sudo pacman -Syyu
sudo pacman -S setconf asp fakeroot binutils
asp checkout dark-reader
cd dark-reader/trunk
setconf PKGBUILD arch '(riscv64 x86_64)'
makepkg -si
And the npm will segment fault at npm test.
See build log
How often does it reproduce? Is there a required condition?
Absolutely every time you build the package.
~~We observe that strip-nondeterminism -t zip build/release/*.xpi in PKGBUILD plays a critical role in reproducing of this BUG.
The subsequent npm test will end up with segment fault at the very beginning at npm test without this command.
The test goes with a single strip-nondeterminism -t zip build/release/*.xpi but segment fault stills,
and npm will pass with adding this command once again(see https://github.com/felixonmars/archriscv-packages/pull/1701/files).~~
Version
v18.9.0 npm 8.19.1
Platform
5.19.7-arch1-1 on qemu-riscv64 version 7.0.0
Subsystem
No response
What steps will reproduce the bug?
On a fresh install ArchLinux RISC-V on qemu
And the npm will segment fault at
npm test
. See build logHow often does it reproduce? Is there a required condition?
Absolutely every time you build the package. ~~We observe that
strip-nondeterminism -t zip build/release/*.xpi
inPKGBUILD
plays a critical role in reproducing of this BUG. The subsequentnpm test
will end up with segment fault at the very beginning atnpm test
without this command. The test goes with a singlestrip-nondeterminism -t zip build/release/*.xpi
but segment fault stills, and npm will pass with adding this command once again(see https://github.com/felixonmars/archriscv-packages/pull/1701/files).~~What is the expected behavior?
npm should not segmentfault
What do you see instead?
npm segmentfault
Additional information
No response