* [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library @ 2019-03-07 13:59 Lukasz Krakowiak 2019-03-07 13:59 ` [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak ` (3 more replies) 0 siblings, 4 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-03-07 13:59 UTC (permalink / raw) To: david.hunt; +Cc: dev, Lukasz Krakowiak This patch rename test_power_acpi_cpufreq.c to test_power_cpufreq.c, remove prefix _acpi from UT power functions/test names, add UT for power turbo feature. Lukasz Krakowiak (3): test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c test: remove prefix _acpi from UT power function/test names test: add UT for power turbo feature app/test/Makefile | 2 +- app/test/autotest_data.py | 4 +- app/test/meson.build | 4 +- ...er_acpi_cpufreq.c => test_power_cpufreq.c} | 84 +++++++++++++++++-- 4 files changed, 83 insertions(+), 11 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (86%) -- 2.17.2 ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c 2019-03-07 13:59 [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library Lukasz Krakowiak @ 2019-03-07 13:59 ` Lukasz Krakowiak 2019-03-27 15:05 ` Burakov, Anatoly 2019-03-07 13:59 ` [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak ` (2 subsequent siblings) 3 siblings, 1 reply; 34+ messages in thread From: Lukasz Krakowiak @ 2019-03-07 13:59 UTC (permalink / raw) To: david.hunt; +Cc: dev, Lukasz Krakowiak This patch rename file test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/Makefile | 2 +- app/test/meson.build | 2 +- app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} | 0 3 files changed, 2 insertions(+), 2 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (100%) diff --git a/app/test/Makefile b/app/test/Makefile index 89949c2bb..d1bcfd456 100644 --- a/app/test/Makefile +++ b/app/test/Makefile @@ -158,7 +158,7 @@ endif SRCS-$(CONFIG_RTE_LIBRTE_METER) += test_meter.c SRCS-$(CONFIG_RTE_LIBRTE_KNI) += test_kni.c -SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_acpi_cpufreq.c +SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_cpufreq.c SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power_kvm_vm.c SRCS-y += test_common.c diff --git a/app/test/meson.build b/app/test/meson.build index 05e5ddeb0..4ae3dffea 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -82,7 +82,7 @@ test_sources = files('commands.c', 'test_pmd_ring.c', 'test_pmd_ring_perf.c', 'test_power.c', - 'test_power_acpi_cpufreq.c', + 'test_power_cpufreq.c', 'test_power_kvm_vm.c', 'test_prefetch.c', 'test_reciprocal_division.c', diff --git a/app/test/test_power_acpi_cpufreq.c b/app/test/test_power_cpufreq.c similarity index 100% rename from app/test/test_power_acpi_cpufreq.c rename to app/test/test_power_cpufreq.c -- 2.17.2 ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c 2019-03-07 13:59 ` [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak @ 2019-03-27 15:05 ` Burakov, Anatoly 2019-03-27 15:05 ` Burakov, Anatoly 0 siblings, 1 reply; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:05 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > This patch rename file test_power_acpi_cpufreq.c -> > app/test/test_power_cpufreq.c. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- Subject line is too long. Perhaps just say "test/power: rename cpufreq autotest" ? Otherwise Acked-by: Anatoly Burakov <anatoly.burakov@intel.com> -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c 2019-03-27 15:05 ` Burakov, Anatoly @ 2019-03-27 15:05 ` Burakov, Anatoly 0 siblings, 0 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:05 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > This patch rename file test_power_acpi_cpufreq.c -> > app/test/test_power_cpufreq.c. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- Subject line is too long. Perhaps just say "test/power: rename cpufreq autotest" ? Otherwise Acked-by: Anatoly Burakov <anatoly.burakov@intel.com> -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names 2019-03-07 13:59 [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-03-07 13:59 ` [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak @ 2019-03-07 13:59 ` Lukasz Krakowiak 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-07 13:59 ` [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak 3 siblings, 1 reply; 34+ messages in thread From: Lukasz Krakowiak @ 2019-03-07 13:59 UTC (permalink / raw) To: david.hunt; +Cc: dev, Lukasz Krakowiak This patch remove prefix _acpi from power UT function/test names. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/autotest_data.py | 4 ++-- app/test/meson.build | 2 +- app/test/test_power_cpufreq.c | 12 ++++++------ 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/app/test/autotest_data.py b/app/test/autotest_data.py index 5f87bb94d..c80801b20 100644 --- a/app/test/autotest_data.py +++ b/app/test/autotest_data.py @@ -622,8 +622,8 @@ "Report": None, }, { - "Name": "Power ACPI cpufreq autotest", - "Command": "power_acpi_cpufreq_autotest", + "Name": "Power cpufreq autotest", + "Command": "power_cpufreq_autotest", "Func": default_autotest, "Report": None, }, diff --git a/app/test/meson.build b/app/test/meson.build index 4ae3dffea..a389a2b10 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -205,7 +205,7 @@ fast_non_parallel_test_names = [ 'member_autotest', 'metrics_autotest', 'pdump_autotest', - 'power_acpi_cpufreq_autotest', + 'power_cpufreq_autotest', 'power_autotest', 'power_kvm_vm_autotest', 'reorder_autotest', diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index 61b1da05a..d099f2f47 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -14,14 +14,14 @@ #ifndef RTE_LIBRTE_POWER static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; } static int -test_power_acpi_caps(void) +test_power_caps(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; @@ -367,7 +367,7 @@ check_power_freq_min(void) } static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { int ret = -1; enum power_management_env env; @@ -539,7 +539,7 @@ test_power_acpi_cpufreq(void) } static int -test_power_acpi_caps(void) +test_power_caps(void) { struct rte_power_core_capabilities caps; int ret; @@ -568,5 +568,5 @@ test_power_acpi_caps(void) #endif -REGISTER_TEST_COMMAND(power_acpi_cpufreq_autotest, test_power_acpi_cpufreq); -REGISTER_TEST_COMMAND(power_acpi_caps_autotest, test_power_acpi_caps); +REGISTER_TEST_COMMAND(power_cpufreq_autotest, test_power_cpufreq); +REGISTER_TEST_COMMAND(power_caps_autotest, test_power_caps); -- 2.17.2 ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names 2019-03-07 13:59 ` [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak @ 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-29 22:40 ` Thomas Monjalon 0 siblings, 2 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:07 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. Again, subject line is IMO too long. How about: "test/power: remove _acpi prefix from test names" ? Otherwise, Acked-by: Anatoly Burakov <anatoly.burakov@intel.com> > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names 2019-03-27 15:07 ` Burakov, Anatoly @ 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-29 22:40 ` Thomas Monjalon 1 sibling, 0 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:07 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. Again, subject line is IMO too long. How about: "test/power: remove _acpi prefix from test names" ? Otherwise, Acked-by: Anatoly Burakov <anatoly.burakov@intel.com> > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-27 15:07 ` Burakov, Anatoly @ 2019-03-29 22:40 ` Thomas Monjalon 2019-03-29 22:40 ` Thomas Monjalon 1 sibling, 1 reply; 34+ messages in thread From: Thomas Monjalon @ 2019-03-29 22:40 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: dev, Burakov, Anatoly, david.hunt 27/03/2019 16:07, Burakov, Anatoly: > On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > > This patch remove prefix _acpi from power UT function/test names. I think it can be merged with patch renaming the file. ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names 2019-03-29 22:40 ` Thomas Monjalon @ 2019-03-29 22:40 ` Thomas Monjalon 0 siblings, 0 replies; 34+ messages in thread From: Thomas Monjalon @ 2019-03-29 22:40 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: dev, Burakov, Anatoly, david.hunt 27/03/2019 16:07, Burakov, Anatoly: > On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > > This patch remove prefix _acpi from power UT function/test names. I think it can be merged with patch renaming the file. ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-03-07 13:59 [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-03-07 13:59 ` [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak 2019-03-07 13:59 ` [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak @ 2019-03-07 13:59 ` Lukasz Krakowiak 2019-03-27 15:13 ` Burakov, Anatoly 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak 3 siblings, 1 reply; 34+ messages in thread From: Lukasz Krakowiak @ 2019-03-07 13:59 UTC (permalink / raw) To: david.hunt; +Cc: dev, Lukasz Krakowiak Add UT check_power_turbo. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ 1 file changed, 72 insertions(+) diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index d099f2f47..c75c9bf1c 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -366,6 +366,59 @@ check_power_freq_min(void) return 0; } +/* Check rte_power_turbo() */ +static int +check_power_turbo(void) +{ + int ret; + + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { + printf("Turbo not available on lcore %u, skipping test\n", + TEST_POWER_LCORE_ID); + return 0; + } + + /* test with an invalid lcore id */ + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly enable turbo successfully " + "on lcore %u\n", TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to enable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); + if (ret < 0) + return -1; + + /* test with an invalid lcore id */ + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly disable turbo successfully " + "on lcore %u\n", TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to disable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); + if (ret < 0) + return -1; + + return 0; +} + static int test_power_cpufreq(void) { @@ -427,6 +480,21 @@ test_power_cpufreq(void) "been initialised\n"); goto fail_all; } + if (rte_power_turbo_status == NULL) { + printf("rte_power_turbo_status should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_enable_turbo == NULL) { + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_disable_turbo == NULL) { + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { @@ -502,6 +570,10 @@ test_power_cpufreq(void) if (ret < 0) goto fail_all; + ret = check_power_turbo(); + if (ret < 0) + goto fail_all; + ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { printf("Cannot exit power management for lcore %u\n", -- 2.17.2 ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-03-07 13:59 ` [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature Lukasz Krakowiak @ 2019-03-27 15:13 ` Burakov, Anatoly 2019-03-27 15:13 ` Burakov, Anatoly 2019-04-03 9:07 ` Krakowiak, LukaszX 0 siblings, 2 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:13 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > Add UT check_power_turbo. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- > app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ > 1 file changed, 72 insertions(+) > > diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c > index d099f2f47..c75c9bf1c 100644 > --- a/app/test/test_power_cpufreq.c > +++ b/app/test/test_power_cpufreq.c > @@ -366,6 +366,59 @@ check_power_freq_min(void) > return 0; > } > > +/* Check rte_power_turbo() */ > +static int > +check_power_turbo(void) > +{ > + int ret; > + > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > + printf("Turbo not available on lcore %u, skipping test\n", > + TEST_POWER_LCORE_ID); Misleading indentation, please add two tabs to the second line of printf() statement. > + return 0; > + } > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly enable turbo successfully " > + "on lcore %u\n", TEST_POWER_LCORE_INVALID); Please don't break up strings to multiple lines. > + return -1; > + } > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to enable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); I wish there was a middle ground between no indentation and way too much indentation... > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > + if (ret < 0) > + return -1; > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly disable turbo successfully " > + "on lcore %u\n", TEST_POWER_LCORE_INVALID); Same as above, don't break up strings. > + return -1; > + } > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to disable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); Same as above, two tabs is enough indentation. > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > + if (ret < 0) > + return -1; > + > + return 0; > +} > + > static int > test_power_cpufreq(void) > { > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > "been initialised\n"); > goto fail_all; > } > + if (rte_power_turbo_status == NULL) { > + printf("rte_power_turbo_status should not be NULL, environment has not " > + "been initialised\n"); Here and below: I don't think saying *why* it should not be NULL - just saying that it shouldn't be NULL is enough. Maybe mention why it's supposed to be not NULL in comments here. Also, i have a suspicion that the message is wrong - why would status be null if the environment was initialized? Presumably it's the other way around? > + goto fail_all; > + } > + if (rte_power_freq_enable_turbo == NULL) { > + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_disable_turbo == NULL) { > + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > if (ret < 0) > goto fail_all; > > + ret = check_power_turbo(); > + if (ret < 0) > + goto fail_all; > + > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > printf("Cannot exit power management for lcore %u\n", > -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-03-27 15:13 ` Burakov, Anatoly @ 2019-03-27 15:13 ` Burakov, Anatoly 2019-04-03 9:07 ` Krakowiak, LukaszX 1 sibling, 0 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-03-27 15:13 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: dev On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > Add UT check_power_turbo. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- > app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ > 1 file changed, 72 insertions(+) > > diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c > index d099f2f47..c75c9bf1c 100644 > --- a/app/test/test_power_cpufreq.c > +++ b/app/test/test_power_cpufreq.c > @@ -366,6 +366,59 @@ check_power_freq_min(void) > return 0; > } > > +/* Check rte_power_turbo() */ > +static int > +check_power_turbo(void) > +{ > + int ret; > + > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > + printf("Turbo not available on lcore %u, skipping test\n", > + TEST_POWER_LCORE_ID); Misleading indentation, please add two tabs to the second line of printf() statement. > + return 0; > + } > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly enable turbo successfully " > + "on lcore %u\n", TEST_POWER_LCORE_INVALID); Please don't break up strings to multiple lines. > + return -1; > + } > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to enable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); I wish there was a middle ground between no indentation and way too much indentation... > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > + if (ret < 0) > + return -1; > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly disable turbo successfully " > + "on lcore %u\n", TEST_POWER_LCORE_INVALID); Same as above, don't break up strings. > + return -1; > + } > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to disable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); Same as above, two tabs is enough indentation. > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > + if (ret < 0) > + return -1; > + > + return 0; > +} > + > static int > test_power_cpufreq(void) > { > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > "been initialised\n"); > goto fail_all; > } > + if (rte_power_turbo_status == NULL) { > + printf("rte_power_turbo_status should not be NULL, environment has not " > + "been initialised\n"); Here and below: I don't think saying *why* it should not be NULL - just saying that it shouldn't be NULL is enough. Maybe mention why it's supposed to be not NULL in comments here. Also, i have a suspicion that the message is wrong - why would status be null if the environment was initialized? Presumably it's the other way around? > + goto fail_all; > + } > + if (rte_power_freq_enable_turbo == NULL) { > + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_disable_turbo == NULL) { > + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > if (ret < 0) > goto fail_all; > > + ret = check_power_turbo(); > + if (ret < 0) > + goto fail_all; > + > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > printf("Cannot exit power management for lcore %u\n", > -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-03-27 15:13 ` Burakov, Anatoly 2019-03-27 15:13 ` Burakov, Anatoly @ 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 10:01 ` Burakov, Anatoly 1 sibling, 2 replies; 34+ messages in thread From: Krakowiak, LukaszX @ 2019-04-03 9:07 UTC (permalink / raw) To: Burakov, Anatoly, Hunt, David; +Cc: dev Hi, > On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > > Add UT check_power_turbo. > > > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > > --- > > app/test/test_power_cpufreq.c | 72 > +++++++++++++++++++++++++++++++++++ > > 1 file changed, 72 insertions(+) > > > > diff --git a/app/test/test_power_cpufreq.c > > b/app/test/test_power_cpufreq.c index d099f2f47..c75c9bf1c 100644 > > --- a/app/test/test_power_cpufreq.c > > +++ b/app/test/test_power_cpufreq.c > > @@ -366,6 +366,59 @@ check_power_freq_min(void) > > return 0; > > } > > > > +/* Check rte_power_turbo() */ > > +static int > > +check_power_turbo(void) > > +{ > > + int ret; > > + > > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > > + printf("Turbo not available on lcore %u, skipping test\n", > > + TEST_POWER_LCORE_ID); > > Misleading indentation, please add two tabs to the second line of > printf() statement. > > > + return 0; > > + } > > + > > + /* test with an invalid lcore id */ > > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > > + if (ret >= 0) { > > + printf("Unexpectedly enable turbo successfully " > > + "on lcore %u\n", > TEST_POWER_LCORE_INVALID); > > Please don't break up strings to multiple lines. > > > + return -1; > > + } > > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > > + if (ret < 0) { > > + printf("Fail to enable turbo on lcore %u\n", > > + > TEST_POWER_LCORE_ID); > > I wish there was a middle ground between no indentation and way too much > indentation... > > > + return -1; > > + } > > + > > + /* Check the current frequency */ > > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > > + if (ret < 0) > > + return -1; > > + > > + /* test with an invalid lcore id */ > > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > > + if (ret >= 0) { > > + printf("Unexpectedly disable turbo successfully " > > + "on lcore %u\n", > TEST_POWER_LCORE_INVALID); > > Same as above, don't break up strings. > > > + return -1; > > + } > > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > > + if (ret < 0) { > > + printf("Fail to disable turbo on lcore %u\n", > > + > TEST_POWER_LCORE_ID); > > Same as above, two tabs is enough indentation. > > > + return -1; > > + } > > + > > + /* Check the current frequency */ > > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > > + if (ret < 0) > > + return -1; > > + > > + return 0; > > +} > > + > > static int > > test_power_cpufreq(void) > > { > > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > > "been initialised\n"); > > goto fail_all; > > } > > + if (rte_power_turbo_status == NULL) { > > + printf("rte_power_turbo_status should not be NULL, > environment has not " > > + "been initialised\n"); > > Here and below: > > I don't think saying *why* it should not be NULL - just saying that it shouldn't be > NULL is enough. Maybe mention why it's supposed to be not NULL in comments > here. > > Also, i have a suspicion that the message is wrong - why would status be null if > the environment was initialized? Presumably it's the other way around? I think this is correctly: if env was initialized, pointer != NULL, otherwise env wasn't initialized. Rest of all, you are right: coding style issues. Thanks. > > > + goto fail_all; > > + } > > + if (rte_power_freq_enable_turbo == NULL) { > > + printf("rte_power_freq_enable_turbo should not be NULL, > environment has not " > > + "been initialised\n"); > > + goto fail_all; > > + } > > + if (rte_power_freq_disable_turbo == NULL) { > > + printf("rte_power_freq_disable_turbo should not be NULL, > environment has not " > > + "been initialised\n"); > > + goto fail_all; > > + } > > > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > > if (ret < 0) { > > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > > if (ret < 0) > > goto fail_all; > > > > + ret = check_power_turbo(); > > + if (ret < 0) > > + goto fail_all; > > + > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > > if (ret < 0) { > > printf("Cannot exit power management for lcore %u\n", > > > > > -- > Thanks, > Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-04-03 9:07 ` Krakowiak, LukaszX @ 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 10:01 ` Burakov, Anatoly 1 sibling, 0 replies; 34+ messages in thread From: Krakowiak, LukaszX @ 2019-04-03 9:07 UTC (permalink / raw) To: Burakov, Anatoly, Hunt, David; +Cc: dev Hi, > On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: > > Add UT check_power_turbo. > > > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > > --- > > app/test/test_power_cpufreq.c | 72 > +++++++++++++++++++++++++++++++++++ > > 1 file changed, 72 insertions(+) > > > > diff --git a/app/test/test_power_cpufreq.c > > b/app/test/test_power_cpufreq.c index d099f2f47..c75c9bf1c 100644 > > --- a/app/test/test_power_cpufreq.c > > +++ b/app/test/test_power_cpufreq.c > > @@ -366,6 +366,59 @@ check_power_freq_min(void) > > return 0; > > } > > > > +/* Check rte_power_turbo() */ > > +static int > > +check_power_turbo(void) > > +{ > > + int ret; > > + > > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > > + printf("Turbo not available on lcore %u, skipping test\n", > > + TEST_POWER_LCORE_ID); > > Misleading indentation, please add two tabs to the second line of > printf() statement. > > > + return 0; > > + } > > + > > + /* test with an invalid lcore id */ > > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > > + if (ret >= 0) { > > + printf("Unexpectedly enable turbo successfully " > > + "on lcore %u\n", > TEST_POWER_LCORE_INVALID); > > Please don't break up strings to multiple lines. > > > + return -1; > > + } > > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > > + if (ret < 0) { > > + printf("Fail to enable turbo on lcore %u\n", > > + > TEST_POWER_LCORE_ID); > > I wish there was a middle ground between no indentation and way too much > indentation... > > > + return -1; > > + } > > + > > + /* Check the current frequency */ > > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > > + if (ret < 0) > > + return -1; > > + > > + /* test with an invalid lcore id */ > > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > > + if (ret >= 0) { > > + printf("Unexpectedly disable turbo successfully " > > + "on lcore %u\n", > TEST_POWER_LCORE_INVALID); > > Same as above, don't break up strings. > > > + return -1; > > + } > > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > > + if (ret < 0) { > > + printf("Fail to disable turbo on lcore %u\n", > > + > TEST_POWER_LCORE_ID); > > Same as above, two tabs is enough indentation. > > > + return -1; > > + } > > + > > + /* Check the current frequency */ > > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > > + if (ret < 0) > > + return -1; > > + > > + return 0; > > +} > > + > > static int > > test_power_cpufreq(void) > > { > > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > > "been initialised\n"); > > goto fail_all; > > } > > + if (rte_power_turbo_status == NULL) { > > + printf("rte_power_turbo_status should not be NULL, > environment has not " > > + "been initialised\n"); > > Here and below: > > I don't think saying *why* it should not be NULL - just saying that it shouldn't be > NULL is enough. Maybe mention why it's supposed to be not NULL in comments > here. > > Also, i have a suspicion that the message is wrong - why would status be null if > the environment was initialized? Presumably it's the other way around? I think this is correctly: if env was initialized, pointer != NULL, otherwise env wasn't initialized. Rest of all, you are right: coding style issues. Thanks. > > > + goto fail_all; > > + } > > + if (rte_power_freq_enable_turbo == NULL) { > > + printf("rte_power_freq_enable_turbo should not be NULL, > environment has not " > > + "been initialised\n"); > > + goto fail_all; > > + } > > + if (rte_power_freq_disable_turbo == NULL) { > > + printf("rte_power_freq_disable_turbo should not be NULL, > environment has not " > > + "been initialised\n"); > > + goto fail_all; > > + } > > > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > > if (ret < 0) { > > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > > if (ret < 0) > > goto fail_all; > > > > + ret = check_power_turbo(); > > + if (ret < 0) > > + goto fail_all; > > + > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > > if (ret < 0) { > > printf("Cannot exit power management for lcore %u\n", > > > > > -- > Thanks, > Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 9:07 ` Krakowiak, LukaszX @ 2019-04-03 10:01 ` Burakov, Anatoly 2019-04-03 10:01 ` Burakov, Anatoly 1 sibling, 1 reply; 34+ messages in thread From: Burakov, Anatoly @ 2019-04-03 10:01 UTC (permalink / raw) To: Krakowiak, LukaszX, Hunt, David; +Cc: dev On 03-Apr-19 10:07 AM, Krakowiak, LukaszX wrote: > Hi, > >> On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: >>> Add UT check_power_turbo. >>> >>> Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> >>> --- >>> app/test/test_power_cpufreq.c | 72 >> +++++++++++++++++++++++++++++++++++ >>> 1 file changed, 72 insertions(+) >>> >>> diff --git a/app/test/test_power_cpufreq.c >>> b/app/test/test_power_cpufreq.c index d099f2f47..c75c9bf1c 100644 >>> --- a/app/test/test_power_cpufreq.c >>> +++ b/app/test/test_power_cpufreq.c >>> @@ -366,6 +366,59 @@ check_power_freq_min(void) >>> return 0; >>> } >>> >>> +/* Check rte_power_turbo() */ >>> +static int >>> +check_power_turbo(void) >>> +{ >>> + int ret; >>> + >>> + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { >>> + printf("Turbo not available on lcore %u, skipping test\n", >>> + TEST_POWER_LCORE_ID); >> >> Misleading indentation, please add two tabs to the second line of >> printf() statement. >> >>> + return 0; >>> + } >>> + >>> + /* test with an invalid lcore id */ >>> + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); >>> + if (ret >= 0) { >>> + printf("Unexpectedly enable turbo successfully " >>> + "on lcore %u\n", >> TEST_POWER_LCORE_INVALID); >> >> Please don't break up strings to multiple lines. >> >>> + return -1; >>> + } >>> + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); >>> + if (ret < 0) { >>> + printf("Fail to enable turbo on lcore %u\n", >>> + >> TEST_POWER_LCORE_ID); >> >> I wish there was a middle ground between no indentation and way too much >> indentation... >> >>> + return -1; >>> + } >>> + >>> + /* Check the current frequency */ >>> + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); >>> + if (ret < 0) >>> + return -1; >>> + >>> + /* test with an invalid lcore id */ >>> + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); >>> + if (ret >= 0) { >>> + printf("Unexpectedly disable turbo successfully " >>> + "on lcore %u\n", >> TEST_POWER_LCORE_INVALID); >> >> Same as above, don't break up strings. >> >>> + return -1; >>> + } >>> + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); >>> + if (ret < 0) { >>> + printf("Fail to disable turbo on lcore %u\n", >>> + >> TEST_POWER_LCORE_ID); >> >> Same as above, two tabs is enough indentation. >> >>> + return -1; >>> + } >>> + >>> + /* Check the current frequency */ >>> + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); >>> + if (ret < 0) >>> + return -1; >>> + >>> + return 0; >>> +} >>> + >>> static int >>> test_power_cpufreq(void) >>> { >>> @@ -427,6 +480,21 @@ test_power_cpufreq(void) >>> "been initialised\n"); >>> goto fail_all; >>> } >>> + if (rte_power_turbo_status == NULL) { >>> + printf("rte_power_turbo_status should not be NULL, >> environment has not " >>> + "been initialised\n"); >> >> Here and below: >> >> I don't think saying *why* it should not be NULL - just saying that it shouldn't be >> NULL is enough. Maybe mention why it's supposed to be not NULL in comments >> here. >> >> Also, i have a suspicion that the message is wrong - why would status be null if >> the environment was initialized? Presumably it's the other way around? > > I think this is correctly: if env was initialized, pointer != NULL, otherwise env wasn't initialized. > Rest of all, you are right: coding style issues. Thanks. The check is if rte_power_turbo_status is equal to NULL. If the check fails (i.e. rte_power_turbo_status is not NULL), we get an error message saying that rte_power_turbo_status should not be NULL. That doesn't compute :) Either it should not be NULL and the check is wrong, or it should be NULL and the error message is wrong. >> >>> + goto fail_all; >>> + } >>> + if (rte_power_freq_enable_turbo == NULL) { >>> + printf("rte_power_freq_enable_turbo should not be NULL, >> environment has not " >>> + "been initialised\n"); >>> + goto fail_all; >>> + } >>> + if (rte_power_freq_disable_turbo == NULL) { >>> + printf("rte_power_freq_disable_turbo should not be NULL, >> environment has not " >>> + "been initialised\n"); >>> + goto fail_all; >>> + } >>> >>> ret = rte_power_exit(TEST_POWER_LCORE_ID); >>> if (ret < 0) { >>> @@ -502,6 +570,10 @@ test_power_cpufreq(void) >>> if (ret < 0) >>> goto fail_all; >>> >>> + ret = check_power_turbo(); >>> + if (ret < 0) >>> + goto fail_all; >>> + >>> ret = rte_power_exit(TEST_POWER_LCORE_ID); >>> if (ret < 0) { >>> printf("Cannot exit power management for lcore %u\n", >>> >> >> >> -- >> Thanks, >> Anatoly -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature 2019-04-03 10:01 ` Burakov, Anatoly @ 2019-04-03 10:01 ` Burakov, Anatoly 0 siblings, 0 replies; 34+ messages in thread From: Burakov, Anatoly @ 2019-04-03 10:01 UTC (permalink / raw) To: Krakowiak, LukaszX, Hunt, David; +Cc: dev On 03-Apr-19 10:07 AM, Krakowiak, LukaszX wrote: > Hi, > >> On 07-Mar-19 1:59 PM, Lukasz Krakowiak wrote: >>> Add UT check_power_turbo. >>> >>> Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> >>> --- >>> app/test/test_power_cpufreq.c | 72 >> +++++++++++++++++++++++++++++++++++ >>> 1 file changed, 72 insertions(+) >>> >>> diff --git a/app/test/test_power_cpufreq.c >>> b/app/test/test_power_cpufreq.c index d099f2f47..c75c9bf1c 100644 >>> --- a/app/test/test_power_cpufreq.c >>> +++ b/app/test/test_power_cpufreq.c >>> @@ -366,6 +366,59 @@ check_power_freq_min(void) >>> return 0; >>> } >>> >>> +/* Check rte_power_turbo() */ >>> +static int >>> +check_power_turbo(void) >>> +{ >>> + int ret; >>> + >>> + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { >>> + printf("Turbo not available on lcore %u, skipping test\n", >>> + TEST_POWER_LCORE_ID); >> >> Misleading indentation, please add two tabs to the second line of >> printf() statement. >> >>> + return 0; >>> + } >>> + >>> + /* test with an invalid lcore id */ >>> + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); >>> + if (ret >= 0) { >>> + printf("Unexpectedly enable turbo successfully " >>> + "on lcore %u\n", >> TEST_POWER_LCORE_INVALID); >> >> Please don't break up strings to multiple lines. >> >>> + return -1; >>> + } >>> + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); >>> + if (ret < 0) { >>> + printf("Fail to enable turbo on lcore %u\n", >>> + >> TEST_POWER_LCORE_ID); >> >> I wish there was a middle ground between no indentation and way too much >> indentation... >> >>> + return -1; >>> + } >>> + >>> + /* Check the current frequency */ >>> + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); >>> + if (ret < 0) >>> + return -1; >>> + >>> + /* test with an invalid lcore id */ >>> + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); >>> + if (ret >= 0) { >>> + printf("Unexpectedly disable turbo successfully " >>> + "on lcore %u\n", >> TEST_POWER_LCORE_INVALID); >> >> Same as above, don't break up strings. >> >>> + return -1; >>> + } >>> + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); >>> + if (ret < 0) { >>> + printf("Fail to disable turbo on lcore %u\n", >>> + >> TEST_POWER_LCORE_ID); >> >> Same as above, two tabs is enough indentation. >> >>> + return -1; >>> + } >>> + >>> + /* Check the current frequency */ >>> + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); >>> + if (ret < 0) >>> + return -1; >>> + >>> + return 0; >>> +} >>> + >>> static int >>> test_power_cpufreq(void) >>> { >>> @@ -427,6 +480,21 @@ test_power_cpufreq(void) >>> "been initialised\n"); >>> goto fail_all; >>> } >>> + if (rte_power_turbo_status == NULL) { >>> + printf("rte_power_turbo_status should not be NULL, >> environment has not " >>> + "been initialised\n"); >> >> Here and below: >> >> I don't think saying *why* it should not be NULL - just saying that it shouldn't be >> NULL is enough. Maybe mention why it's supposed to be not NULL in comments >> here. >> >> Also, i have a suspicion that the message is wrong - why would status be null if >> the environment was initialized? Presumably it's the other way around? > > I think this is correctly: if env was initialized, pointer != NULL, otherwise env wasn't initialized. > Rest of all, you are right: coding style issues. Thanks. The check is if rte_power_turbo_status is equal to NULL. If the check fails (i.e. rte_power_turbo_status is not NULL), we get an error message saying that rte_power_turbo_status should not be NULL. That doesn't compute :) Either it should not be NULL and the check is wrong, or it should be NULL and the error message is wrong. >> >>> + goto fail_all; >>> + } >>> + if (rte_power_freq_enable_turbo == NULL) { >>> + printf("rte_power_freq_enable_turbo should not be NULL, >> environment has not " >>> + "been initialised\n"); >>> + goto fail_all; >>> + } >>> + if (rte_power_freq_disable_turbo == NULL) { >>> + printf("rte_power_freq_disable_turbo should not be NULL, >> environment has not " >>> + "been initialised\n"); >>> + goto fail_all; >>> + } >>> >>> ret = rte_power_exit(TEST_POWER_LCORE_ID); >>> if (ret < 0) { >>> @@ -502,6 +570,10 @@ test_power_cpufreq(void) >>> if (ret < 0) >>> goto fail_all; >>> >>> + ret = check_power_turbo(); >>> + if (ret < 0) >>> + goto fail_all; >>> + >>> ret = rte_power_exit(TEST_POWER_LCORE_ID); >>> if (ret < 0) { >>> printf("Cannot exit power management for lcore %u\n", >>> >> >> >> -- >> Thanks, >> Anatoly -- Thanks, Anatoly ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library 2019-03-07 13:59 [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library Lukasz Krakowiak ` (2 preceding siblings ...) 2019-03-07 13:59 ` [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak ` (5 more replies) 3 siblings, 6 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch rename test_power_acpi_cpufreq.c to test_power_cpufreq.c, remove prefix _acpi from UT power functions/test names, add UT for power turbo feature. --- v2: * coding style improved Lukasz Krakowiak (3): test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c test: remove prefix _acpi from UT power function/test names test: add UT for power turbo feature app/test/Makefile | 2 +- app/test/autotest_data.py | 4 +- app/test/meson.build | 4 +- ...er_acpi_cpufreq.c => test_power_cpufreq.c} | 84 +++++++++++++++++-- 4 files changed, 83 insertions(+), 11 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (86%) ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak ` (4 subsequent siblings) 5 siblings, 0 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch rename test_power_acpi_cpufreq.c to test_power_cpufreq.c, remove prefix _acpi from UT power functions/test names, add UT for power turbo feature. --- v2: * coding style improved Lukasz Krakowiak (3): test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c test: remove prefix _acpi from UT power function/test names test: add UT for power turbo feature app/test/Makefile | 2 +- app/test/autotest_data.py | 4 +- app/test/meson.build | 4 +- ...er_acpi_cpufreq.c => test_power_cpufreq.c} | 84 +++++++++++++++++-- 4 files changed, 83 insertions(+), 11 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (86%) ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak ` (3 subsequent siblings) 5 siblings, 1 reply; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch rename file test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/Makefile | 2 +- app/test/meson.build | 2 +- app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} | 0 3 files changed, 2 insertions(+), 2 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (100%) diff --git a/app/test/Makefile b/app/test/Makefile index 89949c2bb..d1bcfd456 100644 --- a/app/test/Makefile +++ b/app/test/Makefile @@ -158,7 +158,7 @@ endif SRCS-$(CONFIG_RTE_LIBRTE_METER) += test_meter.c SRCS-$(CONFIG_RTE_LIBRTE_KNI) += test_kni.c -SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_acpi_cpufreq.c +SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_cpufreq.c SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power_kvm_vm.c SRCS-y += test_common.c diff --git a/app/test/meson.build b/app/test/meson.build index 05e5ddeb0..4ae3dffea 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -82,7 +82,7 @@ test_sources = files('commands.c', 'test_pmd_ring.c', 'test_pmd_ring_perf.c', 'test_power.c', - 'test_power_acpi_cpufreq.c', + 'test_power_cpufreq.c', 'test_power_kvm_vm.c', 'test_prefetch.c', 'test_reciprocal_division.c', diff --git a/app/test/test_power_acpi_cpufreq.c b/app/test/test_power_cpufreq.c similarity index 100% rename from app/test/test_power_acpi_cpufreq.c rename to app/test/test_power_cpufreq.c -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 0 siblings, 0 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch rename file test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/Makefile | 2 +- app/test/meson.build | 2 +- app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} | 0 3 files changed, 2 insertions(+), 2 deletions(-) rename app/test/{test_power_acpi_cpufreq.c => test_power_cpufreq.c} (100%) diff --git a/app/test/Makefile b/app/test/Makefile index 89949c2bb..d1bcfd456 100644 --- a/app/test/Makefile +++ b/app/test/Makefile @@ -158,7 +158,7 @@ endif SRCS-$(CONFIG_RTE_LIBRTE_METER) += test_meter.c SRCS-$(CONFIG_RTE_LIBRTE_KNI) += test_kni.c -SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_acpi_cpufreq.c +SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power.c test_power_cpufreq.c SRCS-$(CONFIG_RTE_LIBRTE_POWER) += test_power_kvm_vm.c SRCS-y += test_common.c diff --git a/app/test/meson.build b/app/test/meson.build index 05e5ddeb0..4ae3dffea 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -82,7 +82,7 @@ test_sources = files('commands.c', 'test_pmd_ring.c', 'test_pmd_ring_perf.c', 'test_power.c', - 'test_power_acpi_cpufreq.c', + 'test_power_cpufreq.c', 'test_power_kvm_vm.c', 'test_prefetch.c', 'test_reciprocal_division.c', diff --git a/app/test/test_power_acpi_cpufreq.c b/app/test/test_power_cpufreq.c similarity index 100% rename from app/test/test_power_acpi_cpufreq.c rename to app/test/test_power_cpufreq.c -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak ` (2 more replies) 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak ` (2 subsequent siblings) 5 siblings, 3 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch remove prefix _acpi from power UT function/test names. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/autotest_data.py | 4 ++-- app/test/meson.build | 2 +- app/test/test_power_cpufreq.c | 12 ++++++------ 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/app/test/autotest_data.py b/app/test/autotest_data.py index 5f87bb94d..c80801b20 100644 --- a/app/test/autotest_data.py +++ b/app/test/autotest_data.py @@ -622,8 +622,8 @@ "Report": None, }, { - "Name": "Power ACPI cpufreq autotest", - "Command": "power_acpi_cpufreq_autotest", + "Name": "Power cpufreq autotest", + "Command": "power_cpufreq_autotest", "Func": default_autotest, "Report": None, }, diff --git a/app/test/meson.build b/app/test/meson.build index 4ae3dffea..a389a2b10 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -205,7 +205,7 @@ fast_non_parallel_test_names = [ 'member_autotest', 'metrics_autotest', 'pdump_autotest', - 'power_acpi_cpufreq_autotest', + 'power_cpufreq_autotest', 'power_autotest', 'power_kvm_vm_autotest', 'reorder_autotest', diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index 61b1da05a..d099f2f47 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -14,14 +14,14 @@ #ifndef RTE_LIBRTE_POWER static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; } static int -test_power_acpi_caps(void) +test_power_caps(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; @@ -367,7 +367,7 @@ check_power_freq_min(void) } static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { int ret = -1; enum power_management_env env; @@ -539,7 +539,7 @@ test_power_acpi_cpufreq(void) } static int -test_power_acpi_caps(void) +test_power_caps(void) { struct rte_power_core_capabilities caps; int ret; @@ -568,5 +568,5 @@ test_power_acpi_caps(void) #endif -REGISTER_TEST_COMMAND(power_acpi_cpufreq_autotest, test_power_acpi_cpufreq); -REGISTER_TEST_COMMAND(power_acpi_caps_autotest, test_power_acpi_caps); +REGISTER_TEST_COMMAND(power_cpufreq_autotest, test_power_cpufreq); +REGISTER_TEST_COMMAND(power_caps_autotest, test_power_caps); -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:14 ` Hunt, David 2019-04-15 15:32 ` Kevin Traynor 2 siblings, 0 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak This patch remove prefix _acpi from power UT function/test names. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/autotest_data.py | 4 ++-- app/test/meson.build | 2 +- app/test/test_power_cpufreq.c | 12 ++++++------ 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/app/test/autotest_data.py b/app/test/autotest_data.py index 5f87bb94d..c80801b20 100644 --- a/app/test/autotest_data.py +++ b/app/test/autotest_data.py @@ -622,8 +622,8 @@ "Report": None, }, { - "Name": "Power ACPI cpufreq autotest", - "Command": "power_acpi_cpufreq_autotest", + "Name": "Power cpufreq autotest", + "Command": "power_cpufreq_autotest", "Func": default_autotest, "Report": None, }, diff --git a/app/test/meson.build b/app/test/meson.build index 4ae3dffea..a389a2b10 100644 --- a/app/test/meson.build +++ b/app/test/meson.build @@ -205,7 +205,7 @@ fast_non_parallel_test_names = [ 'member_autotest', 'metrics_autotest', 'pdump_autotest', - 'power_acpi_cpufreq_autotest', + 'power_cpufreq_autotest', 'power_autotest', 'power_kvm_vm_autotest', 'reorder_autotest', diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index 61b1da05a..d099f2f47 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -14,14 +14,14 @@ #ifndef RTE_LIBRTE_POWER static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; } static int -test_power_acpi_caps(void) +test_power_caps(void) { printf("Power management library not supported, skipping test\n"); return TEST_SKIPPED; @@ -367,7 +367,7 @@ check_power_freq_min(void) } static int -test_power_acpi_cpufreq(void) +test_power_cpufreq(void) { int ret = -1; enum power_management_env env; @@ -539,7 +539,7 @@ test_power_acpi_cpufreq(void) } static int -test_power_acpi_caps(void) +test_power_caps(void) { struct rte_power_core_capabilities caps; int ret; @@ -568,5 +568,5 @@ test_power_acpi_caps(void) #endif -REGISTER_TEST_COMMAND(power_acpi_cpufreq_autotest, test_power_acpi_cpufreq); -REGISTER_TEST_COMMAND(power_acpi_caps_autotest, test_power_acpi_caps); +REGISTER_TEST_COMMAND(power_cpufreq_autotest, test_power_cpufreq); +REGISTER_TEST_COMMAND(power_caps_autotest, test_power_caps); -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak @ 2019-04-15 15:14 ` Hunt, David 2019-04-15 15:14 ` Hunt, David 2019-04-15 15:32 ` Kevin Traynor 2 siblings, 1 reply; 34+ messages in thread From: Hunt, David @ 2019-04-15 15:14 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: anatoly.burakov, dev On 3/4/2019 11:32 AM, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- I would agree with Thomas, this patch should probably be merged with the file rename patch. Rgds, Dave. ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-15 15:14 ` Hunt, David @ 2019-04-15 15:14 ` Hunt, David 0 siblings, 0 replies; 34+ messages in thread From: Hunt, David @ 2019-04-15 15:14 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: anatoly.burakov, dev On 3/4/2019 11:32 AM, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- I would agree with Thomas, this patch should probably be merged with the file rename patch. Rgds, Dave. ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:14 ` Hunt, David @ 2019-04-15 15:32 ` Kevin Traynor 2019-04-15 15:32 ` Kevin Traynor 2 siblings, 1 reply; 34+ messages in thread From: Kevin Traynor @ 2019-04-15 15:32 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: anatoly.burakov, dev On 03/04/2019 11:32, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. > Hi, why is it better? Asking because it might mean you have to send some future fixes separately to stable branches. > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names 2019-04-15 15:32 ` Kevin Traynor @ 2019-04-15 15:32 ` Kevin Traynor 0 siblings, 0 replies; 34+ messages in thread From: Kevin Traynor @ 2019-04-15 15:32 UTC (permalink / raw) To: Lukasz Krakowiak, david.hunt; +Cc: anatoly.burakov, dev On 03/04/2019 11:32, Lukasz Krakowiak wrote: > This patch remove prefix _acpi from power UT function/test names. > Hi, why is it better? Asking because it might mean you have to send some future fixes separately to stable branches. > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak ` (2 preceding siblings ...) 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak ` (2 more replies) 2019-04-22 20:42 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Thomas Monjalon 2019-07-05 7:35 ` Thomas Monjalon 5 siblings, 3 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak Add UT check_power_turbo. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ 1 file changed, 72 insertions(+) diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index d099f2f47..d203810da 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -366,6 +366,59 @@ check_power_freq_min(void) return 0; } +/* Check rte_power_turbo() */ +static int +check_power_turbo(void) +{ + int ret; + + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { + printf("Turbo not available on lcore %u, skipping test\n", + TEST_POWER_LCORE_ID); + return 0; + } + + /* test with an invalid lcore id */ + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly enable turbo successfully on lcore %u\n", + TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to enable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); + if (ret < 0) + return -1; + + /* test with an invalid lcore id */ + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly disable turbo successfully on lcore %u\n", + TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to disable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); + if (ret < 0) + return -1; + + return 0; +} + static int test_power_cpufreq(void) { @@ -427,6 +480,21 @@ test_power_cpufreq(void) "been initialised\n"); goto fail_all; } + if (rte_power_turbo_status == NULL) { + printf("rte_power_turbo_status should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_enable_turbo == NULL) { + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_disable_turbo == NULL) { + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { @@ -502,6 +570,10 @@ test_power_cpufreq(void) if (ret < 0) goto fail_all; + ret = check_power_turbo(); + if (ret < 0) + goto fail_all; + ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { printf("Cannot exit power management for lcore %u\n", -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak @ 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:22 ` Hunt, David 2019-06-07 8:45 ` Hajkowski, MarcinX 2 siblings, 0 replies; 34+ messages in thread From: Lukasz Krakowiak @ 2019-04-03 10:32 UTC (permalink / raw) To: david.hunt; +Cc: anatoly.burakov, dev, Lukasz Krakowiak Add UT check_power_turbo. Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> --- app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ 1 file changed, 72 insertions(+) diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c index d099f2f47..d203810da 100644 --- a/app/test/test_power_cpufreq.c +++ b/app/test/test_power_cpufreq.c @@ -366,6 +366,59 @@ check_power_freq_min(void) return 0; } +/* Check rte_power_turbo() */ +static int +check_power_turbo(void) +{ + int ret; + + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { + printf("Turbo not available on lcore %u, skipping test\n", + TEST_POWER_LCORE_ID); + return 0; + } + + /* test with an invalid lcore id */ + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly enable turbo successfully on lcore %u\n", + TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to enable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); + if (ret < 0) + return -1; + + /* test with an invalid lcore id */ + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); + if (ret >= 0) { + printf("Unexpectedly disable turbo successfully on lcore %u\n", + TEST_POWER_LCORE_INVALID); + return -1; + } + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); + if (ret < 0) { + printf("Fail to disable turbo on lcore %u\n", + TEST_POWER_LCORE_ID); + return -1; + } + + /* Check the current frequency */ + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); + if (ret < 0) + return -1; + + return 0; +} + static int test_power_cpufreq(void) { @@ -427,6 +480,21 @@ test_power_cpufreq(void) "been initialised\n"); goto fail_all; } + if (rte_power_turbo_status == NULL) { + printf("rte_power_turbo_status should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_enable_turbo == NULL) { + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } + if (rte_power_freq_disable_turbo == NULL) { + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " + "been initialised\n"); + goto fail_all; + } ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { @@ -502,6 +570,10 @@ test_power_cpufreq(void) if (ret < 0) goto fail_all; + ret = check_power_turbo(); + if (ret < 0) + goto fail_all; + ret = rte_power_exit(TEST_POWER_LCORE_ID); if (ret < 0) { printf("Cannot exit power management for lcore %u\n", -- 2.19.2.windows.1 ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak @ 2019-04-15 15:22 ` Hunt, David 2019-04-15 15:22 ` Hunt, David 2019-06-07 8:45 ` Hajkowski, MarcinX 2 siblings, 1 reply; 34+ messages in thread From: Hunt, David @ 2019-04-15 15:22 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: anatoly.burakov, dev On 3/4/2019 11:32 AM, Lukasz Krakowiak wrote: > Add UT check_power_turbo. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- > app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ > 1 file changed, 72 insertions(+) > > diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c > index d099f2f47..d203810da 100644 > --- a/app/test/test_power_cpufreq.c > +++ b/app/test/test_power_cpufreq.c > @@ -366,6 +366,59 @@ check_power_freq_min(void) > return 0; > } > > +/* Check rte_power_turbo() */ > +static int > +check_power_turbo(void) > +{ > + int ret; > + > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > + printf("Turbo not available on lcore %u, skipping test\n", > + TEST_POWER_LCORE_ID); > + return 0; > + } > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly enable turbo successfully on lcore %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to enable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > + if (ret < 0) > + return -1; > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly disable turbo successfully on lcore %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to disable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > + if (ret < 0) > + return -1; > + > + return 0; > +} > + > static int > test_power_cpufreq(void) > { > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > "been initialised\n"); > goto fail_all; > } > + if (rte_power_turbo_status == NULL) { > + printf("rte_power_turbo_status should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_enable_turbo == NULL) { > + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_disable_turbo == NULL) { > + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > if (ret < 0) > goto fail_all; > > + ret = check_power_turbo(); > + if (ret < 0) > + goto fail_all; > + > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > printf("Cannot exit power management for lcore %u\n", Acked-by: David Hunt <david.hunt@intel.com> ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature 2019-04-15 15:22 ` Hunt, David @ 2019-04-15 15:22 ` Hunt, David 0 siblings, 0 replies; 34+ messages in thread From: Hunt, David @ 2019-04-15 15:22 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: anatoly.burakov, dev On 3/4/2019 11:32 AM, Lukasz Krakowiak wrote: > Add UT check_power_turbo. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- > app/test/test_power_cpufreq.c | 72 +++++++++++++++++++++++++++++++++++ > 1 file changed, 72 insertions(+) > > diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c > index d099f2f47..d203810da 100644 > --- a/app/test/test_power_cpufreq.c > +++ b/app/test/test_power_cpufreq.c > @@ -366,6 +366,59 @@ check_power_freq_min(void) > return 0; > } > > +/* Check rte_power_turbo() */ > +static int > +check_power_turbo(void) > +{ > + int ret; > + > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > + printf("Turbo not available on lcore %u, skipping test\n", > + TEST_POWER_LCORE_ID); > + return 0; > + } > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly enable turbo successfully on lcore %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to enable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > + if (ret < 0) > + return -1; > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly disable turbo successfully on lcore %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to disable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > + if (ret < 0) > + return -1; > + > + return 0; > +} > + > static int > test_power_cpufreq(void) > { > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > "been initialised\n"); > goto fail_all; > } > + if (rte_power_turbo_status == NULL) { > + printf("rte_power_turbo_status should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_enable_turbo == NULL) { > + printf("rte_power_freq_enable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_disable_turbo == NULL) { > + printf("rte_power_freq_disable_turbo should not be NULL, environment has not " > + "been initialised\n"); > + goto fail_all; > + } > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > if (ret < 0) > goto fail_all; > > + ret = check_power_turbo(); > + if (ret < 0) > + goto fail_all; > + > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > printf("Cannot exit power management for lcore %u\n", Acked-by: David Hunt <david.hunt@intel.com> ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:22 ` Hunt, David @ 2019-06-07 8:45 ` Hajkowski, MarcinX 2 siblings, 0 replies; 34+ messages in thread From: Hajkowski, MarcinX @ 2019-06-07 8:45 UTC (permalink / raw) To: Krakowiak, LukaszX, Hunt, David; +Cc: Burakov, Anatoly, dev, Krakowiak, LukaszX > -----Original Message----- > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Lukasz Krakowiak > Sent: Wednesday, April 3, 2019 12:33 PM > To: Hunt, David <david.hunt@intel.com> > Cc: Burakov, Anatoly <anatoly.burakov@intel.com>; dev@dpdk.org; > Krakowiak, LukaszX <lukaszx.krakowiak@intel.com> > Subject: [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature > > Add UT check_power_turbo. > > Signed-off-by: Lukasz Krakowiak <lukaszx.krakowiak@intel.com> > --- > app/test/test_power_cpufreq.c | 72 > +++++++++++++++++++++++++++++++++++ > 1 file changed, 72 insertions(+) > > diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c > index d099f2f47..d203810da 100644 > --- a/app/test/test_power_cpufreq.c > +++ b/app/test/test_power_cpufreq.c > @@ -366,6 +366,59 @@ check_power_freq_min(void) > return 0; > } > > +/* Check rte_power_turbo() */ > +static int > +check_power_turbo(void) > +{ > + int ret; > + > + if (rte_power_turbo_status(TEST_POWER_LCORE_ID) == 0) { > + printf("Turbo not available on lcore %u, skipping test\n", > + TEST_POWER_LCORE_ID); > + return 0; > + } > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly enable turbo successfully on lcore > %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_enable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to enable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 0); > + if (ret < 0) > + return -1; > + > + /* test with an invalid lcore id */ > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_INVALID); > + if (ret >= 0) { > + printf("Unexpectedly disable turbo successfully on lcore > %u\n", > + TEST_POWER_LCORE_INVALID); > + return -1; > + } > + ret = rte_power_freq_disable_turbo(TEST_POWER_LCORE_ID); > + if (ret < 0) { > + printf("Fail to disable turbo on lcore %u\n", > + TEST_POWER_LCORE_ID); > + return -1; > + } > + > + /* Check the current frequency */ > + ret = check_cur_freq(TEST_POWER_LCORE_ID, 1); > + if (ret < 0) > + return -1; > + > + return 0; > +} > + > static int > test_power_cpufreq(void) > { > @@ -427,6 +480,21 @@ test_power_cpufreq(void) > "been initialised\n"); > goto fail_all; > } > + if (rte_power_turbo_status == NULL) { > + printf("rte_power_turbo_status should not be NULL, > environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_enable_turbo == NULL) { > + printf("rte_power_freq_enable_turbo should not be NULL, > environment has not " > + "been initialised\n"); > + goto fail_all; > + } > + if (rte_power_freq_disable_turbo == NULL) { > + printf("rte_power_freq_disable_turbo should not be NULL, > environment has not " > + "been initialised\n"); > + goto fail_all; > + } > > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > @@ -502,6 +570,10 @@ test_power_cpufreq(void) > if (ret < 0) > goto fail_all; > > + ret = check_power_turbo(); > + if (ret < 0) > + goto fail_all; > + > ret = rte_power_exit(TEST_POWER_LCORE_ID); > if (ret < 0) { > printf("Cannot exit power management for lcore %u\n", > -- > 2.19.2.windows.1 Tested-by: Marcin Hajkowski <marcinx.hajkowski@intel.com> ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak ` (3 preceding siblings ...) 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak @ 2019-04-22 20:42 ` Thomas Monjalon 2019-04-22 20:42 ` Thomas Monjalon 2019-07-05 7:35 ` Thomas Monjalon 5 siblings, 1 reply; 34+ messages in thread From: Thomas Monjalon @ 2019-04-22 20:42 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: dev, david.hunt, anatoly.burakov > Lukasz Krakowiak (3): > test: rename test_power_acpi_cpufreq.c -> > app/test/test_power_cpufreq.c > test: remove prefix _acpi from UT power function/test names > test: add UT for power turbo feature No review, no merge, sorry. ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library 2019-04-22 20:42 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Thomas Monjalon @ 2019-04-22 20:42 ` Thomas Monjalon 0 siblings, 0 replies; 34+ messages in thread From: Thomas Monjalon @ 2019-04-22 20:42 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: dev, david.hunt, anatoly.burakov > Lukasz Krakowiak (3): > test: rename test_power_acpi_cpufreq.c -> > app/test/test_power_cpufreq.c > test: remove prefix _acpi from UT power function/test names > test: add UT for power turbo feature No review, no merge, sorry. ^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak ` (4 preceding siblings ...) 2019-04-22 20:42 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Thomas Monjalon @ 2019-07-05 7:35 ` Thomas Monjalon 5 siblings, 0 replies; 34+ messages in thread From: Thomas Monjalon @ 2019-07-05 7:35 UTC (permalink / raw) To: Lukasz Krakowiak; +Cc: dev, david.hunt, anatoly.burakov 03/04/2019 12:32, Lukasz Krakowiak: > Lukasz Krakowiak (3): > test: rename test_power_acpi_cpufreq.c -> > app/test/test_power_cpufreq.c > test: remove prefix _acpi from UT power function/test names > test: add UT for power turbo feature I decided to squash first 2 patches and apply, even if we don't have answer to some questions. ^ permalink raw reply [flat|nested] 34+ messages in thread
end of thread, other threads:[~2019-07-05 7:35 UTC | newest] Thread overview: 34+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2019-03-07 13:59 [dpdk-dev] [PATCH 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-03-07 13:59 ` [dpdk-dev] [PATCH 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak 2019-03-27 15:05 ` Burakov, Anatoly 2019-03-27 15:05 ` Burakov, Anatoly 2019-03-07 13:59 ` [dpdk-dev] [PATCH 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-27 15:07 ` Burakov, Anatoly 2019-03-29 22:40 ` Thomas Monjalon 2019-03-29 22:40 ` Thomas Monjalon 2019-03-07 13:59 ` [dpdk-dev] [PATCH 3/3] test: add UT for power turbo feature Lukasz Krakowiak 2019-03-27 15:13 ` Burakov, Anatoly 2019-03-27 15:13 ` Burakov, Anatoly 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 9:07 ` Krakowiak, LukaszX 2019-04-03 10:01 ` Burakov, Anatoly 2019-04-03 10:01 ` Burakov, Anatoly 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 1/3] test: rename test_power_acpi_cpufreq.c -> app/test/test_power_cpufreq.c Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 2/3] test: remove prefix _acpi from UT power function/test names Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:14 ` Hunt, David 2019-04-15 15:14 ` Hunt, David 2019-04-15 15:32 ` Kevin Traynor 2019-04-15 15:32 ` Kevin Traynor 2019-04-03 10:32 ` [dpdk-dev] [PATCH v2 3/3] test: add UT for power turbo feature Lukasz Krakowiak 2019-04-03 10:32 ` Lukasz Krakowiak 2019-04-15 15:22 ` Hunt, David 2019-04-15 15:22 ` Hunt, David 2019-06-07 8:45 ` Hajkowski, MarcinX 2019-04-22 20:42 ` [dpdk-dev] [PATCH v2 0/3] test: improve test coverage for power library Thomas Monjalon 2019-04-22 20:42 ` Thomas Monjalon 2019-07-05 7:35 ` Thomas Monjalon
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).