Re: [PATCH][AArch64] Implement ALU_BRANCH fusion

["Hurugalawadi, Naveen" <Naveen dot Hurugalawadi at cavium dot com>]

Hi,

Thanks for the review and suggestions.

> I think the patch isn't quite complete yet. You will also need changes in
> generic code. Currently sched_macro_fuse_insns() does:

Modified the sched_macro_fuse_insns() as required.

> Basically the idea is to push the check for CC usage into target macros

Done. Pushed the check into target macros.

The modifications were generic and and quite different from ALU+BRANCH
fusion; a separate patch is posted with the above 2 modifications at:-
https://gcc.gnu.org/ml/gcc-patches/2017-03/msg01368.html

> Also in aarch64.c's macro fusion you need check that the branch
>> instruction uses the same register 

Added to check that same registers are used in ALU and Branch instruction.

Bootstrapped and Regression tested on AArch64.

Please review the patch and let us know if its okay?

Thanks,
Naveen

diff --git a/gcc/config/aarch64/aarch64-fusion-pairs.def b/gcc/config/aarch64/aarch64-fusion-pairs.def
index f0e6dbc..300cd00 100644
--- a/gcc/config/aarch64/aarch64-fusion-pairs.def
+++ b/gcc/config/aarch64/aarch64-fusion-pairs.def
@@ -34,5 +34,6 @@ AARCH64_FUSION_PAIR ("movk+movk", MOVK_MOVK)
 AARCH64_FUSION_PAIR ("adrp+ldr", ADRP_LDR)
 AARCH64_FUSION_PAIR ("cmp+branch", CMP_BRANCH)
 AARCH64_FUSION_PAIR ("aes+aesmc", AES_AESMC)
+AARCH64_FUSION_PAIR ("alu+branch", ALU_BRANCH)
 
 #undef AARCH64_FUSION_PAIR
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
index 4f769a4..31bc5f4 100644
--- a/gcc/config/aarch64/aarch64.c
+++ b/gcc/config/aarch64/aarch64.c
@@ -792,7 +792,8 @@ static const struct tune_params thunderx2t99_tunings =
   &generic_approx_modes,
   4, /* memmov_cost.  */
   4, /* issue_rate.  */
-  (AARCH64_FUSE_CMP_BRANCH | AARCH64_FUSE_AES_AESMC), /* fusible_ops  */
+  (AARCH64_FUSE_CMP_BRANCH | AARCH64_FUSE_AES_AESMC
+   | AARCH64_FUSE_ALU_BRANCH), /* fusible_ops  */
   16,	/* function_align.  */
   8,	/* jump_align.  */
   16,	/* loop_align.  */
@@ -13981,6 +13982,50 @@ aarch_macro_fusion_pair_p (rtx_insn *prev, rtx_insn *curr)
         return true;
     }
 
+  if (aarch64_fusion_enabled_p (AARCH64_FUSE_ALU_BRANCH)
+      && any_condjump_p (curr))
+    {
+      /* We're trying to match:
+	  prev (alu_insn) == (set (r0) plus ((r0) (r1/imm)))
+	  curr (cbz) ==  (set (pc) (if_then_else (eq/ne) (r0)
+							 (const_int 0))
+						 (label_ref ("SYM"))
+						 (pc))  */
+
+      if (SET_DEST (curr_set) != (pc_rtx)
+	  || GET_CODE (SET_SRC (curr_set)) != IF_THEN_ELSE
+	  || ! REG_P (XEXP (XEXP (SET_SRC (curr_set), 0), 0))
+	  || ! REG_P (SET_DEST (prev_set))
+	  || REGNO (SET_DEST (prev_set))
+	     != REGNO (XEXP (XEXP (SET_SRC (curr_set), 0), 0)))
+	return false;
+
+      /* Fuse ALU operations followed by conditional branch instruction.  */
+      switch (get_attr_type (prev))
+	{
+	case TYPE_ALU_IMM:
+	case TYPE_ALU_SREG:
+	case TYPE_ADC_REG:
+	case TYPE_ADC_IMM:
+	case TYPE_ADCS_REG:
+	case TYPE_ADCS_IMM:
+	case TYPE_LOGIC_REG:
+	case TYPE_LOGIC_IMM:
+	case TYPE_CSEL:
+	case TYPE_ADR:
+	case TYPE_MOV_IMM:
+	case TYPE_SHIFT_REG:
+	case TYPE_SHIFT_IMM:
+	case TYPE_BFM:
+	case TYPE_RBIT:
+	case TYPE_REV:
+	case TYPE_EXTEND:
+	  return true;
+
+	default:;
+	}
+    }
+
   return false;
 }