RESOLVED FIXED181150
[JSC] JIT requires sizeof(bool) == 1 
https://bugs.webkit.org/show_bug.cgi?id=181150
Summary [JSC] JIT requires sizeof(bool) == 1
Yusuke Suzuki
Reported 2017-12-25 05:27:23 PST
[JSC] Use uint8_t instead of bool if it is used in JIT
Attachments
Patch (6.05 KB, patch)
2017-12-25 05:28 PST, Yusuke Suzuki 		no flags
DetailsFormatted DiffDiff
Archive of layout-test-results from ews106 for mac-elcapitan-wk2 (2.82 MB, application/zip)
2017-12-25 07:58 PST, EWS Watchlist 		no flags
Details
Patch (1.46 KB, patch)
2018-01-21 01:00 PST, Yusuke Suzuki 		no flags
DetailsFormatted DiffDiff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.
Yusuke Suzuki
Comment 1 2017-12-25 05:28:35 PST
Created attachment 330178 [details] Patch
EWS Watchlist
Comment 2 2017-12-25 07:58:08 PST
Comment on attachment 330178 [details] Patch Attachment 330178 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/5826319 New failing tests: imported/w3c/web-platform-tests/service-workers/service-worker/ServiceWorkerGlobalScope/update.https.html
EWS Watchlist
Comment 3 2017-12-25 07:58:09 PST
Created attachment 330183 [details] Archive of layout-test-results from ews106 for mac-elcapitan-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews106 Port: mac-elcapitan-wk2 Platform: Mac OS X 10.11.6
Yusuke Suzuki
Comment 4 2017-12-30 10:38:34 PST
The failed test is not related to this patch.
Alex Christensen
Comment 5 2018-01-02 11:31:13 PST
Comment on attachment 330178 [details] Patch Are you building on a system where sizeof(bool) != 1?
Yusuke Suzuki
Comment 6 2018-01-04 07:59:34 PST
(In reply to Alex Christensen from comment #5) > Comment on attachment 330178 [details] > Patch > > Are you building on a system where sizeof(bool) != 1? No. But it is not defined. To avoid difficult bugs, I think we should do either, 1. Changing bool to uint8_t if it is touched from JIT code 2. Add the requirement to WebKit, `static_assert(sizeof(bool) == 1, "")`
Saam Barati
Comment 7 2018-01-04 08:44:15 PST
(In reply to Yusuke Suzuki from comment #6) > (In reply to Alex Christensen from comment #5) > > Comment on attachment 330178 [details] > > Patch > > > > Are you building on a system where sizeof(bool) != 1? > > No. But it is not defined. To avoid difficult bugs, I think we should do > either, > > 1. Changing bool to uint8_t if it is touched from JIT code > 2. Add the requirement to WebKit, `static_assert(sizeof(bool) == 1, "")` I like (2)
Yusuke Suzuki
Comment 8 2018-01-04 08:46:19 PST
(In reply to Saam Barati from comment #7) > (In reply to Yusuke Suzuki from comment #6) > > (In reply to Alex Christensen from comment #5) > > > Comment on attachment 330178 [details] > > > Patch > > > > > > Are you building on a system where sizeof(bool) != 1? > > > > No. But it is not defined. To avoid difficult bugs, I think we should do > > either, > > > > 1. Changing bool to uint8_t if it is touched from JIT code > > 2. Add the requirement to WebKit, `static_assert(sizeof(bool) == 1, "")` > > I like (2) OK, I'll update the patch.
Yusuke Suzuki
Comment 9 2018-01-21 01:00:11 PST
Created attachment 331862 [details] Patch
WebKit Commit Bot
Comment 10 2018-01-23 07:11:05 PST
Comment on attachment 331862 [details] Patch Clearing flags on attachment: 331862 Committed r227411: <https://trac.webkit.org/changeset/227411>
WebKit Commit Bot
Comment 11 2018-01-23 07:11:07 PST
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 12 2018-01-23 07:12:32 PST
<rdar://problem/36774590>
Note You need to log in before you can comment on or make changes to this bug.