/* Send a backwards direction RPC call.
*
* Caller holds the connection's mutex and has already marshaled
* the RPC/RDMA request.
*
* This is similar to svc_rdma_reply, but takes an rpc_rqst
* instead, does not support chunks, and avoids blocking memory
* allocation.
*
* XXX: There is still an opportunity to block in svc_rdma_send()
* if there are no SQ entries to post the Send. This may occur if
* the adapter has a small maximum SQ depth.
*/
static int svc_rdma_bc_sendto(struct svcxprt_rdma *rdma,
struct rpc_rqst *rqst)
{
struct xdr_buf *sndbuf = &rqst->rq_snd_buf;
struct svc_rdma_op_ctxt *ctxt;
struct svc_rdma_req_map *vec;
struct ib_send_wr send_wr;
int ret;
vec = svc_rdma_get_req_map(rdma);
ret = svc_rdma_map_xdr(rdma, sndbuf, vec, false);
if (ret)
goto out_err;
ret = svc_rdma_repost_recv(rdma, GFP_NOIO);
if (ret)
goto out_err;
ctxt = svc_rdma_get_context(rdma);
ctxt->pages[0] = virt_to_page(rqst->rq_buffer);
ctxt->count = 1;
ctxt->direction = DMA_TO_DEVICE;
ctxt->sge[0].lkey = rdma->sc_pd->local_dma_lkey;
ctxt->sge[0].length = sndbuf->len;
ctxt->sge[0].addr =
ib_dma_map_page(rdma->sc_cm_id->device, ctxt->pages[0], 0,
sndbuf->len, DMA_TO_DEVICE);
if (ib_dma_mapping_error(rdma->sc_cm_id->device, ctxt->sge[0].addr)) {
ret = -EIO;
goto out_unmap;
}
svc_rdma_count_mappings(rdma, ctxt);
memset(&send_wr, 0, sizeof(send_wr));
ctxt->cqe.done = svc_rdma_wc_send;
send_wr.wr_cqe = &ctxt->cqe;
send_wr.sg_list = ctxt->sge;
send_wr.num_sge = 1;
send_wr.opcode = IB_WR_SEND;
send_wr.send_flags = IB_SEND_SIGNALED;
ret = svc_rdma_send(rdma, &send_wr);
if (ret) {
ret = -EIO;
goto out_unmap;
}
out_err:
svc_rdma_put_req_map(rdma, vec);
dprintk("svcrdma: %s returns %d\n", __func__, ret);
return ret;
out_unmap:
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 1);
goto out_err;
}
/*
* Map for DMA and insert a single page into the on-demand paging page tables.
*
* @umem: the umem to insert the page to.
* @page_index: index in the umem to add the page to.
* @page: the page struct to map and add.
* @access_mask: access permissions needed for this page.
* @current_seq: sequence number for synchronization with invalidations.
* the sequence number is taken from
* umem->odp_data->notifiers_seq.
*
* The function returns -EFAULT if the DMA mapping operation fails. It returns
* -EAGAIN if a concurrent invalidation prevents us from updating the page.
*
* The page is released via put_page even if the operation failed. For
* on-demand pinning, the page is released whenever it isn't stored in the
* umem.
*/
static int ib_umem_odp_map_dma_single_page(
struct ib_umem *umem,
int page_index,
struct page *page,
u64 access_mask,
unsigned long current_seq,
enum ib_odp_dma_map_flags flags)
{
struct ib_device *dev = umem->context->device;
dma_addr_t dma_addr;
int stored_page = 0;
int ret = 0;
mutex_lock(&umem->odp_data->umem_mutex);
/*
* Note: we avoid writing if seq is different from the initial seq, to
* handle case of a racing notifier. This check also allows us to bail
* early if we have a notifier running in parallel with us.
*/
if (ib_umem_mmu_notifier_retry(umem, current_seq)) {
ret = -EAGAIN;
goto out;
}
if (!(umem->odp_data->dma_list[page_index])) {
dma_addr = ib_dma_map_page(dev,
page,
0, PAGE_SIZE,
DMA_BIDIRECTIONAL);
if (ib_dma_mapping_error(dev, dma_addr)) {
ret = -EFAULT;
goto out;
}
umem->odp_data->dma_list[page_index] = dma_addr | access_mask;
umem->odp_data->page_list[page_index] = page;
if (flags & IB_ODP_DMA_MAP_FOR_PREFETCH)
atomic_inc(&dev->odp_statistics.num_prefetch_pages);
else
atomic_inc(&dev->odp_statistics.num_page_fault_pages);
stored_page = 1;
} else if (umem->odp_data->page_list[page_index] == page) {
umem->odp_data->dma_list[page_index] |= access_mask;
} else {
pr_err("error: got different pages in IB device and from get_user_pages. IB device page: %p, gup page: %p\n"
, umem->odp_data->page_list[page_index], page);
}
out:
mutex_unlock(&umem->odp_data->umem_mutex);
/* On Demand Paging - avoid pinning the page */
if (umem->context->invalidate_range || !stored_page)
put_page(page);
return ret;
}
static int
svc_rdma_post_recv(struct svcxprt_rdma *xprt)
{
struct ib_recv_wr recv_wr, *bad_recv_wr;
struct svc_rdma_op_ctxt *ctxt;
struct page *page;
dma_addr_t pa;
int sge_no;
int buflen;
int ret;
ctxt = svc_rdma_get_context(xprt);
buflen = 0;
ctxt->direction = DMA_FROM_DEVICE;
ctxt->cqe.done = svc_rdma_wc_receive;
for (sge_no = 0; buflen < xprt->sc_max_req_size; sge_no++) {
if (sge_no >= xprt->sc_max_sge) {
pr_err("svcrdma: Too many sges (%d)\n", sge_no);
goto err_put_ctxt;
}
page = alloc_page(GFP_KERNEL);
if (!page)
goto err_put_ctxt;
ctxt->pages[sge_no] = page;
pa = ib_dma_map_page(xprt->sc_cm_id->device,
page, 0, PAGE_SIZE,
DMA_FROM_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device, pa))
goto err_put_ctxt;
svc_rdma_count_mappings(xprt, ctxt);
ctxt->sge[sge_no].addr = pa;
ctxt->sge[sge_no].length = PAGE_SIZE;
ctxt->sge[sge_no].lkey = xprt->sc_pd->local_dma_lkey;
ctxt->count = sge_no + 1;
buflen += PAGE_SIZE;
}
recv_wr.next = NULL;
recv_wr.sg_list = &ctxt->sge[0];
recv_wr.num_sge = ctxt->count;
recv_wr.wr_cqe = &ctxt->cqe;
svc_xprt_get(&xprt->sc_xprt);
ret = ib_post_recv(xprt->sc_qp, &recv_wr, &bad_recv_wr);
if (ret) {
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 1);
svc_xprt_put(&xprt->sc_xprt);
}
return ret;
err_put_ctxt:
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 1);
return -ENOMEM;
}
static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev,
struct ipoib_cm_rx_buf *rx_ring,
int id, int frags,
u64 mapping[IPOIB_CM_RX_SG])
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct sk_buff *skb;
int i;
skb = dev_alloc_skb(IPOIB_CM_HEAD_SIZE + 12);
if (unlikely(!skb))
return NULL;
/*
* IPoIB adds a 4 byte header. So we need 12 more bytes to align the
* IP header to a multiple of 16.
*/
skb_reserve(skb, 12);
mapping[0] = ib_dma_map_single(priv->ca, skb->data, IPOIB_CM_HEAD_SIZE,
DMA_FROM_DEVICE);
if (unlikely(ib_dma_mapping_error(priv->ca, mapping[0]))) {
dev_kfree_skb_any(skb);
return NULL;
}
for (i = 0; i < frags; i++) {
struct page *page = alloc_page(GFP_ATOMIC);
if (!page)
goto partial_error;
skb_fill_page_desc(skb, i, page, 0, PAGE_SIZE);
mapping[i + 1] = ib_dma_map_page(priv->ca, skb_shinfo(skb)->frags[i].page,
0, PAGE_SIZE, DMA_FROM_DEVICE);
if (unlikely(ib_dma_mapping_error(priv->ca, mapping[i + 1])))
goto partial_error;
}
rx_ring[id].skb = skb;
return skb;
partial_error:
ib_dma_unmap_single(priv->ca, mapping[0], IPOIB_CM_HEAD_SIZE, DMA_FROM_DEVICE);
for (; i > 0; --i)
ib_dma_unmap_single(priv->ca, mapping[i], PAGE_SIZE, DMA_FROM_DEVICE);
dev_kfree_skb_any(skb);
return NULL;
}
int svc_rdma_post_recv(struct svcxprt_rdma *xprt)
{
struct ib_recv_wr recv_wr, *bad_recv_wr;
struct svc_rdma_op_ctxt *ctxt;
struct page *page;
dma_addr_t pa;
int sge_no;
int buflen;
int ret;
ctxt = svc_rdma_get_context(xprt);
buflen = 0;
ctxt->direction = DMA_FROM_DEVICE;
for (sge_no = 0; buflen < xprt->sc_max_req_size; sge_no++) {
BUG_ON(sge_no >= xprt->sc_max_sge);
page = svc_rdma_get_page();
ctxt->pages[sge_no] = page;
pa = ib_dma_map_page(xprt->sc_cm_id->device,
page, 0, PAGE_SIZE,
DMA_FROM_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device, pa))
goto err_put_ctxt;
atomic_inc(&xprt->sc_dma_used);
ctxt->sge[sge_no].addr = pa;
ctxt->sge[sge_no].length = PAGE_SIZE;
ctxt->sge[sge_no].lkey = xprt->sc_dma_lkey;
ctxt->count = sge_no + 1;
buflen += PAGE_SIZE;
}
recv_wr.next = NULL;
recv_wr.sg_list = &ctxt->sge[0];
recv_wr.num_sge = ctxt->count;
recv_wr.wr_id = (u64)(unsigned long)ctxt;
svc_xprt_get(&xprt->sc_xprt);
ret = ib_post_recv(xprt->sc_qp, &recv_wr, &bad_recv_wr);
if (ret) {
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 1);
svc_xprt_put(&xprt->sc_xprt);
}
return ret;
err_put_ctxt:
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 1);
return -ENOMEM;
}
static int rdma_set_ctxt_sge(struct svcxprt_rdma *xprt,
struct svc_rdma_op_ctxt *ctxt,
struct svc_rdma_fastreg_mr *frmr,
struct kvec *vec,
u64 *sgl_offset,
int count)
{
int i;
unsigned long off;
ctxt->count = count;
ctxt->direction = DMA_FROM_DEVICE;
for (i = 0; i < count; i++) {
ctxt->sge[i].length = 0; /* in case map fails */
if (!frmr) {
BUG_ON(0 == virt_to_page(vec[i].iov_base));
off = (unsigned long)vec[i].iov_base & ~PAGE_MASK;
ctxt->sge[i].addr =
ib_dma_map_page(xprt->sc_cm_id->device,
virt_to_page(vec[i].iov_base),
off,
vec[i].iov_len,
DMA_FROM_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device,
ctxt->sge[i].addr))
return -EINVAL;
ctxt->sge[i].lkey = xprt->sc_dma_lkey;
atomic_inc(&xprt->sc_dma_used);
} else {
ctxt->sge[i].addr = (unsigned long)vec[i].iov_base;
ctxt->sge[i].lkey = frmr->mr->lkey;
}
ctxt->sge[i].length = vec[i].iov_len;
*sgl_offset = *sgl_offset + vec[i].iov_len;
}
return 0;
}
/* Map a read-chunk-list to an XDR and fast register the page-list.
*
* Assumptions:
* - chunk[0] position points to pages[0] at an offset of 0
* - pages[] will be made physically contiguous by creating a one-off memory
* region using the fastreg verb.
* - byte_count is # of bytes in read-chunk-list
* - ch_count is # of chunks in read-chunk-list
*
* Output:
* - sge array pointing into pages[] array.
* - chunk_sge array specifying sge index and count for each
* chunk in the read list
*/
static int fast_reg_read_chunks(struct svcxprt_rdma *xprt,
struct svc_rqst *rqstp,
struct svc_rdma_op_ctxt *head,
struct rpcrdma_msg *rmsgp,
struct svc_rdma_req_map *rpl_map,
struct svc_rdma_req_map *chl_map,
int ch_count,
int byte_count)
{
int page_no;
int ch_no;
u32 offset;
struct rpcrdma_read_chunk *ch;
struct svc_rdma_fastreg_mr *frmr;
int ret = 0;
frmr = svc_rdma_get_frmr(xprt);
if (IS_ERR(frmr))
return -ENOMEM;
head->frmr = frmr;
head->arg.head[0] = rqstp->rq_arg.head[0];
head->arg.tail[0] = rqstp->rq_arg.tail[0];
head->arg.pages = &head->pages[head->count];
head->hdr_count = head->count; /* save count of hdr pages */
head->arg.page_base = 0;
head->arg.page_len = byte_count;
head->arg.len = rqstp->rq_arg.len + byte_count;
head->arg.buflen = rqstp->rq_arg.buflen + byte_count;
/* Fast register the page list */
frmr->kva = page_address(rqstp->rq_arg.pages[0]);
frmr->direction = DMA_FROM_DEVICE;
frmr->access_flags = (IB_ACCESS_LOCAL_WRITE|IB_ACCESS_REMOTE_WRITE);
frmr->map_len = byte_count;
frmr->page_list_len = PAGE_ALIGN(byte_count) >> PAGE_SHIFT;
for (page_no = 0; page_no < frmr->page_list_len; page_no++) {
frmr->page_list->page_list[page_no] =
ib_dma_map_page(xprt->sc_cm_id->device,
rqstp->rq_arg.pages[page_no], 0,
PAGE_SIZE, DMA_FROM_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device,
frmr->page_list->page_list[page_no]))
goto fatal_err;
atomic_inc(&xprt->sc_dma_used);
head->arg.pages[page_no] = rqstp->rq_arg.pages[page_no];
}
head->count += page_no;
/* rq_respages points one past arg pages */
rqstp->rq_respages = &rqstp->rq_arg.pages[page_no];
/* Create the reply and chunk maps */
offset = 0;
ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
for (ch_no = 0; ch_no < ch_count; ch_no++) {
rpl_map->sge[ch_no].iov_base = frmr->kva + offset;
rpl_map->sge[ch_no].iov_len = ch->rc_target.rs_length;
chl_map->ch[ch_no].count = 1;
chl_map->ch[ch_no].start = ch_no;
offset += ch->rc_target.rs_length;
ch++;
}
ret = svc_rdma_fastreg(xprt, frmr);
if (ret)
goto fatal_err;
return ch_no;
fatal_err:
printk("svcrdma: error fast registering xdr for xprt %p", xprt);
svc_rdma_put_frmr(xprt, frmr);
return -EIO;
}
/* Issue an RDMA_READ using an FRMR to map the data sink */
static int rdma_read_chunk_frmr(struct svcxprt_rdma *xprt,
struct svc_rqst *rqstp,
struct svc_rdma_op_ctxt *head,
int *page_no,
u32 *page_offset,
u32 rs_handle,
u32 rs_length,
u64 rs_offset,
int last)
{
struct ib_send_wr read_wr;
struct ib_send_wr inv_wr;
struct ib_send_wr fastreg_wr;
u8 key;
int pages_needed = PAGE_ALIGN(*page_offset + rs_length) >> PAGE_SHIFT;
struct svc_rdma_op_ctxt *ctxt = svc_rdma_get_context(xprt);
struct svc_rdma_fastreg_mr *frmr = svc_rdma_get_frmr(xprt);
int ret, read, pno;
u32 pg_off = *page_offset;
u32 pg_no = *page_no;
if (IS_ERR(frmr))
return -ENOMEM;
ctxt->direction = DMA_FROM_DEVICE;
ctxt->frmr = frmr;
pages_needed = min_t(int, pages_needed, xprt->sc_frmr_pg_list_len);
read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
frmr->kva = page_address(rqstp->rq_arg.pages[pg_no]);
frmr->direction = DMA_FROM_DEVICE;
frmr->access_flags = (IB_ACCESS_LOCAL_WRITE|IB_ACCESS_REMOTE_WRITE);
frmr->map_len = pages_needed << PAGE_SHIFT;
frmr->page_list_len = pages_needed;
for (pno = 0; pno < pages_needed; pno++) {
int len = min_t(int, rs_length, PAGE_SIZE - pg_off);
head->arg.pages[pg_no] = rqstp->rq_arg.pages[pg_no];
head->arg.page_len += len;
head->arg.len += len;
if (!pg_off)
head->count++;
rqstp->rq_respages = &rqstp->rq_arg.pages[pg_no+1];
#if (LINUX_VERSION_CODE >= KERNEL_VERSION(3,8,0))
rqstp->rq_next_page = rqstp->rq_respages + 1;
#endif
frmr->page_list->page_list[pno] =
ib_dma_map_page(xprt->sc_cm_id->device,
head->arg.pages[pg_no], 0,
PAGE_SIZE, DMA_FROM_DEVICE);
ret = ib_dma_mapping_error(xprt->sc_cm_id->device,
frmr->page_list->page_list[pno]);
if (ret)
goto err;
atomic_inc(&xprt->sc_dma_used);
/* adjust offset and wrap to next page if needed */
pg_off += len;
if (pg_off == PAGE_SIZE) {
pg_off = 0;
pg_no++;
}
rs_length -= len;
}
if (last && rs_length == 0)
set_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
else
clear_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
/* Bump the key */
key = (u8)(frmr->mr->lkey & 0x000000FF);
ib_update_fast_reg_key(frmr->mr, ++key);
ctxt->sge[0].addr = (unsigned long)frmr->kva + *page_offset;
ctxt->sge[0].lkey = frmr->mr->lkey;
ctxt->sge[0].length = read;
ctxt->count = 1;
ctxt->read_hdr = head;
/* Prepare FASTREG WR */
memset(&fastreg_wr, 0, sizeof(fastreg_wr));
fastreg_wr.opcode = IB_WR_FAST_REG_MR;
fastreg_wr.send_flags = IB_SEND_SIGNALED;
fastreg_wr.wr.fast_reg.iova_start = (unsigned long)frmr->kva;
fastreg_wr.wr.fast_reg.page_list = frmr->page_list;
fastreg_wr.wr.fast_reg.page_list_len = frmr->page_list_len;
fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT;
fastreg_wr.wr.fast_reg.length = frmr->map_len;
fastreg_wr.wr.fast_reg.access_flags = frmr->access_flags;
fastreg_wr.wr.fast_reg.rkey = frmr->mr->lkey;
fastreg_wr.next = &read_wr;
/* Prepare RDMA_READ */
memset(&read_wr, 0, sizeof(read_wr));
read_wr.send_flags = IB_SEND_SIGNALED;
read_wr.wr.rdma.rkey = rs_handle;
read_wr.wr.rdma.remote_addr = rs_offset;
read_wr.sg_list = ctxt->sge;
read_wr.num_sge = 1;
if (xprt->sc_dev_caps & SVCRDMA_DEVCAP_READ_W_INV) {
read_wr.opcode = IB_WR_RDMA_READ_WITH_INV;
read_wr.wr_id = (unsigned long)ctxt;
read_wr.ex.invalidate_rkey = ctxt->frmr->mr->lkey;
} else {
read_wr.opcode = IB_WR_RDMA_READ;
read_wr.next = &inv_wr;
/* Prepare invalidate */
memset(&inv_wr, 0, sizeof(inv_wr));
inv_wr.wr_id = (unsigned long)ctxt;
inv_wr.opcode = IB_WR_LOCAL_INV;
inv_wr.send_flags = IB_SEND_SIGNALED | IB_SEND_FENCE;
inv_wr.ex.invalidate_rkey = frmr->mr->lkey;
}
ctxt->wr_op = read_wr.opcode;
/* Post the chain */
ret = svc_rdma_send(xprt, &fastreg_wr);
if (ret) {
pr_err("svcrdma: Error %d posting RDMA_READ\n", ret);
set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags);
goto err;
}
/* return current location in page array */
*page_no = pg_no;
*page_offset = pg_off;
ret = read;
atomic_inc(&rdma_stat_read);
return ret;
err:
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 0);
svc_rdma_put_frmr(xprt, frmr);
return ret;
}
/* Issue an RDMA_READ using the local lkey to map the data sink */
static int rdma_read_chunk_lcl(struct svcxprt_rdma *xprt,
struct svc_rqst *rqstp,
struct svc_rdma_op_ctxt *head,
int *page_no,
u32 *page_offset,
u32 rs_handle,
u32 rs_length,
u64 rs_offset,
int last)
{
struct ib_send_wr read_wr;
int pages_needed = PAGE_ALIGN(*page_offset + rs_length) >> PAGE_SHIFT;
struct svc_rdma_op_ctxt *ctxt = svc_rdma_get_context(xprt);
int ret, read, pno;
u32 pg_off = *page_offset;
u32 pg_no = *page_no;
ctxt->direction = DMA_FROM_DEVICE;
ctxt->read_hdr = head;
pages_needed =
min_t(int, pages_needed, rdma_read_max_sge(xprt, pages_needed));
read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
for (pno = 0; pno < pages_needed; pno++) {
int len = min_t(int, rs_length, PAGE_SIZE - pg_off);
head->arg.pages[pg_no] = rqstp->rq_arg.pages[pg_no];
head->arg.page_len += len;
head->arg.len += len;
if (!pg_off)
head->count++;
rqstp->rq_respages = &rqstp->rq_arg.pages[pg_no+1];
#if (LINUX_VERSION_CODE >= KERNEL_VERSION(3,8,0))
rqstp->rq_next_page = rqstp->rq_respages + 1;
#endif
ctxt->sge[pno].addr =
ib_dma_map_page(xprt->sc_cm_id->device,
head->arg.pages[pg_no], pg_off,
PAGE_SIZE - pg_off,
DMA_FROM_DEVICE);
ret = ib_dma_mapping_error(xprt->sc_cm_id->device,
ctxt->sge[pno].addr);
if (ret)
goto err;
atomic_inc(&xprt->sc_dma_used);
/* The lkey here is either a local dma lkey or a dma_mr lkey */
ctxt->sge[pno].lkey = xprt->sc_dma_lkey;
ctxt->sge[pno].length = len;
ctxt->count++;
/* adjust offset and wrap to next page if needed */
pg_off += len;
if (pg_off == PAGE_SIZE) {
pg_off = 0;
pg_no++;
}
rs_length -= len;
}
if (last && rs_length == 0)
set_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
else
clear_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags);
memset(&read_wr, 0, sizeof(read_wr));
read_wr.wr_id = (unsigned long)ctxt;
read_wr.opcode = IB_WR_RDMA_READ;
ctxt->wr_op = read_wr.opcode;
read_wr.send_flags = IB_SEND_SIGNALED;
read_wr.wr.rdma.rkey = rs_handle;
read_wr.wr.rdma.remote_addr = rs_offset;
read_wr.sg_list = ctxt->sge;
read_wr.num_sge = pages_needed;
ret = svc_rdma_send(xprt, &read_wr);
if (ret) {
pr_err("svcrdma: Error %d posting RDMA_READ\n", ret);
set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags);
goto err;
}
/* return current location in page array */
*page_no = pg_no;
*page_offset = pg_off;
ret = read;
atomic_inc(&rdma_stat_read);
return ret;
err:
svc_rdma_unmap_dma(ctxt);
svc_rdma_put_context(ctxt, 0);
return ret;
}
void sdp_post_send(struct sdp_sock *ssk, struct sk_buff *skb)
{
struct sdp_buf *tx_req;
struct sdp_bsdh *h = (struct sdp_bsdh *)skb_transport_header(skb);
unsigned long mseq = ring_head(ssk->tx_ring);
int i, rc, frags;
u64 addr;
struct ib_device *dev;
struct ib_send_wr *bad_wr;
struct ib_sge ibsge[SDP_MAX_SEND_SGES];
struct ib_sge *sge = ibsge;
struct ib_send_wr tx_wr = { NULL };
u32 send_flags = IB_SEND_SIGNALED;
SDPSTATS_COUNTER_MID_INC(post_send, h->mid);
SDPSTATS_HIST(send_size, skb->len);
if (!ssk->qp_active)
goto err;
ssk->tx_packets++;
if (h->mid != SDP_MID_SRCAVAIL &&
h->mid != SDP_MID_DATA &&
h->mid != SDP_MID_SRCAVAIL_CANCEL) {
struct sock *sk = sk_ssk(ssk);
sk->sk_wmem_queued += skb->truesize;
sk_mem_charge(sk, skb->truesize);
}
if (unlikely(h->mid == SDP_MID_SRCAVAIL)) {
struct tx_srcavail_state *tx_sa = TX_SRCAVAIL_STATE(skb);
if (ssk->tx_sa != tx_sa) {
sdp_dbg_data(sk_ssk(ssk), "SrcAvail cancelled "
"before being sent!\n");
SDP_WARN_ON(1);
sk_wmem_free_skb(sk_ssk(ssk), skb);
return;
}
TX_SRCAVAIL_STATE(skb)->mseq = mseq;
}
if (unlikely(SDP_SKB_CB(skb)->flags & TCPHDR_URG))
h->flags = SDP_OOB_PRES | SDP_OOB_PEND;
else
h->flags = 0;
h->bufs = htons(rx_ring_posted(ssk));
h->len = htonl(skb->len);
h->mseq = htonl(mseq);
h->mseq_ack = htonl(mseq_ack(ssk));
sdp_prf(sk_ssk(ssk), skb, "TX: %s bufs: %d mseq:%ld ack:%d c: %d",
mid2str(h->mid), rx_ring_posted(ssk), mseq,
ntohl(h->mseq_ack), tx_credits(ssk));
SDP_DUMP_PACKET(sk_ssk(ssk), "TX", skb, h);
tx_req = &ssk->tx_ring.buffer[mseq & (SDP_TX_SIZE - 1)];
tx_req->skb = skb;
dev = ssk->ib_device;
if (skb->len <= ssk->inline_thresh && !skb_shinfo(skb)->nr_frags) {
SDPSTATS_COUNTER_INC(inline_sends);
sge->addr = (u64) skb->data;
sge->length = skb->len;
sge->lkey = 0;
frags = 0;
tx_req->mapping[0] = 0; /* Nothing to be cleaned up by sdp_cleanup_sdp_buf() */
send_flags |= IB_SEND_INLINE;
} else {
addr = ib_dma_map_single(dev, skb->data, skb->len - skb->data_len,
DMA_TO_DEVICE);
tx_req->mapping[0] = addr;
/* TODO: proper error handling */
BUG_ON(ib_dma_mapping_error(dev, addr));
sge->addr = addr;
sge->length = skb->len - skb->data_len;
sge->lkey = ssk->sdp_dev->mr->lkey;
frags = skb_shinfo(skb)->nr_frags;
for (i = 0; i < frags; ++i) {
++sge;
addr = ib_dma_map_page(dev,
skb_shinfo(skb)->frags[i].page.p,
skb_shinfo(skb)->frags[i].page_offset,
skb_shinfo(skb)->frags[i].size,
DMA_TO_DEVICE);
BUG_ON(ib_dma_mapping_error(dev, addr));
tx_req->mapping[i + 1] = addr;
sge->addr = addr;
sge->length = skb_shinfo(skb)->frags[i].size;
sge->lkey = ssk->sdp_dev->mr->lkey;
}
}
tx_wr.next = NULL;
tx_wr.wr_id = ring_head(ssk->tx_ring) | SDP_OP_SEND;
tx_wr.sg_list = ibsge;
tx_wr.num_sge = frags + 1;
tx_wr.opcode = IB_WR_SEND;
tx_wr.send_flags = send_flags;
if (unlikely(SDP_SKB_CB(skb)->flags & TCPHDR_URG))
tx_wr.send_flags |= IB_SEND_SOLICITED;
rc = ib_post_send(ssk->qp, &tx_wr, &bad_wr);
if (unlikely(rc)) {
sdp_dbg(sk_ssk(ssk),
"ib_post_send failed with status %d.\n", rc);
sdp_cleanup_sdp_buf(ssk, tx_req, skb->len - skb->data_len, DMA_TO_DEVICE);
sdp_set_error(sk_ssk(ssk), -ECONNRESET);
goto err;
}
atomic_inc(&ssk->tx_ring.head);
atomic_dec(&ssk->tx_ring.credits);
atomic_set(&ssk->remote_credits, rx_ring_posted(ssk));
return;
err:
sk_wmem_free_skb(sk_ssk(ssk), skb);
}
/* Encode an XDR as an array of IB SGE
*
* Assumptions:
* - head[0] is physically contiguous.
* - tail[0] is physically contiguous.
* - pages[] is not physically or virtually contigous and consists of
* PAGE_SIZE elements.
*
* Output:
* SGE[0] reserved for RCPRDMA header
* SGE[1] data from xdr->head[]
* SGE[2..sge_count-2] data from xdr->pages[]
* SGE[sge_count-1] data from xdr->tail.
*
* The max SGE we need is the length of the XDR / pagesize + one for
* head + one for tail + one for RPCRDMA header. Since RPCSVC_MAXPAGES
* reserves a page for both the request and the reply header, and this
* array is only concerned with the reply we are assured that we have
* on extra page for the RPCRMDA header.
*/
static int fast_reg_xdr(struct svcxprt_rdma *xprt,
struct xdr_buf *xdr,
struct svc_rdma_req_map *vec)
{
int sge_no;
u32 sge_bytes;
u32 page_bytes;
u32 page_off;
int page_no = 0;
u8 *frva;
struct svc_rdma_fastreg_mr *frmr;
frmr = svc_rdma_get_frmr(xprt);
if (IS_ERR(frmr))
return -ENOMEM;
vec->frmr = frmr;
/* Skip the RPCRDMA header */
sge_no = 1;
/* Map the head. */
frva = (void *)((unsigned long)(xdr->head[0].iov_base) & PAGE_MASK);
vec->sge[sge_no].iov_base = xdr->head[0].iov_base;
vec->sge[sge_no].iov_len = xdr->head[0].iov_len;
vec->count = 2;
sge_no++;
/* Build the FRMR */
frmr->kva = frva;
frmr->direction = DMA_TO_DEVICE;
frmr->access_flags = 0;
frmr->map_len = PAGE_SIZE;
frmr->page_list_len = 1;
frmr->page_list->page_list[page_no] =
ib_dma_map_single(xprt->sc_cm_id->device,
(void *)xdr->head[0].iov_base,
PAGE_SIZE, DMA_TO_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device,
frmr->page_list->page_list[page_no]))
goto fatal_err;
atomic_inc(&xprt->sc_dma_used);
page_off = xdr->page_base;
page_bytes = xdr->page_len + page_off;
if (!page_bytes)
goto encode_tail;
/* Map the pages */
vec->sge[sge_no].iov_base = frva + frmr->map_len + page_off;
vec->sge[sge_no].iov_len = page_bytes;
sge_no++;
while (page_bytes) {
struct page *page;
page = xdr->pages[page_no++];
sge_bytes = min_t(u32, page_bytes, (PAGE_SIZE - page_off));
page_bytes -= sge_bytes;
frmr->page_list->page_list[page_no] =
ib_dma_map_page(xprt->sc_cm_id->device, page, 0,
PAGE_SIZE, DMA_TO_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device,
frmr->page_list->page_list[page_no]))
goto fatal_err;
atomic_inc(&xprt->sc_dma_used);
page_off = 0; /* reset for next time through loop */
frmr->map_len += PAGE_SIZE;
frmr->page_list_len++;
}
vec->count++;
encode_tail:
/* Map tail */
if (0 == xdr->tail[0].iov_len)
goto done;
vec->count++;
vec->sge[sge_no].iov_len = xdr->tail[0].iov_len;
if (((unsigned long)xdr->tail[0].iov_base & PAGE_MASK) ==
((unsigned long)xdr->head[0].iov_base & PAGE_MASK)) {
/*
* If head and tail use the same page, we don't need
* to map it again.
*/
vec->sge[sge_no].iov_base = xdr->tail[0].iov_base;
} else {
void *va;
/* Map another page for the tail */
page_off = (unsigned long)xdr->tail[0].iov_base & ~PAGE_MASK;
va = (void *)((unsigned long)xdr->tail[0].iov_base & PAGE_MASK);
vec->sge[sge_no].iov_base = frva + frmr->map_len + page_off;
frmr->page_list->page_list[page_no] =
ib_dma_map_single(xprt->sc_cm_id->device, va, PAGE_SIZE,
DMA_TO_DEVICE);
if (ib_dma_mapping_error(xprt->sc_cm_id->device,
frmr->page_list->page_list[page_no]))
goto fatal_err;
atomic_inc(&xprt->sc_dma_used);
frmr->map_len += PAGE_SIZE;
frmr->page_list_len++;
}
done:
if (svc_rdma_fastreg(xprt, frmr))
goto fatal_err;
return 0;
fatal_err:
printk("svcrdma: Error fast registering memory for xprt %p\n", xprt);
svc_rdma_put_frmr(xprt, frmr);
return -EIO;
}
/* This function prepares the portion of the RPCRDMA message to be
* sent in the RDMA_SEND. This function is called after data sent via
* RDMA has already been transmitted. There are three cases:
* - The RPCRDMA header, RPC header, and payload are all sent in a
* single RDMA_SEND. This is the "inline" case.
* - The RPCRDMA header and some portion of the RPC header and data
* are sent via this RDMA_SEND and another portion of the data is
* sent via RDMA.
* - The RPCRDMA header [NOMSG] is sent in this RDMA_SEND and the RPC
* header and data are all transmitted via RDMA.
* In all three cases, this function prepares the RPCRDMA header in
* sge[0], the 'type' parameter indicates the type to place in the
* RPCRDMA header, and the 'byte_count' field indicates how much of
* the XDR to include in this RDMA_SEND.
*/
static int send_reply(struct svcxprt_rdma *rdma,
struct svc_rqst *rqstp,
struct page *page,
struct rpcrdma_msg *rdma_resp,
struct svc_rdma_op_ctxt *ctxt,
struct svc_rdma_req_map *vec,
int byte_count)
{
struct ib_send_wr send_wr;
struct ib_send_wr inv_wr;
int sge_no;
int sge_bytes;
int page_no;
int ret;
/* Post a recv buffer to handle another request. */
ret = svc_rdma_post_recv(rdma);
if (ret) {
printk(KERN_INFO
"svcrdma: could not post a receive buffer, err=%d."
"Closing transport %p.\n", ret, rdma);
set_bit(XPT_CLOSE, &rdma->sc_xprt.xpt_flags);
svc_rdma_put_context(ctxt, 0);
return -ENOTCONN;
}
/* Prepare the context */
ctxt->pages[0] = page;
ctxt->count = 1;
ctxt->frmr = vec->frmr;
if (vec->frmr)
set_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags);
else
clear_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags);
/* Prepare the SGE for the RPCRDMA Header */
ctxt->sge[0].addr =
ib_dma_map_page(rdma->sc_cm_id->device,
page, 0, PAGE_SIZE, DMA_TO_DEVICE);
if (ib_dma_mapping_error(rdma->sc_cm_id->device, ctxt->sge[0].addr))
goto err;
atomic_inc(&rdma->sc_dma_used);
ctxt->direction = DMA_TO_DEVICE;
ctxt->sge[0].length = svc_rdma_xdr_get_reply_hdr_len(rdma_resp);
ctxt->sge[0].lkey = rdma->sc_dma_lkey;
/* Determine how many of our SGE are to be transmitted */
for (sge_no = 1; byte_count && sge_no < vec->count; sge_no++) {
sge_bytes = min_t(size_t, vec->sge[sge_no].iov_len, byte_count);
byte_count -= sge_bytes;
if (!vec->frmr) {
ctxt->sge[sge_no].addr =
ib_dma_map_single(rdma->sc_cm_id->device,
vec->sge[sge_no].iov_base,
sge_bytes, DMA_TO_DEVICE);
if (ib_dma_mapping_error(rdma->sc_cm_id->device,
ctxt->sge[sge_no].addr))
goto err;
atomic_inc(&rdma->sc_dma_used);
ctxt->sge[sge_no].lkey = rdma->sc_dma_lkey;
} else {
ctxt->sge[sge_no].addr = (unsigned long)
vec->sge[sge_no].iov_base;
ctxt->sge[sge_no].lkey = vec->frmr->mr->lkey;
}
ctxt->sge[sge_no].length = sge_bytes;
}
BUG_ON(byte_count != 0);
/* Save all respages in the ctxt and remove them from the
* respages array. They are our pages until the I/O
* completes.
*/
for (page_no = 0; page_no < rqstp->rq_resused; page_no++) {
ctxt->pages[page_no+1] = rqstp->rq_respages[page_no];
ctxt->count++;
rqstp->rq_respages[page_no] = NULL;
/*
* If there are more pages than SGE, terminate SGE
* list so that svc_rdma_unmap_dma doesn't attempt to
* unmap garbage.
*/
if (page_no+1 >= sge_no)
ctxt->sge[page_no+1].length = 0;
}
BUG_ON(sge_no > rdma->sc_max_sge);
memset(&send_wr, 0, sizeof send_wr);
ctxt->wr_op = IB_WR_SEND;
send_wr.wr_id = (unsigned long)ctxt;
send_wr.sg_list = ctxt->sge;
send_wr.num_sge = sge_no;
send_wr.opcode = IB_WR_SEND;
send_wr.send_flags = IB_SEND_SIGNALED;
if (vec->frmr) {
/* Prepare INVALIDATE WR */
memset(&inv_wr, 0, sizeof inv_wr);
inv_wr.opcode = IB_WR_LOCAL_INV;
inv_wr.send_flags = IB_SEND_SIGNALED;
inv_wr.ex.invalidate_rkey =
vec->frmr->mr->lkey;
send_wr.next = &inv_wr;
}
ret = svc_rdma_send(rdma, &send_wr);
if (ret)
goto err;
return 0;
err:
svc_rdma_put_frmr(rdma, vec->frmr);
svc_rdma_put_context(ctxt, 1);
return -EIO;
}