[Pl-seminar] Fwd: 21st February: Manas Thakur: Compare Less, Defer More: Scaling Value-Contexts Based Whole-Program Heap Analyses

[Aviral Goel]

Hi,

The talk is happening in 43 minutes.

*Location:* Ryder 431
*Time:* 10:00 AM to 11:30 AM
*Speaker:* Manas Thakur
*Faculty Host:* Jan Vitek


---------- Forwarded message ---------
From: Aviral Goel <goel.av at husky.neu.edu>
Date: Tue, Feb 19, 2019 at 1:42 PM
Subject: [Pl-seminar] 21st February: Manas Thakur: Compare Less, Defer
More: Scaling Value-Contexts Based Whole-Program Heap Analyses
To: <pl-seminar at ccs.neu.edu>


*Date:* Thursday, February 21st 2019
*Location:* Ryder 431
*Time:* 10:00 AM to 11:30 AM
*Speaker:* Manas Thakur
*Faculty Host:* Jan Vitek

Compare Less, Defer More: Scaling Value-Contexts Based Whole-Program Heap
Analyses
<http://prl.ccs.neu.edu/seminars.html#thakur-compare-less-defer-more>

*Manas Thakur*

*Abstract*

The precision of heap analyses determines the precision of several
associated optimizations, and has been a prominent area in compiler
research. It has been shown that context-sensitive heap analyses are more
precise than the insensitive ones, but their scalability continues to be a
cause of concern. Though the value-contexts approach improves the
scalability of classical call-string based context-sensitive analyses, it
still does not scale well for several popular whole-program heap analyses.
In this talk, I will discuss our three-staged approach that lets us scale
complex whole-program value-contexts based heap analyses for large
programs. The ideas involved are based on an important observation that we
do not need to compare the complete value-contexts at each call-site. Apart
from this, I will also give an overview of another work, named PYE (for
"precise-yet-efficient”). PYE helps obtain precise results during
just-in-time compilation efficiently, by splitting the work between the
static Java compiler and the C2 compiler of the HotSpot JVM.

*Bio*

Manas is a PhD student under V. Krishna Nandivada at PACE Lab, IIT Madras.
His current research focuses on balancing the precision-scalability
tradeoffs while analyzing programs written in object-oriented languages. He
has worked with the Soot optimization framework for static analyses, and
the HotSpot JVM for just-in-time analyses. Apart from work, he writes
articles on GitHub and Medium (technical and otherwise!), and experiments
with Vim plugins when feeling geeky.

Best,
Aviral
-------------- next part --------------
HTML attachment scrubbed and removed