@@ -508,30 +508,26 @@ static bool rpcrdma_results_inline(struct rpcrdma_xprt *r_xprt,
}
/**
- * rpcrdma_unmap_sendctx - DMA-unmap Send buffers
+ * rpcrdma_sendctx_unmap - DMA-unmap Send buffer
* @sc: sendctx containing SGEs to unmap
*
*/
-void
-rpcrdma_unmap_sendctx(struct rpcrdma_sendctx *sc)
+void rpcrdma_sendctx_unmap(struct rpcrdma_sendctx *sc)
{
- struct rpcrdma_ia *ia = &sc->sc_xprt->rx_ia;
struct ib_sge *sge;
- unsigned int count;
/* The first two SGEs contain the transport header and
* the inline buffer. These are always left mapped so
* they can be cheaply re-used.
*/
- sge = &sc->sc_sges[2];
- for (count = sc->sc_unmap_count; count; ++sge, --count)
- ib_dma_unmap_page(ia->ri_device,
- sge->addr, sge->length, DMA_TO_DEVICE);
+ for (sge = &sc->sc_sges[2]; sc->sc_unmap_count;
+ ++sge, --sc->sc_unmap_count)
+ ib_dma_unmap_page(sc->sc_device, sge->addr, sge->length,
+ DMA_TO_DEVICE);
- if (test_and_clear_bit(RPCRDMA_REQ_F_TX_RESOURCES, &sc->sc_req->rl_flags)) {
- smp_mb__after_atomic();
+ if (test_and_clear_bit(RPCRDMA_REQ_F_TX_RESOURCES,
+ &sc->sc_req->rl_flags))
wake_up_bit(&sc->sc_req->rl_flags, RPCRDMA_REQ_F_TX_RESOURCES);
- }
}
/* Prepare an SGE for the RPC-over-RDMA transport header.
@@ -578,6 +574,7 @@ static bool rpcrdma_prepare_msg_sges(struct rpcrdma_xprt *r_xprt,
*/
if (!rpcrdma_regbuf_dma_map(r_xprt, rb))
goto out_regbuf;
+ sc->sc_device = rdmab_device(rb);
sge_no = 1;
sge[sge_no].addr = rdmab_addr(rb);
sge[sge_no].length = xdr->head[0].iov_len;
@@ -673,12 +670,12 @@ static bool rpcrdma_prepare_msg_sges(struct rpcrdma_xprt *r_xprt,
return false;
out_mapping_overflow:
- rpcrdma_unmap_sendctx(sc);
+ rpcrdma_sendctx_unmap(sc);
pr_err("rpcrdma: too many Send SGEs (%u)\n", sge_no);
return false;
out_mapping_err:
- rpcrdma_unmap_sendctx(sc);
+ rpcrdma_sendctx_unmap(sc);
trace_xprtrdma_dma_maperr(sge[sge_no].addr);
return false;
}
@@ -698,7 +695,7 @@ static bool rpcrdma_prepare_msg_sges(struct rpcrdma_xprt *r_xprt,
struct rpcrdma_req *req, u32 hdrlen,
struct xdr_buf *xdr, enum rpcrdma_chunktype rtype)
{
- req->rl_sendctx = rpcrdma_sendctx_get_locked(&r_xprt->rx_buf);
+ req->rl_sendctx = rpcrdma_sendctx_get_locked(r_xprt);
if (!req->rl_sendctx)
return -EAGAIN;
req->rl_sendctx->sc_wr.num_sge = 0;
@@ -870,20 +870,20 @@ static unsigned long rpcrdma_sendctx_next(struct rpcrdma_buffer *buf,
/**
* rpcrdma_sendctx_get_locked - Acquire a send context
- * @buf: transport buffers from which to acquire an unused context
+ * @r_xprt: controlling transport instance
*
* Returns pointer to a free send completion context; or NULL if
* the queue is empty.
*
* Usage: Called to acquire an SGE array before preparing a Send WR.
*
- * The caller serializes calls to this function (per rpcrdma_buffer),
- * and provides an effective memory barrier that flushes the new value
+ * The caller serializes calls to this function (per transport), and
+ * provides an effective memory barrier that flushes the new value
* of rb_sc_head.
*/
-struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf)
+struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_xprt *r_xprt)
{
- struct rpcrdma_xprt *r_xprt;
+ struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
struct rpcrdma_sendctx *sc;
unsigned long next_head;
@@ -908,7 +908,6 @@ struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf)
* backing up. Cause the caller to pause and try again.
*/
set_bit(RPCRDMA_BUF_F_EMPTY_SCQ, &buf->rb_flags);
- r_xprt = container_of(buf, struct rpcrdma_xprt, rx_buf);
r_xprt->rx_stats.empty_sendctx_q++;
return NULL;
}
@@ -920,7 +919,7 @@ struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf)
* Usage: Called from Send completion to return a sendctxt
* to the queue.
*
- * The caller serializes calls to this function (per rpcrdma_buffer).
+ * The caller serializes calls to this function (per transport).
*/
static void
rpcrdma_sendctx_put_locked(struct rpcrdma_sendctx *sc)
@@ -928,7 +927,7 @@ struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf)
struct rpcrdma_buffer *buf = &sc->sc_xprt->rx_buf;
unsigned long next_tail;
- /* Unmap SGEs of previously completed by unsignaled
+ /* Unmap SGEs of previously completed but unsignaled
* Sends by walking up the queue until @sc is found.
*/
next_tail = buf->rb_sc_tail;
@@ -936,7 +935,7 @@ struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf)
next_tail = rpcrdma_sendctx_next(buf, next_tail);
/* ORDER: item must be accessed _before_ tail is updated */
- rpcrdma_unmap_sendctx(buf->rb_sc_ctxs[next_tail]);
+ rpcrdma_sendctx_unmap(buf->rb_sc_ctxs[next_tail]);
} while (buf->rb_sc_ctxs[next_tail] != sc);
@@ -225,6 +225,7 @@ enum {
struct rpcrdma_sendctx {
struct ib_send_wr sc_wr;
struct ib_cqe sc_cqe;
+ struct ib_device *sc_device;
struct rpcrdma_xprt *sc_xprt;
struct rpcrdma_req *sc_req;
unsigned int sc_unmap_count;
@@ -536,7 +537,7 @@ struct rpcrdma_req *rpcrdma_req_create(struct rpcrdma_xprt *r_xprt, size_t size,
void rpcrdma_req_destroy(struct rpcrdma_req *req);
int rpcrdma_buffer_create(struct rpcrdma_xprt *);
void rpcrdma_buffer_destroy(struct rpcrdma_buffer *);
-struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_buffer *buf);
+struct rpcrdma_sendctx *rpcrdma_sendctx_get_locked(struct rpcrdma_xprt *r_xprt);
struct rpcrdma_mr *rpcrdma_mr_get(struct rpcrdma_xprt *r_xprt);
void rpcrdma_mr_put(struct rpcrdma_mr *mr);
@@ -625,7 +626,7 @@ int rpcrdma_prepare_send_sges(struct rpcrdma_xprt *r_xprt,
struct rpcrdma_req *req, u32 hdrlen,
struct xdr_buf *xdr,
enum rpcrdma_chunktype rtype);
-void rpcrdma_unmap_sendctx(struct rpcrdma_sendctx *sc);
+void rpcrdma_sendctx_unmap(struct rpcrdma_sendctx *sc);
int rpcrdma_marshal_req(struct rpcrdma_xprt *r_xprt, struct rpc_rqst *rqst);
void rpcrdma_set_max_header_sizes(struct rpcrdma_xprt *);
void rpcrdma_reply_handler(struct rpcrdma_rep *rep);
Minor clean-ups I've stumbled on since sendctx was merged last year. In particular, making Send completion processing more efficient appears to have a measurable impact on IOPS throughput. Note: test_and_clear_bit() returns a value, thus an explicit memory barrier is not necessary. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> --- net/sunrpc/xprtrdma/rpc_rdma.c | 27 ++++++++++++--------------- net/sunrpc/xprtrdma/verbs.c | 17 ++++++++--------- net/sunrpc/xprtrdma/xprt_rdma.h | 5 +++-- 3 files changed, 23 insertions(+), 26 deletions(-)