Search found 5 matches
- 2021-09-10, 21:36:17
- Forum: forwardcom forum
- Topic: Branch instruction with a "one-hot" bitmask equality comparison
- Replies: 1
- Views: 19355
Branch instruction with a "one-hot" bitmask equality comparison
This is a proposal for a branch instruction that allows flexible equality comparisons. The proposed branch instruction would have a bitmask, where each bit corresponds to a value of a source register (for example, 32 bit bitmask corresponds to source register range from 0 to 31) - if the correspondi...
- 2021-06-28, 16:26:04
- Forum: forwardcom forum
- Topic: Universal boolean instruction
- Replies: 5
- Views: 25109
Re: Universal boolean instruction
The E2 instruction format would even allow for a four input LUT, since it has a 16 bit immediate that can be used as a truth table. That would mean that the destination register would also serve as a source register (A = A AND B AND C AND D) or alternatively, it could be used as a three input LUT wi...
- 2021-06-08, 22:17:49
- Forum: forwardcom forum
- Topic: Load From Const Array Instruction
- Replies: 3
- Views: 12054
Load From Const Array Instruction
Programs contain many Read Only Arrays/Lookup Tables. Currently they are stored in static data memory. To make a load from such an array, CPU needs to calculate the address as pointer + index and load the value from the dcache. This solution requires a pointer and doesn't exploit locality very well....
- 2020-11-13, 18:44:05
- Forum: forwardcom forum
- Topic: input/output instructions
- Replies: 8
- Views: 35753
Re: input/output instructions
"Code and data memory has a 64-bit address space, while we need much less than 2 64 input/output addresses." I'm not so sure. AMD just announced their Smart Access Memory technology, which essentially gives the CPU direct access to the GPU memory (this improves performance in various appli...
- 2020-05-09, 9:00:32
- Forum: forwardcom forum
- Topic: Putting it on real hardware
- Replies: 7
- Views: 23878
Re: Putting it on real hardware
What about Western Digitals SweRV? https://github.com/chipsalliance/Cores-SweRV