Files
linux/net/core/devmem.h
Bobby Eshleman b27a8560ee net: devmem: allow rx-page-size > PAGE_SIZE per dmabuf binding
Every devmem dmabuf binding today hands the page_pool PAGE_SIZE niovs.
This caps a single RX descriptor at PAGE_SIZE, burning CPU on buffer
churn for large flows.

Add a bind-time netlink attribute, NETDEV_A_DMABUF_RX_PAGE_SIZE, that
lets userspace request a larger niov size. The value must be a power of
two >= PAGE_SIZE.

The TX path is changed to always pass PAGE_SIZE.

Measurements:
Setup: kperf in devmem RX/TX cuda mode, 4 flows, 64 MB messages, 60s,
dctcp, num-rx-queues=4, dmabuf-rx/tx-size-mb=2048, 10 runs per niov
size, mlx5.

CPU Util:

   niov        net sirq %        net idle %         app sys %        app idle %
  -----  ----------------  ----------------  ----------------  ----------------
     4K   62.38 +/-  8.27   33.40 +/-  7.51   54.15 +/- 10.23   43.67 +/- 10.53
    16K   58.91 +/-  5.35   35.23 +/-  5.88   41.05 +/-  8.87   56.42 +/-  9.24
    32K   64.12 +/-  0.68   31.09 +/-  1.48   44.54 +/-  3.51   52.63 +/-  3.65
    64K   54.69 +/-  5.54   39.67 +/-  5.81   35.47 +/-  3.11   61.97 +/-  3.27

RX app sys % drops ~19% from 4K to 64K.

Throughput:

   niov       RX dev Gbps   RX flow avg Gbps
  -----  ----------------  -----------------
     4K  300.63 +/- 53.21    75.16 +/- 13.30
    16K  321.35 +/- 28.20    80.34 +/-  7.05
    32K  347.63 +/-  2.20    86.91 +/-  0.55
    64K  332.11 +/- 14.26    83.03 +/-  3.56

Throughput seems to increase, but the stdev is pretty wide so could just
be noise.

kperf support (not yet merged):
8837577f92

Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
Link: https://patch.msgid.link/20260805-tcpdm-large-niovs-v8-1-3e0225e2808c@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2026-08-07 18:32:39 -07:00

249 lines
6.7 KiB
C

/* SPDX-License-Identifier: GPL-2.0-or-later */
/*
* Device memory TCP support
*
* Authors: Mina Almasry <almasrymina@google.com>
* Willem de Bruijn <willemb@google.com>
* Kaiyuan Zhang <kaiyuanz@google.com>
*
*/
#ifndef _NET_DEVMEM_H
#define _NET_DEVMEM_H
#include <net/netmem.h>
#include <net/netdev_netlink.h>
struct netlink_ext_ack;
struct net_devmem_dmabuf_binding {
struct dma_buf *dmabuf;
struct dma_buf_attachment *attachment;
struct sg_table *sgt;
/* Physical NIC that does the actual DMA for this binding. */
struct net_device *dev;
/* Opaque cookie identifying the virtual device (e.g. netkit) the user
* called bind-tx on. Used only for pointer comparison. Never
* dereferenced.
*/
void *vdev;
struct gen_pool *chunk_pool;
/* Protect dev */
struct mutex lock;
/* The user holds a ref (via the netlink API) for as long as they want
* the binding to remain alive. Each page pool using this binding holds
* a ref to keep the binding alive. The page_pool does not release the
* ref until all the net_iovs allocated from this binding are released
* back to the page_pool.
*
* The binding undos itself and unmaps the underlying dmabuf once all
* those refs are dropped and the binding is no longer desired or in
* use.
*
* net_devmem_get_net_iov() on dmabuf net_iovs will increment this
* reference, making sure that the binding remains alive until all the
* net_iovs are no longer used. net_iovs allocated from this binding
* that are stuck in the TX path for any reason (such as awaiting
* retransmits) hold a reference to the binding until the skb holding
* them is freed.
*/
struct percpu_ref ref;
/* The list of bindings currently active. Used for netlink to notify us
* of the user dropping the bind.
*/
struct list_head list;
/* rxq's this binding is active on. */
struct xarray bound_rxqs;
/* ID of this binding. Globally unique to all bindings currently
* active.
*/
u32 id;
/* DMA direction, FROM_DEVICE for Rx binding, TO_DEVICE for Tx. */
enum dma_data_direction direction;
/* Array of net_iov pointers for this binding, sorted by virtual
* address. This array is convenient to map the virtual addresses to
* net_iovs in the TX path.
*/
struct net_iov **tx_vec;
unsigned int niov_shift;
struct work_struct unbind_w;
};
#if defined(CONFIG_NET_DEVMEM)
/* Owner of the dma-buf chunks inserted into the gen pool. Each scatterlist
* entry from the dmabuf is inserted into the genpool as a chunk, and needs
* this owner struct to keep track of some metadata necessary to create
* allocations from this chunk.
*/
struct dmabuf_genpool_chunk_owner {
struct net_iov_area area;
struct net_devmem_dmabuf_binding *binding;
/* dma_addr of the start of the chunk. */
dma_addr_t base_dma_addr;
};
void __net_devmem_dmabuf_binding_free(struct work_struct *wq);
struct net_devmem_dmabuf_binding *
net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
struct device *dma_dev,
enum dma_data_direction direction,
unsigned int dmabuf_fd, unsigned int niov_shift,
struct netdev_nl_sock *priv,
struct netlink_ext_ack *extack);
struct net_devmem_dmabuf_binding *net_devmem_lookup_dmabuf(u32 id);
void net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding);
int net_devmem_bind_dmabuf_to_queue(struct net_device *dev, u32 rxq_idx,
struct net_devmem_dmabuf_binding *binding,
struct netlink_ext_ack *extack);
static inline struct dmabuf_genpool_chunk_owner *
net_devmem_iov_to_chunk_owner(const struct net_iov *niov)
{
struct net_iov_area *owner = net_iov_owner(niov);
return container_of(owner, struct dmabuf_genpool_chunk_owner, area);
}
static inline struct net_devmem_dmabuf_binding *
net_devmem_iov_binding(const struct net_iov *niov)
{
return net_devmem_iov_to_chunk_owner(niov)->binding;
}
static inline u32 net_devmem_iov_binding_id(const struct net_iov *niov)
{
return net_devmem_iov_binding(niov)->id;
}
static inline unsigned long net_iov_virtual_addr(const struct net_iov *niov)
{
struct dmabuf_genpool_chunk_owner *co =
net_devmem_iov_to_chunk_owner(niov);
return net_iov_owner(niov)->base_virtual +
((unsigned long)net_iov_idx(niov) << co->binding->niov_shift);
}
static inline bool
net_devmem_dmabuf_binding_get(struct net_devmem_dmabuf_binding *binding)
{
return percpu_ref_tryget(&binding->ref);
}
static inline void
net_devmem_dmabuf_binding_put(struct net_devmem_dmabuf_binding *binding)
{
percpu_ref_put(&binding->ref);
}
void net_devmem_get_net_iov(struct net_iov *niov);
void net_devmem_put_net_iov(struct net_iov *niov);
struct net_iov *
net_devmem_alloc_dmabuf(struct net_devmem_dmabuf_binding *binding);
void net_devmem_free_dmabuf(struct net_iov *ppiov);
struct net_devmem_dmabuf_binding *
net_devmem_get_binding(struct sock *sk, unsigned int dmabuf_id);
struct net_iov *
net_devmem_get_niov_at(struct net_devmem_dmabuf_binding *binding, size_t addr,
size_t *off, size_t *size);
#else
struct net_devmem_dmabuf_binding;
static inline void
net_devmem_dmabuf_binding_put(struct net_devmem_dmabuf_binding *binding)
{
}
static inline void net_devmem_get_net_iov(struct net_iov *niov)
{
}
static inline void net_devmem_put_net_iov(struct net_iov *niov)
{
}
static inline struct net_devmem_dmabuf_binding *
net_devmem_bind_dmabuf(struct net_device *dev, void *vdev,
struct device *dma_dev,
enum dma_data_direction direction,
unsigned int dmabuf_fd,
unsigned int niov_shift,
struct netdev_nl_sock *priv,
struct netlink_ext_ack *extack)
{
return ERR_PTR(-EOPNOTSUPP);
}
static inline struct net_devmem_dmabuf_binding *net_devmem_lookup_dmabuf(u32 id)
{
return NULL;
}
static inline void
net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding)
{
}
static inline int
net_devmem_bind_dmabuf_to_queue(struct net_device *dev, u32 rxq_idx,
struct net_devmem_dmabuf_binding *binding,
struct netlink_ext_ack *extack)
{
return -EOPNOTSUPP;
}
static inline struct net_iov *
net_devmem_alloc_dmabuf(struct net_devmem_dmabuf_binding *binding)
{
return NULL;
}
static inline void net_devmem_free_dmabuf(struct net_iov *ppiov)
{
}
static inline unsigned long net_iov_virtual_addr(const struct net_iov *niov)
{
return 0;
}
static inline u32 net_devmem_iov_binding_id(const struct net_iov *niov)
{
return 0;
}
static inline struct net_devmem_dmabuf_binding *
net_devmem_get_binding(struct sock *sk, unsigned int dmabuf_id)
{
return ERR_PTR(-EOPNOTSUPP);
}
static inline struct net_iov *
net_devmem_get_niov_at(struct net_devmem_dmabuf_binding *binding, size_t addr,
size_t *off, size_t *size)
{
return NULL;
}
static inline struct net_devmem_dmabuf_binding *
net_devmem_iov_binding(const struct net_iov *niov)
{
return NULL;
}
#endif
#endif /* _NET_DEVMEM_H */