代码在block/blk_map.c中,代码来自v5.6.1
blk_rq_map_user代码:
int blk_rq_map_user(struct request_queue *q, struct request *rq,
struct rq_map_data *map_data, void __user *ubuf,
unsigned long len, gfp_t gfp_mask)
{
struct iovec iov;
struct iov_iter i;
int ret = import_single_range(rq_data_dir(rq), ubuf, len, &iov, &i);
if (unlikely(ret < 0))
return ret;
return blk_rq_map_user_iov(q, rq, map_data, &i, gfp_mask);
}
其中,import_single_range()根据ubuf和len组装了iter和i
主要是blk_rq_map_user_iov():
根据map_data和边界、对齐等设置确定要不要copy
/**
* blk_rq_map_user_iov - map user data to a request, for passthrough requests
* @q: request queue where request should be inserted
* @rq: request to map data to
* @map_data: pointer to the rq_map_data holding pages (if necessary)
* @iter: iovec iterator
* @gfp_mask: memory allocation flags
*
* Description:
* Data will be mapped directly for zero copy I/O, if possible. Otherwise
* a kernel bounce buffer is used.
*
* A matching blk_rq_unmap_user() must be issued at the end of I/O, while
* still in process context.
*/
int blk_rq_map_user_iov(struct request_queue *q, struct request *rq,
struct rq_map_data *map_data,
const struct iov_iter *iter, gfp_t gfp_mask)
{
bool copy = false;
unsigned long align = q->dma_pad_mask | queue_dma_alignment(q);
struct bio *bio = NULL;
struct iov_iter i;
int ret = -EINVAL;
if (!iter_is_iovec(iter))
goto fail;
if (map_data)
copy = true;
else if (blk_queue_may_bounce(q))
copy = true;
else if (iov_iter_alignment(iter) & align)
copy = true;
else if (queue_virt_boundary(q))
copy = queue_virt_boundary(q) & iov_iter_gap_alignment(iter);
i = *iter;
do {
if (copy)
ret = bio_copy_user_iov(rq, map_data, &i, gfp_mask);
else
ret = bio_map_user_iov(rq, &i, gfp_mask);
if (ret)
goto unmap_rq;
if (!bio)
bio = rq->bio;
} while (iov_iter_count(&i));
return 0;
unmap_rq:
blk_rq_unmap_user(bio);
fail:
rq->bio = NULL;
return ret;
}
如果要copy则进入bio_copy_user_iov:
在这个函数里,分配bio,bio的每一项填写一个page,如果map_data不为空则page从map_data得到,否则调用alloc_page()申请内存页,再调用bio_add_hw_page()将page写进bio。
如果是WRITE操作,则调用bio_copy_from_iter(bio,iter),将iter(最开始传进来的ubuffer相关的结构体)指向的内存拷贝进bio指向的page。
如果不是WRITE,则调用iov_iter_advance()
最后调用blk_rq_append_bio(rq,bio),将bio写进request
static int bio_copy_user_iov(struct request *rq, struct rq_map_data *map_data,
struct iov_iter *iter, gfp_t gfp_mask)
{
struct bio_map_data *bmd;
struct page *page;
struct bio *bio;
int i = 0, ret;
int nr_pages;
unsigned int len = iter->count;
unsigned int offset = map_data ? offset_in_page(map_data->offset) : 0;
bmd = bio_alloc_map_data(iter, gfp_mask);
if (!bmd)
return -ENOMEM;
/*
* We need to do a deep copy of the iov_iter including the iovecs.
* The caller provided iov might point to an on-stack or otherwise
* shortlived one.
*/
bmd->is_our_pages = !map_data;
bmd->is_null_mapped = (map_data && map_data->null_mapped);
nr_pages = bio_max_segs(DIV_ROUND_UP(offset + len, PAGE_SIZE));
ret = -ENOMEM;
bio = bio_kmalloc(gfp_mask, nr_pages);
if (!bio)
goto out_bmd;
bio->bi_opf |= req_op(rq);
if (map_data) {
nr_pages = 1 << map_data->page_order;
i = map_data->offset / PAGE_SIZE;
}
while (len) {
unsigned int bytes = PAGE_SIZE;
bytes -= offset;
if (bytes > len)
bytes = len;
if (map_data) {
if (i == map_data->nr_entries * nr_pages) {
ret = -ENOMEM;
goto cleanup;
}
page = map_data->pages[i / nr_pages];
page += (i % nr_pages);
i++;
} else {
page = alloc_page(GFP_NOIO | gfp_mask);
if (!page) {
ret = -ENOMEM;
goto cleanup;
}
}
if (bio_add_pc_page(rq->q, bio, page, bytes, offset) < bytes) {
if (!map_data)
__free_page(page);
break;
}
len -= bytes;
offset = 0;
}
if (map_data)
map_data->offset += bio->bi_iter.bi_size;
/*
* success
*/
if ((iov_iter_rw(iter) == WRITE &&
(!map_data || !map_data->null_mapped)) ||
(map_data && map_data->from_user)) {
ret = bio_copy_from_iter(bio, iter);
if (ret)
goto cleanup;
} else {
if (bmd->is_our_pages)
zero_fill_bio(bio);
iov_iter_advance(iter, bio->bi_iter.bi_size);
}
bio->bi_private = bmd;
ret = blk_rq_append_bio(rq, bio);
if (ret)
goto cleanup;
return 0;
cleanup:
if (!map_data)
bio_free_pages(bio);
bio_put(bio);
out_bmd:
kfree(bmd);
return ret;
}
如果不要copy,则进入bio_map_user_iov():
这个函数里,从iter得到page,将page信息写进bio,最后调用blk_rq_append_bio(rq,bio)将bio写进request
static int bio_map_user_iov(struct request *rq, struct iov_iter *iter,
gfp_t gfp_mask)
{
unsigned int max_sectors = queue_max_hw_sectors(rq->q);
struct bio *bio;
int ret;
int j;
if (!iov_iter_count(iter))
return -EINVAL;
bio = bio_kmalloc(gfp_mask, iov_iter_npages(iter, BIO_MAX_VECS));
if (!bio)
return -ENOMEM;
bio->bi_opf |= req_op(rq);
while (iov_iter_count(iter)) {
struct page **pages;
ssize_t bytes;
size_t offs, added = 0;
int npages;
bytes = iov_iter_get_pages_alloc(iter, &pages, LONG_MAX, &offs);
if (unlikely(bytes <= 0)) {
ret = bytes ? bytes : -EFAULT;
goto out_unmap;
}
npages = DIV_ROUND_UP(offs + bytes, PAGE_SIZE);
if (unlikely(offs & queue_dma_alignment(rq->q))) {
ret = -EINVAL;
j = 0;
} else {
for (j = 0; j < npages; j++) {
struct page *page = pages[j];
unsigned int n = PAGE_SIZE - offs;
bool same_page = false;
if (n > bytes)
n = bytes;
if (!bio_add_hw_page(rq->q, bio, page, n, offs,
max_sectors, &same_page)) {
if (same_page)
put_page(page);
break;
}
added += n;
bytes -= n;
offs = 0;
}
iov_iter_advance(iter, added);
}
/*
* release the pages we didn't map into the bio, if any
*/
while (j < npages)
put_page(pages[j++]);
kvfree(pages);
/* couldn't stuff something into bio? */
if (bytes)
break;
}
ret = blk_rq_append_bio(rq, bio);
if (ret)
goto out_unmap;
return 0;
out_unmap:
bio_release_pages(bio, false);
bio_put(bio);
return ret;
}
blk_rq_map_kern函数比较简单,和blk_rq_map_user差不多:
/**
* blk_rq_map_kern - map kernel data to a request, for passthrough requests
* @q: request queue where request should be inserted
* @rq: request to fill
* @kbuf: the kernel buffer
* @len: length of user data
* @gfp_mask: memory allocation flags
*
* Description:
* Data will be mapped directly if possible. Otherwise a bounce
* buffer is used. Can be called multiple times to append multiple
* buffers.
*/
int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
unsigned int len, gfp_t gfp_mask)
{
int reading = rq_data_dir(rq) == READ;
unsigned long addr = (unsigned long) kbuf;
struct bio *bio;
int ret;
if (len > (queue_max_hw_sectors(q) << 9))
return -EINVAL;
if (!len || !kbuf)
return -EINVAL;
if (!blk_rq_aligned(q, addr, len) || object_is_on_stack(kbuf) ||
blk_queue_may_bounce(q))
bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading);
else
bio = bio_map_kern(q, kbuf, len, gfp_mask);
if (IS_ERR(bio))
return PTR_ERR(bio);
bio->bi_opf &= ~REQ_OP_MASK;
bio->bi_opf |= req_op(rq);
ret = blk_rq_append_bio(rq, bio);
if (unlikely(ret))
bio_put(bio);
return ret;
}
也是分为了copy和map两个分支。