liblitmus.git/bin, branch 2012.1 [ARCHIVE] The userspace library for LITMUS^RT 2010. Now on GitHub. Export wctime() and cputime() from rtspin to library clients 2011-02-03T22:40:16+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2011-01-30T17:18:26+00:00 86b445e39c3c7eed076b01795b17f34481517288 The timing functions are quite handy when building benchmark tasks. Avoid copy&paste reuse by making them available via the library. 
The timing functions are quite handy when building benchmark tasks.
Avoid copy&paste reuse by making them available via the library.
bugfix: switch operangs in delay computation 2011-02-03T19:52:43+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2011-02-03T19:52:43+00:00 46f3b11a1f49d49120415b56b6f627c66e4b4d1b We want the positive loop length, not the negative, to prevent jobs from overruning their budget. 
We want the positive loop length, not the negative, to prevent jobs
from overruning their budget.
Remove old delay loop configuration cruft from rtspin 2011-01-30T17:28:28+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2011-01-30T17:26:36+00:00 455f8e4ef3fd02d41e2d8ae40918dd22bf7e4f45 Using cputime() is much more accurate than the old delay loop auto-configuration. There is no good reason to keep it around. 
Using cputime() is much more accurate than the old delay loop
auto-configuration. There is no good reason to keep it around.
Modify RTSpin so that it can read execution time data from a CSV file. 2011-01-30T17:06:07+00:00 Christopher Kenna cjk@cs.unc.edu 2011-01-13T17:05:08+00:00 c013cd8625b0ebb28f71337109d3fbfdf0a6b621 Still need to figure out what to do with the period is less than the exec_cost (or budget). 
Still need to figure out what to do with the period is less than the
exec_cost (or budget).
rtspin: transition back to non-real-time mode 2010-11-22T20:00:12+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2010-11-09T00:19:53+00:00 ee91e39147039693f9f4063e319b836adb8401bf rtspin produces prettier schedules if it transitions to non-real-time mode before terminating. 
rtspin produces prettier schedules if it transitions to non-real-time
mode before terminating.
output header to STDERR in measure_syscall 2010-11-09T21:35:33+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2010-11-03T19:31:55+00:00 310881b43bcf20d0a74e94cfcd0090716e90e8ea This hopefully helps a bit to clarify the output. 
This hopefully helps a bit to clarify the output.
add emergency exit to rtspin 2010-11-09T21:35:33+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2010-11-06T20:54:09+00:00 a957bc576d0bdab19f41d367dc662acd3bb82ae9 This should only trigger if the execution time tracking is broken (as it was on District10). With the emergency exit, we at least prevent the system from becoming unresponsive. 
This should only trigger if the execution time tracking is broken (as
it was on District10). With the emergency exit, we at least prevent
the system from becoming unresponsive.
refactor: use architecture-specific includes for cycles.h 2010-11-09T21:35:32+00:00 Bjoern B. Brandenburg bbb@cs.unc.edu 2010-11-06T20:36:31+00:00 f8585fe1fc6f0830f900dad7c8ccc40e17f79644 






rtspin: add -e option to enable PRECISE_ENFORCEMENT
2010-10-19T14:05:36+00:00

Bjoern B. Brandenburg
bbb@cs.unc.edu

2010-10-19T14:05:36+00:00

c3566e210d4e466e97d84d4a190760c4e91766ac

Useful to test how the kernel handles PRECISE_ENFORCEMENT.





Useful to test how the kernel handles PRECISE_ENFORCEMENT.







Support budget enforcement policies.  Allows tasks to specify
2010-05-20T18:17:58+00:00

Glenn Elliott
gelliott@koruna.cs.unc.edu

2010-05-20T18:17:58+00:00

e9c900fae35e4e4730469e189ff17bf30518346a

how their execution budgets should be enforced: NO_ENFORCEMENT,
QUANTUM_ENFORCEMENT, and PRECISE_ENFORCEMENT (unsupported).

NOTE: Users of NO_ENFORCEMENT must call sleep_next_period() at the
end of every job to signal to the kernel that its job is complete.





how their execution budgets should be enforced: NO_ENFORCEMENT,
QUANTUM_ENFORCEMENT, and PRECISE_ENFORCEMENT (unsupported).

NOTE: Users of NO_ENFORCEMENT must call sleep_next_period() at the
end of every job to signal to the kernel that its job is complete.