State of kernel debuggers

[Mathieu Tarral]

Hi Juerg,

thank you for your answer.


On Monday 10 December 2018 09:03, Juerg Haefliger <juerg.haefliger at canonical.com> wrote:

> On Wed, 28 Nov 2018 16:17:53 +0000
> Mathieu Tarral mathieu.tarral at protonmail.com wrote:
>
> > Hi Ubuntu kernel team !
> > I would like to start a small discussion with you about the state
> > of our kernel debuggers in the Linux community.
> > I would like to know what are the tools that you are using to debug the linux kernel
> > in your everyday tasks as developers, and if they could be improved.
>
> printk, ftace, crash and when I'm really desperate single-stepping with QEMU.
>
> > The reason I'm asking is because I'm actually developing a flexible hypervisor-level debugger:
> > https://github.com/Wenzel/pyvmidbg
>
> Hmm... how is that different/better from/than using QEMU's gdb server?

The main goal is to have a GDB stub aware of the operating system internals, via virtual machine introspection.

Therefore you are not limited to the kernel anymore, but you can debug usermode processes too, and reconstruct their execution context, thread list, etc, from the stub.

Thank you for your time, and merry christmas ! 🎄