Sample code for MSP430F5659

Nitesh Padiyar52 said:

I would like to know where i can find the compiler version change path?

It can be changed under PROJECT -> PROPERTIES

Then use one USCIs from your MSP and configure it as UART (example programs are available on the product website for the controller) and connect those pins to pin 12 and 14 of the FET. The device manager shows the COM port that was assigned to the backchannel (application) UART.

Hi Nitesh,

I see what you mean about the navigation issues if you are using the main pane for your navigation - I'll provide this feedback to the team. For now, please use the left hand pane which has a tree/folder structure that you can expand to find your examples. Alternately, try using the Cloud version of Resource Explorer found here: dev.ti.com.

In Resource Explorer MSPware, try looking under:

MSPWare > Libraries > USB Developer's Package > Example Projects. Here on the left tree you should select one of the types of USB example projects, e.g. CDC (Virrual COM Port) Examples, then select one of the examples, e.g. Echo Back to Host. All of this navigation is being done on the tree structure on the left pane. Here is an image:

Hope this helps. Here also is a link to the same place in the Cloud version of TI Resource Explorer: http://dev.ti.com/tirex/#/?link=MSPWare%2FLibraries%2FUSB%20Developers%20Package%2FExamples%2FCDC%20(Virtual%20COM%20Port)%20Examples%2FEcho%20Back%20to%20Host

Regards,

Katie

Hi Nitesh,

For MSP430 USB BSL there is a good application note that talks about considerations for doing a software invocation of the BSL in a USB application: www.ti.com/lit/pdf/slaa452

If the part is currently blank, USB BSL will automatically be called after a BOR reset.

For doing a software entry (instead of the normal hardware method), you can find examples of this throughout the app note, based on what kind of application model you have in your system (e.g., if you use USB in your system you may see section 4.2 and 4.3, but if you don't use USB for anything else, you could see section 4.1.)

Nitesh Padiyar52 said:

- Suppose i want to do firmware upgrade using software invoking the BSL then we need to do the changes in Application right? Please give your suggestion and link  

You would need to add the jump to BSL in your main application code, but you would not have to change anything in the BSL code that's already in the part.  

Nitesh Padiyar52 said:

- How to load the USB BSL if we do some changes?

As said above, you shouldn't have to make any changes to your USB BSL if you want to do the software entry - you can just add this jump to BSL in your main application code. If you need to change the BSL for some other reason (can you share what reason? ) then the source code is provided with www.ti.com/lit/pdf/slaa450 and you would need to load it the first time via JTAG/SBW (you shouldn't try to reprogram the BSL area using the BSL, because in the event of a failed load the BSL will be gone and you won't be able to get back in)

Nitesh Padiyar52 said:

 Can we write the USB BSL Host application as per our requirement? where i can find the host commands for writing the firmware?

USB BSL commands are described in www.ti.com/lit/pdf/slau319 - there you can find the packet structure, command bytes, etc - all the information you should need. We have a few different host example programs, including source code, as well - there is the BSL Scripter software found in www.ti.com/lit/zip/slau319, and there is also the Python USB BSL host software found here: http://software-dl.ti.com/msp430/msp430_public_sw/mcu/msp430/MSP430_USB_Developers_Package/latest/exports/Python_Firmware_Upgrader-5_00_01.zip both include source code so you could see how they work.

Nitesh Padiyar52 said:

- I want to do firmware upgrade by sending some command during run time, can i invoke USB BSL directly or i need any specific step since i am planning to use CDC class for my application where as USB BSL has the HID class. Please need your input.

The app note mentioned above www.ti.com/lit/pdf/slaa452 discusses this - are you having the CDC be communicating with a host PC normally, taking commands, and want to receive a command from your host and enter BSL based on this? In this type of application, I would probably look at section 4.3 Host Control, if you want for example to have your host send you a command over CDC and then invoke the BSL based on that. You could define that command to be anything you like. Your main application code would have to recognize this command, call the USB disconnect function, disable interrupts (clear GIE bit) and then jump to 0x1000. At that point the USB BSL will start and take over and will automatically re-enumerate the device as HID for the BSL without you needing to do anything special. Your host would need to be able to handle seeing the HID device and passing data to it. I've tested the software invocation like this before by using one of the CDC example programs, sending a command from a terminal program on the PC, then seeing the device de-enumerate and re-enumerate in the Device Manager as the HID device with the VID + PID that indicate the USB BSL. Then you can use something like the python host software I linked above, and run this to try using the BSL upgrade. That would show you if it is all working.

Hope this helps get you started.

Regards,

Katie

Hi Nitesh,

Nitesh Padiyar52 said:

Only doubt i had whether we can change the BSL code for some additional handling while firmware upgrade/downgrade? i.e. i have some configurations bytes in flash for v1 f/w and extra bytes added for v2 f/w, for upgrade and downgrade to maintain my configuration bytes as per versions, for this i need to change the BSL.

I'm not sure if I fully understand your need, so you may have to explain more. It sounds like you have two firmware versions, and they have some bytes in flash that are different. However, I don't see why that would mean you need to change the BSL - you can have your BSL host simply read or program more bytes or a different image without having to change the BSL code that is in the MSP. E.g. if you are wanting to do handling like - read bytes at x address, determine if it is v1 or v2 , and then based on this to write some different data or a different image - you could do all of that on the host side using the existing BSL code in the MSP.

Nitesh Padiyar52 said:

One more question - i have around 20-25 parameters to be read and 15-20 parameters to be written, can i use HID for my application? then which type of example project is good to use? For enumeration how i can intimate the host regarding my parameters. Or i can simply use command based HID application to read/write parameters. Please guide me.  

The Examples guide lays out what the different examples do: http://dev.ti.com/tirex/#/?link=MSPWare%2FLibraries%2FUSB%20Developers%20Package%2FExamples%20Guide

There is also a host-side GUI that goes with the MSP430USBDEVPACK that shows host side for HID datapipe and can be used with those examples: http://software-dl.ti.com/msp430/msp430_public_sw/mcu/msp430/MSP430_USB_Developers_Package/latest/exports/Java_HID_Demo-5_00_01.zip

Regards,

Katie

Nitesh Padiyar52 said:

Katie Thanks again for the support.

On my first concern i have few thoughts, See when my configurations are stored in some flash area i need it when i load the new firmware. But when BSL load the new firmware it erases whole flash i.e. all my previous configuration data is lost while writing the new application. Please let me know how i can get my old configurations when i am load with new firmware? Can i use secured flash area to store the my configurations?

On first release its ok, but when we do some bug fixes and we do field upgrade on upgrade we will loose the configurations. Even v1 has x bytes and new v2 y bytes (y>x) on upgrade i have lost previous data's please correct me if i am wrong.   if  downgrade is done where my current version v2 has y bytes and  v1 has x bytes (x<y) here also i will loose the data. How to handle this situation?

Hi Nitesh,

If you see the BSL user's guide www.ti.com/lit/pdf/slau319 section 3.5.2 Command Descriptions, you'll see that the description for the Mass Erase command says that it will not erase the Information Memory. Furthermore, you can provide the correct BSL password to get into the part without erasing it and do your read before doing the mass erase - now, on an incorrect BSL password, a mass erase will be performed - in this case however you can also see in the description for the RX Password command that the mass erase for an incorrect password also will not erase Information Memory. Therefore, I think you should store your version information in the Information Memory - that way you could keep it from being erased, and would have to explicitly use an Erase Block command to erase and then update the version number in Info memory if you ever needed to. I think this should address your concerns.

For placing your version number in INFO memory - you can do it like this in your application code:

#pragma DATA_SECTION(version, ".infoB")
const unsigned char version[4] = {0x12, 0x34, 0xAB, 0xCD};

You can then see if you generate your TI-txt binary image or by looking in your .map file, that the version information has been placed in this case at address 0x1900 which is in INFOB memory (see Memory Organization section of the datasheet www.ti.com/lit/gpn/msp430f5659 ). You could also use INFOA memory that has an additional lock bit if you further want to protect it from accidental erase. I hope that this helps give you some option for your version information to preserve it through the BSL mass erase.

Nitesh Padiyar52 said:

Regarding ADS1292EVM

For this EVM (do you mean the ADC1292ECG-FE board?), I don't really know anything about it or its code - I think you'll get better help for those questions in the Data converters forum where you're more likely to be in touch with the team that worked on this EVM. It is really an EVM for some ADCs it just happens to have an MSP430 on it as the microcontroller to work with them. So that ADC team will more likely be who knows about this code. I hope that this helps.

Regards,

Katie

Hi Nitesh,

Your use of INFOA looks ok to me, but note that INFO A is only 128 bytes (not 2kB that you said) - see www.ti.com/.../msp430f5659 table 6. Memory Organization - INFO each INFO segment A, B, C, D is 128 bytes, so INFO memory as a whole (A + B + C + D) is 512 bytes. How much config data do you have?

Regards,
Katie