This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
prob with "next"
- From: "Vineet Sharma, Noida" <vineets at noida dot hcltech dot com>
- To: gdb at sources dot redhat dot com
- Date: Wed, 24 Mar 2004 20:28:43 +0530
- Subject: prob with "next"
Hi All,
I have some problem with "next" on function calls. My doubts are
Q1 If my target PC register points to the next instruction(to be executed)
not the current one, then does the setting
gdbarch_decr_pc_after_break (gdbarch, 2); (My instruction
size is 2 bytes)
is enough to compenstate the "current pc" requirement of gdb ?
--------------------------------
Q2 How does the "next" work on a function call(on a target which does
not have any hardware support for breakpoint, but single stepping is
supported)? I thought it just sets the breakpoint(at the funct end) and
continues. But that s not happening?
-----------------------------------
Q3I am facing problem with "next" in the following case :
when i execute the "next" command, at x=add(a,b);
at that point my PC is pointing to next instr (i. e address 0x26) and the
current instruction is (0x24)
b=222; //Pseudo
22: 3e 06 MOVE R1,
XXX
24: 0e de MOVE R2,
#deh
x=add(a,b);
PC-> 26: 3e 04 MOVE R3,
#04h
28: 89 0e MOVE R9,
@R3
2a: 3e 06 MOVE R8,
#06h
2c: 99 0e MOVE R1,
@R7
2e: 0b 00 MOVE R1,
#00h
30: 49 28 MOVE R2,
#28h
32: df 0c MOVE R3, PC
34: bd 49 CALL _add
36: c9 09 MOVE R1, R2
38: 3e 02 MOVE R3,
#02h
3a: 8e 49 MOVE @R4,
R4
return 0;
3c: 49 00 MOVE R4,
#00h
}
It executes the whole program and does not halt at "return 0"?
----------------------------------------
Any suggestion help?
Regards
Vineet sharma