diff options
Diffstat (limited to 'js/src/jit-test/tests/debug/clear-old-analyses-01.js')
-rw-r--r-- | js/src/jit-test/tests/debug/clear-old-analyses-01.js | 38 |
1 files changed, 38 insertions, 0 deletions
diff --git a/js/src/jit-test/tests/debug/clear-old-analyses-01.js b/js/src/jit-test/tests/debug/clear-old-analyses-01.js new file mode 100644 index 0000000000..4b580563ce --- /dev/null +++ b/js/src/jit-test/tests/debug/clear-old-analyses-01.js @@ -0,0 +1,38 @@ +// |jit-test| error:AllDone +// When we enter debug mode in a compartment, we must throw away all +// analyses in that compartment (debug mode affects the results of +// analysis, so they become out of date). This is true even when we would +// otherwise be retaining jit code and its related data structures for +// animation timing. + +if (typeof gcPreserveCode != "function") + throw('AllDone'); + +var g = newGlobal({newCompartment: true}); +var dbg = new Debugger; + +g.eval("" + + function fib(n) { + var a = 0, b = 1; + while (n-- > 0) + b = b+a, a = b-a; + return b; + }); + +g.fib(20); // Cause g.fib to be jitted. This creates an analysis with + // debug mode off. + +gcPreserveCode(); // Tell the gc to preserve JIT code and analyses by + // default. A recent call to js::NotifyAnimationActivity + // could have a similar effect in real life. + +dbg.addDebuggee(g); // Put g in debug mode. This triggers a GC which must + // clear all analyses. In the original buggy code, we also + // release all of g's scripts' JIT code, leading to a + // recompilation the next time it was called. + +g.fib(20); // Run g.fib again, causing it to be re-jitted. If the + // original analysis is still present, JM will assert, + // because it is not in debug mode. + +throw('AllDone'); |