RE: Coldfire V2 devices and serial debuggers
Harjit Singh <[email protected]> Wed, 11 Feb 2009 09:58:37 -0800
| Newsgroups | gmane.comp.hardware.motorola.microcontrollers.coldfire |
|---|---|
| Message-ID | <[email protected]> |
X-SpamDetect-Info: ------------- Start ASpam results --------------- X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information X-SpamDetect: *******: 7.000000 From3consonants=0.3, SpamUrl=4.1, Accept Orbs=-1.0, X-Verify-SMTP present=0.6, Aspam=3.0 X-SpamDetect-Info: ------------- End ASpam results ----------------- --_a5ab0339-46cf-4390-aa2b-ea18c3baeb57_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable X-SpamDetect-Info: ------------- Start ASpam results --------------- X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information X-SpamDetect: *******: 7.000000 From3consonants=0.3, SpamUrl=4.1, Accept Orbs=-1.0, X-Verify-SMTP present=0.6, Aspam=3.0 X-SpamDetect-Info: ------------- End ASpam results ----------------- Forgive my lack to astuteness with regards to how to respond to this list. = I'm sure with your assistance I will figure it out soon. =20 Some more info. =20 I'm not using an RTOS and don't have really much of an OS. I boot=2C setup = the HW and run my stuff. Most of my processing is done using state machines= that are triggered by interrupts.=20 =20 In the past=2C I used a CPU32 (68376) with (ten year old) SDS tools where I= used their serial monitor. This allowed me to look at variables=2C single = step code=2C etc. while the servo interrupt was running. The way I did this= was by setting up the servo interrupt to run at a higher interrupt level a= nd so when I broke into the monitor=2C it took over the PC but since it did= n't block interrupts (since the servo interrupt level is higher=2C it will = happily run while I'm poking around the debugger). =20 95% of the time I want to look at variables. Occasionaly I want to step thr= ough code - this happens when I need to debug the state machines. I'm going to use a MC52223 device which has a BDM=2C serial (UART) and USB = interfaces. =20 Tim=2C do you have TargetMon running with a serial port? =20 Oliver=2C I should clarify that I'm not too concerned about precise cycle t= imes. My concern with BDM is that the tools I've read about=2C halt the CPU= for hundreds of microseconds and that causes the servo loop to stop functi= oning and the motors do crazy things. =20 Thanks=2C Harjit From: [email protected]: Digest from [email protected]: ze= [email protected]: Wed=2C 11 Feb 2009 00:00:05 -0600--Forwarded Messag= e Attachment--From: [email protected]: [email protected]: [= ColdFire] Coldfire V2 devices and serial debuggersDate: Mon=2C 9 Feb 2009 2= 2:09:54 -0800 Has anyone got a serial debugger running on a Coldfire V2 MCF52xxx device? = A while back I read that dBug supports the GDB protocol and dBug is a seria= l debugger. But all posts I have seen=2C seem to be about using a BDM based= debugger. The reason a serial debugger is prefered to a BDM based debugger= is that I'm doing some robotics work and I can make the servo code run at = a higher interrupt level than the serial debugger and this will allow me to= debug code with servos running. BDM is too intrusive and it will cause the= servos to go nuts. The CF debug docs say that BDM theoretically will steal= cycles but everything I have read says tha the BDM debuggers halt the CPU.= ... Thanks in advance.Harjit Windows Live=99: Keep your life in sync. See how it works. --Forwarded Mess= age Attachment--Date: Tue=2C 10 Feb 2009 11:06:49 -0800From: stout@blunkmic= ro.comTo: [email protected]: Re: [ColdFire] Coldfire V2 devices = and serial debuggersHi Harjit=2C > Has anyone got a serial debugger runnin= g on a Coldfire V2 MCF52xxx > device? Blunk's TargetTools includes a debug = monitor=2C TargetMon=2C that connectsto the debugger (Insight/GDB) via Ethe= rnet. http://www.blunkmicro.com/tools-cf.htm Currently=2C when a breakpoint= hits=2C all tasks are stopped and only thedebug monitor runs=2C in polled = mode with interrupts masked and stillcommunicating over TCP/IP=2C but we've= had this running in an alternatemode that just suspends the task that hit = the breakpoint and lets othertasks continue running. We've had other reques= ts from this=2C from customers doing motor controlas you are=2C and will be= productizing this mode. If you are interestedin trying our software=2C we = can supply this mode for you. Best regards=2CTim Harjit Singh wrote:> Has a= nyone got a serial debugger running on a Coldfire V2 MCF52xxx device?> > A= while back I read that dBug supports the GDB protocol and dBug is a > seri= al debugger. But all posts I have seen=2C seem to be about using a BDM > ba= sed debugger.> > The reason a serial debugger is prefered to a BDM based d= ebugger is that > I'm doing some robotics work and I can make the servo cod= e run at a > higher interrupt level than the serial debugger and this will = allow me > to debug code with servos running.> > BDM is too intrusive and = it will cause the servos to go nuts.> > The CF debug docs say that BDM the= oretically will steal cycles but > everything I have read says tha the BDM = debuggers halt the CPU...> > Thanks in advance.> Harjit> > ---------------= ---------------------------------------------------------> Windows Live=99:= Keep your life in sync. See how it works. > <http://windowslive.com/howitw= orks?ocid=3DTXT_TAGLM_WL_t1_allup_howitworks_022009>[email protected] >= Send a post to the list. [email protected] Join the list. > coldf= [email protected] Join the list in digest mode. > coldfire-leave@wild= rice.com Leave the list. --------------------------------------------------= ----------Tim Stoutamore=2C Principal EngineerBlunk Microsystems=2C LLC6576= Leyland Park DriveSan Jose=2C CA 95120-4558Tel: 408/323-1758stout@blunkmic= ro.comwww.blunkmicro.com---------------------------------------------------= -----------Forwarded Message Attachment--From: [email protected]: coldfire@= wildrice.comDate: Tue=2C 10 Feb 2009 20:26:44 +0100Subject: Re: [ColdFire] = Coldfire V2 devices and serial debuggersHarjit Singh wrote: [...] > The rea= son a serial debugger is prefered to a BDM based debugger is > that I'm doi= ng some robotics work and I can make the servo code run > at a higher inter= rupt level than the serial debugger and this will > allow me to debug code = with servos running.> > BDM is too intrusive and it will cause the servos t= o go nuts. In most cases=2C a serial debugger will have more impact. Do you= want single step debugging of one task while the interrupt service routine= is still running? This might indeed be harder. If you only want to observe= or maniplate variables during program execution=2C BDM is likely the bette= r way. > The CF debug docs say that BDM theoretically will steal cycles but= > everything I have read says tha the BDM debuggers halt the CPU... Coldfi= re V2 BDM is indeed more intrusive than the HC12 BDM is. Accessing memory s= talls the pipeline=2C causing several cycles delay. I couldn't get Freescal= e giving me official numbers=2C but my measurements showed me that 4..6 cyc= les for every memory access are likely. You will probably see such delays a= lso when using DMA transfers. With architectures like the Coldfire V2=2C yo= u shouldn't rely on precise cycle by cycle instruction timing unless you kn= ow very well what happens. Oliver-- Oliver Betz=2C Muenchen --Forwarded Mes= sage Attachment--Date: Tue=2C 10 Feb 2009 12:53:01 -0700From: Matt.Waddel@f= reescale.comTo: [email protected]: Re: [ColdFire] Coldfire V2 de= vices and serial debuggersHi Harjit=2C Is this a platform that's running uC= linux like M52277? If it is=2C you can rungdbserver/gdb via a serial conne= ction. Best regards=2CMatt Harjit Singh wrote:> Has anyone got a serial deb= ugger running on a Coldfire V2 MCF52xxx device?> > A while back I read tha= t dBug supports the GDB protocol and dBug is a > serial debugger. But all p= osts I have seen=2C seem to be about using a BDM > based debugger.> > The = reason a serial debugger is prefered to a BDM based debugger is that > I'm = doing some robotics work and I can make the servo code run at a > higher in= terrupt level than the serial debugger and this will allow me > to debug co= de with servos running.> > BDM is too intrusive and it will cause the serv= os to go nuts.> > The CF debug docs say that BDM theoretically will steal = cycles but > everything I have read says tha the BDM debuggers halt the CPU= ....> > Thanks in advance.> Harjit> > -------------------------------------= -----------------------------------> Windows Live=99: Keep your life in syn= c. See how it works. > <http://windowslive.com/howitworks?ocid=3DTXT_TAGLM_= WL_t1_allup_howitworks_022009>[email protected] > Send a post to the li= st. [email protected] Join the list. > [email protected]= m Join the list in digest mode. > [email protected] Leave the lis= t. _________________________________________________________________ Windows Live=99: E-mail. Chat. Share. Get more ways to connect.=20 http://windowslive.com/howitworks?ocid=3DTXT_TAGLM_WL_t2_allup_howitworks_0= 22009= --_a5ab0339-46cf-4390-aa2b-ea18c3baeb57_ Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable <html><pre> X-SpamDetect-Info: ------------- Start ASpam results --------------- X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information X-SpamDetect: *******: 7.000000 From3consonants=0.3, SpamUrl=4.1, Accept Orbs=-1.0, X-Verify-SMTP present=0.6, Aspam=3.0 X-SpamDetect-Info: ------------- End ASpam results ----------------- </pre></html> <html> <head> <style> ..hmmessage P { margin:0px=3B padding:0px } body.hmmessage { font-size: 10pt=3B font-family:Verdana } </style> </head> <body class=3D'hmmessage'> Forgive my lack to astuteness with regards to how to respond to this list. = I'm sure with your assistance I will figure it out soon.<BR>  =3B<BR> Some more info.<BR>  =3B<BR> I'm not using an RTOS and don't have really much of an OS. I boot=2C setup = the HW and run my stuff. Most of my processing is done using state machines= that are triggered by interrupts. <BR>  =3B<BR> In the past=2C I used a CPU32 (68376) with (ten year old) SDS tools where I= used their serial monitor. This allowed me to look at variables=2C single = step code=2C etc. while the servo interrupt was running. The way I did this= was by setting up the servo interrupt to run at a higher interrupt level a= nd so when I broke into the monitor=2C it took over the PC but since it did= n't block interrupts (since the servo interrupt level is higher=2C it will = happily run while I'm poking around the debugger).<BR>  =3B<BR> 95% of the time I want to look at variables. Occasionaly I want to step thr= ough code - this happens when I need to debug the state machines.<BR><BR> I'm going to use a MC52223 device which has a BDM=2C serial (UART) and USB = interfaces.<BR>  =3B<BR> Tim=2C do you have TargetMon running with a serial port?<BR>  =3B<BR> Oliver=2C I should clarify that I'm not too concerned about precise cycle t= imes. My concern with BDM is that the tools I've read about=2C halt the CPU= for hundreds of microseconds and that causes the servo loop to stop functi= oning and the motors do crazy things.<BR>  =3B<BR> Thanks=2C<BR> Harjit<BR><BR> <HR id=3DstopSpelling> <BR> From: [email protected]<BR>Subject: Digest from [email protected]<B= R>To: [email protected]<BR>Date: Wed=2C 11 Feb 2009 00:00:05 -0600<BR><B= R><BR><BR>--Forwarded Message Attachment--<BR>From: [email protected]<BR= >To: [email protected]<BR>Subject: [ColdFire] Coldfire V2 devices and s= erial debuggers<BR>Date: Mon=2C 9 Feb 2009 22:09:54 -0800<BR><BR> <STYLE> ..ExternalClass .EC_hmmessage P {padding:0px=3B} ..ExternalClass body.EC_hmmessage {font-size:10pt=3Bfont-family:Verdana=3B} </STYLE> Has anyone got a serial debugger running on a Coldfire V2 MCF52xxx device?<= BR> =3B<BR>A while back I read that dBug supports the GDB protocol and = dBug is a serial debugger. But all posts I have seen=2C seem to be about us= ing a BDM based debugger.<BR> =3B<BR>The reason a serial debugger is pr= efered to a BDM based debugger is that I'm doing some robotics work and I c= an make the servo code run at a higher interrupt level than the serial debu= gger and this will allow me to debug code with servos running.<BR> =3B<= BR>BDM is too intrusive and it will cause the servos to go nuts.<BR> = =3B<BR>The CF debug docs say that BDM theoretically will steal cycles but e= verything I have read says tha the BDM debuggers halt the CPU...<BR> = =3B<BR>Thanks in advance.<BR>Harjit<BR><BR><BR> <HR> <BR> Windows Live=99: Keep your life in sync. <A href=3D"http://windowslive.com/= howitworks?ocid=3DTXT_TAGLM_WL_t1_allup_howitworks_022009">See how it works= ..</A> <BR><BR>--Forwarded Message Attachment--<BR>Date: Tue=2C 10 Feb 2009 = 11:06:49 -0800<BR>From: [email protected]<BR>To: [email protected]<B= R>Subject: Re: [ColdFire] Coldfire V2 devices and serial debuggers<BR><BR><= BR><PRE>Hi Harjit=2C<BR> <BR> >=3B Has anyone got a serial debugger runni= ng on a Coldfire V2 MCF52xxx<BR> >=3B device?<BR> <BR>Blunk's TargetTools= includes a debug monitor=2C TargetMon=2C that connects<BR>to the debugger = (Insight/GDB) via Ethernet.<BR> <BR><A href=3D"http://www.blunkmicro.com/to= ols-cf.htm">http://www.blunkmicro.com/tools-cf.htm</A><BR> <BR>Currently=2C= when a breakpoint hits=2C all tasks are stopped and only the<BR>debug moni= tor runs=2C in polled mode with interrupts masked and still<BR>communicatin= g over TCP/IP=2C but we've had this running in an alternate<BR>mode that ju= st suspends the task that hit the breakpoint and lets other<BR>tasks contin= ue running.<BR> <BR>We've had other requests from this=2C from customers do= ing motor control<BR>as you are=2C and will be productizing this mode. If y= ou are interested<BR>in trying our software=2C we can supply this mode for = you.<BR> <BR>Best regards=2C<BR>Tim<BR> <BR>Harjit Singh wrote:<BR>>=3B H= as anyone got a serial debugger running on a Coldfire V2 MCF52xxx device?<B= R>>=3B <BR>>=3B A while back I read that dBug supports the GDB protoco= l and dBug is a <BR>>=3B serial debugger. But all posts I have seen=2C se= em to be about using a BDM <BR>>=3B based debugger.<BR>>=3B <BR>>=3B= The reason a serial debugger is prefered to a BDM based debugger is that <= BR>>=3B I'm doing some robotics work and I can make the servo code run at= a <BR>>=3B higher interrupt level than the serial debugger and this will= allow me <BR>>=3B to debug code with servos running.<BR>>=3B <BR>>= =3B BDM is too intrusive and it will cause the servos to go nuts.<BR>>=3B= <BR>>=3B The CF debug docs say that BDM theoretically will steal cycles= but <BR>>=3B everything I have read says tha the BDM debuggers halt the = CPU...<BR>>=3B <BR>>=3B Thanks in advance.<BR>>=3B Harjit<BR>>=3B = <BR>>=3B ----------------------------------------------------------------= --------<BR>>=3B Windows Live=99: Keep your life in sync. See how it work= s. <BR>>=3B <=3B<A href=3D"http://windowslive.com/howitworks?ocid=3DTXT= _TAGLM_WL_t1_allup_howitworks_022009">http://windowslive.com/howitworks?oci= d=3DTXT_TAGLM_WL_t1_allup_howitworks_022009</A>&[email protected] = <BR>>=3B Send a post to the list. [email protected] Join the lis= t. <BR>>=3B [email protected] Join the list in digest mode. <B= R>>=3B [email protected] Leave the list.<BR> <BR>--------------= ----------------------------------------------<BR>Tim Stoutamore=2C Princip= al Engineer<BR>Blunk Microsystems=2C LLC<BR>6576 Leyland Park Drive<BR>San = Jose=2C CA 95120-4558<BR>Tel: 408/323-1758<BR>[email protected]<BR>www.b= lunkmicro.com<BR>----------------------------------------------------------= --<BR></PRE><BR><BR>--Forwarded Message Attachment--<BR>From: [email protected]= et<BR>To: [email protected]<BR>Date: Tue=2C 10 Feb 2009 20:26:44 +0100<= BR>Subject: Re: [ColdFire] Coldfire V2 devices and serial debuggers<BR><BR>= <PRE>Harjit Singh wrote:<BR> <BR>[...]<BR> <BR>>=3B The reason a serial d= ebugger is prefered to a BDM based debugger is <BR>>=3B that I'm doing so= me robotics work and I can make the servo code run <BR>>=3B at a higher i= nterrupt level than the serial debugger and this will <BR>>=3B allow me t= o debug code with servos running.<BR>>=3B <BR>>=3B BDM is too intrusive= and it will cause the servos to go nuts.<BR> <BR>In most cases=2C a serial= debugger will have more impact.<BR> <BR>Do you want single step debugging = of one task while the interrupt <BR>service routine is still running? This = might indeed be harder.<BR> <BR>If you only want to observe or maniplate va= riables during program <BR>execution=2C BDM is likely the better way.<BR> <= BR>>=3B The CF debug docs say that BDM theoretically will steal cycles bu= t <BR>>=3B everything I have read says tha the BDM debuggers halt the CPU= ....<BR> <BR>Coldfire V2 BDM is indeed more intrusive than the HC12 BDM is. = <BR>Accessing memory stalls the pipeline=2C causing several cycles delay. I= <BR>couldn't get Freescale giving me official numbers=2C but my <BR>measur= ements showed me that 4..6 cycles for every memory access are <BR>likely.<B= R> <BR>You will probably see such delays also when using DMA transfers.<BR>= <BR>With architectures like the Coldfire V2=2C you shouldn't rely on <BR>p= recise cycle by cycle instruction timing unless you know very well <BR>what= happens.<BR> <BR>Oliver<BR>-- <BR>Oliver Betz=2C Muenchen<BR> <BR></PRE><B= R><BR>--Forwarded Message Attachment--<BR>Date: Tue=2C 10 Feb 2009 12:53:01= -0700<BR>From: [email protected]<BR>To: [email protected]<BR>S= ubject: Re: [ColdFire] Coldfire V2 devices and serial debuggers<BR><BR><PRE= >Hi Harjit=2C<BR> <BR>Is this a platform that's running uClinux like M52277= ? If it is=2C you can run<BR>gdbserver/gdb via a serial connection.<BR> <B= R>Best regards=2C<BR>Matt<BR> <BR>Harjit Singh wrote:<BR>>=3B Has anyone = got a serial debugger running on a Coldfire V2 MCF52xxx device?<BR>>=3B = <BR>>=3B A while back I read that dBug supports the GDB protocol and dBug= is a <BR>>=3B serial debugger. But all posts I have seen=2C seem to be a= bout using a BDM <BR>>=3B based debugger.<BR>>=3B <BR>>=3B The reaso= n a serial debugger is prefered to a BDM based debugger is that <BR>>=3B = I'm doing some robotics work and I can make the servo code run at a <BR>>= =3B higher interrupt level than the serial debugger and this will allow me = <BR>>=3B to debug code with servos running.<BR>>=3B <BR>>=3B BDM is = too intrusive and it will cause the servos to go nuts.<BR>>=3B <BR>>= =3B The CF debug docs say that BDM theoretically will steal cycles but <BR>= >=3B everything I have read says tha the BDM debuggers halt the CPU...<BR= >>=3B <BR>>=3B Thanks in advance.<BR>>=3B Harjit<BR>>=3B <BR>>= =3B -----------------------------------------------------------------------= -<BR>>=3B Windows Live=99: Keep your life in sync. See how it works. <BR>= >=3B <=3B<A href=3D"http://windowslive.com/howitworks?ocid=3DTXT_TAGLM_= WL_t1_allup_howitworks_022009">http://windowslive.com/howitworks?ocid=3DTXT= _TAGLM_WL_t1_allup_howitworks_022009</A>&[email protected] <BR>>= =3B Send a post to the list. [email protected] Join the list. <BR>= >=3B [email protected] Join the list in digest mode. <BR>>= =3B [email protected] Leave the list.<BR></PRE><br /><hr />Window= s Live=99: E-mail. Chat. Share. Get more ways to connect. <a href=3D'http:= //windowslive.com/howitworks?ocid=3DTXT_TAGLM_WL_t2_allup_howitworks_022009= ' target=3D'_new'>See how it works.</a></body> </html>= [email protected]=09Send a post to the list. [email protected]=09Join the list. [email protected]=09Join the list in digest mode. [email protected]=09Leave the list. --_a5ab0339-46cf-4390-aa2b-ea18c3baeb57_--