On 07 Feb 2015, at 17:42, Matthew D. Fuller <fulle...@over-yonder.net> wrote: > > On Sat, Feb 07, 2015 at 12:50:34PM +0000 I heard the voice of > Dimitry Andric, and lo! it spake thus: >> >> Log: >> Pull in r224884 from upstream llvm trunk (by Keno Fischer): > > This blows up the kernel build for me. Apparently reproducible, as my > first build was -j'd and so unreadable, and another non--j build > yielded the below. With the rev patch -R'd out, build completes fine. ... > Cannot emit physreg copy instruction > UNREACHABLE executed at > /usr/src/lib/clang/libllvmx86codegen/../../../contrib/llvm/lib/Target/X86/X86InstrInfo.cpp:3176! > Stack dump: > 0. Program arguments: /usr/obj/usr/src/tmp/usr/bin/cc -cc1 -triple > x86_64-unknown-freebsd11.0 -emit-obj -mrelax-all -disable-free > -main-file-name hptproc.c -mrelocation-model static -mdisable-fp-elim > -masm-verbose -mconstructor-aliases -mcode-model kernel -target-cpu x86-64 > -target-feature -mmx -target-feature -sse -target-feature -aes > -target-feature -avx -disable-red-zone -no-implicit-float -gdwarf-2 > -dwarf-column-info -coverage-file > /usr/obj/usr/src/sys/MORTIS/modules/usr/src/sys/modules/hptmv/hptproc.o > -nostdsysteminc -nobuiltininc -resource-dir > /usr/obj/usr/src/tmp/usr/bin/../lib/clang/3.5.1 -include > /usr/obj/usr/src/sys/MORTIS/opt_global.h -D _KERNEL -D KLD_MODULE -D > HAVE_KERNEL_OPTION_HEADERS -I . -I /usr/src/sys -I /usr/src/sys/contrib/altq > -I /usr/obj/usr/src/sys/MORTIS -isysroot /usr/obj/usr/src/tmp -Werror > -Wno-error-tautological-compare -Wno-error-empty-body > -Wno-error-parentheses-equality -Wno-error-unused-function > -Wno-error-pointer-sign -Wall -Wredundant-decls -Wnested-externs > -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual > -Wundef -Wno-pointer-sign -Wmissing-include-dirs -Wno-unknown-pragmas > -Wno-error-tautological-compare -Wno-error-empty-body > -Wno-error-parentheses-equality -Wno-error-unused-function > -Wno-error-pointer-sign -std=iso9899:1999 -fdebug-compilation-dir > /usr/obj/usr/src/sys/MORTIS/modules/usr/src/sys/modules/hptmv -ferror-limit > 19 -fmessage-length 80 -ffreestanding -fformat-extensions -fwrapv > -stack-protector 1 -mstackrealign -fobjc-runtime=gnustep -fno-common > -fdiagnostics-show-option -fcolor-diagnostics -o hptproc.o -x c > /usr/src/sys/modules/hptmv/../../dev/hptmv/hptproc.c > 1. <eof> parser at end of file > 2. Code generation > 3. Running pass 'Function Pass Manager' on module > '/usr/src/sys/modules/hptmv/../../dev/hptmv/hptproc.c'. > 4. Running pass 'Post-RA pseudo instruction expansion pass' on function > '@hpt_proc_in'
Right, I also got the same report from Sean Bruno. I've backed this out in r278361 for now, since building head is more important than one particular file from LibreOffice. Sorry for the breakage. -Dimitry
signature.asc
Description: Message signed with OpenPGP using GPGMail
