From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Gordeev Subject: [PATCH 0/1] libata: Get rid of ata_port::qc_allocated bitmask Date: Tue, 21 Jan 2014 16:02:25 +0100 Message-ID: Return-path: Received: from mx1.redhat.com ([209.132.183.28]:63902 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754053AbaAUPAq (ORCPT ); Tue, 21 Jan 2014 10:00:46 -0500 Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Tejun Heo Cc: Alexander Gordeev , linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org, "Nicholas A. Bellinger" Hi Tejun, This update is against "scsi-mq" branch in Nicholas tree: git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending.git A quick check with 'fio' tool's random reading indicates an moderate improvement on a single drive, four threads and the queue depth of 32: read : io=131072KB, bw=149785B/s, iops=36, runt=896066msec vs read : io=131072KB, bw=166884B/s, iops=40, runt=804257msec I have not measured cache line bouncing itself. I am kind of limited in hardware configurations, but let me know if you want to see more testing anyway. Thanks! Alexander Gordeev (1): libata: Get rid of ata_port::qc_allocated bitmask drivers/ata/libata-core.c | 82 +++++++++++++++++++++++++++++++------------- include/linux/libata.h | 6 ++- 2 files changed, 62 insertions(+), 26 deletions(-) -- 1.7.7.6