patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Hernan Vargas <hernan.vargas@intel.com>
To: stable@dpdk.org, ktraynor@redhat.com
Cc: nicolas.chautru@intel.com,
	Hernan Vargas <hernan.vargas@intel.com>,
	Maxime Coquelin <maxime.coquelin@redhat.com>
Subject: [PATCH 21.11 2/7] baseband/acc100: check AQ availability
Date: Wed, 16 Nov 2022 20:46:47 -0800	[thread overview]
Message-ID: <20221117044652.163000-3-hernan.vargas@intel.com> (raw)
In-Reply-To: <20221117044652.163000-1-hernan.vargas@intel.com>

[ upstream commit 2df5fe2023c9c7e4a858960f70c7a3b849be1bc1 ]

In some corner case to run more batch enqueue than
supported. A protection is required to avoid that corner case.
Enhance all ACC100 enqueue operations with check to see if there is room
in the atomic queue(AQ) for enqueueing batches into the queue manager.

Fixes: 5ad5060f8f7 ("baseband/acc100: add LDPC processing functions")

Signed-off-by: Hernan Vargas <hernan.vargas@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
---
 drivers/baseband/acc100/rte_acc100_pmd.c | 31 ++++++++++++++++++------
 1 file changed, 24 insertions(+), 7 deletions(-)

diff --git a/drivers/baseband/acc100/rte_acc100_pmd.c b/drivers/baseband/acc100/rte_acc100_pmd.c
index 254f256c80..d13f574b46 100644
--- a/drivers/baseband/acc100/rte_acc100_pmd.c
+++ b/drivers/baseband/acc100/rte_acc100_pmd.c
@@ -3464,12 +3464,27 @@ acc100_enqueue_enc_tb(struct rte_bbdev_queue_data *q_data,
 	return i;
 }
 
+/* Check room in AQ for the enqueues batches into Qmgr */
+static inline int32_t
+acc100_aq_avail(struct rte_bbdev_queue_data *q_data, uint16_t num_ops)
+{
+	struct acc100_queue *q = q_data->queue_private;
+	int32_t aq_avail = q->aq_depth -
+			((q->aq_enqueued - q->aq_dequeued +
+			ACC100_MAX_QUEUE_DEPTH) % ACC100_MAX_QUEUE_DEPTH)
+			- (num_ops >> 7);
+
+	return aq_avail;
+}
+
 /* Enqueue encode operations for ACC100 device. */
 static uint16_t
 acc100_enqueue_enc(struct rte_bbdev_queue_data *q_data,
 		struct rte_bbdev_enc_op **ops, uint16_t num)
 {
-	if (unlikely(num == 0))
+	int32_t aq_avail = acc100_aq_avail(q_data, num);
+
+	if (unlikely((aq_avail <= 0) || (num == 0)))
 		return 0;
 	if (ops[0]->turbo_enc.code_block_mode == RTE_BBDEV_TRANSPORT_BLOCK)
 		return acc100_enqueue_enc_tb(q_data, ops, num);
@@ -3482,7 +3497,9 @@ static uint16_t
 acc100_enqueue_ldpc_enc(struct rte_bbdev_queue_data *q_data,
 		struct rte_bbdev_enc_op **ops, uint16_t num)
 {
-	if (unlikely(num == 0))
+	int32_t aq_avail = acc100_aq_avail(q_data, num);
+
+	if (unlikely((aq_avail <= 0) || (num == 0)))
 		return 0;
 	if (ops[0]->ldpc_enc.code_block_mode == RTE_BBDEV_TRANSPORT_BLOCK)
 		return acc100_enqueue_enc_tb(q_data, ops, num);
@@ -3667,7 +3684,9 @@ static uint16_t
 acc100_enqueue_dec(struct rte_bbdev_queue_data *q_data,
 		struct rte_bbdev_dec_op **ops, uint16_t num)
 {
-	if (unlikely(num == 0))
+	int32_t aq_avail = acc100_aq_avail(q_data, num);
+
+	if (unlikely((aq_avail <= 0) || (num == 0)))
 		return 0;
 	if (ops[0]->turbo_dec.code_block_mode == RTE_BBDEV_TRANSPORT_BLOCK)
 		return acc100_enqueue_dec_tb(q_data, ops, num);
@@ -3680,11 +3699,9 @@ static uint16_t
 acc100_enqueue_ldpc_dec(struct rte_bbdev_queue_data *q_data,
 		struct rte_bbdev_dec_op **ops, uint16_t num)
 {
-	struct acc100_queue *q = q_data->queue_private;
-	int32_t aq_avail = q->aq_depth +
-			(q->aq_dequeued - q->aq_enqueued) / 128;
+	int32_t aq_avail = acc100_aq_avail(q_data, num);
 
-	if (unlikely((aq_avail == 0) || (num == 0)))
+	if (unlikely((aq_avail <= 0) || (num == 0)))
 		return 0;
 
 	if (ops[0]->ldpc_dec.code_block_mode == RTE_BBDEV_TRANSPORT_BLOCK)
-- 
2.37.1


  parent reply	other threads:[~2022-11-16 20:53 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-17  4:46 [PATCH 21.11 0/7] baseband/acc100: backporting patches Hernan Vargas
2022-11-17  4:46 ` [PATCH 21.11 1/7] baseband/acc100: add LDPC encoder padding function Hernan Vargas
2022-11-17  4:46 ` Hernan Vargas [this message]
2022-11-17  4:46 ` [PATCH 21.11 3/7] baseband/acc100: fix ring availability calculation Hernan Vargas
2022-11-17  4:46 ` [PATCH 21.11 4/7] baseband/acc100: enforce additional check on FCW Hernan Vargas
2022-11-17  4:46 ` [PATCH 21.11 5/7] baseband/acc100: fix null HARQ input case Hernan Vargas
2022-11-17  4:46 ` [PATCH 21.11 6/7] baseband/acc100: fix ring/queue allocation Hernan Vargas
2022-11-17  4:46 ` [PATCH 21.11 7/7] baseband/acc100: fix double MSI intr in TB mode Hernan Vargas
2022-11-23 18:13 ` [PATCH 21.11 0/7] baseband/acc100: backporting patches Kevin Traynor

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=20221117044652.163000-3-hernan.vargas@intel.com \
    --to=hernan.vargas@intel.com \
    --cc=ktraynor@redhat.com \
    --cc=maxime.coquelin@redhat.com \
    --cc=nicolas.chautru@intel.com \
    --cc=stable@dpdk.org \
    /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).