| Message ID | 1486649607-31248-1-git-send-email-thuth@redhat.com |
|---|---|
| State | New |
| Headers | show |
On 9 February 2017 at 14:13, Thomas Huth <thuth@redhat.com> wrote: > Peter Maydell recently ran into time-out problems with the > prom-env test on a rather slow ARM board. To tackle this issue, > we can speed up the test by running QEMU with "-nodefaults" here, > so that SLOF has less devices to scan during boot, and by using > the "nvramrc" environment variable instead of "boot-command", > since this variable is evaluated earlier in the boot process. > And to be really sure that we do not face such time out problems > again, let's also increase the time out value from 100s to 120s > instead. > > Signed-off-by: Thomas Huth <thuth@redhat.com> > --- > tests/prom-env-test.c | 8 +++++--- > 1 file changed, 5 insertions(+), 3 deletions(-) Seems to cut the elapsed realtime for the pseries board down from about 90s to 55s. Hopefully that plus the boosted timeout will help. Tested-by: Peter Maydell <peter.maydell@linaro.org> thanks -- PMM
On Thu, Feb 09, 2017 at 02:37:26PM +0000, Peter Maydell wrote: > On 9 February 2017 at 14:13, Thomas Huth <thuth@redhat.com> wrote: > > Peter Maydell recently ran into time-out problems with the > > prom-env test on a rather slow ARM board. To tackle this issue, > > we can speed up the test by running QEMU with "-nodefaults" here, > > so that SLOF has less devices to scan during boot, and by using > > the "nvramrc" environment variable instead of "boot-command", > > since this variable is evaluated earlier in the boot process. > > And to be really sure that we do not face such time out problems > > again, let's also increase the time out value from 100s to 120s > > instead. > > > > Signed-off-by: Thomas Huth <thuth@redhat.com> > > --- > > tests/prom-env-test.c | 8 +++++--- > > 1 file changed, 5 insertions(+), 3 deletions(-) > > Seems to cut the elapsed realtime for the pseries board > down from about 90s to 55s. Hopefully that plus the > boosted timeout will help. > > Tested-by: Peter Maydell <peter.maydell@linaro.org> Peter, will you merge this directly, or do you want me to take it through my tree?
On 9 February 2017 at 23:15, David Gibson <david@gibson.dropbear.id.au> wrote: > On Thu, Feb 09, 2017 at 02:37:26PM +0000, Peter Maydell wrote: >> On 9 February 2017 at 14:13, Thomas Huth <thuth@redhat.com> wrote: >> > Peter Maydell recently ran into time-out problems with the >> > prom-env test on a rather slow ARM board. To tackle this issue, >> > we can speed up the test by running QEMU with "-nodefaults" here, >> > so that SLOF has less devices to scan during boot, and by using >> > the "nvramrc" environment variable instead of "boot-command", >> > since this variable is evaluated earlier in the boot process. >> > And to be really sure that we do not face such time out problems >> > again, let's also increase the time out value from 100s to 120s >> > instead. >> > >> > Signed-off-by: Thomas Huth <thuth@redhat.com> >> > --- >> > tests/prom-env-test.c | 8 +++++--- >> > 1 file changed, 5 insertions(+), 3 deletions(-) >> >> Seems to cut the elapsed realtime for the pseries board >> down from about 90s to 55s. Hopefully that plus the >> boosted timeout will help. >> >> Tested-by: Peter Maydell <peter.maydell@linaro.org> > > Peter, will you merge this directly, or do you want me to take it > through my tree? I just did a full test with the intention of merging it into master, and I noticed it causes new warnings on the sparc prom tests: TEST: tests/prom-env-test... (pid=17171) /sparc/prom-env/SPARCbook: Warning: nic lance.0 has no peer OK /sparc/prom-env/Voyager: Warning: nic lance.0 has no peer OK /sparc/prom-env/SS-20: Warning: nic lance.0 has no peer OK PASS: tests/prom-env-test So it needs some tweaking, I think. thanks -- PMM
On 10.02.2017 13:48, Peter Maydell wrote: > On 9 February 2017 at 23:15, David Gibson <david@gibson.dropbear.id.au> wrote: >> On Thu, Feb 09, 2017 at 02:37:26PM +0000, Peter Maydell wrote: >>> On 9 February 2017 at 14:13, Thomas Huth <thuth@redhat.com> wrote: >>>> Peter Maydell recently ran into time-out problems with the >>>> prom-env test on a rather slow ARM board. To tackle this issue, >>>> we can speed up the test by running QEMU with "-nodefaults" here, >>>> so that SLOF has less devices to scan during boot, and by using >>>> the "nvramrc" environment variable instead of "boot-command", >>>> since this variable is evaluated earlier in the boot process. >>>> And to be really sure that we do not face such time out problems >>>> again, let's also increase the time out value from 100s to 120s >>>> instead. >>>> >>>> Signed-off-by: Thomas Huth <thuth@redhat.com> >>>> --- >>>> tests/prom-env-test.c | 8 +++++--- >>>> 1 file changed, 5 insertions(+), 3 deletions(-) >>> >>> Seems to cut the elapsed realtime for the pseries board >>> down from about 90s to 55s. Hopefully that plus the >>> boosted timeout will help. >>> >>> Tested-by: Peter Maydell <peter.maydell@linaro.org> >> >> Peter, will you merge this directly, or do you want me to take it >> through my tree? > > I just did a full test with the intention of merging it into > master, and I noticed it causes new warnings on the sparc > prom tests: > > TEST: tests/prom-env-test... (pid=17171) > /sparc/prom-env/SPARCbook: > Warning: nic > lance.0 has no peer > OK > /sparc/prom-env/Voyager: > Warning: nic > lance.0 has no peer > OK > /sparc/prom-env/SS-20: > Warning: nic > lance.0 has no peer > OK > PASS: tests/prom-env-test > > So it needs some tweaking, I think. D'oh, sorry, I should have checked that ... it's likely of the new "-nodefaults" flag ... I'll have a look. Thomas
diff --git a/tests/prom-env-test.c b/tests/prom-env-test.c index 0ba6f48..7d97ae4 100644 --- a/tests/prom-env-test.c +++ b/tests/prom-env-test.c @@ -30,8 +30,8 @@ static void check_guest_memory(void) uint32_t signature; int i; - /* Poll until code has run and modified memory. Wait at most 30 seconds */ - for (i = 0; i < 10000; ++i) { + /* Poll until code has run and modified memory. Wait at most 120 seconds */ + for (i = 0; i < 12000; ++i) { signature = readl(ADDRESS); if (signature == MAGIC) { break; @@ -46,7 +46,9 @@ static void test_machine(const void *machine) { char *args; - args = g_strdup_printf("-M %s,accel=tcg -prom-env 'boot-command=%x %x l!'", + args = g_strdup_printf("-M %s,accel=tcg -nodefaults " + "-prom-env 'use-nvramrc?=true' " + "-prom-env 'nvramrc=%x %x l!'", (const char *)machine, MAGIC, ADDRESS); qtest_start(args);
Peter Maydell recently ran into time-out problems with the prom-env test on a rather slow ARM board. To tackle this issue, we can speed up the test by running QEMU with "-nodefaults" here, so that SLOF has less devices to scan during boot, and by using the "nvramrc" environment variable instead of "boot-command", since this variable is evaluated earlier in the boot process. And to be really sure that we do not face such time out problems again, let's also increase the time out value from 100s to 120s instead. Signed-off-by: Thomas Huth <thuth@redhat.com> --- tests/prom-env-test.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)