Paper available: The Predictability of Libraries

Title: The Predictability of Libraries
Authors: Brad Calder, Dirk Grunwald and Amitabh Srivastava
Availability:


http://esl.cs.colorado.edu/BranchPrediction/paper_pubs.html


[ I'm out of the country for a while. Our web server is in a continual
    state of disrepair; if you can't get the file, send email and we'll
    get it to you in 3 weeks or so]


Abstract:


Profile-based optimizations are being used with increasing frequency.
Profile information can be used to improve instruction scheduling,
code layout, and to increase instruction level parallelism. These
optimizations have been shown to be effective when they are applied to
the same program from which the profile was gathered. However, it is
an open question how profile-based optimizations should be applied to
library subroutines. If many programs use libraries in the same way,
it may be possible to ``pre-optimize'' a library, or to use an
optimized shared library.


This study examines the use of commonly used libraries among 43 C and
FORTRAN programs to see if the libraries have common behavior across
different programs. We examine the behavior of the most commonly used
Unix libraries on Digital Unix. We found that libraries have very
predictable behavior between applications. This implies that
profile-based compiler optimizations may be effective for libraries
across applications. Therefore, one can use profile optimizations on
shared and non-shared libraries before they are shipped, allowing a
program using those libraries to take advantage of profile-based
optimization without having to gather any profiles. All results in
this study are shown using branch misprediction rates. We feel this
metric indicates the likelihood that programs have similar behavior,
and allows comparison to earlier studies comparing profile-based
branch prediction between runs of the same program.


--

Return to the comp.compilers page.
Search the comp.compilers archives again.