Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4000

Looks like this has crashed a few times recently: /home/buildnode/jenkins/workspace/oss-swift-incremental-RA-linux-ubuntu-16_10/swift/utils/build-script-impl: line 260: 17061 Segmentation fault "$@" /home/buildnode/jenkins/workspace/oss-swift-incremental-RA-linux-ubuntu-16_10/swift/utils/bu

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4003

Some non-determinsitic problem in sourcekit? *** Error in `/home/buildnode/jenkins/workspace/oss-swift-incremental-RA-linux-ubuntu-16_10/buildbot_incremental/swift-linux-x86_64/bin/sourcekitd-test': corrupted double-linked > On Jun 1, 2017, at 12:02 PM, no-re...@swift.org wrote: > > [FAILURE

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4003

Not sure what this is, David do you have a linux machine you can try this ? > On Jun 1, 2017, at 12:05 PM, Erik Eckstein wrote: > > Some non-determinsitic problem in sourcekit? > > *** Error in > `/home/buildnode/jenkins/workspace/oss-swift-incremental-RA-linux-ubuntu-16_10/buildbot_incrementa

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4000

https://bugs.swift.org/browse/SR-5067 The first failure was 3997, and it’s failed twice more since then, but not every time. #0 __memcpy_ssse3 () at ../sysdeps/x86_64/multiarch/memcpy-ssse3.S:1617 #1 0x7ffbf0d3bfd6 in __atomic_store () from /usr/lib

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4003

On the face of it, this kind of smells a little bit like a use-after-free or heap corruption something. Do we have Linux ASan bots running these tests? > On Jun 1, 2017, at 12:59 PM, David Farler wrote: > > I don't, although maybe rad-main would suffice. > > Mishal, did any libraries or packag

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4003

I don't, although maybe rad-main would suffice. Mishal, did any libraries or packages change on the Linux bots recently? Is rad-main a truly equivalent setup? Erik, do you have a sense for the repro rate of this problem already? David > On Jun 1, 2017, at 12:19 PM, Argyrios Kyrtzidis wrote: >

Re: [swift-dev] Override the path to the Swift compiler?

> On May 30, 2017, at 4:48 PM, Jimmy Yue via swift-dev > wrote: > > Hey Swift Devs, > > Does there happen to be an environment variable, similar to 'CC' for C, that > I could use to override the path to the Swift compiler? If so, is there also > one for ibtool, metal, etc.? > > I want to re

[swift-dev] PR-9995: enable exclusivity runtime checks.

Hi swift-dev, PR-9995 enables exclusivity runtime checks: https://github.com/apple/swift/pull/9995 These checks catch violations of exclusivity that can't be discovered statically [0], printing the message: "Simultaneous accesses to 0x*, but modification requires exclusive access." The error m

Re: [swift-dev] PR-9995: enable exclusivity runtime checks.

Runtime checks only take effect at -Onone. Even if you explicitly ask for checks via -enforce-exclusivity=checked, they will be stripped (with a warning from the driver). Support for dynamic checking at -O is something that will be considered post Swift 4. -Andy > On Jun 1, 2017, at 4:47 PM,

[swift-dev] Questions with SourceKit framework

Hey everyone, I am poking around Swift compiler to see if I am able to get a full AST out to do some preliminary code sanity check and dead code cleanup. So I was looking at open source SourceKitten to see if I can retrieve a full AST from it, however, I was

Re: [swift-dev] [Swift CI] Build Failure: 0. OSS - Swift Incremental RA - Ubuntu 16.10 (master) #4003

Hi David, We are only running ASan for macOS — https://ci.swift.org/job/oss-swift-incremental-ASAN-RA-osx/ . Also, we have not updated any libraries recently. Thanks, Mishal Shah > On Jun 1, 2017, at 1:07 PM, David Farler wrote: