firmware is big endian dwords

[sylware]

In r600_rlc_init, it seems that the firmware would be stored as big endian dwords, and are programmed using cpu endianness?

I thought the GPU worked only in little endian?

What is the story about this?

[bridgman]

In r600_rlc_init, it seems that the firmware would be stored as big endian dwords, and are programmed using cpu endianness? I thought the GPU worked only in little endian? What is the story about this?

The driver code seems to read 32-bit words from the microcode image in memory and write them to the GPU memory mapped register without any intermediate processing, which I *think* avoids endianness issues. I'm a bit rusty on programming on big-endian parts though... I think the 68020 was the last big-endian part I worked on.

[sylware]

Well, it seems the firmware is stored as big endian dwords, but programmed using CPU endianness by using the Linux be32_to_cpup macro.

I noticed in r600_irq_init the use of "ih_cntl |= IH_MC_SWAP(IH_MC_SWAP_32BIT);".

The combination of all that would mean the IH controller work using CPU endianness and not little endian, all the time.

I would conclude the IH microcontroller do work using CPU endianness.