|
From: Greg W. <gvw...@cs...> - 2006-10-04 10:11:05
|
Hi, I have two exceptional undergraduate students who are looking for a one-term final-year project. Both are primarily interested in systems, so we thought that having them help out with Helgrind would be a good fit. I mailed the valgrind-dev list to try to make contact with Helgrind's developers, but have heard nothing back; I'd be grateful if someone could let me know (a) whether Helgrind is in fact being revived, (b) if so, whether there's room for talented young hands, or (c) if not, whether there are tasks in Valgrind's development that could be peeled off and given to a pair of students. We're under a bit of time pressure to get a project nailed down... Thanks, Greg Wilson University of Toronto |
|
From: Tom H. <to...@co...> - 2006-10-04 10:20:55
|
In message <452...@cs...>
Greg Wilson <gvw...@cs...> wrote:
> I have two exceptional undergraduate students who are looking for a
> one-term final-year project. Both are primarily interested in systems,
> so we thought that having them help out with Helgrind would be a good
> fit. I mailed the valgrind-dev list to try to make contact with
> Helgrind's developers, but have heard nothing back; I'd be grateful if
> someone could let me know (a) whether Helgrind is in fact being revived,
> (b) if so, whether there's room for talented young hands, or (c) if not,
> whether there are tasks in Valgrind's development that could be peeled
> off and given to a pair of students. We're under a bit of time pressure
> to get a project nailed down...
There has been some work recently on a new tool called drd which
might take the place of helgrind? Maybe you should be looking at
that instead - there has been some discussion on the dev list.
Alternatively, there is a list of project ideas on the web site
somewhere.
Tom
--
Tom Hughes (to...@co...)
http://www.compton.nu/
|
|
From: Nicholas N. <nj...@cs...> - 2006-10-04 11:15:14
|
On Wed, 4 Oct 2006, Tom Hughes wrote: >> I have two exceptional undergraduate students who are looking for a >> one-term final-year project. Both are primarily interested in systems, >> so we thought that having them help out with Helgrind would be a good >> fit. I mailed the valgrind-dev list to try to make contact with >> Helgrind's developers, but have heard nothing back; I'd be grateful if >> someone could let me know (a) whether Helgrind is in fact being revived, >> (b) if so, whether there's room for talented young hands, or (c) if not, >> whether there are tasks in Valgrind's development that could be peeled >> off and given to a pair of students. We're under a bit of time pressure >> to get a project nailed down... > > There has been some work recently on a new tool called drd which > might take the place of helgrind? Maybe you should be looking at > that instead - there has been some discussion on the dev list. Yes, it could be a very interesting project. It would also be a very challenging one, certainly without any guarantee of a complete final outcome, but maybe that isn't a problem. > Alternatively, there is a list of project ideas on the web site > somewhere. There was a recent good project suggestion that hasn't yet made it to the page, from Brant Gurganus: there are resource allocators at higher levels than malloc/free pairs that would be nice to check, preferrably in an easily extensible fashion. I'd like to see a way of specifying resource allocators and deallocators that are from somebody else's code. For example, I'd like a tool that handles finding unmatched malloc/free, fopen/fclose, open/close, XOpenDisplay/XCloseDisplay, etc.> I don't know if this one would be large enough, however. Nick |
|
From: Dirk M. <dm...@gm...> - 2006-10-04 20:12:34
|
On Wednesday, 4. October 2006 13:14, Nicholas Nethercote wrote: > Yes, it could be a very interesting project. It would also be a very > challenging one, certainly without any guarantee of a complete final > outcome, but maybe that isn't a problem. There is an easier one: implementing lockdep for valgrind: http://www.linuxhq.com/kernel/v2.6/18/Documentation/lockdep-design.txt This should be more or less straight forward, except for the locking order validation algorithm (for which white papers exist). Dirk |