From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp2130.oracle.com ([156.151.31.86]) by Galois.linutronix.de with esmtps (TLS1.2:RSA_AES_256_CBC_SHA256:256) (Exim 4.80) (envelope-from ) id 1fQwsS-0007Hj-BG for speck@linutronix.de; Thu, 07 Jun 2018 17:31:05 +0200 Received: from pps.filterd (userp2130.oracle.com [127.0.0.1]) by userp2130.oracle.com (8.16.0.22/8.16.0.22) with SMTP id w57FQGKO153009 for ; Thu, 7 Jun 2018 15:30:56 GMT Received: from userv0022.oracle.com (userv0022.oracle.com [156.151.31.74]) by userp2130.oracle.com with ESMTP id 2jbvyphh28-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for ; Thu, 07 Jun 2018 15:30:56 +0000 Received: from userv0121.oracle.com (userv0121.oracle.com [156.151.31.72]) by userv0022.oracle.com (8.14.4/8.14.4) with ESMTP id w57FUt7W025810 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for ; Thu, 7 Jun 2018 15:30:55 GMT Received: from abhmp0004.oracle.com (abhmp0004.oracle.com [141.146.116.10]) by userv0121.oracle.com (8.14.4/8.13.8) with ESMTP id w57FUtHI017962 for ; Thu, 7 Jun 2018 15:30:55 GMT Date: Thu, 7 Jun 2018 11:30:54 -0400 From: Konrad Rzeszutek Wilk Subject: [MODERATED] Re: [patch V2 00/12] cpu/hotplug: SMT control Message-ID: <20180607153054.GQ21334@char.us.oracle.com> References: <20180606192714.754943543@linutronix.de> MIME-Version: 1.0 In-Reply-To: <20180606192714.754943543@linutronix.de> Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit To: speck@linutronix.de List-ID: On Wed, Jun 06, 2018 at 09:27:14PM +0200, speck for Thomas Gleixner wrote: > The following series is a reworked version of the initial proof of concept > patch. The main changes are: > > - The primary sibling evaluation has been changed to use APIC ID, so the > hacky stuff is gone. > > - The control has now 3 states: on, off, forceoff > > forceoff is a irreversible operation and if given on the command line > via 'nosmt=force' it makes the processor/APIC enumeration code discard > the non primary siblings. That affects also the number of possible CPUs > and is more or less equivalent to disabling SMT in the BIOS. > > If 'forceoff' is written to the sysfs file, then the non primamry > siblings are offlined as with 'off', but the operation cannot be > undone. That has obviously no effect on num_possible_cpus as that has > been evaluated in the early boot process. > > - Command line and sysfs interface are documented > > Survived testing on various Intel and AMD machines and in VMs of different > flavours and topology configurations. > > Applies on top of Linus tree. Any chance you could provide an git bundle for those lazy folks who like to compile and test it out?