[Skyeye-developer] help simulating at91 timer

[josh jordan]

Thank you for your reply.  I was able to apply the patch and recompile
the binary.  Skyeye will run with 'SIGHUP's, it will still step thru
instructions.  I am still not seeing any effect of writing to IO
space.  Here is the code I am trying to simulate and my skyeye.conf
file.  Please let me know if you see an error in the following test
code.  Best regards, Josh Jordan.


int	main (void) {
	signed char i, j;
	long result = 0;
	volatile long output;
	unsigned int finalcount;
	signed char coeff[9];		// variable to test library function
	// Initialize the Atmel AT91SAM7S256 (watchdog, PLL clock, default
interrupts, etc.)
	LowLevelInit(); //step out of functions that wait on io bits to change

	// enable the Timer0 peripheral clock
	volatile AT91PS_PMC	pPMC = AT91C_BASE_PMC;
	volatile AT91PS_TC pTC = AT91C_BASE_TC0;

	pPMC->PMC_PCER = (1<<AT91C_ID_TC0);	// enable Timer0 peripheral clock	
	pTC->TC_CCR = 0x0005;				// enable the clock	and start it
	pTC->TC_CMR = 0x4000;	//enable waveform mode
	//enableIRQ();
	//enableFIQ();
	for (i = 0; i <= 8; i++){
		coeff[i] = i-4; //manual init puts vars in sram instead of flash.
	}
	for (i = 0; i < 100; i++){
		for (j = 8; j > 0; j--){
			result += coeff[j]*i;
		}
		output = result;	
	}
	finalcount = pTC->TC_CV;
	while (1) {}
}

cpu: arm7tdmi
mach: at91
mem_bank: map=M, type=RW, addr=0x00000000, size=0x00010000
mem_bank: map=M, type=RW, addr=0x00200000, size=0x00004000
mem_bank: map=I, type=RW, addr=0xf0000000, size=0x10000000
net: type=rtl8019, mac=0:4:3:2:1:f, ethmod=tuntap, hostip=10.0.0.1
log: logon=0, logfile=/tmp/sk1.log, start=100000, end=200000



On 7/18/07, Michael. Kang <blackfin.kang at gmail.com> wrote:
>             In the original implementation of at91 simulation, only
> timer1 is simulated. I had done a patch against skyeye_1_2_3_Rel to
> implement three timer channel.Maybe useful for your case.
>          Run "patch -p0 <  skyeye_mach_at91-timer.diff" under skyeye
> source dir to apply my patch in the attachment.


>
> -- Thanks
> -- Michael.Kang
> On 7/17/07, josh jordan <josh.outerspace at gmail.com> wrote:
> > I am trying to get timer0 to work to count cycles in simulation.  Here
> > is some debug output:
> >
> > p/x pTC0
> > $1 = 0xfffa0000
> > p/x *pTC0
> > $2 = {TC_CCR = 0x21, TC_CMR = 0x80000000, Reserved0 = {0x0, 0x0},
> > TC_CV = 0x0, TC_RA = 0x0, TC_RB = 0x0, TC_RC = 0x0, TC_SR = 0x0,
> > TC_IER = 0x0, TC_IDR = 0x0, TC_IMR = 0x0}
> >
> > it recognizes the pTC0 pointer with the correct address.  It decodes
> > the struct.  But the values of the struct are stuck.  My program
> > assigns other values to them but they always read as above.  Even when
> > I try to change it like pTC0->TC_CCR=1 that has no effect either.
> > pTC0->TC_CV always reads zero!
> >
> > It looks like the source code updates timers, I am just not using it correctly.
> >
> > Here is my conf file, trimmed down most everything commented.
> >
> > cpu: arm7tdmi
> >
> > # machine(develop board) maybe at91 or ep7312
> > mach: at91
> >
> > mem_bank: map=M, type=RW, addr=0x00000000, size=0x00010000
> > mem_bank: map=M, type=RW, addr=0x00200000, size=0x00004000
> > mem_bank: map=I, type=RW, addr=0xf0000000, size=0x10000000
> >
> > net: type=rtl8019, mac=0:4:3:2:1:f, ethmod=tuntap, hostip=10.0.0.1
> >
> > log: logon=0, logfile=/tmp/sk1.log, start=100000, end=200000
> >
> > _______________________________________________
> > Skyeye-developer mailing list
> > Skyeye-developer at lists.gro.clinux.org
> > http://lists.gro.clinux.org/cgi-bin/mailman/listinfo/skyeye-developer
> >
> >
>
>
> --
> www.skyeye.org
>
> _______________________________________________
> Skyeye-developer mailing list
> Skyeye-developer at lists.gro.clinux.org
> http://lists.gro.clinux.org/cgi-bin/mailman/listinfo/skyeye-developer
>
>
>