patch 'eal/unix: fix thread creation' has been queued to stable release 22.11.2

[Xueming Li]

Hi,

FYI, your patch has been queued to stable release 22.11.2

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 04/11/23. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.

Queued patches are on a temporary branch at:
https://git.dpdk.org/dpdk-stable/log/?h=22.11-staging

This queued commit can be viewed at:
https://git.dpdk.org/dpdk-stable/log/?h=22.11-staging/commit/db12268a953bbacdb8b518cb1138ed6b79d38491

Thanks.

Xueming Li <xuemingl at nvidia.com>

---
>From db12268a953bbacdb8b518cb1138ed6b79d38491 Mon Sep 17 00:00:00 2001
From: Tyler Retzlaff <roretzla at linux.microsoft.com>
Date: Wed, 15 Mar 2023 17:07:04 -0700
Subject: [PATCH] eal/unix: fix thread creation
Cc: Xueming Li <xuemingl at nvidia.com>

[ upstream commit b28c6196b132d1f25cb8c1bf781520fc41556b3a ]

In rte_thread_create setting affinity after pthread_create may fail.
Such a failure should result in the entire rte_thread_create failing
but doesn't.

Additionally if there is a failure to set affinity a race exists where
the creating thread will free ctx and depending on scheduling of the new
thread it may also free ctx (double free).

Resolve the above by setting the affinity from the newly created thread
using a condition variable to signal the completion of the thread
start wrapper having completed.

Since we are now waiting for the thread start wrapper to complete we can
allocate the thread start wrapper context on the stack. While here clean
up the variable naming in the context to better highlight the fields of
the context require synchronization between the creating and created
thread.

Fixes: ce6e911d20f6 ("eal: add thread lifetime API")

Signed-off-by: Tyler Retzlaff <roretzla at linux.microsoft.com>
Signed-off-by: David Marchand <david.marchand at redhat.com>
---
 lib/eal/unix/rte_thread.c | 73 ++++++++++++++++++++++++---------------
 1 file changed, 45 insertions(+), 28 deletions(-)

diff --git a/lib/eal/unix/rte_thread.c b/lib/eal/unix/rte_thread.c
index 37ebfcfca1..f4076122a4 100644
--- a/lib/eal/unix/rte_thread.c
+++ b/lib/eal/unix/rte_thread.c
@@ -5,6 +5,7 @@
 
 #include <errno.h>
 #include <pthread.h>
+#include <stdbool.h>
 #include <stdlib.h>
 #include <string.h>
 
@@ -16,9 +17,14 @@ struct eal_tls_key {
 	pthread_key_t thread_index;
 };
 
-struct thread_routine_ctx {
+struct thread_start_context {
 	rte_thread_func thread_func;
-	void *routine_args;
+	void *thread_args;
+	const rte_thread_attr_t *thread_attr;
+	pthread_mutex_t wrapper_mutex;
+	pthread_cond_t wrapper_cond;
+	int wrapper_ret;
+	bool wrapper_done;
 };
 
 static int
@@ -81,13 +87,29 @@ thread_map_os_priority_to_eal_priority(int policy, int os_pri,
 }
 
 static void *
-thread_func_wrapper(void *arg)
+thread_start_wrapper(void *arg)
 {
-	struct thread_routine_ctx ctx = *(struct thread_routine_ctx *)arg;
+	struct thread_start_context *ctx = (struct thread_start_context *)arg;
+	rte_thread_func thread_func = ctx->thread_func;
+	void *thread_args = ctx->thread_args;
+	int ret = 0;
+
+	if (ctx->thread_attr != NULL && CPU_COUNT(&ctx->thread_attr->cpuset) > 0) {
+		ret = rte_thread_set_affinity_by_id(rte_thread_self(), &ctx->thread_attr->cpuset);
+		if (ret != 0)
+			RTE_LOG(DEBUG, EAL, "rte_thread_set_affinity_by_id failed\n");
+	}
 
-	free(arg);
+	pthread_mutex_lock(&ctx->wrapper_mutex);
+	ctx->wrapper_ret = ret;
+	ctx->wrapper_done = true;
+	pthread_cond_signal(&ctx->wrapper_cond);
+	pthread_mutex_unlock(&ctx->wrapper_mutex);
 
-	return (void *)(uintptr_t)ctx.thread_func(ctx.routine_args);
+	if (ret != 0)
+		return NULL;
+
+	return (void *)(uintptr_t)thread_func(thread_args);
 }
 
 int
@@ -98,20 +120,18 @@ rte_thread_create(rte_thread_t *thread_id,
 	int ret = 0;
 	pthread_attr_t attr;
 	pthread_attr_t *attrp = NULL;
-	struct thread_routine_ctx *ctx;
 	struct sched_param param = {
 		.sched_priority = 0,
 	};
 	int policy = SCHED_OTHER;
-
-	ctx = calloc(1, sizeof(*ctx));
-	if (ctx == NULL) {
-		RTE_LOG(DEBUG, EAL, "Insufficient memory for thread context allocations\n");
-		ret = ENOMEM;
-		goto cleanup;
-	}
-	ctx->routine_args = args;
-	ctx->thread_func = thread_func;
+	struct thread_start_context ctx = {
+		.thread_func = thread_func,
+		.thread_args = args,
+		.thread_attr = thread_attr,
+		.wrapper_done = false,
+		.wrapper_mutex = PTHREAD_MUTEX_INITIALIZER,
+		.wrapper_cond = PTHREAD_COND_INITIALIZER,
+	};
 
 	if (thread_attr != NULL) {
 		ret = pthread_attr_init(&attr);
@@ -133,7 +153,6 @@ rte_thread_create(rte_thread_t *thread_id,
 			goto cleanup;
 		}
 
-
 		if (thread_attr->priority ==
 				RTE_THREAD_PRIORITY_REALTIME_CRITICAL) {
 			ret = ENOTSUP;
@@ -158,24 +177,22 @@ rte_thread_create(rte_thread_t *thread_id,
 	}
 
 	ret = pthread_create((pthread_t *)&thread_id->opaque_id, attrp,
-		thread_func_wrapper, ctx);
+		thread_start_wrapper, &ctx);
 	if (ret != 0) {
 		RTE_LOG(DEBUG, EAL, "pthread_create failed\n");
 		goto cleanup;
 	}
 
-	if (thread_attr != NULL && CPU_COUNT(&thread_attr->cpuset) > 0) {
-		ret = rte_thread_set_affinity_by_id(*thread_id,
-			&thread_attr->cpuset);
-		if (ret != 0) {
-			RTE_LOG(DEBUG, EAL, "rte_thread_set_affinity_by_id failed\n");
-			goto cleanup;
-		}
-	}
+	pthread_mutex_lock(&ctx.wrapper_mutex);
+	while (!ctx.wrapper_done)
+		pthread_cond_wait(&ctx.wrapper_cond, &ctx.wrapper_mutex);
+	ret = ctx.wrapper_ret;
+	pthread_mutex_unlock(&ctx.wrapper_mutex);
+
+	if (ret != 0)
+		pthread_join((pthread_t)thread_id->opaque_id, NULL);
 
-	ctx = NULL;
 cleanup:
-	free(ctx);
 	if (attrp != NULL)
 		pthread_attr_destroy(&attr);
 
-- 
2.25.1

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2023-04-09 21:45:41.320353400 +0800
+++ 0100-eal-unix-fix-thread-creation.patch	2023-04-09 21:45:38.719042200 +0800
@@ -1 +1 @@
-From b28c6196b132d1f25cb8c1bf781520fc41556b3a Mon Sep 17 00:00:00 2001
+From db12268a953bbacdb8b518cb1138ed6b79d38491 Mon Sep 17 00:00:00 2001
@@ -4,0 +5,3 @@
+Cc: Xueming Li <xuemingl at nvidia.com>
+
+[ upstream commit b28c6196b132d1f25cb8c1bf781520fc41556b3a ]
@@ -25 +27,0 @@
-Cc: stable at dpdk.org