e2forlife / psoc-w5100-driver Goto Github PK

Add a flush for the receiver (and possibly transmitter) buffers, so that there is a quick way to clear out waiting data that is pending in the W5100.

First of all: really awesome work on this library!

I'm using this library to send data from PSoC 4 board to a remote server using TCP(HTTP specifically) protocol. One problem I encountered is: modern HTTP servers will close the connection as soon as the data is sent to the client. Those client may still be in the W5100 shield buffer(hence W5100_SocketRxDataWaiting will give a non-negative result). However, due to this line, the driver will assume the socket is closed, and we cannot get the data returned by the server, even if we already have the data in local buffer.

So, is there any way we can have a TcpReceive function without this restriction? Or can we make that an option to TcpReceive? Thanks!

When waiting for SPI_DONE to be set on the PSoC 5 (SPIM mode) the driver deadlocks.

The READ_WRITE_DELAY causes unnecessary delays after transmitting packets between the PSoC and the W5100.

PSoC5 and PSoC 5LP support DMA transfer, it is desired to use DMA to send data between the PSoC and W5100.

See: https://github.com/lrds/lwip-psoc5, which uses DMA for transfer to MAC.

Pavloven ( Cypress Forum)
For testing, I used Hercules SETUP utility http://www.hw-group.com/products/hercules/index_en.html I do:

1. send a test string and see ETH0_SocketRxDataWaiting(socket) > 0
2. execute commands ETH0_TcpReceive(...
3. and check again ETH0_SocketRxDataWaiting(socket)

only if I install a flag = 1, in W5100.c as here:
str. 1394 $INSTANCE_NAME_ProcessRxData( socket, 0, buffer, RxSize, 1);

i see ETH0_SocketRxDataWaiting(socket) = 0

The v1.0 driver component code does not compile for PSoC3 due to the use of the inline keyword.

When waiting for connection using a BUS_CLK of 64 MHz on the PSoC 5 it was reported that while( !XXX_TcpConnected(socket) ) deadlocks without having the "High Speed Mode" enabled on the SPIM component.