------- Additional Comments From markhobley at yahoo dot co dot uk 2009-01-26 13:16 ------- Subject: Re: Additional switches to disallow processor supplementary instructions
> That is pretty much how it is implemented. But I can't > do it for > any arbitrary instructions. The instruction availability > has to be > able to detected at run-time via CPUID insn. My suggestion was to disable existing instructions only. I didn't mean if enable an instruction if it is not disabled. That functionality would remain as it is now, with availability of instructions being based on the build architecture. The cpuid is not really relevant, because it is possible to be building for the same architecture but the target computer has different flags. (However applying the proposed --nosupplementaryinstructions, would solve that issue with code being portable across all machines of the same architecture.) I must confess that when I posted this, I was one version behind on the binutils and there was a problem with invalid instructions making it into the code. There appears to have been a fix since then. However, the switches would still be useful IMHO. Mark. -- http://sourceware.org/bugzilla/show_bug.cgi?id=9782 ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is. _______________________________________________ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils
