Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Different output on x86_64 and i686 #22

Open
erikrk opened this issue May 6, 2015 · 2 comments
Open

Different output on x86_64 and i686 #22

erikrk opened this issue May 6, 2015 · 2 comments

Comments

@erikrk
Copy link
Contributor

erikrk commented May 6, 2015

Output files generated on different CPU architectures are not identical. This is probably due to rounding errors. The test environment does not handle this. The current MD5 checksums correspond to output generated on x86_64, Consequently, "make check" fails in i686.

@erikrk
Copy link
Contributor Author

erikrk commented May 6, 2015

I think this is the reason.

http://www.network-theory.co.uk/docs/gccintro/gccintro_70.html

It's all about the internal precision of the FPU. The x86_64 code uses SSE instructions which use 64-bit precision internally and the i686 code uses x87 instructions which use 80-bit precision internally.

I don't think it really matters for practical purposes, but it's kind of annoying that you get different results.

@mmroden
Copy link
Contributor

mmroden commented May 6, 2015

Yeah, different outputs for different processors doesn't worry me.
Different outputs for the same processor would be terrifying.

On Wednesday, May 6, 2015, Erik Karlsson [email protected] wrote:

I think this is the reason.

http://www.network-theory.co.uk/docs/gccintro/gccintro_70.html

It's all about the internal precision of the FPU. The x86_64 code uses SSE
instructions which use 64-bit precision internally and the i686 code uses
x87 instructions which use 80-bit precision internally.

I don't think it really matters for practical purposes, but it's kind of
annoying that you get different results.

Reply to this email directly or view it on GitHub
#22 (comment).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants