^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) * blkfront.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) * XenLinux virtual block device driver.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) * Copyright (c) 2004, Christian Limpach
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) * Copyright (c) 2004, Andrew Warfield
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * Copyright (c) 2005, Christopher Clark
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) * Copyright (c) 2005, XenSource Ltd
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) * This program is free software; you can redistribute it and/or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) * modify it under the terms of the GNU General Public License version 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) * as published by the Free Software Foundation; or, when distributed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) * separately from the Linux kernel or incorporated into other
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) * software packages, subject to the following license:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) * Permission is hereby granted, free of charge, to any person obtaining a copy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) * of this source file (the "Software"), to deal in the Software without
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) * restriction, including without limitation the rights to use, copy, modify,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) * merge, publish, distribute, sublicense, and/or sell copies of the Software,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) * and to permit persons to whom the Software is furnished to do so, subject to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) * the following conditions:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * The above copyright notice and this permission notice shall be included in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) * all copies or substantial portions of the Software.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * IN THE SOFTWARE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #include <linux/interrupt.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) #include <linux/blkdev.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #include <linux/blk-mq.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #include <linux/hdreg.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) #include <linux/cdrom.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) #include <linux/mutex.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) #include <linux/scatterlist.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) #include <linux/bitmap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) #include <linux/list.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) #include <linux/workqueue.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) #include <linux/sched/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) #include <xen/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) #include <xen/xenbus.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) #include <xen/grant_table.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) #include <xen/events.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) #include <xen/page.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) #include <xen/platform_pci.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) #include <xen/interface/grant_table.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) #include <xen/interface/io/blkif.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) #include <xen/interface/io/protocols.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) #include <asm/xen/hypervisor.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) * The minimal size of segment supported by the block framework is PAGE_SIZE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) * When Linux is using a different page size than Xen, it may not be possible
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) * to put all the data in a single segment.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) * This can happen when the backend doesn't support indirect descriptor and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) * therefore the maximum amount of data that a request can carry is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE = 44KB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) * Note that we only support one extra request. So the Linux page size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) * should be <= ( 2 * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) * 88KB.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) #define HAS_EXTRA_REQ (BLKIF_MAX_SEGMENTS_PER_REQUEST < XEN_PFN_PER_PAGE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) enum blkif_state {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) BLKIF_STATE_DISCONNECTED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) BLKIF_STATE_CONNECTED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) BLKIF_STATE_SUSPENDED,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) BLKIF_STATE_ERROR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) struct grant {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) grant_ref_t gref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) struct list_head node;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) enum blk_req_status {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) REQ_PROCESSING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) REQ_WAITING,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) REQ_DONE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) REQ_ERROR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) REQ_EOPNOTSUPP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) struct blk_shadow {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) struct blkif_request req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) struct request *request;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) struct grant **grants_used;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) struct grant **indirect_grants;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) struct scatterlist *sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) unsigned int num_sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) enum blk_req_status status;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) #define NO_ASSOCIATED_ID ~0UL
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) * Id of the sibling if we ever need 2 requests when handling a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) * block I/O request
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) unsigned long associated_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) struct blkif_req {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) blk_status_t error;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) static inline struct blkif_req *blkif_req(struct request *rq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) return blk_mq_rq_to_pdu(rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) static DEFINE_MUTEX(blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) static const struct block_device_operations xlvbd_block_fops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) static struct delayed_work blkfront_work;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) static LIST_HEAD(info_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) * Maximum number of segments in indirect requests, the actual value used by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) * the frontend driver is the minimum of this value and the value provided
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) * by the backend driver.
^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 unsigned int xen_blkif_max_segments = 32;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) module_param_named(max_indirect_segments, xen_blkif_max_segments, uint, 0444);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) MODULE_PARM_DESC(max_indirect_segments,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) "Maximum amount of segments in indirect requests (default is 32)");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) static unsigned int xen_blkif_max_queues = 4;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) module_param_named(max_queues, xen_blkif_max_queues, uint, 0444);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) MODULE_PARM_DESC(max_queues, "Maximum number of hardware queues/rings used per virtual disk");
^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) * Maximum order of pages to be used for the shared ring between front and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) * backend, 4KB page granularity is used.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) static unsigned int xen_blkif_max_ring_order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) module_param_named(max_ring_page_order, xen_blkif_max_ring_order, int, 0444);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the shared ring");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) #define BLK_RING_SIZE(info) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) __CONST_RING_SIZE(blkif, XEN_PAGE_SIZE * (info)->nr_ring_pages)
^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) * ring-ref%u i=(-1UL) would take 11 characters + 'ring-ref' is 8, so 19
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) * characters are enough. Define to 20 to keep consistent with backend.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) #define RINGREF_NAME_LEN (20)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) * queue-%u would take 7 + 10(UINT_MAX) = 17 characters.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) #define QUEUE_NAME_LEN (17)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) * Per-ring info.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) * Every blkfront device can associate with one or more blkfront_ring_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) * depending on how many hardware queues/rings to be used.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) struct blkfront_ring_info {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) /* Lock to protect data in every ring buffer. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) spinlock_t ring_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) struct blkif_front_ring ring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) unsigned int ring_ref[XENBUS_MAX_RING_GRANTS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) unsigned int evtchn, irq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) struct work_struct work;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) struct gnttab_free_callback callback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) struct list_head indirect_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) struct list_head grants;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) unsigned int persistent_gnts_c;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) unsigned long shadow_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) struct blkfront_info *dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) struct blk_shadow shadow[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) * We have one of these per vbd, whether ide, scsi or 'other'. They
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) * hang in private_data off the gendisk structure. We may end up
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) * putting all kinds of interesting stuff here :-)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) struct blkfront_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) struct mutex mutex;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) struct xenbus_device *xbdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) struct gendisk *gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) u16 sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) unsigned int physical_sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) int vdevice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) blkif_vdev_t handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) enum blkif_state connected;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) /* Number of pages per ring buffer. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) unsigned int nr_ring_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) struct request_queue *rq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) unsigned int feature_flush:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) unsigned int feature_fua:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) unsigned int feature_discard:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) unsigned int feature_secdiscard:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) unsigned int feature_persistent:1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) unsigned int discard_granularity;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) unsigned int discard_alignment;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) /* Number of 4KB segments handled */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) unsigned int max_indirect_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) int is_ready;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) struct blk_mq_tag_set tag_set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) unsigned int nr_rings;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) unsigned int rinfo_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) /* Save uncomplete reqs and bios for migration. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) struct list_head requests;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) struct bio_list bio_list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) struct list_head info_list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) static unsigned int nr_minors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) static unsigned long *minors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) static DEFINE_SPINLOCK(minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) #define GRANT_INVALID_REF 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) #define PARTS_PER_DISK 16
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) #define PARTS_PER_EXT_DISK 256
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) #define BLKIF_MAJOR(dev) ((dev)>>8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) #define BLKIF_MINOR(dev) ((dev) & 0xff)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) #define EXT_SHIFT 28
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) #define EXTENDED (1<<EXT_SHIFT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) #define VDEV_IS_EXTENDED(dev) ((dev)&(EXTENDED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) #define BLKIF_MINOR_EXT(dev) ((dev)&(~EXTENDED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) #define EMULATED_HD_DISK_MINOR_OFFSET (0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) #define EMULATED_HD_DISK_NAME_OFFSET (EMULATED_HD_DISK_MINOR_OFFSET / 256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) #define EMULATED_SD_DISK_MINOR_OFFSET (0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) #define EMULATED_SD_DISK_NAME_OFFSET (EMULATED_SD_DISK_MINOR_OFFSET / 256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) #define DEV_NAME "xvd" /* name in /dev */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) * Grants are always the same size as a Xen page (i.e 4KB).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) * A physical segment is always the same size as a Linux page.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) * Number of grants per physical segment
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) #define GRANTS_PER_PSEG (PAGE_SIZE / XEN_PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) #define GRANTS_PER_INDIRECT_FRAME \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) (XEN_PAGE_SIZE / sizeof(struct blkif_request_segment))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) #define INDIRECT_GREFS(_grants) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) DIV_ROUND_UP(_grants, GRANTS_PER_INDIRECT_FRAME)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) static void blkfront_gather_backend_features(struct blkfront_info *info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) static int negotiate_mq(struct blkfront_info *info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) #define for_each_rinfo(info, ptr, idx) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) for ((ptr) = (info)->rinfo, (idx) = 0; \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) (idx) < (info)->nr_rings; \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) (idx)++, (ptr) = (void *)(ptr) + (info)->rinfo_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) static inline struct blkfront_ring_info *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) get_rinfo(const struct blkfront_info *info, unsigned int i)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) BUG_ON(i >= info->nr_rings);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) return (void *)info->rinfo + i * info->rinfo_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) static int get_id_from_freelist(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) unsigned long free = rinfo->shadow_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) BUG_ON(free >= BLK_RING_SIZE(rinfo->dev_info));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) rinfo->shadow_free = rinfo->shadow[free].req.u.rw.id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) rinfo->shadow[free].req.u.rw.id = 0x0fffffee; /* debug */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) return free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) static int add_id_to_freelist(struct blkfront_ring_info *rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) unsigned long id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) if (rinfo->shadow[id].req.u.rw.id != id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) if (rinfo->shadow[id].request == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) rinfo->shadow[id].req.u.rw.id = rinfo->shadow_free;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) rinfo->shadow[id].request = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) rinfo->shadow_free = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) static int fill_grant_buffer(struct blkfront_ring_info *rinfo, int num)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) struct page *granted_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) struct grant *gnt_list_entry, *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) int i = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) while (i < num) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) gnt_list_entry = kzalloc(sizeof(struct grant), GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) if (!gnt_list_entry)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) goto out_of_memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) if (info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) granted_page = alloc_page(GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) if (!granted_page) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) kfree(gnt_list_entry);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) goto out_of_memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) gnt_list_entry->page = granted_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) gnt_list_entry->gref = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) list_add(&gnt_list_entry->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) i++;
^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) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) out_of_memory:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) list_for_each_entry_safe(gnt_list_entry, n,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) &rinfo->grants, node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) list_del(&gnt_list_entry->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) if (info->feature_persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) __free_page(gnt_list_entry->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) kfree(gnt_list_entry);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) i--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) BUG_ON(i != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) static struct grant *get_free_grant(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) struct grant *gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) BUG_ON(list_empty(&rinfo->grants));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) gnt_list_entry = list_first_entry(&rinfo->grants, struct grant,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) list_del(&gnt_list_entry->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) if (gnt_list_entry->gref != GRANT_INVALID_REF)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) rinfo->persistent_gnts_c--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) return gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) static inline void grant_foreign_access(const struct grant *gnt_list_entry,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) const struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) gnttab_page_grant_foreign_access_ref_one(gnt_list_entry->gref,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) info->xbdev->otherend_id,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) gnt_list_entry->page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) static struct grant *get_grant(grant_ref_t *gref_head,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) unsigned long gfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) struct grant *gnt_list_entry = get_free_grant(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) if (gnt_list_entry->gref != GRANT_INVALID_REF)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) return gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) /* Assign a gref to this page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) gnt_list_entry->gref = gnttab_claim_grant_reference(gref_head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) BUG_ON(gnt_list_entry->gref == -ENOSPC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) if (info->feature_persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) grant_foreign_access(gnt_list_entry, info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) /* Grant access to the GFN passed by the caller */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) gnttab_grant_foreign_access_ref(gnt_list_entry->gref,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) info->xbdev->otherend_id,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) gfn, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) return gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) static struct grant *get_indirect_grant(grant_ref_t *gref_head,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) struct grant *gnt_list_entry = get_free_grant(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) if (gnt_list_entry->gref != GRANT_INVALID_REF)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) return gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) /* Assign a gref to this page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) gnt_list_entry->gref = gnttab_claim_grant_reference(gref_head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) BUG_ON(gnt_list_entry->gref == -ENOSPC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) if (!info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) struct page *indirect_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) /* Fetch a pre-allocated page to use for indirect grefs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) BUG_ON(list_empty(&rinfo->indirect_pages));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) indirect_page = list_first_entry(&rinfo->indirect_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) struct page, lru);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) list_del(&indirect_page->lru);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) gnt_list_entry->page = indirect_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) grant_foreign_access(gnt_list_entry, info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) return gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) static const char *op_name(int op)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) static const char *const names[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) [BLKIF_OP_READ] = "read",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) [BLKIF_OP_WRITE] = "write",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) [BLKIF_OP_WRITE_BARRIER] = "barrier",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) [BLKIF_OP_FLUSH_DISKCACHE] = "flush",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) [BLKIF_OP_DISCARD] = "discard" };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) if (op < 0 || op >= ARRAY_SIZE(names))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) return "unknown";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) if (!names[op])
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) return "reserved";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) return names[op];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) static int xlbd_reserve_minors(unsigned int minor, unsigned int nr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) unsigned int end = minor + nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) int rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) if (end > nr_minors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) unsigned long *bitmap, *old;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) bitmap = kcalloc(BITS_TO_LONGS(end), sizeof(*bitmap),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) if (bitmap == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) spin_lock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) if (end > nr_minors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) old = minors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) memcpy(bitmap, minors,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) BITS_TO_LONGS(nr_minors) * sizeof(*bitmap));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) minors = bitmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) nr_minors = BITS_TO_LONGS(end) * BITS_PER_LONG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) old = bitmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) spin_unlock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) kfree(old);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) spin_lock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) if (find_next_bit(minors, end, minor) >= end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) bitmap_set(minors, minor, nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) rc = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) rc = -EBUSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) spin_unlock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) return rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) static void xlbd_release_minors(unsigned int minor, unsigned int nr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) unsigned int end = minor + nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) BUG_ON(end > nr_minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) spin_lock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) bitmap_clear(minors, minor, nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) spin_unlock(&minor_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) static void blkif_restart_queue_callback(void *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)arg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) schedule_work(&rinfo->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) static int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) /* We don't have real geometry info, but let's at least return
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) values consistent with the size of the device */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) sector_t nsect = get_capacity(bd->bd_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) sector_t cylinders = nsect;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) hg->heads = 0xff;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) hg->sectors = 0x3f;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) sector_div(cylinders, hg->heads * hg->sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) hg->cylinders = cylinders;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) if ((sector_t)(hg->cylinders + 1) * hg->heads * hg->sectors < nsect)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) hg->cylinders = 0xffff;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) static int blkif_ioctl(struct block_device *bdev, fmode_t mode,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) unsigned command, unsigned long argument)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) struct blkfront_info *info = bdev->bd_disk->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) dev_dbg(&info->xbdev->dev, "command: 0x%x, argument: 0x%lx\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) command, (long)argument);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) switch (command) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) case CDROMMULTISESSION:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) dev_dbg(&info->xbdev->dev, "FIXME: support multisession CDs later\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) for (i = 0; i < sizeof(struct cdrom_multisession); i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) if (put_user(0, (char __user *)(argument + i)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) return -EFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) case CDROM_GET_CAPABILITY: {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) struct gendisk *gd = info->gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) if (gd->flags & GENHD_FL_CD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) /*printk(KERN_ALERT "ioctl %08x not supported by Xen blkdev\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) command);*/
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) return -EINVAL; /* same return as native Linux */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) struct request *req,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) struct blkif_request **ring_req)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) unsigned long id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) *ring_req = RING_GET_REQUEST(&rinfo->ring, rinfo->ring.req_prod_pvt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) rinfo->ring.req_prod_pvt++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) id = get_id_from_freelist(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) rinfo->shadow[id].request = req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) rinfo->shadow[id].status = REQ_PROCESSING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) rinfo->shadow[id].associated_id = NO_ASSOCIATED_ID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) rinfo->shadow[id].req.u.rw.id = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) return id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) struct blkif_request *ring_req, *final_ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) unsigned long id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) /* Fill out a communications ring structure. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) id = blkif_ring_get_request(rinfo, req, &final_ring_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) ring_req = &rinfo->shadow[id].req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) ring_req->operation = BLKIF_OP_DISCARD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) ring_req->u.discard.nr_sectors = blk_rq_sectors(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) ring_req->u.discard.id = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) ring_req->u.discard.sector_number = (blkif_sector_t)blk_rq_pos(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) if (req_op(req) == REQ_OP_SECURE_ERASE && info->feature_secdiscard)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) ring_req->u.discard.flag = BLKIF_DISCARD_SECURE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) ring_req->u.discard.flag = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) /* Copy the request to the ring page. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) *final_ring_req = *ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) rinfo->shadow[id].status = REQ_WAITING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) struct setup_rw_req {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) unsigned int grant_idx;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) struct blkif_request_segment *segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) struct blkif_request *ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) grant_ref_t gref_head;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) unsigned int id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) /* Only used when persistent grant is used and it's a read request */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) bool need_copy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) unsigned int bvec_off;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) char *bvec_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) bool require_extra_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) struct blkif_request *extra_ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) unsigned int len, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) struct setup_rw_req *setup = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) int n, ref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) struct grant *gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) unsigned int fsect, lsect;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) /* Convenient aliases */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) unsigned int grant_idx = setup->grant_idx;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) struct blkif_request *ring_req = setup->ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) struct blkfront_ring_info *rinfo = setup->rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) * We always use the shadow of the first request to store the list
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) * of grant associated to the block I/O request. This made the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) * completion more easy to handle even if the block I/O request is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) * split.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) struct blk_shadow *shadow = &rinfo->shadow[setup->id];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) if (unlikely(setup->require_extra_req &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) grant_idx >= BLKIF_MAX_SEGMENTS_PER_REQUEST)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) * We are using the second request, setup grant_idx
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) * to be the index of the segment array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) grant_idx -= BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) ring_req = setup->extra_ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) if ((ring_req->operation == BLKIF_OP_INDIRECT) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) (grant_idx % GRANTS_PER_INDIRECT_FRAME == 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) if (setup->segments)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) kunmap_atomic(setup->segments);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) n = grant_idx / GRANTS_PER_INDIRECT_FRAME;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) gnt_list_entry = get_indirect_grant(&setup->gref_head, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) shadow->indirect_grants[n] = gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) setup->segments = kmap_atomic(gnt_list_entry->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) ring_req->u.indirect.indirect_grefs[n] = gnt_list_entry->gref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) gnt_list_entry = get_grant(&setup->gref_head, gfn, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) ref = gnt_list_entry->gref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) * All the grants are stored in the shadow of the first
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) * request. Therefore we have to use the global index.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) shadow->grants_used[setup->grant_idx] = gnt_list_entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) if (setup->need_copy) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) void *shared_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) shared_data = kmap_atomic(gnt_list_entry->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) * this does not wipe data stored outside the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) * range sg->offset..sg->offset+sg->length.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) * Therefore, blkback *could* see data from
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) * previous requests. This is OK as long as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) * persistent grants are shared with just one
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) * domain. It may need refactoring if this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) * changes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) memcpy(shared_data + offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) setup->bvec_data + setup->bvec_off,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) kunmap_atomic(shared_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) setup->bvec_off += len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) fsect = offset >> 9;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) lsect = fsect + (len >> 9) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) if (ring_req->operation != BLKIF_OP_INDIRECT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) ring_req->u.rw.seg[grant_idx] =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) (struct blkif_request_segment) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) .gref = ref,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) .first_sect = fsect,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) .last_sect = lsect };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) setup->segments[grant_idx % GRANTS_PER_INDIRECT_FRAME] =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) (struct blkif_request_segment) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) .gref = ref,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) .first_sect = fsect,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) .last_sect = lsect };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) (setup->grant_idx)++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) static void blkif_setup_extra_req(struct blkif_request *first,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) struct blkif_request *second)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) uint16_t nr_segments = first->u.rw.nr_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) * The second request is only present when the first request uses
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) * all its segments. It's always the continuity of the first one.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) first->u.rw.nr_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) second->u.rw.nr_segments = nr_segments - BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) second->u.rw.sector_number = first->u.rw.sector_number +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) (BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) / 512;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) second->u.rw.handle = first->u.rw.handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) second->operation = first->operation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) struct blkif_request *ring_req, *extra_ring_req = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) struct blkif_request *final_ring_req, *final_extra_ring_req = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) unsigned long id, extra_id = NO_ASSOCIATED_ID;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) bool require_extra_req = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) struct setup_rw_req setup = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) .grant_idx = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) .segments = NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) .rinfo = rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) .need_copy = rq_data_dir(req) && info->feature_persistent,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) * Used to store if we are able to queue the request by just using
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) * existing persistent grants, or if we have to get new grants,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) * as there are not sufficiently many free.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) bool new_persistent_gnts = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) struct scatterlist *sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) int num_sg, max_grefs, num_grant;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) max_grefs = req->nr_phys_segments * GRANTS_PER_PSEG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) if (max_grefs > BLKIF_MAX_SEGMENTS_PER_REQUEST)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) * If we are using indirect segments we need to account
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) * for the indirect grefs used in the request.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) max_grefs += INDIRECT_GREFS(max_grefs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) /* Check if we have enough persistent grants to allocate a requests */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) if (rinfo->persistent_gnts_c < max_grefs) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) new_persistent_gnts = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) if (gnttab_alloc_grant_references(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) max_grefs - rinfo->persistent_gnts_c,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) &setup.gref_head) < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) gnttab_request_free_callback(
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) &rinfo->callback,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) blkif_restart_queue_callback,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) max_grefs - rinfo->persistent_gnts_c);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) /* Fill out a communications ring structure. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) id = blkif_ring_get_request(rinfo, req, &final_ring_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) ring_req = &rinfo->shadow[id].req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) num_sg = blk_rq_map_sg(req->q, req, rinfo->shadow[id].sg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) num_grant = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) /* Calculate the number of grant used */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) num_grant += gnttab_count_grant(sg->offset, sg->length);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) require_extra_req = info->max_indirect_segments == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) BUG_ON(!HAS_EXTRA_REQ && require_extra_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) rinfo->shadow[id].num_sg = num_sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) if (num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) likely(!require_extra_req)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) * The indirect operation can only be a BLKIF_OP_READ or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) * BLKIF_OP_WRITE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) BUG_ON(req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) ring_req->operation = BLKIF_OP_INDIRECT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) ring_req->u.indirect.indirect_op = rq_data_dir(req) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) BLKIF_OP_WRITE : BLKIF_OP_READ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) ring_req->u.indirect.sector_number = (blkif_sector_t)blk_rq_pos(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) ring_req->u.indirect.handle = info->handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) ring_req->u.indirect.nr_segments = num_grant;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) ring_req->u.rw.sector_number = (blkif_sector_t)blk_rq_pos(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) ring_req->u.rw.handle = info->handle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) ring_req->operation = rq_data_dir(req) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) BLKIF_OP_WRITE : BLKIF_OP_READ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) if (req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) * Ideally we can do an unordered flush-to-disk.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) * In case the backend onlysupports barriers, use that.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) * A barrier request a superset of FUA, so we can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) * implement it the same way. (It's also a FLUSH+FUA,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) * since it is guaranteed ordered WRT previous writes.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) if (info->feature_flush && info->feature_fua)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) ring_req->operation =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) BLKIF_OP_WRITE_BARRIER;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) else if (info->feature_flush)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) ring_req->operation =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) BLKIF_OP_FLUSH_DISKCACHE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) ring_req->operation = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) ring_req->u.rw.nr_segments = num_grant;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) if (unlikely(require_extra_req)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) extra_id = blkif_ring_get_request(rinfo, req,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) &final_extra_ring_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) extra_ring_req = &rinfo->shadow[extra_id].req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) * Only the first request contains the scatter-gather
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) * list.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) rinfo->shadow[extra_id].num_sg = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) blkif_setup_extra_req(ring_req, extra_ring_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) /* Link the 2 requests together */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) rinfo->shadow[extra_id].associated_id = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) rinfo->shadow[id].associated_id = extra_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) setup.ring_req = ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) setup.id = id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) setup.require_extra_req = require_extra_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) if (unlikely(require_extra_req))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) setup.extra_ring_req = extra_ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) BUG_ON(sg->offset + sg->length > PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) if (setup.need_copy) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) setup.bvec_off = sg->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) setup.bvec_data = kmap_atomic(sg_page(sg));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) gnttab_foreach_grant_in_range(sg_page(sg),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) sg->offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) sg->length,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) blkif_setup_rw_req_grant,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) &setup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) if (setup.need_copy)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) kunmap_atomic(setup.bvec_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) if (setup.segments)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) kunmap_atomic(setup.segments);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) /* Copy request(s) to the ring page. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) *final_ring_req = *ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) rinfo->shadow[id].status = REQ_WAITING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) if (unlikely(require_extra_req)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) *final_extra_ring_req = *extra_ring_req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) rinfo->shadow[extra_id].status = REQ_WAITING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) if (new_persistent_gnts)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) gnttab_free_grant_references(setup.gref_head);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) * Generate a Xen blkfront IO request from a blk layer request. Reads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) * and writes are handled as expected.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) * @req: a request struct
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) static int blkif_queue_request(struct request *req, struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) if (unlikely(rinfo->dev_info->connected != BLKIF_STATE_CONNECTED))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) if (unlikely(req_op(req) == REQ_OP_DISCARD ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) req_op(req) == REQ_OP_SECURE_ERASE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) return blkif_queue_discard_req(req, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) return blkif_queue_rw_req(req, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) static inline void flush_requests(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) int notify;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&rinfo->ring, notify);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) if (notify)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) notify_remote_via_irq(rinfo->irq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) static inline bool blkif_request_flush_invalid(struct request *req,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) return (blk_rq_is_passthrough(req) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) ((req_op(req) == REQ_OP_FLUSH) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) !info->feature_flush) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) ((req->cmd_flags & REQ_FUA) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) !info->feature_fua));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) static blk_status_t blkif_queue_rq(struct blk_mq_hw_ctx *hctx,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) const struct blk_mq_queue_data *qd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) int qid = hctx->queue_num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) struct blkfront_info *info = hctx->queue->queuedata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) struct blkfront_ring_info *rinfo = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) rinfo = get_rinfo(info, qid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) blk_mq_start_request(qd->rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) spin_lock_irqsave(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) if (RING_FULL(&rinfo->ring))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) goto out_busy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) if (blkif_request_flush_invalid(qd->rq, rinfo->dev_info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) goto out_err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) if (blkif_queue_request(qd->rq, rinfo))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) goto out_busy;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) flush_requests(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) return BLK_STS_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) out_err:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) return BLK_STS_IOERR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) out_busy:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) blk_mq_stop_hw_queue(hctx);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) return BLK_STS_DEV_RESOURCE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) static void blkif_complete_rq(struct request *rq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) blk_mq_end_request(rq, blkif_req(rq)->error);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) static const struct blk_mq_ops blkfront_mq_ops = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) .queue_rq = blkif_queue_rq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) .complete = blkif_complete_rq,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) static void blkif_set_queue_limits(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) struct request_queue *rq = info->rq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) struct gendisk *gd = info->gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) unsigned int segments = info->max_indirect_segments ? :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) blk_queue_flag_set(QUEUE_FLAG_VIRT, rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) if (info->feature_discard) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) blk_queue_flag_set(QUEUE_FLAG_DISCARD, rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) blk_queue_max_discard_sectors(rq, get_capacity(gd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) rq->limits.discard_granularity = info->discard_granularity ?:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) info->physical_sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) rq->limits.discard_alignment = info->discard_alignment;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) if (info->feature_secdiscard)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) blk_queue_flag_set(QUEUE_FLAG_SECERASE, rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) /* Hard sector size and max sectors impersonate the equiv. hardware. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) blk_queue_logical_block_size(rq, info->sector_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) blk_queue_physical_block_size(rq, info->physical_sector_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) blk_queue_max_hw_sectors(rq, (segments * XEN_PAGE_SIZE) / 512);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) /* Each segment in a request is up to an aligned page in size. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) blk_queue_segment_boundary(rq, PAGE_SIZE - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) blk_queue_max_segment_size(rq, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) /* Ensure a merged request will fit in a single I/O ring slot. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) blk_queue_max_segments(rq, segments / GRANTS_PER_PSEG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) /* Make sure buffer addresses are sector-aligned. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) blk_queue_dma_alignment(rq, 511);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) unsigned int physical_sector_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) struct request_queue *rq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) struct blkfront_info *info = gd->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) memset(&info->tag_set, 0, sizeof(info->tag_set));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) info->tag_set.ops = &blkfront_mq_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) info->tag_set.nr_hw_queues = info->nr_rings;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) if (HAS_EXTRA_REQ && info->max_indirect_segments == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) * When indirect descriptior is not supported, the I/O request
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) * will be split between multiple request in the ring.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) * To avoid problems when sending the request, divide by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) * 2 the depth of the queue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) info->tag_set.queue_depth = BLK_RING_SIZE(info) / 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) info->tag_set.queue_depth = BLK_RING_SIZE(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) info->tag_set.numa_node = NUMA_NO_NODE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) info->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) info->tag_set.cmd_size = sizeof(struct blkif_req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) info->tag_set.driver_data = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) if (blk_mq_alloc_tag_set(&info->tag_set))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) rq = blk_mq_init_queue(&info->tag_set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) if (IS_ERR(rq)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) blk_mq_free_tag_set(&info->tag_set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) return PTR_ERR(rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) rq->queuedata = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) info->rq = gd->queue = rq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) info->gd = gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) info->sector_size = sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) info->physical_sector_size = physical_sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) blkif_set_queue_limits(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) static const char *flush_info(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) if (info->feature_flush && info->feature_fua)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) return "barrier: enabled;";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) else if (info->feature_flush)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) return "flush diskcache: enabled;";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) return "barrier or flush: disabled;";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) static void xlvbd_flush(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) blk_queue_write_cache(info->rq, info->feature_flush ? true : false,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) info->feature_fua ? true : false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) pr_info("blkfront: %s: %s %s %s %s %s\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) info->gd->disk_name, flush_info(info),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) "persistent grants:", info->feature_persistent ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) "enabled;" : "disabled;", "indirect descriptors:",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) info->max_indirect_segments ? "enabled;" : "disabled;");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) static int xen_translate_vdev(int vdevice, int *minor, unsigned int *offset)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) int major;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) major = BLKIF_MAJOR(vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) *minor = BLKIF_MINOR(vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) switch (major) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) case XEN_IDE0_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) *offset = (*minor / 64) + EMULATED_HD_DISK_NAME_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) *minor = ((*minor / 64) * PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) EMULATED_HD_DISK_MINOR_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) case XEN_IDE1_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) *offset = (*minor / 64) + 2 + EMULATED_HD_DISK_NAME_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) *minor = (((*minor / 64) + 2) * PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) EMULATED_HD_DISK_MINOR_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) case XEN_SCSI_DISK0_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) *offset = (*minor / PARTS_PER_DISK) + EMULATED_SD_DISK_NAME_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) *minor = *minor + EMULATED_SD_DISK_MINOR_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) case XEN_SCSI_DISK1_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) case XEN_SCSI_DISK2_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) case XEN_SCSI_DISK3_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) case XEN_SCSI_DISK4_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) case XEN_SCSI_DISK5_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) case XEN_SCSI_DISK6_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) case XEN_SCSI_DISK7_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) *offset = (*minor / PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) ((major - XEN_SCSI_DISK1_MAJOR + 1) * 16) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) EMULATED_SD_DISK_NAME_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) *minor = *minor +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) ((major - XEN_SCSI_DISK1_MAJOR + 1) * 16 * PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) EMULATED_SD_DISK_MINOR_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) case XEN_SCSI_DISK8_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) case XEN_SCSI_DISK9_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) case XEN_SCSI_DISK10_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) case XEN_SCSI_DISK11_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) case XEN_SCSI_DISK12_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) case XEN_SCSI_DISK13_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) case XEN_SCSI_DISK14_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) case XEN_SCSI_DISK15_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) *offset = (*minor / PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) ((major - XEN_SCSI_DISK8_MAJOR + 8) * 16) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) EMULATED_SD_DISK_NAME_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) *minor = *minor +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) ((major - XEN_SCSI_DISK8_MAJOR + 8) * 16 * PARTS_PER_DISK) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) EMULATED_SD_DISK_MINOR_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) case XENVBD_MAJOR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) *offset = *minor / PARTS_PER_DISK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) printk(KERN_WARNING "blkfront: your disk configuration is "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) "incorrect, please use an xvd device instead\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) static char *encode_disk_name(char *ptr, unsigned int n)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) if (n >= 26)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) ptr = encode_disk_name(ptr, n / 26 - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) *ptr = 'a' + n % 26;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) return ptr + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) struct blkfront_info *info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) u16 vdisk_info, u16 sector_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) unsigned int physical_sector_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) struct gendisk *gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) int nr_minors = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) unsigned int offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) int minor;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) int nr_parts;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) char *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) BUG_ON(info->gd != NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) BUG_ON(info->rq != NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) if ((info->vdevice>>EXT_SHIFT) > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) /* this is above the extended range; something is wrong */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) printk(KERN_WARNING "blkfront: vdevice 0x%x is above the extended range; ignoring\n", info->vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) if (!VDEV_IS_EXTENDED(info->vdevice)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) err = xen_translate_vdev(info->vdevice, &minor, &offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) nr_parts = PARTS_PER_DISK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) minor = BLKIF_MINOR_EXT(info->vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) nr_parts = PARTS_PER_EXT_DISK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) offset = minor / nr_parts;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) if (xen_hvm_domain() && offset < EMULATED_HD_DISK_NAME_OFFSET + 4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) printk(KERN_WARNING "blkfront: vdevice 0x%x might conflict with "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) "emulated IDE disks,\n\t choose an xvd device name"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) "from xvde on\n", info->vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) if (minor >> MINORBITS) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) pr_warn("blkfront: %#x's minor (%#x) out of range; ignoring\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) info->vdevice, minor);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) if ((minor % nr_parts) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) nr_minors = nr_parts;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) err = xlbd_reserve_minors(minor, nr_minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) err = -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) gd = alloc_disk(nr_minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) if (gd == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) goto release;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) strcpy(gd->disk_name, DEV_NAME);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) ptr = encode_disk_name(gd->disk_name + sizeof(DEV_NAME) - 1, offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) BUG_ON(ptr >= gd->disk_name + DISK_NAME_LEN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) if (nr_minors > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) *ptr = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) snprintf(ptr, gd->disk_name + DISK_NAME_LEN - ptr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) "%d", minor & (nr_parts - 1));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) gd->major = XENVBD_MAJOR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) gd->first_minor = minor;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) gd->fops = &xlvbd_block_fops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) gd->private_data = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) set_capacity(gd, capacity);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) if (xlvbd_init_blk_queue(gd, sector_size, physical_sector_size)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) del_gendisk(gd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) goto release;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) xlvbd_flush(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) if (vdisk_info & VDISK_READONLY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) set_disk_ro(gd, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) if (vdisk_info & VDISK_REMOVABLE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) gd->flags |= GENHD_FL_REMOVABLE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) if (vdisk_info & VDISK_CDROM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) gd->flags |= GENHD_FL_CD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) release:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) xlbd_release_minors(minor, nr_minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) static void xlvbd_release_gendisk(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) unsigned int minor, nr_minors, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) if (info->rq == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) /* No more blkif_request(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) blk_mq_stop_hw_queues(info->rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) /* No more gnttab callback work. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) gnttab_cancel_free_callback(&rinfo->callback);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) /* Flush gnttab callback work. Must be done with no locks held. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) flush_work(&rinfo->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) del_gendisk(info->gd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) minor = info->gd->first_minor;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) nr_minors = info->gd->minors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) xlbd_release_minors(minor, nr_minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) blk_cleanup_queue(info->rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) blk_mq_free_tag_set(&info->tag_set);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) info->rq = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) put_disk(info->gd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) info->gd = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) /* Already hold rinfo->ring_lock. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) static inline void kick_pending_request_queues_locked(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) if (!RING_FULL(&rinfo->ring))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) blk_mq_start_stopped_hw_queues(rinfo->dev_info->rq, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) static void kick_pending_request_queues(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) spin_lock_irqsave(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) kick_pending_request_queues_locked(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) static void blkif_restart_queue(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) struct blkfront_ring_info *rinfo = container_of(work, struct blkfront_ring_info, work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) if (rinfo->dev_info->connected == BLKIF_STATE_CONNECTED)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) kick_pending_request_queues(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) static void blkif_free_ring(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) struct grant *persistent_gnt, *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) int i, j, segs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) * Remove indirect pages, this only happens when using indirect
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) * descriptors but not persistent grants
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) if (!list_empty(&rinfo->indirect_pages)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) struct page *indirect_page, *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) BUG_ON(info->feature_persistent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) list_del(&indirect_page->lru);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) __free_page(indirect_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) /* Remove all persistent grants. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) if (!list_empty(&rinfo->grants)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) list_for_each_entry_safe(persistent_gnt, n,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) &rinfo->grants, node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) list_del(&persistent_gnt->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) if (persistent_gnt->gref != GRANT_INVALID_REF) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) gnttab_end_foreign_access(persistent_gnt->gref,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) 0, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) rinfo->persistent_gnts_c--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) if (info->feature_persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) __free_page(persistent_gnt->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295) kfree(persistent_gnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) BUG_ON(rinfo->persistent_gnts_c != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) for (i = 0; i < BLK_RING_SIZE(info); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) * Clear persistent grants present in requests already
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) * on the shared ring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) if (!rinfo->shadow[i].request)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) goto free_shadow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) segs = rinfo->shadow[i].req.operation == BLKIF_OP_INDIRECT ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) rinfo->shadow[i].req.u.indirect.nr_segments :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) rinfo->shadow[i].req.u.rw.nr_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) for (j = 0; j < segs; j++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) persistent_gnt = rinfo->shadow[i].grants_used[j];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) if (info->feature_persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) __free_page(persistent_gnt->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) kfree(persistent_gnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) if (rinfo->shadow[i].req.operation != BLKIF_OP_INDIRECT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) * If this is not an indirect operation don't try to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) * free indirect segments
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) goto free_shadow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326) for (j = 0; j < INDIRECT_GREFS(segs); j++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) persistent_gnt = rinfo->shadow[i].indirect_grants[j];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) __free_page(persistent_gnt->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) kfree(persistent_gnt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) free_shadow:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) kvfree(rinfo->shadow[i].grants_used);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) rinfo->shadow[i].grants_used = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) kvfree(rinfo->shadow[i].indirect_grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) rinfo->shadow[i].indirect_grants = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) kvfree(rinfo->shadow[i].sg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) rinfo->shadow[i].sg = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) /* No more gnttab callback work. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) gnttab_cancel_free_callback(&rinfo->callback);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) /* Flush gnttab callback work. Must be done with no locks held. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) flush_work(&rinfo->work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) /* Free resources associated with old device channel. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) for (i = 0; i < info->nr_ring_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) if (rinfo->ring_ref[i] != GRANT_INVALID_REF) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) gnttab_end_foreign_access(rinfo->ring_ref[i], 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) rinfo->ring_ref[i] = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) free_pages_exact(rinfo->ring.sring,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) info->nr_ring_pages * XEN_PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) rinfo->ring.sring = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) if (rinfo->irq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) unbind_from_irqhandler(rinfo->irq, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) rinfo->evtchn = rinfo->irq = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) static void blkif_free(struct blkfront_info *info, int suspend)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) /* Prevent new requests being issued until we fix things up. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) info->connected = suspend ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) /* No more blkif_request(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) if (info->rq)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) blk_mq_stop_hw_queues(info->rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) for_each_rinfo(info, rinfo, i)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) blkif_free_ring(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) kvfree(info->rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) info->rinfo = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) info->nr_rings = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) struct copy_from_grant {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) const struct blk_shadow *s;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) unsigned int grant_idx;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) unsigned int bvec_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) char *bvec_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) static void blkif_copy_from_grant(unsigned long gfn, unsigned int offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) unsigned int len, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) struct copy_from_grant *info = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) char *shared_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) /* Convenient aliases */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) const struct blk_shadow *s = info->s;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) shared_data = kmap_atomic(s->grants_used[info->grant_idx]->page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401) memcpy(info->bvec_data + info->bvec_offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) shared_data + offset, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) info->bvec_offset += len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) info->grant_idx++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) kunmap_atomic(shared_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) static enum blk_req_status blkif_rsp_to_req_status(int rsp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) switch (rsp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) case BLKIF_RSP_OKAY:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) return REQ_DONE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) case BLKIF_RSP_EOPNOTSUPP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) return REQ_EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) case BLKIF_RSP_ERROR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) return REQ_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) * Get the final status of the block request based on two ring response
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) static int blkif_get_final_status(enum blk_req_status s1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) enum blk_req_status s2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) BUG_ON(s1 < REQ_DONE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) BUG_ON(s2 < REQ_DONE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) if (s1 == REQ_ERROR || s2 == REQ_ERROR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) return BLKIF_RSP_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) else if (s1 == REQ_EOPNOTSUPP || s2 == REQ_EOPNOTSUPP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) return BLKIF_RSP_EOPNOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) return BLKIF_RSP_OKAY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) * Return values:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) * 1 response processed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) * 0 missing further responses.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) * -1 error while processing.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) static int blkif_completion(unsigned long *id,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) struct blkfront_ring_info *rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) struct blkif_response *bret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) int i = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) struct scatterlist *sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) int num_sg, num_grant;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) struct blk_shadow *s = &rinfo->shadow[*id];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) struct copy_from_grant data = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) .grant_idx = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) num_grant = s->req.operation == BLKIF_OP_INDIRECT ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) s->req.u.indirect.nr_segments : s->req.u.rw.nr_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) /* The I/O request may be split in two. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) if (unlikely(s->associated_id != NO_ASSOCIATED_ID)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) struct blk_shadow *s2 = &rinfo->shadow[s->associated_id];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) /* Keep the status of the current response in shadow. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) s->status = blkif_rsp_to_req_status(bret->status);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) /* Wait the second response if not yet here. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) if (s2->status < REQ_DONE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) bret->status = blkif_get_final_status(s->status,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) s2->status);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) * All the grants is stored in the first shadow in order
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) * to make the completion code simpler.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) num_grant += s2->req.u.rw.nr_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) * The two responses may not come in order. Only the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) * first request will store the scatter-gather list.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) if (s2->num_sg != 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) /* Update "id" with the ID of the first response. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) *id = s->associated_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) s = s2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) * We don't need anymore the second request, so recycling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) * it now.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) if (add_id_to_freelist(rinfo, s->associated_id))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) WARN(1, "%s: can't recycle the second part (id = %ld) of the request\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) info->gd->disk_name, s->associated_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) data.s = s;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) num_sg = s->num_sg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) if (bret->operation == BLKIF_OP_READ && info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) for_each_sg(s->sg, sg, num_sg, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) BUG_ON(sg->offset + sg->length > PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) data.bvec_offset = sg->offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) data.bvec_data = kmap_atomic(sg_page(sg));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) gnttab_foreach_grant_in_range(sg_page(sg),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) sg->offset,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) sg->length,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) blkif_copy_from_grant,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) &data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) kunmap_atomic(data.bvec_data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) /* Add the persistent grant into the list of free grants */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521) for (i = 0; i < num_grant; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) if (!gnttab_try_end_foreign_access(s->grants_used[i]->gref)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) * If the grant is still mapped by the backend (the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) * backend has chosen to make this grant persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) * we add it at the head of the list, so it will be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) * reused first.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) if (!info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) pr_alert("backed has not unmapped grant: %u\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) s->grants_used[i]->gref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) return -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) list_add(&s->grants_used[i]->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) rinfo->persistent_gnts_c++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) * If the grant is not mapped by the backend we add it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) * to the tail of the list, so it will not be picked
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) * again unless we run out of persistent grants.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) s->grants_used[i]->gref = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) list_add_tail(&s->grants_used[i]->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) if (s->req.operation == BLKIF_OP_INDIRECT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) for (i = 0; i < INDIRECT_GREFS(num_grant); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) if (!gnttab_try_end_foreign_access(s->indirect_grants[i]->gref)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) if (!info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) pr_alert("backed has not unmapped grant: %u\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) s->indirect_grants[i]->gref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) return -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) list_add(&s->indirect_grants[i]->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) rinfo->persistent_gnts_c++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) struct page *indirect_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) * Add the used indirect page back to the list of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) * available pages for indirect grefs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) if (!info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) indirect_page = s->indirect_grants[i]->page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) list_add(&indirect_page->lru, &rinfo->indirect_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) s->indirect_grants[i]->gref = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) list_add_tail(&s->indirect_grants[i]->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) static irqreturn_t blkif_interrupt(int irq, void *dev_id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) struct request *req;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) struct blkif_response bret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) RING_IDX i, rp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)dev_id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) xen_irq_lateeoi(irq, XEN_EOI_FLAG_SPURIOUS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) return IRQ_HANDLED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) spin_lock_irqsave(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) rp = READ_ONCE(rinfo->ring.sring->rsp_prod);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) virt_rmb(); /* Ensure we see queued responses up to 'rp'. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) if (RING_RESPONSE_PROD_OVERFLOW(&rinfo->ring, rp)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) pr_alert("%s: illegal number of responses %u\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) info->gd->disk_name, rp - rinfo->ring.rsp_cons);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) for (i = rinfo->ring.rsp_cons; i != rp; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) unsigned long id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) unsigned int op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) eoiflag = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) RING_COPY_RESPONSE(&rinfo->ring, i, &bret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) id = bret.id;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) * The backend has messed up and given us an id that we would
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) * never have given to it (we stamp it up to BLK_RING_SIZE -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) * look in get_id_from_freelist.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) if (id >= BLK_RING_SIZE(info)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) pr_alert("%s: response has incorrect id (%ld)\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) info->gd->disk_name, id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) if (rinfo->shadow[id].status != REQ_WAITING) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) pr_alert("%s: response references no pending request\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) info->gd->disk_name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) rinfo->shadow[id].status = REQ_PROCESSING;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) req = rinfo->shadow[id].request;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) op = rinfo->shadow[id].req.operation;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) if (op == BLKIF_OP_INDIRECT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) op = rinfo->shadow[id].req.u.indirect.indirect_op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) if (bret.operation != op) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) pr_alert("%s: response has wrong operation (%u instead of %u)\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) info->gd->disk_name, bret.operation, op);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) if (bret.operation != BLKIF_OP_DISCARD) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) * We may need to wait for an extra response if the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) * I/O request is split in 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) ret = blkif_completion(&id, rinfo, &bret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) if (unlikely(ret < 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) goto err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) if (add_id_to_freelist(rinfo, id)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654) info->gd->disk_name, op_name(bret.operation), id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) if (bret.status == BLKIF_RSP_OKAY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) blkif_req(req)->error = BLK_STS_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) blkif_req(req)->error = BLK_STS_IOERR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) switch (bret.operation) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) case BLKIF_OP_DISCARD:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) struct request_queue *rq = info->rq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) pr_warn_ratelimited("blkfront: %s: %s op failed\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) info->gd->disk_name, op_name(bret.operation));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) blkif_req(req)->error = BLK_STS_NOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) info->feature_discard = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) info->feature_secdiscard = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) blk_queue_flag_clear(QUEUE_FLAG_DISCARD, rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) blk_queue_flag_clear(QUEUE_FLAG_SECERASE, rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) case BLKIF_OP_FLUSH_DISKCACHE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) case BLKIF_OP_WRITE_BARRIER:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) pr_warn_ratelimited("blkfront: %s: %s op failed\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) info->gd->disk_name, op_name(bret.operation));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) blkif_req(req)->error = BLK_STS_NOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) if (unlikely(bret.status == BLKIF_RSP_ERROR &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) rinfo->shadow[id].req.u.rw.nr_segments == 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) pr_warn_ratelimited("blkfront: %s: empty %s op failed\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) info->gd->disk_name, op_name(bret.operation));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) blkif_req(req)->error = BLK_STS_NOTSUPP;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690) if (unlikely(blkif_req(req)->error)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) if (blkif_req(req)->error == BLK_STS_NOTSUPP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692) blkif_req(req)->error = BLK_STS_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) info->feature_fua = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) info->feature_flush = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) xlvbd_flush(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697) fallthrough;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) case BLKIF_OP_READ:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) case BLKIF_OP_WRITE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) if (unlikely(bret.status != BLKIF_RSP_OKAY))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) dev_dbg_ratelimited(&info->xbdev->dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702) "Bad return from blkdev data request: %#x\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) bret.status);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) if (likely(!blk_should_fake_timeout(req->q)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711) blk_mq_complete_request(req);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) rinfo->ring.rsp_cons = i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) if (i != rinfo->ring.req_prod_pvt) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) int more_to_do;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) RING_FINAL_CHECK_FOR_RESPONSES(&rinfo->ring, more_to_do);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) if (more_to_do)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720) goto again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) rinfo->ring.sring->rsp_event = i + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) kick_pending_request_queues_locked(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) xen_irq_lateeoi(irq, eoiflag);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) return IRQ_HANDLED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) err:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) info->connected = BLKIF_STATE_ERROR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) /* No EOI in order to avoid further interrupts. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) pr_alert("%s disabled for further use\n", info->gd->disk_name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740) return IRQ_HANDLED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) static int setup_blkring(struct xenbus_device *dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745) struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747) struct blkif_sring *sring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748) int err, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) unsigned long ring_size = info->nr_ring_pages * XEN_PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) grant_ref_t gref[XENBUS_MAX_RING_GRANTS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) for (i = 0; i < info->nr_ring_pages; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754) rinfo->ring_ref[i] = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) sring = alloc_pages_exact(ring_size, GFP_NOIO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) if (!sring) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) SHARED_RING_INIT(sring);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) FRONT_RING_INIT(&rinfo->ring, sring, ring_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764) err = xenbus_grant_ring(dev, rinfo->ring.sring, info->nr_ring_pages, gref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) if (err < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) free_pages_exact(sring, ring_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767) rinfo->ring.sring = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) for (i = 0; i < info->nr_ring_pages; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) rinfo->ring_ref[i] = gref[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) err = xenbus_alloc_evtchn(dev, &rinfo->evtchn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777) err = bind_evtchn_to_irqhandler_lateeoi(rinfo->evtchn, blkif_interrupt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) 0, "blkif", rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) if (err <= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) xenbus_dev_fatal(dev, err,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) "bind_evtchn_to_irqhandler failed");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784) rinfo->irq = err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788) blkif_free(info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793) * Write out per-ring/queue nodes including ring-ref and event-channel, and each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) * ring buffer may have multi pages depending on ->nr_ring_pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796) static int write_per_ring_nodes(struct xenbus_transaction xbt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) struct blkfront_ring_info *rinfo, const char *dir)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) const char *message = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) if (info->nr_ring_pages == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) err = xenbus_printf(xbt, dir, "ring-ref", "%u", rinfo->ring_ref[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807) message = "writing ring-ref";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811) for (i = 0; i < info->nr_ring_pages; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) char ring_ref_name[RINGREF_NAME_LEN];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) snprintf(ring_ref_name, RINGREF_NAME_LEN, "ring-ref%u", i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) err = xenbus_printf(xbt, dir, ring_ref_name,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) "%u", rinfo->ring_ref[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) message = "writing ring-ref";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) err = xenbus_printf(xbt, dir, "event-channel", "%u", rinfo->evtchn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) message = "writing event-channel";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) abort_transaction:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) xenbus_transaction_end(xbt, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834) if (message)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) xenbus_dev_fatal(info->xbdev, err, "%s", message);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) static void free_info(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842) list_del(&info->info_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) kfree(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846) /* Common code used when first setting up, and when resuming. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) static int talk_to_blkback(struct xenbus_device *dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848) struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) const char *message = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851) struct xenbus_transaction xbt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853) unsigned int i, max_page_order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854) unsigned int ring_page_order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857) if (!info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860) max_page_order = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861) "max-ring-page-order", 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) ring_page_order = min(xen_blkif_max_ring_order, max_page_order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863) info->nr_ring_pages = 1 << ring_page_order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865) err = negotiate_mq(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) /* Create shared ring, alloc event channel. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871) err = setup_blkring(dev, rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876) again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877) err = xenbus_transaction_start(&xbt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879) xenbus_dev_fatal(dev, err, "starting transaction");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) if (info->nr_ring_pages > 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884) err = xenbus_printf(xbt, dev->nodename, "ring-page-order", "%u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885) ring_page_order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887) message = "writing ring-page-order";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892) /* We already got the number of queues/rings in _probe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893) if (info->nr_rings == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) err = write_per_ring_nodes(xbt, info->rinfo, dev->nodename);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898) char *path;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899) size_t pathsize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) err = xenbus_printf(xbt, dev->nodename, "multi-queue-num-queues", "%u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902) info->nr_rings);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) message = "writing multi-queue-num-queues";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908) pathsize = strlen(dev->nodename) + QUEUE_NAME_LEN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) path = kmalloc(pathsize, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910) if (!path) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) err = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) message = "ENOMEM while writing ring references";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917) memset(path, 0, pathsize);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918) snprintf(path, pathsize, "%s/queue-%u", dev->nodename, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) err = write_per_ring_nodes(xbt, rinfo, path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) kfree(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925) kfree(path);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927) err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928) XEN_IO_PROTO_ABI_NATIVE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) message = "writing protocol";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) goto abort_transaction;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933) err = xenbus_printf(xbt, dev->nodename, "feature-persistent", "%u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) info->feature_persistent);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936) dev_warn(&dev->dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) "writing persistent grants feature to xenbus");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939) err = xenbus_transaction_end(xbt, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) if (err == -EAGAIN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) goto again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943) xenbus_dev_fatal(dev, err, "completing transaction");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944) goto destroy_blkring;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) unsigned int j;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) for (j = 0; j < BLK_RING_SIZE(info); j++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951) rinfo->shadow[j].req.u.rw.id = j + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952) rinfo->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954) xenbus_switch_state(dev, XenbusStateInitialised);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958) abort_transaction:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959) xenbus_transaction_end(xbt, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960) if (message)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) xenbus_dev_fatal(dev, err, "%s", message);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962) destroy_blkring:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963) blkif_free(info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966) free_info(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969) dev_set_drvdata(&dev->dev, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974) static int negotiate_mq(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976) unsigned int backend_max_queues;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980) BUG_ON(info->nr_rings);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) /* Check if backend supports multiple queues. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) backend_max_queues = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984) "multi-queue-max-queues", 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985) info->nr_rings = min(backend_max_queues, xen_blkif_max_queues);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986) /* We need at least one ring. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987) if (!info->nr_rings)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988) info->nr_rings = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990) info->rinfo_size = struct_size(info->rinfo, shadow,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) BLK_RING_SIZE(info));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) info->rinfo = kvcalloc(info->nr_rings, info->rinfo_size, GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993) if (!info->rinfo) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995) info->nr_rings = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000) INIT_LIST_HEAD(&rinfo->indirect_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001) INIT_LIST_HEAD(&rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002) rinfo->dev_info = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) INIT_WORK(&rinfo->work, blkif_restart_queue);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) spin_lock_init(&rinfo->ring_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009) /* Enable the persistent grants feature. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010) static bool feature_persistent = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011) module_param(feature_persistent, bool, 0644);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) MODULE_PARM_DESC(feature_persistent,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013) "Enables the persistent grants feature");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016) * Entry point to this code when a new device is created. Allocate the basic
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) * structures and the ring buffer for communication with the backend, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) * inform the backend of the appropriate details for those. Switch to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) * Initialised state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) static int blkfront_probe(struct xenbus_device *dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) const struct xenbus_device_id *id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) int err, vdevice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025) struct blkfront_info *info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) /* FIXME: Use dynamic device id if this is not set. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) err = xenbus_scanf(XBT_NIL, dev->nodename,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029) "virtual-device", "%i", &vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030) if (err != 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) /* go looking in the extended area instead */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032) err = xenbus_scanf(XBT_NIL, dev->nodename, "virtual-device-ext",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033) "%i", &vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) if (err != 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) xenbus_dev_fatal(dev, err, "reading virtual-device");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040) if (xen_hvm_domain()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) char *type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042) int len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) /* no unplug has been done: do not hook devices != xen vbds */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) if (xen_has_pv_and_legacy_disk_devices()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) int major;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047) if (!VDEV_IS_EXTENDED(vdevice))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048) major = BLKIF_MAJOR(vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050) major = XENVBD_MAJOR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052) if (major != XENVBD_MAJOR) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) printk(KERN_INFO
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) "%s: HVM does not support vbd %d as xen block device\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) __func__, vdevice);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059) /* do not create a PV cdrom device if we are an HVM guest */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) type = xenbus_read(XBT_NIL, dev->nodename, "device-type", &len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) if (IS_ERR(type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) if (strncmp(type, "cdrom", 5) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064) kfree(type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067) kfree(type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) info = kzalloc(sizeof(*info), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070) if (!info) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071) xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075) info->xbdev = dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) mutex_init(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078) info->vdevice = vdevice;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079) info->connected = BLKIF_STATE_DISCONNECTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081) info->feature_persistent = feature_persistent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) /* Front end dir is a number, which is used as the id. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) info->handle = simple_strtoul(strrchr(dev->nodename, '/')+1, NULL, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085) dev_set_drvdata(&dev->dev, info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) list_add(&info->info_list, &info_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094) static int blkif_recover(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096) unsigned int r_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) struct request *req, *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098) int rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099) struct bio *bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) unsigned int segs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103) blkfront_gather_backend_features(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) /* Reset limits changed by blk_mq_update_nr_hw_queues(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105) blkif_set_queue_limits(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) segs = info->max_indirect_segments ? : BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107) blk_queue_max_segments(info->rq, segs / GRANTS_PER_PSEG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) for_each_rinfo(info, rinfo, r_index) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110) rc = blkfront_setup_indirect(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111) if (rc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112) return rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) xenbus_switch_state(info->xbdev, XenbusStateConnected);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116) /* Now safe for us to use the shared ring */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117) info->connected = BLKIF_STATE_CONNECTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119) for_each_rinfo(info, rinfo, r_index) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120) /* Kick any other new requests queued since we resumed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) kick_pending_request_queues(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124) list_for_each_entry_safe(req, n, &info->requests, queuelist) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) /* Requeue pending requests (flush or discard) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126) list_del_init(&req->queuelist);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127) BUG_ON(req->nr_phys_segments > segs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128) blk_mq_requeue_request(req, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) blk_mq_start_stopped_hw_queues(info->rq, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131) blk_mq_kick_requeue_list(info->rq);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133) while ((bio = bio_list_pop(&info->bio_list)) != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) /* Traverse the list of pending bios and re-queue them */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) submit_bio(bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) * We are reconnecting to the backend, due to a suspend/resume, or a backend
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143) * driver restart. We tear down our blkif structure and recreate it, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144) * leave the device-layer structures intact so that this is transparent to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) * rest of the kernel.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) static int blkfront_resume(struct xenbus_device *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) struct blkfront_info *info = dev_get_drvdata(&dev->dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150) int err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151) unsigned int i, j;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154) dev_dbg(&dev->dev, "blkfront_resume: %s\n", dev->nodename);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) bio_list_init(&info->bio_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157) INIT_LIST_HEAD(&info->requests);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) struct bio_list merge_bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) struct blk_shadow *shadow = rinfo->shadow;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162) for (j = 0; j < BLK_RING_SIZE(info); j++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163) /* Not in use? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164) if (!shadow[j].request)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168) * Get the bios in the request so we can re-queue them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170) if (req_op(shadow[j].request) == REQ_OP_FLUSH ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) req_op(shadow[j].request) == REQ_OP_DISCARD ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172) req_op(shadow[j].request) == REQ_OP_SECURE_ERASE ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173) shadow[j].request->cmd_flags & REQ_FUA) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175) * Flush operations don't contain bios, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176) * we need to requeue the whole request
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) * XXX: but this doesn't make any sense for a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179) * write with the FUA flag set..
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181) list_add(&shadow[j].request->queuelist, &info->requests);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184) merge_bio.head = shadow[j].request->bio;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) merge_bio.tail = shadow[j].request->biotail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186) bio_list_merge(&info->bio_list, &merge_bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187) shadow[j].request->bio = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188) blk_mq_end_request(shadow[j].request, BLK_STS_OK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192) blkif_free(info, info->connected == BLKIF_STATE_CONNECTED);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194) err = talk_to_blkback(dev, info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195) if (!err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) blk_mq_update_nr_hw_queues(&info->tag_set, info->nr_rings);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199) * We have to wait for the backend to switch to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200) * connected state, since we want to read which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) * features it supports.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) static void blkfront_closing(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) struct xenbus_device *xbdev = info->xbdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) struct block_device *bdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) mutex_lock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) if (xbdev->state == XenbusStateClosing) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219) if (info->gd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220) bdev = bdget_disk(info->gd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) if (!bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225) xenbus_frontend_closed(xbdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229) mutex_lock(&bdev->bd_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) if (bdev->bd_openers) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232) xenbus_dev_error(xbdev, -EBUSY,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) "Device in use; refusing to close");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) xenbus_switch_state(xbdev, XenbusStateClosing);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236) xlvbd_release_gendisk(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237) xenbus_frontend_closed(xbdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240) mutex_unlock(&bdev->bd_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241) bdput(bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244) static void blkfront_setup_discard(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246) info->feature_discard = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247) info->discard_granularity = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) "discard-granularity",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250) info->discard_alignment = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251) "discard-alignment", 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) info->feature_secdiscard =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) !!xenbus_read_unsigned(info->xbdev->otherend, "discard-secure",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254) 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257) static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259) unsigned int psegs, grants, memflags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) int err, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261) struct blkfront_info *info = rinfo->dev_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) memflags = memalloc_noio_save();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) if (info->max_indirect_segments == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) if (!HAS_EXTRA_REQ)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267) grants = BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270) * When an extra req is required, the maximum
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) * grants supported is related to the size of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) * Linux block segment.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) grants = GRANTS_PER_PSEG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) grants = info->max_indirect_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279) psegs = DIV_ROUND_UP(grants, GRANTS_PER_PSEG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) err = fill_grant_buffer(rinfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282) (grants + INDIRECT_GREFS(grants)) * BLK_RING_SIZE(info));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) goto out_of_memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) if (!info->feature_persistent && info->max_indirect_segments) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) * We are using indirect descriptors but not persistent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289) * grants, we need to allocate a set of pages that can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) * used for mapping indirect grefs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292) int num = INDIRECT_GREFS(grants) * BLK_RING_SIZE(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) BUG_ON(!list_empty(&rinfo->indirect_pages));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295) for (i = 0; i < num; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) struct page *indirect_page = alloc_page(GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297) if (!indirect_page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298) goto out_of_memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299) list_add(&indirect_page->lru, &rinfo->indirect_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303) for (i = 0; i < BLK_RING_SIZE(info); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304) rinfo->shadow[i].grants_used =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305) kvcalloc(grants,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306) sizeof(rinfo->shadow[i].grants_used[0]),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308) rinfo->shadow[i].sg = kvcalloc(psegs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) sizeof(rinfo->shadow[i].sg[0]),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311) if (info->max_indirect_segments)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) rinfo->shadow[i].indirect_grants =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313) kvcalloc(INDIRECT_GREFS(grants),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314) sizeof(rinfo->shadow[i].indirect_grants[0]),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) if ((rinfo->shadow[i].grants_used == NULL) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317) (rinfo->shadow[i].sg == NULL) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) (info->max_indirect_segments &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) (rinfo->shadow[i].indirect_grants == NULL)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320) goto out_of_memory;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321) sg_init_table(rinfo->shadow[i].sg, psegs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) memalloc_noio_restore(memflags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328) out_of_memory:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) for (i = 0; i < BLK_RING_SIZE(info); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) kvfree(rinfo->shadow[i].grants_used);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331) rinfo->shadow[i].grants_used = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332) kvfree(rinfo->shadow[i].sg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) rinfo->shadow[i].sg = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) kvfree(rinfo->shadow[i].indirect_grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335) rinfo->shadow[i].indirect_grants = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337) if (!list_empty(&rinfo->indirect_pages)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) struct page *indirect_page, *n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339) list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340) list_del(&indirect_page->lru);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341) __free_page(indirect_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) memalloc_noio_restore(memflags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351) * Gather all backend feature-*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353) static void blkfront_gather_backend_features(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) unsigned int indirect_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357) info->feature_flush = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) info->feature_fua = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) * If there's no "feature-barrier" defined, then it means
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362) * we're dealing with a very old backend which writes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) * synchronously; nothing to do.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365) * If there are barriers, then we use flush.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) if (xenbus_read_unsigned(info->xbdev->otherend, "feature-barrier", 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368) info->feature_flush = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) info->feature_fua = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373) * And if there is "feature-flush-cache" use that above
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) * barriers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376) if (xenbus_read_unsigned(info->xbdev->otherend, "feature-flush-cache",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377) 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378) info->feature_flush = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) info->feature_fua = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382) if (xenbus_read_unsigned(info->xbdev->otherend, "feature-discard", 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) blkfront_setup_discard(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) if (info->feature_persistent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386) info->feature_persistent =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) !!xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388) "feature-persistent", 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) indirect_segments = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391) "feature-max-indirect-segments", 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) if (indirect_segments > xen_blkif_max_segments)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393) indirect_segments = xen_blkif_max_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) if (indirect_segments <= BLKIF_MAX_SEGMENTS_PER_REQUEST)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) indirect_segments = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396) info->max_indirect_segments = indirect_segments;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) if (info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) schedule_delayed_work(&blkfront_work, HZ * 10);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406) * Invoked when the backend is finally 'ready' (and has told produced
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407) * the details about the physical device - #sectors, size, etc).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) static void blkfront_connect(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) unsigned long long sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412) unsigned long sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) unsigned int physical_sector_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414) unsigned int binfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415) int err, i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) switch (info->connected) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) case BLKIF_STATE_CONNECTED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) * Potentially, the back-end may be signalling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422) * a capacity change; update the capacity.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425) "sectors", "%Lu", §ors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426) if (XENBUS_EXIST_ERR(err))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428) printk(KERN_INFO "Setting capacity to %Lu\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429) sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) set_capacity_revalidate_and_notify(info->gd, sectors, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433) case BLKIF_STATE_SUSPENDED:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) * If we are recovering from suspension, we need to wait
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) * for the backend to announce it's features before
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437) * reconnecting, at least we need to know if the backend
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438) * supports indirect descriptors, and how many.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) blkif_recover(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) dev_dbg(&info->xbdev->dev, "%s:%s.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448) __func__, info->xbdev->otherend);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) "sectors", "%llu", §ors,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452) "info", "%u", &binfo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) "sector-size", "%lu", §or_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) xenbus_dev_fatal(info->xbdev, err,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457) "reading backend fields at %s",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) info->xbdev->otherend);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) * physcial-sector-size is a newer field, so old backends may not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) * provide this. Assume physical sector size to be the same as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465) * sector_size in that case.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467) physical_sector_size = xenbus_read_unsigned(info->xbdev->otherend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468) "physical-sector-size",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469) sector_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470) blkfront_gather_backend_features(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) err = blkfront_setup_indirect(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474) xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475) info->xbdev->otherend);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) blkif_free(info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481) err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) physical_sector_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2485) info->xbdev->otherend);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2486) goto fail;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2487) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2488)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2489) xenbus_switch_state(info->xbdev, XenbusStateConnected);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2491) /* Kick pending requests. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2492) info->connected = BLKIF_STATE_CONNECTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2493) for_each_rinfo(info, rinfo, i)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2494) kick_pending_request_queues(rinfo);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2496) device_add_disk(&info->xbdev->dev, info->gd, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2497)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2498) info->is_ready = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2499) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2500)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2501) fail:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2502) blkif_free(info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2503) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2504) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2505)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2506) /**
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2507) * Callback received when the backend's state changes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2508) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2509) static void blkback_changed(struct xenbus_device *dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2510) enum xenbus_state backend_state)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2511) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2512) struct blkfront_info *info = dev_get_drvdata(&dev->dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2514) dev_dbg(&dev->dev, "blkfront:blkback_changed to state %d.\n", backend_state);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2515)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2516) switch (backend_state) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2517) case XenbusStateInitWait:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2518) if (dev->state != XenbusStateInitialising)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2519) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2520) if (talk_to_blkback(dev, info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2521) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2522) case XenbusStateInitialising:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2523) case XenbusStateInitialised:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2524) case XenbusStateReconfiguring:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2525) case XenbusStateReconfigured:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2526) case XenbusStateUnknown:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2527) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2529) case XenbusStateConnected:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2530) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2531) * talk_to_blkback sets state to XenbusStateInitialised
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2532) * and blkfront_connect sets it to XenbusStateConnected
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2533) * (if connection went OK).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2534) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2535) * If the backend (or toolstack) decides to poke at backend
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2536) * state (and re-trigger the watch by setting the state repeatedly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2537) * to XenbusStateConnected (4)) we need to deal with this.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2538) * This is allowed as this is used to communicate to the guest
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2539) * that the size of disk has changed!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2540) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2541) if ((dev->state != XenbusStateInitialised) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2542) (dev->state != XenbusStateConnected)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2543) if (talk_to_blkback(dev, info))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2544) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2545) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2546)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2547) blkfront_connect(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2548) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2549)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2550) case XenbusStateClosed:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2551) if (dev->state == XenbusStateClosed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2552) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2553) fallthrough;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2554) case XenbusStateClosing:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2555) if (info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2556) blkfront_closing(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2557) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2558) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2559) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2560)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2561) static int blkfront_remove(struct xenbus_device *xbdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2562) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2563) struct blkfront_info *info = dev_get_drvdata(&xbdev->dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2564) struct block_device *bdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2565) struct gendisk *disk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2567) dev_dbg(&xbdev->dev, "%s removed", xbdev->nodename);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2569) if (!info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2570) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2572) blkif_free(info, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2574) mutex_lock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2576) disk = info->gd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2577) if (disk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2578) bdev = bdget_disk(disk, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2580) info->xbdev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2581) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2583) if (!bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2584) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2585) free_info(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2586) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2587) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2588) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2589)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2590) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2591) * The xbdev was removed before we reached the Closed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2592) * state. See if it's safe to remove the disk. If the bdev
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2593) * isn't closed yet, we let release take care of it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2594) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2596) mutex_lock(&bdev->bd_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2597) info = disk->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2598)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2599) dev_warn(disk_to_dev(disk),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2600) "%s was hot-unplugged, %d stale handles\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2601) xbdev->nodename, bdev->bd_openers);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2602)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2603) if (info && !bdev->bd_openers) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2604) xlvbd_release_gendisk(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2605) disk->private_data = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2606) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2607) free_info(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2608) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2609) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2610)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2611) mutex_unlock(&bdev->bd_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2612) bdput(bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2614) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2616)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2617) static int blkfront_is_ready(struct xenbus_device *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2618) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2619) struct blkfront_info *info = dev_get_drvdata(&dev->dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2620)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2621) return info->is_ready && info->xbdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2622) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2623)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2624) static int blkif_open(struct block_device *bdev, fmode_t mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2625) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2626) struct gendisk *disk = bdev->bd_disk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2627) struct blkfront_info *info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2628) int err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2629)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2630) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2631)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2632) info = disk->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2633) if (!info) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2634) /* xbdev gone */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2635) err = -ERESTARTSYS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2636) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2637) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2638)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2639) mutex_lock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2641) if (!info->gd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2642) /* xbdev is closed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2643) err = -ERESTARTSYS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2644)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2645) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2647) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2648) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2649) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2650) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2651)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2652) static void blkif_release(struct gendisk *disk, fmode_t mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2653) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2654) struct blkfront_info *info = disk->private_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2655) struct block_device *bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2656) struct xenbus_device *xbdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2658) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2659)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2660) bdev = bdget_disk(disk, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2661)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2662) if (!bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2663) WARN(1, "Block device %s yanked out from us!\n", disk->disk_name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2664) goto out_mutex;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2665) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2666) if (bdev->bd_openers)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2667) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2669) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2670) * Check if we have been instructed to close. We will have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2671) * deferred this request, because the bdev was still open.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2672) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2673)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2674) mutex_lock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2675) xbdev = info->xbdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2676)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2677) if (xbdev && xbdev->state == XenbusStateClosing) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2678) /* pending switch to state closed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2679) dev_info(disk_to_dev(bdev->bd_disk), "releasing disk\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2680) xlvbd_release_gendisk(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2681) xenbus_frontend_closed(info->xbdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2682) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2683)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2684) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2686) if (!xbdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2687) /* sudden device removal */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2688) dev_info(disk_to_dev(bdev->bd_disk), "releasing disk\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2689) xlvbd_release_gendisk(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2690) disk->private_data = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2691) free_info(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2692) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2693)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2694) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2695) bdput(bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2696) out_mutex:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2697) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2698) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2699)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2700) static const struct block_device_operations xlvbd_block_fops =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2701) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2702) .owner = THIS_MODULE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2703) .open = blkif_open,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2704) .release = blkif_release,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2705) .getgeo = blkif_getgeo,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2706) .ioctl = blkif_ioctl,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2707) .compat_ioctl = blkdev_compat_ptr_ioctl,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2708) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2709)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2711) static const struct xenbus_device_id blkfront_ids[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2712) { "vbd" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2713) { "" }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2714) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2716) static struct xenbus_driver blkfront_driver = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2717) .ids = blkfront_ids,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2718) .probe = blkfront_probe,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2719) .remove = blkfront_remove,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2720) .resume = blkfront_resume,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2721) .otherend_changed = blkback_changed,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2722) .is_ready = blkfront_is_ready,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2723) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2725) static void purge_persistent_grants(struct blkfront_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2726) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2727) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2728) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2729) struct blkfront_ring_info *rinfo;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2731) for_each_rinfo(info, rinfo, i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2732) struct grant *gnt_list_entry, *tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2733)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2734) spin_lock_irqsave(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2735)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2736) if (rinfo->persistent_gnts_c == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2737) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2738) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2739) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2741) list_for_each_entry_safe(gnt_list_entry, tmp, &rinfo->grants,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2742) node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2743) if (gnt_list_entry->gref == GRANT_INVALID_REF ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2744) !gnttab_try_end_foreign_access(gnt_list_entry->gref))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2745) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2746)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2747) list_del(&gnt_list_entry->node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2748) rinfo->persistent_gnts_c--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2749) gnt_list_entry->gref = GRANT_INVALID_REF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2750) list_add_tail(&gnt_list_entry->node, &rinfo->grants);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2753) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2754) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2755) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2756)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2757) static void blkfront_delay_work(struct work_struct *work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2758) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2759) struct blkfront_info *info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2760) bool need_schedule_work = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2761)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2762) mutex_lock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2764) list_for_each_entry(info, &info_list, info_list) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2765) if (info->feature_persistent) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2766) need_schedule_work = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2767) mutex_lock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2768) purge_persistent_grants(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2769) mutex_unlock(&info->mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2770) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2771) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2773) if (need_schedule_work)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2774) schedule_delayed_work(&blkfront_work, HZ * 10);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2775)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2776) mutex_unlock(&blkfront_mutex);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2778)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2779) static int __init xlblk_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2780) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2781) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2782) int nr_cpus = num_online_cpus();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2784) if (!xen_domain())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2785) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2786)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2787) if (!xen_has_pv_disk_devices())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2788) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2789)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2790) if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2791) pr_warn("xen_blk: can't get major %d with name %s\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2792) XENVBD_MAJOR, DEV_NAME);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2793) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2794) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2795)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2796) if (xen_blkif_max_segments < BLKIF_MAX_SEGMENTS_PER_REQUEST)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2797) xen_blkif_max_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2798)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2799) if (xen_blkif_max_ring_order > XENBUS_MAX_RING_GRANT_ORDER) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2800) pr_info("Invalid max_ring_order (%d), will use default max: %d.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2801) xen_blkif_max_ring_order, XENBUS_MAX_RING_GRANT_ORDER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2802) xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2803) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2804)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2805) if (xen_blkif_max_queues > nr_cpus) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2806) pr_info("Invalid max_queues (%d), will use default max: %d.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2807) xen_blkif_max_queues, nr_cpus);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2808) xen_blkif_max_queues = nr_cpus;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2809) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2810)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2811) INIT_DELAYED_WORK(&blkfront_work, blkfront_delay_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2812)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2813) ret = xenbus_register_frontend(&blkfront_driver);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2814) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2815) unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2816) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2817) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2818)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2819) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2820) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2821) module_init(xlblk_init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2823)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2824) static void __exit xlblk_exit(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2825) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2826) cancel_delayed_work_sync(&blkfront_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2828) xenbus_unregister_driver(&blkfront_driver);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2829) unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2830) kfree(minors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2831) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2832) module_exit(xlblk_exit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2833)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2834) MODULE_DESCRIPTION("Xen virtual block device frontend");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2835) MODULE_LICENSE("GPL");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2836) MODULE_ALIAS_BLOCKDEV_MAJOR(XENVBD_MAJOR);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2837) MODULE_ALIAS("xen:vbd");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2838) MODULE_ALIAS("xenblk");
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags