Document threads tests that may fail under jitter

[pintos-anon] / doc / threads.texi

diff --git a/doc/threads.texi b/doc/threads.texi
index c1f3bfcd62c2caf43a0b1cbf1118b0cf8ffdb62f..bf8546e20526523a22fec5da30b76d68ddf9364a 100644 (file)
--- a/doc/threads.texi
+++ b/doc/threads.texi
@@ -643,6 +643,17 @@ to cause many of the tests to fail.
 
 @xref{Testing}.
 
 
 @xref{Testing}.
 

+@item Should I try running the tests with jitter?
+
+Using the jitter feature in Bochs (@pxref{Debugging versus Testing})
+is a great way to discover bugs that are timing dependent. However,
+the following tests are known to
+fail with jitter even if your code is correct: @code{alarm-simultaneous},
+@code{mlfqs-recent-1}, @code{mlfqs-fair-2}, @code{mlfqs-fair-20},
+@code{mlfqs-nice-2}, @code{mlfqs-nice-10}, and @code{priority-fifo}.
+The behavior of these tests can sometimes vary based on timing (e.g.,
+if a timer interrupt arrives at an inconvenient time).
+

 @item Why do I get a test failure in @func{pass}?
 
 @anchor{The pass function fails}
 @item Why do I get a test failure in @func{pass}?
 
 @anchor{The pass function fails}


@@ -726,7 +737,7 @@ kernel thread but the first.
 Don't worry about the possibility of timer values overflowing.  Timer
 values are expressed as signed 64-bit numbers, which at 100 ticks per
 second should be good for almost 2,924,712,087 years.  By then, we
 Don't worry about the possibility of timer values overflowing.  Timer
 values are expressed as signed 64-bit numbers, which at 100 ticks per
 second should be good for almost 2,924,712,087 years.  By then, we

-expect Pintos to have been phased out of the @value{coursenumber} curriculum.
+expect Pintos to have been phased out of the Computer Science curriculum.

 @end table
 
 @node Priority Scheduling FAQ
 @end table
 
 @node Priority Scheduling FAQ