[Skiboot] [PATCH v2 2/5] hdata/i2c: ignore i2c devs marked as unknown in the hdat

[Stewart Smith]

Claudio Carvalho <cclaudio at linux.vnet.ibm.com> writes:
> On 01/09/2017 01:03, Oliver wrote:
>> On Thu, Aug 31, 2017 at 5:12 PM, Claudio Carvalho
>> <cclaudio at linux.vnet.ibm.com> wrote:
>>> Some I2C devices are marked as unknown in the hdat. We don't need to add
>>> them to the device tree, so this just log them.
>>>
>>> Signed-off-by: Claudio Carvalho <cclaudio at linux.vnet.ibm.com>
>>> ---
>>>   hdata/i2c.c | 11 +++++++++++
>>>   1 file changed, 11 insertions(+)
>>>
>>> diff --git a/hdata/i2c.c b/hdata/i2c.c
>>> index cc127d5..bfb5b48 100644
>>> --- a/hdata/i2c.c
>>> +++ b/hdata/i2c.c
>>> @@ -226,6 +226,17 @@ int parse_i2c_devs(const struct HDIF_common_hdr *hdr, int idata_index,
>>>                   */
>>>                  i2c_addr = dev->i2c_addr >> 1;
>>>
>>> +               /*
>>> +                * Some i2c devs are marked as unknown in the hdat.
>>> +                * Ignoring them.
>>> +                */
>>> +               if (dev->type == 0xFF) {
>>> +                       prlog(PR_INFO, "HDAT I2C: ignoring e%dp%d - unknown@%x "
>>> +                             "(port 0x%x)\n", dev->i2cm_engine,
>>> +                             dev->i2cm_port, i2c_addr, dev->i2c_port);
>>> +                       continue;
>>> +               }
>>> +
>> I kept these in the device-tree since I figured that we would want to
>> know about them even if we don't know what they are. I don't have any
>> strong opinions about it though so just logging it is fine with me.
>
> I also don't have any strong opinion about that. I proposed this patch 
> because I don't know if an unknown i2c device like that could cause any 
> problem in the kernel. Probably it would be better to keep them in the 
> device tree even if firmware knows nothing about them.
>
> I will drop this patch

I don't think we're in a great position to bet on what is and isn't good
in the HDAT and we'll likely only learn by seeing what's in it and what
happens rather than theorising.

What I might do is take the patch but just have it PR_WARNING out while
still adding the device and just see what shows up for a while.

-- 
Stewart Smith
OPAL Architect, IBM.