[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87r2u8tvu7.fsf@concordia.ellerman.id.au>
Date: Thu, 12 Oct 2017 23:05:36 +1100
From: Michael Ellerman <mpe@...erman.id.au>
To: Anju T Sudhakar <anju@...ux.vnet.ibm.com>,
Stewart Smith <stewart@...ux.vnet.ibm.com>
Cc: linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
maddy@...ux.vnet.ibm.com, hemant@...ux.vnet.ibm.com
Subject: Re: [PATCH] powerpc/powernv: Add kernel cmdline parameter to disable imc
Anju T Sudhakar <anju@...ux.vnet.ibm.com> writes:
> Hi mpe, stewart,
>
>
> On Wednesday 11 October 2017 01:55 AM, Stewart Smith wrote:
>> Michael Ellerman <mpe@...erman.id.au> writes:
>>> Anju T Sudhakar <anju@...ux.vnet.ibm.com> writes:
>>>
>>>> Add a kernel command line parameter option to disable In-Memory Collection
>>>> (IMC) counters and add documentation. This helps in debug.
>>> I'd really rather we didn't. Do we *really* need this?
>>>
>>> We don't have command line parameters to disable any of the other ~20
>>> PMUs, why is this one special?
>
> This one is really helpful in debugging, incase if we want to proceed
> without nest counters OR
> core counters . But if we have the facility to do the same from
> petitboot, its fine.
I understand it's helpful for debugging.
But we can't afford to have a command line option to disable every
~1K SLOC of code.
If there's a compelling reason why we need to disable IMC *on production
systems* - then a command line option is a possibility. Though really if
there is a compelling reason for that maybe we have bigger problems :)
This is also a very new driver, so we're still finding some bugs, in a
few months it should be more stable and then the command line option
will just be cruft.
cheers
Powered by blists - more mailing lists