HRCW location on PowerQUICC 8270 disagrees with FAE's instructions

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

HRCW location on PowerQUICC 8270 disagrees with FAE's instructions

Jump to solution
2,136 Views
Karl_H
Contributor III

I am configuring our application's HRCW.

 

We have flash resident from 0xFC000000 through 0xFFFFFFFF on the 60x bus.

 

60x Address lines A6 through A28 go directly to flash. A6 through A16 go nowhere else.

 

Our FAE told me that the HRCW location will be at the lowest addresses of flash. The implication is that on reset the processor reads the HRCW with A6 through A24 low, and our HRCW should be read from 0xFC000000. This is not what I am seeing. To get the proper reset config I need to program the HRCW at 0xFE000000. What's programmed at 0xFC000000 has no effect whatsoever. The implication is that A7 through A16 are low during the HRCW get, but A6 is high. I've tried it on more than one processor board, and they all behave this way. I have also determined that A6 is NOT stuck, since the processor can correctly resolve different values programmed at locations whose addresses differ only in A6.

 

Is this an expected (i.e., documented) behavior for HRCW in 8270 ??

 

Karl

0 Kudos
1 Solution
1,381 Views
Karl_H
Contributor III

Shortly after posting this query, I found the explanation in table 5-5 and the accompanying text in the reference manual. A6 is held high during the HRCW get under the assumption that there might be a number 7 slave daisy-chained in the system, and if A6 were low during the master's HRCW get, it would incorrectly select number 7 slave as well.

 

The project SW engineer is not happy that the HRCW has to go at the midpoint of flash space rather than at the beginning. He is worried that application code will end up having to straddle that location. I am less worried because I am skeptical that the applications programmers will ever exceed 32 Mbytes of code space. I don't think we have enough programmers to write and test that much code prior to our delivery date.

 

But even if it they do manage to pour in that much code, I expect that I will find a way to force the CodeWarrior linker to configure the application so that nothing collides with HRCW.

 

Karl

 

View solution in original post

0 Kudos
1 Reply
1,382 Views
Karl_H
Contributor III

Shortly after posting this query, I found the explanation in table 5-5 and the accompanying text in the reference manual. A6 is held high during the HRCW get under the assumption that there might be a number 7 slave daisy-chained in the system, and if A6 were low during the master's HRCW get, it would incorrectly select number 7 slave as well.

 

The project SW engineer is not happy that the HRCW has to go at the midpoint of flash space rather than at the beginning. He is worried that application code will end up having to straddle that location. I am less worried because I am skeptical that the applications programmers will ever exceed 32 Mbytes of code space. I don't think we have enough programmers to write and test that much code prior to our delivery date.

 

But even if it they do manage to pour in that much code, I expect that I will find a way to force the CodeWarrior linker to configure the application so that nothing collides with HRCW.

 

Karl

 

0 Kudos