitable state for this test. Alternative to TUN/TAP
+	 * would be e.g. Wireguard which would appear as a pure L3 device to BPF,
+	 * but that requires much more complicated setup.
+	 */
+	nstoken = open_netns(NS_SRC);
+	if (!ASSERT_OK_PTR(nstoken, "setns " NS_SRC))
+		return;
+
+	src_fd = tun_open("tun_src");
+	if (!ASSERT_GE(src_fd, 0, "tun_open tun_src"))
+		goto fail;
+
+	close_netns(nstoken);
+
+	nstoken = open_netns(NS_FWD);
+	if (!ASSERT_OK_PTR(nstoken, "setns " NS_FWD))
+		goto fail;
+
+	target_fd = tun_open("tun_fwd");
+	if (!ASSERT_GE(target_fd, 0, "tun_open tun_fwd"))
+		goto fail;
+
+	tunnel_pid = fork();
+	if (!ASSERT_GE(tunnel_pid, 0, "fork tun_relay_loop"))
+		goto fail;
+
+	if (tunnel_pid == 0)
+		exit(tun_relay_loop(src_fd, target_fd));
+
+	skel = test_tc_peer__open();
+	if (!ASSERT_OK_PTR(skel, "test_tc_peer__open"))
+		goto fail;
+
+	ifindex = get_ifindex("tun_fwd");
+	if (!ASSERT_GE(ifindex, 0, "get_ifindex tun_fwd"))
+		goto fail;
+
+	skel->rodata->IFINDEX_SRC = ifindex;
+	skel->rodata->IFINDEX_DST = setup_result->ifindex_veth_dst_fwd;
+
+	err = test_tc_peer__load(skel);
+	if (!ASSERT_OK(err, "test_tc_peer__load"))
+		goto fail;
+
+	err = bpf_program__pin(skel->progs.tc_src_l3, SRC_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " SRC_PROG_PIN_FILE))
+		goto fail;
+
+	err = bpf_program__pin(skel->progs.tc_dst_l3, DST_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " DST_PROG_PIN_FILE))
+		goto fail;
+
+	err = bpf_program__pin(skel->progs.tc_chk, CHK_PROG_PIN_FILE);
+	if (!ASSERT_OK(err, "pin " CHK_PROG_PIN_FILE))
+		goto fail;
+
+	/* Load "tc_src_l3" to the tun_fwd interface to redirect packets
+	 * towards dst, and "tc_dst" to redirect packets
+	 * and "tc_chk" on veth_dst_fwd to drop non-redirected packets.
+	 */
+	SYS("tc qdisc add dev tun_fwd clsact");
+	SYS("tc filter add dev tun_fwd ingress bpf da object-pinned "
+	    SRC_PROG_PIN_FILE);
+
+	SYS("tc qdisc add dev veth_dst_fwd clsact");
+	SYS("tc filter add dev veth_dst_fwd ingress bpf da object-pinned "
+	    DST_PROG_PIN_FILE);
+	SYS("tc filter add dev veth_dst_fwd egress bpf da object-pinned "
+	    CHK_PROG_PIN_FILE);
+
+	/* Setup route and neigh tables */
+	SYS("ip -netns " NS_SRC " addr add dev tun_src " IP4_TUN_SRC "/24");
+	SYS("ip -netns " NS_FWD " addr add dev tun_fwd " IP4_TUN_FWD "/24");
+
+	SYS("ip -netns " NS_SRC " addr add dev tun_src " IP6_TUN_SRC "/64 nodad");
+	SYS("ip -netns " NS_FWD " addr add dev tun_fwd " IP6_TUN_FWD "/64 nodad");
+
+	SYS("ip -netns " NS_SRC " route del " IP4_DST "/32 dev veth_src scope global");
+	SYS("ip -netns " NS_SRC " route add " IP4_DST "/32 via " IP4_TUN_FWD
+	    " dev tun_src scope global");
+	SYS("ip -netns " NS_DST " route add " IP4_TUN_SRC "/32 dev veth_dst scope global");
+	SYS("ip -netns " NS_SRC " route del " IP6_DST "/128 dev veth_src scope global");
+	SYS("ip -netns " NS_SRC " route add " IP6_DST "/128 via " IP6_TUN_FWD
+	    " dev tun_src scope global");
+	SYS("ip -netns " NS_DST " route add " IP6_TUN_SRC "/128 dev veth_dst scope global");
+
+	SYS("ip -netns " NS_DST " neigh add " IP4_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+	SYS("ip -netns " NS_DST " neigh add " IP6_TUN_SRC " dev veth_dst lladdr " MAC_DST_FWD);
+
+	if (!ASSERT_OK(set_forwarding(false), "disable forwarding"))
+		goto fail;
+
+	test_connectivity();
+
+fail:
+	if (tunnel_pid > 0) {
+		kill(tunnel_pid, SIGTERM);
+		waitpid(tunnel_pid, NULL, 0);
+	}
+	if (src_fd >= 0)
+		close(src_fd);
+	if (target_fd >= 0)
+		close(target_fd);
+	if (skel)
+		test_tc_peer__destroy(skel);
+	if (nstoken)
+		close_netns(nstoken);
+}
+
+#define RUN_TEST(name)                                                                      \
+	({                                                                                  \
+		struct netns_setup_result setup_result;                                     \
+		if (test__start_subtest(#name))                                             \
+			if (ASSERT_OK(netns_setup_namespaces("add"), "setup namespaces")) { \
+				if (ASSERT_OK(netns_setup_links_and_routes(&setup_result),  \
+					      "setup links and routes"))                    \
+					test_ ## name(&setup_result);                       \
+				netns_setup_namespaces("delete");                           \
+			}                                                                   \
+	})
+
+static void *test_tc_redirect_run_tests(void *arg)
+{
+	RUN_TEST(tc_redirect_peer);
+	RUN_TEST(tc_redirect_peer_l3);
+	RUN_TEST(tc_redirect_neigh);
+	RUN_TEST(tc_redirect_neigh_fib);
+	return NULL;
+}
+
+void test_tc_redirect(void)
+{
+	pthread_t test_thread;
+	int err;
+
+	/* Run the tests in their own thread to isolate the namespace changes
+	 * so they do not affect the environment of other tests.
+	 * (specifically needed because of unshare(CLONE_NEWNS) in open_netns())
+	 */
+	err = pthread_create(&test_thread, NULL, &test_tc_redirect_run_tests, NULL);
+	if (ASSERT_OK(err, "pthread_create"))
+		ASSERT_OK(pthread_join(test_thread, NULL), "pthread_join");
+}
diff --git a/tools/testing/selftests/bpf/progs/test_snprintf.c b/tools/testing/selftests/bpf/progs/test_snprintf.c
index 951a0301c553..e35129bea0a0 100644
--- a/tools/testing/selftests/bpf/progs/test_snprintf.c
+++ b/tools/testing/selftests/bpf/progs/test_snprintf.c
@@ -5,6 +5,8 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
+__u32 pid = 0;
+
 char num_out[64] = {};
 long num_ret = 0;
 
@@ -42,6 +44,9 @@ int handler(const void *ctx)
 	static const char str1[] = "str1";
 	static const char longstr[] = "longstr";
 
+	if ((int)bpf_get_current_pid_tgid() != pid)
+		return 0;
+
 	/* Integer types */
 	num_ret  = BPF_SNPRINTF(num_out, sizeof(num_out),
 				"%d %u %x %li %llu %lX",
diff --git a/tools/testing/selftests/bpf/progs/test_tc_neigh.c b/tools/testing/selftests/bpf/progs/test_tc_neigh.c
index b985ac4e7a81..0c93d326a663 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_neigh.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_neigh.c
@@ -33,17 +33,8 @@
 				 a.s6_addr32[3] == b.s6_addr32[3])
 #endif
 
-enum {
-	dev_src,
-	dev_dst,
-};
-
-struct bpf_map_def SEC("maps") ifindex_map = {
-	.type		= BPF_MAP_TYPE_ARRAY,
-	.key_size	= sizeof(int),
-	.value_size	= sizeof(int),
-	.max_entries	= 2,
-};
+volatile const __u32 IFINDEX_SRC;
+volatile const __u32 IFINDEX_DST;
 
 static __always_inline bool is_remote_ep_v4(struct __sk_buff *skb,
 					    __be32 addr)
@@ -79,14 +70,8 @@ static __always_inline bool is_remote_ep_v6(struct __sk_buff *skb,
 	return v6_equal(ip6h->daddr, addr);
 }
 
-static __always_inline int get_dev_ifindex(int which)
-{
-	int *ifindex = bpf_map_lookup_elem(&ifindex_map, &which);
-
-	return ifindex ? *ifindex : 0;
-}
-
-SEC("chk_egress") int tc_chk(struct __sk_buff *skb)
+SEC("classifier/chk_egress")
+int tc_chk(struct __sk_buff *skb)
 {
 	void *data_end = ctx_ptr(skb->data_end);
 	void *data = ctx_ptr(skb->data);
@@ -98,7 +83,8 @@ SEC("chk_egress") int tc_chk(struct __sk_buff *skb)
 	return !raw[0] && !raw[1] && !raw[2] ? TC_ACT_SHOT : TC_ACT_OK;
 }
 
-SEC("dst_ingress") int tc_dst(struct __sk_buff *skb)
+SEC("classifier/dst_ingress")
+int tc_dst(struct __sk_buff *skb)
 {
 	__u8 zero[ETH_ALEN * 2];
 	bool redirect = false;
@@ -119,10 +105,11 @@ SEC("dst_ingress") int tc_dst(struct __sk_buff *skb)
 	if (bpf_skb_store_bytes(skb, 0, &zero, sizeof(zero), 0) < 0)
 		return TC_ACT_SHOT;
 
-	return bpf_redirect_neigh(get_dev_ifindex(dev_src), NULL, 0, 0);
+	return bpf_redirect_neigh(IFINDEX_SRC, NULL, 0, 0);
 }
 
-SEC("src_ingress") int tc_src(struct __sk_buff *skb)
+SEC("classifier/src_ingress")
+int tc_src(struct __sk_buff *skb)
 {
 	__u8 zero[ETH_ALEN * 2];
 	bool redirect = false;
@@ -143,7 +130,7 @@ SEC("src_ingress") int tc_src(struct __sk_buff *skb)
 	if (bpf_skb_store_bytes(skb, 0, &zero, sizeof(zero), 0) < 0)
 		return TC_ACT_SHOT;
 
-	return bpf_redirect_neigh(get_dev_ifindex(dev_dst), NULL, 0, 0);
+	return bpf_redirect_neigh(IFINDEX_DST, NULL, 0, 0);
 }
 
 char __license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c b/tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c
index d82ed3457030..f7ab69cf018e 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c
@@ -75,7 +75,8 @@ static __always_inline int fill_fib_params_v6(struct __sk_buff *skb,
 	return 0;
 }
 
-SEC("chk_egress") int tc_chk(struct __sk_buff *skb)
+SEC("classifier/chk_egress")
+int tc_chk(struct __sk_buff *skb)
 {
 	void *data_end = ctx_ptr(skb->data_end);
 	void *data = ctx_ptr(skb->data);
@@ -142,12 +143,14 @@ static __always_inline int tc_redir(struct __sk_buff *skb)
 /* these are identical, but keep them separate for compatibility with the
  * section names expected by test_tc_redirect.sh
  */
-SEC("dst_ingress") int tc_dst(struct __sk_buff *skb)
+SEC("classifier/dst_ingress")
+int tc_dst(struct __sk_buff *skb)
 {
 	return tc_redir(skb);
 }
 
-SEC("src_ingress") int tc_src(struct __sk_buff *skb)
+SEC("classifier/src_ingress")
+int tc_src(struct __sk_buff *skb)
 {
 	return tc_redir(skb);
 }
diff --git a/tools/testing/selftests/bpf/progs/test_tc_peer.c b/tools/testing/selftests/bpf/progs/test_tc_peer.c
index fc84a7685aa2..fe818cd5f010 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_peer.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_peer.c
@@ -5,41 +5,59 @@
 #include <linux/bpf.h>
 #include <linux/stddef.h>
 #include <linux/pkt_cls.h>
+#include <linux/if_ether.h>
+#include <linux/ip.h>
 
 #include <bpf/bpf_helpers.h>
 
-enum {
-	dev_src,
-	dev_dst,
-};
+volatile const __u32 IFINDEX_SRC;
+volatile const __u32 IFINDEX_DST;
 
-struct bpf_map_def SEC("maps") ifindex_map = {
-	.type		= BPF_MAP_TYPE_ARRAY,
-	.key_size	= sizeof(int),
-	.value_size	= sizeof(int),
-	.max_entries	= 2,
-};
+static const __u8 src_mac[] = {0x00, 0x11, 0x22, 0x33, 0x44, 0x55};
+static const __u8 dst_mac[] = {0x00, 0x22, 0x33, 0x44, 0x55, 0x66};
 
-static __always_inline int get_dev_ifindex(int which)
+SEC("classifier/chk_egress")
+int tc_chk(struct __sk_buff *skb)
 {
-	int *ifindex = bpf_map_lookup_elem(&ifindex_map, &which);
+	return TC_ACT_SHOT;
+}
 
-	return ifindex ? *ifindex : 0;
+SEC("classifier/dst_ingress")
+int tc_dst(struct __sk_buff *skb)
+{
+	return bpf_redirect_peer(IFINDEX_SRC, 0);
 }
 
-SEC("chk_egress") int tc_chk(struct __sk_buff *skb)
+SEC("classifier/src_ingress")
+int tc_src(struct __sk_buff *skb)
 {
-	return TC_ACT_SHOT;
+	return bpf_redirect_peer(IFINDEX_DST, 0);
 }
 
-SEC("dst_ingress") int tc_dst(struct __sk_buff *skb)
+SEC("classifier/dst_ingress_l3")
+int tc_dst_l3(struct __sk_buff *skb)
 {
-	return bpf_redirect_peer(get_dev_ifindex(dev_src), 0);
+	return bpf_redirect(IFINDEX_SRC, 0);
 }
 
-SEC("src_ingress") int tc_src(struct __sk_buff *skb)
+SEC("classifier/src_ingress_l3")
+int tc_src_l3(struct __sk_buff *skb)
 {
-	return bpf_redirect_peer(get_dev_ifindex(dev_dst), 0);
+	__u16 proto = skb->protocol;
+
+	if (bpf_skb_change_head(skb, ETH_HLEN, 0) != 0)
+		return TC_ACT_SHOT;
+
+	if (bpf_skb_store_bytes(skb, 0, &src_mac, ETH_ALEN, 0) != 0)
+		return TC_ACT_SHOT;
+
+	if (bpf_skb_store_bytes(skb, ETH_ALEN, &dst_mac, ETH_ALEN, 0) != 0)
+		return TC_ACT_SHOT;
+
+	if (bpf_skb_store_bytes(skb, ETH_ALEN + ETH_ALEN, &proto, sizeof(__u16), 0) != 0)
+		return TC_ACT_SHOT;
+
+	return bpf_redirect_peer(IFINDEX_DST, 0);
 }
 
 char __license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_tc_redirect.sh b/tools/testing/selftests/bpf/test_tc_redirect.sh
deleted file mode 100755
index 8868aa1ca902..000000000000
--- a/tools/testing/selftests/bpf/test_tc_redirect.sh
+++ /dev/null
@@ -1,216 +0,0 @@
-#!/bin/bash
-# SPDX-License-Identifier: GPL-2.0
-#
-# This test sets up 3 netns (src <-> fwd <-> dst). There is no direct veth link
-# between src and dst. The netns fwd has veth links to each src and dst. The
-# client is in src and server in dst. The test installs a TC BPF program to each
-# host facing veth in fwd which calls into i) bpf_redirect_neigh() to perform the
-# neigh addr population and redirect or ii) bpf_redirect_peer() for namespace
-# switch from ingress side; it also installs a checker prog on the egress side
-# to drop unexpected traffic.
-
-if [[ $EUID -ne 0 ]]; then
-	echo "This script must be run as root"
-	echo "FAIL"
-	exit 1
-fi
-
-# check that needed tools are present
-command -v nc >/dev/null 2>&1 || \
-	{ echo >&2 "nc is not available"; exit 1; }
-command -v dd >/dev/null 2>&1 || \
-	{ echo >&2 "dd is not available"; exit 1; }
-command -v timeout >/dev/null 2>&1 || \
-	{ echo >&2 "timeout is not available"; exit 1; }
-command -v ping >/dev/null 2>&1 || \
-	{ echo >&2 "ping is not available"; exit 1; }
-if command -v ping6 >/dev/null 2>&1; then PING6=ping6; else PING6=ping; fi
-command -v perl >/dev/null 2>&1 || \
-	{ echo >&2 "perl is not available"; exit 1; }
-command -v jq >/dev/null 2>&1 || \
-	{ echo >&2 "jq is not available"; exit 1; }
-command -v bpftool >/dev/null 2>&1 || \
-	{ echo >&2 "bpftool is not available"; exit 1; }
-
-readonly GREEN='\033[0;92m'
-readonly RED='\033[0;31m'
-readonly NC='\033[0m' # No Color
-
-readonly PING_ARG="-c 3 -w 10 -q"
-
-readonly TIMEOUT=10
-
-readonly NS_SRC="ns-src-$(mktemp -u XXXXXX)"
-readonly NS_FWD="ns-fwd-$(mktemp -u XXXXXX)"
-readonly NS_DST="ns-dst-$(mktemp -u XXXXXX)"
-
-readonly IP4_SRC="172.16.1.100"
-readonly IP4_DST="172.16.2.100"
-
-readonly IP6_SRC="::1:dead:beef:cafe"
-readonly IP6_DST="::2:dead:beef:cafe"
-
-readonly IP4_SLL="169.254.0.1"
-readonly IP4_DLL="169.254.0.2"
-readonly IP4_NET="169.254.0.0"
-
-netns_cleanup()
-{
-	ip netns del ${NS_SRC}
-	ip netns del ${NS_FWD}
-	ip netns del ${NS_DST}
-}
-
-netns_setup()
-{
-	ip netns add "${NS_SRC}"
-	ip netns add "${NS_FWD}"
-	ip netns add "${NS_DST}"
-
-	ip link add veth_src type veth peer name veth_src_fwd
-	ip link add veth_dst type veth peer name veth_dst_fwd
-
-	ip link set veth_src netns ${NS_SRC}
-	ip link set veth_src_fwd netns ${NS_FWD}
-
-	ip link set veth_dst netns ${NS_DST}
-	ip link set veth_dst_fwd netns ${NS_FWD}
-
-	ip -netns ${NS_SRC} addr add ${IP4_SRC}/32 dev veth_src
-	ip -netns ${NS_DST} addr add ${IP4_DST}/32 dev veth_dst
-
-	# The fwd netns automatically get a v6 LL address / routes, but also
-	# needs v4 one in order to start ARP probing. IP4_NET route is added
-	# to the endpoints so that the ARP processing will reply.
-
-	ip -netns ${NS_FWD} addr add ${IP4_SLL}/32 dev veth_src_fwd
-	ip -netns ${NS_FWD} addr add ${IP4_DLL}/32 dev veth_dst_fwd
-
-	ip -netns ${NS_SRC} addr add ${IP6_SRC}/128 dev veth_src nodad
-	ip -netns ${NS_DST} addr add ${IP6_DST}/128 dev veth_dst nodad
-
-	ip -netns ${NS_SRC} link set dev veth_src up
-	ip -netns ${NS_FWD} link set dev veth_src_fwd up
-
-	ip -netns ${NS_DST} link set dev veth_dst up
-	ip -netns ${NS_FWD} link set dev veth_dst_fwd up
-
-	ip -netns ${NS_SRC} route add ${IP4_DST}/32 dev veth_src scope global
-	ip -netns ${NS_SRC} route add ${IP4_NET}/16 dev veth_src scope global
-	ip -netns ${NS_FWD} route add ${IP4_SRC}/32 dev veth_src_fwd scope global
-
-	ip -netns ${NS_SRC} route add ${IP6_DST}/128 dev veth_src scope global
-	ip -netns ${NS_FWD} route add ${IP6_SRC}/128 dev veth_src_fwd scope global
-
-	ip -netns ${NS_DST} route add ${IP4_SRC}/32 dev veth_dst scope global
-	ip -netns ${NS_DST} route add ${IP4_NET}/16 dev veth_dst scope global
-	ip -netns ${NS_FWD} route add ${IP4_DST}/32 dev veth_dst_fwd scope global
-
-	ip -netns ${NS_DST} route add ${IP6_SRC}/128 dev veth_dst scope global
-	ip -netns ${NS_FWD} route add ${IP6_DST}/128 dev veth_dst_fwd scope global
-
-	fmac_src=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_src_fwd/address)
-	fmac_dst=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_dst_fwd/address)
-
-	ip -netns ${NS_SRC} neigh add ${IP4_DST} dev veth_src lladdr $fmac_src
-	ip -netns ${NS_DST} neigh add ${IP4_SRC} dev veth_dst lladdr $fmac_dst
-
-	ip -netns ${NS_SRC} neigh add ${IP6_DST} dev veth_src lladdr $fmac_src
-	ip -netns ${NS_DST} neigh add ${IP6_SRC} dev veth_dst lladdr $fmac_dst
-}
-
-netns_test_connectivity()
-{
-	set +e
-
-	ip netns exec ${NS_DST} bash -c "nc -4 -l -p 9004 &"
-	ip netns exec ${NS_DST} bash -c "nc -6 -l -p 9006 &"
-
-	TEST="TCPv4 connectivity test"
-	ip netns exec ${NS_SRC} bash -c "timeout ${TIMEOUT} dd if=/dev/zero bs=1000 count=100 > /dev/tcp/${IP4_DST}/9004"
-	if [ $? -ne 0 ]; then
-		echo -e "${TEST}: ${RED}FAIL${NC}"
-		exit 1
-	fi
-	echo -e "${TEST}: ${GREEN}PASS${NC}"
-
-	TEST="TCPv6 connectivity test"
-	ip netns exec ${NS_SRC} bash -c "timeout ${TIMEOUT} dd if=/dev/zero bs=1000 count=100 > /dev/tcp/${IP6_DST}/9006"
-	if [ $? -ne 0 ]; then
-		echo -e "${TEST}: ${RED}FAIL${NC}"
-		exit 1
-	fi
-	echo -e "${TEST}: ${GREEN}PASS${NC}"
-
-	TEST="ICMPv4 connectivity test"
-	ip netns exec ${NS_SRC} ping  $PING_ARG ${IP4_DST}
-	if [ $? -ne 0 ]; then
-		echo -e "${TEST}: ${RED}FAIL${NC}"
-		exit 1
-	fi
-	echo -e "${TEST}: ${GREEN}PASS${NC}"
-
-	TEST="ICMPv6 connectivity test"
-	ip netns exec ${NS_SRC} $PING6 $PING_ARG ${IP6_DST}
-	if [ $? -ne 0 ]; then
-		echo -e "${TEST}: ${RED}FAIL${NC}"
-		exit 1
-	fi
-	echo -e "${TEST}: ${GREEN}PASS${NC}"
-
-	set -e
-}
-
-hex_mem_str()
-{
-	perl -e 'print join(" ", unpack("(H2)8", pack("L", @ARGV)))' $1
-}
-
-netns_setup_bpf()
-{
-	local obj=$1
-	local use_forwarding=${2:-0}
-
-	ip netns exec ${NS_FWD} tc qdisc add dev veth_src_fwd clsact
-	ip netns exec ${NS_FWD} tc filter add dev veth_src_fwd ingress bpf da obj $obj sec src_ingress
-	ip netns exec ${NS_FWD} tc filter add dev veth_src_fwd egress  bpf da obj $obj sec chk_egress
-
-	ip netns exec ${NS_FWD} tc qdisc add dev veth_dst_fwd clsact
-	ip netns exec ${NS_FWD} tc filter add dev veth_dst_fwd ingress bpf da obj $obj sec dst_ingress
-	ip netns exec ${NS_FWD} tc filter add dev veth_dst_fwd egress  bpf da obj $obj sec chk_egress
-
-	if [ "$use_forwarding" -eq "1" ]; then
-		# bpf_fib_lookup() checks if forwarding is enabled
-		ip netns exec ${NS_FWD} sysctl -w net.ipv4.ip_forward=1
-		ip netns exec ${NS_FWD} sysctl -w net.ipv6.conf.veth_dst_fwd.forwarding=1
-		ip netns exec ${NS_FWD} sysctl -w net.ipv6.conf.veth_src_fwd.forwarding=1
-		return 0
-	fi
-
-	veth_src=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_src_fwd/ifindex)
-	veth_dst=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_dst_fwd/ifindex)
-
-	progs=$(ip netns exec ${NS_FWD} bpftool net --json | jq -r '.[] | .tc | map(.id) | .[]')
-	for prog in $progs; do
-		map=$(bpftool prog show id $prog --json | jq -r '.map_ids | .? | .[]')
-		if [ ! -z "$map" ]; then
-			bpftool map update id $map key hex $(hex_mem_str 0) value hex $(hex_mem_str $veth_src)
-			bpftool map update id $map key hex $(hex_mem_str 1) value hex $(hex_mem_str $veth_dst)
-		fi
-	done
-}
-
-trap netns_cleanup EXIT
-set -e
-
-netns_setup
-netns_setup_bpf test_tc_neigh.o
-netns_test_connectivity
-netns_cleanup
-netns_setup
-netns_setup_bpf test_tc_neigh_fib.o 1
-netns_test_connectivity
-netns_cleanup
-netns_setup
-netns_setup_bpf test_tc_peer.o
-netns_test_connectivity
diff --git a/tools/testing/selftests/bpf/verifier/stack_ptr.c b/tools/testing/selftests/bpf/verifier/stack_ptr.c
index 07eaa04412ae..8ab94d65f3d5 100644
--- a/tools/testing/selftests/bpf/verifier/stack_ptr.c
+++ b/tools/testing/selftests/bpf/verifier/stack_ptr.c
@@ -295,8 +295,6 @@
 	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_1, 0),
 	BPF_EXIT_INSN(),
 	},
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "invalid write to stack R1 off=0 size=1",
 	.result = ACCEPT,
 	.retval = 42,
 },
diff --git a/tools/testing/selftests/bpf/verifier/value_ptr_arith.c b/tools/testing/selftests/bpf/verifier/value_ptr_arith.c
index e5913fd3b903..7ae2859d495c 100644
--- a/tools/testing/selftests/bpf/verifier/value_ptr_arith.c
+++ b/tools/testing/selftests/bpf/verifier/value_ptr_arith.c
@@ -300,8 +300,6 @@
 	},
 	.fixup_map_array_48b = { 3 },
 	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
 	.retval = 1,
 },
 {
@@ -371,8 +369,6 @@
 	},
 	.fixup_map_array_48b = { 3 },
 	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
 	.retval = 1,
 },
 {
@@ -472,8 +468,6 @@
 	},
 	.fixup_map_array_48b = { 3 },
 	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
 	.retval = 1,
 },
 {
@@ -766,8 +760,6 @@
 	},
 	.fixup_map_array_48b = { 3 },
 	.result = ACCEPT,
-	.result_unpriv = REJECT,
-	.errstr_unpriv = "R0 pointer arithmetic of map value goes out of range",
 	.retval = 1,
 },
 {
diff --git a/tools/testing/selftests/cgroup/test_kmem.c b/tools/testing/selftests/cgroup/test_kmem.c
index 0941aa16157e..22b31ebb3513 100644
--- a/tools/testing/selftests/cgroup/test_kmem.c
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -19,12 +19,12 @@
 
 
 /*
- * Memory cgroup charging and vmstat data aggregation is performed using
- * percpu batches 32 pages big (look at MEMCG_CHARGE_BATCH). So the maximum
- * discrepancy between charge and vmstat entries is number of cpus multiplied
- * by 32 pages multiplied by 2.
+ * Memory cgroup charging is performed using percpu batches 32 pages
+ * big (look at MEMCG_CHARGE_BATCH), whereas memory.stat is exact. So
+ * the maximum discrepancy between charge and vmstat entries is number
+ * of cpus multiplied by 32 pages.
  */
-#define MAX_VMSTAT_ERROR (4096 * 32 * 2 * get_nprocs())
+#define MAX_VMSTAT_ERROR (4096 * 32 * get_nprocs())
 
 
 static int alloc_dcache(const char *cgroup, void *arg)
@@ -162,7 +162,7 @@ static int cg_run_in_subcgroups(const char *parent,
  */
 static int test_kmem_memcg_deletion(const char *root)
 {
-	long current, slab, anon, file, kernel_stack, sum;
+	long current, slab, anon, file, kernel_stack, pagetables, percpu, sock, sum;
 	int ret = KSFT_FAIL;
 	char *parent;
 
@@ -184,11 +184,14 @@ static int test_kmem_memcg_deletion(const char *root)
 	anon = cg_read_key_long(parent, "memory.stat", "anon ");
 	file = cg_read_key_long(parent, "memory.stat", "file ");
 	kernel_stack = cg_read_key_long(parent, "memory.stat", "kernel_stack ");
+	pagetables = cg_read_key_long(parent, "memory.stat", "pagetables ");
+	percpu = cg_read_key_long(parent, "memory.stat", "percpu ");
+	sock = cg_read_key_long(parent, "memory.stat", "sock ");
 	if (current < 0 || slab < 0 || anon < 0 || file < 0 ||
-	    kernel_stack < 0)
+	    kernel_stack < 0 || pagetables < 0 || percpu < 0 || sock < 0)
 		goto cleanup;
 
-	sum = slab + anon + file + kernel_stack;
+	sum = slab + anon + file + kernel_stack + pagetables + percpu + sock;
 	if (abs(sum - current) < MAX_VMSTAT_ERROR) {
 		ret = KSFT_PASS;
 	} else {
@@ -198,6 +201,9 @@ static int test_kmem_memcg_deletion(const char *root)
 		printf("anon = %ld\n", anon);
 		printf("file = %ld\n", file);
 		printf("kernel_stack = %ld\n", kernel_stack);
+		printf("pagetables = %ld\n", pagetables);
+		printf("percpu = %ld\n", percpu);
+		printf("sock = %ld\n", sock);
 	}
 
 cleanup:
diff --git a/tools/testing/selftests/dma/dma_map_benchmark.c b/tools/testing/selftests/dma/dma_map_benchmark.c
index fb23ce9617ea..485dff51bad2 100644
--- a/tools/testing/selftests/dma/dma_map_benchmark.c
+++ b/tools/testing/selftests/dma/dma_map_benchmark.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * Copyright (C) 2020 Hisilicon Limited.
+ * Copyright (C) 2020 HiSilicon Limited.
  */
 
 #include <fcntl.h>
@@ -40,7 +40,8 @@ struct map_benchmark {
 	__u32 dma_bits; /* DMA addressing capability */
 	__u32 dma_dir; /* DMA data direction */
 	__u32 dma_trans_ns; /* time for DMA transmission in ns */
-	__u8 expansion[80];	/* For future use */
+	__u32 granule; /* how many PAGE_SIZE will do map/unmap once a time */
+	__u8 expansion[76];	/* For future use */
 };
 
 int main(int argc, char **argv)
@@ -51,11 +52,13 @@ int main(int argc, char **argv)
 	int threads = 1, seconds = 20, node = -1;
 	/* default dma mask 32bit, bidirectional DMA */
 	int bits = 32, xdelay = 0, dir = DMA_MAP_BIDIRECTIONAL;
+	/* default granule 1 PAGESIZE */
+	int granule = 1;
 
 	int cmd = DMA_MAP_BENCHMARK;
 	char *p;
 
-	while ((opt = getopt(argc, argv, "t:s:n:b:d:x:")) != -1) {
+	while ((opt = getopt(argc, argv, "t:s:n:b:d:x:g:")) != -1) {
 		switch (opt) {
 		case 't':
 			threads = atoi(optarg);
@@ -75,6 +78,9 @@ int main(int argc, char **argv)
 		case 'x':
 			xdelay = atoi(optarg);
 			break;
+		case 'g':
+			granule = atoi(optarg);
+			break;
 		default:
 			return -1;
 		}
@@ -110,6 +116,11 @@ int main(int argc, char **argv)
 		exit(1);
 	}
 
+	if (granule < 1 || granule > 1024) {
+		fprintf(stderr, "invalid granule size\n");
+		exit(1);
+	}
+
 	fd = open("/sys/kernel/debug/dma_map_benchmark", O_RDWR);
 	if (fd == -1) {
 		perror("open");
@@ -123,14 +134,15 @@ int main(int argc, char **argv)
 	map.dma_bits = bits;
 	map.dma_dir = dir;
 	map.dma_trans_ns = xdelay;
+	map.granule = granule;
 
 	if (ioctl(fd, cmd, &map)) {
 		perror("ioctl");
 		exit(1);
 	}
 
-	printf("dma mapping benchmark: threads:%d seconds:%d node:%d dir:%s\n",
-			threads, seconds, node, dir[directions]);
+	printf("dma mapping benchmark: threads:%d seconds:%d node:%d dir:%s granule: %d\n",
+			threads, seconds, node, dir[directions], granule);
 	printf("average map latency(us):%.1f standard deviation:%.1f\n",
 			map.avg_map_100ns/10.0, map.map_stddev/10.0);
 	printf("average unmap latency(us):%.1f standard deviation:%.1f\n",
diff --git a/tools/testing/selftests/exec/Makefile b/tools/testing/selftests/exec/Makefile
index cf69b2fcce59..dd61118df66e 100644
--- a/tools/testing/selftests/exec/Makefile
+++ b/tools/testing/selftests/exec/Makefile
@@ -28,8 +28,8 @@ $(OUTPUT)/execveat.denatured: $(OUTPUT)/execveat
 	cp $< $@
 	chmod -x $@
 $(OUTPUT)/load_address_4096: load_address.c
-	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000 -pie $< -o $@
+	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000 -pie -static $< -o $@
 $(OUTPUT)/load_address_2097152: load_address.c
-	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x200000 -pie $< -o $@
+	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x200000 -pie -static $< -o $@
 $(OUTPUT)/load_address_16777216: load_address.c
-	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000000 -pie $< -o $@
+	$(CC) $(CFLAGS) $(LDFLAGS) -Wl,-z,max-page-size=0x1000000 -pie -static $< -o $@
diff --git a/tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c b/tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c
index ad7fabd575f9..65ede506305c 100644
--- a/tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c
+++ b/tools/testing/selftests/filesystems/epoll/epoll_wakeup_test.c
@@ -3449,4 +3449,48 @@ TEST(epoll63)
 	close(sfd[1]);
 }
 
+/*
+ *        t0    t1
+ *     (ew) \  / (ew)
+ *           e0
+ *            | (lt)
+ *           s0
+ */
+TEST(epoll64)
+{
+	pthread_t waiter[2];
+	struct epoll_event e;
+	struct epoll_mtcontext ctx = { 0 };
+
+	signal(SIGUSR1, signal_handler);
+
+	ASSERT_EQ(socketpair(AF_UNIX, SOCK_STREAM, 0, ctx.sfd), 0);
+
+	ctx.efd[0] = epoll_create(1);
+	ASSERT_GE(ctx.efd[0], 0);
+
+	e.events = EPOLLIN;
+	ASSERT_EQ(epoll_ctl(ctx.efd[0], EPOLL_CTL_ADD, ctx.sfd[0], &e), 0);
+
+	/*
+	 * main will act as the emitter once both waiter threads are
+	 * blocked and expects to both be awoken upon the ready event.
+	 */
+	ctx.main = pthread_self();
+	ASSERT_EQ(pthread_create(&waiter[0], NULL, waiter_entry1a, &ctx), 0);
+	ASSERT_EQ(pthread_create(&waiter[1], NULL, waiter_entry1a, &ctx), 0);
+
+	usleep(100000);
+	ASSERT_EQ(write(ctx.sfd[1], "w", 1), 1);
+
+	ASSERT_EQ(pthread_join(waiter[0], NULL), 0);
+	ASSERT_EQ(pthread_join(waiter[1], NULL), 0);
+
+	EXPECT_EQ(ctx.count, 2);
+
+	close(ctx.efd[0]);
+	close(ctx.sfd[0]);
+	close(ctx.sfd[1]);
+}
+
 TEST_HARNESS_MAIN
diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore
index 7bd7e776c266..524c857a049c 100644
--- a/tools/testing/selftests/kvm/.gitignore
+++ b/tools/testing/selftests/kvm/.gitignore
@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0-only
 /aarch64/get-reg-list
 /aarch64/get-reg-list-sve
+/aarch64/vgic_init
 /s390x/memop
 /s390x/resets
 /s390x/sync_regs_test
@@ -38,6 +39,8 @@
 /dirty_log_perf_test
 /hardware_disable_test
 /kvm_create_max_vcpus
+/kvm_page_table_test
 /memslot_modification_stress_test
+/memslot_perf_test
 /set_memory_region_test
 /steal_time
diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index cb95b5bace7b..daaee1888b12 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -33,7 +33,7 @@ ifeq ($(ARCH),s390)
 	UNAME_M := s390x
 endif
 
-LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/sparsebit.c lib/test_util.c lib/guest_modes.c lib/perf_test_util.c
+LIBKVM = lib/assert.c lib/elf.c lib/io.c lib/kvm_util.c lib/rbtree.c lib/sparsebit.c lib/test_util.c lib/guest_modes.c lib/perf_test_util.c
 LIBKVM_x86_64 = lib/x86_64/processor.c lib/x86_64/vmx.c lib/x86_64/svm.c lib/x86_64/ucall.c lib/x86_64/handlers.S
 LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c
 LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c
@@ -72,16 +72,20 @@ TEST_GEN_PROGS_x86_64 += dirty_log_test
 TEST_GEN_PROGS_x86_64 += dirty_log_perf_test
 TEST_GEN_PROGS_x86_64 += hardware_disable_test
 TEST_GEN_PROGS_x86_64 += kvm_create_max_vcpus
+TEST_GEN_PROGS_x86_64 += kvm_page_table_test
 TEST_GEN_PROGS_x86_64 += memslot_modification_stress_test
+TEST_GEN_PROGS_x86_64 += memslot_perf_test
 TEST_GEN_PROGS_x86_64 += set_memory_region_test
 TEST_GEN_PROGS_x86_64 += steal_time
 
 TEST_GEN_PROGS_aarch64 += aarch64/get-reg-list
 TEST_GEN_PROGS_aarch64 += aarch64/get-reg-list-sve
+TEST_GEN_PROGS_aarch64 += aarch64/vgic_init
 TEST_GEN_PROGS_aarch64 += demand_paging_test
 TEST_GEN_PROGS_aarch64 += dirty_log_test
 TEST_GEN_PROGS_aarch64 += dirty_log_perf_test
 TEST_GEN_PROGS_aarch64 += kvm_create_max_vcpus
+TEST_GEN_PROGS_aarch64 += kvm_page_table_test
 TEST_GEN_PROGS_aarch64 += set_memory_region_test
 TEST_GEN_PROGS_aarch64 += steal_time
 
@@ -91,6 +95,7 @@ TEST_GEN_PROGS_s390x += s390x/sync_regs_test
 TEST_GEN_PROGS_s390x += demand_paging_test
 TEST_GEN_PROGS_s390x += dirty_log_test
 TEST_GEN_PROGS_s390x += kvm_create_max_vcpus
+TEST_GEN_PROGS_s390x += kvm_page_table_test
 TEST_GEN_PROGS_s390x += set_memory_region_test
 
 TEST_GEN_PROGS += $(TEST_GEN_PROGS_$(UNAME_M))
diff --git a/tools/testing/selftests/kvm/aarch64/vgic_init.c b/tools/testing/selftests/kvm/aarch64/vgic_init.c
new file mode 100644
index 000000000000..623f31a14326
--- /dev/null
+++ b/tools/testing/selftests/kvm/aarch64/vgic_init.c
@@ -0,0 +1,551 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * vgic init sequence tests
+ *
+ * Copyright (C) 2020, Red Hat, Inc.
+ */
+#define _GNU_SOURCE
+#include <linux/kernel.h>
+#include <sys/syscall.h>
+#include <asm/kvm.h>
+#include <asm/kvm_para.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#define NR_VCPUS		4
+
+#define REDIST_REGION_ATTR_ADDR(count, base, flags, index) (((uint64_t)(count) << 52) | \
+	((uint64_t)((base) >> 16) << 16) | ((uint64_t)(flags) << 12) | index)
+#define REG_OFFSET(vcpu, offset) (((uint64_t)vcpu << 32) | offset)
+
+#define GICR_TYPER 0x8
+
+struct vm_gic {
+	struct kvm_vm *vm;
+	int gic_fd;
+};
+
+static int max_ipa_bits;
+
+/* helper to access a redistributor register */
+static int access_redist_reg(int gicv3_fd, int vcpu, int offset,
+			     uint32_t *val, bool write)
+{
+	uint64_t attr = REG_OFFSET(vcpu, offset);
+
+	return _kvm_device_access(gicv3_fd, KVM_DEV_ARM_VGIC_GRP_REDIST_REGS,
+				  attr, val, write);
+}
+
+/* dummy guest code */
+static void guest_code(void)
+{
+	GUEST_SYNC(0);
+	GUEST_SYNC(1);
+	GUEST_SYNC(2);
+	GUEST_DONE();
+}
+
+/* we don't want to assert on run execution, hence that helper */
+static int run_vcpu(struct kvm_vm *vm, uint32_t vcpuid)
+{
+	ucall_init(vm, NULL);
+	int ret = _vcpu_ioctl(vm, vcpuid, KVM_RUN, NULL);
+	if (ret)
+		return -errno;
+	return 0;
+}
+
+static struct vm_gic vm_gic_create(void)
+{
+	struct vm_gic v;
+
+	v.vm = vm_create_default_with_vcpus(NR_VCPUS, 0, 0, guest_code, NULL);
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	return v;
+}
+
+static void vm_gic_destroy(struct vm_gic *v)
+{
+	close(v->gic_fd);
+	kvm_vm_free(v->vm);
+}
+
+/**
+ * Helper routine that performs KVM device tests in general and
+ * especially ARM_VGIC_V3 ones. Eventually the ARM_VGIC_V3
+ * device gets created, a legacy RDIST region is set at @0x0
+ * and a DIST region is set @0x60000
+ */
+static void subtest_dist_rdist(struct vm_gic *v)
+{
+	int ret;
+	uint64_t addr;
+
+	/* Check existing group/attributes */
+	kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			      KVM_VGIC_V3_ADDR_TYPE_DIST);
+
+	kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			      KVM_VGIC_V3_ADDR_TYPE_REDIST);
+
+	/* check non existing attribute */
+	ret = _kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR, 0);
+	TEST_ASSERT(ret && errno == ENXIO, "attribute not supported");
+
+	/* misaligned DIST and REDIST address settings */
+	addr = 0x1000;
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "GICv3 dist base not 64kB aligned");
+
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "GICv3 redist base not 64kB aligned");
+
+	/* out of range address */
+	if (max_ipa_bits) {
+		addr = 1ULL << max_ipa_bits;
+		ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+					 KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
+		TEST_ASSERT(ret && errno == E2BIG, "dist address beyond IPA limit");
+
+		ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+					 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+		TEST_ASSERT(ret && errno == E2BIG, "redist address beyond IPA limit");
+	}
+
+	/* set REDIST base address @0x0*/
+	addr = 0x00000;
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+
+	/* Attempt to create a second legacy redistributor region */
+	addr = 0xE0000;
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+	TEST_ASSERT(ret && errno == EEXIST, "GICv3 redist base set again");
+
+	/* Attempt to mix legacy and new redistributor regions */
+	addr = REDIST_REGION_ATTR_ADDR(NR_VCPUS, 0x100000, 0, 0);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "attempt to mix GICv3 REDIST and REDIST_REGION");
+
+	/*
+	 * Set overlapping DIST / REDIST, cannot be detected here. Will be detected
+	 * on first vcpu run instead.
+	 */
+	addr = 3 * 2 * 0x10000;
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR, KVM_VGIC_V3_ADDR_TYPE_DIST,
+			  &addr, true);
+}
+
+/* Test the new REDIST region API */
+static void subtest_redist_regions(struct vm_gic *v)
+{
+	uint64_t addr, expected_addr;
+	int ret;
+
+	ret = kvm_device_check_attr(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				     KVM_VGIC_V3_ADDR_TYPE_REDIST);
+	TEST_ASSERT(!ret, "Multiple redist regions advertised");
+
+	addr = REDIST_REGION_ATTR_ADDR(NR_VCPUS, 0x100000, 2, 0);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "redist region attr value with flags != 0");
+
+	addr = REDIST_REGION_ATTR_ADDR(0, 0x100000, 0, 0);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "redist region attr value with count== 0");
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 1);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL,
+		    "attempt to register the first rdist region with index != 0");
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x201000, 0, 1);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "rdist region with misaligned address");
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 0);
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 1);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "register an rdist region with already used index");
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x210000, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL,
+		    "register an rdist region overlapping with another one");
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x240000, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "register redist region with index not +1");
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x240000, 0, 1);
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 1ULL << max_ipa_bits, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == E2BIG,
+		    "register redist region with base address beyond IPA range");
+
+	addr = 0x260000;
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL,
+		    "Mix KVM_VGIC_V3_ADDR_TYPE_REDIST and REDIST_REGION");
+
+	/*
+	 * Now there are 2 redist regions:
+	 * region 0 @ 0x200000 2 redists
+	 * region 1 @ 0x240000 1 redist
+	 * Attempt to read their characteristics
+	 */
+
+	addr = REDIST_REGION_ATTR_ADDR(0, 0, 0, 0);
+	expected_addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 0);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, false);
+	TEST_ASSERT(!ret && addr == expected_addr, "read characteristics of region #0");
+
+	addr = REDIST_REGION_ATTR_ADDR(0, 0, 0, 1);
+	expected_addr = REDIST_REGION_ATTR_ADDR(1, 0x240000, 0, 1);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, false);
+	TEST_ASSERT(!ret && addr == expected_addr, "read characteristics of region #1");
+
+	addr = REDIST_REGION_ATTR_ADDR(0, 0, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, false);
+	TEST_ASSERT(ret && errno == ENOENT, "read characteristics of non existing region");
+
+	addr = 0x260000;
+	kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_DIST, &addr, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x260000, 0, 2);
+	ret = _kvm_device_access(v->gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "register redist region colliding with dist");
+}
+
+/*
+ * VGIC KVM device is created and initialized before the secondary CPUs
+ * get created
+ */
+static void test_vgic_then_vcpus(void)
+{
+	struct vm_gic v;
+	int ret, i;
+
+	v.vm = vm_create_default(0, 0, guest_code);
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	subtest_dist_rdist(&v);
+
+	/* Add the rest of the VCPUs */
+	for (i = 1; i < NR_VCPUS; ++i)
+		vm_vcpu_add_default(v.vm, i, guest_code);
+
+	ret = run_vcpu(v.vm, 3);
+	TEST_ASSERT(ret == -EINVAL, "dist/rdist overlap detected on 1st vcpu run");
+
+	vm_gic_destroy(&v);
+}
+
+/* All the VCPUs are created before the VGIC KVM device gets initialized */
+static void test_vcpus_then_vgic(void)
+{
+	struct vm_gic v;
+	int ret;
+
+	v = vm_gic_create();
+
+	subtest_dist_rdist(&v);
+
+	ret = run_vcpu(v.vm, 3);
+	TEST_ASSERT(ret == -EINVAL, "dist/rdist overlap detected on 1st vcpu run");
+
+	vm_gic_destroy(&v);
+}
+
+static void test_new_redist_regions(void)
+{
+	void *dummy = NULL;
+	struct vm_gic v;
+	uint64_t addr;
+	int ret;
+
+	v = vm_gic_create();
+	subtest_redist_regions(&v);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	ret = run_vcpu(v.vm, 3);
+	TEST_ASSERT(ret == -ENXIO, "running without sufficient number of rdists");
+	vm_gic_destroy(&v);
+
+	/* step2 */
+
+	v = vm_gic_create();
+	subtest_redist_regions(&v);
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x280000, 0, 2);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	ret = run_vcpu(v.vm, 3);
+	TEST_ASSERT(ret == -EBUSY, "running without vgic explicit init");
+
+	vm_gic_destroy(&v);
+
+	/* step 3 */
+
+	v = vm_gic_create();
+	subtest_redist_regions(&v);
+
+	_kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, dummy, true);
+	TEST_ASSERT(ret && errno == EFAULT,
+		    "register a third region allowing to cover the 4 vcpus");
+
+	addr = REDIST_REGION_ATTR_ADDR(1, 0x280000, 0, 2);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	ret = run_vcpu(v.vm, 3);
+	TEST_ASSERT(!ret, "vcpu run");
+
+	vm_gic_destroy(&v);
+}
+
+static void test_typer_accesses(void)
+{
+	struct vm_gic v;
+	uint64_t addr;
+	uint32_t val;
+	int ret, i;
+
+	v.vm = vm_create_default(0, 0, guest_code);
+
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	vm_vcpu_add_default(v.vm, 3, guest_code);
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(ret && errno == EINVAL, "attempting to read GICR_TYPER of non created vcpu");
+
+	vm_vcpu_add_default(v.vm, 1, guest_code);
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(ret && errno == EBUSY, "read GICR_TYPER before GIC initialized");
+
+	vm_vcpu_add_default(v.vm, 2, guest_code);
+
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	for (i = 0; i < NR_VCPUS ; i++) {
+		ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+		TEST_ASSERT(!ret && !val, "read GICR_TYPER before rdist region setting");
+	}
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 0);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	/* The 2 first rdists should be put there (vcpu 0 and 3) */
+	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && !val, "read typer of rdist #0");
+
+	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x310, "read typer of rdist #1");
+
+	addr = REDIST_REGION_ATTR_ADDR(10, 0x100000, 0, 1);
+	ret = _kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+				 KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+	TEST_ASSERT(ret && errno == EINVAL, "collision with previous rdist region");
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x100,
+		    "no redist region attached to vcpu #1 yet, last cannot be returned");
+
+	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x200,
+		    "no redist region attached to vcpu #2, last cannot be returned");
+
+	addr = REDIST_REGION_ATTR_ADDR(10, 0x20000, 0, 1);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #1");
+
+	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x210,
+		    "read typer of rdist #1, last properly returned");
+
+	vm_gic_destroy(&v);
+}
+
+/**
+ * Test GICR_TYPER last bit with new redist regions
+ * rdist regions #1 and #2 are contiguous
+ * rdist region #0 @0x100000 2 rdist capacity
+ *     rdists: 0, 3 (Last)
+ * rdist region #1 @0x240000 2 rdist capacity
+ *     rdists:  5, 4 (Last)
+ * rdist region #2 @0x200000 2 rdist capacity
+ *     rdists: 1, 2
+ */
+static void test_last_bit_redist_regions(void)
+{
+	uint32_t vcpuids[] = { 0, 3, 5, 4, 1, 2 };
+	struct vm_gic v;
+	uint64_t addr;
+	uint32_t val;
+	int ret;
+
+	v.vm = vm_create_default_with_vcpus(6, 0, 0, guest_code, vcpuids);
+
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x100000, 0, 0);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x240000, 0, 1);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	addr = REDIST_REGION_ATTR_ADDR(2, 0x200000, 0, 2);
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION, &addr, true);
+
+	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x000, "read typer of rdist #0");
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #1");
+
+	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x200, "read typer of rdist #2");
+
+	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x310, "read typer of rdist #3");
+
+	ret = access_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x500, "read typer of rdist #5");
+
+	ret = access_redist_reg(v.gic_fd, 4, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x410, "read typer of rdist #4");
+
+	vm_gic_destroy(&v);
+}
+
+/* Test last bit with legacy region */
+static void test_last_bit_single_rdist(void)
+{
+	uint32_t vcpuids[] = { 0, 3, 5, 4, 1, 2 };
+	struct vm_gic v;
+	uint64_t addr;
+	uint32_t val;
+	int ret;
+
+	v.vm = vm_create_default_with_vcpus(6, 0, 0, guest_code, vcpuids);
+
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_CTRL,
+			  KVM_DEV_ARM_VGIC_CTRL_INIT, NULL, true);
+
+	addr = 0x10000;
+	kvm_device_access(v.gic_fd, KVM_DEV_ARM_VGIC_GRP_ADDR,
+			  KVM_VGIC_V3_ADDR_TYPE_REDIST, &addr, true);
+
+	ret = access_redist_reg(v.gic_fd, 0, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x000, "read typer of rdist #0");
+
+	ret = access_redist_reg(v.gic_fd, 3, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x300, "read typer of rdist #1");
+
+	ret = access_redist_reg(v.gic_fd, 5, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x500, "read typer of rdist #2");
+
+	ret = access_redist_reg(v.gic_fd, 1, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x100, "read typer of rdist #3");
+
+	ret = access_redist_reg(v.gic_fd, 2, GICR_TYPER, &val, false);
+	TEST_ASSERT(!ret && val == 0x210, "read typer of rdist #3");
+
+	vm_gic_destroy(&v);
+}
+
+void test_kvm_device(void)
+{
+	struct vm_gic v;
+	int ret, fd;
+
+	v.vm = vm_create_default_with_vcpus(NR_VCPUS, 0, 0, guest_code, NULL);
+
+	/* try to create a non existing KVM device */
+	ret = _kvm_create_device(v.vm, 0, true, &fd);
+	TEST_ASSERT(ret && errno == ENODEV, "unsupported device");
+
+	/* trial mode with VGIC_V3 device */
+	ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, true, &fd);
+	if (ret) {
+		print_skip("GICv3 not supported");
+		exit(KSFT_SKIP);
+	}
+	v.gic_fd = kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false);
+
+	ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, false, &fd);
+	TEST_ASSERT(ret && errno == EEXIST, "create GICv3 device twice");
+
+	kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V3, true);
+
+	if (!_kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V2, true, &fd)) {
+		ret = _kvm_create_device(v.vm, KVM_DEV_TYPE_ARM_VGIC_V2, false, &fd);
+		TEST_ASSERT(ret && errno == EINVAL, "create GICv2 while v3 exists");
+	}
+
+	vm_gic_destroy(&v);
+}
+
+int main(int ac, char **av)
+{
+	max_ipa_bits = kvm_check_cap(KVM_CAP_ARM_VM_IPA_SIZE);
+
+	test_kvm_device();
+	test_vcpus_then_vgic();
+	test_vgic_then_vcpus();
+	test_new_redist_regions();
+	test_typer_accesses();
+	test_last_bit_redist_regions();
+	test_last_bit_single_rdist();
+
+	return 0;
+}
diff --git a/tools/testing/selftests/kvm/demand_paging_test.c b/tools/testing/selftests/kvm/demand_paging_test.c
index 5f7a229c3af1..b74704305835 100644
--- a/tools/testing/selftests/kvm/demand_paging_test.c
+++ b/tools/testing/selftests/kvm/demand_paging_test.c
@@ -9,6 +9,7 @@
 
 #define _GNU_SOURCE /* for pipe2 */
 
+#include <inttypes.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <time.h>
@@ -38,6 +39,7 @@
 
 static int nr_vcpus = 1;
 static uint64_t guest_percpu_mem_size = DEFAULT_PER_VCPU_MEM_SIZE;
+static size_t demand_paging_size;
 static char *guest_data_prototype;
 
 static void *vcpu_worker(void *data)
@@ -71,36 +73,51 @@ static void *vcpu_worker(void *data)
 	return NULL;
 }
 
-static int handle_uffd_page_request(int uffd, uint64_t addr)
+static int handle_uffd_page_request(int uffd_mode, int uffd, uint64_t addr)
 {
-	pid_t tid;
+	pid_t tid = syscall(__NR_gettid);
 	struct timespec start;
 	struct timespec ts_diff;
-	struct uffdio_copy copy;
 	int r;
 
-	tid = syscall(__NR_gettid);
+	clock_gettime(CLOCK_MONOTONIC, &start);
 
-	copy.src = (uint64_t)guest_data_prototype;
-	copy.dst = addr;
-	copy.len = perf_test_args.host_page_size;
-	copy.mode = 0;
+	if (uffd_mode == UFFDIO_REGISTER_MODE_MISSING) {
+		struct uffdio_copy copy;
 
-	clock_gettime(CLOCK_MONOTONIC, &start);
+		copy.src = (uint64_t)guest_data_prototype;
+		copy.dst = addr;
+		copy.len = demand_paging_size;
+		copy.mode = 0;
 
-	r = ioctl(uffd, UFFDIO_COPY, &copy);
-	if (r == -1) {
-		pr_info("Failed Paged in 0x%lx from thread %d with errno: %d\n",
-			addr, tid, errno);
-		return r;
+		r = ioctl(uffd, UFFDIO_COPY, &copy);
+		if (r == -1) {
+			pr_info("Failed UFFDIO_COPY in 0x%lx from thread %d with errno: %d\n",
+				addr, tid, errno);
+			return r;
+		}
+	} else if (uffd_mode == UFFDIO_REGISTER_MODE_MINOR) {
+		struct uffdio_continue cont = {0};
+
+		cont.range.start = addr;
+		cont.range.len = demand_paging_size;
+
+		r = ioctl(uffd, UFFDIO_CONTINUE, &cont);
+		if (r == -1) {
+			pr_info("Failed UFFDIO_CONTINUE in 0x%lx from thread %d with errno: %d\n",
+				addr, tid, errno);
+			return r;
+		}
+	} else {
+		TEST_FAIL("Invalid uffd mode %d", uffd_mode);
 	}
 
 	ts_diff = timespec_elapsed(start);
 
-	PER_PAGE_DEBUG("UFFDIO_COPY %d \t%ld ns\n", tid,
+	PER_PAGE_DEBUG("UFFD page-in %d \t%ld ns\n", tid,
 		       timespec_to_ns(ts_diff));
 	PER_PAGE_DEBUG("Paged in %ld bytes at 0x%lx from thread %d\n",
-		       perf_test_args.host_page_size, addr, tid);
+		       demand_paging_size, addr, tid);
 
 	return 0;
 }
@@ -108,6 +125,7 @@ static int handle_uffd_page_request(int uffd, uint64_t addr)
 bool quit_uffd_thread;
 
 struct uffd_handler_args {
+	int uffd_mode;
 	int uffd;
 	int pipefd;
 	useconds_t delay;
@@ -169,7 +187,7 @@ static void *uffd_handler_thread_fn(void *arg)
 		if (r == -1) {
 			if (errno == EAGAIN)
 				continue;
-			pr_info("Read of uffd gor errno %d", errno);
+			pr_info("Read of uffd got errno %d\n", errno);
 			return NULL;
 		}
 
@@ -184,7 +202,7 @@ static void *uffd_handler_thread_fn(void *arg)
 		if (delay)
 			usleep(delay);
 		addr =  msg.arg.pagefault.address;
-		r = handle_uffd_page_request(uffd, addr);
+		r = handle_uffd_page_request(uffd_args->uffd_mode, uffd, addr);
 		if (r < 0)
 			return NULL;
 		pages++;
@@ -198,43 +216,53 @@ static void *uffd_handler_thread_fn(void *arg)
 	return NULL;
 }
 
-static int setup_demand_paging(struct kvm_vm *vm,
-			       pthread_t *uffd_handler_thread, int pipefd,
-			       useconds_t uffd_delay,
-			       struct uffd_handler_args *uffd_args,
-			       void *hva, uint64_t len)
+static void setup_demand_paging(struct kvm_vm *vm,
+				pthread_t *uffd_handler_thread, int pipefd,
+				int uffd_mode, useconds_t uffd_delay,
+				struct uffd_handler_args *uffd_args,
+				void *hva, void *alias, uint64_t len)
 {
+	bool is_minor = (uffd_mode == UFFDIO_REGISTER_MODE_MINOR);
 	int uffd;
 	struct uffdio_api uffdio_api;
 	struct uffdio_register uffdio_register;
+	uint64_t expected_ioctls = ((uint64_t) 1) << _UFFDIO_COPY;
 
-	uffd = syscall(__NR_userfaultfd, O_CLOEXEC | O_NONBLOCK);
-	if (uffd == -1) {
-		pr_info("uffd creation failed\n");
-		return -1;
+	PER_PAGE_DEBUG("Userfaultfd %s mode, faults resolved with %s\n",
+		       is_minor ? "MINOR" : "MISSING",
+		       is_minor ? "UFFDIO_CONINUE" : "UFFDIO_COPY");
+
+	/* In order to get minor faults, prefault via the alias. */
+	if (is_minor) {
+		size_t p;
+
+		expected_ioctls = ((uint64_t) 1) << _UFFDIO_CONTINUE;
+
+		TEST_ASSERT(alias != NULL, "Alias required for minor faults");
+		for (p = 0; p < (len / demand_paging_size); ++p) {
+			memcpy(alias + (p * demand_paging_size),
+			       guest_data_prototype, demand_paging_size);
+		}
 	}
 
+	uffd = syscall(__NR_userfaultfd, O_CLOEXEC | O_NONBLOCK);
+	TEST_ASSERT(uffd >= 0, "uffd creation failed, errno: %d", errno);
+
 	uffdio_api.api = UFFD_API;
 	uffdio_api.features = 0;
-	if (ioctl(uffd, UFFDIO_API, &uffdio_api) == -1) {
-		pr_info("ioctl uffdio_api failed\n");
-		return -1;
-	}
+	TEST_ASSERT(ioctl(uffd, UFFDIO_API, &uffdio_api) != -1,
+		    "ioctl UFFDIO_API failed: %" PRIu64,
+		    (uint64_t)uffdio_api.api);
 
 	uffdio_register.range.start = (uint64_t)hva;
 	uffdio_register.range.len = len;
-	uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING;
-	if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register) == -1) {
-		pr_info("ioctl uffdio_register failed\n");
-		return -1;
-	}
-
-	if ((uffdio_register.ioctls & UFFD_API_RANGE_IOCTLS) !=
-			UFFD_API_RANGE_IOCTLS) {
-		pr_info("unexpected userfaultfd ioctl set\n");
-		return -1;
-	}
+	uffdio_register.mode = uffd_mode;
+	TEST_ASSERT(ioctl(uffd, UFFDIO_REGISTER, &uffdio_register) != -1,
+		    "ioctl UFFDIO_REGISTER failed");
+	TEST_ASSERT((uffdio_register.ioctls & expected_ioctls) ==
+		    expected_ioctls, "missing userfaultfd ioctls");
 
+	uffd_args->uffd_mode = uffd_mode;
 	uffd_args->uffd = uffd;
 	uffd_args->pipefd = pipefd;
 	uffd_args->delay = uffd_delay;
@@ -243,13 +271,12 @@ static int setup_demand_paging(struct kvm_vm *vm,
 
 	PER_VCPU_DEBUG("Created uffd thread for HVA range [%p, %p)\n",
 		       hva, hva + len);
-
-	return 0;
 }
 
 struct test_params {
-	bool use_uffd;
+	int uffd_mode;
 	useconds_t uffd_delay;
+	enum vm_mem_backing_src_type src_type;
 	bool partition_vcpu_memory_access;
 };
 
@@ -267,14 +294,16 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 	int r;
 
 	vm = perf_test_create_vm(mode, nr_vcpus, guest_percpu_mem_size,
-				 VM_MEM_SRC_ANONYMOUS);
+				 p->src_type);
 
 	perf_test_args.wr_fract = 1;
 
-	guest_data_prototype = malloc(perf_test_args.host_page_size);
+	demand_paging_size = get_backing_src_pagesz(p->src_type);
+
+	guest_data_prototype = malloc(demand_paging_size);
 	TEST_ASSERT(guest_data_prototype,
 		    "Failed to allocate buffer for guest data pattern");
-	memset(guest_data_prototype, 0xAB, perf_test_args.host_page_size);
+	memset(guest_data_prototype, 0xAB, demand_paging_size);
 
 	vcpu_threads = malloc(nr_vcpus * sizeof(*vcpu_threads));
 	TEST_ASSERT(vcpu_threads, "Memory allocation failed");
@@ -282,7 +311,7 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 	perf_test_setup_vcpus(vm, nr_vcpus, guest_percpu_mem_size,
 			      p->partition_vcpu_memory_access);
 
-	if (p->use_uffd) {
+	if (p->uffd_mode) {
 		uffd_handler_threads =
 			malloc(nr_vcpus * sizeof(*uffd_handler_threads));
 		TEST_ASSERT(uffd_handler_threads, "Memory allocation failed");
@@ -296,6 +325,7 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 		for (vcpu_id = 0; vcpu_id < nr_vcpus; vcpu_id++) {
 			vm_paddr_t vcpu_gpa;
 			void *vcpu_hva;
+			void *vcpu_alias;
 			uint64_t vcpu_mem_size;
 
 
@@ -310,8 +340,9 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 			PER_VCPU_DEBUG("Added VCPU %d with test mem gpa [%lx, %lx)\n",
 				       vcpu_id, vcpu_gpa, vcpu_gpa + vcpu_mem_size);
 
-			/* Cache the HVA pointer of the region */
+			/* Cache the host addresses of the region */
 			vcpu_hva = addr_gpa2hva(vm, vcpu_gpa);
+			vcpu_alias = addr_gpa2alias(vm, vcpu_gpa);
 
 			/*
 			 * Set up user fault fd to handle demand paging
@@ -321,13 +352,11 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 				  O_CLOEXEC | O_NONBLOCK);
 			TEST_ASSERT(!r, "Failed to set up pipefd");
 
-			r = setup_demand_paging(vm,
-						&uffd_handler_threads[vcpu_id],
-						pipefds[vcpu_id * 2],
-						p->uffd_delay, &uffd_args[vcpu_id],
-						vcpu_hva, vcpu_mem_size);
-			if (r < 0)
-				exit(-r);
+			setup_demand_paging(vm, &uffd_handler_threads[vcpu_id],
+					    pipefds[vcpu_id * 2], p->uffd_mode,
+					    p->uffd_delay, &uffd_args[vcpu_id],
+					    vcpu_hva, vcpu_alias,
+					    vcpu_mem_size);
 		}
 	}
 
@@ -355,7 +384,7 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 
 	pr_info("All vCPU threads joined\n");
 
-	if (p->use_uffd) {
+	if (p->uffd_mode) {
 		char c;
 
 		/* Tell the user fault fd handler threads to quit */
@@ -377,7 +406,7 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 
 	free(guest_data_prototype);
 	free(vcpu_threads);
-	if (p->use_uffd) {
+	if (p->uffd_mode) {
 		free(uffd_handler_threads);
 		free(uffd_args);
 		free(pipefds);
@@ -387,17 +416,19 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 static void help(char *name)
 {
 	puts("");
-	printf("usage: %s [-h] [-m mode] [-u] [-d uffd_delay_usec]\n"
-	       "          [-b memory] [-v vcpus] [-o]\n", name);
+	printf("usage: %s [-h] [-m vm_mode] [-u uffd_mode] [-d uffd_delay_usec]\n"
+	       "          [-b memory] [-t type] [-v vcpus] [-o]\n", name);
 	guest_modes_help();
-	printf(" -u: use User Fault FD to handle vCPU page\n"
-	       "     faults.\n");
+	printf(" -u: use userfaultfd to handle vCPU page faults. Mode is a\n"
+	       "     UFFD registration mode: 'MISSING' or 'MINOR'.\n");
 	printf(" -d: add a delay in usec to the User Fault\n"
 	       "     FD handler to simulate demand paging\n"
 	       "     overheads. Ignored without -u.\n");
 	printf(" -b: specify the size of the memory region which should be\n"
 	       "     demand paged by each vCPU. e.g. 10M or 3G.\n"
 	       "     Default: 1G\n");
+	printf(" -t: The type of backing memory to use. Default: anonymous\n");
+	backing_src_help();
 	printf(" -v: specify the number of vCPUs to run.\n");
 	printf(" -o: Overlap guest memory accesses instead of partitioning\n"
 	       "     them into a separate region of memory for each vCPU.\n");
@@ -409,19 +440,24 @@ int main(int argc, char *argv[])
 {
 	int max_vcpus = kvm_check_cap(KVM_CAP_MAX_VCPUS);
 	struct test_params p = {
+		.src_type = VM_MEM_SRC_ANONYMOUS,
 		.partition_vcpu_memory_access = true,
 	};
 	int opt;
 
 	guest_modes_append_default();
 
-	while ((opt = getopt(argc, argv, "hm:ud:b:v:o")) != -1) {
+	while ((opt = getopt(argc, argv, "hm:u:d:b:t:v:o")) != -1) {
 		switch (opt) {
 		case 'm':
 			guest_modes_cmdline(optarg);
 			break;
 		case 'u':
-			p.use_uffd = true;
+			if (!strcmp("MISSING", optarg))
+				p.uffd_mode = UFFDIO_REGISTER_MODE_MISSING;
+			else if (!strcmp("MINOR", optarg))
+				p.uffd_mode = UFFDIO_REGISTER_MODE_MINOR;
+			TEST_ASSERT(p.uffd_mode, "UFFD mode must be 'MISSING' or 'MINOR'.");
 			break;
 		case 'd':
 			p.uffd_delay = strtoul(optarg, NULL, 0);
@@ -430,6 +466,9 @@ int main(int argc, char *argv[])
 		case 'b':
 			guest_percpu_mem_size = parse_size(optarg);
 			break;
+		case 't':
+			p.src_type = parse_backing_src_type(optarg);
+			break;
 		case 'v':
 			nr_vcpus = atoi(optarg);
 			TEST_ASSERT(nr_vcpus > 0 && nr_vcpus <= max_vcpus,
@@ -445,6 +484,11 @@ int main(int argc, char *argv[])
 		}
 	}
 
+	if (p.uffd_mode == UFFDIO_REGISTER_MODE_MINOR &&
+	    !backing_src_is_shared(p.src_type)) {
+		TEST_FAIL("userfaultfd MINOR mode requires shared memory; pick a different -t");
+	}
+
 	for_each_guest_mode(run_test, &p);
 
 	return 0;
diff --git a/tools/testing/selftests/kvm/dirty_log_test.c b/tools/testing/selftests/kvm/dirty_log_test.c
index bb2752d78fe3..81edbd23d371 100644
--- a/tools/testing/selftests/kvm/dirty_log_test.c
+++ b/tools/testing/selftests/kvm/dirty_log_test.c
@@ -17,6 +17,7 @@
 #include <linux/bitmap.h>
 #include <linux/bitops.h>
 #include <asm/barrier.h>
+#include <linux/atomic.h>
 
 #include "kvm_util.h"
 #include "test_util.h"
@@ -137,12 +138,20 @@ static uint64_t host_clear_count;
 static uint64_t host_track_next_count;
 
 /* Whether dirty ring reset is requested, or finished */
-static sem_t dirty_ring_vcpu_stop;
-static sem_t dirty_ring_vcpu_cont;
+static sem_t sem_vcpu_stop;
+static sem_t sem_vcpu_cont;
+/*
+ * This is only set by main thread, and only cleared by vcpu thread.  It is
+ * used to request vcpu thread to stop at the next GUEST_SYNC, since GUEST_SYNC
+ * is the only place that we'll guarantee both "dirty bit" and "dirty data"
+ * will match.  E.g., SIG_IPI won't guarantee that if the vcpu is interrupted
+ * after setting dirty bit but before the data is written.
+ */
+static atomic_t vcpu_sync_stop_requested;
 /*
  * This is updated by the vcpu thread to tell the host whether it's a
  * ring-full event.  It should only be read until a sem_wait() of
- * dirty_ring_vcpu_stop and before vcpu continues to run.
+ * sem_vcpu_stop and before vcpu continues to run.
  */
 static bool dirty_ring_vcpu_ring_full;
 /*
@@ -234,6 +243,17 @@ static void clear_log_collect_dirty_pages(struct kvm_vm *vm, int slot,
 	kvm_vm_clear_dirty_log(vm, slot, bitmap, 0, num_pages);
 }
 
+/* Should only be called after a GUEST_SYNC */
+static void vcpu_handle_sync_stop(void)
+{
+	if (atomic_read(&vcpu_sync_stop_requested)) {
+		/* It means main thread is sleeping waiting */
+		atomic_set(&vcpu_sync_stop_requested, false);
+		sem_post(&sem_vcpu_stop);
+		sem_wait_until(&sem_vcpu_cont);
+	}
+}
+
 static void default_after_vcpu_run(struct kvm_vm *vm, int ret, int err)
 {
 	struct kvm_run *run = vcpu_state(vm, VCPU_ID);
@@ -244,6 +264,8 @@ static void default_after_vcpu_run(struct kvm_vm *vm, int ret, int err)
 	TEST_ASSERT(get_ucall(vm, VCPU_ID, NULL) == UCALL_SYNC,
 		    "Invalid guest sync status: exit_reason=%s\n",
 		    exit_reason_str(run->exit_reason));
+
+	vcpu_handle_sync_stop();
 }
 
 static bool dirty_ring_supported(void)
@@ -301,13 +323,13 @@ static void dirty_ring_wait_vcpu(void)
 {
 	/* This makes sure that hardware PML cache flushed */
 	vcpu_kick();
-	sem_wait_until(&dirty_ring_vcpu_stop);
+	sem_wait_until(&sem_vcpu_stop);
 }
 
 static void dirty_ring_continue_vcpu(void)
 {
 	pr_info("Notifying vcpu to continue\n");
-	sem_post(&dirty_ring_vcpu_cont);
+	sem_post(&sem_vcpu_cont);
 }
 
 static void dirty_ring_collect_dirty_pages(struct kvm_vm *vm, int slot,
@@ -361,11 +383,11 @@ static void dirty_ring_after_vcpu_run(struct kvm_vm *vm, int ret, int err)
 		/* Update the flag first before pause */
 		WRITE_ONCE(dirty_ring_vcpu_ring_full,
 			   run->exit_reason == KVM_EXIT_DIRTY_RING_FULL);
-		sem_post(&dirty_ring_vcpu_stop);
+		sem_post(&sem_vcpu_stop);
 		pr_info("vcpu stops because %s...\n",
 			dirty_ring_vcpu_ring_full ?
 			"dirty ring is full" : "vcpu is kicked out");
-		sem_wait_until(&dirty_ring_vcpu_cont);
+		sem_wait_until(&sem_vcpu_cont);
 		pr_info("vcpu continues now.\n");
 	} else {
 		TEST_ASSERT(false, "Invalid guest sync status: "
@@ -377,7 +399,7 @@ static void dirty_ring_after_vcpu_run(struct kvm_vm *vm, int ret, int err)
 static void dirty_ring_before_vcpu_join(void)
 {
 	/* Kick another round of vcpu just to make sure it will quit */
-	sem_post(&dirty_ring_vcpu_cont);
+	sem_post(&sem_vcpu_cont);
 }
 
 struct log_mode {
@@ -505,9 +527,8 @@ static void *vcpu_worker(void *data)
 	 */
 	sigmask->len = 8;
 	pthread_sigmask(0, NULL, sigset);
+	sigdelset(sigset, SIG_IPI);
 	vcpu_ioctl(vm, VCPU_ID, KVM_SET_SIGNAL_MASK, sigmask);
-	sigaddset(sigset, SIG_IPI);
-	pthread_sigmask(SIG_BLOCK, sigset, NULL);
 
 	sigemptyset(sigset);
 	sigaddset(sigset, SIG_IPI);
@@ -768,7 +789,25 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 		usleep(p->interval * 1000);
 		log_mode_collect_dirty_pages(vm, TEST_MEM_SLOT_INDEX,
 					     bmap, host_num_pages);
+
+		/*
+		 * See vcpu_sync_stop_requested definition for details on why
+		 * we need to stop vcpu when verify data.
+		 */
+		atomic_set(&vcpu_sync_stop_requested, true);
+		sem_wait_until(&sem_vcpu_stop);
+		/*
+		 * NOTE: for dirty ring, it's possible that we didn't stop at
+		 * GUEST_SYNC but instead we stopped because ring is full;
+		 * that's okay too because ring full means we're only missing
+		 * the flush of the last page, and since we handle the last
+		 * page specially verification will succeed anyway.
+		 */
+		assert(host_log_mode == LOG_MODE_DIRTY_RING ||
+		       atomic_read(&vcpu_sync_stop_requested) == false);
 		vm_dirty_log_verify(mode, bmap);
+		sem_post(&sem_vcpu_cont);
+
 		iteration++;
 		sync_global_to_guest(vm, iteration);
 	}
@@ -818,9 +857,10 @@ int main(int argc, char *argv[])
 		.interval = TEST_HOST_LOOP_INTERVAL,
 	};
 	int opt, i;
+	sigset_t sigset;
 
-	sem_init(&dirty_ring_vcpu_stop, 0, 0);
-	sem_init(&dirty_ring_vcpu_cont, 0, 0);
+	sem_init(&sem_vcpu_stop, 0, 0);
+	sem_init(&sem_vcpu_cont, 0, 0);
 
 	guest_modes_append_default();
 
@@ -876,6 +916,11 @@ int main(int argc, char *argv[])
 
 	srandom(time(0));
 
+	/* Ensure that vCPU threads start with SIG_IPI blocked.  */
+	sigemptyset(&sigset);
+	sigaddset(&sigset, SIG_IPI);
+	pthread_sigmask(SIG_BLOCK, &sigset, NULL);
+
 	if (host_log_mode_option == LOG_MODE_ALL) {
 		/* Run each log mode */
 		for (i = 0; i < LOG_MODE_NUM; i++) {
diff --git a/tools/testing/selftests/kvm/hardware_disable_test.c b/tools/testing/selftests/kvm/hardware_disable_test.c
index 5aadf84c91c0..4b8db3bce610 100644
--- a/tools/testing/selftests/kvm/hardware_disable_test.c
+++ b/tools/testing/selftests/kvm/hardware_disable_test.c
@@ -132,6 +132,36 @@ static void run_test(uint32_t run)
 	TEST_ASSERT(false, "%s: [%d] child escaped the ninja\n", __func__, run);
 }
 
+void wait_for_child_setup(pid_t pid)
+{
+	/*
+	 * Wait for the child to post to the semaphore, but wake up periodically
+	 * to check if the child exited prematurely.
+	 */
+	for (;;) {
+		const struct timespec wait_period = { .tv_sec = 1 };
+		int status;
+
+		if (!sem_timedwait(sem, &wait_period))
+			return;
+
+		/* Child is still running, keep waiting. */
+		if (pid != waitpid(pid, &status, WNOHANG))
+			continue;
+
+		/*
+		 * Child is no longer running, which is not expected.
+		 *
+		 * If it exited with a non-zero status, we explicitly forward
+		 * the child's status in case it exited with KSFT_SKIP.
+		 */
+		if (WIFEXITED(status))
+			exit(WEXITSTATUS(status));
+		else
+			TEST_ASSERT(false, "Child exited unexpectedly");
+	}
+}
+
 int main(int argc, char **argv)
 {
 	uint32_t i;
@@ -148,7 +178,7 @@ int main(int argc, char **argv)
 			run_test(i); /* This function always exits */
 
 		pr_debug("%s: [%d] waiting semaphore\n", __func__, i);
-		sem_wait(sem);
+		wait_for_child_setup(pid);
 		r = (rand() % DELAY_US_MAX) + 1;
 		pr_debug("%s: [%d] waiting %dus\n", __func__, i, r);
 		usleep(r);
diff --git a/tools/testing/selftests/kvm/include/kvm_util.h b/tools/testing/selftests/kvm/include/kvm_util.h
index 0f4258eaa629..fcd8e3855111 100644
--- a/tools/testing/selftests/kvm/include/kvm_util.h
+++ b/tools/testing/selftests/kvm/include/kvm_util.h
@@ -69,9 +69,6 @@ enum vm_guest_mode {
 #define MIN_PAGE_SIZE		(1U << MIN_PAGE_SHIFT)
 #define PTES_PER_MIN_PAGE	ptes_per_page(MIN_PAGE_SIZE)
 
-#define vm_guest_mode_string(m) vm_guest_mode_string[m]
-extern const char * const vm_guest_mode_string[];
-
 struct vm_guest_mode_params {
 	unsigned int pa_bits;
 	unsigned int va_bits;
@@ -80,11 +77,13 @@ struct vm_guest_mode_params {
 };
 extern const struct vm_guest_mode_params vm_guest_mode_params[];
 
+int open_kvm_dev_path_or_exit(void);
 int kvm_check_cap(long cap);
 int vm_enable_cap(struct kvm_vm *vm, struct kvm_enable_cap *cap);
 int vcpu_enable_cap(struct kvm_vm *vm, uint32_t vcpu_id,
 		    struct kvm_enable_cap *cap);
 void vm_enable_dirty_ring(struct kvm_vm *vm, uint32_t ring_size);
+const char *vm_guest_mode_string(uint32_t i);
 
 struct kvm_vm *vm_create(enum vm_guest_mode mode, uint64_t phy_pages, int perm);
 void kvm_vm_free(struct kvm_vm *vmp);
@@ -148,6 +147,7 @@ void virt_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr,
 void *addr_gpa2hva(struct kvm_vm *vm, vm_paddr_t gpa);
 void *addr_gva2hva(struct kvm_vm *vm, vm_vaddr_t gva);
 vm_paddr_t addr_hva2gpa(struct kvm_vm *vm, void *hva);
+void *addr_gpa2alias(struct kvm_vm *vm, vm_paddr_t gpa);
 
 /*
  * Address Guest Virtual to Guest Physical
@@ -225,6 +225,15 @@ int vcpu_nested_state_set(struct kvm_vm *vm, uint32_t vcpuid,
 #endif
 void *vcpu_map_dirty_ring(struct kvm_vm *vm, uint32_t vcpuid);
 
+int _kvm_device_check_attr(int dev_fd, uint32_t group, uint64_t attr);
+int kvm_device_check_attr(int dev_fd, uint32_t group, uint64_t attr);
+int _kvm_create_device(struct kvm_vm *vm, uint64_t type, bool test, int *fd);
+int kvm_create_device(struct kvm_vm *vm, uint64_t type, bool test);
+int _kvm_device_access(int dev_fd, uint32_t group, uint64_t attr,
+		       void *val, bool write);
+int kvm_device_access(int dev_fd, uint32_t group, uint64_t attr,
+		      void *val, bool write);
+
 const char *exit_reason_str(unsigned int exit_reason);
 
 void virt_pgd_alloc(struct kvm_vm *vm, uint32_t pgd_memslot);
@@ -295,7 +304,7 @@ bool vm_is_unrestricted_guest(struct kvm_vm *vm);
 
 unsigned int vm_get_page_size(struct kvm_vm *vm);
 unsigned int vm_get_page_shift(struct kvm_vm *vm);
-unsigned int vm_get_max_gfn(struct kvm_vm *vm);
+uint64_t vm_get_max_gfn(struct kvm_vm *vm);
 int vm_get_fd(struct kvm_vm *vm);
 
 unsigned int vm_calc_num_guest_pages(enum vm_guest_mode mode, size_t size);
diff --git a/tools/testing/selftests/kvm/include/test_util.h b/tools/testing/selftests/kvm/include/test_util.h
index b7f41399f22c..d79be15dd3d2 100644
--- a/tools/testing/selftests/kvm/include/test_util.h
+++ b/tools/testing/selftests/kvm/include/test_util.h
@@ -17,6 +17,7 @@
 #include <errno.h>
 #include <unistd.h>
 #include <fcntl.h>
+#include <sys/mman.h>
 #include "kselftest.h"
 
 static inline int _no_printf(const char *format, ...) { return 0; }
@@ -71,14 +72,44 @@ enum vm_mem_backing_src_type {
 	VM_MEM_SRC_ANONYMOUS,
 	VM_MEM_SRC_ANONYMOUS_THP,
 	VM_MEM_SRC_ANONYMOUS_HUGETLB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_16KB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_64KB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_512KB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_1MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_2MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_8MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_16MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_32MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_256MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_512MB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_1GB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_2GB,
+	VM_MEM_SRC_ANONYMOUS_HUGETLB_16GB,
+	VM_MEM_SRC_SHMEM,
+	VM_MEM_SRC_SHARED_HUGETLB,
+	NUM_SRC_TYPES,
 };
 
 struct vm_mem_backing_src_alias {
 	const char *name;
-	enum vm_mem_backing_src_type type;
+	uint32_t flag;
 };
 
+bool thp_configured(void);
+size_t get_trans_hugepagesz(void);
+size_t get_def_hugetlb_pagesz(void);
+const struct vm_mem_backing_src_alias *vm_mem_backing_src_alias(uint32_t i);
+size_t get_backing_src_pagesz(uint32_t i);
 void backing_src_help(void);
 enum vm_mem_backing_src_type parse_backing_src_type(const char *type_name);
 
+/*
+ * Whether or not the given source type is shared memory (as opposed to
+ * anonymous).
+ */
+static inline bool backing_src_is_shared(enum vm_mem_backing_src_type t)
+{
+	return vm_mem_backing_src_alias(t)->flag & MAP_SHARED;
+}
+
 #endif /* SELFTEST_KVM_TEST_UTIL_H */
diff --git a/tools/testing/selftests/kvm/kvm_page_table_test.c b/tools/testing/selftests/kvm/kvm_page_table_test.c
new file mode 100644
index 000000000000..1c4753fff19e
--- /dev/null
+++ b/tools/testing/selftests/kvm/kvm_page_table_test.c
@@ -0,0 +1,506 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KVM page table test
+ *
+ * Copyright (C) 2021, Huawei, Inc.
+ *
+ * Make sure that THP has been enabled or enough HUGETLB pages with specific
+ * page size have been pre-allocated on your system, if you are planning to
+ * use hugepages to back the guest memory for testing.
+ */
+
+#define _GNU_SOURCE /* for program_invocation_name */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <time.h>
+#include <pthread.h>
+#include <semaphore.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+#include "guest_modes.h"
+
+#define TEST_MEM_SLOT_INDEX             1
+
+/* Default size(1GB) of the memory for testing */
+#define DEFAULT_TEST_MEM_SIZE		(1 << 30)
+
+/* Default guest test virtual memory offset */
+#define DEFAULT_GUEST_TEST_MEM		0xc0000000
+
+/* Different guest memory accessing stages */
+enum test_stage {
+	KVM_BEFORE_MAPPINGS,
+	KVM_CREATE_MAPPINGS,
+	KVM_UPDATE_MAPPINGS,
+	KVM_ADJUST_MAPPINGS,
+	NUM_TEST_STAGES,
+};
+
+static const char * const test_stage_string[] = {
+	"KVM_BEFORE_MAPPINGS",
+	"KVM_CREATE_MAPPINGS",
+	"KVM_UPDATE_MAPPINGS",
+	"KVM_ADJUST_MAPPINGS",
+};
+
+struct vcpu_args {
+	int vcpu_id;
+	bool vcpu_write;
+};
+
+struct test_args {
+	struct kvm_vm *vm;
+	uint64_t guest_test_virt_mem;
+	uint64_t host_page_size;
+	uint64_t host_num_pages;
+	uint64_t large_page_size;
+	uint64_t large_num_pages;
+	uint64_t host_pages_per_lpage;
+	enum vm_mem_backing_src_type src_type;
+	struct vcpu_args vcpu_args[KVM_MAX_VCPUS];
+};
+
+/*
+ * Guest variables. Use addr_gva2hva() if these variables need
+ * to be changed in host.
+ */
+static enum test_stage guest_test_stage;
+
+/* Host variables */
+static uint32_t nr_vcpus = 1;
+static struct test_args test_args;
+static enum test_stage *current_stage;
+static bool host_quit;
+
+/* Whether the test stage is updated, or completed */
+static sem_t test_stage_updated;
+static sem_t test_stage_completed;
+
+/*
+ * Guest physical memory offset of the testing memory slot.
+ * This will be set to the topmost valid physical address minus
+ * the test memory size.
+ */
+static uint64_t guest_test_phys_mem;
+
+/*
+ * Guest virtual memory offset of the testing memory slot.
+ * Must not conflict with identity mapped test code.
+ */
+static uint64_t guest_test_virt_mem = DEFAULT_GUEST_TEST_MEM;
+
+static void guest_code(int vcpu_id)
+{
+	struct test_args *p = &test_args;
+	struct vcpu_args *vcpu_args = &p->vcpu_args[vcpu_id];
+	enum test_stage *current_stage = &guest_test_stage;
+	uint64_t addr;
+	int i, j;
+
+	/* Make sure vCPU args data structure is not corrupt */
+	GUEST_ASSERT(vcpu_args->vcpu_id == vcpu_id);
+
+	while (true) {
+		addr = p->guest_test_virt_mem;
+
+		switch (READ_ONCE(*current_stage)) {
+		/*
+		 * All vCPU threads will be started in this stage,
+		 * where guest code of each vCPU will do nothing.
+		 */
+		case KVM_BEFORE_MAPPINGS:
+			break;
+
+		/*
+		 * Before dirty logging, vCPUs concurrently access the first
+		 * 8 bytes of each page (host page/large page) within the same
+		 * memory region with different accessing types (read/write).
+		 * Then KVM will create normal page mappings or huge block
+		 * mappings for them.
+		 */
+		case KVM_CREATE_MAPPINGS:
+			for (i = 0; i < p->large_num_pages; i++) {
+				if (vcpu_args->vcpu_write)
+					*(uint64_t *)addr = 0x0123456789ABCDEF;
+				else
+					READ_ONCE(*(uint64_t *)addr);
+
+				addr += p->large_page_size;
+			}
+			break;
+
+		/*
+		 * During dirty logging, KVM will only update attributes of the
+		 * normal page mappings from RO to RW if memory backing src type
+		 * is anonymous. In other cases, KVM will split the huge block
+		 * mappings into normal page mappings if memory backing src type
+		 * is THP or HUGETLB.
+		 */
+		case KVM_UPDATE_MAPPINGS:
+			if (p->src_type == VM_MEM_SRC_ANONYMOUS) {
+				for (i = 0; i < p->host_num_pages; i++) {
+					*(uint64_t *)addr = 0x0123456789ABCDEF;
+					addr += p->host_page_size;
+				}
+				break;
+			}
+
+			for (i = 0; i < p->large_num_pages; i++) {
+				/*
+				 * Write to the first host page in each large
+				 * page region, and triger break of large pages.
+				 */
+				*(uint64_t *)addr = 0x0123456789ABCDEF;
+
+				/*
+				 * Access the middle host pages in each large
+				 * page region. Since dirty logging is enabled,
+				 * this will create new mappings at the smallest
+				 * granularity.
+				 */
+				addr += p->large_page_size / 2;
+				for (j = 0; j < p->host_pages_per_lpage / 2; j++) {
+					READ_ONCE(*(uint64_t *)addr);
+					addr += p->host_page_size;
+				}
+			}
+			break;
+
+		/*
+		 * After dirty logging is stopped, vCPUs concurrently read
+		 * from every single host page. Then KVM will coalesce the
+		 * split page mappings back to block mappings. And a TLB
+		 * conflict abort could occur here if TLB entries of the
+		 * page mappings are not fully invalidated.
+		 */
+		case KVM_ADJUST_MAPPINGS:
+			for (i = 0; i < p->host_num_pages; i++) {
+				READ_ONCE(*(uint64_t *)addr);
+				addr += p->host_page_size;
+			}
+			break;
+
+		default:
+			GUEST_ASSERT(0);
+		}
+
+		GUEST_SYNC(1);
+	}
+}
+
+static void *vcpu_worker(void *data)
+{
+	int ret;
+	struct vcpu_args *vcpu_args = data;
+	struct kvm_vm *vm = test_args.vm;
+	int vcpu_id = vcpu_args->vcpu_id;
+	struct kvm_run *run;
+	struct timespec start;
+	struct timespec ts_diff;
+	enum test_stage stage;
+
+	vcpu_args_set(vm, vcpu_id, 1, vcpu_id);
+	run = vcpu_state(vm, vcpu_id);
+
+	while (!READ_ONCE(host_quit)) {
+		ret = sem_wait(&test_stage_updated);
+		TEST_ASSERT(ret == 0, "Error in sem_wait");
+
+		if (READ_ONCE(host_quit))
+			return NULL;
+
+		clock_gettime(CLOCK_MONOTONIC_RAW, &start);
+		ret = _vcpu_run(vm, vcpu_id);
+		ts_diff = timespec_elapsed(start);
+
+		TEST_ASSERT(ret == 0, "vcpu_run failed: %d\n", ret);
+		TEST_ASSERT(get_ucall(vm, vcpu_id, NULL) == UCALL_SYNC,
+			    "Invalid guest sync status: exit_reason=%s\n",
+			    exit_reason_str(run->exit_reason));
+
+		pr_debug("Got sync event from vCPU %d\n", vcpu_id);
+		stage = READ_ONCE(*current_stage);
+
+		/*
+		 * Here we can know the execution time of every
+		 * single vcpu running in different test stages.
+		 */
+		pr_debug("vCPU %d has completed stage %s\n"
+			 "execution time is: %ld.%.9lds\n\n",
+			 vcpu_id, test_stage_string[stage],
+			 ts_diff.tv_sec, ts_diff.tv_nsec);
+
+		ret = sem_post(&test_stage_completed);
+		TEST_ASSERT(ret == 0, "Error in sem_post");
+	}
+
+	return NULL;
+}
+
+struct test_params {
+	uint64_t phys_offset;
+	uint64_t test_mem_size;
+	enum vm_mem_backing_src_type src_type;
+};
+
+static struct kvm_vm *pre_init_before_test(enum vm_guest_mode mode, void *arg)
+{
+	int ret;
+	struct test_params *p = arg;
+	struct vcpu_args *vcpu_args;
+	enum vm_mem_backing_src_type src_type = p->src_type;
+	uint64_t large_page_size = get_backing_src_pagesz(src_type);
+	uint64_t guest_page_size = vm_guest_mode_params[mode].page_size;
+	uint64_t host_page_size = getpagesize();
+	uint64_t test_mem_size = p->test_mem_size;
+	uint64_t guest_num_pages;
+	uint64_t alignment;
+	void *host_test_mem;
+	struct kvm_vm *vm;
+	int vcpu_id;
+
+	/* Align up the test memory size */
+	alignment = max(large_page_size, guest_page_size);
+	test_mem_size = (test_mem_size + alignment - 1) & ~(alignment - 1);
+
+	/* Create a VM with enough guest pages */
+	guest_num_pages = test_mem_size / guest_page_size;
+	vm = vm_create_with_vcpus(mode, nr_vcpus,
+				  guest_num_pages, 0, guest_code, NULL);
+
+	/* Align down GPA of the testing memslot */
+	if (!p->phys_offset)
+		guest_test_phys_mem = (vm_get_max_gfn(vm) - guest_num_pages) *
+				       guest_page_size;
+	else
+		guest_test_phys_mem = p->phys_offset;
+#ifdef __s390x__
+	alignment = max(0x100000, alignment);
+#endif
+	guest_test_phys_mem &= ~(alignment - 1);
+
+	/* Set up the shared data structure test_args */
+	test_args.vm = vm;
+	test_args.guest_test_virt_mem = guest_test_virt_mem;
+	test_args.host_page_size = host_page_size;
+	test_args.host_num_pages = test_mem_size / host_page_size;
+	test_args.large_page_size = large_page_size;
+	test_args.large_num_pages = test_mem_size / large_page_size;
+	test_args.host_pages_per_lpage = large_page_size / host_page_size;
+	test_args.src_type = src_type;
+
+	for (vcpu_id = 0; vcpu_id < KVM_MAX_VCPUS; vcpu_id++) {
+		vcpu_args = &test_args.vcpu_args[vcpu_id];
+		vcpu_args->vcpu_id = vcpu_id;
+		vcpu_args->vcpu_write = !(vcpu_id % 2);
+	}
+
+	/* Add an extra memory slot with specified backing src type */
+	vm_userspace_mem_region_add(vm, src_type, guest_test_phys_mem,
+				    TEST_MEM_SLOT_INDEX, guest_num_pages, 0);
+
+	/* Do mapping(GVA->GPA) for the testing memory slot */
+	virt_map(vm, guest_test_virt_mem, guest_test_phys_mem, guest_num_pages, 0);
+
+	/* Cache the HVA pointer of the region */
+	host_test_mem = addr_gpa2hva(vm, (vm_paddr_t)guest_test_phys_mem);
+
+	/* Export shared structure test_args to guest */
+	ucall_init(vm, NULL);
+	sync_global_to_guest(vm, test_args);
+
+	ret = sem_init(&test_stage_updated, 0, 0);
+	TEST_ASSERT(ret == 0, "Error in sem_init");
+
+	ret = sem_init(&test_stage_completed, 0, 0);
+	TEST_ASSERT(ret == 0, "Error in sem_init");
+
+	current_stage = addr_gva2hva(vm, (vm_vaddr_t)(&guest_test_stage));
+	*current_stage = NUM_TEST_STAGES;
+
+	pr_info("Testing guest mode: %s\n", vm_guest_mode_string(mode));
+	pr_info("Testing memory backing src type: %s\n",
+		vm_mem_backing_src_alias(src_type)->name);
+	pr_info("Testing memory backing src granularity: 0x%lx\n",
+		large_page_size);
+	pr_info("Testing memory size(aligned): 0x%lx\n", test_mem_size);
+	pr_info("Guest physical test memory offset: 0x%lx\n",
+		guest_test_phys_mem);
+	pr_info("Host  virtual  test memory offset: 0x%lx\n",
+		(uint64_t)host_test_mem);
+	pr_info("Number of testing vCPUs: %d\n", nr_vcpus);
+
+	return vm;
+}
+
+static void vcpus_complete_new_stage(enum test_stage stage)
+{
+	int ret;
+	int vcpus;
+
+	/* Wake up all the vcpus to run new test stage */
+	for (vcpus = 0; vcpus < nr_vcpus; vcpus++) {
+		ret = sem_post(&test_stage_updated);
+		TEST_ASSERT(ret == 0, "Error in sem_post");
+	}
+	pr_debug("All vcpus have been notified to continue\n");
+
+	/* Wait for all the vcpus to complete new test stage */
+	for (vcpus = 0; vcpus < nr_vcpus; vcpus++) {
+		ret = sem_wait(&test_stage_completed);
+		TEST_ASSERT(ret == 0, "Error in sem_wait");
+
+		pr_debug("%d vcpus have completed stage %s\n",
+			 vcpus + 1, test_stage_string[stage]);
+	}
+
+	pr_debug("All vcpus have completed stage %s\n",
+		 test_stage_string[stage]);
+}
+
+static void run_test(enum vm_guest_mode mode, void *arg)
+{
+	int ret;
+	pthread_t *vcpu_threads;
+	struct kvm_vm *vm;
+	int vcpu_id;
+	struct timespec start;
+	struct timespec ts_diff;
+
+	/* Create VM with vCPUs and make some pre-initialization */
+	vm = pre_init_before_test(mode, arg);
+
+	vcpu_threads = malloc(nr_vcpus * sizeof(*vcpu_threads));
+	TEST_ASSERT(vcpu_threads, "Memory allocation failed");
+
+	host_quit = false;
+	*current_stage = KVM_BEFORE_MAPPINGS;
+
+	for (vcpu_id = 0; vcpu_id < nr_vcpus; vcpu_id++) {
+		pthread_create(&vcpu_threads[vcpu_id], NULL, vcpu_worker,
+			       &test_args.vcpu_args[vcpu_id]);
+	}
+
+	vcpus_complete_new_stage(*current_stage);
+	pr_info("Started all vCPUs successfully\n");
+
+	/* Test the stage of KVM creating mappings */
+	*current_stage = KVM_CREATE_MAPPINGS;
+
+	clock_gettime(CLOCK_MONOTONIC_RAW, &start);
+	vcpus_complete_new_stage(*current_stage);
+	ts_diff = timespec_elapsed(start);
+
+	pr_info("KVM_CREATE_MAPPINGS: total execution time: %ld.%.9lds\n\n",
+		ts_diff.tv_sec, ts_diff.tv_nsec);
+
+	/* Test the stage of KVM updating mappings */
+	vm_mem_region_set_flags(vm, TEST_MEM_SLOT_INDEX,
+				KVM_MEM_LOG_DIRTY_PAGES);
+
+	*current_stage = KVM_UPDATE_MAPPINGS;
+
+	clock_gettime(CLOCK_MONOTONIC_RAW, &start);
+	vcpus_complete_new_stage(*current_stage);
+	ts_diff = timespec_elapsed(start);
+
+	pr_info("KVM_UPDATE_MAPPINGS: total execution time: %ld.%.9lds\n\n",
+		ts_diff.tv_sec, ts_diff.tv_nsec);
+
+	/* Test the stage of KVM adjusting mappings */
+	vm_mem_region_set_flags(vm, TEST_MEM_SLOT_INDEX, 0);
+
+	*current_stage = KVM_ADJUST_MAPPINGS;
+
+	clock_gettime(CLOCK_MONOTONIC_RAW, &start);
+	vcpus_complete_new_stage(*current_stage);
+	ts_diff = timespec_elapsed(start);
+
+	pr_info("KVM_ADJUST_MAPPINGS: total execution time: %ld.%.9lds\n\n",
+		ts_diff.tv_sec, ts_diff.tv_nsec);
+
+	/* Tell the vcpu thread to quit */
+	host_quit = true;
+	for (vcpu_id = 0; vcpu_id < nr_vcpus; vcpu_id++) {
+		ret = sem_post(&test_stage_updated);
+		TEST_ASSERT(ret == 0, "Error in sem_post");
+	}
+
+	for (vcpu_id = 0; vcpu_id < nr_vcpus; vcpu_id++)
+		pthread_join(vcpu_threads[vcpu_id], NULL);
+
+	ret = sem_destroy(&test_stage_updated);
+	TEST_ASSERT(ret == 0, "Error in sem_destroy");
+
+	ret = sem_destroy(&test_stage_completed);
+	TEST_ASSERT(ret == 0, "Error in sem_destroy");
+
+	free(vcpu_threads);
+	ucall_uninit(vm);
+	kvm_vm_free(vm);
+}
+
+static void help(char *name)
+{
+	puts("");
+	printf("usage: %s [-h] [-p offset] [-m mode] "
+	       "[-b mem-size] [-v vcpus] [-s mem-type]\n", name);
+	puts("");
+	printf(" -p: specify guest physical test memory offset\n"
+	       "     Warning: a low offset can conflict with the loaded test code.\n");
+	guest_modes_help();
+	printf(" -b: specify size of the memory region for testing. e.g. 10M or 3G.\n"
+	       "     (default: 1G)\n");
+	printf(" -v: specify the number of vCPUs to run\n"
+	       "     (default: 1)\n");
+	printf(" -s: specify the type of memory that should be used to\n"
+	       "     back the guest data region.\n"
+	       "     (default: anonymous)\n\n");
+	backing_src_help();
+	puts("");
+}
+
+int main(int argc, char *argv[])
+{
+	int max_vcpus = kvm_check_cap(KVM_CAP_MAX_VCPUS);
+	struct test_params p = {
+		.test_mem_size = DEFAULT_TEST_MEM_SIZE,
+		.src_type = VM_MEM_SRC_ANONYMOUS,
+	};
+	int opt;
+
+	guest_modes_append_default();
+
+	while ((opt = getopt(argc, argv, "hp:m:b:v:s:")) != -1) {
+		switch (opt) {
+		case 'p':
+			p.phys_offset = strtoull(optarg, NULL, 0);
+			break;
+		case 'm':
+			guest_modes_cmdline(optarg);
+			break;
+		case 'b':
+			p.test_mem_size = parse_size(optarg);
+			break;
+		case 'v':
+			nr_vcpus = atoi(optarg);
+			TEST_ASSERT(nr_vcpus > 0 && nr_vcpus <= max_vcpus,
+				    "Invalid number of vcpus, must be between 1 and %d", max_vcpus);
+			break;
+		case 's':
+			p.src_type = parse_backing_src_type(optarg);
+			break;
+		case 'h':
+		default:
+			help(argv[0]);
+			exit(0);
+		}
+	}
+
+	for_each_guest_mode(run_test, &p);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/kvm/lib/assert.c b/tools/testing/selftests/kvm/lib/assert.c
index 5ebbd0d6b472..71ade6100fd3 100644
--- a/tools/testing/selftests/kvm/lib/assert.c
+++ b/tools/testing/selftests/kvm/lib/assert.c
@@ -71,9 +71,9 @@ test_assert(bool exp, const char *exp_str,
 
 		fprintf(stderr, "==== Test Assertion Failure ====\n"
 			"  %s:%u: %s\n"
-			"  pid=%d tid=%d - %s\n",
+			"  pid=%d tid=%d errno=%d - %s\n",
 			file, line, exp_str, getpid(), _gettid(),
-			strerror(errno));
+			errno, strerror(errno));
 		test_dump_stack();
 		if (fmt) {
 			fputs("  ", stderr);
diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c
index b8849a1aca79..28e528c19d28 100644
--- a/tools/testing/selftests/kvm/lib/kvm_util.c
+++ b/tools/testing/selftests/kvm/lib/kvm_util.c
@@ -18,7 +18,6 @@
 #include <unistd.h>
 #include <linux/kernel.h>
 
-#define KVM_UTIL_PGS_PER_HUGEPG 512
 #define KVM_UTIL_MIN_PFN	2
 
 static int vcpu_mmap_sz(void);
@@ -33,6 +32,34 @@ static void *align(void *x, size_t size)
 }
 
 /*
+ * Open KVM_DEV_PATH if available, otherwise exit the entire program.
+ *
+ * Input Args:
+ *   flags - The flags to pass when opening KVM_DEV_PATH.
+ *
+ * Return:
+ *   The opened file descriptor of /dev/kvm.
+ */
+static int _open_kvm_dev_path_or_exit(int flags)
+{
+	int fd;
+
+	fd = open(KVM_DEV_PATH, flags);
+	if (fd < 0) {
+		print_skip("%s not available, is KVM loaded? (errno: %d)",
+			   KVM_DEV_PATH, errno);
+		exit(KSFT_SKIP);
+	}
+
+	return fd;
+}
+
+int open_kvm_dev_path_or_exit(void)
+{
+	return _open_kvm_dev_path_or_exit(O_RDONLY);
+}
+
+/*
  * Capability
  *
  * Input Args:
@@ -53,10 +80,7 @@ int kvm_check_cap(long cap)
 	int ret;
 	int kvm_fd;
 
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
-
+	kvm_fd = open_kvm_dev_path_or_exit();
 	ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
 	TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
 		"  rc: %i errno: %i", ret, errno);
@@ -129,9 +153,7 @@ void vm_enable_dirty_ring(struct kvm_vm *vm, uint32_t ring_size)
 
 static void vm_open(struct kvm_vm *vm, int perm)
 {
-	vm->kvm_fd = open(KVM_DEV_PATH, perm);
-	if (vm->kvm_fd < 0)
-		exit(KSFT_SKIP);
+	vm->kvm_fd = _open_kvm_dev_path_or_exit(perm);
 
 	if (!kvm_check_cap(KVM_CAP_IMMEDIATE_EXIT)) {
 		print_skip("immediate_exit not available");
@@ -143,17 +165,24 @@ static void vm_open(struct kvm_vm *vm, int perm)
 		"rc: %i errno: %i", vm->fd, errno);
 }
 
-const char * const vm_guest_mode_string[] = {
-	"PA-bits:52,  VA-bits:48,  4K pages",
-	"PA-bits:52,  VA-bits:48, 64K pages",
-	"PA-bits:48,  VA-bits:48,  4K pages",
-	"PA-bits:48,  VA-bits:48, 64K pages",
-	"PA-bits:40,  VA-bits:48,  4K pages",
-	"PA-bits:40,  VA-bits:48, 64K pages",
-	"PA-bits:ANY, VA-bits:48,  4K pages",
-};
-_Static_assert(sizeof(vm_guest_mode_string)/sizeof(char *) == NUM_VM_MODES,
-	       "Missing new mode strings?");
+const char *vm_guest_mode_string(uint32_t i)
+{
+	static const char * const strings[] = {
+		[VM_MODE_P52V48_4K]	= "PA-bits:52,  VA-bits:48,  4K pages",
+		[VM_MODE_P52V48_64K]	= "PA-bits:52,  VA-bits:48, 64K pages",
+		[VM_MODE_P48V48_4K]	= "PA-bits:48,  VA-bits:48,  4K pages",
+		[VM_MODE_P48V48_64K]	= "PA-bits:48,  VA-bits:48, 64K pages",
+		[VM_MODE_P40V48_4K]	= "PA-bits:40,  VA-bits:48,  4K pages",
+		[VM_MODE_P40V48_64K]	= "PA-bits:40,  VA-bits:48, 64K pages",
+		[VM_MODE_PXXV48_4K]	= "PA-bits:ANY, VA-bits:48,  4K pages",
+	};
+	_Static_assert(sizeof(strings)/sizeof(char *) == NUM_VM_MODES,
+		       "Missing new mode strings?");
+
+	TEST_ASSERT(i < NUM_VM_MODES, "Guest mode ID %d too big", i);
+
+	return strings[i];
+}
 
 const struct vm_guest_mode_params vm_guest_mode_params[] = {
 	{ 52, 48,  0x1000, 12 },
@@ -197,7 +226,9 @@ struct kvm_vm *vm_create(enum vm_guest_mode mode, uint64_t phy_pages, int perm)
 	TEST_ASSERT(vm != NULL, "Insufficient Memory");
 
 	INIT_LIST_HEAD(&vm->vcpus);
-	INIT_LIST_HEAD(&vm->userspace_mem_regions);
+	vm->regions.gpa_tree = RB_ROOT;
+	vm->regions.hva_tree = RB_ROOT;
+	hash_init(vm->regions.slot_hash);
 
 	vm->mode = mode;
 	vm->type = 0;
@@ -289,7 +320,7 @@ struct kvm_vm *vm_create_with_vcpus(enum vm_guest_mode mode, uint32_t nr_vcpus,
 	 */
 	uint64_t vcpu_pages = (DEFAULT_STACK_PGS + num_percpu_pages) * nr_vcpus;
 	uint64_t extra_pg_pages = (extra_mem_pages + vcpu_pages) / PTES_PER_MIN_PAGE * 2;
-	uint64_t pages = DEFAULT_GUEST_PHY_PAGES + vcpu_pages + extra_pg_pages;
+	uint64_t pages = DEFAULT_GUEST_PHY_PAGES + extra_mem_pages + vcpu_pages + extra_pg_pages;
 	struct kvm_vm *vm;
 	int i;
 
@@ -349,13 +380,14 @@ struct kvm_vm *vm_create_default(uint32_t vcpuid, uint64_t extra_mem_pages,
  */
 void kvm_vm_restart(struct kvm_vm *vmp, int perm)
 {
+	int ctr;
 	struct userspace_mem_region *region;
 
 	vm_open(vmp, perm);
 	if (vmp->has_irqchip)
 		vm_create_irqchip(vmp);
 
-	list_for_each_entry(region, &vmp->userspace_mem_regions, list) {
+	hash_for_each(vmp->regions.slot_hash, ctr, region, slot_node) {
 		int ret = ioctl(vmp->fd, KVM_SET_USER_MEMORY_REGION, &region->region);
 		TEST_ASSERT(ret == 0, "KVM_SET_USER_MEMORY_REGION IOCTL failed,\n"
 			    "  rc: %i errno: %i\n"
@@ -418,14 +450,21 @@ uint32_t kvm_vm_reset_dirty_ring(struct kvm_vm *vm)
 static struct userspace_mem_region *
 userspace_mem_region_find(struct kvm_vm *vm, uint64_t start, uint64_t end)
 {
-	struct userspace_mem_region *region;
+	struct rb_node *node;
 
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
+	for (node = vm->regions.gpa_tree.rb_node; node; ) {
+		struct userspace_mem_region *region =
+			container_of(node, struct userspace_mem_region, gpa_node);
 		uint64_t existing_start = region->region.guest_phys_addr;
 		uint64_t existing_end = region->region.guest_phys_addr
 			+ region->region.memory_size - 1;
 		if (start <= existing_end && end >= existing_start)
 			return region;
+
+		if (start < existing_start)
+			node = node->rb_left;
+		else
+			node = node->rb_right;
 	}
 
 	return NULL;
@@ -514,7 +553,7 @@ static void vm_vcpu_rm(struct kvm_vm *vm, struct vcpu *vcpu)
 	ret = munmap(vcpu->state, vcpu_mmap_sz());
 	TEST_ASSERT(ret == 0, "munmap of VCPU fd failed, rc: %i "
 		"errno: %i", ret, errno);
-	close(vcpu->fd);
+	ret = close(vcpu->fd);
 	TEST_ASSERT(ret == 0, "Close of VCPU fd failed, rc: %i "
 		"errno: %i", ret, errno);
 
@@ -534,17 +573,22 @@ void kvm_vm_release(struct kvm_vm *vmp)
 	TEST_ASSERT(ret == 0, "Close of vm fd failed,\n"
 		"  vmp->fd: %i rc: %i errno: %i", vmp->fd, ret, errno);
 
-	close(vmp->kvm_fd);
+	ret = close(vmp->kvm_fd);
 	TEST_ASSERT(ret == 0, "Close of /dev/kvm fd failed,\n"
 		"  vmp->kvm_fd: %i rc: %i errno: %i", vmp->kvm_fd, ret, errno);
 }
 
 static void __vm_mem_region_delete(struct kvm_vm *vm,
-				   struct userspace_mem_region *region)
+				   struct userspace_mem_region *region,
+				   bool unlink)
 {
 	int ret;
 
-	list_del(&region->list);
+	if (unlink) {
+		rb_erase(&region->gpa_node, &vm->regions.gpa_tree);
+		rb_erase(&region->hva_node, &vm->regions.hva_tree);
+		hash_del(&region->slot_node);
+	}
 
 	region->region.memory_size = 0;
 	ret = ioctl(vm->fd, KVM_SET_USER_MEMORY_REGION, &region->region);
@@ -563,14 +607,16 @@ static void __vm_mem_region_delete(struct kvm_vm *vm,
  */
 void kvm_vm_free(struct kvm_vm *vmp)
 {
-	struct userspace_mem_region *region, *tmp;
+	int ctr;
+	struct hlist_node *node;
+	struct userspace_mem_region *region;
 
 	if (vmp == NULL)
 		return;
 
 	/* Free userspace_mem_regions. */
-	list_for_each_entry_safe(region, tmp, &vmp->userspace_mem_regions, list)
-		__vm_mem_region_delete(vmp, region);
+	hash_for_each_safe(vmp->regions.slot_hash, ctr, node, region, slot_node)
+		__vm_mem_region_delete(vmp, region, false);
 
 	/* Free sparsebit arrays. */
 	sparsebit_free(&vmp->vpages_valid);
@@ -652,13 +698,64 @@ int kvm_memcmp_hva_gva(void *hva, struct kvm_vm *vm, vm_vaddr_t gva, size_t len)
 	return 0;
 }
 
+static void vm_userspace_mem_region_gpa_insert(struct rb_root *gpa_tree,
+					       struct userspace_mem_region *region)
+{
+	struct rb_node **cur, *parent;
+
+	for (cur = &gpa_tree->rb_node, parent = NULL; *cur; ) {
+		struct userspace_mem_region *cregion;
+
+		cregion = container_of(*cur, typeof(*cregion), gpa_node);
+		parent = *cur;
+		if (region->region.guest_phys_addr <
+		    cregion->region.guest_phys_addr)
+			cur = &(*cur)->rb_left;
+		else {
+			TEST_ASSERT(region->region.guest_phys_addr !=
+				    cregion->region.guest_phys_addr,
+				    "Duplicate GPA in region tree");
+
+			cur = &(*cur)->rb_right;
+		}
+	}
+
+	rb_link_node(&region->gpa_node, parent, cur);
+	rb_insert_color(&region->gpa_node, gpa_tree);
+}
+
+static void vm_userspace_mem_region_hva_insert(struct rb_root *hva_tree,
+					       struct userspace_mem_region *region)
+{
+	struct rb_node **cur, *parent;
+
+	for (cur = &hva_tree->rb_node, parent = NULL; *cur; ) {
+		struct userspace_mem_region *cregion;
+
+		cregion = container_of(*cur, typeof(*cregion), hva_node);
+		parent = *cur;
+		if (region->host_mem < cregion->host_mem)
+			cur = &(*cur)->rb_left;
+		else {
+			TEST_ASSERT(region->host_mem !=
+				    cregion->host_mem,
+				    "Duplicate HVA in region tree");
+
+			cur = &(*cur)->rb_right;
+		}
+	}
+
+	rb_link_node(&region->hva_node, parent, cur);
+	rb_insert_color(&region->hva_node, hva_tree);
+}
+
 /*
  * VM Userspace Memory Region Add
  *
  * Input Args:
  *   vm - Virtual Machine
- *   backing_src - Storage source for this region.
- *                 NULL to use anonymous memory.
+ *   src_type - Storage source for this region.
+ *              NULL to use anonymous memory.
  *   guest_paddr - Starting guest physical address
  *   slot - KVM region slot
  *   npages - Number of physical pages
@@ -681,7 +778,7 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm,
 {
 	int ret;
 	struct userspace_mem_region *region;
-	size_t huge_page_size = KVM_UTIL_PGS_PER_HUGEPG * vm->page_size;
+	size_t backing_src_pagesz = get_backing_src_pagesz(src_type);
 	size_t alignment;
 
 	TEST_ASSERT(vm_adjust_num_guest_pages(vm->mode, npages) == npages,
@@ -716,7 +813,8 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm,
 			(uint64_t) region->region.memory_size);
 
 	/* Confirm no region with the requested slot already exists. */
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
+	hash_for_each_possible(vm->regions.slot_hash, region, slot_node,
+			       slot) {
 		if (region->region.slot != slot)
 			continue;
 
@@ -743,17 +841,36 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm,
 #endif
 
 	if (src_type == VM_MEM_SRC_ANONYMOUS_THP)
-		alignment = max(huge_page_size, alignment);
+		alignment = max(backing_src_pagesz, alignment);
 
 	/* Add enough memory to align up if necessary */
 	if (alignment > 1)
 		region->mmap_size += alignment;
 
+	region->fd = -1;
+	if (backing_src_is_shared(src_type)) {
+		int memfd_flags = MFD_CLOEXEC;
+
+		if (src_type == VM_MEM_SRC_SHARED_HUGETLB)
+			memfd_flags |= MFD_HUGETLB;
+
+		region->fd = memfd_create("kvm_selftest", memfd_flags);
+		TEST_ASSERT(region->fd != -1,
+			    "memfd_create failed, errno: %i", errno);
+
+		ret = ftruncate(region->fd, region->mmap_size);
+		TEST_ASSERT(ret == 0, "ftruncate failed, errno: %i", errno);
+
+		ret = fallocate(region->fd,
+				FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, 0,
+				region->mmap_size);
+		TEST_ASSERT(ret == 0, "fallocate failed, errno: %i", errno);
+	}
+
 	region->mmap_start = mmap(NULL, region->mmap_size,
 				  PROT_READ | PROT_WRITE,
-				  MAP_PRIVATE | MAP_ANONYMOUS
-				  | (src_type == VM_MEM_SRC_ANONYMOUS_HUGETLB ? MAP_HUGETLB : 0),
-				  -1, 0);
+				  vm_mem_backing_src_alias(src_type)->flag,
+				  region->fd, 0);
 	TEST_ASSERT(region->mmap_start != MAP_FAILED,
 		    "test_malloc failed, mmap_start: %p errno: %i",
 		    region->mmap_start, errno);
@@ -762,22 +879,13 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm,
 	region->host_mem = align(region->mmap_start, alignment);
 
 	/* As needed perform madvise */
-	if (src_type == VM_MEM_SRC_ANONYMOUS || src_type == VM_MEM_SRC_ANONYMOUS_THP) {
-		struct stat statbuf;
-
-		ret = stat("/sys/kernel/mm/transparent_hugepage", &statbuf);
-		TEST_ASSERT(ret == 0 || (ret == -1 && errno == ENOENT),
-			    "stat /sys/kernel/mm/transparent_hugepage");
-
-		TEST_ASSERT(ret == 0 || src_type != VM_MEM_SRC_ANONYMOUS_THP,
-			    "VM_MEM_SRC_ANONYMOUS_THP requires THP to be configured in the host kernel");
-
-		if (ret == 0) {
-			ret = madvise(region->host_mem, npages * vm->page_size,
-				      src_type == VM_MEM_SRC_ANONYMOUS ? MADV_NOHUGEPAGE : MADV_HUGEPAGE);
-			TEST_ASSERT(ret == 0, "madvise failed, addr: %p length: 0x%lx src_type: %x",
-				    region->host_mem, npages * vm->page_size, src_type);
-		}
+	if ((src_type == VM_MEM_SRC_ANONYMOUS ||
+	     src_type == VM_MEM_SRC_ANONYMOUS_THP) && thp_configured()) {
+		ret = madvise(region->host_mem, npages * vm->page_size,
+			      src_type == VM_MEM_SRC_ANONYMOUS ? MADV_NOHUGEPAGE : MADV_HUGEPAGE);
+		TEST_ASSERT(ret == 0, "madvise failed, addr: %p length: 0x%lx src_type: %s",
+			    region->host_mem, npages * vm->page_size,
+			    vm_mem_backing_src_alias(src_type)->name);
 	}
 
 	region->unused_phy_pages = sparsebit_alloc();
@@ -796,8 +904,23 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm,
 		ret, errno, slot, flags,
 		guest_paddr, (uint64_t) region->region.memory_size);
 
-	/* Add to linked-list of memory regions. */
-	list_add(&region->list, &vm->userspace_mem_regions);
+	/* Add to quick lookup data structures */
+	vm_userspace_mem_region_gpa_insert(&vm->regions.gpa_tree, region);
+	vm_userspace_mem_region_hva_insert(&vm->regions.hva_tree, region);
+	hash_add(vm->regions.slot_hash, &region->slot_node, slot);
+
+	/* If shared memory, create an alias. */
+	if (region->fd >= 0) {
+		region->mmap_alias = mmap(NULL, region->mmap_size,
+					  PROT_READ | PROT_WRITE,
+					  vm_mem_backing_src_alias(src_type)->flag,
+					  region->fd, 0);
+		TEST_ASSERT(region->mmap_alias != MAP_FAILED,
+			    "mmap of alias failed, errno: %i", errno);
+
+		/* Align host alias address */
+		region->host_alias = align(region->mmap_alias, alignment);
+	}
 }
 
 /*
@@ -820,10 +943,10 @@ memslot2region(struct kvm_vm *vm, uint32_t memslot)
 {
 	struct userspace_mem_region *region;
 
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
+	hash_for_each_possible(vm->regions.slot_hash, region, slot_node,
+			       memslot)
 		if (region->region.slot == memslot)
 			return region;
-	}
 
 	fprintf(stderr, "No mem region with the requested slot found,\n"
 		"  requested slot: %u\n", memslot);
@@ -908,7 +1031,7 @@ void vm_mem_region_move(struct kvm_vm *vm, uint32_t slot, uint64_t new_gpa)
  */
 void vm_mem_region_delete(struct kvm_vm *vm, uint32_t slot)
 {
-	__vm_mem_region_delete(vm, memslot2region(vm, slot));
+	__vm_mem_region_delete(vm, memslot2region(vm, slot), true);
 }
 
 /*
@@ -928,9 +1051,7 @@ static int vcpu_mmap_sz(void)
 {
 	int dev_fd, ret;
 
-	dev_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (dev_fd < 0)
-		exit(KSFT_SKIP);
+	dev_fd = open_kvm_dev_path_or_exit();
 
 	ret = ioctl(dev_fd, KVM_GET_VCPU_MMAP_SIZE, NULL);
 	TEST_ASSERT(ret >= sizeof(struct kvm_run),
@@ -1102,6 +1223,9 @@ vm_vaddr_t vm_vaddr_alloc(struct kvm_vm *vm, size_t sz, vm_vaddr_t vaddr_min,
 	uint64_t pages = (sz >> vm->page_shift) + ((sz % vm->page_size) != 0);
 
 	virt_pgd_alloc(vm, pgd_memslot);
+	vm_paddr_t paddr = vm_phy_pages_alloc(vm, pages,
+					      KVM_UTIL_MIN_PFN * vm->page_size,
+					      data_memslot);
 
 	/*
 	 * Find an unused range of virtual page addresses of at least
@@ -1111,11 +1235,7 @@ vm_vaddr_t vm_vaddr_alloc(struct kvm_vm *vm, size_t sz, vm_vaddr_t vaddr_min,
 
 	/* Map the virtual pages. */
 	for (vm_vaddr_t vaddr = vaddr_start; pages > 0;
-		pages--, vaddr += vm->page_size) {
-		vm_paddr_t paddr;
-
-		paddr = vm_phy_page_alloc(vm,
-				KVM_UTIL_MIN_PFN * vm->page_size, data_memslot);
+		pages--, vaddr += vm->page_size, paddr += vm->page_size) {
 
 		virt_pg_map(vm, vaddr, paddr, pgd_memslot);
 
@@ -1180,16 +1300,14 @@ void *addr_gpa2hva(struct kvm_vm *vm, vm_paddr_t gpa)
 {
 	struct userspace_mem_region *region;
 
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
-		if ((gpa >= region->region.guest_phys_addr)
-			&& (gpa <= (region->region.guest_phys_addr
-				+ region->region.memory_size - 1)))
-			return (void *) ((uintptr_t) region->host_mem
-				+ (gpa - region->region.guest_phys_addr));
+	region = userspace_mem_region_find(vm, gpa, gpa);
+	if (!region) {
+		TEST_FAIL("No vm physical memory at 0x%lx", gpa);
+		return NULL;
 	}
 
-	TEST_FAIL("No vm physical memory at 0x%lx", gpa);
-	return NULL;
+	return (void *)((uintptr_t)region->host_mem
+		+ (gpa - region->region.guest_phys_addr));
 }
 
 /*
@@ -1211,15 +1329,22 @@ void *addr_gpa2hva(struct kvm_vm *vm, vm_paddr_t gpa)
  */
 vm_paddr_t addr_hva2gpa(struct kvm_vm *vm, void *hva)
 {
-	struct userspace_mem_region *region;
+	struct rb_node *node;
+
+	for (node = vm->regions.hva_tree.rb_node; node; ) {
+		struct userspace_mem_region *region =
+			container_of(node, struct userspace_mem_region, hva_node);
+
+		if (hva >= region->host_mem) {
+			if (hva <= (region->host_mem
+				+ region->region.memory_size - 1))
+				return (vm_paddr_t)((uintptr_t)
+					region->region.guest_phys_addr
+					+ (hva - (uintptr_t)region->host_mem));
 
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
-		if ((hva >= region->host_mem)
-			&& (hva <= (region->host_mem
-				+ region->region.memory_size - 1)))
-			return (vm_paddr_t) ((uintptr_t)
-				region->region.guest_phys_addr
-				+ (hva - (uintptr_t) region->host_mem));
+			node = node->rb_right;
+		} else
+			node = node->rb_left;
 	}
 
 	TEST_FAIL("No mapping to a guest physical address, hva: %p", hva);
@@ -1227,6 +1352,42 @@ vm_paddr_t addr_hva2gpa(struct kvm_vm *vm, void *hva)
 }
 
 /*
+ * Address VM physical to Host Virtual *alias*.
+ *
+ * Input Args:
+ *   vm - Virtual Machine
+ *   gpa - VM physical address
+ *
+ * Output Args: None
+ *
+ * Return:
+ *   Equivalent address within the host virtual *alias* area, or NULL
+ *   (without failing the test) if the guest memory is not shared (so
+ *   no alias exists).
+ *
+ * When vm_create() and related functions are called with a shared memory
+ * src_type, we also create a writable, shared alias mapping of the
+ * underlying guest memory. This allows the host to manipulate guest memory
+ * without mapping that memory in the guest's address space. And, for
+ * userfaultfd-based demand paging, we can do so without triggering userfaults.
+ */
+void *addr_gpa2alias(struct kvm_vm *vm, vm_paddr_t gpa)
+{
+	struct userspace_mem_region *region;
+	uintptr_t offset;
+
+	region = userspace_mem_region_find(vm, gpa, gpa);
+	if (!region)
+		return NULL;
+
+	if (!region->host_alias)
+		return NULL;
+
+	offset = gpa - region->region.guest_phys_addr;
+	return (void *) ((uintptr_t) region->host_alias + offset);
+}
+
+/*
  * VM Create IRQ Chip
  *
  * Input Args:
@@ -1734,6 +1895,81 @@ int _kvm_ioctl(struct kvm_vm *vm, unsigned long cmd, void *arg)
 }
 
 /*
+ * Device Ioctl
+ */
+
+int _kvm_device_check_attr(int dev_fd, uint32_t group, uint64_t attr)
+{
+	struct kvm_device_attr attribute = {
+		.group = group,
+		.attr = attr,
+		.flags = 0,
+	};
+
+	return ioctl(dev_fd, KVM_HAS_DEVICE_ATTR, &attribute);
+}
+
+int kvm_device_check_attr(int dev_fd, uint32_t group, uint64_t attr)
+{
+	int ret = _kvm_device_check_attr(dev_fd, group, attr);
+
+	TEST_ASSERT(ret >= 0, "KVM_HAS_DEVICE_ATTR failed, rc: %i errno: %i", ret, errno);
+	return ret;
+}
+
+int _kvm_create_device(struct kvm_vm *vm, uint64_t type, bool test, int *fd)
+{
+	struct kvm_create_device create_dev;
+	int ret;
+
+	create_dev.type = type;
+	create_dev.fd = -1;
+	create_dev.flags = test ? KVM_CREATE_DEVICE_TEST : 0;
+	ret = ioctl(vm_get_fd(vm), KVM_CREATE_DEVICE, &create_dev);
+	*fd = create_dev.fd;
+	return ret;
+}
+
+int kvm_create_device(struct kvm_vm *vm, uint64_t type, bool test)
+{
+	int fd, ret;
+
+	ret = _kvm_create_device(vm, type, test, &fd);
+
+	if (!test) {
+		TEST_ASSERT(ret >= 0,
+			    "KVM_CREATE_DEVICE IOCTL failed, rc: %i errno: %i", ret, errno);
+		return fd;
+	}
+	return ret;
+}
+
+int _kvm_device_access(int dev_fd, uint32_t group, uint64_t attr,
+		      void *val, bool write)
+{
+	struct kvm_device_attr kvmattr = {
+		.group = group,
+		.attr = attr,
+		.flags = 0,
+		.addr = (uintptr_t)val,
+	};
+	int ret;
+
+	ret = ioctl(dev_fd, write ? KVM_SET_DEVICE_ATTR : KVM_GET_DEVICE_ATTR,
+		    &kvmattr);
+	return ret;
+}
+
+int kvm_device_access(int dev_fd, uint32_t group, uint64_t attr,
+		      void *val, bool write)
+{
+	int ret = _kvm_device_access(dev_fd, group, attr, val, write);
+
+	TEST_ASSERT(ret >= 0, "KVM_SET|GET_DEVICE_ATTR IOCTL failed, rc: %i errno: %i", ret, errno);
+	return ret;
+}
+
+/*
  * VM Dump
  *
  * Input Args:
@@ -1750,6 +1986,7 @@ int _kvm_ioctl(struct kvm_vm *vm, unsigned long cmd, void *arg)
  */
 void vm_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
 {
+	int ctr;
 	struct userspace_mem_region *region;
 	struct vcpu *vcpu;
 
@@ -1757,7 +1994,7 @@ void vm_dump(FILE *stream, struct kvm_vm *vm, uint8_t indent)
 	fprintf(stream, "%*sfd: %i\n", indent, "", vm->fd);
 	fprintf(stream, "%*spage_size: 0x%x\n", indent, "", vm->page_size);
 	fprintf(stream, "%*sMem Regions:\n", indent, "");
-	list_for_each_entry(region, &vm->userspace_mem_regions, list) {
+	hash_for_each(vm->regions.slot_hash, ctr, region, slot_node) {
 		fprintf(stream, "%*sguest_phys: 0x%lx size: 0x%lx "
 			"host_virt: %p\n", indent + 2, "",
 			(uint64_t) region->region.guest_phys_addr,
@@ -1943,10 +2180,7 @@ bool vm_is_unrestricted_guest(struct kvm_vm *vm)
 
 	if (vm == NULL) {
 		/* Ensure that the KVM vendor-specific module is loaded. */
-		f = fopen(KVM_DEV_PATH, "r");
-		TEST_ASSERT(f != NULL, "Error in opening KVM dev file: %d",
-			    errno);
-		fclose(f);
+		close(open_kvm_dev_path_or_exit());
 	}
 
 	f = fopen("/sys/module/kvm_intel/parameters/unrestricted_guest", "r");
@@ -1969,7 +2203,7 @@ unsigned int vm_get_page_shift(struct kvm_vm *vm)
 	return vm->page_shift;
 }
 
-unsigned int vm_get_max_gfn(struct kvm_vm *vm)
+uint64_t vm_get_max_gfn(struct kvm_vm *vm)
 {
 	return vm->max_gfn;
 }
diff --git a/tools/testing/selftests/kvm/lib/kvm_util_internal.h b/tools/testing/selftests/kvm/lib/kvm_util_internal.h
index 91ce1b5d480b..a03febc24ba6 100644
--- a/tools/testing/selftests/kvm/lib/kvm_util_internal.h
+++ b/tools/testing/selftests/kvm/lib/kvm_util_internal.h
@@ -8,6 +8,9 @@
 #ifndef SELFTEST_KVM_UTIL_INTERNAL_H
 #define SELFTEST_KVM_UTIL_INTERNAL_H
 
+#include "linux/hashtable.h"
+#include "linux/rbtree.h"
+
 #include "sparsebit.h"
 
 struct userspace_mem_region {
@@ -16,9 +19,13 @@ struct userspace_mem_region {
 	int fd;
 	off_t offset;
 	void *host_mem;
+	void *host_alias;
 	void *mmap_start;
+	void *mmap_alias;
 	size_t mmap_size;
-	struct list_head list;
+	struct rb_node gpa_node;
+	struct rb_node hva_node;
+	struct hlist_node slot_node;
 };
 
 struct vcpu {
@@ -31,6 +38,12 @@ struct vcpu {
 	uint32_t dirty_gfns_count;
 };
 
+struct userspace_mem_regions {
+	struct rb_root gpa_tree;
+	struct rb_root hva_tree;
+	DECLARE_HASHTABLE(slot_hash, 9);
+};
+
 struct kvm_vm {
 	int mode;
 	unsigned long type;
@@ -43,7 +56,7 @@ struct kvm_vm {
 	unsigned int va_bits;
 	uint64_t max_gfn;
 	struct list_head vcpus;
-	struct list_head userspace_mem_regions;
+	struct userspace_mem_regions regions;
 	struct sparsebit *vpages_valid;
 	struct sparsebit *vpages_mapped;
 	bool has_irqchip;
diff --git a/tools/testing/selftests/kvm/lib/perf_test_util.c b/tools/testing/selftests/kvm/lib/perf_test_util.c
index 81490b9b4e32..abf381800a59 100644
--- a/tools/testing/selftests/kvm/lib/perf_test_util.c
+++ b/tools/testing/selftests/kvm/lib/perf_test_util.c
@@ -2,6 +2,7 @@
 /*
  * Copyright (C) 2020, Google LLC.
  */
+#include <inttypes.h>
 
 #include "kvm_util.h"
 #include "perf_test_util.h"
@@ -80,7 +81,8 @@ struct kvm_vm *perf_test_create_vm(enum vm_guest_mode mode, int vcpus,
 	 */
 	TEST_ASSERT(guest_num_pages < vm_get_max_gfn(vm),
 		    "Requested more guest memory than address space allows.\n"
-		    "    guest pages: %lx max gfn: %x vcpus: %d wss: %lx]\n",
+		    "    guest pages: %" PRIx64 " max gfn: %" PRIx64
+		    " vcpus: %d wss: %" PRIx64 "]\n",
 		    guest_num_pages, vm_get_max_gfn(vm), vcpus,
 		    vcpu_memory_bytes);
 
diff --git a/tools/testing/selftests/kvm/lib/rbtree.c b/tools/testing/selftests/kvm/lib/rbtree.c
new file mode 100644
index 000000000000..a703f0194ea3
--- /dev/null
+++ b/tools/testing/selftests/kvm/lib/rbtree.c
@@ -0,0 +1 @@
+#include "../../../../lib/rbtree.c"
diff --git a/tools/testing/selftests/kvm/lib/sparsebit.c b/tools/testing/selftests/kvm/lib/sparsebit.c
index 031ba3c932ed..a0d0c83d83de 100644
--- a/tools/testing/selftests/kvm/lib/sparsebit.c
+++ b/tools/testing/selftests/kvm/lib/sparsebit.c
@@ -1890,7 +1890,6 @@ void sparsebit_validate_internal(struct sparsebit *s)
  */
 
 #include <stdlib.h>
-#include <assert.h>
 
 struct range {
 	sparsebit_idx_t first, last;
diff --git a/tools/testing/selftests/kvm/lib/test_util.c b/tools/testing/selftests/kvm/lib/test_util.c
index 906c955384e2..6ad6c8276b2e 100644
--- a/tools/testing/selftests/kvm/lib/test_util.c
+++ b/tools/testing/selftests/kvm/lib/test_util.c
@@ -10,6 +10,8 @@
 #include <limits.h>
 #include <stdlib.h>
 #include <time.h>
+#include <sys/stat.h>
+#include <linux/mman.h>
 #include "linux/kernel.h"
 
 #include "test_util.h"
@@ -111,28 +113,188 @@ void print_skip(const char *fmt, ...)
 	puts(", skipping test");
 }
 
-const struct vm_mem_backing_src_alias backing_src_aliases[] = {
-	{"anonymous", VM_MEM_SRC_ANONYMOUS,},
-	{"anonymous_thp", VM_MEM_SRC_ANONYMOUS_THP,},
-	{"anonymous_hugetlb", VM_MEM_SRC_ANONYMOUS_HUGETLB,},
-};
+bool thp_configured(void)
+{
+	int ret;
+	struct stat statbuf;
+
+	ret = stat("/sys/kernel/mm/transparent_hugepage", &statbuf);
+	TEST_ASSERT(ret == 0 || (ret == -1 && errno == ENOENT),
+		    "Error in stating /sys/kernel/mm/transparent_hugepage");
+
+	return ret == 0;
+}
+
+size_t get_trans_hugepagesz(void)
+{
+	size_t size;
+	FILE *f;
+
+	TEST_ASSERT(thp_configured(), "THP is not configured in host kernel");
+
+	f = fopen("/sys/kernel/mm/transparent_hugepage/hpage_pmd_size", "r");
+	TEST_ASSERT(f != NULL, "Error in opening transparent_hugepage/hpage_pmd_size");
+
+	fscanf(f, "%ld", &size);
+	fclose(f);
+
+	return size;
+}
+
+size_t get_def_hugetlb_pagesz(void)
+{
+	char buf[64];
+	const char *tag = "Hugepagesize:";
+	FILE *f;
+
+	f = fopen("/proc/meminfo", "r");
+	TEST_ASSERT(f != NULL, "Error in opening /proc/meminfo");
+
+	while (fgets(buf, sizeof(buf), f) != NULL) {
+		if (strstr(buf, tag) == buf) {
+			fclose(f);
+			return strtoull(buf + strlen(tag), NULL, 10) << 10;
+		}
+	}
+
+	if (feof(f))
+		TEST_FAIL("HUGETLB is not configured in host kernel");
+	else
+		TEST_FAIL("Error in reading /proc/meminfo");
+
+	fclose(f);
+	return 0;
+}
+
+const struct vm_mem_backing_src_alias *vm_mem_backing_src_alias(uint32_t i)
+{
+	static const int anon_flags = MAP_PRIVATE | MAP_ANONYMOUS;
+	static const int anon_huge_flags = anon_flags | MAP_HUGETLB;
+
+	static const struct vm_mem_backing_src_alias aliases[] = {
+		[VM_MEM_SRC_ANONYMOUS] = {
+			.name = "anonymous",
+			.flag = anon_flags,
+		},
+		[VM_MEM_SRC_ANONYMOUS_THP] = {
+			.name = "anonymous_thp",
+			.flag = anon_flags,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB] = {
+			.name = "anonymous_hugetlb",
+			.flag = anon_huge_flags,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16KB] = {
+			.name = "anonymous_hugetlb_16kb",
+			.flag = anon_huge_flags | MAP_HUGE_16KB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_64KB] = {
+			.name = "anonymous_hugetlb_64kb",
+			.flag = anon_huge_flags | MAP_HUGE_64KB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_512KB] = {
+			.name = "anonymous_hugetlb_512kb",
+			.flag = anon_huge_flags | MAP_HUGE_512KB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_1MB] = {
+			.name = "anonymous_hugetlb_1mb",
+			.flag = anon_huge_flags | MAP_HUGE_1MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_2MB] = {
+			.name = "anonymous_hugetlb_2mb",
+			.flag = anon_huge_flags | MAP_HUGE_2MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_8MB] = {
+			.name = "anonymous_hugetlb_8mb",
+			.flag = anon_huge_flags | MAP_HUGE_8MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16MB] = {
+			.name = "anonymous_hugetlb_16mb",
+			.flag = anon_huge_flags | MAP_HUGE_16MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_32MB] = {
+			.name = "anonymous_hugetlb_32mb",
+			.flag = anon_huge_flags | MAP_HUGE_32MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_256MB] = {
+			.name = "anonymous_hugetlb_256mb",
+			.flag = anon_huge_flags | MAP_HUGE_256MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_512MB] = {
+			.name = "anonymous_hugetlb_512mb",
+			.flag = anon_huge_flags | MAP_HUGE_512MB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_1GB] = {
+			.name = "anonymous_hugetlb_1gb",
+			.flag = anon_huge_flags | MAP_HUGE_1GB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_2GB] = {
+			.name = "anonymous_hugetlb_2gb",
+			.flag = anon_huge_flags | MAP_HUGE_2GB,
+		},
+		[VM_MEM_SRC_ANONYMOUS_HUGETLB_16GB] = {
+			.name = "anonymous_hugetlb_16gb",
+			.flag = anon_huge_flags | MAP_HUGE_16GB,
+		},
+		[VM_MEM_SRC_SHMEM] = {
+			.name = "shmem",
+			.flag = MAP_SHARED,
+		},
+		[VM_MEM_SRC_SHARED_HUGETLB] = {
+			.name = "shared_hugetlb",
+			/*
+			 * No MAP_HUGETLB, we use MFD_HUGETLB instead. Since
+			 * we're using "file backed" memory, we need to specify
+			 * this when the FD is created, not when the area is
+			 * mapped.
+			 */
+			.flag = MAP_SHARED,
+		},
+	};
+	_Static_assert(ARRAY_SIZE(aliases) == NUM_SRC_TYPES,
+		       "Missing new backing src types?");
+
+	TEST_ASSERT(i < NUM_SRC_TYPES, "Backing src type ID %d too big", i);
+
+	return &aliases[i];
+}
+
+#define MAP_HUGE_PAGE_SIZE(x) (1ULL << ((x >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK))
+
+size_t get_backing_src_pagesz(uint32_t i)
+{
+	uint32_t flag = vm_mem_backing_src_alias(i)->flag;
+
+	switch (i) {
+	case VM_MEM_SRC_ANONYMOUS:
+	case VM_MEM_SRC_SHMEM:
+		return getpagesize();
+	case VM_MEM_SRC_ANONYMOUS_THP:
+		return get_trans_hugepagesz();
+	case VM_MEM_SRC_ANONYMOUS_HUGETLB:
+	case VM_MEM_SRC_SHARED_HUGETLB:
+		return get_def_hugetlb_pagesz();
+	default:
+		return MAP_HUGE_PAGE_SIZE(flag);
+	}
+}
 
 void backing_src_help(void)
 {
 	int i;
 
 	printf("Available backing src types:\n");
-	for (i = 0; i < ARRAY_SIZE(backing_src_aliases); i++)
-		printf("\t%s\n", backing_src_aliases[i].name);
+	for (i = 0; i < NUM_SRC_TYPES; i++)
+		printf("\t%s\n", vm_mem_backing_src_alias(i)->name);
 }
 
 enum vm_mem_backing_src_type parse_backing_src_type(const char *type_name)
 {
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(backing_src_aliases); i++)
-		if (!strcmp(type_name, backing_src_aliases[i].name))
-			return backing_src_aliases[i].type;
+	for (i = 0; i < NUM_SRC_TYPES; i++)
+		if (!strcmp(type_name, vm_mem_backing_src_alias(i)->name))
+			return i;
 
 	backing_src_help();
 	TEST_FAIL("Unknown backing src type: %s", type_name);
diff --git a/tools/testing/selftests/kvm/lib/x86_64/handlers.S b/tools/testing/selftests/kvm/lib/x86_64/handlers.S
index aaf7bc7d2ce1..7629819734af 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/handlers.S
+++ b/tools/testing/selftests/kvm/lib/x86_64/handlers.S
@@ -54,9 +54,9 @@ idt_handlers:
 	.align 8
 
 	/* Fetch current address and append it to idt_handlers. */
-	current_handler = .
+666 :
 .pushsection .rodata
-.quad current_handler
+	.quad 666b
 .popsection
 
 	.if ! \has_error
diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c
index a8906e60a108..efe235044421 100644
--- a/tools/testing/selftests/kvm/lib/x86_64/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c
@@ -657,9 +657,7 @@ struct kvm_cpuid2 *kvm_get_supported_cpuid(void)
 		return cpuid;
 
 	cpuid = allocate_kvm_cpuid2();
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	ret = ioctl(kvm_fd, KVM_GET_SUPPORTED_CPUID, cpuid);
 	TEST_ASSERT(ret == 0, "KVM_GET_SUPPORTED_CPUID failed %d %d\n",
@@ -691,9 +689,7 @@ uint64_t kvm_get_feature_msr(uint64_t msr_index)
 
 	buffer.header.nmsrs = 1;
 	buffer.entry.index = msr_index;
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	r = ioctl(kvm_fd, KVM_GET_MSRS, &buffer.header);
 	TEST_ASSERT(r == 1, "KVM_GET_MSRS IOCTL failed,\n"
@@ -986,9 +982,7 @@ struct kvm_msr_list *kvm_get_msr_index_list(void)
 	struct kvm_msr_list *list;
 	int nmsrs, r, kvm_fd;
 
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	nmsrs = kvm_get_num_msrs_fd(kvm_fd);
 	list = malloc(sizeof(*list) + nmsrs * sizeof(list->indices[0]));
@@ -1312,9 +1306,7 @@ struct kvm_cpuid2 *kvm_get_supported_hv_cpuid(void)
 		return cpuid;
 
 	cpuid = allocate_kvm_cpuid2();
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	ret = ioctl(kvm_fd, KVM_GET_SUPPORTED_HV_CPUID, cpuid);
 	TEST_ASSERT(ret == 0, "KVM_GET_SUPPORTED_HV_CPUID failed %d %d\n",
diff --git a/tools/testing/selftests/kvm/memslot_modification_stress_test.c b/tools/testing/selftests/kvm/memslot_modification_stress_test.c
index 6096bf0a5b34..98351ba0933c 100644
--- a/tools/testing/selftests/kvm/memslot_modification_stress_test.c
+++ b/tools/testing/selftests/kvm/memslot_modification_stress_test.c
@@ -71,14 +71,22 @@ struct memslot_antagonist_args {
 };
 
 static void add_remove_memslot(struct kvm_vm *vm, useconds_t delay,
-			      uint64_t nr_modifications, uint64_t gpa)
+			       uint64_t nr_modifications)
 {
+	const uint64_t pages = 1;
+	uint64_t gpa;
 	int i;
 
+	/*
+	 * Add the dummy memslot just below the perf_test_util memslot, which is
+	 * at the top of the guest physical address space.
+	 */
+	gpa = guest_test_phys_mem - pages * vm_get_page_size(vm);
+
 	for (i = 0; i < nr_modifications; i++) {
 		usleep(delay);
 		vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, gpa,
-					    DUMMY_MEMSLOT_INDEX, 1, 0);
+					    DUMMY_MEMSLOT_INDEX, pages, 0);
 
 		vm_mem_region_delete(vm, DUMMY_MEMSLOT_INDEX);
 	}
@@ -120,11 +128,7 @@ static void run_test(enum vm_guest_mode mode, void *arg)
 	pr_info("Started all vCPUs\n");
 
 	add_remove_memslot(vm, p->memslot_modification_delay,
-			   p->nr_memslot_modifications,
-			   guest_test_phys_mem +
-			   (guest_percpu_mem_size * nr_vcpus) +
-			   perf_test_args.host_page_size +
-			   perf_test_args.guest_page_size);
+			   p->nr_memslot_modifications);
 
 	run_vcpus = false;
 
diff --git a/tools/testing/selftests/kvm/memslot_perf_test.c b/tools/testing/selftests/kvm/memslot_perf_test.c
new file mode 100644
index 000000000000..9307f25d8130
--- /dev/null
+++ b/tools/testing/selftests/kvm/memslot_perf_test.c
@@ -0,0 +1,1037 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * A memslot-related performance benchmark.
+ *
+ * Copyright (C) 2021 Oracle and/or its affiliates.
+ *
+ * Basic guest setup / host vCPU thread code lifted from set_memory_region_test.
+ */
+#include <pthread.h>
+#include <sched.h>
+#include <semaphore.h>
+#include <stdatomic.h>
+#include <stdbool.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <time.h>
+#include <unistd.h>
+
+#include <linux/compiler.h>
+
+#include <test_util.h>
+#include <kvm_util.h>
+#include <processor.h>
+
+#define VCPU_ID 0
+
+#define MEM_SIZE		((512U << 20) + 4096)
+#define MEM_SIZE_PAGES		(MEM_SIZE / 4096)
+#define MEM_GPA		0x10000000UL
+#define MEM_AUX_GPA		MEM_GPA
+#define MEM_SYNC_GPA		MEM_AUX_GPA
+#define MEM_TEST_GPA		(MEM_AUX_GPA + 4096)
+#define MEM_TEST_SIZE		(MEM_SIZE - 4096)
+static_assert(MEM_SIZE % 4096 == 0, "invalid mem size");
+static_assert(MEM_TEST_SIZE % 4096 == 0, "invalid mem test size");
+
+/*
+ * 32 MiB is max size that gets well over 100 iterations on 509 slots.
+ * Considering that each slot needs to have at least one page up to
+ * 8194 slots in use can then be tested (although with slightly
+ * limited resolution).
+ */
+#define MEM_SIZE_MAP		((32U << 20) + 4096)
+#define MEM_SIZE_MAP_PAGES	(MEM_SIZE_MAP / 4096)
+#define MEM_TEST_MAP_SIZE	(MEM_SIZE_MAP - 4096)
+#define MEM_TEST_MAP_SIZE_PAGES (MEM_TEST_MAP_SIZE / 4096)
+static_assert(MEM_SIZE_MAP % 4096 == 0, "invalid map test region size");
+static_assert(MEM_TEST_MAP_SIZE % 4096 == 0, "invalid map test region size");
+static_assert(MEM_TEST_MAP_SIZE_PAGES % 2 == 0, "invalid map test region size");
+static_assert(MEM_TEST_MAP_SIZE_PAGES > 2, "invalid map test region size");
+
+/*
+ * 128 MiB is min size that fills 32k slots with at least one page in each
+ * while at the same time gets 100+ iterations in such test
+ */
+#define MEM_TEST_UNMAP_SIZE		(128U << 20)
+#define MEM_TEST_UNMAP_SIZE_PAGES	(MEM_TEST_UNMAP_SIZE / 4096)
+/* 2 MiB chunk size like a typical huge page */
+#define MEM_TEST_UNMAP_CHUNK_PAGES	(2U << (20 - 12))
+static_assert(MEM_TEST_UNMAP_SIZE <= MEM_TEST_SIZE,
+	      "invalid unmap test region size");
+static_assert(MEM_TEST_UNMAP_SIZE % 4096 == 0,
+	      "invalid unmap test region size");
+static_assert(MEM_TEST_UNMAP_SIZE_PAGES %
+	      (2 * MEM_TEST_UNMAP_CHUNK_PAGES) == 0,
+	      "invalid unmap test region size");
+
+/*
+ * For the move active test the middle of the test area is placed on
+ * a memslot boundary: half lies in the memslot being moved, half in
+ * other memslot(s).
+ *
+ * When running this test with 32k memslots (32764, really) each memslot
+ * contains 4 pages.
+ * The last one additionally contains the remaining 21 pages of memory,
+ * for the total size of 25 pages.
+ * Hence, the maximum size here is 50 pages.
+ */
+#define MEM_TEST_MOVE_SIZE_PAGES	(50)
+#define MEM_TEST_MOVE_SIZE		(MEM_TEST_MOVE_SIZE_PAGES * 4096)
+#define MEM_TEST_MOVE_GPA_DEST		(MEM_GPA + MEM_SIZE)
+static_assert(MEM_TEST_MOVE_SIZE <= MEM_TEST_SIZE,
+	      "invalid move test region size");
+
+#define MEM_TEST_VAL_1 0x1122334455667788
+#define MEM_TEST_VAL_2 0x99AABBCCDDEEFF00
+
+struct vm_data {
+	struct kvm_vm *vm;
+	pthread_t vcpu_thread;
+	uint32_t nslots;
+	uint64_t npages;
+	uint64_t pages_per_slot;
+	void **hva_slots;
+	bool mmio_ok;
+	uint64_t mmio_gpa_min;
+	uint64_t mmio_gpa_max;
+};
+
+struct sync_area {
+	atomic_bool start_flag;
+	atomic_bool exit_flag;
+	atomic_bool sync_flag;
+	void *move_area_ptr;
+};
+
+/*
+ * Technically, we need also for the atomic bool to be address-free, which
+ * is recommended, but not strictly required, by C11 for lockless
+ * implementations.
+ * However, in practice both GCC and Clang fulfill this requirement on
+ * all KVM-supported platforms.
+ */
+static_assert(ATOMIC_BOOL_LOCK_FREE == 2, "atomic bool is not lockless");
+
+static sem_t vcpu_ready;
+
+static bool map_unmap_verify;
+
+static bool verbose;
+#define pr_info_v(...)				\
+	do {					\
+		if (verbose)			\
+			pr_info(__VA_ARGS__);	\
+	} while (0)
+
+static void *vcpu_worker(void *data)
+{
+	struct vm_data *vm = data;
+	struct kvm_run *run;
+	struct ucall uc;
+	uint64_t cmd;
+
+	run = vcpu_state(vm->vm, VCPU_ID);
+	while (1) {
+		vcpu_run(vm->vm, VCPU_ID);
+
+		if (run->exit_reason == KVM_EXIT_IO) {
+			cmd = get_ucall(vm->vm, VCPU_ID, &uc);
+			if (cmd != UCALL_SYNC)
+				break;
+
+			sem_post(&vcpu_ready);
+			continue;
+		}
+
+		if (run->exit_reason != KVM_EXIT_MMIO)
+			break;
+
+		TEST_ASSERT(vm->mmio_ok, "Unexpected mmio exit");
+		TEST_ASSERT(run->mmio.is_write, "Unexpected mmio read");
+		TEST_ASSERT(run->mmio.len == 8,
+			    "Unexpected exit mmio size = %u", run->mmio.len);
+		TEST_ASSERT(run->mmio.phys_addr >= vm->mmio_gpa_min &&
+			    run->mmio.phys_addr <= vm->mmio_gpa_max,
+			    "Unexpected exit mmio address = 0x%llx",
+			    run->mmio.phys_addr);
+	}
+
+	if (run->exit_reason == KVM_EXIT_IO && cmd == UCALL_ABORT)
+		TEST_FAIL("%s at %s:%ld, val = %lu", (const char *)uc.args[0],
+			  __FILE__, uc.args[1], uc.args[2]);
+
+	return NULL;
+}
+
+static void wait_for_vcpu(void)
+{
+	struct timespec ts;
+
+	TEST_ASSERT(!clock_gettime(CLOCK_REALTIME, &ts),
+		    "clock_gettime() failed: %d\n", errno);
+
+	ts.tv_sec += 2;
+	TEST_ASSERT(!sem_timedwait(&vcpu_ready, &ts),
+		    "sem_timedwait() failed: %d\n", errno);
+}
+
+static void *vm_gpa2hva(struct vm_data *data, uint64_t gpa, uint64_t *rempages)
+{
+	uint64_t gpage, pgoffs;
+	uint32_t slot, slotoffs;
+	void *base;
+
+	TEST_ASSERT(gpa >= MEM_GPA, "Too low gpa to translate");
+	TEST_ASSERT(gpa < MEM_GPA + data->npages * 4096,
+		    "Too high gpa to translate");
+	gpa -= MEM_GPA;
+
+	gpage = gpa / 4096;
+	pgoffs = gpa % 4096;
+	slot = min(gpage / data->pages_per_slot, (uint64_t)data->nslots - 1);
+	slotoffs = gpage - (slot * data->pages_per_slot);
+
+	if (rempages) {
+		uint64_t slotpages;
+
+		if (slot == data->nslots - 1)
+			slotpages = data->npages - slot * data->pages_per_slot;
+		else
+			slotpages = data->pages_per_slot;
+
+		TEST_ASSERT(!pgoffs,
+			    "Asking for remaining pages in slot but gpa not page aligned");
+		*rempages = slotpages - slotoffs;
+	}
+
+	base = data->hva_slots[slot];
+	return (uint8_t *)base + slotoffs * 4096 + pgoffs;
+}
+
+static uint64_t vm_slot2gpa(struct vm_data *data, uint32_t slot)
+{
+	TEST_ASSERT(slot < data->nslots, "Too high slot number");
+
+	return MEM_GPA + slot * data->pages_per_slot * 4096;
+}
+
+static struct vm_data *alloc_vm(void)
+{
+	struct vm_data *data;
+
+	data = malloc(sizeof(*data));
+	TEST_ASSERT(data, "malloc(vmdata) failed");
+
+	data->vm = NULL;
+	data->hva_slots = NULL;
+
+	return data;
+}
+
+static bool prepare_vm(struct vm_data *data, int nslots, uint64_t *maxslots,
+		       void *guest_code, uint64_t mempages,
+		       struct timespec *slot_runtime)
+{
+	uint32_t max_mem_slots;
+	uint64_t rempages;
+	uint64_t guest_addr;
+	uint32_t slot;
+	struct timespec tstart;
+	struct sync_area *sync;
+
+	max_mem_slots = kvm_check_cap(KVM_CAP_NR_MEMSLOTS);
+	TEST_ASSERT(max_mem_slots > 1,
+		    "KVM_CAP_NR_MEMSLOTS should be greater than 1");
+	TEST_ASSERT(nslots > 1 || nslots == -1,
+		    "Slot count cap should be greater than 1");
+	if (nslots != -1)
+		max_mem_slots = min(max_mem_slots, (uint32_t)nslots);
+	pr_info_v("Allowed number of memory slots: %"PRIu32"\n", max_mem_slots);
+
+	TEST_ASSERT(mempages > 1,
+		    "Can't test without any memory");
+
+	data->npages = mempages;
+	data->nslots = max_mem_slots - 1;
+	data->pages_per_slot = mempages / data->nslots;
+	if (!data->pages_per_slot) {
+		*maxslots = mempages + 1;
+		return false;
+	}
+
+	rempages = mempages % data->nslots;
+	data->hva_slots = malloc(sizeof(*data->hva_slots) * data->nslots);
+	TEST_ASSERT(data->hva_slots, "malloc() fail");
+
+	data->vm = vm_create_default(VCPU_ID, 1024, guest_code);
+
+	pr_info_v("Adding slots 1..%i, each slot with %"PRIu64" pages + %"PRIu64" extra pages last\n",
+		max_mem_slots - 1, data->pages_per_slot, rempages);
+
+	clock_gettime(CLOCK_MONOTONIC, &tstart);
+	for (slot = 1, guest_addr = MEM_GPA; slot < max_mem_slots; slot++) {
+		uint64_t npages;
+
+		npages = data->pages_per_slot;
+		if (slot == max_mem_slots - 1)
+			npages += rempages;
+
+		vm_userspace_mem_region_add(data->vm, VM_MEM_SRC_ANONYMOUS,
+					    guest_addr, slot, npages,
+					    0);
+		guest_addr += npages * 4096;
+	}
+	*slot_runtime = timespec_elapsed(tstart);
+
+	for (slot = 0, guest_addr = MEM_GPA; slot < max_mem_slots - 1; slot++) {
+		uint64_t npages;
+		uint64_t gpa;
+
+		npages = data->pages_per_slot;
+		if (slot == max_mem_slots - 2)
+			npages += rempages;
+
+		gpa = vm_phy_pages_alloc(data->vm, npages, guest_addr,
+					 slot + 1);
+		TEST_ASSERT(gpa == guest_addr,
+			    "vm_phy_pages_alloc() failed\n");
+
+		data->hva_slots[slot] = addr_gpa2hva(data->vm, guest_addr);
+		memset(data->hva_slots[slot], 0, npages * 4096);
+
+		guest_addr += npages * 4096;
+	}
+
+	virt_map(data->vm, MEM_GPA, MEM_GPA, mempages, 0);
+
+	sync = (typeof(sync))vm_gpa2hva(data, MEM_SYNC_GPA, NULL);
+	atomic_init(&sync->start_flag, false);
+	atomic_init(&sync->exit_flag, false);
+	atomic_init(&sync->sync_flag, false);
+
+	data->mmio_ok = false;
+
+	return true;
+}
+
+static void launch_vm(struct vm_data *data)
+{
+	pr_info_v("Launching the test VM\n");
+
+	pthread_create(&data->vcpu_thread, NULL, vcpu_worker, data);
+
+	/* Ensure the guest thread is spun up. */
+	wait_for_vcpu();
+}
+
+static void free_vm(struct vm_data *data)
+{
+	kvm_vm_free(data->vm);
+	free(data->hva_slots);
+	free(data);
+}
+
+static void wait_guest_exit(struct vm_data *data)
+{
+	pthread_join(data->vcpu_thread, NULL);
+}
+
+static void let_guest_run(struct sync_area *sync)
+{
+	atomic_store_explicit(&sync->start_flag, true, memory_order_release);
+}
+
+static void guest_spin_until_start(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+
+	while (!atomic_load_explicit(&sync->start_flag, memory_order_acquire))
+		;
+}
+
+static void make_guest_exit(struct sync_area *sync)
+{
+	atomic_store_explicit(&sync->exit_flag, true, memory_order_release);
+}
+
+static bool _guest_should_exit(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+
+	return atomic_load_explicit(&sync->exit_flag, memory_order_acquire);
+}
+
+#define guest_should_exit() unlikely(_guest_should_exit())
+
+/*
+ * noinline so we can easily see how much time the host spends waiting
+ * for the guest.
+ * For the same reason use alarm() instead of polling clock_gettime()
+ * to implement a wait timeout.
+ */
+static noinline void host_perform_sync(struct sync_area *sync)
+{
+	alarm(2);
+
+	atomic_store_explicit(&sync->sync_flag, true, memory_order_release);
+	while (atomic_load_explicit(&sync->sync_flag, memory_order_acquire))
+		;
+
+	alarm(0);
+}
+
+static bool guest_perform_sync(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+	bool expected;
+
+	do {
+		if (guest_should_exit())
+			return false;
+
+		expected = true;
+	} while (!atomic_compare_exchange_weak_explicit(&sync->sync_flag,
+							&expected, false,
+							memory_order_acq_rel,
+							memory_order_relaxed));
+
+	return true;
+}
+
+static void guest_code_test_memslot_move(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+	uintptr_t base = (typeof(base))READ_ONCE(sync->move_area_ptr);
+
+	GUEST_SYNC(0);
+
+	guest_spin_until_start();
+
+	while (!guest_should_exit()) {
+		uintptr_t ptr;
+
+		for (ptr = base; ptr < base + MEM_TEST_MOVE_SIZE;
+		     ptr += 4096)
+			*(uint64_t *)ptr = MEM_TEST_VAL_1;
+
+		/*
+		 * No host sync here since the MMIO exits are so expensive
+		 * that the host would spend most of its time waiting for
+		 * the guest and so instead of measuring memslot move
+		 * performance we would measure the performance and
+		 * likelihood of MMIO exits
+		 */
+	}
+
+	GUEST_DONE();
+}
+
+static void guest_code_test_memslot_map(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+
+	GUEST_SYNC(0);
+
+	guest_spin_until_start();
+
+	while (1) {
+		uintptr_t ptr;
+
+		for (ptr = MEM_TEST_GPA;
+		     ptr < MEM_TEST_GPA + MEM_TEST_MAP_SIZE / 2; ptr += 4096)
+			*(uint64_t *)ptr = MEM_TEST_VAL_1;
+
+		if (!guest_perform_sync())
+			break;
+
+		for (ptr = MEM_TEST_GPA + MEM_TEST_MAP_SIZE / 2;
+		     ptr < MEM_TEST_GPA + MEM_TEST_MAP_SIZE; ptr += 4096)
+			*(uint64_t *)ptr = MEM_TEST_VAL_2;
+
+		if (!guest_perform_sync())
+			break;
+	}
+
+	GUEST_DONE();
+}
+
+static void guest_code_test_memslot_unmap(void)
+{
+	struct sync_area *sync = (typeof(sync))MEM_SYNC_GPA;
+
+	GUEST_SYNC(0);
+
+	guest_spin_until_start();
+
+	while (1) {
+		uintptr_t ptr = MEM_TEST_GPA;
+
+		/*
+		 * We can afford to access (map) just a small number of pages
+		 * per host sync as otherwise the host will spend
+		 * a significant amount of its time waiting for the guest
+		 * (instead of doing unmap operations), so this will
+		 * effectively turn this test into a map performance test.
+		 *
+		 * Just access a single page to be on the safe side.
+		 */
+		*(uint64_t *)ptr = MEM_TEST_VAL_1;
+
+		if (!guest_perform_sync())
+			break;
+
+		ptr += MEM_TEST_UNMAP_SIZE / 2;
+		*(uint64_t *)ptr = MEM_TEST_VAL_2;
+
+		if (!guest_perform_sync())
+			break;
+	}
+
+	GUEST_DONE();
+}
+
+static void guest_code_test_memslot_rw(void)
+{
+	GUEST_SYNC(0);
+
+	guest_spin_until_start();
+
+	while (1) {
+		uintptr_t ptr;
+
+		for (ptr = MEM_TEST_GPA;
+		     ptr < MEM_TEST_GPA + MEM_TEST_SIZE; ptr += 4096)
+			*(uint64_t *)ptr = MEM_TEST_VAL_1;
+
+		if (!guest_perform_sync())
+			break;
+
+		for (ptr = MEM_TEST_GPA + 4096 / 2;
+		     ptr < MEM_TEST_GPA + MEM_TEST_SIZE; ptr += 4096) {
+			uint64_t val = *(uint64_t *)ptr;
+
+			GUEST_ASSERT_1(val == MEM_TEST_VAL_2, val);
+			*(uint64_t *)ptr = 0;
+		}
+
+		if (!guest_perform_sync())
+			break;
+	}
+
+	GUEST_DONE();
+}
+
+static bool test_memslot_move_prepare(struct vm_data *data,
+				      struct sync_area *sync,
+				      uint64_t *maxslots, bool isactive)
+{
+	uint64_t movesrcgpa, movetestgpa;
+
+	movesrcgpa = vm_slot2gpa(data, data->nslots - 1);
+
+	if (isactive) {
+		uint64_t lastpages;
+
+		vm_gpa2hva(data, movesrcgpa, &lastpages);
+		if (lastpages < MEM_TEST_MOVE_SIZE_PAGES / 2) {
+			*maxslots = 0;
+			return false;
+		}
+	}
+
+	movetestgpa = movesrcgpa - (MEM_TEST_MOVE_SIZE / (isactive ? 2 : 1));
+	sync->move_area_ptr = (void *)movetestgpa;
+
+	if (isactive) {
+		data->mmio_ok = true;
+		data->mmio_gpa_min = movesrcgpa;
+		data->mmio_gpa_max = movesrcgpa + MEM_TEST_MOVE_SIZE / 2 - 1;
+	}
+
+	return true;
+}
+
+static bool test_memslot_move_prepare_active(struct vm_data *data,
+					     struct sync_area *sync,
+					     uint64_t *maxslots)
+{
+	return test_memslot_move_prepare(data, sync, maxslots, true);
+}
+
+static bool test_memslot_move_prepare_inactive(struct vm_data *data,
+					       struct sync_area *sync,
+					       uint64_t *maxslots)
+{
+	return test_memslot_move_prepare(data, sync, maxslots, false);
+}
+
+static void test_memslot_move_loop(struct vm_data *data, struct sync_area *sync)
+{
+	uint64_t movesrcgpa;
+
+	movesrcgpa = vm_slot2gpa(data, data->nslots - 1);
+	vm_mem_region_move(data->vm, data->nslots - 1 + 1,
+			   MEM_TEST_MOVE_GPA_DEST);
+	vm_mem_region_move(data->vm, data->nslots - 1 + 1, movesrcgpa);
+}
+
+static void test_memslot_do_unmap(struct vm_data *data,
+				  uint64_t offsp, uint64_t count)
+{
+	uint64_t gpa, ctr;
+
+	for (gpa = MEM_TEST_GPA + offsp * 4096, ctr = 0; ctr < count; ) {
+		uint64_t npages;
+		void *hva;
+		int ret;
+
+		hva = vm_gpa2hva(data, gpa, &npages);
+		TEST_ASSERT(npages, "Empty memory slot at gptr 0x%"PRIx64, gpa);
+		npages = min(npages, count - ctr);
+		ret = madvise(hva, npages * 4096, MADV_DONTNEED);
+		TEST_ASSERT(!ret,
+			    "madvise(%p, MADV_DONTNEED) on VM memory should not fail for gptr 0x%"PRIx64,
+			    hva, gpa);
+		ctr += npages;
+		gpa += npages * 4096;
+	}
+	TEST_ASSERT(ctr == count,
+		    "madvise(MADV_DONTNEED) should exactly cover all of the requested area");
+}
+
+static void test_memslot_map_unmap_check(struct vm_data *data,
+					 uint64_t offsp, uint64_t valexp)
+{
+	uint64_t gpa;
+	uint64_t *val;
+
+	if (!map_unmap_verify)
+		return;
+
+	gpa = MEM_TEST_GPA + offsp * 4096;
+	val = (typeof(val))vm_gpa2hva(data, gpa, NULL);
+	TEST_ASSERT(*val == valexp,
+		    "Guest written values should read back correctly before unmap (%"PRIu64" vs %"PRIu64" @ %"PRIx64")",
+		    *val, valexp, gpa);
+	*val = 0;
+}
+
+static void test_memslot_map_loop(struct vm_data *data, struct sync_area *sync)
+{
+	/*
+	 * Unmap the second half of the test area while guest writes to (maps)
+	 * the first half.
+	 */
+	test_memslot_do_unmap(data, MEM_TEST_MAP_SIZE_PAGES / 2,
+			      MEM_TEST_MAP_SIZE_PAGES / 2);
+
+	/*
+	 * Wait for the guest to finish writing the first half of the test
+	 * area, verify the written value on the first and the last page of
+	 * this area and then unmap it.
+	 * Meanwhile, the guest is writing to (mapping) the second half of
+	 * the test area.
+	 */
+	host_perform_sync(sync);
+	test_memslot_map_unmap_check(data, 0, MEM_TEST_VAL_1);
+	test_memslot_map_unmap_check(data,
+				     MEM_TEST_MAP_SIZE_PAGES / 2 - 1,
+				     MEM_TEST_VAL_1);
+	test_memslot_do_unmap(data, 0, MEM_TEST_MAP_SIZE_PAGES / 2);
+
+
+	/*
+	 * Wait for the guest to finish writing the second half of the test
+	 * area and verify the written value on the first and the last page
+	 * of this area.
+	 * The area will be unmapped at the beginning of the next loop
+	 * iteration.
+	 * Meanwhile, the guest is writing to (mapping) the first half of
+	 * the test area.
+	 */
+	host_perform_sync(sync);
+	test_memslot_map_unmap_check(data, MEM_TEST_MAP_SIZE_PAGES / 2,
+				     MEM_TEST_VAL_2);
+	test_memslot_map_unmap_check(data, MEM_TEST_MAP_SIZE_PAGES - 1,
+				     MEM_TEST_VAL_2);
+}
+
+static void test_memslot_unmap_loop_common(struct vm_data *data,
+					   struct sync_area *sync,
+					   uint64_t chunk)
+{
+	uint64_t ctr;
+
+	/*
+	 * Wait for the guest to finish mapping page(s) in the first half
+	 * of the test area, verify the written value and then perform unmap
+	 * of this area.
+	 * Meanwhile, the guest is writing to (mapping) page(s) in the second
+	 * half of the test area.
+	 */
+	host_perform_sync(sync);
+	test_memslot_map_unmap_check(data, 0, MEM_TEST_VAL_1);
+	for (ctr = 0; ctr < MEM_TEST_UNMAP_SIZE_PAGES / 2; ctr += chunk)
+		test_memslot_do_unmap(data, ctr, chunk);
+
+	/* Likewise, but for the opposite host / guest areas */
+	host_perform_sync(sync);
+	test_memslot_map_unmap_check(data, MEM_TEST_UNMAP_SIZE_PAGES / 2,
+				     MEM_TEST_VAL_2);
+	for (ctr = MEM_TEST_UNMAP_SIZE_PAGES / 2;
+	     ctr < MEM_TEST_UNMAP_SIZE_PAGES; ctr += chunk)
+		test_memslot_do_unmap(data, ctr, chunk);
+}
+
+static void test_memslot_unmap_loop(struct vm_data *data,
+				    struct sync_area *sync)
+{
+	test_memslot_unmap_loop_common(data, sync, 1);
+}
+
+static void test_memslot_unmap_loop_chunked(struct vm_data *data,
+					    struct sync_area *sync)
+{
+	test_memslot_unmap_loop_common(data, sync, MEM_TEST_UNMAP_CHUNK_PAGES);
+}
+
+static void test_memslot_rw_loop(struct vm_data *data, struct sync_area *sync)
+{
+	uint64_t gptr;
+
+	for (gptr = MEM_TEST_GPA + 4096 / 2;
+	     gptr < MEM_TEST_GPA + MEM_TEST_SIZE; gptr += 4096)
+		*(uint64_t *)vm_gpa2hva(data, gptr, NULL) = MEM_TEST_VAL_2;
+
+	host_perform_sync(sync);
+
+	for (gptr = MEM_TEST_GPA;
+	     gptr < MEM_TEST_GPA + MEM_TEST_SIZE; gptr += 4096) {
+		uint64_t *vptr = (typeof(vptr))vm_gpa2hva(data, gptr, NULL);
+		uint64_t val = *vptr;
+
+		TEST_ASSERT(val == MEM_TEST_VAL_1,
+			    "Guest written values should read back correctly (is %"PRIu64" @ %"PRIx64")",
+			    val, gptr);
+		*vptr = 0;
+	}
+
+	host_perform_sync(sync);
+}
+
+struct test_data {
+	const char *name;
+	uint64_t mem_size;
+	void (*guest_code)(void);
+	bool (*prepare)(struct vm_data *data, struct sync_area *sync,
+			uint64_t *maxslots);
+	void (*loop)(struct vm_data *data, struct sync_area *sync);
+};
+
+static bool test_execute(int nslots, uint64_t *maxslots,
+			 unsigned int maxtime,
+			 const struct test_data *tdata,
+			 uint64_t *nloops,
+			 struct timespec *slot_runtime,
+			 struct timespec *guest_runtime)
+{
+	uint64_t mem_size = tdata->mem_size ? : MEM_SIZE_PAGES;
+	struct vm_data *data;
+	struct sync_area *sync;
+	struct timespec tstart;
+	bool ret = true;
+
+	data = alloc_vm();
+	if (!prepare_vm(data, nslots, maxslots, tdata->guest_code,
+			mem_size, slot_runtime)) {
+		ret = false;
+		goto exit_free;
+	}
+
+	sync = (typeof(sync))vm_gpa2hva(data, MEM_SYNC_GPA, NULL);
+
+	if (tdata->prepare &&
+	    !tdata->prepare(data, sync, maxslots)) {
+		ret = false;
+		goto exit_free;
+	}
+
+	launch_vm(data);
+
+	clock_gettime(CLOCK_MONOTONIC, &tstart);
+	let_guest_run(sync);
+
+	while (1) {
+		*guest_runtime = timespec_elapsed(tstart);
+		if (guest_runtime->tv_sec >= maxtime)
+			break;
+
+		tdata->loop(data, sync);
+
+		(*nloops)++;
+	}
+
+	make_guest_exit(sync);
+	wait_guest_exit(data);
+
+exit_free:
+	free_vm(data);
+
+	return ret;
+}
+
+static const struct test_data tests[] = {
+	{
+		.name = "map",
+		.mem_size = MEM_SIZE_MAP_PAGES,
+		.guest_code = guest_code_test_memslot_map,
+		.loop = test_memslot_map_loop,
+	},
+	{
+		.name = "unmap",
+		.mem_size = MEM_TEST_UNMAP_SIZE_PAGES + 1,
+		.guest_code = guest_code_test_memslot_unmap,
+		.loop = test_memslot_unmap_loop,
+	},
+	{
+		.name = "unmap chunked",
+		.mem_size = MEM_TEST_UNMAP_SIZE_PAGES + 1,
+		.guest_code = guest_code_test_memslot_unmap,
+		.loop = test_memslot_unmap_loop_chunked,
+	},
+	{
+		.name = "move active area",
+		.guest_code = guest_code_test_memslot_move,
+		.prepare = test_memslot_move_prepare_active,
+		.loop = test_memslot_move_loop,
+	},
+	{
+		.name = "move inactive area",
+		.guest_code = guest_code_test_memslot_move,
+		.prepare = test_memslot_move_prepare_inactive,
+		.loop = test_memslot_move_loop,
+	},
+	{
+		.name = "RW",
+		.guest_code = guest_code_test_memslot_rw,
+		.loop = test_memslot_rw_loop
+	},
+};
+
+#define NTESTS ARRAY_SIZE(tests)
+
+struct test_args {
+	int tfirst;
+	int tlast;
+	int nslots;
+	int seconds;
+	int runs;
+};
+
+static void help(char *name, struct test_args *targs)
+{
+	int ctr;
+
+	pr_info("usage: %s [-h] [-v] [-d] [-s slots] [-f first_test] [-e last_test] [-l test_length] [-r run_count]\n",
+		name);
+	pr_info(" -h: print this help screen.\n");
+	pr_info(" -v: enable verbose mode (not for benchmarking).\n");
+	pr_info(" -d: enable extra debug checks.\n");
+	pr_info(" -s: specify memslot count cap (-1 means no cap; currently: %i)\n",
+		targs->nslots);
+	pr_info(" -f: specify the first test to run (currently: %i; max %zu)\n",
+		targs->tfirst, NTESTS - 1);
+	pr_info(" -e: specify the last test to run (currently: %i; max %zu)\n",
+		targs->tlast, NTESTS - 1);
+	pr_info(" -l: specify the test length in seconds (currently: %i)\n",
+		targs->seconds);
+	pr_info(" -r: specify the number of runs per test (currently: %i)\n",
+		targs->runs);
+
+	pr_info("\nAvailable tests:\n");
+	for (ctr = 0; ctr < NTESTS; ctr++)
+		pr_info("%d: %s\n", ctr, tests[ctr].name);
+}
+
+static bool parse_args(int argc, char *argv[],
+		       struct test_args *targs)
+{
+	int opt;
+
+	while ((opt = getopt(argc, argv, "hvds:f:e:l:r:")) != -1) {
+		switch (opt) {
+		case 'h':
+		default:
+			help(argv[0], targs);
+			return false;
+		case 'v':
+			verbose = true;
+			break;
+		case 'd':
+			map_unmap_verify = true;
+			break;
+		case 's':
+			targs->nslots = atoi(optarg);
+			if (targs->nslots <= 0 && targs->nslots != -1) {
+				pr_info("Slot count cap has to be positive or -1 for no cap\n");
+				return false;
+			}
+			break;
+		case 'f':
+			targs->tfirst = atoi(optarg);
+			if (targs->tfirst < 0) {
+				pr_info("First test to run has to be non-negative\n");
+				return false;
+			}
+			break;
+		case 'e':
+			targs->tlast = atoi(optarg);
+			if (targs->tlast < 0 || targs->tlast >= NTESTS) {
+				pr_info("Last test to run has to be non-negative and less than %zu\n",
+					NTESTS);
+				return false;
+			}
+			break;
+		case 'l':
+			targs->seconds = atoi(optarg);
+			if (targs->seconds < 0) {
+				pr_info("Test length in seconds has to be non-negative\n");
+				return false;
+			}
+			break;
+		case 'r':
+			targs->runs = atoi(optarg);
+			if (targs->runs <= 0) {
+				pr_info("Runs per test has to be positive\n");
+				return false;
+			}
+			break;
+		}
+	}
+
+	if (optind < argc) {
+		help(argv[0], targs);
+		return false;
+	}
+
+	if (targs->tfirst > targs->tlast) {
+		pr_info("First test to run cannot be greater than the last test to run\n");
+		return false;
+	}
+
+	return true;
+}
+
+struct test_result {
+	struct timespec slot_runtime, guest_runtime, iter_runtime;
+	int64_t slottimens, runtimens;
+	uint64_t nloops;
+};
+
+static bool test_loop(const struct test_data *data,
+		      const struct test_args *targs,
+		      struct test_result *rbestslottime,
+		      struct test_result *rbestruntime)
+{
+	uint64_t maxslots;
+	struct test_result result;
+
+	result.nloops = 0;
+	if (!test_execute(targs->nslots, &maxslots, targs->seconds, data,
+			  &result.nloops,
+			  &result.slot_runtime, &result.guest_runtime)) {
+		if (maxslots)
+			pr_info("Memslot count too high for this test, decrease the cap (max is %"PRIu64")\n",
+				maxslots);
+		else
+			pr_info("Memslot count may be too high for this test, try adjusting the cap\n");
+
+		return false;
+	}
+
+	pr_info("Test took %ld.%.9lds for slot setup + %ld.%.9lds all iterations\n",
+		result.slot_runtime.tv_sec, result.slot_runtime.tv_nsec,
+		result.guest_runtime.tv_sec, result.guest_runtime.tv_nsec);
+	if (!result.nloops) {
+		pr_info("No full loops done - too short test time or system too loaded?\n");
+		return true;
+	}
+
+	result.iter_runtime = timespec_div(result.guest_runtime,
+					   result.nloops);
+	pr_info("Done %"PRIu64" iterations, avg %ld.%.9lds each\n",
+		result.nloops,
+		result.iter_runtime.tv_sec,
+		result.iter_runtime.tv_nsec);
+	result.slottimens = timespec_to_ns(result.slot_runtime);
+	result.runtimens = timespec_to_ns(result.iter_runtime);
+
+	/*
+	 * Only rank the slot setup time for tests using the whole test memory
+	 * area so they are comparable
+	 */
+	if (!data->mem_size &&
+	    (!rbestslottime->slottimens ||
+	     result.slottimens < rbestslottime->slottimens))
+		*rbestslottime = result;
+	if (!rbestruntime->runtimens ||
+	    result.runtimens < rbestruntime->runtimens)
+		*rbestruntime = result;
+
+	return true;
+}
+
+int main(int argc, char *argv[])
+{
+	struct test_args targs = {
+		.tfirst = 0,
+		.tlast = NTESTS - 1,
+		.nslots = -1,
+		.seconds = 5,
+		.runs = 1,
+	};
+	struct test_result rbestslottime;
+	int tctr;
+
+	/* Tell stdout not to buffer its content */
+	setbuf(stdout, NULL);
+
+	if (!parse_args(argc, argv, &targs))
+		return -1;
+
+	rbestslottime.slottimens = 0;
+	for (tctr = targs.tfirst; tctr <= targs.tlast; tctr++) {
+		const struct test_data *data = &tests[tctr];
+		unsigned int runctr;
+		struct test_result rbestruntime;
+
+		if (tctr > targs.tfirst)
+			pr_info("\n");
+
+		pr_info("Testing %s performance with %i runs, %d seconds each\n",
+			data->name, targs.runs, targs.seconds);
+
+		rbestruntime.runtimens = 0;
+		for (runctr = 0; runctr < targs.runs; runctr++)
+			if (!test_loop(data, &targs,
+				       &rbestslottime, &rbestruntime))
+				break;
+
+		if (rbestruntime.runtimens)
+			pr_info("Best runtime result was %ld.%.9lds per iteration (with %"PRIu64" iterations)\n",
+				rbestruntime.iter_runtime.tv_sec,
+				rbestruntime.iter_runtime.tv_nsec,
+				rbestruntime.nloops);
+	}
+
+	if (rbestslottime.slottimens)
+		pr_info("Best slot setup time for the whole test area was %ld.%.9lds\n",
+			rbestslottime.slot_runtime.tv_sec,
+			rbestslottime.slot_runtime.tv_nsec);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/kvm/set_memory_region_test.c b/tools/testing/selftests/kvm/set_memory_region_test.c
index f127ed31dba7..978f5b5f4dc0 100644
--- a/tools/testing/selftests/kvm/set_memory_region_test.c
+++ b/tools/testing/selftests/kvm/set_memory_region_test.c
@@ -329,6 +329,22 @@ static void test_zero_memory_regions(void)
 }
 #endif /* __x86_64__ */
 
+static int test_memory_region_add(struct kvm_vm *vm, void *mem, uint32_t slot,
+				   uint32_t size, uint64_t guest_addr)
+{
+	struct kvm_userspace_memory_region region;
+	int ret;
+
+	region.slot = slot;
+	region.flags = 0;
+	region.guest_phys_addr = guest_addr;
+	region.memory_size = size;
+	region.userspace_addr = (uintptr_t) mem;
+	ret = ioctl(vm_get_fd(vm), KVM_SET_USER_MEMORY_REGION, &region);
+
+	return ret;
+}
+
 /*
  * Test it can be added memory slots up to KVM_CAP_NR_MEMSLOTS, then any
  * tentative to add further slots should fail.
@@ -339,9 +355,15 @@ static void test_add_max_memory_regions(void)
 	struct kvm_vm *vm;
 	uint32_t max_mem_slots;
 	uint32_t slot;
-	uint64_t guest_addr = 0x0;
-	uint64_t mem_reg_npages;
-	void *mem;
+	void *mem, *mem_aligned, *mem_extra;
+	size_t alignment;
+
+#ifdef __s390x__
+	/* On s390x, the host address must be aligned to 1M (due to PGSTEs) */
+	alignment = 0x100000;
+#else
+	alignment = 1;
+#endif
 
 	max_mem_slots = kvm_check_cap(KVM_CAP_NR_MEMSLOTS);
 	TEST_ASSERT(max_mem_slots > 0,
@@ -350,30 +372,37 @@ static void test_add_max_memory_regions(void)
 
 	vm = vm_create(VM_MODE_DEFAULT, 0, O_RDWR);
 
-	mem_reg_npages = vm_calc_num_guest_pages(VM_MODE_DEFAULT, MEM_REGION_SIZE);
-
 	/* Check it can be added memory slots up to the maximum allowed */
 	pr_info("Adding slots 0..%i, each memory region with %dK size\n",
 		(max_mem_slots - 1), MEM_REGION_SIZE >> 10);
+
+	mem = mmap(NULL, MEM_REGION_SIZE * max_mem_slots + alignment,
+		   PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+	TEST_ASSERT(mem != MAP_FAILED, "Failed to mmap() host");
+	mem_aligned = (void *)(((size_t) mem + alignment - 1) & ~(alignment - 1));
+
 	for (slot = 0; slot < max_mem_slots; slot++) {
-		vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS,
-					    guest_addr, slot, mem_reg_npages,
-					    0);
-		guest_addr += MEM_REGION_SIZE;
+		ret = test_memory_region_add(vm, mem_aligned +
+					     ((uint64_t)slot * MEM_REGION_SIZE),
+					     slot, MEM_REGION_SIZE,
+					     (uint64_t)slot * MEM_REGION_SIZE);
+		TEST_ASSERT(ret == 0, "KVM_SET_USER_MEMORY_REGION IOCTL failed,\n"
+			    "  rc: %i errno: %i slot: %i\n",
+			    ret, errno, slot);
 	}
 
 	/* Check it cannot be added memory slots beyond the limit */
-	mem = mmap(NULL, MEM_REGION_SIZE, PROT_READ | PROT_WRITE,
-		   MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
-	TEST_ASSERT(mem != MAP_FAILED, "Failed to mmap() host");
+	mem_extra = mmap(NULL, MEM_REGION_SIZE, PROT_READ | PROT_WRITE,
+			 MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+	TEST_ASSERT(mem_extra != MAP_FAILED, "Failed to mmap() host");
 
-	ret = ioctl(vm_get_fd(vm), KVM_SET_USER_MEMORY_REGION,
-		    &(struct kvm_userspace_memory_region) {slot, 0, guest_addr,
-		    MEM_REGION_SIZE, (uint64_t) mem});
+	ret = test_memory_region_add(vm, mem_extra, max_mem_slots, MEM_REGION_SIZE,
+				     (uint64_t)max_mem_slots * MEM_REGION_SIZE);
 	TEST_ASSERT(ret == -1 && errno == EINVAL,
 		    "Adding one more memory slot should fail with EINVAL");
 
-	munmap(mem, MEM_REGION_SIZE);
+	munmap(mem, MEM_REGION_SIZE * max_mem_slots + alignment);
+	munmap(mem_extra, MEM_REGION_SIZE);
 	kvm_vm_free(vm);
 }
 
diff --git a/tools/testing/selftests/kvm/x86_64/evmcs_test.c b/tools/testing/selftests/kvm/x86_64/evmcs_test.c
index ca22ee6d19cb..63096cea26c6 100644
--- a/tools/testing/selftests/kvm/x86_64/evmcs_test.c
+++ b/tools/testing/selftests/kvm/x86_64/evmcs_test.c
@@ -18,6 +18,28 @@
 #include "vmx.h"
 
 #define VCPU_ID		5
+#define NMI_VECTOR	2
+
+static int ud_count;
+
+void enable_x2apic(void)
+{
+	uint32_t spiv_reg = APIC_BASE_MSR + (APIC_SPIV >> 4);
+
+	wrmsr(MSR_IA32_APICBASE, rdmsr(MSR_IA32_APICBASE) |
+	      MSR_IA32_APICBASE_ENABLE | MSR_IA32_APICBASE_EXTD);
+	wrmsr(spiv_reg, rdmsr(spiv_reg) | APIC_SPIV_APIC_ENABLED);
+}
+
+static void guest_ud_handler(struct ex_regs *regs)
+{
+	ud_count++;
+	regs->rip += 3; /* VMLAUNCH */
+}
+
+static void guest_nmi_handler(struct ex_regs *regs)
+{
+}
 
 void l2_guest_code(void)
 {
@@ -25,15 +47,23 @@ void l2_guest_code(void)
 
 	GUEST_SYNC(8);
 
+	/* Forced exit to L1 upon restore */
+	GUEST_SYNC(9);
+
 	/* Done, exit to L1 and never come back.  */
 	vmcall();
 }
 
-void l1_guest_code(struct vmx_pages *vmx_pages)
+void guest_code(struct vmx_pages *vmx_pages)
 {
 #define L2_GUEST_STACK_SIZE 64
 	unsigned long l2_guest_stack[L2_GUEST_STACK_SIZE];
 
+	enable_x2apic();
+
+	GUEST_SYNC(1);
+	GUEST_SYNC(2);
+
 	enable_vp_assist(vmx_pages->vp_assist_gpa, vmx_pages->vp_assist);
 
 	GUEST_ASSERT(vmx_pages->vmcs_gpa);
@@ -55,27 +85,40 @@ void l1_guest_code(struct vmx_pages *vmx_pages)
 	current_evmcs->revision_id = EVMCS_VERSION;
 	GUEST_SYNC(6);
 
+	current_evmcs->pin_based_vm_exec_control |=
+		PIN_BASED_NMI_EXITING;
 	GUEST_ASSERT(!vmlaunch());
 	GUEST_ASSERT(vmptrstz() == vmx_pages->enlightened_vmcs_gpa);
-	GUEST_SYNC(9);
+
+	/*
+	 * NMI forces L2->L1 exit, resuming L2 and hope that EVMCS is
+	 * up-to-date (RIP points where it should and not at the beginning
+	 * of l2_guest_code(). GUEST_SYNC(9) checkes that.
+	 */
 	GUEST_ASSERT(!vmresume());
-	GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL);
+
 	GUEST_SYNC(10);
+
+	GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL);
+	GUEST_SYNC(11);
+
+	/* Try enlightened vmptrld with an incorrect GPA */
+	evmcs_vmptrld(0xdeadbeef, vmx_pages->enlightened_vmcs);
+	GUEST_ASSERT(vmlaunch());
+	GUEST_ASSERT(ud_count == 1);
+	GUEST_DONE();
 }
 
-void guest_code(struct vmx_pages *vmx_pages)
+void inject_nmi(struct kvm_vm *vm)
 {
-	GUEST_SYNC(1);
-	GUEST_SYNC(2);
+	struct kvm_vcpu_events events;
 
-	if (vmx_pages)
-		l1_guest_code(vmx_pages);
+	vcpu_events_get(vm, VCPU_ID, &events);
 
-	GUEST_DONE();
+	events.nmi.pending = 1;
+	events.flags |= KVM_VCPUEVENT_VALID_NMI_PENDING;
 
-	/* Try enlightened vmptrld with an incorrect GPA */
-	evmcs_vmptrld(0xdeadbeef, vmx_pages->enlightened_vmcs);
-	GUEST_ASSERT(vmlaunch());
+	vcpu_events_set(vm, VCPU_ID, &events);
 }
 
 int main(int argc, char *argv[])
@@ -109,6 +152,13 @@ int main(int argc, char *argv[])
 	vcpu_alloc_vmx(vm, &vmx_pages_gva);
 	vcpu_args_set(vm, VCPU_ID, 1, vmx_pages_gva);
 
+	vm_init_descriptor_tables(vm);
+	vcpu_init_descriptor_tables(vm, VCPU_ID);
+	vm_handle_exception(vm, UD_VECTOR, guest_ud_handler);
+	vm_handle_exception(vm, NMI_VECTOR, guest_nmi_handler);
+
+	pr_info("Running L1 which uses EVMCS to run L2\n");
+
 	for (stage = 1;; stage++) {
 		_vcpu_run(vm, VCPU_ID);
 		TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
@@ -124,7 +174,7 @@ int main(int argc, char *argv[])
 		case UCALL_SYNC:
 			break;
 		case UCALL_DONE:
-			goto part1_done;
+			goto done;
 		default:
 			TEST_FAIL("Unknown ucall %lu", uc.cmd);
 		}
@@ -154,12 +204,14 @@ int main(int argc, char *argv[])
 		TEST_ASSERT(!memcmp(&regs1, &regs2, sizeof(regs2)),
 			    "Unexpected register values after vcpu_load_state; rdi: %lx rsi: %lx",
 			    (ulong) regs2.rdi, (ulong) regs2.rsi);
-	}
 
-part1_done:
-	_vcpu_run(vm, VCPU_ID);
-	TEST_ASSERT(run->exit_reason == KVM_EXIT_SHUTDOWN,
-		    "Unexpected successful VMEnter with invalid eVMCS pointer!");
+		/* Force immediate L2->L1 exit before resuming */
+		if (stage == 8) {
+			pr_info("Injecting NMI into L1 before L2 had a chance to run after restore\n");
+			inject_nmi(vm);
+		}
+	}
 
+done:
 	kvm_vm_free(vm);
 }
diff --git a/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c b/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c
index 9b78e8889638..8c77537af5a1 100644
--- a/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c
+++ b/tools/testing/selftests/kvm/x86_64/get_cpuid_test.c
@@ -19,7 +19,12 @@ struct {
 	u32 function;
 	u32 index;
 } mangled_cpuids[] = {
+	/*
+	 * These entries depend on the vCPU's XCR0 register and IA32_XSS MSR,
+	 * which are not controlled for by this test.
+	 */
 	{.function = 0xd, .index = 0},
+	{.function = 0xd, .index = 1},
 };
 
 static void test_guest_cpuids(struct kvm_cpuid2 *guest_cpuid)
diff --git a/tools/testing/selftests/kvm/x86_64/get_msr_index_features.c b/tools/testing/selftests/kvm/x86_64/get_msr_index_features.c
index cb953df4d7d0..8aed0db1331d 100644
--- a/tools/testing/selftests/kvm/x86_64/get_msr_index_features.c
+++ b/tools/testing/selftests/kvm/x86_64/get_msr_index_features.c
@@ -37,9 +37,7 @@ static void test_get_msr_index(void)
 	int old_res, res, kvm_fd, r;
 	struct kvm_msr_list *list;
 
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	old_res = kvm_num_index_msrs(kvm_fd, 0);
 	TEST_ASSERT(old_res != 0, "Expecting nmsrs to be > 0");
@@ -101,9 +99,7 @@ static void test_get_msr_feature(void)
 	int res, old_res, i, kvm_fd;
 	struct kvm_msr_list *feature_list;
 
-	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
-	if (kvm_fd < 0)
-		exit(KSFT_SKIP);
+	kvm_fd = open_kvm_dev_path_or_exit();
 
 	old_res = kvm_num_feature_msrs(kvm_fd, 0);
 	TEST_ASSERT(old_res != 0, "Expecting nmsrs to be > 0");
diff --git a/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c b/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
index 804ff5ff022d..1f4a0599683c 100644
--- a/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
+++ b/tools/testing/selftests/kvm/x86_64/xen_shinfo_test.c
@@ -186,7 +186,7 @@ int main(int argc, char *argv[])
 		vcpu_ioctl(vm, VCPU_ID, KVM_XEN_VCPU_SET_ATTR, &st);
 	}
 
-	struct vcpu_runstate_info *rs = addr_gpa2hva(vm, RUNSTATE_ADDR);;
+	struct vcpu_runstate_info *rs = addr_gpa2hva(vm, RUNSTATE_ADDR);
 	rs->state = 0x5a;
 
 	for (;;) {
diff --git a/tools/testing/selftests/landlock/.gitignore b/tools/testing/selftests/landlock/.gitignore
new file mode 100644
index 000000000000..470203a7cd73
--- /dev/null
+++ b/tools/testing/selftests/landlock/.gitignore
@@ -0,0 +1,2 @@
+/*_test
+/true
diff --git a/tools/testing/selftests/landlock/Makefile b/tools/testing/selftests/landlock/Makefile
new file mode 100644
index 000000000000..a99596ca9882
--- /dev/null
+++ b/tools/testing/selftests/landlock/Makefile
@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: GPL-2.0
+
+CFLAGS += -Wall -O2
+
+src_test := $(wildcard *_test.c)
+
+TEST_GEN_PROGS := $(src_test:.c=)
+
+TEST_GEN_PROGS_EXTENDED := true
+
+KSFT_KHDR_INSTALL := 1
+OVERRIDE_TARGETS := 1
+include ../lib.mk
+
+khdr_dir = $(top_srcdir)/usr/include
+
+$(khdr_dir)/linux/landlock.h: khdr
+	@:
+
+$(OUTPUT)/true: true.c
+	$(LINK.c) $< $(LDLIBS) -o $@ -static
+
+$(OUTPUT)/%_test: %_test.c $(khdr_dir)/linux/landlock.h ../kselftest_harness.h common.h
+	$(LINK.c) $< $(LDLIBS) -o $@ -lcap -I$(khdr_dir)
diff --git a/tools/testing/selftests/landlock/base_test.c b/tools/testing/selftests/landlock/base_test.c
new file mode 100644
index 000000000000..ca40abe9daa8
--- /dev/null
+++ b/tools/testing/selftests/landlock/base_test.c
@@ -0,0 +1,266 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Landlock tests - Common user space base
+ *
+ * Copyright © 2017-2020 Mickaël Salaün <mic@digikod.net>
+ * Copyright © 2019-2020 ANSSI
+ */
+
+#define _GNU_SOURCE
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/landlock.h>
+#include <string.h>
+#include <sys/prctl.h>
+#include <sys/socket.h>
+#include <sys/types.h>
+
+#include "common.h"
+
+#ifndef O_PATH
+#define O_PATH		010000000
+#endif
+
+TEST(inconsistent_attr) {
+	const long page_size = sysconf(_SC_PAGESIZE);
+	char *const buf = malloc(page_size + 1);
+	struct landlock_ruleset_attr *const ruleset_attr = (void *)buf;
+
+	ASSERT_NE(NULL, buf);
+
+	/* Checks copy_from_user(). */
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, 0, 0));
+	/* The size if less than sizeof(struct landlock_attr_enforce). */
+	ASSERT_EQ(EINVAL, errno);
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, 1, 0));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL, 1, 0));
+	/* The size if less than sizeof(struct landlock_attr_enforce). */
+	ASSERT_EQ(EFAULT, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL,
+				sizeof(struct landlock_ruleset_attr), 0));
+	ASSERT_EQ(EFAULT, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, page_size + 1, 0));
+	ASSERT_EQ(E2BIG, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr,
+				sizeof(struct landlock_ruleset_attr), 0));
+	ASSERT_EQ(ENOMSG, errno);
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, page_size, 0));
+	ASSERT_EQ(ENOMSG, errno);
+
+	/* Checks non-zero value. */
+	buf[page_size - 2] = '.';
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, page_size, 0));
+	ASSERT_EQ(E2BIG, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(ruleset_attr, page_size + 1, 0));
+	ASSERT_EQ(E2BIG, errno);
+
+	free(buf);
+}
+
+TEST(abi_version) {
+	const struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE,
+	};
+	ASSERT_EQ(1, landlock_create_ruleset(NULL, 0,
+				LANDLOCK_CREATE_RULESET_VERSION));
+
+	ASSERT_EQ(-1, landlock_create_ruleset(&ruleset_attr, 0,
+				LANDLOCK_CREATE_RULESET_VERSION));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL, sizeof(ruleset_attr),
+				LANDLOCK_CREATE_RULESET_VERSION));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(&ruleset_attr,
+				sizeof(ruleset_attr),
+				LANDLOCK_CREATE_RULESET_VERSION));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL, 0,
+				LANDLOCK_CREATE_RULESET_VERSION | 1 << 31));
+	ASSERT_EQ(EINVAL, errno);
+}
+
+TEST(inval_create_ruleset_flags) {
+	const int last_flag = LANDLOCK_CREATE_RULESET_VERSION;
+	const int invalid_flag = last_flag << 1;
+	const struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE,
+	};
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL, 0, invalid_flag));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(&ruleset_attr, 0, invalid_flag));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(NULL, sizeof(ruleset_attr),
+				invalid_flag));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, landlock_create_ruleset(&ruleset_attr,
+				sizeof(ruleset_attr), invalid_flag));
+	ASSERT_EQ(EINVAL, errno);
+}
+
+TEST(empty_path_beneath_attr) {
+	const struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_EXECUTE,
+	};
+	const int ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	/* Similar to struct landlock_path_beneath_attr.parent_fd = 0 */
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				NULL, 0));
+	ASSERT_EQ(EFAULT, errno);
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+TEST(inval_fd_enforce) {
+	ASSERT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+
+	ASSERT_EQ(-1, landlock_restrict_self(-1, 0));
+	ASSERT_EQ(EBADF, errno);
+}
+
+TEST(unpriv_enforce_without_no_new_privs) {
+	int err;
+
+	drop_caps(_metadata);
+	err = landlock_restrict_self(-1, 0);
+	ASSERT_EQ(EPERM, errno);
+	ASSERT_EQ(err, -1);
+}
+
+TEST(ruleset_fd_io)
+{
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE,
+	};
+	int ruleset_fd;
+	char buf;
+
+	drop_caps(_metadata);
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(-1, write(ruleset_fd, ".", 1));
+	ASSERT_EQ(EINVAL, errno);
+	ASSERT_EQ(-1, read(ruleset_fd, &buf, 1));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+/* Tests enforcement of a ruleset FD transferred through a UNIX socket. */
+TEST(ruleset_fd_transfer)
+{
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_READ_DIR,
+	};
+	struct landlock_path_beneath_attr path_beneath_attr = {
+		.allowed_access = LANDLOCK_ACCESS_FS_READ_DIR,
+	};
+	int ruleset_fd_tx, dir_fd;
+	union {
+		/* Aligned ancillary data buffer. */
+		char buf[CMSG_SPACE(sizeof(ruleset_fd_tx))];
+		struct cmsghdr _align;
+	} cmsg_tx = {};
+	char data_tx = '.';
+	struct iovec io = {
+		.iov_base = &data_tx,
+		.iov_len = sizeof(data_tx),
+	};
+	struct msghdr msg = {
+		.msg_iov = &io,
+		.msg_iovlen = 1,
+		.msg_control = &cmsg_tx.buf,
+		.msg_controllen = sizeof(cmsg_tx.buf),
+	};
+	struct cmsghdr *cmsg;
+	int socket_fds[2];
+	pid_t child;
+	int status;
+
+	drop_caps(_metadata);
+
+	/* Creates a test ruleset with a simple rule. */
+	ruleset_fd_tx = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd_tx);
+	path_beneath_attr.parent_fd = open("/tmp", O_PATH | O_NOFOLLOW |
+			O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, path_beneath_attr.parent_fd);
+	ASSERT_EQ(0, landlock_add_rule(ruleset_fd_tx, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath_attr, 0));
+	ASSERT_EQ(0, close(path_beneath_attr.parent_fd));
+
+	cmsg = CMSG_FIRSTHDR(&msg);
+	ASSERT_NE(NULL, cmsg);
+	cmsg->cmsg_len = CMSG_LEN(sizeof(ruleset_fd_tx));
+	cmsg->cmsg_level = SOL_SOCKET;
+	cmsg->cmsg_type = SCM_RIGHTS;
+	memcpy(CMSG_DATA(cmsg), &ruleset_fd_tx, sizeof(ruleset_fd_tx));
+
+	/* Sends the ruleset FD over a socketpair and then close it. */
+	ASSERT_EQ(0, socketpair(AF_UNIX, SOCK_STREAM | SOCK_CLOEXEC, 0, socket_fds));
+	ASSERT_EQ(sizeof(data_tx), sendmsg(socket_fds[0], &msg, 0));
+	ASSERT_EQ(0, close(socket_fds[0]));
+	ASSERT_EQ(0, close(ruleset_fd_tx));
+
+	child = fork();
+	ASSERT_LE(0, child);
+	if (child == 0) {
+		int ruleset_fd_rx;
+
+		*(char *)msg.msg_iov->iov_base = '\0';
+		ASSERT_EQ(sizeof(data_tx), recvmsg(socket_fds[1], &msg, MSG_CMSG_CLOEXEC));
+		ASSERT_EQ('.', *(char *)msg.msg_iov->iov_base);
+		ASSERT_EQ(0, close(socket_fds[1]));
+		cmsg = CMSG_FIRSTHDR(&msg);
+		ASSERT_EQ(cmsg->cmsg_len, CMSG_LEN(sizeof(ruleset_fd_tx)));
+		memcpy(&ruleset_fd_rx, CMSG_DATA(cmsg), sizeof(ruleset_fd_tx));
+
+		/* Enforces the received ruleset on the child. */
+		ASSERT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+		ASSERT_EQ(0, landlock_restrict_self(ruleset_fd_rx, 0));
+		ASSERT_EQ(0, close(ruleset_fd_rx));
+
+		/* Checks that the ruleset enforcement. */
+		ASSERT_EQ(-1, open("/", O_RDONLY | O_DIRECTORY | O_CLOEXEC));
+		ASSERT_EQ(EACCES, errno);
+		dir_fd = open("/tmp", O_RDONLY | O_DIRECTORY | O_CLOEXEC);
+		ASSERT_LE(0, dir_fd);
+		ASSERT_EQ(0, close(dir_fd));
+		_exit(_metadata->passed ? EXIT_SUCCESS : EXIT_FAILURE);
+		return;
+	}
+
+	ASSERT_EQ(0, close(socket_fds[1]));
+
+	/* Checks that the parent is unrestricted. */
+	dir_fd = open("/", O_RDONLY | O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, dir_fd);
+	ASSERT_EQ(0, close(dir_fd));
+	dir_fd = open("/tmp", O_RDONLY | O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, dir_fd);
+	ASSERT_EQ(0, close(dir_fd));
+
+	ASSERT_EQ(child, waitpid(child, &status, 0));
+	ASSERT_EQ(1, WIFEXITED(status));
+	ASSERT_EQ(EXIT_SUCCESS, WEXITSTATUS(status));
+}
+
+TEST_HARNESS_MAIN
diff --git a/tools/testing/selftests/landlock/common.h b/tools/testing/selftests/landlock/common.h
new file mode 100644
index 000000000000..20e2a9286d71
--- /dev/null
+++ b/tools/testing/selftests/landlock/common.h
@@ -0,0 +1,183 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Landlock test helpers
+ *
+ * Copyright © 2017-2020 Mickaël Salaün <mic@digikod.net>
+ * Copyright © 2019-2020 ANSSI
+ * Copyright © 2021 Microsoft Corporation
+ */
+
+#include <errno.h>
+#include <linux/landlock.h>
+#include <sys/capability.h>
+#include <sys/syscall.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#include "../kselftest_harness.h"
+
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+#endif
+
+/*
+ * TEST_F_FORK() is useful when a test drop privileges but the corresponding
+ * FIXTURE_TEARDOWN() requires them (e.g. to remove files from a directory
+ * where write actions are denied).  For convenience, FIXTURE_TEARDOWN() is
+ * also called when the test failed, but not when FIXTURE_SETUP() failed.  For
+ * this to be possible, we must not call abort() but instead exit smoothly
+ * (hence the step print).
+ */
+#define TEST_F_FORK(fixture_name, test_name) \
+	static void fixture_name##_##test_name##_child( \
+		struct __test_metadata *_metadata, \
+		FIXTURE_DATA(fixture_name) *self, \
+		const FIXTURE_VARIANT(fixture_name) *variant); \
+	TEST_F(fixture_name, test_name) \
+	{ \
+		int status; \
+		const pid_t child = fork(); \
+		if (child < 0) \
+			abort(); \
+		if (child == 0) { \
+			_metadata->no_print = 1; \
+			fixture_name##_##test_name##_child(_metadata, self, variant); \
+			if (_metadata->skip) \
+				_exit(255); \
+			if (_metadata->passed) \
+				_exit(0); \
+			_exit(_metadata->step); \
+		} \
+		if (child != waitpid(child, &status, 0)) \
+			abort(); \
+		if (WIFSIGNALED(status) || !WIFEXITED(status)) { \
+			_metadata->passed = 0; \
+			_metadata->step = 1; \
+			return; \
+		} \
+		switch (WEXITSTATUS(status)) { \
+		case 0: \
+			_metadata->passed = 1; \
+			break; \
+		case 255: \
+			_metadata->passed = 1; \
+			_metadata->skip = 1; \
+			break; \
+		default: \
+			_metadata->passed = 0; \
+			_metadata->step = WEXITSTATUS(status); \
+			break; \
+		} \
+	} \
+	static void fixture_name##_##test_name##_child( \
+		struct __test_metadata __attribute__((unused)) *_metadata, \
+		FIXTURE_DATA(fixture_name) __attribute__((unused)) *self, \
+		const FIXTURE_VARIANT(fixture_name) \
+			__attribute__((unused)) *variant)
+
+#ifndef landlock_create_ruleset
+static inline int landlock_create_ruleset(
+		const struct landlock_ruleset_attr *const attr,
+		const size_t size, const __u32 flags)
+{
+	return syscall(__NR_landlock_create_ruleset, attr, size, flags);
+}
+#endif
+
+#ifndef landlock_add_rule
+static inline int landlock_add_rule(const int ruleset_fd,
+		const enum landlock_rule_type rule_type,
+		const void *const rule_attr, const __u32 flags)
+{
+	return syscall(__NR_landlock_add_rule, ruleset_fd, rule_type,
+			rule_attr, flags);
+}
+#endif
+
+#ifndef landlock_restrict_self
+static inline int landlock_restrict_self(const int ruleset_fd,
+		const __u32 flags)
+{
+	return syscall(__NR_landlock_restrict_self, ruleset_fd, flags);
+}
+#endif
+
+static void _init_caps(struct __test_metadata *const _metadata, bool drop_all)
+{
+	cap_t cap_p;
+	/* Only these three capabilities are useful for the tests. */
+	const cap_value_t caps[] = {
+		CAP_DAC_OVERRIDE,
+		CAP_MKNOD,
+		CAP_SYS_ADMIN,
+		CAP_SYS_CHROOT,
+	};
+
+	cap_p = cap_get_proc();
+	EXPECT_NE(NULL, cap_p) {
+		TH_LOG("Failed to cap_get_proc: %s", strerror(errno));
+	}
+	EXPECT_NE(-1, cap_clear(cap_p)) {
+		TH_LOG("Failed to cap_clear: %s", strerror(errno));
+	}
+	if (!drop_all) {
+		EXPECT_NE(-1, cap_set_flag(cap_p, CAP_PERMITTED,
+					ARRAY_SIZE(caps), caps, CAP_SET)) {
+			TH_LOG("Failed to cap_set_flag: %s", strerror(errno));
+		}
+	}
+	EXPECT_NE(-1, cap_set_proc(cap_p)) {
+		TH_LOG("Failed to cap_set_proc: %s", strerror(errno));
+	}
+	EXPECT_NE(-1, cap_free(cap_p)) {
+		TH_LOG("Failed to cap_free: %s", strerror(errno));
+	}
+}
+
+/* We cannot put such helpers in a library because of kselftest_harness.h . */
+__attribute__((__unused__))
+static void disable_caps(struct __test_metadata *const _metadata)
+{
+	_init_caps(_metadata, false);
+}
+
+__attribute__((__unused__))
+static void drop_caps(struct __test_metadata *const _metadata)
+{
+	_init_caps(_metadata, true);
+}
+
+static void _effective_cap(struct __test_metadata *const _metadata,
+		const cap_value_t caps, const cap_flag_value_t value)
+{
+	cap_t cap_p;
+
+	cap_p = cap_get_proc();
+	EXPECT_NE(NULL, cap_p) {
+		TH_LOG("Failed to cap_get_proc: %s", strerror(errno));
+	}
+	EXPECT_NE(-1, cap_set_flag(cap_p, CAP_EFFECTIVE, 1, &caps, value)) {
+		TH_LOG("Failed to cap_set_flag: %s", strerror(errno));
+	}
+	EXPECT_NE(-1, cap_set_proc(cap_p)) {
+		TH_LOG("Failed to cap_set_proc: %s", strerror(errno));
+	}
+	EXPECT_NE(-1, cap_free(cap_p)) {
+		TH_LOG("Failed to cap_free: %s", strerror(errno));
+	}
+}
+
+__attribute__((__unused__))
+static void set_cap(struct __test_metadata *const _metadata,
+		const cap_value_t caps)
+{
+	_effective_cap(_metadata, caps, CAP_SET);
+}
+
+__attribute__((__unused__))
+static void clear_cap(struct __test_metadata *const _metadata,
+		const cap_value_t caps)
+{
+	_effective_cap(_metadata, caps, CAP_CLEAR);
+}
diff --git a/tools/testing/selftests/landlock/config b/tools/testing/selftests/landlock/config
new file mode 100644
index 000000000000..0f0a65287bac
--- /dev/null
+++ b/tools/testing/selftests/landlock/config
@@ -0,0 +1,7 @@
+CONFIG_OVERLAY_FS=y
+CONFIG_SECURITY_LANDLOCK=y
+CONFIG_SECURITY_PATH=y
+CONFIG_SECURITY=y
+CONFIG_SHMEM=y
+CONFIG_TMPFS_XATTR=y
+CONFIG_TMPFS=y
diff --git a/tools/testing/selftests/landlock/fs_test.c b/tools/testing/selftests/landlock/fs_test.c
new file mode 100644
index 000000000000..10c9a1e4ebd9
--- /dev/null
+++ b/tools/testing/selftests/landlock/fs_test.c
@@ -0,0 +1,2791 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Landlock tests - Filesystem
+ *
+ * Copyright © 2017-2020 Mickaël Salaün <mic@digikod.net>
+ * Copyright © 2020 ANSSI
+ * Copyright © 2020-2021 Microsoft Corporation
+ */
+
+#define _GNU_SOURCE
+#include <fcntl.h>
+#include <linux/landlock.h>
+#include <sched.h>
+#include <string.h>
+#include <sys/capability.h>
+#include <sys/mount.h>
+#include <sys/prctl.h>
+#include <sys/sendfile.h>
+#include <sys/stat.h>
+#include <sys/sysmacros.h>
+#include <unistd.h>
+
+#include "common.h"
+
+#define TMP_DIR		"tmp"
+#define BINARY_PATH	"./true"
+
+/* Paths (sibling number and depth) */
+static const char dir_s1d1[] = TMP_DIR "/s1d1";
+static const char file1_s1d1[] = TMP_DIR "/s1d1/f1";
+static const char file2_s1d1[] = TMP_DIR "/s1d1/f2";
+static const char dir_s1d2[] = TMP_DIR "/s1d1/s1d2";
+static const char file1_s1d2[] = TMP_DIR "/s1d1/s1d2/f1";
+static const char file2_s1d2[] = TMP_DIR "/s1d1/s1d2/f2";
+static const char dir_s1d3[] = TMP_DIR "/s1d1/s1d2/s1d3";
+static const char file1_s1d3[] = TMP_DIR "/s1d1/s1d2/s1d3/f1";
+static const char file2_s1d3[] = TMP_DIR "/s1d1/s1d2/s1d3/f2";
+
+static const char dir_s2d1[] = TMP_DIR "/s2d1";
+static const char file1_s2d1[] = TMP_DIR "/s2d1/f1";
+static const char dir_s2d2[] = TMP_DIR "/s2d1/s2d2";
+static const char file1_s2d2[] = TMP_DIR "/s2d1/s2d2/f1";
+static const char dir_s2d3[] = TMP_DIR "/s2d1/s2d2/s2d3";
+static const char file1_s2d3[] = TMP_DIR "/s2d1/s2d2/s2d3/f1";
+static const char file2_s2d3[] = TMP_DIR "/s2d1/s2d2/s2d3/f2";
+
+static const char dir_s3d1[] = TMP_DIR "/s3d1";
+/* dir_s3d2 is a mount point. */
+static const char dir_s3d2[] = TMP_DIR "/s3d1/s3d2";
+static const char dir_s3d3[] = TMP_DIR "/s3d1/s3d2/s3d3";
+
+/*
+ * layout1 hierarchy:
+ *
+ * tmp
+ * ├── s1d1
+ * │   ├── f1
+ * │   ├── f2
+ * │   └── s1d2
+ * │       ├── f1
+ * │       ├── f2
+ * │       └── s1d3
+ * │           ├── f1
+ * │           └── f2
+ * ├── s2d1
+ * │   ├── f1
+ * │   └── s2d2
+ * │       ├── f1
+ * │       └── s2d3
+ * │           ├── f1
+ * │           └── f2
+ * └── s3d1
+ *     └── s3d2
+ *         └── s3d3
+ */
+
+static void mkdir_parents(struct __test_metadata *const _metadata,
+		const char *const path)
+{
+	char *walker;
+	const char *parent;
+	int i, err;
+
+	ASSERT_NE(path[0], '\0');
+	walker = strdup(path);
+	ASSERT_NE(NULL, walker);
+	parent = walker;
+	for (i = 1; walker[i]; i++) {
+		if (walker[i] != '/')
+			continue;
+		walker[i] = '\0';
+		err = mkdir(parent, 0700);
+		ASSERT_FALSE(err && errno != EEXIST) {
+			TH_LOG("Failed to create directory \"%s\": %s",
+					parent, strerror(errno));
+		}
+		walker[i] = '/';
+	}
+	free(walker);
+}
+
+static void create_directory(struct __test_metadata *const _metadata,
+		const char *const path)
+{
+	mkdir_parents(_metadata, path);
+	ASSERT_EQ(0, mkdir(path, 0700)) {
+		TH_LOG("Failed to create directory \"%s\": %s", path,
+				strerror(errno));
+	}
+}
+
+static void create_file(struct __test_metadata *const _metadata,
+		const char *const path)
+{
+	mkdir_parents(_metadata, path);
+	ASSERT_EQ(0, mknod(path, S_IFREG | 0700, 0)) {
+		TH_LOG("Failed to create file \"%s\": %s", path,
+				strerror(errno));
+	}
+}
+
+static int remove_path(const char *const path)
+{
+	char *walker;
+	int i, ret, err = 0;
+
+	walker = strdup(path);
+	if (!walker) {
+		err = ENOMEM;
+		goto out;
+	}
+	if (unlink(path) && rmdir(path)) {
+		if (errno != ENOENT)
+			err = errno;
+		goto out;
+	}
+	for (i = strlen(walker); i > 0; i--) {
+		if (walker[i] != '/')
+			continue;
+		walker[i] = '\0';
+		ret = rmdir(walker);
+		if (ret) {
+			if (errno != ENOTEMPTY && errno != EBUSY)
+				err = errno;
+			goto out;
+		}
+		if (strcmp(walker, TMP_DIR) == 0)
+			goto out;
+	}
+
+out:
+	free(walker);
+	return err;
+}
+
+static void prepare_layout(struct __test_metadata *const _metadata)
+{
+	disable_caps(_metadata);
+	umask(0077);
+	create_directory(_metadata, TMP_DIR);
+
+	/*
+	 * Do not pollute the rest of the system: creates a private mount point
+	 * for tests relying on pivot_root(2) and move_mount(2).
+	 */
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, unshare(CLONE_NEWNS));
+	ASSERT_EQ(0, mount("tmp", TMP_DIR, "tmpfs", 0, "size=4m,mode=700"));
+	ASSERT_EQ(0, mount(NULL, TMP_DIR, NULL, MS_PRIVATE | MS_REC, NULL));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+}
+
+static void cleanup_layout(struct __test_metadata *const _metadata)
+{
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, umount(TMP_DIR));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, remove_path(TMP_DIR));
+}
+
+static void create_layout1(struct __test_metadata *const _metadata)
+{
+	create_file(_metadata, file1_s1d1);
+	create_file(_metadata, file1_s1d2);
+	create_file(_metadata, file1_s1d3);
+	create_file(_metadata, file2_s1d1);
+	create_file(_metadata, file2_s1d2);
+	create_file(_metadata, file2_s1d3);
+
+	create_file(_metadata, file1_s2d1);
+	create_file(_metadata, file1_s2d2);
+	create_file(_metadata, file1_s2d3);
+	create_file(_metadata, file2_s2d3);
+
+	create_directory(_metadata, dir_s3d2);
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, mount("tmp", dir_s3d2, "tmpfs", 0, "size=4m,mode=700"));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+
+	ASSERT_EQ(0, mkdir(dir_s3d3, 0700));
+}
+
+static void remove_layout1(struct __test_metadata *const _metadata)
+{
+	EXPECT_EQ(0, remove_path(file2_s1d3));
+	EXPECT_EQ(0, remove_path(file2_s1d2));
+	EXPECT_EQ(0, remove_path(file2_s1d1));
+	EXPECT_EQ(0, remove_path(file1_s1d3));
+	EXPECT_EQ(0, remove_path(file1_s1d2));
+	EXPECT_EQ(0, remove_path(file1_s1d1));
+
+	EXPECT_EQ(0, remove_path(file2_s2d3));
+	EXPECT_EQ(0, remove_path(file1_s2d3));
+	EXPECT_EQ(0, remove_path(file1_s2d2));
+	EXPECT_EQ(0, remove_path(file1_s2d1));
+
+	EXPECT_EQ(0, remove_path(dir_s3d3));
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	umount(dir_s3d2);
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, remove_path(dir_s3d2));
+}
+
+FIXTURE(layout1) {
+};
+
+FIXTURE_SETUP(layout1)
+{
+	prepare_layout(_metadata);
+
+	create_layout1(_metadata);
+}
+
+FIXTURE_TEARDOWN(layout1)
+{
+	remove_layout1(_metadata);
+
+	cleanup_layout(_metadata);
+}
+
+/*
+ * This helper enables to use the ASSERT_* macros and print the line number
+ * pointing to the test caller.
+ */
+static int test_open_rel(const int dirfd, const char *const path, const int flags)
+{
+	int fd;
+
+	/* Works with file and directories. */
+	fd = openat(dirfd, path, flags | O_CLOEXEC);
+	if (fd < 0)
+		return errno;
+	/*
+	 * Mixing error codes from close(2) and open(2) should not lead to any
+	 * (access type) confusion for this test.
+	 */
+	if (close(fd) != 0)
+		return errno;
+	return 0;
+}
+
+static int test_open(const char *const path, const int flags)
+{
+	return test_open_rel(AT_FDCWD, path, flags);
+}
+
+TEST_F_FORK(layout1, no_restriction)
+{
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file2_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file2_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(dir_s2d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s2d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s2d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(dir_s3d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d3, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, inval)
+{
+	struct landlock_path_beneath_attr path_beneath = {
+		.allowed_access = LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE,
+		.parent_fd = -1,
+	};
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE,
+	};
+	int ruleset_fd;
+
+	path_beneath.parent_fd = open(dir_s1d2, O_PATH | O_DIRECTORY |
+			O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd);
+
+	ruleset_fd = open(dir_s1d1, O_PATH | O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, ruleset_fd);
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	/* Returns EBADF because ruleset_fd is not a landlock-ruleset FD. */
+	ASSERT_EQ(EBADF, errno);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ruleset_fd = open(dir_s1d1, O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, ruleset_fd);
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	/* Returns EBADFD because ruleset_fd is not a valid ruleset. */
+	ASSERT_EQ(EBADFD, errno);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Gets a real ruleset. */
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd);
+	ASSERT_EQ(0, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+
+	/* Tests without O_PATH. */
+	path_beneath.parent_fd = open(dir_s1d2, O_DIRECTORY | O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd);
+	ASSERT_EQ(0, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+
+	/* Tests with a ruleset FD. */
+	path_beneath.parent_fd = ruleset_fd;
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(EBADFD, errno);
+
+	/* Checks unhandled allowed_access. */
+	path_beneath.parent_fd = open(dir_s1d2, O_PATH | O_DIRECTORY |
+			O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd);
+
+	/* Test with legitimate values. */
+	path_beneath.allowed_access |= LANDLOCK_ACCESS_FS_EXECUTE;
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(EINVAL, errno);
+	path_beneath.allowed_access &= ~LANDLOCK_ACCESS_FS_EXECUTE;
+
+	/* Test with unknown (64-bits) value. */
+	path_beneath.allowed_access |= (1ULL << 60);
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(EINVAL, errno);
+	path_beneath.allowed_access &= ~(1ULL << 60);
+
+	/* Test with no access. */
+	path_beneath.allowed_access = 0;
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(ENOMSG, errno);
+	path_beneath.allowed_access &= ~(1ULL << 60);
+
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+
+	/* Enforces the ruleset. */
+	ASSERT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+	ASSERT_EQ(0, landlock_restrict_self(ruleset_fd, 0));
+
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+#define ACCESS_FILE ( \
+	LANDLOCK_ACCESS_FS_EXECUTE | \
+	LANDLOCK_ACCESS_FS_WRITE_FILE | \
+	LANDLOCK_ACCESS_FS_READ_FILE)
+
+#define ACCESS_LAST LANDLOCK_ACCESS_FS_MAKE_SYM
+
+#define ACCESS_ALL ( \
+	ACCESS_FILE | \
+	LANDLOCK_ACCESS_FS_READ_DIR | \
+	LANDLOCK_ACCESS_FS_REMOVE_DIR | \
+	LANDLOCK_ACCESS_FS_REMOVE_FILE | \
+	LANDLOCK_ACCESS_FS_MAKE_CHAR | \
+	LANDLOCK_ACCESS_FS_MAKE_DIR | \
+	LANDLOCK_ACCESS_FS_MAKE_REG | \
+	LANDLOCK_ACCESS_FS_MAKE_SOCK | \
+	LANDLOCK_ACCESS_FS_MAKE_FIFO | \
+	LANDLOCK_ACCESS_FS_MAKE_BLOCK | \
+	ACCESS_LAST)
+
+TEST_F_FORK(layout1, file_access_rights)
+{
+	__u64 access;
+	int err;
+	struct landlock_path_beneath_attr path_beneath = {};
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = ACCESS_ALL,
+	};
+	const int ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	/* Tests access rights for files. */
+	path_beneath.parent_fd = open(file1_s1d2, O_PATH | O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd);
+	for (access = 1; access <= ACCESS_LAST; access <<= 1) {
+		path_beneath.allowed_access = access;
+		err = landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0);
+		if ((access | ACCESS_FILE) == ACCESS_FILE) {
+			ASSERT_EQ(0, err);
+		} else {
+			ASSERT_EQ(-1, err);
+			ASSERT_EQ(EINVAL, errno);
+		}
+	}
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+}
+
+static void add_path_beneath(struct __test_metadata *const _metadata,
+		const int ruleset_fd, const __u64 allowed_access,
+		const char *const path)
+{
+	struct landlock_path_beneath_attr path_beneath = {
+		.allowed_access = allowed_access,
+	};
+
+	path_beneath.parent_fd = open(path, O_PATH | O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd) {
+		TH_LOG("Failed to open directory \"%s\": %s", path,
+				strerror(errno));
+	}
+	ASSERT_EQ(0, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0)) {
+		TH_LOG("Failed to update the ruleset with \"%s\": %s", path,
+				strerror(errno));
+	}
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+}
+
+struct rule {
+	const char *path;
+	__u64 access;
+};
+
+#define ACCESS_RO ( \
+	LANDLOCK_ACCESS_FS_READ_FILE | \
+	LANDLOCK_ACCESS_FS_READ_DIR)
+
+#define ACCESS_RW ( \
+	ACCESS_RO | \
+	LANDLOCK_ACCESS_FS_WRITE_FILE)
+
+static int create_ruleset(struct __test_metadata *const _metadata,
+		const __u64 handled_access_fs, const struct rule rules[])
+{
+	int ruleset_fd, i;
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = handled_access_fs,
+	};
+
+	ASSERT_NE(NULL, rules) {
+		TH_LOG("No rule list");
+	}
+	ASSERT_NE(NULL, rules[0].path) {
+		TH_LOG("Empty rule list");
+	}
+
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd) {
+		TH_LOG("Failed to create a ruleset: %s", strerror(errno));
+	}
+
+	for (i = 0; rules[i].path; i++) {
+		add_path_beneath(_metadata, ruleset_fd, rules[i].access,
+				rules[i].path);
+	}
+	return ruleset_fd;
+}
+
+static void enforce_ruleset(struct __test_metadata *const _metadata,
+		const int ruleset_fd)
+{
+	ASSERT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+	ASSERT_EQ(0, landlock_restrict_self(ruleset_fd, 0)) {
+		TH_LOG("Failed to enforce ruleset: %s", strerror(errno));
+	}
+}
+
+TEST_F_FORK(layout1, proc_nsfs)
+{
+	const struct rule rules[] = {
+		{
+			.path = "/dev/null",
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	struct landlock_path_beneath_attr path_beneath;
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access |
+			LANDLOCK_ACCESS_FS_READ_DIR, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	ASSERT_EQ(0, test_open("/proc/self/ns/mnt", O_RDONLY));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	ASSERT_EQ(EACCES, test_open("/", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open("/dev", O_RDONLY));
+	ASSERT_EQ(0, test_open("/dev/null", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open("/dev/full", O_RDONLY));
+
+	ASSERT_EQ(EACCES, test_open("/proc", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open("/proc/self", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open("/proc/self/ns", O_RDONLY));
+	/*
+	 * Because nsfs is an internal filesystem, /proc/self/ns/mnt is a
+	 * disconnected path.  Such path cannot be identified and must then be
+	 * allowed.
+	 */
+	ASSERT_EQ(0, test_open("/proc/self/ns/mnt", O_RDONLY));
+
+	/*
+	 * Checks that it is not possible to add nsfs-like filesystem
+	 * references to a ruleset.
+	 */
+	path_beneath.allowed_access = LANDLOCK_ACCESS_FS_READ_FILE |
+		LANDLOCK_ACCESS_FS_WRITE_FILE,
+	path_beneath.parent_fd = open("/proc/self/ns/mnt", O_PATH | O_CLOEXEC);
+	ASSERT_LE(0, path_beneath.parent_fd);
+	ASSERT_EQ(-1, landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH,
+				&path_beneath, 0));
+	ASSERT_EQ(EBADFD, errno);
+	ASSERT_EQ(0, close(path_beneath.parent_fd));
+}
+
+TEST_F_FORK(layout1, unpriv) {
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	int ruleset_fd;
+
+	drop_caps(_metadata);
+
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RO, rules);
+	ASSERT_LE(0, ruleset_fd);
+	ASSERT_EQ(-1, landlock_restrict_self(ruleset_fd, 0));
+	ASSERT_EQ(EPERM, errno);
+
+	/* enforce_ruleset() calls prctl(no_new_privs). */
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+TEST_F_FORK(layout1, effective_access)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = ACCESS_RO,
+		},
+		{
+			.path = file1_s2d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+	char buf;
+	int reg_fd;
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Tests on a directory. */
+	ASSERT_EQ(EACCES, test_open("/", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+
+	/* Tests on a file. */
+	ASSERT_EQ(EACCES, test_open(dir_s2d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d2, O_RDONLY));
+
+	/* Checks effective read and write actions. */
+	reg_fd = open(file1_s2d2, O_RDWR | O_CLOEXEC);
+	ASSERT_LE(0, reg_fd);
+	ASSERT_EQ(1, write(reg_fd, ".", 1));
+	ASSERT_LE(0, lseek(reg_fd, 0, SEEK_SET));
+	ASSERT_EQ(1, read(reg_fd, &buf, 1));
+	ASSERT_EQ('.', buf);
+	ASSERT_EQ(0, close(reg_fd));
+
+	/* Just in case, double-checks effective actions. */
+	reg_fd = open(file1_s2d2, O_RDONLY | O_CLOEXEC);
+	ASSERT_LE(0, reg_fd);
+	ASSERT_EQ(-1, write(reg_fd, &buf, 1));
+	ASSERT_EQ(EBADF, errno);
+	ASSERT_EQ(0, close(reg_fd));
+}
+
+TEST_F_FORK(layout1, unhandled_access)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	/* Here, we only handle read accesses, not write accesses. */
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RO, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/*
+	 * Because the policy does not handle LANDLOCK_ACCESS_FS_WRITE_FILE,
+	 * opening for write-only should be allowed, but not read-write.
+	 */
+	ASSERT_EQ(0, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDWR));
+
+	ASSERT_EQ(0, test_open(file1_s1d2, O_WRONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDWR));
+}
+
+TEST_F_FORK(layout1, ruleset_overlap)
+{
+	const struct rule rules[] = {
+		/* These rules should be ORed among them. */
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_READ_DIR,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks s1d1 hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDWR));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	/* Checks s1d2 hierarchy. */
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_WRONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDWR));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* Checks s1d3 hierarchy. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDWR));
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+}
+
+TEST_F_FORK(layout1, non_overlapping_accesses)
+{
+	const struct rule layer1[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_MAKE_REG,
+		},
+		{}
+	};
+	const struct rule layer2[] = {
+		{
+			.path = dir_s1d3,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_FILE,
+		},
+		{}
+	};
+	int ruleset_fd;
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_MAKE_REG,
+			layer1);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(-1, mknod(file1_s1d1, S_IFREG | 0700, 0));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(0, mknod(file1_s1d2, S_IFREG | 0700, 0));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_REMOVE_FILE,
+			layer2);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Unchanged accesses for file creation. */
+	ASSERT_EQ(-1, mknod(file1_s1d1, S_IFREG | 0700, 0));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(0, mknod(file1_s1d2, S_IFREG | 0700, 0));
+
+	/* Checks file removing. */
+	ASSERT_EQ(-1, unlink(file1_s1d2));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(0, unlink(file1_s1d3));
+}
+
+TEST_F_FORK(layout1, interleaved_masked_accesses)
+{
+	/*
+	 * Checks overly restrictive rules:
+	 * layer 1: allows R   s1d1/s1d2/s1d3/file1
+	 * layer 2: allows RW  s1d1/s1d2/s1d3
+	 *          allows  W  s1d1/s1d2
+	 *          denies R   s1d1/s1d2
+	 * layer 3: allows R   s1d1
+	 * layer 4: allows R   s1d1/s1d2
+	 *          denies  W  s1d1/s1d2
+	 * layer 5: allows R   s1d1/s1d2
+	 * layer 6: allows   X ----
+	 * layer 7: allows  W  s1d1/s1d2
+	 *          denies R   s1d1/s1d2
+	 */
+	const struct rule layer1_read[] = {
+		/* Allows read access to file1_s1d3 with the first layer. */
+		{
+			.path = file1_s1d3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	/* First rule with write restrictions. */
+	const struct rule layer2_read_write[] = {
+		/* Start by granting read-write access via its parent directory... */
+		{
+			.path = dir_s1d3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		/* ...but also denies read access via its grandparent directory. */
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	const struct rule layer3_read[] = {
+		/* Allows read access via its great-grandparent directory. */
+		{
+			.path = dir_s1d1,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	const struct rule layer4_read_write[] = {
+		/*
+		 * Try to confuse the deny access by denying write (but not
+		 * read) access via its grandparent directory.
+		 */
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	const struct rule layer5_read[] = {
+		/*
+		 * Try to override layer2's deny read access by explicitly
+		 * allowing read access via file1_s1d3's grandparent.
+		 */
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	const struct rule layer6_execute[] = {
+		/*
+		 * Restricts an unrelated file hierarchy with a new access
+		 * (non-overlapping) type.
+		 */
+		{
+			.path = dir_s2d1,
+			.access = LANDLOCK_ACCESS_FS_EXECUTE,
+		},
+		{}
+	};
+	const struct rule layer7_read_write[] = {
+		/*
+		 * Finally, denies read access to file1_s1d3 via its
+		 * grandparent.
+		 */
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	int ruleset_fd;
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE,
+			layer1_read);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks that read access is granted for file1_s1d3 with layer 1. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDWR));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file2_s1d3, O_WRONLY));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE, layer2_read_write);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks that previous access rights are unchanged with layer 2. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDWR));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file2_s1d3, O_WRONLY));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE,
+			layer3_read);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks that previous access rights are unchanged with layer 3. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDWR));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file2_s1d3, O_WRONLY));
+
+	/* This time, denies write access for the file hierarchy. */
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE, layer4_read_write);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/*
+	 * Checks that the only change with layer 4 is that write access is
+	 * denied.
+	 */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_WRONLY));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE,
+			layer5_read);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks that previous access rights are unchanged with layer 5. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_EXECUTE,
+			layer6_execute);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks that previous access rights are unchanged with layer 6. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+
+	ruleset_fd = create_ruleset(_metadata, LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE, layer7_read_write);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks read access is now denied with layer 7. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(file2_s1d3, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, inherit_subset)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_READ_DIR,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	/* Write access is forbidden. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	/* Readdir access is allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* Write access is forbidden. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	/* Readdir access is allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/*
+	 * Tests shared rule extension: the following rules should not grant
+	 * any new access, only remove some.  Once enforced, these rules are
+	 * ANDed with the previous ones.
+	 */
+	add_path_beneath(_metadata, ruleset_fd, LANDLOCK_ACCESS_FS_WRITE_FILE,
+			dir_s1d2);
+	/*
+	 * According to ruleset_fd, dir_s1d2 should now have the
+	 * LANDLOCK_ACCESS_FS_READ_FILE and LANDLOCK_ACCESS_FS_WRITE_FILE
+	 * access rights (even if this directory is opened a second time).
+	 * However, when enforcing this updated ruleset, the ruleset tied to
+	 * the current process (i.e. its domain) will still only have the
+	 * dir_s1d2 with LANDLOCK_ACCESS_FS_READ_FILE and
+	 * LANDLOCK_ACCESS_FS_READ_DIR accesses, but
+	 * LANDLOCK_ACCESS_FS_WRITE_FILE must not be allowed because it would
+	 * be a privilege escalation.
+	 */
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	/* Same tests and results as above. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d2. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	/* Readdir access is still allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d3. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	/* Readdir access is still allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/*
+	 * Try to get more privileges by adding new access rights to the parent
+	 * directory: dir_s1d1.
+	 */
+	add_path_beneath(_metadata, ruleset_fd, ACCESS_RW, dir_s1d1);
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	/* Same tests and results as above. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d2. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	/* Readdir access is still allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d3. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	/* Readdir access is still allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/*
+	 * Now, dir_s1d3 get a new rule tied to it, only allowing
+	 * LANDLOCK_ACCESS_FS_WRITE_FILE.  The (kernel internal) difference is
+	 * that there was no rule tied to it before.
+	 */
+	add_path_beneath(_metadata, ruleset_fd, LANDLOCK_ACCESS_FS_WRITE_FILE,
+			dir_s1d3);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/*
+	 * Same tests and results as above, except for open(dir_s1d3) which is
+	 * now denied because the new rule mask the rule previously inherited
+	 * from dir_s1d2.
+	 */
+
+	/* Same tests and results as above. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d2. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	/* Readdir access is still allowed. */
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* It is still forbidden to write in file1_s1d3. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	/*
+	 * Readdir of dir_s1d3 is still allowed because of the OR policy inside
+	 * the same layer.
+	 */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+}
+
+TEST_F_FORK(layout1, inherit_superset)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d3,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	/* Readdir access is denied for dir_s1d2. */
+	ASSERT_EQ(EACCES, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+	/* Readdir access is allowed for dir_s1d3. */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+	/* File access is allowed for file1_s1d3. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+
+	/* Now dir_s1d2, parent of dir_s1d3, gets a new rule tied to it. */
+	add_path_beneath(_metadata, ruleset_fd, LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_READ_DIR, dir_s1d2);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Readdir access is still denied for dir_s1d2. */
+	ASSERT_EQ(EACCES, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+	/* Readdir access is still allowed for dir_s1d3. */
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+	/* File access is still allowed for file1_s1d3. */
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, max_layers)
+{
+	int i, err;
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	for (i = 0; i < 64; i++)
+		enforce_ruleset(_metadata, ruleset_fd);
+
+	for (i = 0; i < 2; i++) {
+		err = landlock_restrict_self(ruleset_fd, 0);
+		ASSERT_EQ(-1, err);
+		ASSERT_EQ(E2BIG, errno);
+	}
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+TEST_F_FORK(layout1, empty_or_same_ruleset)
+{
+	struct landlock_ruleset_attr ruleset_attr = {};
+	int ruleset_fd;
+
+	/* Tests empty handled_access_fs. */
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(-1, ruleset_fd);
+	ASSERT_EQ(ENOMSG, errno);
+
+	/* Enforces policy which deny read access to all files. */
+	ruleset_attr.handled_access_fs = LANDLOCK_ACCESS_FS_READ_FILE;
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+
+	/* Nests a policy which deny read access to all directories. */
+	ruleset_attr.handled_access_fs = LANDLOCK_ACCESS_FS_READ_DIR;
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY));
+
+	/* Enforces a second time with the same ruleset. */
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+TEST_F_FORK(layout1, rule_on_mountpoint)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d1,
+			.access = ACCESS_RO,
+		},
+		{
+			/* dir_s3d2 is a mount point. */
+			.path = dir_s3d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+
+	ASSERT_EQ(EACCES, test_open(dir_s2d1, O_RDONLY));
+
+	ASSERT_EQ(EACCES, test_open(dir_s3d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d3, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, rule_over_mountpoint)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d1,
+			.access = ACCESS_RO,
+		},
+		{
+			/* dir_s3d2 is a mount point. */
+			.path = dir_s3d1,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+
+	ASSERT_EQ(EACCES, test_open(dir_s2d1, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(dir_s3d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s3d3, O_RDONLY));
+}
+
+/*
+ * This test verifies that we can apply a landlock rule on the root directory
+ * (which might require special handling).
+ */
+TEST_F_FORK(layout1, rule_over_root_allow_then_deny)
+{
+	struct rule rules[] = {
+		{
+			.path = "/",
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks allowed access. */
+	ASSERT_EQ(0, test_open("/", O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+
+	rules[0].access = LANDLOCK_ACCESS_FS_READ_FILE;
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks denied access (on a directory). */
+	ASSERT_EQ(EACCES, test_open("/", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, rule_over_root_deny)
+{
+	const struct rule rules[] = {
+		{
+			.path = "/",
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks denied access (on a directory). */
+	ASSERT_EQ(EACCES, test_open("/", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY));
+}
+
+TEST_F_FORK(layout1, rule_inside_mount_ns)
+{
+	const struct rule rules[] = {
+		{
+			.path = "s3d3",
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	int ruleset_fd;
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, syscall(SYS_pivot_root, dir_s3d2, dir_s3d3)) {
+		TH_LOG("Failed to pivot root: %s", strerror(errno));
+	};
+	ASSERT_EQ(0, chdir("/"));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, test_open("s3d3", O_RDONLY));
+	ASSERT_EQ(EACCES, test_open("/", O_RDONLY));
+}
+
+TEST_F_FORK(layout1, mount_and_pivot)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s3d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(-1, mount(NULL, dir_s3d2, NULL, MS_RDONLY, NULL));
+	ASSERT_EQ(EPERM, errno);
+	ASSERT_EQ(-1, syscall(SYS_pivot_root, dir_s3d2, dir_s3d3));
+	ASSERT_EQ(EPERM, errno);
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+}
+
+TEST_F_FORK(layout1, move_mount)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s3d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, syscall(SYS_move_mount, AT_FDCWD, dir_s3d2, AT_FDCWD,
+				dir_s1d2, 0)) {
+		TH_LOG("Failed to move mount: %s", strerror(errno));
+	}
+
+	ASSERT_EQ(0, syscall(SYS_move_mount, AT_FDCWD, dir_s1d2, AT_FDCWD,
+				dir_s3d2, 0));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(-1, syscall(SYS_move_mount, AT_FDCWD, dir_s3d2, AT_FDCWD,
+				dir_s1d2, 0));
+	ASSERT_EQ(EPERM, errno);
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+}
+
+TEST_F_FORK(layout1, release_inodes)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d1,
+			.access = ACCESS_RO,
+		},
+		{
+			.path = dir_s3d2,
+			.access = ACCESS_RO,
+		},
+		{
+			.path = dir_s3d3,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, ACCESS_RW, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	/* Unmount a file hierarchy while it is being used by a ruleset. */
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, umount(dir_s3d2));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s3d2, O_RDONLY));
+	/* This dir_s3d3 would not be allowed and does not exist anyway. */
+	ASSERT_EQ(ENOENT, test_open(dir_s3d3, O_RDONLY));
+}
+
+enum relative_access {
+	REL_OPEN,
+	REL_CHDIR,
+	REL_CHROOT_ONLY,
+	REL_CHROOT_CHDIR,
+};
+
+static void test_relative_path(struct __test_metadata *const _metadata,
+		const enum relative_access rel)
+{
+	/*
+	 * Common layer to check that chroot doesn't ignore it (i.e. a chroot
+	 * is not a disconnected root directory).
+	 */
+	const struct rule layer1_base[] = {
+		{
+			.path = TMP_DIR,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	const struct rule layer2_subs[] = {
+		{
+			.path = dir_s1d2,
+			.access = ACCESS_RO,
+		},
+		{
+			.path = dir_s2d2,
+			.access = ACCESS_RO,
+		},
+		{}
+	};
+	int dirfd, ruleset_fd;
+
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer1_base);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer2_subs);
+
+	ASSERT_LE(0, ruleset_fd);
+	switch (rel) {
+	case REL_OPEN:
+	case REL_CHDIR:
+		break;
+	case REL_CHROOT_ONLY:
+		ASSERT_EQ(0, chdir(dir_s2d2));
+		break;
+	case REL_CHROOT_CHDIR:
+		ASSERT_EQ(0, chdir(dir_s1d2));
+		break;
+	default:
+		ASSERT_TRUE(false);
+		return;
+	}
+
+	set_cap(_metadata, CAP_SYS_CHROOT);
+	enforce_ruleset(_metadata, ruleset_fd);
+
+	switch (rel) {
+	case REL_OPEN:
+		dirfd = open(dir_s1d2, O_DIRECTORY);
+		ASSERT_LE(0, dirfd);
+		break;
+	case REL_CHDIR:
+		ASSERT_EQ(0, chdir(dir_s1d2));
+		dirfd = AT_FDCWD;
+		break;
+	case REL_CHROOT_ONLY:
+		/* Do chroot into dir_s1d2 (relative to dir_s2d2). */
+		ASSERT_EQ(0, chroot("../../s1d1/s1d2")) {
+			TH_LOG("Failed to chroot: %s", strerror(errno));
+		}
+		dirfd = AT_FDCWD;
+		break;
+	case REL_CHROOT_CHDIR:
+		/* Do chroot into dir_s1d2. */
+		ASSERT_EQ(0, chroot(".")) {
+			TH_LOG("Failed to chroot: %s", strerror(errno));
+		}
+		dirfd = AT_FDCWD;
+		break;
+	}
+
+	ASSERT_EQ((rel == REL_CHROOT_CHDIR) ? 0 : EACCES,
+			test_open_rel(dirfd, "..", O_RDONLY));
+	ASSERT_EQ(0, test_open_rel(dirfd, ".", O_RDONLY));
+
+	if (rel == REL_CHROOT_ONLY) {
+		/* The current directory is dir_s2d2. */
+		ASSERT_EQ(0, test_open_rel(dirfd, "./s2d3", O_RDONLY));
+	} else {
+		/* The current directory is dir_s1d2. */
+		ASSERT_EQ(0, test_open_rel(dirfd, "./s1d3", O_RDONLY));
+	}
+
+	if (rel == REL_CHROOT_ONLY || rel == REL_CHROOT_CHDIR) {
+		/* Checks the root dir_s1d2. */
+		ASSERT_EQ(0, test_open_rel(dirfd, "/..", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "/", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "/f1", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "/s1d3", O_RDONLY));
+	}
+
+	if (rel != REL_CHROOT_CHDIR) {
+		ASSERT_EQ(EACCES, test_open_rel(dirfd, "../../s1d1", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "../../s1d1/s1d2", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "../../s1d1/s1d2/s1d3", O_RDONLY));
+
+		ASSERT_EQ(EACCES, test_open_rel(dirfd, "../../s2d1", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "../../s2d1/s2d2", O_RDONLY));
+		ASSERT_EQ(0, test_open_rel(dirfd, "../../s2d1/s2d2/s2d3", O_RDONLY));
+	}
+
+	if (rel == REL_OPEN)
+		ASSERT_EQ(0, close(dirfd));
+	ASSERT_EQ(0, close(ruleset_fd));
+}
+
+TEST_F_FORK(layout1, relative_open)
+{
+	test_relative_path(_metadata, REL_OPEN);
+}
+
+TEST_F_FORK(layout1, relative_chdir)
+{
+	test_relative_path(_metadata, REL_CHDIR);
+}
+
+TEST_F_FORK(layout1, relative_chroot_only)
+{
+	test_relative_path(_metadata, REL_CHROOT_ONLY);
+}
+
+TEST_F_FORK(layout1, relative_chroot_chdir)
+{
+	test_relative_path(_metadata, REL_CHROOT_CHDIR);
+}
+
+static void copy_binary(struct __test_metadata *const _metadata,
+		const char *const dst_path)
+{
+	int dst_fd, src_fd;
+	struct stat statbuf;
+
+	dst_fd = open(dst_path, O_WRONLY | O_TRUNC | O_CLOEXEC);
+	ASSERT_LE(0, dst_fd) {
+		TH_LOG("Failed to open \"%s\": %s", dst_path,
+				strerror(errno));
+	}
+	src_fd = open(BINARY_PATH, O_RDONLY | O_CLOEXEC);
+	ASSERT_LE(0, src_fd) {
+		TH_LOG("Failed to open \"" BINARY_PATH "\": %s",
+				strerror(errno));
+	}
+	ASSERT_EQ(0, fstat(src_fd, &statbuf));
+	ASSERT_EQ(statbuf.st_size, sendfile(dst_fd, src_fd, 0,
+				statbuf.st_size));
+	ASSERT_EQ(0, close(src_fd));
+	ASSERT_EQ(0, close(dst_fd));
+}
+
+static void test_execute(struct __test_metadata *const _metadata,
+		const int err, const char *const path)
+{
+	int status;
+	char *const argv[] = {(char *)path, NULL};
+	const pid_t child = fork();
+
+	ASSERT_LE(0, child);
+	if (child == 0) {
+		ASSERT_EQ(err ? -1 : 0, execve(path, argv, NULL)) {
+			TH_LOG("Failed to execute \"%s\": %s", path,
+					strerror(errno));
+		};
+		ASSERT_EQ(err, errno);
+		_exit(_metadata->passed ? 2 : 1);
+		return;
+	}
+	ASSERT_EQ(child, waitpid(child, &status, 0));
+	ASSERT_EQ(1, WIFEXITED(status));
+	ASSERT_EQ(err ? 2 : 0, WEXITSTATUS(status)) {
+		TH_LOG("Unexpected return code for \"%s\": %s", path,
+				strerror(errno));
+	};
+}
+
+TEST_F_FORK(layout1, execute)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_EXECUTE,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	copy_binary(_metadata, file1_s1d1);
+	copy_binary(_metadata, file1_s1d2);
+	copy_binary(_metadata, file1_s1d3);
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d1, O_RDONLY));
+	test_execute(_metadata, EACCES, file1_s1d1);
+
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	test_execute(_metadata, 0, file1_s1d2);
+
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	test_execute(_metadata, 0, file1_s1d3);
+}
+
+TEST_F_FORK(layout1, link)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_MAKE_REG,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlink(file1_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(-1, link(file2_s1d1, file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	/* Denies linking because of reparenting. */
+	ASSERT_EQ(-1, link(file1_s2d1, file1_s1d2));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, link(file2_s1d2, file1_s1d3));
+	ASSERT_EQ(EXDEV, errno);
+
+	ASSERT_EQ(0, link(file2_s1d2, file1_s1d2));
+	ASSERT_EQ(0, link(file2_s1d3, file1_s1d3));
+}
+
+TEST_F_FORK(layout1, rename_file)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d3,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_FILE,
+		},
+		{
+			.path = dir_s2d2,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_FILE,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/*
+	 * Tries to replace a file, from a directory that allows file removal,
+	 * but to a different directory (which also allows file removal).
+	 */
+	ASSERT_EQ(-1, rename(file1_s2d3, file1_s1d3));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s2d3, AT_FDCWD, file1_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s2d3, AT_FDCWD, dir_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+
+	/*
+	 * Tries to replace a file, from a directory that denies file removal,
+	 * to a different directory (which allows file removal).
+	 */
+	ASSERT_EQ(-1, rename(file1_s2d1, file1_s1d3));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s2d1, AT_FDCWD, file1_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, dir_s2d2, AT_FDCWD, file1_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+
+	/* Exchanges files and directories that partially allow removal. */
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, dir_s2d2, AT_FDCWD, file1_s2d1,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s2d1, AT_FDCWD, dir_s2d2,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EACCES, errno);
+
+	/* Renames files with different parents. */
+	ASSERT_EQ(-1, rename(file1_s2d2, file1_s1d2));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(0, unlink(file1_s1d3));
+	ASSERT_EQ(-1, rename(file1_s2d1, file1_s1d3));
+	ASSERT_EQ(EXDEV, errno);
+
+	/* Exchanges and renames files with same parent. */
+	ASSERT_EQ(0, renameat2(AT_FDCWD, file2_s2d3, AT_FDCWD, file1_s2d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(0, rename(file2_s2d3, file1_s2d3));
+
+	/* Exchanges files and directories with same parent, twice. */
+	ASSERT_EQ(0, renameat2(AT_FDCWD, file1_s2d2, AT_FDCWD, dir_s2d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(0, renameat2(AT_FDCWD, file1_s2d2, AT_FDCWD, dir_s2d3,
+				RENAME_EXCHANGE));
+}
+
+TEST_F_FORK(layout1, rename_dir)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_DIR,
+		},
+		{
+			.path = dir_s2d1,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_DIR,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	/* Empties dir_s1d3 to allow renaming. */
+	ASSERT_EQ(0, unlink(file1_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Exchanges and renames directory to a different parent. */
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, dir_s2d3, AT_FDCWD, dir_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, rename(dir_s2d3, dir_s1d3));
+	ASSERT_EQ(EXDEV, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s2d2, AT_FDCWD, dir_s1d3,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EXDEV, errno);
+
+	/*
+	 * Exchanges directory to the same parent, which doesn't allow
+	 * directory removal.
+	 */
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, dir_s1d1, AT_FDCWD, dir_s2d1,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, renameat2(AT_FDCWD, file1_s1d1, AT_FDCWD, dir_s1d2,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(EACCES, errno);
+
+	/*
+	 * Exchanges and renames directory to the same parent, which allows
+	 * directory removal.
+	 */
+	ASSERT_EQ(0, renameat2(AT_FDCWD, dir_s1d3, AT_FDCWD, file1_s1d2,
+				RENAME_EXCHANGE));
+	ASSERT_EQ(0, unlink(dir_s1d3));
+	ASSERT_EQ(0, mkdir(dir_s1d3, 0700));
+	ASSERT_EQ(0, rename(file1_s1d2, dir_s1d3));
+	ASSERT_EQ(0, rmdir(dir_s1d3));
+}
+
+TEST_F_FORK(layout1, remove_dir)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_DIR,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlink(file1_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(0, rmdir(dir_s1d3));
+	ASSERT_EQ(0, mkdir(dir_s1d3, 0700));
+	ASSERT_EQ(0, unlinkat(AT_FDCWD, dir_s1d3, AT_REMOVEDIR));
+
+	/* dir_s1d2 itself cannot be removed. */
+	ASSERT_EQ(-1, rmdir(dir_s1d2));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, unlinkat(AT_FDCWD, dir_s1d2, AT_REMOVEDIR));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, rmdir(dir_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, unlinkat(AT_FDCWD, dir_s1d1, AT_REMOVEDIR));
+	ASSERT_EQ(EACCES, errno);
+}
+
+TEST_F_FORK(layout1, remove_file)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_REMOVE_FILE,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(-1, unlink(file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, unlinkat(AT_FDCWD, file1_s1d1, 0));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlinkat(AT_FDCWD, file1_s1d3, 0));
+}
+
+static void test_make_file(struct __test_metadata *const _metadata,
+		const __u64 access, const mode_t mode, const dev_t dev)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = access,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, access, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file2_s1d1));
+	ASSERT_EQ(0, mknod(file2_s1d1, mode | 0400, dev)) {
+		TH_LOG("Failed to make file \"%s\": %s",
+				file2_s1d1, strerror(errno));
+	};
+
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlink(file2_s1d2));
+
+	ASSERT_EQ(0, unlink(file1_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(-1, mknod(file1_s1d1, mode | 0400, dev));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, link(file2_s1d1, file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, rename(file2_s1d1, file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+
+	ASSERT_EQ(0, mknod(file1_s1d2, mode | 0400, dev)) {
+		TH_LOG("Failed to make file \"%s\": %s",
+				file1_s1d2, strerror(errno));
+	};
+	ASSERT_EQ(0, link(file1_s1d2, file2_s1d2));
+	ASSERT_EQ(0, unlink(file2_s1d2));
+	ASSERT_EQ(0, rename(file1_s1d2, file2_s1d2));
+
+	ASSERT_EQ(0, mknod(file1_s1d3, mode | 0400, dev));
+	ASSERT_EQ(0, link(file1_s1d3, file2_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+	ASSERT_EQ(0, rename(file1_s1d3, file2_s1d3));
+}
+
+TEST_F_FORK(layout1, make_char)
+{
+	/* Creates a /dev/null device. */
+	set_cap(_metadata, CAP_MKNOD);
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_CHAR, S_IFCHR,
+			makedev(1, 3));
+}
+
+TEST_F_FORK(layout1, make_block)
+{
+	/* Creates a /dev/loop0 device. */
+	set_cap(_metadata, CAP_MKNOD);
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_BLOCK, S_IFBLK,
+			makedev(7, 0));
+}
+
+TEST_F_FORK(layout1, make_reg_1)
+{
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_REG, S_IFREG, 0);
+}
+
+TEST_F_FORK(layout1, make_reg_2)
+{
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_REG, 0, 0);
+}
+
+TEST_F_FORK(layout1, make_sock)
+{
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_SOCK, S_IFSOCK, 0);
+}
+
+TEST_F_FORK(layout1, make_fifo)
+{
+	test_make_file(_metadata, LANDLOCK_ACCESS_FS_MAKE_FIFO, S_IFIFO, 0);
+}
+
+TEST_F_FORK(layout1, make_sym)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_MAKE_SYM,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file2_s1d1));
+	ASSERT_EQ(0, symlink("none", file2_s1d1));
+
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlink(file2_s1d2));
+
+	ASSERT_EQ(0, unlink(file1_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(-1, symlink("none", file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, link(file2_s1d1, file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(-1, rename(file2_s1d1, file1_s1d1));
+	ASSERT_EQ(EACCES, errno);
+
+	ASSERT_EQ(0, symlink("none", file1_s1d2));
+	ASSERT_EQ(0, link(file1_s1d2, file2_s1d2));
+	ASSERT_EQ(0, unlink(file2_s1d2));
+	ASSERT_EQ(0, rename(file1_s1d2, file2_s1d2));
+
+	ASSERT_EQ(0, symlink("none", file1_s1d3));
+	ASSERT_EQ(0, link(file1_s1d3, file2_s1d3));
+	ASSERT_EQ(0, unlink(file2_s1d3));
+	ASSERT_EQ(0, rename(file1_s1d3, file2_s1d3));
+}
+
+TEST_F_FORK(layout1, make_dir)
+{
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_MAKE_DIR,
+		},
+		{}
+	};
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+
+	ASSERT_EQ(0, unlink(file1_s1d1));
+	ASSERT_EQ(0, unlink(file1_s1d2));
+	ASSERT_EQ(0, unlink(file1_s1d3));
+
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Uses file_* as directory names. */
+	ASSERT_EQ(-1, mkdir(file1_s1d1, 0700));
+	ASSERT_EQ(EACCES, errno);
+	ASSERT_EQ(0, mkdir(file1_s1d2, 0700));
+	ASSERT_EQ(0, mkdir(file1_s1d3, 0700));
+}
+
+static int open_proc_fd(struct __test_metadata *const _metadata, const int fd,
+		const int open_flags)
+{
+	static const char path_template[] = "/proc/self/fd/%d";
+	char procfd_path[sizeof(path_template) + 10];
+	const int procfd_path_size = snprintf(procfd_path, sizeof(procfd_path),
+			path_template, fd);
+
+	ASSERT_LT(procfd_path_size, sizeof(procfd_path));
+	return open(procfd_path, open_flags);
+}
+
+TEST_F_FORK(layout1, proc_unlinked_file)
+{
+	const struct rule rules[] = {
+		{
+			.path = file1_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	int reg_fd, proc_fd;
+	const int ruleset_fd = create_ruleset(_metadata,
+			LANDLOCK_ACCESS_FS_READ_FILE |
+			LANDLOCK_ACCESS_FS_WRITE_FILE, rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_RDWR));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	reg_fd = open(file1_s1d2, O_RDONLY | O_CLOEXEC);
+	ASSERT_LE(0, reg_fd);
+	ASSERT_EQ(0, unlink(file1_s1d2));
+
+	proc_fd = open_proc_fd(_metadata, reg_fd, O_RDONLY | O_CLOEXEC);
+	ASSERT_LE(0, proc_fd);
+	ASSERT_EQ(0, close(proc_fd));
+
+	proc_fd = open_proc_fd(_metadata, reg_fd, O_RDWR | O_CLOEXEC);
+	ASSERT_EQ(-1, proc_fd) {
+		TH_LOG("Successfully opened /proc/self/fd/%d: %s",
+				reg_fd, strerror(errno));
+	}
+	ASSERT_EQ(EACCES, errno);
+
+	ASSERT_EQ(0, close(reg_fd));
+}
+
+TEST_F_FORK(layout1, proc_pipe)
+{
+	int proc_fd;
+	int pipe_fds[2];
+	char buf = '\0';
+	const struct rule rules[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	/* Limits read and write access to files tied to the filesystem. */
+	const int ruleset_fd = create_ruleset(_metadata, rules[0].access,
+			rules);
+
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks enforcement for normal files. */
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDWR));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDWR));
+
+	/* Checks access to pipes through FD. */
+	ASSERT_EQ(0, pipe2(pipe_fds, O_CLOEXEC));
+	ASSERT_EQ(1, write(pipe_fds[1], ".", 1)) {
+		TH_LOG("Failed to write in pipe: %s", strerror(errno));
+	}
+	ASSERT_EQ(1, read(pipe_fds[0], &buf, 1));
+	ASSERT_EQ('.', buf);
+
+	/* Checks write access to pipe through /proc/self/fd . */
+	proc_fd = open_proc_fd(_metadata, pipe_fds[1], O_WRONLY | O_CLOEXEC);
+	ASSERT_LE(0, proc_fd);
+	ASSERT_EQ(1, write(proc_fd, ".", 1)) {
+		TH_LOG("Failed to write through /proc/self/fd/%d: %s",
+				pipe_fds[1], strerror(errno));
+	}
+	ASSERT_EQ(0, close(proc_fd));
+
+	/* Checks read access to pipe through /proc/self/fd . */
+	proc_fd = open_proc_fd(_metadata, pipe_fds[0], O_RDONLY | O_CLOEXEC);
+	ASSERT_LE(0, proc_fd);
+	buf = '\0';
+	ASSERT_EQ(1, read(proc_fd, &buf, 1)) {
+		TH_LOG("Failed to read through /proc/self/fd/%d: %s",
+				pipe_fds[1], strerror(errno));
+	}
+	ASSERT_EQ(0, close(proc_fd));
+
+	ASSERT_EQ(0, close(pipe_fds[0]));
+	ASSERT_EQ(0, close(pipe_fds[1]));
+}
+
+FIXTURE(layout1_bind) {
+};
+
+FIXTURE_SETUP(layout1_bind)
+{
+	prepare_layout(_metadata);
+
+	create_layout1(_metadata);
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, mount(dir_s1d2, dir_s2d2, NULL, MS_BIND, NULL));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+}
+
+FIXTURE_TEARDOWN(layout1_bind)
+{
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, umount(dir_s2d2));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+
+	remove_layout1(_metadata);
+
+	cleanup_layout(_metadata);
+}
+
+static const char bind_dir_s1d3[] = TMP_DIR "/s2d1/s2d2/s1d3";
+static const char bind_file1_s1d3[] = TMP_DIR "/s2d1/s2d2/s1d3/f1";
+
+/*
+ * layout1_bind hierarchy:
+ *
+ * tmp
+ * ├── s1d1
+ * │   ├── f1
+ * │   ├── f2
+ * │   └── s1d2
+ * │       ├── f1
+ * │       ├── f2
+ * │       └── s1d3
+ * │           ├── f1
+ * │           └── f2
+ * ├── s2d1
+ * │   ├── f1
+ * │   └── s2d2
+ * │       ├── f1
+ * │       ├── f2
+ * │       └── s1d3
+ * │           ├── f1
+ * │           └── f2
+ * └── s3d1
+ *     └── s3d2
+ *         └── s3d3
+ */
+
+TEST_F_FORK(layout1_bind, no_restriction)
+{
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(dir_s2d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d1, O_RDONLY));
+	ASSERT_EQ(0, test_open(dir_s2d2, O_RDONLY));
+	ASSERT_EQ(0, test_open(file1_s2d2, O_RDONLY));
+	ASSERT_EQ(ENOENT, test_open(dir_s2d3, O_RDONLY));
+	ASSERT_EQ(ENOENT, test_open(file1_s2d3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(bind_dir_s1d3, O_RDONLY));
+	ASSERT_EQ(0, test_open(bind_file1_s1d3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(dir_s3d1, O_RDONLY));
+}
+
+TEST_F_FORK(layout1_bind, same_content_same_file)
+{
+	/*
+	 * Sets access right on parent directories of both source and
+	 * destination mount points.
+	 */
+	const struct rule layer1_parent[] = {
+		{
+			.path = dir_s1d1,
+			.access = ACCESS_RO,
+		},
+		{
+			.path = dir_s2d1,
+			.access = ACCESS_RW,
+		},
+		{}
+	};
+	/*
+	 * Sets access rights on the same bind-mounted directories.  The result
+	 * should be ACCESS_RW for both directories, but not both hierarchies
+	 * because of the first layer.
+	 */
+	const struct rule layer2_mount_point[] = {
+		{
+			.path = dir_s1d2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = dir_s2d2,
+			.access = ACCESS_RW,
+		},
+		{}
+	};
+	/* Only allow read-access to the s1d3 hierarchies. */
+	const struct rule layer3_source[] = {
+		{
+			.path = dir_s1d3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{}
+	};
+	/* Removes all access rights. */
+	const struct rule layer4_destination[] = {
+		{
+			.path = bind_file1_s1d3,
+			.access = LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	int ruleset_fd;
+
+	/* Sets rules for the parent directories. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer1_parent);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks source hierarchy. */
+	ASSERT_EQ(0, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(0, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* Checks destination hierarchy. */
+	ASSERT_EQ(0, test_open(file1_s2d1, O_RDWR));
+	ASSERT_EQ(0, test_open(dir_s2d1, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(file1_s2d2, O_RDWR));
+	ASSERT_EQ(0, test_open(dir_s2d2, O_RDONLY | O_DIRECTORY));
+
+	/* Sets rules for the mount points. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer2_mount_point);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks source hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d1, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	ASSERT_EQ(0, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	/* Checks destination hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s2d1, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s2d1, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s2d1, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(file1_s2d2, O_RDWR));
+	ASSERT_EQ(0, test_open(dir_s2d2, O_RDONLY | O_DIRECTORY));
+	ASSERT_EQ(0, test_open(bind_dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/* Sets a (shared) rule only on the source. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer3_source);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks source hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d2, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d2, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/* Checks destination hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s2d2, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s2d2, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(dir_s2d2, O_RDONLY | O_DIRECTORY));
+
+	ASSERT_EQ(0, test_open(bind_file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(bind_file1_s1d3, O_WRONLY));
+	ASSERT_EQ(EACCES, test_open(bind_dir_s1d3, O_RDONLY | O_DIRECTORY));
+
+	/* Sets a (shared) rule only on the destination. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer4_destination);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks source hierarchy. */
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(file1_s1d3, O_WRONLY));
+
+	/* Checks destination hierarchy. */
+	ASSERT_EQ(EACCES, test_open(bind_file1_s1d3, O_RDONLY));
+	ASSERT_EQ(EACCES, test_open(bind_file1_s1d3, O_WRONLY));
+}
+
+#define LOWER_BASE	TMP_DIR "/lower"
+#define LOWER_DATA	LOWER_BASE "/data"
+static const char lower_fl1[] = LOWER_DATA "/fl1";
+static const char lower_dl1[] = LOWER_DATA "/dl1";
+static const char lower_dl1_fl2[] = LOWER_DATA "/dl1/fl2";
+static const char lower_fo1[] = LOWER_DATA "/fo1";
+static const char lower_do1[] = LOWER_DATA "/do1";
+static const char lower_do1_fo2[] = LOWER_DATA "/do1/fo2";
+static const char lower_do1_fl3[] = LOWER_DATA "/do1/fl3";
+
+static const char (*lower_base_files[])[] = {
+	&lower_fl1,
+	&lower_fo1,
+	NULL
+};
+static const char (*lower_base_directories[])[] = {
+	&lower_dl1,
+	&lower_do1,
+	NULL
+};
+static const char (*lower_sub_files[])[] = {
+	&lower_dl1_fl2,
+	&lower_do1_fo2,
+	&lower_do1_fl3,
+	NULL
+};
+
+#define UPPER_BASE	TMP_DIR "/upper"
+#define UPPER_DATA	UPPER_BASE "/data"
+#define UPPER_WORK	UPPER_BASE "/work"
+static const char upper_fu1[] = UPPER_DATA "/fu1";
+static const char upper_du1[] = UPPER_DATA "/du1";
+static const char upper_du1_fu2[] = UPPER_DATA "/du1/fu2";
+static const char upper_fo1[] = UPPER_DATA "/fo1";
+static const char upper_do1[] = UPPER_DATA "/do1";
+static const char upper_do1_fo2[] = UPPER_DATA "/do1/fo2";
+static const char upper_do1_fu3[] = UPPER_DATA "/do1/fu3";
+
+static const char (*upper_base_files[])[] = {
+	&upper_fu1,
+	&upper_fo1,
+	NULL
+};
+static const char (*upper_base_directories[])[] = {
+	&upper_du1,
+	&upper_do1,
+	NULL
+};
+static const char (*upper_sub_files[])[] = {
+	&upper_du1_fu2,
+	&upper_do1_fo2,
+	&upper_do1_fu3,
+	NULL
+};
+
+#define MERGE_BASE	TMP_DIR "/merge"
+#define MERGE_DATA	MERGE_BASE "/data"
+static const char merge_fl1[] = MERGE_DATA "/fl1";
+static const char merge_dl1[] = MERGE_DATA "/dl1";
+static const char merge_dl1_fl2[] = MERGE_DATA "/dl1/fl2";
+static const char merge_fu1[] = MERGE_DATA "/fu1";
+static const char merge_du1[] = MERGE_DATA "/du1";
+static const char merge_du1_fu2[] = MERGE_DATA "/du1/fu2";
+static const char merge_fo1[] = MERGE_DATA "/fo1";
+static const char merge_do1[] = MERGE_DATA "/do1";
+static const char merge_do1_fo2[] = MERGE_DATA "/do1/fo2";
+static const char merge_do1_fl3[] = MERGE_DATA "/do1/fl3";
+static const char merge_do1_fu3[] = MERGE_DATA "/do1/fu3";
+
+static const char (*merge_base_files[])[] = {
+	&merge_fl1,
+	&merge_fu1,
+	&merge_fo1,
+	NULL
+};
+static const char (*merge_base_directories[])[] = {
+	&merge_dl1,
+	&merge_du1,
+	&merge_do1,
+	NULL
+};
+static const char (*merge_sub_files[])[] = {
+	&merge_dl1_fl2,
+	&merge_du1_fu2,
+	&merge_do1_fo2,
+	&merge_do1_fl3,
+	&merge_do1_fu3,
+	NULL
+};
+
+/*
+ * layout2_overlay hierarchy:
+ *
+ * tmp
+ * ├── lower
+ * │   └── data
+ * │       ├── dl1
+ * │       │   └── fl2
+ * │       ├── do1
+ * │       │   ├── fl3
+ * │       │   └── fo2
+ * │       ├── fl1
+ * │       └── fo1
+ * ├── merge
+ * │   └── data
+ * │       ├── dl1
+ * │       │   └── fl2
+ * │       ├── do1
+ * │       │   ├── fl3
+ * │       │   ├── fo2
+ * │       │   └── fu3
+ * │       ├── du1
+ * │       │   └── fu2
+ * │       ├── fl1
+ * │       ├── fo1
+ * │       └── fu1
+ * └── upper
+ *     ├── data
+ *     │   ├── do1
+ *     │   │   ├── fo2
+ *     │   │   └── fu3
+ *     │   ├── du1
+ *     │   │   └── fu2
+ *     │   ├── fo1
+ *     │   └── fu1
+ *     └── work
+ *         └── work
+ */
+
+FIXTURE(layout2_overlay) {
+};
+
+FIXTURE_SETUP(layout2_overlay)
+{
+	prepare_layout(_metadata);
+
+	create_directory(_metadata, LOWER_BASE);
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	/* Creates tmpfs mount points to get deterministic overlayfs. */
+	ASSERT_EQ(0, mount("tmp", LOWER_BASE, "tmpfs", 0, "size=4m,mode=700"));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	create_file(_metadata, lower_fl1);
+	create_file(_metadata, lower_dl1_fl2);
+	create_file(_metadata, lower_fo1);
+	create_file(_metadata, lower_do1_fo2);
+	create_file(_metadata, lower_do1_fl3);
+
+	create_directory(_metadata, UPPER_BASE);
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	ASSERT_EQ(0, mount("tmp", UPPER_BASE, "tmpfs", 0, "size=4m,mode=700"));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	create_file(_metadata, upper_fu1);
+	create_file(_metadata, upper_du1_fu2);
+	create_file(_metadata, upper_fo1);
+	create_file(_metadata, upper_do1_fo2);
+	create_file(_metadata, upper_do1_fu3);
+	ASSERT_EQ(0, mkdir(UPPER_WORK, 0700));
+
+	create_directory(_metadata, MERGE_DATA);
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	set_cap(_metadata, CAP_DAC_OVERRIDE);
+	ASSERT_EQ(0, mount("overlay", MERGE_DATA, "overlay", 0,
+				"lowerdir=" LOWER_DATA
+				",upperdir=" UPPER_DATA
+				",workdir=" UPPER_WORK));
+	clear_cap(_metadata, CAP_DAC_OVERRIDE);
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+}
+
+FIXTURE_TEARDOWN(layout2_overlay)
+{
+	EXPECT_EQ(0, remove_path(lower_do1_fl3));
+	EXPECT_EQ(0, remove_path(lower_dl1_fl2));
+	EXPECT_EQ(0, remove_path(lower_fl1));
+	EXPECT_EQ(0, remove_path(lower_do1_fo2));
+	EXPECT_EQ(0, remove_path(lower_fo1));
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, umount(LOWER_BASE));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, remove_path(LOWER_BASE));
+
+	EXPECT_EQ(0, remove_path(upper_do1_fu3));
+	EXPECT_EQ(0, remove_path(upper_du1_fu2));
+	EXPECT_EQ(0, remove_path(upper_fu1));
+	EXPECT_EQ(0, remove_path(upper_do1_fo2));
+	EXPECT_EQ(0, remove_path(upper_fo1));
+	EXPECT_EQ(0, remove_path(UPPER_WORK "/work"));
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, umount(UPPER_BASE));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, remove_path(UPPER_BASE));
+
+	set_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, umount(MERGE_DATA));
+	clear_cap(_metadata, CAP_SYS_ADMIN);
+	EXPECT_EQ(0, remove_path(MERGE_DATA));
+
+	cleanup_layout(_metadata);
+}
+
+TEST_F_FORK(layout2_overlay, no_restriction)
+{
+	ASSERT_EQ(0, test_open(lower_fl1, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_dl1, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_dl1_fl2, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_fo1, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_do1, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_do1_fo2, O_RDONLY));
+	ASSERT_EQ(0, test_open(lower_do1_fl3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(upper_fu1, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_du1, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_du1_fu2, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_fo1, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_do1, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_do1_fo2, O_RDONLY));
+	ASSERT_EQ(0, test_open(upper_do1_fu3, O_RDONLY));
+
+	ASSERT_EQ(0, test_open(merge_fl1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_dl1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_dl1_fl2, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_fu1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_du1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_du1_fu2, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_fo1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_do1, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_do1_fo2, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_do1_fl3, O_RDONLY));
+	ASSERT_EQ(0, test_open(merge_do1_fu3, O_RDONLY));
+}
+
+#define for_each_path(path_list, path_entry, i)			\
+	for (i = 0, path_entry = *path_list[i]; path_list[i];	\
+			path_entry = *path_list[++i])
+
+TEST_F_FORK(layout2_overlay, same_content_different_file)
+{
+	/* Sets access right on parent directories of both layers. */
+	const struct rule layer1_base[] = {
+		{
+			.path = LOWER_BASE,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = UPPER_BASE,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = MERGE_BASE,
+			.access = ACCESS_RW,
+		},
+		{}
+	};
+	const struct rule layer2_data[] = {
+		{
+			.path = LOWER_DATA,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = UPPER_DATA,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = MERGE_DATA,
+			.access = ACCESS_RW,
+		},
+		{}
+	};
+	/* Sets access right on directories inside both layers. */
+	const struct rule layer3_subdirs[] = {
+		{
+			.path = lower_dl1,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = lower_do1,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = upper_du1,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = upper_do1,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = merge_dl1,
+			.access = ACCESS_RW,
+		},
+		{
+			.path = merge_du1,
+			.access = ACCESS_RW,
+		},
+		{
+			.path = merge_do1,
+			.access = ACCESS_RW,
+		},
+		{}
+	};
+	/* Tighten access rights to the files. */
+	const struct rule layer4_files[] = {
+		{
+			.path = lower_dl1_fl2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = lower_do1_fo2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = lower_do1_fl3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = upper_du1_fu2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = upper_do1_fo2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = upper_do1_fu3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE,
+		},
+		{
+			.path = merge_dl1_fl2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{
+			.path = merge_du1_fu2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{
+			.path = merge_do1_fo2,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{
+			.path = merge_do1_fl3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{
+			.path = merge_do1_fu3,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	const struct rule layer5_merge_only[] = {
+		{
+			.path = MERGE_DATA,
+			.access = LANDLOCK_ACCESS_FS_READ_FILE |
+				LANDLOCK_ACCESS_FS_WRITE_FILE,
+		},
+		{}
+	};
+	int ruleset_fd;
+	size_t i;
+	const char *path_entry;
+
+	/* Sets rules on base directories (i.e. outside overlay scope). */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer1_base);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks lower layer. */
+	for_each_path(lower_base_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	for_each_path(lower_base_directories, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(lower_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	/* Checks upper layer. */
+	for_each_path(upper_base_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	for_each_path(upper_base_directories, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(upper_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	/*
+	 * Checks that access rights are independent from the lower and upper
+	 * layers: write access to upper files viewed through the merge point
+	 * is still allowed, and write access to lower file viewed (and copied)
+	 * through the merge point is still allowed.
+	 */
+	for_each_path(merge_base_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+	for_each_path(merge_base_directories, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(merge_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+
+	/* Sets rules on data directories (i.e. inside overlay scope). */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer2_data);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks merge. */
+	for_each_path(merge_base_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+	for_each_path(merge_base_directories, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(merge_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+
+	/* Same checks with tighter rules. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer3_subdirs);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks changes for lower layer. */
+	for_each_path(lower_base_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY));
+	}
+	/* Checks changes for upper layer. */
+	for_each_path(upper_base_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY));
+	}
+	/* Checks all merge accesses. */
+	for_each_path(merge_base_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDWR));
+	}
+	for_each_path(merge_base_directories, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(merge_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+
+	/* Sets rules directly on overlayed files. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer4_files);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks unchanged accesses on lower layer. */
+	for_each_path(lower_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	/* Checks unchanged accesses on upper layer. */
+	for_each_path(upper_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDONLY));
+		ASSERT_EQ(EACCES, test_open(path_entry, O_WRONLY));
+	}
+	/* Checks all merge accesses. */
+	for_each_path(merge_base_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDWR));
+	}
+	for_each_path(merge_base_directories, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(merge_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+
+	/* Only allowes access to the merge hierarchy. */
+	ruleset_fd = create_ruleset(_metadata, ACCESS_RW, layer5_merge_only);
+	ASSERT_LE(0, ruleset_fd);
+	enforce_ruleset(_metadata, ruleset_fd);
+	ASSERT_EQ(0, close(ruleset_fd));
+
+	/* Checks new accesses on lower layer. */
+	for_each_path(lower_sub_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY));
+	}
+	/* Checks new accesses on upper layer. */
+	for_each_path(upper_sub_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY));
+	}
+	/* Checks all merge accesses. */
+	for_each_path(merge_base_files, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDWR));
+	}
+	for_each_path(merge_base_directories, path_entry, i) {
+		ASSERT_EQ(EACCES, test_open(path_entry, O_RDONLY | O_DIRECTORY));
+	}
+	for_each_path(merge_sub_files, path_entry, i) {
+		ASSERT_EQ(0, test_open(path_entry, O_RDWR));
+	}
+}
+
+TEST_HARNESS_MAIN
diff --git a/tools/testing/selftests/landlock/ptrace_test.c b/tools/testing/selftests/landlock/ptrace_test.c
new file mode 100644
index 000000000000..15fbef9cc849
--- /dev/null
+++ b/tools/testing/selftests/landlock/ptrace_test.c
@@ -0,0 +1,337 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Landlock tests - Ptrace
+ *
+ * Copyright © 2017-2020 Mickaël Salaün <mic@digikod.net>
+ * Copyright © 2019-2020 ANSSI
+ */
+
+#define _GNU_SOURCE
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/landlock.h>
+#include <signal.h>
+#include <sys/prctl.h>
+#include <sys/ptrace.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#include "common.h"
+
+static void create_domain(struct __test_metadata *const _metadata)
+{
+	int ruleset_fd;
+	struct landlock_ruleset_attr ruleset_attr = {
+		.handled_access_fs = LANDLOCK_ACCESS_FS_MAKE_BLOCK,
+	};
+
+	ruleset_fd = landlock_create_ruleset(&ruleset_attr,
+			sizeof(ruleset_attr), 0);
+	EXPECT_LE(0, ruleset_fd) {
+		TH_LOG("Failed to create a ruleset: %s", strerror(errno));
+	}
+	EXPECT_EQ(0, prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0));
+	EXPECT_EQ(0, landlock_restrict_self(ruleset_fd, 0));
+	EXPECT_EQ(0, close(ruleset_fd));
+}
+
+static int test_ptrace_read(const pid_t pid)
+{
+	static const char path_template[] = "/proc/%d/environ";
+	char procenv_path[sizeof(path_template) + 10];
+	int procenv_path_size, fd;
+
+	procenv_path_size = snprintf(procenv_path, sizeof(procenv_path),
+			path_template, pid);
+	if (procenv_path_size >= sizeof(procenv_path))
+		return E2BIG;
+
+	fd = open(procenv_path, O_RDONLY | O_CLOEXEC);
+	if (fd < 0)
+		return errno;
+	/*
+	 * Mixing error codes from close(2) and open(2) should not lead to any
+	 * (access type) confusion for this test.
+	 */
+	if (close(fd) != 0)
+		return errno;
+	return 0;
+}
+
+FIXTURE(hierarchy) { };
+
+FIXTURE_VARIANT(hierarchy) {
+	const bool domain_both;
+	const bool domain_parent;
+	const bool domain_child;
+};
+
+/*
+ * Test multiple tracing combinations between a parent process P1 and a child
+ * process P2.
+ *
+ * Yama's scoped ptrace is presumed disabled.  If enabled, this optional
+ * restriction is enforced in addition to any Landlock check, which means that
+ * all P2 requests to trace P1 would be denied.
+ */
+
+/*
+ *        No domain
+ *
+ *   P1-.               P1 -> P2 : allow
+ *       \              P2 -> P1 : allow
+ *        'P2
+ */
+FIXTURE_VARIANT_ADD(hierarchy, allow_without_domain) {
+	.domain_both = false,
+	.domain_parent = false,
+	.domain_child = false,
+};
+
+/*
+ *        Child domain
+ *
+ *   P1--.              P1 -> P2 : allow
+ *        \             P2 -> P1 : deny
+ *        .'-----.
+ *        |  P2  |
+ *        '------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, allow_with_one_domain) {
+	.domain_both = false,
+	.domain_parent = false,
+	.domain_child = true,
+};
+
+/*
+ *        Parent domain
+ * .------.
+ * |  P1  --.           P1 -> P2 : deny
+ * '------'  \          P2 -> P1 : allow
+ *            '
+ *            P2
+ */
+FIXTURE_VARIANT_ADD(hierarchy, deny_with_parent_domain) {
+	.domain_both = false,
+	.domain_parent = true,
+	.domain_child = false,
+};
+
+/*
+ *        Parent + child domain (siblings)
+ * .------.
+ * |  P1  ---.          P1 -> P2 : deny
+ * '------'   \         P2 -> P1 : deny
+ *         .---'--.
+ *         |  P2  |
+ *         '------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, deny_with_sibling_domain) {
+	.domain_both = false,
+	.domain_parent = true,
+	.domain_child = true,
+};
+
+/*
+ *         Same domain (inherited)
+ * .-------------.
+ * | P1----.     |      P1 -> P2 : allow
+ * |        \    |      P2 -> P1 : allow
+ * |         '   |
+ * |         P2  |
+ * '-------------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, allow_sibling_domain) {
+	.domain_both = true,
+	.domain_parent = false,
+	.domain_child = false,
+};
+
+/*
+ *         Inherited + child domain
+ * .-----------------.
+ * |  P1----.        |  P1 -> P2 : allow
+ * |         \       |  P2 -> P1 : deny
+ * |        .-'----. |
+ * |        |  P2  | |
+ * |        '------' |
+ * '-----------------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, allow_with_nested_domain) {
+	.domain_both = true,
+	.domain_parent = false,
+	.domain_child = true,
+};
+
+/*
+ *         Inherited + parent domain
+ * .-----------------.
+ * |.------.         |  P1 -> P2 : deny
+ * ||  P1  ----.     |  P2 -> P1 : allow
+ * |'------'    \    |
+ * |             '   |
+ * |             P2  |
+ * '-----------------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, deny_with_nested_and_parent_domain) {
+	.domain_both = true,
+	.domain_parent = true,
+	.domain_child = false,
+};
+
+/*
+ *         Inherited + parent and child domain (siblings)
+ * .-----------------.
+ * | .------.        |  P1 -> P2 : deny
+ * | |  P1  .        |  P2 -> P1 : deny
+ * | '------'\       |
+ * |          \      |
+ * |        .--'---. |
+ * |        |  P2  | |
+ * |        '------' |
+ * '-----------------'
+ */
+FIXTURE_VARIANT_ADD(hierarchy, deny_with_forked_domain) {
+	.domain_both = true,
+	.domain_parent = true,
+	.domain_child = true,
+};
+
+FIXTURE_SETUP(hierarchy)
+{ }
+
+FIXTURE_TEARDOWN(hierarchy)
+{ }
+
+/* Test PTRACE_TRACEME and PTRACE_ATTACH for parent and child. */
+TEST_F(hierarchy, trace)
+{
+	pid_t child, parent;
+	int status, err_proc_read;
+	int pipe_child[2], pipe_parent[2];
+	char buf_parent;
+	long ret;
+
+	/*
+	 * Removes all effective and permitted capabilities to not interfere
+	 * with cap_ptrace_access_check() in case of PTRACE_MODE_FSCREDS.
+	 */
+	drop_caps(_metadata);
+
+	parent = getpid();
+	ASSERT_EQ(0, pipe2(pipe_child, O_CLOEXEC));
+	ASSERT_EQ(0, pipe2(pipe_parent, O_CLOEXEC));
+	if (variant->domain_both) {
+		create_domain(_metadata);
+		if (!_metadata->passed)
+			/* Aborts before forking. */
+			return;
+	}
+
+	child = fork();
+	ASSERT_LE(0, child);
+	if (child == 0) {
+		char buf_child;
+
+		ASSERT_EQ(0, close(pipe_parent[1]));
+		ASSERT_EQ(0, close(pipe_child[0]));
+		if (variant->domain_child)
+			create_domain(_metadata);
+
+		/* Waits for the parent to be in a domain, if any. */
+		ASSERT_EQ(1, read(pipe_parent[0], &buf_child, 1));
+
+		/* Tests PTRACE_ATTACH and PTRACE_MODE_READ on the parent. */
+		err_proc_read = test_ptrace_read(parent);
+		ret = ptrace(PTRACE_ATTACH, parent, NULL, 0);
+		if (variant->domain_child) {
+			EXPECT_EQ(-1, ret);
+			EXPECT_EQ(EPERM, errno);
+			EXPECT_EQ(EACCES, err_proc_read);
+		} else {
+			EXPECT_EQ(0, ret);
+			EXPECT_EQ(0, err_proc_read);
+		}
+		if (ret == 0) {
+			ASSERT_EQ(parent, waitpid(parent, &status, 0));
+			ASSERT_EQ(1, WIFSTOPPED(status));
+			ASSERT_EQ(0, ptrace(PTRACE_DETACH, parent, NULL, 0));
+		}
+
+		/* Tests child PTRACE_TRACEME. */
+		ret = ptrace(PTRACE_TRACEME);
+		if (variant->domain_parent) {
+			EXPECT_EQ(-1, ret);
+			EXPECT_EQ(EPERM, errno);
+		} else {
+			EXPECT_EQ(0, ret);
+		}
+
+		/*
+		 * Signals that the PTRACE_ATTACH test is done and the
+		 * PTRACE_TRACEME test is ongoing.
+		 */
+		ASSERT_EQ(1, write(pipe_child[1], ".", 1));
+
+		if (!variant->domain_parent) {
+			ASSERT_EQ(0, raise(SIGSTOP));
+		}
+
+		/* Waits for the parent PTRACE_ATTACH test. */
+		ASSERT_EQ(1, read(pipe_parent[0], &buf_child, 1));
+		_exit(_metadata->passed ? EXIT_SUCCESS : EXIT_FAILURE);
+		return;
+	}
+
+	ASSERT_EQ(0, close(pipe_child[1]));
+	ASSERT_EQ(0, close(pipe_parent[0]));
+	if (variant->domain_parent)
+		create_domain(_metadata);
+
+	/* Signals that the parent is in a domain, if any. */
+	ASSERT_EQ(1, write(pipe_parent[1], ".", 1));
+
+	/*
+	 * Waits for the child to test PTRACE_ATTACH on the parent and start
+	 * testing PTRACE_TRACEME.
+	 */
+	ASSERT_EQ(1, read(pipe_child[0], &buf_parent, 1));
+
+	/* Tests child PTRACE_TRACEME. */
+	if (!variant->domain_parent) {
+		ASSERT_EQ(child, waitpid(child, &status, 0));
+		ASSERT_EQ(1, WIFSTOPPED(status));
+		ASSERT_EQ(0, ptrace(PTRACE_DETACH, child, NULL, 0));
+	} else {
+		/* The child should not be traced by the parent. */
+		EXPECT_EQ(-1, ptrace(PTRACE_DETACH, child, NULL, 0));
+		EXPECT_EQ(ESRCH, errno);
+	}
+
+	/* Tests PTRACE_ATTACH and PTRACE_MODE_READ on the child. */
+	err_proc_read = test_ptrace_read(child);
+	ret = ptrace(PTRACE_ATTACH, child, NULL, 0);
+	if (variant->domain_parent) {
+		EXPECT_EQ(-1, ret);
+		EXPECT_EQ(EPERM, errno);
+		EXPECT_EQ(EACCES, err_proc_read);
+	} else {
+		EXPECT_EQ(0, ret);
+		EXPECT_EQ(0, err_proc_read);
+	}
+	if (ret == 0) {
+		ASSERT_EQ(child, waitpid(child, &status, 0));
+		ASSERT_EQ(1, WIFSTOPPED(status));
+		ASSERT_EQ(0, ptrace(PTRACE_DETACH, child, NULL, 0));
+	}
+
+	/* Signals that the parent PTRACE_ATTACH test is done. */
+	ASSERT_EQ(1, write(pipe_parent[1], ".", 1));
+	ASSERT_EQ(child, waitpid(child, &status, 0));
+	if (WIFSIGNALED(status) || !WIFEXITED(status) ||
+			WEXITSTATUS(status) != EXIT_SUCCESS)
+		_metadata->passed = 0;
+}
+
+TEST_HARNESS_MAIN
diff --git a/tools/testing/selftests/landlock/true.c b/tools/testing/selftests/landlock/true.c
new file mode 100644
index 000000000000..3f9ccbf52783
--- /dev/null
+++ b/tools/testing/selftests/landlock/true.c
@@ -0,0 +1,5 @@
+// SPDX-License-Identifier: GPL-2.0
+int main(void)
+{
+	return 0;
+}
diff --git a/tools/testing/selftests/mincore/mincore_selftest.c b/tools/testing/selftests/mincore/mincore_selftest.c
index 5a1e85ff5d32..e54106643337 100644
--- a/tools/testing/selftests/mincore/mincore_selftest.c
+++ b/tools/testing/selftests/mincore/mincore_selftest.c
@@ -14,7 +14,6 @@
 #include <sys/mman.h>
 #include <string.h>
 #include <fcntl.h>
-#include <string.h>
 
 #include "../kselftest.h"
 #include "../kselftest_harness.h"
diff --git a/tools/testing/selftests/nci/.gitignore b/tools/testing/selftests/nci/.gitignore
new file mode 100644
index 000000000000..448eeb4590fc
--- /dev/null
+++ b/tools/testing/selftests/nci/.gitignore
@@ -0,0 +1 @@
+/nci_dev
diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.sh b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
index 9236609731b1..3c4cb72ed8a4 100755
--- a/tools/testing/selftests/net/mptcp/mptcp_connect.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.sh
@@ -274,7 +274,7 @@ check_mptcp_disabled()
 	ip netns exec ${disabled_ns} sysctl -q net.mptcp.enabled=0
 
 	local err=0
-	LANG=C ip netns exec ${disabled_ns} ./mptcp_connect -p 10000 -s MPTCP 127.0.0.1 < "$cin" 2>&1 | \
+	LC_ALL=C ip netns exec ${disabled_ns} ./mptcp_connect -p 10000 -s MPTCP 127.0.0.1 < "$cin" 2>&1 | \
 		grep -q "^socket: Protocol not available$" && err=1
 	ip netns delete ${disabled_ns}
 
diff --git a/tools/testing/selftests/perf_events/sigtrap_threads.c b/tools/testing/selftests/perf_events/sigtrap_threads.c
index 78ddf5e11625..8e83cf91513a 100644
--- a/tools/testing/selftests/perf_events/sigtrap_threads.c
+++ b/tools/testing/selftests/perf_events/sigtrap_threads.c
@@ -43,7 +43,7 @@ static struct {
 	siginfo_t first_siginfo;	/* First observed siginfo_t. */
 } ctx;
 
-/* Unique value to check si_perf is correctly set from perf_event_attr::sig_data. */
+/* Unique value to check si_perf_data is correctly set from perf_event_attr::sig_data. */
 #define TEST_SIG_DATA(addr) (~(unsigned long)(addr))
 
 static struct perf_event_attr make_event_attr(bool enabled, volatile void *addr)
@@ -164,8 +164,8 @@ TEST_F(sigtrap_threads, enable_event)
 	EXPECT_EQ(ctx.signal_count, NUM_THREADS);
 	EXPECT_EQ(ctx.tids_want_signal, 0);
 	EXPECT_EQ(ctx.first_siginfo.si_addr, &ctx.iterate_on);
-	EXPECT_EQ(ctx.first_siginfo.si_errno, PERF_TYPE_BREAKPOINT);
-	EXPECT_EQ(ctx.first_siginfo.si_perf, TEST_SIG_DATA(&ctx.iterate_on));
+	EXPECT_EQ(ctx.first_siginfo.si_perf_type, PERF_TYPE_BREAKPOINT);
+	EXPECT_EQ(ctx.first_siginfo.si_perf_data, TEST_SIG_DATA(&ctx.iterate_on));
 
 	/* Check enabled for parent. */
 	ctx.iterate_on = 0;
@@ -183,8 +183,8 @@ TEST_F(sigtrap_threads, modify_and_enable_event)
 	EXPECT_EQ(ctx.signal_count, NUM_THREADS);
 	EXPECT_EQ(ctx.tids_want_signal, 0);
 	EXPECT_EQ(ctx.first_siginfo.si_addr, &ctx.iterate_on);
-	EXPECT_EQ(ctx.first_siginfo.si_errno, PERF_TYPE_BREAKPOINT);
-	EXPECT_EQ(ctx.first_siginfo.si_perf, TEST_SIG_DATA(&ctx.iterate_on));
+	EXPECT_EQ(ctx.first_siginfo.si_perf_type, PERF_TYPE_BREAKPOINT);
+	EXPECT_EQ(ctx.first_siginfo.si_perf_data, TEST_SIG_DATA(&ctx.iterate_on));
 
 	/* Check enabled for parent. */
 	ctx.iterate_on = 0;
@@ -203,8 +203,8 @@ TEST_F(sigtrap_threads, signal_stress)
 	EXPECT_EQ(ctx.signal_count, NUM_THREADS * ctx.iterate_on);
 	EXPECT_EQ(ctx.tids_want_signal, 0);
 	EXPECT_EQ(ctx.first_siginfo.si_addr, &ctx.iterate_on);
-	EXPECT_EQ(ctx.first_siginfo.si_errno, PERF_TYPE_BREAKPOINT);
-	EXPECT_EQ(ctx.first_siginfo.si_perf, TEST_SIG_DATA(&ctx.iterate_on));
+	EXPECT_EQ(ctx.first_siginfo.si_perf_type, PERF_TYPE_BREAKPOINT);
+	EXPECT_EQ(ctx.first_siginfo.si_perf_data, TEST_SIG_DATA(&ctx.iterate_on));
 }
 
 TEST_HARNESS_MAIN
diff --git a/tools/testing/selftests/powerpc/mm/tlbie_test.c b/tools/testing/selftests/powerpc/mm/tlbie_test.c
index f85a0938ab25..48344a74b212 100644
--- a/tools/testing/selftests/powerpc/mm/tlbie_test.c
+++ b/tools/testing/selftests/powerpc/mm/tlbie_test.c
@@ -33,7 +33,6 @@
 #include <sched.h>
 #include <time.h>
 #include <stdarg.h>
-#include <sched.h>
 #include <pthread.h>
 #include <signal.h>
 #include <sys/prctl.h>
diff --git a/tools/testing/selftests/proc/Makefile b/tools/testing/selftests/proc/Makefile
index 8be8a03d2973..1054e40a499a 100644
--- a/tools/testing/selftests/proc/Makefile
+++ b/tools/testing/selftests/proc/Makefile
@@ -12,6 +12,7 @@ TEST_GEN_PROGS += proc-self-map-files-001
 TEST_GEN_PROGS += proc-self-map-files-002
 TEST_GEN_PROGS += proc-self-syscall
 TEST_GEN_PROGS += proc-self-wchan
+TEST_GEN_PROGS += proc-subset-pid
 TEST_GEN_PROGS += proc-uptime-001
 TEST_GEN_PROGS += proc-uptime-002
 TEST_GEN_PROGS += read
diff --git a/tools/testing/selftests/proc/proc-subset-pid.c b/tools/testing/selftests/proc/proc-subset-pid.c
new file mode 100644
index 000000000000..d1052bcab039
--- /dev/null
+++ b/tools/testing/selftests/proc/proc-subset-pid.c
@@ -0,0 +1,121 @@
+/*
+ * Copyright (c) 2021 Alexey Dobriyan <adobriyan@gmail.com>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+/*
+ * Test that "mount -t proc -o subset=pid" hides everything but pids,
+ * /proc/self and /proc/thread-self.
+ */
+#undef NDEBUG
+#include <assert.h>
+#include <errno.h>
+#include <sched.h>
+#include <stdbool.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mount.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <dirent.h>
+#include <unistd.h>
+#include <stdio.h>
+
+static inline bool streq(const char *a, const char *b)
+{
+	return strcmp(a, b) == 0;
+}
+
+static void make_private_proc(void)
+{
+	if (unshare(CLONE_NEWNS) == -1) {
+		if (errno == ENOSYS || errno == EPERM) {
+			exit(4);
+		}
+		exit(1);
+	}
+	if (mount(NULL, "/", NULL, MS_PRIVATE|MS_REC, NULL) == -1) {
+		exit(1);
+	}
+	if (mount(NULL, "/proc", "proc", 0, "subset=pid") == -1) {
+		exit(1);
+	}
+}
+
+static bool string_is_pid(const char *s)
+{
+	while (1) {
+		switch (*s++) {
+		case '0':case '1':case '2':case '3':case '4':
+		case '5':case '6':case '7':case '8':case '9':
+			continue;
+
+		case '\0':
+			return true;
+
+		default:
+			return false;
+		}
+	}
+}
+
+int main(void)
+{
+	make_private_proc();
+
+	DIR *d = opendir("/proc");
+	assert(d);
+
+	struct dirent *de;
+
+	bool dot = false;
+	bool dot_dot = false;
+	bool self = false;
+	bool thread_self = false;
+
+	while ((de = readdir(d))) {
+		if (streq(de->d_name, ".")) {
+			assert(!dot);
+			dot = true;
+			assert(de->d_type == DT_DIR);
+		} else if (streq(de->d_name, "..")) {
+			assert(!dot_dot);
+			dot_dot = true;
+			assert(de->d_type == DT_DIR);
+		} else if (streq(de->d_name, "self")) {
+			assert(!self);
+			self = true;
+			assert(de->d_type == DT_LNK);
+		} else if (streq(de->d_name, "thread-self")) {
+			assert(!thread_self);
+			thread_self = true;
+			assert(de->d_type == DT_LNK);
+		} else {
+			if (!string_is_pid(de->d_name)) {
+				fprintf(stderr, "d_name '%s'\n", de->d_name);
+				assert(0);
+			}
+			assert(de->d_type == DT_DIR);
+		}
+	}
+
+	char c;
+	int rv = readlink("/proc/cpuinfo", &c, 1);
+	assert(rv == -1 && errno == ENOENT);
+
+	int fd = open("/proc/cpuinfo", O_RDONLY);
+	assert(fd == -1 && errno == ENOENT);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/proc/read.c b/tools/testing/selftests/proc/read.c
index b3ef9e14d6cc..35ee78dff144 100644
--- a/tools/testing/selftests/proc/read.c
+++ b/tools/testing/selftests/proc/read.c
@@ -14,7 +14,7 @@
  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  */
 // Test
-// 1) read of every file in /proc
+// 1) read and lseek on every file in /proc
 // 2) readlink of every symlink in /proc
 // 3) recursively (1) + (2) for every directory in /proc
 // 4) write to /proc/*/clear_refs and /proc/*/task/*/clear_refs
@@ -45,6 +45,8 @@ static void f_reg(DIR *d, const char *filename)
 	fd = openat(dirfd(d), filename, O_RDONLY|O_NONBLOCK);
 	if (fd == -1)
 		return;
+	/* struct proc_ops::proc_lseek is mandatory if file is seekable. */
+	(void)lseek(fd, 0, SEEK_SET);
 	rv = read(fd, buf, sizeof(buf));
 	assert((0 <= rv && rv <= sizeof(buf)) || rv == -1);
 	close(fd);
diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c
index 98c3b647f54d..e3d5c77a8612 100644
--- a/tools/testing/selftests/seccomp/seccomp_bpf.c
+++ b/tools/testing/selftests/seccomp/seccomp_bpf.c
@@ -1753,16 +1753,25 @@ TEST_F(TRACE_poke, getpid_runs_normally)
 # define SYSCALL_RET_SET(_regs, _val)				\
 	do {							\
 		typeof(_val) _result = (_val);			\
-		/*						\
-		 * A syscall error is signaled by CR0 SO bit	\
-		 * and the code is stored as a positive value.	\
-		 */						\
-		if (_result < 0) {				\
-			SYSCALL_RET(_regs) = -_result;		\
-			(_regs).ccr |= 0x10000000;		\
-		} else {					\
+		if ((_regs.trap & 0xfff0) == 0x3000) {		\
+			/*					\
+			 * scv 0 system call uses -ve result	\
+			 * for error, so no need to adjust.	\
+			 */					\
 			SYSCALL_RET(_regs) = _result;		\
-			(_regs).ccr &= ~0x10000000;		\
+		} else {					\
+			/*					\
+			 * A syscall error is signaled by the	\
+			 * CR0 SO bit and the code is stored as	\
+			 * a positive value.			\
+			 */					\
+			if (_result < 0) {			\
+				SYSCALL_RET(_regs) = -_result;	\
+				(_regs).ccr |= 0x10000000;	\
+			} else {				\
+				SYSCALL_RET(_regs) = _result;	\
+				(_regs).ccr &= ~0x10000000;	\
+			}					\
 		}						\
 	} while (0)
 # define SYSCALL_RET_SET_ON_PTRACE_EXIT
diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json
index 1cda2e11b3ad..773c5027553d 100644
--- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json
+++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fq_pie.json
@@ -9,11 +9,11 @@
         "setup": [
             "$IP link add dev $DUMMY type dummy || /bin/true"
         ],
-        "cmdUnderTest": "$TC qdisc add dev $DUMMY root fq_pie flows 65536",
-        "expExitCode": "2",
+        "cmdUnderTest": "$TC qdisc add dev $DUMMY handle 1: root fq_pie flows 65536",
+        "expExitCode": "0",
         "verifyCmd": "$TC qdisc show dev $DUMMY",
-        "matchPattern": "qdisc",
-        "matchCount": "0",
+        "matchPattern": "qdisc fq_pie 1: root refcnt 2 limit 10240p flows 65536",
+        "matchCount": "1",
         "teardown": [
             "$IP link del dev $DUMMY"
         ]
diff --git a/tools/testing/selftests/vm/.gitignore b/tools/testing/selftests/vm/.gitignore
index 9a35c3f6a557..1f651e85ed60 100644
--- a/tools/testing/selftests/vm/.gitignore
+++ b/tools/testing/selftests/vm/.gitignore
@@ -22,3 +22,4 @@ map_fixed_noreplace
 write_to_hugetlbfs
 hmm-tests
 local_config.*
+split_huge_page_test
diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile
index 8b0cd421ebd3..73e1cc96d7c2 100644
--- a/tools/testing/selftests/vm/Makefile
+++ b/tools/testing/selftests/vm/Makefile
@@ -42,6 +42,7 @@ TEST_GEN_FILES += on-fault-limit
 TEST_GEN_FILES += thuge-gen
 TEST_GEN_FILES += transhuge-stress
 TEST_GEN_FILES += userfaultfd
+TEST_GEN_FILES += split_huge_page_test
 
 ifeq ($(MACHINE),x86_64)
 CAN_BUILD_I386 := $(shell ./../x86/check_cc.sh $(CC) ../x86/trivial_32bit_program.c -m32)
diff --git a/tools/testing/selftests/vm/gup_test.c b/tools/testing/selftests/vm/gup_test.c
index 6c6336dd3b7f..1e662d59c502 100644
--- a/tools/testing/selftests/vm/gup_test.c
+++ b/tools/testing/selftests/vm/gup_test.c
@@ -13,6 +13,7 @@
 
 /* Just the flags we need, copied from mm.h: */
 #define FOLL_WRITE	0x01	/* check pte is writable */
+#define FOLL_TOUCH	0x02	/* mark page accessed */
 
 static char *cmd_to_str(unsigned long cmd)
 {
@@ -37,13 +38,13 @@ int main(int argc, char **argv)
 {
 	struct gup_test gup = { 0 };
 	unsigned long size = 128 * MB;
-	int i, fd, filed, opt, nr_pages = 1, thp = -1, repeats = 1, write = 0;
+	int i, fd, filed, opt, nr_pages = 1, thp = -1, repeats = 1, write = 1;
 	unsigned long cmd = GUP_FAST_BENCHMARK;
-	int flags = MAP_PRIVATE;
+	int flags = MAP_PRIVATE, touch = 0;
 	char *file = "/dev/zero";
 	char *p;
 
-	while ((opt = getopt(argc, argv, "m:r:n:F:f:abctTLUuwSH")) != -1) {
+	while ((opt = getopt(argc, argv, "m:r:n:F:f:abctTLUuwWSHpz")) != -1) {
 		switch (opt) {
 		case 'a':
 			cmd = PIN_FAST_BENCHMARK;
@@ -65,9 +66,13 @@ int main(int argc, char **argv)
 			 */
 			gup.which_pages[0] = 1;
 			break;
+		case 'p':
+			/* works only with DUMP_USER_PAGES_TEST */
+			gup.test_flags |= GUP_TEST_FLAG_DUMP_PAGES_USE_PIN;
+			break;
 		case 'F':
 			/* strtol, so you can pass flags in hex form */
-			gup.flags = strtol(optarg, 0, 0);
+			gup.gup_flags = strtol(optarg, 0, 0);
 			break;
 		case 'm':
 			size = atoi(optarg) * MB;
@@ -93,6 +98,9 @@ int main(int argc, char **argv)
 		case 'w':
 			write = 1;
 			break;
+		case 'W':
+			write = 0;
+			break;
 		case 'f':
 			file = optarg;
 			break;
@@ -103,6 +111,10 @@ int main(int argc, char **argv)
 		case 'H':
 			flags |= (MAP_HUGETLB | MAP_ANONYMOUS);
 			break;
+		case 'z':
+			/* fault pages in gup, do not fault in userland */
+			touch = 1;
+			break;
 		default:
 			return -1;
 		}
@@ -140,7 +152,7 @@ int main(int argc, char **argv)
 
 	gup.nr_pages_per_call = nr_pages;
 	if (write)
-		gup.flags |= FOLL_WRITE;
+		gup.gup_flags |= FOLL_WRITE;
 
 	fd = open("/sys/kernel/debug/gup_test", O_RDWR);
 	if (fd == -1) {
@@ -160,8 +172,18 @@ int main(int argc, char **argv)
 	else if (thp == 0)
 		madvise(p, size, MADV_NOHUGEPAGE);
 
-	for (; (unsigned long)p < gup.addr + size; p += PAGE_SIZE)
-		p[0] = 0;
+	/*
+	 * FOLL_TOUCH, in gup_test, is used as an either/or case: either
+	 * fault pages in from the kernel via FOLL_TOUCH, or fault them
+	 * in here, from user space. This allows comparison of performance
+	 * between those two cases.
+	 */
+	if (touch) {
+		gup.gup_flags |= FOLL_TOUCH;
+	} else {
+		for (; (unsigned long)p < gup.addr + size; p += PAGE_SIZE)
+			p[0] = 0;
+	}
 
 	/* Only report timing information on the *_BENCHMARK commands: */
 	if ((cmd == PIN_FAST_BENCHMARK) || (cmd == GUP_FAST_BENCHMARK) ||
diff --git a/tools/testing/selftests/vm/mremap_dontunmap.c b/tools/testing/selftests/vm/mremap_dontunmap.c
index 3a7b5ef0b0c6..f01dc4a85b0b 100644
--- a/tools/testing/selftests/vm/mremap_dontunmap.c
+++ b/tools/testing/selftests/vm/mremap_dontunmap.c
@@ -127,6 +127,57 @@ static void mremap_dontunmap_simple()
 	       "unable to unmap source mapping");
 }
 
+// This test validates that MREMAP_DONTUNMAP on a shared mapping works as expected.
+static void mremap_dontunmap_simple_shmem()
+{
+	unsigned long num_pages = 5;
+
+	int mem_fd = memfd_create("memfd", MFD_CLOEXEC);
+	BUG_ON(mem_fd < 0, "memfd_create");
+
+	BUG_ON(ftruncate(mem_fd, num_pages * page_size) < 0,
+			"ftruncate");
+
+	void *source_mapping =
+	    mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE,
+		 MAP_FILE | MAP_SHARED, mem_fd, 0);
+	BUG_ON(source_mapping == MAP_FAILED, "mmap");
+
+	BUG_ON(close(mem_fd) < 0, "close");
+
+	memset(source_mapping, 'a', num_pages * page_size);
+
+	// Try to just move the whole mapping anywhere (not fixed).
+	void *dest_mapping =
+	    mremap(source_mapping, num_pages * page_size, num_pages * page_size,
+		   MREMAP_DONTUNMAP | MREMAP_MAYMOVE, NULL);
+	if (dest_mapping == MAP_FAILED && errno == EINVAL) {
+		// Old kernel which doesn't support MREMAP_DONTUNMAP on shmem.
+		BUG_ON(munmap(source_mapping, num_pages * page_size) == -1,
+			"unable to unmap source mapping");
+		return;
+	}
+
+	BUG_ON(dest_mapping == MAP_FAILED, "mremap");
+
+	// Validate that the pages have been moved, we know they were moved if
+	// the dest_mapping contains a's.
+	BUG_ON(check_region_contains_byte
+	       (dest_mapping, num_pages * page_size, 'a') != 0,
+	       "pages did not migrate");
+
+	// Because the region is backed by shmem, we will actually see the same
+	// memory at the source location still.
+	BUG_ON(check_region_contains_byte
+	       (source_mapping, num_pages * page_size, 'a') != 0,
+	       "source should have no ptes");
+
+	BUG_ON(munmap(dest_mapping, num_pages * page_size) == -1,
+	       "unable to unmap destination mapping");
+	BUG_ON(munmap(source_mapping, num_pages * page_size) == -1,
+	       "unable to unmap source mapping");
+}
+
 // This test validates MREMAP_DONTUNMAP will move page tables to a specific
 // destination using MREMAP_FIXED, also while validating that the source
 // remains intact.
@@ -300,6 +351,7 @@ int main(void)
 	BUG_ON(page_buffer == MAP_FAILED, "unable to mmap a page.");
 
 	mremap_dontunmap_simple();
+	mremap_dontunmap_simple_shmem();
 	mremap_dontunmap_simple_fixed();
 	mremap_dontunmap_partial_mapping();
 	mremap_dontunmap_partial_mapping_overwrite();
diff --git a/tools/testing/selftests/vm/split_huge_page_test.c b/tools/testing/selftests/vm/split_huge_page_test.c
new file mode 100644
index 000000000000..1af16d2c2a0a
--- /dev/null
+++ b/tools/testing/selftests/vm/split_huge_page_test.c
@@ -0,0 +1,390 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * A test of splitting PMD THPs and PTE-mapped THPs from a specified virtual
+ * address range in a process via <debugfs>/split_huge_pages interface.
+ */
+
+#define _GNU_SOURCE
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <unistd.h>
+#include <inttypes.h>
+#include <string.h>
+#include <fcntl.h>
+#include <sys/mman.h>
+#include <sys/mount.h>
+#include <malloc.h>
+#include <stdbool.h>
+
+uint64_t pagesize;
+unsigned int pageshift;
+uint64_t pmd_pagesize;
+
+#define PMD_SIZE_PATH "/sys/kernel/mm/transparent_hugepage/hpage_pmd_size"
+#define SPLIT_DEBUGFS "/sys/kernel/debug/split_huge_pages"
+#define SMAP_PATH "/proc/self/smaps"
+#define INPUT_MAX 80
+
+#define PID_FMT "%d,0x%lx,0x%lx"
+#define PATH_FMT "%s,0x%lx,0x%lx"
+
+#define PFN_MASK     ((1UL<<55)-1)
+#define KPF_THP      (1UL<<22)
+
+int is_backed_by_thp(char *vaddr, int pagemap_file, int kpageflags_file)
+{
+	uint64_t paddr;
+	uint64_t page_flags;
+
+	if (pagemap_file) {
+		pread(pagemap_file, &paddr, sizeof(paddr),
+			((long)vaddr >> pageshift) * sizeof(paddr));
+
+		if (kpageflags_file) {
+			pread(kpageflags_file, &page_flags, sizeof(page_flags),
+				(paddr & PFN_MASK) * sizeof(page_flags));
+
+			return !!(page_flags & KPF_THP);
+		}
+	}
+	return 0;
+}
+
+
+static uint64_t read_pmd_pagesize(void)
+{
+	int fd;
+	char buf[20];
+	ssize_t num_read;
+
+	fd = open(PMD_SIZE_PATH, O_RDONLY);
+	if (fd == -1) {
+		perror("Open hpage_pmd_size failed");
+		exit(EXIT_FAILURE);
+	}
+	num_read = read(fd, buf, 19);
+	if (num_read < 1) {
+		close(fd);
+		perror("Read hpage_pmd_size failed");
+		exit(EXIT_FAILURE);
+	}
+	buf[num_read] = '\0';
+	close(fd);
+
+	return strtoul(buf, NULL, 10);
+}
+
+static int write_file(const char *path, const char *buf, size_t buflen)
+{
+	int fd;
+	ssize_t numwritten;
+
+	fd = open(path, O_WRONLY);
+	if (fd == -1)
+		return 0;
+
+	numwritten = write(fd, buf, buflen - 1);
+	close(fd);
+	if (numwritten < 1)
+		return 0;
+
+	return (unsigned int) numwritten;
+}
+
+static void write_debugfs(const char *fmt, ...)
+{
+	char input[INPUT_MAX];
+	int ret;
+	va_list argp;
+
+	va_start(argp, fmt);
+	ret = vsnprintf(input, INPUT_MAX, fmt, argp);
+	va_end(argp);
+
+	if (ret >= INPUT_MAX) {
+		printf("%s: Debugfs input is too long\n", __func__);
+		exit(EXIT_FAILURE);
+	}
+
+	if (!write_file(SPLIT_DEBUGFS, input, ret + 1)) {
+		perror(SPLIT_DEBUGFS);
+		exit(EXIT_FAILURE);
+	}
+}
+
+#define MAX_LINE_LENGTH 500
+
+static bool check_for_pattern(FILE *fp, const char *pattern, char *buf)
+{
+	while (fgets(buf, MAX_LINE_LENGTH, fp) != NULL) {
+		if (!strncmp(buf, pattern, strlen(pattern)))
+			return true;
+	}
+	return false;
+}
+
+static uint64_t check_huge(void *addr)
+{
+	uint64_t thp = 0;
+	int ret;
+	FILE *fp;
+	char buffer[MAX_LINE_LENGTH];
+	char addr_pattern[MAX_LINE_LENGTH];
+
+	ret = snprintf(addr_pattern, MAX_LINE_LENGTH, "%08lx-",
+		       (unsigned long) addr);
+	if (ret >= MAX_LINE_LENGTH) {
+		printf("%s: Pattern is too long\n", __func__);
+		exit(EXIT_FAILURE);
+	}
+
+
+	fp = fopen(SMAP_PATH, "r");
+	if (!fp) {
+		printf("%s: Failed to open file %s\n", __func__, SMAP_PATH);
+		exit(EXIT_FAILURE);
+	}
+	if (!check_for_pattern(fp, addr_pattern, buffer))
+		goto err_out;
+
+	/*
+	 * Fetch the AnonHugePages: in the same block and check the number of
+	 * hugepages.
+	 */
+	if (!check_for_pattern(fp, "AnonHugePages:", buffer))
+		goto err_out;
+
+	if (sscanf(buffer, "AnonHugePages:%10ld kB", &thp) != 1) {
+		printf("Reading smap error\n");
+		exit(EXIT_FAILURE);
+	}
+
+err_out:
+	fclose(fp);
+	return thp;
+}
+
+void split_pmd_thp(void)
+{
+	char *one_page;
+	size_t len = 4 * pmd_pagesize;
+	uint64_t thp_size;
+	size_t i;
+
+	one_page = memalign(pmd_pagesize, len);
+
+	if (!one_page) {
+		printf("Fail to allocate memory\n");
+		exit(EXIT_FAILURE);
+	}
+
+	madvise(one_page, len, MADV_HUGEPAGE);
+
+	for (i = 0; i < len; i++)
+		one_page[i] = (char)i;
+
+	thp_size = check_huge(one_page);
+	if (!thp_size) {
+		printf("No THP is allocated\n");
+		exit(EXIT_FAILURE);
+	}
+
+	/* split all THPs */
+	write_debugfs(PID_FMT, getpid(), (uint64_t)one_page,
+		(uint64_t)one_page + len);
+
+	for (i = 0; i < len; i++)
+		if (one_page[i] != (char)i) {
+			printf("%ld byte corrupted\n", i);
+			exit(EXIT_FAILURE);
+		}
+
+
+	thp_size = check_huge(one_page);
+	if (thp_size) {
+		printf("Still %ld kB AnonHugePages not split\n", thp_size);
+		exit(EXIT_FAILURE);
+	}
+
+	printf("Split huge pages successful\n");
+	free(one_page);
+}
+
+void split_pte_mapped_thp(void)
+{
+	char *one_page, *pte_mapped, *pte_mapped2;
+	size_t len = 4 * pmd_pagesize;
+	uint64_t thp_size;
+	size_t i;
+	const char *pagemap_template = "/proc/%d/pagemap";
+	const char *kpageflags_proc = "/proc/kpageflags";
+	char pagemap_proc[255];
+	int pagemap_fd;
+	int kpageflags_fd;
+
+	if (snprintf(pagemap_proc, 255, pagemap_template, getpid()) < 0) {
+		perror("get pagemap proc error");
+		exit(EXIT_FAILURE);
+	}
+	pagemap_fd = open(pagemap_proc, O_RDONLY);
+
+	if (pagemap_fd == -1) {
+		perror("read pagemap:");
+		exit(EXIT_FAILURE);
+	}
+
+	kpageflags_fd = open(kpageflags_proc, O_RDONLY);
+
+	if (kpageflags_fd == -1) {
+		perror("read kpageflags:");
+		exit(EXIT_FAILURE);
+	}
+
+	one_page = mmap((void *)(1UL << 30), len, PROT_READ | PROT_WRITE,
+			MAP_ANONYMOUS | MAP_PRIVATE, -1, 0);
+
+	madvise(one_page, len, MADV_HUGEPAGE);
+
+	for (i = 0; i < len; i++)
+		one_page[i] = (char)i;
+
+	thp_size = check_huge(one_page);
+	if (!thp_size) {
+		printf("No THP is allocated\n");
+		exit(EXIT_FAILURE);
+	}
+
+	/* remap the first pagesize of first THP */
+	pte_mapped = mremap(one_page, pagesize, pagesize, MREMAP_MAYMOVE);
+
+	/* remap the Nth pagesize of Nth THP */
+	for (i = 1; i < 4; i++) {
+		pte_mapped2 = mremap(one_page + pmd_pagesize * i + pagesize * i,
+				     pagesize, pagesize,
+				     MREMAP_MAYMOVE|MREMAP_FIXED,
+				     pte_mapped + pagesize * i);
+		if (pte_mapped2 == (char *)-1) {
+			perror("mremap failed");
+			exit(EXIT_FAILURE);
+		}
+	}
+
+	/* smap does not show THPs after mremap, use kpageflags instead */
+	thp_size = 0;
+	for (i = 0; i < pagesize * 4; i++)
+		if (i % pagesize == 0 &&
+		    is_backed_by_thp(&pte_mapped[i], pagemap_fd, kpageflags_fd))
+			thp_size++;
+
+	if (thp_size != 4) {
+		printf("Some THPs are missing during mremap\n");
+		exit(EXIT_FAILURE);
+	}
+
+	/* split all remapped THPs */
+	write_debugfs(PID_FMT, getpid(), (uint64_t)pte_mapped,
+		      (uint64_t)pte_mapped + pagesize * 4);
+
+	/* smap does not show THPs after mremap, use kpageflags instead */
+	thp_size = 0;
+	for (i = 0; i < pagesize * 4; i++) {
+		if (pte_mapped[i] != (char)i) {
+			printf("%ld byte corrupted\n", i);
+			exit(EXIT_FAILURE);
+		}
+		if (i % pagesize == 0 &&
+		    is_backed_by_thp(&pte_mapped[i], pagemap_fd, kpageflags_fd))
+			thp_size++;
+	}
+
+	if (thp_size) {
+		printf("Still %ld THPs not split\n", thp_size);
+		exit(EXIT_FAILURE);
+	}
+
+	printf("Split PTE-mapped huge pages successful\n");
+	munmap(one_page, len);
+	close(pagemap_fd);
+	close(kpageflags_fd);
+}
+
+void split_file_backed_thp(void)
+{
+	int status;
+	int fd;
+	ssize_t num_written;
+	char tmpfs_template[] = "/tmp/thp_split_XXXXXX";
+	const char *tmpfs_loc = mkdtemp(tmpfs_template);
+	char testfile[INPUT_MAX];
+	uint64_t pgoff_start = 0, pgoff_end = 1024;
+
+	printf("Please enable pr_debug in split_huge_pages_in_file() if you need more info.\n");
+
+	status = mount("tmpfs", tmpfs_loc, "tmpfs", 0, "huge=always,size=4m");
+
+	if (status) {
+		printf("Unable to create a tmpfs for testing\n");
+		exit(EXIT_FAILURE);
+	}
+
+	status = snprintf(testfile, INPUT_MAX, "%s/thp_file", tmpfs_loc);
+	if (status >= INPUT_MAX) {
+		printf("Fail to create file-backed THP split testing file\n");
+		goto cleanup;
+	}
+
+	fd = open(testfile, O_CREAT|O_WRONLY);
+	if (fd == -1) {
+		perror("Cannot open testing file\n");
+		goto cleanup;
+	}
+
+	/* write something to the file, so a file-backed THP can be allocated */
+	num_written = write(fd, tmpfs_loc, sizeof(tmpfs_loc));
+	close(fd);
+
+	if (num_written < 1) {
+		printf("Fail to write data to testing file\n");
+		goto cleanup;
+	}
+
+	/* split the file-backed THP */
+	write_debugfs(PATH_FMT, testfile, pgoff_start, pgoff_end);
+
+	status = unlink(testfile);
+	if (status)
+		perror("Cannot remove testing file\n");
+
+cleanup:
+	status = umount(tmpfs_loc);
+	if (status) {
+		printf("Unable to umount %s\n", tmpfs_loc);
+		exit(EXIT_FAILURE);
+	}
+	status = rmdir(tmpfs_loc);
+	if (status) {
+		perror("cannot remove tmp dir");
+		exit(EXIT_FAILURE);
+	}
+
+	printf("file-backed THP split test done, please check dmesg for more information\n");
+}
+
+int main(int argc, char **argv)
+{
+	if (geteuid() != 0) {
+		printf("Please run the benchmark as root\n");
+		exit(EXIT_FAILURE);
+	}
+
+	pagesize = getpagesize();
+	pageshift = ffs(pagesize) - 1;
+	pmd_pagesize = read_pmd_pagesize();
+
+	split_pmd_thp();
+	split_pte_mapped_thp();
+	split_file_backed_thp();
+
+	return 0;
+}
diff --git a/tools/testing/selftests/vm/test_vmalloc.sh b/tools/testing/selftests/vm/test_vmalloc.sh
index 06d2bb109f06..d73b846736f1 100755
--- a/tools/testing/selftests/vm/test_vmalloc.sh
+++ b/tools/testing/selftests/vm/test_vmalloc.sh
@@ -11,6 +11,7 @@
 
 TEST_NAME="vmalloc"
 DRIVER="test_${TEST_NAME}"
+NUM_CPUS=`grep -c ^processor /proc/cpuinfo`
 
 # 1 if fails
 exitcode=1
@@ -22,9 +23,9 @@ ksft_skip=4
 # Static templates for performance, stressing and smoke tests.
 # Also it is possible to pass any supported parameters manualy.
 #
-PERF_PARAM="single_cpu_test=1 sequential_test_order=1 test_repeat_count=3"
-SMOKE_PARAM="single_cpu_test=1 test_loop_count=10000 test_repeat_count=10"
-STRESS_PARAM="test_repeat_count=20"
+PERF_PARAM="sequential_test_order=1 test_repeat_count=3"
+SMOKE_PARAM="test_loop_count=10000 test_repeat_count=10"
+STRESS_PARAM="nr_threads=$NUM_CPUS test_repeat_count=20"
 
 check_test_requirements()
 {
@@ -58,8 +59,8 @@ run_perfformance_check()
 
 run_stability_check()
 {
-	echo "Run stability tests. In order to stress vmalloc subsystem we run"
-	echo "all available test cases on all available CPUs simultaneously."
+	echo "Run stability tests. In order to stress vmalloc subsystem all"
+	echo "available test cases are run by NUM_CPUS workers simultaneously."
 	echo "It will take time, so be patient."
 
 	modprobe $DRIVER $STRESS_PARAM > /dev/null 2>&1
@@ -92,17 +93,17 @@ usage()
 	echo "# Shows help message"
 	echo "./${DRIVER}.sh"
 	echo
-	echo "# Runs 1 test(id_1), repeats it 5 times on all online CPUs"
-	echo "./${DRIVER}.sh run_test_mask=1 test_repeat_count=5"
+	echo "# Runs 1 test(id_1), repeats it 5 times by NUM_CPUS workers"
+	echo "./${DRIVER}.sh nr_threads=$NUM_CPUS run_test_mask=1 test_repeat_count=5"
 	echo
 	echo -n "# Runs 4 tests(id_1|id_2|id_4|id_16) on one CPU with "
 	echo "sequential order"
-	echo -n "./${DRIVER}.sh single_cpu_test=1 sequential_test_order=1 "
+	echo -n "./${DRIVER}.sh sequential_test_order=1 "
 	echo "run_test_mask=23"
 	echo
-	echo -n "# Runs all tests on all online CPUs, shuffled order, repeats "
+	echo -n "# Runs all tests by NUM_CPUS workers, shuffled order, repeats "
 	echo "20 times"
-	echo "./${DRIVER}.sh test_repeat_count=20"
+	echo "./${DRIVER}.sh nr_threads=$NUM_CPUS test_repeat_count=20"
 	echo
 	echo "# Performance analysis"
 	echo "./${DRIVER}.sh performance"
diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c
index 92b8ec423201..f5ab5e0312e7 100644
--- a/tools/testing/selftests/vm/userfaultfd.c
+++ b/tools/testing/selftests/vm/userfaultfd.c
@@ -81,6 +81,8 @@ static volatile bool test_uffdio_copy_eexist = true;
 static volatile bool test_uffdio_zeropage_eexist = true;
 /* Whether to test uffd write-protection */
 static bool test_uffdio_wp = false;
+/* Whether to test uffd minor faults */
+static bool test_uffdio_minor = false;
 
 static bool map_shared;
 static int huge_fd;
@@ -96,6 +98,7 @@ struct uffd_stats {
 	int cpu;
 	unsigned long missing_faults;
 	unsigned long wp_faults;
+	unsigned long minor_faults;
 };
 
 /* pthread_mutex_t starts at page offset 0 */
@@ -153,17 +156,19 @@ static void uffd_stats_reset(struct uffd_stats *uffd_stats,
 		uffd_stats[i].cpu = i;
 		uffd_stats[i].missing_faults = 0;
 		uffd_stats[i].wp_faults = 0;
+		uffd_stats[i].minor_faults = 0;
 	}
 }
 
 static void uffd_stats_report(struct uffd_stats *stats, int n_cpus)
 {
 	int i;
-	unsigned long long miss_total = 0, wp_total = 0;
+	unsigned long long miss_total = 0, wp_total = 0, minor_total = 0;
 
 	for (i = 0; i < n_cpus; i++) {
 		miss_total += stats[i].missing_faults;
 		wp_total += stats[i].wp_faults;
+		minor_total += stats[i].minor_faults;
 	}
 
 	printf("userfaults: %llu missing (", miss_total);
@@ -172,6 +177,9 @@ static void uffd_stats_report(struct uffd_stats *stats, int n_cpus)
 	printf("\b), %llu wp (", wp_total);
 	for (i = 0; i < n_cpus; i++)
 		printf("%lu+", stats[i].wp_faults);
+	printf("\b), %llu minor (", minor_total);
+	for (i = 0; i < n_cpus; i++)
+		printf("%lu+", stats[i].minor_faults);
 	printf("\b)\n");
 }
 
@@ -328,7 +336,7 @@ static struct uffd_test_ops shmem_uffd_test_ops = {
 };
 
 static struct uffd_test_ops hugetlb_uffd_test_ops = {
-	.expected_ioctls = UFFD_API_RANGE_IOCTLS_BASIC,
+	.expected_ioctls = UFFD_API_RANGE_IOCTLS_BASIC & ~(1 << _UFFDIO_CONTINUE),
 	.allocate_area	= hugetlb_allocate_area,
 	.release_pages	= hugetlb_release_pages,
 	.alias_mapping = hugetlb_alias_mapping,
@@ -362,6 +370,22 @@ static void wp_range(int ufd, __u64 start, __u64 len, bool wp)
 	}
 }
 
+static void continue_range(int ufd, __u64 start, __u64 len)
+{
+	struct uffdio_continue req;
+
+	req.range.start = start;
+	req.range.len = len;
+	req.mode = 0;
+
+	if (ioctl(ufd, UFFDIO_CONTINUE, &req)) {
+		fprintf(stderr,
+			"UFFDIO_CONTINUE failed for address 0x%" PRIx64 "\n",
+			(uint64_t)start);
+		exit(1);
+	}
+}
+
 static void *locking_thread(void *arg)
 {
 	unsigned long cpu = (unsigned long) arg;
@@ -569,8 +593,32 @@ static void uffd_handle_page_fault(struct uffd_msg *msg,
 	}
 
 	if (msg->arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_WP) {
+		/* Write protect page faults */
 		wp_range(uffd, msg->arg.pagefault.address, page_size, false);
 		stats->wp_faults++;
+	} else if (msg->arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_MINOR) {
+		uint8_t *area;
+		int b;
+
+		/*
+		 * Minor page faults
+		 *
+		 * To prove we can modify the original range for testing
+		 * purposes, we're going to bit flip this range before
+		 * continuing.
+		 *
+		 * Note that this requires all minor page fault tests operate on
+		 * area_dst (non-UFFD-registered) and area_dst_alias
+		 * (UFFD-registered).
+		 */
+
+		area = (uint8_t *)(area_dst +
+				   ((char *)msg->arg.pagefault.address -
+				    area_dst_alias));
+		for (b = 0; b < page_size; ++b)
+			area[b] = ~area[b];
+		continue_range(uffd, msg->arg.pagefault.address, page_size);
+		stats->minor_faults++;
 	} else {
 		/* Missing page faults */
 		if (bounces & BOUNCE_VERIFY &&
@@ -779,7 +827,7 @@ static int stress(struct uffd_stats *uffd_stats)
 	return 0;
 }
 
-static int userfaultfd_open(int features)
+static int userfaultfd_open_ext(uint64_t *features)
 {
 	struct uffdio_api uffdio_api;
 
@@ -792,7 +840,7 @@ static int userfaultfd_open(int features)
 	uffd_flags = fcntl(uffd, F_GETFD, NULL);
 
 	uffdio_api.api = UFFD_API;
-	uffdio_api.features = features;
+	uffdio_api.features = *features;
 	if (ioctl(uffd, UFFDIO_API, &uffdio_api)) {
 		fprintf(stderr, "UFFDIO_API failed.\nPlease make sure to "
 			"run with either root or ptrace capability.\n");
@@ -804,9 +852,15 @@ static int userfaultfd_open(int features)
 		return 1;
 	}
 
+	*features = uffdio_api.features;
 	return 0;
 }
 
+static int userfaultfd_open(uint64_t features)
+{
+	return userfaultfd_open_ext(&features);
+}
+
 sigjmp_buf jbuf, *sigbuf;
 
 static void sighndl(int sig, siginfo_t *siginfo, void *ptr)
@@ -1112,7 +1166,7 @@ static int userfaultfd_events_test(void)
 	}
 
 	if (!pid)
-		return faulting_process(0);
+		exit(faulting_process(0));
 
 	waitpid(pid, &err, 0);
 	if (err) {
@@ -1215,6 +1269,102 @@ static int userfaultfd_sig_test(void)
 	return userfaults != 0;
 }
 
+static int userfaultfd_minor_test(void)
+{
+	struct uffdio_register uffdio_register;
+	unsigned long expected_ioctls;
+	unsigned long p;
+	pthread_t uffd_mon;
+	uint8_t expected_byte;
+	void *expected_page;
+	char c;
+	struct uffd_stats stats = { 0 };
+	uint64_t features = UFFD_FEATURE_MINOR_HUGETLBFS;
+
+	if (!test_uffdio_minor)
+		return 0;
+
+	printf("testing minor faults: ");
+	fflush(stdout);
+
+	if (uffd_test_ops->release_pages(area_dst))
+		return 1;
+
+	if (userfaultfd_open_ext(&features))
+		return 1;
+	/* If kernel reports the feature isn't supported, skip the test. */
+	if (!(features & UFFD_FEATURE_MINOR_HUGETLBFS)) {
+		printf("skipping test due to lack of feature support\n");
+		fflush(stdout);
+		return 0;
+	}
+
+	uffdio_register.range.start = (unsigned long)area_dst_alias;
+	uffdio_register.range.len = nr_pages * page_size;
+	uffdio_register.mode = UFFDIO_REGISTER_MODE_MINOR;
+	if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) {
+		fprintf(stderr, "register failure\n");
+		exit(1);
+	}
+
+	expected_ioctls = uffd_test_ops->expected_ioctls;
+	expected_ioctls |= 1 << _UFFDIO_CONTINUE;
+	if ((uffdio_register.ioctls & expected_ioctls) != expected_ioctls) {
+		fprintf(stderr, "unexpected missing ioctl(s)\n");
+		exit(1);
+	}
+
+	/*
+	 * After registering with UFFD, populate the non-UFFD-registered side of
+	 * the shared mapping. This should *not* trigger any UFFD minor faults.
+	 */
+	for (p = 0; p < nr_pages; ++p) {
+		memset(area_dst + (p * page_size), p % ((uint8_t)-1),
+		       page_size);
+	}
+
+	if (pthread_create(&uffd_mon, &attr, uffd_poll_thread, &stats)) {
+		perror("uffd_poll_thread create");
+		exit(1);
+	}
+
+	/*
+	 * Read each of the pages back using the UFFD-registered mapping. We
+	 * expect that the first time we touch a page, it will result in a minor
+	 * fault. uffd_poll_thread will resolve the fault by bit-flipping the
+	 * page's contents, and then issuing a CONTINUE ioctl.
+	 */
+
+	if (posix_memalign(&expected_page, page_size, page_size)) {
+		fprintf(stderr, "out of memory\n");
+		return 1;
+	}
+
+	for (p = 0; p < nr_pages; ++p) {
+		expected_byte = ~((uint8_t)(p % ((uint8_t)-1)));
+		memset(expected_page, expected_byte, page_size);
+		if (my_bcmp(expected_page, area_dst_alias + (p * page_size),
+			    page_size)) {
+			fprintf(stderr,
+				"unexpected page contents after minor fault\n");
+			exit(1);
+		}
+	}
+
+	if (write(pipefd[1], &c, sizeof(c)) != sizeof(c)) {
+		perror("pipe write");
+		exit(1);
+	}
+	if (pthread_join(uffd_mon, NULL))
+		return 1;
+
+	close(uffd);
+
+	uffd_stats_report(&stats, 1);
+
+	return stats.missing_faults != 0 || stats.minor_faults != nr_pages;
+}
+
 static int userfaultfd_stress(void)
 {
 	void *area;
@@ -1413,7 +1563,7 @@ static int userfaultfd_stress(void)
 
 	close(uffd);
 	return userfaultfd_zeropage_test() || userfaultfd_sig_test()
-		|| userfaultfd_events_test();
+		|| userfaultfd_events_test() || userfaultfd_minor_test();
 }
 
 /*
@@ -1454,6 +1604,8 @@ static void set_test_type(const char *type)
 		map_shared = true;
 		test_type = TEST_HUGETLB;
 		uffd_test_ops = &hugetlb_uffd_test_ops;
+		/* Minor faults require shared hugetlb; only enable here. */
+		test_uffdio_minor = true;
 	} else if (!strcmp(type, "shmem")) {
 		map_shared = true;
 		test_type = TEST_SHMEM;