mempool/ring: support RTS and HTS ring modes

Two new sync modes were introduced into rte_ring:
relaxed tail sync (RTS) and head/tail sync (HTS).
This change provides user with ability to select these
modes for ring based mempool via mempool ops API.

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Gage Eads <gage.eads@intel.com>

doc/guides/mempool/ring.rst

@@ -12,12 +12,14 @@ and can be selected via mempool ops API:
 - ``ring_mp_mc``
 
   The underlying **rte_ring** operates in multi-thread producer,
-  multi-thread consumer sync mode.
+  multi-thread consumer sync mode. For more information please refer to:
+  :ref:`Ring_Library_MPMC_Mode`.
 
 - ``ring_sp_sc``
 
   The underlying **rte_ring** operates in single-thread producer,
-  single-thread consumer sync mode.
+  single-thread consumer sync mode. For more information please refer to:
+  :ref:`Ring_Library_SPSC_Mode`.
 
 - ``ring_sp_mc``
 
@@ -29,6 +31,22 @@ and can be selected via mempool ops API:
   The underlying **rte_ring** operates in multi-thread producer,
   single-thread consumer sync mode.
 
+- ``ring_mt_rts``
 
+  For underlying **rte_ring** both producer and consumer operate in
+  multi-thread Relaxed Tail Sync (RTS) mode. For more information please
+  refer to: :ref:`Ring_Library_MT_RTS_Mode`.
+
+- ``ring_mt_hts``
+
+  For underlying **rte_ring** both producer and consumer operate in
+  multi-thread Head-Tail Sync (HTS) mode. For more information please
+  refer to: :ref:`Ring_Library_MT_HTS_Mode`.
+
+
+For 'classic' DPDK deployments (with one thread per core) the ``ring_mp_mc``
+mode is usually the most suitable and the fastest one. For overcommitted
+scenarios (multiple threads share same set of cores) the ``ring_mt_rts`` or
+``ring_mt_hts`` modes usually provide a better alternative.
 For more information about ``rte_ring`` structure, behaviour and available
 synchronisation modes please refer to: :doc:`../prog_guide/ring_lib`.

doc/guides/prog_guide/ring_lib.rst

@@ -359,6 +359,8 @@ That should help users to configure ring in the most suitable way for his
 specific usage scenarios.
 Currently supported modes:
 
+.. _Ring_Library_MPMC_Mode:
+
 MP/MC (default one)
 ~~~~~~~~~~~~~~~~~~~
 
@@ -369,11 +371,15 @@ per core) this is usually the most suitable and fastest synchronization mode.
 As a well known limitation - it can perform quite pure on some overcommitted
 scenarios.
 
+.. _Ring_Library_SPSC_Mode:
+
 SP/SC
 ~~~~~
 Single-producer (/single-consumer) mode. In this mode only one thread at a time
 is allowed to enqueue (/dequeue) objects to (/from) the ring.
 
+.. _Ring_Library_MT_RTS_Mode:
+
 MP_RTS/MC_RTS
 ~~~~~~~~~~~~~
 
@@ -390,6 +396,8 @@ one for head update, second for tail update.
 In comparison the original MP/MC algorithm requires one 32-bit CAS
 for head update and waiting/spinning on tail value.
 
+.. _Ring_Library_MT_HTS_Mode:
+
 MP_HTS/MC_HTS
 ~~~~~~~~~~~~~
 

doc/guides/rel_notes/release_20_08.rst

@@ -69,6 +69,12 @@ New Features
   barriers. rte_*mb APIs, for ARMv8 platforms, are changed to use DMB
   instruction to reflect this.
 
+* **Added support for RTS and HTS modes into mempool ring driver.**
+
+  Added ability to select new ring synchronisation modes:
+  ``relaxed tail sync (ring_mt_rts)`` and ``head/tail sync (ring_mt_hts)``
+  via mempool ops API.
+
 * **Added the support for vfio-pci new VF token interface.**
 
   From Linux 5.7, vfio-pci supports to bind both SR-IOV PF and the created VFs,

drivers/mempool/ring/rte_mempool_ring.c

@@ -25,6 +25,22 @@ common_ring_sp_enqueue(struct rte_mempool *mp, void * const *obj_table,
 			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
 }
 
+static int
+rts_ring_mp_enqueue(struct rte_mempool *mp, void * const *obj_table,
+	unsigned int n)
+{
+	return rte_ring_mp_rts_enqueue_bulk(mp->pool_data,
+			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
+}
+
+static int
+hts_ring_mp_enqueue(struct rte_mempool *mp, void * const *obj_table,
+	unsigned int n)
+{
+	return rte_ring_mp_hts_enqueue_bulk(mp->pool_data,
+			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
+}
+
 static int
 common_ring_mc_dequeue(struct rte_mempool *mp, void **obj_table, unsigned n)
 {
@@ -39,17 +55,30 @@ common_ring_sc_dequeue(struct rte_mempool *mp, void **obj_table, unsigned n)
 			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
 }
 
+static int
+rts_ring_mc_dequeue(struct rte_mempool *mp, void **obj_table, unsigned int n)
+{
+	return rte_ring_mc_rts_dequeue_bulk(mp->pool_data,
+			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
+}
+
+static int
+hts_ring_mc_dequeue(struct rte_mempool *mp, void **obj_table, unsigned int n)
+{
+	return rte_ring_mc_hts_dequeue_bulk(mp->pool_data,
+			obj_table, n, NULL) == 0 ? -ENOBUFS : 0;
+}
+
 static unsigned
 common_ring_get_count(const struct rte_mempool *mp)
 {
 	return rte_ring_count(mp->pool_data);
 }
 
-
 static int
-common_ring_alloc(struct rte_mempool *mp)
+ring_alloc(struct rte_mempool *mp, uint32_t rg_flags)
 {
-	int rg_flags = 0, ret;
+	int ret;
 	char rg_name[RTE_RING_NAMESIZE];
 	struct rte_ring *r;
 
@@ -60,12 +89,6 @@ common_ring_alloc(struct rte_mempool *mp)
 		return -rte_errno;
 	}
 
-	/* ring flags */
-	if (mp->flags & MEMPOOL_F_SP_PUT)
-		rg_flags |= RING_F_SP_ENQ;
-	if (mp->flags & MEMPOOL_F_SC_GET)
-		rg_flags |= RING_F_SC_DEQ;
-
 	/*
 	 * Allocate the ring that will be used to store objects.
 	 * Ring functions will return appropriate errors if we are
@@ -82,6 +105,31 @@ common_ring_alloc(struct rte_mempool *mp)
 	return 0;
 }
 
+static int
+common_ring_alloc(struct rte_mempool *mp)
+{
+	uint32_t rg_flags = 0;
+
+	if (mp->flags & MEMPOOL_F_SP_PUT)
+		rg_flags |= RING_F_SP_ENQ;
+	if (mp->flags & MEMPOOL_F_SC_GET)
+		rg_flags |= RING_F_SC_DEQ;
+
+	return ring_alloc(mp, rg_flags);
+}
+
+static int
+rts_ring_alloc(struct rte_mempool *mp)
+{
+	return ring_alloc(mp, RING_F_MP_RTS_ENQ | RING_F_MC_RTS_DEQ);
+}
+
+static int
+hts_ring_alloc(struct rte_mempool *mp)
+{
+	return ring_alloc(mp, RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
 static void
 common_ring_free(struct rte_mempool *mp)
 {
@@ -130,7 +178,29 @@ static const struct rte_mempool_ops ops_sp_mc = {
 	.get_count = common_ring_get_count,
 };
 
+/* ops for mempool with ring in MT_RTS sync mode */
+static const struct rte_mempool_ops ops_mt_rts = {
+	.name = "ring_mt_rts",
+	.alloc = rts_ring_alloc,
+	.free = common_ring_free,
+	.enqueue = rts_ring_mp_enqueue,
+	.dequeue = rts_ring_mc_dequeue,
+	.get_count = common_ring_get_count,
+};
+
+/* ops for mempool with ring in MT_HTS sync mode */
+static const struct rte_mempool_ops ops_mt_hts = {
+	.name = "ring_mt_hts",
+	.alloc = hts_ring_alloc,
+	.free = common_ring_free,
+	.enqueue = hts_ring_mp_enqueue,
+	.dequeue = hts_ring_mc_dequeue,
+	.get_count = common_ring_get_count,
+};
+
 MEMPOOL_REGISTER_OPS(ops_mp_mc);
 MEMPOOL_REGISTER_OPS(ops_sp_sc);
 MEMPOOL_REGISTER_OPS(ops_mp_sc);
 MEMPOOL_REGISTER_OPS(ops_sp_mc);
+MEMPOOL_REGISTER_OPS(ops_mt_rts);
+MEMPOOL_REGISTER_OPS(ops_mt_hts);