1
0
Fork 1
mirror of https://github.com/NixOS/nixpkgs.git synced 2024-12-13 16:15:30 +00:00
Commit graph

505 commits

Author SHA1 Message Date
Daiderd Jordan 222524b5e6
gfortran: fix darwin build on hydra
All the warnings about unused -isystem arguments causes the build log to
exceed the hydra limit.
2018-06-24 15:02:49 +02:00
Vladimír Čunát 392e6de7d0
Merge branch 'master' into staging 2018-05-20 13:20:53 +02:00
Matthew Justin Bauer 2c7c9cbbac
gcc5: delete unused patch 2018-05-19 12:06:16 -05:00
Robert Schütz 90c0100139 Merge branch 'master' into staging 2018-05-19 11:12:19 +02:00
Matthew Justin Bauer bd078c1a31
gcc5: remove patch
The patch fails to apply & may not be necessary any more. Trying to remove it for now just to see.

Hydra log: https://hydra.nixos.org/build/72726891/nixlog/1

Also see #33722

/cc @copumpkin
2018-05-18 17:03:35 -05:00
Tuomas Tynkkynen 6b6a575748 Merge remote-tracking branch 'upstream/master' into staging
Conflicts:
	pkgs/tools/networking/telnet/default.nix
	pkgs/top-level/aliases.nix
2018-05-16 03:09:33 +03:00
Bastian Köcher 780592ca6e gcc7: Enable arm erratum 843419 on aarch64
This fix is required for the raspherry pi 3 with glibc 2.27,
otherwise the kernel panics in initrd with:
```
<<< NixOS Stage 1 >>>

loading module dm_mod...
running udev...
kbd_mode: KDSKBMODE: Inappropriate ioctl for device
Gstarting device mapper and LVM...
[    1.969164] Kernel panic - not syncing: Attempted to kill init!
exitcode=0x0000000b
[    1.969164]
[    1.978476] CPU: 0 PID: 1 Comm: init Not tainted 4.16.8 #1-NixOS
[    1.984580] Hardware name: Raspberry Pi 3 Model B (DT)
[    1.989801] Call trace:
[    1.992301]  dump_backtrace+0x0/0x1c8
[    1.996025]  show_stack+0x24/0x30
[    1.999396]  dump_stack+0x9c/0xc0
[    2.002766]  panic+0x124/0x294
[    2.005872]  complete_and_exit+0x0/0x30
[    2.009771]  do_group_exit+0x40/0xa8
[    2.013406]  get_signal+0x280/0x5b0
[    2.016954]  do_signal+0x88/0x240
[    2.020325]  do_notify_resume+0xd8/0x130
[    2.024311]  work_pending+0x8/0x10
[    2.027774] SMP: stopping secondary CPUs
[    2.031763] Kernel Offset: disabled
[    2.035308] CPU features: 0x0802004
[    2.038850] Memory Limit: none
[    2.041963] ---[ end Kernel panic - not syncing: Attempted to kill
init! exitcode=0x0000000b
[    2.041963]
[    2.865264] random: crng init done
```
Suse has done the same to circumvent crashes with hostname resolving in
glibc 2.27 on aarch64.
2018-05-16 03:04:29 +03:00
John Ericson 01a525b84e Merge remote-tracking branch 'upstream/master' into staging 2018-05-14 13:31:20 -04:00
John Ericson bf7d1bae20 Merge remote-tracking branch 'upstream/master' into gcc-hurd-condition 2018-05-14 13:15:06 -04:00
John Ericson 3a9654bfe2 gcc: Simplify this old Hurd condition
Steps towards getting rid of crossConfig.
2018-05-14 13:13:11 -04:00
John Ericson d00cc1242f
Merge pull request #40040 from obsidiansystems/gnu-config-arm
gnu-config: Update, allowing hacks to be removed
2018-05-14 11:20:09 -04:00
John Ericson f2b575bd7b Merge remote-tracking branch 'upstream/master' into gnu-config-arm 2018-05-14 10:58:15 -04:00
John Ericson f472dd7652 Revert "Revert "gcc, binutils: Get rid of 32-bit ARM configure flag exception""
And there's more reverts too. The previous commmit
d838afbc9376bdadb8c690eb00b425f3eeccdf2d to gnu-config finally solves
it!

This reverts commit 3ed545ab31.
2018-05-14 10:55:32 -04:00
John Ericson 2c5d915200 Merge commit '92b7a814f26ee1d37e989431c18518c67285a332' into staging 2018-05-13 01:02:09 -04:00
John Ericson 92b7a814f2 Merge branch 'fix-gcc-with-float' 2018-05-12 15:21:30 -04:00
John Ericson f2004e6287 lib: Fix float handling for Aarch32
Forgot to adjust default so abi with explicit float attr would be used.
2018-05-12 15:18:31 -04:00
John Ericson ee4b56edd3 Merge remote-tracking branch 'upstream/master' into staging 2018-05-11 14:36:08 -04:00
John Ericson f18ddabee7 Merge remote-tracking branch 'upstream/master' into lib-float 2018-05-10 18:13:00 -04:00
John Ericson 1fe81a4bcd lib: Clean up float/fpu options
ARM ABIs now have a float field. This is used as a fallback to lessen
our use of `platform.gcc.float`. I didn't know what the MIPs convention
is so I kept using `platform.gcc.float` in that case.
2018-05-10 18:02:00 -04:00
John Ericson c9f6a82b61 gcc: Factor out "platform flags" 2018-05-10 18:00:57 -04:00
John Ericson 1d90ea9194
Merge pull request #40294 from obsidiansystems/no-crossAttrs
misc packages: Fewer crossAttrs
2018-05-10 14:38:58 -04:00
John Ericson c42118bb82 gcc: Get rid of crossAttrs
Only a little bit was left
2018-05-10 14:19:38 -04:00
Matthew Bauer 6748534d83 Merge remote-tracking branch 'upstream/master' into staging 2018-05-08 09:36:00 -05:00
John Ericson 4f7cdd35d5
Merge pull request #40139 from obsidiansystems/modular-setup-hooks
treewide: Modular setup hooks
2018-05-07 15:32:10 -04:00
John Ericson 2110c0bd30 treewide: Use pkgs/build-support/roles.bash to remove copy pasta
Also fix some setup hooks that unnecessarily used environment hooks,
which revolted in the same variable being modified too many times.
2018-05-07 15:10:37 -04:00
Jörg Thalheim 1537dca7fc
Merge pull request #39914 from nlewo/pr-retpoline
gcc49: backport retpoline support
2018-05-04 19:14:35 +01:00
John Ericson 88d18d2e04
Merge pull request #39947 from obsidiansystems/libcxx-cross
lib{std,}c++: Fix setup hooks for cross
2018-05-03 18:20:13 -04:00
John Ericson 39df5831ca lib{std,}c++: Fix setup hooks for cross 2018-05-03 17:55:03 -04:00
Patrick Hilhorst 0a4622cfe9 gcc-snapshot: Removed java as a target language 2018-05-03 19:48:00 +02:00
Patrick Hilhorst 719810d474 gcc7: Removed X11 components, since they were only used by langJava
See also:

51ac1d8df0
https://github.com/NixOS/nixpkgs/pull/39866#discussion_r185813776
2018-05-03 16:43:51 +02:00
Patrick Hilhorst 51ac1d8df0 gcc8: Removed X11 components, since they were only used by langJava 2018-05-03 16:30:32 +02:00
Antoine Eiche 5b6129f69c gcc49: remove darwin support 2018-05-03 15:12:01 +02:00
Patrick Hilhorst 3641b35bae gcc7: Removed java as a target language 2018-05-03 15:11:41 +02:00
Patrick Hilhorst d21ff87160 gcc8: Removed java as a target language 2018-05-03 13:07:19 +02:00
Antoine Eiche ada2fc088c gcc49: backport retpoline support
To mitigate Spectre Variant 2, GCC needs to have retpoline
support (-mindirect-branch and -mfunction-return arguments on amd64
and i386).

Patches were pulled from H.J. Lu's backport branch to
4.9 (hjl/indirect/gcc-4_9-branch), available at
https://github.com/hjl-tools/gcc/tree/hjl/indirect/gcc-4_9-branch/master. Upstream
GCC does not apply patches to anything older than the
gcc-6-branch. H.J. Lu is the author of the upstream retpoline commits
as well.

Several Linux distributions already backported these patches to GCC 4
branches and some old kernels (3.13 for instance) have been recompiled
with these GCC patches. These kernels only allow to load kernel
modules that are compiled with the retpoline support.

References:
- Ubuntu bug: https://bugs.launchpad.net/ubuntu/+source/gcc-4.8/+bug/1749261
- Ubuntu package: https://launchpad.net/ubuntu/+source/gcc-4.8/4.8.4-2ubuntu1~14.04.4

Fixes #38394
2018-05-03 11:35:49 +02:00
Matthew Justin Bauer eeb016e8f0
Merge branch 'staging' into fix-ncurses-darwin-extensions 2018-05-02 15:40:38 -05:00
Patrick Hilhorst 5794d94136 gcc8: made the remove-selftests patch more robust, removed old hash 2018-05-02 22:31:58 +02:00
Patrick Hilhorst c7a6b36658 gcc8: removed commented gcc7 hash 2018-05-02 20:53:13 +02:00
Patrick Hilhorst d0708a30c4 gcc8: removed merged patches
Log:
```
patching sources
applying patch /nix/store/6m27y27zvzsjn1ir4y8mm9nc9xnh2sgx-riscv-no-relax.patch
patching file gcc/config/riscv/riscv.c
Reversed (or previously applied) patch detected!  Assume -R? [n]
Apply anyway? [n]
Skipping patch.
1 out of 1 hunk ignored -- saving rejects to file gcc/config/riscv/riscv.c.rej
patching file gcc/config/riscv/riscv.opt
Reversed (or previously applied) patch detected!  Assume -R? [n]
Apply anyway? [n]
Skipping patch.
1 out of 1 hunk ignored -- saving rejects to file gcc/config/riscv/riscv.opt.rej
patching file gcc/doc/invoke.texi
Reversed (or previously applied) patch detected!  Assume -R? [n]
Apply anyway? [n]
Skipping patch.
```
2018-05-02 20:51:01 +02:00
Patrick Hilhorst 728eb27618 gcc8: added myself as maintainer 2018-05-02 18:52:50 +02:00
Patrick Hilhorst 659363fb40 gcc8: init at 8.1.0 2018-05-02 18:34:28 +02:00
John Ericson 591d8c7a19
Merge pull request #39458 from oxij/stdenv/beautifications
stdenv: better names for cc and bintools
2018-04-26 11:41:18 -04:00
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
Jan Malakhovski 7438083a4d tree-wide: disable doCheck and doInstallCheck where it fails (the trivial part) 2018-04-25 04:18:46 +00:00
Jan Malakhovski 2afd3c901e gcc: prepend crossNameAddon instead of appending it (like binutils does) 2018-04-25 00:41:14 +00:00
Shea Levy d0a8866622
Merge branch 'binutils-2.30' into staging 2018-03-22 07:28:10 -04:00
Tuomas Tynkkynen 74720a36c7 Merge remote-tracking branch 'upstream/master' into staging 2018-03-21 23:51:55 +02:00
Clemens Fruhwirth c0cf13e52e Add space for NIX_CXXSTDLIB_COMPILE+=' ...' as NIX_CXXSTDLIB_COMPILE might not be empty. 2018-03-19 13:18:49 +01:00
Shea Levy a5d064826b
gcc: Try to fix /bin/sh dependency 2018-03-17 21:58:15 -04:00