All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: mingo@kernel.org, ravi.bangoria@amd.com, lucas.demarchi@intel.com
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org,
	willy@infradead.org, acme@kernel.org, namhyung@kernel.org,
	mark.rutland@arm.com, alexander.shishkin@linux.intel.com,
	jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com,
	kan.liang@linux.intel.com
Subject: [PATCH v2 07/24] perf: Fix perf_pmu_register() vs perf_init_event()
Date: Wed, 05 Feb 2025 11:21:27 +0100	[thread overview]
Message-ID: <20250205102449.006102855@infradead.org> (raw)
In-Reply-To: 20250205102120.531585416@infradead.org

There is a fairly obvious race between perf_init_event() doing
idr_find() and perf_pmu_register() doing idr_alloc() with an
incompletely initialized pmu pointer.

Avoid by doing idr_alloc() on a NULL pointer to register the id, and
swizzling the real pmu pointer at the end using idr_replace().

Also making sure to not set pmu members after publishing the pmu, duh.

[ introduce idr_cmpxchg() in order to better handle the idr_replace()
  error case -- if it were to return an unexpected pointer, it will
  already have replaced the value and there is no going back. ]

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 kernel/events/core.c |   28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -11739,6 +11739,21 @@ static int pmu_dev_alloc(struct pmu *pmu
 static struct lock_class_key cpuctx_mutex;
 static struct lock_class_key cpuctx_lock;
 
+static bool idr_cmpxchg(struct idr *idr, unsigned long id, void *old, void *new)
+{
+	void *tmp, *val = idr_find(idr, id);
+
+	if (val != old)
+		return false;
+
+	tmp = idr_replace(idr, new, id);
+	if (IS_ERR(tmp))
+		return false;
+
+	WARN_ON_ONCE(tmp != val);
+	return true;
+}
+
 int perf_pmu_register(struct pmu *pmu, const char *name, int type)
 {
 	int cpu, ret, max = PERF_TYPE_MAX;
@@ -11765,7 +11780,7 @@ int perf_pmu_register(struct pmu *pmu, c
 	if (type >= 0)
 		max = type;
 
-	ret = idr_alloc(&pmu_idr, pmu, max, 0, GFP_KERNEL);
+	ret = idr_alloc(&pmu_idr, NULL, max, 0, GFP_KERNEL);
 	if (ret < 0)
 		goto free_pdc;
 
@@ -11773,6 +11788,7 @@ int perf_pmu_register(struct pmu *pmu, c
 
 	type = ret;
 	pmu->type = type;
+	atomic_set(&pmu->exclusive_cnt, 0);
 
 	if (pmu_bus_running && !pmu->dev) {
 		ret = pmu_dev_alloc(pmu);
@@ -11821,14 +11837,22 @@ int perf_pmu_register(struct pmu *pmu, c
 	if (!pmu->event_idx)
 		pmu->event_idx = perf_event_idx_default;
 
+	/*
+	 * Now that the PMU is complete, make it visible to perf_try_init_event().
+	 */
+	if (!idr_cmpxchg(&pmu_idr, pmu->type, NULL, pmu))
+		goto free_context;
 	list_add_rcu(&pmu->entry, &pmus);
-	atomic_set(&pmu->exclusive_cnt, 0);
+
 	ret = 0;
 unlock:
 	mutex_unlock(&pmus_lock);
 
 	return ret;
 
+free_context:
+	free_percpu(pmu->cpu_pmu_context);
+
 free_dev:
 	if (pmu->dev && pmu->dev != PMU_NULL_DEV) {
 		device_del(pmu->dev);



  parent reply	other threads:[~2025-02-05 10:25 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-05 10:21 [PATCH v2 00/24] perf: Make perf_pmu_unregister() usable Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 01/24] lockdep: Fix might_fault() Peter Zijlstra
2025-02-06 18:19   ` David Hildenbrand
2025-02-05 10:21 ` [PATCH v2 02/24] perf: Ensure bpf_perf_link path is properly serialized Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 03/24] perf: Simplify child event tear-down Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 04/24] perf: Simplify perf_event_free_task() wait Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 05/24] perf: Simplify perf_event_release_kernel() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 06/24] perf: Fix pmus_lock vs pmus_srcu ordering Peter Zijlstra
2025-02-27 16:59   ` Lucas De Marchi
2025-02-05 10:21 ` Peter Zijlstra [this message]
2025-02-05 10:21 ` [PATCH v2 08/24] perf: Cleanup perf_try_init_event() Peter Zijlstra
2025-03-05 11:29   ` [tip: perf/core] perf/core: Clean up perf_try_init_event() tip-bot2 for Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 09/24] perf: Simplify perf_event_alloc() error path Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 10/24] perf: Simplify perf_pmu_register() " Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 11/24] perf: Simplify perf_pmu_register() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 12/24] perf: Simplify perf_init_event() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 13/24] perf: Simplify perf_event_alloc() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 14/24] perf: Merge pmu_disable_count into cpu_pmu_context Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 15/24] perf: Add this_cpc() helper Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 16/24] perf: Detach perf_cpu_pmu_context and pmu lifetimes Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 17/24] perf: Introduce perf_free_addr_filters() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 18/24] perf: Robustify perf_event_free_bpf_prog() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 19/24] perf: Simplify perf_mmap() control flow Peter Zijlstra
2025-03-03  5:39   ` Ravi Bangoria
2025-03-03 11:19     ` Ingo Molnar
2025-03-03 13:36       ` Ravi Bangoria
2025-03-04  8:44         ` Ingo Molnar
2025-02-05 10:21 ` [PATCH v2 20/24] perf: Fix perf_mmap() failure path Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 21/24] perf: Further simplify perf_mmap() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 22/24] perf: Remove retry loop from perf_mmap() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 23/24] perf: Lift event->mmap_mutex in perf_mmap() Peter Zijlstra
2025-02-05 10:21 ` [PATCH v2 24/24] perf: Make perf_pmu_unregister() useable Peter Zijlstra
2025-02-10  6:39   ` Ravi Bangoria
2025-02-11 15:46     ` Peter Zijlstra
2025-02-10  6:42   ` Ravi Bangoria
2025-02-12 12:49     ` Peter Zijlstra
2025-02-13  7:52       ` Ravi Bangoria
2025-02-13 13:08         ` Peter Zijlstra
2025-02-14  3:57           ` Ravi Bangoria
2025-02-14 20:24         ` Peter Zijlstra
2025-02-17  8:24           ` Ravi Bangoria
2025-02-17 16:31             ` Ravi Bangoria
2025-02-19 13:23               ` Ravi Bangoria
2025-02-19 14:30                 ` Ravi Bangoria
2025-02-10  6:59   ` Ravi Bangoria
2025-02-13 13:07     ` Peter Zijlstra
2025-03-03  6:01 ` [PATCH v2 00/24] perf: Make perf_pmu_unregister() usable Ravi Bangoria

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=20250205102449.006102855@infradead.org \
    --to=peterz@infradead.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lucas.demarchi@intel.com \
    --cc=mark.rutland@arm.com \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=ravi.bangoria@amd.com \
    --cc=willy@infradead.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.