linux-toradex.git, branch tegra-11.2.13 Linux kernel for Apalis and Colibri modules arm: tegra: clock: Reading APB bus before disabling clock 2011-08-11T10:36:38+00:00 Alok Chauhan alokc@nvidia.com 2011-08-09T05:19:11+00:00 437596050c805c77783e17e6b70cf169f8cb223f It may be possible that write operation on apb bus does not get complete before disabling clock if the clock is disabled just after the write on apb bus. To have proper sequence of operation, it is require to read back the apb bus to make sure the write operation is completed. bug 833341 This is cherry pick of change http://git-master/r/#change,32556 in main but hand-merged. Change-Id: Ifdabd7507d2bd738da3880cff6756a9a7b489e4b Reviewed-on: http://git-master/r/45981 Reviewed-by: Alok Chauhan <alokc@nvidia.com> Tested-by: Alok Chauhan <alokc@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> 
It may be possible that write operation on apb bus does not get
complete before disabling clock if the clock is disabled just after
the write on apb bus.
To have proper sequence of operation, it is require to read back the
apb bus to make sure the write operation is completed.

bug 833341

This is cherry pick of change http://git-master/r/#change,32556
in main but hand-merged.

Change-Id: Ifdabd7507d2bd738da3880cff6756a9a7b489e4b
Reviewed-on: http://git-master/r/45981
Reviewed-by: Alok Chauhan <alokc@nvidia.com>
Tested-by: Alok Chauhan <alokc@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
usb: ehci: tegra: enable hotplug interrupt for port3 2011-08-11T10:36:32+00:00 balam balam@nvidia.com 2011-08-08T06:23:43+00:00 9f46cc5ba4d6f8bcace3bec46a874750740795a8 Hotplug on usb is supported only for usb3(instance 2) Bug 859355 Change-Id: I9f31ebae5bda51beddc23dc6e13fd635cf8cef47 Reviewed-on: http://git-master/r/45763 Reviewed-by: Alex Courbot <acourbot@nvidia.com> Tested-by: Alex Courbot <acourbot@nvidia.com> Reviewed-by: David Schalig <dschalig@nvidia.com> Tested-by: David Schalig <dschalig@nvidia.com> Reviewed-by: Rakesh Bodla <rbodla@nvidia.com> Reviewed-by: Hanumanth Venkateswa Moganty <vmoganty@nvidia.com> 
Hotplug on usb is supported only for usb3(instance 2)

Bug 859355

Change-Id: I9f31ebae5bda51beddc23dc6e13fd635cf8cef47
Reviewed-on: http://git-master/r/45763
Reviewed-by: Alex Courbot <acourbot@nvidia.com>
Tested-by: Alex Courbot <acourbot@nvidia.com>
Reviewed-by: David Schalig <dschalig@nvidia.com>
Tested-by: David Schalig <dschalig@nvidia.com>
Reviewed-by: Rakesh Bodla <rbodla@nvidia.com>
Reviewed-by: Hanumanth Venkateswa Moganty <vmoganty@nvidia.com>
ARM: tegra: ventana: nvidia memory table updates to prevent corruption 2011-08-09T20:22:29+00:00 Varun Wadekar vwadekar@nvidia.com 2011-07-27T06:23:12+00:00 0ee38bb7f7b6720b6c0627a9cf8c334678066d8d To avoid memory corruption when device is operating at full temperature QUSE_EXTRA should always be set to 0 for frequencies 150MHz and less. As extra protection change FBIO_CFG5 to remove the region where there is TriState on the DQS signals thus preventing false DQS pulses (and false reads). (cherry-picked from 2dc075be0e8495654a84a6bc6afa63408e141b02) Original author: James Wylder <james.wylder@motorola.com> Signed-off-by: Varun Wadekar <vwadekar@nvidia.com> Change-Id: I5c711343737edf972251006484ea84661106e0f9 Reviewed-on: http://git-master/r/43401 Reviewed-by: Simone Willett <swillett@nvidia.com> Tested-by: Simone Willett <swillett@nvidia.com> 
To avoid memory corruption when device is operating at full temperature
QUSE_EXTRA should always be set to 0 for frequencies 150MHz and less.
As extra protection change FBIO_CFG5 to remove the region where there
is TriState on the DQS signals thus preventing false DQS pulses (and
false reads).

(cherry-picked from 2dc075be0e8495654a84a6bc6afa63408e141b02)

Original author: James Wylder <james.wylder@motorola.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>

Change-Id: I5c711343737edf972251006484ea84661106e0f9
Reviewed-on: http://git-master/r/43401
Reviewed-by: Simone Willett <swillett@nvidia.com>
Tested-by: Simone Willett <swillett@nvidia.com>
ventana: use separate emc chips for ventana based devices 2011-08-09T00:39:52+00:00 Nitin Kumbhar nkumbhar@nvidia.com 2011-07-25T10:54:11+00:00 a714dd55c49e16173f7cff264ac58b5cbd6c0942 Many platforms are derived from ventana and have different types of memories. Add a separate array for the emc chips being used on those platforms. Be default, the array is not populated with any entries. This disables memory scaling for that device. BUG 854226 Change-Id: Ic200c980c074ce315880c964f08ce1d5482f6766 Reviewed-on: http://git-master/r/43038 Reviewed-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com> Tested-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> 
Many platforms are derived from ventana and have different types
of memories. Add a separate array for the emc chips being used on
those platforms.

Be default, the array is not populated with any entries. This
disables memory scaling for that device.

BUG 854226

Change-Id: Ic200c980c074ce315880c964f08ce1d5482f6766
Reviewed-on: http://git-master/r/43038
Reviewed-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com>
Tested-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
arm: tegra: otg: check connected devices on resume 2011-08-09T00:39:18+00:00 Victor Ryabukhin vryabukhin@nvidia.com 2011-07-11T02:29:43+00:00 ed66cd3fd7774392fd31226dec0a58eb29e1dc69 A device might be connected while CPU is in sleep mode. In this case no interrupt will be triggered. Force irq_work to recheck connected devices. Reviewed-on: http://git-master/r/40331 (cherry picked from commit 749e994e1e0010843e4a2d54e27a7f19274661a7) Change-Id: Ic106a2e3115a87dc9548714ef89383add42fd777 Reviewed-on: http://git-master/r/43433 Reviewed-by: David Schalig <dschalig@nvidia.com> Tested-by: David Schalig <dschalig@nvidia.com> Reviewed-by: Alex Courbot <acourbot@nvidia.com> Reviewed-by: Hanumanth Venkateswa Moganty <vmoganty@nvidia.com> 
A device might be connected while CPU is in sleep mode.
In this case no interrupt will be triggered.
Force irq_work to recheck connected devices.

Reviewed-on: http://git-master/r/40331
(cherry picked from commit 749e994e1e0010843e4a2d54e27a7f19274661a7)

Change-Id: Ic106a2e3115a87dc9548714ef89383add42fd777
Reviewed-on: http://git-master/r/43433
Reviewed-by: David Schalig <dschalig@nvidia.com>
Tested-by: David Schalig <dschalig@nvidia.com>
Reviewed-by: Alex Courbot <acourbot@nvidia.com>
Reviewed-by: Hanumanth Venkateswa Moganty <vmoganty@nvidia.com>
video: tegra: dc: Update hdmi hot-plug status on resume 2011-08-09T00:39:13+00:00 Manjula Gupta magupta@nvidia.com 2011-08-01T11:25:13+00:00 ef32636d02cb3405f7f0232d34fb29a33fc5ad94 - Update the upper layer about hdmi removal in suspended state. - Removes the redundant variable hpd_pending. Bug 857122 Change-Id: I78e22e572a9ff0e475d15bbb94c608cfceeefbee Reviewed-on: http://git-master/r/44326 Tested-by: Manjula Gupta <magupta@nvidia.com> Reviewed-by: Jonathan Mayo <jmayo@nvidia.com> 
- Update the upper layer about hdmi removal in suspended state.
- Removes the redundant variable hpd_pending.

Bug 857122

Change-Id: I78e22e572a9ff0e475d15bbb94c608cfceeefbee
Reviewed-on: http://git-master/r/44326
Tested-by: Manjula Gupta <magupta@nvidia.com>
Reviewed-by: Jonathan Mayo <jmayo@nvidia.com>
asoc: tegra: Fix wired jack resume 2011-08-04T14:12:45+00:00 Sumit Bhattacharya sumitb@nvidia.com 2011-08-03T11:43:43+00:00 54ad1d53ec1f855e133a3c912e0a8c94b4a4c7e1 Calling of snd_soc_jack_add_gpios() from tegra_jack_resume() causes kernel warning since parentless device registration is not allowed during PM transaction. Modify wired_jack supend/ resume functions to disable IRQs during suspend and enable them during resume and report the current jack state to snd_soc_jack interface to ensure jack events happened during suspend state are not missed. Bug 850424 Change-Id: Idde4504096a0793729adaafed3da70f08be91101 Reviewed-on: http://git-master/r/44747 Tested-by: Sumit Bhattacharya <sumitb@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> 
Calling of snd_soc_jack_add_gpios() from tegra_jack_resume()
causes kernel warning since parentless device registration is
not allowed during PM transaction. Modify wired_jack supend/
resume functions to disable IRQs during suspend and enable
them during resume and report the current jack state to
snd_soc_jack interface to ensure jack events happened during
suspend state are not missed.

Bug 850424

Change-Id: Idde4504096a0793729adaafed3da70f08be91101
Reviewed-on: http://git-master/r/44747
Tested-by: Sumit Bhattacharya <sumitb@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
arm:video:dc Audio EDID improvements 2011-08-04T14:12:19+00:00 ScottPeterson speterson@nvidia.com 2011-07-26T21:54:14+00:00 aaadcb7a4f8501dc71c6185e111fab1427bb7286 Finer grain determination of audio capabilities for HDMI devices. TO enable audio we have to have basic audio supported in the EDID plus at least one Short Audio Descriptor block. Change-Id: Icb8b242c80ae580cdcdea1910a7cf7bd1f124239 Reviewed-on: http://git-master/r/43304 Tested-by: Scott Peterson <speterson@nvidia.com> Reviewed-by: Sumit Bhattacharya <sumitb@nvidia.com> Reviewed-by: Scott Peterson <speterson@nvidia.com> 
Finer grain determination of audio capabilities
for HDMI devices. TO enable audio we have to have
basic audio supported in the EDID plus at least one
Short Audio Descriptor block.

Change-Id: Icb8b242c80ae580cdcdea1910a7cf7bd1f124239
Reviewed-on: http://git-master/r/43304
Tested-by: Scott Peterson <speterson@nvidia.com>
Reviewed-by: Sumit Bhattacharya <sumitb@nvidia.com>
Reviewed-by: Scott Peterson <speterson@nvidia.com>
arm: tegra: fuse: Add A04 chip revision. 2011-07-30T00:12:41+00:00 Roger Hsieh rhsieh@nvidia.com 2011-07-27T12:37:28+00:00 6c1f8e8457f7c988ec9fe4e7e902ce83876ec51f tegra_get_revision() returned chip minor version but it doesn't identify A04 chip correctly. Add A04 revision to fix it. Bug 856439 Change-Id: I7249633dccaa1f15dd275feeeb2c528322f1f47f Reviewed-on: http://git-master/r/43481 Reviewed-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com> Tested-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com> Reviewed-by: Varun Wadekar <vwadekar@nvidia.com> Reviewed-by: ChihJen Hsu <chhsu@nvidia.com> Reviewed-by: Yu-Fong (Foster) Cho <ycho@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> 
tegra_get_revision() returned chip minor version but it doesn't identify
A04 chip correctly. Add A04 revision to fix it.

Bug 856439

Change-Id: I7249633dccaa1f15dd275feeeb2c528322f1f47f
Reviewed-on: http://git-master/r/43481
Reviewed-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com>
Tested-by: Ching Kuang (Roger) Hsieh <rhsieh@nvidia.com>
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-by: ChihJen Hsu <chhsu@nvidia.com>
Reviewed-by: Yu-Fong (Foster) Cho <ycho@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
video: tegra: dc: use a delayed_work on hdcp 2011-07-30T00:12:33+00:00 Donghan Ryu dryu@nvidia.com 2011-07-27T22:43:01+00:00 28bb7848e02ce46972cdaad6cab076cf4bc253fe hdmi modeset triggers to re-start hdcp and trying to negotiate too early can cause HDCP failure. using delayed_work can avoid this situation by starting HDCP a little bit late Bug 855002 Change-Id: I64bd84406ebb02ecc33f53eee65e2b3032a69880 Reviewed-on: http://git-master/r/43463 Tested-by: Donghan Ryu <dryu@nvidia.com> Reviewed-by: Jonathan Mayo <jmayo@nvidia.com> Reviewed-by: Donghan Ryu <dryu@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> 
hdmi modeset triggers to re-start hdcp and trying to negotiate
too early can cause HDCP failure. using delayed_work can avoid
this situation by starting HDCP a little bit late

Bug 855002

Change-Id: I64bd84406ebb02ecc33f53eee65e2b3032a69880
Reviewed-on: http://git-master/r/43463
Tested-by: Donghan Ryu <dryu@nvidia.com>
Reviewed-by: Jonathan Mayo <jmayo@nvidia.com>
Reviewed-by: Donghan Ryu <dryu@nvidia.com>
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>