From: David Wei <[email protected]>
To: [email protected], [email protected]
Cc: Jens Axboe <[email protected]>,
Pavel Begunkov <[email protected]>,
Jakub Kicinski <[email protected]>, Paolo Abeni <[email protected]>,
"David S. Miller" <[email protected]>,
Eric Dumazet <[email protected]>,
Jesper Dangaard Brouer <[email protected]>,
David Ahern <[email protected]>,
Mina Almasry <[email protected]>,
Stanislav Fomichev <[email protected]>,
Joe Damato <[email protected]>,
Pedro Tammela <[email protected]>
Subject: [PATCH net-next v11 17/21] io_uring/zcrx: set pp memory provider for an rx queue
Date: Thu, 16 Jan 2025 15:16:59 -0800 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
Set the page pool memory provider for the rx queue configured for zero
copy to io_uring. Then the rx queue is reset using
netdev_rx_queue_restart() and netdev core + page pool will take care of
filling the rx queue from the io_uring zero copy memory provider.
For now, there is only one ifq so its destruction happens implicitly
during io_uring cleanup.
Reviewed-by: Jens Axboe <[email protected]>
Signed-off-by: David Wei <[email protected]>
---
io_uring/zcrx.c | 47 +++++++++++++++++++++++++++++++++++++++--------
1 file changed, 39 insertions(+), 8 deletions(-)
diff --git a/io_uring/zcrx.c b/io_uring/zcrx.c
index cecfb04a117c..0cfa8c0ecff8 100644
--- a/io_uring/zcrx.c
+++ b/io_uring/zcrx.c
@@ -11,11 +11,12 @@
#include <net/page_pool/helpers.h>
#include <net/page_pool/memory_provider.h>
#include <net/netlink.h>
-
-#include <trace/events/page_pool.h>
+#include <net/netdev_rx_queue.h>
#include <net/tcp.h>
#include <net/rps.h>
+#include <trace/events/page_pool.h>
+
#include <uapi/linux/io_uring.h>
#include "io_uring.h"
@@ -275,8 +276,32 @@ static void io_zcrx_drop_netdev(struct io_zcrx_ifq *ifq)
spin_unlock(&ifq->lock);
}
+static void io_close_queue(struct io_zcrx_ifq *ifq)
+{
+ struct net_device *netdev;
+ netdevice_tracker netdev_tracker;
+ struct pp_memory_provider_params p = {
+ .mp_ops = &io_uring_pp_zc_ops,
+ .mp_priv = ifq,
+ };
+
+ if (ifq->if_rxq == -1)
+ return;
+
+ spin_lock(&ifq->lock);
+ netdev = ifq->netdev;
+ netdev_tracker = ifq->netdev_tracker;
+ ifq->netdev = NULL;
+ spin_unlock(&ifq->lock);
+
+ if (netdev)
+ net_mp_close_rxq(netdev, ifq->if_rxq, &p);
+ ifq->if_rxq = -1;
+}
+
static void io_zcrx_ifq_free(struct io_zcrx_ifq *ifq)
{
+ io_close_queue(ifq);
io_zcrx_drop_netdev(ifq);
if (ifq->area)
@@ -291,6 +316,7 @@ static void io_zcrx_ifq_free(struct io_zcrx_ifq *ifq)
int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
struct io_uring_zcrx_ifq_reg __user *arg)
{
+ struct pp_memory_provider_params mp_param = {};
struct io_uring_zcrx_area_reg area;
struct io_uring_zcrx_ifq_reg reg;
struct io_uring_region_desc rd;
@@ -341,7 +367,6 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
goto err;
ifq->rq_entries = reg.rq_entries;
- ifq->if_rxq = reg.if_rxq;
ret = -ENODEV;
ifq->netdev = netdev_get_by_index(current->nsproxy->net_ns, reg.if_idx,
@@ -358,16 +383,20 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
if (ret)
goto err;
+ mp_param.mp_ops = &io_uring_pp_zc_ops;
+ mp_param.mp_priv = ifq;
+ ret = net_mp_open_rxq(ifq->netdev, reg.if_rxq, &mp_param);
+ if (ret)
+ goto err;
+ ifq->if_rxq = reg.if_rxq;
+
reg.offsets.rqes = sizeof(struct io_uring);
reg.offsets.head = offsetof(struct io_uring, head);
reg.offsets.tail = offsetof(struct io_uring, tail);
if (copy_to_user(arg, ®, sizeof(reg)) ||
- copy_to_user(u64_to_user_ptr(reg.region_ptr), &rd, sizeof(rd))) {
- ret = -EFAULT;
- goto err;
- }
- if (copy_to_user(u64_to_user_ptr(reg.area_ptr), &area, sizeof(area))) {
+ copy_to_user(u64_to_user_ptr(reg.region_ptr), &rd, sizeof(rd)) ||
+ copy_to_user(u64_to_user_ptr(reg.area_ptr), &area, sizeof(area))) {
ret = -EFAULT;
goto err;
}
@@ -444,6 +473,8 @@ void io_shutdown_zcrx_ifqs(struct io_ring_ctx *ctx)
if (ctx->ifq)
io_zcrx_scrub(ctx->ifq);
+
+ io_close_queue(ctx->ifq);
}
static inline u32 io_zcrx_rqring_entries(struct io_zcrx_ifq *ifq)
--
2.43.5
next prev parent reply other threads:[~2025-01-16 23:17 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-16 23:16 [PATCH net-next v11 00/21] io_uring zero copy rx David Wei
2025-01-16 23:16 ` [PATCH net-next v11 01/21] net: page_pool: don't cast mp param to devmem David Wei
2025-01-16 23:16 ` [PATCH net-next v11 02/21] net: prefix devmem specific helpers David Wei
2025-01-16 23:16 ` [PATCH net-next v11 03/21] net: generalise net_iov chunk owners David Wei
2025-01-16 23:16 ` [PATCH net-next v11 04/21] net: page_pool: create hooks for custom memory providers David Wei
2025-01-17 1:46 ` Jakub Kicinski
2025-01-17 1:48 ` Jakub Kicinski
2025-01-17 2:15 ` David Wei
2025-01-16 23:16 ` [PATCH net-next v11 05/21] netdev: add io_uring memory provider info David Wei
2025-01-16 23:16 ` [PATCH net-next v11 06/21] net: page_pool: add callback for mp info printing David Wei
2025-01-16 23:16 ` [PATCH net-next v11 07/21] net: page_pool: add a mp hook to unregister_netdevice* David Wei
2025-01-16 23:16 ` [PATCH net-next v11 08/21] net: prepare for non devmem TCP memory providers David Wei
2025-01-16 23:16 ` [PATCH net-next v11 09/21] net: page_pool: add memory provider helpers David Wei
2025-01-17 1:47 ` Jakub Kicinski
2025-01-16 23:16 ` [PATCH net-next v11 10/21] net: add helpers for setting a memory provider on an rx queue David Wei
2025-01-17 1:50 ` Jakub Kicinski
2025-01-17 1:52 ` Jakub Kicinski
2025-01-17 2:17 ` David Wei
2025-01-17 2:25 ` Jakub Kicinski
2025-01-17 2:47 ` Pavel Begunkov
2025-01-17 22:11 ` Jakub Kicinski
2025-01-17 23:20 ` Pavel Begunkov
2025-01-18 2:08 ` Jakub Kicinski
2025-01-18 3:09 ` Pavel Begunkov
2025-01-16 23:16 ` [PATCH net-next v11 11/21] io_uring/zcrx: add interface queue and refill queue David Wei
2025-01-16 23:16 ` [PATCH net-next v11 12/21] io_uring/zcrx: add io_zcrx_area David Wei
2025-01-16 23:16 ` [PATCH net-next v11 13/21] io_uring/zcrx: grab a net device David Wei
2025-01-16 23:16 ` [PATCH net-next v11 14/21] io_uring/zcrx: implement zerocopy receive pp memory provider David Wei
2025-01-17 2:07 ` Jakub Kicinski
2025-01-17 2:17 ` David Wei
2025-01-16 23:16 ` [PATCH net-next v11 15/21] io_uring/zcrx: dma-map area for the device David Wei
2025-01-16 23:16 ` [PATCH net-next v11 16/21] io_uring/zcrx: add io_recvzc request David Wei
2025-01-16 23:16 ` David Wei [this message]
2025-01-17 2:13 ` [PATCH net-next v11 17/21] io_uring/zcrx: set pp memory provider for an rx queue Jakub Kicinski
2025-01-17 2:38 ` Pavel Begunkov
2025-01-16 23:17 ` [PATCH net-next v11 18/21] io_uring/zcrx: throttle receive requests David Wei
2025-01-16 23:17 ` [PATCH net-next v11 19/21] io_uring/zcrx: add copy fallback David Wei
2025-01-16 23:17 ` [PATCH net-next v11 20/21] net: add documentation for io_uring zcrx David Wei
2025-01-16 23:17 ` [PATCH net-next v11 21/21] io_uring/zcrx: add selftest David Wei
2025-01-17 14:28 ` [PATCH net-next v11 00/21] io_uring zero copy rx Paolo Abeni
2025-01-17 14:42 ` Pavel Begunkov
2025-01-17 16:05 ` Paolo Abeni
2025-01-17 16:19 ` Pavel Begunkov
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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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