[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YoSV4CPBXbNlj0Cq@kroah.com>
Date: Wed, 18 May 2022 08:44:48 +0200
From: Greg KH <gregkh@...uxfoundation.org>
To: Kohei Tarumizu <tarumizu.kohei@...itsu.com>
Cc: catalin.marinas@....com, will@...nel.org, tglx@...utronix.de,
mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com,
x86@...nel.org, hpa@...or.com, rafael@...nel.org,
mchehab+huawei@...nel.org, eugenis@...gle.com, tony.luck@...el.com,
pcc@...gle.com, peterz@...radead.org, marcos@...a.pet,
conor.dooley@...rochip.com, nicolas.ferre@...rochip.com,
marcan@...can.st, linus.walleij@...aro.org, arnd@...db.de,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v4 6/8] x86: Add hardware prefetch control support for x86
On Wed, May 18, 2022 at 03:30:30PM +0900, Kohei Tarumizu wrote:
> Adds module init/exit code to create sysfs attributes for x86 with
> "hardware_prefetcher_enable", "ip_prefetcher_enable" and
> "adjacent_cache_line_prefetcher_enable".
>
> This driver works only if a CPU model is mapped to type of register
> specification(e.g. TYPE_L12_BASE) in pfctl_match[].
How will the driver be automatically loaded if this is the case or not?
You can not rely on userspace knowing to load a driver on its own,
please tie into the proper device discovery.
thanks,
greg k-h
Powered by blists - more mailing lists