GLUE on FPGA with H5 Phoenix platform

[ijor]

As I mentioned in the other thread, I like to test not only things that should work, but also things that should not. Often this is more difficult to test, but it is important. Closure might fail in a very rare wakeup state combination. This happens in original hardware and also on these cores, confirming better accuracy:

Bad display at Closure on rare Wakestate 2 combination.

ClosureWs2BadSfifterWs.jpg (319.26 KiB) Viewed 2526 times

Btw, don't try this at home. It might require dozens of power cycles to get the right (wrong) wakestate.

[Cyprian]

this issue looks cool,

[Icky]

Reminds me of one of the states the FPGA Shifter was getting into

IMG_5915.png (864.85 KiB) Viewed 2515 times

[ijor]

Reminds me of one of the states the FPGA Shifter was getting into

Hi Icky. Yes, looks similar although not exactly the same behavior. In any case, it is true that this issue is more Shifter than GLUE or MMU specific. I still need to do a couple of things on the 3 other cores (GLUE, MMMU & BLITTER) and hopefully will try the SHIFTER core on your Phoenix Shifter board shortly.