CCS 4 and 672X

BrianKapitan

Is the 672X family supported under CCS 4.0? I've tried over and over to create a "Target Configuration" for my Spectrum Digital 510 USB (which works just fine in CCS 3.3).

I added the SD 510 USB followed by the 672X CPU and then an 8-bit Bypass (to match the config from CCS 3.3). I keep getting "Bad Target Silicon Revision Number" -- which is similar to what happens in CCS 3.3 when you forget to add the bypass device.

I've tried it with and without the bypass to no avail.

Is there an example for the 672X family? I see no examples for a 672X configuration such as the Lyrtech PADK.

Has 672X support been dropped in CCS 4?

over 16 years ago

0 Ki over 16 years ago

TI__Guru**** 463831 points

Hi Brian

C672x is supported with CCSv4. But you will have to use the advanced setup to create the target (see attached screenshot).

I was successful with my DA708 EVM (C672x) using a SD 510 USB. Try adding the bypass first, then followed by the C672x (see screenshot).

Thanks

0 BrianKapitan over 16 years ago in reply to Ki

Prodigy 80 points

First off, I don't see a screen shot.

Secondly, I have already tried that many times (bypass first, then C672X) and and CCS4 will crash with an error.

Here is the source-form of my config:

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<configurations XML_version="1.2" id="configurations_0">
<configuration XML_version="1.2" id="Spectrum Digital XDS510USB Emulator_0">
        <instance XML_version="1.2" desc="Spectrum Digital XDS510USB Emulator_0" href="connections\SD510USB_Connection.xml" id="Spectrum Digital XDS510USB Emulator_0" xml="SD510USB_Connection.xml" xmlpath="connections"/>
        <connection XML_version="1.2" id="Spectrum Digital XDS510USB Emulator_0">
            <instance XML_version="1.2" href="drivers\sdgo6xusb.xml" id="drivers" xml="sdgo6xusb.xml" xmlpath="drivers"/>
            <platform XML_version="1.2" id="platform_0">
                <cpu HW_revision="1.0" XML_version="1.2" description="Bypass Cpu" id="bypass_0" isa="BYPASS08"/>
                <instance XML_version="1.2" desc="C672x_0" href="cpus\c672x.xml" id="C672x_0" xml="c672x.xml" xmlpath="cpus"/>
            </platform>
        </connection>
    </configuration>
</configurations>

Does your target configuration source match?

Here is the first 50 or so lines of the JVM crash file:

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
# EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x1c0d053f, pid=7992, tid=4404
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_14-b03 mixed mode)
# Problematic frame:
# C [ctools.dll+0x2053f]
#

--------------- T H R E A D ---------------

Current thread (0x1ba65808): JavaThread "pool-3-thread-1" [_thread_in_native, id=4404]

siginfo: ExceptionCode=0xc0000005, reading address 0x00000000

Registers:
EAX=0x1c29fd28, EBX=0x1d70edc8, ECX=0x00000000, EDX=0x00000000
ESP=0x1d70ec98, EBP=0x1d70edd4, ESI=0x1d70edc8, EDI=0x1c103fe0
EIP=0x1c0d053f, EFLAGS=0x00010206

Top of Stack: (sp=0x1d70ec98)
0x1d70ec98:   1c29fd48 211c0020 1e452460 00000000
0x1d70eca8:   1c29fd48 7c00133c 1c290000 1c291d90
0x1d70ecb8:   7c001341 1c29fd28 1e452460 00000000
0x1d70ecc8:   1c293748 1d70edbc 00000000 ffffffff
0x1d70ecd8:   1d70ecbc 1c0c552b 1d70edbc 7c081ecc
0x1d70ece8:   00000000 1e452460 7c081c72 00000001
0x1d70ecf8:   1c0c5608 1c29fd28 1d70e9e0 1d70e9e4
0x1d70ed08:   00000000 1c17e2c0 00000000 00000001

Instructions: (pc=0x1c0d053f)
0x1c0d052f:   e8 fe ff ff 8b 4a 7c 8b 85 e8 fe ff ff 8b 50 7c
0x1c0d053f:   8b 02 ff 10 8b 8d e8 fe ff ff c6 81 20 01 00 00

Stack: [0x1d6d0000,0x1d710000), sp=0x1d70ec98, free space=251k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
C [ctools.dll+0x2053f]
C [ctools.dll+0x15b17]
C [DebugServer.dll+0x7ebfb]

Java frames: (J=compiled Java code, j=interpreted, Vv=VM code)
J org.mozilla.xpcom.internal.XPCOMJavaProxy.callXPCOMMethod(Ljava/lang/Object;Ljava/lang/String;[Ljava/lang/Object;Ljava/lang/String;)Ljava/lang/Object;
J org.mozilla.xpcom.internal.XPCOMJavaProxy$3.call()Ljava/lang/Object;
J java.util.concurrent.FutureTask$Sync.innerRun()V
J java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run()V
J java.util.concurrent.ThreadPoolExecutor$Worker.runTask(Ljava/lang/Runnable;)V
j java.util.concurrent.ThreadPoolExecutor$Worker.run()V+28
j java.lang.Thread.run()V+11
v ~StubRoutines::call_stub

0 Ki over 16 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

The screenshot is attached to my post, just under where is says:

Ki-Soo Lee replied on Tue, Sep 8 2009 2:50 PM | 192.91.75.29

672x.PNG

Anyway, here is a direct link:

http://e2e.ti.com/cfs-filesystemfile.ashx/__key/CommunityServer.Components.PostAttachments/00.00.03.83.02/672x.PNG

The source from your xml looks like mine:

------------------

Since CCSv4 is crashing, a *.dmp file should be getting generated. Could you attach it to this thread?

See #5 of: http://e2e.ti.com/forums/t/6045.aspx for details one where to find *.dmp files. That thread also has infromation on how to attach files to a post.

Thanks

0 BrianKapitan over 16 years ago in reply to Ki

Prodigy 80 points

Here is one of the many dump files...

DebugServer.dll20090908190357.zip

0 Ki over 15 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

Hi Brian,

According to the dmp file, the crash seems to occur during ctools initialization. To determine if this is indeed the cause, you can try to disable ctools by renaming the 'ctools.dll' file to something else (ex: ctools.bak). The file can be found in:

<INSTALL DIR>\ccsv4\emulation\analysis\bin

And then try restarting CCSv4 again and starting a debug session. Not having ctools enabled may cause some warning pop-ups to appear and also some advanced emulation features to not be available so it is not an ideal solution. But it will get you going and also clarify that ctools is the cause and help us investigate this issue.

Thanks

0 BrianKapitan over 15 years ago in reply to Ki

Prodigy 80 points

Renaming ctools.dll allowed me to connect to my C6727 target -- it loaded my code as well.

FYI:

The ctools.dll file is dated 3/27/2009 and looking at properties states that the "File Version" is 2.2.0.0 and "Product Version" is 2.1.0.032709 -- I assume those are the correct versions for CCS V4 release.

0 BrianKapitan over 15 years ago in reply to BrianKapitan

Prodigy 80 points

Ki,

Just for fun I copied the ctools.dll from my CCS 3.3 directory into where the renamed ctools.dll for CCS 4 is. The project loads and debugs without errors. I realize this is an old version of whatever ctool.dll does, but it DOES seem to work and is compatible with at least what I'm doing.

I don't get any warnings from CCS 4 using this "old" DLL.

This may point you guys in the right direction.

0 Ki over 15 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

Thanks for the update. Those versions are correct. You should be able to have most debug functionality. There may be a few issues with some advanced debugging features (HW breakpoints. watchpoints, etc). We will try to get to the bottom of this issue, perhaps is it specific to the Lyrtech EVM as other C672x targets did not exhibit the issue.

Thanks

0 BrianKapitan over 15 years ago in reply to Ki

Prodigy 80 points

I am not using the Lyrtech EVM -- I just mentioned that TI should have provided an example SUCH AS the Lyrtech EVM.

I am debugging my own hardware -- I know of one other person (different company) on totally different hardware with the same issue.

0 Ki over 15 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

BrianKapitan said:

Ki,

Just for fun I copied the ctools.dll from my CCS 3.3 directory into where the renamed ctools.dll for CCS 4 is. The project loads and debugs without errors. I realize this is an old version of whatever ctool.dll does, but it DOES seem to work and is compatible with at least what I'm doing.

I don't get any warnings from CCS 4 using this "old" DLL.

This my point you guys in the right direction.

Thanks for letting me know. This will help with our investigations.

However, be careful - as you mentioned, this is an older version of the DLL and also there have been many changes made in v4 since v3 so having the old DLL may cause more problems then not having any DLL in there.

0 ChrisPerkins over 15 years ago in reply to Ki

Prodigy 210 points

I have been having similar problems (post http://e2e.ti.com/forums/t/9707.aspx). Renaming the ctool.dll to ctool.bak allowed the emulator to connect. CCS 4 has been unusable to me to this point.

Chris

0 BrianKapitan over 15 years ago in reply to ChrisPerkins

Prodigy 80 points

Chris, you may want to try my "old DLL" trick mentioned above. So far, I can load / debug / step / etc.

Let me know if it works for you too.

0 Ki over 15 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

You should not need the old DLL for basic debug (load, step, etc). It will just disable some of the more advanced emulation features.

0 Alan DeMars over 15 years ago in reply to Ki

TI__Mastermind 30830 points

I don't have the CCS4 crashing problem. I have a problem trying to connect to the LYRTECH PADK board with the xds510 USB emulator. My .ccxml file is IDENTICAL to that suggested above.

I get the following error when I try to connect:

Error connecting to the target:
Error 0x80000200/-1063
Fatal Error during: OCS,
Bad Target Silicon Revision Number
Please check your multi-processor configuration.
The number of devices in the JTAG scan path
must be correct for the silicon revision to be read.
Or, you may have selected the wrong DSP device driver.

0 Alan DeMars over 15 years ago in reply to Alan DeMars

TI__Mastermind 30830 points

Alan DeMars said:

I don't have the CCS4 crashing problem. I have a problem trying to connect to the LYRTECH PADK board with the xds510 USB emulator. My .ccxml file is IDENTICAL to that suggested above.

I get the following error when I try to connect:

Error connecting to the target:
Error 0x80000200/-1063
Fatal Error during: OCS,
Bad Target Silicon Revision Number
Please check your multi-processor configuration.
The number of devices in the JTAG scan path
must be correct for the silicon revision to be read.
Or, you may have selected the wrong DSP device driver.

Using the same emulator, I am able to connect successfully with CCS 3.3.

0 Ki over 15 years ago in reply to Alan DeMars

TI__Guru**** 463831 points

Alan,

Though it is a different issue, did you try "disabling" ctools like Brian did?

0 Alan DeMars over 15 years ago in reply to Ki

TI__Mastermind 30830 points

Yes. No change in behavior.

0 Alan DeMars over 15 years ago in reply to Alan DeMars

TI__Mastermind 30830 points

Whats strange is that he started out with the same problem I'm having but somewhere in the process started getting the CCS crash. When he worked around the CCS crash, he was then able to connect...

0 Ki over 15 years ago in reply to Alan DeMars

TI__Guru**** 463831 points

Alan,

Another thing to try is to do some registry cleanup: http://support.spectrumdigital.com/guides/regclean_guide/

Thanks

0 Alan DeMars over 15 years ago in reply to Ki

TI__Mastermind 30830 points

I followed the procedure provided by spectrum digital very carefully and am still having the same problem.

I am still able to connect to the board and run programs using CCS 3.3.

0 TimHarron over 15 years ago in reply to Alan DeMars

TI__Mastermind 18130 points

FWIW I gave this a shot with CCSv4 and my XDS510USB and was able to connect successfully. I built my own configuration in the Advanced tab but it matches up with Ki's. So it seems that the Lyrtech PADK is accessible via SD510USB. I *think* I have a Rev B board.

I am curious if you (Alan) have verified the connection to the board using the SDConfig utility?

*edit* I know it seems redundant at this point but here is my .xml content:

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<configurations XML_version="1.2" id="configurations_0">
   <configuration XML_version="1.2" id="Spectrum Digital XDS510USB Emulator_0">
   <instance XML_version="1.2" desc="Spectrum Digital XDS510USB Emulator_0" href="connections\SD510USB_Connection.xml" id="Spectrum Digital XDS510USB Emulator_0" xml="SD510USB_Connection.xml" xmlpath="connections"/>
      <connection XML_version="1.2" id="Spectrum Digital XDS510USB Emulator_0">
      <instance XML_version="1.2" href="drivers\sdgo6xusb.xml" id="drivers" xml="sdgo6xusb.xml" xmlpath="drivers"/>
         <platform XML_version="1.2" id="platform_0">
            <cpu HW_revision="1.0" XML_version="1.2" description="Bypass Cpu" id="bypass_0" isa="BYPASS08"/>
            <instance XML_version="1.2" desc="C672x_0" href="cpus\c672x.xml" id="C672x_0" xml="c672x.xml" xmlpath="cpus"/>
         </platform>
      </connection>
   </configuration>
</configurations>

0 Alan DeMars over 15 years ago in reply to TimHarron

TI__Mastermind 30830 points

I tried your xml and got the same results.

Yes I have verified the connection with the SDConfig utility. And, remember, I am able to connect to the board successfully by just exiting CCS4 and launching CCS3.3.

0 Alan DeMars over 15 years ago in reply to Alan DeMars

TI__Mastermind 30830 points

It turned out to be a problem with the CCS4 install I was using. When I upgraded to a new engineering release, the problem went away.

0 BrianKapitan over 15 years ago

Prodigy 80 points

Any update on the original problem? This thread has gone a few different directions, but I was wondering on the status of the investigation of ctools.dll.

0 Ki over 15 years ago in reply to BrianKapitan

TI__Guru**** 463831 points

Hi Brian,

Alan's issue was a different problem. Your problem of ctools causing a crash is still unresolved. We are trying to reproduce the issue on our end but seem to be having difficulty doing so.

Code Composer Studio™︎

Code Composer Studio forum

CCS 4 and 672X