All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jiri Olsa <jolsa@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: CAI Qian <caiqian@redhat.com>, Rob Herring <robh@kernel.org>,
	Kan Liang <kan.liang@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Ingo Molnar <mingo@kernel.org>
Subject: Re: [4.9-rc1+] intel_uncore builtin + CONFIG_DEBUG_TEST_DRIVER_REMOVE kernel panic
Date: Thu, 20 Oct 2016 10:58:03 +0200	[thread overview]
Message-ID: <20161020085803.GA31721@krava> (raw)
In-Reply-To: <20161020053944.GQ3102@twins.programming.kicks-ass.net>

On Thu, Oct 20, 2016 at 07:39:44AM +0200, Peter Zijlstra wrote:
> On Wed, Oct 19, 2016 at 09:19:43PM +0200, Jiri Olsa wrote:
> > I think the reason here is that presume pmu devices are always added,
> > but we add them only if pmu_bus_running (in perf_event_sysfs_init)
> > is set which might happen after uncore initcall
> > 
> > attached patch fixes the issue for me
> 
> Right, we never expected to be unloaded before userspace runs.
> 
> Strictly speaking we should only read pmu_bus_running while holding
> pmus_lock, that way we're serialized against perf_event_sysfs_init()
> flipping it while we're being removed etc..
> 
> With the current setup the introduced race is harmless, but who knows
> what other crazy these device people will come up with ;-)
> 

right, did not think of that ;-)

also I did not noticed device_remove_file call for pmu->nr_addr_filters
and we could save one lock/unlock call later.. I'm testing attached patch
now

thanks,
jirka


---
diff --git a/kernel/events/core.c b/kernel/events/core.c
index c6e47e97b33f..224dffbc3b9b 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -8581,24 +8581,24 @@ static void update_pmu_context(struct pmu *pmu, struct pmu *old_pmu)
 	}
 }
 
+/*
+ * The pmus_lock lock must be taken.
+ */
 static void free_pmu_context(struct pmu *pmu)
 {
 	struct pmu *i;
 
-	mutex_lock(&pmus_lock);
 	/*
 	 * Like a real lame refcount.
 	 */
 	list_for_each_entry(i, &pmus, entry) {
 		if (i->pmu_cpu_context == pmu->pmu_cpu_context) {
 			update_pmu_context(i, pmu);
-			goto out;
+			return;
 		}
 	}
 
 	free_percpu(pmu->pmu_cpu_context);
-out:
-	mutex_unlock(&pmus_lock);
 }
 
 /*
@@ -8869,11 +8869,15 @@ void perf_pmu_unregister(struct pmu *pmu)
 	free_percpu(pmu->pmu_disable_count);
 	if (pmu->type >= PERF_TYPE_MAX)
 		idr_remove(&pmu_idr, pmu->type);
-	if (pmu->nr_addr_filters)
-		device_remove_file(pmu->dev, &dev_attr_nr_addr_filters);
-	device_del(pmu->dev);
-	put_device(pmu->dev);
+	mutex_lock(&pmus_lock);
+	if (pmu_bus_running) {
+		if (pmu->nr_addr_filters)
+			device_remove_file(pmu->dev, &dev_attr_nr_addr_filters);
+		device_del(pmu->dev);
+		put_device(pmu->dev);
+	}
 	free_pmu_context(pmu);
+	mutex_unlock(&pmus_lock);
 }
 EXPORT_SYMBOL_GPL(perf_pmu_unregister);
 

  reply	other threads:[~2016-10-20  8:58 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <907882571.66590.1476113724660.JavaMail.zimbra@redhat.com>
2016-10-10 15:37 ` kasan inline + CONFIG_DEBUG_TEST_DRIVER_REMOVE kernel panic CAI Qian
2016-10-10 17:09   ` Rob Herring
2016-10-10 18:25     ` CAI Qian
2016-10-10 17:20   ` Greg Kroah-Hartman
2016-10-10 18:15     ` Rob Herring
2016-10-10 18:22       ` CAI Qian
2016-10-10 19:34         ` Rob Herring
2016-10-10 20:09           ` CAI Qian
2016-10-19 14:45       ` [4.9-rc1+] intel_uncore builtin " CAI Qian
2016-10-19 19:19         ` Jiri Olsa
2016-10-19 20:18           ` CAI Qian
2016-10-20  5:39           ` Peter Zijlstra
2016-10-20  8:58             ` Jiri Olsa [this message]
2016-10-20  9:04               ` Peter Zijlstra
2016-10-20  9:42                 ` Jiri Olsa
2016-10-20 11:10                   ` [PATCH] perf: Protect pmu device removal with pmu_bus_running check " Jiri Olsa
2016-10-20 14:30                     ` CAI Qian
2016-10-28 10:10                     ` [tip:perf/urgent] perf/core: Protect PMU device removal with a 'pmu_bus_running' check, to fix CONFIG_DEBUG_TEST_DRIVER_REMOVE=y " tip-bot for Jiri Olsa

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=20161020085803.GA31721@krava \
    --to=jolsa@redhat.com \
    --cc=caiqian@redhat.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=kan.liang@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=robh@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.