Many thanks for your clarifications!
On Wed, Jan 25, 2012 at 11:43 AM, Joerg Wunsch <j...@uriah.heep.sax.de> wrote:
> As ghedamat wrote:
>
> > So to my understending gdb should send multiple step-into
> > instructions to emulate the step-over functionality...
>
> Yes, exactly.
>
> > I'm asking if you have any info about gdb supporting this behavior
> > and/or if you have any hints on how to deal with this issue
>
> Sure, it does. For example, below is the AVaRICE debugging log for
> single-stepping over an instruction like:
>
> PORTH = (PORTH & ~3) | (~i & 3);
>
> That's an ATmega2560, but the basics should be the same for Xmega
> devices. The log is a bit lengthy, but you can clearly see the
> multiple GDB requests ``GDB: <s>'', followed by inquiries about the
> current PC which is reqported back to GDB. GDB repeats those until
> the desired PC for the next statement (0x1b4) is reached.
>
Mmm I saw the log and of course you're right, avr studio 5 does the same
my point is that for me gdb is not doing it but simply doing skip-into
once.
>
> Single-stepping is sometimes a bit tricky, e.g. when the compiler
> inlines some code. Single-stepping over a lengthy loop (like
> _delay_ms()) can take forever, due to the debugger round-trips. (AVR
> Studio 5 also takes forever for that, they cook with the very same
> water.) Usually, it makes more sense to walk along in larger steps
> usign temporary breakpoints (command `tbreak', or `tb') rather than
> trying to single-step.
>
Right, in fact what I did for the moment is writing this dummy macro
define mynext
tbreak +1
c
end
and from my understandings this seems to be the best option.
>
> > Another question,
> >
> > If you've the time, could you explain this phrase
> >
> > "The JTAG ICE's "step over" instruction can only be used the way AVR
> > Studio 4 uses it, by prior filling the tag memory"
> >
> > I'm not able to understand what the prior filling part is about...
>
> The JTAGICEmkII (and AVR Dragon) contains a kind of tag memory (called
> `event memory' in appnote AVR067). Each bit in this memory
> corresponds to one instruction word (16 bits) in the target's flash
> memory area (so covering 256 KiB = 128 Ki instruction words takes 16
> KiB of tag memory). AVR Studio 4, before starting the debugging
> sessions, fills that tag memory and marks each memory location that
> relates to the beginning of one high-level program line of code. This
> information is then used when the JTAG ICE is asked to `step over'.
>
Right, now it's clear :)
>
>
> Here's the avarice -d log I mentioned above:
>
> GDB: <m16a,27>
>
> GDB: Read 39 bytes from 0x16A
> jtagRead
> command[0x05, 1]: 05 A0 28 00 00 00 6A 01 00 00
> recv: 0x1b
> recv: 0x15
> recv: 0x00
> recv: 0x29
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 41 bytes
> read: 82 1f 93 cf 93 df 93 e9 df 83 e1 92 e0 90 93 25 02 80 93 24 02 c0
> e0 d2 e0 00 d0 0f 92 ed b7 fe b7 31 96 ad b7 be b7 12 96
> recv: 0x1f
> recv: 0xa0
> CRC OK
> Got message seqno 21 (command_sequence == 21)
> response: 82 1F 93 CF 93 DF 93 E9 DF 83 E1 92 E0 90 93 25 02 80 93 24 02
> C0 E0 D2 E0 00 D0 0F 92 ED B7 FE B7 31 96 AD B7 BE B7 12 96
> ->GDB:
> 1f93cf93df93e9df83e192e09093250280932402c0e0d2e000d00f92edb7feb73196adb7beb712
> GDB: <m191,11>
>
> GDB: Read 17 bytes from 0x191
> jtagRead
> command[0x05, 1]: 05 A0 12 00 00 00 90 01 00 00
> recv: 0x1b
> recv: 0x16
> recv: 0x00
> recv: 0x13
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 19 bytes
> read: 82 12 96 dc 93 ce 93 11 97 12 83 13 82 18 d0 90 91 02 01
> recv: 0x52
> recv: 0x16
> CRC OK
> Got message seqno 22 (command_sequence == 22)
> response: 82 12 96 DC 93 CE 93 11 97 12 83 13 82 18 D0 90 91 02 01
> ->GDB: 96dc93ce9311971283138218d090910201
> GDB: <Hc0>
> ->GDB:
> GDB: <s>
>
> command[0x09, 1]: 09 01 01
> recv: 0x1b
> recv: 0x17
> recv: 0x00
> recv: 0x01
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 1 bytes
> read: 80
> recv: 0x04
> recv: 0xc5
> CRC OK
> Got message seqno 23 (command_sequence == 23)
> response: 80
> Waiting for input.
> recv: 0x1b
> recv: 0xff
> recv: 0xff
> recv: 0x06
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 6 bytes
> read: e0 d2 00 00 00 00
> recv: 0xf1
> recv: 0xc4
> CRC OK
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x18
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d2 00 00 00
> recv: 0x3d
> recv: 0x95
> CRC OK
> Got message seqno 24 (command_sequence == 24)
> response: 84 D2 00 00 00
>
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x19
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d2 00 00 00
> recv: 0x68
> recv: 0x10
> CRC OK
> Got message seqno 25 (command_sequence == 25)
> response: 84 D2 00 00 00
> jtagRead
> command[0x05, 1]: 05 20 03 00 00 00 5D 00 00 00
> recv: 0x1b
> recv: 0x1a
> recv: 0x00
> recv: 0x04
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 4 bytes
> read: 82 f5 21 40
> recv: 0x15
> recv: 0xe6
> CRC OK
> Got message seqno 26 (command_sequence == 26)
> response: 82 F5 21 40
> ->GDB: T0520:40;21:f521;22:a4010000;
> GDB: <Z0,1a2,2>
> BP ADD type: 1 addr: 0x1a2 ENABLED
> Slot 0 full
> Slot 1 full
> ->GDB: OK
> GDB: <Hc0>
> ->GDB:
> GDB: <s>
>
> command[0x09, 1]: 09 01 01
> recv: 0x1b
> recv: 0x1b
> recv: 0x00
> recv: 0x01
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 1 bytes
> read: 80
> recv: 0x66
> recv: 0xfe
> CRC OK
> Got message seqno 27 (command_sequence == 27)
> response: 80
> Waiting for input.
> recv: 0x1b
> recv: 0xff
> recv: 0xff
> recv: 0x06
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 6 bytes
> read: e0 d3 00 00 00 00
> recv: 0xb5
> recv: 0xcf
> CRC OK
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x1c
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d3 00 00 00
> recv: 0xf0
> recv: 0x8c
> CRC OK
> Got message seqno 28 (command_sequence == 28)
> response: 84 D3 00 00 00
>
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x1d
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d3 00 00 00
> recv: 0xa5
> recv: 0x09
> CRC OK
> Got message seqno 29 (command_sequence == 29)
> response: 84 D3 00 00 00
> jtagRead
> command[0x05, 1]: 05 20 03 00 00 00 5D 00 00 00
> recv: 0x1b
> recv: 0x1e
> recv: 0x00
> recv: 0x04
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 4 bytes
> read: 82 f5 21 55
> recv: 0x6c
> recv: 0xff
> CRC OK
> Got message seqno 30 (command_sequence == 30)
> response: 82 F5 21 55
> ->GDB: T0520:55;21:f521;22:a6010000;
> GDB: <s>
>
> command[0x09, 1]: 09 01 01
> recv: 0x1b
> recv: 0x1f
> recv: 0x00
> recv: 0x01
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 1 bytes
> read: 80
> recv: 0xb8
> recv: 0xe8
> CRC OK
> Got message seqno 31 (command_sequence == 31)
> response: 80
> Waiting for input.
> recv: 0x1b
> recv: 0xff
> recv: 0xff
> recv: 0x06
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 6 bytes
> read: e0 d4 00 00 00 00
> recv: 0x69
> recv: 0xff
> CRC OK
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x20
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d4 00 00 00
> recv: 0x23
> recv: 0xea
> CRC OK
> Got message seqno 32 (command_sequence == 32)
> response: 84 D4 00 00 00
>
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x21
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d4 00 00 00
> recv: 0x76
> recv: 0x6f
> CRC OK
> Got message seqno 33 (command_sequence == 33)
> response: 84 D4 00 00 00
> jtagRead
> command[0x05, 1]: 05 20 03 00 00 00 5D 00 00 00
> recv: 0x1b
> recv: 0x22
> recv: 0x00
> recv: 0x04
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 4 bytes
> read: 82 f5 21 41
> recv: 0xf9
> recv: 0xd8
> CRC OK
> Got message seqno 34 (command_sequence == 34)
> response: 82 F5 21 41
> ->GDB: T0520:41;21:f521;22:a8010000;
> GDB: <s>
>
> command[0x09, 1]: 09 01 01
> recv: 0x1b
> recv: 0x23
> recv: 0x00
> recv: 0x01
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 1 bytes
> read: 80
> recv: 0x52
> recv: 0x3e
> CRC OK
> Got message seqno 35 (command_sequence == 35)
> response: 80
> Waiting for input.
> recv: 0x1b
> recv: 0xff
> recv: 0xff
> recv: 0x06
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 6 bytes
> read: e0 d5 00 00 00 00
> recv: 0x2d
> recv: 0xf4
> CRC OK
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x24
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d5 00 00 00
> recv: 0xee
> recv: 0xf3
> CRC OK
> Got message seqno 36 (command_sequence == 36)
> response: 84 D5 00 00 00
>
> command[0x07, 1]: 07
> recv: 0x1b
> recv: 0x25
> recv: 0x00
> recv: 0x05
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 5 bytes
> read: 84 d5 00 00 00
> recv: 0xbb
> recv: 0x76
> CRC OK
> Got message seqno 37 (command_sequence == 37)
> response: 84 D5 00 00 00
> jtagRead
> command[0x05, 1]: 05 0x00
> recv: 0x1b
> recv: 0x00
> recv: 0x00
> recv: 0x00
> recv: 0x0e
> sDATA: reading 27 bytes
> read: 82 12 96 dc 93 ce 93 11 97 12 83 13 82 18 d0 90 91 02 01 81 2f 80
> 95 83 70 9c 7f
> recv: 0xe1
> recv: 0xfc
> CRC OK
> Got message seqno 56 (command_sequence == 56)
> response: 82 12 96 DC 93 CE 93 11 97 12 83 13 82 18 D0 90 91 02 01 81 2F
> 80 95 83 70 9C 7F
> ->GDB: 96dc93ce9311971283138218d090910201812f809583709c7f
> GDB: <z0,1a2,2>
> BP DEL type: 1 addr: 0x1a2 DISABLED
> ->GDB: OK
>
> --
> cheers, J"org .-.-. --... ...-- -.. . DL8DTL
>
> http://www.sax.de/~joerg/ NIC: JW11-RIPE
> Never trust an operating system you don't have sources for. ;-)
>
Thanks again
Mattia
------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
avarice-user mailing list
avarice-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/avarice-user
