Re: Jeeni & ARM720T with GDB

[Grant Edwards <grante@visi.com>]

On Wed, Nov 22, 2000 at 10:35:04AM -0000, danish iftikhar wrote:

> After load Pc shows the address : 0x8060
> 
> The start uo code is :
> (gdb) x/30i $pc
> 0x8060 <warm_reset>:    mrs     r7, cpsr
> 0x8064 <warm_reset+4>:  and     r7, r7, #31     ; 0x1f
> 0x8068 <warm_reset+8>:  cmp     r7, #19 ; 0x13
> 0x806c <warm_reset+12>: beq     0x80a0 <start>

That looks right.

>   i am not able to do step or cont but using stepi i went through this 
> startup function . It seems to loop around between 0x815c & 0x8164 and is 
> not going ahead of that .
> the code at this address is as follows :
> 
> 0x8154  <start+180>:            cmp     r1, r2
> -       0x8158  <start+184>:            beq     0x8168 <start+200>
> -       0x815c  <start+188>:            str     r0, [r1], #4
> -       0x8160  <start+192>:            cmp     r1, r2
> -       0x8164  <start+196>:            bne     0x815c <start+188>
> -       0x8168  <start+200>:            bl      0x1f194 <hal_hardware_init>
> -       0x816c  <start+204>:            bl      0x1ee10 
> <cyg_hal_invoke_constructors>
> -       0x8170  <start+208>:            bl      0x1f358 <cyg_start>

It looks to me like the debugger and emulator are working fine.
The loop you're looking at is zeroing out the bss section.  R2
should contain the address of the end of bss, and R1 should
start at the beginning of bss and increment through the whole
section.  This loop will probably execute tens of thousands of
times, so you can get pretty bored doing a stepi...

Do the addresses in R1 and R2 correspond to your bss section?

Can you set a breakpoint immediately following the loop and do
a cont?

-- 
Grant Edwards
grante@visi.com
From jimb@zwingli.cygnus.com Wed Nov 22 07:21:00 2000
From: Jim Blandy <jimb@zwingli.cygnus.com>
To: David Taylor <taylor@cygnus.com>
Cc: gdb@sourceware.cygnus.com, gdb-patches@sourceware.cygnus.com
Subject: Re: pathmap (again)
Date: Wed, 22 Nov 2000 07:21:00 -0000
Message-id: <np1yw43ud3.fsf@zwingli.cygnus.com>
References: <200011212037.PAA22391@texas.cygnus.com>
X-SW-Source: 2000-11/msg00229.html
Content-length: 86

> Comments?

It looks reasonable.  I'd say, give it a try and see what annoys people.
From taylor@cygnus.com Wed Nov 22 08:54:00 2000
From: David Taylor <taylor@cygnus.com>
To: gdb@sourceware.cygnus.com
Cc: gdb-patches@sourceware.cygnus.com
Subject: pathmap semantics issues
Date: Wed, 22 Nov 2000 08:54:00 -0000
Message-id: <200011221653.LAA22540@texas.cygnus.com>
X-SW-Source: 2000-11/msg00230.html
Content-length: 402

Okay, here's some questions concerning how pathmap should behave:

. should it translate all paths?  Or just some paths?  If some, which?
Source paths?  Others?  That is, should it be consulted when looking
for sources?, objects?, executables?, and shared objects?

. should it be searched instead of the existing path?  Or in addition?
If in addition, which should be searched first?

Do people care?
From cgf@redhat.com Wed Nov 22 08:58:00 2000
From: Christopher Faylor <cgf@redhat.com>
To: gdb@sources.redhat.com, gdb-patches@sources.redhat.com
Subject: Re: pathmap semantics issues
Date: Wed, 22 Nov 2000 08:58:00 -0000
Message-id: <20001122115802.B7346@redhat.com>
References: <200011221653.LAA22540@texas.cygnus.com>
X-SW-Source: 2000-11/msg00231.html
Content-length: 769

On Wed, Nov 22, 2000 at 11:53:38AM -0500, David Taylor wrote:
>Okay, here's some questions concerning how pathmap should behave:
>
>. should it translate all paths?  Or just some paths?  If some, which?
>Source paths?  Others?  That is, should it be consulted when looking
>for sources?, objects?, executables?, and shared objects?
>
>. should it be searched instead of the existing path?  Or in addition?
>If in addition, which should be searched first?
>
>Do people care?

I would think that it should be "all paths".  If you've gone to the
effort of issuing the pathmap command then it must be for the reason
that you don't have the specific path on your system and need to
translate it.  So, using the pathmap translation for everything would
make sense, IMO.

cgf
From msnyder@redhat.com Wed Nov 22 09:17:00 2000
From: Michael Snyder <msnyder@redhat.com>
To: David Taylor <taylor@cygnus.com>
Cc: gdb@sourceware.cygnus.com, gdb-patches@sourceware.cygnus.com
Subject: Re: pathmap semantics issues
Date: Wed, 22 Nov 2000 09:17:00 -0000
Message-id: <3A1BFFAF.1E9A@redhat.com>
References: <200011221653.LAA22540@texas.cygnus.com>
X-SW-Source: 2000-11/msg00232.html
Content-length: 550

David Taylor wrote:
> 
> Okay, here's some questions concerning how pathmap should behave:
> 
> . should it translate all paths?  Or just some paths?  If some, which?
> Source paths?  Others?  That is, should it be consulted when looking
> for sources?, objects?, executables?, and shared objects?
> 
> . should it be searched instead of the existing path?  Or in addition?
> If in addition, which should be searched first?


I hope it won't interfere with the path for searching for 
shared libraries.  I've just spent some effort sanifying that...
From ac131313@cygnus.com Wed Nov 22 20:55:00 2000
From: Andrew Cagney <ac131313@cygnus.com>
To: GDB Discussion <gdb@sourceware.cygnus.com>
Subject: gdb@gnu.org Discussion
Date: Wed, 22 Nov 2000 20:55:00 -0000
Message-id: <3A1CA166.A24792EF@cygnus.com>
X-SW-Source: 2000-11/msg00233.html
Content-length: 148

People on this mailing list should be aware of the discussion:

http://sources.redhat.com/ml/overseers/2000-q4/threads.html#00214

	sorry,
		Andrew
From d_iftikhar@hotmail.com Wed Nov 22 23:15:00 2000
From: "danish iftikhar" <d_iftikhar@hotmail.com>
To: grante@visi.com
Cc: fnasser@cygnus.com, gdb@sourceware.cygnus.com
Subject: Re: Jeeni & ARM720T with GDB
Date: Wed, 22 Nov 2000 23:15:00 -0000
Message-id: <F285Ca3Eky00sp0yx2p0000234c@hotmail.com>
X-SW-Source: 2000-11/msg00234.html
Content-length: 3399

  Hi
     i went through the code and checked the addresses of bss as well as 
other sections and they are perfectly o.k.
  ya , grant , it's zeroing the bss address only and i was able to come out 
of that loop .
where it stucks is in <cyg_hal_invoke_constructors> : take a look at code 
below :


> > 0x8154  <start+180>:            cmp     r1, r2
> > -       0x8158  <start+184>:            beq     0x8168 <start+200>
> > -       0x815c  <start+188>:            str     r0, [r1], #4
> > -       0x8160  <start+192>:            cmp     r1, r2
> > -       0x8164  <start+196>:            bne     0x815c <start+188>
> > -       0x8168  <start+200>:            bl      0x1f194 
><hal_hardware_init>
> > -       0x816c  <start+204>:            bl      0x1ee10
<cyg_hal_invoke_constructors>
> > > > -       0x8170  <start+208>:            bl      0x1f358 <cyg_start>


  things seem to work perfectly fine for h/w initialisation .
  i wonder whether some special options have to be configured in ecos build 
to make it work. i have a basic doubt ..how the diagnostic output is handled 
.
i don't think that jeeni will be able to send it back to console ..
or will connecting through the serial port work .
is the diag printf 's are the one causing it to hang.

thanks
danish.



>From: Grant Edwards <grante@visi.com>
>To: danish iftikhar <d_iftikhar@hotmail.com>
>CC: fnasser@cygnus.com, gdb@sourceware.cygnus.com
>Subject: Re: Jeeni & ARM720T with GDB
>Date: Wed, 22 Nov 2000 08:39:21 -0600
>
>On Wed, Nov 22, 2000 at 10:35:04AM -0000, danish iftikhar wrote:
>
> > After load Pc shows the address : 0x8060
> >
> > The start uo code is :
> > (gdb) x/30i $pc
> > 0x8060 <warm_reset>:    mrs     r7, cpsr
> > 0x8064 <warm_reset+4>:  and     r7, r7, #31     ; 0x1f
> > 0x8068 <warm_reset+8>:  cmp     r7, #19 ; 0x13
> > 0x806c <warm_reset+12>: beq     0x80a0 <start>
>
>That looks right.
>
> >   i am not able to do step or cont but using stepi i went through this
> > startup function . It seems to loop around between 0x815c & 0x8164 and 
>is
> > not going ahead of that .
> > the code at this address is as follows :
> >
> > 0x8154  <start+180>:            cmp     r1, r2
> > -       0x8158  <start+184>:            beq     0x8168 <start+200>
> > -       0x815c  <start+188>:            str     r0, [r1], #4
> > -       0x8160  <start+192>:            cmp     r1, r2
> > -       0x8164  <start+196>:            bne     0x815c <start+188>
> > -       0x8168  <start+200>:            bl      0x1f194 
><hal_hardware_init>
> > -       0x816c  <start+204>:            bl      0x1ee10
> > <cyg_hal_invoke_constructors>
> > -       0x8170  <start+208>:            bl      0x1f358 <cyg_start>
>
>It looks to me like the debugger and emulator are working fine.
>The loop you're looking at is zeroing out the bss section.  R2
>should contain the address of the end of bss, and R1 should
>start at the beginning of bss and increment through the whole
>section.  This loop will probably execute tens of thousands of
>times, so you can get pretty bored doing a stepi...
>
>Do the addresses in R1 and R2 correspond to your bss section?
>
>Can you set a breakpoint immediately following the loop and do
>a cont?
>
>--
>Grant Edwards
>grante@visi.com

_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com