ci: Start running wasm32 tests on Travis #46291

alexcrichton · 2017-11-27T00:43:04Z

This commit allocates a builder to running wasm32 tests on Travis. Not all test
suites pass right now so this is starting out with just the run-pass and the
libcore test suites. This'll hopefully give us a pretty broad set of coverage
for integration in rustc itself as well as a somewhat broad coverage of the llvm
backend itself through integration/unit tests.

alexcrichton · 2017-11-27T00:55:38Z

This is the successor of #45905 and #46291 and will help ensure that we don't regress features in wasm over time. Notably this will ensure we don't regress rustc integration itself, the ability to create wasm executables/dylibs/etc, as well as functionality in libstd/libcore. This also relatively importantly actually puts the llvm backend to the test to ensure its correctness.

In testing this turned up a bug previously known about in LLVM but one I hadn't gotten around to fixing yet. I've now backported the fix to our LLVM fork as well as the LLVM 5 branch and it's included here to help the tests pass.

Note that some downsides of this include:

Yet another target which may require liberal // ignore annotations to new tests. The wasm target ignores everything that emscripten would ignore so it's maybe not too bad, but there's certainly a lot of // ignore-wasm32-bare already.
Debugging wasm failures is crazy hard. There's no debugger! There's no println! It's mostly a stare-and-pray game.

I think that despite these downsides though it's good to get ahead of the game and get this target testing sooner rather than later. I think these points will get better over time as more and more functionality is implemented (like debuggers or "real" imports).

kennytm · 2017-11-27T09:28:59Z

src/ci/docker/wasm32-unknown/Dockerfile

+  src/test/run-pass \
+  src/test/compile-fail \
+  src/test/parse-fail \
+  src/test/run-fail \


I think it's better to remove this line over opting-out in bootstrap/check.rs? There needs to be a FIXME comment next to one of the workarounds anyway.

Sure yeah, I preemptively added that to get a ./x.py test --target wasm32-... working, but we can always easily add it back later.

hanna-kruppe · 2017-11-27T10:22:57Z

src/test/run-pass/saturating-float-casts.rs

@@ -10,6 +10,7 @@

 // Tests saturating float->int casts. See u128-as-f32.rs for the opposite direction.
 // compile-flags: -Z saturating-float-casts
+// ignore-wasm32-bare FIXME this should actually work, but doesn't?


Would be good to link #46298 here

alexcrichton · 2017-11-27T18:22:20Z

Updated!

r? @kennytm

kennytm · 2017-11-27T20:07:20Z

@bors r+ p=20

(raising priority since new tests from other PRs may not pass)

bors · 2017-11-27T20:07:21Z

📌 Commit e6977bb has been approved by kennytm

bors · 2017-11-27T21:56:19Z

⌛ Testing commit e6977bb74a1a75bc2ba405c1519f6d39093ddcd8 with merge ee17cf124e93f32e8eb27bb3180dafbcba8db0c9...

bors · 2017-11-27T23:40:13Z

💔 Test failed - status-travis

steveklabnik · 2017-11-27T23:49:04Z

looks spurious?

�[0Ktravis_fold:end:after_failure.1
�[0Ktravis_fold:start:after_failure.2
�[0Ktravis_time:start:28f0a230
�[0K$ cat obj/tmp/sccache.log
WARN:sccache::cache::s3: Got AWS error: Error(BadHTTPStatus(NotFound), State { next_error: None })
WARN:sccache::cache::s3: Got AWS error: Error(BadHTTPStatus(NotFound), State { next_error: None })

travis_time:end:28f0a230:start=1511825893488996918,finish=1511825893494620866,duration=5623948
�[0Ktravis_fold:end:after_failure.2
�[0Ktravis_fold:start:after_failure.3
�[0Ktravis_time:start:093c2a7c
�[0K$ cat /tmp/sccache.log
cat: /tmp/sccache.log: No such file or directory

travis_time:end:093c2a7c:start=1511825893499958946,finish=1511825893504677327,duration=4718381
�[0Ktravis_fold:end:after_failure.3
�[0Ktravis_fold:start:after_failure.4
�[0Ktravis_time:start:024aa040
�[0K$ ls -lat $HOME/Library/Logs/DiagnosticReports/
ls: cannot access /home/travis/Library/Logs/DiagnosticReports/: No such file or directory

travis_time:end:024aa040:start=1511825893510452059,finish=1511825893516475163,duration=6023104
�[0Ktravis_fold:end:after_failure.4
�[0Ktravis_fold:start:after_failure.5
�[0Ktravis_time:start:04998e31
�[0K$ find $HOME/Library/Logs/DiagnosticReports -type f -not -name '*.stage2-*.crash' -not -name 'com.apple.CoreSimulator.CoreSimulatorService-*.crash' -exec printf travis_fold":start:crashlog\n\033[31;1m%s\033[0m\n" {} \; -exec head -750 {} \; -exec echo travis_fold":"end:crashlog \;
find: `/home/travis/Library/Logs/DiagnosticReports': No such file or directory

travis_time:end:04998e31:start=1511825893522040560,finish=1511825893528102089,duration=6061529
�[0Ktravis_fold:end:after_failure.5
�[0Ktravis_fold:start:after_failure.6
�[0Ktravis_time:start:1095a09c
�[0K$ dmesg | grep -i kill
[   17.681956] init: failsafe main process (1133) killed by TERM signal

travis_time:end:1095a09c:start=1511825893532877544,finish=1511825893540881585,duration=8004041
�[0Ktravis_fold:end:after_failure.6
�[0K

kennytm · 2017-11-27T23:54:57Z

No, legit.

[01:39:21] failures:
[01:39:21] 
[01:39:21] ---- [mir-opt] mir-opt/packed-struct-drop-aligned.rs stdout ----
[01:39:21] 	thread '[mir-opt] mir-opt/packed-struct-drop-aligned.rs' panicked at 'Did not find expected line, error: Mismatch in lines
[01:39:21] Currnt block:     bb0: {
[01:39:21] Expected Line: "        drop(_6) -> bb2;                 // scope 1 at /checkout/src/test/mir-opt/packed-struct-drop-aligned.rs:13:5: 13:8"
[01:39:21] Actual Line: "        drop(_6) -> [return: bb4, unwind: bb3];"

alexcrichton · 2017-11-28T00:27:44Z

nah I think that's a test I need to ignore, I'll do that in a bit

This commit allocates a builder to running wasm32 tests on Travis. Not all test suites pass right now so this is starting out with just the run-pass and the libcore test suites. This'll hopefully give us a pretty broad set of coverage for integration in rustc itself as well as a somewhat broad coverage of the llvm backend itself through integration/unit tests.

alexcrichton · 2017-11-28T17:27:46Z

@bors: r=kennytm

bors · 2017-11-28T17:27:46Z

📌 Commit 73970bf has been approved by kennytm

ci: Start running wasm32 tests on Travis This commit allocates a builder to running wasm32 tests on Travis. Not all test suites pass right now so this is starting out with just the run-pass and the libcore test suites. This'll hopefully give us a pretty broad set of coverage for integration in rustc itself as well as a somewhat broad coverage of the llvm backend itself through integration/unit tests.

bors · 2017-11-28T17:59:06Z

⌛ Testing commit 73970bf with merge 71340ca...

bors · 2017-11-28T20:40:44Z

☀️ Test successful - status-appveyor, status-travis
Approved by: kennytm
Pushing 71340ca to master...

steveklabnik · 2017-11-28T21:34:23Z

Not only am I pumped that this landed, but @alexcrichton , apparently this is the 2000th PR you've closed out! 🎊 💯 ❤️

@alexcrichton

…=alexcrichton Rename and fix nolink-with-link-args test There are three problems with the nolink-with-link-args test: * The test fails when using MSVC. It's caused by the `linker-flavor=ld` flag which was added in rust-lang#46291. * In its comment, this test tests that "link_args are indeed passed when nolink is specified", but the `nolink` attribute has been removed [a long time ago](rust-lang#12826). * Pattern has a small typo. At first I was going to completely remove this test, but there is [a closed pull request for that](rust-lang#21090). So: * rename the file as suggested in the closed PR * adjust the comment * fix typo in the pattern * add `ignore-msvc`. r? @alexcrichton

@alexcrichton

…=alexcrichton Rename and fix nolink-with-link-args test There are three problems with the nolink-with-link-args test: * The test fails when using MSVC. It's caused by the `linker-flavor=ld` flag which was added in rust-lang#46291. * In its comment, this test tests that "link_args are indeed passed when nolink is specified", but the `nolink` attribute has been removed [a long time ago](rust-lang#12826). * Pattern has a small typo. At first I was going to completely remove this test, but there is [a closed pull request for that](rust-lang#21090). So: * rename the file as suggested in the closed PR * adjust the comment * fix typo in the pattern * add `ignore-msvc`. r? @alexcrichton

alexcrichton mentioned this pull request Nov 27, 2017

std: Add a new wasm32-unknown-unknown target #45905

Merged

alexcrichton force-pushed the wasm-tests branch from 956483d to edb541d Compare November 27, 2017 01:21

kennytm added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-infra Relevant to the infrastructure team, which will review and decide on the PR/issue. labels Nov 27, 2017

kennytm reviewed Nov 27, 2017

View reviewed changes

hanna-kruppe reviewed Nov 27, 2017

View reviewed changes

alexcrichton force-pushed the wasm-tests branch from edb541d to e6977bb Compare November 27, 2017 18:21

rust-highfive assigned kennytm Nov 27, 2017

kennytm added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 27, 2017

kennytm added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Nov 28, 2017

alexcrichton force-pushed the wasm-tests branch from e6977bb to 73970bf Compare November 28, 2017 17:27

kennytm added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Nov 28, 2017

bors merged commit 73970bf into rust-lang:master Nov 28, 2017

alexcrichton deleted the wasm-tests branch November 29, 2017 00:08

Smibu mentioned this pull request Dec 31, 2018

Rename and fix nolink-with-link-args test #57233

Merged

ci: Start running wasm32 tests on Travis #46291

ci: Start running wasm32 tests on Travis #46291

Uh oh!

Conversation

alexcrichton commented Nov 27, 2017

Uh oh!

alexcrichton commented Nov 27, 2017

Uh oh!

kennytm Nov 27, 2017

Choose a reason for hiding this comment

Uh oh!

alexcrichton Nov 27, 2017

Choose a reason for hiding this comment

Uh oh!

hanna-kruppe Nov 27, 2017

Choose a reason for hiding this comment

Uh oh!

alexcrichton Nov 27, 2017

Choose a reason for hiding this comment

Uh oh!

alexcrichton commented Nov 27, 2017

Uh oh!

kennytm commented Nov 27, 2017

Uh oh!

bors commented Nov 27, 2017

Uh oh!

bors commented Nov 27, 2017

Uh oh!

bors commented Nov 27, 2017

Uh oh!

steveklabnik commented Nov 27, 2017

Uh oh!

kennytm commented Nov 27, 2017

Uh oh!

alexcrichton commented Nov 28, 2017

Uh oh!

alexcrichton commented Nov 28, 2017

Uh oh!

bors commented Nov 28, 2017

Uh oh!

bors commented Nov 28, 2017

Uh oh!

bors commented Nov 28, 2017

Uh oh!

steveklabnik commented Nov 28, 2017

Uh oh!

Uh oh!