development kernel tree: Changes to 'andy'

[git at git.openmoko.org]

Rebased ref, commits from common ancestor:
commit 2048e54a99d8ad4a1403c3fda7be655614f40aee
Author: Sean McNeil <sean at mcneil.com>
Date:   Tue May 13 10:53:58 2008 +0100

    fix-wep-needs-keys-before-ap.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 80c67937b4c3ed4accd3c92cee63fe50c87130d2
Author: matt_hsu <matt_hsu at openmoko.org>
Date:   Sun May 11 12:07:03 2008 +0100

    s3c24xx-pwm-platform-driver.patch
    
    This patch is to register pwm as platform driver to keep the PWM-related
    config when system is in suspend/resume. This could fix the following
    issue after resume:
    
    - HDQ read timeout
    - LEDs blinked abnormally(if LEDs is driven by PWM)
    
    Signed-off-by: Matt Hsu <matt_hsu at openmoko.org>

commit 4f545eeee87ce36b34b1d6b63a9d1e834afaab0a
Author: Sean McNeil <sean at mcneil.com>
Date:   Sun May 11 11:57:29 2008 +0100

    minor problem with LCD
    
    Looking at wrong register to see if the LCD is turned off?

commit 30c79d126371b5aa86eb4c4aad573995a88521d0
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:28 2008 +0100

    clean-remove-build-noise.patch
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 6f5589bc7ee9d9f14248bd05fce82ccdb1a0e5e3
Author: Holger Freyther <zecke at openmoko.org>
Date:   Sun May 11 11:57:28 2008 +0100

    Disable low level debugging, make the s3c24xx PM debug code depend on DEBUG_LL
    
    Hey,
    where are slowly getting into the product state, low level debugging is
    supposed to be turned off there. :)
    
    z
    
    
    
    From 6448f0a521c05c5ba6672f5908c55583974ce1ed Mon Sep 17 00:00:00 2001
    From: Holger Freyther <zecke at openmoko.org>
    Date: Fri, 9 May 2008 19:56:31 +0200
    Subject: [PATCH] [debug] Disable low level debugging for the product
         Disable low level debugging in the defconfig-gta02 and
         the s3c pm code used printascii which is only available
         with low level debugging, update the Kconfig entry for
         this pm debugging support to depend on low level debugging
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit 18c88134312c4c3881c173f9aa8af74592af4c0a
Author: Holger Freyther <zecke at openmoko.org>
Date:   Sun May 11 11:57:28 2008 +0100

    [power] Use the bq27000 battery to provide charging values for apm
        Built the apm emulation for gta02 and say that the gta02 battery
        is supposed to be used for this apm emulation.
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit fbc6c2c9b84fc0d760baada5c73f00c04cd8819e
Author: Holger Freyther <zecke at openmoko.org>
Date:   Sun May 11 11:57:28 2008 +0100

    [pcf50633] Disable debugging statements as the driver is working well
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit b3d66a64ef9d55b0223cc85b64d4e683829664ea
Author: Holger Freyther <ich at tamarin.(none)>
Date:   Sun May 11 11:57:28 2008 +0100

    [pcf50633] Assume that all gta02's have a battery with coulumb counter
        For the gta02 and the bq27000 battery it does not make sense to use the
        ADC to get the current voltage. Under the assumption that all mass
        production gta02's have such batteries it does not make any sense to
        forward this value to APM.
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit e34077d9a5d42f78f2123dc73fed6ffe269d1dc9
Author: Holger Freyther <ich at tamarin.(none)>
Date:   Sun May 11 11:57:28 2008 +0100

    [pcf50633] Add exported symbol to the header file
        Reading the battery volt using the ADC was an exported symbol
        but not in the header file, add it.
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit adcc41fcd210f52d4719fd594a624df308966893
Author: Holger Freyther <ich at tamarin.(none)>
Date:   Sun May 11 11:57:28 2008 +0100

    [pcf50633] Print less debugging messages
        According to the comments in the code we have to read
        INT1 - INT5 in case of any interrupt. So even if we have masked
        out the periodic timer interrupt we were able to see that it
        fired. Only print that we got a SECOND interrupt if we are actually
        interested in the periodic tick. From a brief look at the datasheet there
        is no way to stop the periodic tick.
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit 31f302424bd25815e1cdefef448cfae2ee61a21d
Author: Holger Freyther <ich at tamarin.(none)>
Date:   Sun May 11 11:57:28 2008 +0100

    Explain why we want SECOND to kick.
        We do not mask the SECOND interrupt to make sure that we get
        a interrupt at the beginning. We use this for the coldplug_done
        initisalisation.
        So pcf->flags and INT1M is out of sync at the beginning but after the
        first SECOND interrupt it will be put into sync.
    
    Signed-Off-By: Holger Freyther <zecke at openmoko.org>

commit 7902072cd62ca14db2c411cf8888f226eabbbf1f
Author: Holger Freyther <zecke at openmoko.org>
Date:   Sun May 11 11:57:28 2008 +0100

    Re: [PATCH 6/7] fix-suppress-cpu-suspend-save-restore-messages.patch
    
    Andy Green <andy at ...> writes:
    
    
    I kind of dislike commenting out code. Is this any better?

commit 4d63d061f8b24bf48ec56c052f2a0538f3d3b7d3
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:28 2008 +0100

    fix-backlight-disable-on-zero-intensity.patch
    
    It's never right to put 0 intensity into LEDOUT according to datasheet
    But having a floor at intensity 2 means backlight isn't properly off
    when "dimmed".  So change to intensity 0 --> disable backlight.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 8951eed2ac791cab2270f9c738b880dbc82f530f
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:27 2008 +0100

    add-force-backlight-up-on-resume.patch
    
    The backlight status is restored during resume action, if it was
    off then it resumes like that.
    
    This patch forces it on after resume.
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 5aa1da95dafb5e27eecedd7f48d3e9cde3ac3644
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:27 2008 +0100

    quench-glamofb-err-lock-and-mmc-debug.patch
    
    Remove debugging noise
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit 92ae569b95dada20da0e99911790f1a40abc5a02
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:27 2008 +0100

    fix-suspend-backlight-timing-pm-debug.patch
    
    This patch improves the smoothness of suspend and resume action.
    
    Taking out CONFIG_PM_DEBUG allows much more rapid resume (the low level
    serial traffic appears to be synchronous)
    
    Added a platform callback in jbt driver and support in pcf50633 so we
    can defer bringing up the backlight until the LCM is able to process
    video again (which must happen after the glamo is up and producing
    video beacuse the LCM is hooked to glamo SPI)
    
    GTA01 should not be affected by all this as the callback will default
    to null and it is on pcf50606
    
    Signed-off-by: Andy Green <andy at openmoko.com>

commit c34df299943eff91029a7da9f992f2816a239906
Author: Andy Green <andy at openmoko.com>
Date:   Sun May 11 11:57:27 2008 +0100

    config-add-back-MMC_UNSAFE_RESUME.patch
    
    After LL debug disable, "unsafe" resume allows resume from SD card
    rootfs without corruption, WAH
    
    Signed-off-by: Andy Green <andy at openmoko.com>