Hi,
In gdb 5.1, I get an error message in stack backtraces when certain
functions are on the stack: "Cannot convert floating-point register to
non-floating-point type." The variables in question have a type of
pointer, they are not floating point variables. I checked further, and
the regi
When I try to configure gdb 5.0 for Redhat 7.0, it prints recursively:
Configuring intl...
./configure: cd: intl: No such file or directory
Created "Makefile" in /home/lee/src/gdb-5.0 using "mt-frag"
Configuring intl...
./configure: cd: intl: No such file or directory
Created "Makefile" in /home
Found the problem, my CDPATH did not have the current directory in it,
I just set "CDPATH=:$CDPATH" and that fixed it.
Lee
___
Bug-gdb mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-gdb
The easiest way I know to do this is to use a gdb variable:
set var $my_new_var = (struct a_struct *) 0x123456
Then reference $my_new_var as your new symbol.
Lee
___
Bug-gdb mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-gdb
This is a wish-list item, I would like for gdb to support 32-bit thread IDs
(with the high bit set) in remote mode. When the high bit is set on a
thread ID, strtol returns 0x7fff when it tries to parse it,
because the value won't fit in a long.
Lee
___
As I am developing a remote gdb client, I get gdb core dumps from time
to time. That wouldn't be so bad, but I also cannot debug gdb with gdb,
when the segmentation violation happens, I just get:
Program terminated with signal SIGSEGV, Segmentation fault.
The program no longer exists.
I tried se
I thought there was a way to copy gdb output to a file, but I can't
find any such command in the documentation. Does someone know how to
do this?
Thanks,
Lee
___
Bug-gdb mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-gdb
I'm writing a remote protocol monitor, and I'm not sure what to send
to gdb when a breakpoint is encountered. If I send "S00" it says
"Program received signal 0, Signal 0," which is a bit confusing to the
user. But the only other option (the 'T' response) also has a signal
number.
So how can I se
Found the answer: send an "S05" packet. This is also the correct
response to send for a single-step interrupt, so gdb will process the
single-step correctly.
Lee
___
Bug-gdb mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-gdb
I'm not sure how to answer your question, but I used pthreads with gdb
and had lots of problems until I upgraded to 2.4 Linux (Redhat 7.1).
The pthreads for 2.2 Linux is not very stable, it seems.
Just FYI,
Lee
___
Bug-gdb mailing list
[EMAIL PROTECTED
10 matches
Mail list logo