From patchwork Thu Jun 5 12:44:19 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dolev Raviv X-Patchwork-Id: 4304891 Return-Path: X-Original-To: patchwork-linux-arm-msm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 62A97BEEA7 for ; Thu, 5 Jun 2014 12:44:40 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 4D8B820320 for ; Thu, 5 Jun 2014 12:44:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 42E3E20306 for ; Thu, 5 Jun 2014 12:44:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751365AbaFEMoh (ORCPT ); Thu, 5 Jun 2014 08:44:37 -0400 Received: from smtp.codeaurora.org ([198.145.11.231]:48441 "EHLO smtp.codeaurora.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750939AbaFEMog (ORCPT ); Thu, 5 Jun 2014 08:44:36 -0400 Received: from smtp.codeaurora.org (localhost [127.0.0.1]) by smtp.codeaurora.org (Postfix) with ESMTP id 03F9B13F30A; Thu, 5 Jun 2014 12:44:36 +0000 (UTC) Received: by smtp.codeaurora.org (Postfix, from userid 486) id EB33113F116; Thu, 5 Jun 2014 12:44:35 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Spam-Level: X-Spam-Status: No, score=-7.5 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=ham version=3.3.1 Received: from lx-draviv2.mea.qualcomm.com (unknown [185.23.60.4]) (using TLSv1.1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: draviv@smtp.codeaurora.org) by smtp.codeaurora.org (Postfix) with ESMTPSA id BD6EB13F116; Thu, 5 Jun 2014 12:44:33 +0000 (UTC) From: Dolev Raviv To: James.Bottomley@HansenPartnership.com Cc: linux-scsi@vger.kernel.org, linux-scsi-owner@vger.kernel.org, sthumma@codeaurora.org, linux-arm-msm@vger.kernel.org, santoshsy@gmail.com, Dolev Raviv Subject: [Resend PATCH V1 4/4] scsi: ufs: Fix queue depth handling for best effort cases Date: Thu, 5 Jun 2014 15:44:19 +0300 Message-Id: <1401972259-15471-2-git-send-email-draviv@codeaurora.org> X-Mailer: git-send-email 1.8.5.2 In-Reply-To: <1401972259-15471-1-git-send-email-draviv@codeaurora.org> References: <1401972259-15471-1-git-send-email-draviv@codeaurora.org> X-Virus-Scanned: ClamAV using ClamSMTP Sender: linux-arm-msm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-arm-msm@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Sujit Reddy Thumma Some UFS devices may expose bLUQueueDepth field as zero indicating that the queue depth depends on the number of resources available for LUN at a particular instant to handle the outstanding transfer requests. Currently, when response for SCSI command is TASK_FULL the LLD decrements the queue depth but fails to increment when the resources are available. The scsi mid-layer handles the change in queue depth heuristically and offers simple interface with ->change_queue_depth. Signed-off-by: Sujit Reddy Thumma Signed-off-by: Dolev Raviv --- drivers/scsi/ufs/ufshcd.c | 97 ++++++++++++++++++++--------------------------- 1 file changed, 42 insertions(+), 55 deletions(-) diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index b301ed8..b103e95 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -1991,27 +1991,55 @@ static int ufshcd_slave_alloc(struct scsi_device *sdev) /* allow SCSI layer to restart the device in case of errors */ sdev->allow_restart = 1; + lun_qdepth = ufshcd_read_sdev_qdepth(hba, sdev); - if (lun_qdepth == 0 || lun_qdepth > hba->nutrs) { - dev_info(hba->dev, "%s, lun %d queue depth is %d\n", __func__, - sdev->lun, lun_qdepth); + if (lun_qdepth <= 0) + /* eventually, we can figure out the real queue depth */ lun_qdepth = hba->nutrs; - } else if (lun_qdepth < 0) { - lun_qdepth = 1; - } + else + lun_qdepth = min_t(int, lun_qdepth, hba->nutrs); - /* - * Inform SCSI Midlayer that the LUN queue depth is same as the - * controller queue depth. If a LUN queue depth is less than the - * controller queue depth and if the LUN reports - * SAM_STAT_TASK_SET_FULL, the LUN queue depth will be adjusted - * with scsi_adjust_queue_depth. - */ + dev_dbg(hba->dev, "%s: activate tcq with queue depth %d\n", + __func__, lun_qdepth); scsi_activate_tcq(sdev, lun_qdepth); + return 0; } /** + * ufshcd_change_queue_depth - change queue depth + * @sdev: pointer to SCSI device + * @depth: required depth to set + * @reason: reason for changing the depth + * + * Change queue depth according to the reason and make sure + * the max. limits are not crossed. + */ +int ufshcd_change_queue_depth(struct scsi_device *sdev, int depth, int reason) +{ + struct ufs_hba *hba = shost_priv(sdev->host); + + if (depth > hba->nutrs) + depth = hba->nutrs; + + switch (reason) { + case SCSI_QDEPTH_DEFAULT: + case SCSI_QDEPTH_RAMP_UP: + if (!sdev->tagged_supported) + depth = 1; + scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); + break; + case SCSI_QDEPTH_QFULL: + scsi_track_queue_full(sdev, depth); + break; + default: + return -EOPNOTSUPP; + } + + return depth; +} + +/** * ufshcd_slave_destroy - remove SCSI device configurations * @sdev: pointer to SCSI device */ @@ -2064,42 +2092,6 @@ static int ufshcd_task_req_compl(struct ufs_hba *hba, u32 index, u8 *resp) } /** - * ufshcd_adjust_lun_qdepth - Update LUN queue depth if device responds with - * SAM_STAT_TASK_SET_FULL SCSI command status. - * @cmd: pointer to SCSI command - */ -static void ufshcd_adjust_lun_qdepth(struct scsi_cmnd *cmd) -{ - struct ufs_hba *hba; - int i; - int lun_qdepth = 0; - - hba = shost_priv(cmd->device->host); - - /* - * LUN queue depth can be obtained by counting outstanding commands - * on the LUN. - */ - for (i = 0; i < hba->nutrs; i++) { - if (test_bit(i, &hba->outstanding_reqs)) { - - /* - * Check if the outstanding command belongs - * to the LUN which reported SAM_STAT_TASK_SET_FULL. - */ - if (cmd->device->lun == hba->lrb[i].lun) - lun_qdepth++; - } - } - - /* - * LUN queue depth will be total outstanding commands, except the - * command for which the LUN reported SAM_STAT_TASK_SET_FULL. - */ - scsi_adjust_queue_depth(cmd->device, MSG_SIMPLE_TAG, lun_qdepth - 1); -} - -/** * ufshcd_scsi_cmd_status - Update SCSI command result based on SCSI status * @lrb: pointer to local reference block of completed command * @scsi_status: SCSI command status @@ -2120,12 +2112,6 @@ ufshcd_scsi_cmd_status(struct ufshcd_lrb *lrbp, int scsi_status) scsi_status; break; case SAM_STAT_TASK_SET_FULL: - /* - * If a LUN reports SAM_STAT_TASK_SET_FULL, then the LUN queue - * depth needs to be adjusted to the exact number of - * outstanding commands the LUN can handle at any given time. - */ - ufshcd_adjust_lun_qdepth(lrbp->cmd); case SAM_STAT_BUSY: case SAM_STAT_TASK_ABORTED: ufshcd_copy_sense_data(lrbp); @@ -3156,6 +3142,7 @@ static struct scsi_host_template ufshcd_driver_template = { .queuecommand = ufshcd_queuecommand, .slave_alloc = ufshcd_slave_alloc, .slave_destroy = ufshcd_slave_destroy, + .change_queue_depth = ufshcd_change_queue_depth, .eh_abort_handler = ufshcd_abort, .eh_device_reset_handler = ufshcd_eh_device_reset_handler, .eh_host_reset_handler = ufshcd_eh_host_reset_handler,