Linus Torvalds: Linux 2.6.4-rc1

Kernel.org
Mirrors

Full Changelog

Ok, as usual, there was a lot of stuff for the -rc1, but as
seems to be more and more true it is mainly in the “periphery”.

This is a big patch, but the bulk of it is a long-overdue MIPS
update, and the (working) HFS/HFS+ filesystem update (total
rewrite), and ISDN updates. And some large s390 driver updates too,
for that matter.

So please keep bigger updates to yourself, and let’s calm this
down.

Linus

---

Summary of changes from v2.6.3 to
v2.6.4-rc1

:

• [TIGON3]: Add Apple tigon3 PCI device id

Adam Belay:

• PCI: remove unused defines in pci.h

Alan Stern:

• USB: Remove unneeded and error-provoking variable in UHCI
• USB: Even out distribution of UHCI interrupt transfers
• USB: ERRBUF_LEN compiling error when PAGE_SIZE=64KB on
  IA64
• USB: Simplify locking in UHCI
• USB: Mask “HC Halted” bit in the UHCI status register
• USB Storage: Handle excess 0-length data packets
• USB Storage: Treat STALL as failure for CB[I]
• USB Storage: Reduce auto-sensing for CB transport
• USB Storage: unusual_devs.h fixup
• USB: Improve UHCI root hub code: descriptor, OC bits, etc
• USB: More UHCI root hub code improvements
• USB: Another unusual_devs.h update
• USB: Repair unusual_devs.h entry
• USB: Use driver-model logging in the UHCI driver

Alexander Viro:

• [wan lapb] beginning of cleanups
• [wan lapb] switch to use net_device instead of custom
  token
• [wan lapb] Printks switched from %p lapb->token to %p
  lapb->dev
• [wan lapb] kill now-unused custom token container
• [wan hdlc] hdlc_open() switched to net_device
• [wan hdlc] hdlc_close() switched to net_device
• [wan hdlc] new port_to_dev() helper
• [wan hdlc] switch sca_xxx() to use net_device
• [wan hdlc] hdlc_set_carrier() switched to net_device
• [wan hdlc] hdlc->attach() switched to net_device
• [wan farsync] Eliminated a bunch of port->hdlc and
  hdlc_to_dev() uses
• [wan hdlc] hdlc->proto.*() switched to net_device
• [wan hdlc] hdlc_cisco: killed ->netdev, hdlc_to_name() and
  hdlc_to_dev() uses
• [wan hdlc] hdlc_fr: eliminated ->netdev, hdlc_to_dev() and
  hdlc_to_name() uses
• [wan hdlc_x25] eliminated hdlc_to_dev() and hdlc_to_name()
  uses
• [wan wanxl] eliminated hdlc_to_name() uses and a bunch of
  port->hdlc ones
• [wan hdlc] switch internal ioctl dispatch to net_device
• [wan pc300] more direct use of net_device
• [wan hdlc] new hdlc_stats() helper
• [wan dscc4] Uses of ->hdlc and hdlc_to_dev() encapsulated
  into dscc4_to_dev()
• [wan hdlc] switch register_hdlc_device() to take net_device
  arg
• [wan hdlc] new private struct pointer in hdlc_device, and
  helpers for it
• [wan hdlc] kill embedded struct in various drivers
• [wan pc300] use alloc_hdlcdev()/free_hdlcdev(). Leak fixed
• [wan farsync] embedded struct hdlc_device removal
• [wan dscc4] embedded struct removal
• [wan hdlc] removal hdlc_to_dev()
• [wan hdlc] kill embedding of struct net_device
• [wan hdlc_fr] Switched allocation of net_device to
  alloc_netdev()
• [wan hostess_sv11] sane net_device allocation
• [wan sbni] sane net_device allocation; plug a bunch of
  leaks
• [wan sealevel] Plugged a leak
• [netdrvr tun] Killed bogus ->init()
• [wireless airo] switched to sane allocation
• [netdrvr s390/netiucv] partially sanitized wrt allocation
• [netdrvr fec] switched to sane allocation. It still leaks on
  failure exits, though
• [wan sdla] Fixed leaks and double-free
• [netdrvr isa-skeleton] cleanups and fixes
• [all over] more kfree -> free_netdev
• [netdrvr acenic] Race and leak fixes
• [netdrvr 3c509] Leak fixed
• [netdrvr apne] resource leak fix
• [wireless orinoco] check alloc_etherdev for failure
• [netdrvr s390/lcs] Leak fix
• [wan] leak fixes in hostess_sv11, lapbether
• [netdrvr arm/am79c961] Fix for IO-before-request_region
  race
• [netdrvr saa9730] fix double-free
• [netdrvr arch/uml] leak fix
• [netdrvr dvb/dvb_net] fixes
• [netdrvr shaper] fix double-free
• [netdrvr s390/qeth] Alloc fixes
• Remove unused and invalid ‘struct ppp’ definition from
  if_pppvar.h
• [wan] apply hdlc cleanups to new driver pci200syn
• [PKT_SCHED]: Convert to {subsys,module}_initcall(), fix init
  failure bugs in sch_teql.c
• [IPV6]: Kill MODULE ifdeffing, common init for sysctls
• [IPV4/IPV6]: Convert tunnel drivers to unconditional
  module_init
• [DECNET]: Zap MODULE ifdefs
• removal of bogus CONFIG_BINFMT_ELF uses
• intermezzo ->permission() idiocy
• typo fix in intermezzo patch
• au1k leaks, allocation and free_netdev() fixes
• vlsi_ir leak, allocation and freeing fixes
• [WANROUTER]: Kill MODULE ifdefs
• remove init_{etherdev,netdev} and dev_alloc
• removal of ifdef MODULE from fs/openpromfs
• Clean up __cacheline_aligned

Alexandre Oliva:

• Improve code generation for x86 raid XOR functions

Amir Noam:

• [netdrvr bonding] Cannot remove and re-enslave the original
  active slave
• [netdrvr bonding] Releasing the original active slave causes
  mac address duplication
• [netdrvr bonding] Add support for slaves that use
  ethtool_ops
• [netdrvr bonding] fix build breakage
• [bonding 2.6] Fix compilation warning in bond_alb.c
• [bonding 2.6] Save parameters in a per-bond data structure
• [bonding 2.6] Use the per-bond value of the bond_mode
  parameter
• [bonding 2.6] Use the per-bond values of all remaining
  parameters

Andi Kleen:

• Intel x86-64 support merge
• Update i386 microcode driver for x86-64
• Enable Intel AGP on x86-64
• Allow P4 oprofile code for x86-64
• x86-64 merge for 2.6.3
• New machine check handler for x86-64
• Run 32bit compat ioctl handlers in BKL
• Support AGP bridge on Nvidia Nforce3 + cleanup

Andre:

• [ARM] Optimise ARM720T Thumb abort unwinding

Andreas Gruenbacher:

• CONFIG_REGPARM breaks non-asmlinkage syscalls

Andrew Morton:

• tulip printk warning fix
• ISDN udpate
• i4l: hisax deadlock fix
• Fix for PPP activ/passiv filter
• ia32 early printk
• early printk tweaks
• Remove BDEV_RAW and friends
• msg.h needs list.h
• ppc64: Fix prom.c warnings
• ppc64: fix saved_command_line/cmd_line lengths
• ppc64: fix debugger() warnings
• ppc64: iseries IRQ fix
• loop: remove the bio remapping capability
• remove useless highmem bounce from loop/cryptoloop
• loop: BIO handling fix
• loop.c doesn’t fail init gracefully
• loop: remove redundant initialisation
• ACPI PM timer
• kthread primitive
• Remove kstat cpu notifiers
• Minor workqueue.c cleanup
• Remove More Unneccessary CPU Notifiers
• Use CPU_UP_PREPARE properly
• Limit hashtable sizes
• add Pentium M and Pentium-4 M options
• gcc 2.95 supports -march=k6 (no need for check_gcc)
• AMD Elan is a different subarch
• Documentation: remove /etc/modules.conf refs
• add some more MODULE_ALIASes
• bonding alias revert and documentation fix
• NGROUPS 2.6.2rc2 + fixups
• Mark intermezzo as broken
• bd_set_size i_size handling
• snprintf fixes
• devfs: race fixes and cleanup
• Enable coredumps > 2GB
• MIPS: New 2.6 serial drivers
• #if versus #ifdef cleanup
• kNFSd: Fix possible scheduling_while_atomic in cache.c
• kNFSd: Allow sunrpc/svc cache init function to modify the
  “key”
• kNFSd: ip_map_init does a kmalloc which isn’t checked
• kNFSd: convert NFS /proc interfaces to seq_file
• kNFSd:fix build problems in nfs w/o proc_fs on 2.6.0-test5
• md: Print “deprecated” warning when START_ARRAY is used
• md: Split read and write end_request handlers
• md: Discard the cmd field from r1_bio structure
• md: Remove some un-needed fields from r1bio_s
• md: Avoid unnecessary bio allocation during raid1 resync
• md: Dynamically limit size of bio requests used for raid1
  resync
• md: Allow partitioning of MD devices
• dm: Export dm_vcalloc()
• dm: Move to_bytes() and to_sectors() into dm.h
• dm: Get rid of struct dm_deferred_io in dm.c
• dm: Maintain ordering when deferring bios
• dm: Tidy up the error path for alloc_dev()
• dm: Correct GFP flag in dm_table_create()
• dm: Zero size target sanity check
• dm: Remove redundant spin lock in dec_pending()
• dm: drop BIO_SEG_VALID bit
• 8259 timer ack fix
• Fix printk level on non fatal MCEs
• MCE fixes and cleanups
• Rename bitmap_snprintf() and cpumask_snprintf() to
  *_scnprintf()
• OSS: remove #ifdef’s for kernel 2.0
• remove kernel 2.2 #ifdef’s from {i,}stallion.h
• kbuild documentation fix
• adfs: remove a kernel 2.2 #ifdef
• defer panic for too many items in boot parameter line
• cpufreq_scale() fixes
• Minor cross-compile issues
• /proc thread visibility fixes
• drivers/char/vt possible race
• off_t in nfsd_commit needs to be loff_t
• skip offline CPUs in show_free_areas
• fix display of NBD in /proc/partitions
• cleanup patch that prepares for 4Kb stacks
• 3c59x: bring back the `enable_wol’ option
• Oprofile: fix nmi_timer_int detection
• oprofile: ARM infrastructure
• oprofile: add Pentium Mobile support
• remove max_anon limit
• Fix __release_region() race
• Documentation on how to debug modules
• Module headers cleanup
• add clock_was_set() to all architectures
• Fix race in epoll_ctl(EPOLL_CTL_MOD)
• slab: remove extraneous printk
• do_swap_page() return value fix
• ide-tape: remove obsolete onstream support
• Disable bootmem warning
• dm-crypt
• Fix make rpm when using RH9 or Fedora
• menuconfig: fix the check for ncurses-devel
• Inefficient TLB flush fix
• sf16fmr2 radio card driver
• Remove overenthusiastic BUG in smp_boot_cpus
• Codingstyle update
• smbfs: support the loop driver
• Fix sprintf modifiers in usr/gen_init_cpio.c for cygwin
• wireless/Kconfig enable/select complete replacement
• tuner driver fixes
• crc32.c copyright fix
• Add C99 initializers to arch/i386/pci/fixup.c
• mark ftape un-removable
• aio sysctl parms
• ramdisk cleanup
• slab: print slab name in kmem_cache_init()
• prevent ptrace from altering page permissions
• slab: hexdump for check_poison
• page_add_rmap(): remove meaningless test
• Add CONFIG for -mregparm=3
• Use -funit-at-a-time on ia32
• Add noinline attribute
• use noinline for rest_init()
• gcc-3.5: bonding
• fix access() POSIX compliance
• fix pfn_valid on ia32 discontigmem
• ia32: pfn_to_nid fix
• ia32: disallow NUMA on PC subarch
• config option for irqbalance
• print some x86 build options during oopses
• show_task() fix and cleanup
• show_task() is not SMP safe
• ppc64 compile fix
• v850 ptrace.c task_struct leak
• Fix the display of max-ever-used-stack in sysrq-T output
• smbfs: remove debug code
• x86_64 uniproc build fix
• ia64: fix sched.c compile warning
• MIPS mega-patch
• serial fixups
• more serial driver fixups
• Intel i830 AGP fix
• fix shmat
• dynamic pty allocation
• NBD rmmod oops fix
• m68k: offsets.h generation
• m68k: mm init warning
• m68k: Sun-3 console fix
• m68k: Sun-3 missing sbus_readl()
• m68k: Atari name clashes
• m68k: M68k MCA cleanup
• m68k: M68k uses drivers/Kconfig
• m68k: Amifb modedb bug
• m68k: M68k configuration
• m68k: arch/m68k/mm/Makefile cleanup
• m68k: M68k module loader
• m68k: M68k call trace output
• m68k: M68k cmpxchg
• m68k: M68k FPU emu broken link
• m68k: Mac IOP spelling
• m68k: Dummy dma mapping
• m68k: M68k core spelling
• add range checking to sys_sysctl()
• Kconfig help: dm-crypto && cryptoloop
• nbd: fix set_capacity call
• nbd: remove PARANOIA and other cleanups
• cleanup condsyscall for sysv ipc
• IPMI warning fixes
• mtrr: init section usage
• Fixes to CodingStyle
• Another x86-64 fix for problems from the recent merge
• ext3: fix scheduling-in-spinlock bug
• security oops fix
• From: “Randy.Dunlap” <<A
  HREF=”mailto:rddunlap@osdl.org”>rddunlap@osdl.org>
• janitor: media: use kernel min/max
• telephony: use kernel min/max
• Fix ISDN v.110
• Fix fs/partitions/efi.c printk warnings
• powernow-k8 frequency handling fix
• Require GNU Make version 3.79.1 or later
• runtime PM deadlock fix
• Fix do_shmat() for CONFIG_SYSVIPC=n
• ppc64: remove dump_regs
• ppc64: cleanup lmb code
• ppc64: Potentially avoid an atomic operation in switch_mm
• ppc64: Remove duplicate pcibios_scan_all_fns definition
• ppc64: Fix for ppc64 SMT enablement bug provided by Jimi
  Xenidis and Michael Day
• ppc64: add rtas slot-error-detail information
• ppc64: add log_rtas_error()
• Allow CROSS32_COMPILE to be set via environment variable
• ppc64: Add a ppc64 archhelp
• ppc64: print useful flags in oops, like x86
• ppc64: Add DEBUG_STACK_USAGE
• ppc64: Add -funit-at-a-time
• Add 970FX entry into the cputable
• ppc64: Fix for valid nvram rtas tokens
• fix naming collision with asm-ppc64/vio.h
• ppc64: fix warning and compile error without CONFIG_SMP
• ppc64 cpu spinup fixes
• ppc64: remove useless smp_message_pass args
• ppc64: This cleans up the rtasd logic, and also makes it
  hotplug CPU safe
• ppc64: Add stack overflow debugging
• ppc64: remove get_users in backtrace code
• Add cpus and NUMA memory nodes to sysfs. Also add cpu physical
  id
• ppc64: Make a number of segment table functions static
• ppc64: Clean up per cpu usage in segment table code
• ppc64: PER_CPU irq optimisations
• ppc64: don’t link some non iSeries stuff
• ppc64: Fix __get_SP()
• ppc64: set err to -ENODEV when a new node doesn’t have
  “interrupt” property
• fix for NUMA kernel on non NUMA machine
• trivial oops formatting cleanups
• ppc64: restore cpu names
• ppc64: uniprocessor compile fixes
• ppc64: fix cmd_line bugs
• m68k: Sun-3 LANCE Ethernet
• m68k: Atari Pamsnet warning
• m68k: Amiga A2065 Ethernet new driver model
• m68k: Amiga Ariadne Ethernet new driver model
• m68k: Amiga Hydra Ethernet new driver model
• ppc64: archhelp fix
• Fix make xconfig on /lib64 systems
• Report NGROUPS_MAX via a sysctl (read-only)
• Make insert_resource work for alder IOAPIC resources
• add the Intel Alder IO-APIC PCI device to quirks
• x86: remove THREAD_SIZE assumption cleanups
• cosmetic printk fix
• remove unneeded check from sys_sysctl()
• clarify MSI requirements in Kconfig
• pty changes require procps 3.2
• jffs2: Don’t jump between contexts
• fix the build with CONFIG_UNIX98_PTYS=n
• Eicon isdn driver compile __devexit compile fix
• Change ENOTSUPP to EOPNOTSUPP
• swsusp/s3: Assembly interactions need asmlinkage
• acpi/utils.c warning fix
• drivers/acpi/sleep/proc.c warnings
• add syscalls.h
• asmlinkage fixes
• ppc64: fix de4x5 build
• m68k: Amiga Zorro8390 Ethernet new driver model
• HFS rewrite
• HFS+ support
• superblock fixes
• fix module reference counting in zoran driver
• s390: general update
• s390: common i/o layer
• s390: console driver
• s390: compat_timer_settime
• s390: CTC network driver
• s390: LCS network driver
• s390: IUCV network driver
• s390: DASD device driver
• s390: virtual timer interface
• s390: z/VM monitor stream
• s390: collaborative memory management
• s390: channel measurement block interface
• s390: zfcp host adapter
• s390 syscalls.h update
• s390: DCSS block device driver
• dvb: Update subsystem docs
• dvb: Update saa7146 driver core
• dvb: Minor Skystar2 updates
• dvb: core update
• dvb: Misc frontend updates
• dvb: stv0299 DVB frontend update
• dvb: tda1004x DVB frontend update
• dvb: av7110 DVB driver update
• dvb: TTUSB-Budget DVB driver update
• n_tty.c cleanup
• M68k Macintosh driver config
• request_firmware(): misc fixes
• request_firmware(): more misc fixes
• request_firmware(): add status bitmap
• request_firmware(): fix firmware_priv leak
• request_firmware(): race fixes
• request_firmware(): refactor fw_setup_class_device()
• request_firmware(): fix attribute removal
• early printk documentation fix
• radeon config fix
• Remove unused tty CALLOUT defines
• don’t use floating point in tdfxfb
• mtd locking fix
• C99 patch for fs/afs/inode.c
• Kill bogus __KERNEL_SYSCALLS usage
• C99 initiailzers for drivers/isdn/i4l/isdn_common.c
• C99 initializers for drivers/net/wireless/airo.c
• C99 initializers for drivers/net/wan/wanxl.c
• C99 initializers for drivers/net/wan/pci200syn.c
• C99 initializers for drivers/net/irda/irda-usb.c
• C99 initializers for drivers/media/common/saa7146_video.c
• C99 initializer for drivers/media/dv/frontend/stv0229.c
• C99 initializers for
  drivers/media/dvb/frontends/alps_tdlb7.c
• C99 initializers for drivers/media/dvb/frontends/sp887x.c
• C99 initializer for driver/media/dvb/ttpci/budget-av.c
• V4L: Add new driver for Teletext decoder SAA5246A from
  Philips
• kbuild: add defconfig targets to make help
• wanmain.c build fix
• swsusp locking fix
• PPC64 iSeries virtual disk driver
• Add kallsyms_lookupname()
• ppc64: use kallsyms_lookup_name() in xmon
• ppc64: move sg_dma_{len,address} macros
• ppc64: Fix a sleeping with spinlock bug in ioremap
• knfsd: NGROUPS fixes
• nfsd: don’t modify group_info structures
• kNFSd: Add minimal server-side support for rpcsec_gss
• kNFSd: gss api changes for integrity checking
• kNFSd: IDmap support for the NFSv4 server
• kNFSd: Nfsdv4 pointer cleanup
• kNFSd: NFSv4 locking state fix
• kNFSd: v4 exclusive open fix
• kNFSd: Use higher-resolution time for the changeinfo, instead
  of using time and filesize
• kNFSd: When looking for a shareowner in the nfsd open, make
  sure we don’t get a lockowner instead
• kNFSd: NFSdV4 fixes for replaying open requests
• kNFSd: Use only the uid when deciding whether a setclientid is
  being done with the “same principal”
• kNFSd: When looking for a shareowner in the nfsd open, make
  sure we don’t get a lockowner instead
• kNFSd: readdir error code fix
• kNFSd: correctly tests and sets nfserr_nofilehandle for current
  and save fh
• kNFSd: Fix for lookup-parent at pseudo root
• kNFSd: Correct error returns
• kNFSd: fixes an xdr error by removing the verifier from error
  return
• kNFSd: correct symlink related error returns
• kNFSd: check lock length, return appropriate error
• kNFSd: correct rename error returns
• kNFSd: unlock-on-close fix
• kNFSd: Remove a comment that is no longer accurate
• kNFSd: move fh_dup2 and fix it
• kNFSd: Implement the nfsv4 RELEASE_LOCKOWNER operation
• kNFSd: add OP_ILLEGAL, and fix processing of compounds with out
  of bounds op numbers
• kNFSd: fix an error return for OP_CREATE
• kNFSd: Add a check in OP_LOCK for new lockowners to ensure that
  the open stateid is
• kNFSd: Corrects an error return for OP_OPEN_CONFIRM
• kNFSd: Enforce open_downgrade requirement
• kNFSd: Fix an out-of-spec readlink error return
• kNFSd: Fix an out-of-spec error in nfsd4_remove
• kNFSd: Miscellaneous fixes to stateid-based replay
• kNFSd: Fix out-of-spec error return in attribute decoding
• kNFSd: Make the calculation in the first READ_BUF easier to
  understand
• kNFSd: make sure sunrpc init routines called before gss init
  routines
• kNFSd: return more than one page of directory entries
• Add a MODULE_VERSION macro
• rename other MODULE_VERSION users
• fbdev cursor fix

Andrew Vasquez:

• qla2xxx — Properly schedule mailbox command timeouts
• qla2xxx — FCP_RSP IU check during command completion

Andries E. Brouwer:

• USB: add comments to sddr09.c
• tty utf8 mode

Anton Blanchard:

• ppc64: fix kernel access of user pages
• ppc64: TLB flush rework
• ppc64: fix pci hotplug compile error
• ppc64: remove CONFIG_VETH
• ppc64: defconfig update

Art Haas:

• [NETFILTER]: C99 initializers in ip_conntrack_standalone.c
• [RXRPC]: C99 initialiers for net/rxrpc/connection.c

Bartlomiej Zolnierkiewicz:

• move CONFIG_HOTPLUG to init/Kconfig
• remove dead kernel parameters
• keep documentation of kernel parameters in one place only
• kill useless IDE_SUBDRIVER_VERSION
• kill default_shutdown() and ide_drive_t->shutdown
• kill default_flushcache() and ide_drive_t->flushcache
• remove bogus comment and code from ide_unregister_driver()
• remove dead/unfinished taskfile version of ide_cmd_ioctl()
• remove unused ide_end_taskfile()
• fix /proc/ide/ for IDE PCI modules
• fix /proc/ide/ for IDE PCI modules
• siimage.c: limit requests to 15kB only for Seagate SATA
  drives
• add UDMA6 support to ALi PCI IDE driver
• fix ide_system_bus_speed() causing “Badness in
  pci_find_subsys…”
• ide-io.c: CONFIG_LBD fix
• explicitly define PRD_ENTRIES to 256
• remove redundant ide_setup_pci_device{s}() declarations
• ide-proc.c: remove unused MIN() macro
• ide-taskfile.c: remove debugging placeholders

Ben Collins:

• IEEE1394(r1118): Addition of new config-rom processing
  code
• IEEE1394(r1126): Small cleanup based on patch from Isaac
  Claymore
• IEEE1394(r1128): Implement bus reset handling for hosts (based
  on patch from Oracle)
• IEEE1394(r1129): Initial support for reusing node entries over
  plug/unplug cycles
• IEEE1394(r1130): Conversion of some list_for_each() usages to
  list_for_each_entry()
• IEEE1394/OHCI(r1131): Suspend/resume isn’t dependent on
  CONFIG_PM
• IEEE1394: Sync revisions
• IEEE1394(r1134): Fix some typos introduced over the last few
  changes
• IEEE1394(r1135): Initial ieee1394_host_class
  implementation
• IEEE1394(r1136): Implement node class
• IEEE1394(r1137): Check to make sure we have a hostinfo in
  sbp2_host_reset
• IEEE1394(r1138): Cleanup nodemgr probe, and fix a bug with the
  driver update
• IEEE1394/SBP2(r1139): Some cleanups, and a better, more unique
  id for ieee1394_id attribute
• IEEE1394(r1140): Add a bus rescan bus_attr file, and an
  ignore_driver attr for ud’s
• IEEE1394(r1141): Add an “ignore_drivers” global default
• IEEE1394(r1142): Use a kernel thread to rescan devices so we
  don’t block the writer
• IEEE1394(r1143): Fix FCP requests, broken by my change for
  list_for_each_entry()
• IEEE1394/SBP2(r1144): Convert sbp2 to do one scsi_host per
  unit-directory
• IEEE1394: Revision sync
• IEEE1394 is no longer experimental, but eth1394 is
• IEEE1394(r1146): Make the probe callback return an error if
  problems, and unbind drivers on failure
• IEEE1394/ETH1394(r1147): Make sure to set update_config_rom on
  add host
• IEEE1394/SBP2(r1148): Fix a few bugs, and add set
  blk_queue_dma_alignment to 512
• IEEE1394(r1149): Convert some hardcoded values to constants.
  Add pbook suspend/resume handlers
• IEEE1394/Video1394(r1150): Initialize d->link list so that
  failures cleanup properly
• IEEE1394/Video1394(r1151): Fix typo in last commit
• IEEE1394(r1154): Cleanup dma routines, and use vmalloc for
  sglist allocation
• IEEE1394(r1155): Get rid of another user of
  round_up_to_page()
• eth1394(r1156): Fix oddities when MacOSX is IRM. Also MacOSX
  sets an invalid sspd in the ARP reply
• IEEE1394(r1157): Create a kernel thread to handle all
  run_packet_complete() calls
• SBP2(r1158): We don’t need a remove_host callback in sbp2
• IEEE1394(r1159): Merge run_packet_complete into kernel
  thread
• ohci1394(r1160): Disable cross_bound() check, and add code to
  try to enable connected ports
• IEEE1394(r1161): Major cleanup and addition of a ud-class, to
  make things even cleaner
• SPARC64: Fix debug spinlocks to not trash random memory with
  > 4 cpus’s (or sparse cpu’s)
• SPARC64: Use hard_smp_processor_id() for
  init_irqwork_curcpu()
• IEEE1394(r1162): Check return value for errors from
  hpsb_register_protocol
• IEEE1394(r1163): Fixup nodemgr_{suspend,resume}_ne to use the
  ud class list
• [SPARC64]: Use prom_printf in sun_do_break() instead of printk.
  Avoids lockup
• [SPARC64]: Fix compile warning from RW_LOCK_UNLOCKED with
  spinlock debug enabled

Benjamin Herrenschmidt:

• Be careful about memory ordering in sungem driver
• Fix a DMA underrun problem with pmac IDE
• ppc32: rework l2 cache code
• ppc32: Export cpu_possible_map
• Fix use of sector_t in swim3 driver
• Remove use of “current” identifier in via-pmu
• Disable debugging verbosity in macio_asic.c
• ppc64 oops on /proc/cpuinfo
• ppc64: Fix warning on pmac build
• radeonfb: small cleanup of common register init
• ppc64: physical RAM accounting fix
• ppc64: Fix /dev/mem idea of what is memory
• ppc64: iommu rewrite
• ppc64: Fix drivers/ide when using an IOMMU
• ppc64:Implement support for Apple DART IOMMU (PowerMac G5)

Bjorn Helgaas:

• [SERIAL] Discover ACPI serial ports before plug-in ports
• [SERIAL] Fix /proc serial info for MMIO ports

Brian Childs:

• multicast broken on x86_64

Brian King:

• SCSI: Make retries obey host_self_blocked flag

Carlos Puchol:

• initial support for transmeta’s efficeon processors

Chas Williams:

• [ATM]: horizon: make reset function not __init (from Randy
  Dunlap <<A
  HREF=”mailto:rddunlap@osdl.org”>rddunlap@osdl.org>)
• [ATM]: use clip_tbl instead of clp_tbl_hook (from Francois
  Romieu <<A
  HREF=”mailto:romieu@fr.zoreil.com”>romieu@fr.zoreil.com>)
• [ATM]: [lec] put back pressure on network stack

Christoph Hellwig:

• [SUNGEM]: Kill unused variable on ppc
• remove flush_cache_all() from qla1280
• Remove CONFIG_SCSI_DC390T_NOGENSUPP
• fix up ini9100 interrupt handling
• fix up NCR5380 private data
• move remaining definitions from drivers/scsi/scsi.h to
  include/scsi
• remove useless MOD_{INC,DEC}_USE_COUNT in lasi_82596.c
• remove useless MOD_{INC,DEC}_USE_COUNT in sun3lance
• remove useless MOD_{INC,DEC}_USE_COUNT in sb1250-mac
• remove useless MOD_{INC,DEC}_USE_COUNT in meth
• remove useless MOD_{INC,DEC}_USE_COUNT in wanpipe

Dave Jones:

• [CPUFREQ] Update URL
• [CPUFREQ] bump copyrights
• [CPUFREQ] Add extra __init markers to longhaul driver
• [CPUFREQ] Add extra __init markers to longrun driver
• [CPUFREQ] Don’t set up longhaul voltage scaling too early
• [CPUFREQ] Extra sanity checks in longhaul
• [CPUFREQ] Don’t guess FSB on Nehemiah
• [CPUFREQ] Export powernow-k7 scaling frequencies
• [CPUFREQ] scaling_available_frequencies work for remaining x86
  drivers
• [CPUFREQ] Fix ARM cpufreq governor selection From: Russell King
  <<A
  HREF=”mailto:rmk@arm.linux.org.uk”>rmk@arm.linux.org.uk>
• Fix for a really stupid off by 1 bug

Dave Kleikamp:

• JFS: get_UCSname does not need nls_tab argument
• JFS: Don’t do filename translation by default

David Brownell:

• USB: EHCI updates (mostly periodic schedule scanning)
• USB: usbcore, scatterlist cleanups
• USB: usbcore, hub driver enables TT-per-port mode
• USB: usbcore, avoid RNDIS configs
• USB: usbtest, two more protocol cases
• USB: ehci-hcd, fullspeed iso data structures (1/3)
• USB: ehci-hcd, fullspeed iso data structures (2/3)
• USB: ehci-hcd, scheduler handles TT collisions (3/3)

David Milburn:

• [libata sata_promise] Fix DIMM initialization on PCI-X bus

David Mosberger:

• ia64: Back-port from libunwind: fix off-by-one error in
  kernel-unwinder
• ia64: Fix bug in
  ia64_get_scratch_nat_bits()/ia64_put_scratch_nat_bits()
• ia64: Update defconfig
• ia64: Update defconfig

David S. Miller:

• [IPV6]: Kill 64-bit warnings in ndisc.c
• [IRDA]: Forgot to add stir4200.c in previous commit
• [TG3]: Two more PHY bug workaround, plus fix DMA test on
  big-endian
• [TG3]: Fix early chip programming in
  tg3_setup_copper_phy()
• [TG3]: Bump driver version and reldate
• [APPLETALK]: Use ‘%Z’ for size_t
• [BLUETOOTH]: Use min_t to avoid warning in rfcomm sock.c
• [DECNET]: Make second arg to dn_alloc_send_skb a size_t
  pointer
• [SELINUX]: Forgot to add these in previous commit
• qla2xxx: remove flush_cache_all
• [SPARC64]: Update defconfig
• [I2C]: Fix resource address typing
• [I2C]: Use correct port address types in i2c-velleman.c
• [I2C]: Fix resource address typing in i2c-voodoo3.c
• [FREEVXFS]: Fix u64 printk warnings on some 64-bit
  platforms
• [I2C]: If comparing against ULONG_MAX, use ulong type, in
  lm85.c
• [NFSD]: Fix u64 printk warnings on some 64-bit platforms
• [SMBFS]: Use ‘%z’ printf format for size_t types
• [MEDIA]: Print out pointers correctly in dst.c
• [MEDIA]: Use ‘%z’ printf format for size_t
• [SKFDDI]: Use unsigned long for resource base/size
• [SUNDANCE]: Fix casting so u64 printk does not warn on some
  64-bit platforms
• [MEDIA]: Use ‘%z’ printf format for size_t/ptrdiff_t types in
  w9968cf.c
• [I2C]: Use correct port address typing in i2c-elv.c
• [SPARC64]: Do similar macro casting for {in,out}{b,w,l}() as we
  do for {read,write}{b,w,l,q}()
• [SPARC64]: Update defconfig
• [NETFILTER]: Include net/ipv6.h in ip6_tables.c
• [IPV6]: Make ipv6_skip_exthdrs take const skb arg
• [PKT_SCHED]: Missing linux/init.h includes in
  sch_{arm,dsmark}.c
• Synchronize sungem RX complation path
• [IPV4]: Do not return -EAGAIN on blocking UDP socket, noticed
  by Olaf Kirch
• [IPV6]: Kill unloadable noise in af_inet6.c
• [IPV6]: Fix typo in Al’s module_init changes
• [TIGON3]: Comment out card RAM validation in tg3_test_dma() for
  now
• [TIGON3]: Bump version and reldate
• [NET]: Set default socket rmem/wmem values more sanely and
  consistently
• [IPV6]: UDPv6 needs recvmsg csum error path fix too, thanks
  Olaf
• [SPARC64]: Update defconfig
• [SPARC]: Move ptrace_signal_deliver() implementation out of
  header file
• [SUNRPC]: Use ‘%z’ for size_t printk in svcauth_gss.c

David Stevens:

• [IPV6]: Except MLD packets from source filtering
• [IPV6]: Add sysctl to force MLD version
• [IGMP/MLD]: Validate filter size against optlen
• [IGMP/MLD]: Check for numsrc overflow, plus temp buffer
  tweaks

David Woodhouse:

• [IPV4]: Make ip_auto_config a late initcall

Deepak Saxena:

• USB: Fix USB host code to use generic DMA API
• [ARM PATCH] 1761/1: Remove SA1111_PCI_FAKEDEV (take 2)

Dely Sy:

• PCI Hotplug: Add SHPC and PCI Express hot-plug drivers

Domen Puncer:

• USB: some stv680 fixes

Don Fry:

• [NET]: Fix ethtool oops if device support get but not set
  ringparam
• 2.6.3 pcnet32.c bus master arbitration failure fix
• 2.6.3 pcnet32.c SLAB_DEBUG length error fix
• 2.6.3 pcnet32.c transmit hang fix
• 2.6.3 pcnet32.c wrong vendor ID fix
• 2.6.3 pcnet32.c convert to use netif_msg_*
• 2.6.3 pcnet32.c change to use ethtool_ops
• pcnet32.c handle failures in open
• pcnet32.c non-mii errors with ethtool
• pcnet32.c add PCI hot remove support
• pcnet32.c adds loopback test
• whitespace only change to pcnet32.c
• pcnet32.c fix compile error
• pcnet32 non-mii link state fix

François Romieu:

• [IRDA]: Fix error return status in stir4200 driver
• [IRDA]: In stir4200 driver, defer netif_start_queue() until
  device opening succeeds
• 2.6.3 – drivers/net/sis190.c – Tx desc overflow

Geert Uytterhoeven:

• NCR53C9x slave_{alloc,destroy}()
• Sun-3x ESP SCSI clean up

Grant Grundler:

• [TG3]: Reset GRC, if necessary, before DMA test
• [PKTGEN]: Fix unintentional unaligned access
• [TG3]: Abstract out mailbox workarounds into
  tw32_{rx,tw}_mbox()
• [TG3]: Define MBOX_WRITE_REORDER flag to zero on non-x86

Greg Kroah-Hartman:

• PCI: fix pci quirk for P4B533-V, fixes bug 1720
• USB: fix up compile errors in uhci driver
• USB storage: sync up with some missing unusual_devs entries
  that were in my tree
• PCI Hotplug: fix build warnings on 64 bit processors

Greg Ungerer:

• fix memory leaks in binfmt_flat loader
• allow configuration for shared flat binary support
• add m68k elf relocation types to elf.h
• fixes to ColdFire/5407 startup code
• m68knommu: remove non-existant option from defconfig
• m68knommu: fix interrupt handler return types to be
  irqretur_t
• m68knommu: use KERN_DEBUG in debug printk()’s

Harald Welte:

• [NETFILTER]: Resync with 2.4.x
• [NETFILTER]: Remove unused structure member in NAT, from
  Patrick McHardy

Herbert Xu:

• [IPSEC]: Move hardware headers for decaped packets

Hideaki Yoshifuji:

• [NET]: Kill bogus kmem cache alignment in
  neigh_table_init()
• [IPV6]: Clean up ndisc printks
• [IPV6]: Spelling corrections, and remove some XXX’s
• [IPV6]: Kill remaining in6_u.u6_addrX uses
• [IPV6]: Use udpv6_queue_rcv_skb for multicast delivery

Holger Schurig:

• [ARM PATCH] 1670/1: PXA serial driver

Hollis Blanchard:

• ppc64: virtual IO bus updates

Ian Abbott:

• USB: ftdi_sio new PIDs and name fix for sysfs

Jakub Bogusz:

• 2.6.3 – fix for undefined mdelay in 3c505
• switch alpha to use drivers/Kconfig

James Bottomley:

• MPT Fusion driver 3.00.03 update
• SCSI: 53c700: reduce default tag depth to 4
• fix IRQBALANCE Kconfig dependencies
• Undo SCSI 8-byte alignment relaxation

James Morris:

• [SELINUX]: Event notifications via netlink

Jan Kasprzak:

• [NET]: Do not send negative 2nd arg to skb_put()

Jean Tourrilhes:

• [IRDA]: Add stir4200 driver
• [IRDA]: Rename dongle entry points for consistency

Jeff Garzik:

• [netdrvr e100] include linux/moduleparam.h
• Eliminate ancient and unused
  include/linux/{if_pppvar,ppp}.h
• [libata] catch, and ack, spurious DMA interrupts
• Add Documentation/networking/netif-msg.txt, describing the
  per-network-interface message logging standards for net
  drivers.
• Delete tms380tr firmware, no longer needed now that driver uses
  request_firmware()
• Improvements to the bk-make-sum BitKeeper summary/submission
  script
• default 8139too to PIO
• [libata] Guarantee that S/G entries do not straddle IDE DMA 64k
  boundary
• [libata sata_sil] fix 4-port support on SII 3114
• [libata] Explicitly set max_phys_segments to 128 (current blk
  default), in case MAX_PHYS_SEGMENTS ever changes.
• [libata] bump versions
• Fix PCI MSI build when kirqd is disabled
• [libata] Much better s/g table fill routine
• [libata] limit S/G table size to 128 entries
• [libata ata_piix] add support for ICH6
• [libata ata_piix] Update PCI quirk with new Intel SATA
  devices
• Add Intel ICH6 PCI ids to pci_ids.h
• Add Intel ICH6 irq router
• Add Intel PCI ids to old-OSS driver i810_audio
• Add Intel PCI ids to IDE (PATA) driver
• Fix broken PIIX build
• [IDE] Create DECLARE_PIIX_DEV declarator, to eliminate a large
  amount

Jens Axboe:

• fix SCSI non-sector bio backed IO

Jeremy Higdon:

• ia64: Fix 64 bit DMA mapping problem with PCI cards on SN2

Jeroen Vreeken:

• 6pack reinit bug

Jochen Friedrich:

• tms380tr patch 1/3 (bug fix)
• tms380tr patch 2/3 (queue fix)
• tms380tr patch 3/3 (get firmware out of kernel)

John Rose:

• PCI Hotplug : add DLPAR driver for PPC64 PCI Hotplug slots

John Stultz:

• ia64: add support for time-interpolation via IBM EXA Cyclone
  timer

Joseph Fannin:

• [netdrvr sis900] fix multicast bug

Julian Anastasov:

• [IPV4]: Add configurable restriction of local IP announcements
  in ARP requests
• [IPV4]: Add sophisticated ARP reply control via arp_ignore
  sysctl

Kai Makisara:

• Sysfs class support for SCSI tapes

Kazunori Miyazawa:

• [IPV6]: Unify ipv6 output routine

Keith M. Wesolowski:

• [SPARC32] Uninline atomic_t functions to save space
• [SPARC32]: Do similar macro casting for {in,out}{b,w,l}() as we
  do for {read ,write}{b,w,l,q}().
• [SPARC32]: Nuke a.out build cruft
• [SPARC32]: Remove stale SMP irq implementation
• [SPARC32]: Remove use of PF_USEDFPU
• [SPARC32]: Remove cli/sti from all arch code

Len Brown:

• [ACPI] revert previous AML param patch for ACPICA update
• [ACPI] ACPICA 20040211 udpate from Bob Moore

Linda Xie:

• PCI Hotplug: Add PPC64 PCI Hotplug Driver for RPA

Linus Torvalds:

• Fix up the microcode update on regular 32-bit x86. Our wrmsr()
  is a bit unforgiving and really doesn’t like 64-bit values.
• Include the header file for DMA mapping
• Fix silly thinko in sungem network driver
• Don’t allow memory “lines” argument in “xor_p5_mmx_5()”
• ppc64: fix non-iSeries build
• Add d_type information to legacy readdir system call
• Linux 2.6.4-rc1

long (tlnguyen):

• PCI: add copyright for files msi.c and msi.h

Luiz Capitulino:

• [PPPOE]: Handle disabled PROC_FS properly
• [PPPOE]: Kill unneeded ifdef/endif

MAEDA Naoaki:

• ia64: fix possible memory leak in PCI alloc_resource()

Manfred Spraul:

• rename shmat to make it clear it isn’t a system call
  entrypoint

Marcel Holtmann:

• [Bluetooth] Cleanup drivers Kconfig file
• [Bluetooth] Copy all L2CAP signal frames to the raw
  sockets
• [Bluetooth] Initial sysfs and device class support
• [Bluetooth] Dynamic allocation of HCI device
• [Bluetooth] Use C99 initializer for HCI USB driver
• [Bluetooth] Initialize interval of ISOC URB’s

Marcelo Tosatti:

• cyclades async driver update

Mark A. Greer:

• PCI: Changing ‘GALILEO’ to ‘MARVELL’

Mark Haverkamp:

• add card types to aacraid driver
• aacraid reset handler

Mark Hindley:

• [SERIAL] fix 8250_pnp resource allocation

Martin Hicks:

• ia64: SN2 header file cleanups
• ia64: clean up SN2 setup.c

Martin J. Bligh:

• [NET]: Ditch TSO in loopback driver, it’s had enough
  testing

Martine Silbermann:

• PCI: update MSI Documentation

Matthew Dharm:

• USB Storage: Save the SCSI residue value when auto-sensing
• USB Storage: Reduce unsolicited auto-sense
• USB Storage: Fix small endian-ness bug

Matthew Wilcox:

• [TG32]: Use pci_get_slot() to find 5704 peers, to handle
  domains properly
• PCI: Fix pci_bus_find_capability()
• ia64: SAL version is not zero-padded
• ia64: improve gate generation
• ia64: Add “install” make target

Matthias Andree:

• [NET]: Export sysctl_optmem_max to modules

Michal Ludvig:

• [XFRM_USER]: Fix SKB sizing in xfrm_send_policy_notify()
• [XFRM_USER]: In xfrm_send_{acquire,policy_notify}(), use
  {RTA,NLMSG}_SPACE()

Miguel Freitas:

• [NET_SCHED]: Fix slot leakage in SFQ scheduler

Mike Phillips:

• 3c359_microcode.h clean up – 2.6.3

Mirko Lindner:

• sk98lin: Added Support for Belkin adapter
• [kernel 2.6] sk98lin: Insert revision version and date

Nathan Scott:

• blkdev_open/bd_claim vs BLKBSZSET

Pat Gefre:

• ia64: add “platform_data” to struct pci_controller, update SN2
  accordingly
• ia64: on SN2, use the pda to count interrupts
• ia64: on SN2, skip init_platform_hubinfo() if on the
  simulator
• ia64: cleanup SN2 pci_bus_cvlink.c
• ia64: add Altix hotplug support

Patrick Mansfield:

• have CONFIG_SCSI_PROC_FS depend on CONFIG_PROC_FS

Patrick McHardy:

• [PKTSCHED]: Use queue limit of 1 when tx_queue_len is zero
• [NETFILTER]: Fix amanda helpers, forward port from 2.4.x
  version

Paul Mackerras:

• Clean up IRQ mapping code

Paulo Marques:

• USB: fix usblp.c

Peter Teichmann:

• [ARM PATCH 1719/1] Acornfb update

Petri Koistinen:

• [IRDA]: Fix URLs in Kconfig
• [HAMRADIO]: Fix URLs in Kconfig
• [WAN]: Kconfig clean-up and update URL links

Randy Dunlap:

• depca: remove double semi-colon
• tr/3c359: handle kmalloc failures during init
• ne: eliminate unused var. warning
• ibmtr: use kernel min/max
• strip: remove warnings when !PROC_FS
• strip: use kernel min/max

Robert Picco:

• ia64: add support for NUMA machines with CPU-only (memory-less)
  nodes

Russell King:

• [ARM] Make Acorn I2C build again
• [ARM] Fix ambakmi to use amba_request_regions() and the correct
  IRQ
• ICS IDE is not a PCI IDE interface
• [ARM] Fix Acorn VIDC sound driver
• [ARM] Add safe sa1111 IO handling
• [ARM] Update SA1111 Neponset code to use safe GPIO
  functionality
• [ARM] Dynamically allocate SA1100 PCMCIA sockets
• [ARM] Fix late abort handler for Thumb code
• [ARM] Remove non-existent Kconfig source statement
• [ARM] Add ARM architecture version 6 support
• [ARM] Cleanup MODULE_* macros
• [ARM] Update sa1111-pcibuf for dmapool changes
• [ARM] Add SA11x0 sched_clock() implementation
• [ARM] Remove obsolete sysctl PM interface
• [ARM] Update mach-types definitions file
• [ARM] Improve bad IRQ reporting
• [ARM] Optimise readsl
• [ARM] Allow decompressor to use “cache type” register
• [ARM] Add resources and platform devices for SA11x0 serial
  ports
• [ARM] amba_{request,release}_regions
• [ARM] Make free_memmap() use PFNs instead of physical
  addresses

Rusty Russell:

• Always put cache aligned code in own section, even for
  modules

Santiago Leon:

• Add IBM PowerPC Virtual Ethernet Driver

Scott Feldman:

• [netdrvr] e100 version 3 (complete rewrite)
• [e100] ICH6 IDs + ia64 memcpy fix + module_param
• [netdrvr e100] netpoll + fixes to speed/duplex forced
  settings
• [netdrvr e100] fix slab corruption
• [netdrvr e100] copyright + trailing blanks + misc
• e1000: flow control
• e1000: disable TSO for now
• e1000: disable CSA fix for 82547
• e1000: delay may be too small
• e1000: collision retry count too high
• e1000: handle register_netdev failure
• [netdrvr e100] Response to Jeff’s review plus some minor
  fixes

Sebastian Henschel:

• [NET]: Mention tuxmobil.org in drivers/net/Kconfig
• [ARM] URL change for linux-on-laptops

Shmulik Hen:

• [netdrvr bonding] trivial – Update comment blocks and version
  field
• Add support for HW accel. slaves
• Add VLAN support in TLB mode
• Add VLAN support in ALB mode

Simon Barber:

• [NET]: Capture skb->protocol after invoking bridge

Simon Horman:

• [JHASH]: Make key arg const in jhash()

Sridhar Samudrala:

• [SCTP] Revert back to use kmalloc() for ssnmap allocs of sizes
  <
  128K
• [SCTP] Force enable Crypto options that are needed by SCTP
  config
• [SCTP] Fix incorrect INIT process termination with
  sinit_max_init_timeo

Stefan Rompf:

• Re: Patch: netif_carrier_on()/off() for xircom_tulip_cb
• netif_carrier_on()/off() for xircom_tulip_cb

Stephen Hemminger:

• bugfixes for dgrs.c
• hp100 pci probe problem
• [WAN]: Fix single_open confusion in wandev_show
• Re: IA32 (2.6.3 – 2004-02-18.22.30) – 4 New warnings (gcc
  3.2.2)
• Allow pcnet_cs to work with shared irq
• [IRDA]: Fix handling of shared IRQs in smsc-ircc2 driver
• [IRDA]: COnver irda-usb to dynamic device allocation
• [IRDA]: Rename setup_dma to irda_setup_dma
• [IRDA]: Make more symbols static, to avoid namespace
  pollution
• [IRDA]: Make more symbols static, to avoid namespace
  pollution
• [IRDA]: Make more symbols static, to avoid namespace
  pollution
• [IRDA]: Make more symbols static, to avoid namespace
  pollution
• [IRDA]: No need for volatile type when using set/test_bit
• [IRDA]: Kill infrared_mode, unused
• [IRDA]: Kill dev_flags, unused
• [IRDA]: Hashbin cleanups, remove unused code and add const
  where needed
• [IRDA]: Zap bogus wireless.h include
• [IRDA]: Make irda_device_txqueue_empty inline
• [IRDA]: Mark irport driver as having locking issues

Steve Kinneberg:

• IEEE1394(r1125): Set host field of hpsb_address_serve struct
  when registering address spaces
• IEEE1394(r1132): Update Kconfig description of the eth1394
  driver

Tomasz Torcz:

• Update via-rhine Kconfig entry

Tony Lindgren:

• [ARM PATCH] 1759/1: Add ARM925 support, updated

Walter Harms:

• [NET]: Handle kmem_cache_create() failure in neighbour.c
• [IPV4]: Handle kmem_cache_create() failure in inetpeer.c
• [IPV4]: Handle kmem_cache_create() failure in ipmr.c
• [IPV6]: Handle kmem_cache_create() failure in ip6_fib.c
• [IPV6]: Handle kmem_cache_create() failure in route.c

Wensong Zhang:

• [IPVS] remove the superfluous call of waitpid in sync code
• [IPVS] retry to run kernel_thread when memory is temporarily
  exhausted
• [IPVS] update the version number of code to 1.2.0
• [IPVS] tidy up the header files to include

Yasuyuki Kozakai:

• [IPV6]: In ipv6_skip_exthdr(), access frag header
  correctly
• [NETFILTER]: Fix ipv6 TCP/UDP matching wrt. extension
  headers

Yoshinori Sato:

• H8/300 start_thread problem fix
• H8/300 include cleanup
• H8/300 warning fix
• H8/300 io.h bussizing problem fix
• H8/300 Kconfig / defconfig update