/****************************************************************************** * * Back-end of the driver for virtual block devices. This portion of the * driver exports a 'unified' block-device interface that can be accessed * by any operating system that implements a compatible front end. A * reference front-end implementation can be found in: * drivers/block/xen-blkfront.c * * Copyright (c) 2003-2004, Keir Fraser & Steve Hand * Copyright (c) 2005, Christopher Clark * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License version 2 * as published by the Free Software Foundation; or, when distributed * separately from the Linux kernel or incorporated into other * software packages, subject to the following license: * * Permission is hereby granted, free of charge, to any person obtaining a copy * of this source file (the "Software"), to deal in the Software without * restriction, including without limitation the rights to use, copy, modify, * merge, publish, distribute, sublicense, and/or sell copies of the Software, * and to permit persons to whom the Software is furnished to do so, subject to * the following conditions: * * The above copyright notice and this permission notice shall be included in * all copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS * IN THE SOFTWARE.
*/
/* * Maximum number of unused free pages to keep in the internal buffer. * Setting this to a value too low will reduce memory used in each backend, * but can have a performance penalty. * * A sane value is xen_blkif_reqs * BLKIF_MAX_SEGMENTS_PER_REQUEST, but can * be set to a lower value that might degrade performance on some intensive * IO workloads.
*/
staticint max_buffer_pages = 1024;
module_param_named(max_buffer_pages, max_buffer_pages, int, 0644);
MODULE_PARM_DESC(max_buffer_pages, "Maximum number of free pages to keep in each block backend buffer");
/* * Maximum number of grants to map persistently in blkback. For maximum * performance this should be the total numbers of grants that can be used * to fill the ring, but since this might become too high, specially with * the use of indirect descriptors, we set it to a value that provides good * performance without using too much memory. * * When the list of persistent grants is full we clean it up using a LRU * algorithm.
*/
staticint max_pgrants = 1056;
module_param_named(max_persistent_grants, max_pgrants, int, 0644);
MODULE_PARM_DESC(max_persistent_grants, "Maximum number of grants to map persistently");
/* * How long a persistent grant is allowed to remain allocated without being in * use. The time is in seconds, 0 means indefinitely long.
*/
staticunsignedint pgrant_timeout = 60;
module_param_named(persistent_grant_unused_seconds, pgrant_timeout,
uint, 0644);
MODULE_PARM_DESC(persistent_grant_unused_seconds, "Time in seconds an unused persistent grant is allowed to " "remain allocated. Default is 60, 0 means unlimited.");
/* * Maximum number of rings/queues blkback supports, allow as many queues as there * are CPUs if user has not specified a value.
*/ unsignedint xenblk_max_queues;
module_param_named(max_queues, xenblk_max_queues, uint, 0644);
MODULE_PARM_DESC(max_queues, "Maximum number of hardware queues per virtual disk." \ "By default it is the number of online CPUs.");
/* * Maximum order of pages to be used for the shared ring between front and * backend, 4KB page granularity is used.
*/ unsignedint xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
module_param_named(max_ring_page_order, xen_blkif_max_ring_order, int, 0444);
MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the shared ring"); /* * The LRU mechanism to clean the lists of persistent grants needs to * be executed periodically. The time interval between consecutive executions * of the purge mechanism is set in ms.
*/ #define LRU_INTERVAL 100
/* * When the persistent grants list is full we will remove unused grants * from the list. The percent number of grants to be removed at each LRU * execution.
*/ #define LRU_PERCENT_CLEAN 5
/* * We don't need locking around the persistent grant helpers * because blkback uses a single-thread for each backend, so we * can be sure that this functions will never be called recursively. * * The only exception to that is put_persistent_grant, that can be called * from interrupt context (by xen_blkbk_unmap), so we have to use atomic * bit operations to modify the flags of a persistent grant and to count * the number of used grants.
*/ staticint add_persistent_gnt(struct xen_blkif_ring *ring, struct persistent_gnt *persistent_gnt)
{ struct rb_node **new = NULL, *parent = NULL; struct persistent_gnt *this; struct xen_blkif *blkif = ring->blkif;
if (ring->persistent_gnt_c >= max_pgrants) { if (!blkif->vbd.overflow_max_grants)
blkif->vbd.overflow_max_grants = 1; return -EBUSY;
} /* Figure out where to put new node */ new = &ring->persistent_gnts.rb_node; while (*new) { this = container_of(*new, struct persistent_gnt, node);
parent = *new; if (persistent_gnt->gnt < this->gnt) new = &((*new)->rb_left); elseif (persistent_gnt->gnt > this->gnt) new = &((*new)->rb_right); else {
pr_alert_ratelimited("trying to add a gref that's already in the tree\n"); return -EINVAL;
}
}
if (work_busy(&ring->persistent_purge_work)) {
pr_alert_ratelimited("Scheduled work from previous purge is still busy, cannot purge list\n"); goto out;
}
/* * At this point, we can assure that there will be no calls * to get_persistent_grant (because we are executing this code from * xen_blkif_schedule), there can only be calls to put_persistent_gnt, * which means that the number of currently used grants will go down, * but never up, so we will always be able to remove the requested * number of grants.
*/
if (persistent_gnt->active) continue; if (!scan_used && !persistent_gnt_timeout(persistent_gnt)) continue; if (scan_used && total >= num_clean) continue;
rb_erase(&persistent_gnt->node, root);
list_add(&persistent_gnt->remove_node,
&ring->persistent_purge_list);
total++;
} /* * Check whether we also need to start cleaning * grants that were used since last purge in order to cope * with the requested num
*/ if (!scan_used && total < num_clean) {
pr_debug("Still missing %u purged frames\n", num_clean - total);
scan_used = true; goto purge_list;
}
if (total) {
ring->persistent_gnt_c -= total;
ring->blkif->vbd.overflow_max_grants = 0;
/* We can defer this work */
schedule_work(&ring->persistent_purge_work);
pr_debug("Purged %u/%u\n", num_clean, total);
}
out: return;
}
/* * Retrieve from the 'pending_reqs' a free pending_req structure to be used.
*/ staticstruct pending_req *alloc_req(struct xen_blkif_ring *ring)
{ struct pending_req *req = NULL; unsignedlong flags;
/* * Return the 'pending_req' structure back to the freepool. We also * wake up the thread if it was waiting for a free page.
*/ staticvoid free_req(struct xen_blkif_ring *ring, struct pending_req *req)
{ unsignedlong flags; int was_empty;
pr_info("VBD Resize: Domid: %d, Device: (%d, %d)\n",
blkif->domid, MAJOR(vbd->pdevice), MINOR(vbd->pdevice));
pr_info("VBD Resize: new size %llu\n", new_size);
vbd->size = new_size;
again:
err = xenbus_transaction_start(&xbt); if (err) {
pr_warn("Error starting transaction\n"); return;
}
err = xenbus_printf(xbt, dev->nodename, "sectors", "%llu",
(unsignedlonglong)vbd_sz(vbd)); if (err) {
pr_warn("Error writing new size\n"); goto abort;
} /* * Write the current state; we will use this to synchronize * the front-end. If the current state is "connected" the * front-end will get the new size information online.
*/
err = xenbus_printf(xbt, dev->nodename, "state", "%d", dev->state); if (err) {
pr_warn("Error writing the state\n"); goto abort;
}
err = xenbus_transaction_end(xbt, 0); if (err == -EAGAIN) goto again; if (err)
pr_warn("Error ending transaction\n"); return;
abort:
xenbus_transaction_end(xbt, 1);
}
/* * Notification from the guest OS.
*/ staticvoid blkif_notify_work(struct xen_blkif_ring *ring)
{
ring->waiting_reqs = 1;
wake_up(&ring->wq);
}
/* Shrink the free pages pool if it is too large. */ if (time_before(jiffies, blkif->buffer_squeeze_end))
gnttab_page_cache_shrink(&ring->free_pages, 0); else
gnttab_page_cache_shrink(&ring->free_pages,
max_buffer_pages);
if (log_stats && time_after(jiffies, ring->st_print))
print_stats(ring);
}
/* Drain pending purge work */
flush_work(&ring->persistent_purge_work);
if (log_stats)
print_stats(ring);
ring->xenblkd = NULL;
return 0;
}
/* * Remove persistent grants and empty the pool of free pages
*/ void xen_blkbk_free_caches(struct xen_blkif_ring *ring)
{ /* Free all persistent grant pages */
free_persistent_gnts(ring);
/* Since we are shutting down remove all pages from the buffer */
gnttab_page_cache_shrink(&ring->free_pages, 0 /* All */);
}
/* BUG_ON used to reproduce existing behaviour,
but is this the best way to deal with this? */
BUG_ON(result);
gnttab_page_cache_put(&ring->free_pages, data->pages, data->count);
make_response(ring, pending_req->id,
pending_req->operation, pending_req->status);
free_req(ring, pending_req); /* * Make sure the request is freed before releasing blkif, * or there could be a race between free_req and the * cleanup done in xen_blkif_free during shutdown. * * NB: The fact that we might try to wake up pending_free_wq * before drain_complete (in case there's a drain going on) * it's not a problem with our current implementation * because we can assure there's no thread waiting on * pending_free_wq if there's a drain going on, but it has * to be taken into account if the current model is changed.
*/ if (atomic_dec_and_test(&ring->inflight) && atomic_read(&blkif->drain)) {
complete(&blkif->drain_complete);
}
xen_blkif_put(blkif);
}
/* * Unmap the grant references. * * This could accumulate ops up to the batch size to reduce the number * of hypercalls, but since this is only used in error paths there's * no real need.
*/ staticvoid xen_blkbk_unmap(struct xen_blkif_ring *ring, struct grant_page *pages[], int num)
{ struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct page *unmap_pages[BLKIF_MAX_SEGMENTS_PER_REQUEST]; unsignedint invcount = 0; int ret;
while (num) { unsignedint batch = min(num, BLKIF_MAX_SEGMENTS_PER_REQUEST);
invcount = xen_blkbk_unmap_prepare(ring, pages, batch,
unmap, unmap_pages); if (invcount) {
ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
BUG_ON(ret);
gnttab_page_cache_put(&ring->free_pages, unmap_pages,
invcount);
}
pages += batch;
num -= batch;
}
}
staticint xen_blkbk_map(struct xen_blkif_ring *ring, struct grant_page *pages[], int num, bool ro)
{ struct gnttab_map_grant_ref map[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct page *pages_to_gnt[BLKIF_MAX_SEGMENTS_PER_REQUEST]; struct persistent_gnt *persistent_gnt = NULL;
phys_addr_t addr = 0; int i, seg_idx, new_map_idx; int segs_to_map = 0; int ret = 0; int last_map = 0, map_until = 0; int use_persistent_gnts; struct xen_blkif *blkif = ring->blkif;
/* * Fill out preq.nr_sects with proper amount of sectors, and setup * assign map[..] with the PFN of the page in our domain with the * corresponding grant reference for each page.
*/
again: for (i = map_until; i < num; i++) {
uint32_t flags;
if (use_persistent_gnts) {
persistent_gnt = get_persistent_gnt(
ring,
pages[i]->gref);
}
if (persistent_gnt) { /* * We are using persistent grants and * the grant is already mapped
*/
pages[i]->page = persistent_gnt->page;
pages[i]->persistent_gnt = persistent_gnt;
} else { if (gnttab_page_cache_get(&ring->free_pages,
&pages[i]->page)) {
gnttab_page_cache_put(&ring->free_pages,
pages_to_gnt,
segs_to_map);
ret = -ENOMEM; goto out;
}
addr = vaddr(pages[i]->page);
pages_to_gnt[segs_to_map] = pages[i]->page;
pages[i]->persistent_gnt = NULL;
flags = GNTMAP_host_map; if (!use_persistent_gnts && ro)
flags |= GNTMAP_readonly;
gnttab_set_map_op(&map[segs_to_map++], addr,
flags, pages[i]->gref,
blkif->domid);
}
map_until = i + 1; if (segs_to_map == BLKIF_MAX_SEGMENTS_PER_REQUEST) break;
}
if (segs_to_map)
ret = gnttab_map_refs(map, NULL, pages_to_gnt, segs_to_map);
/* * Now swizzle the MFN in our domain with the MFN from the other domain * so that when we access vaddr(pending_req,i) it has the contents of * the page from the other domain.
*/ for (seg_idx = last_map, new_map_idx = 0; seg_idx < map_until; seg_idx++) { if (!pages[seg_idx]->persistent_gnt) { /* This is a newly mapped grant */
BUG_ON(new_map_idx >= segs_to_map); if (unlikely(map[new_map_idx].status != 0)) {
pr_debug("invalid buffer -- could not remap it\n");
gnttab_page_cache_put(&ring->free_pages,
&pages[seg_idx]->page, 1);
pages[seg_idx]->handle = BLKBACK_INVALID_HANDLE;
ret |= !ret; goto next;
}
pages[seg_idx]->handle = map[new_map_idx].handle;
} else { continue;
} if (use_persistent_gnts &&
ring->persistent_gnt_c < max_pgrants) { /* * We are using persistent grants, the grant is * not mapped but we might have room for it.
*/
persistent_gnt = kmalloc(sizeof(struct persistent_gnt),
GFP_KERNEL); if (!persistent_gnt) { /* * If we don't have enough memory to * allocate the persistent_gnt struct * map this grant non-persistenly
*/ goto next;
}
persistent_gnt->gnt = map[new_map_idx].ref;
persistent_gnt->handle = map[new_map_idx].handle;
persistent_gnt->page = pages[seg_idx]->page; if (add_persistent_gnt(ring,
persistent_gnt)) {
kfree(persistent_gnt);
persistent_gnt = NULL; goto next;
}
pages[seg_idx]->persistent_gnt = persistent_gnt;
pr_debug("grant %u added to the tree of persistent grants, using %u/%u\n",
persistent_gnt->gnt, ring->persistent_gnt_c,
max_pgrants); goto next;
} if (use_persistent_gnts && !blkif->vbd.overflow_max_grants) {
blkif->vbd.overflow_max_grants = 1;
pr_debug("domain %u, device %#x is using maximum number of persistent grants\n",
blkif->domid, blkif->vbd.handle);
} /* * We could not map this grant persistently, so use it as * a non-persistent grant.
*/
next:
new_map_idx++;
}
segs_to_map = 0;
last_map = map_until; if (!ret && map_until != num) goto again;
out: for (i = last_map; i < num; i++) { /* Don't zap current batch's valid persistent grants. */ if (i >= map_until)
pages[i]->persistent_gnt = NULL;
pages[i]->handle = BLKBACK_INVALID_HANDLE;
}
return ret;
}
staticint xen_blkbk_map_seg(struct pending_req *pending_req)
{ int rc;
fail_response: if (err == -EOPNOTSUPP) {
pr_debug("discard op failed, not supported\n");
status = BLKIF_RSP_EOPNOTSUPP;
} elseif (err)
status = BLKIF_RSP_ERROR;
atomic_set(&blkif->drain, 1); do { if (atomic_read(&ring->inflight) == 0) break;
wait_for_completion_interruptible_timeout(
&blkif->drain_complete, HZ);
if (!atomic_read(&blkif->drain)) break;
} while (!kthread_should_stop());
atomic_set(&blkif->drain, 0);
}
staticvoid __end_block_io_op(struct pending_req *pending_req,
blk_status_t error)
{ /* An error fails the entire request. */ if (pending_req->operation == BLKIF_OP_FLUSH_DISKCACHE &&
error == BLK_STS_NOTSUPP) {
pr_debug("flush diskcache op failed, not supported\n");
xen_blkbk_flush_diskcache(XBT_NIL, pending_req->ring->blkif->be, 0);
pending_req->status = BLKIF_RSP_EOPNOTSUPP;
} elseif (pending_req->operation == BLKIF_OP_WRITE_BARRIER &&
error == BLK_STS_NOTSUPP) {
pr_debug("write barrier op failed, not supported\n");
xen_blkbk_barrier(XBT_NIL, pending_req->ring->blkif->be, 0);
pending_req->status = BLKIF_RSP_EOPNOTSUPP;
} elseif (error) {
pr_debug("Buffer not up-to-date at end of operation," " error=%d\n", error);
pending_req->status = BLKIF_RSP_ERROR;
}
/* * If all of the bio's have completed it is time to unmap * the grant references associated with 'request' and provide * the proper response on the ring.
*/ if (atomic_dec_and_test(&pending_req->pendcnt))
xen_blkbk_unmap_and_respond(pending_req);
}
/* * bio callback.
*/ staticvoid end_block_io_op(struct bio *bio)
{
__end_block_io_op(bio->bi_private, bio->bi_status);
bio_put(bio);
}
switch (dst->operation) { case BLKIF_OP_READ: case BLKIF_OP_WRITE: case BLKIF_OP_WRITE_BARRIER: case BLKIF_OP_FLUSH_DISKCACHE:
dst->u.rw.nr_segments = READ_ONCE(src->u.rw.nr_segments);
dst->u.rw.handle = src->u.rw.handle;
dst->u.rw.id = src->u.rw.id;
dst->u.rw.sector_number = src->u.rw.sector_number;
n = min_t(unsignedint, BLKIF_MAX_SEGMENTS_PER_REQUEST,
dst->u.rw.nr_segments); for (i = 0; i < n; i++)
dst->u.rw.seg[i] = src->u.rw.seg[i]; break;
case BLKIF_OP_INDIRECT:
dst->u.indirect.indirect_op = src->u.indirect.indirect_op;
dst->u.indirect.nr_segments =
READ_ONCE(src->u.indirect.nr_segments);
dst->u.indirect.handle = src->u.indirect.handle;
dst->u.indirect.id = src->u.indirect.id;
dst->u.indirect.sector_number = src->u.indirect.sector_number;
n = min(MAX_INDIRECT_PAGES,
INDIRECT_PAGES(dst->u.indirect.nr_segments)); for (i = 0; i < n; i++)
dst->u.indirect.indirect_grefs[i] =
src->u.indirect.indirect_grefs[i]; break;
default: /* * Don't know how to translate this op. Only get the * ID so failure can be reported to the frontend.
*/
dst->u.other.id = src->u.other.id; break;
}
}
switch (dst->operation) { case BLKIF_OP_READ: case BLKIF_OP_WRITE: case BLKIF_OP_WRITE_BARRIER: case BLKIF_OP_FLUSH_DISKCACHE:
dst->u.rw.nr_segments = READ_ONCE(src->u.rw.nr_segments);
dst->u.rw.handle = src->u.rw.handle;
dst->u.rw.id = src->u.rw.id;
dst->u.rw.sector_number = src->u.rw.sector_number;
n = min_t(unsignedint, BLKIF_MAX_SEGMENTS_PER_REQUEST,
dst->u.rw.nr_segments); for (i = 0; i < n; i++)
dst->u.rw.seg[i] = src->u.rw.seg[i]; break;
case BLKIF_OP_INDIRECT:
dst->u.indirect.indirect_op = src->u.indirect.indirect_op;
dst->u.indirect.nr_segments =
READ_ONCE(src->u.indirect.nr_segments);
dst->u.indirect.handle = src->u.indirect.handle;
dst->u.indirect.id = src->u.indirect.id;
dst->u.indirect.sector_number = src->u.indirect.sector_number;
n = min(MAX_INDIRECT_PAGES,
INDIRECT_PAGES(dst->u.indirect.nr_segments)); for (i = 0; i < n; i++)
dst->u.indirect.indirect_grefs[i] =
src->u.indirect.indirect_grefs[i]; break;
default: /* * Don't know how to translate this op. Only get the * ID so failure can be reported to the frontend.
*/
dst->u.other.id = src->u.other.id; break;
}
}
/* * Function to copy the from the ring buffer the 'struct blkif_request' * (which has the sectors we want, number of them, grant references, etc), * and transmute it to the block API to hand it over to the proper block disk.
*/ staticint
__do_block_io_op(struct xen_blkif_ring *ring, unsignedint *eoi_flags)
{ union blkif_back_rings *blk_rings = &ring->blk_rings; struct blkif_request req; struct pending_req *pending_req;
RING_IDX rc, rp; int more_to_do = 0;
rc = blk_rings->common.req_cons;
rp = blk_rings->common.sring->req_prod;
rmb(); /* Ensure we see queued requests up to 'rp'. */
if (RING_REQUEST_PROD_OVERFLOW(&blk_rings->common, rp)) {
rc = blk_rings->common.rsp_prod_pvt;
pr_warn("Frontend provided bogus ring requests (%d - %d = %d). Halting ring processing on dev=%04x\n",
rp, rc, rp - rc, ring->blkif->vbd.pdevice); return -EACCES;
} while (rc != rp) {
if (RING_REQUEST_CONS_OVERFLOW(&blk_rings->common, rc)) break;
/* We've seen a request, so clear spurious eoi flag. */
*eoi_flags &= ~XEN_EOI_FLAG_SPURIOUS;
if (kthread_should_stop()) {
more_to_do = 1; break;
}
switch (ring->blkif->blk_protocol) { case BLKIF_PROTOCOL_NATIVE:
memcpy(&req, RING_GET_REQUEST(&blk_rings->native, rc), sizeof(req)); break; case BLKIF_PROTOCOL_X86_32:
blkif_get_x86_32_req(&req, RING_GET_REQUEST(&blk_rings->x86_32, rc)); break; case BLKIF_PROTOCOL_X86_64:
blkif_get_x86_64_req(&req, RING_GET_REQUEST(&blk_rings->x86_64, rc)); break; default:
BUG();
}
blk_rings->common.req_cons = ++rc; /* before make_response() */
/* Apply all sanity checks to /private copy/ of request. */
barrier();
switch (req.operation) { case BLKIF_OP_READ: case BLKIF_OP_WRITE: case BLKIF_OP_WRITE_BARRIER: case BLKIF_OP_FLUSH_DISKCACHE: case BLKIF_OP_INDIRECT: if (dispatch_rw_block_io(ring, &req, pending_req)) goto done; break; case BLKIF_OP_DISCARD:
free_req(ring, pending_req); if (dispatch_discard_io(ring, &req)) goto done; break; default: if (dispatch_other_io(ring, &req, pending_req)) goto done; break;
}
/* Yield point for this unbounded loop. */
cond_resched();
}
done: return more_to_do;
}
staticint
do_block_io_op(struct xen_blkif_ring *ring, unsignedint *eoi_flags)
{ union blkif_back_rings *blk_rings = &ring->blk_rings; int more_to_do;
do {
more_to_do = __do_block_io_op(ring, eoi_flags); if (more_to_do) break;
RING_FINAL_CHECK_FOR_REQUESTS(&blk_rings->common, more_to_do);
} while (more_to_do);
return more_to_do;
} /* * Transmutation of the 'struct blkif_request' to a proper 'struct bio' * and call the 'submit_bio' to pass it to the underlying storage.
*/ staticint dispatch_rw_block_io(struct xen_blkif_ring *ring, struct blkif_request *req, struct pending_req *pending_req)
{ struct phys_req preq; struct seg_buf *seg = pending_req->seg; unsignedint nseg; struct bio *bio = NULL; struct bio **biolist = pending_req->biolist; int i, nbio = 0; enum req_op operation;
blk_opf_t operation_flags = 0; struct blk_plug plug; bool drain = false; struct grant_page **pages = pending_req->segments; unsignedshort req_operation;
if (xen_vbd_translate(&preq, ring->blkif, operation) != 0) {
pr_debug("access denied: %s of [%llu,%llu] on dev=%04x\n",
operation == REQ_OP_READ ? "read" : "write",
preq.sector_number,
preq.sector_number + preq.nr_sects,
ring->blkif->vbd.pdevice); goto fail_response;
}
/* * This check _MUST_ be done after xen_vbd_translate as the preq.bdev * is set there.
*/ for (i = 0; i < nseg; i++) { if (((int)preq.sector_number|(int)seg[i].nsec) &
((bdev_logical_block_size(preq.bdev) >> 9) - 1)) {
pr_debug("Misaligned I/O request from domain %d\n",
ring->blkif->domid); goto fail_response;
}
}
/* Wait on all outstanding I/O's and once that has been completed * issue the flush.
*/ if (drain)
xen_blk_drain_io(pending_req->ring);
/* * If we have failed at this point, we need to undo the M2P override, * set gnttab_set_unmap_op on all of the grant references and perform * the hypercall to unmap the grants - that is all done in * xen_blkbk_unmap.
*/ if (xen_blkbk_map_seg(pending_req)) goto fail_flush;
/* * This corresponding xen_blkif_put is done in __end_block_io_op, or * below (in "!bio") if we are handling a BLKIF_OP_DISCARD.
*/
xen_blkif_get(ring->blkif);
atomic_inc(&ring->inflight);
for (i = 0; i < nseg; i++) { while ((bio == NULL) ||
(bio_add_page(bio,
pages[i]->page,
seg[i].nsec << 9,
seg[i].offset) == 0)) {
bio = bio_alloc(preq.bdev, bio_max_segs(nseg - i),
operation | operation_flags,
GFP_KERNEL);
biolist[nbio++] = bio;
bio->bi_private = pending_req;
bio->bi_end_io = end_block_io_op;
bio->bi_iter.bi_sector = preq.sector_number;
}
preq.sector_number += seg[i].nsec;
}
/* This will be hit if the operation was a flush or discard. */ if (!bio) {
BUG_ON(operation_flags != REQ_PREFLUSH);
fail_flush:
xen_blkbk_unmap(ring, pending_req->segments,
pending_req->nr_segs);
fail_response: /* Haven't submitted any bio's yet. */
make_response(ring, req->u.rw.id, req_operation, BLKIF_RSP_ERROR);
free_req(ring, pending_req);
msleep(1); /* back off a bit */ return -EIO;
}
/* * Put a response on the ring on how the operation fared.
*/ staticvoid make_response(struct xen_blkif_ring *ring, u64 id, unsignedshort op, int st)
{ struct blkif_response *resp; unsignedlong flags; union blkif_back_rings *blk_rings; int notify;
spin_lock_irqsave(&ring->blk_ring_lock, flags);
blk_rings = &ring->blk_rings; /* Place on the response ring for the relevant domain. */ switch (ring->blkif->blk_protocol) { case BLKIF_PROTOCOL_NATIVE:
resp = RING_GET_RESPONSE(&blk_rings->native,
blk_rings->native.rsp_prod_pvt); break; case BLKIF_PROTOCOL_X86_32:
resp = RING_GET_RESPONSE(&blk_rings->x86_32,
blk_rings->x86_32.rsp_prod_pvt); break; case BLKIF_PROTOCOL_X86_64:
resp = RING_GET_RESPONSE(&blk_rings->x86_64,
blk_rings->x86_64.rsp_prod_pvt); break; default:
BUG();
}
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.