| Message ID | 20210628201203.30974-1-dwagner@suse.de |
|---|---|
| Headers | show |
| Series | Fix for start timestamp | expand |
Hi Daniel, On Mon Jun 28 2021, Daniel Wagner wrote: > Move the get_timestamp() from rt_test_start() to rt_init(). The idea > of rt_test_start() was to get the start timestamp right before the > 'main loop' for the test starts. At least for cyclictest the > rt_test_start() was placed wrongly so that the first test cycle could > hit the pagefault when strftime() wrote into the tsbuf. > > We don't have an exact semantic description what start test timestamp > means, so the simplest thing to avoid any further problems with it, is > to take the timestamp right at the beginning when the program > starts. Most test programs have a very short setup phase anyway. > > Reported-by: Mike Galbraith <efault@gmx.de> > Signed-off-by: Daniel Wagner <dwagner@suse.de> This fixes the issue for me: Tested-by: Kurt Kanzenbach <kurt@linutronix.de> Thanks, Kurt
On Mon, 28 Jun 2021, Daniel Wagner wrote: > Move the get_timestamp() from rt_test_start() to rt_init(). The idea > of rt_test_start() was to get the start timestamp right before the > 'main loop' for the test starts. At least for cyclictest the > rt_test_start() was placed wrongly so that the first test cycle could > hit the pagefault when strftime() wrote into the tsbuf. > > We don't have an exact semantic description what start test timestamp > means, so the simplest thing to avoid any further problems with it, is > to take the timestamp right at the beginning when the program > starts. Most test programs have a very short setup phase anyway. > > Reported-by: Mike Galbraith <efault@gmx.de> > Signed-off-by: Daniel Wagner <dwagner@suse.de> > --- > src/lib/rt-utils.c | 27 ++++++++++++++------------- > 1 file changed, 14 insertions(+), 13 deletions(-) > > diff --git a/src/lib/rt-utils.c b/src/lib/rt-utils.c > index d264c348ad42..11f386098867 100644 > --- a/src/lib/rt-utils.c > +++ b/src/lib/rt-utils.c > @@ -490,6 +490,19 @@ void disable_trace_mark(void) > close_tracemark_fd(); > } > > +static void get_timestamp(char *tsbuf) > +{ > + struct timeval tv; > + struct tm *tm; > + time_t t; > + > + gettimeofday(&tv, NULL); > + t = tv.tv_sec; > + tm = localtime(&t); > + /* RFC 2822-compliant date format */ > + strftime(tsbuf, MAX_TS_SIZE, "%a, %d %b %Y %T %z", tm); > +} > + > void rt_init(int argc, char *argv[]) > { > int offset = 0; > @@ -514,24 +527,12 @@ void rt_init(int argc, char *argv[]) > > offset += len + 1; > } > -} > > -static void get_timestamp(char *tsbuf) > -{ > - struct timeval tv; > - struct tm *tm; > - time_t t; > - > - gettimeofday(&tv, NULL); > - t = tv.tv_sec; > - tm = localtime(&t); > - /* RFC 2822-compliant date format */ > - strftime(tsbuf, MAX_TS_SIZE, "%a, %d %b %Y %T %z", tm); > + get_timestamp(ts_start); > } > > void rt_test_start(void) > { > - get_timestamp(ts_start); > } > > void rt_write_json(const char *filename, int return_code, > -- > 2.32.0 > > Signed-off-by: John Kacur <jkacur@redhat.com>
On Tue, 29 Jun 2021, Kurt Kanzenbach wrote: > Hi Daniel, > > On Mon Jun 28 2021, Daniel Wagner wrote: > > Move the get_timestamp() from rt_test_start() to rt_init(). The idea > > of rt_test_start() was to get the start timestamp right before the > > 'main loop' for the test starts. At least for cyclictest the > > rt_test_start() was placed wrongly so that the first test cycle could > > hit the pagefault when strftime() wrote into the tsbuf. > > > > We don't have an exact semantic description what start test timestamp > > means, so the simplest thing to avoid any further problems with it, is > > to take the timestamp right at the beginning when the program > > starts. Most test programs have a very short setup phase anyway. > > > > Reported-by: Mike Galbraith <efault@gmx.de> > > Signed-off-by: Daniel Wagner <dwagner@suse.de> > > This fixes the issue for me: > > Tested-by: Kurt Kanzenbach <kurt@linutronix.de> > > Thanks, > Kurt > thanks, added