Hacker News Reader

Another aspect of the x86_64 architectural swamp is that Intel deliberately makes their optimized math libraries fail if run on a chip that is not 'INTEL INSIDE'. That block totally ignores cpu feature bits.

I've always wondered if virtualization software vendors made everything claim to be INTEL INSIDE, just to avoid this problem.

View on HN

I know the RPCS3 (PS3 emulator) encountered some differences in a floating point instruction where Intel and AMD had different precision, causing de-synchronizations in LittleBigPlanet multiplayer.

Could be down to this.

Are you talking about the 8087 transcendental instructions? Which are well known and well-written software doesn't have a problem with it.

Nope, there are subtle differences between Intel and AMD that make it a Bad Idea to fake the vendor.

There are? We aren't talking about feature bits, and the only known software that checks the vendor are Intel's math libraries.

They are minor but enough to complicate things. For example Intel rejects the SYSCALL instruction outside 64-bit mode.

Huh. A quick google says that's a feature bit and there's also an OS call to turn it on.

Reply on news.ycombinator.com