Adam C Powell IV wrote:
> Greetings,
>
> Simply running "minicom /dev/ttyS0" or S1 on my machine with 2.2.15pre
> (since pre5, I've tested through pre10) slows it to a crawl (at most 1%
> of CPU is available for everything else, sometimes less, I'm using E
> ripples as a heartbeat though the problem occurs with or without X
> running).
>
> The funny thing is, if I revert to 2.2.13 and try "minicom /dev/ttyS1" it
> says something like ttyS1 is locked. (2.2.14 doesn't boot on PPC.)
Very odd. Minicom now runs for me with both 2.2.15pre19 and 2.2.13, so
whatever was locking that device is no more.
But now with pre19, the following locks up the system:
setserial /dev/ttyS1 baud_base 115200
This works fine on 2.2.13. This is obviously not a problem with minicom or
setserial, as neither should be able to lock the system and the behavior is
kernel version-dependent. I'm very much a serial novice, please let me
know what I might try to help diagnose the real problem here.
> Machine details: StarMax 3000/160 (Motorola pmac 4400 clone) with 160 MHz
> 603e, Debian potato with setserial 2.17-8 running from /dev/hdb6 and 7.
> [By the way, the machine used to *always* panic on power-up boot during
> IDE initialization, and *always* succeed on reboot, now it never panics
> on power-up boot. Thanks very much to whoever's responsible!]
Now that I have quik working, the IDE init kernel panic is back. It always
happens when cold booting from the zip disk at /dev/sda4, never when
booting from the floppy. A single reboot makes it work, so this is not so
important.
> One other thing: I still haven't seen any changes to
> Documentation/md.txt, Documentation/Changes, Documentation/Configure.help
> or drivers/block/Config.in to reflect the unmaintained nature of the RAID
> drivers in the stock kernel in general, and RAID5/IDE/SMP brokenness in
> particular. This is a very serious documentation bug. As I've flamed
> before on this list [linux-kernel], it cost me several days of system
> downtime and of my own time last Fall, and people shouldn't need to be on
> linux-kernel to be able to use features in a stable kernel. Please
> correct this for 2.2.15, with the patch I sent to this list 1/25 or
> something similar.
This persists to pre19. Please treat it as the serious bug that it is.
> Please cc me on all replies.
TIA,
-Adam P.
-
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/
This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:15 EST