Firefox is now failing on i386. 68.0.2 built OK in builds on 20190823
and 20190824 and first failure was on 20190825 (i.e. I did have a successful
build since the last commits to firefox and rust ports).
current build:
/pobj/firefox-68.0.2/bin/c++ -o Unified_cpp_js_src13.o -c
-I/pobj/firefox-68.0.2/build-i386/dist/system_wrappers -inc
lude /pobj/firefox-68.0.2/firefox-68.0.2/config/gcc_hidden.h -DNDEBUG=1
-DTRIMMED=1 -DJS_HAS_CTYPES -DFFI_BUILDING -DE
XPORT_JS_API -DMOZ_HAS_MOZGLUE -I/pobj/firefox-68.0.2/firefox-68.0.2/js/src
-I/pobj/firefox-68.0.2/build-i386/js/src -
I/pobj/firefox-68.0.2/build-i386/js/src/ctypes/libffi/include
-I/pobj/firefox-68.0.2/firefox-68.0.2/js/src/ctypes/libf
fi/src/x86 -I/pobj/firefox-68.0.2/build-i386/dist/include
-I/usr/local/include/nspr -I/usr/include -fPIC -DMOZILLA_CLI
ENT -include /pobj/firefox-68.0.2/build-i386/js/src/js-confdefs.h
-Qunused-arguments -fstack-protector-strong -Qunused
-arguments -Wall -Wbitfield-enum-conversion -Wempty-body -Wignored-qualifiers
-Woverloaded-virtual -Wpointer-arith -Ws
hadow-field-in-constructor-modified -Wsign-compare -Wtype-limits
-Wunreachable-code -Wunreachable-code-return -Wwrite-
strings -Wno-invalid-offsetof -Wclass-varargs -Wfloat-overflow-conversion
-Wfloat-zero-conversion -Wloop-analysis -Wc+
+1z-compat -Wc++2a-compat -Wcomma -Wimplicit-fallthrough -Wstring-conversion
-Wtautological-overlap-compare -Wtautolog
ical-unsigned-enum-zero-compare -Wtautological-unsigned-zero-compare
-Wno-inline-new-delete -Wno-error=deprecated-decl
arations -Wno-error=array-bounds -Wno-error=backend-plugin
-Wno-error=return-std-move -Wno-error=atomic-alignment -Wfo
rmat -Wformat-security -Wno-gnu-zero-variadic-macro-arguments
-Wno-noexcept-type -Wno-unknown-warning-option -Wno-retu
rn-type-c-linkage -fno-sized-deallocation -O2 -pipe -fstack-protector-strong
-fno-rtti -ffunction-sections -fdata-sect
ions -fno-exceptions -fno-math-errno -pthread -pipe -I/usr/local/include
-fno-omit-frame-pointer -funwind-tables -I/us
r/local/include -fno-strict-aliasing -Werror=format -Wno-shadow -MD -MP -MF
.deps/Unified_cpp_js_src13.o.pp /pobj/f
irefox-68.0.2/build-i386/js/src/Unified_cpp_js_src13.cpp
js/src/Unified_cpp_js_src14.o
error: Resource temporarily unavailable (os error 35)
gmake[3]: ***
[/pobj/firefox-68.0.2/firefox-68.0.2/config/makefiles/rust.mk:240:
force-cargo-library-build] Error 101
gmake[3]: Leaving directory
'/pobj/firefox-68.0.2/build-i386/toolkit/library/rust'
previous build:
73 | build_n_args!(SendBoxFnOnce[+Send]: a1: A1, a2: A2, a3: A3, a4: A4, a5:
A5, a6: A6, a7: A7, a8: A8, a9: A9, a10: A10);
error: Resource temporarily unavailable (os error 35)
gmake[3]: ***
[/pobj/firefox-68.0.2/firefox-68.0.2/config/makefiles/rust.mk:240:
force-cargo-library-build] Error 101
gmake[3]: Leaving directory
'/pobj/firefox-68.0.2/build-i386/toolkit/library/rust'
So the failures are all coming from this make target:
force-cargo-library-build:
$(REPORT_BUILD)
$(call CARGO_BUILD) --lib $(cargo_target_flag) $(rust_features_flag) --
$(cargo_rustc_flags)
and it looks like the "error: Resource temporarily unavailable (os error 35)"
message comes from rust's libstd/io.
After restarting the build a few times I have now got past this (slightly
unfortunate as the time it worked I was trying to catch it in ktrace) but
I expect I'll be able to trigger it again if there are any ideas for more
debug.
