[llvm] [AMDGPU][MC] Allow null where 128b or larger dst reg is expected (PR #115200)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Fri Nov 22 11:21:18 PST 2024


================
@@ -281,6 +281,22 @@ s_load_dwordx16 s[20:35], s[2:3], 0x1234 glc dlc
 s_load_dwordx16 s[20:35], s[2:3], s0 offset:0x12345 glc dlc
 // GFX10: encoding: [0x01,0x45,0x11,0xf4,0x45,0x23,0x01,0x00]
 
+// null as dst
+s_load_dword null, s[2:3], s0
+// GFX10: encoding: [0x41,0x1f,0x00,0xf4,0x00,0x00,0x00,0x00]
+
+s_load_dwordx2 null, s[2:3], s0
+// GFX10: encoding: [0x41,0x1f,0x04,0xf4,0x00,0x00,0x00,0x00]
+
----------------
arsenm wrote:

Test the dwordx3 case 

https://github.com/llvm/llvm-project/pull/115200


More information about the llvm-commits mailing list