[Webkit-unassigned] [Bug 177962] New: WebAssembly: optimize signature check on call_indirect

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=177962

            Bug ID: 177962
           Summary: WebAssembly: optimize signature check on call_indirect
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: WebAssembly
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: jfbastien at apple.com
                CC: fpizlo at apple.com, jfbastien at apple.com,
                    keith_miller at apple.com, mark.lam at apple.com,
                    msaboff at apple.com, rmorisset at apple.com,
                    sbarati at apple.com

Keith and I were talking, and we think the signature check could be optimized. We only check that the signature is as expected (so load from table, compare to immediate in code, should be the same). We could instead subtract, not, and then divide by that value (or anything else that traps). That'll SIGFPE on x86, and should only keep the divide unit busy but latency doesn't matter because we don't use the divide result. Maybe there's a similar instruction on ARM (idiv just returns zero on / 0).

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20171005/87376b17/attachment.html>