linux/include/net/xsk_buff_pool.h
Magnus Karlsson 9647c57b11 xsk: i40e: ice: ixgbe: mlx5: Test for dma_need_sync earlier for better performance
Test for dma_need_sync earlier to increase
performance. xsk_buff_dma_sync_for_cpu() takes an xdp_buff as
parameter and from that the xsk_buff_pool reference is dug out. Perf
shows that this dereference causes a lot of cache misses. But as the
buffer pool is now sent down to the driver at zero-copy initialization
time, we might as well use this pointer directly, instead of going via
the xsk_buff and we can do so already in xsk_buff_dma_sync_for_cpu()
instead of in xp_dma_sync_for_cpu. This gets rid of these cache
misses.

Throughput increases with 3% for the xdpsock l2fwd sample application
on my machine.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Link: https://lore.kernel.org/bpf/1598603189-32145-11-git-send-email-magnus.karlsson@intel.com
2020-08-31 21:15:04 +02:00

172 lines
4.5 KiB
C

/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright(c) 2020 Intel Corporation. */
#ifndef XSK_BUFF_POOL_H_
#define XSK_BUFF_POOL_H_
#include <linux/if_xdp.h>
#include <linux/types.h>
#include <linux/dma-mapping.h>
#include <net/xdp.h>
struct xsk_buff_pool;
struct xdp_rxq_info;
struct xsk_queue;
struct xdp_desc;
struct xdp_umem;
struct xdp_sock;
struct device;
struct page;
struct xdp_buff_xsk {
struct xdp_buff xdp;
dma_addr_t dma;
dma_addr_t frame_dma;
struct xsk_buff_pool *pool;
bool unaligned;
u64 orig_addr;
struct list_head free_list_node;
};
struct xsk_dma_map {
dma_addr_t *dma_pages;
struct device *dev;
struct net_device *netdev;
refcount_t users;
struct list_head list; /* Protected by the RTNL_LOCK */
u32 dma_pages_cnt;
bool dma_need_sync;
};
struct xsk_buff_pool {
/* Members only used in the control path first. */
struct device *dev;
struct net_device *netdev;
struct list_head xsk_tx_list;
/* Protects modifications to the xsk_tx_list */
spinlock_t xsk_tx_list_lock;
refcount_t users;
struct xdp_umem *umem;
struct work_struct work;
struct list_head free_list;
u32 heads_cnt;
u16 queue_id;
/* Data path members as close to free_heads at the end as possible. */
struct xsk_queue *fq ____cacheline_aligned_in_smp;
struct xsk_queue *cq;
/* For performance reasons, each buff pool has its own array of dma_pages
* even when they are identical.
*/
dma_addr_t *dma_pages;
struct xdp_buff_xsk *heads;
u64 chunk_mask;
u64 addrs_cnt;
u32 free_list_cnt;
u32 dma_pages_cnt;
u32 free_heads_cnt;
u32 headroom;
u32 chunk_size;
u32 frame_len;
u8 cached_need_wakeup;
bool uses_need_wakeup;
bool dma_need_sync;
bool unaligned;
void *addrs;
struct xdp_buff_xsk *free_heads[];
};
/* AF_XDP core. */
struct xsk_buff_pool *xp_create_and_assign_umem(struct xdp_sock *xs,
struct xdp_umem *umem);
int xp_assign_dev(struct xsk_buff_pool *pool, struct net_device *dev,
u16 queue_id, u16 flags);
void xp_destroy(struct xsk_buff_pool *pool);
void xp_release(struct xdp_buff_xsk *xskb);
void xp_get_pool(struct xsk_buff_pool *pool);
void xp_put_pool(struct xsk_buff_pool *pool);
void xp_clear_dev(struct xsk_buff_pool *pool);
void xp_add_xsk(struct xsk_buff_pool *pool, struct xdp_sock *xs);
void xp_del_xsk(struct xsk_buff_pool *pool, struct xdp_sock *xs);
/* AF_XDP, and XDP core. */
void xp_free(struct xdp_buff_xsk *xskb);
/* AF_XDP ZC drivers, via xdp_sock_buff.h */
void xp_set_rxq_info(struct xsk_buff_pool *pool, struct xdp_rxq_info *rxq);
int xp_dma_map(struct xsk_buff_pool *pool, struct device *dev,
unsigned long attrs, struct page **pages, u32 nr_pages);
void xp_dma_unmap(struct xsk_buff_pool *pool, unsigned long attrs);
struct xdp_buff *xp_alloc(struct xsk_buff_pool *pool);
bool xp_can_alloc(struct xsk_buff_pool *pool, u32 count);
void *xp_raw_get_data(struct xsk_buff_pool *pool, u64 addr);
dma_addr_t xp_raw_get_dma(struct xsk_buff_pool *pool, u64 addr);
static inline dma_addr_t xp_get_dma(struct xdp_buff_xsk *xskb)
{
return xskb->dma;
}
static inline dma_addr_t xp_get_frame_dma(struct xdp_buff_xsk *xskb)
{
return xskb->frame_dma;
}
void xp_dma_sync_for_cpu_slow(struct xdp_buff_xsk *xskb);
static inline void xp_dma_sync_for_cpu(struct xdp_buff_xsk *xskb)
{
xp_dma_sync_for_cpu_slow(xskb);
}
void xp_dma_sync_for_device_slow(struct xsk_buff_pool *pool, dma_addr_t dma,
size_t size);
static inline void xp_dma_sync_for_device(struct xsk_buff_pool *pool,
dma_addr_t dma, size_t size)
{
if (!pool->dma_need_sync)
return;
xp_dma_sync_for_device_slow(pool, dma, size);
}
/* Masks for xdp_umem_page flags.
* The low 12-bits of the addr will be 0 since this is the page address, so we
* can use them for flags.
*/
#define XSK_NEXT_PG_CONTIG_SHIFT 0
#define XSK_NEXT_PG_CONTIG_MASK BIT_ULL(XSK_NEXT_PG_CONTIG_SHIFT)
static inline bool xp_desc_crosses_non_contig_pg(struct xsk_buff_pool *pool,
u64 addr, u32 len)
{
bool cross_pg = (addr & (PAGE_SIZE - 1)) + len > PAGE_SIZE;
if (pool->dma_pages_cnt && cross_pg) {
return !(pool->dma_pages[addr >> PAGE_SHIFT] &
XSK_NEXT_PG_CONTIG_MASK);
}
return false;
}
static inline u64 xp_aligned_extract_addr(struct xsk_buff_pool *pool, u64 addr)
{
return addr & pool->chunk_mask;
}
static inline u64 xp_unaligned_extract_addr(u64 addr)
{
return addr & XSK_UNALIGNED_BUF_ADDR_MASK;
}
static inline u64 xp_unaligned_extract_offset(u64 addr)
{
return addr >> XSK_UNALIGNED_BUF_OFFSET_SHIFT;
}
static inline u64 xp_unaligned_add_offset_to_addr(u64 addr)
{
return xp_unaligned_extract_addr(addr) +
xp_unaligned_extract_offset(addr);
}
#endif /* XSK_BUFF_POOL_H_ */