b9f0f7e | Loong | 30 March 2023, 06:39:48 UTC | fix(REPL): using/import statements should on top-level, #49041 (#49098) (cherry picked from commit e5c2c51c083a33b4357178cb0009bc1e831c004e) | 30 March 2023, 07:32:27 UTC |
5d17f1c | Hans Würfel | 30 March 2023, 06:38:56 UTC | Handle `ProcessChain` in open(f, cmd, ..) (#49166) (cherry picked from commit 329f92c3fab91bc3816bac6ca3b92400c066c5bb) | 30 March 2023, 07:32:16 UTC |
c4ab115 | Valentin Churavy | 27 March 2023, 14:28:49 UTC | Add ITTAPI source for offline (#49022) * Add ITTAPI sources for offline build Co-authored-by: Milan Bouchet-Valat <nalimilan@club.fr> (cherry picked from commit 6b934f91d1b9c50c5b783b6aa36cf2648999461c) | 30 March 2023, 07:06:44 UTC |
5ba487d | Sheehan Olver | 16 March 2023, 10:53:06 UTC | Add *(::Diagonal, ::Diagonal, ::Diagonal) (#49005) (#49007) (cherry picked from commit c37fc2798d9bb7349ff8eadb350ae68cf17cee61) | 30 March 2023, 07:04:01 UTC |
3f04640 | Elliot Saba | 10 March 2023, 06:21:50 UTC | Merge pull request #48933 from JuliaLang/dk/cat_array_number Let Base handle concatenation of arrays and numbers (cherry picked from commit e6c84a1eae1f75cad1c62f62f2074c0f1bf124d1) | 30 March 2023, 07:03:15 UTC |
0a132df | N5N3 | 30 March 2023, 06:57:12 UTC | fix `obviously_disjoint` for Union Types (#49177) (cherry picked from commit 20920007ef3320bd26dbd701e3a259a791acf27e) | 30 March 2023, 06:57:48 UTC |
dd44b58 | Valentin Churavy | 28 March 2023, 18:23:08 UTC | Mark llvm::Any::TypeId as global in julia.expmap (#49124) The dynamic linker needs to unify `llvm::Any::TypeId` across DSOs. In our case `libjulia-codegen` and `libLLVM`. See https://github.com/llvm/llvm-project/blob/2bc4c3e920ee078ef2879b00c40440e0867f0b9e/llvm/include/llvm/ADT/Any.h#L30 Fixes: #49121 (cherry picked from commit d8fa3c85600518045e0ac939ab706cfe0164c7a0) | 30 March 2023, 06:35:58 UTC |
d187689 | Jameson Nash | 24 March 2023, 14:46:55 UTC | fix malloc-stack CI failure (#49082) We see Windows CI fail often here, and we don't actually need this memory--only the first thread should initialize it. (indeed, we were temporarily corrupting these data-structures in the process by reinitializing them, though that was not typically detectable.) From worker 7: Exception: EXCEPTION_ACCESS_VIOLATION at 0x7ffa26726e87 -- jl_makecontext at C:/workdir/src\win32_ucontext.c:67 From worker 7: in expression starting at C:\buildkite-agent\builds\win2k22-amdci6-6\julialang\julia-master\julia-ceffaee345\share\julia\test\ccall.jl:1066 From worker 7: jl_makecontext at C:/workdir/src\win32_ucontext.c:67 From worker 7: jl_install_thread_signal_handler at C:/workdir/src\signals-win.c:490 From worker 7: jl_init_root_task at C:/workdir/src\task.c:1568 From worker 7: ijl_adopt_thread at C:/workdir/src\threading.c:414 From worker 7: unknown function (ip: 000001d791a58969) From worker 7: uv__queue_work at /workspace/srcdir/libuv\src\threadpool.c:305 From worker 7: worker at /workspace/srcdir/libuv\src\threadpool.c:122 From worker 7: uv__thread_start at /workspace/srcdir/libuv\src/win\thread.c:111 From worker 7: beginthreadex at C:\Windows\System32\msvcrt.dll (unknown line) From worker 7: endthreadex at C:\Windows\System32\msvcrt.dll (unknown line) From worker 7: BaseThreadInitThunk at C:\Windows\System32\KERNEL32.DLL (unknown line) From worker 7: RtlUserThreadStart at C:\Windows\SYSTEM32\ntdll.dll (unknown line) From worker 7: Allocations: 352796158 (Pool: 352389694; Big: 406464); GC: 722 (cherry picked from commit 685da8f1fddd1503500a027a782f3bfa57371d66) | 30 March 2023, 06:35:58 UTC |
38f0e29 | Kristoffer Carlsson | 24 March 2023, 21:52:10 UTC | Merge pull request #49032 from N5N3/subtyping-backports Subtyping backports for 1.9 | 24 March 2023, 21:52:10 UTC |
a4cd8d2 | Kristoffer | 24 March 2023, 17:56:54 UTC | bump Pkg to latest v1.9 | 24 March 2023, 17:56:54 UTC |
0bee4f9 | Kristoffer Carlsson | 24 March 2023, 13:50:21 UTC | add back the exception to the error logging when loading an extension fails (#49075) (cherry picked from commit d7df15d1d001192871105d61f5116ddade070e8f) | 24 March 2023, 14:01:20 UTC |
b0cbe1a | woclass | 22 March 2023, 10:39:53 UTC | doc: remove outdata const (#49096) (cherry picked from commit 6d678fec0ad0c78e80f357aac1b1e99ff0ff47ca) | 24 March 2023, 14:01:19 UTC |
8912c53 | Jeff Bezanson | 20 March 2023, 23:45:51 UTC | fix multiq_check_empty to only look at the caller's pool (#49065) Co-authored-by: Jameson Nash <vtjnash@gmail.com> (cherry picked from commit 1323d2f784686815ad7f45cee83c578b91e8e7ba) | 24 March 2023, 14:01:18 UTC |
59c2003 | Tim Holy | 20 March 2023, 23:42:08 UTC | Ensure invalidated target gets logged (#49048) To interpret causes of invalidation, it's helpful to understand what signature got invalidated. #48954 inadvertently dropped this info from the logging stream; this commit restores it. (cherry picked from commit 99fce415079e8545a6b3246d8f998c8a50449ef7) | 24 March 2023, 14:01:17 UTC |
43c1ba2 | Jameson Nash | 16 March 2023, 16:37:26 UTC | fix problem with string_concatenation_hint_handler (#49016) Quiets an error on CI when running missing tests. Refs #45823 (cherry picked from commit 3919a89aaac87fab772721df9f7bbd9d647ce554) | 24 March 2023, 14:01:17 UTC |
cdb4d24 | Jameson Nash | 16 March 2023, 16:36:32 UTC | reuse existing typemap search for method overwrite detection (#48968) It does not really make sense to scan the tree twice, when we already will extra this info near completely from the intersection search. The cost was small, but non-negligible. (cherry picked from commit 38d9d837ad2d4b0438c36c60c17c0c0f8265bf8a) | 24 March 2023, 14:01:16 UTC |
aba1db7 | Kristoffer Carlsson | 16 March 2023, 14:49:59 UTC | delay loading of extensions as much as possible (#48674) (cherry picked from commit e3043a875d432ec4bd7c073874a654b55421438f) | 24 March 2023, 14:01:16 UTC |
09b7504 | Jameson Nash | 15 March 2023, 21:52:50 UTC | codegen: fix use-after-free of module name (#49011) (cherry picked from commit b5aa05706e42d476556d72e9556c5cbe679b7599) | 24 March 2023, 14:01:13 UTC |
20ff1ef | Jameson Nash | 10 March 2023, 23:04:38 UTC | gf: accelerate adding backedges back to old performance (#48966) Since it does not compute and branch on typeof, in my measurements, this implementation is up to 10x faster! (cherry picked from commit 369660fd8e787a999f8a67f6360abee0914dbea4) | 24 March 2023, 14:01:13 UTC |
64ea3e1 | Jameson Nash | 09 March 2023, 17:16:57 UTC | gf: add support for invalidating invoke edges (#48954) Apparently we never actually implemented support for invalidation detection on invoke edges, and furthermore, it had erased part of the support for regular edges. Generalize our code for detecting replacement of a method, to be used when computing replacement of an invoke edge. Fix #48802 (cherry picked from commit 386b1b67d7817a8125eaac5c6c2330dbd663bc5e) | 24 March 2023, 14:01:11 UTC |
e628534 | N5N3 | 08 March 2023, 21:30:47 UTC | Fix `Vararg` methods widening with unused type variable (#48953) Avoid possible nested `Vararg`. Fix #48950 (cherry picked from commit 4da03592d120ce5f8f7ca4ce524da28b2353d727) | 24 March 2023, 14:01:10 UTC |
99b8ff5 | Morten Piibeleht | 08 March 2023, 19:27:06 UTC | Add compat for set_active_project (#48947) (cherry picked from commit 3f7c0467c79d7da6b52a609c99e78127bb2b76a6) | 24 March 2023, 14:01:09 UTC |
18c3eb8 | Milan Bouchet-Valat | 08 March 2023, 16:10:33 UTC | Install 7z, lld and dsymutils to private libexec/julia directory (#48931) * Install 7z, lld, dsymutil to private libexec/julia directory Installing files directly into /usr/libexec pollutes the install tree and can create conflicts if other apps do the same. Instead, install them to `private_libdir`, which defaults to `$(libexecdir)/julia`. This is similar to what we do with `private_libdir`. (cherry picked from commit eb2e9687d0ac694d0aa25434b30396ee2cfa5cd3) | 24 March 2023, 14:01:09 UTC |
06058a5 | Udoh Jeremiah | 07 March 2023, 22:33:48 UTC | close code block for in `PartialQuickSort` docstring (#48937) (cherry picked from commit a3fe0a8c6715f44fcd85591c0a20a2cbaace722c) | 24 March 2023, 14:01:08 UTC |
5341aaf | Shuhei Kadowaki | 23 March 2023, 01:23:34 UTC | fix incorrect static parameter definedness check (#49097) We use the specialized signature of a method for accurate analysis of whether a static parameter is constrained or not. However, it turns out that we can only use it when it doesn't contain any free type variables (which it sometimes does, e.g., when the inference entry is given a signature with a free type variable). In such cases, we should use the method signature, which, by design, never contains any free type variables. | 23 March 2023, 01:52:07 UTC |
13b89aa | K Pamnany | 02 March 2023, 18:22:00 UTC | Fix typo in GC stat name (cherry picked from commit 4ea2d2facb9f7056ab460ee8ee86a1205dc2ee39) | 21 March 2023, 18:15:29 UTC |
af4e5b0 | K Pamnany | 17 February 2023, 02:19:42 UTC | Fix test for threadpool use (cherry picked from commit 55422d988c3a50ca0d272644e456c3959c62b097) | 17 March 2023, 01:18:32 UTC |
a8d8bba | K Pamnany | 17 February 2023, 01:49:31 UTC | Fix `enq_work` behavior when single-threaded If there's only one thread in the task's preferred thread pool, use that thread's work queue. (cherry picked from commit a3a92e8dd82146155c15cb7f5dc411fe29a6fc32) | 17 March 2023, 01:18:24 UTC |
e0b058b | K Pamnany | 06 March 2023, 17:44:15 UTC | Fix task thread pool assignment If a task is spawned with `:interactive` but there are no interactive threads, set the task's thread pool to `:default` so that we don't have to keep checking it in other places. (cherry picked from commit 38727be9d7a1d1ca3a2cd407470cf6374ed3c2c6) | 17 March 2023, 01:18:14 UTC |
64b48f5 | K Pamnany | 06 March 2023, 17:34:29 UTC | Update `nthreads(pool)` Now returns `threadpoolsize(pool)`. (cherry picked from commit 944dce960a8828a63eb9949001b0b7ae958139a0) | 17 March 2023, 01:18:00 UTC |
1dfe63d | K Pamnany | 06 March 2023, 17:31:00 UTC | Extend `Threads.threadpoolsize` Allow specifying which thread pool's size to retrieve. (cherry picked from commit dc68ffcae356554bd18cde6a147e777c7690cf12) | 17 March 2023, 01:17:51 UTC |
b860eca | N5N3 | 09 February 2023, 15:40:10 UTC | Avoid split `y::Union` when comparing with a forall var. If `y` has no free typevar. Then `x <: y` wont cause any env change. Thus there's no need to split `y`. | 17 March 2023, 01:08:55 UTC |
2a2068d | N5N3 | 02 February 2023, 03:41:26 UTC | Subtype: avoid false alarm caused by eager `forall_exists_subtype`. (#48441) * Avoid earsing `Runion` within nested `forall_exists_subtype` If `Runion.more != 0` we‘d better not erase the local `Runion` as we need it if the subtyping fails after. This commit replaces `forall_exists_subtype` with a local version. It first tries `forall_exists_subtype` and estimates the "problem scale". If subtyping fails and the scale looks small then it switches to the slow path. TODO: At present, the "problem scale" only counts the number of checked `Lunion`s. But perhaps we need a more accurate result (e.g. sum of `Runion.depth`) * Change the reversed subtyping into a local check. Make sure we don't forget the bound in `env`. (And we can fuse `local_forall_exists_subtype`) * Optimization for non-union invariant parameter. | 17 March 2023, 01:08:55 UTC |
2117018 | N5N3 | 01 February 2023, 14:13:22 UTC | Subtype: avoid repeated failing `equal_union`. (#48410) skip checked `equal_union` by set `state[i] = 1` | 17 March 2023, 01:06:44 UTC |
1107361 | N5N3 | 01 February 2023, 04:31:22 UTC | Subtype: Code clean for union state stack. (#48479) | 17 March 2023, 01:06:44 UTC |
ab9e44e | N5N3 | 17 March 2023, 01:03:43 UTC | Revert "Revert "fix #47658, state stack overflow on unions containing typevars (#48221)"" This reverts commit 7d335a04dcbfd9705ce353e3cdc7a24211ba5f70. | 17 March 2023, 01:03:43 UTC |
7d335a0 | KristofferC | 15 March 2023, 12:47:28 UTC | Revert "fix #47658, state stack overflow on unions containing typevars (#48221)" This reverts commit 25bad181eb184bce7d3a32a18699fe9f9f9a9325. | 15 March 2023, 12:47:28 UTC |
a82477b | KristofferC | 10 March 2023, 14:35:15 UTC | Revert "effects: taint `nonoverlayed` when bailing out inference (#48838)" This reverts commit 09757f7c83f91b8428f8229375f77810fa829683. | 10 March 2023, 14:35:15 UTC |
fdbba71 | Kristoffer Carlsson | 06 March 2023, 19:21:38 UTC | avoid some invalidations from untyped dict code in TOML print (#48908) (cherry picked from commit 015301ab59d7561ba42596e38574a74485c694cd) | 07 March 2023, 19:57:11 UTC |
e217a22 | Jameson Nash | 04 March 2023, 18:29:57 UTC | array: fix some atomic orderings (#48888) (cherry picked from commit 6124987df34ae5c5e018eb1e44ae619d30a13718) | 07 March 2023, 15:22:40 UTC |
278f86a | Jameson Nash | 04 March 2023, 18:29:00 UTC | ambiguity detection: more optimal code order (#48846) (cherry picked from commit ff5bd4b7e3bcda9681b8878c22881fdefdda1ebb) | 07 March 2023, 15:22:40 UTC |
6c05d8b | Jameson Nash | 01 March 2023, 21:17:45 UTC | generalize insert-backedges to insert in any world (#48785) Rather than a binary valid/not-valid, we track the exact later world that deleted it, relative to when we first assumed it may be valid. (cherry picked from commit 4c323819cfd165222608bb07614a272d043b845d) | 07 March 2023, 15:22:39 UTC |
e970518 | Keno Fischer | 09 December 2022, 08:54:34 UTC | abstract_apply: Don't drop effects of `iterate`'d calls (#47846) We were accidentally dropping the effects of calls from `iterate` calls performed during abstract_iteration. This allowed calls that were not actually eligible for (semi-)concrete evaluation to go through that path anyway. This could cause incorrect results (see test), though it was usually fine, since iterate call tend to not have side effects. It was noticed however in #47688, because it forced irinterp down a path that was not meant to be reachable (resulting in a TODO error message). For good measure, let's also address this todo (since it is reachable by external absint if they want), but the missing effect propagation was the more serious bug here. (cherry picked from commit 2a0d58a32f49573299e1f4cca04bac0f6e6c7717) | 07 March 2023, 15:19:31 UTC |
3b2e0d8 | Kristoffer Carlsson | 07 March 2023, 07:51:45 UTC | set VERSION to 1.9.0-rc1 (#48924) | 07 March 2023, 07:51:45 UTC |
94603a7 | Kristoffer Carlsson | 06 March 2023, 21:13:39 UTC | Merge pull request #48733 from JuliaLang/backports-release-1.9 Backports for 1.9.0-rc1 | 06 March 2023, 21:13:39 UTC |
9ced070 | Gabriel Baraldi | 03 March 2023, 17:32:51 UTC | Fix heapsize hint and use a line for max memory (#48747) * Fix heapsize hint and use a line so that large machines utilize more of their ram (cherry picked from commit 51db8af409f0a386dc26a1fa674f923db0dcc6a3) | 06 March 2023, 08:09:28 UTC |
e6249e7 | Lilith Orion Hafner | 06 March 2023, 03:35:44 UTC | Fix sorting missing values with offset indices (#48864) * Fix sorting missing values with offset indices * Add tests (cherry picked from commit 7eb961588f204e1866aea0bdee63f5dd91d42ac3) | 06 March 2023, 07:59:19 UTC |
9f3b747 | Prem Chintalapudi | 05 March 2023, 08:37:58 UTC | Initialize JIT bytes (cherry picked from commit 39e34ee6f890ce2923aa37ff0a6fb76e17f1c78b) | 06 March 2023, 07:59:19 UTC |
96b3fe3 | Tim Holy | 03 March 2023, 23:21:01 UTC | staticdata: check ci->relocatability before caching external specializations (#48882) Fixes #48837 (cherry picked from commit 89f30fff275a9abbc99c9513922e6cff62622cfa) | 06 March 2023, 07:59:19 UTC |
fd1108b | Tim Holy | 03 March 2023, 23:15:33 UTC | staticdata: set method.nroots_sysimg in `jl_write_values` (#48875) This ensures that the value gets set after all possible compilation has occurred. (cherry picked from commit 0ec704e2669f7190e8b3f346b1ff8bab85205472) | 06 March 2023, 07:59:19 UTC |
8024dd6 | Patrick Bouffard | 03 March 2023, 19:10:13 UTC | Add note to NEWS about keywords in tab completion (#48863) (cherry picked from commit 3850e88dedb511e6329518f00f8d671d9ab750b3) | 06 March 2023, 07:59:19 UTC |
be9df3f | Kristoffer Carlsson | 06 March 2023, 07:52:24 UTC | Merge pull request #48881 from JuliaLang/vc/backports-1.9 [release-1.9] Backport #48751 and #48832 | 06 March 2023, 07:52:24 UTC |
fcde569 | Valentin Churavy | 06 March 2023, 01:47:20 UTC | fixup! gf: cache cache_with_orig decision (#48833) | 06 March 2023, 01:47:20 UTC |
3850e88 | Patrick Bouffard | 03 March 2023, 19:10:13 UTC | Add note to NEWS about keywords in tab completion (#48863) | 03 March 2023, 19:10:13 UTC |
c1cb058 | Jameson Nash | 01 March 2023, 01:10:17 UTC | staticdata: make completeinfo memory-/gc-safe (#48832) There is actually almost no cases where `jl_alloc_svec_uninit` is safe, since if was safe, you would likely would prefer to use the `jl_svec` constructor instead. (cherry picked from commit 0608824bd9553514d71fa3a67d6c9fafecbe15be) | 03 March 2023, 17:54:24 UTC |
65a025f | Jameson Nash | 03 March 2023, 11:11:00 UTC | staticdata: make hookup of code instances correct (#48751) Previously we would double-account for many of these, leading to occasional chaos. Try to avoid serializing code that does not belong to this incremental compilation session package. Refs: #48723 (cherry picked from commit 81f366d7d731ecaaad34ad4dbacdfe53f4f06fa4) | 03 March 2023, 17:54:10 UTC |
ec5e702 | Kristoffer Carlsson | 03 March 2023, 13:47:40 UTC | fixup! fixup! effects: taint `nonoverlayed` when bailing out inference (#48838) | 03 March 2023, 13:47:40 UTC |
d57c4ea | KristofferC | 03 March 2023, 11:06:25 UTC | bump Pkg to latest 1.9 | 03 March 2023, 11:06:25 UTC |
b07fb40 | Jameson Nash | 01 March 2023, 01:09:15 UTC | gf: cache cache_with_orig decision (#48833) Memoizing this can save a lot of repeated effort for queries such as `@which eltype(String)`. Otherwise we repeatedly try to check if `eltype(::Type)` is a good way to cache that result, even though it never gets better the more we check it. (cherry picked from commit bdcd5e2c6d0fdea879a961f1a72d774b743cc846) | 03 March 2023, 10:29:38 UTC |
4710009 | William Moses | 10 February 2023, 20:02:21 UTC | Fix final gc lowering on dynamically sized allocation (#48620) (cherry picked from commit 7e57dc7bcf60eab24d89497d5ac4077c2603d120) | 03 March 2023, 10:29:38 UTC |
570251e | Jameson Nash | 25 February 2023, 22:31:33 UTC | enable re-using external code in pkgimages (#48723) * enable using external code in pkgimages This was unintentionally disabled (and incomplete) in the original PR for pkgimages. Co-authored-by: Valentin Churavy <v.churavy@gmail.com> (cherry picked from commit c136b7e787aee1dffefcea6afce51d0bdd6d5dd1) | 03 March 2023, 10:29:38 UTC |
44085f9 | pchintalapudi | 22 February 2023, 04:07:54 UTC | Atomically order specsigflags, specptr, and invoke (#47832) Co-authored-by: Jameson Nash <vtjnash@gmail.com> Co-authored-by: Dilum Aluthge <dilum@aluthge.com> (cherry picked from commit 6412a56223e38824ce6eff78bf34662637971e1c) | 03 March 2023, 10:29:38 UTC |
df3f686 | Valentin Churavy | 03 March 2023, 08:45:33 UTC | fixup! effects: taint `nonoverlayed` when bailing out inference (#48838) | 03 March 2023, 10:29:38 UTC |
3f040ca | Tim Holy | 03 March 2023, 04:42:18 UTC | Calculate a checksum for the system image (#48869) (cherry picked from commit 48b4caaf94c6dcfc9d28e9929ebb27d480b7b6cc) | 03 March 2023, 10:29:38 UTC |
b710bd2 | Mosè Giordano | 01 March 2023, 15:44:14 UTC | Always use `-Wl,--warn-unresolved-symbols` in MSAN builds (#48835) * Always use `-Wl,--warn-unresolved-symbols` in MSAN builds * Use `-Wl,--warn-unresolved-symbols` only on Linux and FreeBSD (cherry picked from commit a07f51298994299eaeeae3bb254f2079aa6b6e4a) | 03 March 2023, 10:29:38 UTC |
8ba214d | Mark Kittisopikul | 28 February 2023, 19:47:46 UTC | Update MbedTLS to 2.28.2 (#48806) * Update MbedTLS to 2.28.2 * Update deps/mbedtls.mk * Update mbedtls checksum (cherry picked from commit 127fb73ef5804e0c040aefadb47eb07763e36d9c) | 03 March 2023, 10:29:38 UTC |
2497f83 | Cody Tapscott | 28 February 2023, 18:14:40 UTC | Add libjulia-internal to ORCJIT symbol resolution (#48712) We need to make sure that libjulia-internal is loaded and placed first in the DynamicLibrary search order so that calls to runtime intrinsics are resolved to the correct library when multiple libjulia-*'s have been loaded (e.g. when we `ccall` into a PackageCompiler.jl-created shared library). (cherry picked from commit fd45943e993ea851edcc4b1f2cc7e002438be07e) | 03 March 2023, 10:29:38 UTC |
b5243af | Kristoffer Carlsson | 28 February 2023, 11:26:50 UTC | add a type assert to logging to prevent Base.require invalidation (#48810) (cherry picked from commit 85709517be408f32cf102234b906f3d918fafdf2) | 03 March 2023, 10:29:38 UTC |
a671f0c | Valentin Churavy | 26 February 2023, 12:01:21 UTC | Accept weakdeps as a source for UUIDS for Preferences (cherry picked from commit d6431a38913f0b2c44f1147e354fa9a773fabc56) | 03 March 2023, 10:29:38 UTC |
54bdee8 | Mosè Giordano | 25 February 2023, 11:20:42 UTC | [Make.inc] Define new variable `PATCHELF_SET_RPATH_ARG` (cherry picked from commit 219ef2184723ba0c2bb89af96d79c368c4a0f673) | 03 March 2023, 10:29:38 UTC |
e25fcc9 | Ian Butterworth | 21 February 2023, 13:02:28 UTC | Test: add space after test failure for visual clarity (#48526) (cherry picked from commit ba1e568966f82f4732f9a906ab186b02741eccb0) | 03 March 2023, 10:29:38 UTC |
2761b06 | Shuhei Kadowaki | 03 March 2023, 00:40:31 UTC | effects: handle va-method properly when refining `:inaccessiblememonly` (#48854) Previously the `:inaccessiblememonly` effect bit may be wrongly refined when analyzing va-method, e.g.: ```julia julia> callgetfield1(xs...) = getfield(getfield(xs, 1), 1) callgetfield1 (generic function with 1 method) julia> Base.infer_effects(callgetfield1, (Base.RefValue{Symbol},)) (+c,+e,!n,+t,+s,+m,+i) # inaccessiblememonly is wrongly refined here ``` This leads to wrong concrete evaluation of `callgetfield1` and thus may result in a problem like below: ```julia julia> const GLOBAL_XS = Ref(:julia); julia> global_getfield() = callgetfield1(GLOBAL_XS); julia> @test let Base.Experimental.@force_compile global_getfield() end === :julia Test Passed julia> GLOBAL_XS[] = :julia2; julia> @test let Base.Experimental.@force_compile global_getfield() end === :julia2 # this fails Test Failed at REPL[8]:1 Expression: let #= REPL[8]:2 =# Base.Experimental.@force_compile global_getfield() end === :julia2 Evaluated: julia === julia2 ``` This commit fixes it up. | 03 March 2023, 10:29:38 UTC |
09757f7 | Shuhei Kadowaki | 01 March 2023, 10:07:59 UTC | effects: taint `nonoverlayed` when bailing out inference (#48838) | 03 March 2023, 10:29:38 UTC |
a6a19e4 | Shuhei Kadowaki | 01 March 2023, 06:37:54 UTC | inference: fix the correctness of inference bail out interface (#48826) Since we allow overloading of the `bail_out_xxx` hooks, we need to make sure that we widen both type and effects to the top when bailing on inference regardless of the condition presumed by a hook. This commit particularly fixes the correctness of `bail_out_apply` (fixes #48807). I wanted to make a simplified test case for this, but it turns out to be a bit tricky since it relies on the details of multiple match analysis and the bail out logic. | 03 March 2023, 10:29:38 UTC |
d2de99d | Keno Fischer | 29 December 2022, 18:13:15 UTC | Stop renumbering statements in inference proper (#48022) I don't think there's any good reason to try to delete the statements here. The very next thing we do is to convert to IRCode which drops dead code anyway, so this just seems redundant. In addition, it complicates Cthulhu-like analysis, which now has to track an extra set of statement numbers. | 22 February 2023, 10:52:40 UTC |
1c22e77 | Elliot Saba | 07 February 2023, 23:34:49 UTC | [compiler] Teach type inference that `GotoIfNot` can throw (#48583) Previously, the effects system would ignore certain cases where `GotoIfNot` nodes would be capable of throwing; this resulted in simple examples such as the following being marked as `nothrow`: ``` julia> foo(x) = x > 0 ? x : 0 Base.infer_effects(foo, (Missing,)) (+c,+e,+n,+t,+s,+m,+i) ``` With this change, we correctly notice when a `GotoIfNot` node is given a non-`Bool` condition, annotate the basic block as possibly throwing, and further end type inference if the condition is provably non-`Bool`. | 22 February 2023, 09:42:37 UTC |
50b8c13 | Tim Besard | 21 February 2023, 16:05:24 UTC | Add world argument to jl_create_native. (#48746) (cherry picked from commit 09759069203b06024d87dc49071af7a1519c7559) | 21 February 2023, 16:16:24 UTC |
b5d9781 | Jameson Nash | 17 February 2023, 15:59:02 UTC | staticdata: encode link_id in tagged linkage (#48673) On 64-bit, we have enough space to encode (1) the tag, (2) the `depmods` index, and (3) the offset all in a single 64-bit pointer field. This means we don't need the external `link_id` arrays, which reduces the size of many pkgimages by ~5%. On 32-bit, we don't have enough bits to implement this strategy. However, most linkages seem to be against the sysimage, and so by giving that a separate tag we can achieve similar compression because the `link_id` lists will be much shorter. Co-authored-by: Tim Holy <tim.holy@gmail.com> (cherry picked from commit 8e3e97074e34f86e7c796536a46fcdeca53a52dc) | 21 February 2023, 16:16:24 UTC |
956f54b | Mosè Giordano | 20 February 2023, 14:47:04 UTC | [OpenBLAS_jll] Upgrade to new build optimised for PowerPC (#48689) * [OpenBLAS_jll] Upgrade to new build optimised for PowerPC This only difference compared to previous build is that this one enables use of dynamic architecture detection also for the PowerPC architecture. --------- Co-authored-by: Valentin Churavy <v.churavy@gmail.com> (cherry picked from commit aacfcf0afcd325c2ad2e59fc16497405a07b4f85) | 21 February 2023, 16:16:24 UTC |
624c665 | Kristoffer Carlsson | 21 February 2023, 12:52:31 UTC | put back DelimitedFiles as a non-sysimage "stdlib" (#48671) (cherry picked from commit 56667bb47af73c9f6d574dd96ef946983854e1ee) | 21 February 2023, 16:16:24 UTC |
0cb0bb4 | Gabriel Baraldi | 11 February 2023, 11:07:34 UTC | Bump LLD to get dsymutil and use it for pkgimgs (#48628) * Run dsymutil on pkgimgs for apple platforms + LLD bump to get dsymutil from the jll (cherry picked from commit d8c225007498a68d1a1c9f3229d0dbc11b98f0cf) | 21 February 2023, 16:16:24 UTC |
af0e4c1 | Ian Butterworth | 10 February 2023, 12:18:50 UTC | force singlethreading during precompilation (#48592) (cherry picked from commit c0d2c574d44231dba39f90d6a6b27448f71763ca) | 21 February 2023, 16:16:24 UTC |
86b9df2 | Shuhei Kadowaki | 08 February 2023, 09:40:14 UTC | effects: taint `:nothrow` effect on unknown `:static_parameter` (#46791) * effects: taint `:nothrow` effect on unknown `:static_parameter` (conservatively) With this commit, we taint `:nothrow` effect property correctly on access to unknown `:static_parameter`, e.g.: ```julia unknown_sparam_throw(::Union{Nothing, Type{T}}) where T = (T; nothing) @test Core.Compiler.is_nothrow(Base.infer_effects(unknown_sparam_throw, ((Type{Int},)))) @test !Core.Compiler.is_nothrow(Base.infer_effects(unknown_sparam_throw, ((Nothing,)))) ``` This commit implements a very conservative analysis, and thus there is a room for improvement still, e.g.: ```julia unknown_sparam_nothrow(x::Ref{T}) where {T} = (T; nothing) @test_broken Core.Compiler.is_nothrow(Base.infer_effects(unknown_sparam_nothrow, (Ref,))) ``` * inference: improve `:nothrow` modeling for `:static_parameter` (#46820) * Fix test with free type params * Test: Ignore ::Type{T} in detect_unbounded These are only technically unbounded because of the existence of ill-formed types. However, this function is supposed to be an API sanity check and ordinary users should never have ill-formed types, so for the purpose we want here, allow unboundedness in Type{T}. --------- Co-authored-by: Keno Fischer <keno@juliacomputing.com> Co-authored-by: Elliot Saba <staticfloat@gmail.com> (cherry picked from commit b5d17ea2b3d68f41ffca57a87527a337abc8b74d) | 21 February 2023, 16:16:24 UTC |
5453936 | Kristoffer Carlsson | 20 February 2023, 08:49:16 UTC | only load extension triggers from the evnironment where the parent package is loaded (#48703) (cherry picked from commit 200c96272a22a5c9d2434a0d091658cb0e7d0fe4) | 21 February 2023, 16:16:24 UTC |
a66d092 | Jameson Nash | 31 January 2023, 19:01:14 UTC | update loading test to best-practices (#48429) Many of these test were hiding output or using semi-deprecated functions or failing to cleanup output or setting the wrong variables. (cherry picked from commit b48104d5f29fcfd477c72b1baa22145297c08bb6) | 21 February 2023, 16:16:24 UTC |
1121d25 | Gabriel Baraldi | 10 February 2023, 19:20:27 UTC | Allow for use 60% of constrained/total system memory (#48614) Remove the high watermark logic, because it doesn't really make sense, and allow for use of 60% of system memory before aggressive GC kicks in. Should fix #48473 (cherry picked from commit 500f561f728d0be3d5f6916f274cdf14387cf375) | 21 February 2023, 16:16:24 UTC |
6545919 | Fons van der Plas | 13 February 2023, 14:26:31 UTC | julia_cmd() docstring: include pkgimages (#48392) (cherry picked from commit 4a75129efe00ebe1880e3ab86f6901524ca063bc) | 21 February 2023, 16:16:24 UTC |
7734b34 | Kristoffer | 10 February 2023, 14:45:47 UTC | fix printing of parents to extensions in `@time_imports` (cherry picked from commit 6bd658eaf4f3b7495f17c5f28c2d2f1c5c0e3894) | 21 February 2023, 16:16:24 UTC |
2324310 | Kristoffer | 08 February 2023, 10:23:40 UTC | make `insert_extension_triggers` idempotent (cherry picked from commit 2a0b9dd50657beddb7a28d9ecf66b86367ca7084) | 21 February 2023, 16:16:24 UTC |
84e3545 | Keno Fischer | 15 February 2023, 00:51:12 UTC | Put back getfield :boundscheck hack (#48677) We used to have this hack before #48246, but I removed it because I had hoped we don't need. Unfortunately, we weren't able to infer consistency of ``` @inbounds (1,2)[1] ``` With this hack, constprop is able to independently prove inbounded-ness, overriding the usual consistency taintaing that happens for inbounds. (cherry picked from commit 113c2f3c95bf2ae67ac1e8214a17b7d9404dbbfa) | 21 February 2023, 16:16:22 UTC |
5e39fc6 | Shuhei Kadowaki | 16 November 2022, 00:27:26 UTC | lattice: fix minor lattice issues (#47570) I found some lattice issues when implementing `MustAlias` under the new extendable lattice system in another PR. (cherry picked from commit ee0f3fc334a8377da2d2b18e69c538eabc5aec13) | 21 February 2023, 12:23:04 UTC |
f7c4f59 | Jameson Nash | 20 February 2023, 12:08:22 UTC | restore kwcall_mt optimizations | 20 February 2023, 12:08:22 UTC |
25bad18 | Jeff Bezanson | 24 January 2023, 14:05:18 UTC | fix #47658, state stack overflow on unions containing typevars (#48221) (cherry picked from commit 596ce6542624e9b8c3782b19936e2226f307e118) | 20 February 2023, 12:04:07 UTC |
b3d42d1 | Wesley Jenkins | 14 February 2023, 16:35:11 UTC | Fix Base.libllvm_path and jl_get_libllvm don't support non-ASCII characters in path on Windows (#45126) (#45127) * Fix jl_get_libllvm_impl to support non-ASCII characters * Fix jl_get_libllvm_impl to support non-ASCII characters, fix whitespace * Fix jl_get_libllvm_impl to support non-ASCII characters, fix null and buffer (cherry picked from commit 6976bacd66e606865f656312bf6d991f3a6e4507) | 20 February 2023, 11:55:56 UTC |
d43d055 | Valentin Churavy | 13 February 2023, 22:06:17 UTC | Disable frame-pointer-optimiation on Linux (#48660) (cherry picked from commit 9c4a40c7b9e4e1ebb6bfb70fe66fc6fb8e4f4b6d) | 20 February 2023, 11:55:55 UTC |
6213bb8 | Fredrik Ekre | 13 February 2023, 16:23:04 UTC | [REPL] Meta-e: Don't execute the edited code when exiting editor (#46153) (cherry picked from commit aaab409957e32110cbc80318fee7215e616a555c) | 20 February 2023, 11:55:54 UTC |
ceacba1 | Valentin Churavy | 12 February 2023, 23:02:37 UTC | Update checksums for llvm 14.0.6+2 (#48659) (cherry picked from commit ecfaef3394be9d22459c855969dcc655fc7e6e91) | 20 February 2023, 11:55:52 UTC |
d0a7175 | Valentin Churavy | 08 February 2023, 13:12:23 UTC | Update LLVM build to 14.0.6+2 (#48544) (cherry picked from commit 43c6f7568bed8577c3ff93a73c970451859fae9d) | 20 February 2023, 11:55:48 UTC |
1360b3a | Elliot Saba | 07 February 2023, 02:56:36 UTC | Restrict `path` to be an `AbstractString` in `lbt_forward()` Some users tried to pass a `dlopen()` handle into `lbt_forward()` which sadly works just fine, despite `ccall()` declaring its first argument as being a `Cstring`. I guess it's trivial to convert a `Ptr{Cvoid}` into a `Cstring`, so this just goes through. To protect against this, restrict `path` to be an `AbstractString`. (cherry picked from commit 849203173ffaa2f73a4f052e8a832e2c43dcf3b9) | 20 February 2023, 11:55:46 UTC |
b75ddb7 | Kristoffer Carlsson | 07 February 2023, 21:53:19 UTC | set VERSION to 1.9.0-beta4 (#48574) | 07 February 2023, 21:53:19 UTC |