Lines Matching refs:disks
36 * disks. Every stripe_head in the entry must write to the same data disks.
81 * If write-back cache is enabled for any of the disks in the array, its data
161 int disks = sh->disks;
176 * rmw: xor old data and parity from updated disks
182 /* rcw: xor data from all not updated disks */
183 for (i = disks; i--;) {
193 NULL, sh, (void *) (srcs + sh->disks + 2));
299 for (i = 0; i < sh->disks; i++) {
326 * disks. Use bit shift and logarithm to avoid 64-bit division.
479 for (i = 0; i < sh->disks; i++) {
623 rdev = conf->disks[i].rdev;
720 * disks within a stripe and write the result as the new stripe parity. If all
721 * stripe data disks are modified (full stripe write), no partial parity is
722 * available, so just xor the data disks.
724 * Recovery of a PPL entry shall occur only if all modified data disks are
750 * case 2: write to all data disks (also full stripe write):
853 /* iterate through data member disks */
883 rdev = conf->disks[dd_idx].rdev;
935 parity_rdev = conf->disks[sh.pd_idx].rdev;
1350 pr_warn("md/raid:%s PPL doesn't support over %d disks in the array\n",
1402 struct md_rdev *rdev = conf->disks[i].rdev;
1420 /* load and possibly recover the logs from the member disks */