[aarch64-port-dev ] RFR (S) 8131682: C1 should use multibyte nops everywhere
aleksey.shipilev at oracle.com
Mon Jul 27 10:53:20 UTC 2015
On 07/27/2015 01:21 PM, Andrew Haley wrote:
> On 27/07/15 10:13, Aleksey Shipilev wrote:
>> Thanks Goetz! Fixed the assembler_ppc.inline.hpp.
>> Andrew/Edward, are you OK with AArch64 part?
> I agree that it looks good. Please have a look to see how many NOPs take the
> same time as a branch.
I don't quite believe we should spend time trying branches for nops, at
least for x86. The change we are discussing follows the Intel
Optimization Reference Manual 184.108.40.206 "Using NOPs", which
Assembler::align for x86 seems to implement with some bells and
whistles. Agner agrees on using multi-byte nops (0F 1F ...) on modern
x86 chips as well; up to the point he claims 4 insn/clock throughput for
Is there a vendor-recommended strategy for using something else? Even if
it's so, this calls for experimenting with Assembler::align itself (that
also touches C2 usages), and not the C1-specific usages this trivial
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 819 bytes
Desc: OpenPGP digital signature
More information about the ppc-aix-port-dev