mirrors/qemu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

block/nvme: switch to a NVMeRequest freelist

Browse Source 
There are three issues with the current NVMeRequest->busy field:
1. The busy field is accidentally accessed outside q->lock when request
   submission fails.
2. Waiters on free_req_queue are not woken when a request is returned
   early due to submission failure.
2. Finding a free request involves scanning all requests. This makes
   request submission O(n^2).

Switch to an O(1) freelist that is always accessed under the lock.

Also differentiate between NVME_QUEUE_SIZE, the actual SQ/CQ size, and
NVME_NUM_REQS, the number of usable requests. This makes the code
simpler than using NVME_QUEUE_SIZE everywhere and having to keep in mind
that one slot is reserved.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Message-id: 20200617132201.1832152-5-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
This commit is contained in:
Stefan Hajnoczi 2020-06-17 14:21:58 +01:00
parent 04b3fb39c8
commit 1086e95da1
1 changed files with 54 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

81
block/nvme.c
View File

@ -33,6 +33,12 @@
#define NVME_QUEUE_SIZE 128 #define NVME_QUEUE_SIZE 128
#define NVME_BAR_SIZE 8192 #define NVME_BAR_SIZE 8192
/*
* We have to leave one slot empty as that is the full queue case where
* head == tail + 1.
*/
#define NVME_NUM_REQS (NVME_QUEUE_SIZE - 1)
typedef struct { typedef struct {
int32_t head, tail; int32_t head, tail;
uint8_t *queue; uint8_t *queue;
@ -47,7 +53,7 @@ typedef struct {
int cid; int cid;
void *prp_list_page; void *prp_list_page;
uint64_t prp_list_iova; uint64_t prp_list_iova;
bool busy; int free_req_next; /* q->reqs[] index of next free req */
} NVMeRequest; } NVMeRequest;
typedef struct { typedef struct {
@ -61,7 +67,8 @@ typedef struct {
/* Fields protected by @lock */ /* Fields protected by @lock */
NVMeQueue sq, cq; NVMeQueue sq, cq;
int cq_phase; int cq_phase;
NVMeRequest reqs[NVME_QUEUE_SIZE]; int free_req_head;
NVMeRequest reqs[NVME_NUM_REQS];
bool busy; bool busy;
int need_kick; int need_kick;
int inflight; int inflight;
@ -200,19 +207,23 @@ static NVMeQueuePair *nvme_create_queue_pair(BlockDriverState *bs,
qemu_mutex_init(&q->lock); qemu_mutex_init(&q->lock);
q->index = idx; q->index = idx;
qemu_co_queue_init(&q->free_req_queue); qemu_co_queue_init(&q->free_req_queue);
q->prp_list_pages = qemu_blockalign0(bs, s->page_size * NVME_QUEUE_SIZE); q->prp_list_pages = qemu_blockalign0(bs, s->page_size * NVME_NUM_REQS);
r = qemu_vfio_dma_map(s->vfio, q->prp_list_pages, r = qemu_vfio_dma_map(s->vfio, q->prp_list_pages,
s->page_size * NVME_QUEUE_SIZE, s->page_size * NVME_NUM_REQS,
false, &prp_list_iova); false, &prp_list_iova);
if (r) { if (r) {
goto fail; goto fail;
} }
for (i = 0; i < NVME_QUEUE_SIZE; i++) { q->free_req_head = -1;
for (i = 0; i < NVME_NUM_REQS; i++) {
NVMeRequest *req = &q->reqs[i]; NVMeRequest *req = &q->reqs[i];
req->cid = i + 1; req->cid = i + 1;
req->free_req_next = q->free_req_head;
q->free_req_head = i;
req->prp_list_page = q->prp_list_pages + i * s->page_size; req->prp_list_page = q->prp_list_pages + i * s->page_size;
req->prp_list_iova = prp_list_iova + i * s->page_size; req->prp_list_iova = prp_list_iova + i * s->page_size;
} }
nvme_init_queue(bs, &q->sq, size, NVME_SQ_ENTRY_BYTES, &local_err); nvme_init_queue(bs, &q->sq, size, NVME_SQ_ENTRY_BYTES, &local_err);
if (local_err) { if (local_err) {
error_propagate(errp, local_err); error_propagate(errp, local_err);
@ -254,13 +265,11 @@ static void nvme_kick(BDRVNVMeState *s, NVMeQueuePair *q)
*/ */
static NVMeRequest *nvme_get_free_req(NVMeQueuePair *q) static NVMeRequest *nvme_get_free_req(NVMeQueuePair *q)
{ {
int i; NVMeRequest *req;
NVMeRequest *req = NULL;
qemu_mutex_lock(&q->lock); qemu_mutex_lock(&q->lock);
while (q->inflight + q->need_kick > NVME_QUEUE_SIZE - 2) {
/* We have to leave one slot empty as that is the full queue case (head while (q->free_req_head == -1) {
* == tail + 1). */
if (qemu_in_coroutine()) { if (qemu_in_coroutine()) {
trace_nvme_free_req_queue_wait(q); trace_nvme_free_req_queue_wait(q);
qemu_co_queue_wait(&q->free_req_queue, &q->lock); qemu_co_queue_wait(&q->free_req_queue, &q->lock);
@ -269,20 +278,41 @@ static NVMeRequest *nvme_get_free_req(NVMeQueuePair *q)
return NULL; return NULL;
} }
} }
for (i = 0; i < NVME_QUEUE_SIZE; i++) {
if (!q->reqs[i].busy) { req = &q->reqs[q->free_req_head];
q->reqs[i].busy = true; q->free_req_head = req->free_req_next;
req = &q->reqs[i]; req->free_req_next = -1;
break;
}
}
/* We have checked inflight and need_kick while holding q->lock, so one
* free req must be available. */
assert(req);
qemu_mutex_unlock(&q->lock); qemu_mutex_unlock(&q->lock);
return req; return req;
} }
/* With q->lock */
static void nvme_put_free_req_locked(NVMeQueuePair *q, NVMeRequest *req)
{
req->free_req_next = q->free_req_head;
q->free_req_head = req - q->reqs;
}
/* With q->lock */
static void nvme_wake_free_req_locked(BDRVNVMeState *s, NVMeQueuePair *q)
{
if (!qemu_co_queue_empty(&q->free_req_queue)) {
replay_bh_schedule_oneshot_event(s->aio_context,
nvme_free_req_queue_cb, q);
}
}
/* Insert a request in the freelist and wake waiters */
static void nvme_put_free_req_and_wake(BDRVNVMeState *s, NVMeQueuePair *q,
NVMeRequest *req)
{
qemu_mutex_lock(&q->lock);
nvme_put_free_req_locked(q, req);
nvme_wake_free_req_locked(s, q);
qemu_mutex_unlock(&q->lock);
}
static inline int nvme_translate_error(const NvmeCqe *c) static inline int nvme_translate_error(const NvmeCqe *c)
{ {
uint16_t status = (le16_to_cpu(c->status) >> 1) & 0xFF; uint16_t status = (le16_to_cpu(c->status) >> 1) & 0xFF;
@ -344,7 +374,7 @@ static bool nvme_process_completion(BDRVNVMeState *s, NVMeQueuePair *q)
req = *preq; req = *preq;
assert(req.cid == cid); assert(req.cid == cid);
assert(req.cb); assert(req.cb);
preq->busy = false; nvme_put_free_req_locked(q, preq);
preq->cb = preq->opaque = NULL; preq->cb = preq->opaque = NULL;
qemu_mutex_unlock(&q->lock); qemu_mutex_unlock(&q->lock);
req.cb(req.opaque, ret); req.cb(req.opaque, ret);
@ -356,10 +386,7 @@ static bool nvme_process_completion(BDRVNVMeState *s, NVMeQueuePair *q)
/* Notify the device so it can post more completions. */ /* Notify the device so it can post more completions. */
smp_mb_release(); smp_mb_release();
*q->cq.doorbell = cpu_to_le32(q->cq.head); *q->cq.doorbell = cpu_to_le32(q->cq.head);
if (!qemu_co_queue_empty(&q->free_req_queue)) { nvme_wake_free_req_locked(s, q);
replay_bh_schedule_oneshot_event(s->aio_context,
nvme_free_req_queue_cb, q);
}
} }
q->busy = false; q->busy = false;
return progress; return progress;
@ -1001,7 +1028,7 @@ static coroutine_fn int nvme_co_prw_aligned(BlockDriverState *bs,
r = nvme_cmd_map_qiov(bs, &cmd, req, qiov); r = nvme_cmd_map_qiov(bs, &cmd, req, qiov);
qemu_co_mutex_unlock(&s->dma_map_lock); qemu_co_mutex_unlock(&s->dma_map_lock);
if (r) { if (r) {
req->busy = false; nvme_put_free_req_and_wake(s, ioq, req);
return r; return r;
} }
nvme_submit_command(s, ioq, req, &cmd, nvme_rw_cb, &data); nvme_submit_command(s, ioq, req, &cmd, nvme_rw_cb, &data);
@ -1218,7 +1245,7 @@ static int coroutine_fn nvme_co_pdiscard(BlockDriverState *bs,
qemu_co_mutex_unlock(&s->dma_map_lock); qemu_co_mutex_unlock(&s->dma_map_lock);
if (ret) { if (ret) {
req->busy = false; nvme_put_free_req_and_wake(s, ioq, req);
goto out; goto out;
} }