^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) * Copyright (C) 2003 Sistina Software
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Copyright (C) 2006 Red Hat GmbH
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * This file is released under the GPL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include "dm-core.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/device-mapper.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/bio.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/completion.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/mempool.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/dm-io.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #define DM_MSG_PREFIX "io"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #define DM_IO_MAX_REGIONS BITS_PER_LONG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) struct dm_io_client {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) mempool_t pool;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) struct bio_set bios;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * Aligning 'struct io' reduces the number of bits required to store
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) * its address. Refer to store_io_and_region_in_bio() below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) struct io {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) unsigned long error_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) atomic_t count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) struct dm_io_client *client;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) io_notify_fn callback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) void *context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) void *vma_invalidate_address;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) unsigned long vma_invalidate_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) } __attribute__((aligned(DM_IO_MAX_REGIONS)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) static struct kmem_cache *_dm_io_cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * Create a client with mempool and bioset.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) struct dm_io_client *dm_io_client_create(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) struct dm_io_client *client;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) unsigned min_ios = dm_get_reserved_bio_based_ios();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) client = kzalloc(sizeof(*client), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) if (!client)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) ret = mempool_init_slab_pool(&client->pool, min_ios, _dm_io_cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) ret = bioset_init(&client->bios, min_ios, 0, BIOSET_NEED_BVECS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) return client;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) bad:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) mempool_exit(&client->pool);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) kfree(client);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) return ERR_PTR(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) EXPORT_SYMBOL(dm_io_client_create);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) void dm_io_client_destroy(struct dm_io_client *client)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) mempool_exit(&client->pool);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) bioset_exit(&client->bios);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) kfree(client);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) EXPORT_SYMBOL(dm_io_client_destroy);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) /*-----------------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) * We need to keep track of which region a bio is doing io for.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) * To avoid a memory allocation to store just 5 or 6 bits, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) * ensure the 'struct io' pointer is aligned so enough low bits are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) * always zero and then combine it with the region number directly in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) * bi_private.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) *---------------------------------------------------------------*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) static void store_io_and_region_in_bio(struct bio *bio, struct io *io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) unsigned region)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) if (unlikely(!IS_ALIGNED((unsigned long)io, DM_IO_MAX_REGIONS))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) DMCRIT("Unaligned struct io pointer %p", io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) bio->bi_private = (void *)((unsigned long)io | region);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) static void retrieve_io_and_region_from_bio(struct bio *bio, struct io **io,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) unsigned *region)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) unsigned long val = (unsigned long)bio->bi_private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) *io = (void *)(val & -(unsigned long)DM_IO_MAX_REGIONS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) *region = val & (DM_IO_MAX_REGIONS - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) /*-----------------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) * We need an io object to keep track of the number of bios that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) * have been dispatched for a particular io.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) *---------------------------------------------------------------*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) static void complete_io(struct io *io)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) unsigned long error_bits = io->error_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) io_notify_fn fn = io->callback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) void *context = io->context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) if (io->vma_invalidate_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) invalidate_kernel_vmap_range(io->vma_invalidate_address,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) io->vma_invalidate_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) mempool_free(io, &io->client->pool);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) fn(error_bits, context);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) static void dec_count(struct io *io, unsigned int region, blk_status_t error)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) if (error)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) set_bit(region, &io->error_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) if (atomic_dec_and_test(&io->count))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) complete_io(io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) static void endio(struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) struct io *io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) unsigned region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) blk_status_t error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) if (bio->bi_status && bio_data_dir(bio) == READ)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) zero_fill_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) * The bio destructor in bio_put() may use the io object.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) retrieve_io_and_region_from_bio(bio, &io, ®ion);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) error = bio->bi_status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) bio_put(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) dec_count(io, region, error);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) /*-----------------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) * These little objects provide an abstraction for getting a new
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) * destination page for io.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) *---------------------------------------------------------------*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) struct dpages {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) void (*get_page)(struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) struct page **p, unsigned long *len, unsigned *offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) void (*next_page)(struct dpages *dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) union {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) unsigned context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) struct bvec_iter context_bi;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) void *context_ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) void *vma_invalidate_address;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) unsigned long vma_invalidate_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) * Functions for getting the pages from a list.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) static void list_get_page(struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) struct page **p, unsigned long *len, unsigned *offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) unsigned o = dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) struct page_list *pl = (struct page_list *) dp->context_ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) *p = pl->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) *len = PAGE_SIZE - o;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) *offset = o;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) static void list_next_page(struct dpages *dp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) struct page_list *pl = (struct page_list *) dp->context_ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) dp->context_ptr = pl->next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) dp->context_u = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) static void list_dp_init(struct dpages *dp, struct page_list *pl, unsigned offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) dp->get_page = list_get_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) dp->next_page = list_next_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) dp->context_u = offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) dp->context_ptr = pl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) * Functions for getting the pages from a bvec.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) static void bio_get_page(struct dpages *dp, struct page **p,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) unsigned long *len, unsigned *offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) struct bio_vec bvec = bvec_iter_bvec((struct bio_vec *)dp->context_ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) dp->context_bi);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) *p = bvec.bv_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) *len = bvec.bv_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) *offset = bvec.bv_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) /* avoid figuring it out again in bio_next_page() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) dp->context_bi.bi_sector = (sector_t)bvec.bv_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) static void bio_next_page(struct dpages *dp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) unsigned int len = (unsigned int)dp->context_bi.bi_sector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) bvec_iter_advance((struct bio_vec *)dp->context_ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) &dp->context_bi, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) static void bio_dp_init(struct dpages *dp, struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) dp->get_page = bio_get_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) dp->next_page = bio_next_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) * We just use bvec iterator to retrieve pages, so it is ok to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) * access the bvec table directly here
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) dp->context_ptr = bio->bi_io_vec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) dp->context_bi = bio->bi_iter;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) * Functions for getting the pages from a VMA.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) static void vm_get_page(struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) struct page **p, unsigned long *len, unsigned *offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) *p = vmalloc_to_page(dp->context_ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) *offset = dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) *len = PAGE_SIZE - dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) static void vm_next_page(struct dpages *dp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) dp->context_ptr += PAGE_SIZE - dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) dp->context_u = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) static void vm_dp_init(struct dpages *dp, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) dp->get_page = vm_get_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) dp->next_page = vm_next_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) dp->context_u = offset_in_page(data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) dp->context_ptr = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) * Functions for getting the pages from kernel memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) static void km_get_page(struct dpages *dp, struct page **p, unsigned long *len,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) unsigned *offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) *p = virt_to_page(dp->context_ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) *offset = dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) *len = PAGE_SIZE - dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) static void km_next_page(struct dpages *dp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) dp->context_ptr += PAGE_SIZE - dp->context_u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) dp->context_u = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) static void km_dp_init(struct dpages *dp, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) dp->get_page = km_get_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) dp->next_page = km_next_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) dp->context_u = offset_in_page(data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) dp->context_ptr = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) /*-----------------------------------------------------------------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) * IO routines that accept a list of pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) *---------------------------------------------------------------*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) static void do_region(int op, int op_flags, unsigned region,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) struct dm_io_region *where, struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) struct io *io)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) unsigned long len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) unsigned offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) unsigned num_bvecs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) sector_t remaining = where->count;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) struct request_queue *q = bdev_get_queue(where->bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) unsigned short logical_block_size = queue_logical_block_size(q);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) sector_t num_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) unsigned int special_cmd_max_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) * Reject unsupported discard and write same requests.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) if (op == REQ_OP_DISCARD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) special_cmd_max_sectors = q->limits.max_discard_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) else if (op == REQ_OP_WRITE_ZEROES)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) special_cmd_max_sectors = q->limits.max_write_zeroes_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) else if (op == REQ_OP_WRITE_SAME)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) special_cmd_max_sectors = q->limits.max_write_same_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) if ((op == REQ_OP_DISCARD || op == REQ_OP_WRITE_ZEROES ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) op == REQ_OP_WRITE_SAME) && special_cmd_max_sectors == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) atomic_inc(&io->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) dec_count(io, region, BLK_STS_NOTSUPP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) * where->count may be zero if op holds a flush and we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) * send a zero-sized flush.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) * Allocate a suitably sized-bio.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) switch (op) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) case REQ_OP_DISCARD:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) case REQ_OP_WRITE_ZEROES:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) num_bvecs = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) case REQ_OP_WRITE_SAME:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) num_bvecs = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) num_bvecs = min_t(int, BIO_MAX_PAGES,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) dm_sector_div_up(remaining, (PAGE_SIZE >> SECTOR_SHIFT)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) bio = bio_alloc_bioset(GFP_NOIO, num_bvecs, &io->client->bios);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) bio->bi_iter.bi_sector = where->sector + (where->count - remaining);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) bio_set_dev(bio, where->bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) bio->bi_end_io = endio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) bio_set_op_attrs(bio, op, op_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) store_io_and_region_in_bio(bio, io, region);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) if (op == REQ_OP_DISCARD || op == REQ_OP_WRITE_ZEROES) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) num_sectors = min_t(sector_t, special_cmd_max_sectors, remaining);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) bio->bi_iter.bi_size = num_sectors << SECTOR_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) remaining -= num_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) } else if (op == REQ_OP_WRITE_SAME) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) * WRITE SAME only uses a single page.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) dp->get_page(dp, &page, &len, &offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) bio_add_page(bio, page, logical_block_size, offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) num_sectors = min_t(sector_t, special_cmd_max_sectors, remaining);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) bio->bi_iter.bi_size = num_sectors << SECTOR_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) remaining -= num_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) dp->next_page(dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) } else while (remaining) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) * Try and add as many pages as possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) dp->get_page(dp, &page, &len, &offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) len = min(len, to_bytes(remaining));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) if (!bio_add_page(bio, page, len, offset))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) remaining -= to_sector(len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) dp->next_page(dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) atomic_inc(&io->count);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) submit_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) } while (remaining);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) static void dispatch_io(int op, int op_flags, unsigned int num_regions,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) struct dm_io_region *where, struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) struct io *io, int sync)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) struct dpages old_pages = *dp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) BUG_ON(num_regions > DM_IO_MAX_REGIONS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) if (sync)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) op_flags |= REQ_SYNC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) * For multiple regions we need to be careful to rewind
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) * the dp object for each call to do_region.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) for (i = 0; i < num_regions; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) *dp = old_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) if (where[i].count || (op_flags & REQ_PREFLUSH))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) do_region(op, op_flags, i, where + i, dp, io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * Drop the extra reference that we were holding to avoid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) * the io being completed too early.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) dec_count(io, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) struct sync_io {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) unsigned long error_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) struct completion wait;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) static void sync_io_complete(unsigned long error, void *context)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) struct sync_io *sio = context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) sio->error_bits = error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) complete(&sio->wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) static int sync_io(struct dm_io_client *client, unsigned int num_regions,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) struct dm_io_region *where, int op, int op_flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) struct dpages *dp, unsigned long *error_bits)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) struct io *io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) struct sync_io sio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) if (num_regions > 1 && !op_is_write(op)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) init_completion(&sio.wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) io = mempool_alloc(&client->pool, GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) io->error_bits = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) atomic_set(&io->count, 1); /* see dispatch_io() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) io->client = client;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) io->callback = sync_io_complete;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) io->context = &sio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) io->vma_invalidate_address = dp->vma_invalidate_address;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) io->vma_invalidate_size = dp->vma_invalidate_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) dispatch_io(op, op_flags, num_regions, where, dp, io, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) wait_for_completion_io(&sio.wait);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) if (error_bits)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) *error_bits = sio.error_bits;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) return sio.error_bits ? -EIO : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) static int async_io(struct dm_io_client *client, unsigned int num_regions,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) struct dm_io_region *where, int op, int op_flags,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) struct dpages *dp, io_notify_fn fn, void *context)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) struct io *io;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) if (num_regions > 1 && !op_is_write(op)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) fn(1, context);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) io = mempool_alloc(&client->pool, GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) io->error_bits = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) atomic_set(&io->count, 1); /* see dispatch_io() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) io->client = client;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) io->callback = fn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) io->context = context;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) io->vma_invalidate_address = dp->vma_invalidate_address;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) io->vma_invalidate_size = dp->vma_invalidate_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) dispatch_io(op, op_flags, num_regions, where, dp, io, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) static int dp_init(struct dm_io_request *io_req, struct dpages *dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) unsigned long size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) /* Set up dpages based on memory type */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) dp->vma_invalidate_address = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) dp->vma_invalidate_size = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) switch (io_req->mem.type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) case DM_IO_PAGE_LIST:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) list_dp_init(dp, io_req->mem.ptr.pl, io_req->mem.offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) case DM_IO_BIO:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) bio_dp_init(dp, io_req->mem.ptr.bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) case DM_IO_VMA:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) flush_kernel_vmap_range(io_req->mem.ptr.vma, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) if (io_req->bi_op == REQ_OP_READ) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) dp->vma_invalidate_address = io_req->mem.ptr.vma;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) dp->vma_invalidate_size = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) vm_dp_init(dp, io_req->mem.ptr.vma);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) case DM_IO_KMEM:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) km_dp_init(dp, io_req->mem.ptr.addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) * New collapsed (a)synchronous interface.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) * If the IO is asynchronous (i.e. it has notify.fn), you must either unplug
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) * the queue with blk_unplug() some time later or set REQ_SYNC in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) * io_req->bi_opf. If you fail to do one of these, the IO will be submitted to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) * the disk after q->unplug_delay, which defaults to 3ms in blk-settings.c.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) int dm_io(struct dm_io_request *io_req, unsigned num_regions,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) struct dm_io_region *where, unsigned long *sync_error_bits)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) struct dpages dp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) r = dp_init(io_req, &dp, (unsigned long)where->count << SECTOR_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) if (!io_req->notify.fn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) return sync_io(io_req->client, num_regions, where,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) io_req->bi_op, io_req->bi_op_flags, &dp,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) sync_error_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) return async_io(io_req->client, num_regions, where, io_req->bi_op,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) io_req->bi_op_flags, &dp, io_req->notify.fn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) io_req->notify.context);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) EXPORT_SYMBOL(dm_io);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) int __init dm_io_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) _dm_io_cache = KMEM_CACHE(io, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) if (!_dm_io_cache)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) void dm_io_exit(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) kmem_cache_destroy(_dm_io_cache);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) _dm_io_cache = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) }