AMDGPU: Add implicit def of SCC to kill and indirect pseudos

[nhaehnle]

Summary

Kill instructions sometimes do use SCC in unusual circumstances, when
v_cmpx cannot be used due to the operands that are involved.

Additionally, even if SCC was never defined by the expansion, kill pseudos
could previously occur between an s_cmp and an s_cbranch_scc, which breaks
the SCC liveness tracking when the pseudo is expanded to split the basic
block. While it would be possible to explicitly mark the SCC as live-in for
the successor basic block, it's simpler to just mark the pseudo as using SCC,
so that such a sequence is never emitted by instruction selection in the
first place.

A similar issue affects indirect source/dest pseudos in principle, although
I haven't been able to come up with a test case where it actually matters
(this affects instruction selection, so a MIR test can't be used).

Fixes: dEQP-GLES3.functional.shaders.discard.dynamic_loop_always
Change-Id: Ica8d82ecff1a763b892a1112cf1b06c948863a4f

[arsenm]

LGTM

[rampitec]

LGTM

llvm/trunk/lib/Target/AMDGPU/SIInstructions.td

 def SI_ELSE_BREAK : CFPseudoInstSI <
   (outs SReg_64:$dst), (ins SReg_64:$src0, SReg_64:$src1),
   [(set i64:$dst, (int_amdgcn_else_break i64:$src0, i64:$src1))]> {
   let Size = 4;
   let isAsCheapAsAMove = 1;
   let isReMaterializable = 1;
 }
 
-let Uses = [EXEC], Defs = [EXEC,VCC] in {
+let Uses = [EXEC] in {
 
 multiclass PseudoInstKill <dag ins> {
+  // Even though this pseudo can usually be expanded without an SCC def, we
+  // conservatively assume that it has an SCC def, both because it is sometimes
+  // required in degenerate cases (when V_CMPX cannot be used due to constant
+  // bus limitations) and because it allows us to avoid having to track SCC
+  // liveness across basic blocks.
+  let Defs = [EXEC,VCC,SCC] in
   def _PSEUDO : PseudoInstSI <(outs), ins> {
     let isConvergent = 1;
     let usesCustomInserter = 1;
   }
 
+  let Defs = [EXEC,VCC,SCC] in
   def _TERMINATOR : SPseudoInstSI <(outs), ins> {
     let isTerminator = 1;
   }
 }
 
 defm SI_KILL_I1 : PseudoInstKill <(ins SSrc_b64:$src, i1imm:$killvalue)>;
 defm SI_KILL_F32_COND_IMM : PseudoInstKill <(ins VSrc_b32:$src0, i32imm:$src1, i32imm:$cond)>;
 
+let Defs = [EXEC,VCC] in
 def SI_ILLEGAL_COPY : SPseudoInstSI <
   (outs unknown:$dst), (ins unknown:$src),
   [], " ; illegal copy $src to $dst">;
 
 } // End Uses = [EXEC], Defs = [EXEC,VCC]
 
 // Branch on undef scc. Used to avoid intermediate copy from
 // IMPLICIT_DEF to SCC.
def ADJCALLSTACKDOWN : SPseudoInstSI<
   (outs), (ins i32imm:$amt1, i32imm:$amt2),
   [(callseq_end timm:$amt1, timm:$amt2)],
   "; adjcallstackdown $amt1"> {
   let Size = 8; // Worst case. (s_add_u32 + constant)
   let hasSideEffects = 1;
   let usesCustomInserter = 1;
 }
 
-let Defs = [M0, EXEC],
+let Defs = [M0, EXEC, SCC],
   UseNamedOperandTable = 1 in {
 
 class SI_INDIRECT_SRC<RegisterClass rc> : VPseudoInstSI <
   (outs VGPR_32:$vdst),
   (ins rc:$src, VS_32:$idx, i32imm:$offset)> {
   let usesCustomInserter = 1;
 }
 

llvm/trunk/test/CodeGen/AMDGPU/insert-skips-kill-uncond.mir

 body: |
   bb.0:
     successors: %bb.1
     S_CBRANCH_VCCNZ %bb.1, implicit $vcc
 
   bb.1:
     successors: %bb.2
     $vgpr0 = V_MOV_B32_e32 0, implicit $exec
-    SI_KILL_F32_COND_IMM_TERMINATOR $vgpr0, 0, 3, implicit-def $exec, implicit-def $vcc, implicit $exec
+    SI_KILL_F32_COND_IMM_TERMINATOR $vgpr0, 0, 3, implicit-def $exec, implicit-def $vcc, implicit-def $scc, implicit $exec
     S_BRANCH %bb.2
 
   bb.2:
     S_ENDPGM

llvm/trunk/test/CodeGen/AMDGPU/llvm.amdgcn.kill.ll

 ; SI-LABEL: {{^}}test_non_inline_imm_sgpr:
 ; SI-NOT: v_cmpx_ge_f32_e64
 define amdgpu_ps void @test_non_inline_imm_sgpr(float inreg %a) #0 {
   %c = fcmp ole float %a, 1.500000e+00
   call void @llvm.amdgcn.kill(i1 %c) #1
   ret void
 }
 
+; SI-LABEL: {{^}}test_scc_liveness:
+; SI: v_cmp
+; SI: s_and_b64 exec
+; SI: s_cmp
+; SI: s_cbranch_scc
+define amdgpu_ps void @test_scc_liveness() #0 {
+main_body:
+  br label %loop3
+
+loop3:                                            ; preds = %loop3, %main_body
+  %tmp = phi i32 [ 0, %main_body ], [ %tmp5, %loop3 ]
+  %tmp1 = icmp sgt i32 %tmp, 0
+  call void @llvm.amdgcn.kill(i1 %tmp1) #1
+  %tmp5 = add i32 %tmp, 1
+  br i1 %tmp1, label %endloop15, label %loop3
+
+endloop15:                                        ; preds = %loop3
+  ret void
+}
+
 declare void @llvm.amdgcn.kill(i1) #0
 declare void @llvm.amdgcn.exp.f32(i32, i32, float, float, float, float, i1, i1) #0
 declare i1 @llvm.amdgcn.wqm.vote(i1)
 
 attributes #0 = { nounwind }