Linux v2.5.45

Linux v2.5.45

[Linus Torvalds]

Big changes, lots of merges. A number of the merges are fairly
substantial too. 

Device mapper (LVM2), crypto/ipsec stuff for networking, epoll and giving
the new kernel configurator a chance. Big things.

And a _lot_ of maintenance, from various architecture updates to USB and
ISDN and ALSA. Merges with Andrew & Alan etc.. Go out and test, 

		Linus

---

Summary of changes from v2.5.44 to v2.5.45
============================================

<akropel1@rochester.rr.com>:
  o The following patch adds support for ethtool to the ewrk3 driver.
    It is against 2.5-BK but should apply to any recent 2.5 and 2.4 as
    well. In addition to adding ethtool support, it also removes the
    cli/sti fixup attribution from the changelog since that didn't
    actually go in yet and fixes a small style issue I introduced in
    the multi-card support patch.
  o This patch adds some locking fixups to the ewrk3 ioctl routine.
    None of these are critical since the ioctls AFAIK are used only by
    the EEPROM config utility.
  o Last ewrk3 update for now. Updates the changelog to cover previous
    patches, bumps the revision number, and replaces the horrific
    EthwrkSignature function with something (slightly) less horrific.
  o sd.c major number off-by-one

<ambx1@neo.rr.com>:
  o PnP Rewrite Fixes - 2.5.44
  o PnP cleanups and resource changes - 2.5.44 (1/4)
  o PnPBIOS changes - 2.5.44 (2/4)
  o Convert CS4236B driver - 2.5.44 (3/4)
  o update PnP layer to driver model changes - 2.5.44 (4/4)

<arnaud.quette@mgeups.com>:
  o drivers/usb/input/hiddev.c: fix hiddev_connect issue when

<dbrownell@users.sourceforge.net>:
  o ohci-hcd, longer bios handshake timeout
  o usbnet, preliminary zaurus support
  o usb: problem clearing halts

<dipankar@in.ibm.com>:
  o include/asm-sparc64/system.h: Add read_barrier_depends defines

<fw@deneb.enyo.de>:
  o [TCP]: In TCP_LISTEN state, ignore SYNs with RST set

<jbm@joshisanerd.com>:
  o Eliminate Old Prototypes from 2.5.44
  o [PATCH] fix a FIXME in usb.h

<jgarzik@redhat.com>:
  o Remove cli/sti from ewrk3 net driver
  o Fix tulip net driver multi-port board irq assignment
  o Add description of files in Documentation/BK-usage directory
  o Small clarification in BK kernel howto
  o Fix IO API breakage: Make inl() return unsigned int on x86 again
  o Update my email address
  o [IA32] Use -march=pentium{-mmx,3,4} in CFLAGS when available

<johnf@whitsunday.net.au>:
  o In patch-2.5.44 Mike Anderson <andmike@us.ibm.com> made a cleanup
    to the Scsi Host setup.

<jtyner@cs.ucr.edu>:
  o drivers/usb/media/vicam.c: simplify vicam_read
  o drivers/usb/media/vicam.c: simplify vicam_read

<jung-ik.lee@intel.com>:
  o ia64: PCI hotplug changes for 2.5.39 or later

<komujun@nifty.com>:
  o Add PCI id to tulip net driver

<mashirle@us.ibm.com>:
  o [IPV6]: Fix bugs in PMTU handling

<mhopf@innominate.com>:
  o [EBTABLES]: Add tcp/udp port checking

<n0ano@n0ano.com>:
  o ia64: Implement ia32 emulation for SG_IO

<niv@us.ibm.com>:
  o [IPV{4,6}]: Clean up SNMP counter bumping
  o [IPV4]: Add missing IpInUnknownProtos bump

<oliver@oenone.homelinux.org>:
  o USB: microtek driver - remove dead code
  o USB: hpusbscsi - kill wrong error case

<rohit.seth@intel.com>:
  o ia64: protect hugepage-check with mmap_sem

<shaggy@shaggy.austin.ibm.com>:
  o JFS: Add missing byte-swapping macros in xattr.c

<tony.luck@intel.com>:
  o ia64: make kcore work

<venkatesh.pallipadi@intel.com>:
  o ia64: Save/Restore of IA32 fpstate in sigcontext
  o ia64: Clearing of exception status before calling IA32 user signal
    handler

<willy@fc.hp.com>:
  o Simplify MCA date/time printing

Adam J. Richter <adam@yggdrasil.com>:
  o Use pci_[gs]et_drvdata instead of directly referenced ->driver_data
    in struct pci_dev.

Adrian Bunk <bunk@fs.tum.de>:
  o kbuild: Fix soundmodem/Makefile

Alan Cox <alan@lxorguk.ukuu.org.uk>:
  o MCA bus basic cleanups
  o small scsi compile fixes
  o move 53c7,8xx to pci_ not pcibios
  o ressurect the aha1740 driver
  o move advansys from pcibios to pci_
  o fix aic7xxx on gcc 3.2 warning spew
  o initial eata driver updates
  o fix all the IRQ breakage on the in2000
  o inia100 just has to lose a next: NULL
  o ncr53c8xxx needs updating for scsi_hn_get
  o resurrect the NCR53c406a
  o nsp32 needs updating for scsi_hn_get
  o fix scsi irq errors on seagate
  o nsp_cs update from maintainer
  o finish updating sym53c416
  o u14-34f update from maintainer
  o next NCR5380 updates
  o SCSI configure help
  o correct notes on scsi generic release
  o update qlogicfas driver
  o get the right thing out of se401 on gcc 3.2
  o merge befs file system from 2.4 (no core changes)
  o fix umem driver to use pci_get/set
  o xd_open is gone
  o make gscd compile again
  o kill tqueue in dz
  o make bluetooth compile again
  o update i810 tco to C99
  o move ip2 to workqueues
  o kill tqueue in specialix
  o stallion workqueue
  o move stallion to workqueue
  o update the qic02 tape driver to 2.5.44
  o remove tqueue.h from vme_sc
  o move hpt366 to pci_get
  o move siimage to pci_get/set
  o fix IDE compile with SIS5513
  o IDE floppy must be marked removable
  o ARM ide driver updates
  o remove dead ide suspend code
  o make the firewire layer build again
  o fix gcc warnings in eicon
  o kill tqueue in macintosh adb
  o update adv7175 to new i2c bus code
  o cpia driver update from maintainer
  o other minor video updates
  o mpt fusion updates for scsi changes
  o bring i2o_block/i2o_scsi back to life
  o ressurrect the 3c515 driver
  o bring the cops appletalk driver back
  o de620 resurrection
  o depca fix from maintainer
  o dl2k warning fix
  o fix hamradio netdriver builds
  o resurrect the 3c589_cs pcmcia
  o fix up the smc9194 - the extra locks arent needed
  o znet can go from space.c now
  o znet ethernet, back from the dead
  o update APM to match 2.4 features
  o core voyager arch/i386/machine support
  o Documentation for befs
  o remove acorn mfm tqueue.h
  o documentation for voyager
  o fix atm firestream warnings with new gcc
  o drag ATM into the 21st century , part 1
  o Device Mapper, with updates
  o Digital TV framework
  o update wan drivers to new saner ioctls
  o IDE - Andre can't count 8)
  o mempool helpers used by device mapper
  o updated ver_linux
  o DVB drivers AV7110 (Fujitsu, Nova etc)
  o Remove dead code in axnet_cs net driver

Alexander Viro <viro@math.psu.edu>:
  o A couple of compile fixes
  o rd
  o z2ram
  o xpram
  o ps2esdi
  o nftl
  o mtdblock (based on a patch from rmk)
  o hd.c
  o xd.c
  o ftl.c fix
  o dasd.c
  o swim3.c cleanup
  o mtdblock_ro fixes (based on patch from rmk)
  o blk_dev[] is gone
  o removed a bunch of gratuitous ->rq_dev uses
  o randomness made per-disk
  o r/o state moved to gendisks
  o presto cache keyed by superblock instead of kdev_t
  o removed a bunch of gratuitous kdev_t uses
  o saner initialization order in IDE (gendisks allocated slightly
    earlier)
  o block_device_operations always picked from gendisk
  o dasd fixes
  o IO counters - per-partition part
  o IO counters - per-disk part
  o ide-taskfile ioctls prototype cleanup
  o ide-{disk,cd,...} got separate block_device_operations
  o gendisk fixes
  o removal of root_dev_names[]
  o loop/shmfs fixes
  o compile fixes
  o more shm/loop updates
  o loop breakage fix

Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>:
  o [IPV4]: Kill ip_send, use dst_output instead
  o [NET]: Kill reroute from DST ops, unused
  o [IPV4]: Missing ip_rt_put in ip_route_newports
  o include/linux/ip.h: Define AH/ESP header layout
  o [NET]: Fix rtnetlink metric type, should be u32
  o [NET]: Cleanup DST metrics and abstract MSS/PMTU further
  o [NET]: Add DST_NOXFRM and DST_NOPOLICY flags
  o net/ipv4/route.c: Create compare_keys to compare flowi identities
  o [IPV4]: Rework key route lookup interface slightly
  o [IPSEC]: Add transform engine and AH implementation
  o [UDP]: Delete buggy assertion

Andi Kleen <ak@muc.de>:
  o x86-64 updates for 2.5.44

Andrew Morton <akpm@digeo.com>:
  o fix failure to write ext2 indirects under load
  o RCU idle detection fix
  o sparc64 read_barrier_depends fix
  o fid dmi compile warning
  o blkdev_get_block fix
  o move ramfs a_ops into libfs
  o libfs a_ops correctnes
  o invalidate_inode_pages fixes
  o restructure direct-io to suit bio_add_page
  o permit direct IO with finer-than-fs-blocksize alignments
  o add a file_ra_state init function
  o less buslocked operations in the page allocator
  o radix_tree_gang_lookup fix
  o export nr_running and nr_iowait tasks in /proc
  o faster copy_*_user for bad alignments on intel ia32
  o uninline the ia32 copy_*_user functions
  o shrink_slab arith overflow fix
  o thread-aware oom-killer
  o don't invalidate pagecache after direct-IO reads
  o much miscellany
  o tmpfs: shmem_getpage unlock_page
  o tmpfs: shmem_getpage beyond eof
  o tmpfs: shmem_getpage reading holes
  o tmpfs: shmem fs cleanup
  o tmpfs: shmem_file_sendfile
  o tmpfs: shmem_file_write update
  o tmpfs: shmem_getpage missing flush_dcache_page
  o tmpfs: support loopback
  o slab: extended cpu notifiers
  o slab: add_timer_on: add a timer on a particular CPU
  o slab: cleanup: rename static functions
  o slab: enable the cpu arrays on uniprocessor
  o slab: reduce internal fragmentation
  o slab: take the spinlock in the drain function
  o slab: remove spaces from /proc identifiers
  o slab: cleanups and speedups
  o slab: uninline poisoning checks
  o slab: reap timers
  o slab: Rework the slab timer code to use add_timer_on
  o slab: Remove cache_chain_lock
  o slab: additional code cleanup
  o slab: Use CPU notifiers
  o percpu: balance_dirty_pages ratelimit counters
  o percpu: fix compile warning for UP builds
  o percpu: convert RCU
  o percpu: convert timers
  o percpu: convert softirqs
  o percpu: convert buffer.c
  o percpu: create an EXPORT_PER_CPU_SYMBOL() macro
  o percpu: convert global page accounting
  o hot-n-cold pages: bulk page allocator
  o hot-n-cold pages: bulk page freeing
  o hot-n-cold pages: page allocator core
  o hot-n-cold pages: use cold pages for readahead
  o hot-n-cold pages: free and allocate hints

Andy Grover <agrover@groveronline.com>:
  o ACPI: Update to interpreter 20021022
  o ACPI: EC update
  o ACPI: Enable compilation using Intel compiler
  o ACPI: Add needed exports for ACPI-based PCI Hot Plug (J.I. Lee)
  o ACPI: Restore ARB_DIS bit on resume from S1 (Eric Brunet)
  o ACPI: Rename acpi_power_off to acpi_power_off_device (Pavel Machek)
  o ACPI: Remove too-broad blacklist entries
  o ACPI: Use dev->devfn instead of bridge->devfn to determine the pin
    when trying to derive a device's irq from its parent (Ville
    Syrjala)
  o ACPI: Add support for GPE1 block defined with no GPE0 block
  o ACPI: eliminate duplicate lines of code
  o ACPI: implement support for cpufreq interface (Dominik Brodowski)
  o ACPI: Try #2 at fixing the PCI IRQ bridge swizzle (Kai
    Germaschewski)

Arnaldo Carvalho de Melo <acme@conectiva.com.br>:
  o o ipv4: only produce one record per fib_seq_sholl call
  o o ipv4: move /proc/net/udp support back to net/ipv4/udp.c
  o o ipv4: move /proc/net/arp seq_file support back to arp.c
  o ipv4: move proc stuff from net/ipv4/af_inet.c to net/ipv4/proc.c
  o o llc: fix seq_file support

Art Haas <ahaas@neosoft.com>:
  o ia64: C99 designated initializer for include/asm-ia64/thread_info.h

Bart De Schuymer <bart.de.schuymer@pandora.be>:
  o [EBTABLES]: Add byte counter support, plus header cleanup
  o [BRIDGE]: bridge-nf, map IPv4 hooks onto bridge hooks
  o [BRIDGE]: Add ipt_physdev netfilter module

Brian Gerst <bgerst@didntduck.org>:
  o make x86 ptrace use init_fpu()
  o remove __verify_write from sh arch
  o i386 __verify_write fixes
  o factor common GCC options check

Christoph Hellwig <hch@lst.de>:
  o remove dead EH methods
  o Re: [PATCH] fix scsi device/driver model integration
  o get rid of ->finish method for highlevel drivers
  o remove scsi_merge.c
  o sanitize ->bios_param prototype
  o back out bogus init.h change
  o merge sd.h into sd.c and some cleanup
  o remove sd_disks global array from sd.c
  o fix sector_div use in scsicam.c
  o Re: [PATCH] fix sector_div use in scsicam.c
  o misc cleanups for sr
  o remove unused variable in scsi.c
  o get rid of global arrays in sd
  o remove LVM1 leftovers from the tree
  o sanitize intel movsl selection
  o fix xfs build after lvm removal

dan.zink@hp.com <Dan.Zink@hp.com>:
  o Compaq PCI Hotplug bug fix

David Brownell <david-b@pacbell.net>:
  o ehci enumerating full speed devices
  o rm "automagic resubmit" for usb interrupt transfers
  o Zaurus support for usbnet
  o create <linux/usb_ch9.h>
  o USB: clean up usb structures some more
  o ohci td error cleanup
  o usbtest mentions url

David Mosberger <davidm@tiger.hpl.hp.com>:
  o ia64: Incorporate no-flush-needed optimization from Andrew's
    asm-generic/tlb.h
  o ia64: Fix copy_siginfo() to copy all relevant bytes
  o ia64: Fix EFI runtime callbacks so they cannot corrupt fp regs.  A
    few minor
  o ia64: Sync with 2.5.39
  o ia64: Fix 2.5.39 Makefile breakage
  o ia64: Update defconfig
  o ia64: Remove duplicate make targets
  o ia64: Some formatting cleanups
  o ia64: Fix formatting a bit and issue #error when attempting to use
    CONFIG_NUMA without CONFIG_ACPI_NUMA.
  o ia64: Fix up/clean NUMA discontigmem patch
  o ia64: Sync with 2.5.44
  o ia64: Clean up ia64 version of topology.h
  o ia64: Make kernel profiling work again (patch by Peter Chubb)
  o ia64: Minor Makefile cleanup.  Mention CONFIG_NUMA option in
    defconfig
  o ia64: Create dummy offsets.h if it doesn't exist yet.  Patch by
    Keith Owens
  o ia64: Fix Keith's Makefile fix so it actually works

David Mosberger <davidm@wailua.hpl.hp.com>:
  o ia64: Fix perfmon initialization bug (patch by Stephane Eranian)

David S. Miller <davem@nuts.ninka.net>:
  o [IPV4]: Provide full proto/ports in flowi route lookups
  o net/ipv4/af_inet.c: Include net/ip_fib.h
  o net/ipv4/ip_proc.c: Include linux/ax25.h and handle modular AX25
  o arch/sparc64/kernel/ioctl32.c: Handle HDIO_GETGEO_BIG{,_RAW}
  o [IPV4]: When advmss of route is zero, report it as zero not 40
  o [CRYPTO]: Fix compiler warnings and build failures
  o [CRYPTO]: Forgotten file add in previous commit
  o [ip-sysctl.txt]: Clarify conf/*/ behavior
  o [PNP]: Fix build when CONFIG_PNP is not set
  o [SPARC64]: Only HDIO_GETGEO_BIG_RAW exists in 2.5
  o [SPARC64]: Remove silly rule to remove -pg from cflags
  o [SPARC64]: Update defconfig
  o [SPARC]: Bring ESP driver in line with modern EH handling
  o [SPARC]: Bring QlogicPTI driver in line with modern EH handling
  o [FC4]: Kill all references to fcp_old_abort
  o [ESP]: Fix abort return values
  o [CRYPTO]: kunmap does not return a value
  o [CRYPTO]: Build/warning fixups
  o [IPSEC]: Remove debugging code
  o [CRYPTO]: Clean up header file usage
  o include/linux/crypto.h: Include linux/string.h

Davide Libenzi <davidel@xmailserver.org>:
  o sys_epoll 0.15

david_jeffery@adaptec.com <David_Jeffery@adaptec.com>:
  o ips queue depths 2.5.44

Doug Ledford <dledford@aladin.rdu.redhat.com>:
  o Update for new TCQ scheme
  o Fix for scsi host struct change
  o host struct cleanups
  o Compile fixes needed due to host struct change

Erich Focht <efocht@ess.nec.de>:
  o acpi-numa for ia64
  o ia64: topology for ia64

Greg Kroah-Hartman <greg@kroah.com>:
  o IBM PCI Hotplug: fix typos in previous patch
  o USB: added support for Clie NX60 device
  o driver core: add support for calling /sbin/hotplug when classes are
    found and removed from the system
  o USB: fix the usb serial drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb input drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb class drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o fix the usb image drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb media drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb misc drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb net drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o fix the usb storage drivers due to interrupt urb no automatic
    resubmission change to the usb core
  o USB: fix the usb drivers outside the drivers/usb tree due to
    interrupt urb no automatic resubmission change to the usb core
  o USB: fix GFP flags for usb audio driver
  o USB: Fixes for previous USB_* flag patch
  o USB: usb serial driver fixes due to USB structure changes
  o USB: drivers/usb fixups due to USB structure changes
  o USB: sound/usb fixups due to USB structure changes
  o USB: drivers/usb fixups due to USB structure changes
  o USB: drivers/isdn/hisax fixups due to USB structure changes
  o USB: drivers/net/irda fixups due to USB structure changes
  o USB: fix usbmidi driver for no automatic resubmission of interrupt
    urbs

Harald Welte <laforge@gnumonks.org>:
  o [NETFILTER] Add IP unused bit check to ipt_unclean.c, from Maciej
    Soltysiak

Hideaki Yoshifuji <yoshfuji@linux-ipv6.org>:
  o [IPV6]: Add IPV6_V6ONLY socket option support
  o [IPV6]: Add ICMP6 rate limit sysctl
  o [IPV6]: Split ndisc_rcv into helper functions
  o [IPV6]: Avoid garbage sin6_scope_id for MSG_ERRQUEUE messages
  o [IPV6]: Fix for refined IPV6 address validation timer

Hirofumi Ogawa <hirofumi@mail.parknet.co.jp>:
  o remove the conv option of fat (1/3)
  o remove the fat_cvf stuff (2/3)
  o small cleanup of fat  (3/3)

Irene Zubarev <zubarev@us.ibm.com>:
  o IBM PCI Hotplug: small patch

James Bottomley <jejb@mulgrave.(none)>:
  o [SCSI] remove duplicate device registration
  o [SCSI] move build commandblocks to before attach so attach can send
    I/O
  o [PATCH scsi] use sector_div in scsicam.c
  o [SCSI] replace max_host_blocked initialisation lost in hosts rework
  o [SCSI] fix memory etc. leak caused by double preparing requeued
    commands
  o [SCSI] documentation tidy ups and an interface fix in
    mlqueue_insert
  o Merge by hand: recover axboe scsi_init_io() changes
  o Correct sd.c compile by adding } lost in merge
  o patch up scsi mismerge

James Morris <jmorris@intercode.com.au>:
  o [CRYPTO]: Add initial crypto api subsystem
  o [CRYPTO]: Add in 3des implementation
  o [CRYPTO]: Cleanups based upon feedback from Rusty and jgarzik
  o [CRYPTO]: Use try_inc_mod_count and semaphore for alg list
  o [CRYPTO]: Use kmod to try to autoload modules
  o [CRYPTO]: Bug fixes and cleanups
  o [CRYPTO]: More bug fixes and cleanups
  o [CRYPTO]: Add MD4
  o [CRYPTO]: Algorithm lookup API change plus bug fixes
  o [NET]: Backport netlink_set_nonroot changes by Andi Kleen
  o [CRYPTO]: Run tcrypt through lindent, plus doc update
  o [CRYPTO]: Assert that interfaces are called on correct cipher type
  o [CRYPTO]: Cleanups and more consistency checks
  o [CRYPTO]: Update to IV get/set interface
  o [CRYPTO]: Add some documentation
  o [CRYPTO]: Fix some credits
  o [CRYPTO]: Cleanups based upon suggestions by Jeff Garzik
  o [CRYPTO]: Uninline some functions to save some bloat

Jaroslav Kysela <perex@suse.cz>:
  o ALSA updates

Jens Axboe <axboe@suse.de>:
  o scsi patches
  o make deadline_merge prefetch next entry
  o end_io bouncing
  o sr_ioctl must return -EIO, not -EINVAL
  o elv_add_request cleanups
  o make blk_dump_rq_flags a bit more useful
  o make queue prep_rq_fn() a bit more powerful
  o queue dma alignment
  o queue last_merge hint cleanup
  o request references and list deletion/insertion checking
  o add end_request helpers that deal in bytes, not sectors
  o various ide fixes and cleanups
  o queue merge_bvec_fn() changes
  o make bio->bi_end_io() optional
  o bio_map_user() infrastructure
  o misc scsi bits
  o small block bits
  o finally, sgio updates
  o ide-cd updates
  o missed elv_add_request() update
  o bad scsi merge
  o scsi_command_size[] only known when SCSI is enabled
  o remember to export scsi_command_size[]
  o arrange request fiels sanely

John Levon <levon@movementarian.org>:
  o add oprofile to MAINTAINERS
  o fix oprofile multiple counters

Kai Germaschewski <kai@tp1.ruhr-uni-bochum.de>:
  o ISDN/PPP: Separate out VJ header compression
  o ISDN: Remove reference to eth_header
  o ISDN: isdn_netif_rx() helper
  o ISDN/PPP: Separate out and rewrite MPPP code
  o kbuild: Split Rules.make
  o kbuild: Remove some compatibility code, $(echo_target)
  o kbuild: Shut up "make clean" in non-verbose mode
  o kbuild: Switch "make modules_install" to fast mode ;)
  o kbuild: Convert build and modversion phases
  o kbuild: Convert drivers/isdn to be "Rules.make-less"
  o kbuild: Allow for <mod>-y as well as <mod>-objs for multipart
    objects
  o kbuild: Allow UTS_MACHINE to be different from $(ARCH)
  o kbuild: Fix a "make -j<N>" warning
  o ISDN: Move HiSax to spinlocks instead of cli()
  o ISDN: Fix up the introduced spinlocks
  o ISDN: Fix the workqueue changes for the HiSax driver
  o ISDN/PPP: Remove random frame drop
  o ISDN: header cosmetics
  o ISDN: Remove CVS $Revision
  o ISDN/PPP: Pass frame including header to MPPP
  o ISDN: Move drivers/isdn/i4l/isdn_fsm.h include/linux/isdn/fsm.h
  o ISDN: Move ISDN net lib interface related definitions into
    isdn_net_lib.h
  o ISDN: Make raw-IP, CISCO HDLC, ... support optional
  o ISDN: Move isdn_net_lib specific definitions out of linux/isdn.h
  o ISDN: Add missed isdn_net_lib.h
  o ISDN: alloc CISCO HDLC info dynamically
  o ISDN: Convert ISDN/X.25 to inl_priv / ind_priv
  o ISDN: Convert ISDN/PPP to inl_priv / ind_priv
  o ISDN: Remove rcv_waitq/snd_waitq
  o ISDN: Fix AT+FREV command
  o ISDN: improve /dev/isdnctrl read()/write()
  o ISDN: Make array of drivers private to isdn_common.c
  o ISDN: Use a spinlock to protect the list of drivers
  o ISDN: Get rid of global drivers count
  o ISDN: Kill drvid[] array
  o ISDN: State machines for the link layer
  o ISDN: Remove ISDN_STAT_NODCH
  o ISDN: Move driver unload into the state machine
  o ISDN: Remove ISDN_STAT_L1ERR
  o ISDN: STAT_FAXIND and STAT_AUDIO handled by state machine
  o ISDN: Remove isdn_driver::online flags
  o ISDN: Signal incoming calls to ttyI's again
  o ISDN: Remove ttyI specific from global "dev" variable
  o ISDN: Route all driver callbacks through the driver state machine
  o ISDN: Move the tty receive queue out of generic code
  o ISDN: Assorted cleanups
  o ISDN: Make V.110 support less intrusive
  o ISDN: Fix isdnloop for transparent/V.110
  o ISDN: Remove isdn_dc2minor(), isdn_slot_all_eaz()
  o ISDN: stat_callback() and recv_callback() -> event_callback()
  o ISDN: Pass around struct isdn_slot directly
  o ISDN: New timer handling for "+++" escape sequence
  o ISDN: New timer handling for ttyI RING response
  o ISDN: New timer handling for ttyI NO CARRIER response
  o ISDN: Remove delayed ttyI xmit
  o ISDN: New timer handling for read timer
  o ISDN: ttyI cleanups
  o ISDN: lock only used driver
  o kbuild: Fix menuconfig/xconfig and a modversions problem

Kent Yoder <key@austin.ibm.com>:
  o update lanstreamer tokenring driver
  o Check link status in pcnet32 net driver

Kimio Suganuma <k-suganuma@mvj.biglobe.ne.jp>:
  o ia64: ACPI NUMA bugfix
  o ia64: discontigmem patch for 2.5 ia64
  o ia64: allocate all per-CPU pages at BSP-initialization time

Linus Torvalds <torvalds@home.transmeta.com>:
  o Delete old-style config files
  o ia-64 kcore changes broke i386. Guess who gets the shaft?
  o Fix ACPI frequency states to not play games with the configuration
    system, and instead just cleanly show the dependency.
  o Fix up horribly wrong test in new copy-to-user() implementation.
    The optimized versions only work for large areas, make sure we
    don't use them for anything else.

Manfred Spraul <manfred@colorfullife.com>:
  o use correct wakeups in fs/pipe.c

Matt Domsch <Matt_Domsch@dell.com>:
  o EDD: add comments, magic value defines, use snprintf always
  o EDD: cleanups
  o EDD: remove list_head from edd_device, don't delete symlinks
  o EDD: moved attr_test to edd_attribute ->test(), comments

Matthew Dharm <mdharm-usb@one-eyed-alien.net>:
  o USB storage: fix error code
  o USB storage: use scatter-gather core primitives

Matthew Dobson <colpatch@us.ibm.com>:
  o Remove sole CONFIG_MULIQUAD in kernel source

Matthew Wilcox <willy@debian.org>:
  o [NET]: Move more ioctls to top level
  o Update lasi_82596 net driver to replace cli/sti with spinlock
  o PA-RISC math emu
  o include/asm-parisc
  o arch/parisc/mm
  o arch/parisc/kernel
  o perf monitor for PA-RISC
  o parisc64
  o misc PA updates

Mike Anderson <andmike@us.ibm.com>:
  o fix module unload of sg
  o scsi_error device offline fix
  o scsi sync caches w/ dev offline

Neil Brown <neilb@cse.unsw.edu.au>:
  o Define domain_release handle for AUTH_UNIX domains
  o md: factor out MD superblock handling code
  o kNFSd: Fix nfs shutdown problem
  o kNFSd: Make sure export_open cleans up on failure
  o kNFSd: Fix problem with buffer length with rpc/tcp
  o kNFSd: nfsd_readdir changes
  o kNFSd: Convert nfsd to use a list of pages instead of one big
    buffer

Patrick Mochel <mochel@osdl.org>:
  o introduce struct kobject: simple, generic object for embedding in
    other structures
  o sysfs: convert sysfs to use more functions from fs/libfs.c
  o sysfs: marry api with struct kobject
  o sysfs: make symlinks easier
  o Introduce struct subsystem
  o sysfs: kill struct sysfs_dir
  o kobjects: add array of default attributes to subsystems, and create
    on registration

Pavel Machek <pavel@ucw.cz>:
  o swsusp -- small fixes
  o swsusp updates

Pete Zaitcev <zaitcev@redhat.com>:
  o [SPARC]: More -ffunction-sections followups
  o [SPARC]: Some forgotten asm_offsets.h includes

Peter Chubb <peter@chubb.wattle.id.au>:
  o ia64: Fix RAW dependency introduced by HUGETLB patch

Randy Dunlap <rddunlap@osdl.org>:
  o Convert /proc/swaps to use seq_file API
  o remove double-init in /proc/ksyms

Rob Radez <rob@osinvestor.com>:
  o [SPARC32]: Fix build in several spots

Robert Love <rml@tech9.net>:
  o overcommit-accounting doc fix

Roman Zippel <zippel@linux-m68k.org>:
  o new kernel configuration 1-7
  o kconfig update
  o kconfig "choice" fixes

Rusty Russell <rusty@rustcorp.com.au>:
  o Get rid of check_resource() before it becomes a problem

Sam Ravnborg <sam@mars.ravnborg.org>:
  o kbuild: Removed unused definitions
  o kbuild: scrits/Makefile.lib Moved generic definitions to
    Makefile.lib, This allows us to share all generic definitions
    between the different Makefiles.
  o kbuild: Use Makefile.lib for modversion and modules_install Most
    definitions required were present in Makefile.lib, so delete the
    definitions and include Makefile.lib.
  o kbuild: Got rid of $(call descend ...) in top-level Makefile
    Replaced by the more readable $(Q)$(MAKE) construct
  o kbuild: Added Descend to top-level Makefile again It is used by
    arch specific Makefiles

Skip Ford <skip.ford@verizon.net>:
  o net/ipv4/raw.c: Include netfilter_ipv4.h

Stelian Pop <stelian.pop@fr.alcove.com>:
  o sonypi driver update

Stuart MacDonald <stuartm@connecttech.com>:
  o More wh patches

Stéphane Eranian <eranian@hpl.hp.com>:
  o 2.5.35 perfmon update

Takayoshi Koshi <t-kouchi@mvf.biglobe.ne.jp>:
  o ACPI PCI hotplug driver for 2.5
  o ia64: fix fpswa version printing

Tim Hockin <thockin@freakshow.cobalt.com>:
  o drivers/net/eepro100.c: cleanup messages since netif_msg_xxx()
    change
  o drivers/net/eepro100.c: set the PHY ID correctly
  o drivers/net/mii.c: fix flipped logic
  o drivers/net/eepro100.c: set phy_id_mask and reg_num_mask in mii_if

Tim Schmielau <tim@physik3.uni-rostock.de>:
  o fix compares of jiffies

Re: Linux v2.5.45

[Udo A. Steinberg]

[-- Attachment #1: Type: text/plain, Size: 797 bytes --]

On Wed, 30 Oct 2002 16:56:29 -0800 (PST) Linus Torvalds (LT) wrote:

LT> Summary of changes from v2.5.44 to v2.5.45
LT> ============================================

[...]

fs/nfsd/nfs4proc.c: In function `nfsd4_write':
fs/nfsd/nfs4proc.c:484: warning: passing arg 4 of `nfsd_write' from incompatible pointer type
fs/nfsd/nfs4proc.c:484: warning: passing arg 6 of `nfsd_write' makes integer from pointer without a cast
fs/nfsd/nfs4proc.c:484: too few arguments to function `nfsd_write'
fs/nfsd/nfs4proc.c: In function `nfsd4_proc_compound':
fs/nfsd/nfs4proc.c:568: structure has no member named `rq_resbuf'
fs/nfsd/nfs4proc.c:569: structure has no member named `rq_resbuf'
fs/nfsd/nfs4proc.c:569: structure has no member named `rq_resbuf'
make[3]: *** [fs/nfsd/nfs4proc.o] Error 1

Regards,
-Udo.

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux v2.5.45

[Aaron Lehmann]

On Wed, Oct 30, 2002 at 04:56:29PM -0800, Linus Torvalds wrote:
> 
> 
> Big changes, lots of merges. A number of the merges are fairly
> substantial too. 
> 
> Device mapper (LVM2), crypto/ipsec stuff for networking, epoll and giving
> the new kernel configurator a chance. Big things.

Now running 'make oldconfig' or 'make menuconfig' requires a Qt
installation. I believe that this is a bug because these still work
fine without Qt when the -k flag is passed to make.

Re: Linux v2.5.45

[Adam Huffman]

On Wed, 30 Oct 2002, Linus Torvalds wrote:

> 
> 
> Big changes, lots of merges. A number of the merges are fairly
> substantial too. 
> 
> Device mapper (LVM2), crypto/ipsec stuff for networking, epoll and giving
> the new kernel configurator a chance. Big things.
> 
> And a _lot_ of maintenance, from various architecture updates to USB and
> ISDN and ALSA. Merges with Andrew & Alan etc.. Go out and test, 
> 
> 		Linus
> 

  gcc -Wp,-MD,drivers/md/.dm-ioctl.o.d -D__KERNEL__ -Iinclude -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
-march=athlon -Iarch/i386/mach-generic -nostdinc -iwithprefix include
-DKBUILD_BASENAME=dm_ioctl   -c -o drivers/md/dm-ioctl.o
drivers/md/dm-ioctl.c
drivers/md/dm-ioctl.c: In function `create':
drivers/md/dm-ioctl.c:588: incompatible type for argument 1 of
`set_device_ro'
drivers/md/dm-ioctl.c: In function `reload':
drivers/md/dm-ioctl.c:874: incompatible type for argument 1 of
`set_device_ro'
make[2]: *** [drivers/md/dm-ioctl.o] Error 1
make[1]: *** [drivers/md] Error 2
make: *** [drivers] Error 2

Re: Linux v2.5.45

[Alexander Viro]

On Wed, 30 Oct 2002, Aaron Lehmann wrote:

> On Wed, Oct 30, 2002 at 04:56:29PM -0800, Linus Torvalds wrote:
> > 
> > 
> > Big changes, lots of merges. A number of the merges are fairly
> > substantial too. 
> > 
> > Device mapper (LVM2), crypto/ipsec stuff for networking, epoll and giving
> > the new kernel configurator a chance. Big things.
> 
> Now running 'make oldconfig' or 'make menuconfig' requires a Qt
> installation. I believe that this is a bug because these still work
> fine without Qt when the -k flag is passed to make.

Remove "false" from the rule that spits out annoying shit about absence
of QT (_yes_, I _know_ that I don't have that shite installed, thank
you very much for reminder).

Doesn't solve the annoyance problem, though.

Re: Linux v2.5.45

[Jeff Garzik]

Adam Huffman wrote:

>  gcc -Wp,-MD,drivers/md/.dm-ioctl.o.d -D__KERNEL__ -Iinclude -Wall
>-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
>-fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
>-march=athlon -Iarch/i386/mach-generic -nostdinc -iwithprefix include
>-DKBUILD_BASENAME=dm_ioctl   -c -o drivers/md/dm-ioctl.o
>drivers/md/dm-ioctl.c
>drivers/md/dm-ioctl.c: In function `create':
>drivers/md/dm-ioctl.c:588: incompatible type for argument 1 of
>`set_device_ro'
>drivers/md/dm-ioctl.c: In function `reload':
>drivers/md/dm-ioctl.c:874: incompatible type for argument 1 of
>`set_device_ro'
>make[2]: *** [drivers/md/dm-ioctl.o] Error 1
>make[1]: *** [drivers/md] Error 2
>make: *** [drivers] Error 2
>  
>


yeah, don't use it for now, it needs more cleanups.

Re: Linux v2.5.45

[Jan Harkes]

On Wed, Oct 30, 2002 at 08:44:22PM -0500, Alexander Viro wrote:
> > Now running 'make oldconfig' or 'make menuconfig' requires a Qt
> > installation. I believe that this is a bug because these still work
> > fine without Qt when the -k flag is passed to make.
> 
> Remove "false" from the rule that spits out annoying shit about absence
> of QT (_yes_, I _know_ that I don't have that shite installed, thank
> you very much for reminder).
> 
> Doesn't solve the annoyance problem, though.

This patch got rid if the annoying warning for me. It might have broken
the xconfig stuff, but at least oldconfig/config/menuconfig don't depend
on having QT installed anymore.

Jan


--- linux-2.5.45/Makefile	2002-10-30 20:53:08.000000000 -0500
+++ linux/Makefile	2002-10-30 20:54:12.000000000 -0500
@@ -635,7 +635,7 @@
 .PHONY: oldconfig xconfig menuconfig config \
 	make_with_config
 
-scripts/kconfig/conf scripts/kconfig/mconf scripts/kconfig/qconf: scripts/fixdep FORCE
+scripts/kconfig/conf scripts/kconfig/mconf: scripts/fixdep FORCE
 	+@$(call descend,scripts/kconfig,$@)
 
 xconfig: scripts/kconfig/qconf

Re: Linux v2.5.45

[Neil Brown]

On Thursday October 31, us15@os.inf.tu-dresden.de wrote:
> On Wed, 30 Oct 2002 16:56:29 -0800 (PST) Linus Torvalds (LT) wrote:
> 
> LT> Summary of changes from v2.5.44 to v2.5.45
> LT> ============================================
> 
> [...]
> 
> fs/nfsd/nfs4proc.c: In function `nfsd4_write':
> fs/nfsd/nfs4proc.c:484: warning: passing arg 4 of `nfsd_write' from incompatible pointer type
> fs/nfsd/nfs4proc.c:484: warning: passing arg 6 of `nfsd_write' makes integer from pointer without a cast
> fs/nfsd/nfs4proc.c:484: too few arguments to function `nfsd_write'
> fs/nfsd/nfs4proc.c: In function `nfsd4_proc_compound':
> fs/nfsd/nfs4proc.c:568: structure has no member named `rq_resbuf'
> fs/nfsd/nfs4proc.c:569: structure has no member named `rq_resbuf'
> fs/nfsd/nfs4proc.c:569: structure has no member named `rq_resbuf'
> make[3]: *** [fs/nfsd/nfs4proc.o] Error 1
> 

Yes. 

   http://linux.bkbits.com:8080/linux-2.5/cset@1.844.1.75

I did NFSv2 and NFSv3 which are similar.  NFSv4 is quite different
code so I wanted to leave it for a separate patch (which isn't done yet).

NeilBrown

Re: Linux v2.5.45

[Martin J. Bligh]

Just some warnings, if anyone's bored, and wants something to fix ;-)

drivers/base/base.h:64: warning: `class_hotplug' defined but not used
drivers/base/base.h:64: warning: `class_hotplug' defined but not used
drivers/base/base.h:64: warning: `class_hotplug' defined but not used
drivers/base/base.h:64: warning: `class_hotplug' defined but not used
drivers/base/base.h:64: warning: `class_hotplug' defined but not used
mm/slab.c: In function `cache_reap':
mm/slab.c:2061: warning: deprecated use of label at end of compound statement
mm/vmscan.c: In function `shrink_caches':
mm/vmscan.c:733: warning: duplicate `const'
mm/swap_state.c: In function `free_pages_and_swap_cache':
mm/swap_state.c:299: warning: duplicate `const'
drivers/net/starfire.c: In function `netdev_close':
drivers/net/starfire.c:1848: warning: unsigned int format, different type arg (arg 2)
drivers/net/starfire.c:1848: warning: unsigned int format, different type arg (arg 2)
drivers/net/starfire.c:1855: warning: unsigned int format, different type arg (arg 2)
drivers/net/starfire.c:1855: warning: unsigned int format, different type arg (arg 2)
net/ipv4/route.c: In function `ip_rt_init':
net/ipv4/route.c:2544: warning: implicit declaration of function `xfrm_init'

I think the starfire ones will need PAE turned on in the config to
trigger it (printing a phys ptr using an unsigned long format?)

M.

Re: Linux v2.5.45

[Udo A. Steinberg]

[-- Attachment #1: Type: text/plain, Size: 1194 bytes --]

On Wed, 30 Oct 2002 16:56:29 -0800 (PST) Linus Torvalds (LT) wrote:

LT> Summary of changes from v2.5.44 to v2.5.45
LT> ============================================

drivers/acpi/sleep.c: In function `acpi_system_suspend':
drivers/acpi/sleep.c:209: warning: implicit declaration of function `do_suspend_lowlevel'
drivers/acpi/sleep.c: In function `acpi_sleep_init':
drivers/acpi/sleep.c:707: `sysrq_acpi_poweroff_op' undeclared (first use in this function)
drivers/acpi/sleep.c:707: (Each undeclared identifier is reported only once
drivers/acpi/sleep.c:707: for each function it appears in.)
make[3]: *** [drivers/acpi/sleep.o] Error 1

CONFIG_ACPI=y
# CONFIG_ACPI_HT_ONLY is not set
CONFIG_ACPI_BOOT=y
CONFIG_ACPI_SLEEP=y
# CONFIG_ACPI_AC is not set
# CONFIG_ACPI_BATTERY is not set
CONFIG_ACPI_BUTTON=y
CONFIG_ACPI_FAN=y                                         
CONFIG_ACPI_PROCESSOR=y                                   
CONFIG_ACPI_THERMAL=y
# CONFIG_ACPI_TOSHIBA is not set
# CONFIG_ACPI_DEBUG is not set
CONFIG_ACPI_BUS=y
CONFIG_ACPI_INTERPRETER=y
CONFIG_ACPI_EC=y
CONFIG_ACPI_POWER=y
CONFIG_ACPI_PCI=y
CONFIG_ACPI_SYSTEM=y 
# CONFIG_PM is not set                  <=====

Regards,
-Udo.

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux v2.5.45

[Greg KH]

On Wed, Oct 30, 2002 at 06:56:32PM -0800, Martin J. Bligh wrote:
> Just some warnings, if anyone's bored, and wants something to fix ;-)
> 
> drivers/base/base.h:64: warning: `class_hotplug' defined but not used
> drivers/base/base.h:64: warning: `class_hotplug' defined but not used
> drivers/base/base.h:64: warning: `class_hotplug' defined but not used
> drivers/base/base.h:64: warning: `class_hotplug' defined but not used
> drivers/base/base.h:64: warning: `class_hotplug' defined but not used

Why would you want to run a kernel with CONFIG_HOTPLUG turned off!

Grumble, I forgot a "inline" in base.h, I'll fix it later...

greg k-h

Re: Linux v2.5.45

[Paul]

Jeff Garzik <jgarzik@pobox.com>, on Wed Oct 30, 2002 [08:49:43 PM] said:
> Adam Huffman wrote:
> 
> > gcc -Wp,-MD,drivers/md/.dm-ioctl.o.d -D__KERNEL__ -Iinclude -Wall
> >-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
> >-fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
> >-march=athlon -Iarch/i386/mach-generic -nostdinc -iwithprefix include
> >-DKBUILD_BASENAME=dm_ioctl   -c -o drivers/md/dm-ioctl.o
> >drivers/md/dm-ioctl.c
> >drivers/md/dm-ioctl.c: In function `create':
> >drivers/md/dm-ioctl.c:588: incompatible type for argument 1 of
> >`set_device_ro'
> >drivers/md/dm-ioctl.c: In function `reload':
> >drivers/md/dm-ioctl.c:874: incompatible type for argument 1 of
> >`set_device_ro'
> >make[2]: *** [drivers/md/dm-ioctl.o] Error 1
> >make[1]: *** [drivers/md] Error 2
> >make: *** [drivers] Error 2
> > 
> >
> 
> 
> yeah, don't use it for now, it needs more cleanups.
> 
> 
	Hi;

	I need lvm to test this kernel series. I am not using
md stuff, but this is where it breaks for me too. If there
are corrective patches, please let us know...

Paul
set@pobox.com

[PATCH] check QT only if needed

[Roman Zippel]

Hi,

On Wed, 30 Oct 2002, Aaron Lehmann wrote:

> Now running 'make oldconfig' or 'make menuconfig' requires a Qt
> installation. I believe that this is a bug because these still work
> fine without Qt when the -k flag is passed to make.

Yes, it's a bug. The patch below fixes this without breaking xconfig.
Linus, please apply.

bye, Roman

# Only check for the qt installation if a qconf build is requested

--- linux-2.5/scripts/kconfig/Makefile.org	2002-10-28 00:15:29.000000000 +0100
+++ linux-2.5/scripts/kconfig/Makefile	2002-10-31 10:23:07.000000000 +0100
@@ -34,6 +34,7 @@
 
 $(obj)/qconf.o: $(obj)/.tmp_qtcheck
 
+ifeq ($(MAKECMDGOALS),$(obj)/qconf)
 -include $(obj)/.tmp_qtcheck
 
 # QT needs some extra effort...
@@ -52,6 +53,7 @@
 	LIB=qt; \
 	if [ -f $$DIR/lib/libqt-mt.so ]; then LIB=qt-mt; fi; \
 	echo "QTDIR=$$DIR" > $@; echo "QTLIB=$$LIB" >> $@
+endif
 
 $(obj)/zconf.tab.o: $(obj)/lex.zconf.c
 

Re: Linux v2.5.45

[Joe Thornber]

On Thu, Oct 31, 2002 at 01:37:25AM +0000, Adam Huffman wrote:
>   gcc -Wp,-MD,drivers/md/.dm-ioctl.o.d -D__KERNEL__ -Iinclude -Wall
> -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
> -fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
> -march=athlon -Iarch/i386/mach-generic -nostdinc -iwithprefix include
> -DKBUILD_BASENAME=dm_ioctl   -c -o drivers/md/dm-ioctl.o
> drivers/md/dm-ioctl.c
> drivers/md/dm-ioctl.c: In function `create':
> drivers/md/dm-ioctl.c:588: incompatible type for argument 1 of
> `set_device_ro'
> drivers/md/dm-ioctl.c: In function `reload':
> drivers/md/dm-ioctl.c:874: incompatible type for argument 1 of
> `set_device_ro'
> make[2]: *** [drivers/md/dm-ioctl.o] Error 1
> make[1]: *** [drivers/md] Error 2
> make: *** [drivers] Error 2

I've already posted the patches that are needed to the list.  Alternatively get them
from here:

http://people.sistina.com/~thornber/patches/2.5-stable/2.5.45-dm-1.tar.bz2

- Joe

Re: Linux v2.5.45

[Jochen Friedrich]

Hi,

2.4.45 compile on Alpha fails:

  gcc -Wp,-MD,arch/alpha/kernel/.irq_alpha.o.d -D__KERNEL__ -Iinclude
-Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mno-fp-regs -ffixed-8 -mcpu=ev5
-Wa,-mev6 -nostdinc -iwithprefix include    -DKBUILD_BASENAME=irq_alpha
-c -o arch/alpha/kernel/irq_alpha.o arch/alpha/kernel/irq_alpha.c
In file included from arch/alpha/kernel/irq_alpha.c:15:
arch/alpha/kernel/irq_impl.h: In function `alpha_do_profile':
arch/alpha/kernel/irq_impl.h:50: `prof_buffer' undeclared (first use in
this function)
arch/alpha/kernel/irq_impl.h:50: (Each undeclared identifier is reported
only once
arch/alpha/kernel/irq_impl.h:50: for each function it appears in.)
arch/alpha/kernel/irq_impl.h:61: `prof_shift' undeclared (first use in
this function)
arch/alpha/kernel/irq_impl.h:67: `prof_len' undeclared (first use in this
function)
make[1]: *** [arch/alpha/kernel/irq_alpha.o] Error 1
make: *** [arch/alpha/kernel] Error 2

--jochen

Re: Linux v2.5.45

[Adrian Bunk]

On Wed, 30 Oct 2002, Linus Torvalds wrote:

>...
> Summary of changes from v2.5.44 to v2.5.45
> ============================================
>...
> Alexander Viro <viro@math.psu.edu>:
>...
>   o ps2esdi
>...


This patch changed the parameters of ps2esdi_readwrite but didn't change
the function prototype resulting in the following compile error:

<--  snip  -->

...
  gcc -Wp,-MD,drivers/block/.ps2esdi.o.d -D__KERNEL__ -Iinclude -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer -fno-strict-aliasing
-fno-common -pipe -mpreferred-stack-boundary=2 -march=k6 -Iarch/i386/mach-generic
-nostdinc -iwithprefix include    -DKBUILD_BASENAME=ps2esdi   -c -o
drivers/block/ps2esdi.o drivers/block/ps2esdi.c
...
drivers/block/ps2esdi.c:566: conflicting types for `ps2esdi_readwrite'
drivers/block/ps2esdi.c:77: previous declaration of `ps2esdi_readwrite'
make[2]: *** [drivers/block/ps2esdi.o] Error 1

<--  snip   -->


The fix is simple:


--- linux-2.5.45-full/drivers/block/ps2esdi.c.old	2002-10-31 11:42:27.000000000 +0100
+++ linux-2.5.45-full/drivers/block/ps2esdi.c	2002-10-31 12:11:50.000000000 +0100
@@ -74,7 +74,7 @@

 static void do_ps2esdi_request(request_queue_t * q);

-static void ps2esdi_readwrite(int cmd, u_char drive, u_int block, u_int count);
+static void ps2esdi_readwrite(int cmd, struct request *req);

 static void ps2esdi_fill_cmd_block(u_short * cmd_blk, u_short cmd,
 u_short cyl, u_short head, u_short sector, u_short length, u_char drive);



cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux v2.5.45

[Jochen Friedrich]

Hi,

On Thu, 31 Oct 2002, Jochen Friedrich wrote:

> 2.4.45 compile on Alpha fails:

the old patch from Ivan Kokshaysky <ink@jurassic.park.msu.ru> fixed this,
but then i hit the next one:

  gcc -Wp,-MD,drivers/char/.eventpoll.o.d -D__KERNEL__ -Iinclude -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mno-fp-regs -ffixed-8 -mcpu=ev5
-Wa,-mev6 -nostdinc -iwithprefix include    -DKBUILD_BASENAME=eventpoll
-DEXPORT_SYMTAB  -c -o drivers/char/eventpoll.o drivers/char/eventpoll.c
drivers/char/eventpoll.c:226: warning: initialization from incompatible
pointer type
drivers/char/eventpoll.c: In function `write_eventpoll':
drivers/char/eventpoll.c:993: `POLLREMOVE' undeclared (first use in this
function)
drivers/char/eventpoll.c:993: (Each undeclared identifier is reported only
once
drivers/char/eventpoll.c:993: for each function it appears in.)
drivers/char/eventpoll.c: In function `ep_poll':
drivers/char/eventpoll.c:1056: warning: comparison is always false due to
limited range of data type
make[2]: *** [drivers/char/eventpoll.o] Error 1
make[1]: *** [drivers/char] Error 2
make: *** [drivers] Error 2

--jochen

Re: Linux v2.5.45 ipmr.c compile failure

[Helge Hafting]

net/ipv4/ipmr.c: In function `ipmr_forward_finish':
net/ipv4/ipmr.c:1114: structure has no member named `pmtu'
net/ipv4/ipmr.c: In function `ipmr_queue_xmit':
net/ipv4/ipmr.c:1170: structure has no member named `pmtu'

Helge Hafting

Re: Linux v2.5.45

[Adrian Bunk]

On Wed, 30 Oct 2002, Linus Torvalds wrote:

>...
> Summary of changes from v2.5.44 to v2.5.45
> ============================================
>...
> Greg Kroah-Hartman <greg@kroah.com>:
>...
>   o USB: drivers/net/irda fixups due to USB structure changes
>...

<--  snip  -->

...
  gcc -Wp,-MD,drivers/net/irda/.irda-usb.o.d -D__KERNEL__ -Iinclude -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
-march=k6 -Iarch/i386/mach-generic -nostdinc -iwithprefix include
-DKBUILD_BASENAME=irda_usb   -c -o drivers/net/irda/irda-usb.o
drivers/net/irda/irda-usb.c
drivers/net/irda/irda-usb.c: In function `irda_usb_probe':
drivers/net/irda/irda-usb.c:1490: structure has no member named `bInterfaceNumber'
make[3]: *** [drivers/net/irda/irda-usb.o] Error 1

<--  snip  -->

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux v2.5.45

[Adrian Bunk]

Hi Greg,

please ignore this mail, I found the patch that was already sent to l-k...

cu
Adrian


On Fri, 1 Nov 2002, Adrian Bunk wrote:

> On Wed, 30 Oct 2002, Linus Torvalds wrote:
>
> >...
> > Summary of changes from v2.5.44 to v2.5.45
> > ============================================
> >...
> > Greg Kroah-Hartman <greg@kroah.com>:
> >...
> >   o USB: drivers/net/irda fixups due to USB structure changes
> >...
>
> <--  snip  -->
>
> ...
>   gcc -Wp,-MD,drivers/net/irda/.irda-usb.o.d -D__KERNEL__ -Iinclude -Wall
> -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
> -fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
> -march=k6 -Iarch/i386/mach-generic -nostdinc -iwithprefix include
> -DKBUILD_BASENAME=irda_usb   -c -o drivers/net/irda/irda-usb.o
> drivers/net/irda/irda-usb.c
> drivers/net/irda/irda-usb.c: In function `irda_usb_probe':
> drivers/net/irda/irda-usb.c:1490: structure has no member named `bInterfaceNumber'
> make[3]: *** [drivers/net/irda/irda-usb.o] Error 1
>
> <--  snip  -->
>
> cu
> Adrian

Re: Linux v2.5.45 ipmr.c compile failure

[Randy.Dunlap]

| net/ipv4/ipmr.c: In function `ipmr_forward_finish':
| net/ipv4/ipmr.c:1114: structure has no member named `pmtu'
| net/ipv4/ipmr.c: In function `ipmr_queue_xmit':
| net/ipv4/ipmr.c:1170: structure has no member named `pmtu'

Several people reported this problem in 2.5.45.

I don't know how tough it is to check all combinations of config
options for such build problems, and I don't know how much code
and/or options there are in linux/net/ compared to linux/drivers/usb/,
but when I was the USB maintainer, before sending patches to Linus,
I would	use a script to build all of USB in-kernel and then build all
of USB as loadable modules.  This helped me make sure that easy-to-catch
problems didn't happen.

just an idea...
-- 
~Randy
(OK, net/ is about double the size of usb/)

various 2.5.45 compile problems (intermezzo, pmtu, afs, qt) (was: Linux v2.5.45 ipmr.c compile failure)

[Bernd Eckenfels]

In article <Pine.LNX.4.33L2.0211011525300.28320-100000@dragon.pdx.osdl.net> you wrote:
> | net/ipv4/ipmr.c: In function `ipmr_forward_finish':
> Several people reported this problem in 2.5.45.

and this also:

  gcc -Wp,-MD,net/ipv4/netfilter/.ipt_TCPMSS.o.d -D__KERNEL__ -Iinclude
-Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fomit-frame-pointer
-fno-strict-aliasing -fno-common -pipe -mpreferred-stack-boundary=2
-march=i686 -Iarch/i386/mach-generic
-nostdinc -iwithprefix include -DMODULE   -DKBUILD_BASENAME=ipt_TCPMSS   -c
-o net/ipv4/netfilter/ipt_TCPMSS.o net/ipv4/netfilter/ipt_TCPMSS.c
net/ipv4/netfilter/ipt_TCPMSS.c: In function 	pt_tcpmss_target':
net/ipv4/netfilter/ipt_TCPMSS.c:88: structure has no member named \x10mtu'
net/ipv4/netfilter/ipt_TCPMSS.c:91: structure has no member named \x10mtu'
net/ipv4/netfilter/ipt_TCPMSS.c:95: structure has no member named \x10mtu'
make[3]: *** [net/ipv4/netfilter/ipt_TCPMSS.o] Error 1

and diskmapper:

drivers/md/dm-ioctl.c: In function 'create':
drivers/md/dm-ioctl.c:588: incompatible type for argument 1 of 'set_device_ro'
drivers/md/dm-ioctl.c: In function \x12eload':
drivers/md/dm-ioctl.c:874: incompatible type for argument 1 of 'set_device_ro'
make[1]: *** [drivers/md/dm-ioctl.o] Error 1

also intermezzo and afs do not build as a module (maybe even build in)

fs/afs/dir.c:75: warning: unnamed struct/union that defines no instances
fs/afs/dir.c: In function \x01fs_dir_iterate_block':
fs/afs/dir.c:261: union has no member named \x0eame'
fs/afs/dir.c:293: union has no member named \x0eame'
fs/afs/dir.c:296: union has no member named \x16node'
fs/afs/dir.c:296: union has no member named \x16node'
fs/afs/dir.c:296: union has no member named \x16node'
fs/afs/dir.c:297: union has no member named \x15nique'
make[2]: *** [fs/afs/dir.o] Error 1

include/linux/intermezzo_fs.h: In function 	zo_ioctl_packlen':
include/linux/intermezzo_fs.h:864: sizeof applied to an incomplete type
include/linux/intermezzo_fs.h:865: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:866: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h: At top level:
include/linux/intermezzo_fs.h:870: warning: \x13truct izo_ioctl_data' declared
inside parameter list
include/linux/intermezzo_fs.h:871: conflicting types for
zo_ioctl_is_invalid'
include/linux/intermezzo_fs.h:52: previous declaration of
zo_ioctl_is_invalid'
include/linux/intermezzo_fs.h: In function 	zo_ioctl_is_invalid':
include/linux/intermezzo_fs.h:872: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:876: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:880: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:884: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:884: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:888: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:888: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:892: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:892: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:896: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:896: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:900: warning: passing arg 1 of
zo_ioctl_packlen'
from incompatible pointer type
include/linux/intermezzo_fs.h:900: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:904: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:905: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:905: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:909: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:910: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:910: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h:910: dereferencing pointer to incomplete type
include/linux/intermezzo_fs.h: At top level:
include/linux/intermezzo_fs.h:919: warning: \x13truct kml_rec' declared inside
parameter list
include/linux/intermezzo_fs.h:920: warning: \x13truct kml_rec' declared inside
parameter list
make[2]: *** [fs/intermezzo/cache.o] Error 1

In addtion to that make oldconfig depends on the qtlib (i was not realy able
to figure the dependencies out in that makefile but it is enough to touch
the temporary qt found flagfile to run make oldconfig.

Gruss
Bernd