Messages in this thread | | | Date | Fri, 18 Sep 1998 03:24:38 +0000 | From | Trever Adams <> | Subject | Scheduling changes... freezes |
| |
I am still trying to figure out what is going on, so please take this report with a grain of salt. I have had great success with 2.1.119. I didn't try .120 or .121. I am currently running 2.1.122. Nothing else has changed on my system (config available on request).
I am having strange pauses. Yes, I have APM enabled, yes my hard drives spin down. Funny thing is, I listen when these 'hesitations' are happening. The drives are already spun up or don't spin up. Sometimes it is just the programs that freeze up and things like the mouse cursor (X or console) will still move. Sometimes it is just one program. But usually, it tends to freeze the entire system for somewhere between 3 and 30 seconds. I think the average would be somewhere about 7 seconds.
I don't seem to be experience data corruption, disk corruption, or memory corruption (yes I know that the first is a type of second or third nearly always). Networking seems to be working more or less the same as before, even with the hesitations. I am currently working on small programs that are part of a much larger project... a fairly complex (compared to what I have been doing for the past three years... all probably child's play to most on this list). Because of this, I save often, compile often, and test often. I haven't had any programs crash or anything else. Just that pause.
I saw Linus's post about the scheduling changes, and I understand there were a multitude of other changes in general and for the x86 in specific (Cyrix MII-300, 128 meg ram, Fire Port dual 40, Western Digital 3.2 gig drives and two scsi cdroms (one 24x one 6x2 cdrw)). This is my machine at work. I am wondering if there were some "race" (overly loop friendly code that doesn't like to leave a loop or a lock or something when it should) in some of the new code. One last date point, usually it is after something just executed and then while it is in the idle loop in the kernel just before I do something else. APM only has enable apm checked. All else is off, screen blanking, make idle calls, etc.
I hope this bug report is helpful. I will keep trying to figure out what exactly is causing it. If someone else has a similar problem and suspects a specific cause please let me know so I can test to see if that is it here as well.
Thank you much, Trever Adams highlander@teleteam.net
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |