From: "Pandruvada, Srinivas" <srinivas.pandruvada@intel.com>
To: "tglx@linutronix.de" <tglx@linutronix.de>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"peterz@infradead.org" <peterz@infradead.org>,
"Zhang, Rui" <rui.zhang@intel.com>,
"rt@linutronix.de" <rt@linutronix.de>,
"edubezval@gmail.com" <edubezval@gmail.com>,
"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
"x86@kernel.org" <x86@kernel.org>, "bp@alien8.de" <bp@alien8.de>
Subject: Re: [patch 00/12] thermal/x86_pkg_temp: Sanitize yet another hotplug and locking trainwreck
Date: Mon, 21 Nov 2016 23:16:19 +0000 [thread overview]
Message-ID: <1479770177.6544.195.camel@intel.com> (raw)
In-Reply-To: <alpine.DEB.2.20.1611212228220.3677@nanos>
On Mon, 2016-11-21 at 22:34 +0100, Thomas Gleixner wrote:
> On Mon, 21 Nov 2016, Pandruvada, Srinivas wrote:
[...]
> Stupid me. I tested putting a socket offline, which works, but did
> not
> check what happens on module removal. Delta fix below. That needs to
> be
> folded into the series as the wreckage already happens before the
> last
> patch.
Your change below fixes the crash issue. Now I tested a case where the
last cpu offlined from a package, it removed thermal zone and added
zone back once any cpu from the package onlined. So this is working.
I want to try to run some workload on those cpu to bump up the
temperature and check interrupts. I am hitting some issue unrelated to
this change may be. I onlined three cpus from the package 1.
[189443.567728] smpboot: Booting Node 1 Processor 15 APIC 0x2e
[189656.625947] smpboot: Booting Node 1 Processor 8 APIC 0x20
[189829.545851] smpboot: Booting Node 1 Processor 24 APIC 0x21
But I can't schedule anything on those CPUs. For example now can't run
turbostat, it complains
"
turbostat: re-initialized with num_cpus 19
Could not migrate to CPU 8
"
Same with
#taskset 0x100 stress -c 1
taskset: failed to set pid 0's affinity: Invalid argument
I am on the latest linux-pm/linux-next tree on this server. I will
switch to latest main line and try.
Thanks,
Srinivas
8<--------------------
--- a/drivers/thermal/x86_pkg_temp_thermal.c
+++ b/drivers/thermal/x86_pkg_temp_thermal.c
@@ -63,6 +63,7 @@ struct pkg_device {
u32 msr_pkg_therm_high;
struct delayed_work work;
struct thermal_zone_device *tzone;
+ struct cpumask cpumask;
};
static struct thermal_zone_params pkg_temp_tz_params = {
@@ -391,6 +392,7 @@ static int pkg_temp_thermal_device_add(u
rdmsr(MSR_IA32_PACKAGE_THERM_INTERRUPT, pkgdev-
>msr_pkg_therm_low,
pkgdev->msr_pkg_therm_high);
+ cpumask_set_cpu(cpu, &pkgdev->cpumask);
spin_lock_irq(&pkg_temp_lock);
packages[pkgid] = pkgdev;
spin_unlock_irq(&pkg_temp_lock);
@@ -399,13 +401,15 @@ static int pkg_temp_thermal_device_add(u
static int pkg_thermal_cpu_offline(unsigned int cpu)
{
- int target = cpumask_any_but(topology_core_cpumask(cpu), cpu);
struct pkg_device *pkgdev = pkg_temp_thermal_get_dev(cpu);
bool lastcpu, was_target;
+ int target;
if (!pkgdev)
return 0;
+ target = cpumask_any_but(&pkgdev->cpumask, cpu);
+ cpumask_clear_cpu(cpu, &pkgdev->cpumask);
lastcpu = target >= nr_cpu_ids;
/*
@@ -492,8 +496,10 @@ static int pkg_thermal_cpu_online(unsign
return -ENODEV;
/* If the package exists, nothing to do */
- if (pkgdev)
+ if (pkgdev) {
+ cpumask_set_cpu(cpu, &pkgdev->cpumask);
return 0;
+ }
return pkg_temp_thermal_device_add(cpu);
}
next prev parent reply other threads:[~2016-11-21 23:16 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-18 0:03 [patch 00/12] thermal/x86_pkg_temp: Sanitize yet another hotplug and locking trainwreck Thomas Gleixner
2016-11-18 0:03 ` [patch 01/12] thermal/x86_pkg_temp: Cleanup thermal interrupt handling Thomas Gleixner
2016-11-18 0:03 ` [patch 02/12] thermal/x86_pkg_temp: Remove redundant package search Thomas Gleixner
2016-11-18 0:03 ` [patch 03/12] thermal/x86_pkg_temp: Replace open coded cpu search Thomas Gleixner
2016-11-18 0:03 ` [patch 04/12] thermal/x86_pkg_temp: Sanitize callback (de)initialization Thomas Gleixner
2016-11-18 0:03 ` [patch 05/12] thermal/x86_pkg_temp: Get rid of ref counting Thomas Gleixner
2016-11-18 0:03 ` [patch 06/12] thermal/x86_pkg_temp: Cleanup namespace Thomas Gleixner
2016-11-18 0:03 ` [patch 07/12] thermal/x86_pkg_temp: Cleanup code some more Thomas Gleixner
2016-11-18 0:03 ` [patch 08/12] thermal/x86_pkg_temp: Sanitize locking Thomas Gleixner
2016-11-18 0:03 ` [patch 09/12] thermal/x86_pkg_temp: Move work scheduled flag into package struct Thomas Gleixner
2016-11-18 0:03 ` [patch 10/12] thermal/x86_pkg_temp: Move work " Thomas Gleixner
2016-11-18 0:03 ` [patch 11/12] thermal/x86_pkg_temp: Sanitize package management Thomas Gleixner
2016-11-18 0:03 ` [patch 12/12] thermal/x86 pkg temp: Convert to hotplug state machine Thomas Gleixner
2016-11-21 20:02 ` [patch 00/12] thermal/x86_pkg_temp: Sanitize yet another hotplug and locking trainwreck Pandruvada, Srinivas
2016-11-21 21:34 ` Thomas Gleixner
2016-11-21 23:16 ` Pandruvada, Srinivas [this message]
2016-11-22 9:05 ` Thomas Gleixner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1479770177.6544.195.camel@intel.com \
--to=srinivas.pandruvada@intel.com \
--cc=bp@alien8.de \
--cc=edubezval@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=rt@linutronix.de \
--cc=rui.zhang@intel.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).