7a3fff1 | Steven G. Johnson | 08 May 2019, 14:10:03 UTC | pwsh for powershell 6 | 08 May 2019, 14:10:03 UTC |
2b64697 | Steven G. Johnson | 07 May 2019, 21:08:57 UTC | rm missing cross-ref | 07 May 2019, 21:08:57 UTC |
9d4fe42 | GregPlowman | 12 September 2017, 04:48:08 UTC | REPL shell mode for Windows REPL shell mode for Windows Documentation JULIA_SHELL environment variable Interacting with Julia docs - xref JULIA_SHELL Fix whitespace JULIA_SHELL doc Fix JULIA_SHELL cross-reference Address code review comments rebase fix | 26 March 2019, 20:45:52 UTC |
8649af9 | Narendra Patel | 26 March 2019, 17:56:44 UTC | Add inv(::Missing) (#31451) * add inv(::Missing) * update docstring for inv method | 26 March 2019, 17:56:44 UTC |
a90629a | Jameson Nash | 26 March 2019, 15:13:38 UTC | Merge pull request #31440 from JuliaLang/jn/nfc-2 Various other tiny changes and QOL enhancements | 26 March 2019, 15:13:38 UTC |
c0b40d8 | Jameson Nash | 26 March 2019, 15:13:02 UTC | Merge pull request #31413 from JuliaLang/jn/nfc-1 Various tiny changes and QOL enhancements | 26 March 2019, 15:13:02 UTC |
1be1deb | Jeff Bezanson | 26 March 2019, 02:46:04 UTC | small fixes to libuv locks (#31454) - use normal locking instead of nogc - remove locks before threads have started | 26 March 2019, 02:46:04 UTC |
ea0c6e9 | daviehh | 25 March 2019, 22:50:06 UTC | [doc] dlopen docstring (#31475) Documentation for `dlopen_e`: looks like copy-paste error from https://github.com/JuliaLang/julia/commit/d9d2b9cf808aab8c3a90e95e03ffc32761ce5316#diff-4f669dd4cc416a82508956a377e6a7c9 Correct: `dlopen_e` deprecated in favor of `dlopen(...throw_error=false)`; `dlsym_e` deprecated in favor of `dlsym(...throw_error=false)`; Current documentation lists `dlopen_e` deprecated in favor of `dlsym`. | 25 March 2019, 22:50:06 UTC |
92a361b | Valentin Churavy | 25 March 2019, 20:43:40 UTC | Merge pull request #31431 from NHDaly/nhdaly-llvm-ghost-types Fix llvm-assertion: don't instantiate 0-sized structs (ghost types) | 25 March 2019, 20:43:40 UTC |
c46f4bb | Mustafa M | 25 March 2019, 20:37:12 UTC | Return current directory without the trailing slash (#31417) | 25 March 2019, 20:37:12 UTC |
bea863e | Jameson Nash | 25 March 2019, 17:18:00 UTC | Merge pull request #31338 from JuliaLang/jn/opt-faster optimize (a little bit) faster | 25 March 2019, 17:18:00 UTC |
6ba6a83 | Jeff Bezanson | 25 March 2019, 16:13:38 UTC | enable `jl_wake_libuv` on windows (#31450) | 25 March 2019, 16:13:38 UTC |
10141d3 | Thibaut Lienart | 24 March 2019, 23:44:41 UTC | [doc] Adding explicit link to PCRE syntax (#31463) Following a suggestion by @StefanKarpinski [on discourse](https://discourse.julialang.org/t/questions-ss-about-function-import-and-type-parsing-ocaml-code-importing-compiling-and-regex-syntax/22241/10), I added the link to the regex syntax description. | 24 March 2019, 23:44:41 UTC |
f82b81c | Tim Besard | 24 March 2019, 19:35:58 UTC | Merge pull request #31428 from JuliaLang/tb/export_passes Export additional LLVM passes | 24 March 2019, 19:35:58 UTC |
ffa2924 | Jameson Nash | 13 March 2019, 01:03:48 UTC | optimize: optimizations help performance by removing bits of code highlighted by profiling as possible hot-spots. | 22 March 2019, 23:26:55 UTC |
ede38a8 | Jameson Nash | 13 March 2019, 15:33:42 UTC | subtype: add fast-path required by new optimizer This is a new subtype pre-process step that attempts to look for easily computable reasons that the result must be false (or true). Since it can avoid allocating an `stenv` and otherwise setting up the state needed for handling TypeVar constraints precisely, it is hoped to save time (on average) by filtering out easy cases. | 22 March 2019, 23:25:31 UTC |
672bf8b | Mustafa M | 22 March 2019, 21:25:25 UTC | realpath buffer is a vector (#31433) | 22 March 2019, 21:25:25 UTC |
ad40952 | Jeff Bezanson | 22 March 2019, 17:39:09 UTC | add locks for libuv (#31437) | 22 March 2019, 17:39:09 UTC |
6a28aa9 | Jeff Bezanson | 22 March 2019, 17:38:55 UTC | Merge pull request #31398 from JuliaLang/jn/kp/partr2 import partr code, allow using it in threaded loops | 22 March 2019, 17:38:55 UTC |
a84970d | Jameson Nash | 22 March 2019, 16:07:34 UTC | mktempdir: support user-defined filename prefix (#31230) Also switch to using libuv, as it now provides a platform-independent implementation. Co-Authored-By: Jameson Nash <vtjnash@gmail.com> | 22 March 2019, 16:07:34 UTC |
33c15d3 | Nathan Daly | 22 March 2019, 01:54:18 UTC | More tests: can load/store any 0-sized struct | 22 March 2019, 01:54:18 UTC |
3725072 | Nathan Daly | 22 March 2019, 01:51:02 UTC | Add test for Issue #29929: `unsafe_store!(::Ptr{Nothing}, ::Nothing)` | 22 March 2019, 01:51:02 UTC |
d0b0ca3 | Nathan Daly | 22 March 2019, 01:44:55 UTC | Apply vtjnash's suggestion to src/intrinsics.cpp replace `mark_julia_type()` with `return e` always. | 22 March 2019, 01:44:55 UTC |
8df4de4 | Jameson Nash | 22 March 2019, 01:42:25 UTC | Apply vtjnash's suggested change to src/intrinsics.cpp Co-Authored-By: NHDaly <NHDaly@gmail.com> | 22 March 2019, 01:42:25 UTC |
b076d04 | Fredrik Ekre | 21 March 2019, 22:30:28 UTC | Bump Pkg version to 1.2-alpha.3. (#31424) | 21 March 2019, 22:30:28 UTC |
818c576 | Jameson Nash | 21 March 2019, 21:38:07 UTC | Libdl: better support for musl We previously claimed that: "Skip over objects without a path (as they represent this own object)" This was false on two levels: - On everything non-glibc the binary does have a path - On linux, there is also usually also at least one shim object that does not have a path So instead, we filter these out explicitly at the end (and remove some duplicate code paths). | 21 March 2019, 22:17:03 UTC |
a9a681d | Jameson Nash | 21 March 2019, 21:32:53 UTC | build: pass makensis output through iconv This output is known to contain latin-1 characters, (from our using of NSIS Modern User Interface 2.0), but the terminal is probably in utf-8 mode, so this this is known to make some bad terminal emulators (*cough*python*cough*) sad. | 21 March 2019, 22:16:44 UTC |
0531a24 | Jameson Nash | 21 March 2019, 21:56:18 UTC | win: add ./julia "symlink", like on posix While we cannot have real symlinks, we can perhaps fake it enough for convenience purposes. | 21 March 2019, 22:16:44 UTC |
da35c34 | Jameson Nash | 21 March 2019, 21:46:51 UTC | Libdl: fix dlopen uses missed as part of #28953, c.f. #28881 | 21 March 2019, 22:16:44 UTC |
9d4384c | Jameson Nash | 21 March 2019, 21:43:28 UTC | build: fix flisp build rules on Windows | 21 March 2019, 22:16:43 UTC |
4eb6308 | Jeff Bezanson | 21 March 2019, 21:56:21 UTC | fix #31404, lowering `where` inside tuple in kwcall (#31409) | 21 March 2019, 21:56:21 UTC |
604c9b2 | Jameson Nash | 21 March 2019, 21:19:21 UTC | inference: fix typeof-tfunc impl (#31405) | 21 March 2019, 21:19:21 UTC |
d9d8d4c | Jameson Nash | 09 January 2019, 20:47:18 UTC | enable partr backend drop/fix more broken code drop jl_thread_sleep_threshold deadcode improve jl_wakeup_thread | 21 March 2019, 21:13:00 UTC |
58c60a2 | Valentin Churavy | 21 March 2019, 19:14:15 UTC | export custom LLVM passes | 21 March 2019, 19:16:34 UTC |
0d0b853 | Nathan Daly | 21 March 2019, 18:11:41 UTC | Fix llvm-assertion: don't load ghost types As I understand it, LLVM doesn't support 0-sized structs (ghost types), which julia makes significant use of. This fixes an incorrect behavior in the julia compiler accidentally attempting to instantiate such 0-sized structs. When using a debug build of LLVM, this fails with an assertion error, but failed silently on standard LLVM. | 21 March 2019, 18:39:05 UTC |
6ff4729 | Tim Besard | 21 March 2019, 14:17:53 UTC | Export the GC lowering pass. | 21 March 2019, 14:17:53 UTC |
f94eade | Tim Besard | 21 March 2019, 14:16:15 UTC | Fix NVVM reflect pass wrapper. | 21 March 2019, 14:16:15 UTC |
54fe55b | Valentin Churavy | 21 March 2019, 13:47:40 UTC | Merge pull request #31403 from JuliaLang/tb/llvmcall_extern Allow ccall(..., llvmcall, ...) to external functions. | 21 March 2019, 13:47:40 UTC |
f611b46 | Matt Bauman | 20 March 2019, 14:57:26 UTC | Fix 29545: Implement unaliascopy for ReinterpretArray (#30296) * Fix 29545: Implement unaliascopy for ReinterpretArray | 20 March 2019, 14:57:26 UTC |
910e804 | Jameson Nash | 19 March 2019, 19:49:40 UTC | compile: small correctness improvement to generate_precompile.jl | 19 March 2019, 20:01:19 UTC |
78561c3 | Jameson Nash | 19 March 2019, 19:49:12 UTC | test: remove duplicated computation [NFC] | 19 March 2019, 20:01:19 UTC |
35b510d | Jameson Nash | 19 March 2019, 19:48:59 UTC | log: make table data constant [NFC] This allows inference to constant-fold the result! | 19 March 2019, 20:01:19 UTC |
ef9361e | Jameson Nash | 19 March 2019, 19:42:16 UTC | horner macro: add line number info [NFC] Instead of nesting the user code inside the macro, structure it to look very nearly just like inlining would, right down to modeling the inability for lowering to handle the value in return position (by adding a dummy variable `r`), and make `t` local. Before: │ @ REPL[2]:14 within `sin_kernel_f64' │ t@_6 = y² │ %5 = t@_6 │ %6 = Base.Math.muladd(t@_6, Main.DS4, Main.DS3) │ %7 = Base.Math.muladd(%5, %6, Main.DS2) After: │ @ REPL[2]:14 within `sin_kernel_f64' │ t@_3 = y² │ ┌ @ math.jl:101 within `macro expansion' │ │ %5 = t@_3 │ │ %6 = Base.Math.muladd(t@_3, Main.DS4, Main.DS3) │ │ r@_4 = Base.Math.muladd(%5, %6, Main.DS2) │ └ │ %8 = r@_4 | 19 March 2019, 20:01:15 UTC |
b207c01 | Jeff Bezanson | 19 March 2019, 19:38:32 UTC | clean up implementation of `ml_matches_visitor` a bit (#31358) Simplifies the code flow a little, and avoids computing a type intersection that we only used for `include_ambiguous`, which is not the common case. | 19 March 2019, 19:38:32 UTC |
58a0016 | Valentin Churavy | 19 March 2019, 13:28:24 UTC | Merge pull request #31347 from JuliaLang/vc/loopinfo2 [Loopinfo] recursive conversion to metadata | 19 March 2019, 13:28:24 UTC |
b437e61 | Tim Besard | 19 March 2019, 12:47:04 UTC | Allow ccall(..., llvmcall, ...) to external functions. | 19 March 2019, 12:49:47 UTC |
d76a30a | Tanmay Mohapatra | 19 March 2019, 09:11:55 UTC | Merge pull request #31036 from tanmaykm/tan/fix31035 start_worker option to close stdin redirect stdout | 19 March 2019, 09:11:55 UTC |
ca36e57 | Elliot Saba | 19 March 2019, 07:22:13 UTC | Insert missing `\` in `deps/Makefile` | 19 March 2019, 07:22:13 UTC |
4525a6a | Elliot Saba | 19 March 2019, 04:40:15 UTC | Explicitly ask for `HFS+` filesystems when creating `.dmg` files on MacOS (#31385) This is necessary for cross-compatibility, as if a user creates a `.dmg` file on a newer version of MacOS, it may default to `APFS` which is unreadable on MacOS 10.11 and lower. | 19 March 2019, 04:40:15 UTC |
5d6f4ce | Jameson Nash | 19 March 2019, 03:38:18 UTC | SuiteSparse: fix shlib build (Windows) (#31384) The attempt to get env_OS set was thwarted by recursive make. Instead, we just set the value of the cross-compile variable directly to the correct value. | 19 March 2019, 03:38:18 UTC |
43b188c | kpamnany | 09 January 2019, 17:38:50 UTC | import partr branch | 19 March 2019, 02:16:56 UTC |
41c3369 | Jeff Bezanson | 19 March 2019, 02:13:29 UTC | Merge pull request #30806 from JuliaLang/jn/task-ll worklist changes for partr (1/3) | 19 March 2019, 02:13:29 UTC |
bd610a3 | Jameson Nash | 18 March 2019, 19:54:28 UTC | make Workqueue threadsafe (#30838) | 19 March 2019, 02:12:14 UTC |
9902531 | Jameson Nash | 03 January 2019, 20:48:15 UTC | scheduler: use linked list for Workqueue | 19 March 2019, 02:12:09 UTC |
b93fd23 | Elliot Saba | 18 March 2019, 20:58:38 UTC | Backport `libunwind` patch to build on armv7l (#31383) X-ref: https://www.mail-archive.com/libunwind-devel@nongnu.org/msg02617.html X-ref: https://github.com/libunwind/libunwind/commit/a1437a3d27924e17d00021df3cc659bd998e8580 | 18 March 2019, 20:58:38 UTC |
a26e6cc | Max Horn | 18 March 2019, 20:22:49 UTC | Clarify `run` and `success` documentation (#31341) Both of these say "constructed with backticks" without explaining what that means, so we add a link to the "Running External Programs" section. | 18 March 2019, 20:22:49 UTC |
27f8e77 | Jeff Bezanson | 18 March 2019, 20:13:50 UTC | use `String` instead of `AbstractString` in a few places (#31361) | 18 March 2019, 20:13:50 UTC |
4dc1593 | Jeff Bezanson | 18 March 2019, 18:15:50 UTC | some front-end code cleanup (#31360) | 18 March 2019, 18:15:50 UTC |
6aeb7ef | Jeff Bezanson | 18 March 2019, 18:14:53 UTC | fix #31353, `hasmethod` not working with UnionAll types (#31356) | 18 March 2019, 18:14:53 UTC |
be70066 | Matt Bauman | 18 March 2019, 17:54:17 UTC | Merge pull request #31369 from JuliaLang/kf/broadcastedsize Defined size(::Broadcasted) | 18 March 2019, 17:54:17 UTC |
8f5ee84 | Valentin Churavy | 14 March 2019, 14:48:06 UTC | recursive conversion to metadata | 18 March 2019, 15:20:50 UTC |
c9a3710 | Valentin Churavy | 15 March 2019, 03:39:15 UTC | prevent system libjulia from shadowing build libjulia in test/llvmpasses | 18 March 2019, 15:20:50 UTC |
8c0b550 | Nicolau Werneck | 17 March 2019, 17:35:24 UTC | Backporting LLVM patch r355582 to release/6.x (fixes AVX512 max issue) (#31292) * Backporting LLVM patch r355582 to release/6.x (fixes AVX512 max issue | 17 March 2019, 17:35:24 UTC |
cee3030 | Jeff Bezanson | 16 March 2019, 16:55:19 UTC | fix #31326, setting of system image data size when loading (#31364) | 16 March 2019, 16:55:19 UTC |
546eb97 | Keno Fischer | 16 March 2019, 16:54:43 UTC | Fix #31357 - Missing Upsilon node in nested try/catch (#31365) As the issue demonstrates, if we have nested try/catch blocks, we need one upsilon node for every layer of nesting. This just inserts another upsilon node for every layer of nesting. Alternative implementations would be to 1. Re-use the same upsilon node for every phic (but this currently conflicts with our implementation of this feature in the interpreter). 2. To place the additional upsilon nodes in the catch block. This would be slightly faster, but also a bit more complicated, so is left as a future optimization. | 16 March 2019, 16:54:43 UTC |
9579176 | Jeff Bezanson | 16 March 2019, 16:54:04 UTC | fix #31352, make sure errors while printing errors don't kill the REPL (#31366) | 16 March 2019, 16:54:04 UTC |
967c81f | Keno Fischer | 16 March 2019, 00:30:27 UTC | Defined size(::Broadcasted) | 16 March 2019, 00:30:51 UTC |
8376535 | Keno Fischer | 15 March 2019, 00:36:49 UTC | Properly propagate taint in SROA When SROA crosses the compaction line to look at uncompacted IR, it needs to be sure to propagate that taint to all values it pulls out of the uncompacted IR. We had this logic for phi nodes, but we accidentally lacked it for raw SSAValues. This didn't cause problems in practice, because our pipeline compacts out raw SSAValues before running SROA, but it is otherwise legal. Fix that and add a test. | 15 March 2019, 20:08:15 UTC |
00f80f5 | Keno Fischer | 15 March 2019, 00:02:48 UTC | More thoroughly copy IRCode The contents of IRCode get modified in interesting ways all over the optimizer. The `copy(::IRCode)` function did an incomplete job of copying everything that gets modified. We don't really use this function internally much, but it's useful for external tooling, so let's fix it. | 15 March 2019, 20:08:15 UTC |
b71cae2 | Jeff Bezanson | 15 March 2019, 19:51:05 UTC | make a subtyping fast path a bit less aggressive (#31346) | 15 March 2019, 19:51:05 UTC |
6f40203 | Jeff Bezanson | 15 March 2019, 19:36:30 UTC | fix #31295, allow e.g. `view(a,:) .+= view(b,:)` (#31300) | 15 March 2019, 19:36:30 UTC |
9f69b2e | Matt Bauman | 14 March 2019, 18:55:54 UTC | Merge pull request #30403 from tkf/doc-propagate_inbounds Include @​propagate_inbounds docstring in documentation | 14 March 2019, 18:55:54 UTC |
a168bc8 | Valentin Churavy | 14 March 2019, 18:17:49 UTC | Merge pull request #31324 from JuliaLang/vc/tuple_props define getproperty for tuples | 14 March 2019, 18:17:49 UTC |
2d6e183 | Keno Fischer | 14 March 2019, 15:52:07 UTC | Fix a missing widenconst in SROA (#31336) We can now infer Expr(:new) as returning a `PartialStruct` if one of the arguments is constant. However, SROA was missing a corresponding widenconst and thus refusing to process such Expr(:new). This didn't show up very much in julia code, because inlining generally dropped the PartialStruct annotation on the Expr(:new), but it does show up with more aggressive inlining schemes that I'm playing with. Fix it and add a test. | 14 March 2019, 15:52:07 UTC |
97719d7 | Max Horn | 14 March 2019, 15:30:14 UTC | doc: Improve searchsorted{,first,last} examples in documentation (#31340) Use the same inputs for all three functions, and cover various cases, to help the reader grasp how these functions work and relate to each other. Also remove the examples using the `rev` option, which did not really make sense, as the input was not reversed, so these really were examples on "what happens if the input is not sorted", which does not really help much (if at all, an explanation in the text would be better). | 14 March 2019, 15:30:14 UTC |
cc1caa4 | Koustav Chowdhury | 14 March 2019, 15:05:12 UTC | Dates: docstrings for month constants (#31218) | 14 March 2019, 15:05:12 UTC |
2567c3a | ndinsmore | 14 March 2019, 14:40:58 UTC | map!(f, values(dict)): modify Dict values in-place (#31223) | 14 March 2019, 14:40:58 UTC |
95fcde9 | Jan Weidner | 14 March 2019, 08:49:31 UTC | doc: add docstrings for Base.front and Base.tail (#31132) | 14 March 2019, 08:49:31 UTC |
4457264 | Fredrik Ekre | 14 March 2019, 08:47:41 UTC | Remove integer domain example that doesn't work (#31298) * Update integer literal example to reflect smarter interpretation of exponents * Drop mention of integer size * Add an extra example of how to fix an exponent domain error * Fix a sign error in documentation | 14 March 2019, 08:47:41 UTC |
bd80892 | Lyndon White | 14 March 2019, 08:44:44 UTC | doc: highlight that julia integers are machine integers (#31328) | 14 March 2019, 08:44:44 UTC |
f4e029f | Katharine Hyatt | 14 March 2019, 08:43:35 UTC | Document the Base64 module. (#31332) | 14 March 2019, 08:43:35 UTC |
5ccdc99 | Anton Lodder | 14 March 2019, 02:58:31 UTC | Fix a sign error in documentation | 14 March 2019, 02:58:31 UTC |
fd01ff7 | chethega | 13 March 2019, 21:25:56 UTC | fix pessimistic maxprobe in Dict rehash (#31290) | 13 March 2019, 21:25:56 UTC |
5fac1b5 | Lyndon White | 13 March 2019, 20:48:24 UTC | remove codeblock from docstring (#31158) | 13 March 2019, 20:48:24 UTC |
7f7fe9a | Pierre Haessig | 13 March 2019, 20:40:01 UTC | add docstring for Base.isiterable (#31108) | 13 March 2019, 20:40:01 UTC |
af7cf2d | Sebastian Gutsche | 13 March 2019, 19:39:21 UTC | Clarified documentation of `isless` (#31323) | 13 March 2019, 19:39:21 UTC |
4e2c90c | Valentin Churavy | 13 March 2019, 11:59:01 UTC | Merge pull request #31095 from JuliaLang/vc/loopinfo Refactor `Expr(:simdloop)` to `Expr(:loopinfo, ...)` | 13 March 2019, 11:59:01 UTC |
d7fdd75 | Mus | 12 March 2019, 21:46:39 UTC | Capture curl download agent error message (#31209) | 12 March 2019, 21:46:39 UTC |
a85dd55 | Valentin Churavy | 12 March 2019, 20:15:31 UTC | define getproperty for tuples | 12 March 2019, 20:15:31 UTC |
b3ea10a | Valentin Churavy | 16 February 2019, 19:53:54 UTC | Refactor `Expr(:simdloop)` to `Expr(:loopinfo, ...)` - `Expr(:loopinfo, Symbol("julia.simdloop"))` replaces `Expr(:simdloop, false)` - `Expr(:loopinfo, Symbol("julia.simdloop"), Symbol("julia.ivdep"))` replaces `Expr(:simdloop, true)` While currently not directly exposed users can now pass other [loopinfo](https://llvm.org/docs/LangRef.html#llvm-loop) metadata to LLVM. | 12 March 2019, 19:53:51 UTC |
48e7a35 | Jeff Bezanson | 12 March 2019, 15:31:04 UTC | fix #31256, support `catch_stack` in test system (#31280) | 12 March 2019, 15:31:04 UTC |
57b9210 | Valentin Churavy | 12 March 2019, 14:09:58 UTC | Merge pull request #31113 from chethega/simd_specifity Make `simd_inner_length` generic fallback less specific | 12 March 2019, 14:09:58 UTC |
65df511 | Fredrik Ekre | 12 March 2019, 11:42:07 UTC | Bump Pkg to 1.2-alpha.2. (#31317) | 12 March 2019, 11:42:07 UTC |
5c8be73 | chethega | 12 March 2019, 09:47:09 UTC | moved docs to source comment | 12 March 2019, 09:47:09 UTC |
e949f89 | Keno Fischer | 09 March 2019, 22:55:40 UTC | Refactor Inlining. NFC. I'm working on some problems that require some slightly different characteristics than our default inliner. While working on that, I got frustrated by some of the accumulated cruft and structuring of the inliner, so I decided to spend some time to clean it. This doesn't change any behavior, but moves some of the code around, drops now-unused arguments and properties and introduces a `Signature` struct to encapsulate the (f, ft, atypes, atype) tuple we passed around everywhere. | 12 March 2019, 02:35:20 UTC |
86fcbc7 | Katharine Hyatt | 11 March 2019, 20:08:06 UTC | Merge pull request #31297 from JuliaLang/ksh/mem Doc free_memory and total_memory | 11 March 2019, 20:08:06 UTC |
023c8e4 | Kenta Sato | 11 March 2019, 19:06:52 UTC | allow chop to take an empty string (#31312) | 11 March 2019, 19:06:52 UTC |
27cc988 | Katharine Hyatt | 08 March 2019, 13:26:24 UTC | Doc free_memory and total_memory | 11 March 2019, 13:13:29 UTC |
38439d9 | Katharine Hyatt | 11 March 2019, 07:41:02 UTC | add docs for BLAS.iamax (#31277) | 11 March 2019, 07:41:02 UTC |
2f323ee | AnjoMan | 10 March 2019, 04:32:00 UTC | Add an extra example of how to fix an exponent domain error | 10 March 2019, 04:44:28 UTC |