UM0851 datasheet(49/245 Pages) STMICROELECTRONICS

49 / 245 page

UM0851

Communication device drivers

Doc ID 16604 Rev 2

49/245

Note:

The maximum MTU size for SPEAr600 is 4000 and for SPEAr300 is 9000.

●

void (*set_multicast_list)(struct net_device *dev);

Method called when the multicast list for the device changes and when the flags

change.

4.1.4

Concept of socket buffers

Each packet handled by the kernel is contained in a socket buffer structure (struct sk_buff),

whose definition is found in <linux/skbuff.h>. The structure gets its name from the Unix

abstraction used to represent a network connection, the socket. Even if the interface has

nothing to do with sockets, each network packet belongs to a socket in the higher network

layers, and the input/output buffers of any socket are lists of struct sk_buff. The same struct

sk_buff is used to host network data throughout all the Linux network subsystems, but a

socket buffer is just a packet as far as the interface is concerned. A pointer to struct sk_buff

is usually called skb. This practice is used both in the sample code and in the text.

The skb buffers used for the reception or transmission must guarantee cache coherency.

These buffers are allocated in the cached memory regions and therefore there is a

possibility that the memory data is not in synch with cache. The driver uses the following

calls for cache coherency:

dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,enum

dma_data_direction dir);

/* Depending upon the directions of the data transfer the above function either

invalidates or clean the cache contents.

If the argument dir in the above function is set as DMA_FROM_DEVICE, this argument

may do nothing in the above function but invalidates the cache when used in

dma_unmap_single. If the argument dir is set to DMA_TO_DEVICE, it cleans the cache.

Cleaning a cache reestablishes coherence between the cached memory and the main

memory */

void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,enum

dma_data_direction dir);

/* The call to the above function is made when the packet has been received, to see

if a mapped address was really a "safe" buffer and if so, copy the data from the safe

buffer back to the unsafe buffer and free up the safe buffer.

Packet reception

Receiving data from the network is trickier than transmitting it, because a struct sk_buff

must be allocated and handed off to the upper layers within an atomic context. The mode of

packet reception that has been implemented is interrupt driven.

There is a common interrupt registered for both reception and transmission. When the

packet is received a tasklet is scheduled for handing the packet to the upper stacks, as the

packet has been fetched by DMA into the memory buffers. The scheduled tasklet executes

not later than the next timer tick. This scheduling allows to handle more packets in a more

efficient way.

One important thing to note over here is that since the packet handling to the stack is done

through tasklet, it keeps the DMA descriptor occupied till the packet is handed over to the

stack. In case of heavy Ethernet traffic at high speeds, the number of DMA descriptors

UM0851 数据表(PDF) 49 Page - STMicroelectronics

UM0851 数据表(HTML) 49 Page - STMicroelectronics

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 ...More

数据表下载