Hi, This bug seems to be fixed by side-effect after disabling SSE3.
In Makefile.linux we have
ifeq ($(SSE), yes)
ifeq ($(MACHINE), x86_64)
LIB_OBJS+=sha1-mb-x86_64.o sha256-mb-x86_64.o
endif
endif
So if SSE=no then it does not build that two object files. Thus fixes
the FTBFS.
But I still think that we have to check DEB_HOST_ARCH instead of the
MACHINE.
MACHINE=$(shell uname -m | sed 's/i.86/i386/g')
And it seems to me that on x32, "uname -m" will be amd64. Which is still
buggy I think.
Yours,
Paul
--
PaulLiu (劉穎駿)
E-mail: Ying-Chun Liu (PaulLiu) <[email protected]>
signature.asc
Description: OpenPGP digital signature
