Skip to contentSkip to conversation

ARM: Davinci: Fix DM644x timer overflow handling and cleanup

PATCH In review dirk.behme at googlemail.com 2 participants 4 messages Mar 24, 2008 – Mar 25, 2008

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ARM: Davinci: Fix DM644x timer overflow handling and cleanup

Fix ARM based DaVinci DM644x timer overflow handling and cleanup timer code.

Changes:

- Remove *_masked() functions as noted by Wolfgang

- Adapt register naming to recent TI spec (sprue26, March 2007)

- Fix reset_timer() handling

- As reported by Pieter [1] the overflow fix introduced a delay of factor 16 (e.g 2 seconds became 32). While the overflow fix is basically okay, it missed to divide CFG_HZ_CLOCK by 16, too. Fix this.

[1] http://article.gmane.org/gmane.comp.boot-loaders.u-boot/38179

- Remove software division of timer count value (DIV(x) macro) and do it in hardware (TIM_CLK_DIV).

Many thanks to Troy Kisky <[email protected]> for his hints!

Patch is compile tested with davinci_dvevm & sonata & schmoogie configuration. Anybody can test it on real hardware? Pieter?

cpu/arm926ejs/davinci/timer.c +21 -46
Index: uboot_davinci/cpu/arm926ejs/davinci/timer.c
===================================================================
@@ -42,9 +42,9 @@
 
 typedef volatile struct {
 	u_int32_t	pid12;
-	u_int32_t	emumgt_clksped;
-	u_int32_t	gpint_en;
-	u_int32_t	gpdir_dat;
+	u_int32_t	emumgt;
+	u_int32_t	na1;
+	u_int32_t	na2;
 	u_int32_t	tim12;
 	u_int32_t	tim34;
 	u_int32_t	prd12;
@@ -52,21 +52,11 @@ typedef volatile struct {
 	u_int32_t	tcr;
 	u_int32_t	tgcr;
 	u_int32_t	wdtcr;
-	u_int32_t	tlgc;
-	u_int32_t	tlmr;
 } davinci_timer;
 
 davinci_timer		*timer = (davinci_timer *)CFG_TIMERBASE;
 
 #define TIMER_LOAD_VAL	(CFG_HZ_CLOCK / CFG_HZ)
-#define READ_TIMER	timer->tim34
-
-/*
- * Timer runs with CFG_HZ_CLOCK, currently 27MHz. To avoid wrap
- * around of timestamp already after min ~159s, divide it, e.g. by 16.
- * timestamp will then wrap around all min ~42min
- */
-#define DIV(x)		((x) >> 4)
 
 static ulong timestamp;
 static ulong lastinc;
@@ -76,63 +66,51 @@ int timer_init(void)
 	/* We are using timer34 in unchained 32-bit mode, full speed */
 	timer->tcr = 0x0;
 	timer->tgcr = 0x0;
-	timer->tgcr = 0x06;
+	timer->tgcr = 0x06 | ((TIM_CLK_DIV - 1) << 8);
 	timer->tim34 = 0x0;
 	timer->prd34 = TIMER_LOAD_VAL;
 	lastinc = 0;
-	timer->tcr = 0x80 << 16;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 
 	return(0);
 }
 
 void reset_timer(void)
 {
-	reset_timer_masked();
-}
-
-ulong get_timer(ulong base)
-{
-	return(get_timer_masked() - base);
-}
-
-void set_timer(ulong t)
-{
-	timestamp = t;
-}
-
-void udelay(unsigned long usec)
-{
-	udelay_masked(usec);
-}
-
-void reset_timer_masked(void)
-{
-	lastinc = DIV(READ_TIMER);
+	timer->tcr = 0x0;
+	timer->tim34 = 0;
+	lastinc = 0;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 }
 
-ulong get_timer_raw(void)
+static ulong get_timer_raw(void)
 {
-	ulong now = DIV(READ_TIMER);
+	ulong now = timer->tim34;
 
 	if (now >= lastinc) {
 		/* normal mode */
 		timestamp += now - lastinc;
 	} else {
 		/* overflow ... */
-		timestamp += now + DIV(TIMER_LOAD_VAL) - lastinc;
+		timestamp += now + TIMER_LOAD_VAL - lastinc;
 	}
 	lastinc = now;
 	return timestamp;
 }
 
-ulong get_timer_masked(void)
+ulong get_timer(ulong base)
+{
+	return(get_timer_raw() - base);
+}
+
+void set_timer(ulong t)
 {
-	return(get_timer_raw() / DIV(TIMER_LOAD_VAL));
+	timestamp = t;
 }
 
-void udelay_masked(unsigned long usec)
+void udelay(unsigned long usec)
 {
 	ulong tmo;
 	ulong endtime;
@@ -165,8 +143,5 @@ unsigned long long get_ticks(void)
  */
 ulong get_tbclk(void)
 {
-	ulong tbclk;
-
-	tbclk = CFG_HZ;
-	return(tbclk);
+	return CFG_HZ;
 }
include/configs/davinci_dvevm.h +2 -1
Index: uboot_davinci/include/configs/davinci_dvevm.h
===================================================================
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */
include/configs/davinci_schmoogie.h +2 -1
Index: uboot_davinci/include/configs/davinci_schmoogie.h
===================================================================
@@ -33,7 +33,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*=============*/
 /* Memory Info */
include/configs/davinci_sonata.h +2 -1
Index: uboot_davinci/include/configs/davinci_sonata.h
===================================================================
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */
#1 lore raw

Hi Dirk, Troy,

Still some troubles on the DVEVM, i2c and flash are too fast, also dhcp is racing like h#ll. See log below.

Currently the timer rolls over every 1 ms, this means get_timer should be called at least every ms to detect a roll over... Is it better to have a static variable timer than read the timer register directly?

I see that this is standard behavior in other timer code in u-boot, but we don't need it I think. I'll try to implement the 64-bit timer without roll over this afternoon.

Best regards, Pieter

U-Boot > era 03EC0000 +20000

Flash erase timeout at address 3ec0000 data b4f0b4f
Flash erase timeout at address 3ed0000 data f4f0f4f
done
Erased 2 sectors
U-Boot > iprobe
Valid chip addresses: 00 01 02 03 04 05 06 07 08 09 0B 0C 0D 0E 0F 10 11 12 13 1 4 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40 41 42 43 44 45 46 47 48 49
4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60 61 62 63 6
4 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
U-Boot >

-----Original Message-----
From: [email protected] [mailto:dirk.behme at googlemail.com]
Sent: maandag 24 maart 2008 15:58
To: u-boot-users at lists.sourceforge.net
Cc: troy.kisky at boundarydevices.com; Pieter Voorthuijsen
Subject: [PATCH] ARM: Davinci: Fix DM644x timer overflow handling and cleanup

Fix ARM based DaVinci DM644x timer overflow handling and cleanup timer code.

Changes:

- Remove *_masked() functions as noted by Wolfgang

- Adapt register naming to recent TI spec (sprue26, March 2007)

- Fix reset_timer() handling

- As reported by Pieter [1] the overflow fix introduced a delay of factor 16 (e.g 2 seconds became 32). While the overflow fix is basically okay, it missed to divide CFG_HZ_CLOCK by 16, too. Fix this.

[1] http://article.gmane.org/gmane.comp.boot-loaders.u-boot/38179

- Remove software division of timer count value (DIV(x) macro) and do it in hardware (TIM_CLK_DIV).

Many thanks to Troy Kisky <[email protected]> for his hints!

Patch is compile tested with davinci_dvevm & sonata & schmoogie configuration. Anybody can test it on real hardware? Pieter?

cpu/arm926ejs/davinci/timer.c +21 -46
Index: uboot_davinci/cpu/arm926ejs/davinci/timer.c
===================================================================
@@ -42,9 +42,9 @@
 
 typedef volatile struct {
 	u_int32_t	pid12;
-	u_int32_t	emumgt_clksped;
-	u_int32_t	gpint_en;
-	u_int32_t	gpdir_dat;
+	u_int32_t	emumgt;
+	u_int32_t	na1;
+	u_int32_t	na2;
 	u_int32_t	tim12;
 	u_int32_t	tim34;
 	u_int32_t	prd12;
@@ -52,21 +52,11 @@ typedef volatile struct {
 	u_int32_t	tcr;
 	u_int32_t	tgcr;
 	u_int32_t	wdtcr;
-	u_int32_t	tlgc;
-	u_int32_t	tlmr;
 } davinci_timer;
 
 davinci_timer		*timer = (davinci_timer *)CFG_TIMERBASE;
 
 #define TIMER_LOAD_VAL	(CFG_HZ_CLOCK / CFG_HZ)
-#define READ_TIMER	timer->tim34
-
-/*
- * Timer runs with CFG_HZ_CLOCK, currently 27MHz. To avoid wrap
- * around of timestamp already after min ~159s, divide it, e.g. by 16.
- * timestamp will then wrap around all min ~42min
- */
-#define DIV(x)		((x) >> 4)
 
 static ulong timestamp;
 static ulong lastinc;
@@ -76,63 +66,51 @@ int timer_init(void)
 	/* We are using timer34 in unchained 32-bit mode, full speed */
 	timer->tcr = 0x0;
 	timer->tgcr = 0x0;
-	timer->tgcr = 0x06;
+	timer->tgcr = 0x06 | ((TIM_CLK_DIV - 1) << 8);
 	timer->tim34 = 0x0;
 	timer->prd34 = TIMER_LOAD_VAL;
 	lastinc = 0;
-	timer->tcr = 0x80 << 16;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 
 	return(0);
 }
 
 void reset_timer(void)
 {
-	reset_timer_masked();
-}
-
-ulong get_timer(ulong base)
-{
-	return(get_timer_masked() - base);
-}
-
-void set_timer(ulong t)
-{
-	timestamp = t;
-}
-
-void udelay(unsigned long usec)
-{
-	udelay_masked(usec);
-}
-
-void reset_timer_masked(void)
-{
-	lastinc = DIV(READ_TIMER);
+	timer->tcr = 0x0;
+	timer->tim34 = 0;
+	lastinc = 0;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 }
 
-ulong get_timer_raw(void)
+static ulong get_timer_raw(void)
 {
-	ulong now = DIV(READ_TIMER);
+	ulong now = timer->tim34;
 
 	if (now >= lastinc) {
 		/* normal mode */
 		timestamp += now - lastinc;
 	} else {
 		/* overflow ... */
-		timestamp += now + DIV(TIMER_LOAD_VAL) - lastinc;
+		timestamp += now + TIMER_LOAD_VAL - lastinc;
 	}
 	lastinc = now;
 	return timestamp;
 }
 
-ulong get_timer_masked(void)
+ulong get_timer(ulong base)
+{
+	return(get_timer_raw() - base);
+}
+
+void set_timer(ulong t)
 {
-	return(get_timer_raw() / DIV(TIMER_LOAD_VAL));
+	timestamp = t;
 }
 
-void udelay_masked(unsigned long usec)
+void udelay(unsigned long usec)
 {
 	ulong tmo;
 	ulong endtime;
@@ -165,8 +143,5 @@ unsigned long long get_ticks(void)
  */
 ulong get_tbclk(void)
 {
-	ulong tbclk;
-
-	tbclk = CFG_HZ;
-	return(tbclk);
+	return CFG_HZ;
 }
include/configs/davinci_dvevm.h +2 -1
Index: uboot_davinci/include/configs/davinci_dvevm.h
===================================================================
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */
include/configs/davinci_schmoogie.h +2 -1
Index: uboot_davinci/include/configs/davinci_schmoogie.h
===================================================================
@@ -33,7 +33,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*=============*/
 /* Memory Info */
include/configs/davinci_sonata.h +2 -1
Index: uboot_davinci/include/configs/davinci_sonata.h
===================================================================
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */
author
#2 lore raw

PATCH 1/1 ARM: Davinci: Fix DM644x timer overflow handling and cleanup

Quoted 7 lines · Pieter Voorthuijsen

Hi Dirk, Troy,

Still some troubles on the DVEVM, i2c and flash are too fast, also dhcp is racing like h#ll. See log below.

Currently the timer rolls over every 1 ms, this means get_timer should be called at least every ms to detect a roll over... Is it better to have a static variable timer than read the timer register directly?

I see that this is standard behavior in other timer code in u-boot, but we don't need it I think. I'll try to implement the 64-bit timer without roll over this afternoon.

I don't love the idea of 64-bit timer! We *should* get it working with 32bit timer. Do you like spending the time to implement the 64-bit timer instead in thinking/debugging the 32-bit timer?

E.g. what happens if you use

#define TIMER_LOAD_VAL 0xffffffff

instead in patch below?

Dirk

Quoted 221 lines

Best regards, Pieter

U-Boot > era 03EC0000 +20000

Flash erase timeout at address 3ec0000 data b4f0b4f
Flash erase timeout at address 3ed0000 data f4f0f4f
done
Erased 2 sectors
U-Boot > iprobe
Valid chip addresses: 00 01 02 03 04 05 06 07 08 09 0B 0C 0D 0E 0F 10 11 12 13 1 4 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40 41 42 43 44 45 46 47 48 49
4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60 61 62 63 6
4 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
U-Boot >

-----Original Message-----
From: dirk.behme at googlemail.com [mailto:dirk.behme at googlemail.com]
Sent: maandag 24 maart 2008 15:58
To: u-boot-users at lists.sourceforge.net
Cc: troy.kisky at boundarydevices.com; Pieter Voorthuijsen
Subject: [PATCH] ARM: Davinci: Fix DM644x timer overflow handling and cleanup

Fix ARM based DaVinci DM644x timer overflow handling and cleanup timer code.

Changes:

- Remove *_masked() functions as noted by Wolfgang

- Adapt register naming to recent TI spec (sprue26, March 2007)

- Fix reset_timer() handling

- As reported by Pieter [1] the overflow fix introduced a delay of factor 16 (e.g 2 seconds became 32). While the overflow fix is basically okay, it missed to divide CFG_HZ_CLOCK by 16, too. Fix this.

[1] http://article.gmane.org/gmane.comp.boot-loaders.u-boot/38179

- Remove software division of timer count value (DIV(x) macro) and do it in hardware (TIM_CLK_DIV).

Many thanks to Troy Kisky <[email protected]> for his hints!

Patch is compile tested with davinci_dvevm & sonata & schmoogie configuration. Anybody can test it on real hardware? Pieter?

Index: uboot_davinci/cpu/arm926ejs/davinci/timer.c
===================================================================

cpu/arm926ejs/davinci/timer.c +21 -46
@@ -42,9 +42,9 @@
 
 typedef volatile struct {
 	u_int32_t	pid12;
-	u_int32_t	emumgt_clksped;
-	u_int32_t	gpint_en;
-	u_int32_t	gpdir_dat;
+	u_int32_t	emumgt;
+	u_int32_t	na1;
+	u_int32_t	na2;
 	u_int32_t	tim12;
 	u_int32_t	tim34;
 	u_int32_t	prd12;
@@ -52,21 +52,11 @@ typedef volatile struct {
 	u_int32_t	tcr;
 	u_int32_t	tgcr;
 	u_int32_t	wdtcr;
-	u_int32_t	tlgc;
-	u_int32_t	tlmr;
 } davinci_timer;
 
 davinci_timer		*timer = (davinci_timer *)CFG_TIMERBASE;
 
 #define TIMER_LOAD_VAL	(CFG_HZ_CLOCK / CFG_HZ)
-#define READ_TIMER	timer->tim34
-
-/*
- * Timer runs with CFG_HZ_CLOCK, currently 27MHz. To avoid wrap
- * around of timestamp already after min ~159s, divide it, e.g. by 16.
- * timestamp will then wrap around all min ~42min
- */
-#define DIV(x)		((x) >> 4)
 
 static ulong timestamp;
 static ulong lastinc;
@@ -76,63 +66,51 @@ int timer_init(void)
 	/* We are using timer34 in unchained 32-bit mode, full speed */
 	timer->tcr = 0x0;
 	timer->tgcr = 0x0;
-	timer->tgcr = 0x06;
+	timer->tgcr = 0x06 | ((TIM_CLK_DIV - 1) << 8);
 	timer->tim34 = 0x0;
 	timer->prd34 = TIMER_LOAD_VAL;
 	lastinc = 0;
-	timer->tcr = 0x80 << 16;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 
 	return(0);
 }
 
 void reset_timer(void)
 {
-	reset_timer_masked();
-}
-
-ulong get_timer(ulong base)
-{
-	return(get_timer_masked() - base);
-}
-
-void set_timer(ulong t)
-{
-	timestamp = t;
-}
-
-void udelay(unsigned long usec)
-{
-	udelay_masked(usec);
-}
-
-void reset_timer_masked(void)
-{
-	lastinc = DIV(READ_TIMER);
+	timer->tcr = 0x0;
+	timer->tim34 = 0;
+	lastinc = 0;
 	timestamp = 0;
+	timer->tcr = 2 << 22;
 }
 
-ulong get_timer_raw(void)
+static ulong get_timer_raw(void)
 {
-	ulong now = DIV(READ_TIMER);
+	ulong now = timer->tim34;
 
 	if (now >= lastinc) {
 		/* normal mode */
 		timestamp += now - lastinc;
 	} else {
 		/* overflow ... */
-		timestamp += now + DIV(TIMER_LOAD_VAL) - lastinc;
+		timestamp += now + TIMER_LOAD_VAL - lastinc;
 	}
 	lastinc = now;
 	return timestamp;
 }
 
-ulong get_timer_masked(void)
+ulong get_timer(ulong base)
+{
+	return(get_timer_raw() - base);
+}
+
+void set_timer(ulong t)
 {
-	return(get_timer_raw() / DIV(TIMER_LOAD_VAL));
+	timestamp = t;
 }
 
-void udelay_masked(unsigned long usec)
+void udelay(unsigned long usec)
 {
 	ulong tmo;
 	ulong endtime;
@@ -165,8 +143,5 @@ unsigned long long get_ticks(void)
  */
 ulong get_tbclk(void)
 {
-	ulong tbclk;
-
-	tbclk = CFG_HZ;
-	return(tbclk);
+	return CFG_HZ;
 }

Index: uboot_davinci/include/configs/davinci_dvevm.h
===================================================================

include/configs/davinci_dvevm.h +2 -1
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */

Index: uboot_davinci/include/configs/davinci_schmoogie.h
===================================================================

include/configs/davinci_schmoogie.h +2 -1
@@ -33,7 +33,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*=============*/
 /* Memory Info */

Index: uboot_davinci/include/configs/davinci_sonata.h
===================================================================

include/configs/davinci_sonata.h +2 -1
@@ -58,7 +58,8 @@
 #define CONFIG_ARM926EJS			/* arm926ejs CPU core */
 #define CONFIG_SYS_CLK_FREQ	297000000	/* Arm Clock frequency */
 #define CFG_TIMERBASE		0x01c21400	/* use timer 0 */
-#define CFG_HZ_CLOCK		27000000	/* Timer Input clock freq */
+#define TIM_CLK_DIV		16
+#define CFG_HZ_CLOCK		(27000000/TIM_CLK_DIV) /* Timer clock freq */
 #define CFG_HZ			1000
 /*====================================================*/
 /* EEPROM definitions for Atmel 24C256BN SEEPROM chip */
#3 lore raw

Hi Dirk,

Quoted 19 lines

-----Original Message-----
From: Dirk Behme [mailto:dirk.behme at googlemail.com]
Sent: dinsdag 25 maart 2008 11:19
To: Pieter Voorthuijsen
Cc: u-boot-users at lists.sourceforge.net; troy.kisky at boundarydevices.com
Subject: Re: [PATCH] ARM: Davinci: Fix DM644x timer overflow handling and cleanup

Quoted 3 lines · Pieter Voorthuijsen

Hi Dirk, Troy,

Still some troubles on the DVEVM, i2c and flash are too

fast, also dhcp is racing like h#ll. See log below.

Quoted 2 lines

Currently the timer rolls over every 1 ms, this means

get_timer should be called at least every ms to detect a roll over... Is it better to have a static variable timer than read the timer register directly?

Quoted 1 line
Quoted 8 lines
Quoted 1 line

I see that this is standard behavior in other timer code in

u-boot, but we don't need it I think. I'll try to implement the 64-bit timer without roll over this afternoon.

I don't love the idea of 64-bit timer! We *should* get it working with 32bit timer. Do you like spending the time to implement the 64-bit timer instead in thinking/debugging the 32-bit timer?

Ok, let's fix the 32-bit first...

Quoted 6 lines

E.g. what happens if you use

#define TIMER_LOAD_VAL 0xffffffff

instead in patch below?

This has the same behavior a the previous one... I also saw that you did not divide the return value from get_timer, tried that too.

In the meanwhile I found that the previous patch to timer.c was quiet ok, but was missing the following in the udelay function:

cpu/arm926ejs/davinci/timer.c 2008-03-25 16:13:05.000000000 +0100 renamed from cpu/arm926ejs/davinci/timer.c 2008-03-25 09:44:30.000000000 +0100 +1 -0
@@ -141,6 +141,7 @@
        tmo = CFG_HZ_CLOCK / 1000;
        tmo *= usec;
        tmo /= 1000;
+       tmo /= 16;
 
        endtime = get_timer_raw() + tmo;
 

This explains why flash etc operations went OK, but operation requiring udelay failed...

Shall we combine this with the patch you made? Or do we leave thing as they were and insert the diff above?

Best regards, Pieter