Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* [RFA] Prologue detection on x86-64
@ 2003-03-06 22:56 Michal Ludvig
  2003-03-06 23:10 ` Daniel Jacobowitz
  0 siblings, 1 reply; 4+ messages in thread
From: Michal Ludvig @ 2003-03-06 22:56 UTC (permalink / raw)
  To: GDB Patches

[-- Attachment #1: Type: text/plain, Size: 375 bytes --]

Hi all,
the attached patch moves prologue detection from x86_64_skip_prologue() 
to separate function, that could be reused later. I need this separate 
function to implement unwinding of functions without debug info.

It's an almost obvious patch, I believe.

OK to commit?

Michal Ludvig
-- 
* SuSE CR, s.r.o     * mludvig@suse.cz
* (+420) 296.545.373 * http://www.suse.cz

[-- Attachment #2: skip-prologue-1.diff --]
[-- Type: text/plain, Size: 2992 bytes --]

2003-03-06  Michal Ludvig  <mludvig@suse.cz>

	* x86-64-tdep.c (x86_64_function_has_prologue): New function.
	(x86_64_skip_prologue): Move prologue detection to 
	separate function.
	* x86-64-tdep.h (x86_64_function_has_prologue): New prototype.

Index: x86-64-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/x86-64-tdep.c,v
retrieving revision 1.51
diff -u -p -r1.51 x86-64-tdep.c
--- x86-64-tdep.c	2 Mar 2003 04:02:25 -0000	1.51
+++ x86-64-tdep.c	6 Mar 2003 22:50:09 -0000
@@ -851,11 +851,31 @@ x86_64_frameless_function_invocation (st
   return 0;
 }
 
+/* We will handle only functions beginning with:
+   55          pushq %rbp
+   48 89 e5    movq %rsp,%rbp */
+#define PROLOG_BUFSIZE 4
+static int
+x86_64_function_has_prologue (CORE_ADDR pc)
+{
+  int i;
+  unsigned char prolog_expect[PROLOG_BUFSIZE] = { 0x55, 0x48, 0x89, 0xe5 },
+    prolog_buf[PROLOG_BUFSIZE];
+
+  read_memory (pc, (char *) prolog_buf, PROLOG_BUFSIZE);
+
+  /* First check, whether pc points to pushq %rbp, movq %rsp,%rbp.  */
+  for (i = 0; i < PROLOG_BUFSIZE; i++)
+    if (prolog_expect[i] != prolog_buf[i])
+      return 0;		/* ... no, it doesn't. Nothing to skip.  */
+  
+  return 1;
+}
+
 /* If a function with debugging information and known beginning
    is detected, we will return pc of the next line in the source 
    code. With this approach we effectively skip the prolog.  */
 
-#define PROLOG_BUFSIZE 4
 CORE_ADDR
 x86_64_skip_prologue (CORE_ADDR pc)
 {
@@ -863,21 +883,9 @@ x86_64_skip_prologue (CORE_ADDR pc)
   struct symtab_and_line v_sal;
   struct symbol *v_function;
   CORE_ADDR endaddr;
-  unsigned char prolog_buf[PROLOG_BUFSIZE];
-
-  /* We will handle only functions starting with: */
-  static unsigned char prolog_expect[PROLOG_BUFSIZE] =
-  {
-    0x55,			/* pushq %rbp */
-    0x48, 0x89, 0xe5		/* movq %rsp, %rbp */
-  };
 
-  read_memory (pc, (char *) prolog_buf, PROLOG_BUFSIZE);
-
-  /* First check, whether pc points to pushq %rbp, movq %rsp, %rbp.  */
-  for (i = 0; i < PROLOG_BUFSIZE; i++)
-    if (prolog_expect[i] != prolog_buf[i])
-      return pc;		/* ... no, it doesn't.  Nothing to skip.  */
+  if (! x86_64_function_has_prologue (pc))
+    return pc;
 
   /* OK, we have found the prologue and want PC of the first
      non-prologue instruction.  */
Index: x86-64-tdep.h
===================================================================
RCS file: /cvs/src/src/gdb/x86-64-tdep.h,v
retrieving revision 1.10
diff -u -p -r1.10 x86-64-tdep.h
--- x86-64-tdep.h	21 Dec 2002 21:09:58 -0000	1.10
+++ x86-64-tdep.h	6 Mar 2003 22:50:09 -0000
@@ -36,6 +36,7 @@ gdbarch_saved_pc_after_call_ftype x86_64
 gdbarch_pc_in_sigtramp_ftype x86_64_linux_in_sigtramp;
 CORE_ADDR x86_64_linux_frame_chain (struct frame_info *fi);
 CORE_ADDR x86_64_init_frame_pc (int fromleaf, struct frame_info *fi);
+int x86_64_function_has_prologue (CORE_ADDR pc);
 
 void x86_64_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch);
 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [RFA] Prologue detection on x86-64
  2003-03-06 22:56 [RFA] Prologue detection on x86-64 Michal Ludvig
@ 2003-03-06 23:10 ` Daniel Jacobowitz
  2003-03-06 23:14   ` Michal Ludvig
  0 siblings, 1 reply; 4+ messages in thread
From: Daniel Jacobowitz @ 2003-03-06 23:10 UTC (permalink / raw)
  To: GDB Patches

On Thu, Mar 06, 2003 at 11:56:44PM +0100, Michal Ludvig wrote:
> Hi all,
> the attached patch moves prologue detection from x86_64_skip_prologue() 
> to separate function, that could be reused later. I need this separate 
> function to implement unwinding of functions without debug info.
> 
> It's an almost obvious patch, I believe.
> 
> OK to commit?

Just two things:

> +/* We will handle only functions beginning with:
> +   55          pushq %rbp
> +   48 89 e5    movq %rsp,%rbp */

Period and two spaces, please.

> +#define PROLOG_BUFSIZE 4
> +static int
> +x86_64_function_has_prologue (CORE_ADDR pc)


> +int x86_64_function_has_prologue (CORE_ADDR pc);

That's probably a warning.  Static functions don't need to be
prototyped in the tdep.h file; this bit can just be omitted.

Other than that OK.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [RFA] Prologue detection on x86-64
  2003-03-06 23:10 ` Daniel Jacobowitz
@ 2003-03-06 23:14   ` Michal Ludvig
       [not found]     ` <20030306232209.GA10007@nevyn.them.org>
  0 siblings, 1 reply; 4+ messages in thread
From: Michal Ludvig @ 2003-03-06 23:14 UTC (permalink / raw)
  To: Daniel Jacobowitz; +Cc: GDB Patches

Daniel Jacobowitz wrote:
> On Thu, Mar 06, 2003 at 11:56:44PM +0100, Michal Ludvig wrote:
>>+/* We will handle only functions beginning with:
>>+   55          pushq %rbp
>>+   48 89 e5    movq %rsp,%rbp */
> 
> Period and two spaces, please.

Period after %rbp? It wouldn't be an assembler anymore :-(
How about two spaces and no period instead? (compromise :-)

>>+#define PROLOG_BUFSIZE 4
>>+static int
>>+x86_64_function_has_prologue (CORE_ADDR pc)
> 
>>+int x86_64_function_has_prologue (CORE_ADDR pc);
> 
> 
> That's probably a warning.  Static functions don't need to be
> prototyped in the tdep.h file; this bit can just be omitted.

Oops, sorry. Shouldn't be static, because it will be used in 
x86-64-linux-tdep.c, so I need a prototype in x86-64-tdep.h.

Michal Ludvig
-- 
* SuSE CR, s.r.o     * mludvig@suse.cz
* (+420) 296.545.373 * http://www.suse.cz


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [RFA] Prologue detection on x86-64
       [not found]     ` <20030306232209.GA10007@nevyn.them.org>
@ 2003-03-07 10:44       ` Michal Ludvig
  0 siblings, 0 replies; 4+ messages in thread
From: Michal Ludvig @ 2003-03-07 10:44 UTC (permalink / raw)
  To: Daniel Jacobowitz; +Cc: GDB Patches

Daniel Jacobowitz wrote:
> On Fri, Mar 07, 2003 at 12:14:55AM +0100, Michal Ludvig wrote:
> 
>>Daniel Jacobowitz wrote:
>>
>>>On Thu, Mar 06, 2003 at 11:56:44PM +0100, Michal Ludvig wrote:
>>>
>>>>+/* We will handle only functions beginning with:
>>>>+   55          pushq %rbp
>>>>+   48 89 e5    movq %rsp,%rbp */
>>>
>>>Period and two spaces, please.
>>
>>Period after %rbp? It wouldn't be an assembler anymore :-(
>>How about two spaces and no period instead? (compromise :-)
> 
> What I usually do in this case is make up a sentence to go before the
> period :P  No joking.  Something like:
> /* We will handle only functions beginning with
>    55         pushq %rbp
>    48 89 e5   movq %rsp, %rbp
>    Any function that doesn't start with this sequence
>    will be assumed to have no prologue.  */

OK, committed with this change and removed 'static' keyword.

Michal Ludvig
-- 
* SuSE CR, s.r.o     * mludvig@suse.cz
* (+420) 296.545.373 * http://www.suse.cz


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2003-03-07 10:44 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-03-06 22:56 [RFA] Prologue detection on x86-64 Michal Ludvig
2003-03-06 23:10 ` Daniel Jacobowitz
2003-03-06 23:14   ` Michal Ludvig
     [not found]     ` <20030306232209.GA10007@nevyn.them.org>
2003-03-07 10:44       ` Michal Ludvig

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox