release-1.6: Backports for 1.6.6 #43735
Conversation
* Look for package name in `[extras]` When Preferences.jl set's a preferences in a non-top-level package, it adds that package to the `[extras]` entries in the project path. Package loading should have used thhose entries to map the module uuid to the key name in the Preferences.toml Fixes JuliaPackaging/Preferences.jl#24 Co-authored-by: Elliot Saba <staticfloat@gmail.com> (cherry picked from commit 8197c41)
The last entry sentinel for Win64 SEH is `~0L` not NULL. Apparently this doesn't cause issues on windows proper, but does crash wine. Arguably if Windows doesn't have issues then we should just fix this in wine, but since we control the source and nobody else ever seems to have run into this, let's just fix it and save the good Wine folks some headache. Fixes #43569. (cherry picked from commit 722f9d4)
Constructing the Regex touched in this commit can represent a significant fraction (e.g. half or better) of the runtime of the DateFormat method touched in this commit. To make this DateFormat method more efficient, let's lift that Regex construction out of that method body. (cherry picked from commit 7fb5bb8)
See comment in diff for explanation. This fixes test/stacktraces.jl on aarch64 macOS 12, and according to an OpenJDK issue where they ran into the same problem, https://git.openjdk.java.net/jdk/pull/6193, probably also x86_64 macOS 12. (cherry picked from commit 7f27dea)
To avoid the compiler potentially picking the wrong size depending on standards versions. (cherry picked from commit e8d1167)
|
@KristofferC, should #43787 be backported to |
|
Not sure, could be argued it is a bug fix but it isn't really clear to me at least. |
|
Thanks, I'd say it's a bug fix but since it's a long standing one with visual changes, maybe @JeffBezanson can make that call. |
We could also consider adding `defalg(v::AbstractArray{<:Number}) = DEFAULT_UNSTABLE`,
but it is unlikely that someone will want to do `Union` of `<:Number`
and something other than `Missing` that still would support comparison.
Relevant for JuliaStrings/InlineStrings.jl#21
(and other custom types that will want to add `defalg` support allowing
for `Union` with `Missing`).
(cherry picked from commit eb724e0)
(cherry picked from commit 9769024)
(cherry picked from commit c3235cd)
(cherry picked from commit 4c101a1)
(cherry picked from commit 2db86f2)
(cherry picked from commit c839221)
(cherry picked from commit 0c44d23)
Because we might not have synchronized the list again yet. (cherry picked from commit cac90b5)
We have 2 copies of this data, and so need to make sure we are pointing at the correct one for runtime. (cherry picked from commit 2f1f2f6)
vtjnash
force-pushed
the
backports-release-1.6
branch
from
f5ab4fd
to
7452018
Compare
* [CPUID] Rework how current ISA is determined * [CPUID] Add ISA entry for A64FX * [CPUID] Add ISA entry for Apple Silicon M1 * [CPUID] Simplify collection of full set of features for architecture * [CPUID] Remove AES from A64FX ISA, not all chips appear to have it (cherry picked from commit f45b6ad)
…et (#44262) We collect the relocations (i.e. the GOT slots that is used in the code) for each target in `tgt.relocs`. Needing a relocation, however, does not imply that the function is cloned for this target within the group (It does mean that at least one target in the group has it cloned). The previous version would miss the relocation in this case. This was triggerred with the following cloning situation caller: clone_1 callee: clone_1, clone_1.clone_3 Since caller.clone_1 may call either callee.clone_1 or callee.clone_1.clone_3 a relocation for callee will be used and is required to be initialized. In addition to target 1, target 2 (and in fact target 3) within group 1 will also use caller.clone_1. However, since callee isn't cloned for target 2 the previous version wouldn't have saved this slot in the relocation array. (cherry picked from commit 76fc067)
KristofferC
force-pushed
the
backports-release-1.6
branch
from
79388bf
to
cdfbec3
Compare
KristofferC
force-pushed
the
backports-release-1.6
branch
from
cdfbec3
to
c7daef8
Compare
|
@nanosoldier |
|
@nanosoldier |
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
|
Your package evaluation job has completed - possible new issues were detected. A full report can be found here. |
|
@nanosoldier |
|
Your package evaluation job has completed - possible new issues were detected. A full report can be found here. |
KristofferC
force-pushed
the
backports-release-1.6
branch
from
53e02a2
to
8c43f36
Compare
|
@nanosoldier |
|
Your package evaluation job has completed - possible new issues were detected. A full report can be found here. |
|
@nanosoldier |
|
Your package evaluation job has completed - possible new issues were detected. A full report can be found here. |
|
The one remaining Pkg eval "failure" looks like a bug fix which is for some reason is tested on the previous behavior. |
Backported PRs:
[extras]for Preferences #43361throw InexactErrorfromtryparsenext#44004generate_precompile_statements#44196@simdfor non 1 step CartesianPartition #42736Base.runtests: unsetJULIA_PROJECTandJULIA_LOAD_PATH#43575addenvfor environment entries with embedded=#44212open(::Function, ::Cmd)waits for termination #44078GlobalRefofCore.CodeInfoin@generated#43823Base.runtests: set a temporary Pkg depot path #42358setindex(::Int, ::Vector)#43678(::typeof(x))() = ...#43993svdandeigenofDiagonals #43856 <!-- Improve performance ofsvdandeigenofDiagonals --*_URLstdlib definitions into.versionfiles #42474git://protocol in the external stdlib URLs #42906make testallshould run all of the tests #42904git://protocol in Git URLs #42907Need manual backport:
jl_setjmpon windows #42267Contains multiple commits, manual intervention needed:
Base.runtests: rename the--force-netoption to--ci#43168Non-merged PRs with backport label:
transposeinLinearAlgebra.copy_transpose!to handle matmul of isbits array-elements with non-identity transpose #42715