Commit graph

592 commits

Author SHA1 Message Date
John Ericson 5bdfe37daf
Merge pull request #40154 from obsidiansystems/stdenv-arm-alias
stdenv: Put back isArm, with deprecation notice.
2018-05-08 14:24:58 -04:00
John Ericson 302c4c5f2d stdenv: Put back isArm, with deprecation notice.
This was always meant to be deprecated rather than removed.
2018-05-07 20:14:52 -04:00
Michael Raskin 6ca343103b check-meta.nix: specify meta.timeout 2018-04-27 00:25:37 +02:00
John Ericson b9acfb4ecf treewide: isArm -> isAarch32
Following legacy packing conventions, `isArm` was defined just for
32-bit ARM instruction set. This is confusing to non packagers though,
because Aarch64 is an ARM instruction set.

The official ARM overview for ARMv8[1] is surprisingly not confusing,
given the overall state of affairs for ARM naming conventions, and
offers us a solution. It divides the nomenclature into three levels:

```
ISA:             ARMv8   {-A, -R, -M}
                 /    \
Mode:     Aarch32     Aarch64
             |         /   \
Encoding:   A64      A32   T32
```

At the top is the overall v8 instruction set archicture. Second are the
two modes, defined by bitwidth but differing in other semantics too, and
buttom are the encodings, (hopefully?) isomorphic if they encode the
same mode.

The 32 bit encodings are mostly backwards compatible with previous
non-Thumb and Thumb encodings, and if so we can pun the mode names to
instead mean "sets of compatable or isomorphic encodings", and then
voilà we have nice names for 32-bit and 64-bit arm instruction sets
which do not use the word ARM so as to not confused either laymen or
experienced ARM packages.

[1]: https://developer.arm.com/products/architecture/a-profile

(cherry picked from commit ba52ae5048)
2018-04-25 15:50:41 -04:00
John Ericson ba52ae5048 treewide: isArm -> isAarch32
Following legacy packing conventions, `isArm` was defined just for
32-bit ARM instruction set. This is confusing to non packagers though,
because Aarch64 is an ARM instruction set.

The official ARM overview for ARMv8[1] is surprisingly not confusing,
given the overall state of affairs for ARM naming conventions, and
offers us a solution. It divides the nomenclature into three levels:

```
ISA:             ARMv8   {-A, -R, -M}
                 /    \
Mode:     Aarch32     Aarch64
             |         /   \
Encoding:   A64      A32   T32
```

At the top is the overall v8 instruction set archicture. Second are the
two modes, defined by bitwidth but differing in other semantics too, and
buttom are the encodings, (hopefully?) isomorphic if they encode the
same mode.

The 32 bit encodings are mostly backwards compatible with previous
non-Thumb and Thumb encodings, and if so we can pun the mode names to
instead mean "sets of compatable or isomorphic encodings", and then
voilà we have nice names for 32-bit and 64-bit arm instruction sets
which do not use the word ARM so as to not confused either laymen or
experienced ARM packages.

[1]: https://developer.arm.com/products/architecture/a-profile
2018-04-25 15:28:55 -04:00
John Ericson 53686e8995
Merge pull request #38485 from obsidiansystems/nixos-nixpkgs-options
nixpkgs module: Clean up platform options
2018-04-19 14:59:58 -04:00
John Ericson cb212cf549 meta: Add NIXPKGS_ALLOW_UNSUPPORTED_SYSTEM for consistency 2018-04-17 16:02:50 -04:00
Shea Levy 3955b84698 meta: Don't bypass unsupported platforms with allowBroken.
Our platforms are open-world oriented these days, and anyway there's allowUnsupportedSystem.
2018-04-17 16:02:15 -04:00
John Ericson c6f7d43678 nixpkgs module: Clean up platform options
- `localSystem` is added, it strictly supercedes system

 - `crossSystem`'s description mentions `localSystem` (and vice versa).

 - No more weird special casing I don't even understand

TEMP
2018-04-06 12:41:44 -04:00
John Ericson f8ed783f4f meta: Simplify platform check logic
Code golf or readability, you decide
2018-03-27 11:59:59 -04:00
Shea Levy b0482248fe
meta: Add badPlatforms attribute for platform blacklisting. 2018-03-27 08:12:45 -04:00
John Ericson 3c331bff5b
Merge pull request #37395 from obsidiansystems/lib-meta-platform
lib: Factor in tiny bit of `meta.platform` checking
2018-03-19 20:12:50 -04:00
John Ericson e547bd0dc4 lib: Factor in tiny bit of meta.platform checking
I need it in stdenv and release-lib, so that seems motivation enough.
2018-03-19 19:29:16 -04:00
John Ericson 2fa2197a96
Merge pull request #34444 from obsidiansystems/meta-check
lib: Fix #30902
2018-03-18 13:51:03 -04:00
John Ericson c26252af3e lib, stdenv: Check meta.platforms against host platform and be open world
First, we need check against the host platform, not the build platform.
That's simple enough.

Second, we move away from exahustive finite case analysis (i.e.
exhaustively listing all platforms the package builds on). That only
work in a closed-world setting, where we know all platforms we might
build one. But with cross compilation, we may be building for arbitrary
platforms, So we need fancier filters. This is the closed world to open
world change.

The solution is instead of having a list of systems (strings in the form
"foo-bar"), we have a list of of systems or "patterns", i.e. attributes
that partially match the output of the parsers in `lib.systems.parse`.
The "check meta" logic treats the systems strings as an exact whitelist
just as before, but treats the patterns as a fuzzy whitelist,
intersecting the actual `hostPlatform` with the pattern and then
checking for equality. (This is done using `matchAttrs`).

The default convenience lists for `meta.platforms` are now changed to be
lists of patterns (usually a single pattern) in
`lib/systems/for-meta.nix` for maximum flexibility under this new
system.

Fixes #30902
2018-03-15 00:44:34 -04:00
John Ericson 4c52e34ca6 stdenv: Clean up check meta args 2018-03-14 18:58:07 -04:00
Jan Malakhovski 1d4f3fa371 stdenv: check-meta: fix fallout from #36119 2018-03-06 14:14:17 +00:00
Shea Levy 1c1a6dfd23
libgcrypt: Fix cross-compilation 2018-02-24 22:51:22 -05:00
Vladimír Čunát 1d15dadbec
Merge branch 'master' into staging
Larger rebuilds from master.
2018-02-20 20:33:40 +01:00
Jan Malakhovski 810c4702cf stdenv.mkDerivation: rename meta.evaluates -> meta.available
A much better name.
2018-02-18 13:33:25 +00:00
Shea Levy b24ce2ae63
Handle sourceRoots with leading dashes 2018-02-15 11:50:24 -05:00
Tuomas Tynkkynen a6fd03876e check-meta: Use concatStrings 2018-02-11 00:17:47 +02:00
Shea Levy f83b6e1130
unpackPhase: Handle sources starting with a hyphen 2018-01-24 21:58:57 -08:00
John Ericson 5a754e75b4 Merge branch 'ericson2314-cross-master' into staging 2018-01-16 13:05:39 -05:00
Vladimír Čunát 67e8392383
Merge #33057: stdenv meta checks: make them lazy
Closes #22277 - it's superseded;  I have some WIP on evaluation
performance, but best do that in a separate PR/thread.
2018-01-14 21:41:31 +01:00
Tuomas Tynkkynen 6ed0fe7e45 Merge remote-tracking branch 'upstream/master' into staging
Conflicts:
	pkgs/build-support/fetchbower/default.nix
	pkgs/build-support/fetchdarcs/default.nix
	pkgs/build-support/fetchgx/default.nix
	pkgs/development/python-modules/botocore/default.nix
	pkgs/os-specific/linux/firmware/firmware-linux-nonfree/default.nix
	pkgs/tools/admin/awscli/default.nix
2018-01-14 21:18:27 +02:00
Jan Malakhovski fac3d49e48 stdenv: provide meta.name 2018-01-14 13:24:30 +00:00
Jan Malakhovski 50148f0630 stdenv: hide name under check-meta assert
This is a temporary workaround to make `nix-env -qa` and `nix search` ignore
broken packages as they they did before this patchset.

This patch should be reverted after `nix` gets a proper fix for this.
See NixOS/nix#1771.
2018-01-14 13:24:30 +00:00
Jan Malakhovski eaee2a1199 stdenv: implement config.checkMetaRecursively
This option makes `meta.evaluate` into a close approximation of the result of
evaluating `.outPath` by checking all the dependencies recursively at a cost of
2x slowdown. Note that actually evaluating `.outPath` costs some
5x-7x more because `.outPath` also computes all the hashes.
2018-01-14 13:24:30 +00:00
Jan Malakhovski ecd3990cd4 stdenv: provide meta.evaluates
This gives a way to see the result of `check-meta` without triggering any assertions.
2018-01-14 13:24:29 +00:00
Jan Malakhovski 9956687151 stdenv: change some indent 2018-01-14 13:24:22 +00:00
John Ericson c836910e0e
Merge pull request #33670 from obsidiansystems/cross-check
stdenv: Fix doCheck and doInstallCheck logic
2018-01-09 16:37:11 -05:00
John Ericson 567feb3a54 stdenv: Fix doCheck and doInstallCheck logic
It's host != build, not host != target
2018-01-09 16:22:58 -05:00
John Ericson 06a8d66528
Merge pull request #33603 from obsidiansystems/cross-check
stdenv: Force `doCheck` to be false when we are cross compiling
2018-01-09 15:09:54 -05:00
John Ericson 4e907dbca1 stdenv: Force doCheck and doInstallCheck to be false when we are cross compiling
I hope this will be a temporary measure. If there is consensus around
issue #33599, then we can follow an explicit `dontCheck`, but default to
not checking during cross builds when none is given.
2018-01-09 12:37:12 -05:00
Will Dietz 9721ed22e8 schedulingPriority should be an int, fix check-meta type and in-tree use 2018-01-09 07:25:24 -06:00
adisbladis c2316114bc
stdenv: Kill off ensureDir 2018-01-09 11:14:48 +08:00
Jan Malakhovski d1d5ecb3bf stdenv: perform checks only when evaluating .drv and .out
This pushes check-meta evaluation to derivation evaluation step, leaving all other
attributes accessible.

Before this commit:

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen --argstr system aarch64-linux
> Package ‘xen-4.5.5’ in pkgs/applications/virtualization/xen/generic.nix:226 is not supported on ‘aarch64-linux’, refusing to evaluate.

as expected

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen.name --argstr system aarch64-linux
> Package ‘xen-4.5.5’ in pkgs/applications/virtualization/xen/generic.nix:226 is not supported on ‘aarch64-linux’, refusing to evaluate.

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen.meta.description --argstr system aarch64-linux
> Package ‘xen-4.5.5’ in pkgs/applications/virtualization/xen/generic.nix:226 is not supported on ‘aarch64-linux’, refusing to evaluate.

which is unfortunate since its impossible to use packages in autogenerated
documentation on all platforms.

After this commit:

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen --argstr system aarch64-linux

still fails

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen.name --argstr system aarch64-linux
> "xen-4.5.5"

> $ HOME=/homeless-shelter NIX_PATH=nixpkgs=$(pwd) nix-instantiate --eval --strict ./default.nix -A xen.meta.description --argstr system aarch64-linux
> "Xen hypervisor and related components (vanilla)"
2018-01-03 12:20:25 +00:00
John Ericson 469fd89832 stdenv-setup: Ease the transition with native builds
- All deps go on the PATH

 - CC and Bintools wrappers with their host != depender's host still get their
   setup hooks run.

 - Environment hooks get applied to all packages

This isn't so elegent, but eases the transition on a very significant
PR.
2017-12-30 22:04:23 -05:00
John Ericson a036473a0a {bintools,cc}-wrapper: Fix setup hook to respect the role of the cc-compiler
We now have the information to properly determine the role the
cc-wrapper dependency has, by taking advantage of `offset`. No longer
use the soon-to-be-deprecated crossConfig environment variable, the
temp hack used before this change.
2017-12-30 22:04:21 -05:00
John Ericson 7f3ca3e21a stdenv: Fix handling of dependencies and hooks
4 far-reaching changes: Smaller PATH, New vars, different propagation
logic, and different hook logic

Smaller PATH
------------

`buildInputs` no longer go on the PATH at build time, as they cannot be
run when cross compiling and we don't want to special case. Simply make
a `nativeBuildInput` too if one needs them on the PATH. Fixes #21191.

Many new depedendency variables
-------------------------------

See the stdenv chapter of the nixpkgs manual. I pulled out the existing
documentation of dependency specification into a new section, and added
language for these two (and their propagated equivalents) along side
the others'.

More complex propagation logic
------------------------------

Before a propagated*XXX*Input always acted as if it was specified
directly as a *XXX*Input downstream. That's simple enough, but violates
the intended roles of each sort of dep, which has functional and not
just stylistic consequences.

The new algorithm is detailed in the manual, and ensures everything
ends up in the right place. I tried to give both an informal and formal
description, but I suspect in practice it will not make much sense
until one tries cross compiling, after which it will immediately make
sense as the only sane option.

Simplified hook logic
---------------------

Rather than `envHook` and `crossEnvHook`, whose behavior differs
depending on whether we are cross compiling or not, there is now one
hook per sort (or rather non-propagated and propagated pair of sorts)
of dependency. These new hooks have the same meaning regardless of
cross compilation. See the setup hook section of stdenv chapter of the
Nixpkgs manual for more details.
2017-12-30 22:04:21 -05:00
Graham Christensen ae9a7c4969
stdenv: make knownVulnerabilities a known meta-type 2017-12-14 07:36:31 -05:00
Graham Christensen f33a513d2b
stdenv: allow specifying a eval issuee handler 2017-12-12 18:08:11 -05:00
Vladimír Čunát 76bf375a16
stdenv checkMeta: throw -> trace
- tracing seems annoying enough
- we get errors for all packages instead of aborting on the first one
- easier to differentiate from unwanted packages (broken, unfree, etc.)
2017-12-12 18:07:07 -05:00
Will Dietz db7bee240e stdenv: fix use of config-specified allowInsecurePredicate
for some reason we were checking allowUnfreePredicate instead
2017-12-09 20:43:30 +00:00
Jan Malakhovski 1858e8909e tree-wide: fix either check-meta or meta attrs of all the packages I evaluate 2017-12-05 13:46:52 +01:00
Jan Malakhovski 8ae51ff9c1 stdenv/generic/check-meta: fix error message evaluation 2017-12-05 13:46:52 +01:00
John Ericson da19c34d0f stdenv setup: Always use both propagated files
This continues #23374, which always kept around both attributes, by
always including both propagated files: `propgated-native-build-inputs`
and `propagated-build-inputs`. `nativePkgs` and `crossPkgs` are still
defined as before, however, so this change should only barely
observable.

This is an incremental step to fully keeping the dependencies separate
in all cases.
2017-11-21 10:44:44 -05:00
John Ericson 6a5cda5131 stdenv setup: Run setup hooks and other processing after accumulating deps
I find the separation of concerns, accumulating, then processing, easier
to follow. Also, with my yet-to-be-merged cross work, the accumulation
part will become more complex.
2017-11-15 18:51:06 -05:00
Dan Peebles b426c85ce2 Get rid of most @rpath nonsense on Darwin
This requires some small changes in the stdenv, then working around the
weird choice LLVM made to hardcode @rpath in its install name, and then
lets us remove a ton of annoying workaround hacks in many of our Go
packages. With any luck this will mean less hackery going forward.
2017-10-08 16:13:46 -04:00