Hi Shaopeng,
On 9/28/2022 10:28 PM, tan.shaopeng@fujitsu.com wrote:
On 9/27/2022 2:01 AM, tan.shaopeng@fujitsu.com wrote:
On 9/13/2022 6:51 PM, Shaopeng Tan wrote:
Before exiting each test function(run_cmt/cat/mbm/mba_test()), test results are printed by ksft_print_msg() and then temporary result files are cleaned by function cmt/cat/mbm/mba_test_cleanup(). However, before running ksft_print_msg(), function
before -> after?
I think it is "before".
hmmm ... if cmt/cat/mbm/mba_test_cleanup() was run before ksft_print_msg() then there would be no test results to print, no? The current implementation runs cmt/cat/mbm/mba_test_cleanup() after ksft_print_msg() ... albeit twice.
I am sorry I made a mistake in changelog. It should be ksft_test_result() instead of ksft_print_msg().
Changelog: Before exiting each test function(run_cmt/cat/mbm/mba_test()), test results (“ok”/”not ok”) are printed by ksft_test_result() and then temporary result files are cleaned by function cmt/cat/mbm/mba_test_cleanup(). However, before running ksft_test_result(), function cmt/cat/mbm/mba_test_cleanup() has been run in each test function as follows: cmt_resctrl_val() cat_perf_miss_val() mba_schemata_change() mbm_bw_change()
Remove duplicate codes that clear each test result file.
This is clear, thank you very much.
Reinette