| Message ID | 20200626164737.21943-1-desnesn@linux.ibm.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 3337bf41e0dd70b4064cdf60acdfcdc2d050066c |
| Headers | show |
| Series | selftests/powerpc: Purge extra count_pmc() calls of ebb selftests | expand |
> On 26-Jun-2020, at 10:17 PM, Desnes A. Nunes do Rosario <desnesn@linux.ibm.com> wrote: > > An extra count on ebb_state.stats.pmc_count[PMC_INDEX(pmc)] is being per- > formed when count_pmc() is used to reset PMCs on a few selftests. This > extra pmc_count can occasionally invalidate results, such as the ones from > cycles_test shown hereafter. The ebb_check_count() failed with an above > the upper limit error due to the extra value on ebb_state.stats.pmc_count. > > Furthermore, this extra count is also indicated by extra PMC1 trace_log on > the output of the cycle test (as well as on pmc56_overflow_test): > > ========== > ... > [21]: counter = 8 > [22]: register SPRN_MMCR0 = 0x0000000080000080 > [23]: register SPRN_PMC1 = 0x0000000080000004 > [24]: counter = 9 > [25]: register SPRN_MMCR0 = 0x0000000080000080 > [26]: register SPRN_PMC1 = 0x0000000080000004 > [27]: counter = 10 > [28]: register SPRN_MMCR0 = 0x0000000080000080 > [29]: register SPRN_PMC1 = 0x0000000080000004 >>> [30]: register SPRN_PMC1 = 0x000000004000051e > PMC1 count (0x280000546) above upper limit 0x2800003e8 (+0x15e) > [FAIL] Test FAILED on line 52 > failure: cycles > ========== > > Signed-off-by: Desnes A. Nunes do Rosario <desnesn@linux.ibm.com> > --- I too have run into similar failure with cycles_test. I will add that the failure is inconsistent. I have run into this issue 1 out of 25 times. The failure always happen at first instance. Subsequent tries work correctly. With this patch applied the test completes successfully 25 out of 25 times. # ./cycles_test test: cycles ….. ….. [25]: register SPRN_MMCR0 = 0x0000000080000080 [26]: register SPRN_PMC1 = 0x0000000080000004 [27]: counter = 10 [28]: register SPRN_MMCR0 = 0x0000000080000080 [29]: register SPRN_PMC1 = 0x0000000080000004 [30]: register SPRN_PMC1 = 0x000000004000048f PMC1 count (0x2800004b7) above upper limit 0x2800003e8 (+0xcf) [FAIL] Test FAILED on line 52 failure: cycles With the patch # ./cycles_test test: cycles ….. ….. [25]: register SPRN_MMCR0 = 0x0000000080000080 [26]: register SPRN_PMC1 = 0x0000000080000004 [27]: counter = 10 [28]: register SPRN_MMCR0 = 0x0000000080000080 [29]: register SPRN_PMC1 = 0x0000000080000004 PMC1 count (0x280000028) is between 0x27ffffc18 and 0x2800003e8 delta +0x410/-0x3c0 success: cycles # FWIW Tested-by : Sachin Sant <sachinp@linux.vnet.ibm.com> Thanks -Sachin
On 7/8/20 7:38 AM, Sachin Sant wrote: > >> On 26-Jun-2020, at 10:17 PM, Desnes A. Nunes do Rosario <desnesn@linux.ibm.com> wrote: >> >> An extra count on ebb_state.stats.pmc_count[PMC_INDEX(pmc)] is being per- >> formed when count_pmc() is used to reset PMCs on a few selftests. This >> extra pmc_count can occasionally invalidate results, such as the ones from >> cycles_test shown hereafter. The ebb_check_count() failed with an above >> the upper limit error due to the extra value on ebb_state.stats.pmc_count. >> >> Furthermore, this extra count is also indicated by extra PMC1 trace_log on >> the output of the cycle test (as well as on pmc56_overflow_test): >> >> ========== >> ... >> [21]: counter = 8 >> [22]: register SPRN_MMCR0 = 0x0000000080000080 >> [23]: register SPRN_PMC1 = 0x0000000080000004 >> [24]: counter = 9 >> [25]: register SPRN_MMCR0 = 0x0000000080000080 >> [26]: register SPRN_PMC1 = 0x0000000080000004 >> [27]: counter = 10 >> [28]: register SPRN_MMCR0 = 0x0000000080000080 >> [29]: register SPRN_PMC1 = 0x0000000080000004 >>>> [30]: register SPRN_PMC1 = 0x000000004000051e >> PMC1 count (0x280000546) above upper limit 0x2800003e8 (+0x15e) >> [FAIL] Test FAILED on line 52 >> failure: cycles >> ========== >> >> Signed-off-by: Desnes A. Nunes do Rosario <desnesn@linux.ibm.com> >> --- > I too have run into similar failure with cycles_test. I will add that the failure > is inconsistent. I have run into this issue 1 out of 25 times. The failure always > happen at first instance. Subsequent tries work correctly. Indeed; on my tests I was running 100 times to validate. Thanks for the review Sachin > > With this patch applied the test completes successfully 25 out of 25 times. > > # ./cycles_test > test: cycles > ….. > ….. > [25]: register SPRN_MMCR0 = 0x0000000080000080 > [26]: register SPRN_PMC1 = 0x0000000080000004 > [27]: counter = 10 > [28]: register SPRN_MMCR0 = 0x0000000080000080 > [29]: register SPRN_PMC1 = 0x0000000080000004 > [30]: register SPRN_PMC1 = 0x000000004000048f > PMC1 count (0x2800004b7) above upper limit 0x2800003e8 (+0xcf) > [FAIL] Test FAILED on line 52 > failure: cycles > > With the patch > > # ./cycles_test > test: cycles > ….. > ….. > [25]: register SPRN_MMCR0 = 0x0000000080000080 > [26]: register SPRN_PMC1 = 0x0000000080000004 > [27]: counter = 10 > [28]: register SPRN_MMCR0 = 0x0000000080000080 > [29]: register SPRN_PMC1 = 0x0000000080000004 > PMC1 count (0x280000028) is between 0x27ffffc18 and 0x2800003e8 delta +0x410/-0x3c0 > success: cycles > # > > FWIW Tested-by : Sachin Sant <sachinp@linux.vnet.ibm.com> > > Thanks > -Sachin
On Fri, 26 Jun 2020 13:47:37 -0300, Desnes A. Nunes do Rosario wrote: > An extra count on ebb_state.stats.pmc_count[PMC_INDEX(pmc)] is being per- > formed when count_pmc() is used to reset PMCs on a few selftests. This > extra pmc_count can occasionally invalidate results, such as the ones from > cycles_test shown hereafter. The ebb_check_count() failed with an above > the upper limit error due to the extra value on ebb_state.stats.pmc_count. > > Furthermore, this extra count is also indicated by extra PMC1 trace_log on > the output of the cycle test (as well as on pmc56_overflow_test): > > [...] Applied to powerpc/next. [1/1] selftests/powerpc: Purge extra count_pmc() calls of ebb selftests https://git.kernel.org/powerpc/c/3337bf41e0dd70b4064cdf60acdfcdc2d050066c cheers
diff --git a/tools/testing/selftests/powerpc/pmu/ebb/back_to_back_ebbs_test.c b/tools/testing/selftests/powerpc/pmu/ebb/back_to_back_ebbs_test.c index a2d7b0e3dca9..a26ac122c759 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/back_to_back_ebbs_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/back_to_back_ebbs_test.c @@ -91,8 +91,6 @@ int back_to_back_ebbs(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); event_close(&event); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/cycles_test.c b/tools/testing/selftests/powerpc/pmu/ebb/cycles_test.c index bc893813483e..bb9f587fa76e 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/cycles_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/cycles_test.c @@ -42,8 +42,6 @@ int cycles(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); event_close(&event); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_freeze_test.c b/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_freeze_test.c index dcd351d20328..9ae795ce314e 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_freeze_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_freeze_test.c @@ -99,8 +99,6 @@ int cycles_with_freeze(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); printf("EBBs while frozen %d\n", ebbs_while_frozen); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_mmcr2_test.c b/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_mmcr2_test.c index 94c99c12c0f2..4b45a2e70f62 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_mmcr2_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/cycles_with_mmcr2_test.c @@ -71,8 +71,6 @@ int cycles_with_mmcr2(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); event_close(&event); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/ebb.c b/tools/testing/selftests/powerpc/pmu/ebb/ebb.c index dfbc5c3ad52d..21537d6eb6b7 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/ebb.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/ebb.c @@ -396,8 +396,6 @@ int ebb_child(union pipe read_pipe, union pipe write_pipe) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); event_close(&event); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/ebb_on_willing_child_test.c b/tools/testing/selftests/powerpc/pmu/ebb/ebb_on_willing_child_test.c index ca2f7d729155..b208bf6ad58d 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/ebb_on_willing_child_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/ebb_on_willing_child_test.c @@ -38,8 +38,6 @@ static int victim_child(union pipe read_pipe, union pipe write_pipe) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); FAIL_IF(ebb_state.stats.ebb_count == 0); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/lost_exception_test.c b/tools/testing/selftests/powerpc/pmu/ebb/lost_exception_test.c index ac3e6e182614..ba2681a12cc7 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/lost_exception_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/lost_exception_test.c @@ -75,7 +75,6 @@ static int test_body(void) ebb_freeze_pmcs(); ebb_global_disable(); - count_pmc(4, sample_period); mtspr(SPRN_PMC4, 0xdead); dump_summary_ebb_state(); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/multi_counter_test.c b/tools/testing/selftests/powerpc/pmu/ebb/multi_counter_test.c index b8242e9d97d2..791d37ba327b 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/multi_counter_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/multi_counter_test.c @@ -70,13 +70,6 @@ int multi_counter(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - count_pmc(2, sample_period); - count_pmc(3, sample_period); - count_pmc(4, sample_period); - count_pmc(5, sample_period); - count_pmc(6, sample_period); - dump_ebb_state(); for (i = 0; i < 6; i++) diff --git a/tools/testing/selftests/powerpc/pmu/ebb/multi_ebb_procs_test.c b/tools/testing/selftests/powerpc/pmu/ebb/multi_ebb_procs_test.c index a05c0e18ded6..9b0f70d59702 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/multi_ebb_procs_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/multi_ebb_procs_test.c @@ -61,8 +61,6 @@ static int cycles_child(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_summary_ebb_state(); event_close(&event); diff --git a/tools/testing/selftests/powerpc/pmu/ebb/pmae_handling_test.c b/tools/testing/selftests/powerpc/pmu/ebb/pmae_handling_test.c index 153ebc92234f..2904c741e04e 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/pmae_handling_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/pmae_handling_test.c @@ -82,8 +82,6 @@ static int test_body(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(1, sample_period); - dump_ebb_state(); if (mmcr0_mismatch) diff --git a/tools/testing/selftests/powerpc/pmu/ebb/pmc56_overflow_test.c b/tools/testing/selftests/powerpc/pmu/ebb/pmc56_overflow_test.c index eadad75ed7e6..b29f8ba22d1e 100644 --- a/tools/testing/selftests/powerpc/pmu/ebb/pmc56_overflow_test.c +++ b/tools/testing/selftests/powerpc/pmu/ebb/pmc56_overflow_test.c @@ -76,8 +76,6 @@ int pmc56_overflow(void) ebb_global_disable(); ebb_freeze_pmcs(); - count_pmc(2, sample_period); - dump_ebb_state(); printf("PMC5/6 overflow %d\n", pmc56_overflowed);