diff options
Diffstat (limited to 'scripts')
| -rwxr-xr-x | scripts/fuzz_opt.py | 16 |
1 files changed, 13 insertions, 3 deletions
diff --git a/scripts/fuzz_opt.py b/scripts/fuzz_opt.py index 4fc92f367..f272ddf83 100755 --- a/scripts/fuzz_opt.py +++ b/scripts/fuzz_opt.py @@ -729,15 +729,25 @@ def run_bynterp(wasm, args): del os.environ['BINARYEN_MAX_INTERPRETER_DEPTH'] +# Enable even more staged things than V8_OPTS. V8_OPTS are the flags we want to +# use when testing, and enable all features we test against, while --future may +# also enable non-feature things like new JITs and such (which are never needed +# for our normal tests, but do make sense to fuzz for V8's sake). We do this +# randomly for more variety. +def get_v8_extra_flags(): + return ['--future'] if random.random() < 0.5 else [] + + V8_LIFTOFF_ARGS = ['--liftoff', '--no-wasm-tier-up'] -# default to running with liftoff enabled, because we need to pick either +# Default to running with liftoff enabled, because we need to pick either # liftoff or turbo* for consistency (otherwise running the same command twice # may have different results due to NaN nondeterminism), and liftoff is faster -# for small things +# for small things. def run_d8_js(js, args=[], liftoff=True): cmd = [shared.V8] + shared.V8_OPTS + cmd += get_v8_extra_flags() if liftoff: cmd += V8_LIFTOFF_ARGS cmd += [js] @@ -838,7 +848,7 @@ class CompareVMs(TestCaseHandler): name = 'd8' def run(self, wasm, extra_d8_flags=[]): - return run_vm([shared.V8, FUZZ_SHELL_JS] + shared.V8_OPTS + extra_d8_flags + ['--', wasm]) + return run_vm([shared.V8, FUZZ_SHELL_JS] + shared.V8_OPTS + get_v8_extra_flags() + extra_d8_flags + ['--', wasm]) def can_run(self, wasm): # V8 does not support shared memories when running with |