Consider the following example: * The logical block size is 4 KB. * The physical block size is 8 KB. * max_sectors equals (16 KB >> 9) sectors. * A non-aligned 4 KB and an aligned 64 KB bio are merged into a single non-aligned 68 KB bio. The current behavior is to split such a bio into (16 KB + 16 KB + 16 KB + 16 KB + 4 KB). The start of none of these five bio's is aligned to a physical block boundary. This patch ensures that such a bio is split into four aligned and one non-aligned bio instead of being split into five non-aligned bios. This improves performance because most block devices can handle aligned requests faster than non-aligned requests. Since the physical block size is larger than or equal to the logical block size, this patch preserves the guarantee that the returned value is a multiple of the logical block size.
Unified · Split
| @@ -132,16 +132,29 @@ static struct bio *blk_bio_write_same_split(struct request_queue *q, · 131 unmodified lines | |||
| 132 | return bio_split(bio, q->limits.max_write_same_sectors, GFP_NOIO, bs); | 132 | return bio_split(bio, q->limits.max_write_same_sectors, GFP_NOIO, bs); |
| 133 | } | 133 | } |
| 134 | 134 | ||
| 135 | /* | ||
| 136 | * Return the maximum number of sectors from the start of a bio that may be | ||
| 137 | * submitted as a single request to a block device. If enough sectors remain, | ||
| 138 | * align the end to the physical block size. Otherwise align the end to the | ||
| 139 | * logical block size. This approach minimizes the number of non-aligned | ||
| 140 | * requests that are submitted to a block device if the start of a bio is not | ||
| 141 | * aligned to a physical block boundary. | ||
| 142 | */ | ||
| 135 | static inline unsigned get_max_io_size(struct request_queue *q, | 143 | static inline unsigned get_max_io_size(struct request_queue *q, |
| 136 | struct bio *bio) | 144 | struct bio *bio) |
| 137 | { | 145 | { |
| 138 | unsigned sectors = blk_max_size_offset(q, bio->bi_iter.bi_sector); | 146 | unsigned sectors = blk_max_size_offset(q, bio->bi_iter.bi_sector); |
| 139 | unsigned mask = queue_logical_block_size(q) - 1; | 147 | unsigned max_sectors = sectors; |
| 148 | unsigned pbs = queue_physical_block_size(q) >> SECTOR_SHIFT; | ||
| 149 | unsigned lbs = queue_logical_block_size(q) >> SECTOR_SHIFT; | ||
| 150 | unsigned start_offset = bio->bi_iter.bi_sector & (pbs - 1); | ||
| 140 | 151 | ||
| 141 | /* aligned to logical block size */ | 152 | max_sectors += start_offset; |
| 142 | sectors &= ~(mask >> 9); | 153 | max_sectors &= ~(pbs - 1); |
| 154 | if (max_sectors > start_offset) | ||
| 155 | return max_sectors - start_offset; | ||
| 143 | 156 | ||
| 144 | return sectors; | 157 | return sectors & (lbs - 1); |
| 145 | } | 158 | } |
| 146 | 159 | ||
| 147 | static unsigned get_max_segment_size(const struct request_queue *q, | 160 | static unsigned get_max_segment_size(const struct request_queue *q, |