Re: [Valgrind-users] armhf: illegal hardware instruction

[Mathieu M. <ma...@de...>]

On Wed, Jun 29, 2022 at 8:49 PM John Reiser <jr...@bi...> wrote:
>
> > Program received signal SIGILL, Illegal instruction.
> > vgPlain_am_startup (sp_at_startup=3204445696) at
> > m_aspacemgr/aspacemgr-linux.c:1626
> > 1626       init_nsegment(&seg);
> > (gdb) x/i $pc
> > => 0x58071090 <vgPlain_am_startup+20>:  vmov.i32        d16, #0 ; 0x00000000
>
> > As a reminder I do not have neon on this machine:
> >
> > Features        : half thumb fastmult vfp edsp thumbee vfpv3 tls idiva
> > idivt vfpd32 lpae
>
> Therefore this is a gcc configuration problem.  Whoever configured the gcc
> that was used to compile your valgrind assumed that neon would be present,
> but your machine lacks neon.
>
> Run "gcc --verbose".  On my RaspberryPi Model 2B I get (wrapped by hand
> to reasonable line length):
> -----
> Using built-in specs.
> COLLECT_GCC=gcc
> COLLECT_LTO_WRAPPER=/usr/lib/gcc/arm-linux-gnueabihf/10/lto-wrapper
> Target: arm-linux-gnueabihf
> Configured with: ../src/configure -v --with-pkgversion='Debian 10.2.1-6' \
>    --with-bugurl=file:///usr/share/doc/gcc-10/README.Bugs \
>    --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --prefix=/usr \
>    --with-gcc-major-version-only --program-suffix=-10 --program-prefix=arm-linux-gnueabihf- \
>    --enable-shared --enable-linker-build-id --libexecdir=/usr/lib --without-included-gettext \
>    --enable-threads=posix --libdir=/usr/lib --enable-nls --enable-bootstrap --enable-clocale=gnu \
>    --enable-libstdcxx-debug --enable-libstdcxx-time=yes --with-default-libstdcxx-abi=new \
>    --enable-gnu-unique-object --disable-libitm --disable-libquadmath --disable-libquadmath-support \
>    --enable-plugin --enable-default-pie --with-system-zlib --enable-libphobos-checking=release \
>    --with-target-system-zlib=auto --enable-objc-gc=auto --enable-multiarch --disable-sjlj-exceptions \
>    --with-arch=armv7-a --with-fpu=vfpv3-d16 --with-float=hard --with-mode=thumb --disable-werror \
>    --enable-checking=release --build=arm-linux-gnueabihf --host=arm-linux-gnueabihf \
>    --target=arm-linux-gnueabihf
> Thread model: posix
> Supported LTO compression algorithms: zlib zstd
> gcc version 10.2.1 20210110 (Debian 10.2.1-6)
> -----
> where the important part now is "--with-arch=armv7-a --with-fpu=vfpv3-d16"
> for which the "-d16" part restricts gcc to 16 double precision registers
> even though my hardware has "vfpd32".
>
> Consulting "info gcc", then searching for "neon", and examining "ARM Options",
> it seems to me that the fix for gcc to assume vfp3 floating point with 32 double-
> precision registers and the half-precision floating-point conversion operations,
> but omit neon, is the gcc command-line parameter "-march=armv7-a+vfpv3-fp16+nosimd".
> You may want to use "-d16" somewhere, too.

Discussing the issue with Debian/gcc/armhf maintainer resulted in:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1014091#12

So here is my suggested patch for valgrind:

https://bugs.kde.org/show_bug.cgi?id=456200#c1

Thanks everyone for your help !