}
static void efx_unmap_rx_buffer(struct efx_nic *efx,
- struct efx_rx_buffer *rx_buf)
+ struct efx_rx_buffer *rx_buf,
+ unsigned int used_len)
{
if (rx_buf->is_page && rx_buf->u.page) {
struct efx_rx_page_state *state;
state->dma_addr,
efx_rx_buf_size(efx),
PCI_DMA_FROMDEVICE);
+ } else if (used_len) {
+ dma_sync_single_for_cpu(&efx->pci_dev->dev,
+ rx_buf->dma_addr, used_len,
+ DMA_FROM_DEVICE);
}
} else if (!rx_buf->is_page && rx_buf->u.skb) {
pci_unmap_single(efx->pci_dev, rx_buf->dma_addr,
static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
struct efx_rx_buffer *rx_buf)
{
- efx_unmap_rx_buffer(rx_queue->efx, rx_buf);
+ efx_unmap_rx_buffer(rx_queue->efx, rx_buf, 0);
efx_free_rx_buffer(rx_queue->efx, rx_buf);
}
goto out;
}
- /* Release card resources - assumes all RX buffers consumed in-order
- * per RX queue
+ /* Release and/or sync DMA mapping - assumes all RX buffers
+ * consumed in-order per RX queue
*/
- efx_unmap_rx_buffer(efx, rx_buf);
+ efx_unmap_rx_buffer(efx, rx_buf, len);
/* Prefetch nice and early so data will (hopefully) be in cache by
* the time we look at it.