Re: Changing syscall table

On Thu, 03 Sep 2009, Zhenwen Xu wrote:

> On Tue, Sep 01, 2009 at 08:57:58PM +0300, Jonathan Nell wrote:
> > I'm trying to wrap the
> > SG_IO ioctl call (i.e. trap it in the kernel) and have that dump the
> > data from (struct sg_io_hdr).dxferp.
> > Having issues with doing the kernel trap in the newer kernel versions
> > though (trying on 2.6.30). The syscall table is now read-only but for
> > some reason my set_memory_rw() call is failing... Any ideas how to do
> > this properly?
> > 
> > Here are the relevant bits of code:
> 
> try read this:
> http://zhwen.org/xlog/2009/03/%e6%88%aa%e8%8e%b7linux%e7%b3%bb%e7%bb%9f%e8%b0%83%e7%94%a8.htm
> 
> here is the demo.
> http://zhwen.org/coding/cat_syscall.c

while this is quit a nice hack I think that meanwhile standard methods like
kprobes or utrace would do the trick without such intrusive code and allow
very flexible analysis of what is going on.

hofrat