Skip to main content

Browse the archive

https://github.com/JuliaLang/julia
19 April 2024, 06:28:37 UTC
sort by:
Revision Author Date Message Commit Date
b996c77 Kristoffer Carlsson 12 May 2023, 08:00:34 UTC Update stdlib/REPL/src/REPL.jl Co-authored-by: Jameson Nash <vtjnash@gmail.com> 12 May 2023, 08:00:34 UTC
25a11d9 Kristoffer Carlsson 12 May 2023, 08:00:28 UTC Update stdlib/REPL/src/REPL.jl Co-authored-by: Jameson Nash <vtjnash@gmail.com> 12 May 2023, 08:00:28 UTC
a4ac8f7 KristofferC 10 May 2023, 10:09:26 UTC prevent REPL from erroring in numbered mode in some situations 10 May 2023, 10:09:26 UTC
e204e20 Keno Fischer 10 May 2023, 02:57:28 UTC irinterp: Consider cfg information from discovered errors (#49692) If we infer a call to `Union{}`, we can terminate further abstract interpretation. However, this of course also means that we can make use of that information to refine the types of any phis that may have originated from the basic block containing the call that was refined to `Union{}`. Co-authored-by: Shuhei Kadowaki <aviatesk@gmail.com> 10 May 2023, 02:57:28 UTC
b9b8b38 Nathan Daly 09 May 2023, 20:47:27 UTC Move test of stackoverflow with close(::Channel) to stack_overflow.jl (#49702) Per @vtjnash's comment here: https://github.com/JuliaLang/julia/pull/49508/files/bca5ac79d04fb2a95f3b9a7b7448fe5b478f950b#r1186161259 > the handling of this error is quite bad and sometimes ends up breaking > the process (as I found out today trying to debug something completely > unrelated) This is a Tests-only PR 09 May 2023, 20:47:27 UTC
d10d90c Jakob Nybo Nissen 09 May 2023, 20:30:54 UTC Do not reallocate on sizehint to same size (#49703) The sizehint! code currently do not reallocate when sizehinting to a smaller size if 1/8th or fewer elements will be freed. However, the check rounds down, so an array of size e.g. 3 resized to size 3 will "free" 1/8th of the elements, namely zero elements, needlessly resizing. This commit changes the check to only reallocate if MORE than 1/8th of elements will be freed. 09 May 2023, 20:30:54 UTC
e7425d5 Ben Baumgold 09 May 2023, 19:07:02 UTC TimeType subtraction using promote (#49700) 09 May 2023, 19:07:02 UTC
61f6082 Keno Fischer 09 May 2023, 17:10:11 UTC Explicitly note that :consistent includes no-ub (#49693) * Explicitly note that :consistent includes no-ub We may split these effects in the future, but currently :consistent-cy requires the absence for undefined behavior for the function. There were a few questions about this, so explicitly document this in the help text. * Update base/expr.jl Co-authored-by: Sukera <11753998+Seelengrab@users.noreply.github.com> --------- Co-authored-by: Shuhei Kadowaki <40514306+aviatesk@users.noreply.github.com> Co-authored-by: Sukera <11753998+Seelengrab@users.noreply.github.com> 09 May 2023, 17:10:11 UTC
89b47cf Keno Fischer 09 May 2023, 17:09:15 UTC Tweak `donotdelete` effects (#49691) I don't think there's any reason not to mark this intrinsic as `:consistent`. The only real property we need from it at the julia level is that the optimizer does not delete it, which `!:effect_free` will take care of. 09 May 2023, 17:09:15 UTC
7ef78e4 Steve Kelly 09 May 2023, 06:39:44 UTC Add some more Effects system docs (#49652) * add docs for `Base.infer_effects` * add `show` decoder in `Effects` docs Co-authored-by: Shuhei Kadowaki <40514306+aviatesk@users.noreply.github.com> 09 May 2023, 06:39:44 UTC
493a16a Lilith Orion Hafner 09 May 2023, 04:59:56 UTC Micro-optimization in WeakKeyDict constructor (#49690) 09 May 2023, 04:59:56 UTC
33a2a9d DilumAluthgeBot 08 May 2023, 22:53:48 UTC 🤖 [master] Bump the Pkg stdlib from c8249204b to 94f668cee (#49687) 08 May 2023, 22:53:48 UTC
a2d9c00 Katharine Hyatt 08 May 2023, 22:50:38 UTC Some touchups to Task docs (#49682) 08 May 2023, 22:50:38 UTC
0a696a3 Lilith Orion Hafner 08 May 2023, 17:14:16 UTC Remove unnecessary `Base.` prefix (#49668) 08 May 2023, 17:14:16 UTC
8bcea42 Jameson Nash 08 May 2023, 17:13:35 UTC Merge pull request #49556 from JuliaLang/jn/small-typeof convert some typeof tags to small integers 08 May 2023, 17:13:35 UTC
2cad2a7 Jameson Nash 08 May 2023, 14:22:51 UTC avoid allocating a GC frame on the alt stack (#49650) This might be confusing for GC marking, which will not expect to see it running on the altstack. And the link to prev might get corrupted if there is any sort of error that happens during the processing of the signal that causes us to start handling another signal. 08 May 2023, 14:22:51 UTC
43079cf DilumAluthgeBot 08 May 2023, 13:00:30 UTC 🤖 [master] Bump the Pkg stdlib from 7ebf98b43 to c8249204b (#49678) 08 May 2023, 13:00:30 UTC
1ae70ec Jameson Nash 08 May 2023, 12:56:15 UTC test: fix TypeCompareError conflict (#49663) We have a slightly over-eager test in error.jl that requires all types ending in Error to subtype Exception, so this can cause a sporadic failure without this update. 08 May 2023, 12:56:15 UTC
4e782bf Elliot Saba 08 May 2023, 07:21:01 UTC Bump LBT to v5.8.0 (#49658) Includes: https://github.com/JuliaLinearAlgebra/libblastrampoline/commit/81316155d4838392e8462a92bcac3eebe9acd0c7 08 May 2023, 07:21:01 UTC
09adbe3 Sukera 07 May 2023, 20:52:02 UTC Remove mention of TwicePrecision from the docs (#49665) 07 May 2023, 20:52:02 UTC
bdb3ded Kristoffer Carlsson 07 May 2023, 20:11:54 UTC add reference to online tracy profile viewer (#49626) * Add docs for TRACY_CALLSTACKS and Tracy Web Viewer * add reference to online tracy profile viewer * Add `WITH_TRACY_CALLSTACKS` build option This change enables rudimentary call-stack support in Tracy. By default, call stack sampling is turned on for all zones along with random periodic sampling by the system. In the future, we'll probably want to take a more fine-grained approach to judiciously collect stack traces only where there are important, but for now gives users a flag to play with to uncover more information. --------- Co-authored-by: Cody Tapscott <topolarity@tapscott.me> 07 May 2023, 20:11:54 UTC
b69a417 Daniel Karrasch 06 May 2023, 14:08:50 UTC Reduce `mul!` methods related to "banded" matrices (#49666) 06 May 2023, 14:08:50 UTC
83225c5 Valentin Churavy 06 May 2023, 11:35:04 UTC Remove at_time from GC test (#49662) 06 May 2023, 11:35:04 UTC
4be81cd Jameson Nash 27 January 2023, 19:17:47 UTC change CodeInfo compression from Array to String String is simpler and smaller (since we do not yet have a Buffer{UInt8} type, which might be almost as simple) making it easier to manipulate. Combined with the previous changes, this lets us eliminate many more relocations from the image (previously every Array->data), which is quite a few megabytes of array header objects simply gone. MASTER: sysimg size breakdown: sys data: 79207516 isbits data: 58603452 symbols: 463610 tags list: 1078555 reloc list: 4491653 gvar list: 118848 fptr list: 214000 .text 11109176 .data 144184704 BEFORE (moving existing Strings): sysimg size breakdown: sys data: 75146604 isbits data: 62679532 symbols: 463411 tags list: 1015525 reloc list: 4481370 gvar list: 118688 fptr list: 213952 .text 11129192 .data 144126152 AFTER (making CodeInfo into Strings): sys data: 71348124 isbits data: 63789244 symbols: 463411 tags list: 933984 reloc list: 4398377 gvar list: 118688 fptr list: 213904 .text 11132968 .data 141272824 06 May 2023, 04:40:55 UTC
160d261 Jameson Nash 21 April 2023, 13:20:32 UTC move simple (isbits) objects into the constdata section 06 May 2023, 04:40:55 UTC
e26b63c Jameson Nash 05 May 2023, 18:35:48 UTC convert some typeof tags to small integers The value here is two-fold. One, we speed up sysimg load times slightly, since we can entirely skip relocations for these values and avoid faulting in some writable pages. This should let us (later) move more content into the ConstData section, such as String objects. Secondly, some type-manipulation native code becomes simpler, since it is based on small consecutive constants instead of requiring extra pointer loads to check for pointer identity. This is similar to the existing small-union optimization, but for type tags, and only for specific ones. This makes use of the fact that the zero page (about 4096 byes) cannot be allocated in the usual C memory model, which lets us define up to 255 of these special types, after taking into account the 4 gc bits. As a current implementation limitation for performance, these cannot be parametric types. Note there are probably more places in Base that can be updated to use `jl_typetagof` instead of `jl_typeof`, where we know if it is a type or tag. For example, this optimize most of builtins.c file, except for the `jl_object_id` function. 06 May 2023, 04:39:46 UTC
7cf01e5 Jameson Nash 06 May 2023, 04:37:38 UTC Merge pull request #49601 from JuliaLang/kp/channel-task-fix Reduce unnecessary task stickiness from Channel bind calls. Fix also a race condition between scheduling a Task and picking which thread it is sticky to. 06 May 2023, 04:37:38 UTC
633d1ae Alex Arslan 06 May 2023, 04:31:06 UTC Define a method for `hash(::Type, ::UInt)` (#49636) Currently, `hash(::Type, ::UInt)` uses `objectid`, which can have some odd behavior for types: in particular, subsequent identical type-valued variable definitions can have `objectid`s which differ from the first such definition. This has some bizarre downstream effects when e.g. using types as the values of a `Set` or the keys of a `Dict`. See issue 49620 for examples. There is an internal `type_hash` C function used for caching types but isn't exposed to Julia, as Jameson pointed out in the linked issue. This commit exposes it as `jl_type_hash` which is then used via `ccall` to define a method `hash(::Type, ::UInt)`. This method then fixes #49620. Note, however, that this does not affect the differing `objectid`s for otherwise identical types. 06 May 2023, 04:31:06 UTC
9bb0731 Jameson Nash 06 May 2023, 04:27:35 UTC fix some SCC cycle mistakes (#49654) 06 May 2023, 04:27:35 UTC
67dd21c Jameson Nash 06 May 2023, 04:27:06 UTC codegen: move union optimizations into emit_exactly_isa where emit_f_is expects them (#49655) 06 May 2023, 04:27:06 UTC
7ea902c Keno Fischer 06 May 2023, 03:31:57 UTC Extend NamedTuple special case to PartialStruct (#49641) A little later in the `precise_container_type` function, we have a special case for extracting the iteration type from a `NamedTuple`, treating it the same as `Tuple`. However, as a result of this special case we were accidentally discarding any constant information that may have been in a `PartialStruct`. If we didn't have this special case, `abstract_iteration` would have taken care of it properly. To fix this, simply add the same special case to the `PartialStruct` check at the top of the function. Co-authored-by: Shuhei Kadowaki <40514306+aviatesk@users.noreply.github.com> 06 May 2023, 03:31:57 UTC
6296f57 Cédric Belmant 06 May 2023, 03:31:16 UTC Test: improve robustness of stacktrace scrubbing (#49646) 06 May 2023, 03:31:16 UTC
9b7d88b Jameson Nash 05 May 2023, 19:45:47 UTC subtype: more conservative intersection of triangular variables (#49591) When a variable is used triangularly, we need to be careful not to propagate the lower bound implied by the other side to the upper bound implied by the invariant usage of the value--that is only legal when we are intersecting a variable that is used diagonally. Fix #49578 05 May 2023, 19:45:47 UTC
b8c347a Keno Fischer 05 May 2023, 17:50:43 UTC Thread lattice through tuple_tail_elem (#49643) 05 May 2023, 17:50:43 UTC
f4d9416 Keno Fischer 05 May 2023, 17:47:19 UTC `widenconst` before type check in SROA (#49642) * `widenconst` before type check in SROA These days, mutable structs can be `PartialStruct` if one of the fields is annotated as `const`, so we need to widenconst before this check. * Update test/compiler/irpasses.jl Co-authored-by: Shuhei Kadowaki <40514306+aviatesk@users.noreply.github.com> --------- Co-authored-by: Shuhei Kadowaki <40514306+aviatesk@users.noreply.github.com> 05 May 2023, 17:47:19 UTC
9201414 Ian Butterworth 05 May 2023, 17:36:15 UTC fix whitespace in test/ranges.jl (#49648) 05 May 2023, 17:36:15 UTC
6bcdd00 Fredrik Ekre 05 May 2023, 16:22:03 UTC Fix type comparison printing in `MethodError`s from convert. (#49645) This patch removes the erroneous `{}` for types without parameters when printing `MethodError`s from convert. Fixes e.g. the following (note `Float64{}`): ```julia julia> struct A{B, C} end julia> convert(A{Float64,1}, A{Float64,2}()) ERROR: MethodError: Cannot `convert` an object of type A{Float64{},2} to an object of type A{Float64{},1} ``` 05 May 2023, 16:22:03 UTC
6adea08 Daniel Karrasch 05 May 2023, 14:49:44 UTC Merge code that handles `Adjoint` and `Transpose` (#49521) Co-authored-by: Takafumi Arakaki <tkf@users.noreply.github.com> 05 May 2023, 14:49:44 UTC
0c83a23 Cédric Belmant 05 May 2023, 06:47:57 UTC Don't assume macro expansion info is present in stacktrace (#49633) 05 May 2023, 06:47:57 UTC
9c6cfc6 Kristoffer Carlsson 05 May 2023, 06:46:02 UTC silence compiler warning when not building with ITTAPI (#49630) 05 May 2023, 06:46:02 UTC
e70354d Aravindh Krishnamoorthy 05 May 2023, 06:40:01 UTC Fix for #48911: Bugfix for the generic left-divide (ldiv!) algorithm with pivoted QR decomposition and wide matrices (#49570) * Fix for #48911 1. Fix output permutation when A.P is not the identity matrix. 2. Fix the application of the elementary transformation (correct complex conjugation) to the solution as per LAPACK zlarzb https://netlib.org/lapack/explore-html/d5/ddd/zlarzb_8f_source.html 3. Align the final permutation with LAPACK's ZGELSY at https://netlib.org/lapack/explore-html/d8/d6e/zgelsy_8f_source.html 4. Fix a complex conjugate bug in the original solution for #48911, align with LAPACK zlarzb, https://netlib.org/lapack/explore-html/d5/ddd/zlarzb_8f_source.html 5. Improve permutation performance * Integration fix for #48911: Permutation is taken care by ldiv!(QRPivoted, ...) * Tests for #48911: LinearAlgebra/test/qr.jl * Issue #48911: Include original test case mentioned in the issue. --------- Co-authored-by: Daniel Karrasch <daniel.karrasch@posteo.de> 05 May 2023, 06:40:01 UTC
1729d40 Kristoffer Carlsson 05 May 2023, 06:39:27 UTC add another entry to the git ignore blame file (#49638) Co-authored-by: Dilum Aluthge <dilum@aluthge.com> 05 May 2023, 06:39:27 UTC
68cdf29 Cody Tapscott 05 May 2023, 06:34:13 UTC Fix infinite loop on macOS with TIMING enabled (#49632) `jl_pathname_for_handle` calls `jl_load_dynamic_library` internally on macOS, so naively calling it for our TIMING instrumentation causes an infinite loop. This is a simple fix that disables instrumentation when we are RTLD_NOLOAD'ing the library, which probably makes the trace information more meaningful as a set of _opened_ libraries anyway. 05 May 2023, 06:34:13 UTC
6b8ebb3 Jeff Bezanson 05 May 2023, 06:33:58 UTC annotate type of Task.queue field (#48420) This is apparently a common source of false-positives in JET. Co-authored-by: Kristoffer Carlsson <kcarlsson89@gmail.com> 05 May 2023, 06:33:58 UTC
16e10a1 DilumAluthgeBot 05 May 2023, 06:33:40 UTC 🤖 [master] Bump the LibCURL stdlib from fd8af64 to a65b64f (#49091) Co-authored-by: Dilum Aluthge <dilum@aluthge.com> Co-authored-by: Kristoffer Carlsson <kcarlsson89@gmail.com> 05 May 2023, 06:33:40 UTC
dffb9f0 Shuhei Kadowaki 05 May 2023, 06:33:20 UTC optimizer: allow inlining of constant `::TypeName` object (#49225) This allows us to inline `Ref.body.name` and such. This doesn't seem to have much impact on sysimage size: > master ``` ➜ ls -la ./usr/lib/julia total 906640 drwxr-xr-x 7 aviatesk staff 224 Apr 2 20:51 . drwxr-xr-x 276 aviatesk staff 8832 Apr 1 11:22 .. -rw-r--r-- 1 aviatesk staff 18344432 Apr 2 20:49 corecompiler.ji -rw-r--r-- 1 aviatesk staff 192771632 Apr 2 20:51 sys-o.a -rwxr-xr-x 1 aviatesk staff 161034648 Apr 2 20:51 sys.dylib drwxr-xr-x 3 aviatesk staff 96 Apr 1 11:26 sys.dylib.dSYM -rw-r--r-- 1 aviatesk staff 92039864 Apr 2 20:50 sys.ji ``` > this commit ``` ➜ ls -la ./usr/lib/julia total 906912 drwxr-xr-x 7 aviatesk staff 224 Apr 2 20:52 . drwxr-xr-x 276 aviatesk staff 8832 Apr 2 20:48 .. -rw-r--r-- 1 aviatesk staff 18329400 Apr 2 20:49 corecompiler.ji -rw-r--r-- 1 aviatesk staff 192790232 Apr 2 20:52 sys-o.a -rwxr-xr-x 1 aviatesk staff 161051240 Apr 2 20:52 sys.dylib drwxr-xr-x 3 aviatesk staff 96 Mar 30 13:18 sys.dylib.dSYM -rw-r--r-- 1 aviatesk staff 92160128 Apr 2 20:50 sys.ji ``` 05 May 2023, 06:33:20 UTC
08945aa James Wrigley 05 May 2023, 06:32:53 UTC Prettify printing of a small number of completions (#49284) Previously show_completions() tended to maximize the number of columns used, and if there were few enough completions to only use one or two lines the text would be a little difficult to read. Now it tries to minimize the number of columns, and there's a minimum number of completions needed before it'll start using multiple columns. Co-authored-by: Kristoffer Carlsson <kcarlsson89@gmail.com> 05 May 2023, 06:32:53 UTC
ef359e3 Benjamin Lorenz 05 May 2023, 06:32:19 UTC don't add cross-compile prefix for CC twice (#49575) * don't add cross-compile prefix twice make reexports changed environment variables to subprocesses Co-authored-by: Jameson Nash <vtjnash@gmail.com> 05 May 2023, 06:32:19 UTC
356c746 Kristoffer Carlsson 05 May 2023, 06:31:42 UTC Merge pull request #49488 from topolarity/smuggle-task-lineinfo Base.Threads: copy `__source__` LineInfo into closure for @spawn/@async 05 May 2023, 06:31:42 UTC
ee95843 Samuel Badr 05 May 2023, 00:16:10 UTC Allow showing of StepRangLen{T} with generic T (#49516) 05 May 2023, 00:16:10 UTC
ad939df Kristoffer Carlsson 05 May 2023, 00:13:55 UTC Add U+297A (`⥺`) and U+2977 (`⥷`) to binary operators (recreated) (#49623) 05 May 2023, 00:13:55 UTC
c0e12cd Shuhei Kadowaki 04 May 2023, 22:49:25 UTC inference: enable `:call` inference in irinterp (#49191) * inference: enable `:call` inference in irinterp Built on top of #48913, this commit enables `:call` inference in irinterp. In a case when some regression is detected, we can simply revert this commit rather than reverting the whole refactoring from #48913. * fix irinterp lattice Now `LimitedAccuracy` can appear in irinterp, so we should include `InferenceLattice` for `[typeinf|ipo]_lattice` for irinterp. 04 May 2023, 22:49:25 UTC
5032a1a pchintalapudi 04 May 2023, 21:43:33 UTC Initialize hoisted object allocations (#49584) 04 May 2023, 21:43:33 UTC
93ce36c Lilith Orion Hafner 04 May 2023, 17:38:32 UTC Merge pull request #48709 from JuliaLang/prohibit-tabs-in-whitespace-check Prohibit tabs in whitespace check 04 May 2023, 17:38:32 UTC
9dbfc05 K Pamnany 03 May 2023, 17:16:21 UTC Set task tid before scheduling task in `_wait2` 04 May 2023, 16:11:41 UTC
e95416f K Pamnany 01 May 2023, 19:29:37 UTC Close channel task doesn't need to be sticky 04 May 2023, 16:11:41 UTC
5e7d7c3 Jakob Nybo Nissen 04 May 2023, 13:15:34 UTC Remove last references to gc_num.since_sweep (#49628) Commit ebc6776 removed the since_sweep field, but left two references to it, which makes Julia fail to build. This commit removes those references. 04 May 2023, 13:15:34 UTC
1d361c0 Lilith Orion Hafner 04 May 2023, 11:37:28 UTC Remove redundant and unused _extrema in sorting (#48641) Found via coveralls line-by-line coverage. 04 May 2023, 11:37:28 UTC
ebc6776 Cody Tapscott 04 May 2023, 11:36:05 UTC (re-factor) gc: Remove redundant `since_sweep` field (#49195) This field is at all times 0 or identical to allocd, so this change removes it in preference of `gc_num.allocd`. Hopefully this makes the GC metrics a bit easier to interpret for newcomers. 04 May 2023, 11:36:05 UTC
724c93d Taksh Dhabalia 04 May 2023, 11:33:58 UTC Objects.md modified to include outdation warning (#49013) 04 May 2023, 11:33:58 UTC
bb31772 Jeff Bezanson 04 May 2023, 11:20:31 UTC make `map` on tuples accept different lengths (#49336) fix #49299, fix #42216 04 May 2023, 11:20:31 UTC
99e194f Cody Tapscott 24 April 2023, 03:57:29 UTC refactor: add `jl_timing_init_task` This provides a generic entrypoint to initialize any task-specific handles or metadata for profiling back-ends. This change also adds the Module name to the Task name. 04 May 2023, 11:17:06 UTC
0e8c0ea Cody Tapscott 23 April 2023, 13:28:24 UTC Base.Threads: copy `__source__` LineInfo into closure for @spawn/@async This is useful for profilers such as Tracy which need to give a distinguishing name to different tasks. Another useful thing would be to accept an optional `name` parameter in these macros, but that's left for future work. 04 May 2023, 11:15:09 UTC
daea3c3 Ed Hagen 04 May 2023, 10:57:49 UTC character literals are delimited with single quotes (#48998) Immediately inform the reader that character literals are delimited with single quotes, not double quotes. 04 May 2023, 10:57:49 UTC
705bfd3 kamesy 04 May 2023, 10:55:39 UTC Remove unneeded allocations in `ldiv!(::QRPivoted, ...)` (#49282) 04 May 2023, 10:55:39 UTC
c90aa77 Neven Sajko 04 May 2023, 10:48:48 UTC Base: twiceprecision: optimize mul12 (#49568) It is well known and obvious that the algorithm behind `Base.mul12` (sometimes known as "Fast2Mult") doesn't require a "Fast2Sum" (known in the Julia codebase as `Base.canonicalize2`) call at the end, so remove it. This follows from the fact that IEEE-754 floating-point multiplication is required to be well-rounded, so Fast2Sum can't change the result. Reference, for example, the beginning of https://doi.org/10.1145/3121432 by Joldes, Muller, Popescu. Furthermore, `Base.Math.two_mul` already exists, so use it as a kernel function. This required adding a general fallback method for `Base.Math.two_mul`, required, for example, for `BigFloat`. Also removed the `iszero` check that is now obviously not necessary. 04 May 2023, 10:48:48 UTC
ce2c3ae Neven Sajko 04 May 2023, 10:45:22 UTC Base: TwicePrecision: improve constructors (#49616) Correctness is improved for constructing `TwicePrecision{T}` from `TwicePrecision{S}`. Previously a call like `TwicePrecision{Float64}(::TwicePrecision{Float32})` would leave `hi` and `lo` unnormalized; while a call like `TwicePrecision{Float32}(::TwicePrecision{Float64})` would additionally introduce unnecessary truncation. Accuracy is improved for constructing `TwicePrecision` from types like `BigFloat` or `Rational`. Previously a call like `TwicePrecision{Float64}(::BigFloat)` would unconditionally leave `lo` as zero. Example code that tests the improvements (all Boolean values should evaluate to `true`): ```julia const TwicePrecision = Base.TwicePrecision const canonicalize2 = Base.canonicalize2 function twiceprecision_roundtrip_is_not_lossy( ::Type{S}, x::T, ) where {S<:Number, T<:Union{Number,TwicePrecision}} tw = TwicePrecision{S}(x) x == T(tw) end function twiceprecision_is_normalized(tw::Tw) where {Tw<:TwicePrecision} (hi, lo) = (tw.hi, tw.lo) normalized = Tw(canonicalize2(hi, lo)...) (abs(lo) ≤ abs(hi)) & (tw == normalized) end rand_twiceprecision(::Type{T}) where {T<:Number} = TwicePrecision{T}(rand(widen(T))) rand_twiceprecision_is_ok(::Type{T}) where {T<:Number} = !iszero(rand_twiceprecision(T).lo) setprecision(BigFloat, 70) # The mantissa needs to be just a bit larger than for `Float64` rand_twiceprecision_is_ok(Float32) rand_twiceprecision_is_ok(Float32) rand_twiceprecision_is_ok(Float32) rand_twiceprecision_is_ok(Float64) rand_twiceprecision_is_ok(Float64) rand_twiceprecision_is_ok(Float64) twiceprecision_roundtrip_is_not_lossy(Float64, rand(BigFloat)) twiceprecision_roundtrip_is_not_lossy(Float64, rand(BigFloat)) twiceprecision_roundtrip_is_not_lossy(Float64, rand(BigFloat)) twiceprecision_roundtrip_is_not_lossy(Float64, rand_twiceprecision(Float32)) twiceprecision_roundtrip_is_not_lossy(Float64, rand_twiceprecision(Float32)) twiceprecision_roundtrip_is_not_lossy(Float64, rand_twiceprecision(Float32)) twiceprecision_is_normalized(TwicePrecision{Float32}(rand_twiceprecision(Float64))) twiceprecision_is_normalized(TwicePrecision{Float32}(rand_twiceprecision(Float64))) twiceprecision_is_normalized(TwicePrecision{Float32}(rand_twiceprecision(Float64))) twiceprecision_is_normalized(TwicePrecision{Float64}(rand_twiceprecision(Float32))) twiceprecision_is_normalized(TwicePrecision{Float64}(rand_twiceprecision(Float32))) twiceprecision_is_normalized(TwicePrecision{Float64}(rand_twiceprecision(Float32))) ``` Updates #49589 04 May 2023, 10:45:22 UTC
b0db500 Erik Schnetter 04 May 2023, 10:37:52 UTC Hash values are not stable across different Julia invokations (#48408) 04 May 2023, 10:37:52 UTC
36f6826 Lilith Hafner 17 February 2023, 20:01:55 UTC Prohibit tabs in whitespace check 04 May 2023, 10:35:15 UTC
b03e8ab Lilith Hafner 17 February 2023, 20:02:14 UTC replace tabs with spaces and fix indentation 04 May 2023, 10:35:13 UTC
a632def Jeff Bezanson 04 May 2023, 10:26:58 UTC speed up variable lookups in lowering to help #48990 (#49213) Replace some more linear lookups with hash tables. 04 May 2023, 10:26:58 UTC
2e80c0d Zentrik 04 May 2023, 10:26:06 UTC Inline StepRange construction (#49270) This can improve performance when iterating over a range with a step as currently steprange_last is not inlined. About a 12ns improvement I don't think this should slow anything down as at worst you're making a call which is what you'd have to do without inlining. Also, it should be unlikely that branch is taken. This should allow for better automatic inlining. 04 May 2023, 10:26:06 UTC
9a6af18 Gabriel Wu 04 May 2023, 10:20:09 UTC Add fence to avoid JL_CALLABLE being parsed as MarkdownAST.Emph (#49360) 04 May 2023, 10:20:09 UTC
acb2d2d pbouffard 04 May 2023, 10:19:33 UTC Add missing documentation (#49377) 04 May 2023, 10:19:33 UTC
e6616fb Paul Berg 04 May 2023, 10:18:42 UTC Markdown: Reset underline style when wrapping text. (#49454) 04 May 2023, 10:18:42 UTC
37e5aed Neven Sajko 04 May 2023, 10:16:32 UTC doc: manual: command-line interface: fix cross-reference error (#49459) 04 May 2023, 10:16:32 UTC
2a1fa43 Lionel Zoubritzky 04 May 2023, 10:15:27 UTC Fix method show with kwarg and optional arguments (#49475) 04 May 2023, 10:15:27 UTC
59e47b4 José Manuel de Frutos 04 May 2023, 10:15:15 UTC adding docu to sum! explain target should not alias with the source (#49478) 04 May 2023, 10:15:15 UTC
e4622b4 Lilith Orion Hafner 04 May 2023, 10:12:11 UTC Remove unused definition of `_shorthash7` for wrong size UInt (#49498) 04 May 2023, 10:12:11 UTC
425389e Ian Butterworth 04 May 2023, 10:08:03 UTC Various precompile fixes (#49585) 04 May 2023, 10:08:03 UTC
4e70da2 woclass 04 May 2023, 09:42:05 UTC Update HISTORY.md for `DelimitedFiles` (#49582) xref: #48671 04 May 2023, 09:42:05 UTC
dfc74c3 Cody Tapscott 04 May 2023, 09:26:38 UTC Revise `jl_timing` prints for `dlopen` (#49560) By special request, this changes our `dlopen` logic to include the actually-opened filepath. For convenience, the base filename is listed first. This change also fixes a bug in `gnu_basename` where Windows filepaths were not being scanned for '\\' correctly. 04 May 2023, 09:26:38 UTC
827d34a Kristoffer Carlsson 03 May 2023, 21:43:42 UTC make `IRShow.method_name` inferrable (#49607) 03 May 2023, 21:43:42 UTC
4bb3a55 Steve Kelly 03 May 2023, 19:22:13 UTC elide precompilation in Profile.jl (#49613) This saves about 2ms on my system by checking the precompilation state. This also reorganizes the remaining precompilation calls in the module. 03 May 2023, 19:22:13 UTC
a35db92 Gabriel Baraldi 03 May 2023, 18:41:20 UTC Use isb for normal cpu pause on aarch64 (#49481) 03 May 2023, 18:41:20 UTC
252883c Kristoffer Carlsson 03 May 2023, 17:12:36 UTC correct comment about type of `method` in a `LineInfoNode` (#49608) 03 May 2023, 17:12:36 UTC
0fa8f4f Cody Tapscott 03 May 2023, 15:05:13 UTC Restore `modname == NULL` behavior for `jl_load_dynamic_library` (#49611) Within the Julia runtime, this usage has been superseded by the new `jl_find_library_by_addr`, but this is still used downstream. In particular, CBinding.jl ends up using this branch to get a handle to libjulia-internal (perhaps a bit accidentally, since the comments in CBinding.jl suggest it intends to get a handle to the exe) This restores the behavior to avoid downstream breakage. 03 May 2023, 15:05:13 UTC
5304baa Cédric Belmant 03 May 2023, 08:57:59 UTC Extend stacktrace for failed tests not directly under testset (#49451) 03 May 2023, 08:57:59 UTC
43c9202 Kristoffer Carlsson 03 May 2023, 07:22:33 UTC minimal changes to avoid Artifacts to get invalidated by some common definitions in ecosystem (#49603) 03 May 2023, 07:22:33 UTC
5039d8a Kristoffer Carlsson 03 May 2023, 07:21:58 UTC shield `Base.require` from invalidations when loading Symbolics.jl (#49604) 03 May 2023, 07:21:58 UTC
b66f63c Aravindh Krishnamoorthy 03 May 2023, 07:19:12 UTC Ensure length(ipiv)==n before calling LAPACK.getrs! to avoid segfaults (#49602) 03 May 2023, 07:19:12 UTC
404bb1f pchintalapudi 03 May 2023, 04:58:02 UTC Mark up optimization passes with remarks and debug prints (#49588) Add pass remarks and LLVM_DEBUG statements to alloc-opt, muladd, julia-licm, and simdloop passes 03 May 2023, 04:58:02 UTC
e9d678e Jeff Bezanson 02 May 2023, 20:25:53 UTC fix #48870, normalize U+210F to U+0127 (#49559) 02 May 2023, 20:25:53 UTC
0419054 Keno Fischer 02 May 2023, 19:00:51 UTC Allow toplevel-ish MethodInstances in va_process_argtypes (#49592) Usually toplevel methods don't go this way, but we'd like to broaden the use of "toplevel" MethodInstances slightly to other random chunks of IR we'd like to infer (as discussed in #49384), so give something sensible here if we do come this way. 02 May 2023, 19:00:51 UTC
6052e44 pchintalapudi 02 May 2023, 18:00:04 UTC Improve llvmpasses test quality (#49594) 02 May 2023, 18:00:04 UTC
82f76e8 Jameson Nash 02 May 2023, 15:40:47 UTC ml-matches: skip intersection by world-bounds (#49590) Should help speed up package loading some, by avoid complicated self-intersections, which would then be filtered out anyways from the returned results. It is not a frequent pattern, but it was observed in occasional cases, and is such an easy, obvious improvement. 02 May 2023, 15:40:47 UTC
a8474dc Jameson Nash 02 May 2023, 14:50:47 UTC Merge pull request #49597 from JuliaLang/tb/llvm_attributes_bis Improves #49551: Instead of copying attributes manually, just ensure we always call CloneFunctionInto, as it does useful work for declarations too. 02 May 2023, 14:50:47 UTC
b815bb1 Tim Besard 02 May 2023, 12:19:42 UTC Don't explicitly remap the personality function. LLVM does this already. 02 May 2023, 12:19:42 UTC
f77ad0a Tim Besard 02 May 2023, 12:16:57 UTC Ensure LLVM function attributes are preserved by always calling CloneFunctionInto. 02 May 2023, 12:18:04 UTC
1a973c7 Diogo Netto 02 May 2023, 03:46:47 UTC NFC: some cleanup in gc.c (#49577) 02 May 2023, 03:46:47 UTC
back to top