[committed] arm: big-endian issue in gen_cpymem_ldrd_strd [PR105981]

Message ID 20220615150938.3599416-1-rearnsha@arm.com
State Committed
Commit 8aaa948059a8b5f0a62ad010d0aa6346b7ac9cd3
Headers
Series [committed] arm: big-endian issue in gen_cpymem_ldrd_strd [PR105981] |

Commit Message

Richard Earnshaw June 15, 2022, 3:09 p.m. UTC
  The code in gen_cpymem_ldrd_strd has been incorrect for big-endian
since r230663.  The problem is that we use gen_lowpart, etc. to split
the 64-bit quantity, but fail to account for the fact that these
routines are really dealing with 64-bit /values/ and in big-endian the
ordering of the sub-registers changes.

To fix this, I've renamed the conceptually misnamed low_reg and hi_reg
as first_reg and second_reg, and then used different logic for
big-endian targets to initialize these values.  This makes the logic
clearer than trying to think about high bits and low bits.

gcc/ChangeLog:

	PR target/105981
	* config/arm/arm.cc (gen_cpymem_ldrd_strd): Rename low_reg and hi_reg
	to first_reg and second_reg respectively.  Initialize them correctly
	when generating big-endian code.
---
 gcc/config/arm/arm.cc | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)
  

Patch

diff --git a/gcc/config/arm/arm.cc b/gcc/config/arm/arm.cc
index 5aa48fc7664..2925907b436 100644
--- a/gcc/config/arm/arm.cc
+++ b/gcc/config/arm/arm.cc
@@ -15744,13 +15744,21 @@  gen_cpymem_ldrd_strd (rtx *operands)
     {
       len -= 8;
       reg0 = gen_reg_rtx (DImode);
-      rtx low_reg = NULL_RTX;
-      rtx hi_reg = NULL_RTX;
+      rtx first_reg = NULL_RTX;
+      rtx second_reg = NULL_RTX;
 
       if (!src_aligned || !dst_aligned)
 	{
-	  low_reg = gen_lowpart (SImode, reg0);
-	  hi_reg = gen_highpart_mode (SImode, DImode, reg0);
+	  if (BYTES_BIG_ENDIAN)
+	    {
+	      second_reg = gen_lowpart (SImode, reg0);
+	      first_reg = gen_highpart_mode (SImode, DImode, reg0);
+	    }
+	  else
+	    {
+	      first_reg = gen_lowpart (SImode, reg0);
+	      second_reg = gen_highpart_mode (SImode, DImode, reg0);
+	    }
 	}
       if (MEM_ALIGN (src) >= 2 * BITS_PER_WORD)
 	emit_move_insn (reg0, src);
@@ -15758,9 +15766,9 @@  gen_cpymem_ldrd_strd (rtx *operands)
 	emit_insn (gen_unaligned_loaddi (reg0, src));
       else
 	{
-	  emit_insn (gen_unaligned_loadsi (low_reg, src));
+	  emit_insn (gen_unaligned_loadsi (first_reg, src));
 	  src = next_consecutive_mem (src);
-	  emit_insn (gen_unaligned_loadsi (hi_reg, src));
+	  emit_insn (gen_unaligned_loadsi (second_reg, src));
 	}
 
       if (MEM_ALIGN (dst) >= 2 * BITS_PER_WORD)
@@ -15769,9 +15777,9 @@  gen_cpymem_ldrd_strd (rtx *operands)
 	emit_insn (gen_unaligned_storedi (dst, reg0));
       else
 	{
-	  emit_insn (gen_unaligned_storesi (dst, low_reg));
+	  emit_insn (gen_unaligned_storesi (dst, first_reg));
 	  dst = next_consecutive_mem (dst);
-	  emit_insn (gen_unaligned_storesi (dst, hi_reg));
+	  emit_insn (gen_unaligned_storesi (dst, second_reg));
 	}
 
       src = next_consecutive_mem (src);