Improve performance of TSE on Stratix IV GX

Hello,

I'm using the Standard design of Stratix IV on which I put the Nios II Simple Socket Server in order to receive incoming data from my other computer.

I managed to establish a TCP/IP connection between my client and the FPGA server.

My client application send 40 bytes data with the socket send() command.

Everytime my Nios II receive a packet, he aknowledges it by sending back a packet to the client.

Everything is fine except the performance of my Nios II which is about 70kbytes/s.

I've read the "No free buffers for rx" from kaushal which helped me to instantiate the tighlty coupled memory and I modified the ipport.h file according to the help note of the Ethernet_accel_design.

With the AN440 help, I also included the optimization.c and .h file and my ipport.h looks like this:

#define ALTERA_MRAM_FOR_PACKETS 1

# define ALTERA_MRAM_ALLOC_BASE PACKET_MEMORY_BASE

# define ALTERA_MRAM_ALLOC_SPAN PACKET_MEMORY_SPAN

# ifdef ALTERA_TRIPLE_SPEED_MAC

char * ncpalloc(unsigned size);

void ncpfree(void *ptr);

char * ncpballoc(unsigned size);

int ncpbfree(char *ptr);

# if ALTERA_MRAM_FOR_PACKETS

//#include "../net/optimizations.h"

# define BB_ALLOC(size) ncpballoc(size)

# define BB_FREE(ptr) ncpbfree(ptr)

# define LB_ALLOC(size) ncpballoc(size)

# define LB_FREE(ptr) ncpbfree(ptr)

# else

# define BB_ALLOC(size) ncpalloc(size) /* Big packet buffer alloc */

# define BB_FREE(ptr) ncpfree(ptr)

# define LB_ALLOC(size) ncpalloc(size) /* Little packet buffer alloc */

# define LB_FREE(ptr) ncpfree(ptr)

# endif

# else /* Not ALTERA_TRIPLE_SPEED_MAC */

# define BB_ALLOC(size) npalloc(size) /* Big packet buffer alloc */

# define BB_FREE(ptr) npfree(ptr)

# define LB_ALLOC(size) npalloc(size) /* Little packet buffer alloc */

# define LB_FREE(ptr) npfree(ptr)

# endif /* ALTERA_TRIPLE_SPEED_MAC */

and also:

#if ALTERA_MRAM_FOR_PACKETS

# define NUMBIGBUFS 30

# define NUMLILBUFS 30

/* some maximum packet buffer numbers */

# define MAXBIGPKTS 30

# define MAXLILPKTS 30

# define MAXPACKETS (MAXLILPKTS+MAXBIGPKTS)

# define BIGBUFSIZE 1536

# define LILBUFSIZE 128

Please see my .qsys file and my ipport.h file attached.

Except the Hardware checksum calculation, is there is anything I could try in order to improve the Simple Socket Server performance to at least 1Mbytes/s?

Do you think that putting my .stack and .heap directly into a 300kbytes onchip memory could help me having better results?

Thank you in advance

With regards,

Michel