DPDK patches and discussions
 help / color / mirror / Atom feed
From: Hernan Vargas <hernan.vargas@intel.com>
To: dev@dpdk.org, gakhil@marvell.com, trix@redhat.com,
	maxime.coquelin@redhat.com
Cc: nicolas.chautru@intel.com, qi.z.zhang@intel.com,
	Hernan Vargas <hernan.vargas@intel.com>
Subject: [PATCH v4 16/30] baseband/acc100: add scatter-gather support
Date: Tue, 18 Oct 2022 17:39:04 -0700	[thread overview]
Message-ID: <20221019003918.257506-17-hernan.vargas@intel.com> (raw)
In-Reply-To: <20221019003918.257506-1-hernan.vargas@intel.com>

Add flag to support scatter-gather for the mbuf

Signed-off-by: Hernan Vargas <hernan.vargas@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 drivers/baseband/acc/rte_acc100_pmd.c | 38 ++++++++++++++++++---------
 1 file changed, 25 insertions(+), 13 deletions(-)

diff --git a/drivers/baseband/acc/rte_acc100_pmd.c b/drivers/baseband/acc/rte_acc100_pmd.c
index ca92ae906f..0921e9a44d 100644
--- a/drivers/baseband/acc/rte_acc100_pmd.c
+++ b/drivers/baseband/acc/rte_acc100_pmd.c
@@ -1363,6 +1363,8 @@ acc101_fcw_ld_fill(struct rte_bbdev_dec_op *op, struct acc_fcw_ld *fcw,
  *   Store information about device capabilities
  * @param next_triplet
  *   Index for ACC100 DMA Descriptor triplet
+ * @param scattergather
+ *   Flag to support scatter-gather for the mbuf
  *
  * @return
  *   Returns index of next triplet on success, other value if lengths of
@@ -1372,12 +1374,16 @@ acc101_fcw_ld_fill(struct rte_bbdev_dec_op *op, struct acc_fcw_ld *fcw,
 static inline int
 acc100_dma_fill_blk_type_in(struct acc_dma_req_desc *desc,
 		struct rte_mbuf **input, uint32_t *offset, uint32_t cb_len,
-		uint32_t *seg_total_left, int next_triplet)
+		uint32_t *seg_total_left, int next_triplet,
+		bool scattergather)
 {
 	uint32_t part_len;
 	struct rte_mbuf *m = *input;
 
-	part_len = (*seg_total_left < cb_len) ? *seg_total_left : cb_len;
+	if (scattergather)
+		part_len = (*seg_total_left < cb_len) ? *seg_total_left : cb_len;
+	else
+		part_len = cb_len;
 	cb_len -= part_len;
 	*seg_total_left -= part_len;
 
@@ -1468,7 +1474,7 @@ acc100_dma_desc_le_fill(struct rte_bbdev_enc_op *op,
 	}
 
 	next_triplet = acc100_dma_fill_blk_type_in(desc, input, in_offset,
-			pad_le_in(in_length_in_bytes, q), seg_total_left, next_triplet);
+			pad_le_in(in_length_in_bytes, q), seg_total_left, next_triplet, false);
 	if (unlikely(next_triplet < 0)) {
 		rte_bbdev_log(ERR,
 				"Mismatch between data to process and mbuf data length in bbdev_op: %p",
@@ -1556,7 +1562,9 @@ acc100_dma_desc_td_fill(struct rte_bbdev_dec_op *op,
 	}
 
 	next_triplet = acc100_dma_fill_blk_type_in(desc, input, in_offset, kw,
-			seg_total_left, next_triplet);
+			seg_total_left, next_triplet,
+			check_bit(op->turbo_dec.op_flags,
+			RTE_BBDEV_TURBO_DEC_SCATTER_GATHER));
 	if (unlikely(next_triplet < 0)) {
 		rte_bbdev_log(ERR,
 				"Mismatch between data to process and mbuf data length in bbdev_op: %p",
@@ -1658,7 +1666,9 @@ acc100_dma_desc_ld_fill(struct rte_bbdev_dec_op *op,
 
 	next_triplet = acc100_dma_fill_blk_type_in(desc, input,
 			in_offset, input_length,
-			seg_total_left, next_triplet);
+			seg_total_left, next_triplet,
+			check_bit(op->ldpc_dec.op_flags,
+			RTE_BBDEV_LDPC_DEC_SCATTER_GATHER));
 
 	if (unlikely(next_triplet < 0)) {
 		rte_bbdev_log(ERR,
@@ -2726,10 +2736,9 @@ enqueue_ldpc_dec_one_op_cb(struct acc_queue *q, struct rte_bbdev_dec_op *op,
 		fcw = &desc->req.fcw_ld;
 		q->d->fcw_ld_fill(op, fcw, harq_layout);
 
-		/* Special handling when overusing mbuf */
-		if (fcw->rm_e < ACC_MAX_E_MBUF)
-			seg_total_left = rte_pktmbuf_data_len(input)
-					- in_offset;
+		/* Special handling when using mbuf or not */
+		if (check_bit(op->ldpc_dec.op_flags, RTE_BBDEV_LDPC_DEC_SCATTER_GATHER))
+			seg_total_left = rte_pktmbuf_data_len(input) - in_offset;
 		else
 			seg_total_left = fcw->rm_e;
 
@@ -2804,9 +2813,10 @@ enqueue_ldpc_dec_one_op_tb(struct acc_queue *q, struct rte_bbdev_dec_op *op,
 	r = op->ldpc_dec.tb_params.r;
 
 	while (mbuf_total_left > 0 && r < c) {
-
-		seg_total_left = rte_pktmbuf_data_len(input) - in_offset;
-
+		if (check_bit(op->ldpc_dec.op_flags, RTE_BBDEV_LDPC_DEC_SCATTER_GATHER))
+			seg_total_left = rte_pktmbuf_data_len(input) - in_offset;
+		else
+			seg_total_left = op->ldpc_dec.input.length;
 		/* Set up DMA descriptor */
 		desc = acc_desc(q, total_enqueued_cbs);
 		desc->req.data_ptrs[0].address = q->ring_addr_iova + fcw_offset;
@@ -2831,7 +2841,9 @@ enqueue_ldpc_dec_one_op_tb(struct acc_queue *q, struct rte_bbdev_dec_op *op,
 		rte_memdump(stderr, "Req Desc.", desc, sizeof(*desc));
 #endif
 
-		if (seg_total_left == 0) {
+		if (check_bit(op->ldpc_dec.op_flags,
+				RTE_BBDEV_LDPC_DEC_SCATTER_GATHER)
+				&& (seg_total_left == 0)) {
 			/* Go to the next mbuf */
 			input = input->next;
 			in_offset = 0;
-- 
2.37.1


  parent reply	other threads:[~2022-10-18 16:44 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-19  0:38 [PATCH v4 00/30] baseband/acc100: changes for 22.11 Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 01/30] baseband/acc100: fix ring availability calculation Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 02/30] baseband/acc100: add function to check AQ availability Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 03/30] baseband/acc100: memory leak fix Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 04/30] baseband/acc100: add LDPC encoder padding function Hernan Vargas
2022-10-20  9:58   ` [EXT] " Akhil Goyal
2022-10-19  0:38 ` [PATCH v4 05/30] baseband/acc100: check turbo dec/enc input Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 06/30] baseband/acc100: check for unlikely operation vals Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 07/30] baseband/acc100: enforce additional check on FCW Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 08/30] baseband/acc100: allocate ring/queue mem when NULL Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 09/30] baseband/acc100: reduce input length for CRC24B Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 10/30] baseband/acc100: fix clearing PF IR outside handler Hernan Vargas
2022-10-19  0:38 ` [PATCH v4 11/30] baseband/acc100: set device min alignment to 1 Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 12/30] baseband/acc100: add protection for NULL HARQ input Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 13/30] baseband/acc100: reset pointer after rte_free Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 14/30] baseband/acc100: fix debug print for LDPC FCW Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 15/30] baseband/acc100: add enqueue status Hernan Vargas
2022-10-19  0:39 ` Hernan Vargas [this message]
2022-10-19  0:39 ` [PATCH v4 17/30] baseband/acc100: add HARQ index helper function Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 18/30] baseband/acc100: enable input validation by default Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 19/30] baseband/acc100: added LDPC transport block support Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 20/30] baseband/acc100: update validate LDPC enc/dec Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 21/30] baseband/acc100: implement configurable queue depth Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 22/30] baseband/acc100: add queue stop operation Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 23/30] baseband/acc100: update uplink CB input length Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 24/30] baseband/acc100: update log messages Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 25/30] baseband/acc100: store FCW from first CB descriptor Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 26/30] baseband/acc100: update device info Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 27/30] baseband/acc100: add ring companion address Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 28/30] baseband/acc100: add workaround for deRM corner cases Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 29/30] baseband/acc100: configure PMON control registers Hernan Vargas
2022-10-19  0:39 ` [PATCH v4 30/30] baseband/acc100: update guide docs Hernan Vargas
2022-10-20  9:59   ` [EXT] " Akhil Goyal

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20221019003918.257506-17-hernan.vargas@intel.com \
    --to=hernan.vargas@intel.com \
    --cc=dev@dpdk.org \
    --cc=gakhil@marvell.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=nicolas.chautru@intel.com \
    --cc=qi.z.zhang@intel.com \
    --cc=trix@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).