On 6/7/19 2:15 PM, Vladimir Oltean wrote:
On Mon, 3 Jun 2019 at 15:25, Ido Schimmel <ido...@idosch.org> wrote:
From: Shalom Toledo <shal...@mellanox.com>
Test the PTP Physical Hardware Clock functionality using the "phc_ctl" (a
part of "linuxptp").
The test contains three sub-tests:
* "settime" test
* "adjtime" test
* "adjfreq" test
"settime" test:
* set the PHC time to 0 seconds.
* wait for 120.5 seconds.
* check if PHC time equal to 120.XX seconds.
"adjtime" test:
* set the PHC time to 0 seconds.
* adjust the time by 10 seconds.
* check if PHC time equal to 10.XX seconds.
"adjfreq" test:
* adjust the PHC frequency to be 1% faster.
* set the PHC time to 0 seconds.
* wait for 100.5 seconds.
* check if PHC time equal to 101.XX seconds.
Usage:
$ ./phc.sh /dev/ptp<X>
It is possible to run a subset of the tests, for example:
* To run only the "settime" test:
$ TESTS="settime" ./phc.sh /dev/ptp<X>
Signed-off-by: Shalom Toledo <shal...@mellanox.com>
Reviewed-by: Petr Machata <pe...@mellanox.com>
Signed-off-by: Ido Schimmel <ido...@mellanox.com>
---
tools/testing/selftests/ptp/phc.sh | 166 +++++++++++++++++++++++++++++
1 file changed, 166 insertions(+)
create mode 100755 tools/testing/selftests/ptp/phc.sh
diff --git a/tools/testing/selftests/ptp/phc.sh
b/tools/testing/selftests/ptp/phc.sh
new file mode 100755
index 000000000000..ac6e5a6e1d3a
--- /dev/null
+++ b/tools/testing/selftests/ptp/phc.sh
@@ -0,0 +1,166 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+ALL_TESTS="
+ settime
+ adjtime
+ adjfreq
+"
+DEV=$1
+
+##############################################################################
+# Sanity checks
+
+if [[ "$(id -u)" -ne 0 ]]; then
+ echo "SKIP: need root privileges"
+ exit 0
+fi
+
+if [[ "$DEV" == "" ]]; then
+ echo "SKIP: PTP device not provided"
+ exit 0
+fi
+
+require_command()
+{
+ local cmd=$1; shift
+
+ if [[ ! -x "$(command -v "$cmd")" ]]; then
+ echo "SKIP: $cmd not installed"
+ exit 1
+ fi
+}
+
+phc_sanity()
+{
+ phc_ctl $DEV get &> /dev/null
+
+ if [ $? != 0 ]; then
+ echo "SKIP: unknown clock $DEV: No such device"
+ exit 1
+ fi
+}
+
+require_command phc_ctl
+phc_sanity
+
+##############################################################################
+# Helpers
+
+# Exit status to return at the end. Set in case one of the tests fails.
+EXIT_STATUS=0
+# Per-test return value. Clear at the beginning of each test.
+RET=0
+
+check_err()
+{
+ local err=$1
+
+ if [[ $RET -eq 0 && $err -ne 0 ]]; then
+ RET=$err
+ fi
+}
+
+log_test()
+{
+ local test_name=$1
+
+ if [[ $RET -ne 0 ]]; then
+ EXIT_STATUS=1
+ printf "TEST: %-60s [FAIL]\n" "$test_name"
+ return 1
+ fi
+
+ printf "TEST: %-60s [ OK ]\n" "$test_name"
+ return 0
+}
+
+tests_run()
+{
+ local current_test
+
+ for current_test in ${TESTS:-$ALL_TESTS}; do
+ $current_test
+ done
+}
+
+##############################################################################
+# Tests
+
+settime_do()
+{
+ local res
+
+ res=$(phc_ctl $DEV set 0 wait 120.5 get 2> /dev/null \
+ | awk '/clock time is/{print $5}' \
+ | awk -F. '{print $1}')
+
+ (( res == 120 ))
+}
+
+adjtime_do()
+{
+ local res
+
+ res=$(phc_ctl $DEV set 0 adj 10 get 2> /dev/null \
+ | awk '/clock time is/{print $5}' \
+ | awk -F. '{print $1}')
+
+ (( res == 10 ))
+}
+
+adjfreq_do()
+{
+ local res
+
+ # Set the clock to be 1% faster
+ res=$(phc_ctl $DEV freq 10000000 set 0 wait 100.5 get 2> /dev/null \
+ | awk '/clock time is/{print $5}' \
+ | awk -F. '{print $1}')
+
+ (( res == 101 ))
+}
+
+##############################################################################
+
+cleanup()
+{
+ phc_ctl $DEV freq 0.0 &> /dev/null
+ phc_ctl $DEV set &> /dev/null
+}
+
+settime()
+{
+ RET=0
+
+ settime_do
+ check_err $?
+ log_test "settime"
+ cleanup
+}
+
+adjtime()
+{
+ RET=0
+
+ adjtime_do
+ check_err $?
+ log_test "adjtime"
+ cleanup
+}
+
+adjfreq()
+{
+ RET=0
+
+ adjfreq_do
+ check_err $?
+ log_test "adjfreq"
+ cleanup
+}
+
+trap cleanup EXIT
+
+tests_run
+
+exit $EXIT_STATUS
--
2.20.1
Cool testing framework, thanks!
Some things to consider:
- Why the .5 in the wait commands?
- I suspect there's a huge margin of inaccuracy that the test is
missing by only looking at the 'seconds' portion of the PHC time after
the adjfreq operation (up to 10^9 - 1 ppb, in the worst case).
Tested-by: Vladimir Oltean <olte...@gmail.com>
Regards,
-Vladimir
Here, I was actually thinking about something like this:
check_with_tolerance()
{
local res=$1
local expected=$2
local ppb=$3
local expected_min=$(($expected - (($expected * $ppb) / 1000000000)))
local expected_max=$(($expected + (($expected * $ppb) / 1000000000)))
if [ $res -lt $expected_min ]; then
printf "%d is more than %d ppb lower than expected %d (%d)\n" \
$res $ppb $expected $expected_min
return 1
elif [ $res -gt $expected_max ]; then
printf "%d is more than %d ppb higher than expected %d (%d)\n" \
$res $ppb $expected $expected_max
return 1;
else
printf "%d is within the +/-%d ppb tolerance of %d (%d - %d)\n"
\
$res $ppb $expected $expected_min $expected_max
return 0;
fi
}
settime_do()
{
local res
res=$(phc_ctl $DEV set 0 wait 120 get 2> /dev/null \
| awk '/clock time is/{print $5}' \
| awk -F. '{print $1 * 1000000000 + $2}')
check_with_tolerance $res 120000000000 10000
}
adjtime_do()
{
local res
res=$(phc_ctl $DEV set 0 adj 10 get 2> /dev/null \
| awk '/clock time is/{print $5}' \
| awk -F. '{print $1 * 1000000000 + $2}')
check_with_tolerance $res 10000000000 10000
}
adjfreq_do()
{
local res
# Set the clock to be 1% faster
res=$(phc_ctl $DEV freq 10000000 set 0 wait 100 get 2> /dev/null \
| awk '/clock time is/{print $5}' \
| awk -F. '{print $1 * 1000000000 + $2}')
check_with_tolerance $res 101000000000 10000
}
With the above changes:
SJA1105 hardware clock operations (not the timecounter ones that I
submitted):
# ./phc.sh /dev/ptp1
119999611352 is within the +/-1000000 ppb tolerance of 120000000000
(119880000000 - 120120000000)
TEST: settime [ OK ]
10001018472 is within the +/-1000000 ppb tolerance of 10000000000
(9990000000 - 10010000000)
TEST: adjtime [ OK ]
100998300984 is within the +/-1000000 ppb tolerance of 101000000000
(100899000000 - 101101000000)
TEST: adjfreq [ OK ]
But at a lower tolerance of 10000 ppb:
[root@OpenIL:~]# ./phc.sh /dev/ptp1
119998277344 is more than 10000 ppb lower than expected 120000000000
(119998800000)
TEST: settime [FAIL]
10002033840 is more than 10000 ppb higher than expected 10000000000
(10000100000)
TEST: adjtime [FAIL]
100998295304 is more than 10000 ppb lower than expected 101000000000
(100998990000)
TEST: adjfreq [FAIL]
For reference, ptp_qoriq:
[root@OpenIL:~]# ./phc.sh /dev/ptp0
120000960470 is within the +/-10000 ppb tolerance of 120000000000
(119998800000 - 120001200000)
TEST: settime [ OK ]
10000699770 is more than 10000 ppb higher than expected 10000000000
(10000100000)
TEST: adjtime [FAIL]
101000211269 is within the +/-10000 ppb tolerance of 101000000000
(100998990000 - 101001010000)
TEST: adjfreq [ OK ]
Regards,
-Vladimir
