4 ms·
Sufficiently Smart Compilers are also called Superscalar CPU Architectures when they are implemented in hardware and come on a chip.
by garganzol 2y ago
Sufficiently Smart Compilers are also called Superscalar CPU Architectures when they are implemented in hardware and come on a chip.
- duskwuff 2y agoAnd those can succeed where the Sufficiently Smart Compilers failed because: 1) They can adapt to variable-latency memory operations, because they aren't an ahead-of-time compiler. 2) They can adapt to the capabilities of the hardware they're running on, because they ARE the hardware.
- cameldrv 2y agoYeah I think this is the biggest issue. Even if you know precisely the configuration of the target computer, you can’t know if there’s going to be a cache miss. A conventional CPU can reorder instructions at runtime to keep the pipeline full, but a VLIW chip can’t do this. In reality of course you don’t even know the precise configuration of the computer, and you don’t know the exact usage pattern of the software. Even if you do profile guided optimization, someone could use the software with different data that causes different branch patterns than in the profile, and then it runs slow. A branch predictor will notice this at runtime and compensate automatically.