You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The nightly from 6cbd1ac still compiles github.com/meta-leap/langserv repo, but starting with any nightly from f247a90 or newer zig build prints:
TODO buf_write_value_bytes maybe type
Unable to dump stack trace: debug info stripped
dummylangserver...The following command terminated unexpectedly:
/home/_/a/zig/zig build-exe /home/_/c/z/langserv/dummylangserver/main.zig --cache-dir /home/_/c/z/langserv/zig-cache --name dummylangserver --main-pkg-path /home/_/c/z --cache on
Build failed. The following command failed:
/home/_/c/z/langserv/zig-cache/o/xvVeGLArhljJxsswi2Np-jE0aFqC6kTF0kfj9G_N-kEzLzRaN7x5k9oneUVD77RT/build /home/_/a/zig/zig /home/_/c/z/langserv /home/_/c/z/langserv/zig-cache
Further drilling down into the failed commands only yields another
TODO buf_write_value_bytes maybe type
Unable to dump stack trace: debug info stripped
Aborted
(Dependencies of meta-leap/langserv are meta-leap/jsonic + meta-leap/zigsess + meta-leap/zag --- impossible to figure out the exact minimal subset of that sizable code-base that would still repro this regression, I tried for a bit to figure out a way.)
The text was updated successfully, but these errors were encountered:
andrewrk
added
bug
Observed behavior contradicts documented or intended behavior
stage1
The process of building from source via WebAssembly and the C backend.
labels
Mar 8, 2020
I was too impatient and didn't want to stay on 6cbd1ac for too long. So I figured out surgically what to remove from jsonic and langserv repos in order to built on current latest nightly.
This means these repos can no longer be used to provoke the issue: so feel free to close this or leave open to keep a reminder for anyone to some day look into the code-paths that lead down to a TODO buf_write_value_bytes maybe type situation.
As a hint, the removals that matter where basically the contenders that also triggered #4380 --- back then I had to make those fn-typed fields nullable to proceed (with no real need to do so), now I just ditched them and hardcoded my default behaviors, and presto. Above issue was gone. The commit https://github.com/meta-leap/jsonic/commit/31aea7d2e95131d944b02b9f6054eb972ae75b9c illustrates the changes that put me back on the bleeding edge build of zig.
The nightly from 6cbd1ac still compiles
github.com/meta-leap/langserv
repo, but starting with any nightly from f247a90 or newerzig build
prints:Further drilling down into the failed commands only yields another
(Dependencies of
meta-leap/langserv
aremeta-leap/jsonic
+meta-leap/zigsess
+meta-leap/zag
--- impossible to figure out the exact minimal subset of that sizable code-base that would still repro this regression, I tried for a bit to figure out a way.)The text was updated successfully, but these errors were encountered: