(no subject)

[getelson at softhome.net]

Hello, 

I try to "port" a kdb patch to the redhat kernel 2.4.18-5.
Meanwile it looks pretty stable exept this problem:
`ss' command terminates the kdb shell without producing any output.
Next attempt to re-enter kdb shell will hang a system. 

I'm looking for any advise about a couse of such behaviour. 

I work with PII SMP architecture.
Kdb patch files I use are:
 kdb-v2.3-2.4.18-common-2
 kdb-v2.3-2.4.18-i386-3 

Here is the log  of the `ss' command with all kdb debugging flags enabled: 

Entering kdb (current=0xc1e7a000, pid 0) on processor 1 due to Keyboard 
Entry 

[1]kdb> mm kdb_flags 0x5f0000 

0xc039d07c = 0x5f0000 

[1]kdb> ss 

state: kdb_main_loop 3 cpu 1 value -1003 initial 1 state 1000429 

state: kdb 10 cpu 1 value 1 initial 1 state 1000429 

state: kdb 11 cpu 1 value 1 initial 1 state 1000029 

state: kdb 13 cpu 1 value 1 initial 1 state 1000029 

state: kdb 14 cpu 1 value 1 initial 1 state 1000028 

state: kdb 17 cpu 1 value 8 initial 1 state 1000028 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang
Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

Warning: get_irqlock on cpu 1 while kdb is running, may hang 

sh-2.05a#
/**
******************** <re-entry point > ****************************
**/
state: kdb 1 cpu 1 value 8 initial 1 state 1000028
state: kdb 3 cpu 1 value 8 initial 1 state 1000028
kdb: Debugger re-entered on cpu 1, new reason = 8
    Not executing a kdb command
    No longjmp available for recovery
    Cannot recover, allowing event to proceed