added asm optimizer for <= byte (bcc+beq->bcs)

codeGenCpu6502/src/prog8/codegen/cpu6502/AsmOptimizer.kt

@@ -439,6 +439,7 @@ private fun optimizeJsrRtsAndOtherCombinations(linesByFour: List<List<IndexedVal
     // jsr Sub + rts -> jmp Sub
     // rts + jmp -> remove jmp
     // rts + bxx -> remove bxx
+    // and some other optimizations.
 
     val mods = mutableListOf<Modification>()
     for (lines in linesByFour) {
@@ -470,6 +471,32 @@ private fun optimizeJsrRtsAndOtherCombinations(linesByFour: List<List<IndexedVal
             else if (" bvc" in second || "\tbvc" in second)
                 mods += Modification(lines[1].index, true, null)
         }
+
+        /*
+    LDA NUM1
+    CMP NUM2
+    BCC LABEL
+    BEQ LABEL
+
+(or something similar) which branches to LABEL when NUM1 <= NUM2. (In this case NUM1 and NUM2 are unsigned numbers.) However, consider the following sequence:
+
+    LDA NUM2
+    CMP NUM1
+    BCS LABEL
+         */
+        val tfirst = first.trimStart()
+        val tsecond = second.trimStart()
+        val tthird = lines[2].value.trimStart()
+        val tfourth = lines[3].value.trimStart()
+        if(tfirst.startsWith("lda") && tsecond.startsWith("cmp") && tthird.startsWith("bcc") && tfourth.startsWith("beq")) {
+            val label = tthird.substring(4)
+            if(label==tfourth.substring(4)) {
+                mods += Modification(lines[0].index, false, "  lda  ${tsecond.substring(4)}")
+                mods += Modification(lines[1].index, false, "  cmp  ${tfirst.substring(4)}")
+                mods += Modification(lines[2].index, false, "  bcs  $label")
+                mods += Modification(lines[3].index, true, null)
+            }
+        }
     }
     return mods
 }
@@ -523,8 +550,8 @@ private fun optimizeUnneededTempvarInAdd(linesByFour: List<List<IndexedValue<Str
         val fourth = lines[3].value.trimStart()
         if(first.startsWith("sta  P8ZP_SCRATCH_") && second.startsWith("lda") && third.startsWith("clc") && fourth.startsWith("adc  P8ZP_SCRATCH_") ) {
             if(fourth.substring(4)==first.substring(4)) {
-                mods.add(Modification(lines[0].index, false, "    clc"))
+                mods.add(Modification(lines[0].index, false, "  clc"))
-                mods.add(Modification(lines[1].index, false, "    adc  ${second.substring(3).trimStart()}"))
+                mods.add(Modification(lines[1].index, false, "  adc  ${second.substring(3).trimStart()}"))
                 mods.add(Modification(lines[2].index, true, null))
                 mods.add(Modification(lines[3].index, true, null))
             }

docs/source/todo.rst

@@ -1,21 +1,6 @@
 TODO
 ====
 
-add asm optimizer:
-    LDA NUM1
-    CMP NUM2
-    BCC LABEL
-    BEQ LABEL
-
-(or something similar) which branches to LABEL when NUM1 <= NUM2. (In this case NUM1 and NUM2 are unsigned numbers.) However, consider the following sequence:
-
-    LDA NUM2
-    CMP NUM1
-    BCS LABEL
-
-
-
-
 larger programs:
 automatons
 mandelbrot (quite a bit larger)

examples/test.p8

@@ -3,21 +3,28 @@
 %zeropage basicsafe
 
 main {
+    sub rrrr() -> ubyte {
+        cx16.r0L++
+        return cx16.r0L
+    }
+
     sub start() {
-        ubyte[] flakes = [1,2,3]
+        cx16.r0L = rrrr() >= 128
 
-        ubyte @shared idx = 2
+;        ubyte[] flakes = [1,2,3]
-
+;
-        if flakes[idx]==239 {
+;        ubyte @shared idx = 2
-            txt.print("yes")
+;
-        } else {
+;        if flakes[idx]==239 {
-            txt.print("nope")
+;            txt.print("yes")
-        }
+;        } else {
-
+;            txt.print("nope")
-        ubyte @shared xx = 16
+;        }
-        ubyte @shared yy = 20
+;
-
+;        ubyte @shared xx = 16
-        txt.print_ub(xx>79 or yy > 49)
+;        ubyte @shared yy = 20
+;
+;        txt.print_ub(xx>79 or yy > 49)
 
         ; if xx>79 or yy > 49 {
 ;        if xx>79 or yy > 49 {