[llvm] [AArch64] Add scheduling information for flag manipulation instructions in Neoverse-V2 (PR #139765)

[Philipp Rados via llvm-commits]

================
@@ -99,6 +99,8 @@ def V2Write_0c : SchedWriteRes<[]> { let Latency = 0; }
 def V2Write_1c_1B    : SchedWriteRes<[V2UnitB]>   { let Latency = 1; }
 def V2Write_1c_1F    : SchedWriteRes<[V2UnitF]>   { let Latency = 1; }
 def V2Write_1c_1F_1Flg : SchedWriteRes<[V2UnitF, V2UnitFlg]>   { let Latency = 1; }
+def V2Write_1c_1F_1Flg_1T : SchedWriteRes<[V2UnitF, V2UnitFlg]>   { let Latency = 1;
+                                                         let ReleaseAtCycles = [4, 3]; }
----------------
PhilippRados wrote:

I'm not sure I understand what you mean with _more evenly_, since before the resource pressure for the flags was also evenly distributed before just at a higher value (1 vs 0.33).
It does make sense to lower the resource pressure if possible, so I would understand that reasoning.

https://github.com/llvm/llvm-project/pull/139765