RESOLVED FIXED 183812
Inlining of a function that ends in op_unreachable in a non-tail position triggers an ASSERT 
https://bugs.webkit.org/show_bug.cgi?id=183812
Summary Inlining of a function that ends in op_unreachable in a non-tail position tri...
Robin Morisset
Reported 2018-03-21 06:20:23 PDT
The fix I landed for https://bugs.webkit.org/show_bug.cgi?id=181027 was flawed: I tried setting the bytecodeIndex for the new block on line 1679, but it is going to be reset on line 6612. I did not catch this problem at the time, because of a quirk of the test case (there was nothing after the function call). The fix is simply to make the block untargetable by default, and let parseCodeBlock make it targetable afterwards if it is a jump target. rdar://problem/38415497
Attachments
Patch (5.11 KB, patch)
2018-03-21 06:24 PDT, Robin Morisset 		no flags
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Robin Morisset
Comment 1 2018-03-21 06:24:53 PDT
Created attachment 336188 [details] Patch
Robin Morisset
Comment 2 2018-03-21 07:58:11 PDT
The gtk-wk2 bot appears to have crashed/lost connection while running the tests. I strongly doubt it is caused by my patch.
David Kilzer (:ddkilzer)
Comment 3 2018-03-21 09:33:21 PDT
<rdar://problem/38415497>
Keith Miller
Comment 4 2018-03-30 06:38:27 PDT
Comment on attachment 336188 [details] Patch I think this makes sense. r=me.
WebKit Commit Bot
Comment 5 2018-03-30 07:03:15 PDT
Comment on attachment 336188 [details] Patch Clearing flags on attachment: 336188 Committed r230103: <https://trac.webkit.org/changeset/230103>
WebKit Commit Bot
Comment 6 2018-03-30 07:03:16 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.