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

Inline Union Splitting Bugfix #42337

Closed
wants to merge 149 commits into from
Closed

Conversation

ianatol
Copy link
Member

@ianatol ianatol commented Sep 21, 2021

Fix for gridap/Gridap.jl#657 and #42264

This was due to a minor problem with ir_inline_unionsplit! and batch_inline! when dealing with union splitting while inlining.

IanButterworth and others added 30 commits June 10, 2021 09:50
1. `seed!(rng::Xoshiro, seed::NTuple{4,UInt64})` and
   `seed!(rng::TaskLocalRNG, seed::NTuple{4,UInt64})` were doing almost the same thing;
   factor out what was identical;
2. `seed!(rng::Union{TaskLocalRNG, Xoshiro})` was calling the above methods passing
   a random tuple generated by `RandomDevice`: in this case, we don't really need
   to hash the seed, as it's presumably random enough; so use the same algorithm
   as in `Xoshiro()` constructor, and let `Xoshiro()` re-use this new implementation
   of `seed!`.

(cherry picked from commit e35028a)
This now always uses `sparse(I, J, K, m, n)` for showing sparse matrices.
The three-arg `show` used for REPL printing should still behave
the same but has been refactored to share more code with the existing
methods in Base.

closes JuliaLang#41135

(cherry picked from commit f280766)
At least the newly added field `rcondv` may contain `undef` values, so
these can cause the same eigen factorizations not to compare equal. Not
100% sure whether the other fields should be ignored as well, but since
we didn't have them before, it seems at least consistent to ignore them
here.

fixes JuliaDiff/ChainRules.jl#422

(cherry picked from commit bc3ce48)
* Add support for at-overlay with parametric function definitions.

* Fix error message with at-MethodTable.

* Fix function signature to avoid compiler warning.

(cherry picked from commit 27ef9e0)
)

This adds a proof-of-concept demonstration of two new buildkite plugins:

* `cryptic` adds secrets management to privileged pipelines.  These
pipelines cannot be freely modified; their integrity is verified against
a signature maintained by committers with a secret key.  This allows
certain portions of the CI configuration (which are privileged and can
decrypt encrypted files/environment variables) to remain public, but
read-only to the general populace.

*  `sandbox` adds a generic sandboxing mechanism that allows CI steps to
be run within user-provided rootfs images.  We're using these here to
provide compiler toolchains for the `llvm-passes` CI steps, and the plan
is to eventually provide _all_ compiler toolchains through such rootfs
images.

(cherry picked from commit 61ae86d)
I want to use this in Cthulhu, which uses `IRCode` for optimized code
instead of `CodeInfo`.

(cherry picked from commit 083272b)
This was not an external method table, it is just a normal variable
binding. This was causing the precompile files to be corrupted, since we
use normal variables that look like this one at
https://github.com/JuliaLang/julia/blob/dc2befcffc7412768097c2a2a6819724a4745aeb/base/compiler/utilities.jl#L139-L140

Fixes JuliaLang#41156

(cherry picked from commit e3197af)
This includes libraries from GCC 11.

(cherry picked from commit d68e570)
DilumAluthge and others added 25 commits July 26, 2021 11:04
…cumentation, and reorganize/rename/move a bunch of configuration files (JuliaLang#41698)

(cherry picked from commit e6e79f7)
…stead of `signature` (JuliaLang#41706)

* CI (Buildkite): for the signed pipeline test, use `signature_file` instead of `signature`

* Base64-decode the signature before writing it to the signature file

(cherry picked from commit f7f46af)
(cherry picked from commit 5386fa6)
(cherry picked from commit 71308eb)
(cherry picked from commit 8fa8530)
Makes the worker struct threadsafe as well as flushing the GC messages

(cherry picked from commit 0c073cc)
This seems to be a fairly arbitrary case for throwing exceptions, when
the user might often use this value in arithmetic afterwards, which is
not checked. It leads to awkward complexity in the API however, where it
may be unclear which function to reach for, with no particular
justification for why a particular usage is "safe". And it inhibits
optimization and performance due to the additional checks and error
cases (and is not even entirely type-stable).

(cherry picked from commit 3eefaf0)
Allows creating these ranges for any type of integer lengths.

Also need to be careful about using additive identity instead of
multiplicative, and be even more consistent now about types in a
few places.

Fixes JuliaLang#41517

(cherry picked from commit 4f77aba)
This is not a particularly meaningful combination (LLVM dislikes it).

(cherry picked from commit f7fdcac)
…ted_ts" (JuliaLang#41722)

Also reverts "fixup to pull request JuliaLang#38405 (JuliaLang#41641)"

Seems to be causing hanging in CI testing.

This reverts commit 5af1cf0 and this
reverts commit 5a16805, reversing
changes made to 02807b2.

(cherry picked from commit 66f9b55)
…Range{<:Rational})` (JuliaLang#41479)

* Fix length(::AbstractUnitRange), faster length(::AbstractUnitRange{<:Rational})

(cherry picked from commit 1c951f7)
 release-1.7: Backports for 1.7-RC1/1.7-beta4
@ianatol ianatol requested a review from a team as a code owner September 21, 2021 23:18
@DilumAluthge
Copy link
Member

What's going on with the commits here?

@ianatol ianatol closed this Sep 21, 2021
@ianatol
Copy link
Member Author

ianatol commented Sep 21, 2021

Oops, fixing now

@ianatol ianatol deleted the inline-pr branch September 28, 2021 19:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.