From: Mike Christie <michaelc@cs.wisc.edu>
To: axboe@suse.de, Linux SCSI list <linux-scsi@vger.kernel.org>
Subject: [PATCH 1/2] seperate max_sectors from max_hw_sectors
Date: Mon, 05 Dec 2005 02:37:06 -0600 [thread overview]
Message-ID: <1133771827.23619.17.camel@max> (raw)
- export __blk_put_request and blk_execute_rq_nowait
needed for async REQ_BLOCK_PC requests
- seperate max_hw_sectors and max_sectors for block/scsi_ioctl.c and
SG_IO bio.c helpers per Jens's last comments. Since block/scsi_ioctl.c SG_IO was
already testing against max_sectors and SCSI-ml was setting max_sectors and
max_hw_sectors to the same value this does not change any scsi SG_IO behavior. It only
prepares ll_rw_blk.c, scsi_ioctl.c and bio.c for when SCSI-ml begins to set
a valid max_hw_sectors for all LLDs. Today if a LLD does not set it
SCSI-ml sets it to a safe default and some LLDs set it to a artificial low
value to overcome memory and feedback issues.
Note: Since we now cap max_sectors to BLK_DEF_MAX_SECTORS, which is 1024,
drivers that used to call blk_queue_max_sectors with a large value of
max_sectors will now see the fs requests capped to BLK_DEF_MAX_SECTORS.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
index 5f52e30..d57fed9 100644
--- a/block/ll_rw_blk.c
+++ b/block/ll_rw_blk.c
@@ -241,7 +241,7 @@ void blk_queue_make_request(request_queu
q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
q->backing_dev_info.state = 0;
q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY;
- blk_queue_max_sectors(q, MAX_SECTORS);
+ blk_queue_max_sectors(q, SAFE_MAX_SECTORS);
blk_queue_hardsect_size(q, 512);
blk_queue_dma_alignment(q, 511);
blk_queue_congestion_threshold(q);
@@ -557,7 +557,12 @@ void blk_queue_max_sectors(request_queue
printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors);
}
- q->max_sectors = q->max_hw_sectors = max_sectors;
+ if (BLK_DEF_MAX_SECTORS > max_sectors)
+ q->max_hw_sectors = q->max_sectors = max_sectors;
+ else {
+ q->max_sectors = BLK_DEF_MAX_SECTORS;
+ q->max_hw_sectors = max_sectors;
+ }
}
EXPORT_SYMBOL(blk_queue_max_sectors);
@@ -659,8 +664,8 @@ EXPORT_SYMBOL(blk_queue_hardsect_size);
void blk_queue_stack_limits(request_queue_t *t, request_queue_t *b)
{
/* zero is "infinity" */
- t->max_sectors = t->max_hw_sectors =
- min_not_zero(t->max_sectors,b->max_sectors);
+ t->max_sectors = min_not_zero(t->max_sectors,b->max_sectors);
+ t->max_hw_sectors = min_not_zero(t->max_hw_sectors,b->max_hw_sectors);
t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments);
t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments);
@@ -1295,9 +1300,15 @@ static inline int ll_new_hw_segment(requ
static int ll_back_merge_fn(request_queue_t *q, struct request *req,
struct bio *bio)
{
+ unsigned short max_sectors;
int len;
- if (req->nr_sectors + bio_sectors(bio) > q->max_sectors) {
+ if (unlikely(blk_pc_request(req)))
+ max_sectors = q->max_hw_sectors;
+ else
+ max_sectors = q->max_sectors;
+
+ if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
req->flags |= REQ_NOMERGE;
if (req == q->last_merge)
q->last_merge = NULL;
@@ -1327,9 +1338,16 @@ static int ll_back_merge_fn(request_queu
static int ll_front_merge_fn(request_queue_t *q, struct request *req,
struct bio *bio)
{
+ unsigned short max_sectors;
int len;
- if (req->nr_sectors + bio_sectors(bio) > q->max_sectors) {
+ if (unlikely(blk_pc_request(req)))
+ max_sectors = q->max_hw_sectors;
+ else
+ max_sectors = q->max_sectors;
+
+
+ if (req->nr_sectors + bio_sectors(bio) > max_sectors) {
req->flags |= REQ_NOMERGE;
if (req == q->last_merge)
q->last_merge = NULL;
@@ -2146,7 +2164,7 @@ int blk_rq_map_user(request_queue_t *q,
struct bio *bio;
int reading;
- if (len > (q->max_sectors << 9))
+ if (len > (q->max_hw_sectors << 9))
return -EINVAL;
if (!len || !ubuf)
return -EINVAL;
@@ -2261,7 +2279,7 @@ int blk_rq_map_kern(request_queue_t *q,
{
struct bio *bio;
- if (len > (q->max_sectors << 9))
+ if (len > (q->max_hw_sectors << 9))
return -EINVAL;
if (!len || !kbuf)
return -EINVAL;
@@ -2308,6 +2326,8 @@ void blk_execute_rq_nowait(request_queue
generic_unplug_device(q);
}
+EXPORT_SYMBOL_GPL(blk_execute_rq_nowait);
+
/**
* blk_execute_rq - insert a request into queue for execution
* @q: queue to insert the request in
@@ -2446,7 +2466,7 @@ void disk_round_stats(struct gendisk *di
/*
* queue lock must be held
*/
-static void __blk_put_request(request_queue_t *q, struct request *req)
+void __blk_put_request(request_queue_t *q, struct request *req)
{
struct request_list *rl = req->rl;
@@ -2475,6 +2495,8 @@ static void __blk_put_request(request_qu
}
}
+EXPORT_SYMBOL_GPL(__blk_put_request);
+
void blk_put_request(struct request *req)
{
unsigned long flags;
diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c
index 382dea7..4e390df 100644
--- a/block/scsi_ioctl.c
+++ b/block/scsi_ioctl.c
@@ -233,7 +233,7 @@ static int sg_io(struct file *file, requ
if (verify_command(file, cmd))
return -EPERM;
- if (hdr->dxfer_len > (q->max_sectors << 9))
+ if (hdr->dxfer_len > (q->max_hw_sectors << 9))
return -EIO;
if (hdr->dxfer_len)
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index a6d3baa..a6f2dc6 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -638,7 +638,7 @@ int dm_split_args(int *argc, char ***arg
static void check_for_valid_limits(struct io_restrictions *rs)
{
if (!rs->max_sectors)
- rs->max_sectors = MAX_SECTORS;
+ rs->max_sectors = SAFE_MAX_SECTORS;
if (!rs->max_phys_segments)
rs->max_phys_segments = MAX_PHYS_SEGMENTS;
if (!rs->max_hw_segments)
diff --git a/fs/bio.c b/fs/bio.c
index 460554b..38d3e80 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -313,7 +313,8 @@ int bio_get_nr_vecs(struct block_device
}
static int __bio_add_page(request_queue_t *q, struct bio *bio, struct page
- *page, unsigned int len, unsigned int offset)
+ *page, unsigned int len, unsigned int offset,
+ unsigned short max_sectors)
{
int retried_segments = 0;
struct bio_vec *bvec;
@@ -327,7 +328,7 @@ static int __bio_add_page(request_queue_
if (bio->bi_vcnt >= bio->bi_max_vecs)
return 0;
- if (((bio->bi_size + len) >> 9) > q->max_sectors)
+ if (((bio->bi_size + len) >> 9) > max_sectors)
return 0;
/*
@@ -386,6 +387,25 @@ static int __bio_add_page(request_queue_
}
/**
+ * bio_add_pc_page - attempt to add page to bio
+ * @bio: destination bio
+ * @page: page to add
+ * @len: vec entry length
+ * @offset: vec entry offset
+ *
+ * Attempt to add a page to the bio_vec maplist. This can fail for a
+ * number of reasons, such as the bio being full or target block
+ * device limitations. The target block device must allow bio's
+ * smaller than PAGE_SIZE, so it is always possible to add a single
+ * page to an empty bio. This should only be used by REQ_PC bios.
+ */
+int bio_add_pc_page(request_queue_t *q, struct bio *bio, struct page *page,
+ unsigned int len, unsigned int offset)
+{
+ return __bio_add_page(q, bio, page, len, offset, q->max_hw_sectors);
+}
+
+/**
* bio_add_page - attempt to add page to bio
* @bio: destination bio
* @page: page to add
@@ -401,8 +421,8 @@ static int __bio_add_page(request_queue_
int bio_add_page(struct bio *bio, struct page *page, unsigned int len,
unsigned int offset)
{
- return __bio_add_page(bdev_get_queue(bio->bi_bdev), bio, page,
- len, offset);
+ struct request_queue *q = bdev_get_queue(bio->bi_bdev);
+ return __bio_add_page(q, bio, page, len, offset, q->max_sectors);
}
struct bio_map_data {
@@ -514,7 +534,7 @@ struct bio *bio_copy_user(request_queue_
break;
}
- if (__bio_add_page(q, bio, page, bytes, 0) < bytes) {
+ if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) {
ret = -EINVAL;
break;
}
@@ -628,7 +648,8 @@ static struct bio *__bio_map_user_iov(re
/*
* sorry...
*/
- if (__bio_add_page(q, bio, pages[j], bytes, offset) < bytes)
+ if (bio_add_pc_page(q, bio, pages[j], bytes, offset) <
+ bytes)
break;
len -= bytes;
@@ -801,8 +822,8 @@ static struct bio *__bio_map_kern(reques
if (bytes > len)
bytes = len;
- if (__bio_add_page(q, bio, virt_to_page(data), bytes,
- offset) < bytes)
+ if (bio_add_pc_page(q, bio, virt_to_page(data), bytes,
+ offset) < bytes)
break;
data += bytes;
@@ -1228,6 +1249,7 @@ EXPORT_SYMBOL(bio_clone);
EXPORT_SYMBOL(bio_phys_segments);
EXPORT_SYMBOL(bio_hw_segments);
EXPORT_SYMBOL(bio_add_page);
+EXPORT_SYMBOL(bio_add_pc_page);
EXPORT_SYMBOL(bio_get_nr_vecs);
EXPORT_SYMBOL(bio_map_user);
EXPORT_SYMBOL(bio_unmap_user);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 685fd37..b60ffe3 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -292,6 +292,8 @@ extern struct bio *bio_clone(struct bio
extern void bio_init(struct bio *);
extern int bio_add_page(struct bio *, struct page *, unsigned int,unsigned int);
+extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *,
+ unsigned int, unsigned int);
extern int bio_get_nr_vecs(struct block_device *);
extern struct bio *bio_map_user(struct request_queue *, struct block_device *,
unsigned long, unsigned int, int);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index a33a31e..856e76d 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -558,6 +558,7 @@ extern void blk_unregister_queue(struct
extern void register_disk(struct gendisk *dev);
extern void generic_make_request(struct bio *bio);
extern void blk_put_request(struct request *);
+extern void __blk_put_request(request_queue_t *, struct request *);
extern void blk_end_sync_rq(struct request *rq);
extern void blk_attempt_remerge(request_queue_t *, struct request *);
extern struct request *blk_get_request(request_queue_t *, int, gfp_t);
@@ -579,6 +580,10 @@ extern int blk_rq_map_kern(request_queue
extern int blk_rq_map_user_iov(request_queue_t *, struct request *, struct sg_iovec *, int);
extern int blk_execute_rq(request_queue_t *, struct gendisk *,
struct request *, int);
+extern void blk_execute_rq_nowait(request_queue_t *, struct gendisk *,
+ struct request *, int,
+ void (*done)(struct request *));
+
static inline request_queue_t *bdev_get_queue(struct block_device *bdev)
{
return bdev->bd_disk->queue;
@@ -696,7 +701,8 @@ extern int blkdev_issue_flush(struct blo
#define MAX_PHYS_SEGMENTS 128
#define MAX_HW_SEGMENTS 128
-#define MAX_SECTORS 255
+#define SAFE_MAX_SECTORS 255
+#define BLK_DEF_MAX_SECTORS 1024
#define MAX_SEGMENT_SIZE 65536
reply other threads:[~2005-12-05 8:37 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1133771827.23619.17.camel@max \
--to=michaelc@cs.wisc.edu \
--cc=axboe@suse.de \
--cc=linux-scsi@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.