[Sbcl-devel] Interested in GSoC Vectorized sequence operations project

[Nitika Daga]

Hello!

My name is Nitika Daga. I'm currently a second year student at UC Berkeley, studying Computer Science. I'm interested in participating in GSoC with SBCL this summer, especially the project involving vectorizing operations using SSE intrinsics.  In one of my current CS classes, we just finished a project that involved parallelization with SSE intrinsics (the project involved performance optimization for matrix multiplying), which makes me uniquely suited for this project. I also worked a bit in X86 assembly, although not as much as I have in SSE intrinsics in C.

In terms of my general technical background I am experienced in Python, Java, C, and Scheme. I have taken classes in data structures and machine structures, both of which would help me in understanding how your compiler works.

I'm currently in the process of filling out my student proposal for this project, but I was hoping you all could provide some advice/further information about this project that could help me, specifically some concrete examples of kinds of operations you were hoping to vectorize.

Thank you!

Nitika Daga

[Paul Khuong]

Nitika Daga wrote:
> I'm currently in the process of filling out my student proposal for this
> project, but I was hoping you all could provide some advice/further
> information about this project that could help me, specifically some
> concrete examples of kinds of operations you were hoping to vectorize.

Here' what I wrote for another student who's thinking of working on that
project:
http://thread.gmane.org/gmane.lisp.steel-bank.devel/17199/focus=17201 .

There's been a lot of discussion about GSoC in the past weeks, on
sbcl-devel. Combing through the archives might be helpful.

Paul Khuong

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr
_______________________________________________
Sbcl-devel mailing list
Sbcl-...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sbcl-devel