| Message ID | 20180528043803.4824-1-bhole_prashant_q7@lab.ntt.co.jp |
|---|---|
| Headers | show |
| Series | fix test_sockmap | expand |
On 05/27/2018 09:37 PM, Prashant Bhole wrote: > This series fixes error handling, timeout and data verification in > test_sockmap. Previously it was not able to detect failure/timeout in > RX/TX thread because error was not notified to the main thread. > > Also slightly improved test output by printing parameter values (cork, > apply, start, end) so that parameters for all tests are displayed. > > Prashant Bhole (5): > selftests/bpf: test_sockmap, check test failure > selftests/bpf: test_sockmap, join cgroup in selftest mode > selftests/bpf: test_sockmap, fix test timeout > selftests/bpf: test_sockmap, fix data verification > selftests/bpf: test_sockmap, print additional test options > > tools/testing/selftests/bpf/test_sockmap.c | 76 +++++++++++++++++----- > 1 file changed, 58 insertions(+), 18 deletions(-) > After first patch "check test failure" how do we handle the case where test is known to cause timeouts because we are specifically testing these cases. This is the 'cork' parameter we discussed in the last series. It looks like with this series the test may still throw an error? Thanks, John
On 5/30/2018 12:48 AM, John Fastabend wrote: > On 05/27/2018 09:37 PM, Prashant Bhole wrote: >> This series fixes error handling, timeout and data verification in >> test_sockmap. Previously it was not able to detect failure/timeout in >> RX/TX thread because error was not notified to the main thread. >> >> Also slightly improved test output by printing parameter values (cork, >> apply, start, end) so that parameters for all tests are displayed. >> >> Prashant Bhole (5): >> selftests/bpf: test_sockmap, check test failure >> selftests/bpf: test_sockmap, join cgroup in selftest mode >> selftests/bpf: test_sockmap, fix test timeout >> selftests/bpf: test_sockmap, fix data verification >> selftests/bpf: test_sockmap, print additional test options >> >> tools/testing/selftests/bpf/test_sockmap.c | 76 +++++++++++++++++----- >> 1 file changed, 58 insertions(+), 18 deletions(-) >> > > After first patch "check test failure" how do we handle the case > where test is known to cause timeouts because we are specifically testing > these cases. This is the 'cork' parameter we discussed in the last > series. It looks like with this series the test may still throw an > error? Sorry. In your comment in last series, did you mean to skip error checking only for all cork tests (for now)? -Prashant
On 05/29/2018 05:44 PM, Prashant Bhole wrote: > > > On 5/30/2018 12:48 AM, John Fastabend wrote: >> On 05/27/2018 09:37 PM, Prashant Bhole wrote: >>> This series fixes error handling, timeout and data verification in >>> test_sockmap. Previously it was not able to detect failure/timeout in >>> RX/TX thread because error was not notified to the main thread. >>> >>> Also slightly improved test output by printing parameter values (cork, >>> apply, start, end) so that parameters for all tests are displayed. >>> >>> Prashant Bhole (5): >>> selftests/bpf: test_sockmap, check test failure >>> selftests/bpf: test_sockmap, join cgroup in selftest mode >>> selftests/bpf: test_sockmap, fix test timeout >>> selftests/bpf: test_sockmap, fix data verification >>> selftests/bpf: test_sockmap, print additional test options >>> >>> tools/testing/selftests/bpf/test_sockmap.c | 76 +++++++++++++++++----- >>> 1 file changed, 58 insertions(+), 18 deletions(-) >>> >> >> After first patch "check test failure" how do we handle the case >> where test is known to cause timeouts because we are specifically testing >> these cases. This is the 'cork' parameter we discussed in the last >> series. It looks like with this series the test may still throw an >> error? > > Sorry. In your comment in last series, did you mean to skip error > checking only for all cork tests (for now)? > > -Prashant > Hi, After this is applied are any errors returned from test_sockmap? When I read the first patch it looked like timeouts from the cork tests may result in errors "FAILED" tests. If this is the case then yes we need skip error checking on all tests or just the corked tests. .John
On 5/30/2018 2:12 PM, John Fastabend wrote: > On 05/29/2018 05:44 PM, Prashant Bhole wrote: >> >> >> On 5/30/2018 12:48 AM, John Fastabend wrote: >>> On 05/27/2018 09:37 PM, Prashant Bhole wrote: >>>> This series fixes error handling, timeout and data verification in >>>> test_sockmap. Previously it was not able to detect failure/timeout in >>>> RX/TX thread because error was not notified to the main thread. >>>> >>>> Also slightly improved test output by printing parameter values (cork, >>>> apply, start, end) so that parameters for all tests are displayed. >>>> >>>> Prashant Bhole (5): >>>> selftests/bpf: test_sockmap, check test failure >>>> selftests/bpf: test_sockmap, join cgroup in selftest mode >>>> selftests/bpf: test_sockmap, fix test timeout >>>> selftests/bpf: test_sockmap, fix data verification >>>> selftests/bpf: test_sockmap, print additional test options >>>> >>>> tools/testing/selftests/bpf/test_sockmap.c | 76 +++++++++++++++++----- >>>> 1 file changed, 58 insertions(+), 18 deletions(-) >>>> >>> >>> After first patch "check test failure" how do we handle the case >>> where test is known to cause timeouts because we are specifically testing >>> these cases. This is the 'cork' parameter we discussed in the last >>> series. It looks like with this series the test may still throw an >>> error? >> >> Sorry. In your comment in last series, did you mean to skip error >> checking only for all cork tests (for now)? >> >> -Prashant >> > > Hi, After this is applied are any errors returned from test_sockmap? > When I read the first patch it looked like timeouts from the cork > tests may result in errors "FAILED" tests. If this is the case then > yes we need skip error checking on all tests or just the corked > tests. Yes errors returned after applying this series. I will skip error checking on just corked tests. -Prashant