[Fwd: multiple location in C++ constructors]

[Fwd: multiple location in C++ constructors]

[Denis PILAT]

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



[-- Attachment #2: multiple location in C++ constructors.eml --]
[-- Type: message/rfc822, Size: 4160 bytes --]

From: Denis PILAT <denis.pilat@st.com>
To: gdb@sourceware.org
Subject: multiple location in C++ constructors
Date: Tue, 14 Oct 2008 17:12:55 +0200
Message-ID: <48F4B6F7.2010109@st.com>

We (at ST) have a compiler that generates DW_AT_MIPS_linkage_name 
attributes for constructors.
Unfortunately, that prevents breakpoint in constructor to have multiple 
locations since the gdb heuristics eliminates one location in the 
expand_line_sal_maybe() function of breakpoint.c: see bellow, we go thru 
the remove_sal () line.

(from breakpoint.c)
struct symtabs_and_lines
expand_line_sal_maybe (struct symtab_and_line sal)
...
	  if (find_pc_partial_function (pc, &this_function, 
					&func_addr, &func_end))
	    {
	      if (this_function && 
		  strcmp (this_function, original_function) != 0)
		{
		  remove_sal (&expanded, i);
		  --i;
		}
	      else if (func_addr == pc)	    
...



If present into dwarf2 debug information, the find_pc_partial_function() 
returns the DW_AT_MIPS_linkage_name in this_function, which is alway 
different from the original_function. Therefore there is always only one 
location for constructor breakpoints since the DW_AT_MIPS_linkage_name 
is filled with the mangled function name.


My question is:
Is the GDB heuristic to find multiple location for breakpoint wrong ?
Or is the DW_AT_MIPS_linkage_name attribute for constructors useless in 
debug informations ?

Thanks for your feedback
Denis

Re: [Fwd: multiple location in C++ constructors]

[Daniel Jacobowitz]

On Thu, Oct 16, 2008 at 02:45:47PM +0200, Denis PILAT wrote:
> Date: Tue, 14 Oct 2008 17:12:55 +0200
> From: Denis PILAT <denis.pilat@st.com>
> Subject: multiple location in C++ constructors
> To: gdb@sourceware.org

This isn't a patch, so gdb@ was the right list.  I just haven't had
time to think through a response yet; I am hoping Vladimir will have a
chance to look at this, since IIRC he knew more about it.

-- 
Daniel Jacobowitz
CodeSourcery