linux-toradex.git, branch tegra-9.12.9 Linux kernel for Apalis and Colibri modules ARM/tegra: delete SDIO DDK and DDK-based MMC host driver 2010-04-03T00:56:13+00:00 Gary King gking@nvidia.com 2010-04-02T00:41:35+00:00 1c3dac0f5908cbce08ae7cdcc82ff19b4827ecd6 Change-Id: I30872e9905b7bdf1cca56e53089e5cfde7586798 Reviewed-on: http://git-master/r/1026 Reviewed-by: Gary King <gking@nvidia.com> Tested-by: Gary King <gking@nvidia.com> 
Change-Id: I30872e9905b7bdf1cca56e53089e5cfde7586798
Reviewed-on: http://git-master/r/1026
Reviewed-by: Gary King <gking@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>
Fix NvRmMemUnpin so it ignores NULL memhandles 2010-04-02T04:06:54+00:00 Acorn Pooley apooley@nvidia.com 2010-04-02T03:28:06+00:00 293044ba71d73024a971387dc41716f6c650b619 Change-Id: Iffad8a511bb6434b26e58d8dacf24a5751c23610 Reviewed-on: http://git-master/r/1028 Reviewed-by: Acorn Pooley <apooley@nvidia.com> Tested-by: Acorn Pooley <apooley@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> 
Change-Id: Iffad8a511bb6434b26e58d8dacf24a5751c23610
Reviewed-on: http://git-master/r/1028
Reviewed-by: Acorn Pooley <apooley@nvidia.com>
Tested-by: Acorn Pooley <apooley@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
tegra mmc: enable MMC_UNSAFE_RESUME config for tegra_whistler_gnu_linux 2010-04-02T02:05:40+00:00 Deepesh Gujarathi dgujarathi@nvidia.com 2010-03-31T07:55:48+00:00 75bcb77344c4f242f595107d7b89e6a316ac0690 Normally, when the system goes into suspend, all the storage cards are removed and later re-enumerated upon resume. During re-enumeration the device name and number are not guaranteed to be the same as they were during bootup. This can cause issues if rootfs falls on one such device. When the MMC_UNSAFE_RESUME config is enabled, it will prevent kernel from unmounting the mmc cards before suspend. Upon resume the cards will still be using the same device id as they were before suspend. Change-Id: Ibc1511c5feb0391dc644df2d710e386ec0d353ce Reviewed-on: http://git-master/r/994 Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> Tested-by: Gary King <gking@nvidia.com> 
Normally, when the system goes into suspend, all the storage cards are
removed and later re-enumerated upon resume. During re-enumeration the
device name and number are not guaranteed to be the same as they were
during bootup. This can cause issues if rootfs falls on one such
device.

When the MMC_UNSAFE_RESUME config is enabled, it will prevent kernel
from unmounting the mmc cards before suspend. Upon resume the cards
will still be using the same device id as they were before suspend.

Change-Id: Ibc1511c5feb0391dc644df2d710e386ec0d353ce
Reviewed-on: http://git-master/r/994
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>
tegra RM: Updated VDE clock configuration policy. 2010-04-02T00:49:30+00:00 Alex Frid afrid@nvidia.com 2010-04-01T03:06:27+00:00 ccbaa3ec2afb825c315d11e34b30faffa10c199e Updated VDE clock configuration policy - allowed to use high frequency PLLC for VDE targets within 100MHz-200MHz range. This range was covered by low frequency PLLP0, but better divider granularity achieved with PLLC results in lower voltage requirements. Change-Id: I922dbd8db19dc19339db5bfbf2651604e28e789d Reviewed-on: http://git-master/r/1007 Tested-by: Aleksandr Frid <afrid@nvidia.com> Reviewed-by: Sharad Ranjan <shranjan@nvidia.com> Tested-by: Sharad Ranjan <shranjan@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> 
Updated VDE clock configuration policy - allowed to use high frequency
PLLC for VDE targets within 100MHz-200MHz range. This range was covered
by low frequency PLLP0, but better divider granularity achieved with PLLC
results in lower voltage requirements.

Change-Id: I922dbd8db19dc19339db5bfbf2651604e28e789d
Reviewed-on: http://git-master/r/1007
Tested-by: Aleksandr Frid <afrid@nvidia.com>
Reviewed-by: Sharad Ranjan <shranjan@nvidia.com>
Tested-by: Sharad Ranjan <shranjan@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
tegra pcie: add the VDD_1V5 power rail to the pcie guid 2010-04-02T00:48:00+00:00 Deepesh Gujarathi dgujarathi@nvidia.com 2010-04-01T13:12:19+00:00 0f29062f22c119d305eec613bfae1b2c8a6983f4 As per the harmony power tree VDDIO_PEX_CLK is the T20's 3.3V PCIe supply. It is sourced from LDO0 of the PMU. By default, LDO0 is off. VDD_1V5 is the 1.5V supply for the PCI Express Mini Card connectors. It turns on automatically in tandem with the general-purpose 3.3V domain. If there are no active pcie devices on the bus turn off the VDD_1V5 power rail. Change-Id: I518368937aa36e7c73fbb93c28a39706903a614d Reviewed-on: http://git-master/r/1019 Reviewed-by: Kenneth A Radtke <kradtke@nvidia.com> Reviewed-by: Narendra Damahe <ndamahe@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> Tested-by: Gary King <gking@nvidia.com> 
As per the harmony power tree VDDIO_PEX_CLK is the T20's
3.3V PCIe supply. It is sourced from LDO0 of the PMU. By
default, LDO0 is off.

VDD_1V5 is the 1.5V supply for the PCI Express Mini Card
connectors. It turns on automatically in tandem with the
general-purpose 3.3V domain.

If there are no active pcie devices on the bus turn off
the VDD_1V5 power rail.

Change-Id: I518368937aa36e7c73fbb93c28a39706903a614d
Reviewed-on: http://git-master/r/1019
Reviewed-by: Kenneth A Radtke <kradtke@nvidia.com>
Reviewed-by: Narendra Damahe <ndamahe@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>
tegra ODM: Changing pinmux configuration for UART-A when ULPI is used. 2010-04-02T00:44:42+00:00 Steve Lin stlin@nvidia.com 2010-04-01T01:04:07+00:00 4b115fddc0a3358f6f7ef5c85ad13c45ea5ba27c This fix corrects the pinmux configuration so UART-A is mapped to SDIO1 pins when ULPI is used. AP will be able to communicate with the ULPI modem through both UART and ULPI. Change-Id: I268a76496bd5371c2b757b0133eb0f276234504f Reviewed-on: http://git-master/r/1005 Tested-by: Szming Lin <stlin@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> 
This fix corrects the pinmux configuration so UART-A is mapped to SDIO1 pins
when ULPI is used. AP will be able to communicate with the ULPI modem through
both UART and ULPI.

Change-Id: I268a76496bd5371c2b757b0133eb0f276234504f
Reviewed-on: http://git-master/r/1005
Tested-by: Szming Lin <stlin@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
tegra usb: USB Host devices are not working on Whistler 2010-04-02T00:44:42+00:00 sgadagottu sgadagottu@nvidia.com 2010-04-01T04:16:23+00:00 b90f10ffdc6693997d8bcd9a5a9a2ba1aa518868 With CL#5705540 USB host functionality on OTG port has broken on Whistler.Also Device dis-connection and connection on root port(USB3),makes host functionality broken. So this needs to debugged further. However for Harmony both USB1 device and USB3 integrated hub are working fine with-out any issue. So reverting the change for Whistler configuration only. Bug 669288 :[AP20/Android/Whistler]: USB devices not working. Tested on : Whistler and Harmony Tested for USB host functionality on OTG port Tested for USB host functionality on USB3 with multiple connection/dis-conenction. Change-Id: Id1c2157d796e095ff66e5002b5d00958a89a7bf7 Reviewed-on: http://git-master/r/1008 Reviewed-by: Narendra Damahe <ndamahe@nvidia.com> Tested-by: Gary King <gking@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> 
With CL#5705540 USB host functionality on OTG port has broken
on Whistler.Also Device dis-connection and connection on root
port(USB3),makes host functionality broken. So this needs to
debugged further. However for Harmony both USB1 device and USB3
integrated hub are working fine with-out any issue. So reverting
the change for Whistler configuration only.

Bug 669288 :[AP20/Android/Whistler]: USB devices not working.

Tested on : Whistler and Harmony
Tested for USB host functionality on OTG port
Tested for USB host functionality on USB3 with
multiple connection/dis-conenction.

Change-Id: Id1c2157d796e095ff66e5002b5d00958a89a7bf7
Reviewed-on: http://git-master/r/1008
Reviewed-by: Narendra Damahe <ndamahe@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
tegra: Fixed error module bond-out index calculation 2010-04-01T23:06:04+00:00 Scott Williams scwilliams@nvidia.com 2010-04-01T22:55:02+00:00 f1a4aa601cfe0689cedc77cbc01abf02a276bee1 Fixed an incorrect calculation for skipping to the next byte of bond-out information. Bailing out of a byte early would result in an incorrect offset being used for subsequent module ids resulting in incorrect modules being marked as bonded out. Change-Id: Ia5cb449006c2d3b825097c852c21f4e1346e6696 Reviewed-on: http://git-master/r/1024 Reviewed-by: Scott Williams <scwilliams@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> Tested-by: Gary King <gking@nvidia.com> 
Fixed an incorrect calculation for skipping to the next byte of bond-out
information. Bailing out of a byte early would result in an incorrect offset
being used for subsequent module ids resulting in incorrect modules being
marked as bonded out.

Change-Id: Ia5cb449006c2d3b825097c852c21f4e1346e6696
Reviewed-on: http://git-master/r/1024
Reviewed-by: Scott Williams <scwilliams@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>
[ALSA]: Fix Record Hang 2010-03-31T22:48:08+00:00 Manjula Gupta magupta@nvidia.com 2010-03-31T12:30:56+00:00 f7993d029dca1275c292b11e4e04006ce470d98b - Don't signal the shutdown thread until the thread process the SNDRV_PCM_TRIGGER_STOP call. The Kthread_stop() call is a blocking call that waits until the kthread_should_stop() returns. - Set the prtd->state to SNDRV_PCM_TRIGGER_STOP in pcm_common_close specifically to ensure proper clean-up and to handle the case where application exits without triggering the STOP call. - wait for the buffer done notification for the buffers queued in before setting mixer state to stop. - For Bug 667787 [LDK/Whistler/ALSA]"arecord" app hung with DVD quality .raw audio recording (Mono/Stereo). Change-Id: Id9c13bbd2dd0b5f90b3c545677eddfa05849c45f Reviewed-on: http://git-master/r/997 Reviewed-by: Vijay Mali <vmali@nvidia.com> Tested-by: Vijay Mali <vmali@nvidia.com> Reviewed-by: Gary King <gking@nvidia.com> 
- Don't signal the shutdown thread until the thread
  process the SNDRV_PCM_TRIGGER_STOP call. The Kthread_stop()
  call is a blocking call that waits until the kthread_should_stop()
  returns.

- Set the prtd->state to SNDRV_PCM_TRIGGER_STOP in pcm_common_close
  specifically to ensure proper clean-up and to handle the case
  where application exits without triggering the STOP call.

- wait for the buffer done notification for the buffers queued in
  before setting mixer state to stop.

- For Bug 667787 [LDK/Whistler/ALSA]"arecord" app hung
  with DVD quality .raw audio recording (Mono/Stereo).

Change-Id: Id9c13bbd2dd0b5f90b3c545677eddfa05849c45f
Reviewed-on: http://git-master/r/997
Reviewed-by: Vijay Mali <vmali@nvidia.com>
Tested-by: Vijay Mali <vmali@nvidia.com>
Reviewed-by: Gary King <gking@nvidia.com>
ARM/tegra: add missing copyright attribution to cpuidle driver 2010-03-31T21:36:58+00:00 Gary King gking@nvidia.com 2010-03-31T21:35:53+00:00 2552f1cb5f0efd10509ea92bceb477b90fc7e7f5 Change-Id: I1db2c4cba29e0d4c34cc7d34c4a1be0cfb13caa5 Reviewed-on: http://git-master/r/1002 Reviewed-by: Gary King <gking@nvidia.com> Tested-by: Gary King <gking@nvidia.com> 
Change-Id: I1db2c4cba29e0d4c34cc7d34c4a1be0cfb13caa5
Reviewed-on: http://git-master/r/1002
Reviewed-by: Gary King <gking@nvidia.com>
Tested-by: Gary King <gking@nvidia.com>