[llvm] [AMDGPU][True16][MC] true16 for v_cmpx_xx_f16 (PR #123419)

[Brox Chen via llvm-commits]

================
@@ -127,10 +127,30 @@
 # GFX11: v_cmpx_eq_u32 v255, v255 dpp8:[0,0,0,0,0,0,0,0] fi:1 ; encoding: [0xea,0xfe,0x95,0x7d,0xff,0x00,0x00,0x00]
 
 0xe9,0x04,0x00,0x7d,0x01,0x77,0x39,0x05
-# GFX11: v_cmpx_f_f16 v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x00,0x7d,0x01,0x77,0x39,0x05]
+# GFX11-REAL16: v_cmpx_f_f16 v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x00,0x7d,0x01,0x77,0x39,0x05]
+# GFX11-FAKE16: v_cmpx_f_f16 v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0x04,0x00,0x7d,0x01,0x77,0x39,0x05]
 
 0xea,0xfe,0x00,0x7d,0x7f,0x00,0x00,0x00
-# GFX11: v_cmpx_f_f16 v127, v127 dpp8:[0,0,0,0,0,0,0,0] fi:1 ; encoding: [0xea,0xfe,0x00,0x7d,0x7f,0x00,0x00,0x00]
+# GFX11-REAL16: v_cmpx_f_f16 v127.l, v127.l dpp8:[0,0,0,0,0,0,0,0] fi:1 ; encoding: [0xea,0xfe,0x00,0x7d,0x7f,0x00,0x00,0x00]
+# GFX11-FAKE16: v_cmpx_f_f16 v127, v127 dpp8:[0,0,0,0,0,0,0,0] fi:1 ; encoding: [0xea,0xfe,0x00,0x7d,0x7f,0x00,0x00,0x00]
+
+0xe9,0xfe,0x00,0x7d,0x7f,0x77,0x39,0x05
+# GFX11-REAL16: v_cmpx_f_f16 v127.l, v127.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0xfe,0x00,0x7d,0x7f,0x77,0x39,0x05]
+# GFX11-FAKE16: v_cmpx_f_f16 v127, v127 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0xfe,0x00,0x7d,0x7f,0x77,0x39,0x05]
+# GFX12-REAL16: v_cmpx_f_f16 v127.l, v127.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0xfe,0x00,0x7d,0x7f,0x77,0x39,0x05]
+# GFX12-FAKE16: v_cmpx_f_f16 v127, v127 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xe9,0xfe,0x00,0x7d,0x7f,0x77,0x39,0x05]
----------------
broxigarchen wrote:

Right. Seems I somehow still missed this.

I will create a new PR and cleanup the test altogether

https://github.com/llvm/llvm-project/pull/123419