Skip to content

Conversation

@JeffBezanson
Copy link
Member

@JeffBezanson JeffBezanson commented Mar 17, 2020

This speeds up codegen of functions that use many duplicate llvmcalls, such as in LoopVectorization/SIMDPirates. From issue #35131

For me this takes the time to run the LoopVectorization tests from 5'40" to 5'15" (note this is a somewhat slow machine).

@JeffBezanson JeffBezanson added compiler:codegen Generation of LLVM IR and native code latency Latency labels Mar 17, 2020
@JeffBezanson JeffBezanson requested a review from vtjnash March 18, 2020 23:55
Copy link
Member

@vtjnash vtjnash left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This variable needs to go away (soon), but might as well use the existing name, so that we can grep for it.

@JeffBezanson JeffBezanson merged commit 5d5f799 into master Mar 20, 2020
@JeffBezanson JeffBezanson deleted the jb/llvmcallnames branch March 20, 2020 19:30
KristofferC pushed a commit that referenced this pull request Mar 23, 2020
@KristofferC KristofferC mentioned this pull request Mar 23, 2020
27 tasks
KristofferC pushed a commit that referenced this pull request Mar 23, 2020
oxinabox pushed a commit to oxinabox/julia that referenced this pull request Apr 8, 2020
ravibitsgoa pushed a commit to ravibitsgoa/julia that referenced this pull request Apr 9, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

compiler:codegen Generation of LLVM IR and native code latency Latency

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants