i386: Use xor to write zero to memory with -Os even for more than 4 stores [PR11877]

> > 2021-06-20  Roger Sayle  <roger@nextmovesoftware.com>
> >
> > gcc/ChangeLog
> >         PR target/11877
> >         * config/i386/i386.md: New define_peephole2s to shrink writing
> >         1, 2 or 4 consecutive zeros to memory when optimizing for size.

It unfortunately doesn't extend well to larger memory clearing.
Consider e.g.
void
foo (int *p)
{
  p[0] = 0;
  p[7] = 0;
  p[23] = 0;
  p[41] = 0;
  p[48] = 0;
  p[59] = 0;
  p[69] = 0;
  p[78] = 0;
  p[83] = 0;
  p[89] = 0;
  p[98] = 0;
  p[121] = 0;
  p[132] = 0;
  p[143] = 0;
  p[154] = 0;
}
where with the patch we emit:
        xorl    %eax, %eax
        xorl    %edx, %edx
        xorl    %ecx, %ecx
        xorl    %esi, %esi
        xorl    %r8d, %r8d
        movl    %eax, (%rdi)
        movl    %eax, 28(%rdi)
        movl    %eax, 92(%rdi)
        movl    %eax, 164(%rdi)
        movl    %edx, 192(%rdi)
        movl    %edx, 236(%rdi)
        movl    %edx, 276(%rdi)
        movl    %edx, 312(%rdi)
        movl    %ecx, 332(%rdi)
        movl    %ecx, 356(%rdi)
        movl    %ecx, 392(%rdi)
        movl    %ecx, 484(%rdi)
        movl    %esi, 528(%rdi)
        movl    %esi, 572(%rdi)
        movl    %r8d, 616(%rdi)
Here is an incremental patch that emits:
        xorl    %eax, %eax
        movl    %eax, (%rdi)
        movl    %eax, 28(%rdi)
        movl    %eax, 92(%rdi)
        movl    %eax, 164(%rdi)
        movl    %eax, 192(%rdi)
        movl    %eax, 236(%rdi)
        movl    %eax, 276(%rdi)
        movl    %eax, 312(%rdi)
        movl    %eax, 332(%rdi)
        movl    %eax, 356(%rdi)
        movl    %eax, 392(%rdi)
        movl    %eax, 484(%rdi)
        movl    %eax, 528(%rdi)
        movl    %eax, 572(%rdi)
        movl    %eax, 616(%rdi)
instead.

2021-06-22  Jakub Jelinek  <jakub@redhat.com>

	PR target/11877
	* config/i386/i386-protos.h (ix86_last_zero_store_uid): Declare.
	* config/i386/i386-expand.c (ix86_last_zero_store_uid): New variable.
	* config/i386/i386.c (ix86_expand_prologue): Clear it.
	* config/i386/i386.md (peephole2s for 1/2/4 stores of const0_rtx):
	Remove "" from match_operand.  Emit new insns using emit_move_insn and
	set ix86_last_zero_store_uid to INSN_UID of the last store.
	Add peephole2s for 1/2/4 stores of const0_rtx following previous
	successful peep2s.

	* gcc.target/i386/pr11877-2.c: New test.

5 files changed, 104 insertions, 14 deletions

diff --git a/gcc/config/i386/i386-expand.c b/gcc/config/i386/i386-expand.c
index cc2eaee..2986b49 100644
--- a/gcc/config/i386/i386-expand.c
+++ b/gcc/config/i386/i386-expand.c
@@ -1316,6 +1316,9 @@ find_nearest_reg_def (rtx_insn *insn, int regno1, int regno2)
   return false;
 }
 
+/* INSN_UID of the last insn emitted by zero store peephole2s.  */
+int ix86_last_zero_store_uid;
+
 /* Split lea instructions into a sequence of instructions
    which are executed on ALU to avoid AGU stalls.
    It is assumed that it is allowed to clobber flags register
diff --git a/gcc/config/i386/i386-protos.h b/gcc/config/i386/i386-protos.h
index e6ac939..1d05206 100644
--- a/gcc/config/i386/i386-protos.h
+++ b/gcc/config/i386/i386-protos.h
@@ -111,6 +111,7 @@ extern bool ix86_use_lea_for_mov (rtx_insn *, rtx[]);
 extern bool ix86_avoid_lea_for_addr (rtx_insn *, rtx[]);
 extern void ix86_split_lea_for_addr (rtx_insn *, rtx[], machine_mode);
 extern bool ix86_lea_for_add_ok (rtx_insn *, rtx[]);
+extern int ix86_last_zero_store_uid;
 extern bool ix86_vec_interleave_v2df_operator_ok (rtx operands[3], bool high);
 extern bool ix86_dep_by_shift_count (const_rtx set_insn, const_rtx use_insn);
 extern bool ix86_agi_dependent (rtx_insn *set_insn, rtx_insn *use_insn);
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index c3740ff..3d5883b 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -8196,6 +8196,7 @@ ix86_expand_prologue (void)
   bool save_stub_call_needed;
   rtx static_chain = NULL_RTX;
 
+  ix86_last_zero_store_uid = 0;
   if (ix86_function_naked (current_function_decl))
     {
       if (flag_stack_usage_info)
diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 9116828..700c158 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -19360,37 +19360,96 @@
 ;; When optimizing for size, zeroing memory should use a register.
 (define_peephole2
   [(match_scratch:SWI48 0 "r")
-   (set (match_operand:SWI48 1 "memory_operand" "") (const_int 0))
-   (set (match_operand:SWI48 2 "memory_operand" "") (const_int 0))
-   (set (match_operand:SWI48 3 "memory_operand" "") (const_int 0))
-   (set (match_operand:SWI48 4 "memory_operand" "") (const_int 0))]
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 2 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 3 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 4 "memory_operand") (const_int 0))]
   "optimize_insn_for_size_p () && peep2_regno_dead_p (0, FLAGS_REG)"
-  [(set (match_dup 1) (match_dup 0))
-   (set (match_dup 2) (match_dup 0))
-   (set (match_dup 3) (match_dup 0))
-   (set (match_dup 4) (match_dup 0))]
+  [(const_int 0)]
 {
   ix86_expand_clear (operands[0]);
+  emit_move_insn (operands[1], operands[0]);
+  emit_move_insn (operands[2], operands[0]);
+  emit_move_insn (operands[3], operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[4], operands[0]));
+  DONE;
 })
 
 (define_peephole2
   [(match_scratch:SWI48 0 "r")
-   (set (match_operand:SWI48 1 "memory_operand" "") (const_int 0))
-   (set (match_operand:SWI48 2 "memory_operand" "") (const_int 0))]
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 2 "memory_operand") (const_int 0))]
   "optimize_insn_for_size_p () && peep2_regno_dead_p (0, FLAGS_REG)"
-  [(set (match_dup 1) (match_dup 0))
-   (set (match_dup 2) (match_dup 0))]
+  [(const_int 0)]
 {
   ix86_expand_clear (operands[0]);
+  emit_move_insn (operands[1], operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[2], operands[0]));
+  DONE;
 })
 
 (define_peephole2
   [(match_scratch:SWI48 0 "r")
-   (set (match_operand:SWI48 1 "memory_operand" "") (const_int 0))]
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))]
   "optimize_insn_for_size_p () && peep2_regno_dead_p (0, FLAGS_REG)"
-  [(set (match_dup 1) (match_dup 0))]
+  [(const_int 0)]
 {
   ix86_expand_clear (operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[1], operands[0]));
+  DONE;
+})
+
+(define_peephole2
+  [(set (match_operand:SWI48 5 "memory_operand")
+	(match_operand:SWI48 0 "general_reg_operand"))
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 2 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 3 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 4 "memory_operand") (const_int 0))]
+  "optimize_insn_for_size_p ()
+   && INSN_UID (peep2_next_insn (0)) == ix86_last_zero_store_uid"
+  [(const_int 0)]
+{
+  emit_move_insn (operands[5], operands[0]);
+  emit_move_insn (operands[1], operands[0]);
+  emit_move_insn (operands[2], operands[0]);
+  emit_move_insn (operands[3], operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[4], operands[0]));
+  DONE;
+})
+
+(define_peephole2
+  [(set (match_operand:SWI48 3 "memory_operand")
+	(match_operand:SWI48 0 "general_reg_operand"))
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))
+   (set (match_operand:SWI48 2 "memory_operand") (const_int 0))]
+  "optimize_insn_for_size_p ()
+   && INSN_UID (peep2_next_insn (0)) == ix86_last_zero_store_uid"
+  [(const_int 0)]
+{
+  emit_move_insn (operands[3], operands[0]);
+  emit_move_insn (operands[1], operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[2], operands[0]));
+  DONE;
+})
+
+(define_peephole2
+  [(set (match_operand:SWI48 2 "memory_operand")
+	(match_operand:SWI48 0 "general_reg_operand"))
+   (set (match_operand:SWI48 1 "memory_operand") (const_int 0))]
+  "optimize_insn_for_size_p ()
+   && INSN_UID (peep2_next_insn (0)) == ix86_last_zero_store_uid"
+  [(const_int 0)]
+{
+  emit_move_insn (operands[2], operands[0]);
+  ix86_last_zero_store_uid
+    = INSN_UID (emit_move_insn (operands[1], operands[0]));
+  DONE;
 })
 
 ;; Reload dislikes loading constants directly into class_likely_spilled
diff --git a/gcc/testsuite/gcc.target/i386/pr11877-2.c b/gcc/testsuite/gcc.target/i386/pr11877-2.c
new file mode 100644
index 0000000..4782dd2
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr11877-2.c
@@ -0,0 +1,26 @@
+/* PR target/11877 */
+/* { dg-do compile } */
+/* { dg-options "-Os" } */
+
+void
+foo (int *p)
+{
+  p[0] = 0;
+  p[7] = 0;
+  p[23] = 0;
+  p[41] = 0;
+  p[48] = 0;
+  p[59] = 0;
+  p[69] = 0;
+  p[78] = 0;
+  p[83] = 0;
+  p[89] = 0;
+  p[98] = 0;
+  p[121] = 0;
+  p[132] = 0;
+  p[143] = 0;
+  p[154] = 0;
+}
+
+/* { dg-final { scan-assembler-times "xorl\[ \t\]" 1 } } */
+/* { dg-final { scan-assembler-not "\\\$0," } } */