From 19d9e36fa53f9a87efe167b7962fa6fe05e490a7 Mon Sep 17 00:00:00 2001
From: zbyti <zbyti@yahoo.com>
Date: Sat, 19 Sep 2020 01:34:29 +0200
Subject: [PATCH 1/5] A8 more examples

---
 examples/README.md                        |  10 +-
 examples/a8/countdown_for_benchmark.mfk   |  98 ++++++++++++++++++
 examples/a8/countdown_while_benchmark.mfk | 119 ++++++++++++++++++++++
 examples/a8/vertical_scroll.mfk           |  37 +++++++
 4 files changed, 262 insertions(+), 2 deletions(-)
 create mode 100644 examples/a8/countdown_for_benchmark.mfk
 create mode 100644 examples/a8/countdown_while_benchmark.mfk
 create mode 100644 examples/a8/vertical_scroll.mfk

diff --git a/examples/README.md b/examples/README.md
index 4a786f53..b7045b6b 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -84,11 +84,17 @@ how to create a program made of multiple files loaded on demand
 
 * [DLI example](a8/dli_example.mfk) – simple display list and display list interrupt example
 
-* [Scroll example](a8/endless_scroll.mfk) – simple horizontal scroll example
+* [Horizontal scroll example](a8/endless_scroll.mfk) – simple horizontal scroll example
+
+* [Vertical scroll example](a8/vertical_scroll.mfk) – simple vertical scroll example
 
 * [System Off example](a8/systemoff_example.mfk) – Programming example with ROM off
 
-* [GR.8 Chessboard Benchmark](a8/gr8_chessboard_benchmark.mfk) – Chessboard drawing benchmark in GR.0
+* [GR.8 Chessboard Benchmark](a8/gr8_chessboard_benchmark.mfk) – Chessboard drawing benchmark in GR.8
+
+* [FOR Countdown Benchmark](a8/countdown_for_benchmark.mfk) – Countdown from 1,999,999 to 0 (FOR loop)
+
+* [WHILE Countdown Benchmark](a8/countdown_while_benchmark.mfk) – Countdown from 1,999,999 to 0 (WHILE loop)
 
 ## Game Boy examples
 
diff --git a/examples/a8/countdown_for_benchmark.mfk b/examples/a8/countdown_for_benchmark.mfk
new file mode 100644
index 00000000..2af77dd9
--- /dev/null
+++ b/examples/a8/countdown_for_benchmark.mfk
@@ -0,0 +1,98 @@
+byte RTCLOK @ 0
+byte iter0B @ 1
+word iter0W @ 2
+bool run_counter @ 4
+
+byte zpr_0 @ $24, zpr_1 @ $23, zpr_2 @ $22, zpr_3 @ $21, zpr_4 @ $20
+byte zpc_0 @ $47, zpc_1 @ $46, zpc_2 @ $45, zpc_3 @ $44, zpc_4 @ $43, zpc_5 @ $42, zpc_6 @ $41
+
+const array(byte) dl align(16) = [
+  $70,$70,$70,
+  $42,$20,0,
+  $41,@word[dl.addr]
+]
+
+void system_off(){
+  asm { sei }
+  antic_nmien = 0
+  pia_portb = $fe
+  os_NMIVEC = vbi.addr
+  run_counter = false
+  antic_nmien = $40
+}
+
+asm void pause() {
+  lda RTCLOK
+  .rt_check:
+  cmp RTCLOK
+  beq .rt_check
+  rts
+}
+
+interrupt void vbi(){
+  RTCLOK += 1
+  if run_counter{
+    zpr_0 += 1
+    if zpr_0 == 10 {
+      zpr_1 += 1
+      zpr_0 = 0
+    }
+    if zpr_1 == 10 {
+      zpr_2 += 1
+      zpr_1 = 0
+    }
+    if zpr_2 == 10 {
+      zpr_3 += 1
+      zpr_2 = 0
+    }
+    if zpr_3 == 10 {
+      zpr_4 += 1
+      zpr_3 = 0
+    }
+  }
+}
+
+void copy_block(pointer src, pointer dsc, word size){
+  for iter0W,0,to,size-1{
+    dsc[iter0W] = src[iter0W]
+  }
+}
+
+void set_block(pointer from, word size, byte val){
+  for iter0W,0,to,size-1{
+    from[iter0W] = val
+  }
+}
+
+void main(){
+  copy_block($e080,$4000,80)
+  system_off()
+  set_block($20,40,255)
+  set_block($20,5,0)
+  set_block($41,7,0)
+
+  pause()
+  antic_chbase = $40
+  antic_dlist = dl.addr
+
+  run_counter = true
+
+  for zpc_6,1,downto,0{
+    for zpc_5,9,downto,0{
+      for zpc_4,9,downto,0{
+        for zpc_3,9,downto,0{
+          for zpc_2,9,downto,0{
+            for zpc_1,9,downto,0{
+              for zpc_0,9,downto,0{
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+
+  run_counter = false
+
+  while true {}
+}
diff --git a/examples/a8/countdown_while_benchmark.mfk b/examples/a8/countdown_while_benchmark.mfk
new file mode 100644
index 00000000..11fbea6b
--- /dev/null
+++ b/examples/a8/countdown_while_benchmark.mfk
@@ -0,0 +1,119 @@
+byte RTCLOK @ 0
+byte iter0B @ 1
+word iter0W @ 2
+bool run_counter @ 4
+
+byte zpr_0 @ $24, zpr_1 @ $23, zpr_2 @ $22, zpr_3 @ $21, zpr_4 @ $20
+byte zpc_0 @ $47, zpc_1 @ $46, zpc_2 @ $45, zpc_3 @ $44, zpc_4 @ $43, zpc_5 @ $42, zpc_6 @ $41
+
+const array(byte) dl align(16) = [
+  $70,$70,$70,
+  $42,$20,0,
+  $41,@word[dl.addr]
+]
+
+void system_off(){
+  asm { sei }
+  antic_nmien = 0
+  pia_portb = $fe
+  os_NMIVEC = vbi.addr
+  run_counter = false
+  antic_nmien = $40
+}
+
+asm void pause() {
+  lda RTCLOK
+  .rt_check:
+  cmp RTCLOK
+  beq .rt_check
+  rts
+}
+
+interrupt void vbi(){
+  RTCLOK += 1
+  if run_counter{
+    zpr_0 += 1
+    if zpr_0 == 10 {
+      zpr_1 += 1
+      zpr_0 = 0
+    }
+    if zpr_1 == 10 {
+      zpr_2 += 1
+      zpr_1 = 0
+    }
+    if zpr_2 == 10 {
+      zpr_3 += 1
+      zpr_2 = 0
+    }
+    if zpr_3 == 10 {
+      zpr_4 += 1
+      zpr_3 = 0
+    }
+  }
+}
+
+void copy_block(pointer src, pointer dsc, word size){
+  for iter0W,0,to,size-1{
+    dsc[iter0W] = src[iter0W]
+  }
+}
+
+void set_block(pointer from, word size, byte val){
+  for iter0W,0,to,size-1{
+    from[iter0W] = val
+  }
+}
+
+void main(){
+  copy_block($e080,$4000,80)
+  system_off()
+  set_block($20,40,255)
+  set_block($20,5,0)
+  set_block($41,7,0)
+
+  zpc_0 = 9
+  zpc_1 = 9
+  zpc_2 = 9
+  zpc_3 = 9
+  zpc_4 = 9
+  zpc_5 = 9
+  zpc_6 = 1
+
+  pause()
+  antic_chbase = $40
+  antic_dlist = dl.addr
+
+  run_counter = true
+
+  while(zpc_6 != $ff){
+    zpc_5 = 9
+    while(zpc_5 != $ff){
+      zpc_4 = 9
+      while(zpc_4 != $ff){
+        zpc_3 = 9
+        while(zpc_3 != $ff){
+          zpc_2 = 9
+          while(zpc_2 != $ff){
+            zpc_1 = 9
+            while(zpc_1 != $ff){
+              zpc_0 = 9
+              while(zpc_0 != $ff){
+                zpc_0 -= 1
+              }
+              zpc_1 -= 1
+            }
+            zpc_2 -= 1
+          }
+          zpc_3 -= 1
+        }
+        zpc_4 -= 1
+      }
+      zpc_5 -= 1
+    }
+    zpc_6 -= 1
+  }
+
+  run_counter = false
+
+  while true {}
+}
diff --git a/examples/a8/vertical_scroll.mfk b/examples/a8/vertical_scroll.mfk
new file mode 100644
index 00000000..b195819c
--- /dev/null
+++ b/examples/a8/vertical_scroll.mfk
@@ -0,0 +1,37 @@
+const array text align(64) = "...MILLFORK RULEZ..." atariscr
+
+const array(byte) dl align(16) = [
+  $70,$70,$70,$70,
+  $67,@word[text.addr],
+  $41,@word[dl.addr]
+]
+
+noinline asm void wait(byte register(a) f) {
+  clc
+  adc os_RTCLOK.b2
+  .rt_check:
+  cmp os_RTCLOK.b2
+  bne .rt_check
+  rts
+}
+
+void main(){
+  byte i0B @ $80
+
+  i0B = $f
+  os_SDLST = dl.addr
+
+  while true {
+    while (i0B != 0){
+      i0B -= 1
+      antic_vscrol = i0B
+      wait(3)
+    }
+    wait(50)
+    while (i0B < $f){
+      i0B += 1
+      antic_vscrol = i0B
+      wait(2)
+    }
+  }
+}
\ No newline at end of file

From 3234bbd3d9ffcbd1fb9786687f5c5a27452a7f7a Mon Sep 17 00:00:00 2001
From: zbyti <zbyti@yahoo.com>
Date: Sat, 19 Sep 2020 12:52:41 +0200
Subject: [PATCH 2/5] two more becnhmarks

---
 examples/README.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/examples/README.md b/examples/README.md
index b7045b6b..2f8ab1f4 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -96,6 +96,10 @@ how to create a program made of multiple files loaded on demand
 
 * [WHILE Countdown Benchmark](a8/countdown_while_benchmark.mfk) – Countdown from 1,999,999 to 0 (WHILE loop)
 
+* [Sieve of Eratosthenes 1899 Benchmark](a8/sieve1899.mfk) – Sieve of Eratosthenes, 1899 primes algorithm
+
+* [Monte Carlo PI estimation Benchmark](a8/montecarlo_pi_benchmark.mfk) – measures the efficiency of multiplication
+
 ## Game Boy examples
 
 * [GB test example](gb/gbtest.mfk) – a partial port of the NES example, with a rudimentary experimental text output implementation

From 269519715b0ccebf5a4ba57ddf191ee1a9ab2ed6 Mon Sep 17 00:00:00 2001
From: zbyti <zbyti@yahoo.com>
Date: Sat, 19 Sep 2020 12:59:14 +0200
Subject: [PATCH 3/5] two more benchmarks

---
 examples/a8/montecarlo_pi_benchmark.mfk | 64 +++++++++++++++++++++++
 examples/a8/sieve1899.mfk               | 68 +++++++++++++++++++++++++
 2 files changed, 132 insertions(+)
 create mode 100644 examples/a8/montecarlo_pi_benchmark.mfk
 create mode 100644 examples/a8/sieve1899.mfk

diff --git a/examples/a8/montecarlo_pi_benchmark.mfk b/examples/a8/montecarlo_pi_benchmark.mfk
new file mode 100644
index 00000000..9e840953
--- /dev/null
+++ b/examples/a8/montecarlo_pi_benchmark.mfk
@@ -0,0 +1,64 @@
+const word probe = 9999
+const word radius = 127 * 127
+pointer screen @ $80
+
+asm void pause() {
+  lda os_RTCLOK.b2
+  .rt_check:
+  cmp os_RTCLOK.b2
+  beq .rt_check
+  rts
+}
+
+// print in HEX
+void printScore(word val) {
+  array(byte) tmp[4]
+  byte iter
+
+  tmp[0] = val.hi >> 4
+  tmp[1] = val.hi & %00001111
+  tmp[2] = val.lo >> 4
+  tmp[3] = val.lo & %00001111
+
+  for iter:tmp {
+    if tmp[iter] < 10 {
+      screen[iter] = tmp[iter] + $10
+    } else {
+      screen[iter] = tmp[iter] + $17
+    }
+  }
+  screen += 40
+}
+
+void main() {
+  array(bool) flags[size] align(1024)
+  word i@$e0, bingo@$e2
+  word x@$e4, y@$e6, n@$e8, p@$ea
+
+  screen = os_SAVMSC
+
+  x = 0
+  y = 0
+  bingo = 0
+
+  pause()
+  os_RTCLOK = 0
+
+  for i,0,to,probe {
+    n = pokey_random & 127
+    x = n * n
+    n = pokey_random & 127
+    y = n * n
+    if ((x + y) <= radius) {
+      bingo += 1
+    }
+  }
+  p = 4 * bingo
+
+  n = os_RTCLOK.b2 + (os_RTCLOK.b1 * 256)
+
+  printScore(n)
+  printScore(p)
+
+  while true {}
+}
diff --git a/examples/a8/sieve1899.mfk b/examples/a8/sieve1899.mfk
new file mode 100644
index 00000000..62bd754b
--- /dev/null
+++ b/examples/a8/sieve1899.mfk
@@ -0,0 +1,68 @@
+const word size = 8192
+
+word RTCLOK @ $13, SAVMSC @ $58
+word i@$e0, prime@$e2, k@$e4, count@$e6
+pointer screen@$e8
+
+asm void pause() {
+  lda $14
+  .rt_check:
+  cmp $14
+  beq .rt_check
+  rts
+}
+
+// print in HEX
+void printScore() {
+  array(byte) tmp[4]
+  byte iter
+
+  screen = SAVMSC
+
+  tmp[0] = RTCLOK.lo >> 4
+  tmp[1] = RTCLOK.lo & %00001111
+  tmp[2] = RTCLOK.hi >> 4
+  tmp[3] = RTCLOK.hi & %00001111
+
+  for iter:tmp {
+    if tmp[iter] < 10 {
+      screen[iter] = tmp[iter] + $10
+    } else {
+      screen[iter] = tmp[iter] + $17
+    }
+  }
+}
+
+void main() {
+  array(bool) flags[size] align(1024)
+  byte iter
+
+  pause()
+  RTCLOK = 0
+
+  for iter,9,downto,0 {
+
+    count = 0
+
+    for i:flags {
+      flags[i] = true
+    }
+
+    for i:flags {
+      if flags[i] {
+        prime = (i * 2) + 3
+        k = i + prime
+        while k <= size {
+          flags[k] = false
+          k += prime
+        }
+        count += 1
+      }
+    }
+
+  }
+
+  printScore()
+
+  while true {}
+}

From 07b6adf6bccbeea0d791a1359d32fa876046fc5f Mon Sep 17 00:00:00 2001
From: zbyti <zbyti@yahoo.com>
Date: Sat, 19 Sep 2020 13:02:55 +0200
Subject: [PATCH 4/5] two more benchmarks

---
 examples/README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/README.md b/examples/README.md
index 2f8ab1f4..e9035f2e 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -96,7 +96,7 @@ how to create a program made of multiple files loaded on demand
 
 * [WHILE Countdown Benchmark](a8/countdown_while_benchmark.mfk) – Countdown from 1,999,999 to 0 (WHILE loop)
 
-* [Sieve of Eratosthenes 1899 Benchmark](a8/sieve1899.mfk) – Sieve of Eratosthenes, 1899 primes algorithm
+* [Sieve of Eratosthenes (1899) Benchmark](a8/sieve1899.mfk) – Sieve of Eratosthenes, 1899 primes algorithm
 
 * [Monte Carlo PI estimation Benchmark](a8/montecarlo_pi_benchmark.mfk) – measures the efficiency of multiplication
 

From e9d4359bc2021d6e4a7257e34cbacd4ed83e694a Mon Sep 17 00:00:00 2001
From: zbyti <zbyti@yahoo.com>
Date: Sun, 20 Sep 2020 22:58:20 +0200
Subject: [PATCH 5/5] A8 Monte Carlo PI refactor (__mul_u16u8u16)

---
 examples/a8/montecarlo_pi_benchmark.mfk | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/examples/a8/montecarlo_pi_benchmark.mfk b/examples/a8/montecarlo_pi_benchmark.mfk
index 9e840953..846ed70e 100644
--- a/examples/a8/montecarlo_pi_benchmark.mfk
+++ b/examples/a8/montecarlo_pi_benchmark.mfk
@@ -33,7 +33,8 @@ void printScore(word val) {
 void main() {
   array(bool) flags[size] align(1024)
   word i@$e0, bingo@$e2
-  word x@$e4, y@$e6, n@$e8, p@$ea
+  word x@$e4, y@$e6, p@$e8, t@$ea
+  byte n@$82
 
   screen = os_SAVMSC
 
@@ -46,19 +47,19 @@ void main() {
 
   for i,0,to,probe {
     n = pokey_random & 127
-    x = n * n
+    x = n * word(n)
     n = pokey_random & 127
-    y = n * n
+    y = n * word(n)
     if ((x + y) <= radius) {
       bingo += 1
     }
   }
   p = 4 * bingo
 
-  n = os_RTCLOK.b2 + (os_RTCLOK.b1 * 256)
+  t = os_RTCLOK.b2 + (os_RTCLOK.b1 * 256)
 
-  printScore(n)
+  printScore(t)
   printScore(p)
 
   while true {}
-}
+}
\ No newline at end of file