Virtual Interrupt Vector for bootloader project in IAR

The RL78 does not support relocating the interrupt vectors, any interrupt will vector to the lower 64KB of the code-flash (due to only 16-bit intr vectors).  As such, where the interrupt processing needs to go has to be handled in software by some method.

If your boot-loader doesn't use any interrupts, the solution is pretty easy - vector to a long-jump that resides somewhere in the application.

If the boot-loader uses an interrupt, the ISR will need to check something to decide whether to handle the ISR in the boot-loader's code or the application's code.

Hello JimB and thanks for your answer.

Both the bootloader and the application are using interruptions.

If I understood you correctly, I need to define a handler for every type of interruptions in the bootloader. And each time and interruption occures, I check if I'm running the bootloader or the application and I forward the interruption into the handler of the bootloader or the application. Am I right?

Or do you have an example showing how to control the ISR to handle boot code or app code? (I can easily get a flag from the flash to check whether am I in the boot or the app).

This topic is a little blurry for me, sorry if it basic questions

Thanks !

EDIT : I went through this documentation : https://www.renesas.com/jp/ja/document/apn/rl78-dual-image-bootloader-example . I understand theorically what to do but in the practice I don't know how to implement it. I checked 3.4 Vector Table and 3.5 ISR Processing

Hi, dear Lazy Dridri. 

Please check the post blow, it might clarify the topic. 

 https://community.renesas.com/mcu-mpu/rl78/f/rl78-forum/31553/2-project-codes-to-be-placed-in-the-same-rl78g23/112182#112182

If you needed further assistant let us know. and about the linker, please clarify the compiler that you are using in IAR. 
I hope you find that useful. 

Best regards. 

If this response or any other user's response answers your concern, kindly verify the answer. Thank you!

Renesas Engineering Community Moderator
https://community.renesas.com/
https://academy.renesas.com/
https://en-support.renesas.com/knowledgeBase/

Hello Hossein and thanks for your answer

I read this thread but it seems like it handle the update of the bootloader itself, not the application. So I still have an issue when dealing with an interruption, it always goes back to the bootloader, even if I'm in the application. Because I can't swap full application with bootloader I believe.

I'm using IAR Embedded Workbench 7.0.5.3137

Ok I may have undestand why the boot swap is useful in my case thanks to this post : https://community.renesas.com/mcu-mpu/rl78/f/rl78-forum/29913/rl78-l13-bootloader-jump-to-app-problem/110915#110915 

So, I'll try this : 

• Update flash linker to make sure bootloader is contained in 0x0000-0x3FFF area
• Flash the bootloader using classic E1 tool
• Send data to bootloader and make sure that the Start Program of the application is contained in 0x4000-0x7FFF
• Jump to application from bootloader AND perform boot swap (so Start Program of the application will be connected to the ISR)

This way, when I want to update application, I can swap again both boot to use the one of the bootloader.

Might not be that clear but I'll try this and keep you update. Do you think this process can make the job?

The boot swap is to be used to update the existing boot cluster area with a new boot cluster. It is not used to swap the application with the boot cluster. The boot cluster always exists at 0x0000 0000.

Ok I see. Then I really don't know how to make sure interrupts are connected to application when I'm running the application and connected to the bootloader when I'm running the bootloader.

Do I need to hardcode an interrupt handler like this?

if(Interrupt_X)
{
        if(is_running_boot)
            Boot_Interrupt_X();
        else
            App_Interrupt_X();
}
        else if(Interrupt_Y)
{
        ...
}
...

Do you need to have the same interrupts on both bootloader and application with different functionality ?

It is the same interruption with the same functionality.

For example, I have this interrupt : 

#pragma vector = INTTM00_vect
static __interrupt void TM00_isr (void) {
	Schedule = 1;
	TMIF00 = 0;
}

It is the same in the bootloader AND in the application. But when I start application from bootloader, it seems like we the interruption is calling the bootloader

All interrupts are always handled by the vector table which resides at the boot cluster, this cannot be changed. Please see section 4 of this app note:

https://www.renesas.com/us/en/document/apn/rl78-bootloader-one-image-r11an0470eu0100-rev100