/explore
|
You are here 
|
projectf.io | ||
| | | | |
domipheus.com
|
|
| | | | | One of the things which RPU has done from the start is keep the cpu pipeline very simple. It's a Fetch, Decode, Execute, [Memory], Writeback pipeline, but it does not run pipelined. Some operations really do require multiple cycles to execute, and in this post we figure out how to add this functionality into the CPU | |
| | | | |
pchaigno.github.io
|
|
| | | | | This post details the different versions of the eBPF instruction set. Their impact on program size and eBPF complexity is then evaluated. | |
| | | | |
probablydance.com
|
|
| | | | | Here is a rough approximation of float multiplication (source): float rough_float_multiply(float a, float b) { constexpr uint32_t bias = 0x3f76d000; return bit_cast(bit_cast(a) + bit_cast(b) - bias); } We're casting the floats to ints, adding them, adjusting the exponent, and returning as float. If you think about it for a second you will realize that since... | |
| | | | |
www.typescriptlang.org
|
|
| | | |||
