test/py: drain console log at the end of any failed test

Tests may fail for a number of reasons, and in particular for reasons
other than a timeout waiting for U-Boot to print expected data. If the
last operation that a failed test performs is not waiting for U-Boot to
print something, then any trailing output from U-Boot during that test's
operation will not be logged as part of that test, but rather either
along with the next test, or even thrown away, potentiall hiding clues
re: the test failure reason.

Solve this by explicitly draining (and hence logging) the U-Boot output
in the case of failed tests.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
master
Stephen Warren 9 years ago committed by Simon Glass
parent 636f38d83a
commit c10eb9d39f
  1. 1
      test/py/conftest.py
  2. 38
      test/py/u_boot_console_base.py

@ -386,6 +386,7 @@ def pytest_runtest_protocol(item, nextitem):
skipped = report
if failed:
console.drain_console()
tests_failed.add(item.name)
elif skipped:
tests_skipped.add(item.name)

@ -14,6 +14,7 @@ import os
import pytest
import re
import sys
import u_boot_spawn
# Regexes for text we expect U-Boot to send to the console.
pattern_u_boot_spl_signon = re.compile('(U-Boot SPL \\d{4}\\.\\d{2}-[^\r\n]*)')
@ -213,6 +214,43 @@ class ConsoleBase(object):
self.run_command(chr(3), wait_for_echo=False, send_nl=False)
def drain_console(self):
'''Read from and log the U-Boot console for a short time.
U-Boot's console output is only logged when the test code actively
waits for U-Boot to emit specific data. There are cases where tests
can fail without doing this. For example, if a test asks U-Boot to
enable USB device mode, then polls until a host-side device node
exists. In such a case, it is useful to log U-Boot's console output
in case U-Boot printed clues as to why the host-side even did not
occur. This function will do that.
Args:
None.
Returns:
Nothing.
'''
# If we are already not connected to U-Boot, there's nothing to drain.
# This should only happen when a previous call to run_command() or
# wait_for() failed (and hence the output has already been logged), or
# the system is shutting down.
if not self.p:
return
orig_timeout = self.p.timeout
try:
# Drain the log for a relatively short time.
self.p.timeout = 1000
# Wait for something U-Boot will likely never send. This will
# cause the console output to be read and logged.
self.p.expect(['This should never match U-Boot output'])
except u_boot_spawn.Timeout:
pass
finally:
self.p.timeout = orig_timeout
def ensure_spawned(self):
'''Ensure a connection to a correctly running U-Boot instance.

Loading…
Cancel
Save