I have been trying to work out why my code would not support SPP0 and in particular SPP0_SCK pin which is on P3_0.
After a lot of code research I think that link 65 in libs\Pin.cpp is the issue:
map<int, PinName> pins = {{(0x00<<8)+0,P0_0},{(0x00<<8)+1,P0_1},{(0x01<<8)+0,P1_0},{(0x01<<8)+1,P1_1},{(0x01<<8)+2,P1_2},{(0x01<<8)+3,P1_3},{(0x01<<8)+4,P1_4},{(0x01<<8)+5,P1_5},{(0x01<<8)+6,P1_6},{(0x01<<8)+7,P1_7},{(0x01<<8)+8,P1_8},{(0x01<<8)+9,P1_9},{(0x01<<8)+10,P1_10},{(0x01<<8)+11,P1_11},{(0x01<<8)+12,P1_12},{(0x01<<8)+13,P1_13},{(0x01<<8)+14,P1_14},{(0x01<<8)+15,P1_15},{(0x01<<8)+16,P1_16},{(0x01<<8)+17,P1_17},{(0x01<<8)+18,P1_18},{(0x01<<8)+20,P1_20},{(0x02<<8)+0,P2_0},{(0x02<<8)+1,P2_1},{(0x02<<8)+2,P2_2},{(0x02<<8)+3,P2_3},{(0x02<<8)+4,P2_4},{(0x02<<8)+5,P2_5},{(0x02<<8)+6,P2_6},{(0x02<<8)+7,P2_7},{(0x02<<8)+8,P2_8},{(0x02<<8)+9,P2_9},{(0x02<<8)+10,P2_10},{(0x02<<8)+11,P2_11},{(0x02<<8)+12,P2_12},{(0x02<<8)+13,P2_13},{(0x03<<8)+1,P3_1},{(0x03<<8)+2,P3_2},{(0x03<<8)+4,P3_4},{(0x03<<8)+5,P3_5},{(0x03<<8)+6,P3_6},{(0x03<<8)+7,P3_7},{(0x03<<8)+8,P3_8},{(0x04<<8)+0,P4_0},{(0x04<<8)+1,P4_1},{(0x04<<8)+2,P4_2},{(0x04<<8)+3,P4_3},{(0x04<<8)+4,P4_4},{(0x04<<8)+5,P4_5},{(0x04<<8)+6,P4_6},{(0x04<<8)+8,P4_8},{(0x04<<8)+9,P4_9},{(0x04<<8)+10,P4_10},{(0x05<<8)+0,P5_0},{(0x05<<8)+1,P5_1},{(0x05<<8)+2,P5_2},{(0x05<<8)+3,P5_3},{(0x05<<8)+4,P5_4},{(0x05<<8)+5,P5_5},{(0x05<<8)+6,P5_6},{(0x05<<8)+7,P5_7},{(0x06<<8)+1,P6_1},{(0x06<<8)+2,P6_2},{(0x06<<8)+3,P6_3},{(0x06<<8)+4,P6_4},{(0x06<<8)+5,P6_5},{(0x06<<8)+6,P6_6},{(0x06<<8)+7,P6_7},{(0x06<<8)+8,P6_8},{(0x06<<8)+9,P6_9},{(0x06<<8)+10,P6_10},{(0x06<<8)+11,P6_11},{(0x06<<8)+12,P6_12},{(0x07<<8)+0,P7_0},{(0x07<<8)+1,P7_1},{(0x07<<8)+2,P7_2},{(0x07<<8)+3,P7_3},{(0x07<<8)+4,P7_4},{(0x07<<8)+5,P7_5},{(0x07<<8)+6,P7_6},{(0x07<<8)+7,P7_7},{(0x08<<8)+8,P8_0},{(0x09<<8)+0,P8_1},{(0x09<<8)+1,P8_2},{(0x09<<8)+2,P8_3},{(0x08<<8)+4,P8_4},{(0x08<<8)+5,P8_5},{(0x08<<8)+6,P8_6},{(0x08<<8)+7,P8_7},{(0x09<<8)+0,P9_0},{(0x09<<8)+1,P9_1},{(0x09<<8)+2,P9_2},{(0x09<<8)+3,P9_3},{(0x09<<8)+4,P9_4},{(0x09<<8)+5,P9_5},{(0x09<<8)+6,P9_6},{(0x0A<<8)+1,PA_1},{(0x0A<<8)+2,PA_2},{(0x0A<<8)+3,PA_3},{(0x0A<<8)+4,PA_4},{(0x0B<<8)+0,PB_0},{(0x0B<<8)+1,PB_1},{(0x0B<<8)+2,PB_2},{(0x0B<<8)+3,PB_3},{(0x0B<<8)+4,PB_4},{(0x0B<<8)+5,PB_5},{(0x0B<<8)+6,PB_6},{(0x0C<<8)+1,PC_1},{(0x0C<<8)+2,PC_2},{(0x0C<<8)+3,PC_3},{(0x0C<<8)+4,PC_4},{(0x0C<<8)+5,PC_5},{(0x0C<<8)+6,PC_6},{(0x0C<<8)+7,PC_7},{(0x0C<<8)+8,PC_8},{(0x0C<<8)+9,PC_9},{(0x0C<<8)+10,PC_10},{(0x0C<<8)+11,PC_11},{(0x0C<<8)+12,PC_12},{(0x0C<<8)+13,PC_13},{(0x0C<<8)+14,PC_14},{(0x0D<<8)+0,PD_0},{(0x0D<<8)+1,PD_1},{(0x0D<<8)+2,PD_2},{(0x0D<<8)+3,PD_3},{(0x0D<<8)+4,PD_4},{(0x0D<<8)+5,PD_5},{(0x0D<<8)+6,PD_6},{(0x0D<<8)+7,PD_7},{(0x0D<<8)+8,PD_8},{(0x0D<<8)+9,PD_9},{(0x0D<<8)+10,PD_10},{(0x0D<<8)+11,PD_11},{(0x0D<<8)+12,PD_12},{(0x0D<<8)+13,PD_13},{(0x0D<<8)+14,PD_14},{(0x0D<<8)+15,PD_15},{(0x0D<<8)+16,PD_16},{(0x0E<<8)+0,PE_0},{(0x0E<<8)+1,PE_1},{(0x0E<<8)+2,PE_2},{(0x0E<<8)+3,PE_3},{(0x0E<<8)+4,PE_4},{(0x0E<<8)+5,PE_5},{(0x0E<<8)+6,PE_6},{(0x0E<<8)+7,PE_7},{(0x0E<<8)+8,PE_8},{(0x0E<<8)+9,PE_9},{(0x0E<<8)+10,PE_10},{(0x0E<<8)+11,PE_11},{(0x0E<<8)+12,PE_12},{(0x0E<<8)+13,PE_13},{(0x0E<<8)+14,PE_14},{(0x0E<<8)+15,PE_15},{(0x0F<<8)+1,PF_1},{(0x0F<<8)+2,PF_2},{(0x0F<<8)+3,PF_3},{(0x0F<<8)+5,PF_5},{(0x0F<<8)+6,PF_6},{(0x0F<<8)+7,PF_7},{(0x0F<<8)+8,PF_8},{(0x0F<<8)+9,PF_9},{(0x0F<<8)+10,PF_10},{(0x0F<<8)+11,PF_11}};
I have not systematically compared this with the LPC4330 chip itself, before I embark on this exercise, am I correct in this assumption?
Note: I realise that I should change to a 'channel' approach as used in max31855.cpp, which for the LPC4330 is:
SPI1_SSEL P1_5
SPI1_MOSI P1_4
SPI1_MISO P1_3
SPI1_SCK PF_4
SPI0_SSEL P1_0
SPI0_MOSI P1_2
SPI0_MISO P1_1
SPI0_SCK P3_0
And I think I need a method of changing from 16bit mode to 8 bit mode to support different devices and I probably need to introduce mutexes to ensure they are thread safe for a proper RTOS implementation too.
I am wondering if I should create a wrapper for an SPI object that deals with all these issues so future processor ports will be simpler, not to mention different usage for the same peripheral?
Cheers
Douglas
I have been trying to work out why my code would not support SPP0 and in particular SPP0_SCK pin which is on P3_0.
After a lot of code research I think that link 65 in libs\Pin.cpp is the issue:
map<int, PinName> pins = {{(0x00<<8)+0,P0_0},{(0x00<<8)+1,P0_1},{(0x01<<8)+0,P1_0},{(0x01<<8)+1,P1_1},{(0x01<<8)+2,P1_2},{(0x01<<8)+3,P1_3},{(0x01<<8)+4,P1_4},{(0x01<<8)+5,P1_5},{(0x01<<8)+6,P1_6},{(0x01<<8)+7,P1_7},{(0x01<<8)+8,P1_8},{(0x01<<8)+9,P1_9},{(0x01<<8)+10,P1_10},{(0x01<<8)+11,P1_11},{(0x01<<8)+12,P1_12},{(0x01<<8)+13,P1_13},{(0x01<<8)+14,P1_14},{(0x01<<8)+15,P1_15},{(0x01<<8)+16,P1_16},{(0x01<<8)+17,P1_17},{(0x01<<8)+18,P1_18},{(0x01<<8)+20,P1_20},{(0x02<<8)+0,P2_0},{(0x02<<8)+1,P2_1},{(0x02<<8)+2,P2_2},{(0x02<<8)+3,P2_3},{(0x02<<8)+4,P2_4},{(0x02<<8)+5,P2_5},{(0x02<<8)+6,P2_6},{(0x02<<8)+7,P2_7},{(0x02<<8)+8,P2_8},{(0x02<<8)+9,P2_9},{(0x02<<8)+10,P2_10},{(0x02<<8)+11,P2_11},{(0x02<<8)+12,P2_12},{(0x02<<8)+13,P2_13},{(0x03<<8)+1,P3_1},{(0x03<<8)+2,P3_2},{(0x03<<8)+4,P3_4},{(0x03<<8)+5,P3_5},{(0x03<<8)+6,P3_6},{(0x03<<8)+7,P3_7},{(0x03<<8)+8,P3_8},{(0x04<<8)+0,P4_0},{(0x04<<8)+1,P4_1},{(0x04<<8)+2,P4_2},{(0x04<<8)+3,P4_3},{(0x04<<8)+4,P4_4},{(0x04<<8)+5,P4_5},{(0x04<<8)+6,P4_6},{(0x04<<8)+8,P4_8},{(0x04<<8)+9,P4_9},{(0x04<<8)+10,P4_10},{(0x05<<8)+0,P5_0},{(0x05<<8)+1,P5_1},{(0x05<<8)+2,P5_2},{(0x05<<8)+3,P5_3},{(0x05<<8)+4,P5_4},{(0x05<<8)+5,P5_5},{(0x05<<8)+6,P5_6},{(0x05<<8)+7,P5_7},{(0x06<<8)+1,P6_1},{(0x06<<8)+2,P6_2},{(0x06<<8)+3,P6_3},{(0x06<<8)+4,P6_4},{(0x06<<8)+5,P6_5},{(0x06<<8)+6,P6_6},{(0x06<<8)+7,P6_7},{(0x06<<8)+8,P6_8},{(0x06<<8)+9,P6_9},{(0x06<<8)+10,P6_10},{(0x06<<8)+11,P6_11},{(0x06<<8)+12,P6_12},{(0x07<<8)+0,P7_0},{(0x07<<8)+1,P7_1},{(0x07<<8)+2,P7_2},{(0x07<<8)+3,P7_3},{(0x07<<8)+4,P7_4},{(0x07<<8)+5,P7_5},{(0x07<<8)+6,P7_6},{(0x07<<8)+7,P7_7},{(0x08<<8)+8,P8_0},{(0x09<<8)+0,P8_1},{(0x09<<8)+1,P8_2},{(0x09<<8)+2,P8_3},{(0x08<<8)+4,P8_4},{(0x08<<8)+5,P8_5},{(0x08<<8)+6,P8_6},{(0x08<<8)+7,P8_7},{(0x09<<8)+0,P9_0},{(0x09<<8)+1,P9_1},{(0x09<<8)+2,P9_2},{(0x09<<8)+3,P9_3},{(0x09<<8)+4,P9_4},{(0x09<<8)+5,P9_5},{(0x09<<8)+6,P9_6},{(0x0A<<8)+1,PA_1},{(0x0A<<8)+2,PA_2},{(0x0A<<8)+3,PA_3},{(0x0A<<8)+4,PA_4},{(0x0B<<8)+0,PB_0},{(0x0B<<8)+1,PB_1},{(0x0B<<8)+2,PB_2},{(0x0B<<8)+3,PB_3},{(0x0B<<8)+4,PB_4},{(0x0B<<8)+5,PB_5},{(0x0B<<8)+6,PB_6},{(0x0C<<8)+1,PC_1},{(0x0C<<8)+2,PC_2},{(0x0C<<8)+3,PC_3},{(0x0C<<8)+4,PC_4},{(0x0C<<8)+5,PC_5},{(0x0C<<8)+6,PC_6},{(0x0C<<8)+7,PC_7},{(0x0C<<8)+8,PC_8},{(0x0C<<8)+9,PC_9},{(0x0C<<8)+10,PC_10},{(0x0C<<8)+11,PC_11},{(0x0C<<8)+12,PC_12},{(0x0C<<8)+13,PC_13},{(0x0C<<8)+14,PC_14},{(0x0D<<8)+0,PD_0},{(0x0D<<8)+1,PD_1},{(0x0D<<8)+2,PD_2},{(0x0D<<8)+3,PD_3},{(0x0D<<8)+4,PD_4},{(0x0D<<8)+5,PD_5},{(0x0D<<8)+6,PD_6},{(0x0D<<8)+7,PD_7},{(0x0D<<8)+8,PD_8},{(0x0D<<8)+9,PD_9},{(0x0D<<8)+10,PD_10},{(0x0D<<8)+11,PD_11},{(0x0D<<8)+12,PD_12},{(0x0D<<8)+13,PD_13},{(0x0D<<8)+14,PD_14},{(0x0D<<8)+15,PD_15},{(0x0D<<8)+16,PD_16},{(0x0E<<8)+0,PE_0},{(0x0E<<8)+1,PE_1},{(0x0E<<8)+2,PE_2},{(0x0E<<8)+3,PE_3},{(0x0E<<8)+4,PE_4},{(0x0E<<8)+5,PE_5},{(0x0E<<8)+6,PE_6},{(0x0E<<8)+7,PE_7},{(0x0E<<8)+8,PE_8},{(0x0E<<8)+9,PE_9},{(0x0E<<8)+10,PE_10},{(0x0E<<8)+11,PE_11},{(0x0E<<8)+12,PE_12},{(0x0E<<8)+13,PE_13},{(0x0E<<8)+14,PE_14},{(0x0E<<8)+15,PE_15},{(0x0F<<8)+1,PF_1},{(0x0F<<8)+2,PF_2},{(0x0F<<8)+3,PF_3},{(0x0F<<8)+5,PF_5},{(0x0F<<8)+6,PF_6},{(0x0F<<8)+7,PF_7},{(0x0F<<8)+8,PF_8},{(0x0F<<8)+9,PF_9},{(0x0F<<8)+10,PF_10},{(0x0F<<8)+11,PF_11}};I have not systematically compared this with the LPC4330 chip itself, before I embark on this exercise, am I correct in this assumption?
Note: I realise that I should change to a 'channel' approach as used in max31855.cpp, which for the LPC4330 is:
And I think I need a method of changing from 16bit mode to 8 bit mode to support different devices and I probably need to introduce mutexes to ensure they are thread safe for a proper RTOS implementation too.
I am wondering if I should create a wrapper for an SPI object that deals with all these issues so future processor ports will be simpler, not to mention different usage for the same peripheral?
Cheers
Douglas