EBI Bus error causing IVOR1trap, but only after SRAM write

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

EBI Bus error causing IVOR1trap, but only after SRAM write

990 Views
laurayindra
Contributor I

I was testing internal SRAM on a MPC567xK when the EBI stopped working:

 

There is 96KB of ramtest SRAM at 0x40021A00, and this error only occurs after a march12 memory test (bother reads and writes to ram) on ramtest.

 

The error, an IVOR1trap, only occurs on reading from EBI after the test on ramtest.

 

MCSR is 0x0008800C

BUS_WRERR is also set

MCSSR0 is  the last of the following:

    e_lis           r8,0x30000

    e_add16i   r3,r8,0x1408

    se_lwz       r3,0x0(r3)

    se_lhz        r31,0x4(r3)

MCSRR1 is 0x00008000

 

Here are the rest of my EBI settings:

MCR: 0x06000004

BMCR: 0x00001980

BR0: 0x20000827

OR0: 0xFFB00000

 

Why does this IVOR1trap only happen after the test on ramtest SRAM?  I don't understand how reading and writing to SRAM is changing the behavior of the EBI bus read.

Labels (1)
0 Kudos
3 Replies

794 Views
laurayindra
Contributor I

More info:

it is not the SRAM read/write causing the IVOR1trap on an EBI read.

the code called before the SRAM read/write is:

asm void mask_external_interrupts()

{

nofralloc

wrteei 0

blr

}

and the code called after the SRAM read/write but before the IVOR1trap is

asm void unmask_external_interrupts()

{

nofralloc

wrteei 1

blr

}

I can tell you that the EBI read that caused the IVOR1trap is in an interrupt based on an internal timer.

I don't know what this assembly code is doing.

0 Kudos

794 Views
davidtosenovjan
NXP TechSupport
NXP TechSupport

How it is related to EBI? You are testing internal SRAM, is it so?

0 Kudos

794 Views
laurayindra
Contributor I

The test of internal SRAM was wrapped by mask_external_interrupts() and unmask_external_interrupts().

Removing mask_external_interrupts() and unmask_external_interrupts(), and reducing internal SRAM testing to the unused section of internal SRAM (ramtest) stopped the IVOR1trap on read from EBI.

I don't know, and haven't had time to research, what the assembly in mask_external_interrupts() and unmask_external_interrupts() is doing to cause the IVOR1trap on read from EBI.  This read from EBI is being performed from an interrupt routine based on an internal timer.

0 Kudos