mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-28 11:18:45 +07:00
1e42807918
blk_recalc_rq_segments() requires a request structure passed in, which we don't have from blk_recount_segments(). So the latter allocates one on the stack, using > 400 bytes of stack for that. This can cause us to spill over one page of stack from ext4 at least: 0) 4560 400 blk_recount_segments+0x43/0x62 1) 4160 32 bio_phys_segments+0x1c/0x24 2) 4128 32 blk_rq_bio_prep+0x2a/0xf9 3) 4096 32 init_request_from_bio+0xf9/0xfe 4) 4064 112 __make_request+0x33c/0x3f6 5) 3952 144 generic_make_request+0x2d1/0x321 6) 3808 64 submit_bio+0xb9/0xc3 7) 3744 48 submit_bh+0xea/0x10e 8) 3696 368 ext4_mb_init_cache+0x257/0xa6a [ext4] 9) 3328 288 ext4_mb_regular_allocator+0x421/0xcd9 [ext4] 10) 3040 160 ext4_mb_new_blocks+0x211/0x4b4 [ext4] 11) 2880 336 ext4_ext_get_blocks+0xb61/0xd45 [ext4] 12) 2544 96 ext4_get_blocks_wrap+0xf2/0x200 [ext4] 13) 2448 80 ext4_da_get_block_write+0x6e/0x16b [ext4] 14) 2368 352 mpage_da_map_blocks+0x7e/0x4b3 [ext4] 15) 2016 352 ext4_da_writepages+0x2ce/0x43c [ext4] 16) 1664 32 do_writepages+0x2d/0x3c 17) 1632 144 __writeback_single_inode+0x162/0x2cd 18) 1488 96 generic_sync_sb_inodes+0x1e3/0x32b 19) 1392 16 sync_sb_inodes+0xe/0x10 20) 1376 48 writeback_inodes+0x69/0xb3 21) 1328 208 balance_dirty_pages_ratelimited_nr+0x187/0x2f9 22) 1120 224 generic_file_buffered_write+0x1d4/0x2c4 23) 896 176 __generic_file_aio_write_nolock+0x35f/0x393 24) 720 80 generic_file_aio_write+0x6c/0xc8 25) 640 80 ext4_file_write+0xa9/0x137 [ext4] 26) 560 320 do_sync_write+0xf0/0x137 27) 240 48 vfs_write+0xb3/0x13c 28) 192 64 sys_write+0x4c/0x74 29) 128 128 system_call_fastpath+0x16/0x1b Split the segment counting out into a __blk_recalc_rq_segments() helper to avoid allocating an onstack request just for checking the physical segment count. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
436 lines
10 KiB
C
436 lines
10 KiB
C
/*
|
|
* Functions related to segment and merge handling
|
|
*/
|
|
#include <linux/kernel.h>
|
|
#include <linux/module.h>
|
|
#include <linux/bio.h>
|
|
#include <linux/blkdev.h>
|
|
#include <linux/scatterlist.h>
|
|
|
|
#include "blk.h"
|
|
|
|
void blk_recalc_rq_sectors(struct request *rq, int nsect)
|
|
{
|
|
if (blk_fs_request(rq) || blk_discard_rq(rq)) {
|
|
rq->hard_sector += nsect;
|
|
rq->hard_nr_sectors -= nsect;
|
|
|
|
/*
|
|
* Move the I/O submission pointers ahead if required.
|
|
*/
|
|
if ((rq->nr_sectors >= rq->hard_nr_sectors) &&
|
|
(rq->sector <= rq->hard_sector)) {
|
|
rq->sector = rq->hard_sector;
|
|
rq->nr_sectors = rq->hard_nr_sectors;
|
|
rq->hard_cur_sectors = bio_cur_sectors(rq->bio);
|
|
rq->current_nr_sectors = rq->hard_cur_sectors;
|
|
rq->buffer = bio_data(rq->bio);
|
|
}
|
|
|
|
/*
|
|
* if total number of sectors is less than the first segment
|
|
* size, something has gone terribly wrong
|
|
*/
|
|
if (rq->nr_sectors < rq->current_nr_sectors) {
|
|
printk(KERN_ERR "blk: request botched\n");
|
|
rq->nr_sectors = rq->current_nr_sectors;
|
|
}
|
|
}
|
|
}
|
|
|
|
static unsigned int __blk_recalc_rq_segments(struct request_queue *q,
|
|
struct bio *bio,
|
|
unsigned int *seg_size_ptr)
|
|
{
|
|
unsigned int phys_size;
|
|
struct bio_vec *bv, *bvprv = NULL;
|
|
int cluster, i, high, highprv = 1;
|
|
unsigned int seg_size, nr_phys_segs;
|
|
struct bio *fbio;
|
|
|
|
if (!bio)
|
|
return 0;
|
|
|
|
fbio = bio;
|
|
cluster = test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
|
|
seg_size = 0;
|
|
phys_size = nr_phys_segs = 0;
|
|
for_each_bio(bio) {
|
|
bio_for_each_segment(bv, bio, i) {
|
|
/*
|
|
* the trick here is making sure that a high page is
|
|
* never considered part of another segment, since that
|
|
* might change with the bounce page.
|
|
*/
|
|
high = page_to_pfn(bv->bv_page) > q->bounce_pfn;
|
|
if (high || highprv)
|
|
goto new_segment;
|
|
if (cluster) {
|
|
if (seg_size + bv->bv_len > q->max_segment_size)
|
|
goto new_segment;
|
|
if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv))
|
|
goto new_segment;
|
|
if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv))
|
|
goto new_segment;
|
|
|
|
seg_size += bv->bv_len;
|
|
bvprv = bv;
|
|
continue;
|
|
}
|
|
new_segment:
|
|
if (nr_phys_segs == 1 && seg_size >
|
|
fbio->bi_seg_front_size)
|
|
fbio->bi_seg_front_size = seg_size;
|
|
|
|
nr_phys_segs++;
|
|
bvprv = bv;
|
|
seg_size = bv->bv_len;
|
|
highprv = high;
|
|
}
|
|
}
|
|
|
|
if (seg_size_ptr)
|
|
*seg_size_ptr = seg_size;
|
|
|
|
return nr_phys_segs;
|
|
}
|
|
|
|
void blk_recalc_rq_segments(struct request *rq)
|
|
{
|
|
unsigned int seg_size = 0, phys_segs;
|
|
|
|
phys_segs = __blk_recalc_rq_segments(rq->q, rq->bio, &seg_size);
|
|
|
|
if (phys_segs == 1 && seg_size > rq->bio->bi_seg_front_size)
|
|
rq->bio->bi_seg_front_size = seg_size;
|
|
if (seg_size > rq->biotail->bi_seg_back_size)
|
|
rq->biotail->bi_seg_back_size = seg_size;
|
|
|
|
rq->nr_phys_segments = phys_segs;
|
|
}
|
|
|
|
void blk_recount_segments(struct request_queue *q, struct bio *bio)
|
|
{
|
|
struct bio *nxt = bio->bi_next;
|
|
|
|
bio->bi_next = NULL;
|
|
bio->bi_phys_segments = __blk_recalc_rq_segments(q, bio, NULL);
|
|
bio->bi_next = nxt;
|
|
bio->bi_flags |= (1 << BIO_SEG_VALID);
|
|
}
|
|
EXPORT_SYMBOL(blk_recount_segments);
|
|
|
|
static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio,
|
|
struct bio *nxt)
|
|
{
|
|
if (!test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags))
|
|
return 0;
|
|
|
|
if (bio->bi_seg_back_size + nxt->bi_seg_front_size >
|
|
q->max_segment_size)
|
|
return 0;
|
|
|
|
if (!bio_has_data(bio))
|
|
return 1;
|
|
|
|
if (!BIOVEC_PHYS_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)))
|
|
return 0;
|
|
|
|
/*
|
|
* bio and nxt are contiguous in memory; check if the queue allows
|
|
* these two to be merged into one
|
|
*/
|
|
if (BIO_SEG_BOUNDARY(q, bio, nxt))
|
|
return 1;
|
|
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* map a request to scatterlist, return number of sg entries setup. Caller
|
|
* must make sure sg can hold rq->nr_phys_segments entries
|
|
*/
|
|
int blk_rq_map_sg(struct request_queue *q, struct request *rq,
|
|
struct scatterlist *sglist)
|
|
{
|
|
struct bio_vec *bvec, *bvprv;
|
|
struct req_iterator iter;
|
|
struct scatterlist *sg;
|
|
int nsegs, cluster;
|
|
|
|
nsegs = 0;
|
|
cluster = test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
|
|
|
|
/*
|
|
* for each bio in rq
|
|
*/
|
|
bvprv = NULL;
|
|
sg = NULL;
|
|
rq_for_each_segment(bvec, rq, iter) {
|
|
int nbytes = bvec->bv_len;
|
|
|
|
if (bvprv && cluster) {
|
|
if (sg->length + nbytes > q->max_segment_size)
|
|
goto new_segment;
|
|
|
|
if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec))
|
|
goto new_segment;
|
|
if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec))
|
|
goto new_segment;
|
|
|
|
sg->length += nbytes;
|
|
} else {
|
|
new_segment:
|
|
if (!sg)
|
|
sg = sglist;
|
|
else {
|
|
/*
|
|
* If the driver previously mapped a shorter
|
|
* list, we could see a termination bit
|
|
* prematurely unless it fully inits the sg
|
|
* table on each mapping. We KNOW that there
|
|
* must be more entries here or the driver
|
|
* would be buggy, so force clear the
|
|
* termination bit to avoid doing a full
|
|
* sg_init_table() in drivers for each command.
|
|
*/
|
|
sg->page_link &= ~0x02;
|
|
sg = sg_next(sg);
|
|
}
|
|
|
|
sg_set_page(sg, bvec->bv_page, nbytes, bvec->bv_offset);
|
|
nsegs++;
|
|
}
|
|
bvprv = bvec;
|
|
} /* segments in rq */
|
|
|
|
|
|
if (unlikely(rq->cmd_flags & REQ_COPY_USER) &&
|
|
(rq->data_len & q->dma_pad_mask)) {
|
|
unsigned int pad_len = (q->dma_pad_mask & ~rq->data_len) + 1;
|
|
|
|
sg->length += pad_len;
|
|
rq->extra_len += pad_len;
|
|
}
|
|
|
|
if (q->dma_drain_size && q->dma_drain_needed(rq)) {
|
|
if (rq->cmd_flags & REQ_RW)
|
|
memset(q->dma_drain_buffer, 0, q->dma_drain_size);
|
|
|
|
sg->page_link &= ~0x02;
|
|
sg = sg_next(sg);
|
|
sg_set_page(sg, virt_to_page(q->dma_drain_buffer),
|
|
q->dma_drain_size,
|
|
((unsigned long)q->dma_drain_buffer) &
|
|
(PAGE_SIZE - 1));
|
|
nsegs++;
|
|
rq->extra_len += q->dma_drain_size;
|
|
}
|
|
|
|
if (sg)
|
|
sg_mark_end(sg);
|
|
|
|
return nsegs;
|
|
}
|
|
EXPORT_SYMBOL(blk_rq_map_sg);
|
|
|
|
static inline int ll_new_hw_segment(struct request_queue *q,
|
|
struct request *req,
|
|
struct bio *bio)
|
|
{
|
|
int nr_phys_segs = bio_phys_segments(q, bio);
|
|
|
|
if (req->nr_phys_segments + nr_phys_segs > q->max_hw_segments
|
|
|| req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
|
|
req->cmd_flags |= REQ_NOMERGE;
|
|
if (req == q->last_merge)
|
|
q->last_merge = NULL;
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* This will form the start of a new hw segment. Bump both
|
|
* counters.
|
|
*/
|
|
req->nr_phys_segments += nr_phys_segs;
|
|
return 1;
|
|
}
|
|
|
|
int ll_back_merge_fn(struct request_queue *q, struct request *req,
|
|
struct bio *bio)
|
|
{
|
|
unsigned short 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->cmd_flags |= REQ_NOMERGE;
|
|
if (req == q->last_merge)
|
|
q->last_merge = NULL;
|
|
return 0;
|
|
}
|
|
if (!bio_flagged(req->biotail, BIO_SEG_VALID))
|
|
blk_recount_segments(q, req->biotail);
|
|
if (!bio_flagged(bio, BIO_SEG_VALID))
|
|
blk_recount_segments(q, bio);
|
|
|
|
return ll_new_hw_segment(q, req, bio);
|
|
}
|
|
|
|
int ll_front_merge_fn(struct request_queue *q, struct request *req,
|
|
struct bio *bio)
|
|
{
|
|
unsigned short 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->cmd_flags |= REQ_NOMERGE;
|
|
if (req == q->last_merge)
|
|
q->last_merge = NULL;
|
|
return 0;
|
|
}
|
|
if (!bio_flagged(bio, BIO_SEG_VALID))
|
|
blk_recount_segments(q, bio);
|
|
if (!bio_flagged(req->bio, BIO_SEG_VALID))
|
|
blk_recount_segments(q, req->bio);
|
|
|
|
return ll_new_hw_segment(q, req, bio);
|
|
}
|
|
|
|
static int ll_merge_requests_fn(struct request_queue *q, struct request *req,
|
|
struct request *next)
|
|
{
|
|
int total_phys_segments;
|
|
unsigned int seg_size =
|
|
req->biotail->bi_seg_back_size + next->bio->bi_seg_front_size;
|
|
|
|
/*
|
|
* First check if the either of the requests are re-queued
|
|
* requests. Can't merge them if they are.
|
|
*/
|
|
if (req->special || next->special)
|
|
return 0;
|
|
|
|
/*
|
|
* Will it become too large?
|
|
*/
|
|
if ((req->nr_sectors + next->nr_sectors) > q->max_sectors)
|
|
return 0;
|
|
|
|
total_phys_segments = req->nr_phys_segments + next->nr_phys_segments;
|
|
if (blk_phys_contig_segment(q, req->biotail, next->bio)) {
|
|
if (req->nr_phys_segments == 1)
|
|
req->bio->bi_seg_front_size = seg_size;
|
|
if (next->nr_phys_segments == 1)
|
|
next->biotail->bi_seg_back_size = seg_size;
|
|
total_phys_segments--;
|
|
}
|
|
|
|
if (total_phys_segments > q->max_phys_segments)
|
|
return 0;
|
|
|
|
if (total_phys_segments > q->max_hw_segments)
|
|
return 0;
|
|
|
|
/* Merge is OK... */
|
|
req->nr_phys_segments = total_phys_segments;
|
|
return 1;
|
|
}
|
|
|
|
/*
|
|
* Has to be called with the request spinlock acquired
|
|
*/
|
|
static int attempt_merge(struct request_queue *q, struct request *req,
|
|
struct request *next)
|
|
{
|
|
if (!rq_mergeable(req) || !rq_mergeable(next))
|
|
return 0;
|
|
|
|
/*
|
|
* not contiguous
|
|
*/
|
|
if (req->sector + req->nr_sectors != next->sector)
|
|
return 0;
|
|
|
|
if (rq_data_dir(req) != rq_data_dir(next)
|
|
|| req->rq_disk != next->rq_disk
|
|
|| next->special)
|
|
return 0;
|
|
|
|
if (blk_integrity_rq(req) != blk_integrity_rq(next))
|
|
return 0;
|
|
|
|
/*
|
|
* If we are allowed to merge, then append bio list
|
|
* from next to rq and release next. merge_requests_fn
|
|
* will have updated segment counts, update sector
|
|
* counts here.
|
|
*/
|
|
if (!ll_merge_requests_fn(q, req, next))
|
|
return 0;
|
|
|
|
/*
|
|
* At this point we have either done a back merge
|
|
* or front merge. We need the smaller start_time of
|
|
* the merged requests to be the current request
|
|
* for accounting purposes.
|
|
*/
|
|
if (time_after(req->start_time, next->start_time))
|
|
req->start_time = next->start_time;
|
|
|
|
req->biotail->bi_next = next->bio;
|
|
req->biotail = next->biotail;
|
|
|
|
req->nr_sectors = req->hard_nr_sectors += next->hard_nr_sectors;
|
|
|
|
elv_merge_requests(q, req, next);
|
|
|
|
if (req->rq_disk) {
|
|
struct hd_struct *part;
|
|
int cpu;
|
|
|
|
cpu = part_stat_lock();
|
|
part = disk_map_sector_rcu(req->rq_disk, req->sector);
|
|
|
|
part_round_stats(cpu, part);
|
|
part_dec_in_flight(part);
|
|
|
|
part_stat_unlock();
|
|
}
|
|
|
|
req->ioprio = ioprio_best(req->ioprio, next->ioprio);
|
|
if (blk_rq_cpu_valid(next))
|
|
req->cpu = next->cpu;
|
|
|
|
__blk_put_request(q, next);
|
|
return 1;
|
|
}
|
|
|
|
int attempt_back_merge(struct request_queue *q, struct request *rq)
|
|
{
|
|
struct request *next = elv_latter_request(q, rq);
|
|
|
|
if (next)
|
|
return attempt_merge(q, rq, next);
|
|
|
|
return 0;
|
|
}
|
|
|
|
int attempt_front_merge(struct request_queue *q, struct request *rq)
|
|
{
|
|
struct request *prev = elv_former_request(q, rq);
|
|
|
|
if (prev)
|
|
return attempt_merge(q, prev, rq);
|
|
|
|
return 0;
|
|
}
|