I've tracked down the problem to something in mca.c: the second call to get SAL error logs causes a partial reboot instead of returning error logs. I instrumented bits of mca.c to see what was going on (hope the printk output is self-explanatory). This is on a uniprocessor i2000. calling initcall 0xa00000010070fbe8 ( __initcall_ia64_mca_check_errors ) ia64_mca_log_sal_error_record ia64_log_get() ia64_log_get -- log_lock ia64_log_get -- sal_get_state_info SAL_CALL(isrv, SAL_GET_STATE_INFO, ...) sal_get_state_info returned 0 log_get returned null calling initcall 0xa00000010070fbf0 ( __initcall_ia64_mca_late_init ) init_timer(cmc_poll_timer) init_tiemr(cpe_poll_timer) acpi_request_vector(ACPI_INTERRUPT_CPE1) (obviously couldn't get it...) ia64_mca_cpe_poll set start_count smp call ia64_mca_cpe_int_caller local call ia64_mca_cpe_int_caller ia64_mca_cpe_int_handler ia64_mca_log_sal_error_record ia64_log_get() ia64_log_get -- log_lock ia64_log_get -- sal_get_state_info SAL_CALL(isrv, SAL_GET_STATE_INFO, ...) (Monitor goes blank, I hear the floppy drive seek, and machine becomes totally unresponsive and needs to be power cycled to continue) Does anyone have any ideas? -- Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au You are lost in a maze of BitKeeper repositories, all slightly different.Received on Mon May 26 14:21:25 2003
This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:15 EST