MPAM branch verification (was RE: [RFC PATCH 2/2] ACPI / PPTT: cacheinfo: Label caches based on fw_token)

From: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
To: James Morse <james.morse@arm.com>
Cc: "linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
	Vijaya Kumar K <vkilari@codeaurora.org>,
	Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
	Jeffrey Hugo <jhugo@codeaurora.org>,
	Sudeep Holla <sudeep.holla@arm.com>,
	Jeremy Linton <jeremy.linton@arm.com>,
	Tomasz Nowicki <Tomasz.Nowicki@cavium.com>,
	Richard Ruigrok <rruigrok@qti.qualcomm.com>,
	"Guohanjun (Hanjun Guo)" <guohanjun@huawei.com>,
	"wangxiongfeng (C)" <wangxiongfeng2@huawei.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	Linuxarm <linuxarm@huawei.com>
Subject: MPAM branch verification (was RE: [RFC PATCH 2/2] ACPI / PPTT: cacheinfo: Label caches based on fw_token)
Date: Fri, 21 Jun 2019 15:57:14 +0000	[thread overview]
Message-ID: <5FC3163CFD30C246ABAA99954A238FA83F2B38FD@lhreml524-mbs.china.huawei.com> (raw)

Hi James,

> -----Original Message-----
> From: James Morse [mailto:james.morse@arm.com]
> Sent: 19 June 2019 14:31
> To: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
> Cc: linux-acpi@vger.kernel.org; Vijaya Kumar K <vkilari@codeaurora.org>;
> Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>; Jeffrey Hugo
> <jhugo@codeaurora.org>; Sudeep Holla <sudeep.holla@arm.com>; Jeremy
> Linton <jeremy.linton@arm.com>; Tomasz Nowicki
> <Tomasz.Nowicki@cavium.com>; Richard Ruigrok
> <rruigrok@qti.qualcomm.com>; Guohanjun (Hanjun Guo)
> <guohanjun@huawei.com>; wangxiongfeng (C)
> <wangxiongfeng2@huawei.com>; linux-arm-kernel@lists.infradead.org;
> Linuxarm <linuxarm@huawei.com>
> Subject: Re: [RFC PATCH 2/2] ACPI / PPTT: cacheinfo: Label caches based on
> fw_token
> 
> Hi Shameer,
> 
[...] 
> > I was just trying out the latest public MPAM branch available here[1]
> 
> Great!
> 
> 
> > and noted that
> > on our HiSilicon platform all the L3 cache were labeled with the same Id.
> Debugging> revealed that the above leaf node check was removed in this
> branch[2] which makes
> > the min_physid calculation going wrong.
> 
> Thanks for debugging this,
> 
> > Just wondering is there any particular reason
> > for removing the check or the branch is not carrying the latest patch?
> 
> Nope, that's a bug.
> 
> Jeremy Linton's review feedback[0] was that that PROCESSOR_ID_VALID flag
> can't be relied
> on. It looks like I over-zealously removed the whole if(), and this doesn't cause a
> problem with my pptt so I didn't notice.
> 
> I've fixed it locally, I've also pushed a fix to those branches, but it will get folded
> in
> next time I push a branch.

Thanks for that.

Apart from the above, I have come across few other issues as well and had some
temporary fixes to the branch here[0]. This is encountered while trying to get the
resctrl fs mounted and attempted a cqm test run using resctrl_tests tool. 

The fixes may not be proper ones, but I think it will give you an idea. Please take a
look and let me know your thoughts.

Thanks,
Shameer

[0] https://github.com/hisilicon/kernel-dev.git  branch: private-dbg-mpam-5.2-rc1

> 
> Thanks!
> 
> James
> 
> [0] lore.kernel.org/r/a68abfd2-1e28-d9e7-919a-8b3133db4d20@arm.com