2.1.109ac1 is out

Alan Cox (alan@lxorguk.ukuu.org.uk)
Fri, 17 Jul 98 03:56 BST


Despite the best efforts of a quantum bigfoot drive (yes I know everyone
told me they suck, now I know they were right) 2.1.109ac1 is now
available on ftp.linux.org.uk:/pub/linux/alan/

Its differs from Linus tree by the following (and some small bits of
"noise" I'll clean up in ac2)

Changes between 2.1.109 and 2.1.109ac1

o Asynchronous I/O via realtime signals
o Large file arrays per process (10K is just fine)
o 2Gig swap files
o Loop device supports sparse backing files
o Runtime settable mount array size
o request_module signal handling fixed
o Better menuconfig
o NFS handles "access denied" from clients right
o Appletalk handles PROBE and name changes correctly
o Network layer meets oversized buffer size requests as best it
can by limits, not by using default
o Much more extensive joystick support
o Tropez+ Wavefront sound driver
o Zoltrix radio driver
o Bill Hawes tty driver cleanup
o Parallel port doesnt panic the kernel if it unloads while you
are in /proc
o /proc inode counts are right
o Bitops ffs doesnt leak into userspace for libc5 users
o BTTV doesnt use old BIOS interface (still not up to date with
Ralph)
o Parport glitch fixed
o 3c59x patches - should stabilise the 3c59x/3c90x driver
o Amateur radio hfmodem compiles (syntax error in Linus tree)
o msp3400 and i2c tuners work if compiled in
o i2c tuner doesnt clash with other code on debug symbol
o i2c works compiled in
o Videodev supports KMOD
o Appletalk uses older field names for back compatibility
o Displays %cr0 %cr2 %cr3 on an Oops
o Bill Hawes "is root busy" checker
o MTRR builds uniproc with debugging spinlocks
o Network probe order cleaned up
o Apricot driver sort out
o Aims lab radio driver fixes
o SF16FMI radio driver fixes
o Old legacy radio interface killed (with its authors permission)
o SAA5249 doesnt reference PCI. Its not PCI it shouldnt have done
o More __initfuncs
o Tiny 3c509 cleanups
o Small spelling fixes to 3c501
o Warning fix to baycom_epp
o BPQ ether transmit path fix. RX path still busted
o MAD16 crashes fixed
o Turtle Beach Multisound update
o DECnet end node support
o Dump unused netbeui header
o X.25 fixes
o init knows root=loop (for initrd users)
o Econet, netlink, packet sockets don't cause NULL pointer crashes (NULL
not sock_no_* vectors)
o IP firewalling TOS change works again
o Sendmsg on UDP IPV6 fix
o Schd_api compiles

-
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.altern.org/andrebalsa/doc/lkml-faq.html