Really install cargo#16

vext01 · 2019-04-25T15:11:17Z

No description provided.

vext01 · 2019-04-25T15:11:25Z

bors try

ltratt · 2019-04-25T16:49:24Z

.buildbot.sh


 # Archive the build and put it in /opt
-TARBALL_TOPDIR=ykrustc-stage2
+TARBALL_TOPDIR=ykrustc-stage2 # To match [install] in the build config.


This comment is a bit too cryptic, I think.

It has to match the string used in the toml file:

[install] prefix = ...

(i'll push a better comment once it works)

ltratt · 2019-04-25T16:49:57Z

.buildbot.sh

-cd build/x86_64-unknown-linux-gnu
-ln -sf stage2 ${TARBALL_TOPDIR}
+mkdir -p ${TARBALL_TOPDIR} # Or you get "cannot canonicalize..."
+env RUST_BACKTRACE=1 ./x.py install --config .buildbot.toml


Do we need the RUST_BACKTRACE thing on? IIRC, it does have some sort of performance consequence.

I found a post where someone claims it does, but no-one knows why.

I'd like to have backtraces if possible, because otherwise I may have to build again to know what went wrong.

bors · 2019-04-25T20:07:07Z

try

Build succeeded

buildbot/buildbot-build-script

vext01 · 2019-04-25T21:20:41Z

This worked, but we added about an hour to the build time.

The last pushes (amongst a couple of other bits) prevents rust from building documentation. Hopefully it'll be faster.

bors try

bors · 2019-04-26T02:01:18Z

try

Build failed

buildbot/buildbot-build-script

vext01 · 2019-04-26T08:59:56Z

+ git show -s HEAD^2
fatal: ambiguous argument 'HEAD^2': unknown revision or path not in the working tree.

That is caused by one of the other commits.

So it looks like we might shave 15 minutes off without building docs.

vext01 · 2019-04-26T09:11:18Z

Looking at the logs, the install stage builds:

the installer (whatever that is)
cargo
rustfmt

And those three items take:

Build completed successfully in 0:49:36

So I think that's all time accounted for.

We might be able to invoke individual targets for cargo and rustfmt and then not use cargo install to avoid building the installer.

Do you have any thoughts before I go ahead with that?

ltratt · 2019-04-26T09:23:40Z

Let's do it.

vext01 · 2019-04-26T11:43:54Z

bors try

bors · 2019-04-26T11:44:51Z

try

Build failed

buildbot/buildbot-build-script

vext01 · 2019-04-26T12:44:05Z

bors try

bors · 2019-04-26T18:16:33Z

try

Build succeeded

buildbot/buildbot-build-script

vext01 · 2019-04-26T19:59:35Z

I don't understand why that took longer than what we had before.

ltratt · 2019-04-26T20:02:00Z

Do you want to squash?

vext01 · 2019-04-26T20:03:16Z

Do you want to squash?

Well no, because the build took ages compared to what we had before.

I'm starting to wonder if there's a lot of variation...

I'm going to kick of the build again on the same version and see if it takes the same time.

bors try

ltratt · 2019-04-26T20:04:11Z

OK.

bors · 2019-04-27T00:37:53Z

try

Build succeeded

buildbot/buildbot-build-script

vext01 · 2019-04-27T14:12:14Z

So we ran two builds on the same commit, the first took 5.5 hours, the second took 4.5. Looks to me like there's significant non-determinism in the compiler, or perhaps the buildbot was busier during the first build?

What do you think we should do? Good news is, once SSA is gone the build should be a bit faster.

ltratt · 2019-04-27T15:38:36Z

I wouldn't read too much into the times: who knows what else the machine was doing at that point? Can we merge this now or ...?

vext01 · 2019-04-28T18:47:33Z

bors try

vext01 · 2019-04-28T18:56:53Z

Can we merge this now or ...?

Not just yet.

I did a test run on yk, and it bombed out because rustdoc wasn't present. The last commit invokes that target. Bors is testing that now.

Another annoying quirk is that when you don't use cargo install (but instead invoke targets manually as we have here), cargo fmt doesn't work. You have to invoke rustfmt manually with a file list. A bit annoying...

bors · 2019-04-28T23:22:31Z

try

Build succeeded

buildbot/buildbot-build-script

vext01 · 2019-04-29T09:04:59Z

I'm going to revert to using cargo install to build the tarball because I don't want to layer on any more hacks to make this work.

ltratt · 2019-04-29T09:06:55Z

OK.

vext01 · 2019-04-29T10:49:08Z

bors try

bors · 2019-04-29T15:44:47Z

try

Build succeeded

buildbot/buildbot-build-script

vext01 · 2019-04-29T15:45:55Z

OK, so this works. Let me now test the resulting tarball on yk.

vext01 · 2019-04-29T16:29:01Z

ykjit/yk#1 is now passing.

I think this is ready to squash, unless you have any comments?

ltratt · 2019-04-29T21:44:18Z

Please squash.

Also ensure rustdoc is installed.

vext01 · 2019-04-30T09:03:26Z

splat

ltratt · 2019-04-30T09:12:42Z

bors r+

16: Really install cargo r=ltratt a=vext01 Co-authored-by: Edd Barrett <vext01@gmail.com>

bors · 2019-04-30T13:52:23Z

Build succeeded

buildbot/buildbot-build-script

This is a combination of 18 commits. Commit softdevteam#2: Additional examples and some small improvements. Commit softdevteam#3: fixed mir-opt non-mir extensions and spanview title elements Corrected a fairly recent assumption in runtest.rs that all MIR dump files end in .mir. (It was appending .mir to the graphviz .dot and spanview .html file names when generating blessed output files. That also left outdated files in the baseline alongside the files with the incorrect names, which I've now removed.) Updated spanview HTML title elements to match their content, replacing a hardcoded and incorrect name that was left in accidentally when originally submitted. Commit softdevteam#4: added more test examples also improved Makefiles with support for non-zero exit status and to force validation of tests unless a specific test overrides it with a specific comment. Commit softdevteam#5: Fixed rare issues after testing on real-world crate Commit softdevteam#6: Addressed PR feedback, and removed temporary -Zexperimental-coverage -Zinstrument-coverage once again supports the latest capabilities of LLVM instrprof coverage instrumentation. Also fixed a bug in spanview. Commit softdevteam#7: Fix closure handling, add tests for closures and inner items And cleaned up other tests for consistency, and to make it more clear where spans start/end by breaking up lines. Commit softdevteam#8: renamed "typical" test results "expected" Now that the `llvm-cov show` tests are improved to normally expect matching actuals, and to allow individual tests to override that expectation. Commit softdevteam#9: test coverage of inline generic struct function Commit softdevteam#10: Addressed review feedback * Removed unnecessary Unreachable filter. * Replaced a match wildcard with remining variants. * Added more comments to help clarify the role of successors() in the CFG traversal Commit softdevteam#11: refactoring based on feedback * refactored `fn coverage_spans()`. * changed the way I expand an empty coverage span to improve performance * fixed a typo that I had accidently left in, in visit.rs Commit softdevteam#12: Optimized use of SourceMap and SourceFile Commit softdevteam#13: Fixed a regression, and synched with upstream Some generated test file names changed due to some new change upstream. Commit softdevteam#14: Stripping out crate disambiguators from demangled names These can vary depending on the test platform. Commit softdevteam#15: Ignore llvm-cov show diff on test with generics, expand IO error message Tests with generics produce llvm-cov show results with demangled names that can include an unstable "crate disambiguator" (hex value). The value changes when run in the Rust CI Windows environment. I added a sed filter to strip them out (in a prior commit), but sed also appears to fail in the same environment. Until I can figure out a workaround, I'm just going to ignore this specific test result. I added a FIXME to follow up later, but it's not that critical. I also saw an error with Windows GNU, but the IO error did not specify a path for the directory or file that triggered the error. I updated the error messages to provide more info for next, time but also noticed some other tests with similar steps did not fail. Looks spurious. Commit softdevteam#16: Modify rust-demangler to strip disambiguators by default Commit softdevteam#17: Remove std::process::exit from coverage tests Due to Issue #77553, programs that call std::process::exit() do not generate coverage results on Windows MSVC. Commit softdevteam#18: fix: test file paths exceeding Windows max path len

bors bot added a commit that referenced this pull request Apr 25, 2019

Try #16:

571ae77

ltratt reviewed Apr 25, 2019

View reviewed changes

bors bot added a commit that referenced this pull request Apr 25, 2019

Try #16:

6d5d302

bors bot added a commit that referenced this pull request Apr 26, 2019

Try #16:

ba0a26c

bors bot added a commit that referenced this pull request Apr 26, 2019

Try #16:

b445c7a

bors bot added a commit that referenced this pull request Apr 26, 2019

Try #16:

3c194d0

bors bot added a commit that referenced this pull request Apr 28, 2019

Try #16:

6665bda

bors bot added a commit that referenced this pull request Apr 29, 2019

Try #16:

6c7c879

Use ./x.py install to ensure installation of extended tools.

37225c0

Also ensure rustdoc is installed.

vext01 force-pushed the yk-really-install-cargo branch from 63e9bfc to 37225c0 Compare April 30, 2019 09:03

bors bot added a commit that referenced this pull request Apr 30, 2019

Merge #16

1d3e5d8

16: Really install cargo r=ltratt a=vext01 Co-authored-by: Edd Barrett <vext01@gmail.com>

bors bot merged commit 37225c0 into master Apr 30, 2019

bors bot deleted the yk-really-install-cargo branch April 30, 2019 13:52

Conversation

vext01 commented Apr 25, 2019

Uh oh!

vext01 commented Apr 25, 2019

Uh oh!

ltratt Apr 25, 2019

Choose a reason for hiding this comment

Uh oh!

vext01 Apr 25, 2019

Choose a reason for hiding this comment

Uh oh!

vext01 Apr 25, 2019

Choose a reason for hiding this comment

Uh oh!

ltratt Apr 25, 2019

Choose a reason for hiding this comment

Uh oh!

vext01 Apr 25, 2019

Choose a reason for hiding this comment

Uh oh!

bors bot commented Apr 25, 2019

try

Build succeeded

Uh oh!

vext01 commented Apr 25, 2019

Uh oh!

bors bot commented Apr 26, 2019

try

Build failed

Uh oh!

vext01 commented Apr 26, 2019

Uh oh!

vext01 commented Apr 26, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

ltratt commented Apr 26, 2019

Uh oh!

vext01 commented Apr 26, 2019

Uh oh!

bors bot commented Apr 26, 2019

try

Build failed

Uh oh!

vext01 commented Apr 26, 2019

Uh oh!

bors bot commented Apr 26, 2019

try

Build succeeded

Uh oh!

vext01 commented Apr 26, 2019

Uh oh!

ltratt commented Apr 26, 2019

Uh oh!

vext01 commented Apr 26, 2019

Uh oh!

ltratt commented Apr 26, 2019

Uh oh!

bors bot commented Apr 27, 2019

try

Build succeeded

Uh oh!

vext01 commented Apr 27, 2019

Uh oh!

ltratt commented Apr 27, 2019

Uh oh!

vext01 commented Apr 28, 2019

Uh oh!

vext01 commented Apr 28, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

bors bot commented Apr 28, 2019

try

Build succeeded

Uh oh!

vext01 commented Apr 29, 2019

Uh oh!

ltratt commented Apr 29, 2019

Uh oh!

vext01 commented Apr 29, 2019

Uh oh!

vext01 commented Apr 26, 2019 •

edited

Loading

vext01 commented Apr 28, 2019 •

edited

Loading