linux-kernel - Re: [PATCH v8 05/10] powerpc/perf: IMC pmu cpumask and cpuhotplug support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87o9uybvg3.fsf@concordia.ellerman.id.au>
Date:   Fri, 12 May 2017 13:33:32 +1000
From:   Michael Ellerman <mpe@...erman.id.au>
To:     Stewart Smith <stewart@...ux.vnet.ibm.com>,
        Madhavan Srinivasan <maddy@...ux.vnet.ibm.com>,
        Daniel Axtens <dja@...ens.net>,
        Anju T Sudhakar <anju@...ux.vnet.ibm.com>
Cc:     linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
        ego@...ux.vnet.ibm.com, bsingharora@...il.com, anton@...ba.org,
        sukadev@...ux.vnet.ibm.com, mikey@...ling.org, eranian@...gle.com,
        hemant@...ux.vnet.ibm.com
Subject: Re: [PATCH v8 05/10] powerpc/perf: IMC pmu cpumask and cpuhotplug support

Stewart Smith <stewart@...ux.vnet.ibm.com> writes:

> Madhavan Srinivasan <maddy@...ux.vnet.ibm.com> writes:
>>>      * in patch 9 should opal_imc_counters_init return something other
>>>        than OPAL_SUCCESS in the case on invalid arguments? Maybe
>>>        OPAL_PARAMETER? (I think you fix this in a later patch anyway?)
>>
>> So, init call will return OPAL_PARAMETER for the unsupported
>> domains (core and nest are supported). And if the init operation
>> fails for any reason, it would return OPAL_HARDWARE. And this is
>> documented.
>
> (I'll comment on the skiboot one too), but I think that if the class
> exists but init is a no-op, then OPAL_IMC_COUNTERS_INIT should return
> OPAL_SUCCESS and just do nothing. This future proofs everything, and the
> API is that one *must* call _INIT before start.

Yes, 100%.

That's what I described in my replies to a previous version, if it
doesn't do that we need to fix it.

cheers