Bug 128789
| Summary: | A number of js/regress tests simply take way too long to run in debug on Windows and should be skipped for now | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Roger Fong <roger_fong> |
| Component: | Tools / Tests | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | bfulgham, ggaren, mark.lam, msaboff, roger_fong |
| Priority: | P2 | Keywords: | LayoutTestFailure |
| Version: | 528+ (Nightly build) | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Roger Fong
For example, one test has the following code.
var cycles = 20000
var numberObject = Number
function foo() {
var result = 0;
var innerCycles = cycles;
var Number = numberObject;
for (var i = 0; i < innerCycles; ++i)
result += 0 | isNaN(Number.NaN);
return result;
}
Perhaps we should consider decreasing the number of cycles. It has no issue, if I set the cycles to say, 2000.
js/regress/Int16Array-bubble-sort-with-byteLength.html [ Timeout ]
js/regress/Int16Array-bubble-sort.html [ Timeout ]
js/regress/chain-custom-getter.html [ Timeout ]
js/regress/chain-getter-access.html [ Timeout ]
js/regress/marsaglia.html [ Timeout ]
js/regress/proto-custom-getter.html [ Timeout ]
js/regress/proto-getter-access.html [ Timeout ]
js/regress/simple-custom-getter.html [ Timeout ]
js/regress/simple-getter-access.html [ Timeout ]]
Skipping for now
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Roger Fong
I don't know when this regressed.
It seems to be fine on release.