bswap: Improve perform_symbolic_merge [PR103376]

Thinking more about it, perhaps we could do more for BIT_XOR_EXPR.
We could allow masked1 == masked2 case for it, but would need to
do something different than the
  n->n = n1->n | n2->n;
we do on all the bytes together.
In particular, for masked1 == masked2 if masked1 != 0 (well, for 0
both variants are the same) and masked1 != 0xff we would need to
clear corresponding n->n byte instead of setting it to the input
as x ^ x = 0 (but if we don't know what x and y are, the result is
also don't know).  Now, for plus it is much harder, because not only
for non-zero operands we don't know what the result is, but it can
modify upper bytes as well.  So perhaps only if current's byte
masked1 && masked2 set the resulting byte to 0xff (unknown) iff
the byte above it is 0 and 0, and set that resulting byte to 0xff too.
Also, even for | we could instead of return NULL just set the resulting
byte to 0xff if it is different, perhaps it will be masked off later on.

This patch just punts on plus if both corresponding bytes are non-zero,
otherwise implements the above.

2021-11-25  Jakub Jelinek  <jakub@redhat.com>

	PR tree-optimization/103376
	* gimple-ssa-store-merging.c (perform_symbolic_merge): For
	BIT_IOR_EXPR, if masked1 && masked2 && masked1 != masked2, don't
	punt, but set the corresponding result byte to MARKER_BYTE_UNKNOWN.
	For BIT_XOR_EXPR similarly and if masked1 == masked2 and the
	byte isn't MARKER_BYTE_UNKNOWN, set the corresponding result byte to
	0.

	* gcc.dg/optimize-bswapsi-7.c: New test.

diff --git a/gcc/gimple-ssa-store-merging.c b/gcc/gimple-ssa-store-merging.c
index 35df0fee9559225eec604884557bde4f880caeed..e7c90ba8b59c33326ad0c67a8c7ba70872316f7e 100644 (file)
--- a/gcc/gimple-ssa-store-merging.c
+++ b/gcc/gimple-ssa-store-merging.c
@@ -556,6 +556,7 @@ perform_symbolic_merge (gimple *source_stmt1, struct symbolic_number *n1,
   n->bytepos = n_start->bytepos;
   n->type = n_start->type;
   size = TYPE_PRECISION (n->type) / BITS_PER_UNIT;
+  uint64_t res_n = n1->n | n2->n;
 
   for (i = 0, mask = MARKER_MASK; i < size; i++, mask <<= BITS_PER_MARKER)
     {
@@ -563,12 +564,33 @@ perform_symbolic_merge (gimple *source_stmt1, struct symbolic_number *n1,
 
       masked1 = n1->n & mask;
       masked2 = n2->n & mask;
-      /* For BIT_XOR_EXPR or PLUS_EXPR, at least one of masked1 and masked2
-        has to be 0, for BIT_IOR_EXPR x | x is still x.  */
-      if (masked1 && masked2 && (code != BIT_IOR_EXPR || masked1 != masked2))
-       return NULL;
+      /* If at least one byte is 0, all of 0 | x == 0 ^ x == 0 + x == x.  */
+      if (masked1 && masked2)
+       {
+         /* + can carry into upper bits, just punt.  */
+         if (code == PLUS_EXPR)
+           return NULL;
+         /* x | x is still x.  */
+         if (code == BIT_IOR_EXPR && masked1 == masked2)
+           continue;
+         if (code == BIT_XOR_EXPR)
+           {
+             /* x ^ x is 0, but MARKER_BYTE_UNKNOWN stands for
+                unknown values and unknown ^ unknown is unknown.  */
+             if (masked1 == masked2
+                 && masked1 != ((uint64_t) MARKER_BYTE_UNKNOWN
+                                << i * BITS_PER_MARKER))
+               {
+                 res_n &= ~mask;
+                 continue;
+               }
+           }
+         /* Otherwise set the byte to unknown, it might still be
+            later masked off.  */
+         res_n |= mask;
+       }
     }
-  n->n = n1->n | n2->n;
+  n->n = res_n;
   n->n_ops = n1->n_ops + n2->n_ops;
 
   return source_stmt;

diff --git a/gcc/testsuite/gcc.dg/optimize-bswapsi-7.c b/gcc/testsuite/gcc.dg/optimize-bswapsi-7.c
new file mode 100644 (file)
index 0000000..1e4db5e
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/optimize-bswapsi-7.c
@@ -0,0 +1,37 @@
+/* PR tree-optimization/103376 */
+/* { dg-do compile } */
+/* { dg-require-effective-target bswap } */
+/* { dg-options "-O2 -fno-tree-vectorize -fdump-tree-optimized" } */
+/* { dg-additional-options "-march=z900" { target s390-*-* } } */
+
+static unsigned int
+f1 (unsigned int x)
+{
+  return (x << 24) | (x >> 8);
+}
+
+unsigned int
+f2 (unsigned *p)
+{
+  return ((f1 (p[0]) | (p[0] >> 8)) & 0xff000000U) | (p[0] >> 24) | ((p[0] & 0xff00U) << 8) | ((p[0] & 0xff0000U) >> 8);
+}
+
+unsigned int
+f3 (unsigned *p)
+{
+  return ((f1 (p[0]) | (p[0] & 0x00ff00ffU)) & 0xff00ff00U) | (f1 (f1 (f1 (p[0]))) & 0x00ff00ffU);
+}
+
+unsigned int
+f4 (unsigned *p)
+{
+  return (f1 (p[0]) ^ (p[0] >> 8)) ^ (p[0] >> 24) ^ ((p[0] & 0xff00U) << 8) ^ ((p[0] & 0xff0000U) >> 8);
+}
+
+unsigned int
+f5 (unsigned *p)
+{
+  return (((f1 (p[0]) | (p[0] >> 16)) ^ (p[0] >> 8)) & 0xffff0000U) ^ (p[0] >> 24) ^ ((p[0] & 0xff00U) << 8) ^ ((p[0] & 0xff0000U) >> 8);
+}
+
+/* { dg-final { scan-tree-dump-times "= __builtin_bswap32 \\\(" 4 "optimized" } } */