From patchwork Thu Mar 29 22:04:55 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Matias Bjorling X-Patchwork-Id: 10316589 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id BD63F60383 for ; Thu, 29 Mar 2018 22:15:18 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id A2C0C2A4B9 for ; Thu, 29 Mar 2018 22:15:18 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 970AB2A536; Thu, 29 Mar 2018 22:15:18 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id B6EA12A4B9 for ; Thu, 29 Mar 2018 22:15:17 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752731AbeC2WPI (ORCPT ); Thu, 29 Mar 2018 18:15:08 -0400 Received: from mail-lf0-f65.google.com ([209.85.215.65]:33161 "EHLO mail-lf0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752234AbeC2WHF (ORCPT ); Thu, 29 Mar 2018 18:07:05 -0400 Received: by mail-lf0-f65.google.com with SMTP id x70-v6so3501312lfa.0 for ; Thu, 29 Mar 2018 15:07:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=lightnvm-io.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=KbWGL89TkiJvR3PhvZPn4MQnPkIwmLlljeGPQ/sdLjc=; b=i409WwvdYa2mvqC4zOr//G/Z55Vlf6sphuXRgaYQBxznYkg3uekvnC+qBW/Mfm9VMs eq0V2vge5zdzexDFhadsb/70AMLcPQPG2x/7opnKwSUGiJQlQtAPOq7MNySW4unrJBTL lCFCwJjHmhF4STTV89zgbcceTUQvsq6SpBmIK3/IG/LrxRmRg9PdohO1d4LaJYhk60sS oB5hcEtthH1g5HpsASg4rrNp1yoy2ZT+vtHM8kgzVv3CGRDLp8z8kOblfKXma02S6/vm 6w4xQpDB5hnoPtC26g+2uOHxZlX+iBVOluaFH9raEDJisBYAkJwxHe8xxYAgKqCM6193 AmIQ== 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:in-reply-to :references:mime-version:content-transfer-encoding; bh=KbWGL89TkiJvR3PhvZPn4MQnPkIwmLlljeGPQ/sdLjc=; b=M9sAe6axhSG19a0a19tZXT1cLNqhKxWaE39W9bB4DhCyCE7e/2zxkvekqiMHXTx+5c 49QSjK7W33z9ziKGQVkLlnRSKD2auUJDesvLovoe7T2lM2dg6cTiyaP1LJYpQNNxy+Sh jzx/plx1Zt/VCy4fzE3muFheIhOq6niPYPd8+5sxbsuNETutfICjG43p1/FTcQIXKUi8 IzohNk9l4Vt1fwQjCAKLZBbqZhxGuIXgOYsAYSdM03N4MTM42t73SAmHIuTeuu2mQcvN X901qSpzzYrCRX7n0waXUITK2WgIUYHpUqTnS3Lo5OpyvIkSttwOpQr8bfL16i76EUS/ QpIQ== X-Gm-Message-State: ALQs6tCeJijF1nI91sZTCHvAbYvIOxuzcGFvIYUzzLVUXfQttCvkjVKz sfI65646dYGoAXnOBn8NvXWExhh7 X-Google-Smtp-Source: AIpwx49s4nZ2VSuQt8brYbkFPl5/6m0Oe9IUUiTUiMylZc2xt3Vz1mD1dn3jP6Jw8vS+lS2oJKYC+A== X-Received: by 2002:a19:d085:: with SMTP id h127-v6mr2767368lfg.29.1522361224177; Thu, 29 Mar 2018 15:07:04 -0700 (PDT) Received: from Macroninja.cnexlabs.com (x1-6-a4-08-f5-18-3c-3a.cpe.webspeed.dk. [188.176.29.198]) by smtp.gmail.com with ESMTPSA id i89-v6sm1365508lfk.12.2018.03.29.15.07.03 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 29 Mar 2018 15:07:03 -0700 (PDT) From: =?UTF-8?q?Matias=20Bj=C3=B8rling?= To: axboe@fb.com Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, keith.busch@intel.com, javier@cnexlabs.com, Hans Holmberg , =?UTF-8?q?Matias=20Bj=C3=B8rling?= Subject: [GIT PULL 08/37] lightnvm: pblk: add padding distribution sysfs attribute Date: Fri, 30 Mar 2018 00:04:55 +0200 Message-Id: <20180329220524.30363-9-mb@lightnvm.io> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20180329220524.30363-1-mb@lightnvm.io> References: <20180329220524.30363-1-mb@lightnvm.io> MIME-Version: 1.0 Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Hans Holmberg When pblk receives a sync, all data up to that point in the write buffer must be comitted to persistent storage, and as flash memory comes with a minimal write size there is a significant cost involved both in terms of time for completing the sync and in terms of write amplification padded sectors for filling up to the minimal write size. In order to get a better understanding of the costs involved for syncs, Add a sysfs attribute to pblk: padded_dist, showing a normalized distribution of sectors padded. In order to facilitate measurements of specific workloads during the lifetime of the pblk instance, the distribution can be reset by writing 0 to the attribute. Do this by introducing counters for each possible padding: {0..(minimal write size - 1)} and calculate the normalized distribution when showing the attribute. Signed-off-by: Hans Holmberg Signed-off-by: Javier González Rearranged total_buckets statement in pblk_sysfs_get_padding_dist Signed-off-by: Matias Bjørling --- drivers/lightnvm/pblk-init.c | 16 +++++++- drivers/lightnvm/pblk-rb.c | 17 +++++---- drivers/lightnvm/pblk-sysfs.c | 87 ++++++++++++++++++++++++++++++++++++++++++- drivers/lightnvm/pblk.h | 6 ++- 4 files changed, 113 insertions(+), 13 deletions(-) diff --git a/drivers/lightnvm/pblk-init.c b/drivers/lightnvm/pblk-init.c index 0ffc17ccf1cc..8416910ee8bf 100644 --- a/drivers/lightnvm/pblk-init.c +++ b/drivers/lightnvm/pblk-init.c @@ -921,6 +921,7 @@ static void pblk_free(struct pblk *pblk) { pblk_luns_free(pblk); pblk_lines_free(pblk); + kfree(pblk->pad_dist); pblk_line_meta_free(pblk); pblk_core_free(pblk); pblk_l2p_free(pblk); @@ -998,11 +999,13 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, pblk->pad_rst_wa = 0; pblk->gc_rst_wa = 0; + atomic64_set(&pblk->nr_flush, 0); + pblk->nr_flush_rst = 0; + #ifdef CONFIG_NVM_DEBUG atomic_long_set(&pblk->inflight_writes, 0); atomic_long_set(&pblk->padded_writes, 0); atomic_long_set(&pblk->padded_wb, 0); - atomic_long_set(&pblk->nr_flush, 0); atomic_long_set(&pblk->req_writes, 0); atomic_long_set(&pblk->sub_writes, 0); atomic_long_set(&pblk->sync_writes, 0); @@ -1034,10 +1037,17 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, goto fail_free_luns; } + pblk->pad_dist = kzalloc((pblk->min_write_pgs - 1) * sizeof(atomic64_t), + GFP_KERNEL); + if (!pblk->pad_dist) { + ret = -ENOMEM; + goto fail_free_line_meta; + } + ret = pblk_core_init(pblk); if (ret) { pr_err("pblk: could not initialize core\n"); - goto fail_free_line_meta; + goto fail_free_pad_dist; } ret = pblk_l2p_init(pblk); @@ -1097,6 +1107,8 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, pblk_l2p_free(pblk); fail_free_core: pblk_core_free(pblk); +fail_free_pad_dist: + kfree(pblk->pad_dist); fail_free_line_meta: pblk_line_meta_free(pblk); fail_free_luns: diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c index 7044b5599cc4..8b1434060fb3 100644 --- a/drivers/lightnvm/pblk-rb.c +++ b/drivers/lightnvm/pblk-rb.c @@ -437,9 +437,7 @@ static int pblk_rb_may_write_flush(struct pblk_rb *rb, unsigned int nr_entries, if (bio->bi_opf & REQ_PREFLUSH) { struct pblk *pblk = container_of(rb, struct pblk, rwb); -#ifdef CONFIG_NVM_DEBUG - atomic_long_inc(&pblk->nr_flush); -#endif + atomic64_inc(&pblk->nr_flush); if (pblk_rb_flush_point_set(&pblk->rwb, bio, mem)) *io_ret = NVM_IO_OK; } @@ -620,14 +618,17 @@ unsigned int pblk_rb_read_to_bio(struct pblk_rb *rb, struct nvm_rq *rqd, pr_err("pblk: could not pad page in write bio\n"); return NVM_IO_ERR; } + + if (pad < pblk->min_write_pgs) + atomic64_inc(&pblk->pad_dist[pad - 1]); + else + pr_warn("pblk: padding more than min. sectors\n"); + + atomic64_add(pad, &pblk->pad_wa); } - atomic64_add(pad, &((struct pblk *) - (container_of(rb, struct pblk, rwb)))->pad_wa); - #ifdef CONFIG_NVM_DEBUG - atomic_long_add(pad, &((struct pblk *) - (container_of(rb, struct pblk, rwb)))->padded_writes); + atomic_long_add(pad, &pblk->padded_writes); #endif return NVM_IO_OK; diff --git a/drivers/lightnvm/pblk-sysfs.c b/drivers/lightnvm/pblk-sysfs.c index beed99936c89..c2cf6c939752 100644 --- a/drivers/lightnvm/pblk-sysfs.c +++ b/drivers/lightnvm/pblk-sysfs.c @@ -340,15 +340,62 @@ static ssize_t pblk_sysfs_get_write_amp_trip(struct pblk *pblk, char *page) atomic64_read(&pblk->pad_wa) - pblk->pad_rst_wa, page); } +static long long bucket_percentage(unsigned long long bucket, + unsigned long long total) +{ + int p = bucket * 100; + + p = div_u64(p, total); + + return p; +} + +static ssize_t pblk_sysfs_get_padding_dist(struct pblk *pblk, char *page) +{ + int sz = 0; + unsigned long long total; + unsigned long long total_buckets = 0; + int buckets = pblk->min_write_pgs - 1; + int i; + + total = atomic64_read(&pblk->nr_flush) - pblk->nr_flush_rst; + if (!total) { + for (i = 0; i < (buckets + 1); i++) + sz += snprintf(page + sz, PAGE_SIZE - sz, + "%d:0 ", i); + sz += snprintf(page + sz, PAGE_SIZE - sz, "\n"); + + return sz; + } + + for (i = 0; i < buckets; i++) + total_buckets += atomic64_read(&pblk->pad_dist[i]); + + sz += snprintf(page + sz, PAGE_SIZE - sz, "0:%lld%% ", + bucket_percentage(total - total_buckets, total)); + + for (i = 0; i < buckets; i++) { + unsigned long long p; + + p = bucket_percentage(atomic64_read(&pblk->pad_dist[i]), + total); + sz += snprintf(page + sz, PAGE_SIZE - sz, "%d:%lld%% ", + i + 1, p); + } + sz += snprintf(page + sz, PAGE_SIZE - sz, "\n"); + + return sz; +} + #ifdef CONFIG_NVM_DEBUG static ssize_t pblk_sysfs_stats_debug(struct pblk *pblk, char *page) { return snprintf(page, PAGE_SIZE, - "%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\n", + "%lu\t%lu\t%ld\t%llu\t%ld\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\t%lu\n", atomic_long_read(&pblk->inflight_writes), atomic_long_read(&pblk->inflight_reads), atomic_long_read(&pblk->req_writes), - atomic_long_read(&pblk->nr_flush), + (u64)atomic64_read(&pblk->nr_flush), atomic_long_read(&pblk->padded_writes), atomic_long_read(&pblk->padded_wb), atomic_long_read(&pblk->sub_writes), @@ -426,6 +473,32 @@ static ssize_t pblk_sysfs_set_write_amp_trip(struct pblk *pblk, } +static ssize_t pblk_sysfs_set_padding_dist(struct pblk *pblk, + const char *page, size_t len) +{ + size_t c_len; + int reset_value; + int buckets = pblk->min_write_pgs - 1; + int i; + + c_len = strcspn(page, "\n"); + if (c_len >= len) + return -EINVAL; + + if (kstrtouint(page, 0, &reset_value)) + return -EINVAL; + + if (reset_value != 0) + return -EINVAL; + + for (i = 0; i < buckets; i++) + atomic64_set(&pblk->pad_dist[i], 0); + + pblk->nr_flush_rst = atomic64_read(&pblk->nr_flush); + + return len; +} + static struct attribute sys_write_luns = { .name = "write_luns", .mode = 0444, @@ -486,6 +559,11 @@ static struct attribute sys_write_amp_trip = { .mode = 0644, }; +static struct attribute sys_padding_dist = { + .name = "padding_dist", + .mode = 0644, +}; + #ifdef CONFIG_NVM_DEBUG static struct attribute sys_stats_debug_attr = { .name = "stats", @@ -506,6 +584,7 @@ static struct attribute *pblk_attrs[] = { &sys_lines_info_attr, &sys_write_amp_mileage, &sys_write_amp_trip, + &sys_padding_dist, #ifdef CONFIG_NVM_DEBUG &sys_stats_debug_attr, #endif @@ -539,6 +618,8 @@ static ssize_t pblk_sysfs_show(struct kobject *kobj, struct attribute *attr, return pblk_sysfs_get_write_amp_mileage(pblk, buf); else if (strcmp(attr->name, "write_amp_trip") == 0) return pblk_sysfs_get_write_amp_trip(pblk, buf); + else if (strcmp(attr->name, "padding_dist") == 0) + return pblk_sysfs_get_padding_dist(pblk, buf); #ifdef CONFIG_NVM_DEBUG else if (strcmp(attr->name, "stats") == 0) return pblk_sysfs_stats_debug(pblk, buf); @@ -557,6 +638,8 @@ static ssize_t pblk_sysfs_store(struct kobject *kobj, struct attribute *attr, return pblk_sysfs_set_sec_per_write(pblk, buf, len); else if (strcmp(attr->name, "write_amp_trip") == 0) return pblk_sysfs_set_write_amp_trip(pblk, buf, len); + else if (strcmp(attr->name, "padding_dist") == 0) + return pblk_sysfs_set_padding_dist(pblk, buf, len); return 0; } diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h index 4b7d8618631f..17e2f242f7da 100644 --- a/drivers/lightnvm/pblk.h +++ b/drivers/lightnvm/pblk.h @@ -626,12 +626,16 @@ struct pblk { u64 gc_rst_wa; u64 pad_rst_wa; + /* Counters used for calculating padding distribution */ + atomic64_t *pad_dist; /* Padding distribution buckets */ + u64 nr_flush_rst; /* Flushes reset value for pad dist.*/ + atomic64_t nr_flush; /* Number of flush/fua I/O */ + #ifdef CONFIG_NVM_DEBUG /* Non-persistent debug counters, 4kb sector I/Os */ atomic_long_t inflight_writes; /* Inflight writes (user and gc) */ atomic_long_t padded_writes; /* Sectors padded due to flush/fua */ atomic_long_t padded_wb; /* Sectors padded in write buffer */ - atomic_long_t nr_flush; /* Number of flush/fua I/O */ atomic_long_t req_writes; /* Sectors stored on write buffer */ atomic_long_t sub_writes; /* Sectors submitted from buffer */ atomic_long_t sync_writes; /* Sectors synced to media */