Loading…
This event has ended. View the official site or create your own event → Check it out
This event has ended. Create your own
View analytic
Thursday, November 3 • 4:15pm - 5:00pm
Enhancing LLVM’s Floating-Point Exception and Rounding Mode Support

Sign up or log in to save this to your schedule and see who's attending!

In this BoF session, we will discuss ways in which LLVM can be extended to support non-default floating-point behavior. Topics will include respecting FP rounding modes in optimization passes, preserving FP exception status, avoiding false FP exceptions and enabling run-time handling of FP exceptions. 

Speakers
avatar for Andy Kaylor

Andy Kaylor

Sr. Software Engineer, Intel
I've been a tools developer at Intel for 17 years and have been working with LLVM since 2012, contributing to areas such as MCJIT, LLDB and Windows exception handling. I'm about to dive into LLVM's representation and handling of floating point operations.
avatar for David Kreitzer

David Kreitzer

Principal Engineer, Intel
I have spent the majority of my adult life getting the Intel compiler to generate superb code for x86 processors. I developed many major pieces of functionality in the Intel compiler's back end including its register allocator. My current focus is to draw on that experience to help improve LLVM's generated code for x86.



Thursday November 3, 2016 4:15pm - 5:00pm
3 - BoF (Rm LL21CD)

Attendees (18)