The issue arose due to changes in the DPDK read-write lock
implementation. Following these changes, the RW-lock no longer supports
recursion, implying that a single thread shouldn't obtain a read lock if
it already possesses one. The problem arises during initialization: the
rte_eal_init() function acquires the memory_hotplug_lock, and later on,
there are sequences of calls that acquire it again without releasing it.
* rte_eal_memory_init() -> eal_memalloc_init() -> rte_memseg_list_walk()
* rte_eal_memory_init() -> rte_eal_hugepage_init() ->
eal_dynmem_hugepage_init() -> rte_memseg_list_walk()
This scenario introduces the risk of a potential deadlock when concurrent
write locks are applied to the same memory_hotplug_lock. To address this
locally, we resolved the issue by replacing rte_memseg_list_walk() with
rte_memseg_list_walk_thread_unsafe().
Bugzilla ID: 1277
Fixes: 832cecc03d77 ("rwlock: prevent readers from starving writers")
Cc: sta...@dpdk.org
Signed-off-by: Artemy Kovalyov <artem...@nvidia.com>
---
lib/eal/common/eal_common_dynmem.c | 5 ++++-
lib/eal/include/generic/rte_rwlock.h | 4 ++++
lib/eal/linux/eal_memalloc.c | 7 +++++--
3 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/lib/eal/common/eal_common_dynmem.c
b/lib/eal/common/eal_common_dynmem.c
index bdbbe23..95da55d 100644
--- a/lib/eal/common/eal_common_dynmem.c
+++ b/lib/eal/common/eal_common_dynmem.c
@@ -251,7 +251,10 @@
*/
memset(&dummy, 0, sizeof(dummy));
dummy.hugepage_sz = hpi->hugepage_sz;
- if (rte_memseg_list_walk(hugepage_count_walk, &dummy) < 0)
+ /* memory_hotplug_lock is held during initialization, so it's
+ * safe to call thread-unsafe version.
+ */
+ if (rte_memseg_list_walk_thread_unsafe(hugepage_count_walk,
&dummy) < 0)
return -1;
for (i = 0; i < RTE_DIM(dummy.num_pages); i++) {
diff --git a/lib/eal/include/generic/rte_rwlock.h
b/lib/eal/include/generic/rte_rwlock.h
index 9e083bb..c98fc7d 100644
--- a/lib/eal/include/generic/rte_rwlock.h
+++ b/lib/eal/include/generic/rte_rwlock.h
@@ -80,6 +80,10 @@
/**
* Take a read lock. Loop until the lock is held.
*
+ * @note The RW lock isn't recursive, so calling this function on the same
+ * lock twice without releasing it could potentially result in a deadlock
+ * scenario when a write lock is involved.
+ *
* @param rwl
* A pointer to a rwlock structure.
*/
diff --git a/lib/eal/linux/eal_memalloc.c b/lib/eal/linux/eal_memalloc.c
index f8b1588..9853ec7 100644
--- a/lib/eal/linux/eal_memalloc.c
+++ b/lib/eal/linux/eal_memalloc.c
@@ -1740,7 +1740,10 @@ struct rte_memseg *
eal_get_internal_configuration();
if (rte_eal_process_type() == RTE_PROC_SECONDARY)
- if (rte_memseg_list_walk(secondary_msl_create_walk, NULL) < 0)
+ /* memory_hotplug_lock is held during initialization, so it's
+ * safe to call thread-unsafe version.
+ */
+ if
(rte_memseg_list_walk_thread_unsafe(secondary_msl_create_walk, NULL) < 0)
return -1;
if (rte_eal_process_type() == RTE_PROC_PRIMARY &&
internal_conf->in_memory) {
@@ -1778,7 +1781,7 @@ struct rte_memseg *
}
/* initialize all of the fd lists */
- if (rte_memseg_list_walk(fd_list_create_walk, NULL))
+ if (rte_memseg_list_walk_thread_unsafe(fd_list_create_walk, NULL))
return -1;
return 0;
}
--
1.8.3.1
