DPDK patches and discussions
 help / color / mirror / Atom feed
From: <pbhagavatula@marvell.com>
To: <jerinj@marvell.com>, Pavan Nikhilesh <pbhagavatula@marvell.com>,
	"Shijith Thotton" <sthotton@marvell.com>,
	Nithin Dabilpuram <ndabilpuram@marvell.com>,
	Kiran Kumar K <kirankumark@marvell.com>,
	Sunil Kumar Kori <skori@marvell.com>,
	Satha Rao <skoteshwar@marvell.com>,
	Harman Kalra <hkalra@marvell.com>,
	Anatoly Burakov <anatoly.burakov@intel.com>
Cc: <dev@dpdk.org>
Subject: [PATCH 02/20] event/cnxk: add CN20K specific device probe
Date: Thu, 3 Oct 2024 18:52:19 +0530	[thread overview]
Message-ID: <20241003132237.20193-2-pbhagavatula@marvell.com> (raw)
In-Reply-To: <20241003132237.20193-1-pbhagavatula@marvell.com>

From: Pavan Nikhilesh <pbhagavatula@marvell.com>

Add platform specific event device probe and remove, also add
event device info get function.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
---
 doc/guides/eventdevs/cnxk.rst          | 30 +++++----
 doc/guides/rel_notes/release_24_11.rst |  4 ++
 drivers/common/cnxk/roc_sso.c          | 10 ++-
 drivers/event/cnxk/cn20k_eventdev.c    | 93 ++++++++++++++++++++++++++
 drivers/event/cnxk/meson.build         |  8 ++-
 5 files changed, 128 insertions(+), 17 deletions(-)
 create mode 100644 drivers/event/cnxk/cn20k_eventdev.c

diff --git a/doc/guides/eventdevs/cnxk.rst b/doc/guides/eventdevs/cnxk.rst
index d038930594..3b79efc7f8 100644
--- a/doc/guides/eventdevs/cnxk.rst
+++ b/doc/guides/eventdevs/cnxk.rst
@@ -16,6 +16,7 @@ Supported OCTEON cnxk SoCs
 
 - CN9XX
 - CN10XX
+- CN20XX
 
 Features
 --------
@@ -36,7 +37,7 @@ Features of the OCTEON cnxk SSO PMD are:
   DRAM
 - HW accelerated dequeue timeout support to enable power management
 - HW managed event timers support through TIM, with high precision and
-  time granularity of 2.5us on CN9K and 1us on CN10K.
+  time granularity of 2.5us on CN9K and 1us on CN10K/CN20K.
 - Up to 256 TIM rings a.k.a event timer adapters.
 - Up to 8 rings traversed in parallel.
 - HW managed packets enqueued from ethdev to eventdev exposed through event eth
@@ -45,8 +46,8 @@ Features of the OCTEON cnxk SSO PMD are:
 - Lockfree Tx from event eth Tx adapter using ``RTE_ETH_TX_OFFLOAD_MT_LOCKFREE``
   capability while maintaining receive packet order.
 - Full Rx/Tx offload support defined through ethdev queue configuration.
-- HW managed event vectorization on CN10K for packets enqueued from ethdev to
-  eventdev configurable per each Rx queue in Rx adapter.
+- HW managed event vectorization on CN10K/CN20K for packets enqueued from ethdev
+  to eventdev configurable per each Rx queue in Rx adapter.
 - Event vector transmission via Tx adapter.
 - Up to 2 event link profiles.
 
@@ -78,10 +79,11 @@ Runtime Config Options
 
     -a 0002:0e:00.0,single_ws=1
 
-- ``CN10K Getwork mode``
+- ``CN10K/CN20K Getwork mode``
 
-  CN10K supports three getwork prefetch modes no prefetch[0], prefetch
-  immediately[1] and delayed prefetch on forward progress event[2].
+  CN10K/CN20K supports three getwork prefetch modes no prefetch[0],
+  prefetch immediately[1] and delayed prefetch on forward progress
+  event[2].
   The default getwork mode is 2.
 
   For example::
@@ -103,13 +105,13 @@ Runtime Config Options
 
     -a 0002:0e:00.0,qos=[1-50-50]
 
-- ``CN10K WQE stashing support``
+- ``CN10K/CN20K WQE stashing support``
 
-  CN10K supports stashing the scheduled WQE carried by `rte_event` to the
-  cores L2 Dcache. The number of cache lines to be stashed and the offset
-  is configurable per HWGRP i.e. event queue. The dictionary format is as
-  follows `[Qx|stash_offset|stash_length]` here the stash offset can be
-  a negative integer.
+  CN10K/CN20K supports stashing the scheduled WQE carried by `rte_event`
+  to the cores L2 Dcache. The number of cache lines to be stashed and the
+  offset is configurable per HWGRP i.e. event queue. The dictionary format
+  is as follows `[Qx|stash_offset|stash_length]` here the stash offset can
+  be a negative integer.
   By default, stashing is enabled on queues which have been connected to
   Rx adapter. Both MBUF and NIX_RX_WQE_HDR + NIX_RX_PARSE_S are stashed.
 
@@ -198,8 +200,8 @@ Runtime Config Options
 
     -a 0002:0e:00.0,tim_eclk_freq=122880000-1000000000-0
 
-Power Saving on CN10K
----------------------
+Power Saving on CN10K/CN20K
+---------------------------
 
 ARM cores can additionally use WFE when polling for transactions on SSO bus
 to save power i.e., in the event dequeue call ARM core can enter WFE and exit
diff --git a/doc/guides/rel_notes/release_24_11.rst b/doc/guides/rel_notes/release_24_11.rst
index 023f357d94..9d65594f2f 100644
--- a/doc/guides/rel_notes/release_24_11.rst
+++ b/doc/guides/rel_notes/release_24_11.rst
@@ -63,6 +63,10 @@ New Features
 
   * Added ethdev driver support for CN20K SoC.
 
+* **Updated Marvell cnxk event device driver.**
+
+  * Added eventdev driver support for CN20K SoC.
+
 
 Removed Items
 -------------
diff --git a/drivers/common/cnxk/roc_sso.c b/drivers/common/cnxk/roc_sso.c
index ff76565af9..b4cfe04d40 100644
--- a/drivers/common/cnxk/roc_sso.c
+++ b/drivers/common/cnxk/roc_sso.c
@@ -870,7 +870,10 @@ sso_update_msix_vec_count(struct roc_sso *roc_sso, uint16_t sso_vec_cnt)
 	if (idev == NULL)
 		return -ENODEV;
 
-	mbox_vec_cnt = RVU_PF_INT_VEC_AFPF_MBOX + 1;
+	if (roc_model_is_cn20k())
+		mbox_vec_cnt = RVU_MBOX_PF_INT_VEC_AFPF_MBOX + 1;
+	else
+		mbox_vec_cnt = RVU_PF_INT_VEC_AFPF_MBOX + 1;
 
 	/* Allocating vectors for the first time */
 	if (plt_intr_max_intr_get(pci_dev->intr_handle) == 0) {
@@ -1017,7 +1020,10 @@ roc_sso_rsrc_init(struct roc_sso *roc_sso, uint8_t nb_hws, uint16_t nb_hwgrp, ui
 	}
 
 	/* 2 error interrupt per TIM LF */
-	sso_vec_cnt += 2 * nb_tim_lfs;
+	if (roc_model_is_cn20k())
+		sso_vec_cnt += 3 * nb_tim_lfs;
+	else
+		sso_vec_cnt += 2 * nb_tim_lfs;
 
 	rc = sso_update_msix_vec_count(roc_sso, sso_vec_cnt);
 	if (rc < 0) {
diff --git a/drivers/event/cnxk/cn20k_eventdev.c b/drivers/event/cnxk/cn20k_eventdev.c
new file mode 100644
index 0000000000..c4b80f64f3
--- /dev/null
+++ b/drivers/event/cnxk/cn20k_eventdev.c
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2024 Marvell.
+ */
+
+#include "roc_api.h"
+
+#include "cnxk_eventdev.h"
+
+static void
+cn20k_sso_set_rsrc(void *arg)
+{
+	struct cnxk_sso_evdev *dev = arg;
+
+	dev->max_event_ports = dev->sso.max_hws;
+	dev->max_event_queues = dev->sso.max_hwgrp > RTE_EVENT_MAX_QUEUES_PER_DEV ?
+					RTE_EVENT_MAX_QUEUES_PER_DEV :
+					dev->sso.max_hwgrp;
+}
+
+static void
+cn20k_sso_info_get(struct rte_eventdev *event_dev, struct rte_event_dev_info *dev_info)
+{
+	struct cnxk_sso_evdev *dev = cnxk_sso_pmd_priv(event_dev);
+
+	dev_info->driver_name = RTE_STR(EVENTDEV_NAME_CN20K_PMD);
+	cnxk_sso_info_get(dev, dev_info);
+	dev_info->max_event_port_enqueue_depth = UINT32_MAX;
+}
+
+static struct eventdev_ops cn20k_sso_dev_ops = {
+	.dev_infos_get = cn20k_sso_info_get,
+};
+
+static int
+cn20k_sso_init(struct rte_eventdev *event_dev)
+{
+	struct cnxk_sso_evdev *dev = cnxk_sso_pmd_priv(event_dev);
+	int rc;
+
+	rc = roc_plt_init();
+	if (rc < 0) {
+		plt_err("Failed to initialize platform model");
+		return rc;
+	}
+
+	event_dev->dev_ops = &cn20k_sso_dev_ops;
+	/* For secondary processes, the primary has done all the work */
+	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+		return 0;
+
+	rc = cnxk_sso_init(event_dev);
+	if (rc < 0)
+		return rc;
+
+	cn20k_sso_set_rsrc(cnxk_sso_pmd_priv(event_dev));
+	if (!dev->max_event_ports || !dev->max_event_queues) {
+		plt_err("Not enough eventdev resource queues=%d ports=%d", dev->max_event_queues,
+			dev->max_event_ports);
+		cnxk_sso_fini(event_dev);
+		return -ENODEV;
+	}
+
+	plt_sso_dbg("Initializing %s max_queues=%d max_ports=%d", event_dev->data->name,
+		    dev->max_event_queues, dev->max_event_ports);
+
+	return 0;
+}
+
+static int
+cn20k_sso_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
+{
+	return rte_event_pmd_pci_probe(pci_drv, pci_dev, sizeof(struct cnxk_sso_evdev),
+				       cn20k_sso_init);
+}
+
+static const struct rte_pci_id cn20k_pci_sso_map[] = {
+	CNXK_PCI_ID(PCI_SUBSYSTEM_DEVID_CN20KA, PCI_DEVID_CNXK_RVU_SSO_TIM_PF),
+	CNXK_PCI_ID(PCI_SUBSYSTEM_DEVID_CN20KA, PCI_DEVID_CNXK_RVU_SSO_TIM_VF),
+	{
+		.vendor_id = 0,
+	},
+};
+
+static struct rte_pci_driver cn20k_pci_sso = {
+	.id_table = cn20k_pci_sso_map,
+	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_NEED_IOVA_AS_VA,
+	.probe = cn20k_sso_probe,
+	.remove = cnxk_sso_remove,
+};
+
+RTE_PMD_REGISTER_PCI(event_cn20k, cn20k_pci_sso);
+RTE_PMD_REGISTER_PCI_TABLE(event_cn20k, cn20k_pci_sso_map);
+RTE_PMD_REGISTER_KMOD_DEP(event_cn20k, "vfio-pci");
diff --git a/drivers/event/cnxk/meson.build b/drivers/event/cnxk/meson.build
index e0e350570d..ec74a0d6ce 100644
--- a/drivers/event/cnxk/meson.build
+++ b/drivers/event/cnxk/meson.build
@@ -14,7 +14,7 @@ else
         soc_type = platform
 endif
 
-if soc_type != 'cn9k' and soc_type != 'cn10k'
+if soc_type != 'cn9k' and soc_type != 'cn10k' and soc_type != 'cn20k'
         soc_type = 'all'
 endif
 
@@ -325,6 +325,12 @@ sources += files(
 endif
 endif
 
+if soc_type == 'cn20k' or soc_type == 'all'
+sources += files(
+        'cn20k_eventdev.c',
+)
+endif
+
 extra_flags = ['-flax-vector-conversions', '-Wno-strict-aliasing']
 if cc.get_id() == 'clang'
         extra_flags += ['-Wno-asm-operand-widths']
-- 
2.25.1


  reply	other threads:[~2024-10-03 13:22 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-03 13:22 [PATCH 01/20] common/cnxk: implement SSO HW info pbhagavatula
2024-10-03 13:22 ` pbhagavatula [this message]
2024-10-03 13:22 ` [PATCH 03/20] event/cnxk: add CN20K device config pbhagavatula
2024-10-03 13:22 ` [PATCH 04/20] event/cnxk: add CN20k event queue config pbhagavatula
2024-10-03 13:22 ` [PATCH 05/20] event/cnxk: add CN20K event port configuration pbhagavatula
2024-10-03 13:22 ` [PATCH 06/20] event/cnxk: add CN20K SSO enqueue fast path pbhagavatula
2024-10-03 13:22 ` [PATCH 07/20] event/cnxk: add CN20K SSO dequeue " pbhagavatula
2024-10-03 13:22 ` [PATCH 08/20] event/cnxk: add CN20K event port quiesce pbhagavatula
2024-10-03 13:22 ` [PATCH 09/20] event/cnxk: add CN20K event port profile switch pbhagavatula
2024-10-03 13:22 ` [PATCH 10/20] event/cnxk: add CN20K device start pbhagavatula
2024-10-03 13:22 ` [PATCH 11/20] event/cnxk: add CN20K device stop and close pbhagavatula
2024-10-03 13:22 ` [PATCH 12/20] event/cnxk: add CN20K xstats, selftest and dump pbhagavatula
2024-10-03 13:22 ` [PATCH 13/20] event/cnxk: support CN20K Rx adapter pbhagavatula
2024-10-03 13:22 ` [PATCH 14/20] event/cnxk: support CN20K Rx adapter fast path pbhagavatula
2024-10-03 13:22 ` [PATCH 15/20] event/cnxk: support CN20K Tx adapter pbhagavatula
2024-10-03 13:22 ` [PATCH 16/20] event/cnxk: support CN20K Tx adapter fast path pbhagavatula
2024-10-03 13:22 ` [PATCH 17/20] common/cnxk: add SSO event aggregator pbhagavatula
2024-10-03 13:22 ` [PATCH 18/20] event/cnxk: add Rx/Tx event vector support pbhagavatula
2024-10-03 13:22 ` [PATCH 19/20] common/cnxk: update timer base code pbhagavatula
2024-10-03 13:22 ` [PATCH 20/20] event/cnxk: add CN20K timer adapter pbhagavatula

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=20241003132237.20193-2-pbhagavatula@marvell.com \
    --to=pbhagavatula@marvell.com \
    --cc=anatoly.burakov@intel.com \
    --cc=dev@dpdk.org \
    --cc=hkalra@marvell.com \
    --cc=jerinj@marvell.com \
    --cc=kirankumark@marvell.com \
    --cc=ndabilpuram@marvell.com \
    --cc=skori@marvell.com \
    --cc=skoteshwar@marvell.com \
    --cc=sthotton@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).