/*
* eio_mem_init
*/
int eio_mem_init(struct cache_c *dmc)
{
u_int32_t lsb_bits;
u_int32_t msb_bits_24; /* most significant bits in shrunk dbn */
u_int64_t max_dbn;
u_int64_t num_sets_64;
/*
* Sanity check the number of sets.
*/
num_sets_64 = EIO_DIV(dmc->size, dmc->assoc);
if (num_sets_64 > UINT_MAX) {
pr_err("Number of cache sets (%lu) greater than maximum"
"allowed (%u)",
(long unsigned int)num_sets_64, UINT_MAX);
return -1;
}
/*
* Find the number of bits required to encode the set number and
* its corresponding mask value.
*/
dmc->num_sets = (u_int32_t)num_sets_64;
for (dmc->num_sets_bits = 0; (dmc->num_sets >> dmc->num_sets_bits) != 0;
dmc->num_sets_bits++) ;
dmc->num_sets_mask = ULLONG_MAX >> (64 - dmc->num_sets_bits);
/*
* If we don't have at least 16 bits to save,
* we can't use small metadata.
*/
if (dmc->num_sets_bits < 16) {
dmc->cache_flags |= CACHE_FLAGS_MD8;
pr_info("Not enough sets to use small metadata");
return 1;
}
/*
* Now compute the largest sector number that we can shrink; then see
* if the source volume is smaller.
*/
lsb_bits = dmc->consecutive_shift + dmc->block_shift;
msb_bits_24 = 24 - 1 - lsb_bits; /* 1 for wrapped bit */
max_dbn =
((u_int64_t)1) << (msb_bits_24 + dmc->num_sets_bits + lsb_bits);
if (to_sector(eio_get_device_size(dmc->disk_dev)) > max_dbn) {
dmc->cache_flags |= CACHE_FLAGS_MD8;
pr_info("Source volume too big to use small metadata");
return 1;
}
return 0;
}
/*-----------------------------------------------------------------
* IO routines that accept a list of pages.
*---------------------------------------------------------------*/
static void do_region(int rw, unsigned region, struct dm_io_region *where,
struct dpages *dp, struct io *io)
{
struct bio *bio;
struct page *page;
unsigned long len;
unsigned offset;
unsigned num_bvecs;
sector_t remaining = where->count;
/*
* where->count may be zero if rw holds a flush and we need to
* send a zero-sized flush.
*/
do {
/*
* Allocate a suitably sized-bio.
*/
num_bvecs = dm_sector_div_up(remaining,
(PAGE_SIZE >> SECTOR_SHIFT));
num_bvecs = min_t(int, bio_get_nr_vecs(where->bdev), num_bvecs);
bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios);
if (!bio) {
printk(KERN_WARNING "%s : %s() failed\n", __FILE__, __func__);
BUG_ON(1);
}
bio->bi_sector = where->sector + (where->count - remaining);
bio->bi_bdev = where->bdev;
bio->bi_end_io = endio;
bio->bi_destructor = dm_bio_destructor;
store_io_and_region_in_bio(bio, io, region);
/*
* Try and add as many pages as possible.
*/
while (remaining) {
dp->get_page(dp, &page, &len, &offset);
len = min(len, to_bytes(remaining));
if (!bio_add_page(bio, page, len, offset))
break;
offset = 0;
remaining -= to_sector(len);
dp->next_page(dp);
}
atomic_inc(&io->count);
submit_bio(rw, bio);
} while (remaining);
}
static void do_region(int rw, unsigned region, struct dm_io_region *where,
struct dpages *dp, struct io *io)
{
struct bio *bio;
struct page *page;
unsigned long len;
unsigned offset;
unsigned num_bvecs;
sector_t remaining = where->count;
struct request_queue *q = bdev_get_queue(where->bdev);
sector_t discard_sectors;
do {
if (rw & REQ_DISCARD)
num_bvecs = 1;
else
num_bvecs = min_t(int, bio_get_nr_vecs(where->bdev),
dm_sector_div_up(remaining, (PAGE_SIZE >> SECTOR_SHIFT)));
bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios);
bio->bi_sector = where->sector + (where->count - remaining);
bio->bi_bdev = where->bdev;
bio->bi_end_io = endio;
bio->bi_destructor = dm_bio_destructor;
store_io_and_region_in_bio(bio, io, region);
if (rw & REQ_DISCARD) {
discard_sectors = min_t(sector_t, q->limits.max_discard_sectors, remaining);
bio->bi_size = discard_sectors << SECTOR_SHIFT;
remaining -= discard_sectors;
} else while (remaining) {
dp->get_page(dp, &page, &len, &offset);
len = min(len, to_bytes(remaining));
if (!bio_add_page(bio, page, len, offset))
break;
offset = 0;
remaining -= to_sector(len);
dp->next_page(dp);
}
atomic_inc(&io->count);
submit_bio(rw, bio);
} while (remaining);
}
static struct iostash_bio *_io_alloc(struct hdd_info * hdd, struct ssd_info * ssd, uint32_t fragnum,
struct bio *bio, sector_t psn)
{
struct iostash_bio *io = mempool_alloc(hdd->io_pool, GFP_NOIO);
if (io) {
atomic_inc(&hdd->io_pending);
io->hdd = hdd;
io->ssd = ssd;
io->fragnum = fragnum;
io->base_bio = bio;
io->psn = psn;
io->nr_sctr = to_sector(BIO_SIZE(bio));
io->error = 0;
io->ssd_werr = 0; /* SSD write error */
atomic_set(&io->io_pending, 0);
}
return io;
}
static void __clone_and_map(struct clone_info *ci)
{
struct bio *clone, *bio = ci->bio;
struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti);
struct target_io *tio;
/*
* Allocate a target io object.
*/
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
if (ci->sector_count <= max) {
/*
* Optimise for the simple case where we can do all of
* the remaining io with a single clone.
*/
clone = clone_bio(bio, ci->sector, ci->idx,
bio->bi_vcnt - ci->idx, ci->sector_count);
__map_bio(ti, clone, tio);
ci->sector_count = 0;
} else if (to_sector(bio->bi_io_vec[ci->idx].bv_len) <= max) {
/*
* There are some bvecs that don't span targets.
* Do as many of these as possible.
*/
int i;
sector_t remaining = max;
sector_t bv_len;
for (i = ci->idx; remaining && (i < bio->bi_vcnt); i++) {
bv_len = to_sector(bio->bi_io_vec[i].bv_len);
if (bv_len > remaining)
break;
remaining -= bv_len;
len += bv_len;
}
clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len);
__map_bio(ti, clone, tio);
ci->sector += len;
ci->sector_count -= len;
ci->idx = i;
} else {
/*
* Create two copy bios to deal with io that has
* been split across a target.
*/
struct bio_vec *bv = bio->bi_io_vec + ci->idx;
clone = split_bvec(bio, ci->sector, ci->idx,
bv->bv_offset, max);
__map_bio(ti, clone, tio);
ci->sector += max;
ci->sector_count -= max;
ti = dm_table_find_target(ci->map, ci->sector);
len = to_sector(bv->bv_len) - max;
clone = split_bvec(bio, ci->sector, ci->idx,
bv->bv_offset + to_bytes(max), len);
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
__map_bio(ti, clone, tio);
ci->sector += len;
ci->sector_count -= len;
ci->idx++;
}
}
static void __clone_and_map(struct clone_info *ci)
{
struct bio *clone, *bio = ci->bio;
struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti);
struct target_io *tio;
/*
* Allocate a target io object.
*/
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
if (ci->sector_count <= max) {
/*
* Optimise for the simple case where we can do all of
* the remaining io with a single clone.
*/
clone = clone_bio(bio, ci->sector, ci->idx,
bio->bi_vcnt - ci->idx, ci->sector_count,
ci->md->bs);
__map_bio(ti, clone, tio);
ci->sector_count = 0;
} else if (to_sector(bio->bi_io_vec[ci->idx].bv_len) <= max) {
/*
* There are some bvecs that don't span targets.
* Do as many of these as possible.
*/
int i;
sector_t remaining = max;
sector_t bv_len;
for (i = ci->idx; remaining && (i < bio->bi_vcnt); i++) {
bv_len = to_sector(bio->bi_io_vec[i].bv_len);
if (bv_len > remaining)
break;
remaining -= bv_len;
len += bv_len;
}
clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len,
ci->md->bs);
__map_bio(ti, clone, tio);
ci->sector += len;
ci->sector_count -= len;
ci->idx = i;
} else {
/*
* Handle a bvec that must be split between two or more targets.
*/
struct bio_vec *bv = bio->bi_io_vec + ci->idx;
sector_t remaining = to_sector(bv->bv_len);
unsigned int offset = 0;
do {
if (offset) {
ti = dm_table_find_target(ci->map, ci->sector);
max = max_io_len(ci->md, ci->sector, ti);
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
}
len = min(remaining, max);
clone = split_bvec(bio, ci->sector, ci->idx,
bv->bv_offset + offset, len,
ci->md->bs);
__map_bio(ti, clone, tio);
ci->sector += len;
ci->sector_count -= len;
offset += to_bytes(len);
} while (remaining -= len);
ci->idx++;
}
}
int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmd, unsigned int flags, get_block_t get_block,
dax_iodone_t complete_unwritten)
{
struct file *file = vma->vm_file;
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
struct buffer_head bh;
unsigned blkbits = inode->i_blkbits;
unsigned long pmd_addr = address & PMD_MASK;
bool write = flags & FAULT_FLAG_WRITE;
struct block_device *bdev;
pgoff_t size, pgoff;
sector_t block;
int error, result = 0;
bool alloc = false;
/* dax pmd mappings require pfn_t_devmap() */
if (!IS_ENABLED(CONFIG_FS_DAX_PMD))
return VM_FAULT_FALLBACK;
/* Fall back to PTEs if we're going to COW */
if (write && !(vma->vm_flags & VM_SHARED)) {
split_huge_pmd(vma, pmd, address);
dax_pmd_dbg(NULL, address, "cow write");
return VM_FAULT_FALLBACK;
}
/* If the PMD would extend outside the VMA */
if (pmd_addr < vma->vm_start) {
dax_pmd_dbg(NULL, address, "vma start unaligned");
return VM_FAULT_FALLBACK;
}
if ((pmd_addr + PMD_SIZE) > vma->vm_end) {
dax_pmd_dbg(NULL, address, "vma end unaligned");
return VM_FAULT_FALLBACK;
}
pgoff = linear_page_index(vma, pmd_addr);
size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (pgoff >= size)
return VM_FAULT_SIGBUS;
/* If the PMD would cover blocks out of the file */
if ((pgoff | PG_PMD_COLOUR) >= size) {
dax_pmd_dbg(NULL, address,
"offset + huge page size > file size");
return VM_FAULT_FALLBACK;
}
memset(&bh, 0, sizeof(bh));
bh.b_bdev = inode->i_sb->s_bdev;
block = (sector_t)pgoff << (PAGE_SHIFT - blkbits);
bh.b_size = PMD_SIZE;
if (get_block(inode, block, &bh, 0) != 0)
return VM_FAULT_SIGBUS;
if (!buffer_mapped(&bh) && write) {
if (get_block(inode, block, &bh, 1) != 0)
return VM_FAULT_SIGBUS;
alloc = true;
}
bdev = bh.b_bdev;
/*
* If the filesystem isn't willing to tell us the length of a hole,
* just fall back to PTEs. Calling get_block 512 times in a loop
* would be silly.
*/
if (!buffer_size_valid(&bh) || bh.b_size < PMD_SIZE) {
dax_pmd_dbg(&bh, address, "allocated block too small");
return VM_FAULT_FALLBACK;
}
/*
* If we allocated new storage, make sure no process has any
* zero pages covering this hole
*/
if (alloc) {
loff_t lstart = pgoff << PAGE_SHIFT;
loff_t lend = lstart + PMD_SIZE - 1; /* inclusive */
truncate_pagecache_range(inode, lstart, lend);
}
i_mmap_lock_read(mapping);
/*
* If a truncate happened while we were allocating blocks, we may
* leave blocks allocated to the file that are beyond EOF. We can't
* take i_mutex here, so just leave them hanging; they'll be freed
* when the file is deleted.
*/
size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (pgoff >= size) {
result = VM_FAULT_SIGBUS;
goto out;
}
if ((pgoff | PG_PMD_COLOUR) >= size) {
dax_pmd_dbg(&bh, address,
"offset + huge page size > file size");
goto fallback;
}
if (!write && !buffer_mapped(&bh) && buffer_uptodate(&bh)) {
spinlock_t *ptl;
pmd_t entry;
struct page *zero_page = get_huge_zero_page();
if (unlikely(!zero_page)) {
dax_pmd_dbg(&bh, address, "no zero page");
goto fallback;
}
ptl = pmd_lock(vma->vm_mm, pmd);
if (!pmd_none(*pmd)) {
spin_unlock(ptl);
dax_pmd_dbg(&bh, address, "pmd already present");
goto fallback;
}
dev_dbg(part_to_dev(bdev->bd_part),
"%s: %s addr: %lx pfn: <zero> sect: %llx\n",
__func__, current->comm, address,
(unsigned long long) to_sector(&bh, inode));
entry = mk_pmd(zero_page, vma->vm_page_prot);
entry = pmd_mkhuge(entry);
set_pmd_at(vma->vm_mm, pmd_addr, pmd, entry);
result = VM_FAULT_NOPAGE;
spin_unlock(ptl);
} else {
struct blk_dax_ctl dax = {
.sector = to_sector(&bh, inode),
.size = PMD_SIZE,
};
long length = dax_map_atomic(bdev, &dax);
if (length < 0) {
result = VM_FAULT_SIGBUS;
goto out;
}
if (length < PMD_SIZE) {
dax_pmd_dbg(&bh, address, "dax-length too small");
dax_unmap_atomic(bdev, &dax);
goto fallback;
}
if (pfn_t_to_pfn(dax.pfn) & PG_PMD_COLOUR) {
dax_pmd_dbg(&bh, address, "pfn unaligned");
dax_unmap_atomic(bdev, &dax);
goto fallback;
}
if (!pfn_t_devmap(dax.pfn)) {
dax_unmap_atomic(bdev, &dax);
dax_pmd_dbg(&bh, address, "pfn not in memmap");
goto fallback;
}
if (buffer_unwritten(&bh) || buffer_new(&bh)) {
clear_pmem(dax.addr, PMD_SIZE);
wmb_pmem();
count_vm_event(PGMAJFAULT);
mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT);
result |= VM_FAULT_MAJOR;
}
dax_unmap_atomic(bdev, &dax);
/*
* For PTE faults we insert a radix tree entry for reads, and
* leave it clean. Then on the first write we dirty the radix
* tree entry via the dax_pfn_mkwrite() path. This sequence
* allows the dax_pfn_mkwrite() call to be simpler and avoid a
* call into get_block() to translate the pgoff to a sector in
* order to be able to create a new radix tree entry.
*
* The PMD path doesn't have an equivalent to
* dax_pfn_mkwrite(), though, so for a read followed by a
* write we traverse all the way through __dax_pmd_fault()
* twice. This means we can just skip inserting a radix tree
* entry completely on the initial read and just wait until
* the write to insert a dirty entry.
*/
if (write) {
error = dax_radix_entry(mapping, pgoff, dax.sector,
true, true);
if (error) {
dax_pmd_dbg(&bh, address,
"PMD radix insertion failed");
goto fallback;
}
}
dev_dbg(part_to_dev(bdev->bd_part),
"%s: %s addr: %lx pfn: %lx sect: %llx\n",
__func__, current->comm, address,
pfn_t_to_pfn(dax.pfn),
(unsigned long long) dax.sector);
result |= vmf_insert_pfn_pmd(vma, address, pmd,
dax.pfn, write);
}
out:
i_mmap_unlock_read(mapping);
if (buffer_unwritten(&bh))
complete_unwritten(&bh, !(result & VM_FAULT_ERROR));
return result;
fallback:
count_vm_event(THP_FAULT_FALLBACK);
result = VM_FAULT_FALLBACK;
goto out;
}
EXPORT_SYMBOL_GPL(__dax_pmd_fault);
/**
* dax_pmd_fault - handle a PMD fault on a DAX file
* @vma: The virtual memory area where the fault occurred
* @vmf: The description of the fault
* @get_block: The filesystem method used to translate file offsets to blocks
*
* When a page fault occurs, filesystems may call this helper in their
* pmd_fault handler for DAX files.
*/
int dax_pmd_fault(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmd, unsigned int flags, get_block_t get_block,
dax_iodone_t complete_unwritten)
{
int result;
struct super_block *sb = file_inode(vma->vm_file)->i_sb;
if (flags & FAULT_FLAG_WRITE) {
sb_start_pagefault(sb);
file_update_time(vma->vm_file);
}
result = __dax_pmd_fault(vma, address, pmd, flags, get_block,
complete_unwritten);
if (flags & FAULT_FLAG_WRITE)
sb_end_pagefault(sb);
return result;
}
static ssize_t dax_io(struct inode *inode, struct iov_iter *iter,
loff_t start, loff_t end, get_block_t get_block,
struct buffer_head *bh)
{
loff_t pos = start, max = start, bh_max = start;
bool hole = false, need_wmb = false;
struct block_device *bdev = NULL;
int rw = iov_iter_rw(iter), rc;
long map_len = 0;
struct blk_dax_ctl dax = {
.addr = (void __pmem *) ERR_PTR(-EIO),
};
if (rw == READ)
end = min(end, i_size_read(inode));
while (pos < end) {
size_t len;
if (pos == max) {
unsigned blkbits = inode->i_blkbits;
long page = pos >> PAGE_SHIFT;
sector_t block = page << (PAGE_SHIFT - blkbits);
unsigned first = pos - (block << blkbits);
long size;
if (pos == bh_max) {
bh->b_size = PAGE_ALIGN(end - pos);
bh->b_state = 0;
rc = get_block(inode, block, bh, rw == WRITE);
if (rc)
break;
if (!buffer_size_valid(bh))
bh->b_size = 1 << blkbits;
bh_max = pos - first + bh->b_size;
bdev = bh->b_bdev;
} else {
unsigned done = bh->b_size -
(bh_max - (pos - first));
bh->b_blocknr += done >> blkbits;
bh->b_size -= done;
}
hole = rw == READ && !buffer_written(bh);
if (hole) {
size = bh->b_size - first;
} else {
dax_unmap_atomic(bdev, &dax);
dax.sector = to_sector(bh, inode);
dax.size = bh->b_size;
map_len = dax_map_atomic(bdev, &dax);
if (map_len < 0) {
rc = map_len;
break;
}
if (buffer_unwritten(bh) || buffer_new(bh)) {
dax_new_buf(dax.addr, map_len, first,
pos, end);
need_wmb = true;
}
dax.addr += first;
size = map_len - first;
}
max = min(pos + size, end);
}
if (iov_iter_rw(iter) == WRITE) {
len = copy_from_iter_pmem(dax.addr, max - pos, iter);
need_wmb = true;
} else if (!hole)
len = copy_to_iter((void __force *) dax.addr, max - pos,
iter);
else
len = iov_iter_zero(max - pos, iter);
if (!len) {
rc = -EFAULT;
break;
}
pos += len;
if (!IS_ERR(dax.addr))
dax.addr += len;
}
if (need_wmb)
wmb_pmem();
dax_unmap_atomic(bdev, &dax);
return (pos == start) ? rc : pos - start;
}
void iostash_mkrequest(struct request_queue *q, struct bio *bio)
#endif
{
struct hdd_info *hdd;
struct ssd_info *ssd;
struct iostash_bio *io;
sce_fmap_t fmap;
uint32_t nr_sctr;
sector_t psn;
make_request_fn *org_mapreq = NULL;
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
blk_qc_t ret = BLK_QC_T_NONE;
#endif
DBG("Got bio=%p bio->bi_rw(%lu) request at s=%lu l=%u.\n",
bio, bio->bi_rw, BIO_SECTOR(bio), bio_sectors(bio));
rcu_read_lock();
hdd = hdd_search(bio);
if (hdd) {
atomic_inc(&hdd->nr_ref);
org_mapreq = hdd->org_mapreq;
}
rcu_read_unlock();
if (unlikely(NULL == hdd)) {
/* have to requeue the request, somebody was holding a
* dangling reference */
ERR("Request holding a dangling make_request_fn pointer\n.");
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
bio->bi_error = -EAGAIN;
return ret;
#elif LINUX_VERSION_CODE <= KERNEL_VERSION(3,1,0)
rmb(); /* read the change in make_request_fn */
return -EAGAIN; /* retry */
#else
/* no retry possible in newer kernels since the return
* of make_request_fn is no longer checked and retried
* if not zero, we cannot unload the module */
BUG();
return;
#endif
}
if (!hdd->online) {
ERR("request re-routed due to hdd not being online.\n");
/* being unloaded, re-route */
goto out;
}
hdd->request_q = q;
/* calculate physical sector number -- offset partition information */
psn = BIO_SECTOR(bio) + bio->bi_bdev->bd_part->start_sect;
nr_sctr = to_sector(BIO_SIZE(bio));
do {
if (bio_sectors(bio) == 0)
break;
/* partition boundary check */
if ((psn < hdd->part_start) ||
((psn + nr_sctr) > hdd->part_end))
break;
if (bio_data_dir(bio) == WRITE) {
gctx.st_write++;
#ifdef SCE_AWT
/* make sure the request is only for one fragment */
if (((psn + nr_sctr - 1) / SCE_SCTRPERFRAG) !=
(psn / SCE_SCTRPERFRAG)) {
sce_invalidate(hdd->lun, psn, nr_sctr);
break;
}
rcu_read_lock();
if (sce_get4write(hdd->lun, psn, nr_sctr, &fmap)
== SCE_SUCCESS) {
ssd = (struct ssd_info *)fmap.cdevctx;
atomic_inc(&ssd->nr_ref);
rcu_read_unlock();
if (!ssd->online) {
sce_put4write(hdd->lun, psn,
nr_sctr, 1);
atomic_dec(&ssd->nr_ref);
} else {
io = _io_alloc(hdd, ssd, fmap.fragnum, bio, psn);
if (NULL == io) {
atomic_dec(&ssd->nr_ref);
break;
}
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
ret = _io_worker_run(&io->work);
#else
_io_queue(io);
#endif
/* lose the reference to hdd, not needed anymore */
atomic_dec(&hdd->nr_ref);
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
return ret;
#elif LINUX_VERSION_CODE <= KERNEL_VERSION(3,1,0)
return 0;
#else
return;
#endif
}
} else
rcu_read_unlock();
#else
sce_invalidate(hdd->lun, psn, nr_sctr);
#endif
break;
}
else
{
/* Read handling */
gctx.st_read++;
/* make sure the request is only for one fragment */
if (((psn + nr_sctr - 1) / SCE_SCTRPERFRAG) !=
(psn / SCE_SCTRPERFRAG))
break;
/* cache hit/miss check */
rcu_read_lock();
if (sce_get4read(hdd->lun, psn, nr_sctr, &fmap) != SCE_SUCCESS) {
rcu_read_unlock();
break;
}
BUG_ON(NULL == fmap.cdevctx);
ssd = (struct ssd_info *) fmap.cdevctx;
atomic_inc(&ssd->nr_ref);
rcu_read_unlock();
/* make sure the request is within the SSD limits and the SSD is online */
if (!ssd->online || ssd->queue_max_hw_sectors < nr_sctr) {
sce_put4read(hdd->lun, psn, nr_sctr);
atomic_dec(&ssd->nr_ref);
break;
}
/* cache hit */
io = _io_alloc(hdd, ssd, fmap.fragnum, bio, psn);
if (NULL == io) {
atomic_dec(&ssd->nr_ref);
break;
}
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
ret = _io_worker_run(&io->work);
#else
_io_queue(io);
#endif
/* lose the reference to hdd , not needed anymore */
atomic_dec(&hdd->nr_ref);
}
#if KERNEL_VERSION(4,4,0) <= LINUX_VERSION_CODE
return ret;
#elif LINUX_VERSION_CODE <= KERNEL_VERSION(3,1,0)
return 0;
#else
return;
#endif
} while (0);
out:
/* lose the reference to hdd , not needed anymore */
atomic_dec(&hdd->nr_ref);
return (org_mapreq) (q, bio);
}
static ssize_t dax_io(struct inode *inode, struct iov_iter *iter,
loff_t start, loff_t end, get_block_t get_block,
struct buffer_head *bh)
{
loff_t pos = start, max = start, bh_max = start;
bool hole = false, need_wmb = false;
struct block_device *bdev = NULL;
int rw = iov_iter_rw(iter), rc;
long map_len = 0;
struct blk_dax_ctl dax = {
.addr = (void __pmem *) ERR_PTR(-EIO),
};
unsigned blkbits = inode->i_blkbits;
sector_t file_blks = (i_size_read(inode) + (1 << blkbits) - 1)
>> blkbits;
if (rw == READ)
end = min(end, i_size_read(inode));
while (pos < end) {
size_t len;
if (pos == max) {
long page = pos >> PAGE_SHIFT;
sector_t block = page << (PAGE_SHIFT - blkbits);
unsigned first = pos - (block << blkbits);
long size;
if (pos == bh_max) {
bh->b_size = PAGE_ALIGN(end - pos);
bh->b_state = 0;
rc = get_block(inode, block, bh, rw == WRITE);
if (rc)
break;
if (!buffer_size_valid(bh))
bh->b_size = 1 << blkbits;
bh_max = pos - first + bh->b_size;
bdev = bh->b_bdev;
/*
* We allow uninitialized buffers for writes
* beyond EOF as those cannot race with faults
*/
WARN_ON_ONCE(
(buffer_new(bh) && block < file_blks) ||
(rw == WRITE && buffer_unwritten(bh)));
} else {
unsigned done = bh->b_size -
(bh_max - (pos - first));
bh->b_blocknr += done >> blkbits;
bh->b_size -= done;
}
hole = rw == READ && !buffer_written(bh);
if (hole) {
size = bh->b_size - first;
} else {
dax_unmap_atomic(bdev, &dax);
dax.sector = to_sector(bh, inode);
dax.size = bh->b_size;
map_len = dax_map_atomic(bdev, &dax);
if (map_len < 0) {
rc = map_len;
break;
}
dax.addr += first;
size = map_len - first;
}
/*
* pos + size is one past the last offset for IO,
* so pos + size can overflow loff_t at extreme offsets.
* Cast to u64 to catch this and get the true minimum.
*/
max = min_t(u64, pos + size, end);
}
if (iov_iter_rw(iter) == WRITE) {
len = copy_from_iter_pmem(dax.addr, max - pos, iter);
need_wmb = true;
} else if (!hole)
len = copy_to_iter((void __force *) dax.addr, max - pos,
iter);
else
len = iov_iter_zero(max - pos, iter);
if (!len) {
rc = -EFAULT;
break;
}
pos += len;
if (!IS_ERR(dax.addr))
dax.addr += len;
}
if (need_wmb)
wmb_pmem();
dax_unmap_atomic(bdev, &dax);
return (pos == start) ? rc : pos - start;
}
/*-----------------------------------------------------------------
* IO routines that accept a list of pages.
*---------------------------------------------------------------*/
static void do_region(int rw, unsigned region, struct dm_io_region *where,
struct dpages *dp, struct io *io)
{
struct bio *bio;
struct page *page;
unsigned long len;
unsigned offset;
unsigned num_bvecs;
sector_t remaining = where->count;
struct request_queue *q = bdev_get_queue(where->bdev);
unsigned short logical_block_size = queue_logical_block_size(q);
sector_t num_sectors;
unsigned int uninitialized_var(special_cmd_max_sectors);
/*
* Reject unsupported discard and write same requests.
*/
if (rw & REQ_DISCARD)
special_cmd_max_sectors = q->limits.max_discard_sectors;
else if (rw & REQ_WRITE_SAME)
special_cmd_max_sectors = q->limits.max_write_same_sectors;
if ((rw & (REQ_DISCARD | REQ_WRITE_SAME)) && special_cmd_max_sectors == 0) {
dec_count(io, region, -EOPNOTSUPP);
return;
}
/*
* where->count may be zero if rw holds a flush and we need to
* send a zero-sized flush.
*/
do {
/*
* Allocate a suitably sized-bio.
*/
if ((rw & REQ_DISCARD) || (rw & REQ_WRITE_SAME))
num_bvecs = 1;
else
num_bvecs = min_t(int, BIO_MAX_PAGES,
dm_sector_div_up(remaining, (PAGE_SIZE >> SECTOR_SHIFT)));
bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, io->client->bios);
bio->bi_iter.bi_sector = where->sector + (where->count - remaining);
bio->bi_bdev = where->bdev;
bio->bi_end_io = endio;
store_io_and_region_in_bio(bio, io, region);
if (rw & REQ_DISCARD) {
num_sectors = min_t(sector_t, special_cmd_max_sectors, remaining);
bio->bi_iter.bi_size = num_sectors << SECTOR_SHIFT;
remaining -= num_sectors;
} else if (rw & REQ_WRITE_SAME) {
/*
* WRITE SAME only uses a single page.
*/
dp->get_page(dp, &page, &len, &offset);
bio_add_page(bio, page, logical_block_size, offset);
num_sectors = min_t(sector_t, special_cmd_max_sectors, remaining);
bio->bi_iter.bi_size = num_sectors << SECTOR_SHIFT;
offset = 0;
remaining -= num_sectors;
dp->next_page(dp);
} else while (remaining) {
/*
* Try and add as many pages as possible.
*/
dp->get_page(dp, &page, &len, &offset);
len = min(len, to_bytes(remaining));
if (!bio_add_page(bio, page, len, offset))
break;
offset = 0;
remaining -= to_sector(len);
dp->next_page(dp);
}
atomic_inc(&io->count);
submit_bio(rw, bio);
} while (remaining);
}
/**ltl
* 功能: 将bio请求分割成多个bio，并分发到多个目标设备中
* 参数:
* 返回值:
* 说明: 此
*/
static void __clone_and_map(struct clone_info *ci)
{
struct bio *clone, *bio = ci->bio;
/* 根据请求的起始扇区获取目标设备 */
struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti); /* 可以请求的最大数据长度 */
struct target_io *tio;
/*
* Allocate a target io object.
*/
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
/* [step1]请求的数据长度没有超出目标设备的剩下的数据长度，则一次操作就可以完成 */
if (ci->sector_count <= max) {
/*
* Optimise for the simple case where we can do all of
* the remaining io with a single clone.
*/
/* 拷贝bio对象 */
clone = clone_bio(bio, ci->sector, ci->idx,
bio->bi_vcnt - ci->idx, ci->sector_count);
/* 将新的bio请求映射为目标设备的请求 */
__map_bio(ti, clone, tio);
ci->sector_count = 0;
}/*[step2]请求的数据长度已经超过目标设备空间，但是idx下标所指的数组项在目标设备空间范围之内 */
else if (to_sector(bio->bi_io_vec[ci->idx].bv_len) <= max) {
/*
* There are some bvecs that don't span targets.
* Do as many of these as possible.
*/
int i;
sector_t remaining = max;
sector_t bv_len;
/* 求出在此目标设备可以传输的最大数据长度 */
for (i = ci->idx; remaining && (i < bio->bi_vcnt); i++) {
bv_len = to_sector(bio->bi_io_vec[i].bv_len);
/* 此bio_vec的数据长度已经超出目标设备的空间
* 表明bio->bi_io_vec[i]中的数据跨越两个目标设备，因此剩余数据要走[step3]中的流程
*/
if (bv_len > remaining)
break;
remaining -= bv_len;
len += bv_len;
}
/* 克隆bio对象 */
clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len);
/* 将bio对象映射到目标设备 */
__map_bio(ti, clone, tio);
/* 剩下请求的起始扇区 */
ci->sector += len;
/* 剩下的扇区数，若此字段不为0，则会转到[step3]执行 */
ci->sector_count -= len;
/* 数据下标 */
ci->idx = i;
}
else
{/* [step3]表示bio->bi_io_vec[i]请求数据跨越了两个目标设备，因此要对其分割 */
/*
* Handle a bvec that must be split between two or more targets.
*/
/* 跨越两个目标设备的bio_vec对象 */
struct bio_vec *bv = bio->bi_io_vec + ci->idx;
/* 数据长度 */
sector_t remaining = to_sector(bv->bv_len);
unsigned int offset = 0;
/* 将bio_vec分割成多份，分发到目标设备中 */
do {
if (offset) {
ti = dm_table_find_target(ci->map, ci->sector); /* 在btree中查找 */
max = max_io_len(ci->md, ci->sector, ti);
tio = alloc_tio(ci->md);
tio->io = ci->io;
tio->ti = ti;
memset(&tio->info, 0, sizeof(tio->info));
}
len = min(remaining, max); /* 数据长度 */
/* 分割bio_vec请求 */
clone = split_bvec(bio, ci->sector, ci->idx,
bv->bv_offset + offset, len);
/* 将请求映射到目标设备 */
__map_bio(ti, clone, tio);
ci->sector += len; /* dm的起始扇区号 */
ci->sector_count -= len; /* 剩下的扇区数 */
offset += to_bytes(len); /* bio_vec的数据偏移量 */
} while (remaining -= len);
/* 将下标指向下了个bio_vec数组项 */
ci->idx++;
}
}
