The first one -- something is really odd.
retiu must be issued.
Second one -- can't do anything. Will have a look anyway.
~d
On Thursday 08 May 2003 21:24, Mark Stokes wrote:
> I found two things that are not good:
> First is a major problem that I'm not sure how it got this far.
> It looks like, when you use Timer_A CCR0 IRQ there is a missing iret
> instruction. Here is a snapshot of what I have:
> 00001a8a <TimerACC0IRQ>:
>
> /***********************************************************************
> *******
> Timer_A Capture Compare Interrupt Service Routine (CCIFG0):
> ************************************************************************
> ******/
> interrupt (TIMERA0_VECTOR) TimerACC0IRQ( void )
> {
> // Clear TAR, this is not 100% accurate, because of IRQ latency.
> // But will do for a test
> TACTL |= TACLR; // Clear the TAR register.
> 1a8a: a2 d2 60 01 bis #4, &0x0160 ;r2 As==10
> count8m = TACCR0; // Get current value of CCR0
> 1a8e: 92 42 72 01 mov &0x0172,&0x0200 ;0x0172
> 1a92: 00 02
> 1a94: 82 43 02 02 mov #0, &0x0202 ;r3 As==00
> __asm__( "reti" );
> 1a98: 00 13 reti
>
> 00001a9a <BasicTimerIRQ>:
> }
>
> As you can see, the reti is only there because I specifically added the
> __asm__ instruction. Since this is an interrupt, shouldn't the reti be
> there automatically?
>
> Second thing I found is an optimizing issue. The very handy
> "port4.out.pin7" method of accessing the ports has a slight side effect.
> When using it to manually set a pin high, or low, etc. it works great
> and is very small code. However, in this example, it is amazing what is
> generated. The net effect I want here it to "toggle" the output pin:
>
> port4.out.pin7 = ~port4.out.pin7;
> 1a52: fb b0 80 00 bit.b #128, 0(r11) ;#0x0080
> 1a56: 00 00
> 1a58: 4e 43 clr.b r14 ;
> 1a5a: 4e 63 adc.b r14 ;
> 1a5c: 7e e3 xor.b #-1, r14 ;r3 As==11
> 1a5e: 4e 11 rra.b r14 ;
> 1a60: 4e 43 clr.b r14 ;
> 1a62: 4e 10 rrc.b r14 ;
> 1a64: 6f 4b mov.b @r11, r15 ;
> 1a66: 7f f0 7f 00 and.b #127, r15 ;#0x007f
> 1a6a: 4f de bis.b r14, r15 ;
> 1a6c: cb 4f 00 00 mov.b r15, 0(r11);
>
> I can't even tell what this is doing. This is quite a bit of code for a
> simple toggle. A more efficient way would be:
>
> __asm__( " xor.b #0x0080, &0x001d ;\n" );
>
> Both examples were compiled w/ 20030506 version of Win32 release. I
> used the following command line:
> msp430-gcc -lm -mmcu=msp430x449 -W -Wall -g %2 -o %1.elf %1.c
> (%1 is filename and %2 is "-O2")
>
> Any thoughts?
> -Mark
>
>
>
>
> -------------------------------------------------------
> Enterprise Linux Forum Conference & Expo, June 4-6, 2003, Santa Clara
> The only event dedicated to issues related to Linux enterprise solutions
> www.enterpriselinuxforum.com
>
> _______________________________________________
> Mspgcc-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/mspgcc-users
--
/********************************************************************
("`-''-/").___..--''"`-._ (\ Dimmy the Wild UA1ACZ
`6_ 6 ) `-. ( ).`-.__.`) Enterprise Information Sys
(_Y_.)' ._ ) `._ `. ``-..-' Nevsky prospekt, 20 / 44
_..`--'_..-_/ /--'_.' ,' Saint Petersburg, Russia
(il),-'' (li),' ((!.-' +7 (812) 3468202, 5585314
********************************************************************/
