development kernel tree: Changes to 'andy'

[git at git.openmoko.org]

Rebased ref, commits from common ancestor:
commit 80397f460d1cfdad9cc74e083164fd314c9c67ca
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:27:06 2008 +0100

    fix-glamo-crank-memory-to-90MHz.patch
    
    This changes Glamo memory and now host bus PLL to 90MHz from
    80MHz, as recommended by S-Media.  Bandwidth should go up
    by 12.5% by this in raw terms anyway.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit b7292e3909e86a15485ee31ee827f98afe0d496c
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:27:04 2008 +0100

    fix-glamo-turbo-host-interface.patch
    
    Until now, we forced Glamo wait states to 3 on read and write
    actions -- on the GLAMO side of the fence.  This reduces them
    to "0", although nWAIT is still briefly operated I saw.  This
    should reduce all memory bus timing by 3 Glamo Host bus clocks,
    60ns at 50MHz.
    
    In addition, until now we used PLL1 (50MHz) for host bus clock.
    Now, we use PLL2 (80MHz) for host bus clock, so all transactions
    complete quicker on Glamo side, it's a raw 60% improvement.
    
    These changes do not appear to impact correct operation of Glamo
    SD Card or video after testing for a week.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit eac79e9b466f1a24f89fc65d385d2f20037f8691
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:26:27 2008 +0100

    fix-reote-install-for-ext3-only-sd.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 0cc74fbc0f37e37f70e43d8c780edeff2678372b
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:26:27 2008 +0100

    defconfig-nuke-cruft.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 6d7de9e407239cbad0a3a9184e6bcd46e21a527a
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:26:27 2008 +0100

    defconfig-kill-hxd8.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit f2c5feee0a7eefcf8c247a943ba4038962d20256
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:26:27 2008 +0100

    fix-touchscreen-driver-gta01-missing-includes.patch
    
        Add missing initialization for the touchscreen driver for the
        gta01 platform.
    
        Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 952009f92410fbcb6ddd7653ba35af1570867746
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:26:26 2008 +0100

    test-touchscreen-median.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 78621054e25f8dd7700288816b9cda40dbb10b76
Author: Andy Green <andy at openmoko.com>
Date:   Fri Aug 8 13:26:26 2008 +0100

    config-touchscreen-filters.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 2d68fa06d5cfcf38ce9bd551b478dab408141cce
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:25:02 2008 +0100

    gta01-uart-fifo-trigger-sooner.patch
    
        Set the UART FIFO to trigger earlier on the GTA01 device to minimize
        UART overruns from the GSM.
    
        Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 9ee405d79609106ea330b21d54f30b853e262296
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:10:59 2008 +0100

    fix-suspend-backlight-timing-gta01.patch
    
        This patch adds the gta01 backlight callback that defers the
        restoring of the backlight until after the jbt driver has
        resumed.  This doesn't eliminate the flashing of the LCD on
        the gta01, but it reduces it considerably.
    
    Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 0c136b451381e760c056a7683b7fc28901f54da4
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:10:59 2008 +0100

    gta01-pcf50606-disable-irq-from-suspend-until-resume.patch
    
        This patch is the pcf50606 equivalent of the pcf50633 patch that
        disables interrupts from the chip until after resume is complete.
        In order to ensure no data is lost, the work function is called
        post-resume to process any pending interrupts.
    
        Most of the code was quite literally re-used from Andy Green's
        original patch.
    
    Signed-off-by: Mike Westerhof <mwester at dls.net>

commit bea12b6beb76a483145b07d30602b0dd18ff2484
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:10:18 2008 +0100

    silence-serial-console-gta01.patch
    
        This patch ensures that no console data will go the UART while
        the GSM mux is switched to the GSM.  This is necessary despite
        the code that disables the console due to the fact that the
        console resumes before the neo1973_pm_gsm driver, and consoles
        always resume in the "on" state.
    
    Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 749594928cc1a0e452c4c791fcbea2f8f4a7e259
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:08:03 2008 +0100

    commit 830ea3d0c27c0c750b7bf1b56c002ee7943f3edc
    
        gta0x-log-serial-rx-error.patch
    
        This patch causes a KERN_DEBUG message to be printed each time an
        error status is read from a UART.  This is intended to facilitate
        the reporting of more useful problem and bug reports from users
        in the field.
    
        Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 7e5f528cb1448ac31a4e3a45c7c953ea95869d4d
Author: Mike Westerhof <mwester at dls.net>
Date:   Fri Aug 8 13:08:03 2008 +0100

    gta0x-add-minimal-GSM-flowcontrol.patch
    
        Add the basic GSM flowcontrol code.
    
    Signed-off-by: Mike Westerhof <mwester at dls.net>

commit 436204281bcd1fe5999ad6589ea7ab1b5360c352
Author: Andy Green <andy at openmoko.com>
Date:   Thu Aug 7 16:27:33 2008 +0100

    config-add-udf-support.patch
    
    Timo noticed that his compound USB device wasn't getting recognized properly,
    it was stuck on the CD-ROM class part of it.  This was accepted by the GTA02
    but no scsi device was spawned for it.  He figured out in the end that it is
    the lack of a UDF filesystem module that was the problem.
    
    This patch adds UDF and CDROM as a built-in on defconfig-2.6.24, for the
    modular configs it just adds it as a module.
    
    Reported-by: Timo Drick <timo at drick.de>
    Signed-off-by: Andy Green <andy at openmoko.com>

commit e8780c02d307380e9e08d1b85888a599d53078ac
Author: Andy Green <andy at openmoko.com>
Date:   Thu Aug 7 16:27:21 2008 +0100

    config-remove-evbug-module.patch
    
    Carsten mentioned that evbug module is getting modprobed in
    to ASU somehow and making havoc.  This patch stops us building
    it because there's no time to figure out why it gets modprobed
    in and it doesn't seem to do anything but spew syslog -- and we
    never needed it before now.
    
    Reported-by: Carsten Haitzler <raster at openmoko.org>
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 2f36cf3d6a324af5c43f9d6bfded95d7c1a9c532
Author: Holger Freyther <zecke at openmoko.org>
Date:   Thu Aug 7 16:27:12 2008 +0100

    From fa269b44512a03523b164c3cebc20312748c524b Mon Sep 17 00:00:00 2001
    Subject: [PATCH] [ar6k] Build the wireless driver without DEBUG
        - Remove DEBUG from the Makefile
    
        - Do not send events through netlink to userspace. We might
          need to reevaluate this. But we seem to use wireless_send_event
          at the right places.
          (SEND_EVENT_TO_APP)
    
        - Do not report debug logs to apps
          (REPORT_DEBUG_LOGS_TO_APP)
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit c596d9067aa789a62e51df89917680eabe6453e2
Author: Andy Green <andy at openmoko.com>
Date:   Thu Aug 7 12:20:37 2008 +0100

    add-build-makerecovery.patch
    
    This patch adds a script that creates a "recovery kernel" image
    from a recovery-uImage-moredrivers-... image.  The use of these
    recovery images is explained here:
    
    http://wiki.openmoko.org/wiki/Freerunner_backup_kernel
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 2c55337cb0cde08203df321a108409b7e33e6904
Author: Andy Green <andy at openmoko.com>
Date:   Wed Aug 6 23:13:28 2008 +0100

    fix-build-hash-name-no-colon-separator.patch
    Suggested-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006 at gmx.net>
    
    Colons are not very friendly for tar... scp has trouble too but
    you can override that by prepending any path like ./ ... still
    better to dilute the colon down to a _ avoiding - because it can
    appear in branch names we use.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit b79f79448a70b5718db2428939e00150315b5eda
Author: Andy Green <andy at openmoko.com>
Date:   Wed Aug 6 20:46:42 2008 +0100

    add-buildkeep-hash-named-copy.patch
    
    Keep a copy of last "moredrivers" config kernel built stamped with
    branch and git head at build time.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 0c67caf29d655f9ce1c53f8e1433c59d86e04193
Author: Andy Green <andy at openmoko.com>
Date:   Wed Aug 6 20:27:23 2008 +0100

    add-build-git-head-info.patch
    
    This patch adds the branch and truncated head has to the version of the
    kernel
    
    # cat /proc/version
    Linux version 2.6.26-andy-2.6.26:2b3bf342baac52b9-mokodev (agreen at pads.home.warmcat.com) (gcc version 4.1.2) #878 PREEMPT Sun Jul 27 14:45:25 BST 2008
    
    Note this versioning is visible down /lib/modules
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 27fc8d82e365c47065f3a9240bfe21e67a50edf2
Author: Andy Green <andy at openmoko.com>
Date:   Wed Aug 6 15:08:34 2008 +0100

    fix-pcf50633-usbrem-charging-led.patch
    
    If the charger was removed, we fell through back to old
    hdq-driven code with normally wrong but slightly random
    results for charging LED behaviour in that circumstance
    
    This patch makes us use the tracked charger status
    callbacks alone if they are defined in the platform
    data.
    
    Signed-off-by: Andy Green <andy at openmoko.com>