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 v9 16/20] io_uring/zcrx: set pp memory provider for an rx queue
Date: Tue, 17 Dec 2024 16:37:42 -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 | 83 ++++++++++++++++++++++++++++++++++++++++++++-----
1 file changed, 75 insertions(+), 8 deletions(-)
diff --git a/io_uring/zcrx.c b/io_uring/zcrx.c
index 1122c80502d6..756c78c0920e 100644
--- a/io_uring/zcrx.c
+++ b/io_uring/zcrx.c
@@ -10,11 +10,12 @@
#include <net/page_pool/helpers.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"
@@ -119,6 +120,65 @@ static void io_zcrx_get_niov_uref(struct net_iov *niov)
atomic_inc(io_get_user_counter(niov));
}
+static int io_open_zc_rxq(struct io_zcrx_ifq *ifq, unsigned ifq_idx)
+{
+ struct netdev_rx_queue *rxq;
+ struct net_device *dev = ifq->dev;
+ int ret;
+
+ ASSERT_RTNL();
+
+ if (ifq_idx >= dev->num_rx_queues)
+ return -EINVAL;
+ ifq_idx = array_index_nospec(ifq_idx, dev->num_rx_queues);
+
+ rxq = __netif_get_rx_queue(ifq->dev, ifq_idx);
+ if (rxq->mp_params.mp_priv)
+ return -EEXIST;
+
+ ifq->if_rxq = ifq_idx;
+ rxq->mp_params.mp_ops = &io_uring_pp_zc_ops;
+ rxq->mp_params.mp_priv = ifq;
+ ret = netdev_rx_queue_restart(ifq->dev, ifq->if_rxq);
+ if (ret)
+ goto fail;
+ return 0;
+fail:
+ rxq->mp_params.mp_ops = NULL;
+ rxq->mp_params.mp_priv = NULL;
+ ifq->if_rxq = -1;
+ return ret;
+}
+
+static void io_close_zc_rxq(struct io_zcrx_ifq *ifq)
+{
+ struct netdev_rx_queue *rxq;
+ int err;
+
+ if (ifq->if_rxq == -1)
+ return;
+
+ rtnl_lock();
+ if (WARN_ON_ONCE(ifq->if_rxq >= ifq->dev->num_rx_queues)) {
+ rtnl_unlock();
+ return;
+ }
+
+ rxq = __netif_get_rx_queue(ifq->dev, ifq->if_rxq);
+
+ WARN_ON_ONCE(rxq->mp_params.mp_priv != ifq);
+
+ rxq->mp_params.mp_ops = NULL;
+ rxq->mp_params.mp_priv = NULL;
+
+ err = netdev_rx_queue_restart(ifq->dev, ifq->if_rxq);
+ if (err)
+ pr_devel("io_uring: can't restart a queue on zcrx close\n");
+
+ rtnl_unlock();
+ ifq->if_rxq = -1;
+}
+
static int io_allocate_rbuf_ring(struct io_zcrx_ifq *ifq,
struct io_uring_zcrx_ifq_reg *reg,
struct io_uring_region_desc *rd)
@@ -255,6 +315,8 @@ static struct io_zcrx_ifq *io_zcrx_ifq_alloc(struct io_ring_ctx *ctx)
static void io_zcrx_ifq_free(struct io_zcrx_ifq *ifq)
{
+ io_close_zc_rxq(ifq);
+
if (ifq->area)
io_zcrx_free_area(ifq->area);
@@ -317,7 +379,6 @@ 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->dev = netdev_get_by_index(current->nsproxy->net_ns, reg.if_idx,
@@ -329,16 +390,20 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
if (ret)
goto err;
+ rtnl_lock();
+ ret = io_open_zc_rxq(ifq, reg.if_rxq);
+ rtnl_unlock();
+ if (ret)
+ goto err;
+
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))) {
+ io_close_zc_rxq(ifq);
ret = -EFAULT;
goto err;
}
@@ -415,6 +480,8 @@ void io_shutdown_zcrx_ifqs(struct io_ring_ctx *ctx)
if (ctx->ifq)
io_zcrx_scrub(ctx->ifq);
+
+ io_close_zc_rxq(ctx->ifq);
}
static inline u32 io_zcrx_rqring_entries(struct io_zcrx_ifq *ifq)
--
2.43.5
next prev parent reply other threads:[~2024-12-18 0:38 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-18 0:37 [PATCH RESEND net-next v9 00/21] io_uring zero copy rx David Wei
2024-12-18 0:37 ` [PATCH net-next v9 01/20] net: page_pool: don't cast mp param to devmem David Wei
2024-12-20 22:04 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 02/20] net: prefix devmem specific helpers David Wei
2024-12-20 22:05 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 03/20] net: generalise net_iov chunk owners David Wei
2024-12-20 22:14 ` Jakub Kicinski
2024-12-21 0:50 ` Pavel Begunkov
2024-12-21 2:17 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 04/20] net: page_pool: create hooks for custom page providers David Wei
2024-12-18 0:37 ` [PATCH net-next v9 05/20] net: page_pool: add mp op for netlink reporting David Wei
2024-12-20 22:16 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 06/20] net: page_pool: add a mp hook to unregister_netdevice* David Wei
2024-12-20 22:18 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 07/20] net: prepare for non devmem TCP memory providers David Wei
2024-12-20 22:18 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 08/20] net: expose page_pool_{set,clear}_pp_info David Wei
2024-12-20 22:31 ` Jakub Kicinski
2024-12-21 1:07 ` Pavel Begunkov
2024-12-21 2:23 ` Jakub Kicinski
2024-12-18 0:37 ` [PATCH net-next v9 09/20] net: page_pool: introduce page_pool_mp_return_in_cache David Wei
2024-12-18 0:37 ` [PATCH net-next v9 10/20] io_uring/zcrx: add interface queue and refill queue David Wei
2024-12-18 0:37 ` [PATCH net-next v9 11/20] io_uring/zcrx: add io_zcrx_area David Wei
2024-12-18 0:37 ` [PATCH net-next v9 12/20] io_uring/zcrx: grab a net device David Wei
2024-12-18 0:37 ` [PATCH net-next v9 13/20] net: page pool: export page_pool_set_dma_addr_netmem() David Wei
2024-12-18 0:37 ` [PATCH net-next v9 14/20] io_uring/zcrx: dma-map area for the device David Wei
2024-12-20 22:38 ` Jakub Kicinski
2024-12-21 1:04 ` Pavel Begunkov
2024-12-18 0:37 ` [PATCH net-next v9 15/20] io_uring/zcrx: add io_recvzc request David Wei
2024-12-18 0:37 ` David Wei [this message]
2024-12-18 0:37 ` [PATCH net-next v9 17/20] io_uring/zcrx: throttle receive requests David Wei
2024-12-18 0:37 ` [PATCH net-next v9 18/20] io_uring/zcrx: add copy fallback David Wei
2024-12-18 0:37 ` [PATCH net-next v9 19/20] net: add documentation for io_uring zcrx David Wei
2024-12-18 0:37 ` [PATCH net-next v9 20/20] io_uring/zcrx: add selftest David Wei
2024-12-18 14:40 ` [PATCH RESEND net-next v9 00/21] io_uring zero copy rx Jens Axboe
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