[En-Nut-Discussion] LCD_DELAY in avr/dev/hd44780.c

[Uwe Bonnes]

Hallo,

the datasheet for the Hitachi HD447800 controller demands for a minimum of
500 ns Enable Cycle time and a minimum Enable pulse width (high level) of
230 ns.

As I understand the code, our code doesn't guarantee that timing. I think
the patch appended would guarantee the timing up to 16 MHz clock.

Bye
-- 
Uwe Bonnes                bon at elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
--- nut/arch/avr/dev/hd44780.c.~1.1.~	2005-07-26 20:02:27.000000000 +0200
+++ nut/arch/avr/dev/hd44780.c	2006-09-04 18:25:26.000000000 +0200
@@ -122,7 +122,7 @@
  */
 
 static u_char during_init = 1;
-#define LCD_DELAY _NOP(); _NOP()
+#define LCD_DELAY _NOP(); _NOP(); _NOP(); _NOP() /*Three Nops would fit too */
 
 #ifdef LCD_RW_BIT
 
@@ -134,6 +134,7 @@
     sbi(LCD_ENABLE_PORT, LCD_ENABLE_BIT);
     LCD_DELAY;
     cbi(LCD_ENABLE_PORT, LCD_ENABLE_BIT);
+    LCD_DELAY;
     return inp(LCD_DATA_PIN) & LCD_DATA_BITS;
 }
 
@@ -142,12 +143,9 @@
     u_char data;
 #if LCD_DATA_BITS == 0x0F
     data = LcdReadNibble();
-    LCD_DELAY;
     data = data | (LcdReadNibble() << 4);
-    LCD_DELAY;
 #elif LCD_DATA_BITS == 0xF0
     data = LcdReadNibble() >> 4;
-    LCD_DELAY;
     data |= LcdReadNibble();
 #elif LCD_DATA_BITS == 0xFF
     data = LcdReadNibble();
@@ -208,6 +206,7 @@
     sbi(LCD_ENABLE_PORT, LCD_ENABLE_BIT);
     LCD_DELAY; 
     cbi(LCD_ENABLE_PORT, LCD_ENABLE_BIT); 
+    LCD_DELAY; 
 }
 
 /*!