* [PATCH 01/11] io_uring: add interface queue
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 02/11] io_uring: add mmap support for shared ifq ringbuffers David Wei
` (10 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch introduces a new object in io_uring called an interface queue
(ifq) which contains:
* A pool region allocated by userspace and registered w/ io_uring where
RX data is written to.
* A net device and one specific RX queue in it that will be configured
for ZC RX.
* A pair of shared ringbuffers w/ userspace, dubbed registered buf
(rbuf) rings. Each entry contains a pool region id and an offset + len
within that region. The kernel writes entries into the completion ring
to tell userspace where RX data is relative to the start of a region.
Userspace writes entries into the refill ring to tell the kernel when
it is done with the data.
For now, each io_uring instance has a single ifq, and each ifq has a
single pool region associated with one RX queue.
Add a new opcode to io_uring_register that sets up an ifq. Size and
offsets of shared ringbuffers are returned to userspace for it to mmap.
The implementation will be added in a later patch.
Signed-off-by: David Wei <[email protected]>
---
include/linux/io_uring_types.h | 6 ++++
include/uapi/linux/io_uring.h | 50 ++++++++++++++++++++++++++
io_uring/Makefile | 3 +-
io_uring/io_uring.c | 7 ++++
io_uring/kbuf.c | 30 ++++++++++++++++
io_uring/kbuf.h | 5 +++
io_uring/zc_rx.c | 65 ++++++++++++++++++++++++++++++++++
io_uring/zc_rx.h | 21 +++++++++++
8 files changed, 186 insertions(+), 1 deletion(-)
create mode 100644 io_uring/zc_rx.c
create mode 100644 io_uring/zc_rx.h
diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h
index 1b2a20a42413..8f6068da185c 100644
--- a/include/linux/io_uring_types.h
+++ b/include/linux/io_uring_types.h
@@ -151,6 +151,10 @@ struct io_rings {
struct io_uring_cqe cqes[] ____cacheline_aligned_in_smp;
};
+struct io_rbuf_ring {
+ struct io_uring rq, cq;
+};
+
struct io_restriction {
DECLARE_BITMAP(register_op, IORING_REGISTER_LAST);
DECLARE_BITMAP(sqe_op, IORING_OP_LAST);
@@ -330,6 +334,8 @@ struct io_ring_ctx {
struct io_rsrc_data *file_data;
struct io_rsrc_data *buf_data;
+ struct io_zc_rx_ifq *ifq;
+
/* protected by ->uring_lock */
struct list_head rsrc_ref_list;
struct io_alloc_cache rsrc_node_cache;
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 0716cb17e436..8f2a1061629b 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -523,6 +523,9 @@ enum {
/* register a range of fixed file slots for automatic slot allocation */
IORING_REGISTER_FILE_ALLOC_RANGE = 25,
+ /* register a network interface queue for zerocopy */
+ IORING_REGISTER_ZC_RX_IFQ = 26,
+
/* this goes last */
IORING_REGISTER_LAST,
@@ -703,6 +706,53 @@ struct io_uring_recvmsg_out {
__u32 flags;
};
+struct io_uring_rbuf_rqe {
+ __u32 off;
+ __u32 len;
+ __u16 region;
+ __u8 __pad[6];
+};
+
+struct io_uring_rbuf_cqe {
+ __u32 off;
+ __u32 len;
+ __u16 region;
+ __u8 flags;
+ __u8 __pad[3];
+};
+
+struct io_rbuf_rqring_offsets {
+ __u32 head;
+ __u32 tail;
+ __u32 rqes;
+ __u8 __pad[4];
+};
+
+struct io_rbuf_cqring_offsets {
+ __u32 head;
+ __u32 tail;
+ __u32 cqes;
+ __u8 __pad[4];
+};
+
+/*
+ * Argument for IORING_REGISTER_ZC_RX_IFQ
+ */
+struct io_uring_zc_rx_ifq_reg {
+ __u32 if_idx;
+ /* hw rx descriptor ring id */
+ __u32 if_rxq_id;
+ __u32 region_id;
+ __u32 rq_entries;
+ __u32 cq_entries;
+ __u32 flags;
+ __u16 cpu;
+
+ __u32 mmap_sz;
+ struct io_rbuf_rqring_offsets rq_off;
+ struct io_rbuf_cqring_offsets cq_off;
+};
+
#ifdef __cplusplus
}
#endif
diff --git a/io_uring/Makefile b/io_uring/Makefile
index 8cc8e5387a75..7818b015a1f2 100644
--- a/io_uring/Makefile
+++ b/io_uring/Makefile
@@ -7,5 +7,6 @@ obj-$(CONFIG_IO_URING) += io_uring.o xattr.o nop.o fs.o splice.o \
openclose.o uring_cmd.o epoll.o \
statx.o net.o msg_ring.o timeout.o \
sqpoll.o fdinfo.o tctx.o poll.o \
- cancel.o kbuf.o rsrc.o rw.o opdef.o notif.o
+ cancel.o kbuf.o rsrc.o rw.o opdef.o \
+ notif.o zc_rx.o
obj-$(CONFIG_IO_WQ) += io-wq.o
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 3bca7a79efda..7705d18dceff 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -92,6 +92,7 @@
#include "cancel.h"
#include "net.h"
#include "notif.h"
+#include "zc_rx.h"
#include "timeout.h"
#include "poll.h"
@@ -4418,6 +4419,12 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
break;
ret = io_register_file_alloc_range(ctx, arg);
break;
+ case IORING_REGISTER_ZC_RX_IFQ:
+ ret = -EINVAL;
+ if (!arg || nr_args != 1)
+ break;
+ ret = io_register_zc_rx_ifq(ctx, arg);
+ break;
default:
ret = -EINVAL;
break;
diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c
index 2f0181521c98..d7499e7b34bd 100644
--- a/io_uring/kbuf.c
+++ b/io_uring/kbuf.c
@@ -634,3 +634,33 @@ void *io_pbuf_get_address(struct io_ring_ctx *ctx, unsigned long bgid)
return bl->buf_ring;
}
+
+int io_allocate_rbuf_ring(struct io_zc_rx_ifq *ifq,
+ struct io_uring_zc_rx_ifq_reg *reg)
+{
+ gfp_t gfp = GFP_KERNEL_ACCOUNT | __GFP_ZERO | __GFP_NOWARN | __GFP_COMP;
+ size_t off, size, rq_size, cq_size;
+ void *ptr;
+
+ off = sizeof(struct io_rbuf_ring);
+ rq_size = reg->rq_entries * sizeof(struct io_uring_rbuf_rqe);
+ cq_size = reg->cq_entries * sizeof(struct io_uring_rbuf_cqe);
+ size = off + rq_size + cq_size;
+ ptr = (void *) __get_free_pages(gfp, get_order(size));
+ if (!ptr)
+ return -ENOMEM;
+ ifq->ring = (struct io_rbuf_ring *)ptr;
+ ifq->rqes = (struct io_uring_rbuf_rqe *)((char *)ptr + off);
+ ifq->cqes = (struct io_uring_rbuf_cqe *)((char *)ifq->rqes + rq_size);
+
+ return 0;
+}
+
+void io_free_rbuf_ring(struct io_zc_rx_ifq *ifq)
+{
+ struct page *page;
+
+ page = virt_to_head_page(ifq->ring);
+ if (put_page_testzero(page))
+ free_compound_page(page);
+}
diff --git a/io_uring/kbuf.h b/io_uring/kbuf.h
index d14345ef61fc..6c8afda93646 100644
--- a/io_uring/kbuf.h
+++ b/io_uring/kbuf.h
@@ -4,6 +4,8 @@
#include <uapi/linux/io_uring.h>
+#include "zc_rx.h"
+
struct io_buffer_list {
/*
* If ->buf_nr_pages is set, then buf_pages/buf_ring are used. If not,
@@ -57,6 +59,9 @@ void io_kbuf_recycle_legacy(struct io_kiocb *req, unsigned issue_flags);
void *io_pbuf_get_address(struct io_ring_ctx *ctx, unsigned long bgid);
+int io_allocate_rbuf_ring(struct io_zc_rx_ifq *ifq, struct io_uring_zc_rx_ifq_reg *reg);
+void io_free_rbuf_ring(struct io_zc_rx_ifq *ifq);
+
static inline void io_kbuf_recycle_ring(struct io_kiocb *req)
{
/*
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
new file mode 100644
index 000000000000..63bc6cd7d205
--- /dev/null
+++ b/io_uring/zc_rx.c
@@ -0,0 +1,65 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/io_uring.h>
+
+#include <uapi/linux/io_uring.h>
+
+#include "io_uring.h"
+#include "kbuf.h"
+#include "zc_rx.h"
+
+static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
+{
+ struct io_zc_rx_ifq *ifq;
+
+ ifq = kzalloc(sizeof(*ifq), GFP_KERNEL);
+ if (!ifq)
+ return NULL;
+
+ ifq->ctx = ctx;
+
+ return ifq;
+}
+
+static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
+{
+ io_free_rbuf_ring(ifq);
+ kfree(ifq);
+}
+
+int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
+ struct io_uring_zc_rx_ifq_reg __user *arg)
+{
+ struct io_uring_zc_rx_ifq_reg reg;
+ struct io_zc_rx_ifq *ifq;
+ int ret;
+
+ if (copy_from_user(®, arg, sizeof(reg)))
+ return -EFAULT;
+ if (ctx->ifq)
+ return -EBUSY;
+
+ ifq = io_zc_rx_ifq_alloc(ctx);
+ if (!ifq)
+ return -ENOMEM;
+
+ /* TODO: initialise network interface */
+
+ ret = io_allocate_rbuf_ring(ifq, ®);
+ if (ret)
+ goto err;
+
+ /* TODO: map zc region and initialise zc pool */
+
+ ifq->rq_entries = reg.rq_entries;
+ ifq->cq_entries = reg.cq_entries;
+ ifq->if_rxq_id = reg.if_rxq_id;
+ ctx->ifq = ifq;
+
+ return 0;
+err:
+ io_zc_rx_ifq_free(ifq);
+ return ret;
+}
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
new file mode 100644
index 000000000000..4363734f3d98
--- /dev/null
+++ b/io_uring/zc_rx.h
@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0
+#ifndef IOU_ZC_RX_H
+#define IOU_ZC_RX_H
+
+struct io_zc_rx_ifq {
+ struct io_ring_ctx *ctx;
+ struct net_device *dev;
+ struct io_rbuf_ring *ring;
+ struct io_uring_rbuf_rqe *rqes;
+ struct io_uring_rbuf_cqe *cqes;
+ u32 rq_entries, cq_entries;
+ void *pool;
+
+ /* hw rx descriptor ring id */
+ u32 if_rxq_id;
+};
+
+int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
+ struct io_uring_zc_rx_ifq_reg __user *arg);
+
+#endif
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 02/11] io_uring: add mmap support for shared ifq ringbuffers
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
2023-08-26 1:19 ` [PATCH 01/11] io_uring: add interface queue David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command David Wei
` (9 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds mmap support for ifq rbuf rings. There are two rings and
a struct io_rbuf_ring that contains the head and tail ptrs into each
ring.
Just like the io_uring SQ/CQ rings, userspace issues a single mmap call
using the io_uring fd w/ magic offset IORING_OFF_RBUF_RING. An opaque
ptr is returned to userspace, which is then expected to use the offsets
returned in the registration struct to get access to the head/tail and
rings.
Signed-off-by: David Wei <[email protected]>
---
include/uapi/linux/io_uring.h | 2 ++
io_uring/io_uring.c | 5 +++++
io_uring/zc_rx.c | 17 +++++++++++++++++
3 files changed, 24 insertions(+)
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 8f2a1061629b..28154abfe6f4 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -393,6 +393,8 @@ enum {
#define IORING_OFF_PBUF_RING 0x80000000ULL
#define IORING_OFF_PBUF_SHIFT 16
#define IORING_OFF_MMAP_MASK 0xf8000000ULL
+#define IORING_OFF_RBUF_RING 0x20000000ULL
+#define IORING_OFF_RBUF_SHIFT 16
/*
* Filled with the offset for mmap(2)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 7705d18dceff..0b6c5508b1ca 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -3368,6 +3368,11 @@ static void *io_uring_validate_mmap_request(struct file *file,
return ERR_PTR(-EINVAL);
break;
}
+ case IORING_OFF_RBUF_RING:
+ if (!ctx->ifq)
+ return ERR_PTR(-EINVAL);
+ ptr = ctx->ifq->ring;
+ break;
default:
return ERR_PTR(-EINVAL);
}
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 63bc6cd7d205..6c57c9b06e05 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -34,6 +34,7 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
{
struct io_uring_zc_rx_ifq_reg reg;
struct io_zc_rx_ifq *ifq;
+ size_t ring_sz, rqes_sz, cqes_sz;
int ret;
if (copy_from_user(®, arg, sizeof(reg)))
@@ -58,6 +59,22 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
ifq->if_rxq_id = reg.if_rxq_id;
ctx->ifq = ifq;
+ ring_sz = sizeof(struct io_rbuf_ring);
+ rqes_sz = sizeof(struct io_uring_rbuf_rqe) * ifq->rq_entries;
+ cqes_sz = sizeof(struct io_uring_rbuf_cqe) * ifq->cq_entries;
+ reg.mmap_sz = ring_sz + rqes_sz + cqes_sz;
+ reg.rq_off.rqes = ring_sz;
+ reg.cq_off.cqes = ring_sz + rqes_sz;
+ reg.rq_off.head = offsetof(struct io_rbuf_ring, rq.head);
+ reg.rq_off.tail = offsetof(struct io_rbuf_ring, rq.tail);
+ reg.cq_off.head = offsetof(struct io_rbuf_ring, cq.head);
+ reg.cq_off.tail = offsetof(struct io_rbuf_ring, cq.tail);
+
+ if (copy_to_user(arg, ®, sizeof(reg))) {
+ ret = -EFAULT;
+ goto err;
+ }
+
return 0;
err:
io_zc_rx_ifq_free(ifq);
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
2023-08-26 1:19 ` [PATCH 01/11] io_uring: add interface queue David Wei
2023-08-26 1:19 ` [PATCH 02/11] io_uring: add mmap support for shared ifq ringbuffers David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 2:21 ` David Ahern
2023-08-26 1:19 ` [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq David Wei
` (8 subsequent siblings)
11 siblings, 1 reply; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds a new XDP_SETUP_ZC_RX command that will be used in a
later patch to enable or disable ZC RX for a specific RX queue.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/linux/netdevice.h | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 08fbd4622ccf..a20a5c847916 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1000,6 +1000,7 @@ enum bpf_netdev_command {
BPF_OFFLOAD_MAP_ALLOC,
BPF_OFFLOAD_MAP_FREE,
XDP_SETUP_XSK_POOL,
+ XDP_SETUP_ZC_RX,
};
struct bpf_prog_offload_ops;
@@ -1038,6 +1039,11 @@ struct netdev_bpf {
struct xsk_buff_pool *pool;
u16 queue_id;
} xsk;
+ /* XDP_SETUP_ZC_RX */
+ struct {
+ struct io_zc_rx_ifq *ifq;
+ u16 queue_id;
+ } zc_rx;
};
};
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* Re: [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command
2023-08-26 1:19 ` [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command David Wei
@ 2023-08-26 2:21 ` David Ahern
2023-08-26 21:37 ` David Wei
0 siblings, 1 reply; 18+ messages in thread
From: David Ahern @ 2023-08-26 2:21 UTC (permalink / raw)
To: David Wei, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 8/25/23 6:19 PM, David Wei wrote:
> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> index 08fbd4622ccf..a20a5c847916 100644
> --- a/include/linux/netdevice.h
> +++ b/include/linux/netdevice.h
> @@ -1000,6 +1000,7 @@ enum bpf_netdev_command {
> BPF_OFFLOAD_MAP_ALLOC,
> BPF_OFFLOAD_MAP_FREE,
> XDP_SETUP_XSK_POOL,
> + XDP_SETUP_ZC_RX,
Why XDP in the name? Packets go from nic to driver to stack to io_uring,
no? That is not XDP.
> };
>
> struct bpf_prog_offload_ops;
> @@ -1038,6 +1039,11 @@ struct netdev_bpf {
> struct xsk_buff_pool *pool;
> u16 queue_id;
> } xsk;
> + /* XDP_SETUP_ZC_RX */
> + struct {
> + struct io_zc_rx_ifq *ifq;
> + u16 queue_id;
> + } zc_rx;
> };
> };
>
^ permalink raw reply [flat|nested] 18+ messages in thread* Re: [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command
2023-08-26 2:21 ` David Ahern
@ 2023-08-26 21:37 ` David Wei
0 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 21:37 UTC (permalink / raw)
To: David Ahern, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 25/08/2023 19:21, David Ahern wrote:
> On 8/25/23 6:19 PM, David Wei wrote:
>> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>> index 08fbd4622ccf..a20a5c847916 100644
>> --- a/include/linux/netdevice.h
>> +++ b/include/linux/netdevice.h
>> @@ -1000,6 +1000,7 @@ enum bpf_netdev_command {
>> BPF_OFFLOAD_MAP_ALLOC,
>> BPF_OFFLOAD_MAP_FREE,
>> XDP_SETUP_XSK_POOL,
>> + XDP_SETUP_ZC_RX,
>
> Why XDP in the name? Packets go from nic to driver to stack to io_uring,
> no? That is not XDP.
This new bpf_netdev_command XDP_SETUP_ZC_RX mirrors XDP_SETUP_XSK_POOL above,
both in terms of naming and purpose (the arg structs are almost identical).
However, I dislike this way of toggling ZC RX support anyway and am happy to
change it to a better method e.g. properly supports namespaces.
>
>
>> };
>>
>> struct bpf_prog_offload_ops;
>> @@ -1038,6 +1039,11 @@ struct netdev_bpf {
>> struct xsk_buff_pool *pool;
>> u16 queue_id;
>> } xsk;
>> + /* XDP_SETUP_ZC_RX */
>> + struct {
>> + struct io_zc_rx_ifq *ifq;
>> + u16 queue_id;
>> + } zc_rx;
>> };
>> };
>>
>
^ permalink raw reply [flat|nested] 18+ messages in thread
* [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (2 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 03/11] netdev: add XDP_SETUP_ZC_RX command David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 2:26 ` David Ahern
2023-08-26 1:19 ` [PATCH 05/11] io_uring: add ZC buf and pool David Wei
` (7 subsequent siblings)
11 siblings, 1 reply; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch sets up ZC for an RX queue in a net device when an ifq is
registered with io_uring. The RX queue is specified in the registration
struct. The XDP command added in the previous patch is used to enable or
disable ZC RX.
For now since there is only one ifq, its destruction is implicit during
io_uring cleanup.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
io_uring/io_uring.c | 1 +
io_uring/zc_rx.c | 62 +++++++++++++++++++++++++++++++++++++++++++--
io_uring/zc_rx.h | 1 +
3 files changed, 62 insertions(+), 2 deletions(-)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 0b6c5508b1ca..f2ec0a454307 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -3098,6 +3098,7 @@ static __cold void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
percpu_ref_kill(&ctx->refs);
xa_for_each(&ctx->personalities, index, creds)
io_unregister_personality(ctx, index);
+ io_unregister_zc_rx_ifq(ctx);
if (ctx->rings)
io_poll_remove_all(ctx, NULL, true);
mutex_unlock(&ctx->uring_lock);
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 6c57c9b06e05..8cc66731af5b 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -3,6 +3,7 @@
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/io_uring.h>
+#include <linux/netdevice.h>
#include <uapi/linux/io_uring.h>
@@ -10,6 +11,35 @@
#include "kbuf.h"
#include "zc_rx.h"
+typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
+
+static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
+ u16 queue_id)
+{
+ struct netdev_bpf cmd;
+ bpf_op_t ndo_bpf;
+
+ ndo_bpf = dev->netdev_ops->ndo_bpf;
+ if (!ndo_bpf)
+ return -EINVAL;
+
+ cmd.command = XDP_SETUP_ZC_RX;
+ cmd.zc_rx.ifq = ifq;
+ cmd.zc_rx.queue_id = queue_id;
+
+ return ndo_bpf(dev, &cmd);
+}
+
+static int io_open_zc_rxq(struct io_zc_rx_ifq *ifq)
+{
+ return __io_queue_mgmt(ifq->dev, ifq, ifq->if_rxq_id);
+}
+
+static int io_close_zc_rxq(struct io_zc_rx_ifq *ifq)
+{
+ return __io_queue_mgmt(ifq->dev, NULL, ifq->if_rxq_id);
+}
+
static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
{
struct io_zc_rx_ifq *ifq;
@@ -19,12 +49,17 @@ static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
return NULL;
ifq->ctx = ctx;
+ ifq->if_rxq_id = -1;
return ifq;
}
static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
{
+ if (ifq->if_rxq_id != -1)
+ io_close_zc_rxq(ifq);
+ if (ifq->dev)
+ dev_put(ifq->dev);
io_free_rbuf_ring(ifq);
kfree(ifq);
}
@@ -41,17 +76,22 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
return -EFAULT;
if (ctx->ifq)
return -EBUSY;
+ if (reg.if_rxq_id == -1)
+ return -EINVAL;
ifq = io_zc_rx_ifq_alloc(ctx);
if (!ifq)
return -ENOMEM;
- /* TODO: initialise network interface */
-
ret = io_allocate_rbuf_ring(ifq, ®);
if (ret)
goto err;
+ ret = -ENODEV;
+ ifq->dev = dev_get_by_index(&init_net, reg.if_idx);
+ if (!ifq->dev)
+ goto err;
+
/* TODO: map zc region and initialise zc pool */
ifq->rq_entries = reg.rq_entries;
@@ -59,6 +99,10 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
ifq->if_rxq_id = reg.if_rxq_id;
ctx->ifq = ifq;
+ ret = io_open_zc_rxq(ifq);
+ if (ret)
+ goto err;
+
ring_sz = sizeof(struct io_rbuf_ring);
rqes_sz = sizeof(struct io_uring_rbuf_rqe) * ifq->rq_entries;
cqes_sz = sizeof(struct io_uring_rbuf_cqe) * ifq->cq_entries;
@@ -80,3 +124,17 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
io_zc_rx_ifq_free(ifq);
return ret;
}
+
+int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx)
+{
+ struct io_zc_rx_ifq *ifq;
+
+ ifq = ctx->ifq;
+ if (!ifq)
+ return -EINVAL;
+
+ ctx->ifq = NULL;
+ io_zc_rx_ifq_free(ifq);
+
+ return 0;
+}
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
index 4363734f3d98..340ececa9f9c 100644
--- a/io_uring/zc_rx.h
+++ b/io_uring/zc_rx.h
@@ -17,5 +17,6 @@ struct io_zc_rx_ifq {
int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
struct io_uring_zc_rx_ifq_reg __user *arg);
+int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx);
#endif
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* Re: [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq
2023-08-26 1:19 ` [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq David Wei
@ 2023-08-26 2:26 ` David Ahern
2023-08-26 22:00 ` David Wei
0 siblings, 1 reply; 18+ messages in thread
From: David Ahern @ 2023-08-26 2:26 UTC (permalink / raw)
To: David Wei, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 8/25/23 6:19 PM, David Wei wrote:
> diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
> index 6c57c9b06e05..8cc66731af5b 100644
> --- a/io_uring/zc_rx.c
> +++ b/io_uring/zc_rx.c
> @@ -10,6 +11,35 @@
> #include "kbuf.h"
> #include "zc_rx.h"
>
> +typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
> +
> +static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
> + u16 queue_id)
> +{
> + struct netdev_bpf cmd;
> + bpf_op_t ndo_bpf;
> +
> + ndo_bpf = dev->netdev_ops->ndo_bpf;
This is not bpf related, so it seems wrong to be overloading this ndo.
> + if (!ndo_bpf)
> + return -EINVAL;
> +
> + cmd.command = XDP_SETUP_ZC_RX;
> + cmd.zc_rx.ifq = ifq;
> + cmd.zc_rx.queue_id = queue_id;
> +
> + return ndo_bpf(dev, &cmd);
> +}
> +
> +static int io_open_zc_rxq(struct io_zc_rx_ifq *ifq)
> +{
> + return __io_queue_mgmt(ifq->dev, ifq, ifq->if_rxq_id);
> +}
> +
> +static int io_close_zc_rxq(struct io_zc_rx_ifq *ifq)
> +{
> + return __io_queue_mgmt(ifq->dev, NULL, ifq->if_rxq_id);
> +}
> +
> static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
> {
> struct io_zc_rx_ifq *ifq;
> @@ -19,12 +49,17 @@ static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
> return NULL;
>
> ifq->ctx = ctx;
> + ifq->if_rxq_id = -1;
>
> return ifq;
> }
>
> static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
> {
> + if (ifq->if_rxq_id != -1)
> + io_close_zc_rxq(ifq);
> + if (ifq->dev)
> + dev_put(ifq->dev);
> io_free_rbuf_ring(ifq);
> kfree(ifq);
> }
> @@ -41,17 +76,22 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
> return -EFAULT;
> if (ctx->ifq)
> return -EBUSY;
> + if (reg.if_rxq_id == -1)
> + return -EINVAL;
>
> ifq = io_zc_rx_ifq_alloc(ctx);
> if (!ifq)
> return -ENOMEM;
>
> - /* TODO: initialise network interface */
> -
> ret = io_allocate_rbuf_ring(ifq, ®);
> if (ret)
> goto err;
>
> + ret = -ENODEV;
> + ifq->dev = dev_get_by_index(&init_net, reg.if_idx);
What's the plan for other namespaces? Ideally the device bind comes from
a socket and that gives you the namespace.
> + if (!ifq->dev)
> + goto err;
> +
> /* TODO: map zc region and initialise zc pool */
>
> ifq->rq_entries = reg.rq_entries;
^ permalink raw reply [flat|nested] 18+ messages in thread* Re: [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq
2023-08-26 2:26 ` David Ahern
@ 2023-08-26 22:00 ` David Wei
0 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 22:00 UTC (permalink / raw)
To: David Ahern, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 25/08/2023 19:26, David Ahern wrote:
> On 8/25/23 6:19 PM, David Wei wrote:
>> diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
>> index 6c57c9b06e05..8cc66731af5b 100644
>> --- a/io_uring/zc_rx.c
>> +++ b/io_uring/zc_rx.c
>> @@ -10,6 +11,35 @@
>> #include "kbuf.h"
>> #include "zc_rx.h"
>>
>> +typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
>> +
>> +static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
>> + u16 queue_id)
>> +{
>> + struct netdev_bpf cmd;
>> + bpf_op_t ndo_bpf;
>> +
>> + ndo_bpf = dev->netdev_ops->ndo_bpf;
>
> This is not bpf related, so it seems wrong to be overloading this ndo.
Agreed. I believe the original author (Jonathan) was inspired by
XDP_SETUP_XSK_POOL. I would also prefer a better way of setting up an RX queue
for ZC. Mina's proposal I think uses netlink for this.
Do you have any other suggestions? We want to keep resource registration
inline, so we need to be able to call it from within io_uring.
>
>
>> + if (!ndo_bpf)
>> + return -EINVAL;
>> +
>> + cmd.command = XDP_SETUP_ZC_RX;
>> + cmd.zc_rx.ifq = ifq;
>> + cmd.zc_rx.queue_id = queue_id;
>> +
>> + return ndo_bpf(dev, &cmd);
>> +}
>> +
>> +static int io_open_zc_rxq(struct io_zc_rx_ifq *ifq)
>> +{
>> + return __io_queue_mgmt(ifq->dev, ifq, ifq->if_rxq_id);
>> +}
>> +
>> +static int io_close_zc_rxq(struct io_zc_rx_ifq *ifq)
>> +{
>> + return __io_queue_mgmt(ifq->dev, NULL, ifq->if_rxq_id);
>> +}
>> +
>> static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
>> {
>> struct io_zc_rx_ifq *ifq;
>> @@ -19,12 +49,17 @@ static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
>> return NULL;
>>
>> ifq->ctx = ctx;
>> + ifq->if_rxq_id = -1;
>>
>> return ifq;
>> }
>>
>> static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
>> {
>> + if (ifq->if_rxq_id != -1)
>> + io_close_zc_rxq(ifq);
>> + if (ifq->dev)
>> + dev_put(ifq->dev);
>> io_free_rbuf_ring(ifq);
>> kfree(ifq);
>> }
>> @@ -41,17 +76,22 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
>> return -EFAULT;
>> if (ctx->ifq)
>> return -EBUSY;
>> + if (reg.if_rxq_id == -1)
>> + return -EINVAL;
>>
>> ifq = io_zc_rx_ifq_alloc(ctx);
>> if (!ifq)
>> return -ENOMEM;
>>
>> - /* TODO: initialise network interface */
>> -
>> ret = io_allocate_rbuf_ring(ifq, ®);
>> if (ret)
>> goto err;
>>
>> + ret = -ENODEV;
>> + ifq->dev = dev_get_by_index(&init_net, reg.if_idx);
>
> What's the plan for other namespaces? Ideally the device bind comes from
> a socket and that gives you the namespace.
Sorry, I did not consider namespaces yet. I'll look into how namespaces work
and then get back to you.
>
>> + if (!ifq->dev)
>> + goto err;
>> +
>> /* TODO: map zc region and initialise zc pool */
>>
>> ifq->rq_entries = reg.rq_entries;
>
>
^ permalink raw reply [flat|nested] 18+ messages in thread
* [PATCH 05/11] io_uring: add ZC buf and pool
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (3 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 04/11] io_uring: setup ZC for an RX queue when registering an ifq David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 06/11] io_uring: add ZC pool API David Wei
` (6 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds two objects:
* Zero copy buffer representation, holding a page, its mapped dma_addr,
and a refcount for lifetime management.
* Zero copy pool, spiritually similar to page pool, that holds ZC bufs
and hands them out to net devices.
The ZC pool is tiered with currently two tiers: a fast lockless cache
that should only be accessed from the NAPI context of a single RX queue,
and a freelist. When a ZC pool region is first mapped, it is added to
the freelist.
During normal operation, bufs are moved from the freelist into the cache
in POOL_CACHE_SIZE blocks before being given out.
Pool regions are registered w/ io_uring using the registered buffer API,
with a 1:1 mapping between region and nr_iovec in
io_uring_register_buffers. This does the heavy lifting of pinning and
chunking into bvecs into a struct io_mapped_ubuf for us.
For now as there is only one pool region per ifq, there is no separate
API for adding/removing regions yet and it is mapped implicitly during
ifq registration.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/linux/io_uring.h | 6 ++
io_uring/zc_rx.c | 173 ++++++++++++++++++++++++++++++++++++++-
io_uring/zc_rx.h | 1 +
3 files changed, 179 insertions(+), 1 deletion(-)
diff --git a/include/linux/io_uring.h b/include/linux/io_uring.h
index 7fe31b2cd02f..cf1993befa6a 100644
--- a/include/linux/io_uring.h
+++ b/include/linux/io_uring.h
@@ -41,6 +41,12 @@ static inline const void *io_uring_sqe_cmd(const struct io_uring_sqe *sqe)
return sqe->cmd;
}
+struct io_zc_rx_buf {
+ dma_addr_t dma;
+ struct page *page;
+ atomic_t refcount;
+};
+
#if defined(CONFIG_IO_URING)
int io_uring_cmd_import_fixed(u64 ubuf, unsigned long len, int rw,
struct iov_iter *iter, void *ioucmd);
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 8cc66731af5b..317127d0d4e7 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -4,13 +4,43 @@
#include <linux/mm.h>
#include <linux/io_uring.h>
#include <linux/netdevice.h>
+#include <linux/nospec.h>
#include <uapi/linux/io_uring.h>
#include "io_uring.h"
#include "kbuf.h"
+#include "rsrc.h"
#include "zc_rx.h"
+#define POOL_CACHE_SIZE 128
+
+struct io_zc_rx_pool {
+ struct io_zc_rx_ifq *ifq;
+ struct io_zc_rx_buf *bufs;
+ u16 pool_id;
+ u32 nr_pages;
+
+ /* fast cache */
+ u32 cache_count;
+ u32 cache[POOL_CACHE_SIZE];
+
+ /* freelist */
+ spinlock_t freelist_lock;
+ u32 free_count;
+ u32 freelist[];
+};
+
+static struct device *netdev2dev(struct net_device *dev)
+{
+ return dev->dev.parent;
+}
+
+static u64 mk_page_info(u16 pool_id, u32 pgid)
+{
+ return (u64)0xface << 48 | (u64)pool_id << 32 | (u64)pgid;
+}
+
typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
@@ -40,6 +70,143 @@ static int io_close_zc_rxq(struct io_zc_rx_ifq *ifq)
return __io_queue_mgmt(ifq->dev, NULL, ifq->if_rxq_id);
}
+static int io_zc_rx_map_buf(struct device *dev, struct page *page, u16 pool_id,
+ u32 pgid, struct io_zc_rx_buf *buf)
+{
+ dma_addr_t addr;
+
+ SetPagePrivate(page);
+ set_page_private(page, mk_page_info(pool_id, pgid));
+
+ addr = dma_map_page_attrs(dev, page, 0, PAGE_SIZE,
+ DMA_BIDIRECTIONAL,
+ DMA_ATTR_SKIP_CPU_SYNC);
+ if (dma_mapping_error(dev, addr)) {
+ set_page_private(page, 0);
+ ClearPagePrivate(page);
+ return -ENOMEM;
+ }
+
+ buf->dma = addr;
+ buf->page = page;
+ atomic_set(&buf->refcount, 0);
+ get_page(page);
+
+ return 0;
+}
+
+static void io_zc_rx_unmap_buf(struct device *dev, struct io_zc_rx_buf *buf)
+{
+ struct page *page;
+
+ page = buf->page;
+ set_page_private(page, 0);
+ ClearPagePrivate(page);
+ dma_unmap_page_attrs(dev, buf->dma, PAGE_SIZE,
+ DMA_BIDIRECTIONAL,
+ DMA_ATTR_SKIP_CPU_SYNC);
+ put_page(page);
+}
+
+static int io_zc_rx_map_pool(struct io_zc_rx_pool *pool,
+ struct io_mapped_ubuf *imu,
+ struct device *dev)
+{
+ struct io_zc_rx_buf *buf;
+ struct page *page;
+ int i, ret;
+
+ for (i = 0; i < imu->nr_bvecs; i++) {
+ page = imu->bvec[i].bv_page;
+ if (PagePrivate(page)) {
+ ret = -EEXIST;
+ goto err;
+ }
+
+ buf = &pool->bufs[i];
+ ret = io_zc_rx_map_buf(dev, page, pool->pool_id, i, buf);
+ if (ret)
+ goto err;
+
+ pool->freelist[i] = i;
+ }
+
+ return 0;
+err:
+ while (i--) {
+ buf = &pool->bufs[i];
+ io_zc_rx_unmap_buf(dev, buf);
+ }
+
+ return ret;
+}
+
+int io_zc_rx_create_pool(struct io_ring_ctx *ctx,
+ struct io_zc_rx_ifq *ifq,
+ u16 id)
+{
+ struct device *dev = netdev2dev(ifq->dev);
+ struct io_mapped_ubuf *imu;
+ struct io_zc_rx_pool *pool;
+ int nr_pages;
+ int ret;
+
+ if (ifq->pool)
+ return -EFAULT;
+
+ if (unlikely(id >= ctx->nr_user_bufs))
+ return -EFAULT;
+ id = array_index_nospec(id, ctx->nr_user_bufs);
+ imu = ctx->user_bufs[id];
+ if (imu->ubuf & ~PAGE_MASK || imu->ubuf_end & ~PAGE_MASK)
+ return -EFAULT;
+
+ ret = -ENOMEM;
+ nr_pages = imu->nr_bvecs;
+ pool = kvmalloc(struct_size(pool, freelist, nr_pages), GFP_KERNEL);
+ if (!pool)
+ goto err;
+
+ pool->bufs = kvmalloc_array(nr_pages, sizeof(*pool->bufs), GFP_KERNEL);
+ if (!pool->bufs)
+ goto err_buf;
+
+ ret = io_zc_rx_map_pool(pool, imu, dev);
+ if (ret)
+ goto err_map;
+
+ pool->ifq = ifq;
+ pool->pool_id = id;
+ pool->nr_pages = nr_pages;
+ pool->cache_count = 0;
+ spin_lock_init(&pool->freelist_lock);
+ pool->free_count = nr_pages;
+ ifq->pool = pool;
+
+ return 0;
+
+err_map:
+ kvfree(pool->bufs);
+err_buf:
+ kvfree(pool);
+err:
+ return ret;
+}
+
+static void io_zc_rx_destroy_pool(struct io_zc_rx_pool *pool)
+{
+ struct device *dev = netdev2dev(pool->ifq->dev);
+ struct io_zc_rx_buf *buf;
+
+ for (int i = 0; i < pool->nr_pages; i++) {
+ buf = &pool->bufs[i];
+
+ io_zc_rx_unmap_buf(dev, buf);
+ }
+ kvfree(pool->bufs);
+ kvfree(pool);
+}
+
static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
{
struct io_zc_rx_ifq *ifq;
@@ -58,6 +225,8 @@ static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
{
if (ifq->if_rxq_id != -1)
io_close_zc_rxq(ifq);
+ if (ifq->pool)
+ io_zc_rx_destroy_pool(ifq->pool);
if (ifq->dev)
dev_put(ifq->dev);
io_free_rbuf_ring(ifq);
@@ -92,7 +261,9 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
if (!ifq->dev)
goto err;
- /* TODO: map zc region and initialise zc pool */
+ ret = io_zc_rx_create_pool(ctx, ifq, reg.region_id);
+ if (ret)
+ goto err;
ifq->rq_entries = reg.rq_entries;
ifq->cq_entries = reg.cq_entries;
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
index 340ececa9f9c..3cd0e730115d 100644
--- a/io_uring/zc_rx.h
+++ b/io_uring/zc_rx.h
@@ -18,5 +18,6 @@ struct io_zc_rx_ifq {
int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
struct io_uring_zc_rx_ifq_reg __user *arg);
int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx);
+int io_zc_rx_pool_create(struct io_zc_rx_ifq *ifq, u16 id);
#endif
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 06/11] io_uring: add ZC pool API
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (4 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 05/11] io_uring: add ZC buf and pool David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 07/11] skbuff: add SKBFL_FIXED_FRAG and skb_fixed() David Wei
` (5 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds an API to get/put bufs from a ZC pool added in the
previous patch.
Recall that there is an rbuf refill ring in an ifq that is shared w/
userspace, which puts bufs it is done with back into it. A new tier is
added to the ZC pool that drains entries from the refill ring to put
into the cache. So when the cache is empty, it is refilled from the
refill ring first, then the freelist.
ZC bufs are refcounted, with both a kref and a uref. Userspace is given
an off + len into the entire ZC pool region, not individual pages from
ZC bufs. A net device may pack multiple packets into the same page it
gets from a ZC buf, so it is possible for the same ZC buf to be handed
out to userspace multiple times.
This means it is possible to drain the entire refill ring, and have no
usable free bufs. Suggestions for dealing w/ this are very welcome!
Only up to POOL_REFILL_COUNT entries are refilled from the refill ring.
Given the above, we may want to limit the amount of work being done
since refilling happens inside the NAPI softirq context.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/linux/io_uring.h | 18 ++++++++
io_uring/zc_rx.c | 98 ++++++++++++++++++++++++++++++++++++++++
io_uring/zc_rx.h | 13 ++++++
3 files changed, 129 insertions(+)
diff --git a/include/linux/io_uring.h b/include/linux/io_uring.h
index cf1993befa6a..61eae25a8f1d 100644
--- a/include/linux/io_uring.h
+++ b/include/linux/io_uring.h
@@ -60,6 +60,17 @@ void __io_uring_free(struct task_struct *tsk);
void io_uring_unreg_ringfd(void);
const char *io_uring_get_opcode(u8 opcode);
+struct io_zc_rx_ifq;
+struct io_zc_rx_buf *io_zc_rx_get_buf(struct io_zc_rx_ifq *ifq);
+void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf);
+static inline dma_addr_t io_zc_rx_buf_dma(struct io_zc_rx_buf *buf)
+{
+ return buf->dma;
+}
+static inline struct page *io_zc_rx_buf_page(struct io_zc_rx_buf *buf)
+{
+ return buf->page;
+}
static inline void io_uring_files_cancel(void)
{
if (current->io_uring) {
@@ -108,6 +119,13 @@ static inline const char *io_uring_get_opcode(u8 opcode)
{
return "";
}
+static inline struct io_zc_rx_buf *io_zc_rx_get_buf(struct io_zc_rx_ifq *ifq)
+{
+ return NULL;
+}
+void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf)
+{
+}
#endif
#endif
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 317127d0d4e7..14bc063f1c6c 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -14,6 +14,9 @@
#include "zc_rx.h"
#define POOL_CACHE_SIZE 128
+#define POOL_REFILL_COUNT 64
+#define IO_ZC_RX_UREF 0x10000
+#define IO_ZC_RX_KREF_MASK (IO_ZC_RX_UREF - 1)
struct io_zc_rx_pool {
struct io_zc_rx_ifq *ifq;
@@ -267,6 +270,8 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
ifq->rq_entries = reg.rq_entries;
ifq->cq_entries = reg.cq_entries;
+ ifq->cached_rq_head = 0;
+ ifq->cached_cq_tail = 0;
ifq->if_rxq_id = reg.if_rxq_id;
ctx->ifq = ifq;
@@ -309,3 +314,96 @@ int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx)
return 0;
}
+
+static bool io_zc_rx_put_buf_uref(struct io_zc_rx_buf *buf)
+{
+ if (atomic_read(&buf->refcount) < IO_ZC_RX_UREF)
+ return false;
+
+ return atomic_sub_and_test(IO_ZC_RX_UREF, &buf->refcount);
+}
+
+static void io_zc_rx_refill_cache(struct io_zc_rx_ifq *ifq, int count)
+{
+ unsigned int entries = io_zc_rx_rqring_entries(ifq);
+ unsigned int mask = ifq->rq_entries - 1;
+ struct io_zc_rx_pool *pool = ifq->pool;
+ struct io_uring_rbuf_rqe *rqe;
+ struct io_zc_rx_buf *buf;
+ int i, filled;
+
+ if (!entries)
+ return;
+
+ for (i = 0, filled = 0; i < entries && filled < count; i++) {
+ unsigned int rq_idx = ifq->cached_rq_head++ & mask;
+ u32 pgid;
+
+ rqe = &ifq->rqes[rq_idx];
+ pgid = rqe->off / PAGE_SIZE;
+ buf = &pool->bufs[pgid];
+ if (!io_zc_rx_put_buf_uref(buf))
+ continue;
+ pool->cache[filled++] = pgid;
+ }
+
+ smp_store_release(&ifq->ring->rq.head, ifq->cached_rq_head);
+ pool->cache_count += filled;
+}
+
+struct io_zc_rx_buf *io_zc_rx_get_buf(struct io_zc_rx_ifq *ifq)
+{
+ struct io_zc_rx_pool *pool;
+ struct io_zc_rx_buf *buf;
+ int count;
+ u16 pgid;
+
+ pool = ifq->pool;
+ if (pool->cache_count)
+ goto out;
+
+ io_zc_rx_refill_cache(ifq, POOL_REFILL_COUNT);
+ if (pool->cache_count)
+ goto out;
+
+ spin_lock(&pool->freelist_lock);
+
+ count = min_t(u32, pool->free_count, POOL_CACHE_SIZE);
+ pool->free_count -= count;
+ pool->cache_count += count;
+ memcpy(pool->cache, &pool->freelist[pool->free_count],
+ count * sizeof(u32));
+
+ spin_unlock(&pool->freelist_lock);
+
+ if (pool->cache_count)
+ goto out;
+
+ return NULL;
+out:
+ pgid = pool->cache[--pool->cache_count];
+ buf = &pool->bufs[pgid];
+ atomic_set(&buf->refcount, 1);
+
+ return buf;
+}
+EXPORT_SYMBOL(io_zc_rx_get_buf);
+
+static void io_zc_rx_recycle_buf(struct io_zc_rx_pool *pool,
+ struct io_zc_rx_buf *buf)
+{
+ spin_lock(&pool->freelist_lock);
+ pool->freelist[pool->free_count++] = buf - pool->bufs;
+ spin_unlock(&pool->freelist_lock);
+}
+
+void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf)
+{
+ struct io_zc_rx_pool *pool = ifq->pool;
+
+ if (!atomic_dec_and_test(&buf->refcount))
+ return;
+
+ io_zc_rx_recycle_buf(pool, buf);
+}
+EXPORT_SYMBOL(io_zc_rx_put_buf);
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
index 3cd0e730115d..b063a3c81ccb 100644
--- a/io_uring/zc_rx.h
+++ b/io_uring/zc_rx.h
@@ -2,6 +2,8 @@
#ifndef IOU_ZC_RX_H
#define IOU_ZC_RX_H
+#include <linux/io_uring_types.h>
+
struct io_zc_rx_ifq {
struct io_ring_ctx *ctx;
struct net_device *dev;
@@ -9,12 +11,23 @@ struct io_zc_rx_ifq {
struct io_uring_rbuf_rqe *rqes;
struct io_uring_rbuf_cqe *cqes;
u32 rq_entries, cq_entries;
+ u32 cached_rq_head;
+ u32 cached_cq_tail;
void *pool;
/* hw rx descriptor ring id */
u32 if_rxq_id;
};
+static inline u32 io_zc_rx_rqring_entries(struct io_zc_rx_ifq *ifq)
+{
+ struct io_rbuf_ring *ring = ifq->ring;
+ u32 entries;
+
+ entries = smp_load_acquire(&ring->rq.tail) - ifq->cached_rq_head;
+ return min(entries, ifq->rq_entries);
+}
+
int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
struct io_uring_zc_rx_ifq_reg __user *arg);
int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx);
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 07/11] skbuff: add SKBFL_FIXED_FRAG and skb_fixed()
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (5 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 06/11] io_uring: add ZC pool API David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 08/11] io_uring: allocate a uarg for freeing zero copy skbs David Wei
` (4 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
When an skb that is marked as zero copy goes up the network stack during
RX, skb_orphan_frags_rx is called which then calls skb_copy_ubufs and
defeat the purpose of ZC.
This is because currently zero copy is only for TX and this behaviour is
designed to prevent TX zero copy data being redirected up the network
stack rather than new zero copy RX data coming from the driver.
This patch adds a new flag SKBFL_FIXED_FRAG and checks for this in
skb_orphan_frags, not calling skb_copy_ubufs if it is set.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/linux/skbuff.h | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 8cff3d817131..d7ef998df4a5 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -518,6 +518,9 @@ enum {
* use frags only up until ubuf_info is released
*/
SKBFL_MANAGED_FRAG_REFS = BIT(4),
+
+ /* don't move or copy the fragment */
+ SKBFL_FIXED_FRAG = BIT(5),
};
#define SKBFL_ZEROCOPY_FRAG (SKBFL_ZEROCOPY_ENABLE | SKBFL_SHARED_FRAG)
@@ -1674,6 +1677,11 @@ static inline bool skb_zcopy_managed(const struct sk_buff *skb)
return skb_shinfo(skb)->flags & SKBFL_MANAGED_FRAG_REFS;
}
+static inline bool skb_fixed(const struct sk_buff *skb)
+{
+ return skb_shinfo(skb)->flags & SKBFL_FIXED_FRAG;
+}
+
static inline bool skb_pure_zcopy_same(const struct sk_buff *skb1,
const struct sk_buff *skb2)
{
@@ -3135,7 +3143,7 @@ static inline int skb_orphan_frags(struct sk_buff *skb, gfp_t gfp_mask)
/* Frags must be orphaned, even if refcounted, if skb might loop to rx path */
static inline int skb_orphan_frags_rx(struct sk_buff *skb, gfp_t gfp_mask)
{
- if (likely(!skb_zcopy(skb)))
+ if (likely(!skb_zcopy(skb) || skb_fixed(skb)))
return 0;
return skb_copy_ubufs(skb, gfp_mask);
}
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 08/11] io_uring: allocate a uarg for freeing zero copy skbs
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (6 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 07/11] skbuff: add SKBFL_FIXED_FRAG and skb_fixed() David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 09/11] io_uring: delay ZC pool destruction David Wei
` (3 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
As ZC skbs are marked as zero copy, they will bypass the default skb
frag destructor. This patch adds a static uarg that is attached to ZC
bufs and a callback that returns them to the freelist of a ZC pool.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/linux/io_uring.h | 7 +++++++
include/linux/netdevice.h | 1 +
io_uring/zc_rx.c | 44 +++++++++++++++++++++++++++++++++++++++
io_uring/zc_rx.h | 2 ++
4 files changed, 54 insertions(+)
diff --git a/include/linux/io_uring.h b/include/linux/io_uring.h
index 61eae25a8f1d..e2a4f92df814 100644
--- a/include/linux/io_uring.h
+++ b/include/linux/io_uring.h
@@ -62,6 +62,8 @@ const char *io_uring_get_opcode(u8 opcode);
struct io_zc_rx_ifq;
struct io_zc_rx_buf *io_zc_rx_get_buf(struct io_zc_rx_ifq *ifq);
+struct io_zc_rx_buf *io_zc_rx_buf_from_page(struct io_zc_rx_ifq *ifq,
+ struct page *page);
void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf);
static inline dma_addr_t io_zc_rx_buf_dma(struct io_zc_rx_buf *buf)
{
@@ -123,6 +125,11 @@ static inline struct io_zc_rx_buf *io_zc_rx_get_buf(struct io_zc_rx_ifq *ifq)
{
return NULL;
}
+static inline struct io_zc_rx_buf *io_zc_rx_buf_from_page(struct io_zc_rx_ifq *ifq,
+ struct page *page)
+{
+ return NULL;
+}
void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf)
{
}
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a20a5c847916..bf133cbee721 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1043,6 +1043,7 @@ struct netdev_bpf {
struct {
struct io_zc_rx_ifq *ifq;
u16 queue_id;
+ struct ubuf_info *uarg;
} zc_rx;
};
};
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 14bc063f1c6c..b8dd699e2777 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -44,6 +44,11 @@ static u64 mk_page_info(u16 pool_id, u32 pgid)
return (u64)0xface << 48 | (u64)pool_id << 32 | (u64)pgid;
}
+static bool is_zc_rx_page(struct page *page)
+{
+ return PagePrivate(page) && ((page_private(page) >> 48) == 0xface);
+}
+
typedef int (*bpf_op_t)(struct net_device *dev, struct netdev_bpf *bpf);
static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
@@ -59,6 +64,7 @@ static int __io_queue_mgmt(struct net_device *dev, struct io_zc_rx_ifq *ifq,
cmd.command = XDP_SETUP_ZC_RX;
cmd.zc_rx.ifq = ifq;
cmd.zc_rx.queue_id = queue_id;
+ cmd.zc_rx.uarg = &ifq->uarg;
return ndo_bpf(dev, &cmd);
}
@@ -73,6 +79,26 @@ static int io_close_zc_rxq(struct io_zc_rx_ifq *ifq)
return __io_queue_mgmt(ifq->dev, NULL, ifq->if_rxq_id);
}
+static void io_zc_rx_skb_free(struct sk_buff *skb, struct ubuf_info *uarg,
+ bool success)
+{
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
+ struct io_zc_rx_ifq *ifq;
+ struct io_zc_rx_buf *buf;
+ struct page *page;
+ int i;
+
+ ifq = container_of(uarg, struct io_zc_rx_ifq, uarg);
+ for (i = 0; i < shinfo->nr_frags; i++) {
+ page = skb_frag_page(&shinfo->frags[i]);
+ buf = io_zc_rx_buf_from_page(ifq, page);
+ if (likely(buf))
+ io_zc_rx_put_buf(ifq, buf);
+ else
+ __skb_frag_unref(&shinfo->frags[i], skb->pp_recycle);
+ }
+}
+
static int io_zc_rx_map_buf(struct device *dev, struct page *page, u16 pool_id,
u32 pgid, struct io_zc_rx_buf *buf)
{
@@ -268,6 +294,8 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
if (ret)
goto err;
+ ifq->uarg.callback = io_zc_rx_skb_free;
+ ifq->uarg.flags = SKBFL_ALL_ZEROCOPY | SKBFL_FIXED_FRAG;
ifq->rq_entries = reg.rq_entries;
ifq->cq_entries = reg.cq_entries;
ifq->cached_rq_head = 0;
@@ -407,3 +435,19 @@ void io_zc_rx_put_buf(struct io_zc_rx_ifq *ifq, struct io_zc_rx_buf *buf)
io_zc_rx_recycle_buf(pool, buf);
}
EXPORT_SYMBOL(io_zc_rx_put_buf);
+
+struct io_zc_rx_buf *io_zc_rx_buf_from_page(struct io_zc_rx_ifq *ifq,
+ struct page *page)
+{
+ struct io_zc_rx_pool *pool;
+ int pgid;
+
+ if (!is_zc_rx_page(page))
+ return NULL;
+
+ pool = ifq->pool;
+ pgid = page_private(page) & 0xffffffff;
+
+ return &pool->bufs[pgid];
+}
+EXPORT_SYMBOL(io_zc_rx_buf_from_page);
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
index b063a3c81ccb..ee7e36295f91 100644
--- a/io_uring/zc_rx.h
+++ b/io_uring/zc_rx.h
@@ -3,6 +3,7 @@
#define IOU_ZC_RX_H
#include <linux/io_uring_types.h>
+#include <linux/skbuff.h>
struct io_zc_rx_ifq {
struct io_ring_ctx *ctx;
@@ -10,6 +11,7 @@ struct io_zc_rx_ifq {
struct io_rbuf_ring *ring;
struct io_uring_rbuf_rqe *rqes;
struct io_uring_rbuf_cqe *cqes;
+ struct ubuf_info uarg;
u32 rq_entries, cq_entries;
u32 cached_rq_head;
u32 cached_cq_tail;
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 09/11] io_uring: delay ZC pool destruction
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (7 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 08/11] io_uring: allocate a uarg for freeing zero copy skbs David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 10/11] netdev/bnxt: add data pool and use it in BNXT driver David Wei
` (2 subsequent siblings)
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
At a point in time, a ZC buf may be in:
* RX queue
* Socket
* One of the ifq ringbufs
* Userspace
The ZC pool region and the pool itself cannot be destroyed until all
bufs have been returned.
This patch changes the ZC pool destruction to be delayed work, waiting
for up to 10 seconds for bufs to be returned before unconditionally
destroying the pool.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
io_uring/zc_rx.c | 50 ++++++++++++++++++++++++++++++++++++++++++------
1 file changed, 44 insertions(+), 6 deletions(-)
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index b8dd699e2777..70e39f851e47 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -28,6 +28,10 @@ struct io_zc_rx_pool {
u32 cache_count;
u32 cache[POOL_CACHE_SIZE];
+ /* delayed destruction */
+ unsigned long delay_end;
+ struct delayed_work destroy_work;
+
/* freelist */
spinlock_t freelist_lock;
u32 free_count;
@@ -222,20 +226,56 @@ int io_zc_rx_create_pool(struct io_ring_ctx *ctx,
return ret;
}
-static void io_zc_rx_destroy_pool(struct io_zc_rx_pool *pool)
+static void io_zc_rx_destroy_ifq(struct io_zc_rx_ifq *ifq)
+{
+ if (ifq->dev)
+ dev_put(ifq->dev);
+ io_free_rbuf_ring(ifq);
+ kfree(ifq);
+}
+
+static void io_zc_rx_destroy_pool_work(struct work_struct *work)
{
+ struct io_zc_rx_pool *pool = container_of(
+ to_delayed_work(work), struct io_zc_rx_pool, destroy_work);
struct device *dev = netdev2dev(pool->ifq->dev);
struct io_zc_rx_buf *buf;
+ int i, refc, count;
- for (int i = 0; i < pool->nr_pages; i++) {
+ for (i = 0; i < pool->nr_pages; i++) {
buf = &pool->bufs[i];
+ refc = atomic_read(&buf->refcount) & IO_ZC_RX_KREF_MASK;
+ if (refc) {
+ if (time_before(jiffies, pool->delay_end)) {
+ schedule_delayed_work(&pool->destroy_work, HZ);
+ return;
+ }
+ count++;
+ }
+ }
+
+ if (count)
+ pr_debug("freeing pool with %d/%d outstanding pages\n",
+ count, pool->nr_pages);
+
+ for (i = 0; i < pool->nr_pages; i++) {
+ buf = &pool->bufs[i];
io_zc_rx_unmap_buf(dev, buf);
}
+
+ io_zc_rx_destroy_ifq(pool->ifq);
kvfree(pool->bufs);
kvfree(pool);
}
+static void io_zc_rx_destroy_pool(struct io_zc_rx_pool *pool)
+{
+ pool->delay_end = jiffies + HZ * 10;
+ INIT_DELAYED_WORK(&pool->destroy_work, io_zc_rx_destroy_pool_work);
+ schedule_delayed_work(&pool->destroy_work, 0);
+}
+
static struct io_zc_rx_ifq *io_zc_rx_ifq_alloc(struct io_ring_ctx *ctx)
{
struct io_zc_rx_ifq *ifq;
@@ -256,10 +296,8 @@ static void io_zc_rx_ifq_free(struct io_zc_rx_ifq *ifq)
io_close_zc_rxq(ifq);
if (ifq->pool)
io_zc_rx_destroy_pool(ifq->pool);
- if (ifq->dev)
- dev_put(ifq->dev);
- io_free_rbuf_ring(ifq);
- kfree(ifq);
+ else
+ io_zc_rx_destroy_ifq(ifq);
}
int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 10/11] netdev/bnxt: add data pool and use it in BNXT driver
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (8 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 09/11] io_uring: delay ZC pool destruction David Wei
@ 2023-08-26 1:19 ` David Wei
2023-08-26 1:19 ` [PATCH 11/11] io_uring: add io_recvzc request David Wei
2023-10-22 19:06 ` [RFC RESEND 00/11] Zero copy network RX using io_uring Gal Pressman
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds a thin wrapper called data pool that wraps the existing
page pool and the newly added ZC pool.
There is one struct netdev_rx_queue per logical RX queue. This patch
adds ZC ifq, page pool, and uarg (set during skb construction) to a
netdev_rx_queue. The data pool wrapper uses the ZC pool if an ifq is
present, otherwise using the page pool.
The BNXT driver is modified to use data pool in order to support ZC RX.
A setup function bnxt_zc_rx is added that is called on XDP_SETUP_ZC_RX
XDP command which sets fields in netdev_rx_queue. Calls to get/put bufs
from the page pool are related w/ the data pool.
Signed-off-by: David Wei <[email protected]>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 59 ++++++++----
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 4 +
drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 3 +
include/linux/netdevice.h | 4 +
include/net/data_pool.h | 96 +++++++++++++++++++
5 files changed, 149 insertions(+), 17 deletions(-)
create mode 100644 include/net/data_pool.h
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 48f584f78561..5c1dabaf07f9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -57,6 +57,7 @@
#include <net/page_pool.h>
#include <linux/align.h>
#include <net/netdev_queues.h>
+#include <net/data_pool.h>
#include "bnxt_hsi.h"
#include "bnxt.h"
@@ -724,7 +725,7 @@ static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
__netif_txq_completed_wake(txq, nr_pkts, tx_bytes,
bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
- READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING);
+ READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING);
}
static struct page *__bnxt_alloc_rx_64k_page(struct bnxt *bp, dma_addr_t *mapping,
@@ -738,13 +739,7 @@ static struct page *__bnxt_alloc_rx_64k_page(struct bnxt *bp, dma_addr_t *mappin
if (!page)
return NULL;
- *mapping = dma_map_page_attrs(&bp->pdev->dev, page, offset,
- BNXT_RX_PAGE_SIZE, DMA_FROM_DEVICE,
- DMA_ATTR_WEAK_ORDERING);
- if (dma_mapping_error(&bp->pdev->dev, *mapping)) {
- page_pool_recycle_direct(rxr->page_pool, page);
- return NULL;
- }
+ *mapping = page_pool_get_dma_addr(page);
if (page_offset)
*page_offset = offset;
@@ -757,19 +752,14 @@ static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping,
struct bnxt_rx_ring_info *rxr,
gfp_t gfp)
{
- struct device *dev = &bp->pdev->dev;
struct page *page;
- page = page_pool_dev_alloc_pages(rxr->page_pool);
+ page = data_pool_alloc_page(rxr->rx_queue);
if (!page)
return NULL;
- *mapping = dma_map_page_attrs(dev, page, 0, PAGE_SIZE, bp->rx_dir,
- DMA_ATTR_WEAK_ORDERING);
- if (dma_mapping_error(dev, *mapping)) {
- page_pool_recycle_direct(rxr->page_pool, page);
- return NULL;
- }
+ *mapping = data_pool_get_dma_addr(rxr->rx_queue, page);
+
return page;
}
@@ -1787,6 +1777,8 @@ static void bnxt_deliver_skb(struct bnxt *bp, struct bnxt_napi *bnapi,
return;
}
skb_record_rx_queue(skb, bnapi->index);
+ if (bnapi->rx_ring->rx_queue->zc_ifq)
+ skb_zcopy_init(skb, bnapi->rx_ring->rx_queue->zc_uarg);
skb_mark_for_recycle(skb);
napi_gro_receive(&bnapi->napi, skb);
}
@@ -3016,7 +3008,7 @@ static void bnxt_free_one_rx_ring_skbs(struct bnxt *bp, int ring_nr)
rx_agg_buf->page = NULL;
__clear_bit(i, rxr->rx_agg_bmap);
- page_pool_recycle_direct(rxr->page_pool, page);
+ data_pool_put_page(rxr->rx_queue, page);
}
skip_rx_agg_free:
@@ -3225,6 +3217,8 @@ static void bnxt_free_rx_rings(struct bnxt *bp)
page_pool_destroy(rxr->page_pool);
rxr->page_pool = NULL;
+ rxr->rx_queue->page_pool = NULL;
+ rxr->rx_queue->zc_ifq = NULL;
kfree(rxr->rx_agg_bmap);
rxr->rx_agg_bmap = NULL;
@@ -3251,12 +3245,16 @@ static int bnxt_alloc_rx_page_pool(struct bnxt *bp,
pp.dma_dir = DMA_BIDIRECTIONAL;
if (PAGE_SIZE > BNXT_RX_PAGE_SIZE)
pp.flags |= PP_FLAG_PAGE_FRAG;
+ pp.flags |= PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+ pp.max_len = PAGE_SIZE;
rxr->page_pool = page_pool_create(&pp);
+ data_pool_set_page_pool(bp->dev, rxr->bnapi->index, rxr->page_pool);
if (IS_ERR(rxr->page_pool)) {
int err = PTR_ERR(rxr->page_pool);
rxr->page_pool = NULL;
+ data_pool_set_page_pool(bp->dev, rxr->bnapi->index, NULL);
return err;
}
return 0;
@@ -4620,6 +4618,8 @@ static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
rxr->rx_agg_ring_struct.ring_mem.flags =
BNXT_RMEM_RING_PTE_FLAG;
}
+
+ rxr->rx_queue = data_pool_get_rx_queue(bp->dev, bp->bnapi[i]->index);
rxr->bnapi = bp->bnapi[i];
bp->bnapi[i]->rx_ring = &bp->rx_ring[i];
}
@@ -13904,6 +13904,31 @@ void bnxt_print_device_info(struct bnxt *bp)
pcie_print_link_status(bp->pdev);
}
+int bnxt_zc_rx(struct bnxt *bp, struct netdev_bpf *xdp)
+{
+ if (xdp->zc_rx.queue_id >= bp->rx_nr_rings)
+ return -EINVAL;
+
+ bnxt_rtnl_lock_sp(bp);
+ if (netif_running(bp->dev)) {
+ struct netdev_rx_queue *rxq;
+ int rc;
+
+ bnxt_ulp_stop(bp);
+ bnxt_close_nic(bp, true, false);
+
+ rxq = data_pool_get_rx_queue(bp->dev, xdp->zc_rx.queue_id);
+ rxq->queue_id = xdp->zc_rx.queue_id;
+ rxq->zc_ifq = xdp->zc_rx.ifq;
+ rxq->zc_uarg = xdp->zc_rx.uarg;
+
+ rc = bnxt_open_nic(bp, true, false);
+ bnxt_ulp_start(bp, rc);
+ }
+ bnxt_rtnl_unlock_sp(bp);
+ return 0;
+}
+
static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
{
struct net_device *dev;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 9fd85ebd8ae8..554c0abc0d44 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -949,6 +949,7 @@ struct bnxt_rx_ring_info {
struct bnxt_ring_struct rx_agg_ring_struct;
struct xdp_rxq_info xdp_rxq;
struct page_pool *page_pool;
+ struct netdev_rx_queue *rx_queue;
};
struct bnxt_rx_sw_stats {
@@ -2454,4 +2455,7 @@ int bnxt_get_port_parent_id(struct net_device *dev,
void bnxt_dim_work(struct work_struct *work);
int bnxt_hwrm_set_ring_coal(struct bnxt *bp, struct bnxt_napi *bnapi);
void bnxt_print_device_info(struct bnxt *bp);
+
+int bnxt_zc_rx(struct bnxt *bp, struct netdev_bpf *xdp);
+
#endif
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
index 4efa5fe6972b..1387f0e1fff5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
@@ -455,6 +455,9 @@ int bnxt_xdp(struct net_device *dev, struct netdev_bpf *xdp)
case XDP_SETUP_PROG:
rc = bnxt_xdp_set(bp, xdp->prog);
break;
+ case XDP_SETUP_ZC_RX:
+ return bnxt_zc_rx(bp, xdp);
+ break;
default:
rc = -EINVAL;
break;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index bf133cbee721..994237e92cbc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -789,6 +789,10 @@ struct netdev_rx_queue {
struct kobject kobj;
struct net_device *dev;
netdevice_tracker dev_tracker;
+ unsigned int queue_id;
+ struct page_pool *page_pool;
+ struct io_zc_rx_ifq *zc_ifq;
+ struct ubuf_info *zc_uarg;
#ifdef CONFIG_XDP_SOCKETS
struct xsk_buff_pool *pool;
diff --git a/include/net/data_pool.h b/include/net/data_pool.h
new file mode 100644
index 000000000000..84c96aa1c542
--- /dev/null
+++ b/include/net/data_pool.h
@@ -0,0 +1,96 @@
+#ifndef _DATA_POOL_H
+#define _DATA_POOL_H
+
+#include <linux/io_uring.h>
+#include <linux/io_uring_types.h>
+#include <linux/mm_types.h>
+#include <linux/netdevice.h>
+
+static inline struct netdev_rx_queue *
+data_pool_get_rx_queue(struct net_device *dev, unsigned int q_idx)
+{
+ if (q_idx >= dev->num_rx_queues)
+ return NULL;
+ return __netif_get_rx_queue(dev, q_idx);
+}
+
+static inline int data_pool_set_page_pool(struct net_device *dev,
+ unsigned int q_idx,
+ struct page_pool *pool)
+{
+ struct netdev_rx_queue *rxq;
+
+ rxq = data_pool_get_rx_queue(dev, q_idx);
+ if (!rxq)
+ return -EINVAL;
+
+ rxq->page_pool = pool;
+ return 0;
+}
+
+static inline int data_pool_set_zc_ifq(struct net_device *dev,
+ unsigned int q_idx,
+ struct io_zc_rx_ifq *ifq)
+{
+ struct netdev_rx_queue *rxq;
+
+ rxq = data_pool_get_rx_queue(dev, q_idx);
+ if (!rxq)
+ return -EINVAL;
+
+ rxq->zc_ifq = ifq;
+ return 0;
+}
+
+static inline struct page *data_pool_alloc_page(struct netdev_rx_queue *rxq)
+{
+ if (rxq->zc_ifq) {
+ struct io_zc_rx_buf *buf;
+ buf = io_zc_rx_get_buf(rxq->zc_ifq);
+ if (!buf)
+ return NULL;
+ return buf->page;
+ } else {
+ return page_pool_dev_alloc_pages(rxq->page_pool);
+ }
+}
+
+static inline void data_pool_fragment_page(struct netdev_rx_queue *rxq,
+ struct page *page,
+ unsigned long bias)
+{
+ if (rxq->zc_ifq) {
+ struct io_zc_rx_buf *buf;
+ buf = io_zc_rx_buf_from_page(rxq->zc_ifq, page);
+ atomic_set(&buf->refcount, bias);
+ } else {
+ page_pool_fragment_page(page, bias);
+ }
+}
+
+static inline void data_pool_put_page(struct netdev_rx_queue *rxq,
+ struct page *page)
+{
+ if (rxq->zc_ifq) {
+ struct io_zc_rx_buf *buf;
+ buf = io_zc_rx_buf_from_page(rxq->zc_ifq, page);
+ io_zc_rx_put_buf(rxq->zc_ifq, buf);
+ } else {
+ WARN_ON_ONCE(page->pp_magic != PP_SIGNATURE);
+ page_pool_recycle_direct(rxq->page_pool, page);
+ }
+}
+
+static inline dma_addr_t data_pool_get_dma_addr(struct netdev_rx_queue *rxq,
+ struct page *page)
+{
+ if (rxq->zc_ifq) {
+ struct io_zc_rx_buf *buf;
+ buf = io_zc_rx_buf_from_page(rxq->zc_ifq, page);
+ return io_zc_rx_buf_dma(buf);
+ } else {
+ return page_pool_get_dma_addr(page);
+ }
+}
+
+#endif
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* [PATCH 11/11] io_uring: add io_recvzc request
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (9 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 10/11] netdev/bnxt: add data pool and use it in BNXT driver David Wei
@ 2023-08-26 1:19 ` David Wei
2023-10-22 19:06 ` [RFC RESEND 00/11] Zero copy network RX using io_uring Gal Pressman
11 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-08-26 1:19 UTC (permalink / raw)
To: Jens Axboe, Pavel Begunkov; +Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
From: David Wei <[email protected]>
This patch adds an io_uring opcode OP_RECV_ZC for doing ZC reads from a
socket that is set up for ZC RX. The request reads skbs from a socket
where its page frags are tagged w/ a magic cookie in their page private
field. For each frag, entries are written into the ifq rbuf completion
ring, and the total number of bytes read is returned to user as an
io_uring completion event.
Multishot requests work. There is no need to specify provided buffers as
data is returned in the ifq rbuf completion rings.
Userspace is expected to look into the ifq rbuf completion ring when it
receives an io_uring completion event.
The addr3 field is used to encode params in the following format:
addr3 = (readlen << 32) | ifq_id;
readlen is the max amount of data to read from the socket. ifq_id is the
interface queue id, and currently only 0 is supported.
Signed-off-by: David Wei <[email protected]>
Co-developed-by: Jonathan Lemon <[email protected]>
---
include/uapi/linux/io_uring.h | 1 +
io_uring/net.c | 83 +++++++++++-
io_uring/opdef.c | 16 +++
io_uring/zc_rx.c | 232 ++++++++++++++++++++++++++++++++++
io_uring/zc_rx.h | 4 +
5 files changed, 335 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index 28154abfe6f4..c43e5cc7de0a 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -223,6 +223,7 @@ enum io_uring_op {
IORING_OP_URING_CMD,
IORING_OP_SEND_ZC,
IORING_OP_SENDMSG_ZC,
+ IORING_OP_RECV_ZC,
/* this goes last, obviously */
IORING_OP_LAST,
diff --git a/io_uring/net.c b/io_uring/net.c
index 89e839013837..9a0a008418ec 100644
--- a/io_uring/net.c
+++ b/io_uring/net.c
@@ -69,6 +69,12 @@ struct io_sr_msg {
struct io_kiocb *notif;
};
+struct io_recvzc {
+ struct io_sr_msg sr;
+ u32 datalen;
+ u16 ifq_id;
+};
+
static inline bool io_check_multishot(struct io_kiocb *req,
unsigned int issue_flags)
{
@@ -574,7 +580,8 @@ int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
if (sr->msg_flags & MSG_ERRQUEUE)
req->flags |= REQ_F_CLEAR_POLLIN;
if (sr->flags & IORING_RECV_MULTISHOT) {
- if (!(req->flags & REQ_F_BUFFER_SELECT))
+ if (!(req->flags & REQ_F_BUFFER_SELECT)
+ && req->opcode != IORING_OP_RECV_ZC)
return -EINVAL;
if (sr->msg_flags & MSG_WAITALL)
return -EINVAL;
@@ -931,6 +938,80 @@ int io_recv(struct io_kiocb *req, unsigned int issue_flags)
return ret;
}
+int io_recvzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+ struct io_recvzc *zc = io_kiocb_to_cmd(req, struct io_recvzc);
+ u64 recvzc_cmd;
+
+ recvzc_cmd = READ_ONCE(sqe->addr3);
+ zc->datalen = recvzc_cmd >> 32;
+ zc->ifq_id = recvzc_cmd & 0xffff;
+ if (zc->ifq_id != 0)
+ return -EINVAL;
+
+ return io_recvmsg_prep(req, sqe);
+}
+
+int io_recvzc(struct io_kiocb *req, unsigned int issue_flags)
+{
+ struct io_recvzc *zc = io_kiocb_to_cmd(req, struct io_recvzc);
+ struct socket *sock;
+ unsigned flags;
+ int ret, min_ret = 0;
+ bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
+
+ if (!(req->flags & REQ_F_POLLED) &&
+ (zc->sr.flags & IORING_RECVSEND_POLL_FIRST))
+ return -EAGAIN;
+
+ sock = sock_from_file(req->file);
+ if (unlikely(!sock))
+ return -ENOTSOCK;
+
+retry_multishot:
+ flags = zc->sr.msg_flags;
+ if (force_nonblock)
+ flags |= MSG_DONTWAIT;
+ if (flags & MSG_WAITALL)
+ min_ret = zc->sr.len;
+
+ ret = io_zc_rx_recv(sock, zc->datalen, flags);
+ if (ret < min_ret) {
+ if (ret == -EAGAIN && force_nonblock) {
+ if (issue_flags & IO_URING_F_MULTISHOT) {
+ io_kbuf_recycle(req, issue_flags);
+ return IOU_ISSUE_SKIP_COMPLETE;
+ }
+
+ return -EAGAIN;
+ }
+ if (ret > 0 && io_net_retry(sock, flags)) {
+ zc->sr.len -= ret;
+ zc->sr.buf += ret;
+ zc->sr.done_io += ret;
+ req->flags |= REQ_F_PARTIAL_IO;
+ return -EAGAIN;
+ }
+ if (ret == -ERESTARTSYS)
+ ret = -EINTR;
+ req_set_fail(req);
+ } else if ((flags & MSG_WAITALL) && (flags & (MSG_TRUNC | MSG_CTRUNC))) {
+ req_set_fail(req);
+ }
+
+ if (ret > 0)
+ ret += zc->sr.done_io;
+ else if (zc->sr.done_io)
+ ret = zc->sr.done_io;
+ else
+ io_kbuf_recycle(req, issue_flags);
+
+ if (!io_recv_finish(req, &ret, 0, ret <= 0, issue_flags))
+ goto retry_multishot;
+
+ return ret;
+}
+
void io_send_zc_cleanup(struct io_kiocb *req)
{
struct io_sr_msg *zc = io_kiocb_to_cmd(req, struct io_sr_msg);
diff --git a/io_uring/opdef.c b/io_uring/opdef.c
index 3b9c6489b8b6..4dee7f83222f 100644
--- a/io_uring/opdef.c
+++ b/io_uring/opdef.c
@@ -33,6 +33,7 @@
#include "poll.h"
#include "cancel.h"
#include "rw.h"
+#include "zc_rx.h"
static int io_no_issue(struct io_kiocb *req, unsigned int issue_flags)
{
@@ -426,6 +427,18 @@ const struct io_issue_def io_issue_defs[] = {
.issue = io_sendmsg_zc,
#else
.prep = io_eopnotsupp_prep,
+#endif
+ },
+ [IORING_OP_RECV_ZC] = {
+ .needs_file = 1,
+ .unbound_nonreg_file = 1,
+ .pollin = 1,
+ .ioprio = 1,
+#if defined(CONFIG_NET)
+ .prep = io_recvzc_prep,
+ .issue = io_recvzc,
+#else
+ .prep = io_eopnotsupp_prep,
#endif
},
};
@@ -648,6 +661,9 @@ const struct io_cold_def io_cold_defs[] = {
.fail = io_sendrecv_fail,
#endif
},
+ [IORING_OP_RECV_ZC] = {
+ .name = "RECV_ZC",
+ },
};
const char *io_uring_get_opcode(u8 opcode)
diff --git a/io_uring/zc_rx.c b/io_uring/zc_rx.c
index 70e39f851e47..a861be50fd61 100644
--- a/io_uring/zc_rx.c
+++ b/io_uring/zc_rx.c
@@ -5,6 +5,7 @@
#include <linux/io_uring.h>
#include <linux/netdevice.h>
#include <linux/nospec.h>
+#include <net/tcp.h>
#include <uapi/linux/io_uring.h>
@@ -38,6 +39,13 @@ struct io_zc_rx_pool {
u32 freelist[];
};
+static inline u32 io_zc_rx_cqring_entries(struct io_zc_rx_ifq *ifq)
+{
+ struct io_rbuf_ring *ring = ifq->ring;
+
+ return ifq->cached_cq_tail - READ_ONCE(ring->cq.head);
+}
+
static struct device *netdev2dev(struct net_device *dev)
{
return dev->dev.parent;
@@ -381,6 +389,14 @@ int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx)
return 0;
}
+static void io_zc_rx_get_buf_uref(struct io_zc_rx_pool *pool, u16 pgid)
+{
+ if (WARN_ON(pgid >= pool->nr_pages))
+ return;
+
+ atomic_add(IO_ZC_RX_UREF, &pool->bufs[pgid].refcount);
+}
+
static bool io_zc_rx_put_buf_uref(struct io_zc_rx_buf *buf)
{
if (atomic_read(&buf->refcount) < IO_ZC_RX_UREF)
@@ -489,3 +505,219 @@ struct io_zc_rx_buf *io_zc_rx_buf_from_page(struct io_zc_rx_ifq *ifq,
return &pool->bufs[pgid];
}
EXPORT_SYMBOL(io_zc_rx_buf_from_page);
+
+static struct io_zc_rx_ifq *io_zc_rx_ifq_skb(struct sk_buff *skb)
+{
+ struct ubuf_info *uarg = skb_zcopy(skb);
+
+ if (uarg && uarg->callback == io_zc_rx_skb_free)
+ return container_of(uarg, struct io_zc_rx_ifq, uarg);
+ return NULL;
+}
+
+static int zc_rx_recv_frag(struct io_zc_rx_ifq *ifq, const skb_frag_t *frag,
+ int off, int len)
+{
+ struct io_uring_rbuf_cqe *cqe;
+ unsigned int pgid, cq_idx, queued, free, entries;
+ struct page *page;
+ unsigned int mask;
+
+ page = skb_frag_page(frag);
+ off += skb_frag_off(frag);
+
+ if (likely(ifq && is_zc_rx_page(page))) {
+ mask = ifq->cq_entries - 1;
+ pgid = page_private(page) & 0xffffffff;
+ io_zc_rx_get_buf_uref(ifq->pool, pgid);
+ cq_idx = ifq->cached_cq_tail & mask;
+ smp_rmb();
+ queued = min(io_zc_rx_cqring_entries(ifq), ifq->cq_entries);
+ free = ifq->cq_entries - queued;
+ entries = min(free, ifq->cq_entries - cq_idx);
+ if (!entries)
+ return -ENOBUFS;
+ cqe = &ifq->cqes[cq_idx];
+ ifq->cached_cq_tail++;
+ cqe->region = 0;
+ cqe->off = pgid * PAGE_SIZE + off;
+ cqe->len = len;
+ cqe->flags = 0;
+ } else {
+ /* TODO: copy frags that aren't backed by zc pages */
+ WARN_ON_ONCE(1);
+ return -ENOMEM;
+ }
+
+ return len;
+}
+
+static int
+zc_rx_recv_skb(read_descriptor_t *desc, struct sk_buff *skb,
+ unsigned int offset, size_t len)
+{
+ struct io_zc_rx_ifq *ifq;
+ struct sk_buff *frag_iter;
+ unsigned start, start_off;
+ int i, copy, end, off;
+ int ret = 0;
+
+ ifq = io_zc_rx_ifq_skb(skb);
+ start = skb_headlen(skb);
+ start_off = offset;
+
+ // TODO: copy payload in skb linear data */
+ WARN_ON_ONCE(offset < start);
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+ const skb_frag_t *frag;
+
+ WARN_ON(start > offset + len);
+
+ frag = &skb_shinfo(skb)->frags[i];
+ end = start + skb_frag_size(frag);
+
+ if (offset < end) {
+ copy = end - offset;
+ if (copy > len)
+ copy = len;
+
+ off = offset - start;
+ ret = zc_rx_recv_frag(ifq, frag, off, copy);
+ if (ret < 0)
+ goto out;
+
+ offset += ret;
+ len -= ret;
+ if (len == 0 || ret != copy)
+ goto out;
+ }
+ start = end;
+ }
+
+ skb_walk_frags(skb, frag_iter) {
+ WARN_ON(start > offset + len);
+
+ end = start + frag_iter->len;
+ if (offset < end) {
+ copy = end - offset;
+ if (copy > len)
+ copy = len;
+
+ off = offset - start;
+ ret = zc_rx_recv_skb(desc, frag_iter, off, copy);
+ if (ret < 0)
+ goto out;
+
+ offset += ret;
+ len -= ret;
+ if (len == 0 || ret != copy)
+ goto out;
+ }
+ start = end;
+ }
+
+out:
+ smp_store_release(&ifq->ring->cq.tail, ifq->cached_cq_tail);
+ if (offset == start_off)
+ return ret;
+ return offset - start_off;
+}
+
+static int io_zc_rx_tcp_read(struct sock *sk)
+{
+ read_descriptor_t rd_desc = {
+ .count = 1,
+ };
+
+ return tcp_read_sock(sk, &rd_desc, zc_rx_recv_skb);
+}
+
+static int io_zc_rx_tcp_recvmsg(struct sock *sk, unsigned int recv_limit,
+ int flags, int *addr_len)
+{
+ size_t used;
+ long timeo;
+ int ret;
+
+ ret = used = 0;
+
+ lock_sock(sk);
+
+ timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+ while (recv_limit) {
+ ret = io_zc_rx_tcp_read(sk);
+ if (ret < 0)
+ break;
+ if (!ret) {
+ if (used)
+ break;
+ if (sock_flag(sk, SOCK_DONE))
+ break;
+ if (sk->sk_err) {
+ ret = sock_error(sk);
+ break;
+ }
+ if (sk->sk_shutdown & RCV_SHUTDOWN)
+ break;
+ if (sk->sk_state == TCP_CLOSE) {
+ ret = -ENOTCONN;
+ break;
+ }
+ if (!timeo) {
+ ret = -EAGAIN;
+ break;
+ }
+ if (!skb_queue_empty(&sk->sk_receive_queue))
+ break;
+ sk_wait_data(sk, &timeo, NULL);
+ if (signal_pending(current)) {
+ ret = sock_intr_errno(timeo);
+ break;
+ }
+ continue;
+ }
+ recv_limit -= ret;
+ used += ret;
+
+ if (!timeo)
+ break;
+ release_sock(sk);
+ lock_sock(sk);
+
+ if (sk->sk_err || sk->sk_state == TCP_CLOSE ||
+ (sk->sk_shutdown & RCV_SHUTDOWN) ||
+ signal_pending(current))
+ break;
+ }
+
+ release_sock(sk);
+
+ /* TODO: handle timestamping */
+
+ if (used)
+ return used;
+
+ return ret;
+}
+
+int io_zc_rx_recv(struct socket *sock, unsigned int limit, unsigned int flags)
+{
+ struct sock *sk = sock->sk;
+ const struct proto *prot;
+ int addr_len = 0;
+ int ret;
+
+ if (flags & MSG_ERRQUEUE)
+ return -EOPNOTSUPP;
+
+ prot = READ_ONCE(sk->sk_prot);
+ if (prot->recvmsg != tcp_recvmsg)
+ return -EPROTONOSUPPORT;
+
+ sock_rps_record_flow(sk);
+
+ ret = io_zc_rx_tcp_recvmsg(sk, limit, flags, &addr_len);
+
+ return ret;
+}
diff --git a/io_uring/zc_rx.h b/io_uring/zc_rx.h
index ee7e36295f91..63ff81a7900a 100644
--- a/io_uring/zc_rx.h
+++ b/io_uring/zc_rx.h
@@ -35,4 +35,8 @@ int io_register_zc_rx_ifq(struct io_ring_ctx *ctx,
int io_unregister_zc_rx_ifq(struct io_ring_ctx *ctx);
int io_zc_rx_pool_create(struct io_zc_rx_ifq *ifq, u16 id);
+int io_recvzc(struct io_kiocb *req, unsigned int issue_flags);
+int io_recvzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
+int io_zc_rx_recv(struct socket *sock, unsigned int limit, unsigned int flags);
+
#endif
--
2.39.3
^ permalink raw reply related [flat|nested] 18+ messages in thread* Re: [RFC RESEND 00/11] Zero copy network RX using io_uring
2023-08-26 1:19 [RFC RESEND 00/11] Zero copy network RX using io_uring David Wei
` (10 preceding siblings ...)
2023-08-26 1:19 ` [PATCH 11/11] io_uring: add io_recvzc request David Wei
@ 2023-10-22 19:06 ` Gal Pressman
2023-10-23 3:35 ` David Wei
11 siblings, 1 reply; 18+ messages in thread
From: Gal Pressman @ 2023-10-22 19:06 UTC (permalink / raw)
To: David Wei, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 26/08/2023 4:19, David Wei wrote:
> From: David Wei <[email protected]>
>
> This patchset is a proposal that adds zero copy network RX to io_uring.
> With it, userspace can register a region of host memory for receiving
> data directly from a NIC using DMA, without needing a kernel to user
> copy.
>
> Software support is added to the Broadcom BNXT driver. Hardware support
> for receive flow steering and header splitting is required.
>
> On the userspace side, a sample server is added in this branch of
> liburing:
> https://github.com/spikeh/liburing/tree/zcrx2
>
> Build liburing as normal, and run examples/zcrx. Then, set flow steering
> rules using ethtool. A sample shell script is included in
> examples/zcrx_flow.sh, but you need to change the source IP. Finally,
> connect a client using e.g. netcat and send data.
>
> This patchset + userspace code was tested on an Intel Xeon Platinum
> 8321HC CPU and Broadcom BCM57504 NIC.
>
> Early benchmarks using this prototype, with iperf3 as a load generator,
> showed a ~50% reduction in overall system memory bandwidth as measured
> using perf counters. Note that DDIO must be disabled on Intel systems.
>
> Mina et al. from Google and Kuba are collaborating on a similar proposal
> to ZC from NIC to devmem. There are many shared functionality in netdev
> that we can collaborate on e.g.:
> * Page pool memory provider backend and resource registration
> * Page pool refcounted iov/buf representation and lifecycle
> * Setting receive flow steering
>
> As mentioned earlier, this is an early prototype. It is brittle, some
> functionality is missing and there's little optimisation. We're looking
> for feedback on the overall approach and points of collaboration in
> netdev.
> * No copy fallback, if payload ends up in linear part of skb then the
> code will not work
> * No way to pin an RX queue to a specific CPU
> * Only one ifq, one pool region, on RX queue...
>
> This patchset is based on the work by Jonathan Lemon
> <[email protected]>:
> https://lore.kernel.org/io-uring/[email protected]/
Hello David,
This work looks interesting, is there anywhere I can read about it some
more? Maybe it was presented (and hopefully recorded) in a recent
conference?
Maybe something geared towards adding more drivers support?
I took a brief look at the bnxt patch and saw you converted the page
pool allocation to data pool allocation, I assume this is done for data
pages only, right? Headers are still allocated on page pool pages?
Thanks
^ permalink raw reply [flat|nested] 18+ messages in thread* Re: [RFC RESEND 00/11] Zero copy network RX using io_uring
2023-10-22 19:06 ` [RFC RESEND 00/11] Zero copy network RX using io_uring Gal Pressman
@ 2023-10-23 3:35 ` David Wei
0 siblings, 0 replies; 18+ messages in thread
From: David Wei @ 2023-10-23 3:35 UTC (permalink / raw)
To: Gal Pressman, Jens Axboe, Pavel Begunkov
Cc: io-uring, netdev, Mina Almasry, Jakub Kicinski
On 2023-10-22 12:06, Gal Pressman wrote:
> On 26/08/2023 4:19, David Wei wrote:
>> From: David Wei <[email protected]>
>>
>> This patchset is a proposal that adds zero copy network RX to io_uring.
>> With it, userspace can register a region of host memory for receiving
>> data directly from a NIC using DMA, without needing a kernel to user
>> copy.
>>
>> Software support is added to the Broadcom BNXT driver. Hardware support
>> for receive flow steering and header splitting is required.
>>
>> On the userspace side, a sample server is added in this branch of
>> liburing:
>> https://github.com/spikeh/liburing/tree/zcrx2
>>
>> Build liburing as normal, and run examples/zcrx. Then, set flow steering
>> rules using ethtool. A sample shell script is included in
>> examples/zcrx_flow.sh, but you need to change the source IP. Finally,
>> connect a client using e.g. netcat and send data.
>>
>> This patchset + userspace code was tested on an Intel Xeon Platinum
>> 8321HC CPU and Broadcom BCM57504 NIC.
>>
>> Early benchmarks using this prototype, with iperf3 as a load generator,
>> showed a ~50% reduction in overall system memory bandwidth as measured
>> using perf counters. Note that DDIO must be disabled on Intel systems.
>>
>> Mina et al. from Google and Kuba are collaborating on a similar proposal
>> to ZC from NIC to devmem. There are many shared functionality in netdev
>> that we can collaborate on e.g.:
>> * Page pool memory provider backend and resource registration
>> * Page pool refcounted iov/buf representation and lifecycle
>> * Setting receive flow steering
>>
>> As mentioned earlier, this is an early prototype. It is brittle, some
>> functionality is missing and there's little optimisation. We're looking
>> for feedback on the overall approach and points of collaboration in
>> netdev.
>> * No copy fallback, if payload ends up in linear part of skb then the
>> code will not work
>> * No way to pin an RX queue to a specific CPU
>> * Only one ifq, one pool region, on RX queue...
>>
>> This patchset is based on the work by Jonathan Lemon
>> <[email protected]>:
>> https://lore.kernel.org/io-uring/[email protected]/
>
> Hello David,
>
> This work looks interesting, is there anywhere I can read about it some
> more? Maybe it was presented (and hopefully recorded) in a recent
> conference?
> Maybe something geared towards adding more drivers support?
>
Hi Gal,
Thank you for your interest in our work! We will be publishing a paper
and presenting this work at NetDev conference on 1 Nov.
Support for more drivers (e.g. mlx5) is definitely on our radar. We are
collaborating with Mina and others from Google who are working on a
similar proposal but targetting NIC -> ZC RX into GPU memory. We both
require shared bits of infra e.g. page pool memory providers that will
replace the use of a one-off data_pool in this patchset. This would
minimise driver changes needed to support this feature.
> I took a brief look at the bnxt patch and saw you converted the page
> pool allocation to data pool allocation, I assume this is done for data
> pages only, right? Headers are still allocated on page pool pages?
>
> Thanks
Yes, that's right.
David
^ permalink raw reply [flat|nested] 18+ messages in thread