[PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[Florian Fainelli]

This patch makes the GPIO pin multiplexing configuration
read the initial GPIO mode register value instead of
setting it initially to 0, then setting the correct
bits, this is safer.

Signed-off-by: Florian Fainelli <florian@openwrt.org>
---
diff --git a/arch/mips/bcm63xx/boards/board_bcm963xx.c b/arch/mips/bcm63xx/boards/board_bcm963xx.c
index cfe32af..6ae4242 100644
--- a/arch/mips/bcm63xx/boards/board_bcm963xx.c
+++ b/arch/mips/bcm63xx/boards/board_bcm963xx.c
@@ -634,7 +634,7 @@ void __init board_prom_init(void)
 	/* setup pin multiplexing depending on board enabled device,
 	 * this has to be done this early since PCI init is done
 	 * inside arch_initcall */
-	val = 0;
+	val = bcm_gpio_readl(GPIO_MODE_REG);
 
 #ifdef CONFIG_PCI
 	if (board.has_pci) {

Re: [PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[Maxime Bizon]

On Mon, 2009-08-31 at 20:28 +0200, Florian Fainelli wrote:

> This patch makes the GPIO pin multiplexing configuration
> read the initial GPIO mode register value instead of
> setting it initially to 0, then setting the correct
> bits, this is safer.

I disagree, now we get working or not working devices depending on the
CFE version used, for which we don't have any public source code nor
changelog.

I cleared the register for that purpose.

If a particular pin muxing config is needed for some board, we should
add this to the board specific struct instead.

-- 
Maxime

Re: [PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[Florian Fainelli]

[-- Attachment #1: Type: text/plain, Size: 956 bytes --]

Le lundi 31 août 2009 22:26:29, Maxime Bizon a écrit :
> On Mon, 2009-08-31 at 20:28 +0200, Florian Fainelli wrote:
> > This patch makes the GPIO pin multiplexing configuration
> > read the initial GPIO mode register value instead of
> > setting it initially to 0, then setting the correct
> > bits, this is safer.
>
> I disagree, now we get working or not working devices depending on the
> CFE version used, for which we don't have any public source code nor
> changelog.

I did not think this could break other boards, it is required on BCM6345 
though.

>
> I cleared the register for that purpose.
>
> If a particular pin muxing config is needed for some board, we should
> add this to the board specific struct instead.

Then I will make it BCM6345 specific.

Thanks.
-- 
Best regards, Florian Fainelli
Email: florian@openwrt.org
Web: http://openwrt.org
IRC: [florian] on irc.freenode.net
-------------------------------

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 201 bytes --]

Re: [PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[Ralf Baechle]

On Mon, Aug 31, 2009 at 11:37:31PM +0200, Florian Fainelli wrote:

> Le lundi 31 août 2009 22:26:29, Maxime Bizon a écrit :
> > On Mon, 2009-08-31 at 20:28 +0200, Florian Fainelli wrote:
> > > This patch makes the GPIO pin multiplexing configuration
> > > read the initial GPIO mode register value instead of
> > > setting it initially to 0, then setting the correct
> > > bits, this is safer.
> >
> > I disagree, now we get working or not working devices depending on the
> > CFE version used, for which we don't have any public source code nor
> > changelog.
> 
> I did not think this could break other boards, it is required on BCM6345 
> though.
> 
> >
> > I cleared the register for that purpose.
> >
> > If a particular pin muxing config is needed for some board, we should
> > add this to the board specific struct instead.
> 
> Then I will make it BCM6345 specific.

I thought you were going to send an update to this?

  Ralf

Re: [PATCH 2/2] bcm63xx: only set the proper GPIO overlay settings

[Florian Fainelli]

On Monday 12 October 2009 00:53:20 Ralf Baechle wrote:
> On Mon, Aug 31, 2009 at 11:37:31PM +0200, Florian Fainelli wrote:
> > Le lundi 31 août 2009 22:26:29, Maxime Bizon a écrit :
> > > On Mon, 2009-08-31 at 20:28 +0200, Florian Fainelli wrote:
> > > > This patch makes the GPIO pin multiplexing configuration
> > > > read the initial GPIO mode register value instead of
> > > > setting it initially to 0, then setting the correct
> > > > bits, this is safer.
> > >
> > > I disagree, now we get working or not working devices depending on the
> > > CFE version used, for which we don't have any public source code nor
> > > changelog.
> >
> > I did not think this could break other boards, it is required on BCM6345
> > though.
> >
> > > I cleared the register for that purpose.
> > >
> > > If a particular pin muxing config is needed for some board, we should
> > > add this to the board specific struct instead.
> >
> > Then I will make it BCM6345 specific.
>
> I thought you were going to send an update to this?

Not at the moment, the user who reported that issue though it was related to 
that and I also saw a potential for breaking BCM6345 GPIO config, but since 
that one has a totally different GPIO registers layout it needs a bigger fix 
which deals with different registers.