From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752073AbcEJNTT (ORCPT ); Tue, 10 May 2016 09:19:19 -0400 Received: from mga01.intel.com ([192.55.52.88]:40471 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751937AbcEJNTQ (ORCPT ); Tue, 10 May 2016 09:19:16 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,604,1455004800"; d="scan'208";a="803030268" From: Alexander Shishkin To: Peter Zijlstra Cc: Thomas Gleixner , x86@kernel.org, Borislav Petkov , Ingo Molnar , linux-kernel@vger.kernel.org, vince@deater.net, eranian@google.com, Arnaldo Carvalho de Melo , Alexander Shishkin Subject: [PATCH 2/2] perf: Disable the event on a truncated AUX record Date: Tue, 10 May 2016 16:18:33 +0300 Message-Id: <1462886313-13660-3-git-send-email-alexander.shishkin@linux.intel.com> X-Mailer: git-send-email 2.8.0.rc3 In-Reply-To: <1462886313-13660-1-git-send-email-alexander.shishkin@linux.intel.com> References: <1462886313-13660-1-git-send-email-alexander.shishkin@linux.intel.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When the PMU driver reports a truncated AUX record, it effectively means that there is no more usable room in the event's AUX buffer (even though there may still be some room, so that perf_aux_output_begin() doesn't take action). At this point the consumer still has to be woken up and the event has to be disabled, otherwise the event will just keep spinning between perf_aux_output_begin() and perf_aux_output_end() until its context gets unscheduled. Again, for cpu-wide events this means never, so once in this condition, they will be forever losing data. Fix this by disabling the event and waking up the consumer in case of a truncated AUX record. Reported-by: Markus Metzger Signed-off-by: Alexander Shishkin --- kernel/events/ring_buffer.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 0f7f011710..5dd23b2d03 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -405,6 +405,7 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, bool truncated) { struct ring_buffer *rb = handle->rb; + bool wakeup = truncated; unsigned long aux_head; u64 flags = 0; @@ -433,9 +434,16 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, aux_head = rb->user_page->aux_head = local_read(&rb->aux_head); if (aux_head - local_read(&rb->aux_wakeup) >= rb->aux_watermark) { - perf_output_wakeup(handle); + wakeup = true; local_add(rb->aux_watermark, &rb->aux_wakeup); } + + if (wakeup) { + if (truncated) + handle->event->pending_disable = 1; + perf_output_wakeup(handle); + } + handle->event = NULL; local_set(&rb->aux_nest, 0); -- 2.8.1