[RFA] mingw64: Skip over __main in main

[RFA] mingw64: Skip over __main in main

[Pierre Muller]

  This patch enables recognition of call to __main
as part of the prologue of main for mingw64 binaries.

  It's almost a copy of the i386-tdep.c counterpart
(expect that I removed the 32-bit specific overflow code).

  It makes things more consistent with the way 32-bit binaries
are treated and allows to run testsuite on Cygwin
(after a few tricks ...).

  Is this patch OK?




2010-09-27  Pierre Muller  <muller@ics.u-strasbg.fr>

	* src/gdb/amd64-windows-tdep.c (amd64_skip_main_prologue): New function.
	(amd64_windows_init_abi): Register amd64_skip_main_prologue
	as gdbarch skip_main_prologue method.

Index: src/gdb/amd64-windows-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/amd64-windows-tdep.c,v
retrieving revision 1.6
diff -u -p -r1.6 amd64-windows-tdep.c
--- src/gdb/amd64-windows-tdep.c	29 Jan 2010 05:29:21 -0000	1.6
+++ src/gdb/amd64-windows-tdep.c	27 Sep 2010 16:15:28 -0000
@@ -122,6 +122,38 @@ amd64_windows_return_value (struct gdbar
     }
 }
 
+/* Check that the code pointed to by PC corresponds to a call to
+   __main, skip it if so.  Return PC otherwise.  */
+
+static CORE_ADDR
+amd64_skip_main_prologue (struct gdbarch *gdbarch, CORE_ADDR pc)
+{
+  enum bfd_endian byte_order = gdbarch_byte_order (gdbarch);
+  gdb_byte op;
+
+  target_read_memory (pc, &op, 1);
+  if (op == 0xe8)
+    {
+      gdb_byte buf[4];
+
+      if (target_read_memory (pc + 1, buf, sizeof buf) == 0)
+ 	{
+ 	  struct minimal_symbol *s;
+ 	  CORE_ADDR call_dest;
+
+	  call_dest = pc + 5 + extract_signed_integer (buf, 4, byte_order);
+ 	  s = lookup_minimal_symbol_by_pc (call_dest);
+ 	  if (s != NULL
+ 	      && SYMBOL_LINKAGE_NAME (s) != NULL
+ 	      && strcmp (SYMBOL_LINKAGE_NAME (s), "__main") == 0)
+ 	    pc += 5;
+ 	}
+    }
+
+  return pc;
+}
+
+
 static void
 amd64_windows_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
 {
@@ -140,6 +172,7 @@ amd64_windows_init_abi (struct gdbarch_i
   tdep->memory_args_by_pointer = 1;
   tdep->integer_param_regs_saved_in_caller_frame = 1;
   set_gdbarch_return_value (gdbarch, amd64_windows_return_value);
+  set_gdbarch_skip_main_prologue (gdbarch, amd64_skip_main_prologue);
 
   set_solib_ops (gdbarch, &solib_target_so_ops);
 }

Re: [RFA] mingw64: Skip over __main in main

[Mark Kettenis]

> From: "Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>
> Date: Mon, 27 Sep 2010 18:25:25 +0200
> 
>   This patch enables recognition of call to __main
> as part of the prologue of main for mingw64 binaries.
> 
>   It's almost a copy of the i386-tdep.c counterpart
> (expect that I removed the 32-bit specific overflow code).
> 
>   It makes things more consistent with the way 32-bit binaries
> are treated and allows to run testsuite on Cygwin
> (after a few tricks ...).
> 
>   Is this patch OK?

Fine with me.

> 2010-09-27  Pierre Muller  <muller@ics.u-strasbg.fr>
> 
> 	* src/gdb/amd64-windows-tdep.c (amd64_skip_main_prologue): New function.
> 	(amd64_windows_init_abi): Register amd64_skip_main_prologue
> 	as gdbarch skip_main_prologue method.

RE: [RFA] mingw64: Skip over __main in main

[Pierre Muller]

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Mark Kettenis
> Envoyé : Monday, September 27, 2010 7:43 PM
> À : pierre.muller@ics-cnrs.unistra.fr
> Cc : gdb-patches@sourceware.org
> Objet : Re: [RFA] mingw64: Skip over __main in main
> 
> > From: "Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>
> > Date: Mon, 27 Sep 2010 18:25:25 +0200
> >
> >   This patch enables recognition of call to __main
> > as part of the prologue of main for mingw64 binaries.
> >
> >   It's almost a copy of the i386-tdep.c counterpart
> > (expect that I removed the 32-bit specific overflow code).
> >
> >   It makes things more consistent with the way 32-bit binaries
> > are treated and allows to run testsuite on Cygwin
> > (after a few tricks ...).
> >
> >   Is this patch OK?
> 
> Fine with me.

  Thanks for the approval,
patch committed.
http://sourceware.org/ml/gdb-cvs/2010-09/msg00166.html

Pierre