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.

Sign up for GitHub

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

Jump to bottom

Consider deprecating SSE3 code #260

Open
myurkin opened this issue Jun 25, 2019 · 0 comments
Open

Consider deprecating SSE3 code #260

myurkin opened this issue Jun 25, 2019 · 0 comments
Assignees
@myurkin
Labels
comp-Logic Related to internal code logic maintainability Simplifies further code development (standardization, robustness) performance Simulation speed, memory consumption pri-Medium Worth assigning to a milestone sparse Sparse (non-FFT) mode
Milestone
1.5

Comments

@myurkin
Copy link
Member

myurkin commented Jun 25, 2019

Recent optimizations of imExp (#169) made SSE3 code only marginally faster than the standard c99 one (with compiler optimizations). It also showed that part of the SSE3 speedup is due to unsafe optimizations (can lead to precision loss).

Thus, better maintainability can be obtained if SSE3 code is removed altogether. However, it is worth studying where the remaining 10% speedup due to SSE3 comes from and optimizing the standard code accordingly.

/cc @jleinonen
@myurkin myurkin added comp-Logic Related to internal code logic performance Simulation speed, memory consumption pri-Medium Worth assigning to a milestone maintainability Simplifies further code development (standardization, robustness) sparse Sparse (non-FFT) mode labels Jun 25, 2019
@myurkin myurkin added this to the 1.5 milestone Jun 25, 2019
@myurkin myurkin self-assigned this Jun 25, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@myurkin myurkin

Labels
comp-Logic Related to internal code logic maintainability Simplifies further code development (standardization, robustness) performance Simulation speed, memory consumption pri-Medium Worth assigning to a milestone sparse Sparse (non-FFT) mode
Projects
None yet
Milestone
1.5
Development

No branches or pull requests
1 participant
@myurkin