aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/hw/hfi1/user_sdma.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/hw/hfi1/user_sdma.c')
-rw-r--r--drivers/infiniband/hw/hfi1/user_sdma.c137
1 files changed, 64 insertions, 73 deletions
diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c
index 5c88706121c1..3f0aadccd9f6 100644
--- a/drivers/infiniband/hw/hfi1/user_sdma.c
+++ b/drivers/infiniband/hw/hfi1/user_sdma.c
@@ -1,5 +1,5 @@
/*
- * Copyright(c) 2015 - 2017 Intel Corporation.
+ * Copyright(c) 2015 - 2018 Intel Corporation.
*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
@@ -76,8 +76,7 @@ MODULE_PARM_DESC(sdma_comp_size, "Size of User SDMA completion ring. Default: 12
static unsigned initial_pkt_count = 8;
-static int user_sdma_send_pkts(struct user_sdma_request *req,
- unsigned maxpkts);
+static int user_sdma_send_pkts(struct user_sdma_request *req, u16 maxpkts);
static void user_sdma_txreq_cb(struct sdma_txreq *txreq, int status);
static inline void pq_update(struct hfi1_user_sdma_pkt_q *pq);
static void user_sdma_free_request(struct user_sdma_request *req, bool unpin);
@@ -101,7 +100,7 @@ static inline u32 get_lrh_len(struct hfi1_pkt_header, u32 len);
static int defer_packet_queue(
struct sdma_engine *sde,
- struct iowait *wait,
+ struct iowait_work *wait,
struct sdma_txreq *txreq,
uint seq,
bool pkts_sent);
@@ -124,13 +123,13 @@ static struct mmu_rb_ops sdma_rb_ops = {
static int defer_packet_queue(
struct sdma_engine *sde,
- struct iowait *wait,
+ struct iowait_work *wait,
struct sdma_txreq *txreq,
uint seq,
bool pkts_sent)
{
struct hfi1_user_sdma_pkt_q *pq =
- container_of(wait, struct hfi1_user_sdma_pkt_q, busy);
+ container_of(wait->iow, struct hfi1_user_sdma_pkt_q, busy);
struct hfi1_ibdev *dev = &pq->dd->verbs_dev;
struct user_sdma_txreq *tx =
container_of(txreq, struct user_sdma_txreq, txreq);
@@ -187,13 +186,12 @@ int hfi1_user_sdma_alloc_queues(struct hfi1_ctxtdata *uctxt,
pq->ctxt = uctxt->ctxt;
pq->subctxt = fd->subctxt;
pq->n_max_reqs = hfi1_sdma_comp_ring_size;
- pq->state = SDMA_PKT_Q_INACTIVE;
atomic_set(&pq->n_reqs, 0);
init_waitqueue_head(&pq->wait);
atomic_set(&pq->n_locked, 0);
pq->mm = fd->mm;
- iowait_init(&pq->busy, 0, NULL, defer_packet_queue,
+ iowait_init(&pq->busy, 0, NULL, NULL, defer_packet_queue,
activate_packet_queue, NULL);
pq->reqidx = 0;
@@ -276,7 +274,7 @@ int hfi1_user_sdma_free_queues(struct hfi1_filedata *fd,
/* Wait until all requests have been freed. */
wait_event_interruptible(
pq->wait,
- (READ_ONCE(pq->state) == SDMA_PKT_Q_INACTIVE));
+ !atomic_read(&pq->n_reqs));
kfree(pq->reqs);
kfree(pq->req_in_use);
kmem_cache_destroy(pq->txreq_cache);
@@ -312,6 +310,13 @@ static u8 dlid_to_selector(u16 dlid)
return mapping[hash];
}
+/**
+ * hfi1_user_sdma_process_request() - Process and start a user sdma request
+ * @fd: valid file descriptor
+ * @iovec: array of io vectors to process
+ * @dim: overall iovec array size
+ * @count: number of io vector array entries processed
+ */
int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
struct iovec *iovec, unsigned long dim,
unsigned long *count)
@@ -328,7 +333,6 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
u8 opcode, sc, vl;
u16 pkey;
u32 slid;
- int req_queued = 0;
u16 dlid;
u32 selector;
@@ -392,7 +396,6 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
req->data_len = 0;
req->pq = pq;
req->cq = cq;
- req->status = -1;
req->ahg_idx = -1;
req->iov_idx = 0;
req->sent = 0;
@@ -400,12 +403,14 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
req->seqcomp = 0;
req->seqsubmitted = 0;
req->tids = NULL;
- req->done = 0;
req->has_error = 0;
INIT_LIST_HEAD(&req->txps);
memcpy(&req->info, &info, sizeof(info));
+ /* The request is initialized, count it */
+ atomic_inc(&pq->n_reqs);
+
if (req_opcode(info.ctrl) == EXPECTED) {
/* expected must have a TID info and at least one data vector */
if (req->data_iovs < 2) {
@@ -500,7 +505,6 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
ret = pin_vector_pages(req, &req->iovs[i]);
if (ret) {
req->data_iovs = i;
- req->status = ret;
goto free_req;
}
req->data_len += req->iovs[i].iov.iov_len;
@@ -561,23 +565,11 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
req->ahg_idx = sdma_ahg_alloc(req->sde);
set_comp_state(pq, cq, info.comp_idx, QUEUED, 0);
- atomic_inc(&pq->n_reqs);
- req_queued = 1;
+ pq->state = SDMA_PKT_Q_ACTIVE;
/* Send the first N packets in the request to buy us some time */
ret = user_sdma_send_pkts(req, pcount);
- if (unlikely(ret < 0 && ret != -EBUSY)) {
- req->status = ret;
+ if (unlikely(ret < 0 && ret != -EBUSY))
goto free_req;
- }
-
- /*
- * It is possible that the SDMA engine would have processed all the
- * submitted packets by the time we get here. Therefore, only set
- * packet queue state to ACTIVE if there are still uncompleted
- * requests.
- */
- if (atomic_read(&pq->n_reqs))
- xchg(&pq->state, SDMA_PKT_Q_ACTIVE);
/*
* This is a somewhat blocking send implementation.
@@ -588,14 +580,8 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
while (req->seqsubmitted != req->info.npkts) {
ret = user_sdma_send_pkts(req, pcount);
if (ret < 0) {
- if (ret != -EBUSY) {
- req->status = ret;
- WRITE_ONCE(req->has_error, 1);
- if (READ_ONCE(req->seqcomp) ==
- req->seqsubmitted - 1)
- goto free_req;
- return ret;
- }
+ if (ret != -EBUSY)
+ goto free_req;
wait_event_interruptible_timeout(
pq->busy.wait_dma,
(pq->state == SDMA_PKT_Q_ACTIVE),
@@ -606,10 +592,19 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd,
*count += idx;
return 0;
free_req:
- user_sdma_free_request(req, true);
- if (req_queued)
+ /*
+ * If the submitted seqsubmitted == npkts, the completion routine
+ * controls the final state. If sequbmitted < npkts, wait for any
+ * outstanding packets to finish before cleaning up.
+ */
+ if (req->seqsubmitted < req->info.npkts) {
+ if (req->seqsubmitted)
+ wait_event(pq->busy.wait_dma,
+ (req->seqcomp == req->seqsubmitted - 1));
+ user_sdma_free_request(req, true);
pq_update(pq);
- set_comp_state(pq, cq, info.comp_idx, ERROR, req->status);
+ set_comp_state(pq, cq, info.comp_idx, ERROR, ret);
+ }
return ret;
}
@@ -760,9 +755,10 @@ static int user_sdma_txadd(struct user_sdma_request *req,
return ret;
}
-static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts)
+static int user_sdma_send_pkts(struct user_sdma_request *req, u16 maxpkts)
{
- int ret = 0, count;
+ int ret = 0;
+ u16 count;
unsigned npkts = 0;
struct user_sdma_txreq *tx = NULL;
struct hfi1_user_sdma_pkt_q *pq = NULL;
@@ -864,8 +860,10 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts)
changes = set_txreq_header_ahg(req, tx,
datalen);
- if (changes < 0)
+ if (changes < 0) {
+ ret = changes;
goto free_tx;
+ }
}
} else {
ret = sdma_txinit(&tx->txreq, 0, sizeof(req->hdr) +
@@ -914,10 +912,11 @@ static int user_sdma_send_pkts(struct user_sdma_request *req, unsigned maxpkts)
npkts++;
}
dosend:
- ret = sdma_send_txlist(req->sde, &pq->busy, &req->txps, &count);
+ ret = sdma_send_txlist(req->sde,
+ iowait_get_ib_work(&pq->busy),
+ &req->txps, &count);
req->seqsubmitted += count;
if (req->seqsubmitted == req->info.npkts) {
- WRITE_ONCE(req->done, 1);
/*
* The txreq has already been submitted to the HW queue
* so we can free the AHG entry now. Corruption will not
@@ -1365,11 +1364,15 @@ static int set_txreq_header_ahg(struct user_sdma_request *req,
return idx;
}
-/*
- * SDMA tx request completion callback. Called when the SDMA progress
- * state machine gets notification that the SDMA descriptors for this
- * tx request have been processed by the DMA engine. Called in
- * interrupt context.
+/**
+ * user_sdma_txreq_cb() - SDMA tx request completion callback.
+ * @txreq: valid sdma tx request
+ * @status: success/failure of request
+ *
+ * Called when the SDMA progress state machine gets notification that
+ * the SDMA descriptors for this tx request have been processed by the
+ * DMA engine. Called in interrupt context.
+ * Only do work on completed sequences.
*/
static void user_sdma_txreq_cb(struct sdma_txreq *txreq, int status)
{
@@ -1378,7 +1381,7 @@ static void user_sdma_txreq_cb(struct sdma_txreq *txreq, int status)
struct user_sdma_request *req;
struct hfi1_user_sdma_pkt_q *pq;
struct hfi1_user_sdma_comp_q *cq;
- u16 idx;
+ enum hfi1_sdma_comp_state state = COMPLETE;
if (!tx->req)
return;
@@ -1391,39 +1394,25 @@ static void user_sdma_txreq_cb(struct sdma_txreq *txreq, int status)
SDMA_DBG(req, "SDMA completion with error %d",
status);
WRITE_ONCE(req->has_error, 1);
+ state = ERROR;
}
req->seqcomp = tx->seqnum;
kmem_cache_free(pq->txreq_cache, tx);
- tx = NULL;
-
- idx = req->info.comp_idx;
- if (req->status == -1 && status == SDMA_TXREQ_S_OK) {
- if (req->seqcomp == req->info.npkts - 1) {
- req->status = 0;
- user_sdma_free_request(req, false);
- pq_update(pq);
- set_comp_state(pq, cq, idx, COMPLETE, 0);
- }
- } else {
- if (status != SDMA_TXREQ_S_OK)
- req->status = status;
- if (req->seqcomp == (READ_ONCE(req->seqsubmitted) - 1) &&
- (READ_ONCE(req->done) ||
- READ_ONCE(req->has_error))) {
- user_sdma_free_request(req, false);
- pq_update(pq);
- set_comp_state(pq, cq, idx, ERROR, req->status);
- }
- }
+
+ /* sequence isn't complete? We are done */
+ if (req->seqcomp != req->info.npkts - 1)
+ return;
+
+ user_sdma_free_request(req, false);
+ set_comp_state(pq, cq, req->info.comp_idx, state, status);
+ pq_update(pq);
}
static inline void pq_update(struct hfi1_user_sdma_pkt_q *pq)
{
- if (atomic_dec_and_test(&pq->n_reqs)) {
- xchg(&pq->state, SDMA_PKT_Q_INACTIVE);
+ if (atomic_dec_and_test(&pq->n_reqs))
wake_up(&pq->wait);
- }
}
static void user_sdma_free_request(struct user_sdma_request *req, bool unpin)
@@ -1448,6 +1437,8 @@ static void user_sdma_free_request(struct user_sdma_request *req, bool unpin)
if (!node)
continue;
+ req->iovs[i].node = NULL;
+
if (unpin)
hfi1_mmu_rb_remove(req->pq->handler,
&node->rb);
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.