[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <87wp9fdifx.fsf@linux.vnet.ibm.com>
Date: Wed, 17 May 2017 17:57:22 +1000
From: Stewart Smith <stewart@...ux.vnet.ibm.com>
To: Anju T Sudhakar <anju@...ux.vnet.ibm.com>, mpe@...erman.id.au
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, dja@...ens.net,
eranian@...gle.com, hemant@...ux.vnet.ibm.com,
maddy@...ux.vnet.ibm.com, anju@...ux.vnet.ibm.com
Subject: Re: [PATCH v8 07/10] powerpc/perf: PMU functions for Core IMC and hotplugging
Anju T Sudhakar <anju@...ux.vnet.ibm.com> writes:
> --- a/arch/powerpc/include/asm/imc-pmu.h
> +++ b/arch/powerpc/include/asm/imc-pmu.h
> @@ -24,6 +24,7 @@
> */
> #define IMC_MAX_CHIPS 32
> #define IMC_MAX_PMUS 32
> +#define IMC_MAX_CORES 32
>
> /*
> * This macro is used for memory buffer allocation of
> @@ -38,6 +39,11 @@
> #define IMC_NEST_MAX_PAGES 64
>
> /*
> + * IMC Core engine expects 8K bytes of memory for counter collection.
> + */
> +#define IMC_CORE_COUNTER_MEM 8192
Any reason for this not to be in the device tree?
This is the size of memory that Linux needs to give OPAL, so the size of
that should probably come from OPAL rather than Linux.
Otherwise, if in the future, we had a counter with an offset greater
than 8192 in, we'd have no way to detect that and we'd fail silently by
overwriting memory.
--
Stewart Smith
OPAL Architect, IBM.
Powered by blists - more mailing lists