This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c/44534] New: _builtin_ia32_vextractf128_si256(X,N) acts as if N was always set to 1.
- From: "jtlapreste at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 14 Jun 2010 09:08:15 -0000
- Subject: [Bug c/44534] New: _builtin_ia32_vextractf128_si256(X,N) acts as if N was always set to 1.
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
the avx built-in _builtin_ia32_vextractf128_si256(X,N) acts as if the
immediate N was always set to 1.
to isolate the problem I compiled 2 programs:
---------------------------
#include <immintrin.h>
int main()
{
__m256i c = _mm256_set_epi32(1, 2, 3, 4, 5, 6, 7, 8);
__m128i d = (__m128i)__builtin_ia32_vextractf128_si256( (__v8si)c, 1);
return 0;
}
----------------------------
and
--------------------------
#include <immintrin.h>
int main()
{
__m256i c = _mm256_set_epi32(1, 2, 3, 4, 5, 6, 7, 8);
__m128i d = (__m128i)__builtin_ia32_vextractf128_si256( (__v8si)c, 0);
return 0;
}
----------------------------
with
gcc-4.5 pipo0.cpp -mavx -o pipo0
and
gc-4.5 pipo1.cpp -mavx -o pipo1
the two binaries generated are identical
and both use
vextractf128 $0x1,%ymm0,%xmm0
in the assembler output from embroider
programs using ouputs were also tested and provide always the ouput
5 6 7 8 for d (never 1 2 3 4)
--------------------------------------------------------
gcc-4.5 -v
Using built-in specs.
COLLECT_GCC=gcc-4.5
COLLECT_LTO_WRAPPER=/usr/lib/gcc/i486-linux-gnu/4.5.1/lto-wrapper
Target: i486-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian 4.5.0-2'
--with-bugurl=file:///usr/share/doc/gcc-4.5/README.Bugs
--enable-languages=c,c++,java,fortran,objc,obj-c++ --prefix=/usr
--enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib
--libexecdir=/usr/lib --without-included-gettext --enable-threads=posix
--with-gxx-include-dir=/usr/include/c++/4.5 --program-suffix=-4.5 --enable-nls
--enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes
--enable-plugin --enable-gold --with-plugin-ld=ld.gold --disable-browser-plugin
--enable-java-awt=gtk --enable-gtk-cairo
--with-java-home=/usr/lib/jvm/java-1.5.0-gcj-4.5/jre --enable-java-home
--with-jvm-root-dir=/usr/lib/jvm/java-1.5.0-gcj-4.5
--with-jvm-jar-dir=/usr/lib/jvm-exports/java-1.5.0-gcj-4.5
--with-arch-directory=i386 --with-ecj-jar=/usr/share/java/eclipse-ecj.jar
--enable-objc-gc --enable-targets=all --with-arch-32=i486 --with-tune=generic
--enable-checking=yes --build=i486-linux-gnu --host=i486-linux-gnu
--target=i486-linux-gnu
Thread model: posix
gcc version 4.5.1 20100419 (prerelease) (Debian 4.5.0-2)
------------------------------------------------------------------
thanks for your attention, and sorry if I missed something
J.T. Lapresté
--
Summary: _builtin_ia32_vextractf128_si256(X,N) acts as if N was
always set to 1.
Product: gcc
Version: 4.5.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: jtlapreste at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44534