Commit 5dff3c4b68 made rpm use autoreconfHook, so the patch that we
are making to `configure` gets lost when the file is regenerated.
To fix this, just patch the equivalent string in the `configure.ac` file
instead.
Fixes#15287
The set/env fix in #14907 wasn't very good, so let's use a null-delimited
approach. Suggested by Aszlig.
In particular, this should fix a mass-breakage on Darwin, though I was
unable to test that.
The go tests get tripped up due to error messages along the lines of:
ld: warning: /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation, ignoring unexpected dylib file
Which is due to us passing that along via $NIX_LDFLAGS in the `clang` wrapper.
To keep `go` from getting confused, I create a small `clang` wrapper that
filters out that warning.
Also, the strip.patch is no longer necessary, and only causes problems when
testing DWARF support:
--- FAIL: TestDwarfAranges (0.59s)
runtime-lldb_test.go:218: Missing aranges section
FAIL
FAIL runtime 17.123s
Also, I disable the misc/cgo/errors test, as I suspect it is also due to similar
problems regarding `ld`:
##### ../misc/cgo/errors
misc/cgo/errors/test.bash: BUG: expected error output to contain "err1.go:11:" but saw:
# command-line-arguments
cannot parse gcc output $WORK/command-line-arguments/_obj//_cgo_.o as ELF, Mach-O, PE object
2016/05/07 02:07:58 Failed: exit status 1
Closes#14208
Fixes#15184. Install everything, including documentation, into one
output, increasing package size by 10%. Otherwise, the help commands for
CMake do not work. This is a good trade because CMake should be a
build-only dependency. The only reason the docs should ever make it to
runtime is if the user has actually installed CMake, in which case
there's a pretty good chance they want the docs, too.
Merges pull request #15275:
This addresses #15226 and fixes killing of processes before
switching from the initrd to the real root.
Right now, the pkill that is issued not only kills user space
processes but also sends a SIGKILL to kernel threads as well.
Usually these threads ignore signals, but some of these processes do
handle signals, like for example the md module, which happened in
#15226.
It also adds a small check for the swraid installer test and a
standalone test which checks on just that problem, so in the future
this shouldn't happen again.
This has been acked by @edolstra on IRC.
As @edolstra pointed out that the kernel module might be painful to
maintain. I strongly disagree because it's only a small module and it's
good to have such a canary in the tests no matter how the bootup process
looks like, so I'm going the masochistic route and try to maintain it.
If it *really* becomes too much maintenance burden, we can still drop or
disable kcanary.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
It takes some extra 13MB (and in dev, not out), but allows perf to show kernel
symbols when profiling. I think it is worth it.
In my NixOS, I refer to it in the system derivation, for easy telling to perf
through /run/booted-system/vmlinux:
system.extraSystemBuilderCmds = ''
ln -s ${config.boot.kernelPackages.kernel.dev}/vmlinux $out/vmlinux
'';
We don't want to push out a channel update whenever this test fails,
because that might have unexpected and confused side effects and it
*really* means that stage 1 of our boot up is broken.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
