DPDK patches and discussions
 help / color / mirror / Atom feed
From: Steven Lariau <steven.lariau@arm.com>
To: Gage Eads <gage.eads@intel.com>, Olivier Matz <olivier.matz@6wind.com>
Cc: dev@dpdk.org, honnappa.nagarahalli@arm.com, nd@arm.com,
	Steven Lariau <steven.lariau@arm.com>
Subject: [dpdk-dev] [PATCH 4/4] test/stack: remove atomics operations
Date: Wed,  5 Aug 2020 16:46:01 +0100	[thread overview]
Message-ID: <20200805154601.19609-5-steven.lariau@arm.com> (raw)
In-Reply-To: <20200805154601.19609-1-steven.lariau@arm.com>

Remove the part that checks if there is enough room in the stack, it's
always true as long as size of stack >= MAX_BULK*rte_lcore_count().
This check used an atomic cmpset, and read / write to a shared size
variable. These operations result in some form of synchronization
that might get in the way of the actual stack testing.

Signed-off-by: Steven Lariau <steven.lariau@arm.com>
Reviewed-by: Dharmik Thakkar <dharmik.thakkar@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
---
 app/test/test_stack.c | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

diff --git a/app/test/test_stack.c b/app/test/test_stack.c
index b9d6befd4..02bea7ec5 100644
--- a/app/test/test_stack.c
+++ b/app/test/test_stack.c
@@ -273,7 +273,6 @@ test_free_null(void)
 
 struct test_args {
 	struct rte_stack *s;
-	rte_atomic64_t *sz;
 };
 
 static int
@@ -284,21 +283,9 @@ stack_thread_push_pop(void *args)
 	int i;
 
 	for (i = 0; i < NUM_ITERS_PER_THREAD; i++) {
-		unsigned int success, num;
+		unsigned int num;
 
-		/* Reserve up to min(MAX_BULK, available slots) stack entries,
-		 * then push and pop those stack entries.
-		 */
-		do {
-			uint64_t sz = rte_atomic64_read(t->sz);
-			volatile uint64_t *sz_addr;
-
-			sz_addr = (volatile uint64_t *)t->sz;
-
-			num = RTE_MIN(rte_rand() % MAX_BULK, STACK_SIZE - sz);
-
-			success = rte_atomic64_cmpset(sz_addr, sz, sz + num);
-		} while (success == 0);
+		num = rte_rand() % MAX_BULK;
 
 		if (rte_stack_push(t->s, obj_table, num) != num) {
 			printf("[%s():%u] Failed to push %u pointers\n",
@@ -312,7 +299,6 @@ stack_thread_push_pop(void *args)
 			return -1;
 		}
 
-		rte_atomic64_sub(t->sz, num);
 	}
 
 	return 0;
@@ -324,7 +310,6 @@ test_stack_multithreaded(uint32_t flags)
 	struct test_args args;
 	unsigned int lcore_id;
 	struct rte_stack *s;
-	rte_atomic64_t size;
 	int result = 0;
 
 	if (rte_lcore_count() < 2) {
@@ -335,16 +320,14 @@ test_stack_multithreaded(uint32_t flags)
 	printf("[%s():%u] Running with %u lcores\n",
 	       __func__, __LINE__, rte_lcore_count());
 
-	s = rte_stack_create("test", STACK_SIZE, rte_socket_id(), flags);
+	s = rte_stack_create("test", MAX_BULK * rte_lcore_count(), rte_socket_id(), flags);
 	if (s == NULL) {
 		printf("[%s():%u] Failed to create a stack\n",
 		       __func__, __LINE__);
 		return -1;
 	}
 
-	rte_atomic64_init(&size);
 	args.s = s;
-	args.sz = &size;
 
 	if (rte_eal_mp_remote_launch(stack_thread_push_pop, &args, CALL_MASTER))
 		rte_panic("Failed to launch tests\n");
-- 
2.17.1


  parent reply	other threads:[~2020-08-06 10:28 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-05 15:45 [dpdk-dev] [PATCH 0/4] test/stack: improve multithreaded test Steven Lariau
2020-08-05 15:45 ` [dpdk-dev] [PATCH 1/4] test/stack: avoid trivial memory allocations Steven Lariau
2020-08-05 15:45 ` [dpdk-dev] [PATCH 2/4] test/stack: launch tests with mp remote launch API Steven Lariau
2020-08-05 15:46 ` [dpdk-dev] [PATCH 3/4] test/stack: propagate errors to main core Steven Lariau
2020-08-05 15:46 ` Steven Lariau [this message]
2020-08-05 15:57 [dpdk-dev] [PATCH 0/4] test/stack: improve multithreaded test Steven Lariau
2020-08-05 15:57 ` [dpdk-dev] [PATCH 4/4] test/stack: remove atomics operations Steven Lariau
2020-08-11 20:13   ` Eads, Gage

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=20200805154601.19609-5-steven.lariau@arm.com \
    --to=steven.lariau@arm.com \
    --cc=dev@dpdk.org \
    --cc=gage.eads@intel.com \
    --cc=honnappa.nagarahalli@arm.com \
    --cc=nd@arm.com \
    --cc=olivier.matz@6wind.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).