with structuredAttrs lists will be bash arrays which cannot be exported
which will be a issue with some patches and some wrappers like cc-wrapper
this makes it clearer that NIX_CFLAGS_COMPILE must be a string as lists
in env cause a eval failure
/proc/uptime support can't be detected automatically when cross-compiling, so
it is disabled by default. This causes extremely large uptime values (e.g.
19167 days) on systems without RTCs because the boot time is set to the
epoch.
I suspect this issue is caused by the very old version of diffutils we
have in the aarch64+musl bootstrap tools, so hopefully we can
re-enable the tests once we have newer bootstrap tools.
Until a recent patch in coreutils, gnus uname -p returned a different
arch on apple silicon macs, compared to the uname shipped with macos.
This causes config.guess to produce incorrect build system triplets
for various projects on these systems when in a nix-shell.[23][42]
As coreutils only releases every few months/years and no release is
planned at the moment, I'm introducing the patch here. We can drop it
once the next coreutils version is released.
I made a tiny adjustment to the patch from [23]. I removed the usage
of MAYBE_UNUSED, which currently only compiles against HEAD.
[23] https://debbugs.gnu.org/cgi/bugreport.cgi?bug=52330
[42] https://github.com/NixOS/nixpkgs/issues/147914
The arm patch tries to modify files that are nowhere to be found in the
entire source tree. The cygwin patch also doesn't work so good anymore:
1 out of 1 hunk FAILED -- saving rejects to file ChangeLog.rej
1 out of 1 hunk ignored -- saving rejects to file configure.ac.rej
1 out of 1 hunk FAILED -- saving rejects to file lib/hash-triple.c.rej
2 out of 2 hunks ignored
2 out of 2 hunks FAILED -- saving rejects to file src/chroot.c.rej
1 out of 1 hunk FAILED -- saving rejects to file src/cksum.c.rej
6 out of 9 hunks FAILED -- saving rejects to file src/copy.c.rej
1 out of 4 hunks FAILED -- saving rejects to file src/dd.c.rej
1 out of 1 hunk ignored
1 out of 2 hunks FAILED -- saving rejects to file src/mv.c.rej
4 out of 7 hunks FAILED -- saving rejects to file src/stat.c.rej
Tested on 16-core machine with 10 rebuilds:
$ nix build -f. coreutils --repeat 10
No failures. Original failure is probably fixed upstream
with a69e54cfdf7 ("maint: fix dependency of man/arch.1".)
I made a mistake merge. Reverting it in c778945806 undid the state
on master, but now I realize it crippled the git merge mechanism.
As the merge contained a mix of commits from `master..staging-next`
and other commits from `staging-next..staging`, it got the
`staging-next` branch into a state that was difficult to recover.
I reconstructed the "desired" state of staging-next tree by:
- checking out the last commit of the problematic range: 4effe769e2
- `git rebase -i --preserve-merges a8a018ddc0` - dropping the mistaken
merge commit and its revert from that range (while keeping
reapplication from 4effe769e2)
- merging the last unaffected staging-next commit (803ca85c20)
- fortunately no other commits have been pushed to staging-next yet
- applying a diff on staging-next to get it into that state
