DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ashwin Sekhar T K <asekhar@marvell.com>
To: <dev@dpdk.org>, Nithin Dabilpuram <ndabilpuram@marvell.com>,
	Kiran Kumar K <kirankumark@marvell.com>,
	Sunil Kumar Kori <skori@marvell.com>,
	Satha Rao <skoteshwar@marvell.com>,
	Ashwin Sekhar T K <asekhar@marvell.com>,
	"Pavan Nikhilesh" <pbhagavatula@marvell.com>
Cc: <jerinj@marvell.com>, <psatheesh@marvell.com>,
	<anoobj@marvell.com>, <gakhil@marvell.com>, <hkalra@marvell.com>
Subject: [PATCH v2 1/2] mempool/cnxk: fix indefinite wait in batch alloc
Date: Mon, 29 May 2023 14:55:44 +0530	[thread overview]
Message-ID: <20230529092545.959180-1-asekhar@marvell.com> (raw)
In-Reply-To: <20230526134507.885354-1-asekhar@marvell.com>

Avoid waiting indefinitely when counting batch allocated
pointers by adding a wait timeout.

Fixes: 50d08d3934ec ("common/cnxk: fix batch alloc completion poll logic")

Signed-off-by: Ashwin Sekhar T K <asekhar@marvell.com>
---
 drivers/common/cnxk/roc_npa.h            | 15 +++++++++------
 drivers/mempool/cnxk/cn10k_mempool_ops.c |  3 ++-
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/common/cnxk/roc_npa.h b/drivers/common/cnxk/roc_npa.h
index 21608a40d9..d3caa71586 100644
--- a/drivers/common/cnxk/roc_npa.h
+++ b/drivers/common/cnxk/roc_npa.h
@@ -241,19 +241,23 @@ roc_npa_aura_batch_alloc_issue(uint64_t aura_handle, uint64_t *buf,
 }
 
 static inline void
-roc_npa_batch_alloc_wait(uint64_t *cache_line)
+roc_npa_batch_alloc_wait(uint64_t *cache_line, unsigned int wait_us)
 {
+	const uint64_t ticks = (uint64_t)wait_us * plt_tsc_hz() / (uint64_t)1E6;
+	const uint64_t start = plt_tsc_cycles();
+
 	/* Batch alloc status code is updated in bits [5:6] of the first word
 	 * of the 128 byte cache line.
 	 */
 	while (((__atomic_load_n(cache_line, __ATOMIC_RELAXED) >> 5) & 0x3) ==
 	       ALLOC_CCODE_INVAL)
-		;
+		if (wait_us && (plt_tsc_cycles() - start) >= ticks)
+			break;
 }
 
 static inline unsigned int
 roc_npa_aura_batch_alloc_count(uint64_t *aligned_buf, unsigned int num,
-			       unsigned int do_wait)
+			       unsigned int wait_us)
 {
 	unsigned int count, i;
 
@@ -267,8 +271,7 @@ roc_npa_aura_batch_alloc_count(uint64_t *aligned_buf, unsigned int num,
 
 		status = (struct npa_batch_alloc_status_s *)&aligned_buf[i];
 
-		if (do_wait)
-			roc_npa_batch_alloc_wait(&aligned_buf[i]);
+		roc_npa_batch_alloc_wait(&aligned_buf[i], wait_us);
 
 		count += status->count;
 	}
@@ -293,7 +296,7 @@ roc_npa_aura_batch_alloc_extract(uint64_t *buf, uint64_t *aligned_buf,
 
 		status = (struct npa_batch_alloc_status_s *)&aligned_buf[i];
 
-		roc_npa_batch_alloc_wait(&aligned_buf[i]);
+		roc_npa_batch_alloc_wait(&aligned_buf[i], 0);
 
 		line_count = status->count;
 
diff --git a/drivers/mempool/cnxk/cn10k_mempool_ops.c b/drivers/mempool/cnxk/cn10k_mempool_ops.c
index ba826f0f01..ff0015d8de 100644
--- a/drivers/mempool/cnxk/cn10k_mempool_ops.c
+++ b/drivers/mempool/cnxk/cn10k_mempool_ops.c
@@ -9,6 +9,7 @@
 
 #define BATCH_ALLOC_SZ              ROC_CN10K_NPA_BATCH_ALLOC_MAX_PTRS
 #define BATCH_OP_DATA_TABLE_MZ_NAME "batch_op_data_table_mz"
+#define BATCH_ALLOC_WAIT_US         5
 
 enum batch_op_status {
 	BATCH_ALLOC_OP_NOT_ISSUED = 0,
@@ -178,7 +179,7 @@ cn10k_mempool_get_count(const struct rte_mempool *mp)
 
 		if (mem->status == BATCH_ALLOC_OP_ISSUED)
 			count += roc_npa_aura_batch_alloc_count(
-				mem->objs, BATCH_ALLOC_SZ, 1);
+				mem->objs, BATCH_ALLOC_SZ, BATCH_ALLOC_WAIT_US);
 
 		if (mem->status == BATCH_ALLOC_OP_DONE)
 			count += mem->sz;
-- 
2.25.1


  parent reply	other threads:[~2023-05-29  9:25 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-26 13:45 [PATCH 1/2] mempool/cnxk: avoid indefinite wait Ashwin Sekhar T K
2023-05-26 13:45 ` [PATCH 2/2] common/cnxk: add new APIs for batch operations Ashwin Sekhar T K
2023-05-29  9:14 ` [PATCH 1/2] mempool/cnxk: avoid indefinite wait Jerin Jacob
2023-05-29  9:25 ` Ashwin Sekhar T K [this message]
2023-05-29  9:25   ` [PATCH v2 2/2] common/cnxk: add new APIs for batch operations Ashwin Sekhar T K
2023-05-30  9:12 ` [PATCH v3] " Ashwin Sekhar T K
2023-05-30 16:51   ` Jerin Jacob

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=20230529092545.959180-1-asekhar@marvell.com \
    --to=asekhar@marvell.com \
    --cc=anoobj@marvell.com \
    --cc=dev@dpdk.org \
    --cc=gakhil@marvell.com \
    --cc=hkalra@marvell.com \
    --cc=jerinj@marvell.com \
    --cc=kirankumark@marvell.com \
    --cc=ndabilpuram@marvell.com \
    --cc=pbhagavatula@marvell.com \
    --cc=psatheesh@marvell.com \
    --cc=skori@marvell.com \
    --cc=skoteshwar@marvell.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).