[linux-2.6-block.git] / drivers / lightnvm / pblk-read.c

// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (C) 2016 CNEX Labs
 * Initial release: Javier Gonzalez <javier@cnexlabs.com>
 *                  Matias Bjorling <matias@cnexlabs.com>
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License version
 * 2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * pblk-read.c - pblk's read path
 */

#include "pblk.h"

/*
 * There is no guarantee that the value read from cache has not been updated and
 * resides at another location in the cache. We guarantee though that if the
 * value is read from the cache, it belongs to the mapped lba. In order to
 * guarantee and order between writes and reads are ordered, a flush must be
 * issued.
 */
static int pblk_read_from_cache(struct pblk *pblk, struct bio *bio,
				sector_t lba, struct ppa_addr ppa,
				int bio_iter, bool advanced_bio)
{
#ifdef CONFIG_NVM_PBLK_DEBUG
	/* Callers must ensure that the ppa points to a cache address */
	BUG_ON(pblk_ppa_empty(ppa));
	BUG_ON(!pblk_addr_in_cache(ppa));
#endif

	return pblk_rb_copy_to_bio(&pblk->rwb, bio, lba, ppa,
						bio_iter, advanced_bio);
}

static void pblk_read_ppalist_rq(struct pblk *pblk, struct nvm_rq *rqd,
				 struct bio *bio, sector_t blba,
				 unsigned long *read_bitmap)
{
	struct pblk_sec_meta *meta_list = rqd->meta_list;
	struct ppa_addr ppas[NVM_MAX_VLBA];
	int nr_secs = rqd->nr_ppas;
	bool advanced_bio = false;
	int i, j = 0;

	pblk_lookup_l2p_seq(pblk, ppas, blba, nr_secs);

	for (i = 0; i < nr_secs; i++) {
		struct ppa_addr p = ppas[i];
		sector_t lba = blba + i;

retry:
		if (pblk_ppa_empty(p)) {
			WARN_ON(test_and_set_bit(i, read_bitmap));
			meta_list[i].lba = cpu_to_le64(ADDR_EMPTY);

			if (unlikely(!advanced_bio)) {
				bio_advance(bio, (i) * PBLK_EXPOSED_PAGE_SIZE);
				advanced_bio = true;
			}

			goto next;
		}

		/* Try to read from write buffer. The address is later checked
		 * on the write buffer to prevent retrieving overwritten data.
		 */
		if (pblk_addr_in_cache(p)) {
			if (!pblk_read_from_cache(pblk, bio, lba, p, i,
								advanced_bio)) {
				pblk_lookup_l2p_seq(pblk, &p, lba, 1);
				goto retry;
			}
			WARN_ON(test_and_set_bit(i, read_bitmap));
			meta_list[i].lba = cpu_to_le64(lba);
			advanced_bio = true;
#ifdef CONFIG_NVM_PBLK_DEBUG
			atomic_long_inc(&pblk->cache_reads);
#endif
		} else {
			/* Read from media non-cached sectors */
			rqd->ppa_list[j++] = p;
		}

next:
		if (advanced_bio)
			bio_advance(bio, PBLK_EXPOSED_PAGE_SIZE);
	}

	if (pblk_io_aligned(pblk, nr_secs))
		rqd->is_seq = 1;

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_add(nr_secs, &pblk->inflight_reads);
#endif
}


static void pblk_read_check_seq(struct pblk *pblk, struct nvm_rq *rqd,
				sector_t blba)
{
	struct pblk_sec_meta *meta_lba_list = rqd->meta_list;
	int nr_lbas = rqd->nr_ppas;
	int i;

	for (i = 0; i < nr_lbas; i++) {
		u64 lba = le64_to_cpu(meta_lba_list[i].lba);

		if (lba == ADDR_EMPTY)
			continue;

		if (lba != blba + i) {
#ifdef CONFIG_NVM_PBLK_DEBUG
			struct ppa_addr *ppa_list = nvm_rq_to_ppa_list(rqd);

			print_ppa(pblk, &ppa_list[i], "seq", i);
#endif
			pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n",
							lba, (u64)blba + i);
			WARN_ON(1);
		}
	}
}

/*
 * There can be holes in the lba list.
 */
static void pblk_read_check_rand(struct pblk *pblk, struct nvm_rq *rqd,
				 u64 *lba_list, int nr_lbas)
{
	struct pblk_sec_meta *meta_lba_list = rqd->meta_list;
	int i, j;

	for (i = 0, j = 0; i < nr_lbas; i++) {
		u64 lba = lba_list[i];
		u64 meta_lba;

		if (lba == ADDR_EMPTY)
			continue;

		meta_lba = le64_to_cpu(meta_lba_list[j].lba);

		if (lba != meta_lba) {
#ifdef CONFIG_NVM_PBLK_DEBUG
			struct ppa_addr *ppa_list = nvm_rq_to_ppa_list(rqd);

			print_ppa(pblk, &ppa_list[j], "rnd", j);
#endif
			pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n",
							meta_lba, lba);
			WARN_ON(1);
		}

		j++;
	}

	WARN_ONCE(j != rqd->nr_ppas, "pblk: corrupted random request\n");
}

static void pblk_end_user_read(struct bio *bio)
{
#ifdef CONFIG_NVM_PBLK_DEBUG
	WARN_ONCE(bio->bi_status, "pblk: corrupted read bio\n");
#endif
	bio_endio(bio);
}

static void __pblk_end_io_read(struct pblk *pblk, struct nvm_rq *rqd,
			       bool put_line)
{
	struct nvm_tgt_dev *dev = pblk->dev;
	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	struct bio *int_bio = rqd->bio;
	unsigned long start_time = r_ctx->start_time;

	generic_end_io_acct(dev->q, REQ_OP_READ, &pblk->disk->part0, start_time);

	if (rqd->error)
		pblk_log_read_err(pblk, rqd);

	pblk_read_check_seq(pblk, rqd, r_ctx->lba);

	if (int_bio)
		bio_put(int_bio);

	if (put_line)
		pblk_rq_to_line_put(pblk, rqd);

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_add(rqd->nr_ppas, &pblk->sync_reads);
	atomic_long_sub(rqd->nr_ppas, &pblk->inflight_reads);
#endif

	pblk_free_rqd(pblk, rqd, PBLK_READ);
	atomic_dec(&pblk->inflight_io);
}

static void pblk_end_io_read(struct nvm_rq *rqd)
{
	struct pblk *pblk = rqd->private;
	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	struct bio *bio = (struct bio *)r_ctx->private;

	pblk_end_user_read(bio);
	__pblk_end_io_read(pblk, rqd, true);
}

static void pblk_end_partial_read(struct nvm_rq *rqd)
{
	struct pblk *pblk = rqd->private;
	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	struct pblk_pr_ctx *pr_ctx = r_ctx->private;
	struct bio *new_bio = rqd->bio;
	struct bio *bio = pr_ctx->orig_bio;
	struct bio_vec src_bv, dst_bv;
	struct pblk_sec_meta *meta_list = rqd->meta_list;
	int bio_init_idx = pr_ctx->bio_init_idx;
	unsigned long *read_bitmap = pr_ctx->bitmap;
	int nr_secs = pr_ctx->orig_nr_secs;
	int nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs);
	__le64 *lba_list_mem, *lba_list_media;
	void *src_p, *dst_p;
	int hole, i;

	if (unlikely(nr_holes == 1)) {
		struct ppa_addr ppa;

		ppa = rqd->ppa_addr;
		rqd->ppa_list = pr_ctx->ppa_ptr;
		rqd->dma_ppa_list = pr_ctx->dma_ppa_list;
		rqd->ppa_list[0] = ppa;
	}

	/* Re-use allocated memory for intermediate lbas */
	lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size);
	lba_list_media = (((void *)rqd->ppa_list) + 2 * pblk_dma_ppa_size);

	for (i = 0; i < nr_secs; i++) {
		lba_list_media[i] = meta_list[i].lba;
		meta_list[i].lba = lba_list_mem[i];
	}

	/* Fill the holes in the original bio */
	i = 0;
	hole = find_first_zero_bit(read_bitmap, nr_secs);
	do {
		struct pblk_line *line;

		line = pblk_ppa_to_line(pblk, rqd->ppa_list[i]);
		kref_put(&line->ref, pblk_line_put);

		meta_list[hole].lba = lba_list_media[i];

		src_bv = new_bio->bi_io_vec[i++];
		dst_bv = bio->bi_io_vec[bio_init_idx + hole];

		src_p = kmap_atomic(src_bv.bv_page);
		dst_p = kmap_atomic(dst_bv.bv_page);

		memcpy(dst_p + dst_bv.bv_offset,
			src_p + src_bv.bv_offset,
			PBLK_EXPOSED_PAGE_SIZE);

		kunmap_atomic(src_p);
		kunmap_atomic(dst_p);

		mempool_free(src_bv.bv_page, &pblk->page_bio_pool);

		hole = find_next_zero_bit(read_bitmap, nr_secs, hole + 1);
	} while (hole < nr_secs);

	bio_put(new_bio);
	kfree(pr_ctx);

	/* restore original request */
	rqd->bio = NULL;
	rqd->nr_ppas = nr_secs;

	bio_endio(bio);
	__pblk_end_io_read(pblk, rqd, false);
}

static int pblk_setup_partial_read(struct pblk *pblk, struct nvm_rq *rqd,
			    unsigned int bio_init_idx,
			    unsigned long *read_bitmap,
			    int nr_holes)
{
	struct pblk_sec_meta *meta_list = rqd->meta_list;
	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	struct pblk_pr_ctx *pr_ctx;
	struct bio *new_bio, *bio = r_ctx->private;
	__le64 *lba_list_mem;
	int nr_secs = rqd->nr_ppas;
	int i;

	/* Re-use allocated memory for intermediate lbas */
	lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size);

	new_bio = bio_alloc(GFP_KERNEL, nr_holes);

	if (pblk_bio_add_pages(pblk, new_bio, GFP_KERNEL, nr_holes))
		goto fail_bio_put;

	if (nr_holes != new_bio->bi_vcnt) {
		WARN_ONCE(1, "pblk: malformed bio\n");
		goto fail_free_pages;
	}

	pr_ctx = kmalloc(sizeof(struct pblk_pr_ctx), GFP_KERNEL);
	if (!pr_ctx)
		goto fail_free_pages;

	for (i = 0; i < nr_secs; i++)
		lba_list_mem[i] = meta_list[i].lba;

	new_bio->bi_iter.bi_sector = 0; /* internal bio */
	bio_set_op_attrs(new_bio, REQ_OP_READ, 0);

	rqd->bio = new_bio;
	rqd->nr_ppas = nr_holes;

	pr_ctx->ppa_ptr = NULL;
	pr_ctx->orig_bio = bio;
	bitmap_copy(pr_ctx->bitmap, read_bitmap, NVM_MAX_VLBA);
	pr_ctx->bio_init_idx = bio_init_idx;
	pr_ctx->orig_nr_secs = nr_secs;
	r_ctx->private = pr_ctx;

	if (unlikely(nr_holes == 1)) {
		pr_ctx->ppa_ptr = rqd->ppa_list;
		pr_ctx->dma_ppa_list = rqd->dma_ppa_list;
		rqd->ppa_addr = rqd->ppa_list[0];
	}
	return 0;

fail_free_pages:
	pblk_bio_free_pages(pblk, new_bio, 0, new_bio->bi_vcnt);
fail_bio_put:
	bio_put(new_bio);

	return -ENOMEM;
}

static int pblk_partial_read_bio(struct pblk *pblk, struct nvm_rq *rqd,
				 unsigned int bio_init_idx,
				 unsigned long *read_bitmap, int nr_secs)
{
	int nr_holes;
	int ret;

	nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs);

	if (pblk_setup_partial_read(pblk, rqd, bio_init_idx, read_bitmap,
				    nr_holes))
		return NVM_IO_ERR;

	rqd->end_io = pblk_end_partial_read;

	ret = pblk_submit_io(pblk, rqd);
	if (ret) {
		bio_put(rqd->bio);
		pblk_err(pblk, "partial read IO submission failed\n");
		goto err;
	}

	return NVM_IO_OK;

err:
	pblk_err(pblk, "failed to perform partial read\n");

	/* Free allocated pages in new bio */
	pblk_bio_free_pages(pblk, rqd->bio, 0, rqd->bio->bi_vcnt);
	__pblk_end_io_read(pblk, rqd, false);
	return NVM_IO_ERR;
}

static void pblk_read_rq(struct pblk *pblk, struct nvm_rq *rqd, struct bio *bio,
			 sector_t lba, unsigned long *read_bitmap)
{
	struct pblk_sec_meta *meta_list = rqd->meta_list;
	struct ppa_addr ppa;

	pblk_lookup_l2p_seq(pblk, &ppa, lba, 1);

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_inc(&pblk->inflight_reads);
#endif

retry:
	if (pblk_ppa_empty(ppa)) {
		WARN_ON(test_and_set_bit(0, read_bitmap));
		meta_list[0].lba = cpu_to_le64(ADDR_EMPTY);
		return;
	}

	/* Try to read from write buffer. The address is later checked on the
	 * write buffer to prevent retrieving overwritten data.
	 */
	if (pblk_addr_in_cache(ppa)) {
		if (!pblk_read_from_cache(pblk, bio, lba, ppa, 0, 1)) {
			pblk_lookup_l2p_seq(pblk, &ppa, lba, 1);
			goto retry;
		}

		WARN_ON(test_and_set_bit(0, read_bitmap));
		meta_list[0].lba = cpu_to_le64(lba);

#ifdef CONFIG_NVM_PBLK_DEBUG
		atomic_long_inc(&pblk->cache_reads);
#endif
	} else {
		rqd->ppa_addr = ppa;
	}
}

int pblk_submit_read(struct pblk *pblk, struct bio *bio)
{
	struct nvm_tgt_dev *dev = pblk->dev;
	struct request_queue *q = dev->q;
	sector_t blba = pblk_get_lba(bio);
	unsigned int nr_secs = pblk_get_secs(bio);
	struct pblk_g_ctx *r_ctx;
	struct nvm_rq *rqd;
	unsigned int bio_init_idx;
	DECLARE_BITMAP(read_bitmap, NVM_MAX_VLBA);
	int ret = NVM_IO_ERR;

	generic_start_io_acct(q, REQ_OP_READ, bio_sectors(bio),
			      &pblk->disk->part0);

	bitmap_zero(read_bitmap, nr_secs);

	rqd = pblk_alloc_rqd(pblk, PBLK_READ);

	rqd->opcode = NVM_OP_PREAD;
	rqd->nr_ppas = nr_secs;
	rqd->bio = NULL; /* cloned bio if needed */
	rqd->private = pblk;
	rqd->end_io = pblk_end_io_read;

	r_ctx = nvm_rq_to_pdu(rqd);
	r_ctx->start_time = jiffies;
	r_ctx->lba = blba;
	r_ctx->private = bio; /* original bio */

	/* Save the index for this bio's start. This is needed in case
	 * we need to fill a partial read.
	 */
	bio_init_idx = pblk_get_bi_idx(bio);

	if (pblk_alloc_rqd_meta(pblk, rqd))
		goto fail_rqd_free;

	if (nr_secs > 1)
		pblk_read_ppalist_rq(pblk, rqd, bio, blba, read_bitmap);
	else
		pblk_read_rq(pblk, rqd, bio, blba, read_bitmap);

	if (bitmap_full(read_bitmap, nr_secs)) {
		atomic_inc(&pblk->inflight_io);
		__pblk_end_io_read(pblk, rqd, false);
		return NVM_IO_DONE;
	}

	/* All sectors are to be read from the device */
	if (bitmap_empty(read_bitmap, rqd->nr_ppas)) {
		struct bio *int_bio = NULL;

		/* Clone read bio to deal with read errors internally */
		int_bio = bio_clone_fast(bio, GFP_KERNEL, &pblk_bio_set);
		if (!int_bio) {
			pblk_err(pblk, "could not clone read bio\n");
			goto fail_end_io;
		}

		rqd->bio = int_bio;

		if (pblk_submit_io(pblk, rqd)) {
			pblk_err(pblk, "read IO submission failed\n");
			ret = NVM_IO_ERR;
			goto fail_end_io;
		}

		return NVM_IO_OK;
	}

	/* The read bio request could be partially filled by the write buffer,
	 * but there are some holes that need to be read from the drive.
	 */
	ret = pblk_partial_read_bio(pblk, rqd, bio_init_idx, read_bitmap,
				    nr_secs);
	if (ret)
		goto fail_meta_free;

	return NVM_IO_OK;

fail_meta_free:
	nvm_dev_dma_free(dev->parent, rqd->meta_list, rqd->dma_meta_list);
fail_rqd_free:
	pblk_free_rqd(pblk, rqd, PBLK_READ);
	return ret;
fail_end_io:
	__pblk_end_io_read(pblk, rqd, false);
	return ret;
}

static int read_ppalist_rq_gc(struct pblk *pblk, struct nvm_rq *rqd,
			      struct pblk_line *line, u64 *lba_list,
			      u64 *paddr_list_gc, unsigned int nr_secs)
{
	struct ppa_addr ppa_list_l2p[NVM_MAX_VLBA];
	struct ppa_addr ppa_gc;
	int valid_secs = 0;
	int i;

	pblk_lookup_l2p_rand(pblk, ppa_list_l2p, lba_list, nr_secs);

	for (i = 0; i < nr_secs; i++) {
		if (lba_list[i] == ADDR_EMPTY)
			continue;

		ppa_gc = addr_to_gen_ppa(pblk, paddr_list_gc[i], line->id);
		if (!pblk_ppa_comp(ppa_list_l2p[i], ppa_gc)) {
			paddr_list_gc[i] = lba_list[i] = ADDR_EMPTY;
			continue;
		}

		rqd->ppa_list[valid_secs++] = ppa_list_l2p[i];
	}

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_add(valid_secs, &pblk->inflight_reads);
#endif

	return valid_secs;
}

static int read_rq_gc(struct pblk *pblk, struct nvm_rq *rqd,
		      struct pblk_line *line, sector_t lba,
		      u64 paddr_gc)
{
	struct ppa_addr ppa_l2p, ppa_gc;
	int valid_secs = 0;

	if (lba == ADDR_EMPTY)
		goto out;

	/* logic error: lba out-of-bounds */
	if (lba >= pblk->rl.nr_secs) {
		WARN(1, "pblk: read lba out of bounds\n");
		goto out;
	}

	spin_lock(&pblk->trans_lock);
	ppa_l2p = pblk_trans_map_get(pblk, lba);
	spin_unlock(&pblk->trans_lock);

	ppa_gc = addr_to_gen_ppa(pblk, paddr_gc, line->id);
	if (!pblk_ppa_comp(ppa_l2p, ppa_gc))
		goto out;

	rqd->ppa_addr = ppa_l2p;
	valid_secs = 1;

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_inc(&pblk->inflight_reads);
#endif

out:
	return valid_secs;
}

int pblk_submit_read_gc(struct pblk *pblk, struct pblk_gc_rq *gc_rq)
{
	struct nvm_tgt_dev *dev = pblk->dev;
	struct nvm_geo *geo = &dev->geo;
	struct bio *bio;
	struct nvm_rq rqd;
	int data_len;
	int ret = NVM_IO_OK;

	memset(&rqd, 0, sizeof(struct nvm_rq));

	ret = pblk_alloc_rqd_meta(pblk, &rqd);
	if (ret)
		return ret;

	if (gc_rq->nr_secs > 1) {
		gc_rq->secs_to_gc = read_ppalist_rq_gc(pblk, &rqd, gc_rq->line,
							gc_rq->lba_list,
							gc_rq->paddr_list,
							gc_rq->nr_secs);
		if (gc_rq->secs_to_gc == 1)
			rqd.ppa_addr = rqd.ppa_list[0];
	} else {
		gc_rq->secs_to_gc = read_rq_gc(pblk, &rqd, gc_rq->line,
							gc_rq->lba_list[0],
							gc_rq->paddr_list[0]);
	}

	if (!(gc_rq->secs_to_gc))
		goto out;

	data_len = (gc_rq->secs_to_gc) * geo->csecs;
	bio = pblk_bio_map_addr(pblk, gc_rq->data, gc_rq->secs_to_gc, data_len,
						PBLK_VMALLOC_META, GFP_KERNEL);
	if (IS_ERR(bio)) {
		pblk_err(pblk, "could not allocate GC bio (%lu)\n",
								PTR_ERR(bio));
		ret = PTR_ERR(bio);
		goto err_free_dma;
	}

	bio->bi_iter.bi_sector = 0; /* internal bio */
	bio_set_op_attrs(bio, REQ_OP_READ, 0);

	rqd.opcode = NVM_OP_PREAD;
	rqd.nr_ppas = gc_rq->secs_to_gc;
	rqd.bio = bio;

	if (pblk_submit_io_sync(pblk, &rqd)) {
		ret = -EIO;
		pblk_err(pblk, "GC read request failed\n");
		goto err_free_bio;
	}

	pblk_read_check_rand(pblk, &rqd, gc_rq->lba_list, gc_rq->nr_secs);

	atomic_dec(&pblk->inflight_io);

	if (rqd.error) {
		atomic_long_inc(&pblk->read_failed_gc);
#ifdef CONFIG_NVM_PBLK_DEBUG
		pblk_print_failed_rqd(pblk, &rqd, rqd.error);
#endif
	}

#ifdef CONFIG_NVM_PBLK_DEBUG
	atomic_long_add(gc_rq->secs_to_gc, &pblk->sync_reads);
	atomic_long_add(gc_rq->secs_to_gc, &pblk->recov_gc_reads);
	atomic_long_sub(gc_rq->secs_to_gc, &pblk->inflight_reads);
#endif

out:
	pblk_free_rqd_meta(pblk, &rqd);
	return ret;

err_free_bio:
	bio_put(bio);
err_free_dma:
	pblk_free_rqd_meta(pblk, &rqd);
	return ret;
}
Commit	Line	Data
02a1520d	1	// SPDX-License-Identifier: GPL-2.0
a4bd217b JG	2	/*
	3	* Copyright (C) 2016 CNEX Labs
	4	* Initial release: Javier Gonzalez <javier@cnexlabs.com>
	5	* Matias Bjorling <matias@cnexlabs.com>
	6	*
	7	* This program is free software; you can redistribute it and/or
	8	* modify it under the terms of the GNU General Public License version
	9	* 2 as published by the Free Software Foundation.
	10	*
	11	* This program is distributed in the hope that it will be useful, but
	12	* WITHOUT ANY WARRANTY; without even the implied warranty of
	13	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	14	* General Public License for more details.
	15	*
	16	* pblk-read.c - pblk's read path
	17	*/
	18
	19	#include "pblk.h"
	20
	21	/*
	22	* There is no guarantee that the value read from cache has not been updated and
	23	* resides at another location in the cache. We guarantee though that if the
	24	* value is read from the cache, it belongs to the mapped lba. In order to
	25	* guarantee and order between writes and reads are ordered, a flush must be
	26	* issued.
	27	*/
	28	static int pblk_read_from_cache(struct pblk pblk, struct bio bio,
	29	sector_t lba, struct ppa_addr ppa,
75cb8e93	30	int bio_iter, bool advanced_bio)
a4bd217b	31	{
880eda54	32	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	33	/* Callers must ensure that the ppa points to a cache address */
	34	BUG_ON(pblk_ppa_empty(ppa));
	35	BUG_ON(!pblk_addr_in_cache(ppa));
	36	#endif
	37
75cb8e93 JG	38	return pblk_rb_copy_to_bio(&pblk->rwb, bio, lba, ppa,
75cb8e93 JG	39	bio_iter, advanced_bio);
a4bd217b JG	40	}
	41
	42	static void pblk_read_ppalist_rq(struct pblk pblk, struct nvm_rq rqd,
87cc40bb JG	43	struct bio *bio, sector_t blba,
87cc40bb JG	44	unsigned long *read_bitmap)
a4bd217b	45	{
a4809fee	46	struct pblk_sec_meta *meta_list = rqd->meta_list;
afdc23c9	47	struct ppa_addr ppas[NVM_MAX_VLBA];
a4bd217b	48	int nr_secs = rqd->nr_ppas;
75cb8e93	49	bool advanced_bio = false;
a4bd217b JG	50	int i, j = 0;
a4bd217b JG	51
a4bd217b JG	52	pblk_lookup_l2p_seq(pblk, ppas, blba, nr_secs);
	53
	54	for (i = 0; i < nr_secs; i++) {
	55	struct ppa_addr p = ppas[i];
	56	sector_t lba = blba + i;
	57
	58	retry:
	59	if (pblk_ppa_empty(p)) {
	60	WARN_ON(test_and_set_bit(i, read_bitmap));
a4809fee	61	meta_list[i].lba = cpu_to_le64(ADDR_EMPTY);
75cb8e93 JG	62
	63	if (unlikely(!advanced_bio)) {
	64	bio_advance(bio, (i) * PBLK_EXPOSED_PAGE_SIZE);
	65	advanced_bio = true;
	66	}
	67
	68	goto next;
a4bd217b JG	69	}
	70
	71	/* Try to read from write buffer. The address is later checked
	72	* on the write buffer to prevent retrieving overwritten data.
	73	*/
	74	if (pblk_addr_in_cache(p)) {
75cb8e93 JG	75	if (!pblk_read_from_cache(pblk, bio, lba, p, i,
75cb8e93 JG	76	advanced_bio)) {
a4bd217b JG	77	pblk_lookup_l2p_seq(pblk, &p, lba, 1);
	78	goto retry;
	79	}
	80	WARN_ON(test_and_set_bit(i, read_bitmap));
a4809fee	81	meta_list[i].lba = cpu_to_le64(lba);
75cb8e93	82	advanced_bio = true;
880eda54	83	#ifdef CONFIG_NVM_PBLK_DEBUG
db7ada33 JG	84	atomic_long_inc(&pblk->cache_reads);
db7ada33 JG	85	#endif
a4bd217b JG	86	} else {
	87	/* Read from media non-cached sectors */
	88	rqd->ppa_list[j++] = p;
	89	}
	90
75cb8e93	91	next:
a4bd217b JG	92	if (advanced_bio)
	93	bio_advance(bio, PBLK_EXPOSED_PAGE_SIZE);
	94	}
	95
f9c10152	96	if (pblk_io_aligned(pblk, nr_secs))
d7b68016	97	rqd->is_seq = 1;
f9c10152	98
880eda54	99	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	100	atomic_long_add(nr_secs, &pblk->inflight_reads);
	101	#endif
	102	}
	103
a4bd217b	104
03a34b2d JG	105	static void pblk_read_check_seq(struct pblk pblk, struct nvm_rq rqd,
03a34b2d JG	106	sector_t blba)
a4809fee	107	{
03a34b2d JG	108	struct pblk_sec_meta *meta_lba_list = rqd->meta_list;
03a34b2d JG	109	int nr_lbas = rqd->nr_ppas;
a4809fee JG	110	int i;
	111
	112	for (i = 0; i < nr_lbas; i++) {
310df582	113	u64 lba = le64_to_cpu(meta_lba_list[i].lba);
a4809fee JG	114
	115	if (lba == ADDR_EMPTY)
	116	continue;
	117
03a34b2d	118	if (lba != blba + i) {
880eda54	119	#ifdef CONFIG_NVM_PBLK_DEBUG
d68a9344	120	struct ppa_addr *ppa_list = nvm_rq_to_ppa_list(rqd);
03a34b2d	121
d68a9344	122	print_ppa(pblk, &ppa_list[i], "seq", i);
03a34b2d	123	#endif
4e495a46	124	pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n",
03a34b2d JG	125	lba, (u64)blba + i);
	126	WARN_ON(1);
	127	}
a4809fee JG	128	}
	129	}
	130
310df582 JG	131	/*
	132	* There can be holes in the lba list.
	133	*/
03a34b2d JG	134	static void pblk_read_check_rand(struct pblk pblk, struct nvm_rq rqd,
03a34b2d JG	135	u64 *lba_list, int nr_lbas)
310df582	136	{
03a34b2d	137	struct pblk_sec_meta *meta_lba_list = rqd->meta_list;
310df582 JG	138	int i, j;
	139
	140	for (i = 0, j = 0; i < nr_lbas; i++) {
	141	u64 lba = lba_list[i];
	142	u64 meta_lba;
	143
	144	if (lba == ADDR_EMPTY)
	145	continue;
	146
03a34b2d	147	meta_lba = le64_to_cpu(meta_lba_list[j].lba);
310df582 JG	148
310df582 JG	149	if (lba != meta_lba) {
880eda54	150	#ifdef CONFIG_NVM_PBLK_DEBUG
d68a9344	151	struct ppa_addr *ppa_list = nvm_rq_to_ppa_list(rqd);
03a34b2d	152
ea1d24bc	153	print_ppa(pblk, &ppa_list[j], "rnd", j);
03a34b2d	154	#endif
4e495a46	155	pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n",
ea1d24bc	156	meta_lba, lba);
310df582 JG	157	WARN_ON(1);
310df582 JG	158	}
03a34b2d JG	159
03a34b2d JG	160	j++;
310df582	161	}
03a34b2d JG	162
03a34b2d JG	163	WARN_ONCE(j != rqd->nr_ppas, "pblk: corrupted random request\n");
310df582 JG	164	}
310df582 JG	165
7bd4d370 JG	166	static void pblk_end_user_read(struct bio *bio)
7bd4d370 JG	167	{
880eda54	168	#ifdef CONFIG_NVM_PBLK_DEBUG
7bd4d370 JG	169	WARN_ONCE(bio->bi_status, "pblk: corrupted read bio\n");
	170	#endif
	171	bio_endio(bio);
7bd4d370 JG	172	}
	173
	174	static void __pblk_end_io_read(struct pblk pblk, struct nvm_rq rqd,
	175	bool put_line)
a4bd217b	176	{
998ba629	177	struct nvm_tgt_dev *dev = pblk->dev;
084ec9ba	178	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
87cc40bb	179	struct bio *int_bio = rqd->bio;
998ba629 JG	180	unsigned long start_time = r_ctx->start_time;
998ba629 JG	181
ddcf35d3	182	generic_end_io_acct(dev->q, REQ_OP_READ, &pblk->disk->part0, start_time);
a4bd217b JG	183
	184	if (rqd->error)
	185	pblk_log_read_err(pblk, rqd);
a4bd217b	186
03a34b2d	187	pblk_read_check_seq(pblk, rqd, r_ctx->lba);
a4809fee	188
87cc40bb JG	189	if (int_bio)
87cc40bb JG	190	bio_put(int_bio);
084ec9ba	191
7bd4d370	192	if (put_line)
ae14cc04	193	pblk_rq_to_line_put(pblk, rqd);
a4bd217b	194
880eda54	195	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	196	atomic_long_add(rqd->nr_ppas, &pblk->sync_reads);
	197	atomic_long_sub(rqd->nr_ppas, &pblk->inflight_reads);
	198	#endif
	199
e2cddf20	200	pblk_free_rqd(pblk, rqd, PBLK_READ);
588726d3	201	atomic_dec(&pblk->inflight_io);
a4bd217b JG	202	}
a4bd217b JG	203
7bd4d370 JG	204	static void pblk_end_io_read(struct nvm_rq *rqd)
	205	{
	206	struct pblk *pblk = rqd->private;
87cc40bb JG	207	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
87cc40bb JG	208	struct bio bio = (struct bio )r_ctx->private;
7bd4d370	209
87cc40bb	210	pblk_end_user_read(bio);
7bd4d370 JG	211	__pblk_end_io_read(pblk, rqd, true);
	212	}
	213
11f6ad69	214	static void pblk_end_partial_read(struct nvm_rq *rqd)
a4bd217b	215	{
11f6ad69 HL	216	struct pblk *pblk = rqd->private;
	217	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	218	struct pblk_pr_ctx *pr_ctx = r_ctx->private;
	219	struct bio *new_bio = rqd->bio;
	220	struct bio *bio = pr_ctx->orig_bio;
a4bd217b	221	struct bio_vec src_bv, dst_bv;
11f6ad69 HL	222	struct pblk_sec_meta *meta_list = rqd->meta_list;
	223	int bio_init_idx = pr_ctx->bio_init_idx;
	224	unsigned long *read_bitmap = pr_ctx->bitmap;
	225	int nr_secs = pr_ctx->orig_nr_secs;
a4bd217b	226	int nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs);
11f6ad69 HL	227	__le64 lba_list_mem, lba_list_media;
	228	void src_p, dst_p;
	229	int hole, i;
a4bd217b	230
0f9248cf	231	if (unlikely(nr_holes == 1)) {
7bd4d370 JG	232	struct ppa_addr ppa;
	233
	234	ppa = rqd->ppa_addr;
11f6ad69 HL	235	rqd->ppa_list = pr_ctx->ppa_ptr;
11f6ad69 HL	236	rqd->dma_ppa_list = pr_ctx->dma_ppa_list;
7bd4d370	237	rqd->ppa_list[0] = ppa;
a4bd217b JG	238	}
a4bd217b JG	239
11f6ad69 HL	240	/* Re-use allocated memory for intermediate lbas */
	241	lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size);
	242	lba_list_media = (((void )rqd->ppa_list) + 2 pblk_dma_ppa_size);
	243
a4809fee JG	244	for (i = 0; i < nr_secs; i++) {
	245	lba_list_media[i] = meta_list[i].lba;
	246	meta_list[i].lba = lba_list_mem[i];
	247	}
	248
a4bd217b JG	249	/* Fill the holes in the original bio */
	250	i = 0;
	251	hole = find_first_zero_bit(read_bitmap, nr_secs);
	252	do {
cb21665c	253	struct pblk_line *line;
7bd4d370	254
cb21665c	255	line = pblk_ppa_to_line(pblk, rqd->ppa_list[i]);
7bd4d370 JG	256	kref_put(&line->ref, pblk_line_put);
7bd4d370 JG	257
a4809fee JG	258	meta_list[hole].lba = lba_list_media[i];
a4809fee JG	259
a4bd217b	260	src_bv = new_bio->bi_io_vec[i++];
11f6ad69	261	dst_bv = bio->bi_io_vec[bio_init_idx + hole];
a4bd217b JG	262
	263	src_p = kmap_atomic(src_bv.bv_page);
	264	dst_p = kmap_atomic(dst_bv.bv_page);
	265
	266	memcpy(dst_p + dst_bv.bv_offset,
	267	src_p + src_bv.bv_offset,
	268	PBLK_EXPOSED_PAGE_SIZE);
	269
	270	kunmap_atomic(src_p);
	271	kunmap_atomic(dst_p);
	272
b906bbb6	273	mempool_free(src_bv.bv_page, &pblk->page_bio_pool);
a4bd217b JG	274
	275	hole = find_next_zero_bit(read_bitmap, nr_secs, hole + 1);
	276	} while (hole < nr_secs);
	277
	278	bio_put(new_bio);
11f6ad69	279	kfree(pr_ctx);
a4bd217b	280
87cc40bb JG	281	/* restore original request */
87cc40bb JG	282	rqd->bio = NULL;
a4bd217b	283	rqd->nr_ppas = nr_secs;
a4bd217b	284
11f6ad69	285	bio_endio(bio);
7bd4d370	286	__pblk_end_io_read(pblk, rqd, false);
11f6ad69	287	}
a4bd217b	288
11f6ad69 HL	289	static int pblk_setup_partial_read(struct pblk pblk, struct nvm_rq rqd,
	290	unsigned int bio_init_idx,
	291	unsigned long *read_bitmap,
	292	int nr_holes)
	293	{
	294	struct pblk_sec_meta *meta_list = rqd->meta_list;
	295	struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd);
	296	struct pblk_pr_ctx *pr_ctx;
	297	struct bio new_bio, bio = r_ctx->private;
	298	__le64 *lba_list_mem;
	299	int nr_secs = rqd->nr_ppas;
	300	int i;
	301
	302	/* Re-use allocated memory for intermediate lbas */
	303	lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size);
	304
	305	new_bio = bio_alloc(GFP_KERNEL, nr_holes);
	306
	307	if (pblk_bio_add_pages(pblk, new_bio, GFP_KERNEL, nr_holes))
	308	goto fail_bio_put;
	309
	310	if (nr_holes != new_bio->bi_vcnt) {
	311	WARN_ONCE(1, "pblk: malformed bio\n");
	312	goto fail_free_pages;
	313	}
	314
	315	pr_ctx = kmalloc(sizeof(struct pblk_pr_ctx), GFP_KERNEL);
	316	if (!pr_ctx)
	317	goto fail_free_pages;
	318
	319	for (i = 0; i < nr_secs; i++)
	320	lba_list_mem[i] = meta_list[i].lba;
	321
	322	new_bio->bi_iter.bi_sector = 0; /* internal bio */
	323	bio_set_op_attrs(new_bio, REQ_OP_READ, 0);
	324
	325	rqd->bio = new_bio;
	326	rqd->nr_ppas = nr_holes;
11f6ad69 HL	327
	328	pr_ctx->ppa_ptr = NULL;
	329	pr_ctx->orig_bio = bio;
	330	bitmap_copy(pr_ctx->bitmap, read_bitmap, NVM_MAX_VLBA);
	331	pr_ctx->bio_init_idx = bio_init_idx;
	332	pr_ctx->orig_nr_secs = nr_secs;
	333	r_ctx->private = pr_ctx;
	334
	335	if (unlikely(nr_holes == 1)) {
	336	pr_ctx->ppa_ptr = rqd->ppa_list;
	337	pr_ctx->dma_ppa_list = rqd->dma_ppa_list;
	338	rqd->ppa_addr = rqd->ppa_list[0];
	339	}
	340	return 0;
	341
	342	fail_free_pages:
fbadca73	343	pblk_bio_free_pages(pblk, new_bio, 0, new_bio->bi_vcnt);
11f6ad69 HL	344	fail_bio_put:
	345	bio_put(new_bio);
	346
	347	return -ENOMEM;
	348	}
	349
	350	static int pblk_partial_read_bio(struct pblk pblk, struct nvm_rq rqd,
	351	unsigned int bio_init_idx,
	352	unsigned long *read_bitmap, int nr_secs)
	353	{
	354	int nr_holes;
	355	int ret;
	356
	357	nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs);
	358
	359	if (pblk_setup_partial_read(pblk, rqd, bio_init_idx, read_bitmap,
	360	nr_holes))
	361	return NVM_IO_ERR;
	362
	363	rqd->end_io = pblk_end_partial_read;
	364
	365	ret = pblk_submit_io(pblk, rqd);
	366	if (ret) {
	367	bio_put(rqd->bio);
	368	pblk_err(pblk, "partial read IO submission failed\n");
	369	goto err;
	370	}
	371
	372	return NVM_IO_OK;
	373
	374	err:
4e495a46	375	pblk_err(pblk, "failed to perform partial read\n");
11f6ad69 HL	376
	377	/* Free allocated pages in new bio */
	378	pblk_bio_free_pages(pblk, rqd->bio, 0, rqd->bio->bi_vcnt);
7bd4d370	379	__pblk_end_io_read(pblk, rqd, false);
a4bd217b JG	380	return NVM_IO_ERR;
	381	}
	382
87cc40bb	383	static void pblk_read_rq(struct pblk pblk, struct nvm_rq rqd, struct bio *bio,
84454e6d	384	sector_t lba, unsigned long *read_bitmap)
a4bd217b	385	{
a4809fee	386	struct pblk_sec_meta *meta_list = rqd->meta_list;
a4bd217b	387	struct ppa_addr ppa;
a4bd217b JG	388
	389	pblk_lookup_l2p_seq(pblk, &ppa, lba, 1);
	390
880eda54	391	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	392	atomic_long_inc(&pblk->inflight_reads);
	393	#endif
	394
	395	retry:
	396	if (pblk_ppa_empty(ppa)) {
	397	WARN_ON(test_and_set_bit(0, read_bitmap));
a4809fee	398	meta_list[0].lba = cpu_to_le64(ADDR_EMPTY);
a4bd217b JG	399	return;
	400	}
	401
	402	/* Try to read from write buffer. The address is later checked on the
	403	* write buffer to prevent retrieving overwritten data.
	404	*/
	405	if (pblk_addr_in_cache(ppa)) {
75cb8e93	406	if (!pblk_read_from_cache(pblk, bio, lba, ppa, 0, 1)) {
a4bd217b JG	407	pblk_lookup_l2p_seq(pblk, &ppa, lba, 1);
	408	goto retry;
	409	}
a4809fee	410
7bd4d370	411	WARN_ON(test_and_set_bit(0, read_bitmap));
a4809fee JG	412	meta_list[0].lba = cpu_to_le64(lba);
a4809fee JG	413
880eda54	414	#ifdef CONFIG_NVM_PBLK_DEBUG
7bd4d370	415	atomic_long_inc(&pblk->cache_reads);
db7ada33	416	#endif
a4bd217b JG	417	} else {
	418	rqd->ppa_addr = ppa;
	419	}
	420	}
	421
	422	int pblk_submit_read(struct pblk pblk, struct bio bio)
	423	{
	424	struct nvm_tgt_dev *dev = pblk->dev;
998ba629	425	struct request_queue *q = dev->q;
84454e6d	426	sector_t blba = pblk_get_lba(bio);
5bf1e1ee	427	unsigned int nr_secs = pblk_get_secs(bio);
a4809fee	428	struct pblk_g_ctx *r_ctx;
a4bd217b	429	struct nvm_rq *rqd;
a4bd217b	430	unsigned int bio_init_idx;
921aebfa	431	DECLARE_BITMAP(read_bitmap, NVM_MAX_VLBA);
a4bd217b JG	432	int ret = NVM_IO_ERR;
a4bd217b JG	433
ddcf35d3 MC	434	generic_start_io_acct(q, REQ_OP_READ, bio_sectors(bio),
ddcf35d3 MC	435	&pblk->disk->part0);
998ba629	436
921aebfa	437	bitmap_zero(read_bitmap, nr_secs);
a4bd217b	438
e2cddf20	439	rqd = pblk_alloc_rqd(pblk, PBLK_READ);
a4bd217b JG	440
a4bd217b JG	441	rqd->opcode = NVM_OP_PREAD;
a4bd217b	442	rqd->nr_ppas = nr_secs;
87cc40bb	443	rqd->bio = NULL; /* cloned bio if needed */
a4bd217b JG	444	rqd->private = pblk;
	445	rqd->end_io = pblk_end_io_read;
	446
a4809fee	447	r_ctx = nvm_rq_to_pdu(rqd);
998ba629	448	r_ctx->start_time = jiffies;
a4809fee	449	r_ctx->lba = blba;
87cc40bb	450	r_ctx->private = bio; /* original bio */
a4809fee	451
a4bd217b JG	452	/* Save the index for this bio's start. This is needed in case
	453	* we need to fill a partial read.
	454	*/
	455	bio_init_idx = pblk_get_bi_idx(bio);
	456
45dcf29b	457	if (pblk_alloc_rqd_meta(pblk, rqd))
63e3809c	458	goto fail_rqd_free;
a4bd217b	459
45dcf29b	460	if (nr_secs > 1)
921aebfa	461	pblk_read_ppalist_rq(pblk, rqd, bio, blba, read_bitmap);
45dcf29b	462	else
921aebfa	463	pblk_read_rq(pblk, rqd, bio, blba, read_bitmap);
a4bd217b	464
921aebfa	465	if (bitmap_full(read_bitmap, nr_secs)) {
588726d3	466	atomic_inc(&pblk->inflight_io);
7bd4d370	467	__pblk_end_io_read(pblk, rqd, false);
87cc40bb	468	return NVM_IO_DONE;
a4bd217b JG	469	}
	470
	471	/* All sectors are to be read from the device */
921aebfa	472	if (bitmap_empty(read_bitmap, rqd->nr_ppas)) {
a4bd217b	473	struct bio *int_bio = NULL;
a4bd217b JG	474
a4bd217b JG	475	/* Clone read bio to deal with read errors internally */
b906bbb6	476	int_bio = bio_clone_fast(bio, GFP_KERNEL, &pblk_bio_set);
a4bd217b	477	if (!int_bio) {
4e495a46	478	pblk_err(pblk, "could not clone read bio\n");
998ba629	479	goto fail_end_io;
a4bd217b JG	480	}
	481
	482	rqd->bio = int_bio;
a4bd217b	483
e13f421b	484	if (pblk_submit_io(pblk, rqd)) {
4e495a46	485	pblk_err(pblk, "read IO submission failed\n");
e13f421b	486	ret = NVM_IO_ERR;
998ba629	487	goto fail_end_io;
a4bd217b JG	488	}
	489
	490	return NVM_IO_OK;
	491	}
	492
	493	/* The read bio request could be partially filled by the write buffer,
	494	* but there are some holes that need to be read from the drive.
	495	*/
11f6ad69 HL	496	ret = pblk_partial_read_bio(pblk, rqd, bio_init_idx, read_bitmap,
	497	nr_secs);
	498	if (ret)
	499	goto fail_meta_free;
	500
	501	return NVM_IO_OK;
a4bd217b	502
11f6ad69 HL	503	fail_meta_free:
11f6ad69 HL	504	nvm_dev_dma_free(dev->parent, rqd->meta_list, rqd->dma_meta_list);
a4bd217b	505	fail_rqd_free:
e2cddf20	506	pblk_free_rqd(pblk, rqd, PBLK_READ);
a4bd217b	507	return ret;
998ba629 JG	508	fail_end_io:
	509	__pblk_end_io_read(pblk, rqd, false);
	510	return ret;
a4bd217b JG	511	}
	512
	513	static int read_ppalist_rq_gc(struct pblk pblk, struct nvm_rq rqd,
	514	struct pblk_line line, u64 lba_list,
d340121e	515	u64 *paddr_list_gc, unsigned int nr_secs)
a4bd217b	516	{
afdc23c9	517	struct ppa_addr ppa_list_l2p[NVM_MAX_VLBA];
d340121e	518	struct ppa_addr ppa_gc;
a4bd217b JG	519	int valid_secs = 0;
	520	int i;
	521
d340121e	522	pblk_lookup_l2p_rand(pblk, ppa_list_l2p, lba_list, nr_secs);
a4bd217b JG	523
a4bd217b JG	524	for (i = 0; i < nr_secs; i++) {
d340121e JG	525	if (lba_list[i] == ADDR_EMPTY)
	526	continue;
	527
	528	ppa_gc = addr_to_gen_ppa(pblk, paddr_list_gc[i], line->id);
	529	if (!pblk_ppa_comp(ppa_list_l2p[i], ppa_gc)) {
	530	paddr_list_gc[i] = lba_list[i] = ADDR_EMPTY;
a4bd217b JG	531	continue;
	532	}
	533
d340121e	534	rqd->ppa_list[valid_secs++] = ppa_list_l2p[i];
a4bd217b JG	535	}
a4bd217b JG	536
880eda54	537	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	538	atomic_long_add(valid_secs, &pblk->inflight_reads);
a4bd217b JG	539	#endif
d340121e	540
a4bd217b JG	541	return valid_secs;
	542	}
	543
	544	static int read_rq_gc(struct pblk pblk, struct nvm_rq rqd,
d340121e JG	545	struct pblk_line *line, sector_t lba,
d340121e JG	546	u64 paddr_gc)
a4bd217b	547	{
d340121e	548	struct ppa_addr ppa_l2p, ppa_gc;
a4bd217b JG	549	int valid_secs = 0;
a4bd217b JG	550
659226eb DC	551	if (lba == ADDR_EMPTY)
	552	goto out;
	553
a4bd217b	554	/* logic error: lba out-of-bounds */
2a79efd8 DC	555	if (lba >= pblk->rl.nr_secs) {
2a79efd8 DC	556	WARN(1, "pblk: read lba out of bounds\n");
a4bd217b JG	557	goto out;
	558	}
	559
a4bd217b	560	spin_lock(&pblk->trans_lock);
d340121e	561	ppa_l2p = pblk_trans_map_get(pblk, lba);
a4bd217b JG	562	spin_unlock(&pblk->trans_lock);
a4bd217b JG	563
d340121e JG	564	ppa_gc = addr_to_gen_ppa(pblk, paddr_gc, line->id);
d340121e JG	565	if (!pblk_ppa_comp(ppa_l2p, ppa_gc))
a4bd217b JG	566	goto out;
a4bd217b JG	567
d340121e	568	rqd->ppa_addr = ppa_l2p;
a4bd217b JG	569	valid_secs = 1;
a4bd217b JG	570
880eda54	571	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	572	atomic_long_inc(&pblk->inflight_reads);
	573	#endif
	574
	575	out:
	576	return valid_secs;
	577	}
	578
d340121e	579	int pblk_submit_read_gc(struct pblk pblk, struct pblk_gc_rq gc_rq)
a4bd217b JG	580	{
	581	struct nvm_tgt_dev *dev = pblk->dev;
	582	struct nvm_geo *geo = &dev->geo;
a4bd217b JG	583	struct bio *bio;
a4bd217b JG	584	struct nvm_rq rqd;
d340121e JG	585	int data_len;
d340121e JG	586	int ret = NVM_IO_OK;
a4bd217b JG	587
	588	memset(&rqd, 0, sizeof(struct nvm_rq));
	589
45dcf29b JG	590	ret = pblk_alloc_rqd_meta(pblk, &rqd);
	591	if (ret)
	592	return ret;
63e3809c	593
d340121e	594	if (gc_rq->nr_secs > 1) {
d340121e JG	595	gc_rq->secs_to_gc = read_ppalist_rq_gc(pblk, &rqd, gc_rq->line,
	596	gc_rq->lba_list,
	597	gc_rq->paddr_list,
	598	gc_rq->nr_secs);
	599	if (gc_rq->secs_to_gc == 1)
63e3809c	600	rqd.ppa_addr = rqd.ppa_list[0];
a4bd217b	601	} else {
d340121e JG	602	gc_rq->secs_to_gc = read_rq_gc(pblk, &rqd, gc_rq->line,
	603	gc_rq->lba_list[0],
	604	gc_rq->paddr_list[0]);
a4bd217b JG	605	}
a4bd217b JG	606
d340121e	607	if (!(gc_rq->secs_to_gc))
a4bd217b JG	608	goto out;
a4bd217b JG	609
e46f4e48	610	data_len = (gc_rq->secs_to_gc) * geo->csecs;
d340121e	611	bio = pblk_bio_map_addr(pblk, gc_rq->data, gc_rq->secs_to_gc, data_len,
7d327a9e	612	PBLK_VMALLOC_META, GFP_KERNEL);
a4bd217b	613	if (IS_ERR(bio)) {
4e495a46	614	pblk_err(pblk, "could not allocate GC bio (%lu)\n",
45dcf29b JG	615	PTR_ERR(bio));
45dcf29b JG	616	ret = PTR_ERR(bio);
a4bd217b JG	617	goto err_free_dma;
	618	}
	619
	620	bio->bi_iter.bi_sector = 0; /* internal bio */
	621	bio_set_op_attrs(bio, REQ_OP_READ, 0);
	622
	623	rqd.opcode = NVM_OP_PREAD;
d340121e	624	rqd.nr_ppas = gc_rq->secs_to_gc;
a4bd217b JG	625	rqd.bio = bio;
a4bd217b JG	626
1a94b2d4	627	if (pblk_submit_io_sync(pblk, &rqd)) {
d340121e	628	ret = -EIO;
4e495a46	629	pblk_err(pblk, "GC read request failed\n");
7d327a9e	630	goto err_free_bio;
a4bd217b JG	631	}
a4bd217b JG	632
03a34b2d	633	pblk_read_check_rand(pblk, &rqd, gc_rq->lba_list, gc_rq->nr_secs);
310df582	634
588726d3	635	atomic_dec(&pblk->inflight_io);
a4bd217b JG	636
	637	if (rqd.error) {
	638	atomic_long_inc(&pblk->read_failed_gc);
880eda54	639	#ifdef CONFIG_NVM_PBLK_DEBUG
a4bd217b JG	640	pblk_print_failed_rqd(pblk, &rqd, rqd.error);
	641	#endif
	642	}
	643
880eda54	644	#ifdef CONFIG_NVM_PBLK_DEBUG
d340121e JG	645	atomic_long_add(gc_rq->secs_to_gc, &pblk->sync_reads);
	646	atomic_long_add(gc_rq->secs_to_gc, &pblk->recov_gc_reads);
	647	atomic_long_sub(gc_rq->secs_to_gc, &pblk->inflight_reads);
a4bd217b JG	648	#endif
	649
	650	out:
45dcf29b	651	pblk_free_rqd_meta(pblk, &rqd);
d340121e	652	return ret;
a4bd217b	653
7d327a9e JG	654	err_free_bio:
7d327a9e JG	655	bio_put(bio);
a4bd217b	656	err_free_dma:
45dcf29b	657	pblk_free_rqd_meta(pblk, &rqd);
d340121e	658	return ret;
a4bd217b	659	}