[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
Hi Everyone, Some advise please, I do a fair bit of kernel driver development on the rt2x00 project (wireless lan) a side effect of kernel devleopment is that bugs in your code can oops the kernel with results varying from a simple bug() call to a random walk through kernel memory and a distroyed file system. At the minimum its a reboot of the machine cause even a minor oops locks that driver up and you cannot rmmod and re insmod it. Virtulisation seems the way to go but i know *nothing* about it. The device I am playing with is USB based so can install some virtulisation software and have a VM running that can access the physical USB device and if it goes bang only upset the VM so i can quickly restart the VM from an image? Secondly in the past i have used two system to catch really nasty bugs by setting one system to have its kernel dmesg output sent up /dev/ttyS0 and the 2nd system runs a terminal to capture the output for post crash analysis. Is this possible with a VM and a host system. So any advice on software to run and what to do and not to do would be greatly thanked. At least a pointer in the correct direction to start. Also anything that i can apt-get (from debian unstable/testing or stable i run unstable) is prefered over compiling from source, not that i mined compiling but i prefer packages where possible. Thanks! Robin (avoiding work again) -- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/linux_adm/list-faq.html