|
From: Nicholas N. <nj...@ca...> - 2004-02-22 17:27:32
|
Hi, Valgrind 2.1.0 was released on December 14. Since then, FV has entered the head, plus a ton of other bug fixes and improvements. I think it's time for a new development release. However, I'm not in a position to do it because I don't have access to a range of machines and distros. Normally Julian does it (he at least is the only one who can put the tarballs onto the site). To de-burden Julian, and get more thorough testing, Perhaps we could try to do some more distributed pre-release testing now -- ie. get everyone to run the regtests on all their machines, plus generally play around with a few big apps like Mozilla, OpenOffice, or Unreal Tournament. Thoughts? N |
|
From: Dirk M. <dm...@gm...> - 2004-02-22 19:27:08
|
On Sunday 22 February 2004 18:19, Nicholas Nethercote wrote: > plus generally play around with a few big apps like Mozilla, OpenOffice, > or Unreal Tournament. HEAD does not work at all for me since I have ulimits on address space. imho as long as this is unsolved its not a good idea to do a 2.1.1. Also, as you correctly mentioned, valgrind does not compile with newer glibc releases, so it would be a good idea to solve #69616 first. Dirk |
|
From: Dirk M. <dm...@gm...> - 2004-02-23 02:33:20
|
On Sunday 22 February 2004 20:19, Dirk Mueller wrote: > Also, as you correctly mentioned, valgrind does not compile with newer > glibc releases, so it would be a good idea to solve #69616 first. BTW, do we plan to try to include the freebsd port into 2.1.1 or 2.1.2 ? I think we can go with the additional x86-common subdirectory, and I'd like to hide it in an arch subdir to avoid the cluttering of the coregrind directory. and regarding i386 vs x86 vs ia32: http://www.ussg.iu.edu/hypermail/linux/kernel/0402.2/0795.html :) |
|
From: Nicholas N. <nj...@ca...> - 2004-02-23 14:19:57
|
On Mon, 23 Feb 2004, Dirk Mueller wrote: > BTW, do we plan to try to include the freebsd port into 2.1.1 or 2.1.2 ? I would suggest 2.1.2 -- 2.1.1 already has plenty of changes, esp. FV. > and regarding i386 vs x86 vs ia32: > > http://www.ussg.iu.edu/hypermail/linux/kernel/0402.2/0795.html Woo, I like 'x86' the best too. N |
|
From: Jeremy F. <je...@go...> - 2004-02-23 20:02:08
|
On Mon, 2004-02-23 at 06:11, Nicholas Nethercote wrote: > On Mon, 23 Feb 2004, Dirk Mueller wrote: > > > BTW, do we plan to try to include the freebsd port into 2.1.1 or 2.1.2 ? > > I would suggest 2.1.2 -- 2.1.1 already has plenty of changes, esp. FV. I agree. I think we should fix any little bugs which are show-stoppers, and then release it. > > and regarding i386 vs x86 vs ia32: > > > > http://www.ussg.iu.edu/hypermail/linux/kernel/0402.2/0795.html > > Woo, I like 'x86' the best too. Except that Linux uses i386 everywhere for ia32... J |
|
From: Nicholas N. <nj...@ca...> - 2004-02-24 23:35:51
|
On Mon, 23 Feb 2004, Jeremy Fitzhardinge wrote: > > > BTW, do we plan to try to include the freebsd port into 2.1.1 or 2.1.2 ? > > > > I would suggest 2.1.2 -- 2.1.1 already has plenty of changes, esp. FV. > > I agree. I think we should fix any little bugs which are > show-stoppers, and then release it. Ok, so what's the definition of a show-stopping bug? Does Dirk's ulimit problem count? (What is the problem, Dirk -- have you filed a bug report?) Does anything else? Can the release proceed? N |
|
From: Tom H. <th...@cy...> - 2004-02-24 23:37:54
|
In message <Pin...@ye...>
Nicholas Nethercote <nj...@ca...> wrote:
> Ok, so what's the definition of a show-stopping bug? Does Dirk's ulimit
> problem count? (What is the problem, Dirk -- have you filed a bug
> report?) Does anything else? Can the release proceed?
If you have a data limit set then nothing will start under valgrind
although it does give an explanatory error message. I thought Jeremy
said he knew how to rework it to avoid the problem but that was a
while ago.
Tom
--
Tom Hughes (th...@cy...)
Software Engineer, Cyberscience Corporation
http://www.cyberscience.com/
|
|
From: Jeremy F. <je...@go...> - 2004-02-25 00:20:38
|
On Tue, 2004-02-24 at 15:34, Nicholas Nethercote wrote: > Ok, so what's the definition of a show-stopping bug? Does Dirk's ulimit > problem count? Not really, since it only applies to setting a datasize ulimit, which is a pretty useless thing to set anyway (since glibc malloc will always get around it by using mmap). More seriously, the brk extension in stage1 was pretty sensitive to the kernel's limits on how rapidly you can grow the brk, which depends on things like how much physical memory there is, how busy the machine is, and so on. Anyway, I just checked in a fix for it. J |
|
From: Dirk M. <dm...@gm...> - 2004-02-26 18:50:19
|
On Wednesday 25 February 2004 01:17, Jeremy Fitzhardinge wrote: > Anyway, I just checked in a fix for it. Another (imho) showstopper is that there are stale threads staying around if the valgrind process dies an unexpected death (e.g. by hitting an assertion failure). |
|
From: Tom H. <th...@cy...> - 2004-02-26 18:58:51
|
In message <200...@gm...>
Dirk Mueller <dm...@gm...> wrote:
> On Wednesday 25 February 2004 01:17, Jeremy Fitzhardinge wrote:
>
> > Anyway, I just checked in a fix for it.
>
> Another (imho) showstopper is that there are stale threads staying around
> if the valgrind process dies an unexpected death (e.g. by hitting an
> assertion failure).
Have you seen that as well then, or are you just going on the report
on valgrind-users this afternoon? Only that is the first report I remember
seeing of it and I certainly haven't noticed it.
Tom
--
Tom Hughes (th...@cy...)
Software Engineer, Cyberscience Corporation
http://www.cyberscience.com/
|
|
From: Nicholas N. <nj...@ca...> - 2004-02-26 19:06:31
|
On Thu, 26 Feb 2004, Tom Hughes wrote: > > > Anyway, I just checked in a fix for it. > > > > Another (imho) showstopper is that there are stale threads staying around > > if the valgrind process dies an unexpected death (e.g. by hitting an > > assertion failure). > > Have you seen that as well then, or are you just going on the report > on valgrind-users this afternoon? Only that is the first report I remember > seeing of it and I certainly haven't noticed it. I used to have this problem (or something that looked a lot like it), but don't seem to anymore. N |
|
From: Dirk M. <dm...@gm...> - 2004-02-26 19:11:36
|
On Thursday 26 February 2004 19:57, Tom Hughes wrote: > Have you seen that as well then, or are you just going on the report > on valgrind-users this afternoon? Both, I've seen this as well. Its kind of annoying as you don't get rid of those processes with kill (or kill -9). you can only reboot. it happened when I was trying to debug an assertion failure I'm hitting sometimes. each time one stale process stayed around. Dirk |
|
From: Nicholas N. <nj...@ca...> - 2004-02-26 19:17:13
|
On Thu, 26 Feb 2004, Dirk Mueller wrote: > Both, I've seen this as well. Its kind of annoying as you don't get rid of > those processes with kill (or kill -9). you can only reboot. Hmm, when I had my problem I could kill the zombies. So it must have been different. |