This is MAC module routine. It initializes ENC28J60 controller. This function is internaly splited into 2 parts to help linker when coming short of memory.

• receive buffer start address : 0x0000.
• receive buffer end address : 0x19AD.
• transmit buffer start address: 0x19AE.
• transmit buffer end address : 0x1FFF.
• RAM buffer read/write pointers in auto-increment mode.
• receive filters set to default: CRC + MAC Unicast + MAC Broadcast in OR mode.
• flow control with TX and RX pause frames in full duplex mode.
• frames are padded to 60 bytes + CRC.
• maximum packet size is set to 1518.
• Back-to-Back Inter-Packet Gap: 0x15 in full duplex mode; 0x12 in half duplex mode.
• Non-Back-to-Back Inter-Packet Gap: 0x0012 in full duplex mode; 0x0C12 in half duplex mode.
• Collision window is set to 63 in half duplex mode to accomodate some ENC28J60 revisions silicon bugs.
• CLKOUT output is disabled to reduce EMI generation.
• half duplex loopback disabled.
• LED configuration: default (LEDA-link status, LEDB-link activity).

• mac: RAM buffer containing valid MAC address.
• ip: RAM buffer containing valid IP address.
• fullDuplex: ethernet duplex mode switch. Valid values: 0 (half duplex mode) and 1 (full duplex mode).

Nothing.

External dependencies of the library from the top of the page must be defined before using this function.
The SPI module needs to be initialized. See the SPIx_Init and SPIx_Init_Advanced routines.

None.

This is MAC module routine. This routine enables appropriate network traffic on the ENC28J60 module by the means of it's receive filters (unicast, multicast, broadcast, crc). Specific type of network traffic will be enabled if a corresponding bit of this routine's input parameter is set. Therefore, more than one type of network traffic can be enabled at the same time. For this purpose, predefined library constants (see the table below) can be ORed to form appropriate input value.

Advanced filtering available in the ENC28J60 module such as Pattern Match, Magic Packet and Hash Table can not be enabled by this routine. Additionaly, all filters, except CRC, enabled with this routine will work in OR mode, which means that packet will be received if any of the enabled filters accepts it.

This routine will change receive filter configuration on-the-fly. It will not, in any way, mess with enabling/disabling receive/transmit logic or any other part of the ENC28J60 module. The ENC28J60 module should be properly cofigured by the means of SPI_Ethernet_Init routine.

• enFlt: network traffic/receive filter flags. Each bit corresponds to the appropriate network traffic/receive filter:

Bit	Mask	Description	Predefined library const
0	0x01	MAC Broadcast traffic/receive filter flag. When set, MAC broadcast traffic will be enabled.	_SPI_Ethernet_BROADCAST
1	0x02	MAC Multicast traffic/receive filter flag. When set, MAC multicast traffic will be enabled.	_SPI_Ethernet_MULTICAST
2	0x04	not used	none
3	0x08	not used	none
4	0x10	not used	none
5	0x20	CRC check flag. When set, packets with invalid CRC field will be discarded.	_SPI_Ethernet_CRC
6	0x40	not used	none
7	0x80	MAC Unicast traffic/receive filter flag. When set, MAC unicast traffic will be enabled.	_SPI_Ethernet_UNICAST

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

SPI_Ethernet_Enable(_SPI_Ethernet_CRC | _SPI_Ethernet_UNICAST); // enable CRC checking and Unicast traffic

Advanced filtering available in the ENC28J60 module such as Pattern Match, Magic Packet and Hash Table can not be enabled by this routine. Additionaly, all filters, except CRC, enabled with this routine will work in OR mode, which means that packet will be received if any of the enabled filters accepts it.

This routine will change receive filter configuration on-the-fly. It will not, in any way, mess with enabling/disabling receive/transmit logic or any other part of the ENC28J60 module. The ENC28J60 module should be properly cofigured by the means of SPI_Ethernet_Init routine.

This is MAC module routine. This routine disables appropriate network traffic on the ENC28J60 module by the means of it's receive filters (unicast, multicast, broadcast, crc). Specific type of network traffic will be disabled if a corresponding bit of this routine's input parameter is set. Therefore, more than one type of network traffic can be disabled at the same time. For this purpose, predefined library constants (see the table below) can be ORed to form appropriate input value.

• disFlt: network traffic/receive filter flags. Each bit corresponds to the appropriate network traffic/receive filter:

Bit	Mask	Description	Predefined library const
0	0x01	MAC Broadcast traffic/receive filter flag. When set, MAC broadcast traffic will be disabled.	_SPI_Ethernet_BROADCAST
1	0x02	MAC Multicast traffic/receive filter flag. When set, MAC multicast traffic will be disabled.	_SPI_Ethernet_MULTICAST
2	0x04	not used	none
3	0x08	not used	none
4	0x10	not used	none
5	0x20	CRC check flag. When set, CRC check will be disabled and packets with invalid CRC field will be accepted.	_SPI_Ethernet_CRC
6	0x40	not used	none
7	0x80	MAC Unicast traffic/receive filter flag. When set, MAC unicast traffic will be disabled.	_SPI_Ethernet_UNICAST

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

SPI_Ethernet_Disable(_SPI_Ethernet_CRC | _SPI_Ethernet_UNICAST); // disable CRC checking and Unicast traffic

Advanced filtering available in the ENC28J60 module such as Pattern Match, Magic Packet and Hash Table can not be disabled by this routine.

This routine will change receive filter configuration on-the-fly. It will not, in any way, mess with enabling/disabling receive/transmit logic or any other part of the ENC28J60 module. The ENC28J60 module should be properly cofigured by the means of SPI_Ethernet_Init routine.

This is MAC module routine. It processes next received packet if such exists. Packets are processed in the following manner:

• ARP & ICMP requests are replied automatically.
• upon TCP request the SPI_Ethernet_UserTCP function is called for further processing.
• upon UDP request the SPI_Ethernet_UserUDP function is called for further processing.

None.

• 0 - upon successful packet processing (zero packets received or received packet processed successfully).
• 1 - upon reception error or receive buffer corruption. ENC28J60 controller needs to be restarted.
• 2 - received packet was not sent to us (not our IP, nor IP broadcast address).
• 3 - received IP packet was not IPv4.
• 4 - received packet was of type unknown to the library.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

if (SPI_Ethernet_doPacket() == 0)(1) {  // process received packets
  ...
}

SPI_Ethernet_doPacket must be called as often as possible in user's code.

This is MAC module routine. It stores one byte to address pointed by the current ENC28J60 write pointer (EWRPT).

• v: value to store

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char data_;
...
SPI_Ethernet_putByte(data); // put an byte into ENC28J60 buffer

None.

This is MAC module routine. It stores requested number of bytes into ENC28J60 RAM starting from current ENC28J60 write pointer (EWRPT) location.

• ptr: RAM buffer containing bytes to be written into ENC28J60 RAM.
• n: number of bytes to be written.

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char *buffer =  "mikroElektronika";	 
...
SPI_Ethernet_putBytes(buffer, 16); // put an RAM array into ENC28J60 buffer

None.

This is MAC module routine. It stores requested number of const bytes into ENC28J60 RAM starting from current ENC28J60 write pointer (EWRPT) location.

• ptr: const buffer containing bytes to be written into ENC28J60 RAM.
• n: number of bytes to be written.

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

const char *buffer =  "mikroElektronika";	 
...
SPI_Ethernet_putConstBytes(buffer, 16); // put a const array into ENC28J60 buffer

None.

This is MAC module routine. It stores whole string (excluding null termination) into ENC28J60 RAM starting from current ENC28J60 write pointer (EWRPT) location.

• ptr: string to be written into ENC28J60 RAM.

Number of bytes written into ENC28J60 RAM.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char *buffer =  "mikroElektronika";	 
...
SPI_Ethernet_putString(buffer); // put a RAM string into ENC28J60 buffer

None.

This is MAC module routine. It stores whole const string (excluding null termination) into ENC28J60 RAM starting from current ENC28J60 write pointer (EWRPT) location.

• ptr: const string to be written into ENC28J60 RAM.

Number of bytes written into ENC28J60 RAM.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

const char *buffer =  "mikroElektronika"; 
...
SPI_Ethernet_putConstString(buffer); // put a const string into ENC28J60 buffer

None.

This is MAC module routine. It fetches a byte from address pointed to by current ENC28J60 read pointer (ERDPT).

None.

Byte read from ENC28J60 RAM.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char buffer; 
...
buffer = SPI_Ethernet_getByte(); // read a byte from ENC28J60 buffer

None.

This is MAC module routine. It fetches equested number of bytes from ENC28J60 RAM starting from given address. If value of 0xFFFF is passed as the address parameter, the reading will start from current ENC28J60 read pointer (ERDPT) location.

• ptr: buffer for storing bytes read from ENC28J60 RAM.
• addr: ENC28J60 RAM start address. Valid values: 0..8192.
• n: number of bytes to be read.

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char buffer[16];	 
...
SPI_Ethernet_getBytes(buffer, 0x100, 16); // read 16 bytes, starting from address 0x100

None.

This is TCP module routine. It is internally called by the library. The user accesses to the TCP request by using some of the SPI_Ethernet_get routines. The user puts data in the transmit buffer by using some of the SPI_Ethernet_put routines. The function must return the length in bytes of the TCP reply, or 0 if there is nothing to transmit. If there is no need to reply to the TCP requests, just define this function with return(0) as a single statement.

• remoteHost: client's IP address.
• remotePort: client's TCP port.
• localPort: port to which the request is sent.
• reqLength: TCP request data field length.
• flags: structure consisted of two bit fields :
  Copy Code To Clipboard

  typedef struct {
    unsigned canCloseTCP: 1;  // flag which closes socket
    unsigned isBroadcast: 1;  // flag which denotes that the IP package has been received via subnet broadcast address
  } TEthPktFlags;
  

• 0 - there should not be a reply to the request.
• Length of TCP reply data field - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

This function is internally called by the library and should not be called by the user's code.

The function source code is provided with appropriate example projects. The code should be adjusted by the user to achieve desired reply.

This is UDP module routine. It is internally called by the library. The user accesses to the UDP request by using some of the SPI_Ethernet_get routines. The user puts data in the transmit buffer by using some of the SPI_Ethernet_put routines. The function must return the length in bytes of the UDP reply, or 0 if nothing to transmit. If you don't need to reply to the UDP requests, just define this function with a return(0) as single statement.

• remoteHost: client's IP address.
• remotePort: client's port.
• localPort: port to which the request is sent.
• reqLength: UDP request data field length.
• flags: structure consisted of two bit fields :
  Copy Code To Clipboard

  typedef struct {
    unsigned canCloseTCP: 1;  // flag which closes TCP socket (not relevant to UDP)
    unsigned isBroadcast: 1;  // flag which denotes that the IP package has been received via subnet broadcast address
  } TEthPktFlags;
  

• 0 - there should not be a reply to the request.
• Length of UDP reply data field - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

This function is internally called by the library and should not be called by the user's code.

The function source code is provided with appropriate example projects. The code should be adjusted by the user to achieve desired reply.

This routine should be used when DHCP server is present on the network to fetch assigned IP address.

None.

Pointer to the global variable holding IP address.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char ipAddr[4];  // user IP address buffer
...	
memcpy(ipAddr, SPI_Ethernet_getIpAddress(), 4); // fetch IP address

User should always copy the IP address from the RAM location returned by this routine into it's own IP address buffer. These locations should not be altered by the user in any case!

This routine should be used when DHCP server is present on the network to fetch assigned gateway IP address.

None.

Pointer to the global variable holding gateway IP address.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char gwIpAddr[4];  // user gateway IP address buffer
...	
memcpy(gwIpAddr, SPI_Ethernet_getGwIpAddress(), 4); // fetch gateway IP address 

User should always copy the IP address from the RAM location returned by this routine into it's own gateway IP address buffer. These locations should not be altered by the user in any case!

This routine should be used when DHCP server is present on the network to fetch assigned DNS IP address.

None.

Pointer to the global variable holding DNS IP address.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char dnsIpAddr[4];  // user DNS IP address buffer
...	
memcpy(dnsIpAddr, SPI_Ethernet_getDnsIpAddress(), 4); // fetch DNS server address 

User should always copy the IP address from the RAM location returned by this routine into it's own DNS IP address buffer. These locations should not be altered by the user in any case!

This routine should be used when DHCP server is present on the network to fetch assigned IP subnet mask.

None.

Pointer to the global variable holding IP subnet mask.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char IpMask[4];  // user IP subnet mask buffer
...	
memcpy(IpMask, SPI_Ethernet_getIpMask(), 4); // fetch IP subnet mask

User should always copy the IP address from the RAM location returned by this routine into it's own IP subnet mask buffer. These locations should not be altered by the user in any case!

Configures network parameters (IP subnet mask, gateway IP address, DNS IP address) when DHCP is not used.

• ipMask: IP subnet mask.
• gwIpAddr gateway IP address.
• dnsIpAddr: DNS IP address.

Nothing.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

char ipMask[4]    = {255, 255, 255,  0 };  // network mask (for example : 255.255.255.0)
char gwIpAddr[4]  = {192, 168,   1,  1 };  // gateway (router) IP address
char dnsIpAddr[4] = {192, 168,   1,  1 };  // DNS server IP address
...	
SPI_Ethernet_confNetwork(ipMask, gwIpAddr, dnsIpAddr); // set network configuration parameters

The above mentioned network parameters should be set by this routine only if DHCP module is not used. Otherwise DHCP will override these settings.

This is ARP module routine. It sends an ARP request for given IP address and waits for ARP reply. If the requested IP address was resolved, an ARP cash entry is used for storing the configuration. ARP cash can store up to 3 entries. For ARP cash structure refer to "eth_enc28j60LibDef.h" header file in the compiler's Uses folder.

• ip: IP address to be resolved.
• tmax: time in seconds to wait for an reply.

• MAC address behind the IP address - the requested IP address was resolved.
• 0 - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char IpAddr[4]  = {192, 168,   1,  1 };  // IP address
...	
SPI_Ethernet_arpResolve(IpAddr, 5); // get MAC address behind the above IP address, wait 5 secs for the response

The Ethernet services are not stopped while this routine waits for ARP reply. The incoming packets will be processed normaly during this time.

This is UDP module routine. It sends an UDP packet on the network.

• destIP: remote host IP address.
• sourcePort: local UDP source port number.
• destPort: destination UDP port number.
• pkt: packet to transmit.
• pktLen: length in bytes of packet to transmit.

• 1 - UDP packet was sent successfully.
• 0 - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char IpAddr[4]  = {192, 168,   1,  1 };  // remote IP address
...	
SPI_Ethernet_sendUDP(IpAddr, 10001, 10001, "Hello", 5); // send Hello message to the above IP address, from UDP port 10001 to UDP port 10001 

None.

This is DNS module routine. It sends an DNS request for given host name and waits for DNS reply. If the requested host name was resolved, it's IP address is stored in library global variable and a pointer containing this address is returned by the routine. UDP port 53 is used as DNS port.

• host: host name to be resolved.
• tmax: time in seconds to wait for an reply.

• pointer to the location holding the IP address - the requested host name was resolved.
• 0 - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

unsigned char * remoteHostIpAddr[4];	// user host IP address buffer
...
// SNTP server:
// Zurich, Switzerland: Integrated Systems Lab, Swiss Fed. Inst. of Technology
// 129.132.2.21: swisstime.ethz.ch
// Service Area: Switzerland and Europe	
memcpy(remoteHostIpAddr, SPI_Ethernet_dnsResolve("swisstime.ethz.ch", 5), 4);

The Ethernet services are not stopped while this routine waits for DNS reply. The incoming packets will be processed normaly during this time.

User should always copy the IP address from the RAM location returned by this routine into it's own resolved host IP address buffer. These locations should not be altered by the user in any case!

This is DHCP module routine. It sends an DHCP request for network parameters (IP, gateway, DNS addresses and IP subnet mask) and waits for DHCP reply. If the requested parameters were obtained successfully, their values are stored into the library global variables.

These parameters can be fetched by using appropriate library IP get routines:

• SPI_Ethernet_getIpAddress - fetch IP address.
• SPI_Ethernet_getGwIpAddress - fetch gateway IP address.
• SPI_Ethernet_getDnsIpAddress - fetch DNS IP address.
• SPI_Ethernet_getIpMask - fetch IP subnet mask.

UDP port 68 is used as DHCP client port and UDP port 67 is used as DHCP server port.

• tmax: time in seconds to wait for an reply.

• 1 - network parameters were obtained successfully.
• 0 - otherwise.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

...	
SPI_Ethernet_initDHCP(5); // get network configuration from DHCP server, wait 5 sec for the response 
...

The Ethernet services are not stopped while this routine waits for DNS reply. The incoming packets will be processed normaly during this time.

When DHCP module is used, global library variable SPI_Ethernet_userTimerSec is used to keep track of time. It is user responsibility to increment this variable each second in it's code.

This is DHCP module routine. It takes care of IP address lease time by decrementing the global lease time library counter. When this time expires, it's time to contact DHCP server and renew the lease.

None.

• 0 - lease time has not expired yet.
• 1 - lease time has expired, it's time to renew it.

Ethernet module has to be initialized. See SPI_Ethernet_Init.

while(1) {	
  ...
  if (SPI_Ethernet_doDHCPLeaseTime())
    ... // it's time to renew the IP address lease                  
}

None.

This is DHCP module routine. It sends IP address lease time renewal request to DHCP server.

• tmax: time in seconds to wait for an reply.

• 1 - upon success (lease time was renewed).
• 0 - otherwise (renewal request timed out).

Ethernet module has to be initialized. See SPI_Ethernet_Init.

while(1) {	
  ...
  if (SPI_Ethernet_doDHCPLeaseTime())
    SPI_Ethernet_renewDHCP(5);   // it's time to renew the IP address lease, with 5 secs for a reply                  
  ...  
}

None.