RE: [PATCH v5 0/6] Add hardware prefetch control driver for A64FX and x86

From: "Luck, Tony" <tony.luck@intel.com>
To: "tarumizu.kohei@fujitsu.com" <tarumizu.kohei@fujitsu.com>,
	'Linus Walleij' <linus.walleij@linaro.org>
Cc: "rafael@kernel.org" <rafael@kernel.org>,
	"peterz@infradead.org" <peterz@infradead.org>,
	"catalin.marinas@arm.com" <catalin.marinas@arm.com>,
	"dave.hansen@linux.intel.com" <dave.hansen@linux.intel.com>,
	"ast@kernel.org" <ast@kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"conor.dooley@microchip.com" <conor.dooley@microchip.com>,
	"hpa@zytor.com" <hpa@zytor.com>,
	"will@kernel.org" <will@kernel.org>,
	"mchehab+huawei@kernel.org" <mchehab+huawei@kernel.org>,
	"x86@kernel.org" <x86@kernel.org>,
	"linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
	"mingo@redhat.com" <mingo@redhat.com>,
	"kuba@kernel.org" <kuba@kernel.org>,
	"eugenis@google.com" <eugenis@google.com>,
	"lenb@kernel.org" <lenb@kernel.org>,
	"arnd@arndb.de" <arnd@arndb.de>, "bp@alien8.de" <bp@alien8.de>,
	"tglx@linutronix.de" <tglx@linutronix.de>,
	"pcc@google.com" <pcc@google.com>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"marcos@orca.pet" <marcos@orca.pet>,
	Greg KH <gregkh@linuxfoundation.org>,
	"peter.chen@kernel.org" <peter.chen@kernel.org>,
	"marcan@marcan.st" <marcan@marcan.st>
Subject: RE: [PATCH v5 0/6] Add hardware prefetch control driver for A64FX and x86
Date: Tue, 28 Jun 2022 16:39:31 +0000	[thread overview]
Message-ID: <c2fb1376536d471f88b58ad82859a183@intel.com> (raw)
In-Reply-To: <OSBPR01MB203739DD69FB8C50990C356A80B89@OSBPR01MB2037.jpnprd01.prod.outlook.com>

>> The right way to solve this is to make the Linux kernel contain the necessary
>> heuristics to identify which tasks and thus cores need this to improve efficiency
>> and then apply it automatically.
>>
>> Putting it in userspace is making a human do a machines job which isn't
>> sustainable.
>>
>> By putting the heuristics in kernelspace Linux will improve performance also on
>> workloads the human operator didn't think of as the machine will detect them from
>> statictical or other behaviour patterns.
>
>In order to put the heuristics into kernelspace Linux, I think it
>necessary to consider the following two points.
>
>1) Which cores are tied with the process?
>This is different from the core on which the process can run. It
>probably need to combine some CPU resource limit to avoid affecting
>non-target processes.
>
>2) How to derive the value to set in the register?
>It is necessary to verify whether an appropriate set value can be
>derived using statistical information, etc. In addition, to prevent
>the cost of automatic derivation from exceeding the value that would
>be improved by it.
>
>I don't have a prospect for resolving these issues yet. I will
>continue these considerations.

Another approach would be to make the set of prefetch settings
a task attribute. Then set them in the context switch code when
the process is about to run on a CPU.

But that assumes you can cheaply change the attributes. If doing
so requires multiple MSR writes (on x86) it might be a non-starter.

-Tony
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel