d4cdde1 | Jeff Bezanson | 04 September 2018, 18:43:43 UTC | fix #28915, identity conversion to union of tuple types (#28951) | 04 September 2018, 18:43:43 UTC |
14fb104 | wfrgra | 04 September 2018, 18:15:02 UTC | Fix download agent search relying on throwing of Sys.which(). Update t #28157 (#28682) * Fix download agent search relying on throwing of Sys.which() | 04 September 2018, 18:15:02 UTC |
e7d7259 | Keno Fischer | 03 September 2018, 00:13:48 UTC | Fix out of bounds write in array deserialization The array was allocated based on the serialized `elsize` of the array, however, unions get an extra selector array after the regular storage which was not allocated (because we didn't know it was gonna be a union array at the time when we allocated it). According to a48eeefd6207090898f5b1877684aa10f217800a we cannot look at the element type to allocate the array, so we need to serialize a bit to indicate that we will have a union array. Fixes #28998 | 04 September 2018, 16:55:32 UTC |
c8450d8 | Pablo San-Jose | 04 September 2018, 10:24:29 UTC | CartesianIndices and LinearIndices docstring (#29023) | 04 September 2018, 10:24:29 UTC |
49e58ba | Rafael Fourquet | 03 September 2018, 19:19:59 UTC | doc: fix accumulate examples not using the init keyword (#28910) * doc: fix accumulate examples not using the init keyword * add links (KristofferC) | 03 September 2018, 19:19:59 UTC |
b1dac9f | Keno Fischer | 30 August 2018, 18:49:56 UTC | Vector lowering improvents in GC placement Support for vectors of tracked pointer was incomplete in the GC placement pass. Try to fix as many cases as possible and add some tests. A refactor to make all of this nicer (vectors weren't originally part of the implementation might be good), but for now, let's get it correct first. Fixes #28536 | 03 September 2018, 17:58:27 UTC |
2e603aa | Jan Magnusson | 03 September 2018, 08:58:15 UTC | Docs for logging: write to file (#28927) * write log to file * Use julia-repl for highlighting, fix indentation | 03 September 2018, 08:58:15 UTC |
9f6dbec | Keno Fischer | 01 September 2018, 22:41:46 UTC | Also fix predicate in stmt_effect_free | 02 September 2018, 19:40:29 UTC |
48c6846 | Keno Fischer | 01 September 2018, 20:13:14 UTC | Fix a small typo in sroa This caused the optimizer to miss some optimization opportunities. | 02 September 2018, 19:40:29 UTC |
7453e59 | Keno Fischer | 01 September 2018, 19:49:19 UTC | Generalize condition in is_known_call We're not consistent about whether singletons are represented as Consts or as types (particularly after round-tripping through a representation that doesn't have `Const`). As such, it is easy for the compiler to miss calls. | 02 September 2018, 19:40:29 UTC |
9ab0261 | Mark | 02 September 2018, 18:50:36 UTC | ismatch() is now occursin() not contains() (#29005) | 02 September 2018, 18:50:36 UTC |
f77c36b | Stephan Hilb | 02 September 2018, 11:27:46 UTC | doc/faq: move question to correct section (#28997) | 02 September 2018, 11:27:46 UTC |
5ee5de4 | Simon Byrne | 01 September 2018, 23:50:31 UTC | Fix spelling of kernel in html id (#28936) | 01 September 2018, 23:50:31 UTC |
c36c7ff | angusmoore | 01 September 2018, 23:49:25 UTC | Cross reference hasmethod with applicable in docs (#28918) | 01 September 2018, 23:49:25 UTC |
226a4ec | Keno Fischer | 01 September 2018, 19:52:14 UTC | Indirect checking for return_type through an extra function (#26836) Compiler.return_type is a bit of an odd beast because we basically treat it like a built-in, but it's defined in the compiler. When working on inference, it can be useful to load a second copy of inference with the new changes (e.g. to test changes before the changes are capable of running correctly on all input). This works quite well, but was causing problems with return type, because Base's notion of return_type and the second inference copy's notion of return_type were different. This patch adds a simple is_return_type(f) predicate that we can overload in our second inference copy in order to make it recognize both Core.Compiler.return_type and return_type (from the second copy's perspective as the return_type builtin). | 01 September 2018, 19:52:14 UTC |
d9d2b9c | Elliot Saba | 01 September 2018, 17:16:19 UTC | Fix Libdl changes (#28953) * Update `NEWS.md` for `Libdl.{dlopen,dlsym}` changes * Fix `dlopen()` and `dlsym()` backwards incompatibility * Update tests | 01 September 2018, 17:16:19 UTC |
8acad0c | Valentin Churavy | 01 September 2018, 03:55:36 UTC | Merge pull request #28841 from JuliaLang/vc/utils Add `lit` and `FileCheck` to tools and run llvmpasses by default | 01 September 2018, 03:55:36 UTC |
455e669 | Keno Fischer | 28 July 2018, 19:45:33 UTC | Run the llvmpasses tests on Travis | 31 August 2018, 21:50:07 UTC |
d30b3df | Keno Fischer | 28 July 2018, 19:42:30 UTC | Fix alloc opt tests Alloc opt got smarter. This makes a bunch of tests basically useless because they get optimized out entirely, but keep them to make sure it stays that way. | 31 August 2018, 21:50:07 UTC |
5bf885c | Valentin Churavy | 31 August 2018, 06:01:46 UTC | correct PATH override for test/llvmpasses | 31 August 2018, 21:50:07 UTC |
d1a27f4 | Valentin Churavy | 31 August 2018, 05:48:41 UTC | add make dep on build_staging for BINARYBUILDER_LLVM | 31 August 2018, 21:50:07 UTC |
9158101 | Valentin Churavy | 31 August 2018, 05:47:58 UTC | make llvmpasses test work for out-of-source | 31 August 2018, 21:50:07 UTC |
eebb4af | Valentin Churavy | 22 August 2018, 21:28:42 UTC | fix llvmpasses Makefile to use lit from tools | 31 August 2018, 21:50:07 UTC |
7fa6f1b | Valentin Churavy | 22 August 2018, 21:26:05 UTC | bump LLVMBuilder to rel 2 | 31 August 2018, 21:50:07 UTC |
f6fcd82 | Valentin Churavy | 22 August 2018, 18:38:26 UTC | install lit and FileCheck to tools | 31 August 2018, 21:50:07 UTC |
804d12e | Sacha Verweij | 31 August 2018, 02:54:48 UTC | Merge pull request #28977 from NHDaly/patch-2 Fix History `eye` deprecation PR: should be 24415 | 31 August 2018, 02:54:48 UTC |
dabf70e | å¼ å®žå”¯ | 30 August 2018, 03:56:04 UTC | fix typo in `?@specialize` | 30 August 2018, 18:40:10 UTC |
8555aa0 | Andrew Janke | 29 August 2018, 19:55:16 UTC | Getting Started: Capitalize "CPU" for consistency | 30 August 2018, 17:45:24 UTC |
13cfbfe | Nathan Daly | 30 August 2018, 17:14:19 UTC | Fix History `eye` deprecation PR: should be 24415 The History.md file linked to #24413 for `eye` deprecation in 0.7. This commit changes it to link to #24415. | 30 August 2018, 17:14:19 UTC |
2346be8 | Israel Herraiz | 29 August 2018, 21:26:58 UTC | Fix call to round for Julia 1.0 (#28944) | 29 August 2018, 21:26:58 UTC |
bd1e46a | timkittel | 29 August 2018, 21:26:31 UTC | removed outdated paragraph on automatic broadcasting for binary operators (#28950) | 29 August 2018, 21:26:31 UTC |
0dd1a9a | Ian Allison | 29 August 2018, 11:45:51 UTC | Pkg docs: consistent home dir; show build.jl contents (#28932) * Update index.md In the "adding a build step" section of "Creating your own packages", the output of the build step is shown, but the contents of `build.jl` are not shown. This is where the file is first described so I've added an example which should give the output shown. * remove trailing whitespace make check-whitespace caught an extra space in the first line of the change. This has been removed. Signed-off-by: Ian Allison <iana@pims.math.ca> * Make the homedir used consistent throughout example The first path given in the DEPOT_PATH is `/users/me/.julia`, but when the search paths are expanded later in the example `/home/me/.julia` is used instead. Signed-off-by: Ian Allison <iana@pims.math.ca> | 29 August 2018, 11:45:51 UTC |
39e5742 | Jim Garrison | 29 August 2018, 05:09:04 UTC | Fix typo in accumulate docstring (#28939) | 29 August 2018, 05:09:04 UTC |
fc27b48 | staticfloat | 28 August 2018, 22:31:53 UTC | Use lambda-initialization for `ptls_getter` to reap rewards of `static` again | 29 August 2018, 04:24:06 UTC |
595e53d | staticfloat | 26 August 2018, 22:51:22 UTC | Explain `ifunc` AArch64 woes a little more thoroughly. | 29 August 2018, 04:24:06 UTC |
4a5de48 | staticfloat | 26 August 2018, 22:49:30 UTC | Add `dlclose(::Nothing)` that does nothing | 29 August 2018, 04:24:06 UTC |
7a3c9df | staticfloat | 26 August 2018, 22:49:18 UTC | Add `JL_THREAD_LOCAL` `#define` | 29 August 2018, 04:24:06 UTC |
99776ad | staticfloat | 26 August 2018, 22:08:31 UTC | Fix botching of compile-time link bug | 29 August 2018, 04:24:06 UTC |
01e0c62 | staticfloat | 26 August 2018, 22:01:11 UTC | Make `jl_dlerror()` thread-safe | 29 August 2018, 04:24:06 UTC |
e6f1bd7 | staticfloat | 24 August 2018, 23:36:47 UTC | Perform the same treatment to `dlopen()` for consistency's sake This changes `dlopen()` to mimic `dlsym()` so that a negative result returns `nothing`. While not necessary in the same way as in the `dlsym()` case (there are no valid `NULL` `dlopen()` results) the consistency is worth it. | 29 August 2018, 04:24:06 UTC |
68cc9f0 | staticfloat | 24 August 2018, 23:09:30 UTC | Rework `Libdl.dlsym()` and `jl_dlsym()`. * Remove `jl_dlsym_e()` to instead be rolled into `jl_dlsym()` with a `throw_err` parameter, similar to `jl_load_dynamic_library()`. * Fix https://github.com/JuliaLang/julia/issues/28881 by having `Libdl.dlsym()` return `nothing` on missing symbol, rather than `C_NULL`. | 29 August 2018, 04:24:06 UTC |
255030e | Páll Haraldsson | 28 August 2018, 08:42:56 UTC | PTX/CUDAnative no longer needs a source build | 28 August 2018, 17:01:10 UTC |
59c7ced | Keno Fischer | 27 August 2018, 14:51:06 UTC | Add a helpful assertion in jl_gc_big_alloc Would have caught some bringup mistakes that cost me a bunch of time today. | 28 August 2018, 15:47:52 UTC |
b553d86 | Keno Fischer | 27 August 2018, 00:12:24 UTC | Make jl_ more robust during early startup When bringing up the runtime on a new architecture, this can be quite broken early on (including mesed up memory allocation, etc). In those situations, jl_ (along with regular printf) is an indispensible tool. However, it was assuming that variables are already initialized and crashed during very early bootstrap when this wasn't the case. Add some extra checks to make it a bit more robust. | 28 August 2018, 15:47:52 UTC |
3ab56f1 | Rafael Fourquet | 27 August 2018, 10:53:42 UTC | fix typo in append! doc | 27 August 2018, 10:53:42 UTC |
5013cfa | Rafael Fourquet | 27 August 2018, 08:15:56 UTC | accumulate: fix typo in docstring | 27 August 2018, 08:15:56 UTC |
a2a1506 | Andreas Noack | 26 August 2018, 10:17:16 UTC | Compare Symbols and Chars more efficiently in Cholesky (#28873) | 26 August 2018, 10:17:16 UTC |
7314249 | Kristoffer Carlsson | 26 August 2018, 10:16:46 UTC | improve performance for string(...) (#28876) * improve performance for various string methods | 26 August 2018, 10:16:46 UTC |
24de931 | Tim Holy | 26 August 2018, 09:51:56 UTC | issubset: check IteratorSize trait before calling length (#28871) | 26 August 2018, 09:51:56 UTC |
304cbd9 | Jan Magnusson | 26 August 2018, 09:50:28 UTC | improve docstring of pathof (#28817) * improve docstring of pathof | 26 August 2018, 09:50:28 UTC |
b18f589 | Scott Thomas | 25 August 2018, 23:43:25 UTC | Fix typo in length(::AbstractString) doc [ci skip] (#28867) | 25 August 2018, 23:43:25 UTC |
9107365 | Keno Fischer | 25 August 2018, 22:06:05 UTC | Update variables after normalizing varargs (#28812) This code was a bit odd. It started by normalzing the varargs, but then used `va0` and `va1` from the original, unnormalized type. Either the unormalized version was intended in which case normalizing first was wasteful, or the normalized version was intended in which case the code was incorrct. Talking with Jeff, he believes the latter was intended, so update the code to do that instead. | 25 August 2018, 22:06:05 UTC |
a5bfabc | Ronan Arraes Jardim Chagas | 25 August 2018, 15:46:24 UTC | Fix note tag in the documentation (#28877) | 25 August 2018, 15:46:24 UTC |
7e1f739 | Kristoffer Carlsson | 25 August 2018, 15:45:59 UTC | fix doccing call overload without named argument (#28875) | 25 August 2018, 15:45:59 UTC |
145224f | Seth Bromberger | 24 August 2018, 22:25:31 UTC | Support ranges with different integer widths in `preduce` (#28651) | 24 August 2018, 22:25:31 UTC |
35c67e5 | Harmen Stoppels | 24 August 2018, 20:56:39 UTC | Make sure iterate(::Tuple) does not throw (#28847) | 24 August 2018, 20:56:39 UTC |
eb8a933 | Fredrik Ekre | 24 August 2018, 19:12:14 UTC | Absolutify project path specified with --project and JULIA_PROJECT. (#28625) | 24 August 2018, 19:12:14 UTC |
d55b044 | Matt Bauman | 23 August 2018, 23:25:13 UTC | Fix ambiguity test failure when arrayops and LinearAlgebra run on the same worker (#28836) | 23 August 2018, 23:25:13 UTC |
293c904 | Paul Bartlett | 23 August 2018, 23:24:03 UTC | Correct typeof result for one of the Union types (#28843) This union is type-equivalent to `Real`, so the output is a bit misleading, since this just evaluates `typeof(Real)` in disguised form. Remove as suggested by yuyichao. | 23 August 2018, 23:24:03 UTC |
157fb34 | tchr | 23 August 2018, 19:00:26 UTC | fix doc-string references to bit-rotted variables x(->v) and v(->itr) in var, varm, std, stdm, & quantiles | 23 August 2018, 23:21:30 UTC |
cd850de | Jeff Bezanson | 23 August 2018, 19:21:11 UTC | fix #28593, macro hygiene problems in type definitions (#28706) | 23 August 2018, 19:21:11 UTC |
6d4ff5e | Israel Herraiz | 23 August 2018, 10:03:13 UTC | @schedule is deprecated, the code is no longer valid in 1.0 (#28800) * @schedule is deprecated, the code is no longer valid in 1.0 * Additional changes for 1.0 compliance | 23 August 2018, 10:03:13 UTC |
472c178 | Rafael Fourquet | 23 August 2018, 10:02:46 UTC | make cmp with BigInt return in [-1, 0, 1] (#28780) * make cmp with BigInt return in [-1, 0, 1] | 23 August 2018, 10:02:46 UTC |
b6756c5 | Sukera | 23 August 2018, 10:01:59 UTC | Added seekstart to getpass in base/util.jl (#28828) (#28832) | 23 August 2018, 10:01:59 UTC |
31c9ae9 | Christian Kurz | 23 August 2018, 10:00:57 UTC | add `Regex` documentation (#28703) add `Regex` documentation. (#26919) | 23 August 2018, 10:00:57 UTC |
b411a58 | Nikita Sirgienko | 23 August 2018, 07:16:28 UTC | Update info about `versioninfo` in devdocs (#28810) | 23 August 2018, 07:16:28 UTC |
f9f3c4f | Fredrik Ekre | 22 August 2018, 21:05:50 UTC | Patch Documenter v0.19.6 to create a correct version selector Co-authored-by: Morten Piibeleht <morten.piibeleht@gmail.com> Co-authored-by: Fredrik Ekre <fredrik.ekre@chalmers.se> | 23 August 2018, 05:51:15 UTC |
c366143 | Fredrik Ekre | 21 August 2018, 12:16:38 UTC | doc: change the canonical url to v1 instead of stable. | 23 August 2018, 05:51:15 UTC |
f31e28a | Takafumi Arakaki | 22 August 2018, 03:56:41 UTC | Run ambiguity test inside a clean process; fixes 28804 | 22 August 2018, 23:20:46 UTC |
6a2dcc6 | Jutho | 22 August 2018, 22:09:59 UTC | Make copyto!(A,I) work for rectangular matrices (#28790) * Make copyto!(A,I) work for rectangular matrices * add tests for `copyto!` add tests for `copyto!(::Matrix, ::UniformScaling)` * fix whitespace | 22 August 2018, 22:09:59 UTC |
85cc84f | Fredrik Ekre | 22 August 2018, 21:36:24 UTC | Add a short warning to the doc-frontpage if the docs (#28791) are for an in-development version of Julia. | 22 August 2018, 21:36:24 UTC |
351bcd5 | Keno Fischer | 22 August 2018, 00:20:55 UTC | More analyzer annotations This gets us far enough to pass all the .c files without warning. The .cpp files are another issue (partly because C++ is more complicated, partly because they play a bit fast and lose with GC rooting). However, once this is merged, we should be able to set up CI for the .c files at least. | 22 August 2018, 21:16:58 UTC |
f2f1f43 | Keno Fischer | 21 August 2018, 18:30:04 UTC | Switch jl_typemap_t from unions to bare pointers The clang static analyzer has very poor support for unions. Thus in an attempt to get the GC analysis through without any false positives, stop using unions for typemaps. This commit should have no changes other than that. GC annotations for this code will come after I rip out the hacks to support unions from the static analyzer. | 22 August 2018, 14:51:41 UTC |
81850b6 | Kristoffer Carlsson | 22 August 2018, 10:10:22 UTC | improve performance of parse (#28787) | 22 August 2018, 10:10:22 UTC |
81b04cf | Keno Fischer | 20 August 2018, 02:15:10 UTC | Some more work on gf.c | 21 August 2018, 16:26:02 UTC |
2caef76 | Keno Fischer | 20 August 2018, 02:04:19 UTC | Make a bunch of progress on anntations for typemap.c | 21 August 2018, 16:26:02 UTC |
8af9d37 | Keno Fischer | 19 August 2018, 21:44:11 UTC | Add static analysis annotations for toplevel.c | 21 August 2018, 16:26:02 UTC |
5a1d3b5 | Keno Fischer | 19 August 2018, 21:01:37 UTC | Add static analysis annotations for threading.c | 21 August 2018, 16:26:02 UTC |
f82084c | Keno Fischer | 19 August 2018, 20:51:59 UTC | Add static analysis annotations for task.c | 21 August 2018, 16:26:02 UTC |
4eeb8f7 | Keno Fischer | 19 August 2018, 19:48:28 UTC | Add static analysis annotations for sys.c | 21 August 2018, 16:26:02 UTC |
4d6d216 | Keno Fischer | 19 August 2018, 19:46:28 UTC | Add static analysis annotations for subtype.c | 21 August 2018, 16:26:02 UTC |
13820d0 | Keno Fischer | 18 August 2018, 22:21:59 UTC | Add static analysis annotations for stackwalk.c | 21 August 2018, 16:26:02 UTC |
06968c1 | Keno Fischer | 18 August 2018, 22:20:46 UTC | Silence analyzer complaints about signal-handling.c | 21 August 2018, 16:26:02 UTC |
db21a81 | Keno Fischer | 18 August 2018, 22:13:07 UTC | Add annotations for runtime_intrinsics.c | 21 August 2018, 16:26:02 UTC |
54dae01 | Keno Fischer | 18 August 2018, 21:28:26 UTC | Add annotations for rtutils | 21 August 2018, 16:26:02 UTC |
9ab3d8f | Keno Fischer | 18 August 2018, 21:10:02 UTC | Add static analysis annotations for precompile.c | 21 August 2018, 16:26:02 UTC |
20792fe | Keno Fischer | 18 August 2018, 21:03:44 UTC | Add static analysis annotations for module.c | 21 August 2018, 16:26:02 UTC |
9ca53d0 | Keno Fischer | 18 August 2018, 20:42:02 UTC | Add static analysis annotations for src/method.c | 21 August 2018, 16:26:02 UTC |
a759c11 | Keno Fischer | 18 August 2018, 20:37:27 UTC | Add gc annotations to jltypes.c | 21 August 2018, 16:26:02 UTC |
ee88b92 | Keno Fischer | 18 August 2018, 19:28:35 UTC | Add static analysis annotations for jloptions.c | 21 August 2018, 16:26:02 UTC |
3708c64 | Keno Fischer | 18 August 2018, 19:21:00 UTC | Add GC static analysis annotations in interpreter.c | 21 August 2018, 16:26:02 UTC |
b108d51 | Keno Fischer | 18 August 2018, 18:51:07 UTC | static analysis annotations for init.c | 21 August 2018, 16:26:02 UTC |
dce8cc2 | Keno Fischer | 18 August 2018, 18:48:57 UTC | Partial GC annotations for gf.c This file is hard. Let's come back to it later. | 21 August 2018, 16:26:02 UTC |
9a380fa | Keno Fischer | 18 August 2018, 02:04:52 UTC | Add static analysis annotations for gc.c | 21 August 2018, 16:26:02 UTC |
4498d27 | Takafumi Arakaki | 21 August 2018, 13:12:58 UTC | Resolve all method ambiguities in LinearAlgebra (#28749) * Add a test to check method ambiguities * Use StridedMatrix to avoid method ambiguity; fixes #27405 * Avoid method ambiguity in promote_leaf_eltypes This change is only for making Test.detect_ambiguities and shouldn't introduce any change in working code (unless it depends on MethodError). Before this change, we have: julia> LinearAlgebra.promote_leaf_eltypes(()) ERROR: MethodError: LinearAlgebra.promote_leaf_eltypes(::Tuple{}) is ambiguous. Candidates: With this change, we have: julia> LinearAlgebra.promote_leaf_eltypes(()) Bool | 21 August 2018, 13:12:58 UTC |
ae0738e | annainfo | 20 August 2018, 17:15:38 UTC | Update integers-and-floating-point-numbers.md (#28744) | 20 August 2018, 17:15:38 UTC |
302a507 | James Shaw | 20 August 2018, 15:46:04 UTC | fix two typos (#28731) | 20 August 2018, 15:46:04 UTC |
f6c48eb | ExpandingMan | 20 August 2018, 09:16:47 UTC | added documentation for read command methods (#28748) * added documentation for read command methods * fixed docs | 20 August 2018, 09:16:47 UTC |
3a54c14 | Kenta Sato | 20 August 2018, 09:09:45 UTC | fix description of kwargs... (#28757) * fix description of kwargs... * clarify the key type of dicts passed as kwargs | 20 August 2018, 09:09:45 UTC |
f7733f8 | Valentin Churavy | 20 August 2018, 04:03:57 UTC | Merge pull request #28745 from JuliaLang/vc/6.0.1 upgrade to LLVM 6.0.1 | 20 August 2018, 04:03:57 UTC |