We are running a survey to help us improve the experience for all of our members. If you see the survey appear, please take the time to tell us about your experience if you can.
Hi All,
I need to set an conditional breakpoint to identify the location of the memory overwrite on a variable. So I am trying to place a breakpoint with Expression as mentioned below,
sabcAppQue.tx_queue_id == 0x00000000
but returning an error as "undefined identifier" even after I selected the option as Write,objects and mentioned its size as 4.
Note:The above variable sabcAppQue.tx_queue_id is belong to datatype ULONG and its value is getting changed from 0x51545555 to 0x00000000.
hence I have to identify the exact location.
kindly suggest me whether the above expression is correct and to solve the error.
Is the identifier in scope at the time you try to set the breakpoint?
If not, have you tried using the "fully-qualified" name:
http://www.keil.com/support/man/docs/uv4/uv4_db_exp_symb_fq.htm
www.keil.com/.../uv4_db_exp_progvariables.htm
http://www.keil.com/support/man/docs/uv4/uv4_db_dbg_breakpnts.htm
http://www.keil.com/support/man/docs/uv4/uv4_db_expressions.htm
Hi Neil,
Thanks for reply. The problem gets solved.
The problem is that I tried to set breakpoint during the variable is in out of scope.
Even though the execution stops on that condition I can't able to view or find out the exact location/file of memory corruption of that variable.
Is it too possible,if my project contains more that 100 files?
Vignesh.
Just out of scope, or possibly out of existence, too? Automatic variables (most people will speak of them as "local") don't exist while they're out of scope. Trying to install a data breakpoint for one of those is quite pointless, because they don't actually own any memory address all to themselves, so other parts of the program will modify the same memory, but that doesn't mean anything happened to the variable you're studying.
You may be on a wild goose chace there. I suggest you take a couple steps back and analyze what actual symptoms you have, and whether the current hypothesis, of data corruption by something else in the program, is really the only one that could explain those symptoms.