Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
Obtain sha256 via nix-prefetch-url
Update sha256
|
|
Julia does not actually require python2; it just requires any version of python
[that is equal to or greater than 2.7](https://github.com/JuliaLang/julia/blob/master/doc/src/devdocs/build/build.md#required-build-tools-and-external-libraries).
This changes the python dependency from python2 to python3.
I have verified that julia-1.0 builds with this change. julia-1.5 is marked as broken
in nixpkgs, so I have not tested it (but I would be very surprised if this change
broke it further).
Motivated by #148779
|
|
Thank you @7c6f434c for spotting these.
|
|
No longer supported upstream:
https://discourse.julialang.org/t/julia-v1-7-0-has-been-released-and-v1-6-is-now-lts/72336
|
|
https://github.com/JuliaLang/julia/blob/v1.7.0/NEWS.md
|
|
Patches: https://github.com/JuliaLang/julia/compare/v1.6.3...v1.6.4
Broken `tempname` tests disabled upstream, dropping workaround patch:
https://github.com/JuliaLang/julia/pull/43012
The issue itself however, persists:
https://github.com/JuliaLang/julia/issues/35785
|
|
Patches: https://github.com/JuliaLang/julia/compare/v1.6.2...v1.6.3
Dropping `chown` test workaround patch as it is now fixed upstream:
https://github.com/JuliaLang/julia/commit/3e8eb9e8deb72f5bf231dbde3a48eae69c558947
|
|
Patches: https://github.com/JuliaLang/julia/compare/v1.6.1...v1.6.2
|
|
|
|
|
|
Fixes https://github.com/NixOS/nixpkgs/issues/121101.
|
|
|
|
Almost as clean as it could be, but was forced to set `LD_LIBRARY_PATH`
to work around a `Libdl` failure that is unique to v1.0.x.
|
|
Relevant patches lifted from the `release-1.6` branch of my Julia fork:
https://git.sr.ht/~ninjin/julia-nix/tree/release-1.6
|
|
Provides a few hopefully helpful pointers that would not work well as
inline comments in the expressions themselves. Most likely the README
will need to be expanded upon over time to cover how we handle the Julia
release process, but I hope this is a good starting point.
|
|
Provides very little comfort compared what is outlined in the
manual [1], only supports a single version, and would probably be better
to implement as a general Nixpkg tool.
[1]: https://nixos.org/manual/nixpkgs/stable/#sec-source-hashes
|
|
Makes the top-level directory organisation easier with an increasing
number of patches.
|
|
Fix merged upstream and backported to Julia 1.0.5:
https://github.com/JuliaLang/julia/pull/31443
|
|
As far as I can tell this patch is redundant as all pre-compiled code
generated at build time is baked into the Julia system image and will
thus never get invalidated: Note that for both julia_10 and julia_15
there are no `.ji` files produced in the derivations.
|
|
Erroneously disabled by 3ae5e6ce03f9dbf3f0a0a3e3161c83e28c1b45af as it
mistook Julia for using CMake (it is used by some of the vendored
dependencies).
|
|
Closes #82008.
|
|
|
|
|
|
top-level, lib: Remove platform attribute of platforms
|
|
Second attempt of 8929989614589ee3acd070a6409b2b9700c92d65; see that
commit for details.
This reverts commit 0bc275e63423456d6deb650e146120c39c1e0723.
|
|
|
|
This is a stdenv-rebuild, and should not be merged
into master
This reverts commit 8929989614589ee3acd070a6409b2b9700c92d65.
|
|
The `platform` field is pointless nesting: it's just stuff that happens
to be defined together, and that should be an implementation detail.
This instead makes `linux-kernel` and `gcc` top level fields in platform
configs. They join `rustc` there [all are optional], which was put there
and not in `platform` in anticipation of a change like this.
`linux-kernel.arch` in particular also becomes `linuxArch`, to match the
other `*Arch`es.
The next step after is this to combine the *specific* machines from
`lib.systems.platforms` with `lib.systems.examples`, keeping just the
"multiplatform" ones for defaulting.
|
|
|
|
|
|
Don't use ./shared.nix as it's not shared anymore.
Fix tests with an upstream patch and a few commands in postPatch.
Assert that blas and lapack are compatible.
Rename utf8proc patch.
|
|
|
|
|
|
|
|
|
|
This is a better name since we have multiple 64-bit things that could
be referred to.
LP64 : integer=32, long=64, pointer=64
ILP64 : integer=64, long=64, pointer=64
|
|
This makes packages use lapack and blas, which can wrap different
BLAS/LAPACK implementations.
treewide: cleanup from blas/lapack changes
A few issues in the original treewide:
- can’t assume blas64 is a bool
- unused commented code
|
|
|
|
|
|
Added all-packages.nix
|
|
|
|
|
|
There ver very many conflicts, basically all due to
name -> pname+version. Fortunately, almost everything was auto-resolved
by kdiff3, and for now I just fixed up a couple evaluation problems,
as verified by the tarball job. There might be some fallback to these
conflicts, but I believe it should be minimal.
Hydra nixpkgs: ?compare=1538299
|
|
|
|
|