On 6/19/07, via RT Ron Blaschke <[EMAIL PROTECTED]> wrote:
# New Ticket Created by Ron Blaschke # Please include the string: [perl #43250] # in the subject line of all future correspondence about this issue. # <URL: http://rt.perl.org/rt3/Ticket/Display.html?id=43250 >Test 5 of F<t/dynoplibs/myops.t> has some very strong timing constraint. Two sleeps of a second each (and a few minor int/print ops) may not be faster than a two seconds alarm. This often doesn't work out on my box (Windows XP, Visual C++). t\dynoplibs\myops....NOK 5# Failed test (t\dynoplibs\myops.t at line 77) # got: '1 # 2 # alarm # 3 # done. # ' # expected: '1 # alarm # 2 # 3 # done. # ' Attached patch unrolls the loop, which may or may not be more readable, but more importantly opens a +/- 1 second time window for the alarm. $ prove t\dynoplibs\myops.t t\dynoplibs\myops....ok All tests successful. Files=1, Tests=8, 10 wallclock secs ( 0.00 cusr + 0.00 csys = 0.00 CPU) Ron Index: t/dynoplibs/myops.t =================================================================== --- t/dynoplibs/myops.t (revision 19083) +++ t/dynoplibs/myops.t (working copy) @@ -80,14 +80,16 @@ .sub main :main find_global P0, "_alarm" alarm 2.0, P0 - set I0, 1 -loop: sleep 1 - print I0 - print "\n" - inc I0 - # check_events - le I0, 3, loop + print "1\n" + + # alarm should be triggered half-way + # during this sleep + sleep 2 + print "2\n" + + sleep 1 + print "3\n" print "done.\n" .end
thanks, applied as r19131. ~jerry
