Thanks Cai fpr pointing that out. Forwarding your Patch to Poornima who is also working on LTP-KDUMP for some automation. She will integrate yourś and herś and then give it to me. ** Poornima, can you please look in to this patch as well.
--Subrata-- -------- Forwarded Message -------- From: Cai Qian <[EMAIL PROTECTED]> To: [email protected] Subject: [LTP] [PATCH] fix endless loop hit messages for kdump test case. Date: Tue, 30 Oct 2007 15:09:57 +0800 (CST) Hi, In regards to the ltp-kdump issue where we see endless "loop hit" messages, NMI Watchdog detected LOCKUP on CPU 1 lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit lkdtm : Crash point INT_HARDWARE_ENTRY of type LOOP hit ... The following patch from Don Zickus will fix the issue. Not sure if it is correct with regards to intention of the design, but nonetheless it gets the test by. The reason for the problem seems to stem from the fact there are lots of irqs going on so when the test performs the correct NMI call, executing crash_kexec takes forever because interrupts are still happening on other cpus. The patch just squashes the printks once count falls below zero. This allows the code path to continue at an somewhat sane pace. I'm not entirely sure why after 5 hours the kdump code couldn't execute (despite the numerous printks for every interrupt). Regards, CQ diff -u testcases/kdump/rhtools/lkdtm_mod/lkdtm.c.orig testcases/kdump/rhtools/lkdtm_mod/lkdtm.c --- testcases/kdump/rhtools/lkdtm_mod/lkdtm.c.orig 2007-10-30 15:04:15.000000000 +0800 +++ testcases/kdump/rhtools/lkdtm_mod/lkdtm.c 2007-10-30 15:06:44.000000000 +0800 @@ -235,6 +235,9 @@ void lkdtm_handler(void) { + if (count < 0) + return; + printk(KERN_INFO "lkdtm : Crash point %s of type %s hit\n", cpoint_name, cpoint_type); --count; ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Ltp-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ltp-list ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Ltp-list mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ltp-list
