If you build --with-static-checking, spidermonkey breaks because it's looking for $(topsrcdir)/xpcom/analysis/static-checking.js which obviously doesn't exist. Since the main-tree static checking isn't much use in spidermonkey, I think we should just disable static checking in Spidermonkey until jorendorff has the cx->fp write barrier analysis ready to land.
I think 'make check' will complain about the difference between js/src/config/static-checking-config.mk and config/static-checking-config.mk. Couldn't we have the two be identical, and have js/src/Makefile.in set some variable that says, "I decline static checking?"
Comment on attachment 350996 [details] [diff] [review] Add static-checking-config.mk to the exceptions list, rev. 1 r=me, but if the intent is to have different static-checking-config.mk files setting different defaults for $topsrcdir and js/src, then it seems like js/src/static-check-config.mk should pretty much just be empty, except for the comment explaining that it's a placeholder until we have some analyses we want to run there. It seems odd to have references to xpcom/analysis anywhere in js/src; I think it will confuse people about how independent js/src really is.
ok, pushed http://hg.mozilla.org/mozilla-central/rev/4b4ee8b2dc54 without any dangling references to xpcom
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Comment on attachment 350996 [details] [diff] [review] Add static-checking-config.mk to the exceptions list, rev. 1 No-risk, and we really should have static checking available on the branch ;-)
Comment on attachment 350996 [details] [diff] [review] Add static-checking-config.mk to the exceptions list, rev. 1 a191=beltzner
Attachment #350996 - Flags: approval1.9.1? → approval1.9.1+
Pushed to the 1.9.1 branch: http://hg.mozilla.org/releases/mozilla-1.9.1/rev/734900e6f341
You need to log in before you can comment on or make changes to this bug.