* gdbserver-6.6 threading support
@ 2007-03-06 16:59 Jon Ringle
2007-03-06 17:09 ` Daniel Jacobowitz
0 siblings, 1 reply; 9+ messages in thread
From: Jon Ringle @ 2007-03-06 16:59 UTC (permalink / raw)
To: gdb
Hello all,
I found that during my testing of gdbserver and signals that I was not
actually varying the gdbserver used on the target. I was always using
gdbserver-6.5 and never gdbserver-6.6. When I actually did use
gdbserver-6.6 or gdbserver-6.6.50.20070306-cvs, I get the following from
my crossgdb:
ringlej@crossdev:~/src/gdbsignals-test$
~/build/armeb-linux-gdb-6.6/gdb/gdb ./gdbsignal-test
GNU gdb 6.6
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "--host=i686-pc-linux-gnu
--target=armeb-linux"...
(gdb) handle SIGUSR2 nostop noprint
Signal Stop Print Pass to program Description
SIGUSR2 No No Yes User defined signal 2
(gdb) b 106
Breakpoint 1 at 0x8b1c: file gdbsignal-test.c, line 106.
(gdb) target remote 192.168.205.10:5555
Remote debugging using 192.168.205.10:5555
warning: Remote failure reply: E01
0x400009f0 in _start () from
/scratchbox/compilers/armeb-linux-uclibc/lib/ld-uClibc.so.0
(gdb) cont
Continuing.
Program received signal SIG32, Real-time event 32.
0x400849a4 in __rt_sigsuspend () from
/scratchbox/compilers/armeb-linux-uclibc/lib/libc.so.0
(gdb) show solib-absolute-prefix
Prefix for loading absolute shared library symbol files is
"/scratchbox/compilers/armeb-linux-uclibc".
From other mailing list threads that I had read this seem to imply that
there was some mismatch between my target's libpthread.so and
libthread_db.so and the one in my host's solib-absolute-prefix. I found
that the two sets matched. Further, if I rebuild gdbserver-6.5 and use
the rebuilt gdbserver-6.5, it once again works.
Jon
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: gdbserver-6.6 threading support
2007-03-06 16:59 gdbserver-6.6 threading support Jon Ringle
@ 2007-03-06 17:09 ` Daniel Jacobowitz
2007-03-06 17:23 ` [SPAM] " Jon Ringle
0 siblings, 1 reply; 9+ messages in thread
From: Daniel Jacobowitz @ 2007-03-06 17:09 UTC (permalink / raw)
To: Jon Ringle; +Cc: gdb
On Tue, Mar 06, 2007 at 11:59:09AM -0500, Jon Ringle wrote:
> (gdb) target remote 192.168.205.10:5555
> Remote debugging using 192.168.205.10:5555
> warning: Remote failure reply: E01
Use "set debug remote 1" to see what failed.
> From other mailing list threads that I had read this seem to imply that
> there was some mismatch between my target's libpthread.so and
> libthread_db.so and the one in my host's solib-absolute-prefix. I found
> that the two sets matched. Further, if I rebuild gdbserver-6.5 and use
> the rebuilt gdbserver-6.5, it once again works.
I suggest you debug gdbserver.
Thread debugging is extremely sensitive to your environment.
Especially when uClibc is involved, it is very difficult for us to
diagnose problems.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 17:09 ` Daniel Jacobowitz
@ 2007-03-06 17:23 ` Jon Ringle
2007-03-06 17:38 ` Daniel Jacobowitz
0 siblings, 1 reply; 9+ messages in thread
From: Jon Ringle @ 2007-03-06 17:23 UTC (permalink / raw)
To: Jon Ringle, gdb
Daniel Jacobowitz wrote:
> On Tue, Mar 06, 2007 at 11:59:09AM -0500, Jon Ringle wrote:
>
>> (gdb) target remote 192.168.205.10:5555
>> Remote debugging using 192.168.205.10:5555
>> warning: Remote failure reply: E01
>>
>
> Use "set debug remote 1" to see what failed.
>
>
ringlej@crossdev:~/src/gdbsignals-test$
~/build/armeb-linux-gdb-6.6/gdb/gdb ./gdbsignal-test
GNU gdb 6.6
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "--host=i686-pc-linux-gnu
--target=armeb-linux"...
(gdb) set debug remote 1
(gdb) target remote 192.168.205.10:5555
Remote debugging using 192.168.205.10:5555
Sending packet: $qSupported#37...Ack
Packet received: PacketSize=7cf;qXfer:auxv:read+
Packet qSupported (supported-packets) is supported
Sending packet: $Hc-1#09...Ack
Packet received: E01
Sending packet: $qC#b4...Ack
Packet received:
Sending packet: $qOffsets#4b...Ack
Packet received: E01
warning: Remote failure reply: E01
Sending packet: $?#3f...Ack
Packet received: T050b:00000000;0d:beed6ad0;0f:400009f0;
Sending packet: $m10bfc,e0#ba...Ack
Packet received:
00000001000000010000000100000022000000010000009900000001000000d50000000c000085e40000000d00008b9800000004000080e8000000050000838c00000006000081ac0000000a000001610000000b0000001000000015000000000000000300010cdc00000002000000780000001400000011000000170000856c000000110000854c000000120000002000000013000000086ffffffe0000852c6fffffff000000016ffffff0000084ee000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
Sending packet: $m10bfc,e0#ba...Ack
Packet received:
00000001000000010000000100000022000000010000009900000001000000d50000000c000085e40000000d00008b9800000004000080e8000000050000838c00000006000081ac0000000a000001610000000b0000001000000015000000000000000300010cdc00000002000000780000001400000011000000170000856c000000110000854c000000120000002000000013000000086ffffffe0000852c6fffffff000000016ffffff0000084ee000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
Sending packet: $m10bfc,e0#ba...Ack
Packet received:
00000001000000010000000100000022000000010000009900000001000000d50000000c000085e40000000d00008b9800000004000080e8000000050000838c00000006000081ac0000000a000001610000000b0000001000000015000000000000000300010cdc00000002000000780000001400000011000000170000856c000000110000854c000000120000002000000013000000086ffffffe0000852c6fffffff000000016ffffff0000084ee000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
Sending packet: $qSymbol::#5b...Ack
Packet received:
Packet qSymbol (symbol-lookup) is NOT supported
Sending packet: $qXfer:auxv:read::0,7ca#a5...Ack
Packet received:
l\000\000\000\020\000\000\000\227\000\000\000\006\000\000\020\000\000\000\000\021\000\000\000d\000\000\000\003\000\000\2004\000\000\000\004\000\000\000
\000\000\000\005\000\000\000\005\000\000\000\a@\000\000\000\000\000\000\b\000\000\000\000\000\000\000\t\000\000\206\000\000\000\013\000\000\000\000\000\000\000\f\000\000\000\000\000\000\000\r\000\000\000\000\000\000\000\016\000\000\000\000\000\000\000\027\000\000\000\000\000\000\000\017??000\000\000\000\000\000\000\000
Sending packet: $m400009f0,4#90...Ack
Packet received: e1a0000d
Sending packet: $m400009ec,4#c2...Ack
Packet received: e1a00000
Sending packet: $m400009f0,4#90...Ack
Packet received: e1a0000d
Sending packet: $m400009ec,4#c2...Ack
Packet received: e1a00000
Sending packet: $m400009f0,4#90...Ack
Packet received: e1a0000d
Sending packet: $m400009f0,4#90...Ack
Packet received: e1a0000d
Sending packet: $m400009f0,4#90...Ack
Packet received: e1a0000d
0x400009f0 in _start () from
/scratchbox/compilers/armeb-linux-uclibc/lib/ld-uClibc.so.0
(gdb)
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 17:23 ` [SPAM] " Jon Ringle
@ 2007-03-06 17:38 ` Daniel Jacobowitz
2007-03-06 17:52 ` Jon Ringle
0 siblings, 1 reply; 9+ messages in thread
From: Daniel Jacobowitz @ 2007-03-06 17:38 UTC (permalink / raw)
To: Jon Ringle; +Cc: gdb
On Tue, Mar 06, 2007 at 12:22:29PM -0500, Jon Ringle wrote:
> Sending packet: $qOffsets#4b...Ack
> Packet received: E01
> warning: Remote failure reply: E01
#if defined(__UCLIBC__) && !defined(__UCLIBC_HAS_MMU__)
Is this really Linux, or uClinux? It seems to be configured as if it
were uClinux, but it looks like Linux.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 17:38 ` Daniel Jacobowitz
@ 2007-03-06 17:52 ` Jon Ringle
2007-03-06 18:02 ` Jon Ringle
0 siblings, 1 reply; 9+ messages in thread
From: Jon Ringle @ 2007-03-06 17:52 UTC (permalink / raw)
To: Jon Ringle, gdb
Daniel Jacobowitz wrote:
> On Tue, Mar 06, 2007 at 12:22:29PM -0500, Jon Ringle wrote:
>
>> Sending packet: $qOffsets#4b...Ack
>> Packet received: E01
>> warning: Remote failure reply: E01
>>
>
> #if defined(__UCLIBC__) && !defined(__UCLIBC_HAS_MMU__)
>
> Is this really Linux, or uClinux? It seems to be configured as if it
> were uClinux, but it looks like Linux.
>
>
The target is Linux-2.6.16.29. Processor is IXP455 (it has an MMU). The
rootfs was built with buildroot using uClibc-0.9.28. Perhaps, I've got
something misconfigured in buildroot or uClibc then...
Jon
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 17:52 ` Jon Ringle
@ 2007-03-06 18:02 ` Jon Ringle
2007-03-06 18:17 ` Daniel Jacobowitz
2007-03-06 18:26 ` Jon Ringle
0 siblings, 2 replies; 9+ messages in thread
From: Jon Ringle @ 2007-03-06 18:02 UTC (permalink / raw)
To: Jon Ringle; +Cc: gdb
Jon Ringle wrote:
> Daniel Jacobowitz wrote:
>> On Tue, Mar 06, 2007 at 12:22:29PM -0500, Jon Ringle wrote:
>>
>>> Sending packet: $qOffsets#4b...Ack
>>> Packet received: E01
>>> warning: Remote failure reply: E01
>>>
>>
>> #if defined(__UCLIBC__) && !defined(__UCLIBC_HAS_MMU__)
>>
>> Is this really Linux, or uClinux? It seems to be configured as if it
>> were uClinux, but it looks like Linux.
>>
>>
> The target is Linux-2.6.16.29. Processor is IXP455 (it has an MMU).
> The rootfs was built with buildroot using uClibc-0.9.28. Perhaps, I've
> got something misconfigured in buildroot or uClibc then...
>
I just checked my uClibc-0.9.28 source tree and I can't find any
reference to __UCLIBC_HAS_MMU__. So, __UCLIBC_HAS_MMU__ will not be
defined for anyone that uses uClibc-0.9.28. Where does
__UCLIBC_HAS_MMU__ come from?
ringlej@crossdev:~/src/uClibc-0.9.28$ grep -r __UCLIBC_HAS_MMU__ .
ringlej@crossdev:~/src/uClibc-0.9.28$ grep -r __UCLIBC__ .
./libc/stdio/vfprintf.c:#if defined(__UCLIBC__) &&
!defined(__UCLIBC_HAS_FLOATS__)
./libc/stdio/_stdio.c:#ifndef __UCLIBC__
./libc/stdio/_stdio.c:#endif /* __UCLIBC__ */
./libc/sysdeps/linux/common/ssp.c:#ifdef __UCLIBC__
./test/dlopen/test1.c:#ifdef __UCLIBC__
./test/dlopen/test2.c:#ifdef __UCLIBC__
./test/dlopen/dltest.c:#ifdef __UCLIBC__
./extra/locale/tst_nl_langinfo.c:#if !defined(__UCLIBC__) && 0
./README: #ifdef __UCLIBC__
./include/features.h:#define __UCLIBC__ 1
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 18:02 ` Jon Ringle
@ 2007-03-06 18:17 ` Daniel Jacobowitz
2007-03-07 18:34 ` Nathan Sidwell
2007-03-06 18:26 ` Jon Ringle
1 sibling, 1 reply; 9+ messages in thread
From: Daniel Jacobowitz @ 2007-03-06 18:17 UTC (permalink / raw)
To: Jon Ringle, Nathan Sidwell; +Cc: gdb
On Tue, Mar 06, 2007 at 01:02:03PM -0500, Jon Ringle wrote:
> I just checked my uClibc-0.9.28 source tree and I can't find any
> reference to __UCLIBC_HAS_MMU__. So, __UCLIBC_HAS_MMU__ will not be
> defined for anyone that uses uClibc-0.9.28. Where does
> __UCLIBC_HAS_MMU__ come from?
Good question. Nathan, do you know?
Meanwhile, this is probably not related to your problem.
--
Daniel Jacobowitz
CodeSourcery
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [SPAM] Re: gdbserver-6.6 threading support
2007-03-06 18:02 ` Jon Ringle
2007-03-06 18:17 ` Daniel Jacobowitz
@ 2007-03-06 18:26 ` Jon Ringle
1 sibling, 0 replies; 9+ messages in thread
From: Jon Ringle @ 2007-03-06 18:26 UTC (permalink / raw)
To: Jon Ringle; +Cc: gdb
Jon Ringle wrote:
> Jon Ringle wrote:
>> Daniel Jacobowitz wrote:
>>> On Tue, Mar 06, 2007 at 12:22:29PM -0500, Jon Ringle wrote:
>>>
>>>> Sending packet: $qOffsets#4b...Ack
>>>> Packet received: E01
>>>> warning: Remote failure reply: E01
>>>>
>>>
>>> #if defined(__UCLIBC__) && !defined(__UCLIBC_HAS_MMU__)
>>>
>>> Is this really Linux, or uClinux? It seems to be configured as if it
>>> were uClinux, but it looks like Linux.
>>>
>>>
>> The target is Linux-2.6.16.29. Processor is IXP455 (it has an MMU).
>> The rootfs was built with buildroot using uClibc-0.9.28. Perhaps,
>> I've got something misconfigured in buildroot or uClibc then...
>>
> I just checked my uClibc-0.9.28 source tree and I can't find any
> reference to __UCLIBC_HAS_MMU__. So, __UCLIBC_HAS_MMU__ will not be
> defined for anyone that uses uClibc-0.9.28. Where does
> __UCLIBC_HAS_MMU__ come from?
I do however, have __ARCH_HAS_MMU__ defined.
I found the following Changelog entry in uClibc-0.9.28/Changelog.full
interesting:
---------------------
PatchSet 3180
Date: 2004/01/16 12:43:01
Author: andersen
Branch: HEAD
Tag: (none)
Log:
s/UCLIBC_HAS_MMU/ARCH_HAS_MMU/g
Members:
extra/Configs/Config.alpha:1.17->1.18
extra/Configs/Config.cris:1.11->1.12
extra/Configs/Config.in.arch:1.13->1.14
extra/Configs/Config.powerpc:1.34->1.35
extra/Configs/Config.sh:1.48->1.49
libc/inet/rpc/rcmd.c:1.9->1.10
libc/sysdeps/linux/common/bits/posix_opt.h:1.2->1.3
libc/unistd/daemon.c:1.7->1.8
libpthread/linuxthreads/manager.c:1.7->1.8
Jon
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2007-03-07 18:34 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-03-06 16:59 gdbserver-6.6 threading support Jon Ringle
2007-03-06 17:09 ` Daniel Jacobowitz
2007-03-06 17:23 ` [SPAM] " Jon Ringle
2007-03-06 17:38 ` Daniel Jacobowitz
2007-03-06 17:52 ` Jon Ringle
2007-03-06 18:02 ` Jon Ringle
2007-03-06 18:17 ` Daniel Jacobowitz
2007-03-07 18:34 ` Nathan Sidwell
2007-03-06 18:26 ` Jon Ringle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox