[PATCH] set namespace prefix to threads

[PATCH] set namespace prefix to threads

[Thomas Monjalon]

When looking at threads in a system, it can be confusing
to find some unknown threads without a clue it is started by DPDK.

Let's start all thread names with "dpdk-"
and the driver name if it comes from a driver.

One more constraint: the thread names are generally limited
to 16 characters, including NUL character.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
---
Please review carefully, I may have missed things.

Note: we should use rte_ctrl_thread_create() as much as possible.

---
 app/test/test_lcores.c                  | 2 +-
 drivers/event/dlb2/pf/base/dlb2_osdep.h | 2 +-
 drivers/net/ark/ark_ethdev.c            | 4 ++--
 drivers/net/ipn3ke/ipn3ke_representor.c | 2 +-
 drivers/net/kni/rte_eth_kni.c           | 2 +-
 drivers/net/mlx5/mlx5_hws_cnt.c         | 2 +-
 drivers/net/sfc/sfc_mae_counter.c       | 2 +-
 drivers/raw/ifpga/ifpga_rawdev.c        | 2 +-
 drivers/vdpa/ifc/ifcvf_vdpa.c           | 6 +++---
 drivers/vdpa/mlx5/mlx5_vdpa_cthread.c   | 2 +-
 drivers/vdpa/mlx5/mlx5_vdpa_event.c     | 2 +-
 examples/vhost/main.c                   | 4 ++--
 examples/vhost_blk/vhost_blk.c          | 2 +-
 lib/eal/common/eal_common_proc.c        | 2 +-
 lib/eal/freebsd/eal.c                   | 2 +-
 lib/eal/freebsd/eal_interrupts.c        | 2 +-
 lib/eal/linux/eal.c                     | 2 +-
 lib/eal/linux/eal_interrupts.c          | 2 +-
 lib/eal/linux/eal_timer.c               | 2 +-
 lib/eal/windows/eal.c                   | 2 +-
 lib/eventdev/rte_event_eth_rx_adapter.c | 2 +-
 lib/telemetry/telemetry.c               | 4 ++--
 lib/vhost/socket.c                      | 4 ++--
 lib/vhost/vduse.c                       | 2 +-
 24 files changed, 30 insertions(+), 30 deletions(-)

diff --git a/app/test/test_lcores.c b/app/test/test_lcores.c
index 2c945b0136..9522663f1a 100644
--- a/app/test/test_lcores.c
+++ b/app/test/test_lcores.c
@@ -362,7 +362,7 @@ test_ctrl_thread(void)
 	/* Create one control thread */
 	t = &ctrl_thread_context;
 	t->state = Thread_INIT;
-	if (rte_ctrl_thread_create((pthread_t *)&t->id, "test_ctrl_threads",
+	if (rte_ctrl_thread_create((pthread_t *)&t->id, "dpdk-test-ctrlt",
 					NULL, ctrl_thread_loop, t) != 0)
 		return -1;
 
diff --git a/drivers/event/dlb2/pf/base/dlb2_osdep.h b/drivers/event/dlb2/pf/base/dlb2_osdep.h
index cffe22f3c5..75860cca4d 100644
--- a/drivers/event/dlb2/pf/base/dlb2_osdep.h
+++ b/drivers/event/dlb2/pf/base/dlb2_osdep.h
@@ -200,7 +200,7 @@ static inline void os_schedule_work(struct dlb2_hw *hw)
 	dlb2_dev = container_of(hw, struct dlb2_dev, hw);
 
 	ret = rte_ctrl_thread_create(&complete_queue_map_unmap_thread,
-				     "dlb_queue_unmap_waiter",
+				     "dpdk-dlb-qunmap",
 				     NULL,
 				     dlb2_complete_queue_map_unmap,
 				     dlb2_dev);
diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_ethdev.c
index b2995427c8..30814650bf 100644
--- a/drivers/net/ark/ark_ethdev.c
+++ b/drivers/net/ark/ark_ethdev.c
@@ -600,11 +600,11 @@ eth_ark_dev_start(struct rte_eth_dev *dev)
 	if (!ark->isvf && ark->start_pg && !ark->pg_running) {
 		pthread_t thread;
 
-		/* Delay packet generatpr start allow the hardware to be ready
+		/* Delay packet generator start allow the hardware to be ready
 		 * This is only used for sanity checking with internal generator
 		 */
 		char tname[32];
-		snprintf(tname, sizeof(tname), "ark-delay-pg-%d",
+		snprintf(tname, sizeof(tname), "dpdk-ark-pg%d",
 			 dev->data->port_id);
 
 		if (rte_ctrl_thread_create(&thread, tname, NULL,
diff --git a/drivers/net/ipn3ke/ipn3ke_representor.c b/drivers/net/ipn3ke/ipn3ke_representor.c
index e50fc73f43..c82f8b5333 100644
--- a/drivers/net/ipn3ke/ipn3ke_representor.c
+++ b/drivers/net/ipn3ke/ipn3ke_representor.c
@@ -2593,7 +2593,7 @@ ipn3ke_rpst_scan_check(void)
 
 	if (ipn3ke_rpst_scan_num == 1) {
 		ret = rte_ctrl_thread_create(&ipn3ke_rpst_scan_thread,
-			"ipn3ke scanner",
+			"dpdk-ipn3ke-scn",
 			NULL,
 			ipn3ke_rpst_scan_handle_request, NULL);
 		if (ret) {
diff --git a/drivers/net/kni/rte_eth_kni.c b/drivers/net/kni/rte_eth_kni.c
index c0e1f8db40..6b90c4e9b5 100644
--- a/drivers/net/kni/rte_eth_kni.c
+++ b/drivers/net/kni/rte_eth_kni.c
@@ -163,7 +163,7 @@ eth_kni_dev_start(struct rte_eth_dev *dev)
 		internals->stop_thread = 0;
 
 		ret = rte_ctrl_thread_create(&internals->thread,
-			"kni_handle_req", NULL,
+			"dpdk-kni-req", NULL,
 			kni_handle_request, internals);
 		if (ret) {
 			PMD_LOG(ERR,
diff --git a/drivers/net/mlx5/mlx5_hws_cnt.c b/drivers/net/mlx5/mlx5_hws_cnt.c
index d98df68f39..415bb7dae4 100644
--- a/drivers/net/mlx5/mlx5_hws_cnt.c
+++ b/drivers/net/mlx5/mlx5_hws_cnt.c
@@ -453,7 +453,7 @@ mlx5_hws_cnt_service_thread_create(struct mlx5_dev_ctx_shared *sh)
 		DRV_LOG(ERR, "Failed to create HW steering's counter service thread.");
 		return -ENOSYS;
 	}
-	snprintf(name, CNT_THREAD_NAME_MAX - 1, "%s/svc@%d",
+	snprintf(name, CNT_THREAD_NAME_MAX - 1, "dpdk-mlx5-hws-%s/svc@%d",
 		 sh->ibdev_name, service_core);
 	rte_thread_set_name((rte_thread_t){(uintptr_t)sh->cnt_svc->service_thread},
 		name);
diff --git a/drivers/net/sfc/sfc_mae_counter.c b/drivers/net/sfc/sfc_mae_counter.c
index 79043ff7d7..631b54afeb 100644
--- a/drivers/net/sfc/sfc_mae_counter.c
+++ b/drivers/net/sfc/sfc_mae_counter.c
@@ -711,7 +711,7 @@ sfc_mae_counter_thread_spawn(struct sfc_adapter *sa,
 	counter_registry->polling.thread.run = true;
 
 	rc = rte_ctrl_thread_create(&sa->mae.counter_registry.polling.thread.id,
-				    "mae_counter_thread", NULL,
+				    "dpdk-sfc-maecnt", NULL,
 				    sfc_mae_counter_thread, sa);
 
 	return rc;
diff --git a/drivers/raw/ifpga/ifpga_rawdev.c b/drivers/raw/ifpga/ifpga_rawdev.c
index b9f6f432c8..a7b18299de 100644
--- a/drivers/raw/ifpga/ifpga_rawdev.c
+++ b/drivers/raw/ifpga/ifpga_rawdev.c
@@ -551,7 +551,7 @@ ifpga_monitor_start_func(struct ifpga_rawdev *dev)
 
 	if (!__atomic_fetch_add(&ifpga_monitor_refcnt, 1, __ATOMIC_RELAXED)) {
 		ret = rte_ctrl_thread_create(&ifpga_monitor_start_thread,
-					     "ifpga-monitor", NULL,
+					     "dpdk-ifpga-mon", NULL,
 					     ifpga_rawdev_gsd_handle, NULL);
 		if (ret != 0) {
 			ifpga_monitor_start_thread = 0;
diff --git a/drivers/vdpa/ifc/ifcvf_vdpa.c b/drivers/vdpa/ifc/ifcvf_vdpa.c
index e4133568c1..55346bbc08 100644
--- a/drivers/vdpa/ifc/ifcvf_vdpa.c
+++ b/drivers/vdpa/ifc/ifcvf_vdpa.c
@@ -570,7 +570,7 @@ setup_notify_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-notify-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-noti%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->tid, name, NULL, notify_relay,
 				     (void *)internal);
 	if (ret != 0) {
@@ -681,7 +681,7 @@ setup_intr_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-intr-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-int%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->intr_tid, name, NULL,
 				     intr_relay, (void *)internal);
 	if (ret) {
@@ -1019,7 +1019,7 @@ setup_vring_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-vring-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-ring%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->tid, name, NULL, vring_relay,
 				     (void *)internal);
 	if (ret != 0) {
diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c b/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
index 6e6624e5a3..a1d291bf71 100644
--- a/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
+++ b/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
@@ -312,7 +312,7 @@ mlx5_vdpa_c_thread_create(int cpu_core)
 			"vdpa multi-threads %d.", thrd_idx);
 			goto c_thread_err;
 		}
-		snprintf(name, sizeof(name), "vDPA-mthread-%d", thrd_idx);
+		snprintf(name, sizeof(name), "dpdk-vmlx5-c%d", thrd_idx);
 		ret = pthread_setname_np(
 				conf_thread_mng.cthrd[thrd_idx].tid, name);
 		if (ret)
diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_event.c b/drivers/vdpa/mlx5/mlx5_vdpa_event.c
index f3d392c73d..c486ded598 100644
--- a/drivers/vdpa/mlx5/mlx5_vdpa_event.c
+++ b/drivers/vdpa/mlx5/mlx5_vdpa_event.c
@@ -546,7 +546,7 @@ mlx5_vdpa_cqe_event_setup(struct mlx5_vdpa_priv *priv)
 		DRV_LOG(ERR, "Failed to set thread affinity.");
 		goto out;
 	}
-	snprintf(name, sizeof(name), "vDPA-mlx5-%d", priv->vid);
+	snprintf(name, sizeof(name), "dpdk-vmlx5-%d", priv->vid);
 	rte_thread_set_name((rte_thread_t){(uintptr_t)priv->timer_tid}, name);
 out:
 	if (attrp != NULL)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index bfe466ff77..9ff2c1f31b 100644
--- a/examples/vhost/main.c
+++ b/examples/vhost/main.c
@@ -1986,11 +1986,11 @@ main(int argc, char *argv[])
 
 	/* Enable stats if the user option is set. */
 	if (enable_stats) {
-		ret = rte_ctrl_thread_create(&tid, "print-stats", NULL,
+		ret = rte_ctrl_thread_create(&tid, "dpdk-vhost-stat", NULL,
 					print_stats, NULL);
 		if (ret < 0)
 			rte_exit(EXIT_FAILURE,
-				"Cannot create print-stats thread\n");
+				"Cannot create dpdk-vhost-stat thread\n");
 	}
 
 	/* Launch all data cores. */
diff --git a/examples/vhost_blk/vhost_blk.c b/examples/vhost_blk/vhost_blk.c
index 3709d7ed06..45699eb0b3 100644
--- a/examples/vhost_blk/vhost_blk.c
+++ b/examples/vhost_blk/vhost_blk.c
@@ -686,7 +686,7 @@ new_device(int vid)
 	/* start polling vring */
 	worker_thread_status = WORKER_STATE_START;
 	fprintf(stdout, "New Device %s, Device ID %d\n", path, vid);
-	if (rte_ctrl_thread_create(&tid, "vhostblk-ctrlr", NULL,
+	if (rte_ctrl_thread_create(&tid, "dpdk-vhost-blk", NULL,
 				   &ctrlr_worker, ctrlr) != 0) {
 		fprintf(stderr, "Worker Thread Started Failed\n");
 		return -1;
diff --git a/lib/eal/common/eal_common_proc.c b/lib/eal/common/eal_common_proc.c
index 7a038e0a3c..39ee9bcd7e 100644
--- a/lib/eal/common/eal_common_proc.c
+++ b/lib/eal/common/eal_common_proc.c
@@ -635,7 +635,7 @@ rte_mp_channel_init(void)
 		return -1;
 	}
 
-	if (rte_ctrl_thread_create(&mp_handle_tid, "rte_mp_handle",
+	if (rte_ctrl_thread_create(&mp_handle_tid, "dpdk-mp-msg",
 			NULL, mp_handle, NULL) < 0) {
 		RTE_LOG(ERR, EAL, "failed to create mp thread: %s\n",
 			strerror(errno));
diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
index 7008303e11..e201acd1c2 100644
--- a/lib/eal/freebsd/eal.c
+++ b/lib/eal/freebsd/eal.c
@@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread_name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-				"rte-worker-%d", i);
+				"dpdk-worker-%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eal/freebsd/eal_interrupts.c b/lib/eal/freebsd/eal_interrupts.c
index ffcf97ad4c..68cdb9deb3 100644
--- a/lib/eal/freebsd/eal_interrupts.c
+++ b/lib/eal/freebsd/eal_interrupts.c
@@ -637,7 +637,7 @@ rte_eal_intr_init(void)
 	}
 
 	/* create the host thread to wait/handle the interrupt */
-	ret = rte_ctrl_thread_create(&intr_thread, "eal-intr-thread", NULL,
+	ret = rte_ctrl_thread_create(&intr_thread, "dpdk-intr", NULL,
 			eal_intr_thread_main, NULL);
 	if (ret != 0) {
 		rte_errno = -ret;
diff --git a/lib/eal/linux/eal.c b/lib/eal/linux/eal.c
index 145afafde2..655653e2c1 100644
--- a/lib/eal/linux/eal.c
+++ b/lib/eal/linux/eal.c
@@ -1272,7 +1272,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread_name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-			"rte-worker-%d", i);
+			"dpdk-worker-%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eal/linux/eal_interrupts.c b/lib/eal/linux/eal_interrupts.c
index c9881143be..19b36787b8 100644
--- a/lib/eal/linux/eal_interrupts.c
+++ b/lib/eal/linux/eal_interrupts.c
@@ -1188,7 +1188,7 @@ rte_eal_intr_init(void)
 	}
 
 	/* create the host thread to wait/handle the interrupt */
-	ret = rte_ctrl_thread_create(&intr_thread, "eal-intr-thread", NULL,
+	ret = rte_ctrl_thread_create(&intr_thread, "dpdk-intr", NULL,
 			eal_intr_thread_main, NULL);
 	if (ret != 0) {
 		rte_errno = -ret;
diff --git a/lib/eal/linux/eal_timer.c b/lib/eal/linux/eal_timer.c
index 95c8cca992..1898709700 100644
--- a/lib/eal/linux/eal_timer.c
+++ b/lib/eal/linux/eal_timer.c
@@ -176,7 +176,7 @@ rte_eal_hpet_init(int make_default)
 
 	/* create a thread that will increment a global variable for
 	 * msb (hpet is 32 bits by default under linux) */
-	ret = rte_ctrl_thread_create(&msb_inc_thread_id, "hpet-msb-inc", NULL,
+	ret = rte_ctrl_thread_create(&msb_inc_thread_id, "dpdk-hpet-msb", NULL,
 				     hpet_msb_inc, NULL);
 	if (ret != 0) {
 		RTE_LOG(ERR, EAL, "ERROR: Cannot create HPET timer thread!\n");
diff --git a/lib/eal/windows/eal.c b/lib/eal/windows/eal.c
index 2d7a0e9ab2..d3c9de70f8 100644
--- a/lib/eal/windows/eal.c
+++ b/lib/eal/windows/eal.c
@@ -468,7 +468,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-			"rte-worker-%d", i);
+			"dpdk-worker-%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eventdev/rte_event_eth_rx_adapter.c b/lib/eventdev/rte_event_eth_rx_adapter.c
index f7f93ccdfd..ec5bec020d 100644
--- a/lib/eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/eventdev/rte_event_eth_rx_adapter.c
@@ -1620,7 +1620,7 @@ rxa_create_intr_thread(struct event_eth_rx_adapter *rx_adapter)
 	rte_spinlock_init(&rx_adapter->intr_ring_lock);
 
 	snprintf(thread_name, RTE_MAX_THREAD_NAME_LEN,
-			"rx-intr-thread-%d", rx_adapter->id);
+			"dpdk-evt-rx%d", rx_adapter->id);
 
 	err = rte_ctrl_thread_create(&rx_adapter->rx_intr_thread, thread_name,
 				NULL, rxa_intr_thread, rx_adapter);
diff --git a/lib/telemetry/telemetry.c b/lib/telemetry/telemetry.c
index 590720bfa6..76b1677250 100644
--- a/lib/telemetry/telemetry.c
+++ b/lib/telemetry/telemetry.c
@@ -563,7 +563,7 @@ telemetry_legacy_init(void)
 		return -1;
 	}
 	pthread_setaffinity_np(t_old, sizeof(*thread_cpuset), thread_cpuset);
-	set_thread_name(t_old, "telemetry-v1");
+	set_thread_name(t_old, "dpdk-telemet-v1");
 	TMTY_LOG(DEBUG, "Legacy telemetry socket initialized ok\n");
 	pthread_detach(t_old);
 	return 0;
@@ -617,7 +617,7 @@ telemetry_v2_init(void)
 		return -1;
 	}
 	pthread_setaffinity_np(t_new, sizeof(*thread_cpuset), thread_cpuset);
-	set_thread_name(t_new, "telemetry-v2");
+	set_thread_name(t_new, "dpdk-telemet-v2");
 	pthread_detach(t_new);
 	atexit(unlink_sockets);
 
diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c
index 19a7469e45..906db3871d 100644
--- a/lib/vhost/socket.c
+++ b/lib/vhost/socket.c
@@ -513,7 +513,7 @@ vhost_user_reconnect_init(void)
 	}
 	TAILQ_INIT(&reconn_list.head);
 
-	ret = rte_ctrl_thread_create(&reconn_tid, "vhost_reconn", NULL,
+	ret = rte_ctrl_thread_create(&reconn_tid, "dpdk-vhost-reco", NULL,
 			     vhost_user_client_reconnect, NULL);
 	if (ret != 0) {
 		VHOST_LOG_CONFIG("thread", ERR, "failed to create reconnect thread\n");
@@ -1239,7 +1239,7 @@ rte_vhost_driver_start(const char *path)
 		}
 
 		int ret = rte_ctrl_thread_create(&fdset_tid,
-			"vhost-events", NULL, fdset_event_dispatch,
+			"dpdk-vhost-evts", NULL, fdset_event_dispatch,
 			&vhost_user.fdset);
 		if (ret != 0) {
 			VHOST_LOG_CONFIG(path, ERR, "failed to create fdset handling thread\n");
diff --git a/lib/vhost/vduse.c b/lib/vhost/vduse.c
index a509daf80c..50dad7022f 100644
--- a/lib/vhost/vduse.c
+++ b/lib/vhost/vduse.c
@@ -455,7 +455,7 @@ vduse_device_create(const char *path)
 			return -1;
 		}
 
-		ret = rte_ctrl_thread_create(&fdset_tid, "vduse-events", NULL,
+		ret = rte_ctrl_thread_create(&fdset_tid, "dpdk-vduse-evt", NULL,
 				fdset_event_dispatch, &vduse.fdset);
 		if (ret != 0) {
 			VHOST_LOG_CONFIG(path, ERR, "failed to create vduse fdset handling thread\n");
-- 
2.41.0

Re: [PATCH] set namespace prefix to threads

[David Marchand]

On Mon, Jul 3, 2023 at 1:24 PM Thomas Monjalon <thomas@monjalon.net> wrote:
>
> When looking at threads in a system, it can be confusing
> to find some unknown threads without a clue it is started by DPDK.
>
> Let's start all thread names with "dpdk-"
> and the driver name if it comes from a driver.
>
> One more constraint: the thread names are generally limited
> to 16 characters, including NUL character.

Yes, and no way to discover this limit.


>
> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> ---
> Please review carefully, I may have missed things.
>
> Note: we should use rte_ctrl_thread_create() as much as possible.
>

As for this comment, we should advertise rte_thread_create_control()
instead, as it is thread library agnostic.
Which leads me to suggest looking at rte_thread_create_control() calls too.


> diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
> index 7008303e11..e201acd1c2 100644
> --- a/lib/eal/freebsd/eal.c
> +++ b/lib/eal/freebsd/eal.c
> @@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
>
>                 /* Set thread_name for aid in debugging. */
>                 snprintf(thread_name, sizeof(thread_name),
> -                               "rte-worker-%d", i);
> +                               "dpdk-worker-%d", i);
>                 rte_thread_set_name(lcore_config[i].thread_id, thread_name);
>
>                 ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,

We changed this not so long ago with 8ae946970ed3 ("eal: fix thread
name for high order lcores") which offered up to 4 digits for lcores.
With your proposal, we are back to only 3 digits, which is probably
enough, but I preferred to raise it.



-- 
David Marchand

Re: [PATCH] set namespace prefix to threads

[Thomas Monjalon]

03/07/2023 13:38, David Marchand:
> On Mon, Jul 3, 2023 at 1:24 PM Thomas Monjalon <thomas@monjalon.net> wrote:
> >
> > When looking at threads in a system, it can be confusing
> > to find some unknown threads without a clue it is started by DPDK.
> >
> > Let's start all thread names with "dpdk-"
> > and the driver name if it comes from a driver.
> >
> > One more constraint: the thread names are generally limited
> > to 16 characters, including NUL character.
> 
> Yes, and no way to discover this limit.
> 
> 
> >
> > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > ---
> > Please review carefully, I may have missed things.
> >
> > Note: we should use rte_ctrl_thread_create() as much as possible.
> >
> 
> As for this comment, we should advertise rte_thread_create_control()
> instead, as it is thread library agnostic.
> Which leads me to suggest looking at rte_thread_create_control() calls too.

You're right, I'll do a second pass.

> > diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
> > index 7008303e11..e201acd1c2 100644
> > --- a/lib/eal/freebsd/eal.c
> > +++ b/lib/eal/freebsd/eal.c
> > @@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
> >
> >                 /* Set thread_name for aid in debugging. */
> >                 snprintf(thread_name, sizeof(thread_name),
> > -                               "rte-worker-%d", i);
> > +                               "dpdk-worker-%d", i);
> >                 rte_thread_set_name(lcore_config[i].thread_id, thread_name);
> >
> >                 ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
> 
> We changed this not so long ago with 8ae946970ed3 ("eal: fix thread
> name for high order lcores") which offered up to 4 digits for lcores.
> With your proposal, we are back to only 3 digits, which is probably
> enough, but I preferred to raise it.

In general no need of hyphen before a number.
I will change to dpdk-worker%d.

Re: [PATCH] set namespace prefix to threads

[Bruce Richardson]

On Mon, Jul 03, 2023 at 01:49:55PM +0200, Thomas Monjalon wrote:
> 03/07/2023 13:38, David Marchand:
> > On Mon, Jul 3, 2023 at 1:24 PM Thomas Monjalon <thomas@monjalon.net> wrote:
> > >
> > > When looking at threads in a system, it can be confusing
> > > to find some unknown threads without a clue it is started by DPDK.
> > >
> > > Let's start all thread names with "dpdk-"
> > > and the driver name if it comes from a driver.
> > >
> > > One more constraint: the thread names are generally limited
> > > to 16 characters, including NUL character.
> > 
> > Yes, and no way to discover this limit.
> > 
> > 
> > >
> > > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > > ---
> > > Please review carefully, I may have missed things.
> > >
> > > Note: we should use rte_ctrl_thread_create() as much as possible.
> > >
> > 
> > As for this comment, we should advertise rte_thread_create_control()
> > instead, as it is thread library agnostic.
> > Which leads me to suggest looking at rte_thread_create_control() calls too.
> 
> You're right, I'll do a second pass.
> 
> > > diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
> > > index 7008303e11..e201acd1c2 100644
> > > --- a/lib/eal/freebsd/eal.c
> > > +++ b/lib/eal/freebsd/eal.c
> > > @@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
> > >
> > >                 /* Set thread_name for aid in debugging. */
> > >                 snprintf(thread_name, sizeof(thread_name),
> > > -                               "rte-worker-%d", i);
> > > +                               "dpdk-worker-%d", i);
> > >                 rte_thread_set_name(lcore_config[i].thread_id, thread_name);
> > >
> > >                 ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
> > 
> > We changed this not so long ago with 8ae946970ed3 ("eal: fix thread
> > name for high order lcores") which offered up to 4 digits for lcores.
> > With your proposal, we are back to only 3 digits, which is probably
> > enough, but I preferred to raise it.
> 
> In general no need of hyphen before a number.
> I will change to dpdk-worker%d.
> 
While I know that "dpdk-" is probably the clearer prefix, "rte-" is also
well-know as the standard API prefix for DPDK. Give we have a 16-char
limit, I would suggest keeping using the "rte" prefix on the basis of
brevity!
A third alternative is to just use "dp-" as prefix...

/Bruce

Re: [PATCH] set namespace prefix to threads

[Thomas Monjalon]

17/07/2023 16:09, Bruce Richardson:
> On Mon, Jul 03, 2023 at 01:49:55PM +0200, Thomas Monjalon wrote:
> > 03/07/2023 13:38, David Marchand:
> > > On Mon, Jul 3, 2023 at 1:24 PM Thomas Monjalon <thomas@monjalon.net> wrote:
> > > >
> > > > When looking at threads in a system, it can be confusing
> > > > to find some unknown threads without a clue it is started by DPDK.
> > > >
> > > > Let's start all thread names with "dpdk-"
> > > > and the driver name if it comes from a driver.
> > > >
> > > > One more constraint: the thread names are generally limited
> > > > to 16 characters, including NUL character.
> > > 
> > > Yes, and no way to discover this limit.
> > > 
> > > 
> > > >
> > > > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > > > ---
> > > > Please review carefully, I may have missed things.
> > > >
> > > > Note: we should use rte_ctrl_thread_create() as much as possible.
> > > >
> > > 
> > > As for this comment, we should advertise rte_thread_create_control()
> > > instead, as it is thread library agnostic.
> > > Which leads me to suggest looking at rte_thread_create_control() calls too.
> > 
> > You're right, I'll do a second pass.
> > 
> > > > diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
> > > > index 7008303e11..e201acd1c2 100644
> > > > --- a/lib/eal/freebsd/eal.c
> > > > +++ b/lib/eal/freebsd/eal.c
> > > > @@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
> > > >
> > > >                 /* Set thread_name for aid in debugging. */
> > > >                 snprintf(thread_name, sizeof(thread_name),
> > > > -                               "rte-worker-%d", i);
> > > > +                               "dpdk-worker-%d", i);
> > > >                 rte_thread_set_name(lcore_config[i].thread_id, thread_name);
> > > >
> > > >                 ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
> > > 
> > > We changed this not so long ago with 8ae946970ed3 ("eal: fix thread
> > > name for high order lcores") which offered up to 4 digits for lcores.
> > > With your proposal, we are back to only 3 digits, which is probably
> > > enough, but I preferred to raise it.
> > 
> > In general no need of hyphen before a number.
> > I will change to dpdk-worker%d.
> > 
> While I know that "dpdk-" is probably the clearer prefix, "rte-" is also
> well-know as the standard API prefix for DPDK. Give we have a 16-char
> limit, I would suggest keeping using the "rte" prefix on the basis of
> brevity!

It's one character less :)
"rte" is used for API.
"dpdk-" is used for file prefixes.
When running an application, users may understand what is a "dpdk-xyz" thread,
while "rte-xyz" is more cryptic if they are not involved with DPDK at all.

> A third alternative is to just use "dp-" as prefix...

A third alternative looks like more confusion.
The idea of the thread prefix is to indicate,
to anyone monitoring a system, where the threads come from.

Re: [PATCH] set namespace prefix to threads

[Bruce Richardson]

On Mon, Jul 17, 2023 at 06:07:59PM +0200, Thomas Monjalon wrote:
> 17/07/2023 16:09, Bruce Richardson:
> > On Mon, Jul 03, 2023 at 01:49:55PM +0200, Thomas Monjalon wrote:
> > > 03/07/2023 13:38, David Marchand:
> > > > On Mon, Jul 3, 2023 at 1:24 PM Thomas Monjalon <thomas@monjalon.net> wrote:
> > > > >
> > > > > When looking at threads in a system, it can be confusing
> > > > > to find some unknown threads without a clue it is started by DPDK.
> > > > >
> > > > > Let's start all thread names with "dpdk-"
> > > > > and the driver name if it comes from a driver.
> > > > >
> > > > > One more constraint: the thread names are generally limited
> > > > > to 16 characters, including NUL character.
> > > > 
> > > > Yes, and no way to discover this limit.
> > > > 
> > > > 
> > > > >
> > > > > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > > > > ---
> > > > > Please review carefully, I may have missed things.
> > > > >
> > > > > Note: we should use rte_ctrl_thread_create() as much as possible.
> > > > >
> > > > 
> > > > As for this comment, we should advertise rte_thread_create_control()
> > > > instead, as it is thread library agnostic.
> > > > Which leads me to suggest looking at rte_thread_create_control() calls too.
> > > 
> > > You're right, I'll do a second pass.
> > > 
> > > > > diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
> > > > > index 7008303e11..e201acd1c2 100644
> > > > > --- a/lib/eal/freebsd/eal.c
> > > > > +++ b/lib/eal/freebsd/eal.c
> > > > > @@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
> > > > >
> > > > >                 /* Set thread_name for aid in debugging. */
> > > > >                 snprintf(thread_name, sizeof(thread_name),
> > > > > -                               "rte-worker-%d", i);
> > > > > +                               "dpdk-worker-%d", i);
> > > > >                 rte_thread_set_name(lcore_config[i].thread_id, thread_name);
> > > > >
> > > > >                 ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
> > > > 
> > > > We changed this not so long ago with 8ae946970ed3 ("eal: fix thread
> > > > name for high order lcores") which offered up to 4 digits for lcores.
> > > > With your proposal, we are back to only 3 digits, which is probably
> > > > enough, but I preferred to raise it.
> > > 
> > > In general no need of hyphen before a number.
> > > I will change to dpdk-worker%d.
> > > 
> > While I know that "dpdk-" is probably the clearer prefix, "rte-" is also
> > well-know as the standard API prefix for DPDK. Give we have a 16-char
> > limit, I would suggest keeping using the "rte" prefix on the basis of
> > brevity!
> 
> It's one character less :)

Yes, but when you are limited to 16 characters, each char counts! "dpdk-"
uses 31% of the available chars, vs 25% for the shorter prefix.

> "rte" is used for API.
> "dpdk-" is used for file prefixes.
> When running an application, users may understand what is a "dpdk-xyz" thread,
> while "rte-xyz" is more cryptic if they are not involved with DPDK at all.
> 
> > A third alternative is to just use "dp-" as prefix...
> 
> A third alternative looks like more confusion.
> The idea of the thread prefix is to indicate,
> to anyone monitoring a system, where the threads come from.
> 
Sure. No massive objections either way, just keen to see the names being as
meaningful as possible.

[PATCH v2] set namespace prefix to threads

[Thomas Monjalon]

When looking at threads in a system, it can be confusing
to find some unknown threads without a clue it is started by DPDK.

Let's start all thread names with "dpdk-"
plus the driver name if it comes from a driver.

One more constraint: the thread names are generally limited
to 16 characters, including the NUL character.

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
---
Please review carefully, I may have missed things.

Note: we should use rte_thread_create_control() as much as possible.

---
 app/test/test_lcores.c                  | 2 +-
 app/test/test_threads.c                 | 2 +-
 drivers/dma/skeleton/skeleton_dmadev.c  | 2 +-
 drivers/event/dlb2/pf/base/dlb2_osdep.h | 2 +-
 drivers/net/ark/ark_ethdev.c            | 4 ++--
 drivers/net/iavf/iavf_vchnl.c           | 2 +-
 drivers/net/ice/ice_dcf_parent.c        | 2 +-
 drivers/net/ipn3ke/ipn3ke_representor.c | 2 +-
 drivers/net/ixgbe/ixgbe_ethdev.c        | 2 +-
 drivers/net/kni/rte_eth_kni.c           | 2 +-
 drivers/net/mlx5/mlx5_hws_cnt.c         | 6 ++----
 drivers/net/sfc/sfc_mae_counter.c       | 2 +-
 drivers/raw/ifpga/ifpga_rawdev.c        | 2 +-
 drivers/vdpa/ifc/ifcvf_vdpa.c           | 6 +++---
 drivers/vdpa/mlx5/mlx5_vdpa_cthread.c   | 2 +-
 drivers/vdpa/mlx5/mlx5_vdpa_event.c     | 2 +-
 examples/vhost/main.c                   | 4 ++--
 examples/vhost_blk/vhost_blk.c          | 2 +-
 lib/eal/common/eal_common_proc.c        | 2 +-
 lib/eal/freebsd/eal.c                   | 2 +-
 lib/eal/freebsd/eal_interrupts.c        | 2 +-
 lib/eal/linux/eal.c                     | 2 +-
 lib/eal/linux/eal_interrupts.c          | 2 +-
 lib/eal/linux/eal_timer.c               | 2 +-
 lib/eal/windows/eal.c                   | 2 +-
 lib/eal/windows/eal_interrupts.c        | 2 +-
 lib/eventdev/rte_event_eth_rx_adapter.c | 2 +-
 lib/telemetry/telemetry.c               | 4 ++--
 lib/vhost/socket.c                      | 4 ++--
 lib/vhost/vduse.c                       | 2 +-
 30 files changed, 37 insertions(+), 39 deletions(-)

diff --git a/app/test/test_lcores.c b/app/test/test_lcores.c
index 2c945b0136..9522663f1a 100644
--- a/app/test/test_lcores.c
+++ b/app/test/test_lcores.c
@@ -362,7 +362,7 @@ test_ctrl_thread(void)
 	/* Create one control thread */
 	t = &ctrl_thread_context;
 	t->state = Thread_INIT;
-	if (rte_ctrl_thread_create((pthread_t *)&t->id, "test_ctrl_threads",
+	if (rte_ctrl_thread_create((pthread_t *)&t->id, "dpdk-test-ctrlt",
 					NULL, ctrl_thread_loop, t) != 0)
 		return -1;
 
diff --git a/app/test/test_threads.c b/app/test/test_threads.c
index a4c4f651a4..4ad5902969 100644
--- a/app/test/test_threads.c
+++ b/app/test/test_threads.c
@@ -239,7 +239,7 @@ test_thread_control_create_join(void)
 	rte_thread_t thread_main_id;
 
 	thread_id_ready = 0;
-	RTE_TEST_ASSERT(rte_thread_create_control(&thread_id, "test_control_threads",
+	RTE_TEST_ASSERT(rte_thread_create_control(&thread_id, "dpdk-test-thcc",
 		NULL, thread_main, &thread_main_id) == 0,
 		"Failed to create thread.");
 
diff --git a/drivers/dma/skeleton/skeleton_dmadev.c b/drivers/dma/skeleton/skeleton_dmadev.c
index c2d776dbbd..493224542f 100644
--- a/drivers/dma/skeleton/skeleton_dmadev.c
+++ b/drivers/dma/skeleton/skeleton_dmadev.c
@@ -127,7 +127,7 @@ skeldma_start(struct rte_dma_dev *dev)
 
 	rte_mb();
 
-	snprintf(name, sizeof(name), "dma_skel_%d", dev->data->dev_id);
+	snprintf(name, sizeof(name), "dpdk-dma-skel%d", dev->data->dev_id);
 	ret = rte_thread_create_control(&hw->thread, name, NULL,
 				     cpucopy_thread, dev);
 	if (ret) {
diff --git a/drivers/event/dlb2/pf/base/dlb2_osdep.h b/drivers/event/dlb2/pf/base/dlb2_osdep.h
index cffe22f3c5..75860cca4d 100644
--- a/drivers/event/dlb2/pf/base/dlb2_osdep.h
+++ b/drivers/event/dlb2/pf/base/dlb2_osdep.h
@@ -200,7 +200,7 @@ static inline void os_schedule_work(struct dlb2_hw *hw)
 	dlb2_dev = container_of(hw, struct dlb2_dev, hw);
 
 	ret = rte_ctrl_thread_create(&complete_queue_map_unmap_thread,
-				     "dlb_queue_unmap_waiter",
+				     "dpdk-dlb-qunmap",
 				     NULL,
 				     dlb2_complete_queue_map_unmap,
 				     dlb2_dev);
diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_ethdev.c
index b2995427c8..30814650bf 100644
--- a/drivers/net/ark/ark_ethdev.c
+++ b/drivers/net/ark/ark_ethdev.c
@@ -600,11 +600,11 @@ eth_ark_dev_start(struct rte_eth_dev *dev)
 	if (!ark->isvf && ark->start_pg && !ark->pg_running) {
 		pthread_t thread;
 
-		/* Delay packet generatpr start allow the hardware to be ready
+		/* Delay packet generator start allow the hardware to be ready
 		 * This is only used for sanity checking with internal generator
 		 */
 		char tname[32];
-		snprintf(tname, sizeof(tname), "ark-delay-pg-%d",
+		snprintf(tname, sizeof(tname), "dpdk-ark-pg%d",
 			 dev->data->port_id);
 
 		if (rte_ctrl_thread_create(&thread, tname, NULL,
diff --git a/drivers/net/iavf/iavf_vchnl.c b/drivers/net/iavf/iavf_vchnl.c
index 524732f67d..c31a6d5c98 100644
--- a/drivers/net/iavf/iavf_vchnl.c
+++ b/drivers/net/iavf/iavf_vchnl.c
@@ -135,7 +135,7 @@ iavf_dev_event_handler_init(void)
 	TAILQ_INIT(&handler->pending);
 	pthread_mutex_init(&handler->lock, NULL);
 
-	if (rte_thread_create_control(&handler->tid, "iavf-event-thread",
+	if (rte_thread_create_control(&handler->tid, "dpdk-iavf-event",
 				NULL, iavf_dev_event_handle, NULL)) {
 		__atomic_fetch_sub(&handler->ndev, 1, __ATOMIC_RELAXED);
 		return -1;
diff --git a/drivers/net/ice/ice_dcf_parent.c b/drivers/net/ice/ice_dcf_parent.c
index d62837840d..ad98a531de 100644
--- a/drivers/net/ice/ice_dcf_parent.c
+++ b/drivers/net/ice/ice_dcf_parent.c
@@ -181,7 +181,7 @@ start_vsi_reset_thread(struct ice_dcf_hw *dcf_hw, bool vfr, uint16_t vf_id)
 	param->vfr = vfr;
 	param->vf_id = vf_id;
 
-	snprintf(name, sizeof(name), "ice-reset-%u", vf_id);
+	snprintf(name, sizeof(name), "dpdk-ice-rst%u", vf_id);
 	ret = rte_thread_create_control(&thread, name, NULL,
 				     ice_dcf_vsi_update_service_handler, param);
 	if (ret != 0) {
diff --git a/drivers/net/ipn3ke/ipn3ke_representor.c b/drivers/net/ipn3ke/ipn3ke_representor.c
index e50fc73f43..c82f8b5333 100644
--- a/drivers/net/ipn3ke/ipn3ke_representor.c
+++ b/drivers/net/ipn3ke/ipn3ke_representor.c
@@ -2593,7 +2593,7 @@ ipn3ke_rpst_scan_check(void)
 
 	if (ipn3ke_rpst_scan_num == 1) {
 		ret = rte_ctrl_thread_create(&ipn3ke_rpst_scan_thread,
-			"ipn3ke scanner",
+			"dpdk-ipn3ke-scn",
 			NULL,
 			ipn3ke_rpst_scan_handle_request, NULL);
 		if (ret) {
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index 14a7d571e0..7eac05ee60 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -4329,7 +4329,7 @@ ixgbe_dev_link_update_share(struct rte_eth_dev *dev,
 				 */
 				intr->flags |= IXGBE_FLAG_NEED_LINK_CONFIG;
 				if (rte_thread_create_control(&ad->link_thread_tid,
-					"ixgbe-link-handler",
+					"dpdk-ixgbe-link",
 					NULL,
 					ixgbe_dev_setup_link_thread_handler,
 					dev) < 0) {
diff --git a/drivers/net/kni/rte_eth_kni.c b/drivers/net/kni/rte_eth_kni.c
index c0e1f8db40..6b90c4e9b5 100644
--- a/drivers/net/kni/rte_eth_kni.c
+++ b/drivers/net/kni/rte_eth_kni.c
@@ -163,7 +163,7 @@ eth_kni_dev_start(struct rte_eth_dev *dev)
 		internals->stop_thread = 0;
 
 		ret = rte_ctrl_thread_create(&internals->thread,
-			"kni_handle_req", NULL,
+			"dpdk-kni-req", NULL,
 			kni_handle_request, internals);
 		if (ret) {
 			PMD_LOG(ERR,
diff --git a/drivers/net/mlx5/mlx5_hws_cnt.c b/drivers/net/mlx5/mlx5_hws_cnt.c
index d98df68f39..732c4598c4 100644
--- a/drivers/net/mlx5/mlx5_hws_cnt.c
+++ b/drivers/net/mlx5/mlx5_hws_cnt.c
@@ -439,8 +439,7 @@ mlx5_hws_cnt_pool_init(struct mlx5_dev_ctx_shared *sh,
 int
 mlx5_hws_cnt_service_thread_create(struct mlx5_dev_ctx_shared *sh)
 {
-#define CNT_THREAD_NAME_MAX 256
-	char name[CNT_THREAD_NAME_MAX];
+	char name[RTE_MAX_THREAD_NAME_LEN];
 	rte_cpuset_t cpuset;
 	int ret;
 	uint32_t service_core = sh->cnt_svc->service_core;
@@ -453,8 +452,7 @@ mlx5_hws_cnt_service_thread_create(struct mlx5_dev_ctx_shared *sh)
 		DRV_LOG(ERR, "Failed to create HW steering's counter service thread.");
 		return -ENOSYS;
 	}
-	snprintf(name, CNT_THREAD_NAME_MAX - 1, "%s/svc@%d",
-		 sh->ibdev_name, service_core);
+	snprintf(name, RTE_MAX_THREAD_NAME_LEN, "dpdk-mlx5-%d", service_core);
 	rte_thread_set_name((rte_thread_t){(uintptr_t)sh->cnt_svc->service_thread},
 		name);
 	CPU_SET(service_core, &cpuset);
diff --git a/drivers/net/sfc/sfc_mae_counter.c b/drivers/net/sfc/sfc_mae_counter.c
index 79043ff7d7..631b54afeb 100644
--- a/drivers/net/sfc/sfc_mae_counter.c
+++ b/drivers/net/sfc/sfc_mae_counter.c
@@ -711,7 +711,7 @@ sfc_mae_counter_thread_spawn(struct sfc_adapter *sa,
 	counter_registry->polling.thread.run = true;
 
 	rc = rte_ctrl_thread_create(&sa->mae.counter_registry.polling.thread.id,
-				    "mae_counter_thread", NULL,
+				    "dpdk-sfc-maecnt", NULL,
 				    sfc_mae_counter_thread, sa);
 
 	return rc;
diff --git a/drivers/raw/ifpga/ifpga_rawdev.c b/drivers/raw/ifpga/ifpga_rawdev.c
index b9f6f432c8..a7b18299de 100644
--- a/drivers/raw/ifpga/ifpga_rawdev.c
+++ b/drivers/raw/ifpga/ifpga_rawdev.c
@@ -551,7 +551,7 @@ ifpga_monitor_start_func(struct ifpga_rawdev *dev)
 
 	if (!__atomic_fetch_add(&ifpga_monitor_refcnt, 1, __ATOMIC_RELAXED)) {
 		ret = rte_ctrl_thread_create(&ifpga_monitor_start_thread,
-					     "ifpga-monitor", NULL,
+					     "dpdk-ifpga-mon", NULL,
 					     ifpga_rawdev_gsd_handle, NULL);
 		if (ret != 0) {
 			ifpga_monitor_start_thread = 0;
diff --git a/drivers/vdpa/ifc/ifcvf_vdpa.c b/drivers/vdpa/ifc/ifcvf_vdpa.c
index e4133568c1..55346bbc08 100644
--- a/drivers/vdpa/ifc/ifcvf_vdpa.c
+++ b/drivers/vdpa/ifc/ifcvf_vdpa.c
@@ -570,7 +570,7 @@ setup_notify_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-notify-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-noti%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->tid, name, NULL, notify_relay,
 				     (void *)internal);
 	if (ret != 0) {
@@ -681,7 +681,7 @@ setup_intr_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-intr-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-int%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->intr_tid, name, NULL,
 				     intr_relay, (void *)internal);
 	if (ret) {
@@ -1019,7 +1019,7 @@ setup_vring_relay(struct ifcvf_internal *internal)
 	char name[THREAD_NAME_LEN];
 	int ret;
 
-	snprintf(name, sizeof(name), "ifc-vring-%d", internal->vid);
+	snprintf(name, sizeof(name), "dpdk-ifc-ring%d", internal->vid);
 	ret = rte_ctrl_thread_create(&internal->tid, name, NULL, vring_relay,
 				     (void *)internal);
 	if (ret != 0) {
diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c b/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
index 6e6624e5a3..a1d291bf71 100644
--- a/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
+++ b/drivers/vdpa/mlx5/mlx5_vdpa_cthread.c
@@ -312,7 +312,7 @@ mlx5_vdpa_c_thread_create(int cpu_core)
 			"vdpa multi-threads %d.", thrd_idx);
 			goto c_thread_err;
 		}
-		snprintf(name, sizeof(name), "vDPA-mthread-%d", thrd_idx);
+		snprintf(name, sizeof(name), "dpdk-vmlx5-c%d", thrd_idx);
 		ret = pthread_setname_np(
 				conf_thread_mng.cthrd[thrd_idx].tid, name);
 		if (ret)
diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_event.c b/drivers/vdpa/mlx5/mlx5_vdpa_event.c
index f3d392c73d..c486ded598 100644
--- a/drivers/vdpa/mlx5/mlx5_vdpa_event.c
+++ b/drivers/vdpa/mlx5/mlx5_vdpa_event.c
@@ -546,7 +546,7 @@ mlx5_vdpa_cqe_event_setup(struct mlx5_vdpa_priv *priv)
 		DRV_LOG(ERR, "Failed to set thread affinity.");
 		goto out;
 	}
-	snprintf(name, sizeof(name), "vDPA-mlx5-%d", priv->vid);
+	snprintf(name, sizeof(name), "dpdk-vmlx5-%d", priv->vid);
 	rte_thread_set_name((rte_thread_t){(uintptr_t)priv->timer_tid}, name);
 out:
 	if (attrp != NULL)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index bfe466ff77..9ff2c1f31b 100644
--- a/examples/vhost/main.c
+++ b/examples/vhost/main.c
@@ -1986,11 +1986,11 @@ main(int argc, char *argv[])
 
 	/* Enable stats if the user option is set. */
 	if (enable_stats) {
-		ret = rte_ctrl_thread_create(&tid, "print-stats", NULL,
+		ret = rte_ctrl_thread_create(&tid, "dpdk-vhost-stat", NULL,
 					print_stats, NULL);
 		if (ret < 0)
 			rte_exit(EXIT_FAILURE,
-				"Cannot create print-stats thread\n");
+				"Cannot create dpdk-vhost-stat thread\n");
 	}
 
 	/* Launch all data cores. */
diff --git a/examples/vhost_blk/vhost_blk.c b/examples/vhost_blk/vhost_blk.c
index 3709d7ed06..45699eb0b3 100644
--- a/examples/vhost_blk/vhost_blk.c
+++ b/examples/vhost_blk/vhost_blk.c
@@ -686,7 +686,7 @@ new_device(int vid)
 	/* start polling vring */
 	worker_thread_status = WORKER_STATE_START;
 	fprintf(stdout, "New Device %s, Device ID %d\n", path, vid);
-	if (rte_ctrl_thread_create(&tid, "vhostblk-ctrlr", NULL,
+	if (rte_ctrl_thread_create(&tid, "dpdk-vhost-blk", NULL,
 				   &ctrlr_worker, ctrlr) != 0) {
 		fprintf(stderr, "Worker Thread Started Failed\n");
 		return -1;
diff --git a/lib/eal/common/eal_common_proc.c b/lib/eal/common/eal_common_proc.c
index 7a038e0a3c..39ee9bcd7e 100644
--- a/lib/eal/common/eal_common_proc.c
+++ b/lib/eal/common/eal_common_proc.c
@@ -635,7 +635,7 @@ rte_mp_channel_init(void)
 		return -1;
 	}
 
-	if (rte_ctrl_thread_create(&mp_handle_tid, "rte_mp_handle",
+	if (rte_ctrl_thread_create(&mp_handle_tid, "dpdk-mp-msg",
 			NULL, mp_handle, NULL) < 0) {
 		RTE_LOG(ERR, EAL, "failed to create mp thread: %s\n",
 			strerror(errno));
diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c
index 7008303e11..1c02aef716 100644
--- a/lib/eal/freebsd/eal.c
+++ b/lib/eal/freebsd/eal.c
@@ -843,7 +843,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread_name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-				"rte-worker-%d", i);
+				"dpdk-worker%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eal/freebsd/eal_interrupts.c b/lib/eal/freebsd/eal_interrupts.c
index ffcf97ad4c..68cdb9deb3 100644
--- a/lib/eal/freebsd/eal_interrupts.c
+++ b/lib/eal/freebsd/eal_interrupts.c
@@ -637,7 +637,7 @@ rte_eal_intr_init(void)
 	}
 
 	/* create the host thread to wait/handle the interrupt */
-	ret = rte_ctrl_thread_create(&intr_thread, "eal-intr-thread", NULL,
+	ret = rte_ctrl_thread_create(&intr_thread, "dpdk-intr", NULL,
 			eal_intr_thread_main, NULL);
 	if (ret != 0) {
 		rte_errno = -ret;
diff --git a/lib/eal/linux/eal.c b/lib/eal/linux/eal.c
index 145afafde2..44b8ba6897 100644
--- a/lib/eal/linux/eal.c
+++ b/lib/eal/linux/eal.c
@@ -1272,7 +1272,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread_name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-			"rte-worker-%d", i);
+			"dpdk-worker%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eal/linux/eal_interrupts.c b/lib/eal/linux/eal_interrupts.c
index c9881143be..19b36787b8 100644
--- a/lib/eal/linux/eal_interrupts.c
+++ b/lib/eal/linux/eal_interrupts.c
@@ -1188,7 +1188,7 @@ rte_eal_intr_init(void)
 	}
 
 	/* create the host thread to wait/handle the interrupt */
-	ret = rte_ctrl_thread_create(&intr_thread, "eal-intr-thread", NULL,
+	ret = rte_ctrl_thread_create(&intr_thread, "dpdk-intr", NULL,
 			eal_intr_thread_main, NULL);
 	if (ret != 0) {
 		rte_errno = -ret;
diff --git a/lib/eal/linux/eal_timer.c b/lib/eal/linux/eal_timer.c
index 95c8cca992..1898709700 100644
--- a/lib/eal/linux/eal_timer.c
+++ b/lib/eal/linux/eal_timer.c
@@ -176,7 +176,7 @@ rte_eal_hpet_init(int make_default)
 
 	/* create a thread that will increment a global variable for
 	 * msb (hpet is 32 bits by default under linux) */
-	ret = rte_ctrl_thread_create(&msb_inc_thread_id, "hpet-msb-inc", NULL,
+	ret = rte_ctrl_thread_create(&msb_inc_thread_id, "dpdk-hpet-msb", NULL,
 				     hpet_msb_inc, NULL);
 	if (ret != 0) {
 		RTE_LOG(ERR, EAL, "ERROR: Cannot create HPET timer thread!\n");
diff --git a/lib/eal/windows/eal.c b/lib/eal/windows/eal.c
index 2d7a0e9ab2..4fdb74c9cc 100644
--- a/lib/eal/windows/eal.c
+++ b/lib/eal/windows/eal.c
@@ -468,7 +468,7 @@ rte_eal_init(int argc, char **argv)
 
 		/* Set thread name for aid in debugging. */
 		snprintf(thread_name, sizeof(thread_name),
-			"rte-worker-%d", i);
+			"dpdk-worker%d", i);
 		rte_thread_set_name(lcore_config[i].thread_id, thread_name);
 
 		ret = rte_thread_set_affinity_by_id(lcore_config[i].thread_id,
diff --git a/lib/eal/windows/eal_interrupts.c b/lib/eal/windows/eal_interrupts.c
index 49c4b9620a..b1b1228c2b 100644
--- a/lib/eal/windows/eal_interrupts.c
+++ b/lib/eal/windows/eal_interrupts.c
@@ -98,7 +98,7 @@ rte_eal_intr_init(void)
 		return -1;
 	}
 
-	ret = rte_thread_create_control(&intr_thread, "eal-intr-thread", NULL,
+	ret = rte_thread_create_control(&intr_thread, "dpdk-intr", NULL,
 			eal_intr_thread_main, NULL);
 	if (ret != 0) {
 		rte_errno = -ret;
diff --git a/lib/eventdev/rte_event_eth_rx_adapter.c b/lib/eventdev/rte_event_eth_rx_adapter.c
index f7f93ccdfd..ec5bec020d 100644
--- a/lib/eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/eventdev/rte_event_eth_rx_adapter.c
@@ -1620,7 +1620,7 @@ rxa_create_intr_thread(struct event_eth_rx_adapter *rx_adapter)
 	rte_spinlock_init(&rx_adapter->intr_ring_lock);
 
 	snprintf(thread_name, RTE_MAX_THREAD_NAME_LEN,
-			"rx-intr-thread-%d", rx_adapter->id);
+			"dpdk-evt-rx%d", rx_adapter->id);
 
 	err = rte_ctrl_thread_create(&rx_adapter->rx_intr_thread, thread_name,
 				NULL, rxa_intr_thread, rx_adapter);
diff --git a/lib/telemetry/telemetry.c b/lib/telemetry/telemetry.c
index 590720bfa6..76b1677250 100644
--- a/lib/telemetry/telemetry.c
+++ b/lib/telemetry/telemetry.c
@@ -563,7 +563,7 @@ telemetry_legacy_init(void)
 		return -1;
 	}
 	pthread_setaffinity_np(t_old, sizeof(*thread_cpuset), thread_cpuset);
-	set_thread_name(t_old, "telemetry-v1");
+	set_thread_name(t_old, "dpdk-telemet-v1");
 	TMTY_LOG(DEBUG, "Legacy telemetry socket initialized ok\n");
 	pthread_detach(t_old);
 	return 0;
@@ -617,7 +617,7 @@ telemetry_v2_init(void)
 		return -1;
 	}
 	pthread_setaffinity_np(t_new, sizeof(*thread_cpuset), thread_cpuset);
-	set_thread_name(t_new, "telemetry-v2");
+	set_thread_name(t_new, "dpdk-telemet-v2");
 	pthread_detach(t_new);
 	atexit(unlink_sockets);
 
diff --git a/lib/vhost/socket.c b/lib/vhost/socket.c
index 19a7469e45..cef67c771d 100644
--- a/lib/vhost/socket.c
+++ b/lib/vhost/socket.c
@@ -513,7 +513,7 @@ vhost_user_reconnect_init(void)
 	}
 	TAILQ_INIT(&reconn_list.head);
 
-	ret = rte_ctrl_thread_create(&reconn_tid, "vhost_reconn", NULL,
+	ret = rte_ctrl_thread_create(&reconn_tid, "dpdk-vhost-reco", NULL,
 			     vhost_user_client_reconnect, NULL);
 	if (ret != 0) {
 		VHOST_LOG_CONFIG("thread", ERR, "failed to create reconnect thread\n");
@@ -1239,7 +1239,7 @@ rte_vhost_driver_start(const char *path)
 		}
 
 		int ret = rte_ctrl_thread_create(&fdset_tid,
-			"vhost-events", NULL, fdset_event_dispatch,
+			"dpdk-vhost-evt", NULL, fdset_event_dispatch,
 			&vhost_user.fdset);
 		if (ret != 0) {
 			VHOST_LOG_CONFIG(path, ERR, "failed to create fdset handling thread\n");
diff --git a/lib/vhost/vduse.c b/lib/vhost/vduse.c
index a509daf80c..50dad7022f 100644
--- a/lib/vhost/vduse.c
+++ b/lib/vhost/vduse.c
@@ -455,7 +455,7 @@ vduse_device_create(const char *path)
 			return -1;
 		}
 
-		ret = rte_ctrl_thread_create(&fdset_tid, "vduse-events", NULL,
+		ret = rte_ctrl_thread_create(&fdset_tid, "dpdk-vduse-evt", NULL,
 				fdset_event_dispatch, &vduse.fdset);
 		if (ret != 0) {
 			VHOST_LOG_CONFIG(path, ERR, "failed to create vduse fdset handling thread\n");
-- 
2.41.0

Re: [PATCH v2] set namespace prefix to threads

[fengchengwen]

+1 for same prefix.
Acked-by: Chengwen Feng <fengchengwen@huawei.com>

On 2023/7/3 20:14, Thomas Monjalon wrote:
> When looking at threads in a system, it can be confusing
> to find some unknown threads without a clue it is started by DPDK.
> 
> Let's start all thread names with "dpdk-"
> plus the driver name if it comes from a driver.
> 
> One more constraint: the thread names are generally limited
> to 16 characters, including the NUL character.
> 
> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> ---
> Please review carefully, I may have missed things.
> 
> Note: we should use rte_thread_create_control() as much as possible.
> 

...

> 

Re: [PATCH v2] set namespace prefix to threads

[Tyler Retzlaff]

On Mon, Jul 03, 2023 at 02:14:06PM +0200, Thomas Monjalon wrote:
> When looking at threads in a system, it can be confusing
> to find some unknown threads without a clue it is started by DPDK.
> 
> Let's start all thread names with "dpdk-"
> plus the driver name if it comes from a driver.
> 
> One more constraint: the thread names are generally limited
> to 16 characters, including the NUL character.
> 
> Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> ---
> Please review carefully, I may have missed things.

i think the only caution here is that some of the new names may be
truncated where the valuable information is lost in the truncation?

#define RTE_MAX_THREAD_NAME_LEN 16

a limit inherited from linux and adopted as the dpdk standard/maximum.
by prefixing dpdk- we end up with a bit less name length budget to spend
potentially causing the most important part of the thread name to be lost.

a couple of notes though.
  * some of the existing names may already suffer from truncation. i
    didn't check but can see some that look close.
  * some of your new names have reduced the length so they are in fact
    better.

Acked-by: Tyler Retzlaff <roretzla@linux.microsoft.com>

Re: [PATCH v2] set namespace prefix to threads

[Thomas Monjalon]

10/07/2023 20:42, Tyler Retzlaff:
> On Mon, Jul 03, 2023 at 02:14:06PM +0200, Thomas Monjalon wrote:
> > When looking at threads in a system, it can be confusing
> > to find some unknown threads without a clue it is started by DPDK.
> > 
> > Let's start all thread names with "dpdk-"
> > plus the driver name if it comes from a driver.
> > 
> > One more constraint: the thread names are generally limited
> > to 16 characters, including the NUL character.
> > 
> > Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
> > ---
> > Please review carefully, I may have missed things.
> 
> i think the only caution here is that some of the new names may be
> truncated where the valuable information is lost in the truncation?
> 
> #define RTE_MAX_THREAD_NAME_LEN 16
> 
> a limit inherited from linux and adopted as the dpdk standard/maximum.
> by prefixing dpdk- we end up with a bit less name length budget to spend
> potentially causing the most important part of the thread name to be lost.
> 
> a couple of notes though.
>   * some of the existing names may already suffer from truncation. i
>     didn't check but can see some that look close.
>   * some of your new names have reduced the length so they are in fact
>     better.
> 
> Acked-by: Tyler Retzlaff <roretzla@linux.microsoft.com>

Updated for the recently added txgbe thread,
and applied.