|
From: Mayank J. <con...@gm...> - 2008-12-14 14:45:04
|
Hi Developers, Valgrind is such a useful tool to detect the memory leak. I want to use this on FreeBSD amd64 bit architecture. Is there any plans to port this for amd64. I am very interested to get this feature and I guess there are lots of other people who are also interested with this. Can any body let me know or share some information how can I port valgrind for amd64. Any suggestions and help will be greately appreciated. Thank You Mayank Jain |
|
From: Tom H. <to...@co...> - 2008-12-14 15:47:36
|
Mayank Jain wrote: > Valgrind is such a useful tool to detect the memory leak. I want to use > this on FreeBSD amd64 bit architecture. Is there any plans to port this > for amd64. I am very interested to get this feature and I guess there > are lots of other people who are also interested with this. Can any body > let me know or share some information how can I port valgrind for amd64. > Any suggestions and help will be greately appreciated. Valgrind works find on amd64 and has done for a long time. What it doesn't do is work on FreeBSD though. There was an attempt to port it a long time ago, but it was never merged back and I don't know what the current status of it is. Perhaps you're talking about that old FreeBSD port, which probably doesn't support amd64, rather than the current core valgrind code, which does? Tom -- Tom Hughes (to...@co...) http://www.compton.nu/ |
|
From: Filipe C. <fi...@gm...> - 2008-12-14 16:25:18
|
Also, valgrind is already running (albeit a bit unstable and incomplete) on Mac OS X, maybe those efforts can be adapted to FreeBSD. There was a patch this week to start merging the work and abstracting some things from valgrind's core so the Mac OS X port could be "plugged in". Regards, F On 14 Dec, 2008, at 15:47, Tom Hughes wrote: > Mayank Jain wrote: > >> Valgrind is such a useful tool to detect the memory leak. I want to >> use >> this on FreeBSD amd64 bit architecture. Is there any plans to port >> this >> for amd64. I am very interested to get this feature and I guess there >> are lots of other people who are also interested with this. Can any >> body >> let me know or share some information how can I port valgrind for >> amd64. >> Any suggestions and help will be greately appreciated. > > Valgrind works find on amd64 and has done for a long time. > > What it doesn't do is work on FreeBSD though. There was an attempt to > port it a long time ago, but it was never merged back and I don't know > what the current status of it is. > > Perhaps you're talking about that old FreeBSD port, which probably > doesn't support amd64, rather than the current core valgrind code, > which > does? > > Tom > > -- > Tom Hughes (to...@co...) > http://www.compton.nu/ > > ------------------------------------------------------------------------------ > SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, > Nevada. > The future of the web can't happen without you. Join us at MIX09 to > help > pave the way to the Next Web now. Learn more and register at > http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/ > _______________________________________________ > Valgrind-developers mailing list > Val...@li... > https://lists.sourceforge.net/lists/listinfo/valgrind-developers |
|
From: Peter W. <pe...@we...> - 2008-12-17 03:37:36
|
On Sun, Dec 14, 2008 at 7:47 AM, Tom Hughes <to...@co...> wrote: > Mayank Jain wrote: > >> Valgrind is such a useful tool to detect the memory leak. I want to use >> this on FreeBSD amd64 bit architecture. Is there any plans to port this >> for amd64. I am very interested to get this feature and I guess there >> are lots of other people who are also interested with this. Can any body >> let me know or share some information how can I port valgrind for amd64. >> Any suggestions and help will be greately appreciated. > > Valgrind works find on amd64 and has done for a long time. > > What it doesn't do is work on FreeBSD though. There was an attempt to > port it a long time ago, but it was never merged back and I don't know > what the current status of it is. > > Perhaps you're talking about that old FreeBSD port, which probably > doesn't support amd64, rather than the current core valgrind code, which > does? I've got a nearly complete FreeBSD port that does both 32 and 64 bit. I'm in the process of making some changes to FreeBSD itself to better support some things that Valgrind needs. Other than that it runs very well. It does not work reliably with pthreads though due to an architectural issue (libvex doesn't do atomic operations, which doesn't mix well with the umtx syscalls that FreeBSD's threads uses.) It is well past time to make this more public. While it is sitting in the FreeBSD perforce tree, that isn't exactly easy to get to for random people. I'll tidy things up and make it available. -- Peter Wemm - pe...@we...; peter@FreeBSD.org; pe...@ya...; KI6FJV "All of this is for nothing if we don't go to the stars" - JMS/B5 "If Java had true garbage collection, most programs would delete themselves upon execution." -- Robert Sewell |
|
From: Mayank J. <con...@gm...> - 2008-12-17 12:07:40
|
Hi Peter, That's a gr8 news you have done so much for porting it for FreeBSD. Can you please share the source code with me so that I can help you with solving the issue. Thank You Mayank Jain On 12/17/08, Peter Wemm <pe...@we...> wrote: > On Sun, Dec 14, 2008 at 7:47 AM, Tom Hughes <to...@co...> wrote: >> Mayank Jain wrote: >> >>> Valgrind is such a useful tool to detect the memory leak. I want to use >>> this on FreeBSD amd64 bit architecture. Is there any plans to port this >>> for amd64. I am very interested to get this feature and I guess there >>> are lots of other people who are also interested with this. Can any body >>> let me know or share some information how can I port valgrind for amd64. >>> Any suggestions and help will be greately appreciated. >> >> Valgrind works find on amd64 and has done for a long time. >> >> What it doesn't do is work on FreeBSD though. There was an attempt to >> port it a long time ago, but it was never merged back and I don't know >> what the current status of it is. >> >> Perhaps you're talking about that old FreeBSD port, which probably >> doesn't support amd64, rather than the current core valgrind code, which >> does? > > I've got a nearly complete FreeBSD port that does both 32 and 64 bit. > I'm in the process of making some changes to FreeBSD itself to better > support some things that Valgrind needs. Other than that it runs very > well. It does not work reliably with pthreads though due to an > architectural issue (libvex doesn't do atomic operations, which > doesn't mix well with the umtx syscalls that FreeBSD's threads uses.) > > It is well past time to make this more public. While it is sitting > in the FreeBSD perforce tree, that isn't exactly easy to get to for > random people. I'll tidy things up and make it available. > > -- > Peter Wemm - pe...@we...; peter@FreeBSD.org; pe...@ya...; KI6FJV > "All of this is for nothing if we don't go to the stars" - JMS/B5 > "If Java had true garbage collection, most programs would delete > themselves upon execution." -- Robert Sewell > |
|
From: Bart V. A. <bar...@gm...> - 2008-12-17 12:24:51
|
On Wed, Dec 17, 2008 at 4:05 AM, Peter Wemm <pe...@we...> wrote: > [...] It does not work reliably with pthreads though due to an > architectural issue (libvex doesn't do atomic operations, which > doesn't mix well with the umtx syscalls that FreeBSD's threads uses.) [ ...] Both the Helgrind and DRD tools have some support for atomic operations. I don't know from the top of my head when exactly support for atomic operations was added to VEX. Bart. |
|
From: Julian S. <js...@ac...> - 2008-12-17 12:51:11
|
On Wednesday 17 December 2008, Bart Van Assche wrote: > On Wed, Dec 17, 2008 at 4:05 AM, Peter Wemm <pe...@we...> wrote: > > [...] It does not work reliably with pthreads though due to an > > architectural issue (libvex doesn't do atomic operations, which > > doesn't mix well with the umtx syscalls that FreeBSD's threads uses.) [ > > ...] > > Both the Helgrind and DRD tools have some support for atomic > operations. I don't know from the top of my head when exactly support > for atomic operations was added to VEX. This is a different problem, that Peter explained on this list a couple of weeks back. The problem is that VEX breaks up LOCK-prefixed read-modify-write instructions so that they are no longer atomic. This causes Valgrinded processes using such instructions to lose races against other processes (or the FreeBSD kernel) when the other processes also use LOCK-prefixed instructions on those same locations. Fixing this will require some considerable hacking around inside both VEX and various of the Valgrind tools. I plan to have a look at it over the Christmas break. J |
|
From: Mayank J. <con...@gm...> - 2008-12-19 04:56:54
|
Hi, is there anything i can help with can u plz let me know what exactly is the problem and can u plz give me access to the source code On 12/17/08, Julian Seward <js...@ac...> wrote: > On Wednesday 17 December 2008, Bart Van Assche wrote: >> On Wed, Dec 17, 2008 at 4:05 AM, Peter Wemm <pe...@we...> wrote: >> > [...] It does not work reliably with pthreads though due to an >> > architectural issue (libvex doesn't do atomic operations, which >> > doesn't mix well with the umtx syscalls that FreeBSD's threads uses.) [ >> > ...] >> >> Both the Helgrind and DRD tools have some support for atomic >> operations. I don't know from the top of my head when exactly support >> for atomic operations was added to VEX. > > This is a different problem, that Peter explained on this list a couple > of weeks back. The problem is that VEX breaks up LOCK-prefixed > read-modify-write instructions so that they are no longer atomic. This > causes Valgrinded processes using such instructions to lose races against > other processes (or the FreeBSD kernel) when the other processes also > use LOCK-prefixed instructions on those same locations. > > Fixing this will require some considerable hacking around inside both > VEX and various of the Valgrind tools. I plan to have a look at it > over the Christmas break. > > J > |