[Crash-utility] display function parameters for call stack

[Lei Wen]

Dave,

On Fri, Sep 21, 2012 at 9:44 PM, Dave Anderson <anderson at redhat.com> wrote:

>
>
> ----- Original Message -----
> >
> > Hi Dave,
> >
> >
> > I notice there was some discussion around on enabling
> > the display function parameters feature for crash tool.
> > Like this one:
> > http://www.redhat.com/archives/crash-utility/2009-May/msg00016.html
>
> Yes, it seems to come up every few years.  People have tried, but it
> has never come to fruition.
>
> >
> > Although the answer seems to be quite clear, it should be hard to
> > support such feature. But does this feature cannot be done anyway?
> > I think this feature would be great in assisting kernel debug.
> >
> > The main feature I require, first should be know each param's value
> > in the call stack.
> > Then it is better to know each call stack's param's name.
> >
> > Like for do_vfs_ioctl function, it is better to display as:
> > do_vfs_ioctl(filp=xxx, fd=xxx, cmd=xxx, arg=xxx)
> >
> > Do you have any idea on how this could be implemented?
>
> No I don't.
>

Sorry for asking the stupid question, but I notice that we could
get the structure member in crash by "struct" command:
crash> struct stackframe
struct stackframe {
    long unsigned int fp;
    long unsigned int sp;
    long unsigned int lr;
    long unsigned int pc;
}
SIZE: 16

I check the crash code, and find cmd_struct use gdb interface to parse it
out.

And the gdb itself seems already could parse out function argument[1], so
could
we still use similar mechanism as the "struct" command to implement a "func"
command, which could shows the function declaration?


Reference:
[1] http://www.chemie.fu-berlin.de/chemnet/use/info/gdb/gdb_7.html

GDB could shows the function param string and corresponding value

#0  m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)

    at builtin.c:993
#1  0x6e38 in expand_macro (sym=0x2b600) at macro.c:242
#2  0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
    at macro.c:71
(More stack frames follow...)

Thanks,
Lei
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/crash-utility/attachments/20120924/25e500d3/attachment.htm>