gdb-5.0 disassembler output in console window

[Jay Bromley <jaywb at pacbell dot net>]

Hello,

I seem to have come across a problem with disassembler output in the
latest gdb (I updated my cygwin today, 8/9.) The problem is that if I
open a console window while using Insight and then at some later point
in program execution try to get a disassembly what I get is a list of
addresses without any assembly code.

I tried the same thing using "gdb -nw theprogram.exe" and disassemble
yields a list of addresses which is then followed by the assembler
source code smushed together on a few lines. Below is an example of what
happens.

(gdb) disass
Dump of assembler code for function
add_subscription__Q27TestMkt6TradedPQ27TestM
kt8TestSubs:
0x410e7c <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs>:
0x410e7d <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+1>:
0x410e7f <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+3>:
0x410e85 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+9>:
0x410e86 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+10>:
0x410e87 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+11>:
0x410e88 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+12>:
0x410e8d <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+17>:
0x410e8f <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+19>:
0x410e91 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+21>:
0x410e93 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+23>:
0x410e99 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+29>:
0x410e9c <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+32>:
0x410e9f <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+35>:
0x410ea4 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+40>:
0x410ea5 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+41>:
0x410ea8 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+44>:
0x410ea9 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+45>:
0x410eae <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+50>:
0x410eb1 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+53>:
0x410eb7 <add_subscription__Q27TestMkt6TradedPQ27TestMkt8TestSubs+59>:
push   %ebpmov    %esp,%ebpsub    $0xfc,%esppush   %edipush   %esipush
%ebxcal
l   0x4dbfb4 <__get_eh_context>mov    %eax,%eaxmov    %eax,%edxmov
%edx,%eaxm
ov    %eax,0xffffff1c(%ebp)add    $0xfffffff8,%espmov
0x8(%ebp),%eaxadd    $0
xcc,%eaxpush   %eaxlea    0xfffffff0(%ebp),%eaxpush   %eaxcall
0x501d14 <__Q26
MWUtil9WriteLockPQ26MWUtil4Lock>add    $0x10,%espmov
0xffffff1c(%ebp),%edxadd
    $0x4,%edx---Type <return> to continue, or q <return> to quit--- q
Quit

It seems something is tweaked in the display loop and this causes the
Insight gdb console to spit out the plain addresses.

So the question is is this a known bug, a new bug, or is my gdb just
tweaked? If a new bug a pointer to the code to check out to fix this
would be handy.

Thanks and regards,
Jay Bromley


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com