Lines Matching +full:b +full:- +full:side

1 /* SPDX-License-Identifier: GPL-2.0-only */
30 	P_RECV_ACK	      = 0x15, /* Used in protocol B */
33 	P_SUPERSEDED	      = 0x18, /* Used in proto C, two-primaries conflict detection */
55 	P_CONN_ST_CHG_REPLY   = 0x2b, /* meta sock: Connection side state req reply */
69 	 * On a receiving side without REQ_WRITE_SAME,
74 	P_ZEROES              = 0x36, /* data sock: zero-out, WRITE_ZEROES */
135 #define DP_SEND_RECEIVE_ACK 128 /* This is a proto B write request */
147 	u64	    block_id;  /* to identify the request in protocol B&C */
154 	u32	    size;	/* == bio->bi_size */
159 	u32           size;     /* == bio->bi_size */
165  *   P_RECV_ACK (proto B), P_WRITE_ACK (proto C),
166  *   P_SUPERSEDED (proto C, two-primaries conflict detection)
196 /* Detect all-zeros during resync, and rather TRIM/UNMAP/DISCARD those blocks
203  *   - indicates support for 128 MiB "batch bios",
206  *   - indicates that we exchange additional settings in p_sizes
213  * We used to map that to "discard" on the sending side, and if we cannot
214  * guarantee that discard zeroes data, the receiving side would map discard
215  * back to zero-out.
220  * UNMAP if possible, zero-out the rest.
224  * While an un-allocated block on dm-thin reads as zeroes, on a dm-thin
229  * If we cannot distinguish between zero-out and discard on the receiving
230  * side, to avoid "undefined old garbage" to pop up randomly at later times
231  * on supposedly zero-initialized blocks, we'd need to map all discards to
232  * zero-out on the receiving side.  But that would potentially do a full
234  * unmap/trim/discard/de-allocate.
237  * zeroes (and thus use zero-out, potentially doing the mentioned full-alloc),
239  * zeroing" (by "discarding" blocks aligned to discard-granularity, and zeroing
241  * avoid "false positives" in an online-verify later, hoping that someone
333 	 * which may be translated to several bio on the receiving side.
346 	 * Receiving side uses "blkdev_issue_discard()", no need to communicate