io_uring/zcrx: implement device-less mode for zcrx

Allow creating a zcrx instance without attaching it to a net device.
All data will be copied through the fallback path. The user is also
expected to use ZCRX_CTRL_FLUSH_RQ to handle overflows as it normally
should even with a netdev, but it becomes even more relevant as there
will likely be no one to automatically pick up buffers.

Apart from that, it follows the zcrx uapi for the I/O path, and is
useful for testing, experimentation, and potentially for the copy
receive path in the future if improved.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://patch.msgid.link/674f8ad679c5a0bc79d538352b3042cf0999596e.1774261953.git.asml.silence@gmail.com
[axboe: fix spelling error in uapi header and commit message]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Pavel Begunkov
2026-03-23 12:43:54 +00:00
committed by Jens Axboe
parent 06fc3b6d38
commit 825f276491
3 changed files with 36 additions and 16 deletions

View File

@@ -49,7 +49,14 @@ struct io_uring_zcrx_area_reg {
};
enum zcrx_reg_flags {
ZCRX_REG_IMPORT = 1,
ZCRX_REG_IMPORT = 1,
/*
* Register a zcrx instance without a net device. All data will be
* copied. The refill queue entries might not be automatically
* consumed and need to be flushed, see ZCRX_CTRL_FLUSH_RQ.
*/
ZCRX_REG_NODEV = 2,
};
enum zcrx_features {

View File

@@ -127,10 +127,10 @@ static int io_import_dmabuf(struct io_zcrx_ifq *ifq,
int dmabuf_fd = area_reg->dmabuf_fd;
int i, ret;
if (!ifq->dev)
return -EINVAL;
if (off)
return -EINVAL;
if (WARN_ON_ONCE(!ifq->dev))
return -EFAULT;
if (!IS_ENABLED(CONFIG_DMA_SHARED_BUFFER))
return -EINVAL;
@@ -211,11 +211,13 @@ static int io_import_umem(struct io_zcrx_ifq *ifq,
if (ret)
goto out_err;
ret = dma_map_sgtable(ifq->dev, &mem->page_sg_table,
DMA_FROM_DEVICE, IO_DMA_ATTR);
if (ret < 0)
goto out_err;
mapped = true;
if (ifq->dev) {
ret = dma_map_sgtable(ifq->dev, &mem->page_sg_table,
DMA_FROM_DEVICE, IO_DMA_ATTR);
if (ret < 0)
goto out_err;
mapped = true;
}
mem->account_pages = io_count_account_pages(pages, nr_pages);
ret = io_account_mem(ifq->user, ifq->mm_account, mem->account_pages);
@@ -450,7 +452,8 @@ static int io_zcrx_create_area(struct io_zcrx_ifq *ifq,
ret = io_import_area(ifq, &area->mem, area_reg);
if (ret)
goto err;
area->is_mapped = true;
if (ifq->dev)
area->is_mapped = true;
if (buf_size_shift > io_area_max_shift(&area->mem)) {
ret = -ERANGE;
@@ -486,9 +489,11 @@ static int io_zcrx_create_area(struct io_zcrx_ifq *ifq,
niov->type = NET_IOV_IOURING;
}
ret = io_populate_area_dma(ifq, area);
if (ret)
goto err;
if (ifq->dev) {
ret = io_populate_area_dma(ifq, area);
if (ret)
goto err;
}
area->free_count = nr_iovs;
/* we're only supporting one area per ifq for now */
@@ -826,6 +831,8 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
return -EFAULT;
if (reg.if_rxq == -1 || !reg.rq_entries)
return -EINVAL;
if ((reg.if_rxq || reg.if_idx) && (reg.flags & ZCRX_REG_NODEV))
return -EINVAL;
if (reg.rq_entries > IO_RQ_MAX_ENTRIES) {
if (!(ctx->flags & IORING_SETUP_CLAMP))
return -EINVAL;
@@ -861,9 +868,15 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
if (ret)
goto err;
ret = zcrx_register_netdev(ifq, &reg, &area);
if (ret)
goto err;
if (!(reg.flags & ZCRX_REG_NODEV)) {
ret = zcrx_register_netdev(ifq, &reg, &area);
if (ret)
goto err;
} else {
ret = io_zcrx_create_area(ifq, &area, &reg);
if (ret)
goto err;
}
reg.zcrx_id = id;

View File

@@ -8,7 +8,7 @@
#include <net/page_pool/types.h>
#include <net/net_trackers.h>
#define ZCRX_SUPPORTED_REG_FLAGS (ZCRX_REG_IMPORT)
#define ZCRX_SUPPORTED_REG_FLAGS (ZCRX_REG_IMPORT | ZCRX_REG_NODEV)
#define ZCRX_FEATURES (ZCRX_FEATURE_RX_PAGE_SIZE)
struct io_zcrx_mem {