Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Backports 1.6-rc1 #39160

Merged
merged 79 commits into from
Feb 1, 2021
Merged

Backports 1.6-rc1 #39160

merged 79 commits into from
Feb 1, 2021

Conversation

KristofferC
Copy link
Sponsor Member

@KristofferC KristofferC commented Jan 9, 2021

Need manual backport:

Non-merged PRs with backport label:

belamenso and others added 8 commits January 9, 2021 11:38
* added automatic keyword assignment support to @test macro

* added some tests for test macro using atol keyword

* x = a.x syntax support added

* Update stdlib/Test/src/Test.jl

Co-authored-by: Simeon Schaub <simeondavidschaub99@gmail.com>

Co-authored-by: Simeon Schaub <simeondavidschaub99@gmail.com>
(cherry picked from commit 6c42190)
(cherry picked from commit 0949eb2)
Instead of writing out to actual `$HOME`, create a temporary directory
and set that as `$HOME`/`$USERPROFILE`.

(cherry picked from commit 7a6f423)
Sometimes we want to widen the compilation signature, but then end up
with something which does not fit the original pattern. This then can
cause problems later, when we try to use the Method (from the cache),
but discover it does not actually match the call.

Fixes #38999

(cherry picked from commit 8937f7e)
Not sure why this worked before #37461, perhaps #9498?

(cherry picked from commit ca07546)
@KristofferC KristofferC added the release Release management and versioning. label Jan 9, 2021
@KristofferC
Copy link
Sponsor Member Author

@nanosoldier runtests(ALL, vs = ":release-1.5")

@KristofferC
Copy link
Sponsor Member Author

@nanosoldier runbenchmarks(ALL, vs = ":release-1.5")

@nanosoldier
Copy link
Collaborator

Something went wrong when running your job:

NanosoldierError: failed to run benchmarks against comparison commit: failed process: Process(`/home/nanosoldier/workdir/jl_AqTIX5/julia -e 'VERSION >= v"0.7.0-DEV.3656" && using Pkg; Pkg.update()'`, ProcessExited(1)) [1]

Logs and partial data can be found here
cc @christopher-dG

@nanosoldier
Copy link
Collaborator

Your package evaluation job has completed - possible new issues were detected. A full report can be found here. cc @maleadt

…39226)

Co-authored-by: Ian McInerney <Ian.S.McInerney@ieee.org>
staticfloat and others added 6 commits January 15, 2021 08:57
Co-authored-by: Dilum Aluthge <dilum@aluthge.com>
(cherry picked from commit 2bbd5fb)
GCC 11 has changed header dependencies again, so the build of LLVM
fails because <limits> wasn't being included when it should be. This
is already fixed upstream in LLVM 12.

(cherry picked from commit 4935775)
(cherry picked from commit 918ab6d)
omus and others added 20 commits January 26, 2021 08:14
mbedtls-cmake-findpy.patch lacks some whitespace that exists in
the target file, which causes patch application failure when the
patch tool is strict about whitespace. This teensy diff makes
the patch's whitespace strictly match that in the target file.

(cherry picked from commit 6416a3e)
This has actually been broken for a long time, so it only got noticed
when it caused a regression.

Fixes #39361

(cherry picked from commit fd8f97e)
Error introduced by #36733

Fixes #38435

(cherry picked from commit cdaf740)
Co-authored-by: Julian Samaroo <jpsamaroo@jpsamaroo.me>
(cherry picked from commit 0fe05cb)
If you had 2 aliases that both matched, we might print
```S{T} where T (alias for S{T} where T)```
which is clearly unnecessary.

(cherry picked from commit 48ad01d)
Always a bit more compact in this form, and somewhat easier to implement
too (thus keeping this consistent with the corrected typealias printing).

(cherry picked from commit ee816ef)
(cherry picked from commit 80ace52)
(cherry picked from commit 8631089)
With constant-propagation, inference (and Varargs runtime) is likely
better able to handle this version now (and it removes the n^2 behavior
definitions for semi-low argument counts).

Now we also need to force Vararg specialization up to 16 arguments
manually, so we do that explicitly (and slightly hack-y).

Fixes regression in #39175

(cherry picked from commit 5cd1e3e)
Co-authored-by: Dilum Aluthge <dilum@aluthge.com>
(cherry picked from commit f0e8ee6)
This was inserting a method with signature [Int, Int], rather than
Tuple{typeof(+), Int, Int}. Fix that and add a test for it so it doesn't
happen again.

(cherry picked from commit b9f8b8b)
@KristofferC
Copy link
Sponsor Member Author

@nanosoldier runbenchmarks(ALL, vs = ":release-1.5")

@KristofferC
Copy link
Sponsor Member Author

@nanosoldier runtests(ALL, vs = ":release-1.6")

@nanosoldier
Copy link
Collaborator

Your package evaluation job has completed - possible new issues were detected. A full report can be found here. cc @maleadt

@krrutkow
Copy link

@fingolfin The incoming CBindingGen v0.4.5 release supports LLVM 11.0.1, so this should resolve the CBindingGen, P4est, and PLCTag package tests.

@KristofferC
Copy link
Sponsor Member Author

I ran the benchmarks locally and didn't see anything completely crazy. There are some things that might be real regressions but with the server situation being what it is I think we have to go on with RC1 and then perhaps we can do a proper check for RC2.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
release Release management and versioning.
Projects
None yet
Development

Successfully merging this pull request may close these issues.