Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Nix
On 14 May 2008, Jeff Dike verbalised: > On Wed, May 14, 2008 at 09:12:12PM +0100, Nix wrote: >> Oh, and, it's pedantic of me, I know, but what does this do if time goes >> backwards in the NO_HZ case? (Or is handling that a 2.6.26 thing?) > > In all cases, it holds time steady until the host catch

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Jeff Dike
On Wed, May 14, 2008 at 09:12:12PM +0100, Nix wrote: > Oh, and, it's pedantic of me, I know, but what does this do if time goes > backwards in the NO_HZ case? (Or is handling that a 2.6.26 thing?) In all cases, it holds time steady until the host catches up with what the guest thinks the time shou

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Nix
[vincent-perrier <[EMAIL PROTECTED]> removed from Cc;, his MTA says `Client host rejected: AP0002 Please use your ISP mailserver' only I don't *have* an ISP mailserver.] On 14 May 2008, [EMAIL PROTECTED] verbalised: > Annoyingly, now I've upgraded the host to 2.6.25 (hence sans skas3), > timings

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Nix
On 14 May 2008, Jeff Dike verbalised: > I finally reproduced this using 2.6.25.1 (2.6.25-mm1 was no good) with > your config. YAY! (I wonder why this was so .config-dependent? You'd think it would trigger on anything, but I couldn't even make it happen on all my hosts...) > The patch below fixes

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Jeff Dike
I finally reproduced this using 2.6.25.1 (2.6.25-mm1 was no good) with your config. The patch below fixes it for me. You'll notice a certain similarity between this and a previous patch that you posted. It's not clear to me why yours didn't work. Jeff -- Work email - jdike

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread vincent-perrier
Sorry if you receive this message twice, but I sent 2 messages following each other by a few minutes, and received only the second one, this was the first one: Hello, I know that I should not work as root, but I am too used to it, so my example is done with the root user. I am not sure that the

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread vincent-perrier
I forgot to tell you, if you launch the daemon as a normal user, the RSA files of this normal user will be used, and not the root ones. And also, the daemon has to mount a file when it creates the ubdb config file, one more reason to have the root sticky bit. This tool is made for network study a

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread vincent-perrier
Hello, I know that I should not work as root, but I am too used to it, so my example is done with the root user. I am not sure that the network simulator works with another user for the version you have, but I have checked that for next one, a normal user will do as long as the daemon is owned by

Re: [uml-devel] actually useful backtrace from a CPU-chewing hang

2008-05-14 Thread Jeff Dike
On Fri, May 09, 2008 at 07:13:22PM +0200, vincent-perrier wrote: > The download is long, the untar too, but the freeze is garanteed! > > to start after download: > > "./start_clownix_net virtual_platform_configs/single_machine" > > Hope you try it! Why is it mucking with root's ssh keys: RS

Re: [uml-devel] umls unresponsive & consuming 100% cpu time

2008-05-14 Thread Sakari Ailus
Bram Matthys (Syzop) wrote: > Thanks for your reply. > > $ grep HZ .config > CONFIG_HZ=100 > # CONFIG_NO_HZ is not set > > I've applied your patch against my 2.6.25 (vanilla)... > patching file arch/um/os-Linux/time.c > patching file arch/um/kernel/time.c > Hunk #1 succeeded at 74 (offset -1 line