From patchwork Thu Oct 11 07:09:51 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Damien Le Moal X-Patchwork-Id: 10636191 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id D3EEB69B1 for ; Thu, 11 Oct 2018 07:10:21 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id B664D2AED2 for ; Thu, 11 Oct 2018 07:10:21 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id A80692AF3A; Thu, 11 Oct 2018 07:10:21 +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=-7.7 required=2.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,MAILING_LIST_MULTI,RCVD_IN_DNSWL_HI autolearn=ham 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 7BCB52AE79 for ; Thu, 11 Oct 2018 07:10:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726223AbeJKOgO (ORCPT ); Thu, 11 Oct 2018 10:36:14 -0400 Received: from esa2.hgst.iphmx.com ([68.232.143.124]:49663 "EHLO esa2.hgst.iphmx.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726205AbeJKOgO (ORCPT ); Thu, 11 Oct 2018 10:36:14 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=wdc.com; i=@wdc.com; q=dns/txt; s=dkim.wdc.com; t=1539241838; x=1570777838; h=from:to:cc:subject:date:message-id:in-reply-to: references; bh=c6+tmQ0OOkMKZ4pkgXOCXGrf+9QchbZ2F4xTLVxKViA=; b=Y5PuIsGZfhHfDofvtnWGaIxAqlEvNJpOMFxdJbEV1/TgNjNPZFhwr+FM E+de/dmT5VptCE7fObLObV/ds8xyYWGcKK2Ioy8aHs83lB7sRMU7ugvo3 vBVal6bdvOhNJRG/dy3NCbf8wb9A9dIOJJOUz09+j7IbCcj/ogn0zKVFn VXpP48CQgt9+T1+MlgdAdOZxmTlft16Jcqg/35tUkayYeviIXOFtGEff9 SFNrLAA0rW3mOOdoFM2d6hZ9MGwrh/L3v/hVO72KAFGrMd+jCI43mEIHE HwKgNDADwC5Kw3xLV/Q6SPbFOpnyfM4HPUZ4IS2KXmPBuHpSEum2OUAbP Q==; X-IronPort-AV: E=Sophos;i="5.54,367,1534780800"; d="scan'208";a="189389141" Received: from h199-255-45-14.hgst.com (HELO uls-op-cesaep01.wdc.com) ([199.255.45.14]) by ob1.hgst.iphmx.com with ESMTP; 11 Oct 2018 15:10:37 +0800 Received: from uls-op-cesaip02.wdc.com ([10.248.3.37]) by uls-op-cesaep01.wdc.com with ESMTP; 10 Oct 2018 23:55:42 -0700 Received: from washi.fujisawa.hgst.com ([10.149.53.254]) by uls-op-cesaip02.wdc.com with ESMTP; 11 Oct 2018 00:10:17 -0700 From: Damien Le Moal To: linux-block@vger.kernel.org, Jens Axboe , linux-scsi@vger.kernel.org, "Martin K . Petersen" , dm-devel@redhat.com, Mike Snitzer Cc: Christoph Hellwig , Matias Bjorling Subject: [PATCH v2 10/11] block: add a report_zones method Date: Thu, 11 Oct 2018 16:09:51 +0900 Message-Id: <20181011070952.13248-11-damien.lemoal@wdc.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20181011070952.13248-1-damien.lemoal@wdc.com> References: <20181011070952.13248-1-damien.lemoal@wdc.com> 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: Christoph Hellwig Dispatching a report zones command through the request queue is a major pain due to the command reply payload rewriting necessary. Given that blkdev_report_zones() is executing everything synchronously, implement report zones as a block device file operation instead, allowing major simplification of the code in many places. sd, null-blk, dm-linear and dm-flakey being the only block device drivers supporting exposing zoned block devices, these drivers are modified to provide the device side implementation of the report_zones() block device file operation. For device mappers, a new report_zones() target type operation is defined so that the upper block layer calls blkdev_report_zones() can be propagated down to the underlying devices of the dm targets. Implementation for this new operation is added to the dm-linear and dm-flakey targets. Signed-off-by: Christoph Hellwig [Damien] * Changed method block_device argument to gendisk * Various bug fixes and improvements * Added support for null_blk, dm-linear and dm-flakey. Signed-off-by: Damien Le Moal --- block/blk-core.c | 1 - block/blk-mq-debugfs.c | 1 - block/blk-zoned.c | 164 ++++++++++---------------------- drivers/block/null_blk.h | 11 ++- drivers/block/null_blk_main.c | 23 +---- drivers/block/null_blk_zoned.c | 57 +++-------- drivers/md/dm-flakey.c | 30 ++++-- drivers/md/dm-linear.c | 35 ++++--- drivers/md/dm.c | 169 ++++++++++++++++----------------- drivers/scsi/sd.c | 13 +-- drivers/scsi/sd.h | 11 +-- drivers/scsi/sd_zbc.c | 153 +++++++++-------------------- include/linux/blk_types.h | 2 - include/linux/blkdev.h | 8 +- include/linux/device-mapper.h | 12 ++- include/trace/events/f2fs.h | 1 - 16 files changed, 266 insertions(+), 425 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index cff0a60ee200..18e7050eb5a4 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -2307,7 +2307,6 @@ generic_make_request_checks(struct bio *bio) if (!q->limits.max_write_same_sectors) goto not_supported; break; - case REQ_OP_ZONE_REPORT: case REQ_OP_ZONE_RESET: if (!blk_queue_is_zoned(q)) goto not_supported; diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c index cb1e6cf7ac48..a14ed04c1ff7 100644 --- a/block/blk-mq-debugfs.c +++ b/block/blk-mq-debugfs.c @@ -275,7 +275,6 @@ static const char *const op_name[] = { REQ_OP_NAME(WRITE), REQ_OP_NAME(FLUSH), REQ_OP_NAME(DISCARD), - REQ_OP_NAME(ZONE_REPORT), REQ_OP_NAME(SECURE_ERASE), REQ_OP_NAME(ZONE_RESET), REQ_OP_NAME(WRITE_SAME), diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 5d967fd39fbd..90cf503091d5 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -93,13 +93,10 @@ unsigned int blkdev_nr_zones(struct block_device *bdev) EXPORT_SYMBOL_GPL(blkdev_nr_zones); /* - * Check that a zone report belongs to the partition. - * If yes, fix its start sector and write pointer, copy it in the - * zone information array and return true. Return false otherwise. + * Check that a zone report belongs to this partition, and if yes, fix its start + * sector and write pointer and return true. Return false otherwise. */ -static bool blkdev_report_zone(struct block_device *bdev, - struct blk_zone *rep, - struct blk_zone *zone) +static bool blkdev_report_zone(struct block_device *bdev, struct blk_zone *rep) { sector_t offset = get_start_sect(bdev); @@ -114,11 +111,36 @@ static bool blkdev_report_zone(struct block_device *bdev, rep->wp = rep->start + rep->len; else rep->wp -= offset; - memcpy(zone, rep, sizeof(struct blk_zone)); - return true; } +static int blk_report_zones(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) +{ + struct request_queue *q = disk->queue; + unsigned int z = 0, n, nrz = *nr_zones; + sector_t capacity = get_capacity(disk); + int ret; + + while (z < nrz && sector < capacity) { + n = nrz - z; + ret = disk->fops->report_zones(disk, sector, &zones[z], &n, + gfp_mask); + if (ret) + return ret; + if (!n) + break; + sector += blk_queue_zone_sectors(q) * n; + z += n; + } + + WARN_ON(z > *nr_zones); + *nr_zones = z; + + return 0; +} + /** * blkdev_report_zones - Get zones information * @bdev: Target block device @@ -133,130 +155,46 @@ static bool blkdev_report_zone(struct block_device *bdev, * requested by @nr_zones. The number of zones actually reported is * returned in @nr_zones. */ -int blkdev_report_zones(struct block_device *bdev, - sector_t sector, - struct blk_zone *zones, - unsigned int *nr_zones, +int blkdev_report_zones(struct block_device *bdev, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, gfp_t gfp_mask) { struct request_queue *q = bdev_get_queue(bdev); - struct blk_zone_report_hdr *hdr; - unsigned int nrz = *nr_zones; - struct page *page; - unsigned int nr_rep; - size_t rep_bytes; - unsigned int nr_pages; - struct bio *bio; - struct bio_vec *bv; - unsigned int i, n, nz; - unsigned int ofst; - void *addr; + unsigned int i, nrz; int ret; - if (!q) - return -ENXIO; - if (!blk_queue_is_zoned(q)) return -EOPNOTSUPP; - if (!nrz) - return 0; - - if (sector > bdev->bd_part->nr_sects) { - *nr_zones = 0; - return 0; - } - /* - * The zone report has a header. So make room for it in the - * payload. Also make sure that the report fits in a single BIO - * that will not be split down the stack. + * A block device that advertized itself as zoned must have a + * report_zones method. If it does not have one defined, the device + * driver has a bug. So warn about that. */ - rep_bytes = sizeof(struct blk_zone_report_hdr) + - sizeof(struct blk_zone) * nrz; - rep_bytes = (rep_bytes + PAGE_SIZE - 1) & PAGE_MASK; - if (rep_bytes > (queue_max_sectors(q) << 9)) - rep_bytes = queue_max_sectors(q) << 9; - - nr_pages = min_t(unsigned int, BIO_MAX_PAGES, - rep_bytes >> PAGE_SHIFT); - nr_pages = min_t(unsigned int, nr_pages, - queue_max_segments(q)); - - bio = bio_alloc(gfp_mask, nr_pages); - if (!bio) - return -ENOMEM; + if (WARN_ON_ONCE(!bdev->bd_disk->fops->report_zones)) + return -EOPNOTSUPP; - bio_set_dev(bio, bdev); - bio->bi_iter.bi_sector = blk_zone_start(q, sector); - bio_set_op_attrs(bio, REQ_OP_ZONE_REPORT, 0); - - for (i = 0; i < nr_pages; i++) { - page = alloc_page(gfp_mask); - if (!page) { - ret = -ENOMEM; - goto out; - } - if (!bio_add_page(bio, page, PAGE_SIZE, 0)) { - __free_page(page); - break; - } + if (!*nr_zones || sector >= bdev->bd_part->nr_sects) { + *nr_zones = 0; + return 0; } - if (i == 0) - ret = -ENOMEM; - else - ret = submit_bio_wait(bio); + nrz = min(*nr_zones, + __blkdev_nr_zones(q, bdev->bd_part->nr_sects - sector)); + ret = blk_report_zones(bdev->bd_disk, get_start_sect(bdev) + sector, + zones, &nrz, gfp_mask); if (ret) - goto out; - - /* - * Process the report result: skip the header and go through the - * reported zones to fixup and fixup the zone information for - * partitions. At the same time, return the zone information into - * the zone array. - */ - n = 0; - nz = 0; - nr_rep = 0; - bio_for_each_segment_all(bv, bio, i) { + return ret; - if (!bv->bv_page) + for (i = 0; i < nrz; i++) { + if (!blkdev_report_zone(bdev, zones)) break; - - addr = kmap_atomic(bv->bv_page); - - /* Get header in the first page */ - ofst = 0; - if (!nr_rep) { - hdr = addr; - nr_rep = hdr->nr_zones; - ofst = sizeof(struct blk_zone_report_hdr); - } - - /* Fixup and report zones */ - while (ofst < bv->bv_len && - n < nr_rep && nz < nrz) { - if (blkdev_report_zone(bdev, addr + ofst, &zones[nz])) - nz++; - ofst += sizeof(struct blk_zone); - n++; - } - - kunmap_atomic(addr); - - if (n >= nr_rep || nz >= nrz) - break; - + zones++; } - *nr_zones = nz; -out: - bio_for_each_segment_all(bv, bio, i) - __free_page(bv->bv_page); - bio_put(bio); + *nr_zones = i; - return ret; + return 0; } EXPORT_SYMBOL_GPL(blkdev_report_zones); diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h index 34e0030f0592..7685df43f1ef 100644 --- a/drivers/block/null_blk.h +++ b/drivers/block/null_blk.h @@ -87,7 +87,9 @@ struct nullb { #ifdef CONFIG_BLK_DEV_ZONED int null_zone_init(struct nullb_device *dev); void null_zone_exit(struct nullb_device *dev); -blk_status_t null_zone_report(struct nullb *nullb, struct bio *bio); +int null_zone_report(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask); void null_zone_write(struct nullb_cmd *cmd, sector_t sector, unsigned int nr_sectors); void null_zone_reset(struct nullb_cmd *cmd, sector_t sector); @@ -97,10 +99,11 @@ static inline int null_zone_init(struct nullb_device *dev) return -EINVAL; } static inline void null_zone_exit(struct nullb_device *dev) {} -static inline blk_status_t null_zone_report(struct nullb *nullb, - struct bio *bio) +static inline int null_zone_report(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, + unsigned int *nr_zones, gfp_t gfp_mask) { - return BLK_STS_NOTSUPP; + return -EOPNOTSUPP; } static inline void null_zone_write(struct nullb_cmd *cmd, sector_t sector, unsigned int nr_sectors) diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c index 093b614d6524..f5759ca768d1 100644 --- a/drivers/block/null_blk_main.c +++ b/drivers/block/null_blk_main.c @@ -1157,34 +1157,12 @@ static void null_restart_queue_async(struct nullb *nullb) } } -static bool cmd_report_zone(struct nullb *nullb, struct nullb_cmd *cmd) -{ - struct nullb_device *dev = cmd->nq->dev; - - if (dev->queue_mode == NULL_Q_BIO) { - if (bio_op(cmd->bio) == REQ_OP_ZONE_REPORT) { - cmd->error = null_zone_report(nullb, cmd->bio); - return true; - } - } else { - if (req_op(cmd->rq) == REQ_OP_ZONE_REPORT) { - cmd->error = null_zone_report(nullb, cmd->rq->bio); - return true; - } - } - - return false; -} - static blk_status_t null_handle_cmd(struct nullb_cmd *cmd) { struct nullb_device *dev = cmd->nq->dev; struct nullb *nullb = dev->nullb; int err = 0; - if (cmd_report_zone(nullb, cmd)) - goto out; - if (test_bit(NULLB_DEV_FL_THROTTLED, &dev->flags)) { struct request *rq = cmd->rq; @@ -1528,6 +1506,7 @@ static const struct block_device_operations null_fops = { .owner = THIS_MODULE, .open = null_open, .release = null_release, + .report_zones = null_zone_report, }; static void null_init_queue(struct nullb *nullb, struct nullb_queue *nq) diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c index 7c6b86d98700..c0b0e4a3fa8f 100644 --- a/drivers/block/null_blk_zoned.c +++ b/drivers/block/null_blk_zoned.c @@ -48,54 +48,27 @@ void null_zone_exit(struct nullb_device *dev) kvfree(dev->zones); } -static void null_zone_fill_bio(struct nullb_device *dev, struct bio *bio, - unsigned int zno, unsigned int nr_zones) +int null_zone_report(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) { - struct blk_zone_report_hdr *hdr = NULL; - struct bio_vec bvec; - struct bvec_iter iter; - void *addr; - unsigned int zones_to_cpy; - - bio_for_each_segment(bvec, bio, iter) { - addr = kmap_atomic(bvec.bv_page); - - zones_to_cpy = bvec.bv_len / sizeof(struct blk_zone); - - if (!hdr) { - hdr = (struct blk_zone_report_hdr *)addr; - hdr->nr_zones = nr_zones; - zones_to_cpy--; - addr += sizeof(struct blk_zone_report_hdr); - } - - zones_to_cpy = min_t(unsigned int, zones_to_cpy, nr_zones); - - memcpy(addr, &dev->zones[zno], - zones_to_cpy * sizeof(struct blk_zone)); - - kunmap_atomic(addr); + struct nullb *nullb = disk->private_data; + struct nullb_device *dev = nullb->dev; + unsigned int zno, nrz = 0; - nr_zones -= zones_to_cpy; - zno += zones_to_cpy; + if (!dev->zoned) + /* Not a zoned null device */ + return -EOPNOTSUPP; - if (!nr_zones) - break; + zno = null_zone_no(dev, sector); + if (zno < dev->nr_zones) { + nrz = min_t(unsigned int, *nr_zones, dev->nr_zones - zno); + memcpy(zones, &dev->zones[zno], nrz * sizeof(struct blk_zone)); } -} -blk_status_t null_zone_report(struct nullb *nullb, struct bio *bio) -{ - struct nullb_device *dev = nullb->dev; - unsigned int zno = null_zone_no(dev, bio->bi_iter.bi_sector); - unsigned int nr_zones = dev->nr_zones - zno; - unsigned int max_zones; + *nr_zones = nrz; - max_zones = (bio->bi_iter.bi_size / sizeof(struct blk_zone)) - 1; - nr_zones = min_t(unsigned int, nr_zones, max_zones); - null_zone_fill_bio(nullb->dev, bio, zno, nr_zones); - - return BLK_STS_OK; + return 0; } void null_zone_write(struct nullb_cmd *cmd, sector_t sector, diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index 32aabe27b37c..3cb97fa4c11d 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -315,10 +315,6 @@ static int flakey_map(struct dm_target *ti, struct bio *bio) if (bio_op(bio) == REQ_OP_ZONE_RESET) goto map_bio; - /* We need to remap reported zones, so remember the BIO iter */ - if (bio_op(bio) == REQ_OP_ZONE_REPORT) - goto map_bio; - /* Are we alive ? */ elapsed = (jiffies - fc->start_time) / HZ; if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval) { @@ -380,11 +376,6 @@ static int flakey_end_io(struct dm_target *ti, struct bio *bio, if (bio_op(bio) == REQ_OP_ZONE_RESET) return DM_ENDIO_DONE; - if (bio_op(bio) == REQ_OP_ZONE_REPORT) { - dm_remap_zone_report(ti, bio, fc->start); - return DM_ENDIO_DONE; - } - if (!*error && pb->bio_submitted && (bio_data_dir(bio) == READ)) { if (fc->corrupt_bio_byte && (fc->corrupt_bio_rw == READ) && all_corrupt_bio_flags_match(bio, fc)) { @@ -457,6 +448,26 @@ static int flakey_prepare_ioctl(struct dm_target *ti, struct block_device **bdev return 0; } +#ifdef CONFIG_BLK_DEV_ZONED +static int flakey_report_zones(struct dm_target *ti, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) +{ + struct flakey_c *fc = ti->private; + int ret; + + /* Do report and remap it */ + ret = blkdev_report_zones(fc->dev->bdev, flakey_map_sector(ti, sector), + zones, nr_zones, gfp_mask); + if (ret != 0) + return ret; + + if (*nr_zones) + dm_remap_zone_report(ti, fc->start, zones, nr_zones); + return 0; +} +#endif + static int flakey_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data) { struct flakey_c *fc = ti->private; @@ -469,6 +480,7 @@ static struct target_type flakey_target = { .version = {1, 5, 0}, #ifdef CONFIG_BLK_DEV_ZONED .features = DM_TARGET_ZONED_HM, + .report_zones = flakey_report_zones, #endif .module = THIS_MODULE, .ctr = flakey_ctr, diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index 2f7c44a006c4..8d7ddee6ac4d 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -102,19 +102,6 @@ static int linear_map(struct dm_target *ti, struct bio *bio) return DM_MAPIO_REMAPPED; } -#ifdef CONFIG_BLK_DEV_ZONED -static int linear_end_io(struct dm_target *ti, struct bio *bio, - blk_status_t *error) -{ - struct linear_c *lc = ti->private; - - if (!*error && bio_op(bio) == REQ_OP_ZONE_REPORT) - dm_remap_zone_report(ti, bio, lc->start); - - return DM_ENDIO_DONE; -} -#endif - static void linear_status(struct dm_target *ti, status_type_t type, unsigned status_flags, char *result, unsigned maxlen) { @@ -148,6 +135,26 @@ static int linear_prepare_ioctl(struct dm_target *ti, struct block_device **bdev return 0; } +#ifdef CONFIG_BLK_DEV_ZONED +static int linear_report_zones(struct dm_target *ti, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) +{ + struct linear_c *lc = (struct linear_c *) ti->private; + int ret; + + /* Do report and remap it */ + ret = blkdev_report_zones(lc->dev->bdev, linear_map_sector(ti, sector), + zones, nr_zones, gfp_mask); + if (ret != 0) + return ret; + + if (*nr_zones) + dm_remap_zone_report(ti, lc->start, zones, nr_zones); + return 0; +} +#endif + static int linear_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data) { @@ -211,8 +218,8 @@ static struct target_type linear_target = { .name = "linear", .version = {1, 4, 0}, #ifdef CONFIG_BLK_DEV_ZONED - .end_io = linear_end_io, .features = DM_TARGET_PASSES_INTEGRITY | DM_TARGET_ZONED_HM, + .report_zones = linear_report_zones, #else .features = DM_TARGET_PASSES_INTEGRITY, #endif diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 45abb54037fc..6be21dc210a1 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -458,6 +458,57 @@ static int dm_blk_getgeo(struct block_device *bdev, struct hd_geometry *geo) return dm_get_geometry(md, geo); } +static int dm_blk_report_zones(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) +{ +#ifdef CONFIG_BLK_DEV_ZONED + struct mapped_device *md = disk->private_data; + struct dm_target *tgt; + struct dm_table *map; + int srcu_idx, ret; + + if (dm_suspended_md(md)) + return -EAGAIN; + + map = dm_get_live_table(md, &srcu_idx); + if (!map) + return -EIO; + + tgt = dm_table_find_target(map, sector); + if (!dm_target_is_valid(tgt)) { + ret = -EIO; + goto out; + } + + /* + * If we are executing this, we already know that the block device + * is a zoned device and so each target should have support for that + * type of drive. A missing report_zones method means that the target + * driver has a problem. + */ + if (WARN_ON(!tgt->type->report_zones)) { + ret = -EIO; + goto out; + } + + /* + * blkdev_report_zones() will loop and call this again to cover all the + * zones of the target, eventually moving on to the next target. + * So there is no need to loop here trying to fill the entire array + * of zones. + */ + ret = tgt->type->report_zones(tgt, sector, zones, + nr_zones, gfp_mask); + +out: + dm_put_live_table(md, srcu_idx); + return ret; +#else + return -ENOTSUPP; +#endif +} + static int dm_prepare_ioctl(struct mapped_device *md, int *srcu_idx, struct block_device **bdev) __acquires(md->io_barrier) @@ -1155,93 +1206,49 @@ void dm_accept_partial_bio(struct bio *bio, unsigned n_sectors) EXPORT_SYMBOL_GPL(dm_accept_partial_bio); /* - * The zone descriptors obtained with a zone report indicate zone positions - * within the target backing device, regardless of that device is a partition - * and regardless of the target mapping start sector on the device or partition. - * The zone descriptors start sector and write pointer position must be adjusted - * to match their relative position within the dm device. - * A target may call dm_remap_zone_report() after completion of a - * REQ_OP_ZONE_REPORT bio to remap the zone descriptors obtained from the - * backing device. + * The zone descriptors obtained with a zone report indicate + * zone positions within the underlying device of the target. The zone + * descriptors must be remapped to match their position within the dm device. + * The caller target should obtain the zones information using + * blkdev_report_zones() to ensure that remapping for partition offset is + * already handled. */ -void dm_remap_zone_report(struct dm_target *ti, struct bio *bio, sector_t start) +void dm_remap_zone_report(struct dm_target *ti, sector_t start, + struct blk_zone *zones, unsigned int *nr_zones) { #ifdef CONFIG_BLK_DEV_ZONED - struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone); - struct bio *report_bio = tio->io->orig_bio; - struct blk_zone_report_hdr *hdr = NULL; struct blk_zone *zone; - unsigned int nr_rep = 0; - unsigned int ofst; - sector_t part_offset; - struct bio_vec bvec; - struct bvec_iter iter; - void *addr; - - if (bio->bi_status) - return; - - /* - * bio sector was incremented by the request size on completion. Taking - * into account the original request sector, the target start offset on - * the backing device and the target mapping offset (ti->begin), the - * start sector of the backing device. The partition offset is always 0 - * if the target uses a whole device. - */ - part_offset = bio->bi_iter.bi_sector + ti->begin - (start + bio_end_sector(report_bio)); + unsigned int nrz = *nr_zones; + int i; /* - * Remap the start sector of the reported zones. For sequential zones, - * also remap the write pointer position. + * Remap the start sector and write pointer position of the zones in + * the array. Since we may have obtained from the target underlying + * device more zones that the target size, also adjust the number + * of zones. */ - bio_for_each_segment(bvec, report_bio, iter) { - addr = kmap_atomic(bvec.bv_page); - - /* Remember the report header in the first page */ - if (!hdr) { - hdr = addr; - ofst = sizeof(struct blk_zone_report_hdr); - } else - ofst = 0; - - /* Set zones start sector */ - while (hdr->nr_zones && ofst < bvec.bv_len) { - zone = addr + ofst; - zone->start -= part_offset; - if (zone->start >= start + ti->len) { - hdr->nr_zones = 0; - break; - } - zone->start = zone->start + ti->begin - start; - if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL) { - if (zone->cond == BLK_ZONE_COND_FULL) - zone->wp = zone->start + zone->len; - else if (zone->cond == BLK_ZONE_COND_EMPTY) - zone->wp = zone->start; - else - zone->wp = zone->wp + ti->begin - start - part_offset; - } - ofst += sizeof(struct blk_zone); - hdr->nr_zones--; - nr_rep++; + for (i = 0; i < nrz; i++) { + zone = zones + i; + if (zone->start >= start + ti->len) { + memset(zone, 0, sizeof(struct blk_zone) * (nrz - i)); + break; } - if (addr != hdr) - kunmap_atomic(addr); + zone->start = zone->start + ti->begin - start; + if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL) + continue; - if (!hdr->nr_zones) - break; - } - - if (hdr) { - hdr->nr_zones = nr_rep; - kunmap_atomic(hdr); + if (zone->cond == BLK_ZONE_COND_FULL) + zone->wp = zone->start + zone->len; + else if (zone->cond == BLK_ZONE_COND_EMPTY) + zone->wp = zone->start; + else + zone->wp = zone->wp + ti->begin - start; } - bio_advance(report_bio, report_bio->bi_iter.bi_size); - + *nr_zones = i; #else /* !CONFIG_BLK_DEV_ZONED */ - bio->bi_status = BLK_STS_NOTSUPP; + *nr_zones = 0; #endif } EXPORT_SYMBOL_GPL(dm_remap_zone_report); @@ -1327,8 +1334,7 @@ static int clone_bio(struct dm_target_io *tio, struct bio *bio, return r; } - if (bio_op(bio) != REQ_OP_ZONE_REPORT) - bio_advance(clone, to_bytes(sector - clone->bi_iter.bi_sector)); + bio_advance(clone, to_bytes(sector - clone->bi_iter.bi_sector)); clone->bi_iter.bi_size = to_bytes(len); if (unlikely(bio_integrity(bio) != NULL)) @@ -1541,7 +1547,6 @@ static bool __process_abnormal_io(struct clone_info *ci, struct dm_target *ti, */ static int __split_and_process_non_flush(struct clone_info *ci) { - struct bio *bio = ci->bio; struct dm_target *ti; unsigned len; int r; @@ -1553,11 +1558,7 @@ static int __split_and_process_non_flush(struct clone_info *ci) if (unlikely(__process_abnormal_io(ci, ti, &r))) return r; - if (bio_op(bio) == REQ_OP_ZONE_REPORT) - len = ci->sector_count; - else - len = min_t(sector_t, max_io_len(ci->sector, ti), - ci->sector_count); + len = min_t(sector_t, max_io_len(ci->sector, ti), ci->sector_count); r = __clone_and_map_data_bio(ci, ti, ci->sector, &len); if (r < 0) @@ -1616,9 +1617,6 @@ static blk_qc_t __split_and_process_bio(struct mapped_device *md, * We take a clone of the original to store in * ci.io->orig_bio to be used by end_io_acct() and * for dec_pending to use for completion handling. - * As this path is not used for REQ_OP_ZONE_REPORT, - * the usage of io->orig_bio in dm_remap_zone_report() - * won't be affected by this reassignment. */ struct bio *b = bio_split(bio, bio_sectors(bio) - ci.sector_count, GFP_NOIO, &md->queue->bio_split); @@ -3167,6 +3165,7 @@ static const struct block_device_operations dm_blk_dops = { .release = dm_blk_close, .ioctl = dm_blk_ioctl, .getgeo = dm_blk_getgeo, + .report_zones = dm_blk_report_zones, .pr_ops = &dm_pr_ops, .owner = THIS_MODULE }; diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 4a57ffecc7e6..718aaf15d812 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1271,8 +1271,6 @@ static int sd_init_command(struct scsi_cmnd *cmd) case REQ_OP_READ: case REQ_OP_WRITE: return sd_setup_read_write_cmnd(cmd); - case REQ_OP_ZONE_REPORT: - return sd_zbc_setup_report_cmnd(cmd); case REQ_OP_ZONE_RESET: return sd_zbc_setup_reset_cmnd(cmd); default: @@ -1801,6 +1799,7 @@ static const struct block_device_operations sd_fops = { .check_events = sd_check_events, .revalidate_disk = sd_revalidate_disk, .unlock_native_capacity = sd_unlock_native_capacity, + .report_zones = sd_zbc_report_zones, .pr_ops = &sd_pr_ops, }; @@ -1952,16 +1951,6 @@ static int sd_done(struct scsi_cmnd *SCpnt) scsi_set_resid(SCpnt, blk_rq_bytes(req)); } break; - case REQ_OP_ZONE_REPORT: - if (!result) { - good_bytes = scsi_bufflen(SCpnt) - - scsi_get_resid(SCpnt); - scsi_set_resid(SCpnt, 0); - } else { - good_bytes = 0; - scsi_set_resid(SCpnt, blk_rq_bytes(req)); - } - break; default: /* * In case of bogus fw or device, we could end up having diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h index a7d4f50b67d4..f72f20fd0d8b 100644 --- a/drivers/scsi/sd.h +++ b/drivers/scsi/sd.h @@ -273,10 +273,12 @@ static inline int sd_is_zoned(struct scsi_disk *sdkp) extern int sd_zbc_read_zones(struct scsi_disk *sdkp, unsigned char *buffer); extern void sd_zbc_remove(struct scsi_disk *sdkp); extern void sd_zbc_print_zones(struct scsi_disk *sdkp); -extern int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd); extern int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd); extern void sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes, struct scsi_sense_hdr *sshdr); +extern int sd_zbc_report_zones(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask); #else /* CONFIG_BLK_DEV_ZONED */ @@ -290,11 +292,6 @@ static inline void sd_zbc_remove(struct scsi_disk *sdkp) {} static inline void sd_zbc_print_zones(struct scsi_disk *sdkp) {} -static inline int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd) -{ - return BLKPREP_INVALID; -} - static inline int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd) { return BLKPREP_INVALID; @@ -304,6 +301,8 @@ static inline void sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes, struct scsi_sense_hdr *sshdr) {} +#define sd_zbc_report_zones NULL + #endif /* CONFIG_BLK_DEV_ZONED */ #endif /* _SCSI_DISK_H */ diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c index 44b64b4a922a..1e8274f473b0 100644 --- a/drivers/scsi/sd_zbc.c +++ b/drivers/scsi/sd_zbc.c @@ -62,7 +62,7 @@ static void sd_zbc_parse_report(struct scsi_disk *sdkp, u8 *buf, } /** - * sd_zbc_report_zones - Issue a REPORT ZONES scsi command. + * sd_zbc_do_report_zones - Issue a REPORT ZONES scsi command. * @sdkp: The target disk * @buf: Buffer to use for the reply * @buflen: the buffer size @@ -75,9 +75,9 @@ static void sd_zbc_parse_report(struct scsi_disk *sdkp, u8 *buf, * zones and will only report the count of zones fitting in the command reply * buffer. */ -static int sd_zbc_report_zones(struct scsi_disk *sdkp, unsigned char *buf, - unsigned int buflen, sector_t lba, - bool partial) +static int sd_zbc_do_report_zones(struct scsi_disk *sdkp, unsigned char *buf, + unsigned int buflen, sector_t lba, + bool partial) { struct scsi_device *sdp = sdkp->device; const int timeout = sdp->request_queue->rq_timeout; @@ -118,108 +118,56 @@ static int sd_zbc_report_zones(struct scsi_disk *sdkp, unsigned char *buf, } /** - * sd_zbc_setup_report_cmnd - Prepare a REPORT ZONES scsi command - * @cmd: The command to setup + * sd_zbc_report_zones - Disk report zones operation. + * @disk: The target disk + * @sector: Start 512B sector of the report + * @zones: Array of zone descriptors + * @nr_zones: Number of descriptors in the array + * @gfp_mask: Memory allocation mask * - * Call in sd_init_command() for a REQ_OP_ZONE_REPORT request. + * Execute a report zones command on the target disk. */ -int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd) +int sd_zbc_report_zones(struct gendisk *disk, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask) { - struct request *rq = cmd->request; - struct scsi_disk *sdkp = scsi_disk(rq->rq_disk); - sector_t lba, sector = blk_rq_pos(rq); - unsigned int nr_bytes = blk_rq_bytes(rq); - int ret; - - WARN_ON(nr_bytes == 0); + struct scsi_disk *sdkp = scsi_disk(disk); + unsigned int i, buflen, nrz = *nr_zones; + unsigned char *buf; + size_t offset = 0; + int ret = 0; if (!sd_is_zoned(sdkp)) /* Not a zoned device */ - return BLKPREP_KILL; - - ret = scsi_init_io(cmd); - if (ret != BLKPREP_OK) - return ret; - - cmd->cmd_len = 16; - memset(cmd->cmnd, 0, cmd->cmd_len); - cmd->cmnd[0] = ZBC_IN; - cmd->cmnd[1] = ZI_REPORT_ZONES; - lba = sectors_to_logical(sdkp->device, sector); - put_unaligned_be64(lba, &cmd->cmnd[2]); - put_unaligned_be32(nr_bytes, &cmd->cmnd[10]); - /* Do partial report for speeding things up */ - cmd->cmnd[14] = ZBC_REPORT_ZONE_PARTIAL; - - cmd->sc_data_direction = DMA_FROM_DEVICE; - cmd->sdb.length = nr_bytes; - cmd->transfersize = sdkp->device->sector_size; - cmd->allowed = 0; + return -EOPNOTSUPP; - return BLKPREP_OK; -} - -/** - * sd_zbc_report_zones_complete - Process a REPORT ZONES scsi command reply. - * @scmd: The completed report zones command - * @good_bytes: reply size in bytes - * - * Convert all reported zone descriptors to struct blk_zone. The conversion - * is done in-place, directly in the request specified sg buffer. - */ -static void sd_zbc_report_zones_complete(struct scsi_cmnd *scmd, - unsigned int good_bytes) -{ - struct request *rq = scmd->request; - struct scsi_disk *sdkp = scsi_disk(rq->rq_disk); - struct sg_mapping_iter miter; - struct blk_zone_report_hdr hdr; - struct blk_zone zone; - unsigned int offset, bytes = 0; - unsigned long flags; - u8 *buf; - - if (good_bytes < 64) - return; - - memset(&hdr, 0, sizeof(struct blk_zone_report_hdr)); - - sg_miter_start(&miter, scsi_sglist(scmd), scsi_sg_count(scmd), - SG_MITER_TO_SG | SG_MITER_ATOMIC); + /* + * Get a reply buffer for the number of requested zones plus a header. + * For ATA, buffers must be aligned to 512B. + */ + buflen = roundup((nrz + 1) * 64, 512); + buf = kmalloc(buflen, gfp_mask); + if (!buf) + return -ENOMEM; - local_irq_save(flags); - while (sg_miter_next(&miter) && bytes < good_bytes) { + ret = sd_zbc_do_report_zones(sdkp, buf, buflen, + sectors_to_logical(sdkp->device, sector), true); + if (ret) + goto out_free_buf; - buf = miter.addr; - offset = 0; + nrz = min(nrz, get_unaligned_be32(&buf[0]) / 64); + for (i = 0; i < nrz; i++) { + offset += 64; + sd_zbc_parse_report(sdkp, buf + offset, zones); + zones++; + } - if (bytes == 0) { - /* Set the report header */ - hdr.nr_zones = min_t(unsigned int, - (good_bytes - 64) / 64, - get_unaligned_be32(&buf[0]) / 64); - memcpy(buf, &hdr, sizeof(struct blk_zone_report_hdr)); - offset += 64; - bytes += 64; - } + *nr_zones = nrz; - /* Parse zone descriptors */ - while (offset < miter.length && hdr.nr_zones) { - WARN_ON(offset > miter.length); - buf = miter.addr + offset; - sd_zbc_parse_report(sdkp, buf, &zone); - memcpy(buf, &zone, sizeof(struct blk_zone)); - offset += 64; - bytes += 64; - hdr.nr_zones--; - } - - if (!hdr.nr_zones) - break; +out_free_buf: + kfree(buf); - } - sg_miter_stop(&miter); - local_irq_restore(flags); + return ret; } /** @@ -302,13 +250,6 @@ void sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes, case REQ_OP_WRITE_ZEROES: case REQ_OP_WRITE_SAME: break; - - case REQ_OP_ZONE_REPORT: - - if (!result) - sd_zbc_report_zones_complete(cmd, good_bytes); - break; - } } @@ -390,7 +331,7 @@ static s32 sd_zbc_check_zones(struct scsi_disk *sdkp) return -ENOMEM; /* Do a report zone to get max_lba and the same field */ - ret = sd_zbc_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, 0, false); + ret = sd_zbc_do_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, 0, false); if (ret) goto out_free; @@ -447,8 +388,8 @@ static s32 sd_zbc_check_zones(struct scsi_disk *sdkp) } if (block < sdkp->capacity) { - ret = sd_zbc_report_zones(sdkp, buf, - SD_ZBC_BUF_SIZE, block, true); + ret = sd_zbc_do_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, + block, true); if (ret) goto out_free; } @@ -564,8 +505,8 @@ sd_zbc_setup_seq_zones_bitmap(struct scsi_disk *sdkp, u32 zone_shift, goto out; while (lba < sdkp->capacity) { - ret = sd_zbc_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, - lba, true); + ret = sd_zbc_do_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, lba, + true); if (ret) goto out; lba = sd_zbc_get_seq_zones(sdkp, buf, SD_ZBC_BUF_SIZE, diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index f6dfb30737d8..1dcf652ba0aa 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -284,8 +284,6 @@ enum req_opf { REQ_OP_FLUSH = 2, /* discard sectors */ REQ_OP_DISCARD = 3, - /* get zone information */ - REQ_OP_ZONE_REPORT = 4, /* securely erase sectors */ REQ_OP_SECURE_ERASE = 5, /* seset a zone write pointer */ diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 23ab53d2d4ca..ab8aec3be639 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -396,11 +396,6 @@ struct queue_limits { #ifdef CONFIG_BLK_DEV_ZONED -struct blk_zone_report_hdr { - unsigned int nr_zones; - u8 padding[60]; -}; - extern unsigned int blkdev_nr_zones(struct block_device *bdev); extern int blkdev_report_zones(struct block_device *bdev, sector_t sector, struct blk_zone *zones, @@ -1992,6 +1987,9 @@ struct block_device_operations { int (*getgeo)(struct block_device *, struct hd_geometry *); /* this callback is with swap_lock and sometimes page table lock held */ void (*swap_slot_free_notify) (struct block_device *, unsigned long); + int (*report_zones)(struct gendisk *, sector_t sector, + struct blk_zone *zones, unsigned int *nr_zones, + gfp_t gfp_mask); struct module *owner; const struct pr_ops *pr_ops; }; diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 6fb0808e87c8..a23b396a8edc 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -92,6 +92,11 @@ typedef int (*dm_message_fn) (struct dm_target *ti, unsigned argc, char **argv, typedef int (*dm_prepare_ioctl_fn) (struct dm_target *ti, struct block_device **bdev); +typedef int (*dm_report_zones_fn) (struct dm_target *ti, sector_t sector, + struct blk_zone *zones, + unsigned int *nr_zones, + gfp_t gfp_mask); + /* * These iteration functions are typically used to check (and combine) * properties of underlying devices. @@ -180,6 +185,9 @@ struct target_type { dm_status_fn status; dm_message_fn message; dm_prepare_ioctl_fn prepare_ioctl; +#ifdef CONFIG_BLK_DEV_ZONED + dm_report_zones_fn report_zones; +#endif dm_busy_fn busy; dm_iterate_devices_fn iterate_devices; dm_io_hints_fn io_hints; @@ -420,8 +428,8 @@ struct gendisk *dm_disk(struct mapped_device *md); int dm_suspended(struct dm_target *ti); int dm_noflush_suspending(struct dm_target *ti); void dm_accept_partial_bio(struct bio *bio, unsigned n_sectors); -void dm_remap_zone_report(struct dm_target *ti, struct bio *bio, - sector_t start); +void dm_remap_zone_report(struct dm_target *ti, sector_t start, + struct blk_zone *zones, unsigned int *nr_zones); union map_info *dm_get_rq_mapinfo(struct request *rq); struct queue_limits *dm_get_queue_limits(struct mapped_device *md); diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 795698925d20..3ec73f17ee2a 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -82,7 +82,6 @@ TRACE_DEFINE_ENUM(CP_TRIMMED); { REQ_OP_WRITE, "WRITE" }, \ { REQ_OP_FLUSH, "FLUSH" }, \ { REQ_OP_DISCARD, "DISCARD" }, \ - { REQ_OP_ZONE_REPORT, "ZONE_REPORT" }, \ { REQ_OP_SECURE_ERASE, "SECURE_ERASE" }, \ { REQ_OP_ZONE_RESET, "ZONE_RESET" }, \ { REQ_OP_WRITE_SAME, "WRITE_SAME" }, \