Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* [RFA] fix inferior call for avr
@ 2009-10-23 15:16 Tristan Gingold
  2009-10-26 10:44 ` Joel Brobecker
  0 siblings, 1 reply; 3+ messages in thread
From: Tristan Gingold @ 2009-10-23 15:16 UTC (permalink / raw)
  To: gdb-patches ml

Hi,

there are two issues in avr_push_dummy_call:
* avr6 architecture (were PC size is 3 bytes instead of 2) was not  
correctly handled
* the returned value was not correct, which made gdb assertions failed.

Ok for trunk ?

Tristan.

2009-10-23  Tristan Gingold  <gingold@adacore.com>

	* avr-tdep.c (avr_push_dummy_call): Handle avr6 architecture.
	Fix the returned value.  Fix style.



--- avr-tdep.c	2 Jul 2009 17:25:52 -0000	1.115
+++ avr-tdep.c	23 Oct 2009 15:13:35 -0000
@@ -1181,7 +1181,8 @@
  {
    enum bfd_endian byte_order = gdbarch_byte_order (gdbarch);
    int i;
-  unsigned char buf[2];
+  unsigned char buf[3];
+  int call_length = gdbarch_tdep (gdbarch)->call_length;
    CORE_ADDR return_pc = avr_convert_iaddr_to_raw (bp_addr);
    int regnum = AVR_ARGN_REGNUM;
    struct stack_item *si = NULL;
@@ -1219,11 +1220,9 @@
              regnum--;

            val = extract_unsigned_integer (contents, len, byte_order);
-          for (j=0; j<len; j++)
-            {
-              regcache_cooked_write_unsigned (regcache, regnum--,
-                                              val >> (8*(len-j-1)));
-            }
+          for (j = 0; j < len; j++)
+            regcache_cooked_write_unsigned
+              (regcache, regnum--, val >> (8 * (len - j - 1)));
          }
        /* No registers available, push the args onto the stack. */
        else
@@ -1245,17 +1244,23 @@
    /* Set the return address.  For the avr, the return address is the  
BP_ADDR.
       Need to push the return address onto the stack noting that it  
needs to be
       in big-endian order on the stack.  */
-  buf[0] = (return_pc >> 8) & 0xff;
-  buf[1] = return_pc & 0xff;
+  for (i = 1; i <= call_length; i++)
+    {
+      buf[call_length - i] = return_pc & 0xff;
+      return_pc >>= 8;
+    }

-  sp -= 2;
-  write_memory (sp + 1, buf, 2);  /* Add one since pushes are post  
decr ops. */
+  sp -= call_length;
+  /* Use 'sp + 1' since pushes are post decr ops. */
+  write_memory (sp + 1, buf, call_length);

    /* Finally, update the SP register. */
    regcache_cooked_write_unsigned (regcache, AVR_SP_REGNUM,
  				  avr_convert_saddr_to_raw (sp));

-  return sp;
+  /* Return SP value for the dummy frame, where the return address  
hasn't been
+     pushed.  */
+  return sp + call_length;
  }

  /* Initialize the gdbarch structure for the AVR's. */


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFA] fix inferior call for avr
  2009-10-23 15:16 [RFA] fix inferior call for avr Tristan Gingold
@ 2009-10-26 10:44 ` Joel Brobecker
  2009-10-26 16:56   ` Tristan Gingold
  0 siblings, 1 reply; 3+ messages in thread
From: Joel Brobecker @ 2009-10-26 10:44 UTC (permalink / raw)
  To: Tristan Gingold; +Cc: gdb-patches ml

> 2009-10-23  Tristan Gingold  <gingold@adacore.com>
>
> 	* avr-tdep.c (avr_push_dummy_call): Handle avr6 architecture.
> 	Fix the returned value.  Fix style.

OK.

-- 
Joel


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFA] fix inferior call for avr
  2009-10-26 10:44 ` Joel Brobecker
@ 2009-10-26 16:56   ` Tristan Gingold
  0 siblings, 0 replies; 3+ messages in thread
From: Tristan Gingold @ 2009-10-26 16:56 UTC (permalink / raw)
  To: Joel Brobecker; +Cc: gdb-patches ml


On Oct 26, 2009, at 11:44 AM, Joel Brobecker wrote:

>> 2009-10-23  Tristan Gingold  <gingold@adacore.com>
>>
>> 	* avr-tdep.c (avr_push_dummy_call): Handle avr6 architecture.
>> 	Fix the returned value.  Fix style.
>
> OK.

Thanks, committed.

Tristan.



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2009-10-26 16:56 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-10-23 15:16 [RFA] fix inferior call for avr Tristan Gingold
2009-10-26 10:44 ` Joel Brobecker
2009-10-26 16:56   ` Tristan Gingold

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