Re: [PATCH v4] powerpc/pci: Assign fixed PHB number based on device-tree properties

Wed, 06 Apr 2016 15:57:03 -0700

Hey guys - our system test team opened a defect on this, since Ubuntu (as it stands) is broken now with CAPI Flash cards.
 
 
So - this is important. You need to tested to confirm that this is in-fact the root cause of Dion's bug, fix it, verify that a Surelock AFU boots correctly (and cxlflash loads), and then push a patch in 16.04 in time for the release (in a few weeks).
 
~ Mike
 

Michael C. Hollinger
和宇喆
Master Inventor
Power Open Source Solutions
IBM Systems
Austin, TX Development Lab
Phone: 1-512-286-6688 | Tie-Line: 363-6688 | Mobile: 1-512-850-6153
E-mail: mchol...@us.ibm.com
Chat:Sametime:  mchol...@us.ibm.com
Find me on: LinkedIn: http://www.linkedin.com/in/mikehollinger Twitter: http://www.twitter.com/mike_hollinger GooglePlus: https://plus.google.com/100310751329925639140/ and within IBM on: IBM Connections: http://w3.ibm.com/connections/profiles/html/profileView.do?email=mchol...@us.ibm.com&lang=en 
IBM

11400 Burnet Road
Austin, TX 78758
 
 
----- Original message -----
From: "Guilherme G. Piccoli" <gpicc...@linux.vnet.ibm.com>
To: Ian Munsie <imun...@au1.ibm.com>
Cc: mikey <mi...@neuling.org>, Michael C Hollinger/Austin/IBM@IBMUS, Frederic Barrat <frederic.bar...@fr.ibm.com>, linux-pci <linux-...@vger.kernel.org>, "Matthew R. Ochs" <mro...@linux.vnet.ibm.com>, gwshan <gws...@linux.vnet.ibm.com>, Manoj Kumar/Austin/IBM@IBMUS, paulus <pau...@samba.org>, "andrew.donnellan" <andrew.donnel...@au1.ibm.com>, bhelgaas <bhelg...@google.com>, linuxppc-dev <linuxppc-dev@lists.ozlabs.org>, Michael Ellerman <m...@ellerman.id.au>
Subject: Re: [PATCH v4] powerpc/pci: Assign fixed PHB number based on device-tree properties
Date: Wed, Apr 6, 2016 4:51 PM
 
On 04/06/2016 04:38 PM, Ian Munsie wrote:
>> +    /* try fixed PHB numbering first, by checking archs and reading
>> +     * the respective device-tree property. */
>> +    if (machine_is(pseries)) {
>> +        regs = of_get_property(dn, "reg", NULL);
>> +        if (regs)
>> +            return (int)(be32_to_cpu(regs[1]) & 0xFFFF);
>> +    } else if (machine_is(powernv)) {
>> +        prop64 = of_get_property(dn, "ibm,opal-phbid", NULL);
>> +        if (prop64)
>> +            return (int)(be64_to_cpup(prop64) & 0xFFFF);
>> +    }
>
> I think these cases should still set the bit in phb_bitmap, otherwise a
> virtual PHB (e.g. as used in cxl/cxlflash) will be assigned PHB 0, and
> since that is already taken it will fail - we're already seeing a
> failure in Ubuntu Xenial since Canonical picked this patch up already
> (though have not confirmed that this is definitely the cause yet).
>
> There might also be some interesting races to think about here if a
> virtual PHB grabs a PHB number before the real one gets a chance.

This is a very interesting case I didn't think before. Thanks for
pointing this Ian.

We can, as you suggested, set the bitmap in any case to avoid conflicts
with virtual PHBs.

And in the case a virtual PHB grabs the bitmap before, we just need to
add Michael's suggested check and fallback to bitmap PHB numbering in
this case.

Do you think this is enough to avoid issues with cxl'a virtual PHBs?

Thanks,


Guilherme

_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev

Reply via email to