<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Sep 26, 2017, at 3:57 PM, Maksym Grebenets <<a href="mailto:mgrebenets@gmail.com" class="">mgrebenets@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class="">Hi Michael<div class=""><br class=""></div><div class="">Thanks for following up!</div><div class=""><br class=""></div><div class="">I've found some extra time too to give it few more tries.</div><div class="">As a matter of fact, I've tried quite a few things.</div><div class="">Spoiler alert - none of them worked :)</div><div class="">Here's the list of things anyways:</div><div class="">- A completely clean checkout of sift repo and all submodules (still same laptop, but to be 100% there's no left overs of any kind)</div><div class="">- Change LANG env var from en_GB.UTF-8 to en_US.UTF-8</div><div class="">- Use exactly the same version of CMake as on CI: 3.6.3</div><div class="">- Remove all extra software which is not on CI: python v3, Z3, cmake from Homebrew</div><div class="">- Make sure python 2.x version is almost the same as on CI (Jenkins has 2.7.10, while I have 2.7.13, recently bumped to 2.7.14)</div><div class=""><br class=""></div><div class="">- Checkout 100% identical source code as it is used by Jenkins</div><div class="">For this one I checked build #354 <a href="https://ci.swift.org/job/oss-swift-package-osx/354/parameters/" class="">https://ci.swift.org/job/oss-swift-package-osx/354/parameters/</a> and made sure all submodules have identical commits.</div><div class="">This is where I noticed that I was using latest ninja version 1.8.2, while Jenkins uses 1.7.2.</div><div class="">I also noticed that llvm is build from 1 commit above the swift-DEVELOPMENT-SNAPSHOT-2017-09-25-a tag somehow... (418d91a243e)</div><div class=""><br class=""></div><div class="">This one wasn't intentional, but I ended up trying to build 2017-09-25-a snapshot using 2017-09-25-a toolchain downloaded from <a href="http://swift.org/" class="">swift.org</a>.</div><div class=""><br class=""></div><div class="">All of that lands me with the same 3 failing tests for iPhone 5 simulator.</div><div class=""><br class=""></div><div class="">I know that I can work on code base without running the tests.</div><div class=""><br class=""></div><div class="">But I really wanted to be able to build the toolchain and try it out in Xcode, especially since blog post on refactoring just says "Use build-toolchain script to build your own toolchain"...</div></div></div></blockquote><div><br class=""></div><div>Quick update: I just checked my side machine. I was able to build everything successfully, pass tests, and build the package/toolchain for tag 'swift-DEVELOPMENT-SNAPSHOT-2017-09-18-a'.</div><div><br class=""></div><div>I am going to swing back and read the rest of this email later today. I need to finish a few things today.</div><div><br class=""></div><div>Michael</div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class=""><br class=""></div><div class="">Regards,</div><div class="">Max</div>
</div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Wed, Sep 27, 2017 at 3:11 AM, Michael Gottesman <span dir="ltr" class=""><<a href="mailto:mgottesman@apple.com" target="_blank" class="">mgottesman@apple.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class="">Ok. My side machine info:<div class=""><br class=""></div><div class="">```</div><div class="">$ sw_vers<br class="">ProductName:<span class="m_-5153177594578927515Apple-tab-span" style="white-space:pre-wrap"> </span>Mac OS X<br class="">ProductVersion:<span class="m_-5153177594578927515Apple-tab-span" style="white-space:pre-wrap"> </span>10.12.6<br class="">BuildVersion:<span class="m_-5153177594578927515Apple-tab-span" style="white-space:pre-wrap"> </span>16G26<br class="">$ xcodebuild -version<span class=""><br class="">Xcode 9.0<br class="">Build version 9M214v</span></div><div class="">$ cmake --version<br class="">cmake version 3.8.2<br class=""><br class="">CMake suite maintained and supported by Kitware (<a href="http://kitware.com/cmake" target="_blank" class="">kitware.com/cmake</a>).</div><div class="">$</div><div class="">```</div><div class=""><br class=""></div><div class="">So I think I should be able to reproduce since we have the same xcode/os version. I am starting a build now.</div><div class=""><div class="h5"><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Sep 26, 2017, at 9:44 AM, Michael Gottesman <<a href="mailto:mgottesman@apple.com" target="_blank" class="">mgottesman@apple.com</a>> wrote:</div><br class="m_-5153177594578927515Apple-interchange-newline"><div class=""><div style="word-wrap:break-word" class="">Ok. I am looking at this now on my side machine.<div class=""><br class=""></div><div class="">Sorry for the delay, I had to land code and then I had vacation = (.</div><div class=""><br class=""></div><div class="">Michael</div><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Sep 20, 2017, at 10:05 PM, Michael Gottesman via swift-dev <<a href="mailto:swift-dev@swift.org" target="_blank" class="">swift-dev@swift.org</a>> wrote:</div><br class="m_-5153177594578927515Apple-interchange-newline"><div class=""><div style="word-wrap:break-word" class=""><div class="">FYI, I haven't forgotten you! I just need to land some code tonight ; ). I want to try to reproduce exactly what you are doing.</div><div class=""><br class=""></div><div class="">Michael</div><div class=""><br class=""></div><div class=""><blockquote type="cite" class=""><div class="">On Sep 20, 2017, at 6:08 AM, Maksym Grebenets <<a href="mailto:mgrebenets@gmail.com" target="_blank" class="">mgrebenets@gmail.com</a>> wrote:</div><br class="m_-5153177594578927515Apple-interchange-newline"><div class=""><div dir="ltr" class="">I tried few more options.<div class=""><br class=""></div><div class="">I tried switching my shell to bash and giving it a go.</div><div class="">This time I got some unusual configuration error.</div><div class="">The error came from /usr/local/Cellar/cmake/<wbr class="">3.9.2/share/cmake/Modules/<wbr class="">FindPythonInterp.cmake.</div><div class="">Unfortunately the error message was in stderr, while I only captured stdout: <a href="https://gist.github.com/mgrebenets/bf3a54cb90141c17c715f5ddfddf6c04" target="_blank" class="">https://gist.github.<wbr class="">com/mgrebenets/<wbr class="">bf3a54cb90141c17c715f5ddfddf6c<wbr class="">04</a></div><div class=""><br class=""></div><div class="">So then I added CMake.app bin to the path</div><div class="">export PATH=/Applications/CMake.app/<wbr class="">Contents/bin:PATH<br class=""></div><div class="">and ran the build again.</div><div class="">This time it didn't fail with configuration error, but failed with "too many files open" and somehow still using cmake 3.9.2 from Homebrew installation:</div><div class=""><a href="https://gist.github.com/mgrebenets/100cf995fbb9c5415a26e0968aa7ed58" target="_blank" class="">https://gist.github.com/<wbr class="">mgrebenets/<wbr class="">100cf995fbb9c5415a26e0968aa7ed<wbr class="">58</a><br class=""></div><div class=""><br class=""></div><div class="">So finally, I decided to use CMake.app only:</div><div class="">brew unlink cmake<br class=""></div><div class="">Run again, and same 3 tests fail on same i386 simulator: </div><div class=""><a href="https://gist.github.com/mgrebenets/cf722b3e19894e7cbeeb3c4fa6d2511e" target="_blank" class="">https://gist.github.com/<wbr class="">mgrebenets/<wbr class="">cf722b3e19894e7cbeeb3c4fa6d251<wbr class="">1e</a><br class=""></div><div class=""><br class=""></div><div class="">Regards,</div><div class="">Max</div><div class=""><br class=""></div><div class=""><br class=""></div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Wed, Sep 20, 2017 at 2:59 PM, Maksym Grebenets <span dir="ltr" class=""><<a href="mailto:mgrebenets@gmail.com" target="_blank" class="">mgrebenets@gmail.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr" class="">I see.<div class=""><br class=""></div><div class="">I gave it another go: <a href="https://gist.github.com/mgrebenets/e2c981951586910c679df17d377c9e69" target="_blank" class="">https://gist.github.com/mg<wbr class="">rebenets/e2c981951586910c679df<wbr class="">17d377c9e69</a></div><div class=""><br class=""></div><div class="">Looking at the build logs, I can definitely see llvm source code being built.<br class="">I've looked up a bunch of llvm .cpp files in the build log, e.g. LLVMTargetMachine.cpp.</div><div class="">Unless I'm not looking in the right place.</div><div class=""><br class=""></div><div class="">Still same 3 failures in the end.</div><div class=""><br class=""></div><div class="">I have noticed that in certain places Jenkings builds are using /Applications/CMake.app, while on my machine it is <b style="color:rgb(69,69,69);font-family:"Helvetica Neue";font-size:12px" class="">/usr/local/Cellar/cmake/3.9<wbr class="">.1/bin/cmake.</b></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">So I've tried to upgrade to 3.9.2 and run same test command manually - same failures. </span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Same for using CMake.app downloaded from <a href="http://cmake.org/" target="_blank" class="">cmake.org</a> (3.9.2) version.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><b class=""><br class=""></b></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">I've double checked python version that I have, it's 2.7.10, same as on Jenkins instances.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><br class=""></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">I do have python3 installed in /usr/local/bin/python3, not sure if this could be related.<br class=""><br class=""></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Looking at the failing tests, it seems like all of them are related to encoding.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">The first one is related to Unicode 9 graphemes.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">For a string </span></font><span class="m_-5153177594578927515m_8825358364639058251gmail-s1" style="color:rgb(69,69,69)">"</span><span style="color:rgb(69,69,69);font-family:".Apple Color Emoji UI";font-size:12px" class="">πΊπΈπ¨π¦π©π°π³οΈβπ</span><span class="m_-5153177594578927515m_8825358364639058251gmail-s1" style="color:rgb(69,69,69)">" the expected count is 4, while on my machine it's 5.</span></div><div class=""><span class="m_-5153177594578927515m_8825358364639058251gmail-s1" style="color:rgb(69,69,69)">Other failures are in CodableTests.swift, but all of them report an error like:</span></div><div class=""><span class="m_-5153177594578927515m_8825358364639058251gmail-s1" style="color:rgb(69,69,69)"><div style="color:rgb(212,212,212);background-color:rgb(30,30,30);font-family:Menlo,Monaco,"Courier New",monospace;font-size:12px;line-height:18px;white-space:pre-wrap" class="">Decoded URLComponents <//<a href="http://0.0.0.0/" target="_blank" class="">0.0.0.0</a>> not equal to original <//<a href="http://0.0.0.0/" target="_blank" class="">0.0.0.0</a>></div></span></div>
<div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">I can't really spot the difference, unless there's some invisible escape character...</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><b class=""><br class=""></b></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Last failing test is for NSValue bridging:</span></font></div><span class=""><div class=""><div style="color:rgb(212,212,212);background-color:rgb(30,30,30);font-family:Menlo,Monaco,"Courier New",monospace;font-size:12px;line-height:18px;white-space:pre-wrap" class=""><div class="">stdout>>> check failed at /Users/grebenma/Projects/oss/s<wbr class="">wift/swift/stdlib/private/Stdl<wbr class="">ibUnittestFoundationExtras/Std<wbr class="">libUnittestFoundationExtras.<wbr class="">swift, line 130</div><div class="">stdout>>> expected: <00000000 00003140 00000000 00004340> (of type NSConcreteValue)</div><div class="">stdout>>> actual: <00000000 00003140 00000000 00004340> (of type NSConcreteValue)</div></div></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><b class=""><br class=""></b></span></font></div></span><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">It appears that all failures are for iPhone Simulator i386.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">So it must be related to 32-bit platforms only.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><br class=""></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">I don't have much ideas on what it could be.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Is it something set in my bash (zsh) profile?</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">E.g. something in the path, or some 3rd party tool installed.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Are there any hints in the log that may point me in the right direction?</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><br class=""></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class="">Thanks.</span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><br class=""></span></font></div><div class=""><font color="#454545" face="Helvetica Neue" class=""><span style="font-size:12px" class=""><br class=""></span></font></div>
</div>
</blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></div>______________________________<wbr class="">_________________<br class="">swift-dev mailing list<br class=""><a href="mailto:swift-dev@swift.org" target="_blank" class="">swift-dev@swift.org</a><br class=""><a href="https://lists.swift.org/mailman/listinfo/swift-dev" target="_blank" class="">https://lists.swift.org/<wbr class="">mailman/listinfo/swift-dev</a><br class=""></div></blockquote></div><br class=""></div></div></div></blockquote></div><br class=""></div></div></div></div></blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></body></html>