FAQ - CCSv6

Installation

Q: Can I install CCSv6 on a computer that has previous versions?

Yes you can install multiple versions of CCS on the same machine. Just make sure to install them in different directories (i.e. not on top of one another).

Q: What is the difference between the web installer and the offline installer?

The web installer is a small installation program that you download and execute. You then make your installation selections (device families and features desired) and the installer then downloads and installs only those selected packages. The offline installer is a large package that includes all packages (except for those only available via the App Center). The offline installer is generally only recommended if you have issues with your firewall or anti-virus software blocking the web installer. The offline installer is also useful if you need to install CCS on a machine that does not have internet access.

Q: CCS says "Not Responding" when installing additional Add-On's. How do I resolve this?

This is a known issue with the installation dialog. However, the installation of the Add-On's is actually progressing (only the dialog itself is hung). Simply leave it alone and the installation will eventually succeed and the dialog will be responsive again.

Q: My username has Unicode characters and CCS fails to install. How do I resolve this?

Due to a limitation in eclipse, the username cannot contain any Unicode characters. In order to install the product, please create a temporary admin user and install CCS for all users.

Q: On Linux the installer just exits without doing anything, why is this?

The installer requires some packages to be installed in order to run. You are likely missing one or more of these packages. Please refer to Linux_Host_Support_CCSv6 for more information on installing on Linux.

Q: On Windows the installer just exits without doing anything, why is this?

There is some software that sets up the TCL_LIBRARY environment variable which is interfering with the CCS installer. Please unset this variable temporarily, before trying the install.

Q: After installing on Linux, CCS simply refuses to open

This may happen when updating or installing a newer version of the same stream of CCS (6.0.0 to 6.0.1, for example). Nothing happens when double-clicking the desktop icon and, if launching from a command line, the following error is shown:

user@host:/opt/ti/ccsv6/eclipse$ ./ccstudio 
Permission denied
If this continues, please run fsclean or set TI_APPDATA_DIR to directory you have permissions to access

The cause is that, if CCS was previously launched as root, the temporary directory in the user area may have become tainted and its ownership was set to root.
In other words, the temporary CCS directory would look similar to:

user@host:~/.ti$ ls -al
total 20
drwxr-xr-x   3 root     root      4096 Jan 24  2014 .
drwxr-xr-x 135 user     user      12288 Jan 26 11:25 ..
drwxr-xr-x   3 root     root      4096 Jan 24  2014 CCSv6
-rw-r--r--   1 root     root         0 Jan 24  2014 .lock

To fix this, use chown to change the ownership of all the directories and the .lock file back to the user

user@host:~/.ti$ sudo chown -hR user:user * .lock

Q: After installing CCS, I found out some features (such as certain processor families or debug probes) are missing. Is there a way to add these features?

Yes. Simply re-run the installer executable (either online or offline). Follow the installer screens and select the same directory as the original install; after that, keep following the installer screens and select the missing components.

You can also add new features from inside CCS, but this is more limited (support for additional device families cannot be added this way, for example). For details, check Updating CCSv6

Q: When installing CCS it says "An error occured: can't read "msg": no such variable". How do I resolve this?

Make sure you are installing CCS to a clean directory where a previous installation does not exist.

Q: How can I perform a silent installation of CCS?

For CCSv6, you can first create a response file by running the installer with these options:
<installername> --save-response-file <full path to response file\responsefile name> --skip-install true

For example:

 ccs_setup_6.1.1.00022.exe --save-response-file c:\temp\response.txt --skip-install true 

Note that the directory for the response file must already exist.

You can then click through the installer, picking the options you'd like to install. The parameter --skip-install true will skip the installation part and just create the response.txt file. If you omit that option it will continue with the installation.

To run the installation using the response.txt file:

ccs_setup_6.1.1.00022.exe --response-file c:\temp\response.txt 

Other options you can use with the above command are --prefix <installdirectory> and --mode unattended

Licensing

Q: Where is the link to the 90 day eval license?

• CCSv6 extension

Q: If I upgrade my CCS license to CCSv6 can I still use CCSv5?

Your CCSv6 license will work with both CCSv6 and CCSv5. Also your CCSv5 license will not stop working just because you generated the CCSv6 license.

Q: I ordered CCS and received CCSv5, how do I get CCSv6?

If you ordered CCS shortly after the launch of CCSv6 it is possible that you may have received a CCSv5 package as it takes a few weeks to update inventory. In this case activate your license according to the instructions in the package. When you have done that there will be a manage button beside your license entry in myregistered software (www.ti.com/myregisteredsoftware). Make sure that you select the entry for your prouduct line item and not your subscription.  Click on it and then select the option to upgrade your license. This will result in a CCSv6 license being generated for you. Then you can download the CCSv6 image and drop in your v6 license, no need to even install CCSv5.

Q: How do I get CCSv5 or CCSv4?

If you purchase CCSv6 your license file will also work with CCSv5 or CCSv4. Simply download the version of CCS that you want to use from the download page and place your license file in the /DebugServer/license folder.

Q: How do I get CCSv3.3?

If you purchase CCSv6 you request a CCSv3.3 image from ccs_license_file_help@list.ti.com make sure to include your CCSv6 activation code. You will be sent a download link for CCSv3.3. You will not be able to register CCSv3.3 as your upgrades are determined by the subscription associated with your CCSv6 purchase.

Q: What happens if the license server goes down or there is a network problem?

Also if for some reason the license server goes down or there is a network issue we still let you use CCS it just says “unlicensed” at the top.

Q: What happens to CCS when my subscription expires?

CCS comes with 12 months of subscription. This is in no way related to your license to use CCS. If you have purchased a CCS license there is no time limit on how long you can use it. Subscription is used to determine if you receive major upgrades (v7, v8...) for free.

Q: I didn't receive the instructions along with my license file, where can I get them?

Some mail filters may block the license installation instructions from being delivered. There are a few ways that you can install a license file. If you created a node locked license the easiest thing to do is to just copy your license file into \ccsv6\ccs_base\DebugServer\License CCS will automatically look in that folder for licenses.  If you need to setup a floating license on a license server you can get the instructions here: License_Server_Administration_for_CCS

Q: How long is my CCS license good for?

Unless you are using one of the 90 day evaluation licenses your CCS license is a perpetual license, i.e. there is no time limit on the license.

Q: How do I add a second computer to my node-locked license?

With node-locked licenses you can specify 2 MAC addresses, i.e. you can run CCS on two different machines. You can specify these addresses when initially creating your license or if you need to do it later on you can add a second MAC address using the re-host feature at www.ti.com/myregisteredsoftware

Q: How do I upgrade from the code size limited tools to the full tools?

Just purchase your full license and then add that license to CCS via the help menu. CCS will then start using the license with no limit.

Q: Is license "borrowing" supported?

We have not implemented license borrowing in CCS. Borrowing is a license feature where you can reserve or checkout a license from the license server for a period of time. Some vendors require you to do this if your machine is going to be out of contact with the license server. However with CCS we just change the title bar to unlicensed so there is no need to implement license borrowing. If you are going to be constantly running CCS disconnected from your companies network you should consider getting a node locked license.

Q: I installed a second copy of CCS on my machine how do I get a license for it?

You can install multiple copies of CCS on the same machine use the same license for all. Just copy the license file from your existing installation into your new one. You can get another copy of your license file from www.ti.com/myregisteredsoftware

Q: What advantages does a paid license give me?

The paid license allows using more advanced JTAG debuggers (XDS200, XDS510, XDS560, XDS560v2) and their advanced features like System Trace. This does not affect support for the XDS100 JTAG debuggers and the ones that are built into development boards that are available with a free license.

For MSP430, the advantage is more obvious. The paid license allows using the unlimited version of the compiler, useful for code larger than 16kB.

IMPORTANT! CCS free license does not impose any limitations on the optimizer, thus avoiding any impact to the code size limited version for MSP430.

Q: I have a CCSv6 node locked license, but when I start a CCS debug session the full license is replaced by the free license. Why?

CCS may just be finding the free license first and if that is good enough to debug with ( as in the case of XDS100v2 or on-board emulators) then it will just use that. You can follow these steps to disable the free license so the full license is always used.

Go to Help -> Code Composer Studio Licensing Information. Then go to the Manage tab and look at the list of licenses. One of them should be the free license (unless the free license being used is the 90 day evaluation). You can either remove that license file from the dialog or better yet go to the location that it is loading the license from and move the file somewhere else (just in case there is a need to bring it back at some point).

Q: How to enter my license information on Mac OSX or CCSv6.2 for Linux?

These releases use a different licensing implementation. You will not see the Code Composer Studio licensing information menu item that launches the license configuration dialog. The license status is also not indicated in the bottom status bar when the license is accepted. Simply copy your license file into /ccsv6/ccs_base/DebugServer/license. If you are using the free license there is no action required. CCS will report if there are any problems, if it is not complaining then everything is working fine.

Updating

Q: How do I get updates to CCS?

By default CCS automatically checks for updates each time it starts. This update check will find updates to features that are already installed. For example it will find updates to the CCS IDE, compiler bug fix release, device support... It will not find new features including new feature revisions of the compiler.

Q: How do I search for new features for CCS?

The App Center which is accessible from the View menu is useful for finding new features. The App Center will highlight a selection of new features that are relevant to the device families that you have installed. This is the best way to install completely new features. For installing new compilers I prefer to use Help -> Install New Software and then select the Code Generation Tools update site. This will then find all the new compiler releases.

Q: How do I get TI-RTOS Updates?

The automatic update check will find incremental updates to TI-RTOS. For more significant releases you can use the App Center or Help -> Install New Software.

Q: When installing CCS updates I get a dailog that warns me that the feature is unsigned. Is this a problem?

Most but not all of the updates released by TI are signed, however not all of them are. Similarly if you are installing an update or feature from a partner it may not have been digitally signed. When this happens you get a warning from CCS noting that the feature has not been signed. In general these are safe to ignore but it is best to always make sure you know where the update is coming from before installing it.

License Server Administration

Q: Where do I go to get help with licensing?

First check the FAQs here. If that does not solve the problem please try the CCS forum. If you have an issue with an activation code or re-hosting you can email ccs_license_file_help@list.ti.com.

Q: What does the error "No socket connection to license server manager" mean?

In this case CCS was able to connect to the license server but there is an issue with the daemon that is running. One thing to check is to ensure that the manifest file that is included inside the zip file with the daemon executable was also extracted. If this was not done you will see "unlicensed" in the CCS title bar and if you go to Help -> Licensing Options you will get a dialog box with the error message.

Q: What OS can the license server run on?

Windows, Linux and Solaris are supported.

Q: Can we use a single license server to handle multiple geographic locations?

Yes you could use a single license server to distribute licenses to different locations as long as all locations are on the same network. CCS performs a limited number of license checks and thus does not significant network bandwidth.

Q: Can I run CCS on the same machine that is running the license server?

Yes. One way to achieve this is to edit the first line of the licenselocation.txt file to read @localhost as described in License_Server_Administration_for_CCS#Advanced_license_configurations

Q: How do I setup a 3 server redundant configuration?

Note that we do not recommend setting up a 3 server redundant configuration for CCS as it is not necessary. It is not necessary since if CCS cannot contact your license server it will still continue to operate and will just inform you that there is a problem. Thus the main purpose of the redundant configuration to protect against a server going down is not necessary. However if you still want to set this up the information on how to do so is below.

Important: When you activate your license specify the name of your primary server in the description box. This will ensure that this name gets populated within the "VENDOR_STRING" field for each feature in the license file. It is important to do this at the time of activation as that section of the file cannot be edited afterwards.

After receiving your license file open it up in a text editor.

Edit your license file so that the top of the file looks like this

1. Please Do not delete this comment line.

SERVER <name of primary server> <MAC address of primary Server> PRIMARY_IS_MASTER

SERVER <name of 2nd server> <MAC address of 2nd server>

SERVER <name of 3rd server> <MAC address of 3rd server>

VENDOR tidspssw USE_SERVER

Here is an example:

#Please Do not delete this comment line.

SERVER server1 0000ffee1111 PRIMARY_IS_MASTER

SERVER server2 0000ddee1111

SERVER server3 0000ccee1111

VENDOR tidspssw USE_SERVER

The server names other than the primary server are outside of the signed region of the license file so you can edit this file directly without needing TI to sign the file. It is also possible to change the MAC address associated with each server this way.

Q: I received a second floating license file. Can I combine the two?

Yes. The procedure is similar as if you were adding the first license file in your system.

• If using Windows, open a web browser and type the URL 127.0.0.1:8080 (if using the default port) to open the license manager GUI. Then follow steps 7 through 10 of the procedure above. Restart the license server for the changes to take effect.
• If using UNIX command line setup, stop the license server (./lmutil lmdown -vendor tidspssw) and follow steps 6 through 10 of the procedure above.
• If using UNIX Graphical setup, open a web browser and type the URL 127.0.0.1:8080 (if using the default port) to open the license manager GUI. Then follow steps 6 through 9 of the procedure above. Restart the vendor daemon for the changes to take effect.
  • Depending on the configuration of your system, you may have problems in importing the license due to the lack of write permissions to the licenses/tidspssw directory. Thus you will need to give momentarily write permissions to this directory before proceeding.

Not recommended. If you prefer to merge the two or more license files received, check an interesting discussion in this forum thread on how to do this by directly editing the license file. Be extremely careful when manually editing the files, as this is very error prone. Always make backup copies of your received license files!

If you have multiple license files from different vendors, check Chapter 18 of the The License Administration Guide (located here).

Q: How do I get the license server to start when the machine boots?

If you are finding that on a windows license server the license server is only starting when you are logged in and shuts down when you log out there could be a problem with your settings in LMTOOLS. Run LMTOOLS, go to the "Config Services" tab and make sure that the two check boxes at the bottom are checked.

Q: I keep getting "CreateProcess error code: 0x36b1 File=tidspssw.exe", why is this?

If this is a Windows machine running the license server the cause could be that you do not have the required version of some Microsoft runtime libraries installed. They can be obtained here. Specifically we need version 8.0.50727.762 of msvcrt80.dll

Q: What does "Port Number is not numeric" mean?

When entering a port number, it is likely there may be a trailing blank after the number. For example, port 27009 entered as '27009 ' instead of '27009'. The port number will unfortunately not be accepted if such a trailing blank exists. Once removed, the wizard should accept the port entry and connect to the license.

Q: I have multiple licenses from different vendors. Can I use them on the same server?

Yes. Check chapter 18 of the License Administration Guide.
IMPORTANT! Pay special attention to the port configuration of each license: they must be different for each vendor.

Q: I have a server running FreeBSD. Can I install the license server software?

Note: these instructions were kindly provided by a developer external to TI.

Yes. The FreeBSD version tested is:

FreeBSD <HOSTNAME> 9.3-STABLE FreeBSD 9.3-STABLE #0: Sun Oct 26 05:28:09 UTC 2014
root@snap.freebsd.org:/usr/obj/usr/src/sys/GENERIC  i386  32-bit

By default it does not have the LSB compatibility, thus when trying to run the setup program the following error occurs:

opt/license_server/bin # ./lmgrd
ELF binary type "0" not known.
./lmgrd: Exec format error. Binary file not executable.

Therefore additional packages must be installed:

1) pkg_add -r linux_base-f10
2)Edit /etc/rc.conf and add line linux_enable="YES"
3)Create symbolic link to mask the reguired library: ln -s /compat/linux/lib/ld-linux.so.2 /lib/ld-lsb.so.3
4)Brand the licesnce server executables so it runs with FreeBSD linux base libs:
brandelf -t linux /opt/license_server/bin/lmdiag
brandelf -t linux /opt/license_server/bin/lmgrd
brandelf -t linux /opt/license_server/bin/lmutil
brandelf -t linux /opt/license_server/bin/tidspssw
5) run lmgrd.

Projects

Q: I get a 'java.lang.NullPointerException' error when I try to import an existing project. How to I resolve this?

The most common cause of this error, when importing a project, is that the device specified in the project is not supported by the CCS version. Please check for updates to make sure you have the latest device support installed. Note that the error message has been improved starting with CCSv6.1.3 - instead of giving a vague 'java.lang.NullPointerException' message, it will explicitly say that the required device support is missing.

Q: Is it possible for CCS to override the TEMP and TMP environment variables that are used to determine where temporary files are placed during CCS build process?

During the CCS build process, gmake creates temporary batch files in the directory that corresponds to the user's TEMP and TMP System Environment Variables. The compiler also uses this directory for temporary files during compilation. CCS has the ability to let the user override these environment variables within the scope of CCS, so you don't have to modify your global System variables.

In CCS, open the menu Window->Preferences->C/C++->Build->Environment, and use the Add button to add the two variables:



Set the appropriate button to either "append" or "replace" variables to native environment.
On Windows, make sure to use backslashes (not forward slashes) in the paths. Also make sure the directory specified for the variables exists before running the build.

Q: Is it possible to convert a non-RTSC CCS project to a RTSC project, and vice-versa??

To convert a non-RTSC project to a RTSC project, simply add a RTSC configuration file (.cfg) to the project. CCS will ask if you want to enable RTSC support for the project and if you click "Yes", will convert it to a RTSC project.

It is not possible to convert a RTSC project to a non-RTSC project. In this case, you would need to start over by creating a new project.

Q: In some cases, during project build, I see the file ccsLinker.opt getting passed to the linker instead of the list of object files . What is this file and when is it used?

The 'ccsLinker.opt' file is generated by CCS and passed to the linker if the list of linker options and input files on the linker command line exceeds the maximum command line length of the host operating system. So, on Windows 7, a project with a large number of object files (.obj) may pass the linker options and input files to the linker via the ccsLinker.opt file rather than directly on the command line.

Debugging

Q: I can connect to my target but I get errors when I try to load my program. How do I fix this?

It is likely that a data verification error occurred. See the below FAQ for details.

Q: What does 'data verification error' mean?

See: Troubleshooting_CCS_-_Data_Verification_Errors

Q: What is the difference between Run/Resume and Free Run?

Free Run disables breakpoints before executing code on the target. Essentially you are running free of the debugger. You will not see any updates to any views or printfs if you perform a Free Run.

Q: How can I view non debuggable devices and routers in the Debug view?

The 'Debug' view hides non-debuggable devices by default. If you wish to view these devices:
Right-click in the Debug view and select "Show All Cores" in the context menu. A node called "Non Debuggable Devices" will then appear which you can expand to see the devices.

You can also set this within a given debug configuration so every time a debug session is launched with that target configuration (with or wihout associated project), the settings will take effect.

Go to CCS menu Run->Debug Configurations, and enable Non Debuggable Devices/Routers for the desired debug configuration.

Q: In multi-core devices, the CIO console prints out the CPU-core name after every output character when using assert() statements. Is there a way to prevent this?

In CCS all the CIO output goes to a single console by default, so CCS inserts the CPU-core name in front so that you can tell which core it is coming from. For assert() statements, the CPU-core name is added whenever a given CPU flushes its CIO buffer, making the output very difficult to read. There is an option to control this. To change the setting, go to menu Run->Debug Configurations. Select your debug configuration and uncheck the option to use the same console for all cores.

Q: I can't connect to my BeagleBone in CCS. What is happening?

Starting with revision A5, BeagleBone has a different USB Product ID (PID) of the embedded XDS100v2/Serial device which caused a compatibility break with CCS versions prior to CCSv5.1.1.

The supported versions are:

• CCSv5.1.0 for Linux: Revisions A3 and A4 can be used
• CCSv5.1.0 for Windows: Revisions A3 and A4 can be used
• CCSv5.1.1 for Linux: Revision A5 can be used
• CCSv5.1.1 for Windows: Revisions A3, A4 and A5 can be used

As a rule of thumb: update the Board's VID and use the latest version of CCSv5 and everything will be fine.

Note: the differences between Linux and Windows versions are due to updates to the TI emulation packs.

Note for Windows: the Rev A5 board does not use the XDS100 device driver installed with CCS. You need to download and install the standard FTDI D2XX device drivers from their website.

Q: My BeagleBone is Rev A5 and greater, but I still can't connect to my CCS . Why?

If you are trying to connect to the board and is getting this error:

Error connecting to the target:
(Error -151 @ 0x0)
One of the FTDI driver functions used during
the connect returned bad status or an error.
The cause may one or more of: invalid emulator serial number,
blank emulator EEPROM, missing FTDI drivers, faulty USB cable.
Use the xds100serial command-line utility in the 'common/uscif'
folder to verify the emulator can be located.
(Emulation package n.n.nnn.n)

Or is running the xds100serial utility and is getting:

C:\ti\ccsv6\ccs_base\common\uscif>xds100serial.exe
Scanning for XDS100 emulators...

VID/PID     Type         Serial #  Description
0403/6010   blank EEPROM unknown   no description available

Or the Windows control panel is showing two USB serial ports when you connect the board to the PC, then you have a BeagleBone board with a blank (non-programmed) FTDI device.
Therefore you need to program the board with a known good configuration. One can be found at CircuitCo's BeagleBone page.

Q: Can I have a debug two boards using two separate emulators at the same time?

Yes, this is possible. See this wiki topic for more details: Multi-Emulator Debug with CCSv5

Q: When debugging program in cache, I see cache lines getting invalidated. Why is this happening?

For any kind of action that involves setting software breakpoints (this also includes single-stepping or running to an address) on C6x, the corresponding cache line is invalidated. This is expected behavior since software breakpoints require the breakpoint opcode being written to the program memory at that address.

Q: I can connect to the target and load my program but the menus/icons for target execution (resume, suspend, step, etc) are not available (greyed out). What happened?

Most likely the Debug view was closed. See this FAQ.

If the Debug view was not closed, other potential issues could be related to the CCS Simple perspective as detailed in this E2E forum post.

Q: I get an error "Could not fetch next record for SQL statement" when trying to launch a debug session and running fsclean does not help. What else can I do?

Try deleting the target cache files as described in the troubleshooting guide.

Q: "Cannot read System Setup data from XML file <file path/name>.ccxml XML configuration file has no <configurations> root node." How do I resolve this?

Try creating and using a new target configuration file.

Q: "Is it possible for the Graphs view and properties to carry over from one CCS session to another, so when CCS is restarted the graphs come up again?

CCS saves the arrangement and contents of graph views when a debug session is terminated and restarted, but not when CCS itself is shut down and restarted. The solution to help streamline the input of graphing properties is the "Export" and "Import" buttons at the bottom of the graph properties. You could export the graph properties to a file and then import it back when you restart CCS.

General

Q: Where are the simulators? They don't seem to come with CCSv6

See: CCSv6 Changes.

Q: Are C54x devices supported?

CCSv6 does not support C54x devices. The last release to support C54x is CCSv5.5.

Q: Can I use DSP/BIOS with CCSv6?

Yes you can use DSP/BIOS with CCSv6 and build projects. Note that the Real-time analysis features are not supported with DSP/BIOS.

Q: How do I modify the maximum Java heap size for Eclipse?

Open <ccstudio.ini> in /ccsv6/eclipse (/ccsv6/eclipse/Eclipse.app/Contents/Eclipse for MacOS). Adjust the "-Xmx" argument. Note that CCS is a 32-bit application hence there is a cap on how much you can increase this value to with it having any effect. And also, having too high a value may cause other IDE stability problems depending on how much content is open in the workspace and what features are being used. If you have 4+ GB of system RAM, the "sweet spot" seems to be between 512m to 768m, and it works best for most users.

Q: Does CCSv6 support Windows 10?

Windows 10 will officially be supported starting with CCSv6.1.3 (available end of March 2016) and greater. However, many people have been working with older CCSv6 versions on Windows 10 with minimal issues.

Q: Is the XDS510LC emulator from Spectrum Digital supported with F2837xS/D and F2807x devices?

These devices have an Icepick router which is NOT supported by the XDS510LC. Please use one of the supported emulators as mentioned in this forum post.