NetBSD-Bugs archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
RE: toolchain/51147: Segmentation fault with build.sh on OpenBSD current (trying rump)
The following reply was made to PR toolchain/51147; it has been noted by GNATS.
From: <Tomas.T.Bodzar%tieto.com@localhost>
To: <gnats-bugs%NetBSD.org@localhost>, <toolchain-manager%netbsd.org@localhost>,
<gnats-admin%netbsd.org@localhost>, <netbsd-bugs%netbsd.org@localhost>
Cc:
Subject: RE: toolchain/51147: Segmentation fault with build.sh on OpenBSD
current (trying rump)
Date: Mon, 23 May 2016 13:44:42 +0000
-----Original Message-----
From: Christos Zoulas [mailto:christos%zoulas.com@localhost]=20
Sent: Monday, May 23, 2016 3:20 PM
To: Bodzar Tomas T <Tomas.T.Bodzar%tieto.com@localhost>; gnats-bugs%NetBSD.org@localhost; toolc=
hain-manager%netbsd.org@localhost; gnats-admin%netbsd.org@localhost; netbsd-bugs%netbsd.org@localhost
Subject: RE: toolchain/51147: Segmentation fault with build.sh on OpenBSD c=
urrent (trying rump)
On May 23, 12:42pm, Tomas.T.Bodzar%tieto.com@localhost (<Tomas.T.Bodzar%tieto.com@localhost>) w=
rote:
-- Subject: RE: toolchain/51147: Segmentation fault with build.sh on OpenBS=
D=20
|=20
| -----Original Message-----
| From: Christos Zoulas [mailto:christos%zoulas.com@localhost]=3D20
| Sent: Friday, May 20, 2016 4:15 PM
| To: toolchain-manager%netbsd.org@localhost; gnats-admin%netbsd.org@localhost;=20
| netbsd-bugs@netbs=3D d.org; Bodzar Tomas T <Tomas.T.Bodzar%tieto.com@localhost>
| Subject: Re: toolchain/51147: Segmentation fault with build.sh on=20
| OpenBSD c=3D urrent (trying rump)
|=20
| The following reply was made to PR toolchain/51147; it has been noted=20
| by GN=3D ATS.
|=20
| From: christos%zoulas.com@localhost (Christos Zoulas)
| To: <Tomas.T.Bodzar%tieto.com@localhost>, <toolchain-manager%netbsd.org@localhost>,=3D20
| <gnats-admin%netbsd.org@localhost>, <netbsd-bugs%netbsd.org@localhost>,=3D20
| <gnats-bugs%NetBSD.org@localhost>
| Cc:=3D20
| Subject: Re: toolchain/51147: Segmentation fault with build.sh on=20
| OpenBSD c=3D urrent (trying rump)
| Date: Fri, 20 May 2016 10:11:36 -0400
|=20
| On May 20, 1:02pm, Tomas.T.Bodzar%tieto.com@localhost=20
| (<Tomas.T.Bodzar%tieto.com@localhost>) =3D
| wrote:
| -- Subject: Re: toolchain/51147: Segmentation fault with build.sh on=20
| OpenB=3D
| SD=3D20
| =3D20
| | Not sure if I understand you properly now. It is not NetBSD=20
| machine. I a=3D m o=3D3D | n OpenBSD and following tutorial for=20
| Rumpkernel. Part of that is =3D download o=3D3D | f src-netbsd via git=20
| and then trying to build it with buil=3D d.sh (necessary t=3D3D | ools).=
=20
| However it always crash on same place no mat=3D ter if gcc or egcc is=20
| us=3D3D | ed. There are no binaries inside relevant to=3D ols directory,=
=20
| not are others i=3D3D | n bin=3D3D0A=3D3D | =3D3D0A=3D3D | $ pwd=3D3D=
=3D 0A=3D3D |=20
| /home/username/rump/rumprun/src-netbsd=3D3D0A=3D3D
| | $ ls -F bin/=3D3D0A=3D3D
| | cat/ chmod/ cp/ dd/ df/ ed/ ln/ ls/ mkdir/ mv/ =
pa=3D
| x/ =3D3D
| | rm/ rmdir/=3D3D0A=3D3D
| | $ ls -F elftools.x86_64/bin/=3D3D0A=3D3D
| | nbmake* nbmake-amd64*=3D3D0A=3D3D
| | $=3D3D0A=3D3D
| =3D20
| =3D20
| I understand. in src-netbsd there is a tools subdirectory. It contains t=
he=3D
| toolchain to build the system. In tools there is a texinfo=20
| directory. Und=3D er that there should be the binaries for the tool=20
| that coredumps. Alternat=3D ively cp /usr/bin/true to the tool that=20
| core-dumps so the build can procee=3D d :-)
| =3D20
| Christos
|=20
| Tried fresh run to avoid any leftovers from playing. Followed this one=20
| http=3D=20
| s://github.com/rumpkernel/wiki/wiki/Tutorial%3A-Building-Rumprun-Unike
| rnels=3D with OpenBSD -current updated just before run, CC=3D3Degcc (fro=
m=20
| packages) a=3D nd ld line about version changed from die to echo in=20
| build-rr.sh
|=20
| # create libc/fgetws.d =
=3D
| =
=3D
| =20
| CC=3D3D/home/username/rumprun/obj-amd64-hw/rumptools/bin/x8=3D
| 6_64--netbsd-gcc /home/username/rumprun/obj-amd64-hw/rumptools/bin/nbmkde=
p =3D
| -f fgetws.d.tmp -- -std=3D3Dgnu99 -D
| __NetBSD__ -U__OpenBSD__=20
| --sysroot=3D3D/home/username/rumprun/obj-amd64-hw/ru=3D
| mptools/dest -D_LIBC -DLIBC_SCCS -DSYSLIBC_SCCS -D_REENTRANT=20
| -D_DIAGNOSTIC=3D -DHESIOD -DINET6 -D NLS=20
| -I/home/username/rumprun/src-netbsd/lib/libc/include=20
| -I/home/username/r=3D umprun/src-netbsd/lib/libc=20
| -I/home/username/rumprun/src-netbsd/sys -I/home/=3D username/rumprun/s=20
| rc-netbsd/lib/libc/compat/../locale=20
| -I/home/username/rumprun/src-netbsd/lib=3D
| /libc/compat/stdlib=20
| -I/home/username/rumprun/src-netbsd/lib/libc/compat/../=3D
| stdlib -D__BUILD_L
| EGACY=20
| -I/home/username/rumprun/src-netbsd/lib/libc/../../common/lib/libc/qu=3D
| ad=20
| -I/home/username/rumprun/src-netbsd/lib/libc/../../common/lib/libc/str
| in=3D
| g -I/home/username
| /rumprun/src-netbsd/lib/libc/../../common/lib/libc/arch/x86_64/string=20
| -D__D=3D BINTERFACE_PRIVATE=20
| -I/home/username/rumprun/src-netbsd/libexec/ld.elf_so -I=3D=20
| /home/username/rum prun/src-netbsd/lib/libc/dlfcn=20
| -I/home/username/rumprun/src-netbsd/lib/libc=3D
| /gdtoa -I/home/username/rumprun/src-netbsd/lib/libc/locale -DNO_FENV_H=20
| -I/h=3D ome/username/rumpr un/src-netbsd/lib/libc/arch/x86_64/gdtoa=20
| -DWITH_RUNE -I/home/username/rumpr=3D un/src-netbsd/lib/libc=20
| -DPOSIX_MISTAKE -DCOMPAT__RES -DUSE_POLL -DPORTMAP -=3D DWIDE_DOUBLE=20
| -DALL
| _STATE -DUSG_COMPAT /home/username/rumprun/src-netbsd/lib/libc/stdio=
/f=3D
| getws.c && mv fgetws.d.tmp fgetws.d
| --- fgets_chk.d ---
| In file included from=20
| /home/username/rumprun/src-netbsd/lib/libc/ssp/fgets_=3D
| chk.c:41:0:
| /usr/local/lib/gcc/x86_64-unknown-openbsd6.0/4.9.3/include/ssp/stdio.h
| :38:1=3D
| 7: fatal error: ssp.h: No such file or directory #include <ssp.h>
| ^
| compilation terminated.
| nbmkdep: compile failed.
| *** [fgets_chk.d] Error code 1
|=20
| nbmake: stopped in /home/username/rumprun/src-netbsd/lib/libc
| 1 error
|=20
| nbmake: stopped in /home/username/rumprun/src-netbsd/lib/libc
This is building the NetBSD libc with the OpenBSD compiler which seems to b=
e breaking because of header inconsistencies in the OpenBSD gcc.
Is there an ssp.h in /usr/local/lib/gcc/x86_64-unknown-openbsd6.0/4.9.3/inc=
lude/
Christos
It is further down the road
$ cd /usr/local/lib/gcc/x86_64-unknown-openbsd6.0/4.9.3/include
$ ls -F
adxintrin.h cpuid.h mm_malloc.h ssp/
ammintrin.h cross-stdarg.h mmintrin.h tbmintrin.h
avx2intrin.h emmintrin.h nmmintrin.h tgmath.h
avx512cdintrin.h f16cintrin.h pmmintrin.h tmmintrin.h
avx512erintrin.h fma4intrin.h popcntintrin.h unwind.h
avx512fintrin.h fmaintrin.h prfchwintrin.h wmmintrin.h
avx512pfintrin.h fxsrintrin.h quadmath.h x86intrin.h
avxintrin.h ia32intrin.h quadmath_weak.h xmmintrin.h
bmi2intrin.h immintrin.h rdseedintrin.h xopintrin.h
bmiintrin.h lwpintrin.h rtmintrin.h xsaveintrin.h
bmmintrin.h lzcntintrin.h shaintrin.h xsaveoptintrin.=
h
cilk/ mm3dnow.h smmintrin.h xtestintrin.h
$ ls -lF ssp/
total 36
-rw-r--r-- 1 root bin 2402 May 14 02:17 ssp.h
-rw-r--r-- 1 root bin 3473 May 14 02:17 stdio.h
-rw-r--r-- 1 root bin 5720 May 14 02:17 string.h
-rw-r--r-- 1 root bin 2816 May 14 02:17 unistd.h
$
Home |
Main Index |
Thread Index |
Old Index