Re: [Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread Matthias Weingart
On Wed, Oct 20, 2004 at 07:55:27PM +0400, Dmitry wrote: > > Oops.. > > this is the one of mysterious problems which happens with large files. > I think there is a memory leak somewhere in cgen code... > Marcus, Just split your code in a couple of files and the bug should > disappear. Interestin

Re: [Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread Dmitry
Oops.. this is the one of mysterious problems which happens with large files. I think there is a memory leak somewhere in cgen code... Marcus, Just split your code in a couple of files and the bug should disappear. ~d On Wednesday 20 October 2004 18:35, Dmitry wrote: > nope... > e: b2 40 b4

[Mspgcc-users] ADC12 on 449

2004-10-20 Thread Sašo pošta
Hi, having some questions about ADC12 on F449. I' ve compiled some examples, but the one with using channel 10 named 'Fet440_ADC12_09.c' (MSP-FET430P440 Demo - ADC12, Converison of the Temperature Diode), is giving me 4095 as a result of the conversion (ADCresult). Other examples I compiled are wor

Re: [Mspgcc-users] Bug Report: Internal compiler error in gen_lowpart

2004-10-20 Thread Dmitry
Hi, could you please strip down this monster file just toMgmtQueryM$Timer$fired() function in order to allow me to debug a compiler? :) Thanks, ~d > > I'm compiling on Windows XP, and have found a fatal compiler bug. > > Here is the GCC version information, from msp430-gcc: > > Reading specs fr

Re: [Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread Dmitry
nope... e: b2 40 b4 ff mov #-76, &0x0138 ;#0xffb4 12: 38 01 14: c2 43 39 01 mov.b #0, &0x0139 ;r3 As==00 check the third line... or the result of mult does not depend on high nibble of OP2 ? If there is a bug I'll fix this shortly. ~d On Wednesday 20 October 2004 15:20, dkorov...@luxoft.com

AW: [Mspgcc-users] Something strange with multiplication of const ant byte numbers

2004-10-20 Thread Else Marcus
Thank You for Your answer, but: Operations order? The job of this little programme is to multiply an unsigned char "wert" with an unsigned number 180 to an unsigned int. With a further pair of brackets it is the same problem: --

RE: [Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread DKorovkin
Dmitry, Will you explain me the thing. ((unsigned int)180) treated as a one byte variable? Why? Am I wrong in assumption that first wert should be converted to unsigned int, next 180 must be converted to unsigned int and at last multiplied. The result must be converted to unsigned int (no idea, wha

Re: [Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread Dmitry
this is a feature actually:) you forgot about operations order. ~d On Wednesday 20 October 2004 12:48, Else Marcus wrote: > Hello! > > There is something strange with multiplication of constant byte numbers. > > --- >- --

[Mspgcc-users] Something strange with multiplication of constant byte numbers

2004-10-20 Thread Else Marcus
Hello! There is something strange with multiplication of constant byte numbers. -- Example 1.1: unsigned int multiplikation (unsigned char wert) { return (unsigned int)((unsigned int)wert*((unsigned int)180));

[Mspgcc-users] Re: Stack initialized to incorrect address on MSP430F1611

2004-10-20 Thread Robert Carlsson
Hi Steve I am using the latest windows installer from the mspgcc web site (the one from 20040723), so I guess I should try the latest binutils. Thanks for your help. Regards, Robert