[PATCH v5 0/4] Fix i2c bus hang on A0 version of the Armada XP SoCs

[Jason Cooper]

On Mon, Jan 20, 2014 at 12:20:40PM +0100, Gregory CLEMENT wrote:
> On 14/01/2014 09:46, Gregory CLEMENT wrote:
> > On 14/01/2014 03:14, Jason Cooper wrote:
> >> Gregory,
> >>
> >> On Wed, Jan 08, 2014 at 04:06:25PM +0100, Gregory CLEMENT wrote:
> >>> Hi,
> >>>
> >>> Here come the 5th version of the series fixing the i2c bus hang on A0
> >>> version of the Armada XP SoCs. It occurred on the early release of the
> >>> OpenBlocks AX3-4 boards. Indeed the first variants of Armada XP SoCs
> >>> (A0 stepping) have issues related to the i2c controller which prevent
> >>> to use the offload mechanism and lead to a kernel hang during boot.
> >>>
> >>> The main change are the use of marvell,mv78230-a0-i2c and that the
> >>> function mvebu_get_soc_id() is now local to mach-mvebu.
> >>>
> >>> The first patch add a mean to detect the SoCs version at run-time and
> >>> the second one use this feature in the driver.
> >>>
> >>> The 3 first patches should be applied on 3.13-rc and on stable kernel
> >>> 3.12 as it fixes a regression introduce by the commit 930ab3d403ae
> >>> "i2c: mv64xxx: Add I2C Transaction Generator support".
> >>
> >> Ok, I've pulled this in by cherrypicking the commits.  I needed to add
> >> the 'Fixes: ...' and 'Cc: stable ...' language, so the commit ids were
> >> going to change anyhow.  I also added the note to the binding as we
> >> discussed.  I've also based this against v3.13-rc1 as there doesn't
> >> appear to be any need to drag in everything up to -rc6.
> >>
> >> I've pushed this to mvebu/fixes.  Please take a look.  If it all looks
> >> good I'll send the pull request off tomorrow.
> > Hi Jason,
> > 
> > Everything looks perfect!
> > 
> > Thanks for having improved it,
> > 
> > Gregory
> > 
> 
> Hi Jason,
> 
> Eventually we didn't finish with it!
> Ezequiel found an issue when we try to access pci_base if it have
> not been properly mapped, in this case the kernel hang. Indeed the check
> of  the return of of_iomap was wrong.
> 
> The fix would be something like:
> 
> -o<---o<---o<---o<---o<---o<---o<---o<---o<---o<---o<--
> --- a/arch/arm/mach-mvebu/mvebu-soc-id.c
> +++ b/arch/arm/mach-mvebu/mvebu-soc-id.c
> @@ -88,7 +88,7 @@ static int __init mvebu_soc_id_init(void)
>         }
> 
>         pci_base = of_iomap(child, 0);
> -       if (IS_ERR(pci_base)) {
> +       if (pci_base == NULL) {
>                 pr_err("cannot map registers\n");
>                 ret = -ENOMEM;
>                 goto res_ioremap;
> -o<---o<---o<---o<---o<---o<---o<---o<---o<---o<---o<--
> 
> How do you want we handle it?
> 
> Do you want a proper patch to amend?
> A new series?
> Something else?
> 

submit a patch, please.  We'll push it as a fix.

thx,

Jason.