Is bit-banding possible on the Kinetis? Any examples?

markosiponen — Tue, 16 Jul 2013 09:19:45 GMT

I read here: https://en.wikipedia.org/wiki/ARM_Cortex-M that bit-banding is optional on the Cortex M0+.

Re: Is bit-banding possible on the Kinetis? Any examples?

antonioconcio — Tue, 16 Jul 2013 12:29:26 GMT

Hi,

You are right, it is optional and so bit-banding was not implemented in Kinetis-L because we have Bit Manipulation Engine (BME) which is much more powerful compared to bit-banding. You can find a chapter about it on each Kinetis L Reference Manual.

We have some examples in KL05/KL25 sample code, you can also refer to this THREAD.

Regards,

Antonio

Re: Is bit-banding possible on the Kinetis? Any examples?

markosiponen — Tue, 16 Jul 2013 13:56:06 GMT

Thank you. I also downloaded the FRDM-KL25Z and TWRKL25Z48M Sample Code Package and looked at the BME examples.

Re: Is bit-banding possible on the Kinetis? Any examples?

egoodii — Tue, 16 Jul 2013 17:51:42 GMT

Specific 'Bit Banding', as per that ARM document, IS enabled in (at least) the Kinetis 'K' M4 devices. Pretty easy to use on peripherals; there is even a macro for it in the Kinetis headers as 'BITBAND_REG'. It is a little harder to use in RAM, mostly because the tools (compilers and linkers) won't help! But here is an example for a graphics display RAM, in this case using the least-significant-byte of each 32-bit word as screen data (so that the upper bits can be preset for a SPI DMA refresh).

So, you have to pre-allocate the space yourself (fixed location, subtracted from the linker-space):

#pragma location=0x2000EF00

__no_init u32_t Display_RAM[Y_PITCH*8]; //Pre-allocated fixed-space for Display RAM

//Necessary, in SRAM-U, to use bit-banding!

Then the SRAM bit-band macro is pretty similar to theirs:

#define BITBAND_SRAM_A(Addr,Bit) (0x22000000u + (((u32_t)(Addr) - (u32_t)0x20000000u)<<5) + (((u32_t)(Bit))<<2))

and here is a line-drawing process using that bit-banding to avoid detailed RMW of display RAM bits, preferring instead to just write each single bit

///Draw a line from (xstrt,ystrt) to (xfin,yfin) using 32bit dashing 'pattern'

// modified Bresenham algorithm, using ARM M4 'bit banding' RAM access to allow direct bit-writes (no RMW)

#define Y_PITCH 132

void OLED_Line(u16_t xstrt, u16_t ystrt, u16_t xfin, u16_t yfin, u32_t pattern)

{

s32_t eterm=0; //Error-term from fractional accumulations

///Some kind of 'set to half' should improve centering the steps

u16_t length; //# of dots to draw

s32_t diffx=xfin-xstrt;

s32_t diffy=yfin-ystrt;

u32_t _PTR_ offset;

s16_t stepx=1<<5; //Byte increment point for bit-band address

u16_t ybit;

if( xstrt > 127) xstrt = 127; //A little sanity checking!

if( xfin > 127) xfin = 127;

if( ystrt > 63) ystrt = 63;

if( yfin > 63) yfin = 63;

ybit = ystrt&7;

offset = &Display_RAM[xstrt];

offset += (ystrt>>3)*Y_PITCH; //8 bits of pixels per page, so add the page offset

offset = (u32_t _PTR_)BITBAND_SRAM_A(offset,ybit); //Convert to bit-band address, adding-in bit offset within byte for pixel row

diffx=xfin-xstrt;

diffy=yfin-ystrt;

if (diffx<0)

{

stepx=-stepx;

diffx=-diffx;

}

if (diffy<0)

{

diffy=-diffy;

if (diffx>diffy) //'mostly horizontal', move integer x, fractional y

{

length=diffx+1;

// eterm = diffy/2;

for (u16_t i=0;i<length;i++)

{

*offset=pattern; //LSB-only is written

tmp = (pattern&1)<<31; //Compiler is SMART -- makes a proper ROR out of this pair!!!

pattern = (pattern>>1) | tmp;

offset+=stepx;

eterm+=diffy;

if (eterm>diffx)

{

eterm-=diffx;

offset-=1;//<<2; //Each step in y moves to the next bit, BUT once we underflow the byte

ybit = (ybit-1)&7; // where bitcount in bitband address is *4

if(ybit == 7) // we must move to the LSB or MSB of the next row up/down.

offset -= (Y_PITCH<<5)-(8);//<<2); //To next bitband byte location, less the 8-bit-count already done

}

else //'mostly vertical', move integer y, fractional x

{

length=diffy+1;

// eterm = diffx/2;

for (u16_t i=0;i<length;i++)

{

*offset=pattern; //LSB-only is written

tmp = (pattern&1)<<31;

pattern = (pattern>>1) | tmp;

offset-=1;//<<2; //Each step in y moves to the next bit, BUT once we underflow the byte

ybit = (ybit-1)&7; // where bitcount in bitband address is *4

if(ybit == 7) // we must move to the LSB or MSB of the next row up/down.

offset -= (Y_PITCH<<5)-(8);//<<2); //To next bitband byte location, less the 8-bit-count already done

eterm+=diffx;

if (eterm>diffy)

{

eterm-=diffy;

offset+=stepx;

}

}else

{ //Positive Y-direction:

if (diffx>diffy) //'mostly horizontal', move integer x, fractional y

{

length=diffx+1;

// eterm = diffy/2;

for (u16_t i=0;i<length;i++)

{

*offset=pattern; //LSB-only is written

tmp = (pattern&1)<<31;

pattern = (pattern>>1) | tmp;

offset+=stepx;

eterm+=diffy;

if (eterm>diffx)

{

eterm-=diffx;

// Note: bitcount in bitband address is *4, but since 'offset' is a u32 pointer, this is accounted for

offset+=1;//<<2; //Each step in y moves to the next bit, BUT once we overflow the byte

ybit = (ybit+1)&7;

if(ybit == 0) // we must move to the LSB or MSB of the next row up/down.

offset += (Y_PITCH<<5)-(8);//<<2); //To next bitband byte location, less the 8-bit-count already done

}

else //'mostly vertical', move integer y, fractional x

{

length=diffy+1;

// eterm = diffx/2;

for (u16_t i=0;i<length;i++)

{

*offset=pattern; //LSB-only is written

tmp = (pattern&1)<<31;

pattern = (pattern>>1) | tmp; //Rotate-right 1

offset+=1;//<<2; //Each step in y moves to the next bit, BUT once we overflow the byte

ybit = (ybit+1)&7; // where bitcount in bitband address is *4

if(ybit == 0) // we must move to the LSB or MSB of the next row up/down.

offset += (Y_PITCH<<5)-(8);//<<2); //To next bitband byte location, less the 8-bit-count already done

eterm+=diffx;

if (eterm>diffy)

{

eterm-=diffy;

offset+=stepx;

}

topic Re: Is bit-banding possible on the Kinetis? Any examples? in Kinetis Microcontrollers

Is bit-banding possible on the Kinetis? Any examples?

Re: Is bit-banding possible on the Kinetis? Any examples?

Re: Is bit-banding possible on the Kinetis? Any examples?

Re: Is bit-banding possible on the Kinetis? Any examples?