BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160728Z
LOCATION:D175
DTSTART;TZID=America/Chicago:20181112T120000
DTEND;TZID=America/Chicago:20181112T123000
UID:submissions.supercomputing.org_SC18_sess176_ws_llvmf106@linklings.com
SUMMARY:LLVM and the Automatic Vectorization of Loops Invoking Math Routin
 es: -fsimdmath
DESCRIPTION:Workshop\nProgram Transformation, Programming Systems, Worksho
 p Reg Pass\n\nLLVM and the Automatic Vectorization of Loops Invoking Math 
 Routines: -fsimdmath\n\nPetrogalli, Walker\n\nThe vectorization of loops i
 nvoking math function is an important optimization that is available in mo
 st commercial compilers. This paper describes a new command line option, -
 fsimdmath, available in Arm Compiler for HPC, that enables auto-vectorizat
 ion of math functions in C and C++ code, and that will also be applicable 
 to Fortran code in a future version.\n\nThe design of -fsimdath is based o
 n open standards and public architectural specifications. The library that
  provides the vector implementation of the math routines, libsimdmath.so, 
 is shipped with the compiler and based on the SLEEF library libsleefgnuabi
 .so. SLEEF is a project that aims to provide a vector implementation of al
 l C99 math functions, for a wide variety of vector extensions and architec
 tures, across multiple platforms.\n\nThis feature is very important for HP
 C programmers, because the vector units of new CPUs are getting wider. Whe
 ther you are targeting Intel architectures with the AVX512 vector extensio
 n, or Arm architectures with the Scalable Vector Extension, good quality a
 uto-vectorization is of increasing importance.\n\nAlthough -fsimdmath has 
 been implemented in a commercial compiler, it has been designed with porta
 bility and compatibility in mind, so that its use is not limited only to t
 he vector extensions of the Arm architectures, but can be easily introduce
 d as a major optimization for all the vector extensions that LLVM supports
 .\n\nIf accepted upstream, this new feature will enlarge the set of loops 
 that LLVM will be able to auto-vectorize.
URL:https://sc18.supercomputing.org/presentation/?id=ws_llvmf106&sess=sess
 176
END:VEVENT
END:VCALENDAR