gettext failed to link against the GCC 16 toolchain with
ld: libc.a(...rcgu.o):(.bss.error_message_count+0x0): multiple
definition of `error_message_count'; libgrt.a(libgrt_a-error.o):
first defined here
Root cause is a cross-compile artefact, not a compiler change. gnulib's
check for a working error() is a RUN test, so cross-compiling reports
'checking for working error function... guessing no' and turns on
GNULIB_REPLACE_ERROR. Two things then go wrong at once: consumers are
redirected to rpl_error, and gnulib's own error.c -- listed in
am__objects_5 with no GL_COND_OBJ_ERROR guard -- still defines the
unrenamed error_* globals, which collide with relibc's.
relibc genuinely implements error(), error_at_line() and all three
globals (verified with nm). Two coordinated fixes:
- gl_cv_func_working_error=yes in the recipe. Stating the answer a run
test cannot reach while cross-compiling is the standard autoconf
mechanism, and is exactly what the neighbouring ac_cv_/gt_cv_ entries
already do for the same reason. Consumers now call relibc's error()
and no rpl_error reference remains.
- Guard gnulib's error.c on HAVE_ERROR, which configure already defines
as 1 -- upstream gnulib later made this module conditional on the
system lacking error(); this snapshot computes the right answer and
does not act on it. Without this the definitions still collide, since
relibc is Rust and one codegen unit carries several symbols, so
referencing any of them drags the whole object in. glibc escapes this
only by giving each its own object file.
gettext cooks clean. Patch generated by diff, applies at --fuzz=0.
The build system uses `patch --fuzz=0` to apply patches atomically.
Previous patches had extra context that included line counts in the
hunk header that didn't match the current source state due to upstream
drift. The result was atomic rollbacks.
Reduce the hunk context to 1-2 lines around the actual change. This
keeps the patches minimal and ensures they apply cleanly under fuzz=0.
- P0-redox-scheme-bump-0.11.1.patch: hunk now starts at line 96
(where redox-scheme actually is in the current source)
- P0-relibc-syscall-0.8.1.patch: hunk now starts at line 71
(where redox_syscall actually is in the current relibc source)
The previous P0-redox-ioctl-syscall-0.8.1.patch on relibc was useless
because base uses `redox-ioctl = { git = ... relibc.git }` which
bypasses the relibc recipe's local source.
Add a [patch] override in base/Cargo.toml to use the local relibc
source's redox-ioctl instead, which has the syscall 0.8.1 pin.
Base build was failing with:
error[E0277]: `?` couldn't convert the error to `syscall::error::Error`
--> drivers/graphics/driver-graphics/src/lib.rs
because relibc's redox-ioctl sub-crate pins `redox_syscall = "0.7"`
which uses one syscall::Error type, but the rest of the build
(redox-scheme 0.11.1, base workspace redox_syscall 0.8.1) uses
another. Two versions of the same type can't unify across crate
boundaries.
Bump the pin in relibc/redox-ioctl/Cargo.toml to 0.8.1 to match.
vasilito