RFR(L): 8031321 Support Intel bit manipulation instructions
Igor Veresov
igor.veresov at oracle.com
Fri Feb 14 12:27:18 PST 2014
Thanks for the review, Roland!
On Feb 14, 2014, at 1:34 AM, Roland Westrelin <roland.westrelin at oracle.com> wrote:
>> Can I please get a second review for this?
>>
>> http://cr.openjdk.java.net/~iveresov/8031321/webrev.04/
>
> Can’t we drop the #ifdef X86 in matcher.[ch]pp because UseBMI1Instructions is only true on x86?
The flag is currently defined only for x86, to remove the ifdefs I’ll have to move it to a platform-independent file. I will try to work on extending ADL and matcher to work with patterns like that, so these hacks in matcher will likely go away in the future.
igor
>
> Otherwise looks good to me.
>
> Roland.
More information about the hotspot-compiler-dev
mailing list