| Message ID | 20094a678ea3d76fc1b8817ae0dd6d136cdc3860.1578225300.git.lorenzo@kernel.org |
|---|---|
| State | RFC |
| Delegated to: | David Miller |
| Headers | show |
| Series | [RFC/RFT,net-next] net: socionext: get rid of huge dma sync in netsec_alloc_rx_data | expand |
Hi Lorenzo, Although the box using thei NIC usually runs with coherent DMA, there's a configuration that disables that. So having this is has some meaning. Minor comments below. On Sun, Jan 05, 2020 at 12:57:56PM +0100, Lorenzo Bianconi wrote: > Socionext driver can run on dma coherent and non-coherent devices. > Get rid of huge dma_sync_single_for_device in netsec_alloc_rx_data since > now the driver can let page_pool API to managed needed DMA sync > > Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> > --- > drivers/net/ethernet/socionext/netsec.c | 60 ++++++++++++++----------- > 1 file changed, 33 insertions(+), 27 deletions(-) > > diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c > index b5a9e947a4a8..7a2eb0e71d2a 100644 > --- a/drivers/net/ethernet/socionext/netsec.c > +++ b/drivers/net/ethernet/socionext/netsec.c > @@ -243,6 +243,7 @@ > NET_IP_ALIGN) > #define NETSEC_RX_BUF_NON_DATA (NETSEC_RXBUF_HEADROOM + \ > SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) > +#define NETSEC_RX_BUF_SIZE (PAGE_SIZE - NETSEC_RX_BUF_NON_DATA) > > #define DESC_SZ sizeof(struct netsec_de) > > @@ -714,12 +715,11 @@ static void netsec_process_tx(struct netsec_priv *priv) > } > > static void *netsec_alloc_rx_data(struct netsec_priv *priv, > - dma_addr_t *dma_handle, u16 *desc_len) i'd prefer having this function fill in the size, insetad of defining it every time we refill the descriptors You can keep the new define for PAGE_SIZE - NETSEC_RX_BUF_NON_DATA, it looks cleaner > + dma_addr_t *dma_handle) > > { > > struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; > - enum dma_data_direction dma_dir; > struct page *page; > > page = page_pool_dev_alloc_pages(dring->page_pool); > @@ -734,10 +734,6 @@ static void *netsec_alloc_rx_data(struct netsec_priv *priv, > /* Make sure the incoming payload fits in the page for XDP and non-XDP > * cases and reserve enough space for headroom + skb_shared_info > */ > - *desc_len = PAGE_SIZE - NETSEC_RX_BUF_NON_DATA; > - dma_dir = page_pool_get_dma_dir(dring->page_pool); > - dma_sync_single_for_device(priv->dev, *dma_handle, *desc_len, dma_dir); > - > return page_address(page); > } > > @@ -883,6 +879,7 @@ static u32 netsec_xdp_xmit_back(struct netsec_priv *priv, struct xdp_buff *xdp) > static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, > struct xdp_buff *xdp) > { > + struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; > u32 ret = NETSEC_XDP_PASS; > int err; > u32 act; > @@ -896,7 +893,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, > case XDP_TX: > ret = netsec_xdp_xmit_back(priv, xdp); > if (ret != NETSEC_XDP_TX) > - xdp_return_buff(xdp); > + __page_pool_put_page(dring->page_pool, > + virt_to_head_page(xdp->data), > + xdp->data_end - xdp->data_hard_start, > + true); > break; > case XDP_REDIRECT: > err = xdp_do_redirect(priv->ndev, xdp, prog); > @@ -904,7 +904,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, > ret = NETSEC_XDP_REDIR; > } else { > ret = NETSEC_XDP_CONSUMED; > - xdp_return_buff(xdp); > + __page_pool_put_page(dring->page_pool, > + virt_to_head_page(xdp->data), > + xdp->data_end - xdp->data_hard_start, > + true); > } > break; > default: > @@ -915,7 +918,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, > /* fall through -- handle aborts by dropping packet */ > case XDP_DROP: > ret = NETSEC_XDP_CONSUMED; > - xdp_return_buff(xdp); > + __page_pool_put_page(dring->page_pool, > + virt_to_head_page(xdp->data), > + xdp->data_end - xdp->data_hard_start, > + true); > break; > } > > @@ -944,10 +950,10 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) > struct netsec_desc *desc = &dring->desc[idx]; > struct page *page = virt_to_page(desc->addr); > u32 xdp_result = XDP_PASS; > - u16 pkt_len, desc_len; > dma_addr_t dma_handle; > struct xdp_buff xdp; > void *buf_addr; > + u16 pkt_len; > > if (de->attr & (1U << NETSEC_RX_PKT_OWN_FIELD)) { > /* reading the register clears the irq */ > @@ -982,8 +988,7 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) > /* allocate a fresh buffer and map it to the hardware. > * This will eventually replace the old buffer in the hardware > */ > - buf_addr = netsec_alloc_rx_data(priv, &dma_handle, &desc_len); > - > + buf_addr = netsec_alloc_rx_data(priv, &dma_handle); > if (unlikely(!buf_addr)) > break; > > @@ -1014,7 +1019,8 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) > * cache state. Since we paid the allocation cost if > * building an skb fails try to put the page into cache > */ > - page_pool_recycle_direct(dring->page_pool, page); > + __page_pool_put_page(dring->page_pool, page, > + desc->len, true); > netif_err(priv, drv, priv->ndev, > "rx failed to build skb\n"); > break; > @@ -1037,7 +1043,7 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) > } > > /* Update the descriptor with fresh buffers */ > - desc->len = desc_len; > + desc->len = NETSEC_RX_BUF_SIZE; Similar comment here, i'd prefer having a sinlge fucntion calculate the length. > desc->dma_addr = dma_handle; > desc->addr = buf_addr; > > @@ -1272,17 +1278,19 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv) > { > struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; > struct bpf_prog *xdp_prog = READ_ONCE(priv->xdp_prog); > - struct page_pool_params pp_params = { 0 }; > + struct page_pool_params pp_params = { > + .order = 0, > + /* internal DMA mapping in page_pool */ > + .flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV, > + .pool_size = DESC_NUM, > + .nid = NUMA_NO_NODE, > + .dev = priv->dev, > + .dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE, > + .offset = NETSEC_RXBUF_HEADROOM, > + .max_len = NETSEC_RX_BUF_SIZE, > + }; > int i, err; > > - pp_params.order = 0; > - /* internal DMA mapping in page_pool */ > - pp_params.flags = PP_FLAG_DMA_MAP; > - pp_params.pool_size = DESC_NUM; > - pp_params.nid = NUMA_NO_NODE; > - pp_params.dev = priv->dev; > - pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; > - > dring->page_pool = page_pool_create(&pp_params); > if (IS_ERR(dring->page_pool)) { > err = PTR_ERR(dring->page_pool); > @@ -1303,17 +1311,15 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv) > struct netsec_desc *desc = &dring->desc[i]; > dma_addr_t dma_handle; > void *buf; > - u16 len; > - > - buf = netsec_alloc_rx_data(priv, &dma_handle, &len); > > + buf = netsec_alloc_rx_data(priv, &dma_handle); > if (!buf) { > err = -ENOMEM; > goto err_out; > } > + desc->len = NETSEC_RX_BUF_SIZE; > desc->dma_addr = dma_handle; > desc->addr = buf; > - desc->len = len; > } > > netsec_rx_fill(priv, 0, DESC_NUM); > -- > 2.21.1 > Other than that this looks good, re-send it as a non RFC patch and i can test it Thanks! /Ilias
diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c index b5a9e947a4a8..7a2eb0e71d2a 100644 --- a/drivers/net/ethernet/socionext/netsec.c +++ b/drivers/net/ethernet/socionext/netsec.c @@ -243,6 +243,7 @@ NET_IP_ALIGN) #define NETSEC_RX_BUF_NON_DATA (NETSEC_RXBUF_HEADROOM + \ SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) +#define NETSEC_RX_BUF_SIZE (PAGE_SIZE - NETSEC_RX_BUF_NON_DATA) #define DESC_SZ sizeof(struct netsec_de) @@ -714,12 +715,11 @@ static void netsec_process_tx(struct netsec_priv *priv) } static void *netsec_alloc_rx_data(struct netsec_priv *priv, - dma_addr_t *dma_handle, u16 *desc_len) + dma_addr_t *dma_handle) { struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; - enum dma_data_direction dma_dir; struct page *page; page = page_pool_dev_alloc_pages(dring->page_pool); @@ -734,10 +734,6 @@ static void *netsec_alloc_rx_data(struct netsec_priv *priv, /* Make sure the incoming payload fits in the page for XDP and non-XDP * cases and reserve enough space for headroom + skb_shared_info */ - *desc_len = PAGE_SIZE - NETSEC_RX_BUF_NON_DATA; - dma_dir = page_pool_get_dma_dir(dring->page_pool); - dma_sync_single_for_device(priv->dev, *dma_handle, *desc_len, dma_dir); - return page_address(page); } @@ -883,6 +879,7 @@ static u32 netsec_xdp_xmit_back(struct netsec_priv *priv, struct xdp_buff *xdp) static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, struct xdp_buff *xdp) { + struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; u32 ret = NETSEC_XDP_PASS; int err; u32 act; @@ -896,7 +893,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, case XDP_TX: ret = netsec_xdp_xmit_back(priv, xdp); if (ret != NETSEC_XDP_TX) - xdp_return_buff(xdp); + __page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), + xdp->data_end - xdp->data_hard_start, + true); break; case XDP_REDIRECT: err = xdp_do_redirect(priv->ndev, xdp, prog); @@ -904,7 +904,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, ret = NETSEC_XDP_REDIR; } else { ret = NETSEC_XDP_CONSUMED; - xdp_return_buff(xdp); + __page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), + xdp->data_end - xdp->data_hard_start, + true); } break; default: @@ -915,7 +918,10 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, /* fall through -- handle aborts by dropping packet */ case XDP_DROP: ret = NETSEC_XDP_CONSUMED; - xdp_return_buff(xdp); + __page_pool_put_page(dring->page_pool, + virt_to_head_page(xdp->data), + xdp->data_end - xdp->data_hard_start, + true); break; } @@ -944,10 +950,10 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) struct netsec_desc *desc = &dring->desc[idx]; struct page *page = virt_to_page(desc->addr); u32 xdp_result = XDP_PASS; - u16 pkt_len, desc_len; dma_addr_t dma_handle; struct xdp_buff xdp; void *buf_addr; + u16 pkt_len; if (de->attr & (1U << NETSEC_RX_PKT_OWN_FIELD)) { /* reading the register clears the irq */ @@ -982,8 +988,7 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) /* allocate a fresh buffer and map it to the hardware. * This will eventually replace the old buffer in the hardware */ - buf_addr = netsec_alloc_rx_data(priv, &dma_handle, &desc_len); - + buf_addr = netsec_alloc_rx_data(priv, &dma_handle); if (unlikely(!buf_addr)) break; @@ -1014,7 +1019,8 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) * cache state. Since we paid the allocation cost if * building an skb fails try to put the page into cache */ - page_pool_recycle_direct(dring->page_pool, page); + __page_pool_put_page(dring->page_pool, page, + desc->len, true); netif_err(priv, drv, priv->ndev, "rx failed to build skb\n"); break; @@ -1037,7 +1043,7 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) } /* Update the descriptor with fresh buffers */ - desc->len = desc_len; + desc->len = NETSEC_RX_BUF_SIZE; desc->dma_addr = dma_handle; desc->addr = buf_addr; @@ -1272,17 +1278,19 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv) { struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; struct bpf_prog *xdp_prog = READ_ONCE(priv->xdp_prog); - struct page_pool_params pp_params = { 0 }; + struct page_pool_params pp_params = { + .order = 0, + /* internal DMA mapping in page_pool */ + .flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV, + .pool_size = DESC_NUM, + .nid = NUMA_NO_NODE, + .dev = priv->dev, + .dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE, + .offset = NETSEC_RXBUF_HEADROOM, + .max_len = NETSEC_RX_BUF_SIZE, + }; int i, err; - pp_params.order = 0; - /* internal DMA mapping in page_pool */ - pp_params.flags = PP_FLAG_DMA_MAP; - pp_params.pool_size = DESC_NUM; - pp_params.nid = NUMA_NO_NODE; - pp_params.dev = priv->dev; - pp_params.dma_dir = xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; - dring->page_pool = page_pool_create(&pp_params); if (IS_ERR(dring->page_pool)) { err = PTR_ERR(dring->page_pool); @@ -1303,17 +1311,15 @@ static int netsec_setup_rx_dring(struct netsec_priv *priv) struct netsec_desc *desc = &dring->desc[i]; dma_addr_t dma_handle; void *buf; - u16 len; - - buf = netsec_alloc_rx_data(priv, &dma_handle, &len); + buf = netsec_alloc_rx_data(priv, &dma_handle); if (!buf) { err = -ENOMEM; goto err_out; } + desc->len = NETSEC_RX_BUF_SIZE; desc->dma_addr = dma_handle; desc->addr = buf; - desc->len = len; } netsec_rx_fill(priv, 0, DESC_NUM);
Socionext driver can run on dma coherent and non-coherent devices. Get rid of huge dma_sync_single_for_device in netsec_alloc_rx_data since now the driver can let page_pool API to managed needed DMA sync Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> --- drivers/net/ethernet/socionext/netsec.c | 60 ++++++++++++++----------- 1 file changed, 33 insertions(+), 27 deletions(-)