Bug 18182

 
 	Since about mid-April, remote kernel debugging has become
 	non-functional: there are so many transmission errors on the
 	link that communication is effectively impossible.  Debugging
 	printouts in the kernel show that the input routines are
 	dropping several characters in sucession.
 
 	The following examples are from a dual processor Celeron, but
 	this does not appear to be an SMP bug.  Older SP kernels work,
 	and current SP kernels do not.  No older SMP kernel is
 	available for testing, but I am currently building one.
 
 	This example shows that there is no trouble entering gdb from
 	the keyboard (ctrl-alt-esc).  This works reliably.  On the
 	other hand, a breakpoint placed on execve repeatedly fails to
 	communicate, to the point where it is no longer possible to
 	resume execution.  This looks like an interrupt problem, but I
 	checked the ipl at the execve entry point, and as expected it
 	was set to 0.
 
 	$ gdb -k
 	GNU gdb 4.18
 	Copyright 1998 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 "i386-unknown-freebsd".
 	Debugger (msg=0xc0354709 "manual escape to debugger") at ../../i386/i386/db_interface.c:325
 	325     }
 	(kgdb) bt
 	#0  Debugger (msg=0xc0354709 "manual escape to debugger") at ../../i386/i386/db_interface.c:325
 	#1  0xc02e088a in scgetc (sc=0xc03aca80, flags=0x2) at ../../dev/syscons/syscons.c:3124
 	#2  0xc02dd2d1 in sckbdevent (thiskbd=0xc03a51a0, event=0x0, arg=0xc03aca80) at ../../dev/syscons/syscons.c:634
 	#3  0xc02d52c6 in atkbd_intr (kbd=0xc03a51a0, arg=0x0) at ../../dev/kbd/atkbd.c:462
 	#4  0xc0306814 in atkbd_isa_intr (arg=0xc03a51a0) at ../../isa/atkbd_isa.c:125
 	(kgdb) b execve
 	Breakpoint 1 at 0xc01aa2bc: file ../../kern/kern_exec.c, line 103.
 	(kgdb) c
 	Continuing.
 
 	Breakpoint 1, 0xc01aa2bc in execve (p=0xcf56b3c0, uap=0xcff72f80) at ../../kern/kern_exec.c:103
 	103     {
 	(kgdb) bt
 	#0  0xc01aa2bc in execve (Ignoring packet error, continuing...
 	p=Reply contains invalid hex digit 116
 	) at ../../kern/kern_exec.c:103
 	#1  0xc02f49e5 in syscall2 (frame={tf_fs = 0x2f, tf_es = 0x2f, tf_ds = 0x2f, tf_edi = 0x7c1c0, tf_esi = 0x7c0a0, 
 	      tf_ebp = 0xbfbffc4c, tf_isp = 0xcff72fd4, tf_ebx = 0x7c170, tf_edx = 0x7c010, tf_ecx = 0x4, tf_eax = 0x3b, 
 	      tf_trapno = 0xc, tf_err = 0x7, tf_eip = 0x20098411, tf_cs = 0x1f, tf_eflags = 0x292, tf_esp = 0xbfbffb7c, 
 	      tf_ss = 0x2f}) at ../../i386/i386/trap.c:1130
 	#2  0x20098411 in ?? ()
 	#3  0xe0ff in ?? ()
 	#4  0xd90b in ??Ignoring packet error, continuing...
 	 ()Reply contains invalid hex digit 116
 	(kgdb) bt
 	#0  0xc01aa2bc in execve (Ignoring packet error, continuing...
 	p=Reply contains invalid hex digit 116
 	) at ../../kern/kern_exec.c:103
 	^C#1  0xc02f49e5 in syscall2 (Ignoring packet error, continuing...
 	frame=Reply contains invalid hex digit 116
 	) at ../../i386/i386/trap.c:1130
 	#2  0x20098411 in ??Ignoring packet error, continuing...
 	 ()Reply contains invalid hex digit 116
 	(kgdb)

Fix: 

No fix known.
How-To-Repeat:  
 	As shown above.