Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754011Ab0HZPOX (ORCPT ); Thu, 26 Aug 2010 11:14:23 -0400 Received: from gir.skynet.ie ([193.1.99.77]:45004 "EHLO gir.skynet.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753953Ab0HZPOS (ORCPT ); Thu, 26 Aug 2010 11:14:18 -0400 From: Mel Gorman To: linux-mm@kvack.org, linux-fsdevel@vger.kernel.org Cc: Mel Gorman , Andrew Morton , Christian Ehrhardt , Johannes Weiner , Wu Fengguang , Jan Kara , linux-kernel@vger.kernel.org Subject: [PATCH 2/3] writeback: Record if the congestion was unnecessary Date: Thu, 26 Aug 2010 16:14:15 +0100 Message-Id: <1282835656-5638-3-git-send-email-mel@csn.ul.ie> X-Mailer: git-send-email 1.7.1 In-Reply-To: <1282835656-5638-1-git-send-email-mel@csn.ul.ie> References: <1282835656-5638-1-git-send-email-mel@csn.ul.ie> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3531 Lines: 108 If congestion_wait() is called when there is no congestion, the caller will wait for the full timeout. This can cause unreasonable and unnecessary stalls. There are a number of potential modifications that could be made to wake sleepers but this patch measures how serious the problem is. It keeps count of how many congested BDIs there are. If congestion_wait() is called with no BDIs congested, the tracepoint will record that the wait was unnecessary. Signed-off-by: Mel Gorman --- include/trace/events/writeback.h | 11 ++++++++--- mm/backing-dev.c | 15 ++++++++++++--- 2 files changed, 20 insertions(+), 6 deletions(-) diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h index e3bee61..03bb04b 100644 --- a/include/trace/events/writeback.h +++ b/include/trace/events/writeback.h @@ -155,19 +155,24 @@ DEFINE_WBC_EVENT(wbc_writepage); TRACE_EVENT(writeback_congest_waited, - TP_PROTO(unsigned int usec_delayed), + TP_PROTO(unsigned int usec_delayed, bool unnecessary), - TP_ARGS(usec_delayed), + TP_ARGS(usec_delayed, unnecessary), TP_STRUCT__entry( __field( unsigned int, usec_delayed ) + __field( unsigned int, unnecessary ) ), TP_fast_assign( __entry->usec_delayed = usec_delayed; + __entry->unnecessary = unnecessary; ), - TP_printk("usec_delayed=%u", __entry->usec_delayed) + TP_printk("usec_delayed=%u unnecessary=%d", + __entry->usec_delayed, + __entry->unnecessary + ) ); #endif /* _TRACE_WRITEBACK_H */ diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 7ae33e2..a49167f 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -724,6 +724,7 @@ static wait_queue_head_t congestion_wqh[2] = { __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) }; +static atomic_t nr_bdi_congested[2]; void clear_bdi_congested(struct backing_dev_info *bdi, int sync) { @@ -731,7 +732,8 @@ void clear_bdi_congested(struct backing_dev_info *bdi, int sync) wait_queue_head_t *wqh = &congestion_wqh[sync]; bit = sync ? BDI_sync_congested : BDI_async_congested; - clear_bit(bit, &bdi->state); + if (test_and_clear_bit(bit, &bdi->state)) + atomic_dec(&nr_bdi_congested[sync]); smp_mb__after_clear_bit(); if (waitqueue_active(wqh)) wake_up(wqh); @@ -743,7 +745,8 @@ void set_bdi_congested(struct backing_dev_info *bdi, int sync) enum bdi_state bit; bit = sync ? BDI_sync_congested : BDI_async_congested; - set_bit(bit, &bdi->state); + if (!test_and_set_bit(bit, &bdi->state)) + atomic_inc(&nr_bdi_congested[sync]); } EXPORT_SYMBOL(set_bdi_congested); @@ -760,14 +763,20 @@ long congestion_wait(int sync, long timeout) { long ret; unsigned long start = jiffies; + bool unnecessary = false; DEFINE_WAIT(wait); wait_queue_head_t *wqh = &congestion_wqh[sync]; + /* Check if this call to congestion_wait was necessary */ + if (atomic_read(&nr_bdi_congested[sync]) == 0) + unnecessary = true; + prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); ret = io_schedule_timeout(timeout); finish_wait(wqh, &wait); - trace_writeback_congest_waited(jiffies_to_usecs(jiffies - start)); + trace_writeback_congest_waited(jiffies_to_usecs(jiffies - start), + unnecessary); return ret; } -- 1.7.1 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/