Linux Today: Linux News On Internet Time.
Search Linux Today
Linux News Sections:  Developer -  High Performance -  Infrastructure -  IT Management -  Security -  Storage -
Linux Today Navigation
LT Home
Link to Us
Linux Jobs

Top White Papers

More on LinuxToday

Linus Torvalds: Linux 2.5.15

May 10, 2002, 00:17 (4 Talkback[s]) Mirrors

Summary of changes from v2.5.14 to v2.5.15

<> (02/04/15 1.447.14.6)
        Move arch/i386/kernel/pci-*.c to arch/i386/kernel/pci/; prepare for further cleanups

<> (02/04/16 1.447.14.7)
        Further split/cleanup of x86 PCI code

<> (02/04/16 1.447.14.8)
        Further break-up-age
        Make ACPI PCI IRQ routing work a bit better...

<> (02/04/16 1.447.14.9)
        Split drivers/pci/pci.c into smaller, bite-size chunks

<> (02/04/16 1.447.14.10)
        don't enable debug in arch/i386/kernel/pci/

<> (02/04/16 1.447.26.3)
        Fix NUMA compile after PCI cleanup

<> (02/04/26 1.447.52.1)
        Move the IPv6 symbols into the ipv6 module
        Currently, we're trying to export the IPv6 modules from netsyms.c,
        which can't work when netsyms is built-in, but IPv6 is modular. Fix is to
        move the IPv6 symbols to the ipv6 module. Actually, netsyms.c could use a
        lot more of this kind of untangling.
        Also, move exporting the secure_*v6* symbols to drivers/char/random.c, 
        since that's where they're defined.
        The reason why things worked before is that the ipv6 module doesn't have
        a EXPORT_NO_SYMBOLS directive, so _all_ of its symbols where exported by

<trini@bill-the-cat.bloom.county> (02/04/30 1.447.66.1)
        PPC32: Minor cleanups to the i8259 and OpenPIC code.  From myself and
        Hollis Blanchard.

<trini@bill-the-cat.bloom.county> (02/04/30 1.447.66.2)
        PPC32: Modify the OpenPIC code to allow for all combinations of sense
        and polarity of IRQs.  Update the machines in CONFIG_ALL_PPC for this change.

<trini@bill-the-cat.bloom.county> (02/04/30 1.447.66.3)
        PPC32: Update Motorola LoPEC support to match recent changes.

<trini@bill-the-cat.bloom.county> (02/05/01 1.447.66.4)
        PPC32: Modify openpic_get_irq() to call i8259_irq() for cascaded IRQs.  This
        requires that i8259_init() be called with the real address where PCI
        interrupt-acknowledge transactions will be generated.

<trini@bill-the-cat.bloom.county> (02/05/02 1.447.79.1)
        PPC32: define some important magic constants.

<trini@bill-the-cat.bloom.county> (02/05/02 1.447.79.2)
        Fix building of CONFIG_XMON && !CONFIG_MAGIC_SYSRQ

<> (02/05/03 1.447.80.1)
        This fixes a possible buffer overflow in hid-core.c in case a
        device would have very long string descriptors (vendor and device

<> (02/05/03 1.447.80.2)
        Fix for the previous fix. hid->name is 128 bytes, not 64 bytes long.

<> (02/05/03 1.447.77.2)
        - htons cleanups
        - remove duplicated code
        - apply CodingStyle

<> (02/05/04 1.513.6.1)
        drivers/net/Makefile cleanup
        Group selecting subdirs and linking them together. (Doesn't change
        the link order)

<> (02/05/04 1.513.6.2)
        Link drivers/net/* from drivers/net/Makefile
        Group knowledge about building objects and linking them together
        instead of splitting it between the top-level Makefile and
        the per-directory Makefiles.
        This change tries to keep the link order the same as it was, at least
        as far as possible. One critical point is that we now link
        the pcmcia / wireless netdrivers before the pcmcia subsystem.
        However, the pcmcia subsystem is initialized via a late_initcall(),
        so this shouldn't make a difference.
        drivers/net/Makefile looks a bit long now, but that can be improved
        a lot, patch yet to come.

<> (02/05/04 1.513.7.1)
        Documentation/kbuild/makefiles.txt polish
        Some improvements to the wording by Toshiyasu Morita.

<> (02/05/04 1.369.106.40)
        Sparc32 sun4c:
        - In sun4c_pmd_set, actually set the thing.
        - In ld_mmu_sun4c, pmd_set no longer BTFIXUPCALL_NOP.

<> (02/05/05 1.447.77.3)
        net/core/neighbour.c: Delete ancient ASSERT_WL debugging.

<> (02/05/05 1.513.1.37)
        Tigon3: Handle Netgear GA302T correctly.

<> (02/05/05 1.369.106.41)
        Sparc64 updates:
        - Implement changes for pfn_to_page et al.
        - Update defconfig.

<> (02/05/05 1.369.106.42)
        Sparc64: Implement pfn_pte.

<> (02/05/06 1.369.106.43)
        Sparc64: Fix thinko in page_to_pfn

<> (02/05/06 1.369.106.44)
        Sparc64: Update ioctl32 for bluetooth updates.

<> (02/05/06 1.513.10.1)
        NTFS: Fix compilation on 2.5.14 ({set_,clear_,}buffer_async was renamed
        to {set_,clear_,}buffer_async_read).

<> (02/05/06 1.513.4.3)
        [PATCH] USB printer freeing minors in probe error path
        fix a failure to free a minor in the error path of probe

<> (02/05/06 1.513.4.4)
        [PATCH] PATCH ehci -- interrupt xfer requeue
        The fix basically removes a bit-complement that shouldn't have
        been there.  (Plus related simplification.)  That changed the PID,
        so that (for one example) only the first hub plug/unplug event could
        work.  So it's essential for anyone using high speed hubs.
        I'm still not quite sure why this hasn't affected more testing, but
        that's life ... :)

<> (02/05/06 1.513.12.1)
        [PATCH] tcq initialization fix
        Transposed the last two arguments to memset, causing a slab poisoned
        kernel not to use tagging correctly... Brown paper bag stuff.

<> (02/05/06 1.513.12.2)
        [PATCH] 2.5.14 IDE 55
         - Update HPT374 driver carried over from 2.4.xx series by Andrew Morton.
           Resync it with the recent host chip driver changes, or better the
           introduction of an API at all.

         - Consolidate the handling of device ID byte order in one place.
           This was spotted and patched by Bartomiej onierkiewicz.
         - Eliminate CONFIG_BLK_DEV_IDEPCI - it's duplicating the functionality of the
           already present and fine CONFIG_PCI flag and if we are a PCI host, we are
           indeed very likely to need host chip support anyway.
         - Remove some redundant info about the model and channel number from
           /proc/ide. Remove the binary entries not helpful to the user, and not used
           by any program and redundant to corresponding ioctls.
         - Properly return udma_read and udma_write values in taskfile.
         - Only initialize XXX_udma to the default handlers if it has not been
           initialized by the host chip initialization.
        I have enabled spin lock debugging and can see that on device
        flush the spin locks get wrong counts... no problems elsewher ethus
        far. I will re check them next time around.

<> (02/05/06 1.513.10.3)
        [PATCH] small filemap.c/pagemap.h cleanups
         - remove page_cache_entry - I think it never ever was actually used.
         - remmove wake_up_page (as in 2.4) - never used
         - remove misleading comment ontop of pagemap.h
         - make wait_on_page* directly call wait_on_page_bit
         - some reordering and additional comments in pagemap.h

<> (02/05/06 1.369.106.45)
        soft-fp fix:
        - When converting from integer to float, rounding is done
        incorrectly due to off-by-one shift in conversion

<> (02/05/06 1.513.13.1)
        - remove spurious spaces and tabs at end of lines
        - make sure if, while, for, switch has a space before the opening '('
        - make sure no line has more than 80 chars
        - move initializations to the declaration line where possible
        - bitwise, logical and arithmetic operators have spaces before and after,
          improving readability of complex expressions
        - remove uneeded () in returns
        - other minor cleanups

<> (02/05/06 1.513.14.2)
        add back NCR53c810 PCI quirk code from davej that was lost in the merge

<> (02/05/06 1.513.14.3)
        hand merge of davej's x86 pci-pc janitor work

<> (02/05/06 1.513.17.1)
        ISDN: CAPI cleanup: register/release
        Introduce register_appl() / release_appl() which are basically equivalent 
        to open()/close() for net_device's.
        Remove the registered() / released() callbacks and do the work immediately

<> (02/05/06 1.513.14.4)
        PCI hotplug update 
        include file location changed due to pci changes

<> (02/05/06 1.513.17.2)
        ISDN: CAPI maintain driver MOD_USE_COUNT
        Apart from the indentation changes, add an owner field to the hardware
        driver struct and use that to automatically inc/dec the module use count
        on register_appl()/release_appl()

<> (02/05/06 1.513.17.3)
        ISDN: CAPI cleanup
        Remove unused header file.

<> (02/05/06 1.513.17.4)
        ISDN: CAPI: use <linux/list.h> lists

<> (02/05/06 1.513.17.5)
        ISDN: export callbacks directly
        Just EXPORT_SYMBOL() the callbacks, instead of passing them to the
        drivers in a struct.

<> (02/05/06 1.513.17.6)
        ISDN: CAPI: alloc struct capi_ctr dynamically
        First step in actually having the hardware drivers alloc this as part
        of their per-controller data.

<> (02/05/06 1.513.4.5)
        [PATCH] usage count handling during disconnect
        in usb.c during disconnect handling a reference to a driver is passed
        after the module usage count is decremented. In theory this is a race.

<> (02/05/06 1.513.11.3)
        USB pl2303 driver
        added support for another pl2303 device thanks to lutz rothhardt

<> (02/05/07 1.513.10.4)
        [PATCH] PATCH - kNFSd - Remove flowcontrol problems with lockd/tcp
        The sunrpc/tcp layer currently allocates some resources linearly with
        number of server threads.  This causes problems when there are very
        small numbers of threads (e.g. lockd with one thread), so this patch
        adds a small constant to the number of threads before calculating
        resoures to allocate.
        This patch also provide proper upper-limits for the response sizes for
        lockd requests, so resources are not over-allocated.
        Together these resolve problems with heavy lockd load over tcp.

<> (02/05/07 1.513.10.5)
        [PATCH] PATCH - kNFSd in 2.5.14 - Two small changes to nfsd/nfssvc.c
        1/ Return correct error code if nfs server creation fails
        2/ Increase limit on number of nfsd threads
           There is no real justification for stopping a sysadmin
           from running lots and lots of nfsd threads, so we raise the
           imposed limit from 128 to 8192... maybe it should just go away...

<> (02/05/07 1.513.10.6)
        [PATCH] PATCH - kNFSd in 2.5.14 - Tidy up comments in expfs.c
        Chris Wright <>  suggested these to limit to 80
        columns and such.

<> (02/05/07 1.513.10.7)
        [PATCH] PATCH - kNFSd in 2.5.14 - Add a kernel_lock in d_splice_alias
        d_move wants the kernel to be locked, so
        d_splice_alias now takes that lock.

<> (02/05/07 1.513.10.8)
        [PATCH] PATCH - kNFSd/ext3 in 2.5.14 - export_operations support for ext3
        This patch converts ext3 to use the new export_operations for
        exporting a filesystem.
        In particular it
         - defines "ext3_get_parent" to do a lookup(".."),
         - calls d_splice_alias when ext3_lookup finds something,
           so that build a dcache path from the bottom up works, and
         - sets sb->s_export_op to point to an (almost empty)
           export_operations structure.  The fact that it is mostly empty
           means that the default lookup operations are used, which match the
           previous approach (i.e. use iget, and then check the generation

<> (02/05/07 1.513.10.9)
        [PATCH] PATHC - kNFSd/reiserfs in 2.5.14 - Convert reisferfs to use export_operations
        This patch converts reiserfs to use the new export_operations
        interface for exporting a filesystem instead of the old
        fh_to_dentry and dentry_to_fh.
        Essentially it:
          - renamed reiserfs_dentry_to_fh to reiserfs_encode_fh
          - split reiserfs_fh_to_dentry into reiserfs_get_dentry
            and reiserfs_decode_fh
          - calls d_splice_alias from reiserfs_lookup so that building
            a dcache path from the bottom works well
          - addes reiserfs_get_parent to do a lookup of ".." without going
            through the VFS interfaces.
          - sets sb->s_export_op to be an appropriately initialised
            struct export_operations
        This has been tested and works.

<> (02/05/07 1.513.10.10)
        [PATCH] PATCH - kNFSd/FAT in 2.5.14 - Add export_operations support for FAT
        This actually contains a lot of code that is never used,
        as we don't currently attempt to load an inode that isn't
        in cache, so the fat_get_parent is never actually reached.
        However it is there (and should be right) incase some brave
        soul does decide to enhance fat_get_dentry accordingly.

<> (02/05/07 1.513.10.11)
        [PATCH] 2.5.14 IDE 56

         - Push poll_timeout down from the hwgroup to the channel. We are resetting the
           channel and not a whole hwgroup. This way using multiple pdc202xxx cards
           should magically start to work with multiple performance and resets will no
           longer lock the system.
         - Updates for PDC4030 by Peter Denison <>.
         - Make ide_raw_taskfile don't care about request buffers. They where always
         - Port set multi mode count over from the special setting interface to
           ide_raw_taskfile. Fix errors in the corresponding interrupt handler in one go
           as well. It turned out that this is precisely the same code as in
           task_no_data_intr, so we can nuke it altogether. And finally we have found
           some problems with the set_pio_mode() command which can fail with -EBUSY -
           this is in esp. probably *very* common during boot hdparm usage those days!
           (OK it was masked by reportig too early that it finished...  Crap Crap utter
           crap it was!!!) Right now hdparm should just be extendid to properly
           sync and retry on   -EBUSY and everything should be fine.
           And now the 1 Milion EUR question for everybody who loves to put driver
           settings in to /proc:
           How the hell could echo > /proc/ide/ide0/settings blah blah blah blah handle
           properly cases like -EIO, -EBUSY and so on??? Having the possibility o do it
           does not mean that it is a good idea to use it.
           OK. After realizing the simple fact that quite a lot of low level hardware
           manipulating ioctls may require assistance in usage from proper logic which is
           *very* unlikely to be implemented in a bash (for me preferable still ksh) I
           have made my mind up.
                /proc/ide will be nuked.
         - Execute the recalibration for error recovery on precisely the same request as
           the one which failed.
         - Remove set geometry.  It's crap by means of standard specification. Because:
           1. We rely on the existence of the identify command anyway.
           2. This command was obsoleted *before* the identify command existed as far
           as I can see.
           2. I'm able to have a look at what other ATA/ATAPI drivers in the wild do:
           They don't do it.
         - Just call tuneproc in set_pio_mode() directly - we are already behind the rq
           queue there.
         - After we have uncovered the broken logics behind the whole ioctl handling we
           now just have made ide_spin_wait_hwgroup() waiting for a proper somehow
           longer timeout before giving up. This was previously just hiding the broken
           concept of setting ioctl values through /proc/ide/ideX/settings - now it just
           really helps hdparm to not to give up too early. (It shouldn't probably play
           wreck havock on the global driver spin lock as well. I will look in to this
         - Scrap the non necessary, to say the least, disabling of interrupts for 3,
           read it again please, 3 seconds, on the local CPU inside
           ide_spin_wait_hwgroup().  Spin lock handling needs checking there badly as I
           see now as well...
        Hey apparently any "special" requests are gone. We now have only
        to deal with REQ_DEVICE_ACB and REQ_DEVICE_CMD. One of them is still too
        much and will be killed.

<> (02/05/07 1.513.10.12)
        [PATCH] 2.5.14 IDE 57
        Nuke /proc/ide. For explanations why, please see the frustrated comments in the
        previous change log. If one still don't see why it wasn't a good thing,
        well please just take a look at the following:
        Kernel size before:
            text    data     bss     dec     hex filename
        1716049  403968  470252 2590269  27863d vmlinux
        Kernel size after:
            text    data     bss     dec     hex filename
        1680993  403488  470124 2554605  26faed vmlinux
        2% of overall size! And this is not exactly an minimalistic setup.

<> (02/05/07 1.513.10.13)
        [PATCH] IDE 58
         - m68k fixes by Roman Zippel.
         - CDROM PIO mode fix by Osamu Tamita.
           (You are true "Hawk-eye" hovering over my head! Respect - and many Thanks.)
         - Virtualize the udma_enable method as well to help ARM and PPC people.  Please
           please if you would like to have some other methods virtualized in a similar
           way - just tell me or even better do it yourself at the end of ide-dma.c.
           I *don't mind* patches.
         - Fix the pmac code to adhere to the new API. It's supposed to work again.
           However this is blind coding... I give myself 80% chances for it to work ;-).

<> (02/05/07 1.513.10.14)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.15)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.16)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.17)
        [PATCH] capabilities for mtrr driver.

<> (02/05/07 1.513.10.18)
        [PATCH] DMI update.
        Another Sony for the apm blacklist.

<> (02/05/07 1.513.10.19)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.20)
        [PATCH] motioneye driver update.
        Syncs up with Stelians latest

<> (02/05/07 1.513.10.21)
        [PATCH] update NBD URL

<> (02/05/07 1.513.10.22)
        [PATCH] check various return codes in advantechwdt.
        Adds missing checks. Probably another from William Stinson

<> (02/05/07 1.513.10.23)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.24)
        [PATCH] IA64 ZX1 AGPGart support.
        From 2.4

<> (02/05/07 1.513.10.25)
        [PATCH] CREDITS update

<> (02/05/07 1.513.10.26)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.27)
        [PATCH] typo.

<> (02/05/07 1.513.10.28)
        [PATCH] VIA quirk update.
        Brings up to date with the quirk in 2.4
        This makes sure we only apply it on the right chipsets,
        and adjusts the bits that get poked depending on which we find.

<> (02/05/07 1.513.10.29)
        [PATCH] sonypi update
        Syncs up with latest from Stelian

<> (02/05/07 1.513.10.30)
        [PATCH] prefetching too far in mem copies
        This patch from 2.4 makes sure we don't prefetch past the
        end of a range to be copied (in case its at the end of a memrange)
        i386 case looks safe already, we just weren't optimal for the last
        chunk to be copied.
        Andi. same change needed for x86-64.

<> (02/05/07 1.513.10.31)
        [PATCH] region handling cleanup
        Done by William Stinson.
        Adds error handling to request_region() calls,
        and converts some old check_region() calls too.

<> (02/05/07 1.513.10.32)
        [PATCH] documentation for Cyrix IRQ routing

<> (02/05/07 1.513.10.33)
        [PATCH] simplify wdt285 copying.
        copy_to_user already does the relevant checking.

<> (02/05/07 1.513.10.34)
        [PATCH] typo in z2ram driver

<> (02/05/07 1.513.10.35)
        [PATCH] PNPBIOS / floppy conflict.
        The PNPBIOS driver and the floppy driver both fight over reservation
        of the floppy port. This patch makes things friendly again.
        As the comment suggests, this fix could be done better, but until
        someone steps forward to fix this, this is the best we have.

<> (02/05/07 1.513.10.36)
        [PATCH] NCR5380 update.
        From 2.4
        - Lindented
        - Add Docbook comments
        - Remove duplicate MODULE_LICENSE tag

<> (02/05/07 1.513.10.37)
        [PATCH] dmasound update.
        The largest part of the OSS update from 2.4
        Various changelog entries scattered throughout the patch.

<> (02/05/07 1.513.10.39)
        Re-apply Dave's VIA quirk update from 2.4.x
        after the manual merge of the PCI re-organization
        removed it.

<> (02/05/07 1.513.18.1)
        [PATCH] Fix missing #includes
        There are missing #includes which will break compilation on some non-x86
        platforms. With following patch this compiles and works on alpha.

<> (02/05/07 1.513.10.40)
        [PATCH] aha1542 update
        This ISA scsi adapter still seems to be popular
        given the number of people that supply patches
        for it collected in Dave's tree.
        The latest patch was from William Stinson (first item):
          - request_region cleanup
          - removal of old scsi error handling (leaving newer version)
          - scatterlist::address replaced
          - host_lock replacing io_request_lock
        Doug Gilbert
        P.S. This is a transfer from Dave's tree. It runs fine for
        me on a dual Celeron SMP box.

<> (02/05/07 1.513.10.41)
        [PATCH] osst update
        This is the OnStream variant of the st driver that
        has been in Dave's tree for some time.
        The change looks like it is from Willem Riede (
        and that it has been forwarded ported from lk 2.4 .

<> (02/05/07 1.513.10.42)
        [PATCH] sg update
        Here is one fix and one enhancement to the scsi generic (sg)
        driver in 2.5.14
            - off by one fix for last scatter gather element
            - if possible compact kiobuf_map into scatter gather list

<> (02/05/07 1.513.17.7)
        ISDN: CAPI: alloc struct capi_appl dynamically
        First step in having the users alloc this as part of the state info
        they have to maintain anyway.

<> (02/05/07 1.513.10.43)
        [PATCH] m68k: AFFS update [1/20]
        - sizeof changes from Kernel Janitor Project
        - several bug fixes found with fsx

<> (02/05/07 1.513.10.44)
        [PATCH] m68k: atari updates [2/20]
        Some compile fixes.

<> (02/05/07 1.513.10.45)
        [PATCH] m68k: bitops update [3/20]
        Add __clear_bit/__ffs/sched_find_first_bit.

<> (02/05/07 1.513.10.46)
        [PATCH] m68k: add cacheflush.h [4/20]
        This moves several cache flush function into cacheflush.h.

<> (02/05/07 1.513.14.6)
        Move arch/i386/kernel/pci/ to arch/i386/pci/

<> (02/05/07 1.513.10.47)
        [PATCH] m68k: update [5/20]
        Update to the latest

<> (02/05/07 1.513.10.48)
        [PATCH] m68k: license updates [7/20]
        Add missing license tags.

<> (02/05/07 1.513.10.49)
        [PATCH] m68k: hp300 update [8/20]
        compile fix

<> (02/05/07 1.513.10.50)
        [PATCH] m68k: remove hwclk_time/gettod [9/20]
        - replace hwclk_time with rtc_time
        - use hwclk instead of gettod to set initial time

<> (02/05/07 1.513.10.51)
        [PATCH] m68k: idle update [10/20]
        Fix idle functions.

<> (02/05/07 1.513.10.52)
        [PATCH] m68k: add missing include [11/20]
        Add missing include files.

<> (02/05/07 1.513.10.53)
        [PATCH] m68k: Rename KTHREAD_SIZE [12/20]
        This patch renames KTHREAD_SIZE into THREAD_SIZE, similiar to all other archs.

<> (02/05/07 1.513.10.54)
        [PATCH] m68k: math-emu updates [13/20]
        Compile fixes

<> (02/05/07 1.513.10.55)
        [PATCH] m68k: rename p_pptr [14/20]
        Rename p_pptr into parent.

<> (02/05/07 1.513.10.56)
        [PATCH] m68k: readd lost proc functions [15/20]
        These function were removed by a "cleanup", but they are still used.

<> (02/05/07 1.513.10.57)
        [PATCH] m68k: rlimits update [16/20]
        Update INIT_RLIMITS.

<> (02/05/07 1.513.10.58)
        [PATCH] m68k: seq_file fixes [17/20]
        update some proc functions to use the seq_file interface

<> (02/05/07 1.513.10.59)
        [PATCH] m68k: add tlbflush.h [18/20]
        Move several tlb functions into tlbflush.h.

<> (02/05/07 1.513.10.60)
        [PATCH] m68k: add show_trace/show_trace_task [19/20]
        extract show_trace()/show_trace_task() out of dump_stack().

<> (02/05/07 1.513.10.61)
        [PATCH] m68k: zorro updates [20/20]
        - small compile fixes
        - id update
        - add asm-ppc/zorro.h

<> (02/05/07 1.513.10.62)
        [PATCH] m68k: driver updates [6/20]
        - kdev_t fixes
        - bio fixes
        - other cleanups and compile fixes

<> (02/05/07 1.513.18.2)
        Move arch/i386/kernel/pci/ to arch/i386/pci/

<> (02/05/07 1.513.10.63)
        [PATCH] NLS: Invalid koi8-ru return values
        During my lurking around NLS code I found that KOI8-RU returns character
        code instead of length of character for two characters.
                                                Petr Vandrovec

<> (02/05/07 1.513.10.64)
        [PATCH] NLS: Allow user to select 1:1 mapping
        This allows user to select 'default' encoding, which has defined mapping
        for each of 255 byte values, so one can use it as a fallback when it
        finds filesystem with unknown encoding (EBCDIC for example) just to get
        characters through filesystem.
                                        Petr Vandrovec

<> (02/05/07 1.513.19.1)
        PCI Update: Fix oops on boot w/ ACPI enabled

<> (02/05/07 1.513.10.65)
        [PATCH] 2.5.14: New PCI allocation code (alpha, arm, parisc) [1/2]
        This changes PCI resource allocation algorithm to 3 passes vs.
        current 2 passes. Extra pass is used for calculation of required
        size and alignment of PCI buses behind PCI-PCI bridges. After
        that, in the pass #3, these buses get allocated like regular
        PCI devices. This gives tighter PCI IO and memory packing -
        for instance, this fixes allocation problems on certain alphas
        with very small (112Mb) PCI memory range. Also, the new code
        - will allow mixed approach to resource allocation:
          architecture can keep BIOS settings for some devices,
          and re-allocate resources for others, including improperly
          initialized bridges;
        - makes prefetchable ranges support much simpler;
        - allows sizing of IO and memory ranges for the host
          bridges, which might be very useful in some situations.
        It was tested on various alphas; I haven't heard any complaints
        from rmk and rth, so probably all of this is ok. :-)
        Part 1:
        - for all archs, 4th argument (align) added to
          pcibios_align_resource (and its callers).
          It's necessary because this function will be called for
          bus resources as well, and in this case size != alignment.
        - for several archs, dead/bogus code removed from

<> (02/05/07 1.513.10.66)
        [PATCH] 2.5.14: New PCI allocation code (alpha, arm, parisc) [2/2]
        Summary of changes:
        - alpha, arm: code related to PCI-PCI bridges from pcibios_fixup_bus()
          removed - now it's generic;
        - pdev_sort_resource: sort resources all together, no matter IO or memory;
        - pbus_assign_resources_sorted: ditto;
        - pci_bridge_check_ranges, pci_setup_bridge: changed for prefetch support;
        - pbus_size_io, pbus_size_mem: core stuff; tested with randomly generated
          sets of resources;
        - pbus_size_bridges: pass #2 (pass #1 is PCI probing, common for all archs);
        - pbus_assign_resources: pass #3.

<> (02/05/07 1.513.10.67)
        [PATCH] fixes for PCI reorg changes
        There are missing #includes which will break compilation on some non-x86
        platforms. With following patch this compiles and works on alpha.

<> (02/05/07 1.513.10.68)
        [PATCH] USE_STANDARD_AS_RULE in i386 arch Makefiles
        In Rules.make, a comment says:
        # Old makefiles define their own rules for compiling .S files,
        # but these standard rules are available for any Makefile that
        # wants to use them.  Our plan is to incrementally convert all
        # the Makefiles to these standard rules.  -- rmk, mec
        This patch does that for the i386 arch Makefiles.

<> (02/05/07 1.513.10.69)
        [PATCH] suser() -> capable() checks
          Trivial patch to change some instances of suser() and fsuser() to
        proper capable() checks.

<> (02/05/07 1.513.10.70)
        [PATCH] Fix: writing to the swap space
        Ah.  That's some left-over code.  Reads will be OK, but writes will be
        unexpectedly asynchronous.  Nothing in the kernel uses that function for
        writes so it didn't show up.

<> (02/05/07 1.513.10.72)
        Update kernel version

<> (02/05/07 1.513.11.4)
        [PATCH] USB usbnet driver update
            - generalizes/cleans keventd support to also handle
                * rx stalls (and usb 2.0 transaction translator unplug)
                * rx memory shortfalls (latent bug Oliver noticed)
                * cleanup on device disconnect (quiesce first)
            - merges Brad's patch to use the IEEE802 "locally assigned" bit
            - fixes a couple minor bugs on error paths (leak, bogus diagnostic)
            - updates some comments

<> (02/05/07 1.513.14.9)
        PCI hotplug drivers
        change due to moved location of i386's pci.h

<> (02/05/07 1.513.17.8)
        ISDN: CAPI remove NCCI up/down notification
        Applications really ought to take care of maintaing the state themselves,
        instead of cheating by listening to controller private messages.

<> (02/05/07 1.513.17.9)
        ISDN: CAPI fix new "ncci up"
        There's actually two cases when a new NCCI is created:
        outgoing and incoming connections (had missed the latter)

<> (02/05/07 1.513.11.5)
        USB and Makefile minor changes
        sync up with -dj tree

<> (02/05/07 1.513.11.6)
        USB ohci driver update
        add PMAC changes found in -dj tree

<> (02/05/07 1.513.11.7)
        minor -dj tree updates.

<> (02/05/07 1.513.11.8)
        [PATCH] PATCH 2.5.14 -- ehci misc fixes
            - Report better errors, and in one case the correct one.
            - Adds strategic wmb() calls
            - Claims the right (scaled) ISO bandwidth
            - Uses non-CVS version ID
        This will likely resolve that Archos MP3 Jukebox issue, where "-104"
        (-ECONNRESET) was appearing mysteriously.

<> (02/05/07 1.513.14.10)
        Re-add check for valid acpi routing information, instead of assuming it works

<> (02/05/07 1.369.106.46)
        Sparc64: Fix pfn_foo macros and add descriptive commentary
        above explaining the {phys,pfn}_base stuff.

<> (02/05/07 1.513.20.1)
        Netfilter ipt_ULOG fix:
        - If timers were pending during module unload, we would OOPS.

<> (02/05/07 1.513.21.1)
        copy_mm fix:
        - If dup_mmap fails we will try to destroy_context before
        init_new_context occurs.  Platforms with non-trivial
        init_new_context can explode because of this.  The fix
        is to invoke init_new_context before dup_mmap.

<> (02/05/07 1.513.20.2)
        Tigon3 fix:
        - Fix typo in setting MR_LP_ADV duplex flags.

<> (02/05/08 1.513.15.3)
        Link drivers/char/* from drivers/char/Makefile
        Link the subdirs of drivers/char from drivers/char/Makefile instead the
        top-level Makefile. 
        Link order changes slightly, shouldn't case any problems, though.
        Fix drivers/char/pcmcia/
        Cleanup drivers/char/pcmcia/Makefile.
        Fix rio build rules in drivers/char/Makefile.

<> (02/05/08 1.513.15.4)
        Fix compile time warning in serial.c

<> (02/05/08 1.513.15.5)
        Cleanup drivers/fusion/* build
        Don't shortcut from drivers/Makefile to 
        drivers/message/{fusion,i2o}/Makefile, but use intermediate
        Cleanup drivers/message/fusion/ and get rid of unnecessary

<> (02/05/08 1.513.15.6)
        Remove PCMCIA backwards compatibility rule.
        Old pcmcia-cs versions expect the pcmcia modules in 
        /lib/modules/`uname -r`/pcmcia, so we symlinked them for compatibility.
        According to the comment, the generation of these links should have
        been removed in 2.4, so it's definitely time for it to go now.

<> (02/05/08 1.513.15.7)
        Cleanup drivers/pcmcia/Makefile
        Use the usual list-based approach. Still not too nice.

<> (02/05/08 1.513.15.8)
        Cleanup drivers/pcmcia/Makefile part 2
        Adding some more flexibility to (Allow for core built-in but
        socket driver modular) cleans up the Makefile as well ;-)

<> (02/05/08 1.513.15.9)
        Fix cut'n'paste error in drivers/message/Makefile

<> (02/05/08 1.513.15.10)
        Fix the Makefile cleanups

<> (02/05/08 1.513.1.40)
        Add SIGURG to list of ignore-by-default signals

<> (02/05/08 1.513.1.41)
        Fix makefile errors

<> (02/05/09 1.516)
        Update the for ppc32.  References to 486s and pentiums aren't
        really helpful here. :)

<> (02/05/09 1.517)
        PPC32: Rename ppc_defs.h to asm-offsets.h to ease the transition to kbuild-2.5 later.

<> (02/05/08 1.369.106.47)
        Soft-fp fix:
        - Fix handling of implicit 1 bit of fraction part when converting
        from int to float.  We cannot handle it properly unless we
        defer implicit 1 bit handling till after rounding as
        rounding may move where the implicit 1 bit is in
        the final converted integer.

<> (02/05/08 1.369.106.48)
        Sparc64: Make use of USE_STANDARD_AS_RULE.

<> (02/05/08 1.369.106.49)
        Sparc64: Enter KERNEL_DS when invoking do_sigaltstack with
        kernel pointers.

<> (02/05/08 1.513.21.2)
        SLAB: When using get_user on kernel pointers, enter KERNEL_DS.

<> (02/05/09 1.519)
        PPC32: This changeset adds preemptible kernel support for ppc32
        and also streamlines the exception entry/exit code by not saving
        all the GPRs on the common exceptions (system call, external
        interrupt and decrementer).

<> (02/05/09 1.520)
        Define pfn_to_page, page_to_pfn, pte_pfn, pfn_pte for ppc32.

<> (02/05/09 1.521)
        PPC32: minor cosmetic changes, eliminate compile warnings

<> (02/05/09 1.513.21.3)
        drivers/usb/net/rtl8150.c: Include linux/init.h

<> (02/05/09 1.522)
        PPC32: fix serial clock for embedded EP405 board.

<> (02/05/09 1.523)
        PPC32: remove some compiler warnings, xmon update

<> (02/05/09 1.524)
        PPC32: powermac update from Ben Herrenschmidt.
        This adds support for recent powermac models and fixes a few bugs.

<> (02/05/09 1.513.1.45)
        Sparc64: Make pcibios_init a subsys_initcall.

<> (02/05/09 1.513.1.46)
        Sparc64: Keep totalram_pages equal to num_physpages.

<> (02/05/09 1.513.20.3)
        IPV4: Add statistics for route cache GC monitoring.

<> (02/05/09 1.513.1.47)
        Sparc: Kill references to buffermem_pages.

<> (02/05/09 1.525)
        PPC32: add SIGURG to the list of signals ignored by default

<> (02/05/09 1.526)
        PPC32: add a subsys_initcall so pcibios_init gets called

<> (02/05/09 1.513.1.48)
        asm-sparc64/thread_info.h: Include asm/page.h even if __ASSEMBLY__

<> (02/05/09 1.513.22.1)
        [PATCH] Fix JFS file system corruption
        JFS: Flush dirty metadata to disk when remounting from read-write
        to read-only.  Also fix umount ordering to make sure metadata is
        written before journal closed.
        With Andrew Morton's recent changes, JFS is no longer writing much
        of its dirty metadata when remounting from read-write to read-only.
        This causes severe file system corruption.  A JFS root file system
        will be corrupted on shutdown.
        This patch fixes the problem by explicitly writing the dirty metadata
        before the journal is closed.  It also fixes the ordering so that
        the dirty metadata is completely written before the journal is closed
        for the normal unmount case as well.

<> (02/05/09 1.513.22.2)
        [PATCH] acenic driver and 867 MHz G4 sound problems
        Fix acenic driver

<> (02/05/09 1.513.23.1)
        moved the pci_alloc_consistent() and pci_free_consistent() functions back
        into arch/i386/kernel as they are needed even if CONFIG_PCI is not enabled.

<> (02/05/09 1.513.22.3)
        [PATCH] list macro conversion.
        Use listwalking macro..

<> (02/05/09 1.513.22.4)
        [PATCH] swap file missing error check
        Found by Andries Brouwer a while back iirc.

<> (02/05/09 1.513.22.5)
        [PATCH] ptrace bug
        We weren't incrementing the address when walking a processes
        address space.  From 2.4

<> (02/05/09 1.513.22.6)
        [PATCH] HP ZX1 AGP/DRM support.
        From 2.4

<> (02/05/09 1.513.22.8)
        [PATCH] fix drivers/scsi/sd.c for ppc32

        This patch fixes the compile errors in sd_find_target, which is only
        used on 32-bit PPC systems, and changes the ifdef around from
        CONFIG_PPC to CONFIG_PPC32 since ppc64 doesn't need this routine.

<> (02/05/09 1.513.22.9)
        [PATCH] fix drivers/pci/Makefile for PPC
        On 32-bit PPC we don't need setup-bus.o but we do need setup-irq.o.
        This patch changes drivers/pci/Makefile to reflect that.

<> (02/05/09 1.513.22.10)
        [PATCH] Directory Notification Fix
        This patch fixes directory notification so that notifications get
        cancelled when a process exits.  This make dnotify MUCH more stable and
        usable :-)

<> (02/05/09 1.513.24.1)
        IBM PCI Hotplug driver
        update the ibm pci hotplug driver to the latest version.  Contains lots of
        small bugfixes and added features for new hardware.

<> (02/05/09 1.513.24.2)
        PCI Hotplug core
        removed pcihpfs_statfs(), as it's not used anymore.

<> (02/05/09 1.528)
        [PATCH] 2.5.14 IDE 59
        Basically PCI driver handling reorganization. This is one step further
        ahead toward the goal of fully modularized host chip drivers.
         - Adjust ide-scsi to the new error handling. Just don't try any device
           resets there.
         - Add unmasking of IRQ per default to the PMac PCI code.
         - Split up the crap table from ide-pci. Let the corresponding drivers do
           registration of the functions they provide. This small change makes
           this patch rather big.
         - Hard-code the number of ports requested for DMA engines. They are always
           precisely 8 on PCs. If you hove something different to deal with,
           well then please just provide your own init_dma method.
         - Remove the HDIO_GETGEO_BIG ioctl. Patch by Andries Brouwer. Applies
         - Make ON_BOARD be equal 0, so we can spare ourself some typing in structure
         - Normalize the terminology in the host chip drivers. It will make spotting
           the tons of common code found there later easier.