On Sat, 2007-03-17 at 15:29 -0800, Chris Lattner wrote:> On Sat, 17 Mar 2007, Vikram S. Adve wrote: > > Google's SoC is becoming a huge program, and I expect we'll get a few > > students wanting to do LLVM projects. (I already got one and asked him to > > write to llvmdev.) Is there anyone who can volunteer to put up a simple set > > of guidelines for LLVM SoC projects? (That will also get LLVM noticed by > > anyone searching for SoC.) It can be modeled after (but much shorter than) > > GNU's SoC page: > > I don't see any need for a formal set of guidelines, but if you want to > write one up, go for it. I expect the requirements to vary greatly based > on what the specific project is.I think about the best we could do is list some of the projects that we're interested in so candidates know what kinds of things to propose. However, we already have an OpenProjects page. What else is needed?> > -Chris >
On Mar 17, 2007, at 4:44 PM, Reid Spencer wrote:> I think about the best we could do is list some of the projects that > we're interested in so candidates know what kinds of things to > propose. > However, we already have an OpenProjects page. What else is needed?I think just two things are needed: (1) Just having a SoC page will make more people consider LLVM because, without it, they may just go to some other project that does. (2) Some of the projects on the OpenProjects page are too large for a summer. It would be helpful to identify suitable ones for SoC. In fact, I think #1 is more important than #2. We could just add a prominent notice to our home page asking interested students to post to llvmdev. --Vikram http://www.cs.uiuc.edu/~vadve http://llvm.cs.uiuc.edu/
A decent page about what projects might be completed in a summer would be helpful. Speaking as a prospective applicant, and recently inspired by the "New LLVM talk" posting (regarding dynamic languages), I wanted to know what might be within my grasp. I've only been lurking in this message group for a little while and reviewing some of the code, but it's difficult to gauge the potential complexity of projects. -Gabe