Vitor Sessak <[email protected]> writes:
> ---
> configure | 5 +++++
> 1 files changed, 5 insertions(+), 0 deletions(-)
>
> diff --git a/configure b/configure
> index 4b1e551..7a5c788 100755
> --- a/configure
> +++ b/configure
> @@ -243,6 +243,7 @@ Optimization options (experts only):
> --disable-sse disable SSE optimizations
> --disable-ssse3 disable SSSE3 optimizations
> --disable-avx disable AVX optimizations
> + --disable-xop disable XOP optimizations
> --disable-armv5te disable armv5te optimizations
> --disable-armv6 disable armv6 optimizations
> --disable-armv6t2 disable armv6t2 optimizations
> @@ -1032,6 +1033,7 @@ ARCH_EXT_LIST='
> armv6t2
> armvfp
> avx
> + xop
> mmi
> mmx
> mmx2
Diego probably has opinions on the ordering here. While he'll insist on
strict alphabetic ordering, I think it's fine to group the x86 options
together.
> @@ -1254,6 +1256,7 @@ mmx2_deps="mmx"
> sse_deps="mmx"
> ssse3_deps="sse"
> avx_deps="ssse3"
> +xop_deps="avx"
>
> aligned_stack_if_any="ppc x86"
> fast_64bit_if_any="alpha ia64 mips64 parisc64 ppc64 sparc64 x86_64"
> @@ -2817,6 +2820,7 @@ EOF
> check_yasm "pextrd [eax], xmm0, 1" && enable yasm ||
> die "yasm not found, use --disable-yasm for a crippled build"
> check_yasm "vextractf128 xmm0, ymm0, 0" || disable avx
> + check_yasm "vfmaddps xmm0, xmm0, xmm0, xmm0" || disable xop
> fi
>
> case "$cpu" in
> @@ -3213,6 +3217,7 @@ if enabled x86; then
> echo "SSE enabled ${sse-no}"
> echo "SSSE3 enabled ${ssse3-no}"
> echo "AVX enabled ${avx-no}"
> + echo "XOP enabled ${xop-no}"
> echo "CMOV enabled ${cmov-no}"
> echo "CMOV is fast ${fast_cmov-no}"
> echo "EBX available ${ebx_available-no}"
> --
Looks OK, assuming that instruction is a suitable one to test.
--
Måns Rullgård
[email protected]
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
