[U-Boot-Users] mpc83xx watchdog

[U-Boot-Users] mpc83xx watchdog

[David Updegraff]

Hi.

Watchdog could work with this.

-dbu
-------------- next part --------------
A non-text attachment was scrubbed...
Name: wd.patch
Type: text/x-patch
Size: 914 bytes
Desc: not available
Url : http://lists.denx.de/pipermail/u-boot/attachments/20061109/56743104/attachment.bin 

[U-Boot-Users] mpc83xx watchdog

[Wolfgang Denk]

In message <eivm9p$njd$2@sea.gmane.org> you wrote:
>
> Watchdog could work with this.

Maybe. But I have  to  reject  the  patch  because  of  coding  style
problems. Please indent by TABs.

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Accident: A condition in which presence of mind is good, but  absence
of body is better.

[U-Boot-Users] mpc83xx watchdog

[Timur Tabi]

David Updegraff wrote:
> Hi.
> 
> Watchdog could work with this.

A patch that does something similar is already in Freescale's 83xx tree that 
is waiting for Wolfgang to pull.

Here's the relevant bit:

@@ -150,9 +213,21 @@ unsigned long get_tbclk(void)
  #if defined(CONFIG_WATCHDOG)
  void watchdog_reset (void)
  {
-	hang();		/* FIXME: implement watchdog_reset()? */
+#ifdef CONFIG_MPC834X
+	int re_enable = disable_interrupts();
+
+	/* Reset the 83xx watchdog */
+	volatile immap_t *immr = (immap_t *) CFG_IMMRBAR;
+	immr->wdt.swsrr = 0x556c;
+	immr->wdt.swsrr = 0xaa39;
+
+	if (re_enable)
+		enable_interrupts ();
+#else
+	hang();
+#endif
  }
-#endif /* CONFIG_WATCHDOG */
+#endif


We don't have any changes for start.S, though.  Also, our tree renames 
CFG_IMMRBAR to CFG_IMMR.

I'm going to NACK your patch, simply because it's incompatible with ours.  I'm 
willing to work with you to fix that, though, if you like.

-- 
Timur Tabi
Linux Kernel Developer @ Freescale

[U-Boot-Users] mpc83xx watchdog

[David Updegraff]

Timur

Good to hear you already have one; fine by me.

But.. I dont know how it can work w/o the tweek to start.S, since by my
reading of the PPC 603 asm.ref., the existing instruction there will not
accept full 16bit value.  Did you have another workaround for it?

> David Updegraff wrote:
>> Hi.
>>
>> Watchdog could work with this.
> 
> A patch that does something similar is already in Freescale's 83xx tree
> that is waiting for Wolfgang to pull.
> 
> Here's the relevant bit:
> 
> @@ -150,9 +213,21 @@ unsigned long get_tbclk(void)
>  #if defined(CONFIG_WATCHDOG)
>  void watchdog_reset (void)
>  {
> -    hang();        /* FIXME: implement watchdog_reset()? */
> +#ifdef CONFIG_MPC834X
> +    int re_enable = disable_interrupts();
> +
> +    /* Reset the 83xx watchdog */
> +    volatile immap_t *immr = (immap_t *) CFG_IMMRBAR;
> +    immr->wdt.swsrr = 0x556c;
> +    immr->wdt.swsrr = 0xaa39;
> +
> +    if (re_enable)
> +        enable_interrupts ();
> +#else
> +    hang();
> +#endif
>  }
> -#endif /* CONFIG_WATCHDOG */
> +#endif
> 
> 
> We don't have any changes for start.S, though.  Also, our tree renames
> CFG_IMMRBAR to CFG_IMMR.
> 
> I'm going to NACK your patch, simply because it's incompatible with
> ours.  I'm willing to work with you to fix that, though, if you like.
>