The version of rtcheck in MRG has a number of issues which have resolved in this latest version (attached below). They include: Updates include: o whitespace cleanups o reimplemented output macros (greatly reduced code size) o removal of hrtimer loop (runs much faster now) o use clock_gettime(CLOCK_MONOTONIC,...) instead of gtod() o use clock_nanosleep() instead of nanosleep() o sleep for measurable period of time (50us) rather than 1ns (which is a no-op system call basically) o verbose flag always prints reported values (not just on fail) o verbose flag always reruns tests (ignores cache) o new -p priority option o test is run by default as SCHED_FIFO 95 (rather than SCHED_OTHER) o test is run with mlockall(CURRENT|FUTURE) to avoid paging induced latencies o cleaned up output formatting for verbose flag (more intuitive messages) rtcheck.c version 7-4 (updated) Updated version of rtcheck.c with ns aware clock resolution reporting.
Created attachment 318126 [details] rtcheck.c version 7-4 (updated)
can I get a CHANGELOG updated for 0.7.4? Better yet, do you have a tarball with all the things needed (Makefile, rtcheck.c, CHANGELOG, INSTALL, README, TODO)?
build rtcheck-0.7.4-2 from tarball supplied by IBM Scheduled to deliver in MRG 1.1
Created attachment 322448 [details] Updated rtcheck with a changelog and init script Please update to this latest version of rtcheck (0.7.5) which includes a full CHANGELOG and a proper init script. I have tested it on MRG-GA.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2009-0034.html