From patchwork Wed Oct 14 12:27:34 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kashyap Desai X-Patchwork-Id: 11837597 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id DB72A921 for ; Wed, 14 Oct 2020 12:28:02 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id A8FCA208B3 for ; Wed, 14 Oct 2020 12:28:00 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=broadcom.com header.i=@broadcom.com header.b="hXbEtSB4" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730173AbgJNM2A (ORCPT ); Wed, 14 Oct 2020 08:28:00 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:40796 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1730076AbgJNM17 (ORCPT ); Wed, 14 Oct 2020 08:27:59 -0400 Received: from mail-ej1-x641.google.com (mail-ej1-x641.google.com [IPv6:2a00:1450:4864:20::641]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 52A00C061755 for ; Wed, 14 Oct 2020 05:27:59 -0700 (PDT) Received: by mail-ej1-x641.google.com with SMTP id lw21so4553698ejb.6 for ; Wed, 14 Oct 2020 05:27:59 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=broadcom.com; s=google; h=from:to:cc:subject:date:message-id; bh=iBBVQn92GDGWJOGXAvSWVXVKRMAwlAPva6OrF0Oo8QM=; b=hXbEtSB4X2AtSY+q3EJ8G6lN5lYTm/4pOvB/WsiShV/ih4ahkTe6cbG97oG5sFTBUb Kh08K2BDRrE9EAy3N1OnBgwLw8hO3IsoJwi0WAiDYs9e4s7UozfMhb2D8CEt10cqxaio 7OUCwtfyrkUok+U46VZriYRhxZxJUriYMVmm0= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id; bh=iBBVQn92GDGWJOGXAvSWVXVKRMAwlAPva6OrF0Oo8QM=; b=PbxJauesttv5tpkudIf8KY/CrlxeGI+ytKFEgLC/3iSvxY2DuZ+tBGFgBPC/1ZXTC1 vNcACkCE/TIt+7xcU8C1au+aoH6yAMcvVWeD9N8FMnfZbvKyk0Jv4u38srHnYi5hcKnI 2OWZSgscUYpgxHavuuLNO6JcfkPTQcuqvT3/hENgIaJwYwT85Sms1mIUeNu7F5Q7WzsT hBy3037rZPguV3ApwAwLrKz+1vQ/rj6f6eHZ2szYSpPz1oXRn1k6vw5RXFSh7Pmu92Qs mWhi3FNJ4VdTYxGYyyvm+1pN21RLQnWTDauLZA+nS8MhYZrIUAcafkkmRKa4uZd3IIgL aIRQ== X-Gm-Message-State: AOAM531Z+EFcAr0F5GC5t52gEoYEcjn/KpxPZfO3vJUyXM9sTrv94ydd obNHtDh7xZlGpAvDwL36/9ZgHQ== X-Google-Smtp-Source: ABdhPJxa8YXtCHj8mJYiSVYJdNLmyfslgoeV30FJ03rrdJPJO//OJUSKcMKWQm+slRhgUkK/yHK/gQ== X-Received: by 2002:a17:906:6bce:: with SMTP id t14mr5194238ejs.118.1602678477888; Wed, 14 Oct 2020 05:27:57 -0700 (PDT) Received: from drv-bst-rhel8.static.broadcom.net ([192.19.234.250]) by smtp.gmail.com with ESMTPSA id k21sm1505779edv.31.2020.10.14.05.27.55 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 14 Oct 2020 05:27:57 -0700 (PDT) From: Kashyap Desai To: linux-scsi@vger.kernel.org Cc: Kashyap Desai , dgilbert@interlog.com, linux-block@vger.kernel.org Subject: [PATCH 3/3] scsi_debug: iouring iopoll support Date: Wed, 14 Oct 2020 17:57:34 +0530 Message-Id: <20201014122734.201426-1-kashyap.desai@broadcom.com> X-Mailer: git-send-email 2.18.1 Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org Add support of iouring iopoll interface in scsi_debug. This feature requires shared hosttag support in kernel and driver. Signed-off-by: Kashyap Desai Cc: dgilbert@interlog.com Cc: linux-block@vger.kernel.org Reported-by: kernel test robot Reported-by: kernel test robot --- drivers/scsi/scsi_debug.c | 123 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 123 insertions(+) diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c index a87e40aec11f..4e441aaa60db 100644 --- a/drivers/scsi/scsi_debug.c +++ b/drivers/scsi/scsi_debug.c @@ -826,6 +826,7 @@ static int sdeb_zbc_max_open = DEF_ZBC_MAX_OPEN_ZONES; static int sdeb_zbc_nr_conv = DEF_ZBC_NR_CONV_ZONES; static int submit_queues = DEF_SUBMIT_QUEUES; /* > 1 for multi-queue (mq) */ +static int poll_queues; /* iouring iopoll interface.*/ static struct sdebug_queue *sdebug_q_arr; /* ptr to array of submit queues */ static DEFINE_RWLOCK(atomic_rw); @@ -5422,6 +5423,14 @@ static int schedule_resp(struct scsi_cmnd *cmnd, struct sdebug_dev_info *devip, cmnd->host_scribble = (unsigned char *)sqcp; sd_dp = sqcp->sd_dp; spin_unlock_irqrestore(&sqp->qc_lock, iflags); + + /* Do not complete IO from default completion path. + * Let it to be on queue. + * Completion should happen from mq_poll interface. + */ + if ((sqp - sdebug_q_arr) >= (submit_queues - poll_queues)) + return; + if (!sd_dp) { sd_dp = kzalloc(sizeof(*sd_dp), GFP_ATOMIC); if (!sd_dp) { @@ -5604,6 +5613,7 @@ module_param_named(sector_size, sdebug_sector_size, int, S_IRUGO); module_param_named(statistics, sdebug_statistics, bool, S_IRUGO | S_IWUSR); module_param_named(strict, sdebug_strict, bool, S_IRUGO | S_IWUSR); module_param_named(submit_queues, submit_queues, int, S_IRUGO); +module_param_named(poll_queues, poll_queues, int, S_IRUGO); module_param_named(tur_ms_to_ready, sdeb_tur_ms_to_ready, int, S_IRUGO); module_param_named(unmap_alignment, sdebug_unmap_alignment, int, S_IRUGO); module_param_named(unmap_granularity, sdebug_unmap_granularity, int, S_IRUGO); @@ -5673,6 +5683,7 @@ MODULE_PARM_DESC(sector_size, "logical block size in bytes (def=512)"); MODULE_PARM_DESC(statistics, "collect statistics on commands, queues (def=0)"); MODULE_PARM_DESC(strict, "stricter checks: reserved field in cdb (def=0)"); MODULE_PARM_DESC(submit_queues, "support for block multi-queue (def=1)"); +MODULE_PARM_DESC(poll_queues, "support for iouring iopoll queues"); MODULE_PARM_DESC(tur_ms_to_ready, "TEST UNIT READY millisecs before initial good status (def=0)"); MODULE_PARM_DESC(unmap_alignment, "lowest aligned thin provisioning lba (def=0)"); MODULE_PARM_DESC(unmap_granularity, "thin provisioning granularity in blocks (def=1)"); @@ -7140,6 +7151,104 @@ static int resp_not_ready(struct scsi_cmnd *scp, struct sdebug_dev_info *devip) return check_condition_result; } +static int sdebug_map_queues(struct Scsi_Host *shost) +{ + int i, qoff; + + if (shost->nr_hw_queues == 1) + return 0; + + for (i = 0, qoff = 0; i < HCTX_MAX_TYPES; i++) { + struct blk_mq_queue_map *map = &shost->tag_set.map[i]; + + map->nr_queues = 0; + + if (i == HCTX_TYPE_DEFAULT) + map->nr_queues = submit_queues - poll_queues; + else if (i == HCTX_TYPE_POLL) + map->nr_queues = poll_queues; + + if (!map->nr_queues) { + BUG_ON(i == HCTX_TYPE_DEFAULT); + continue; + } + + map->queue_offset = qoff; + blk_mq_map_queues(map); + + qoff += map->nr_queues; + } + + return 0; + +} + +int sdebug_blk_mq_poll(struct Scsi_Host *shost, unsigned int queue_num) +{ + int qc_idx; + int retiring = 0; + unsigned long iflags; + struct sdebug_queue *sqp; + struct sdebug_queued_cmd *sqcp; + struct scsi_cmnd *scp; + struct sdebug_dev_info *devip; + int num_entries = 0; + + sqp = sdebug_q_arr + queue_num; + + do { + spin_lock_irqsave(&sqp->qc_lock, iflags); + qc_idx = find_first_bit(sqp->in_use_bm, sdebug_max_queue); + if (unlikely((qc_idx < 0) || (qc_idx >= SDEBUG_CANQUEUE))) + goto out; + + sqcp = &sqp->qc_arr[qc_idx]; + scp = sqcp->a_cmnd; + if (unlikely(scp == NULL)) { + pr_err("scp is NULL, queue_num=%d, qc_idx=%d from %s\n", + queue_num, qc_idx, __func__); + goto out; + } + devip = (struct sdebug_dev_info *)scp->device->hostdata; + if (likely(devip)) + atomic_dec(&devip->num_in_q); + else + pr_err("devip=NULL from %s\n", __func__); + if (unlikely(atomic_read(&retired_max_queue) > 0)) + retiring = 1; + + sqcp->a_cmnd = NULL; + if (unlikely(!test_and_clear_bit(qc_idx, sqp->in_use_bm))) { + pr_err("Unexpected completion sqp %p queue_num=%d qc_idx=%d from %s\n", + sqp, queue_num, qc_idx, __func__); + goto out; + } + + if (unlikely(retiring)) { /* user has reduced max_queue */ + int k, retval; + + retval = atomic_read(&retired_max_queue); + if (qc_idx >= retval) { + pr_err("index %d too large\n", retval); + goto out; + } + k = find_last_bit(sqp->in_use_bm, retval); + if ((k < sdebug_max_queue) || (k == retval)) + atomic_set(&retired_max_queue, 0); + else + atomic_set(&retired_max_queue, k + 1); + } + spin_unlock_irqrestore(&sqp->qc_lock, iflags); + scp->scsi_done(scp); /* callback to mid level */ + num_entries++; + } while (1); + +out: + spin_unlock_irqrestore(&sqp->qc_lock, iflags); + return num_entries; +} + + static int scsi_debug_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scp) { @@ -7318,6 +7427,8 @@ static struct scsi_host_template sdebug_driver_template = { .ioctl = scsi_debug_ioctl, .queuecommand = scsi_debug_queuecommand, .change_queue_depth = sdebug_change_qdepth, + .map_queues = sdebug_map_queues, + .mq_poll = sdebug_blk_mq_poll, .eh_abort_handler = scsi_debug_abort, .eh_device_reset_handler = scsi_debug_device_reset, .eh_target_reset_handler = scsi_debug_target_reset, @@ -7365,6 +7476,18 @@ static int sdebug_driver_probe(struct device *dev) if (sdebug_host_max_queue) hpnt->host_tagset = 1; + /* poll queues are possible for nr_hw_queues > 1 */ + if (hpnt->nr_hw_queues == 1) + poll_queues = 0; + + /* poll queues */ + if (poll_queues >= submit_queues) { + pr_warn("%s: trim poll_queues to 1\n", my_name); + poll_queues = 1; + } + if (poll_queues) + hpnt->nr_maps = 3; + sdbg_host->shost = hpnt; *((struct sdebug_host_info **)hpnt->hostdata) = sdbg_host; if ((hpnt->this_id >= 0) && (sdebug_num_tgts > hpnt->this_id))