arm.c (arm_override_options): Revert change of arm_constant_limit when optimizing...

* arm.c (arm_override_options): Revert change of arm_constant_limit
when optimizing for size.

From-SVN: r73082

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index b7f9c38b13f9b492aff869ee409fff8893ace005..7f2ea308792916d9c49389c0e36e2fa5545bad4d 100644 (file)
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,8 @@
+2003-10-30  Richard Earnshaw  <rearnsha@arm.com>
+
+       * arm.c (arm_override_options): Revert change of arm_constant_limit
+       when optimizing for size.
+
 2003-10-29  Richard Henderson  <rth@redhat.com>
 
        * fold-const.c (fold_single_bit_test): Convert the input to the

diff --git a/gcc/config/arm/arm.c b/gcc/config/arm/arm.c
index cab1d96fd4f5e13b2580f7345cec8c4d219f0fee..73ed8822e9759ac8a569486151b8334a708123d0 100644 (file)
--- a/gcc/config/arm/arm.c
+++ b/gcc/config/arm/arm.c
@@ -847,11 +847,11 @@ arm_override_options (void)
 
   if (optimize_size)
     {
-      /* If optimizing for space, we let the compiler synthesize constants
-         with up to 2 insns, which uses the same space as a load from memory.
-         This gives the opportunity to take even less space when different
-         offsets can be factorized into multiple pre-indexed loads or stores. */
-      arm_constant_limit = 2;
+      /* There's some dispute as to whether this should be 1 or 2.  However,
+        experiments seem to show that in pathological cases a setting of
+        1 degrades less severly than a setting of 2.  This could change if
+        other parts of the compiler change their behavior.  */
+      arm_constant_limit = 1;
 
       /* If optimizing for size, bump the number of instructions that we
          are prepared to conditionally execute (even on a StrongARM). */