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
f7de6d2 Michael Abbott 07 May 2020, 14:36:32 UTC Make kron(a,b) preserve Adjoint-ness (#35755) * kron(adjoint vector) * tests * docstring for kron * the implementation * better idea Co-authored-by: Michael Abbott <me@escbook> 07 May 2020, 14:36:32 UTC
a47cf00 Kristoffer Carlsson 07 May 2020, 11:54:28 UTC add better implementations for generic findprev findnext for AbstractString (#35742) * add better implementations for generic findprev findnext for AbstractString 07 May 2020, 11:54:28 UTC
d4d9303 Alex Arslan 07 May 2020, 00:56:34 UTC Document and export Base.peek (#28811) 07 May 2020, 00:56:34 UTC
52c9e3d Harmen Stoppels 06 May 2020, 21:07:07 UTC Bump OpenBLAS 0.3.9 (#35113) * Bump OpenBLAS to 0.3.9 (builder still at 0.3.7) * Append USE_BLAS_FFLAGS to OPENBLAS_FFLAGS * Use the same Windows patch for OpenBLAS as used in Yggdrasil * Add OpenBLAS -ofast power patch * Bump lapack to 3.9.0 to match OpenBLAS version * Bump BB OpenBLAS to 0.3.9-0 * Bump OpenBLAS BB to version 2 * Add aarch64 patch for OpenBLAS 0.3.9 * Add OpenBLAS BB v0.3.9-3 * Actually bump the BB version to 3 * Update Versions.make * Add 0.3.9-4 checksums * Add the new patch to base as well Co-authored-by: Viral B. Shah <ViralBShah@users.noreply.github.com> 06 May 2020, 21:07:07 UTC
bb1a6ce Jeff Bezanson 06 May 2020, 20:57:27 UTC fix #35771, `global a` as repl input (#35775) 06 May 2020, 20:57:27 UTC
7abd5ad Stefan Karpinski 06 May 2020, 20:56:57 UTC PCRE: use correct type for error constants (Cint), audit C calls (#35773) 06 May 2020, 20:56:57 UTC
a4641c8 Jameson Nash 06 May 2020, 19:12:05 UTC threading: improve safety posture of modifying global bindings (#35761) Ensures that we are holding some lock (a new one) while mutating the internal global `jl_current_modules` table. While reading these binding values is not memory-safe to do simultaneously (it may invent pointers from thin-air and segfault), this commit ensures that a data conflict during the writes to them will not corrupt the values written into memory. 06 May 2020, 19:12:05 UTC
c3d6a46 Jeff Bezanson 06 May 2020, 16:24:31 UTC Partly revert "Reduce consequences of method invalidation for Dict and Set" (#35762) This partly reverts commit 1da1b8d35b060a0cebe94d6413a5a3c184fd08de. 06 May 2020, 16:24:31 UTC
99dbd2d Shan Sikdar 06 May 2020, 14:57:35 UTC add <: to add-dots (#35085) 06 May 2020, 14:57:35 UTC
816df51 Tim Besard 06 May 2020, 14:54:11 UTC Make llvmcall errors dynamic. (#35767) 06 May 2020, 14:54:11 UTC
097f0b0 Kristoffer Carlsson 06 May 2020, 12:56:45 UTC Revert "manually destruct and reconstruct subarray when throwing boundserror (#29867)" (#35748) This reverts commit 13694f4210f42a8e7a0d282296422b1d0a07b0f4. 06 May 2020, 12:56:45 UTC
60b46a3 Fredrik Ekre 06 May 2020, 12:15:00 UTC Update Documenter to 0.24.11, fixes #35724. (#35765) 06 May 2020, 12:15:00 UTC
5c5ab49 Fredrik Ekre 06 May 2020, 07:38:25 UTC Bump Pkg to 1.5.0. (#35746) 06 May 2020, 07:38:25 UTC
b472f88 Jeff Bezanson 06 May 2020, 01:04:41 UTC NEWS edits for 1.5 (#35754) 06 May 2020, 01:04:41 UTC
34d95d0 Jeff Bezanson 06 May 2020, 01:01:32 UTC Revert "add a test for big tuple conversion (#35736)" (#35756) This reverts commit b72e191ea9ad98aee0e8f8fa4688491906316f50. 06 May 2020, 01:01:32 UTC
dd9d1f4 Harry Scholes 06 May 2020, 01:01:13 UTC Update @deprecate docs (#35749) 06 May 2020, 01:01:13 UTC
98d1c31 Jeff Bezanson 05 May 2020, 22:17:30 UTC fix #35740, mistake in sysimg generation with `ccallable` (#35752) 05 May 2020, 22:17:30 UTC
9c2492e Tim Besard 05 May 2020, 21:17:06 UTC Fix and test PCRE error message getter. (#35728) 05 May 2020, 21:17:06 UTC
6b53e02 Jameson Nash 05 May 2020, 21:08:54 UTC clarify warning for constant redefinition (#35119) Before it sounded like a linter message, now it should be clearer that this warning means Julia may now be partially dysfunctional. 05 May 2020, 21:08:54 UTC
84ae4a4 Keno Fischer 05 May 2020, 20:01:42 UTC Revert "Update PCRE2 version to 10.34, fixes #35322, fixes #35459. (#35607)" (#35750) This reverts commit fbab705782742505762e3bb6d03e68f2b2ebe574. 05 May 2020, 20:01:42 UTC
3da2c52 Jeff Bezanson 05 May 2020, 18:39:24 UTC fix #35708, GC seeing uninitialized struct in `jl_new_struct*` (#35737) 05 May 2020, 18:39:24 UTC
5a06515 Rami 05 May 2020, 18:37:30 UTC Inference of circcopy and circshift (#35398) fixes #35396, fixes #35248 05 May 2020, 18:37:30 UTC
39134c5 Tim Holy 05 May 2020, 17:33:20 UTC Revert "Support `⊙` and `⊗` as elementwise- and tensor-product operators (#35150)" (#35744) This reverts commit f36036cc628df2cc12e0933b099d3ac935423f21. This was viewed by some as premature, and there seems to be willingness to develop it in a package first with an eye towards future inclusion. 05 May 2020, 17:33:20 UTC
b7cacf7 Matt Bauman 05 May 2020, 17:13:59 UTC Strided array doc: Remove reference to getindex (#26150) * Strided array doc: Remove reference to getindex As noticed by Jutho, we don't assume that `getindex(A, i)` is the same as accessing the `i`th element as computed by the strides. * Rewrite with motivating examples... instead of consise theoretical arithmetic. * Emphasize the distinction from linear indexing Co-authored-by: Tim Holy <tim.holy@gmail.com> 05 May 2020, 17:13:59 UTC
2253843 Valentin Churavy 05 May 2020, 15:41:14 UTC [GPUCodegen] Re-infer deleted methods if needed (#35655) 05 May 2020, 15:41:14 UTC
b72e191 Kristoffer Carlsson 05 May 2020, 13:43:37 UTC add a test for big tuple conversion (#35736) 05 May 2020, 13:43:37 UTC
33823ec cossio 05 May 2020, 08:50:59 UTC Fix typo in rand doc string (#35734) 05 May 2020, 08:50:59 UTC
7b7f63b Elliot Saba 05 May 2020, 08:00:32 UTC Merge pull request #35739 from JuliaLang/revert-34101-inplace-copy-dict 05 May 2020, 08:00:32 UTC
fbc4c31 Keno Fischer 05 May 2020, 06:19:11 UTC Revert "Optimize copy!(::Dict, ::Dict) (#34101)" This reverts commit cf123a8e402b2eb83e3d532b42f07415ee0803d7. 05 May 2020, 06:19:11 UTC
d07fadf Jeff Bezanson 05 May 2020, 05:59:36 UTC Merge pull request #35686 from JuliaLang/jb/threadstatus upgrade threads from experimental to stable-with-caveats 05 May 2020, 05:59:36 UTC
d68243c Paul Söderlind 05 May 2020, 04:15:32 UTC varargs/keywords doc (#35725) 05 May 2020, 04:15:32 UTC
cf123a8 Takafumi Arakaki 05 May 2020, 04:13:56 UTC Optimize copy!(::Dict, ::Dict) (#34101) Co-authored-by: Jameson Nash <vtjnash@gmail.com> 05 May 2020, 04:13:56 UTC
c879d1a Jameson Nash 03 May 2020, 17:58:10 UTC add documentation on finalizer interactions with threads 05 May 2020, 03:17:31 UTC
d541c2d Jeff Bezanson 01 May 2020, 17:54:00 UTC upgrade threads from experimental to stable-with-caveats more threading doc edits 05 May 2020, 03:16:31 UTC
faafb78 Jeff Bezanson 05 May 2020, 01:05:08 UTC add `threads static` to aid thread API evolution (#35646) 05 May 2020, 01:05:08 UTC
5e25ed9 Jameson Nash 04 May 2020, 23:55:57 UTC Merge pull request #35595 from JuliaLang/jn/gcchecker-atomics [GCChecker] teach it how to deal with more complex expressions 04 May 2020, 23:55:57 UTC
ab5853f Jeff Bezanson 04 May 2020, 23:25:47 UTC make at-sync thread safe (#35641) fixes #34666 04 May 2020, 23:25:47 UTC
7ba7daf Jeff Bezanson 04 May 2020, 23:24:13 UTC make `ccallable` work with system images and precompilation (#35574) fixes #35014 04 May 2020, 23:24:13 UTC
84d7e67 Tim Holy 04 May 2020, 21:00:13 UTC Merge pull request #35714 from JuliaLang/teh/less_invalidation A grab bag of latency reductions 04 May 2020, 21:00:13 UTC
a6e27b3 Sebastian Stock 04 May 2020, 20:05:21 UTC Make empty ranges compare equal (#32348) 04 May 2020, 20:05:21 UTC
6418be9 Elliot Saba 04 May 2020, 19:16:00 UTC Merge pull request #35521 from JuliaLang/sf/udp_multicast 04 May 2020, 19:16:00 UTC
44f3bb3 Matt Bauman 04 May 2020, 18:32:13 UTC Make transposes of StridedArrays strided (#29135) * Make transposes of StridedArrays strided but only in cases where the transpose is actually stored in memory. 04 May 2020, 18:32:13 UTC
39d7a18 Elliot Saba 04 May 2020, 18:19:02 UTC Merge pull request #35626 from JuliaLang/sf/makefile_caching 04 May 2020, 18:19:02 UTC
de42716 Jameson Nash 22 April 2020, 21:04:33 UTC [GCChecker] teach it how to deal with more complex expressions and atomics 04 May 2020, 18:08:32 UTC
78af72f Liozou 04 May 2020, 18:04:05 UTC Faster Rationals by avoiding unnecessary divgcd (#35492) 04 May 2020, 18:04:05 UTC
ad907d5 Elliot Saba 21 April 2020, 21:04:51 UTC Add Makefile caching helpers for lazy, expensive evaluations Makefiles natively offer two variable expansion modes: immediate and deferred. When expanding variables that require invocations of external programs (such as `llvm-config`) immediate expansion is almost always preferred, as it will run the external command once, exactly when the makefile variable is defined. Deferred mode, on the other hand, will expand the variable every time it is used, running the external program again and again. When the external program is expensive, this cost can slow down the build significantly, however when the external program requires some setup (for instance, when it itself is downloaded through other rules in the Makefile) it cannot always be immediately expanded. To address this, we build a caching layer that allows for deferred expansion, but once it has been expanded once, the variable is replaced with the result of running the command, and further hits to the same variable will return the cached value. (With the slight caveat that an empty result will cause the external command to be run again in the future). As an example usecase, this commit converts our relative path calculation to use a python script and showcases how to cache this operation. This will be used for further JLL stdlib work where the invocation is much more expensive. 04 May 2020, 18:00:43 UTC
a0048c8 Jameson Nash 25 April 2020, 05:10:07 UTC the do_apply growth strategy was possibly inefficient for the first step 04 May 2020, 17:57:43 UTC
0fa43ed Jameson Nash 25 April 2020, 04:46:44 UTC fix missing gc roots in subtyping 04 May 2020, 17:57:43 UTC
5ec0608 Jameson Nash 04 May 2020, 17:50:54 UTC Export jl_debug_method_invalidation. (#35729) * Export jl_debug_method_invalidation. * Don't write trailing null to stdout. 04 May 2020, 17:50:54 UTC
9d175a0 Tim Besard 04 May 2020, 17:49:40 UTC Include llvm-version.h first to avoid compiler warnings. (#35730) 04 May 2020, 17:49:40 UTC
6a2224b Milan Bouchet-Valat 04 May 2020, 16:00:01 UTC Bump Statistics stdlib module (#35709) 04 May 2020, 16:00:01 UTC
ffde1b2 Tim Besard 04 May 2020, 12:05:18 UTC Don't write trailing null to stdout. 04 May 2020, 12:05:18 UTC
f1d10e7 Michael Jared Lumpe 04 May 2020, 11:41:25 UTC Increment header level of TerminalMenus section in REPL docs. (#35718) 04 May 2020, 11:41:25 UTC
cc37c4a Tim Besard 04 May 2020, 11:13:31 UTC Export jl_debug_method_invalidation. 04 May 2020, 11:13:31 UTC
f83a84c Tim Holy 03 May 2020, 19:57:18 UTC Cut backedges to CoreLogging.shouldlog and CoreLogging.min_enabled_level 04 May 2020, 10:53:34 UTC
41264e4 Tim Holy 17 April 2020, 13:59:56 UTC Add more pkg-related precompile statements 04 May 2020, 10:53:34 UTC
4409d44 Kristoffer 17 April 2020, 13:24:03 UTC make sure stale_cachefile is precompiled 04 May 2020, 10:53:34 UTC
5206f93 Tim Holy 03 May 2020, 13:16:15 UTC Avoid specialization in logging 04 May 2020, 10:53:34 UTC
1da1b8d Tim Holy 03 May 2020, 13:15:49 UTC Reduce consequences of method invalidation for Dict and Set This aims to reduce disruption when adding methods to `convert` and `copymutable`. 04 May 2020, 10:53:18 UTC
b88d6df Tim Holy 03 May 2020, 13:12:24 UTC Improve dispatch in code loading CodeTracking.jl (a foundation for Revise & JuliaInterpreter) defines additional `PkgId` methods; leaving this untyped leads to invalidation of the entire code-loading path. 04 May 2020, 10:18:44 UTC
f345c5f Tim Holy 03 May 2020, 13:10:01 UTC Better codegen for unspecialized `copyto!(::Array, ::Array)` When the eltype and/or dimensionality is unspecified, there were 2 or 3 methods that might apply here (see `code_typed(copyto!, (Array, Array))`). Consequently partially-specified codegen tends to be poor for this operation. This harmonizes our implementations and eliminates eltype and dimensionality as a consideration for the dispatch. By default, it dispatches on `IndexStyle(src)` after a call to `unalias`, which makes sense because copying an array can change its preferred `IndexStyle`. It also fixes an aliasing bug. 04 May 2020, 10:16:25 UTC
a1f6448 Rafael Fourquet 04 May 2020, 10:04:00 UTC faster rand!(::MersenneTwister, ::Array{Bool}) (#33721) This uses the same optimizations as for other bits types, and gives equivalent performance as for `UInt8` (at least 7x to 9x speedup in few tested cases). 04 May 2020, 10:04:00 UTC
bb2aa52 Erik Schnetter 04 May 2020, 06:52:45 UTC Update VecElement description (#34536) * Update VecElement description Remove restriction that does not apply any more. 04 May 2020, 06:52:45 UTC
4297097 Benjamin Lungwitz 04 May 2020, 05:09:48 UTC Enhance Noteworthy Differences to Python in Documentation (#35268) Co-Authored-By: Jonas Schulze <jonas.schulze7@t-online.de> Co-Authored-By: Waldir Pimenta <waldyrious@gmail.com> Co-authored-by: Stefan Karpinski <stefan@karpinski.org> Co-Authored-By: Takafumi Arakaki <aka.tkf@gmail.com> 04 May 2020, 05:09:48 UTC
c2787b1 Keno Fischer 04 May 2020, 04:17:49 UTC Revert "Add a `@ref` to the Main docstring (#33357)" (#35722) This reverts commit dc8973c6464518056b7ae00b304657013aa4e6ab. 04 May 2020, 04:17:49 UTC
bf22b5a Matt Bauman 04 May 2020, 02:59:11 UTC Make tuple t[2:end] inferrable through constant prop (#31138) Spurred by #31132, I was curious if our constant propagation was robust enough to handle a peephole optimization here -- and it is! This removes the need to export front/tail by simply allowing folks to get inferrable results when they spell these operations as `t[2:end]` and `t[1:end-1]`. I additionally allowed for `t[3:end]`, `t[1:end-2]`, and the trivial `t[1:end]` as I have used `tail(tail(t))` in the past. 04 May 2020, 02:59:11 UTC
af324c7 Takafumi Arakaki 04 May 2020, 00:30:07 UTC Mention --bug-report in --help (#35712) * Mention --bug-report in --help * Mention --bug-report=help 04 May 2020, 00:30:07 UTC
dc8973c Anshul Singhvi 03 May 2020, 23:18:32 UTC Add a `@ref` to the Main docstring (#33357) 03 May 2020, 23:18:32 UTC
4495b76 Erik Schnetter 03 May 2020, 23:10:56 UTC Extend description of allowed `@fastmath` transformations (#17452) 03 May 2020, 23:10:56 UTC
ba9a003 Yijia Chen 03 May 2020, 20:46:00 UTC Clarify stdlib package for identity matrix (#29277) * Clarify the use of LinearAlgebra stdlib for identity matrices. Co-authored-by: Viral B. Shah <ViralBShah@users.noreply.github.com> 03 May 2020, 20:46:00 UTC
cf3d8b5 David P. Sanders 03 May 2020, 16:50:37 UTC Correct docstring of rem_pio2_kernel (#29335) 03 May 2020, 16:50:37 UTC
8ab87d2 Keno Fischer 03 May 2020, 08:07:24 UTC Provide high performance fallback for Pipe readbytes! (#35696) Without this it was falling back to the byte-by-byte version, which is incredibly slow. 03 May 2020, 08:07:24 UTC
b8d9e78 Tim Besard 03 May 2020, 06:32:48 UTC Merge pull request #35645 from JuliaLang/tb/strip_addrspace Add an LLVM pass to remove Julia-specific address space information. 03 May 2020, 06:32:48 UTC
dd1e525 Tim Besard 03 May 2020, 06:31:47 UTC Merge pull request #35680 from JuliaLang/tb/llvm_pass_timings Report LLVM timings and statistics during teardown. 03 May 2020, 06:31:47 UTC
61dd0ae Tim Besard 03 May 2020, 06:30:33 UTC Merge pull request #35380 from JuliaLang/tb/asan2 Put ASAN options in the executable 03 May 2020, 06:30:33 UTC
3a84b51 Keno Fischer 03 May 2020, 06:29:50 UTC Fix dropped bytes when bulk reading from uv streams with oversized buffers (#35695) In readbytes!(s::UVStream, buf::AbstractVector{UInt8}, nb), we wrap `buf` in a PipeBuffer of maxsize nb and swap it in as the main buffer for the stream `s`. However, when we inform libuv of the size of the buffer, we instead use the size of the underlying array, which can be larger. In that case, we probably either drop bytes or override something that the user did not want overriden. 03 May 2020, 06:29:50 UTC
d5b329a kimikage 03 May 2020, 06:29:29 UTC Reduce estimated cost of `not_int` to zero (#35707) This reduces the estimated cost of `not_int` to zero for more reasonable inlining. This change is not a perfect solution, but has little side-effects on the existing cost model. 03 May 2020, 06:29:29 UTC
301db97 Stephan Hilb 02 May 2020, 22:36:59 UTC implement `count` and `count!` using `mapreduce` (#34048) This creates the same calling interface for `count` as for other mapreduce-type functions like e.g. `sum`, namely allowing the `dims` keyword. The implementation itself is shorter than before without sacrificing performance. More detailed documentation for `count` was added too. 02 May 2020, 22:36:59 UTC
dddda07 Jeff Bezanson 02 May 2020, 22:14:24 UTC Merge pull request #35618 from JuliaLang/jb/iosbufsize change default ios buffer size to 32k 02 May 2020, 22:14:24 UTC
d192a73 Jeff Bezanson 02 May 2020, 22:11:29 UTC adjust deprecations for v1.5 (#35674) fixes #35665 02 May 2020, 22:11:29 UTC
7be230b Tim Holy 02 May 2020, 20:00:56 UTC Add note about removal of AST struct-defining Expr-heads (#35700) xref #33553 02 May 2020, 20:00:56 UTC
e1d8da1 Elliot Saba 19 April 2020, 04:39:29 UTC Sockets: Add UDP multicast group support Adds two functions `join_multicast_group()` and `leave_multicast_group` for usage with UDP multicast groups. 02 May 2020, 17:41:56 UTC
40cc4ad Tim Holy 02 May 2020, 14:31:57 UTC Move custom error hints to Experimental (#35679) Closes #35671 02 May 2020, 14:31:57 UTC
68c8708 Mateusz Baran 02 May 2020, 12:36:07 UTC Faster matrix sqrt for upper triangular matrices. (#31100) Replacing Val(::Bool) in an argument list with an explicit if makes the call type stable. The issue was discussed in #31007. 02 May 2020, 12:36:07 UTC
f36036c Tim Holy 02 May 2020, 09:29:02 UTC Support `⊙` and `⊗` as elementwise- and tensor-product operators (#35150) While we have broadcasting and `a*b'`, sometimes you need to pass an operator as an argument to a function. Since we already have `dot` or `⋅` for the inner product, these elementwise and tensor products fill out the space of possibilities. 02 May 2020, 09:29:02 UTC
9507225 Pedro Bruel 02 May 2020, 07:50:31 UTC Add inline documentation for Type and DataType. (#24561) * Add inline documentation for Type and DataType. Closes #23655. * Fix type annotation and clean up examples * Apply suggestions from code review Co-authored-by: Matt Bauman <mbauman@gmail.com> Co-authored-by: Matt Bauman <mbauman@gmail.com> 02 May 2020, 07:50:31 UTC
f2fa6cf Viral B. Shah 02 May 2020, 01:31:13 UTC doc: fix two reference links in the manual (#35693) 02 May 2020, 01:31:13 UTC
165a37b Matt Bauman 01 May 2020, 23:51:39 UTC Preserve structured matrix types more often when broadcasting (#32762) * Preserve structured matrix types more often when broadcasting The primary motivation here is exponentiation (fixes #32759). Previously `X .^ 2` would return a Matrix but `two=2; X .^ two` would return the same structured matrix type as `X`. This simply teaches LinearAlgebra a little bit more about broadcast so it can safely compute the zero-preservation property of broadcasts involving `Ref`s, which is sufficient to handle the literal pow argument list. * fixup negative exponents 01 May 2020, 23:51:39 UTC
3f9faf3 simeonschaub 01 May 2020, 22:41:36 UTC lower a' as var"'"(a) (#34634) 01 May 2020, 22:41:36 UTC
7583d87 Matt Bauman 01 May 2020, 22:15:26 UTC Make StridedArray display as StridedArray (#31149) Instead of a huge union. This is just a special-casing for this particular union instead of pursuing the general (and hard\!) typealias search. But given the ubiquity of StridedArray in method lists, I find this special case to be warranted and hugely beneficial. Here is a sampling from `methods(*)`: ```julia [242] *(A::LinearAlgebra.AbstractQ, b::StridedArray{T, 1} where T) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/qr.jl:556 [243] *(A::LinearAlgebra.AbstractQ, B::StridedArray{T, 2} where T) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/qr.jl:568 [244] *(Q::LinearAlgebra.AbstractQ, adjB::LinearAlgebra.Adjoint{#s614,#s613} where #s613<:(StridedVecOrMat{T} where T) where #s614) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/qr.jl:623 [245] *(A::StridedArray{T, 2} where T, Q::LinearAlgebra.AbstractQ) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/qr.jl:668 [246] *(A::StridedArray{T, 2} where T, adjB::LinearAlgebra.Adjoint{#s614,#s613} where #s613<:LinearAlgebra.AbstractQ where #s614) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/qr.jl:708 [247] *(A::LinearAlgebra.LQPackedQ, B::StridedVecOrMat{T} where T) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/lq.jl:153 [248] *(A::LinearAlgebra.LQPackedQ, adjB::LinearAlgebra.Adjoint{#s614,#s613} where #s613<:(StridedVecOrMat{T} where T) where #s614) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/lq.jl:177 [249] *(A::StridedVecOrMat{T} where T, adjQ::LinearAlgebra.Adjoint{#s614,#s613} where #s613<:LinearAlgebra.LQPackedQ where #s614) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/lq.jl:216 [250] *(A::StridedVecOrMat{T} where T, Q::LinearAlgebra.LQPackedQ) in LinearAlgebra at /home/mbauman/julia/usr/share/julia/stdlib/v1.2/LinearAlgebra/src/lq.jl:240 ``` 01 May 2020, 22:15:26 UTC
ab9ff0f Keno Fischer 01 May 2020, 21:00:29 UTC Disable flaky rmprocs test (#35677) 01 May 2020, 21:00:29 UTC
7a1f294 Viral B. Shah 01 May 2020, 18:53:49 UTC Remove doc reference to Base.add_sum 01 May 2020, 18:53:49 UTC
e58c6e6 Viral B. Shah 01 May 2020, 18:39:41 UTC Fix whitespace in #27014 01 May 2020, 18:39:41 UTC
e3f91bd Matt Bauman 01 May 2020, 17:06:04 UTC Optimize Float.([32 64]) exponentiation by.(-1:3) (#34863) We perform these optimizations in the literal pow case, so I'd contend that they're just as worthwhile to do in the non-literal case. Calling LLVM's pow intrinsic takes ~16ns on my computer; adding these quick exits get compiled down to a simple switch statement that seems to add ~0.5ns in the cases where we don't hit the quick exits. In exchange, we get consistent numbers everywhere, they're better numbers at that (they're closer to the real answer), and these small common powers become essentially free. 01 May 2020, 17:06:04 UTC
a93e45e Matt Bauman 01 May 2020, 17:00:51 UTC Merge pull request #35620 from JuliaLang/mb/yet-another-indirection-for-broadcast Enable combined BroadcastStyle dispatch in materialize! 01 May 2020, 17:00:51 UTC
4296239 Stefan Karpinski 01 May 2020, 16:42:41 UTC implement "nearly division less" algorithm for rand(a:b) (#29240) * implement "nearly division less" algorithm for rand(a:b) Cf. https://arxiv.org/abs/1805.10941. Closes #20582, #29004. * fix overflow error in tests * make NDL the default algo * update NEWS.md * try make tests pass on 32-bits machines * add a comment for mod(-s, s) * remove vestigial transient `fast` function, and update comments 01 May 2020, 16:42:41 UTC
594ff49 Fredrik Ekre 01 May 2020, 16:40:13 UTC Add extra example to fill docstring. (#35683) Co-authored-by: Paul Söderlind <PaulSoderlind@users.noreply.github.com> 01 May 2020, 16:40:13 UTC
c466e8a cormullion 01 May 2020, 16:39:09 UTC Update example in Functions/Arg destructuring (#27014) The example in Argument destructuring didn't work when I typed it in, so I changed the name to `gap()`. 01 May 2020, 16:39:09 UTC
5def3e6 Simon Etter 01 May 2020, 16:28:49 UTC Comment on byte order for `read` and `write` (#26471) * Comment on byte order for `read` and `write` * Clarify endianness conversion * Use htol and ltoh consistently * Change from `/tmp/filename` to `tempname()` * plattform -> platform 01 May 2020, 16:28:49 UTC
285a46b Jeff Bezanson 01 May 2020, 16:23:54 UTC revise parallel computing manual chapters (#35650) 01 May 2020, 16:23:54 UTC
back to top