2.2.14 "stuck on TLB IPI wait (CPU#1)"

From: Frank Baumgart (f.baumgart@myokay.net)
Date: Tue Jan 11 2000 - 07:13:03 EST


linux 2.2.14 + reiserfs 3.5.14
Gigabyte GA686BXD (MP 1.4)
Dual PIII/450, 256 M, 1.1 GB swap
6 + 13 + 40 GB EIDE, ATAPI CD-ROM, 4 SCSI CD-ROMs, SCSI Tape.
AGP TNT (not used), PCI Matrox G200 (used), NCR 810, Miro bttv card,
AWE32, 10 MBit PCI NE2k clone (RTL), ISDN Teles 16.3
XFree 3.3.6, glibc 2.1.2, gcc 2.95.2

The system was *completely* unresponsive for about a minute after which
the following appeared in messages:

Jan 11 01:11:54 dose kernel: stuck on TLB IPI wait (CPU#1)
Jan 11 01:11:55 dose last message repeated 25 times

The problem occured once more since then but has never occured on
2.2.13.
Hardware did not change since 2.2.13 apart from the new 40 GB EIDE disc.

As far as I can tell I was not doing any fancy stuff when this occured
(no vmware, no inserting/removing of modules, no scsi drives mounted).
There might have been some -j2 compiling at that moment.

All mounted partitions (apart from a small ext2 /boot) at that time were
made with reiserfs 3.5.14 but I have not seen any problems with that.
There were no problems with reiserfs and 2.2.13, too.

Frank

-
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 : Sat Jan 15 2000 - 21:00:17 EST