|
@@ -1927,253 +1927,6 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous)
|
|
}
|
|
}
|
|
|
|
|
|
|
|
|
|
-
|
|
|
|
-/*
|
|
|
|
- * Copy data between a page in the stripe cache, and one or more bion
|
|
|
|
- * The page could align with the middle of the bio, or there could be
|
|
|
|
- * several bion, each with several bio_vecs, which cover part of the page
|
|
|
|
- * Multiple bion are linked together on bi_next. There may be extras
|
|
|
|
- * at the end of this list. We ignore them.
|
|
|
|
- */
|
|
|
|
-static void copy_data(int frombio, struct bio *bio,
|
|
|
|
- struct page *page,
|
|
|
|
- sector_t sector)
|
|
|
|
-{
|
|
|
|
- char *pa = page_address(page);
|
|
|
|
- struct bio_vec *bvl;
|
|
|
|
- int i;
|
|
|
|
- int page_offset;
|
|
|
|
-
|
|
|
|
- if (bio->bi_sector >= sector)
|
|
|
|
- page_offset = (signed)(bio->bi_sector - sector) * 512;
|
|
|
|
- else
|
|
|
|
- page_offset = (signed)(sector - bio->bi_sector) * -512;
|
|
|
|
- bio_for_each_segment(bvl, bio, i) {
|
|
|
|
- int len = bio_iovec_idx(bio,i)->bv_len;
|
|
|
|
- int clen;
|
|
|
|
- int b_offset = 0;
|
|
|
|
-
|
|
|
|
- if (page_offset < 0) {
|
|
|
|
- b_offset = -page_offset;
|
|
|
|
- page_offset += b_offset;
|
|
|
|
- len -= b_offset;
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- if (len > 0 && page_offset + len > STRIPE_SIZE)
|
|
|
|
- clen = STRIPE_SIZE - page_offset;
|
|
|
|
- else clen = len;
|
|
|
|
-
|
|
|
|
- if (clen > 0) {
|
|
|
|
- char *ba = __bio_kmap_atomic(bio, i, KM_USER0);
|
|
|
|
- if (frombio)
|
|
|
|
- memcpy(pa+page_offset, ba+b_offset, clen);
|
|
|
|
- else
|
|
|
|
- memcpy(ba+b_offset, pa+page_offset, clen);
|
|
|
|
- __bio_kunmap_atomic(ba, KM_USER0);
|
|
|
|
- }
|
|
|
|
- if (clen < len) /* hit end of page */
|
|
|
|
- break;
|
|
|
|
- page_offset += len;
|
|
|
|
- }
|
|
|
|
-}
|
|
|
|
-
|
|
|
|
-#define check_xor() do { \
|
|
|
|
- if (count == MAX_XOR_BLOCKS) { \
|
|
|
|
- xor_blocks(count, STRIPE_SIZE, dest, ptr);\
|
|
|
|
- count = 0; \
|
|
|
|
- } \
|
|
|
|
- } while(0)
|
|
|
|
-
|
|
|
|
-static void compute_parity6(struct stripe_head *sh, int method)
|
|
|
|
-{
|
|
|
|
- raid5_conf_t *conf = sh->raid_conf;
|
|
|
|
- int i, pd_idx, qd_idx, d0_idx, disks = sh->disks, count;
|
|
|
|
- int syndrome_disks = sh->ddf_layout ? disks : (disks - 2);
|
|
|
|
- struct bio *chosen;
|
|
|
|
- /**** FIX THIS: This could be very bad if disks is close to 256 ****/
|
|
|
|
- void *ptrs[syndrome_disks+2];
|
|
|
|
-
|
|
|
|
- pd_idx = sh->pd_idx;
|
|
|
|
- qd_idx = sh->qd_idx;
|
|
|
|
- d0_idx = raid6_d0(sh);
|
|
|
|
-
|
|
|
|
- pr_debug("compute_parity, stripe %llu, method %d\n",
|
|
|
|
- (unsigned long long)sh->sector, method);
|
|
|
|
-
|
|
|
|
- switch(method) {
|
|
|
|
- case READ_MODIFY_WRITE:
|
|
|
|
- BUG(); /* READ_MODIFY_WRITE N/A for RAID-6 */
|
|
|
|
- case RECONSTRUCT_WRITE:
|
|
|
|
- for (i= disks; i-- ;)
|
|
|
|
- if ( i != pd_idx && i != qd_idx && sh->dev[i].towrite ) {
|
|
|
|
- chosen = sh->dev[i].towrite;
|
|
|
|
- sh->dev[i].towrite = NULL;
|
|
|
|
-
|
|
|
|
- if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
|
|
|
|
- wake_up(&conf->wait_for_overlap);
|
|
|
|
-
|
|
|
|
- BUG_ON(sh->dev[i].written);
|
|
|
|
- sh->dev[i].written = chosen;
|
|
|
|
- }
|
|
|
|
- break;
|
|
|
|
- case CHECK_PARITY:
|
|
|
|
- BUG(); /* Not implemented yet */
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- for (i = disks; i--;)
|
|
|
|
- if (sh->dev[i].written) {
|
|
|
|
- sector_t sector = sh->dev[i].sector;
|
|
|
|
- struct bio *wbi = sh->dev[i].written;
|
|
|
|
- while (wbi && wbi->bi_sector < sector + STRIPE_SECTORS) {
|
|
|
|
- copy_data(1, wbi, sh->dev[i].page, sector);
|
|
|
|
- wbi = r5_next_bio(wbi, sector);
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- set_bit(R5_LOCKED, &sh->dev[i].flags);
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[i].flags);
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- /* Note that unlike RAID-5, the ordering of the disks matters greatly.*/
|
|
|
|
-
|
|
|
|
- for (i = 0; i < disks; i++)
|
|
|
|
- ptrs[i] = (void *)raid6_empty_zero_page;
|
|
|
|
-
|
|
|
|
- count = 0;
|
|
|
|
- i = d0_idx;
|
|
|
|
- do {
|
|
|
|
- int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
|
|
|
|
-
|
|
|
|
- ptrs[slot] = page_address(sh->dev[i].page);
|
|
|
|
- if (slot < syndrome_disks &&
|
|
|
|
- !test_bit(R5_UPTODATE, &sh->dev[i].flags)) {
|
|
|
|
- printk(KERN_ERR "block %d/%d not uptodate "
|
|
|
|
- "on parity calc\n", i, count);
|
|
|
|
- BUG();
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- i = raid6_next_disk(i, disks);
|
|
|
|
- } while (i != d0_idx);
|
|
|
|
- BUG_ON(count != syndrome_disks);
|
|
|
|
-
|
|
|
|
- raid6_call.gen_syndrome(syndrome_disks+2, STRIPE_SIZE, ptrs);
|
|
|
|
-
|
|
|
|
- switch(method) {
|
|
|
|
- case RECONSTRUCT_WRITE:
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
|
|
|
|
- set_bit(R5_LOCKED, &sh->dev[pd_idx].flags);
|
|
|
|
- set_bit(R5_LOCKED, &sh->dev[qd_idx].flags);
|
|
|
|
- break;
|
|
|
|
- case UPDATE_PARITY:
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
|
|
|
|
- break;
|
|
|
|
- }
|
|
|
|
-}
|
|
|
|
-
|
|
|
|
-
|
|
|
|
-/* Compute one missing block */
|
|
|
|
-static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
|
|
|
|
-{
|
|
|
|
- int i, count, disks = sh->disks;
|
|
|
|
- void *ptr[MAX_XOR_BLOCKS], *dest, *p;
|
|
|
|
- int qd_idx = sh->qd_idx;
|
|
|
|
-
|
|
|
|
- pr_debug("compute_block_1, stripe %llu, idx %d\n",
|
|
|
|
- (unsigned long long)sh->sector, dd_idx);
|
|
|
|
-
|
|
|
|
- if ( dd_idx == qd_idx ) {
|
|
|
|
- /* We're actually computing the Q drive */
|
|
|
|
- compute_parity6(sh, UPDATE_PARITY);
|
|
|
|
- } else {
|
|
|
|
- dest = page_address(sh->dev[dd_idx].page);
|
|
|
|
- if (!nozero) memset(dest, 0, STRIPE_SIZE);
|
|
|
|
- count = 0;
|
|
|
|
- for (i = disks ; i--; ) {
|
|
|
|
- if (i == dd_idx || i == qd_idx)
|
|
|
|
- continue;
|
|
|
|
- p = page_address(sh->dev[i].page);
|
|
|
|
- if (test_bit(R5_UPTODATE, &sh->dev[i].flags))
|
|
|
|
- ptr[count++] = p;
|
|
|
|
- else
|
|
|
|
- printk("compute_block() %d, stripe %llu, %d"
|
|
|
|
- " not present\n", dd_idx,
|
|
|
|
- (unsigned long long)sh->sector, i);
|
|
|
|
-
|
|
|
|
- check_xor();
|
|
|
|
- }
|
|
|
|
- if (count)
|
|
|
|
- xor_blocks(count, STRIPE_SIZE, dest, ptr);
|
|
|
|
- if (!nozero) set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
|
|
|
|
- else clear_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
|
|
|
|
- }
|
|
|
|
-}
|
|
|
|
-
|
|
|
|
-/* Compute two missing blocks */
|
|
|
|
-static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
|
|
|
|
-{
|
|
|
|
- int i, count, disks = sh->disks;
|
|
|
|
- int syndrome_disks = sh->ddf_layout ? disks : disks-2;
|
|
|
|
- int d0_idx = raid6_d0(sh);
|
|
|
|
- int faila = -1, failb = -1;
|
|
|
|
- /**** FIX THIS: This could be very bad if disks is close to 256 ****/
|
|
|
|
- void *ptrs[syndrome_disks+2];
|
|
|
|
-
|
|
|
|
- for (i = 0; i < disks ; i++)
|
|
|
|
- ptrs[i] = (void *)raid6_empty_zero_page;
|
|
|
|
- count = 0;
|
|
|
|
- i = d0_idx;
|
|
|
|
- do {
|
|
|
|
- int slot = raid6_idx_to_slot(i, sh, &count, syndrome_disks);
|
|
|
|
-
|
|
|
|
- ptrs[slot] = page_address(sh->dev[i].page);
|
|
|
|
-
|
|
|
|
- if (i == dd_idx1)
|
|
|
|
- faila = slot;
|
|
|
|
- if (i == dd_idx2)
|
|
|
|
- failb = slot;
|
|
|
|
- i = raid6_next_disk(i, disks);
|
|
|
|
- } while (i != d0_idx);
|
|
|
|
- BUG_ON(count != syndrome_disks);
|
|
|
|
-
|
|
|
|
- BUG_ON(faila == failb);
|
|
|
|
- if ( failb < faila ) { int tmp = faila; faila = failb; failb = tmp; }
|
|
|
|
-
|
|
|
|
- pr_debug("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n",
|
|
|
|
- (unsigned long long)sh->sector, dd_idx1, dd_idx2,
|
|
|
|
- faila, failb);
|
|
|
|
-
|
|
|
|
- if (failb == syndrome_disks+1) {
|
|
|
|
- /* Q disk is one of the missing disks */
|
|
|
|
- if (faila == syndrome_disks) {
|
|
|
|
- /* Missing P+Q, just recompute */
|
|
|
|
- compute_parity6(sh, UPDATE_PARITY);
|
|
|
|
- return;
|
|
|
|
- } else {
|
|
|
|
- /* We're missing D+Q; recompute D from P */
|
|
|
|
- compute_block_1(sh, ((dd_idx1 == sh->qd_idx) ?
|
|
|
|
- dd_idx2 : dd_idx1),
|
|
|
|
- 0);
|
|
|
|
- compute_parity6(sh, UPDATE_PARITY); /* Is this necessary? */
|
|
|
|
- return;
|
|
|
|
- }
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- /* We're missing D+P or D+D; */
|
|
|
|
- if (failb == syndrome_disks) {
|
|
|
|
- /* We're missing D+P. */
|
|
|
|
- raid6_datap_recov(syndrome_disks+2, STRIPE_SIZE, faila, ptrs);
|
|
|
|
- } else {
|
|
|
|
- /* We're missing D+D. */
|
|
|
|
- raid6_2data_recov(syndrome_disks+2, STRIPE_SIZE, faila, failb,
|
|
|
|
- ptrs);
|
|
|
|
- }
|
|
|
|
-
|
|
|
|
- /* Both the above update both missing blocks */
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[dd_idx1].flags);
|
|
|
|
- set_bit(R5_UPTODATE, &sh->dev[dd_idx2].flags);
|
|
|
|
-}
|
|
|
|
-
|
|
|
|
static void
|
|
static void
|
|
schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s,
|
|
schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s,
|
|
int rcw, int expand)
|
|
int rcw, int expand)
|
|
@@ -2331,13 +2084,6 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
|
|
|
|
|
|
static void end_reshape(raid5_conf_t *conf);
|
|
static void end_reshape(raid5_conf_t *conf);
|
|
|
|
|
|
-static int page_is_zero(struct page *p)
|
|
|
|
-{
|
|
|
|
- char *a = page_address(p);
|
|
|
|
- return ((*(u32*)a) == 0 &&
|
|
|
|
- memcmp(a, a+4, STRIPE_SIZE-4)==0);
|
|
|
|
-}
|
|
|
|
-
|
|
|
|
static void stripe_set_idx(sector_t stripe, raid5_conf_t *conf, int previous,
|
|
static void stripe_set_idx(sector_t stripe, raid5_conf_t *conf, int previous,
|
|
struct stripe_head *sh)
|
|
struct stripe_head *sh)
|
|
{
|
|
{
|