How to determine the order of the registers got, their width and count?

2014-06-06 Thread Vardhana, Ananda
I am newbie in GDB. I had couple of questions

1.  When I send out 'g' I get a full dump of the registers. Is there 'q' 
query packet I can send and ask the GDB server in what order(eax, ebx, ecx ...) 
will it dump the registers?

2.  Before we do a 'g' can we tell the server that we are interested in 32 
bit (EAX, ECX etc ...) only?

3.  Can I send out a 'q' packet and ask the server whether it will give me 
back 32 or 64 bit registers?

4.  Nowadays it is mostly 64 bit systems and they support both 32 and 64 
bit. So my idea was to ask whatever I want.

Please check out these transactions:

Sending packet: $qSupported:multiprocess+;xmlRegisters=i386;qRelocInsn+#b5...Ack
Packet received: ReverseStep+;ReverseContinue+;qXfer:features:read+
Packet qSupported (supported-packets) is supported
Sending packet: $Hg0#df...Ack
Packet received: OK
Sending packet: $qXfer:features:read:target.xml:0,283#e8...Ack
Packet received: m\n  
i386:x86-64\n\n
Sending packet: $qXfer:features:read:target.xml:4c,283#4f...Ack
Packet received: l

1. Question is how did it come up with the number 0,283
2. What range does 0,283 imply? Is it the number of lines in the target.xml 
file?
3. How can I get the total count?
4. Please see that the next time it does it from 4c,283. Where did it get these 
numbers?
 >> Packet received: m\n  
 >> i386:x86-64\n\n
5.  I guess i386:x86-64 means the system is an 
Intel system(i386), it supports both 32 bit(x86) and 64 bit (64). Is that right?


Sorry for the long list of questions, just wanted to spill out ... . Many 
thanks in advance

Cheers
 Ananda


___
bug-gdb mailing list
bug-gdb@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gdb


RE: Usage of vCont packet please

2014-05-26 Thread Vardhana, Ananda
Hi Sergio,

Thanks for replying. It compiles fine but the server complains with the 
following error:

[gdb0 info] got message: "vCont?"
[gdb0 info] Sending packet: "vCont;c;C;s;S"
[gdb0 info] got message: "vCont;c:-1"
[gdb0 error] Malformed vCont packet ";c:-1", ignoring

I guess the server is dated and I shall see if I can fix the server side. The 
comment for vCont in the GDB website says vCont is new and using H is being 
obsoleted. I did try H and the server liked it. If you have any comments on my 
conlcusion please let me know

Thanks
Ananda

-Original Message-
From: Sergio Durigan Junior [mailto:sergi...@sergiodj.net] 
Sent: Thursday, May 22, 2014 5:34 PM
To: Vardhana, Ananda
Cc: bug-gdb@gnu.org
Subject: Re: Usage of vCont packet please

On Thursday, May 22 2014, Ananda Vardhana wrote:

> Right now I send the packet Hc-1 meaning continue or make the target 
> run and the -1 signifies do it on all process. It works fine.  What 
> would be the equivalent for that in vCont? I do check and vCont?
> Support and the server does support it.

Hello Ananda,

You mean to use vCont to resume all threads of the inferior, right?

> vCont;c-1 does not work and whatever I do with vcont it hangs. The 
> documentation in 
> https://sourceware.org/gdb/onlinedocs/gdb/Packets.html#thread%2did%20s
> yntax
>
> Does not seem to help or I am misinterpreting.  If you point me to URL 
> with usage, meaning and example if vCont that would be ideal

It should be "vCont;c:-1" (note the colon between "c" and "-1").  Could you try 
that?

Thanks,

--
Sergio

___
bug-gdb mailing list
bug-gdb@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gdb


Usage of vCont packet please

2014-05-22 Thread Vardhana, Ananda
Right now I send the packet Hc-1 meaning continue or make the target run and 
the -1 signifies do it on all process. It works fine.  What would be the 
equivalent for that in vCont? I do check and vCont? Support and the server does 
support it.

vCont;c-1 does not work and whatever I do with vcont it hangs. The 
documentation in
https://sourceware.org/gdb/onlinedocs/gdb/Packets.html#thread%2did%20syntax

Does not seem to help or I am misinterpreting.  If you point me to URL with 
usage, meaning and example if vCont that would be ideal

Thanks





___
bug-gdb mailing list
bug-gdb@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gdb