user/sven/linux.git/tools/testing/selftests/resctrl/mbm_test.c, branch v6.7.2

selftests/resctrl: Fix MBM test failure when MBA unavailable

2023-10-18T20:33:44Z

Commit 20d96b25cc4c ("selftests/resctrl: Fix schemata write error check") exposed a problem in feature detection logic in MBM selftest. If schemata does not support MB:x=x entries, the schemata write to initialize 100% memory bandwidth allocation in mbm_setup() will now fail with -EINVAL due to the error handling corrected by the commit 20d96b25cc4c ("selftests/resctrl: Fix schemata write error check"). That commit just uncovers the failed write, it is not wrong itself. If MB:x=x is not supported by schemata, it is safe to assume 100% memory bandwidth is always set. Therefore, the previously ignored error does not make the MBM test itself wrong. Restore the previous behavior of MBM test by checking MB support before attempting to write it into schemata which results in behavior equivalent to ignoring the write error. Fixes: 20d96b25cc4c ("selftests/resctrl: Fix schemata write error check") Signed-off-by: Ilpo Järvinen Reviewed-by: Reinette Chatre Signed-off-by: Shuah Khan

selftests/resctrl: Reduce failures due to outliers in MBA/MBM tests

2023-10-13T20:54:38Z

The initial value of 5% chosen for the maximum allowed percentage difference between resctrl mbm value and IMC mbm value in commit 06bd03a57f8c ("selftests/resctrl: Fix MBA/MBM results reporting format") was "randomly chosen value" (as admitted by the changelog). When running tests in our lab across a large number platforms, 5% difference upper bound for success seems a bit on the low side for the MBA and MBM tests. Some platforms produce outliers that are slightly above that, typically 6-7%, which leads MBA/MBM test frequently failing. Replace the "randomly chosen value" with a success bound that is based on those measurements across large number of platforms by relaxing the MBA/MBM success bound to 8%. The relaxed bound removes the failures due the frequent outliers. Fixed commit description style error during merge: Shuah Khan Fixes: 06bd03a57f8c ("selftests/resctrl: Fix MBA/MBM results reporting format") Signed-off-by: Ilpo Järvinen Tested-by: Shaopeng Tan Reviewed-by: Reinette Chatre Reviewed-by: Shaopeng Tan Cc: Signed-off-by: Shuah Khan

selftests/resctrl: Make benchmark command const and build it with pointers

2023-10-13T20:28:55Z

Benchmark command is used in multiple tests so it should not be mutated by the tests but CMT test alters span argument. Due to the order of tests (CMT test runs last), mutating the span argument in CMT test does not trigger any real problems currently. Mark benchmark_cmd strings as const and setup the benchmark command using pointers. Because the benchmark command becomes const, the input arguments can be used directly. Besides being simpler, using the input arguments directly also removes the internal size restriction. CMT test has to create a copy of the benchmark command before altering the benchmark command. Signed-off-by: Ilpo Järvinen Tested-by: Shaopeng Tan Reviewed-by: Shaopeng Tan Reviewed-by: Reinette Chatre Reviewed-by: "Wieczor-Retman, Maciej" Signed-off-by: Shuah Khan

selftests/resctrl: Simplify span lifetime

2023-10-13T20:28:44Z

struct resctrl_val_param contains span member. resctrl_val(), however, never uses it because the value of span is embedded into the default benchmark command and parsed from it by run_benchmark(). Remove span from resctrl_val_param. Provide DEFAULT_SPAN for the code that needs it. CMT and CAT tests communicate span that is different from the DEFAULT_SPAN between their internal functions which is converted into passing it directly as a parameter. Signed-off-by: Ilpo Järvinen Tested-by: Shaopeng Tan Reviewed-by: Reinette Chatre Reviewed-by: Shaopeng Tan Reviewed-by: "Wieczor-Retman, Maciej" Signed-off-by: Shuah Khan

selftests/resctrl: Remove bw_report and bm_type from main()

2023-10-13T20:28:38Z

bw_report is always set to "reads" and bm_type is set to "fill_buf" but is never used. Set bw_report directly to "reads" in MBA/MBM test and remove bm_type. Signed-off-by: Ilpo Järvinen Tested-by: Shaopeng Tan Reviewed-by: Reinette Chatre Reviewed-by: Shaopeng Tan Reviewed-by: "Wieczor-Retman, Maciej" Signed-off-by: Shuah Khan

selftests/resctrl: Don't use variable argument list for ->setup()

2023-07-25T14:53:53Z

struct resctrl_val_param has ->setup() function that accepts variable argument list. All test cases use only 1 argument as input and it's the struct resctrl_val_param pointer. Instead of variable argument list, directly pass struct resctrl_val_param pointer as the only parameter to ->setup(). Co-developed-by: Fenghua Yu Signed-off-by: Fenghua Yu Signed-off-by: Ilpo Järvinen Reviewed-by: Reinette Chatre Tested-by: Babu Moger Tested-by: Shaopeng Tan (Fujitsu) Signed-off-by: Shuah Khan

selftests/resctrl: Express span internally in bytes

2023-07-25T14:53:18Z

MBA and MBM tests to use megabytes to represent span. CMT test uses bytes. The difference requires run_benchmark() to size the buffer differently based on the test name, which in turn requires passing the test name into run_benchmark(). Convert MBA and MBM tests to use span internally in bytes like CMT test to remove the internal inconsistency between the tests. Remove the test dependent buffer sizing from run_benchmark(). This change eliminates one of the reasons why the test name has to be passed around but there are still other users too so the test name passing cannot yet be removed. Co-developed-by: Fenghua Yu Signed-off-by: Fenghua Yu Signed-off-by: Ilpo Järvinen Tested-by: Babu Moger Tested-by: Shaopeng Tan (Fujitsu) Signed-off-by: Shuah Khan

selftests/resctrl: Convert span to size_t

2023-07-25T14:53:13Z

Span is defined either as unsigned long or int. Consistently use size_t everywhere for span as it refers to size of the memory block. Co-developed-by: Fenghua Yu Signed-off-by: Fenghua Yu Signed-off-by: Ilpo Järvinen Tested-by: Babu Moger Tested-by: Shaopeng Tan (Fujitsu) Signed-off-by: Shuah Khan

selftests/resctrl: Remove mum_resctrlfs from struct resctrl_val_param

2023-07-25T14:53:06Z

Resctrl FS mount/umount are now cleanly paired leaving .mum_resctrlfs in the struct resctrl_val_param unused. Remove .mum_resctrlfs from struct resctrl_val_param. Signed-off-by: Ilpo Järvinen Reviewed-by: Reinette Chatre Tested-by: Babu Moger Tested-by: Shaopeng Tan (Fujitsu) Signed-off-by: Shuah Khan

selftests/resctrl: Fix incorrect error return on test complete

2023-04-14T17:13:18Z

An error snuck in between two recent conflicting changes: Until recently ->setup() used negative values to indicate normal test termination. This was changed in commit fa10366cc6f4 ("selftests/resctrl: Allow ->setup() to return errors") that transitioned ->setup() to use negative values to indicate errors and a new END_OF_TESTS to indicate normal termination. commit 42e3b093eb7c ("selftests/resctrl: Fix set up schemata with 100% allocation on first run in MBM test") continued to use negative return to indicate normal test termination. Fix mbm_setup() to use the new END_OF_TESTS to indicate error-free test termination. Fixes: 42e3b093eb7c ("selftests/resctrl: Fix set up schemata with 100% allocation on first run in MBM test") Reported-by: Ilpo Järvinen Link: https://lore.kernel.org/lkml/bb65cce8-54d7-68c5-ef19-3364ec95392a@linux.intel.com/ Signed-off-by: Reinette Chatre Signed-off-by: Shuah Khan