Re: AW: [Simulavr-devel] [patch #3453] gdbserver.c gdb_extract_hex_nu m(...) bug fix (for more recent gdb versions)

[Paul Schlie]

As a clarification, I was attempting to answer the same question as Frost
answered; I didn't mean to imply any contradiction; the patch and his
example are both correct.

(as a heads up, I believe similar problems may still exist in gdbserver, as
I'm still in the process of attempting to track down and fix what appear to
be remaining simulavr-gdb command/request misinterpretations, although the
fix did allow substantial progress.)

Thanks -paul-


> From: Paul Schlie <address@hidden>
> 
> num is an integer, not a string, hex nibbles are 4 bit's wide,
> hence to build an integer out of a sequence of hex-chars->nibbles
> one shifts the integer by 4.
> 
> 
>> From: Frost_Tobias <address@hidden>
>> Date: Tue, 19 Oct 2004 09:13:37 +0200
>> To: <address@hidden>
>> Subject: AW: [Simulavr-devel] [patch #3453] gdbserver.c gdb_extract_hex_nu
>> m(...) bug fix (for more recent gdb versions)
>> 
>> No, the patch's correct.
>> Num will be shifted by 4 every time the while is executed and then stored.
>> Then the low nibble will be ored into num
>> 
>> Example:
>> 
>> *p = ABCDEF
>> 
>> will be: (each line is one while-execution)
>> 
>>     num         
>> 0x0000 000A     
>> 0x0000 00AB
>> 0x0000 0ABC
>> 0x0000 ABCD
>> 0x000A BCDE
>> 0x00AB CDEF 
>> 
>>  num = (num << (i * 4)) | hex2nib (*p);  (replaced by following line) */
>> 
>> without the patch you'll get:
>> 
>> (i=0) -> no shift         0x0000 000A
>> (i=1 -> shifted with 4) ->     num = 0x0000 00A0  | 0x0B  -> num = 0x0000
>> 00AB
>> (i=2 -> num shifted with 8) -> num = 0x0000 AB00 | 0x0C -> num = 0x0000 AB0C
>> (ERROR!)
>> 
>>   
>>  
>> 
>> -----Ursprüngliche Nachricht-----
>> Von: Klaus Rudolph [mailto:address@hidden
>> Gesendet: Montag, 18. Oktober 2004 17:16
>> An: address@hidden
>> Betreff: [Simulavr-devel] [patch #3453] gdbserver.c
>> gdb_extract_hex_num(...) bug fix (for more recent gdb versions)
>> 
>> 
>> This mail is an automated notification from the patch tracker
>>  of the project: Simulavr: an AVR simulator.
>> 
>> /**************************************************************************/
>> [patch #3453] Latest Modifications:
>> 
>> Changes by: 
>> Klaus Rudolph <address@hidden>
>> 'Date: 
>> Mon 18.10.2004 at 15:07 (Europe/Berlin)
>> 
>> ------------------ Additional Follow-up Comments
>> ----------------------------
>> Can you give me an idea what problem the patch should fix?
>> 
>> Your new line:
>>  num = (num << 4) | hex2nib (*p);
>> could never work I think. This will always change
>> the "0000000X0" position but nothing else!?.
>> If the tranmitted var is longer you have to change
>> the init of max_shifts but that is not done....
>> 
>> maybe I am totally confused today, but please let us know
>> what your problem is.
>> 
>> Why you have not send a patch file as patch?
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> /**************************************************************************/
>> [patch #3453] Full Item Snapshot:
>> 
>> URL: <http://savannah.nongnu.org/patch/?func=detailitem&item_id=3453>
>> Project: Simulavr: an AVR simulator
>> Submitted by: Paul Schlie
>> On: Mon 18.10.2004 at 12:48
>> 
>> Category:  bug fix
>> Priority:  9 - Immediate
>> Resolution:  None
>> Privacy:  Public
>> Assigned to:  None
>> Originator Email:
>> Status:  Open
>> 
>> 
>> Summary:  gdbserver.c gdb_extract_hex_num(...) bug fix (for more recent gdb
>> versions)
>> 
>> Original Submission:  static int
>> gdb_extract_hex_num (char **pkt, char stop)
>> {
>>     int i = 0;
>>     int num = 0;
>>     char *p = *pkt;
>>     int max_shifts = sizeof (int) * 2 - 1; /* max number of nibbles to shift
>>                                               through */
>> 
>>     while ((*p != stop) && (*p != ''))
>>     {
>>         if (i > max_shifts)
>>             avr_error ("number too large");
>> 
>> /* This is broken for values encoded with greater than two hex digits, which
>>     is typical in gdb sessions when leading 0's are encoded for integer
>> values
>>         num = (num << (i * 4)) | hex2nib (*p);  (replaced by following line)
>> */
>>         num = (num << 4) | hex2nib (*p);
>>         i++;
>>         p++;
>>     }
>> 
>>     *pkt = p;
>>     return num;
>> }
>> 
>> 
>> Follow-up Comments
>> ------------------
>> 
>> 
>> -------------------------------------------------------
>> Date: Mon 18.10.2004 at 15:07       By: Klaus Rudolph <zfrdh>
>> Can you give me an idea what problem the patch should fix?
>> 
>> Your new line:
>>  num = (num << 4) | hex2nib (*p);
>> could never work I think. This will always change
>> the "0000000X0" position but nothing else!?.
>> If the tranmitted var is longer you have to change
>> the init of max_shifts but that is not done....
>> 
>> maybe I am totally confused today, but please let us know
>> what your problem is.
>> 
>> Why you have not send a patch file as patch?
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> For detailed info, follow this link:
>> <http://savannah.nongnu.org/patch/?func=detailitem&item_id=3453>
>> 
>> _______________________________________________
>>   Message sent via/by Savannah
>>   http://savannah.nongnu.org/
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> Simulavr-devel mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/simulavr-devel
>> 
>> 
>> _______________________________________________
>> Simulavr-devel mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/simulavr-devel