Re: PATCH: PR 834: IA64: Change br to brl for "far" branches when possible

["H. J. Lu" <hjl at lucon dot org>]

On Thu, May 12, 2005 at 09:11:41AM -0700, H. J. Lu wrote:
> This patch will try to turn br into brl when it is possible.
> 

Slot 0 has to be NOP only for BBB. Here is an update.


H.J.
----
2005-05-12  H.J. Lu  <hongjiu.lu@intel.com>

	PR 834
	* elfxx-ia64.c (elfNN_ia64_relax_br): New.
	(elfNN_ia64_relax_brl): Use it.

--- bfd/elfxx-ia64.c.brl	2005-05-07 06:58:12.000000000 -0700
+++ bfd/elfxx-ia64.c	2005-05-12 09:34:52.000000000 -0700
@@ -681,6 +681,114 @@ bfd_elfNN_ia64_after_parse (int itanium)
   oor_branch_size = itanium ? sizeof (oor_ip) : sizeof (oor_brl);
 }
 
+static bfd_boolean
+elfNN_ia64_relax_br (bfd_byte *contents, bfd_vma off)
+{
+  unsigned int template, t0, t1, t2, t3, br_code1, br_code2, mlx;
+  long br_slot;
+  bfd_byte *hit_addr;
+
+  hit_addr = (bfd_byte *) (contents + off);
+  br_slot = (long) hit_addr & 0x3;
+  hit_addr -= br_slot;
+  t0 = bfd_getl32 (hit_addr + 0);
+  t1 = bfd_getl32 (hit_addr + 4);
+  t2 = bfd_getl32 (hit_addr + 8);
+  t3 = bfd_getl32 (hit_addr + 12);
+
+  /* Check if we can turn br into brl.  A label is always at the start
+     of the bundle.  Even if there are predicates on NOPs, we still
+     perform this optimization.  */
+  template = t0 & 0x1e;
+  switch (br_slot)
+    {
+    case 0:
+      /* Check if slot 1 and slot 2 are NOPs. Possible template is
+         BBB.  We only need to check nop.b.  */
+      if (!((t2 & 0x787e00) == 0x100000
+	    && (t3 & 0xf0fc0000) == 0x20000000))
+	return FALSE;
+      br_code1 = (t0 & 0xffffffe0) >> 5 | (t1 & 0x1f) << 27;
+      br_code2 = (t1 & 0x3fe0) >> 5;
+      break;
+    case 1:
+      /* Check if slot 2 is NOP. Possible templates are MBB and BBB.
+	 For BBB, slot 0 also has to be nop.b.  */
+      if (!((template == 0x12				/* MBB */
+	     && (t3 & 0xf0fc0000) == 0x20000000)
+	    || (template == 0x16			/* BBB */
+		&& (t1 & 0x3c3f) == 0x800
+		&& (t3 & 0xf0fc0000) == 0x20000000)))
+	return FALSE;
+      br_code1 = (t1 & 0xffffc000) >> 14 | (t2 & 0x3fff) << 18;
+      br_code2 = (t2 & 0x7fc000) >> 14;
+      break;
+    case 2:
+      /* Check if slot 1 is NOP. Possible templates are MIB, MBB, BBB,
+	 MMB and MFB. For BBB, slot 0 also has to be nop.b.  */
+      if (!((template == 0x10				/* MIB */
+	     && (t2 & 0x7bff00) == 0x200)
+	    || (template == 0x12			/* MBB */
+		&& (t2 & 0x787e00) == 0x100000)
+	    || (template == 0x16			/* BBB */
+		&& (t1 & 0x3c3f) == 0x800
+		&& (t2 & 0x787e00) == 0x100000)
+	    || (template == 0x18			/* MMB */
+		&& (t2 & 0x7bff00) == 0x200)
+	    || (template == 0x1c			/* MFB */
+		&& (t2 & 0x78ff00) == 0x200)))
+	return FALSE;
+      br_code1 = (t2 & 0xff800000) >> 23 | (t3 & 0x7fffff) << 9;
+      br_code2 = (t3 & 0xff800000) >> 23;
+      break;
+    default:
+      /* It should never happen.  */
+      abort ();
+    }
+  
+  /* Check if we can turn br into brl.  */
+  if (!(((br_code2 & 0x1e0) == 0xa0)
+	|| ((br_code1 & 0x1c0) == 0 && (br_code2 & 0x1e0) == 0x80)))
+    return FALSE;
+
+  /* Turn br into brl by setting bit 40.  */
+  br_code2 |= 0x100;
+
+  /* Turn the old bundle into a MLX bundle with the same stop-bit
+     variety.  */
+  if (t0 & 0x1)
+    mlx = 0x5;
+  else
+    mlx = 0x4;
+
+  if (template == 0x16)
+    {
+      /* For BBB, we need to put nop.m in slot 0 and keep the original
+	 predicate.  */
+      t0 &= 0x7e0;
+      t1 &= ~0x3fff;
+      t1 |= 0x1;
+    }
+  else
+    {
+      /* Keep the original instruction in slot 0.  */
+      t0 &= 0xffffffe0;
+      t1 &= 0x3fff;
+    }
+
+  t0 |= mlx;
+
+  /* Put brl in slot 1.  */
+  t2 = br_code1 << 23;
+  t3 = (br_code1 >> 9) | (br_code2 << 23);
+
+  bfd_putl32 (t0, hit_addr);
+  bfd_putl32 (t1, hit_addr + 4);
+  bfd_putl32 (t2, hit_addr + 8);
+  bfd_putl32 (t3, hit_addr + 12);
+  return TRUE;
+}
+
 static void
 elfNN_ia64_relax_brl (bfd_byte *contents, bfd_vma off)
 {
@@ -985,6 +1093,16 @@ elfNN_ia64_relax_section (abfd, sec, lin
 	    }
 	  else if (r_type == R_IA64_PCREL60B)
 	    continue;
+	  else if (elfNN_ia64_relax_br (contents, roff))
+	    {
+	      irel->r_info
+		= ELFNN_R_INFO (ELFNN_R_SYM (irel->r_info),
+				R_IA64_PCREL60B);
+
+	      /* Make the relocation offset point to slot 1.  */
+	      irel->r_offset = (irel->r_offset & ~((bfd_vma) 0x3)) + 1;
+	      continue;
+	    }
 
 	  /* We can't put a trampoline in a .init/.fini section. Issue
 	     an error.  */