RFR: 8343789: Move mutable nmethod data out of CodeCache [v9]

[Andrew Haley]

On 2/8/25 10:38, Andrew Haley wrote:
> On 2/7/25 18:33, Boris Ulasevich wrote:
>> I think ADP+MOVK is better both in terms of performance and code density.
> 
> Good work, you may be right.
> 
> Neoverse N1 has a fairly narrow (4 wide) decoder, so I guess it's more likely
> to be limited by instruction count.
> 
> That benchmark isn't valid for GCC on my machine, because its outputs aren't
> used so GCC doesn't generate code for the asm.
> 
> However, if we change the benchmark to actually *do something* with the data
> (simply add the results together) we get this for movz+movk on Apple M1:

Sorry, I messed up:

      2,916,015,951      cycles:u                         #    3.151 GHz                         (95.33%)
     22,773,874,629      instructions:u                   #    7.81  insn per cycle              (95.33%)

adrp:

      3,132,136,323      cycles:u                         #    3.114 GHz                         (96.60%)
     18,420,464,246      instructions:u                   #    5.88  insn per cycle              (96.60%)

-- 
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671