This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: RFC: one approach to fixing PR 14100
On Thu, 09 Aug 2012 11:42:28 +0200, Mark Kettenis wrote:
> Sorry, but I really think you're working around a problem in the
> tailcall sniffer here. The tailcall sniffer seems to violate several
> of the design principles of the frame unwinder framework. It should
> be fixed instead.
Could you be more specific? The tailcall sniffer is unusual that it needs to
unwind frame (PC) X+1 from frame X and then insert arbitrary number of frames
in between. So it needs to do "preliminary unwind" first.
I have tried various ways while coding it and I found the current
implementation to fit most the GDB codebase.
Thanks,
Jan