[Linux-ia64] unw_init_frame_info() and activation records

[Linux-ia64] unw_init_frame_info() and activation records

[Piet/Pete Delaney]

[-- Attachment #1: Type: text/plain, Size: 724 bytes --]

Hi Keith:

Our linux lcrash code support for stack frames needs improvement, I'm
looking around for code to pinch and was wondering about the kdb code.

The kdb code at least seems to support the NEW_UNWIND code by using 
unw_init_frame_info().  You code mentions that it should be using activation 
records and I was wondering what that's all about.

I've been reading Davie Mosberger chapter on Stack Unwinding and I'll check the
kernel code for perspectives on the code. I suspect that the gdb unwind code
and the HP-IPF-unwind library (see attached) are likely an overkill and only
needed for C++.

Perhaps you could take a few minutes and provide a few pointers on how
to save time in adding unwind support. 

-piet




[-- Attachment #2: unwind.sources.tar --]
[-- Type: application/x-tar, Size: 425512 bytes --]

Re: [Linux-ia64] unw_init_frame_info() and activation records

[n0ano]

All-

Just a friendly note from the mailing list administrator.  I let this
post go through but this is probably something that should have been
put up on an `ftp' server somewhere with a note about it's existence
sent to the list.  I don't want to put hard size limits on anyone but
we all should use common sense and use `ftp' servers where appropriate.

Tnx.

On Tue, Apr 16, 2002 at 02:55:21PM -0700, Piet/Pete Delaney wrote:
> Hi Keith:
> 
> Our linux lcrash code support for stack frames needs improvement, I'm
> looking around for code to pinch and was wondering about the kdb code.
> 
> The kdb code at least seems to support the NEW_UNWIND code by using 
> unw_init_frame_info().  You code mentions that it should be using activation 
> records and I was wondering what that's all about.
> 
> I've been reading Davie Mosberger chapter on Stack Unwinding and I'll check the
> kernel code for perspectives on the code. I suspect that the gdb unwind code
> and the HP-IPF-unwind library (see attached) are likely an overkill and only
> needed for C++.
> 
> Perhaps you could take a few minutes and provide a few pointers on how
> to save time in adding unwind support. 
> 
> -piet
> 
> 
> 



-- 
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
n0ano@indstorage.com
Ph: 303/652-0870x117