gnu-config: Update, allowing hacks to be removed #40040
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dezgeg
reviewed
May 6, 2018
| if buildPlatform == hostPlatform && hostPlatform == targetPlatform && targetPlatform.isAarch32 | ||
| then [] | ||
| else [ "build" "host" ] ++ stdenv.lib.optional (targetPlatform != hostPlatform) "target"; | ||
| configurePlatforms = [ "build" "host" ] ++ stdenv.lib.optional (targetPlatform != hostPlatform) "target"; |
There was a problem hiding this comment.
Is it runtime-tested that this fixes it?
There was a problem hiding this comment.
No it is not. Unless we have some CI that does this, I was hoping you might :).
There was a problem hiding this comment.
The only other issue I remember is ld picking a wrong emulation. If this also doesn't solve that we have the means to force it from nix now.
There was a problem hiding this comment.
IIRC the LD emulation thing was the actual thing that prevented it from working.
There was a problem hiding this comment.
Also as far as I know we are using the same triplets that other distros do, or maybe that was one of the things to double-check - IIRC gnu-efi had some problems that I had to work around on ARM. I can still try to build this though once some other builds finish.
There was a problem hiding this comment.
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=8f50e6eddb95 is the culprit. Looks like they meant to blacklist odd-end arm vendors, so unknown is being caught by mistake.
There was a problem hiding this comment.
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 7f5ba50d4b2..0acd953f119 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -254,6 +254,7 @@ esac
case ${target} in
# Avoid special cases that are not obsolete
arm*-*-*eabi* \
+ | arm*-unknown-* \
)
;;
arm*-wince-pe* \I'll submit
There was a problem hiding this comment.
Are you sure the real problem isn't that your quadruplet is missing the *eabi* part? What do other distros do?
There was a problem hiding this comment.
(There is an older 'OABI' ABI which has been dropped by GCC by now, I think).
There was a problem hiding this comment.
@dezgeg oh I see, interesting.
dezgeg
reviewed
May 6, 2018
lib/systems/parse.nix
Outdated
| @@ -69,8 +69,13 @@ rec { | |||
| cpuTypes = with significantBytes; setTypes types.openCpuType { | |||
| arm = { bits = 32; significantByte = littleEndian; family = "arm"; }; | |||
| armv5tel = { bits = 32; significantByte = littleEndian; family = "arm"; }; | |||
| armv6a = { bits = 32; significantByte = littleEndian; family = "arm"; }; | |||
There was a problem hiding this comment.
How are the R- and M- parts tested given that they probably don't run Linux at all or without a nommu-configured libc etc.?
There was a problem hiding this comment.
I merely wanted to use gccCrossStageStatic. This does allow gcc and binutils to configure.
There was a problem hiding this comment.
I also built binutils.
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Failure on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
Ericson2314
force-pushed
the
gnu-config-arm
branch
2 times, most recently
from
eca653b
to
189614e
Compare
|
@dezgeg OK I prohibited |
|
Well, now I get that error message on |
|
@dezgeg ah thanks, I'll change the defaulting. Would |
Ericson2314
force-pushed
the
gnu-config-arm
branch
2 times, most recently
from
bd04b0c
to
b3a67c0
Compare
|
Doing the lib changes in #40297 |
|
Failure on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Failure on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
Ericson2314
force-pushed
the
gnu-config-arm
branch
from
b3a67c0
to
84c279f
Compare
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Failure on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Success on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Build-testing this version now... |
|
This one builds stdenv now at least. |
|
Great, Thanks so much for testing! I might rebase again but I'll merge today. |
Not just Aarch64. Other non-x86 platforms might be old enough, but I am about to update gnu-config to handle things better across the board.
…xception"" And there's more reverts too. The previous commmit d838afbc9376bdadb8c690eb00b425f3eeccdf2d to gnu-config finally solves it! This reverts commit 3ed545a.
Ericson2314
force-pushed
the
gnu-config-arm
branch
from
84c279f
to
f2b575b
Compare
|
Success on x86_64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
8 tasks
|
Success on x86_64-darwin (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
|
Failure on aarch64-linux (full log) Attempted: binutils, gnu-config Partial log (click to expand)
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation for this change
gnu-configfailed to parse certain ARM cpu types and most 4-component configs, so I fixed it upstream. Now after updating it here, we should be able to always pass--buildand--hostto binutils and GCC like other platforms, and also support more ARM cpu types ourselves.Things done
build-use-sandboxinnix.confon non-NixOS)nix-shell -p nox --run "nox-review wip"./result/bin/)