SI-6616 Check that unsafe operations are only called on the presentation... #1583
Conversation
…ion compiler thread. The method that checks the actual constraint is @elidable, expecting it to be used for nightly builds but stripped-off in release builds. This way we don't lose any performance, but 'fail-fast' in IDE nightlies. This assumes that release builds will have at least `-Xelide-below ASSERTION`, but this pull request does not do that.
|
Started jenkins job pr-rangepos at https://scala-webapps.epfl.ch/jenkins/job/pr-rangepos/856/ |
|
Started jenkins job pr-scala-testsuite-linux-opt at https://scala-webapps.epfl.ch/jenkins/job/pr-scala-testsuite-linux-opt/1563/ |
|
jenkins job pr-rangepos: Success - https://scala-webapps.epfl.ch/jenkins/job/pr-rangepos/856/ |
|
jenkins job pr-scala-testsuite-linux-opt: Success - https://scala-webapps.epfl.ch/jenkins/job/pr-scala-testsuite-linux-opt/1563/ |
|
Yet another initialization tracker.... @paulp didn't you add another one of these recently in typer? |
|
Yes, that is either redundant with 'isGlobalInitialized', or we should make it so. (By which I mean, if isGlobalInitialized doesn't have exactly the desired semantics for this, I am sure we can find something which satisfies all uses.) |
|
Thanks for pointing that out. My guess is that I can use the other method, I only need to eliminate some false positives. Even if it's not as precise as my variable, it would still be ok. I'll make the changes ASAP. |
|
..the only problem being that |
|
@paulp is there any reason not to cherry-pick isGlobalInitialzied from master -> 2.10.x? |
|
No reason. |
|
I'll close this pull request and open another one with the cherry-pick. |
|
The commit itself is small, but depends on a bunch of other refactorings and can't be easily backported to 2.10.x anymore. I decided to keep this pull request for 2.10.x, and open a new one for 2.11 using the new support. |
|
So guys, WDYT? |
|
Looks good to me. |
SI-6616 Check that unsafe operations are only called on the presentation...
| @@ -51,6 +56,7 @@ class Global(settings: Settings, _reporter: Reporter, projectName: String = "") | |||
| import log.logreplay | |||
| debugLog("logger: " + log.getClass + " writing to " + (new java.io.File(logName)).getAbsolutePath) | |||
| debugLog("classpath: "+classPath) | |||
| Console.err.println("\n ======= CHECK THREAD ACCESS compiler build ========\n") | |||
There was a problem hiding this comment.
oops -- how come all reviewers missed this? :-)
There was a problem hiding this comment.
Clearly the reviewers in this case were a bunch of scrubs.
There was a problem hiding this comment.
It's only printing that in the presentation compiler, if that's any consolation :)
... compiler thread.
The method that checks the actual constraint is @elidable, expecting it to be used
for nightly builds but stripped-off in release builds. This way we don't lose any
performance, but 'fail-fast' in IDE nightlies.
This assumes that release builds will have at least
-Xelide-below ASSERTION, butthis pull request does not do that.
review by @adriaanm, @paulp