Changes from 1.3.45 to 1.3.46
These changes are archived on ftp.crynwr.com:pub/kchanges
and http://www.crynwr.com/kchanges.
I cannot answer questions about Linux -- I merely summarize the kernel
patches after reading them. I try to make them useful to everyone,
but kernel hackers should take them with a grain of salt and read the
patches themselves. The patch file is patch-1.3.46.gz
These are from mec@treflan.shout.net (Michael Elizabeth Chastain)
[he probably can't answer questions about Linux either :) -russ]
- arch/i386/boot/setup.S: backed out recent video mode changes.
- arch/i386/boot/setup.S: APM BIOS checking [Stephen Rothwell].
- arch/i386/kernel/process.c: APM support.
- drivers/block/cmd640.c: Fixes for vlb initialization code, enable read-ahead
for versions 'B' and 'C' of chip by default, some code cleanup.
- drivers/block/ide-tape.c: new driver in alpha stage [Gadi Oxman].
- drivers/block/ide.c: fix potential infinite timeout on resets; extend reset
poll into a general purpose polling scheme; add atapi tape drive support
from Gadi Oxman; simplify exit from _intr routines: no IDE_DO_REQUEST;
leave current rq on blkdev request list during I/O; generalized
ide_do_drive_cmd() for tape/cdrom driver use; fix nasty cdrom/tape bug
(ide_preempt was messed up); fix ide_xlate_1024() to work with/without
drive->id.
- drivers/block/ll_rw_blk.c: change handling of READA/WRITEA.
- drivers/cdrom/aztcd.c: started to modify azt_poll to prevent reading beyond
end of tracks [Werner Zimmerman].
- drivers/char/apm_bios.c: new driver [Stephen Rothwell].
- drivers/char/console.c: APM support.
- drivers/char/lp.c: new LPGETFLAGS ioctl.
- drivers/net/arcnet.c: "Unknown protocol ID" messages now also indicate the
station which sent the unrecognized packet, to aid in debugging network
confustion. Also, if anyone knows why Novell servers send packets with
protocol ID 0xEC, be sure to tell me. For now they're ignored.
Rearranged ARC_P_* handling a bit, so it makes slightly more sense.
We were clearing irq2dev_map too soon, and causing spurious "irq %d
for unknown device" messages. Moved all the set/clear irq2dev_map
operations to more intelligent places. 1.2.x kernels really didn't
work with 2.20 ALPHA. Maybe this will fix it. Fixed the setting of
set_multicast_list. Since we don't have multicast support, there's
no point in using this at all. [Avery Pennarun]
- drivers/net/de4x5.c: change multicast.
- drivers/net/slhc.c: fix memory clobber when rslots != tslots.
- drivers/pci/pci.c: add CIRRUS_7543, SPECIALIX_XIO, SPECIALIX_RIO.
- drivers/scsi/eata_dma.c: major changes to int handler.
- drivers/scsi/hosts.c: BusLogic must come before aha1542.c.
- drivers/scsi/scsi.c: major changes to scsi scanning.
- drivers/scsi/sg.c: change some EWOULDBLOCK to EAGAIN.
- fs/dquot.c: disk quotas [Marco van Wieringen].
- net/Changes:
- Faster buffer copy/clone [Linus]
- AX.25/NetROM fixes/changes [John Naylor]
- Further attempts to fix the IPX memory bug
- ARP fixes (Assorted)
- Driver fixes for multicast lists