Chuanxiao Dong [Thu, 8 Dec 2011 09:30:37 +0000 (17:30 +0800)]
mmc: block: kill block request if card is removed
BZ: 18012
Kill block requests when the host realizes that the card is
removed from the slot and is sure that subsequent requests
are bound to fail. Do this silently so that the block
layer doesn't output unnecessary error messages.
Change-Id: I3e5555f92db8a05424018947effeb251ff4a97eb
Signed-off-by: Chuanxiao Dong <chuanxiao.dong@intel.com>
Reviewed-on: http://android.intel.com:8080/29374
Reviewed-by: Tang, Richard <richard.tang@intel.com>
Tested-by: Sun, Jianhua <jianhua.sun@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Chuanxiao Dong [Thu, 8 Dec 2011 09:10:27 +0000 (17:10 +0800)]
mmc: allow upper layers to know immediately if card has been removed
BZ: 18012
[PORT FROM UPSTREAM]
Add a function mmc_detect_card_removed() which upper layers can use to
determine immediately if a card has been removed. This function should
be called after an I/O request fails so that all queued I/O requests
can be errored out immediately instead of waiting for the card device
to be removed.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Sujit Reddy Thumma <sthumma@codeaurora.org>
Signed-off-by: Chris Ball <cjb@laptop.org>
Change-Id: I3b602d69e0cc3c5a49245a35adcd227d849103b9
Signed-off-by: Chuanxiao Dong <chuanxiao.dong@intel.com>
Reviewed-on: http://android.intel.com:8080/29373
Reviewed-by: Tang, Richard <richard.tang@intel.com>
Tested-by: Sun, Jianhua <jianhua.sun@intel.com>
Reviewed-by: Zayet, AymenX <aymenx.zayet@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Vaibhav Agarwal [Mon, 12 Dec 2011 06:28:50 +0000 (11:58 +0530)]
HDMI-Audio: Make HAD as static module in default i386_mfld_defconfig
BZ: 17087
Modify default i386_mfld_defconfig to build HDMI Audio (HAD) module
statically.
Change-Id: I508f6ff76fcfac7985624807bab0d792849b2394
Signed-off-by: Vaibhav Agarwal <vaibhav.agarwal@intel.com>
Reviewed-on: http://android.intel.com:8080/26983
Reviewed-by: Centelles, Sylvain <sylvain.centelles@intel.com>
Reviewed-by: Koul, Vinod <vinod.koul@intel.com>
Reviewed-by: Babu, Ramesh <ramesh.babu@intel.com>
Reviewed-by: Gupta, ArvindX K <arvindx.k.gupta@intel.com>
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Vaibhav Agarwal [Wed, 7 Dec 2011 15:33:51 +0000 (21:03 +0530)]
HDMI-Audio: Ported to ICS
BZ: 17078
Steps to verify:
1. Ensure device is not in sleep state
2. alsa_aplay -l
3. alsa_aplay -Dplughw:IntelHDMI,0 -F 23220 <file_name>
Change-Id: I48626a0dec8fdccfdec90a2b9a8747bfe11ea250
Signed-off-by: Vaibhav Agarwal <vaibhav.agarwal@intel.com>
Reviewed-on: http://android.intel.com:8080/26982
Reviewed-by: Koul, Vinod <vinod.koul@intel.com>
Reviewed-by: Centelles, Sylvain <sylvain.centelles@intel.com>
Reviewed-by: Babu, Ramesh <ramesh.babu@intel.com>
Reviewed-by: Gupta, ArvindX K <arvindx.k.gupta@intel.com>
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Austin Hu [Fri, 16 Dec 2011 05:10:16 +0000 (13:10 +0800)]
Gfx-Display: Enhance DDC to read extention block in EDID
BZ: 17008
Replacing late_initcall with device_initcall for I2C GPIO driver init, to
resolve the I2C adapter detection issue for HDMI.
Change-Id: Ia14c923ef855c4de1242b753fe9c08fc2d1875f6
Signed-off-by: Lei Zhang <lei.zhang@intel.com>
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/27744
Reviewed-by: Hu, Austin <austin.hu@intel.com>
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Tong Bo [Tue, 20 Dec 2011 10:04:47 +0000 (18:04 +0800)]
Gfx-Display: Tag the eng-build release version
BZ: 17941
change PSB_PACKAGE_VERSION to 5.6.0.1152 to tag the eng-build release on the ww52 of 2011.
Change-Id: Ia33b2ca0358e9586a28044c9e97acc1d00016975
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/29355
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Randy Xu [Tue, 20 Dec 2011 01:10:12 +0000 (09:10 +0800)]
[Graphics/Display]: Port some patches from HC/DDK1.7 branch
BZ: 17881
GFX-Display-Enable-DSR-POC-with-runtime_pm-support.patch
GFX-DISPLAY-Add-support-to-CMI-video-mode-MIPI-disp.patch
GFX-DISPLAY-Add-supports-for-CMI-720p-command-mode.patch
Change-Id: I7602bab208bb2ea7648ec7e772bc834667a63665
Signed-off-by: Randy Xu <randy.xu@intel.com>
Reviewed-on: http://android.intel.com:8080/29176
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Jackie Li [Mon, 12 Dec 2011 14:14:09 +0000 (22:14 +0800)]
hwc: added a wait for vblank flag to sprite operations
BZ: 17189
we need wait for vblank start when doing sprite plane flip to
avoid tearing issues.
Change-Id: I786827f8a5ec80bfec94babb6bf84b65e5264f7c
Signed-off-by: Jackie Li <yaodong.li@intel.com>
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/27728
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Austin Hu [Tue, 13 Dec 2011 13:33:18 +0000 (21:33 +0800)]
Gfx-Display: fixed a potential panic issue.
BZ: 17268
When HDMI flip chain isn't initialized, de-queue buffer from it would
cause panic as its buffer number may be 0.
Fixed by checking whether the flip chain is initialized, before getting
the buffer handle.
Change-Id: I61d9f17bd484c352b2039a61b99b6947f7c36347
Signed-off-by: Austin Hu <austin.hu@intel.com>
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/27729
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Yun Tu(Mark Tu) [Mon, 31 Oct 2011 07:00:46 +0000 (15:00 +0800)]
GFX-3D: add yv12 support to video buffer class driver
BZ:13400
add yv12 format support to video buffer class driver
Change-Id: Id7c4c217f683e7c2671b19531456c0ced1974e28
Signed-off-by: Yun Tu(Mark Tu) <yun.tu@intel.com>
Reviewed-on: http://android.intel.com:8080/27732
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Lei Zhang [Tue, 13 Dec 2011 09:21:15 +0000 (17:21 +0800)]
Gfx-Display: Added an ioctl to report the HDCP Ri mismatch
BZ: 17008
Ported from Jim Liu's patch in hc-ddk-1.7 to support HDCP.
An ioctl function - get_hdcp_link_status_ioctl is added into display driver. It
is used to report the Ri mismatch to upper level SW. The upper level SW can
stop playing premium content and disable HDCP when Ri mismatch occurs.
Change-Id: I0d43e35ff45791269da4bbb6b652eaf38e0f6c41
Signed-off-by: Lei Zhang <lei.zhang@intel.com>
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/27241
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: Ai, Ke <ke.ai@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ryan VanSickle [Wed, 14 Dec 2011 00:51:29 +0000 (16:51 -0800)]
Checking in color correction IOCTL call
BZ: 17761
This IOCTL call exposes the csc_program_DC function to allow a user
to send in a 3x3 color correction matrix to the driver.
The basics are that you have a new struct defined in psb_drm.h as follows:
struct drm_psb_csc_matrix {
int pipe;
int64_t matrix[9];
};
The matrix variable is define such that the 3x3 matrix will be represented
in a vector with 9 elements.
M00 = V0, M01 = V1, M02 = V2,
M10 = V3, M11 = V4, M12 = V5,
M20 = V6, M21 = V7, M22 = V8.
The pipe variable is defined as the offset to the PIPEASRC register value
as defined in psb_intel_reg.h (see the PIPEB and PIPEC definitions in that
file. The function that does the color correction checks against PIPEB
and PIPEC values and assumes you are setting the correction to PIPEA if no
match is found.
FYI
PIPEB = 0x1
PIPEC =0x2
PIPEA has no formal definition. I used 0x0 for my purposes.
The ioctl number is defined as
\#define DRM_IOCTL_PSB_SET_CSC DRM_IOW(DRM_PSB_SET_CSC + DRM_COMMAND_BASE, \
struct drm_psb_csc_matrix (found in psb_drv.c)
\#define DRM_PSB_SET_CSC 0x29 (found in psb_drm.h)
Finally the actual function doing the work looks like this:
static int psb_set_csc_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv)
{
struct drm_psb_csc_matrix *csc_matrix = data;
csc_program_DC(dev, csc_matrix->matrix, csc_matrix->pipe);
return 0;
}
Change-Id: I84c6cd1b79226af4afea23b80036d0d8e07da64d
Signed-off-by: Randy Xu <randy.xu@intel.com>
Reviewed-on: http://android.intel.com:8080/27730
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
gwang23 [Wed, 19 Oct 2011 08:52:10 +0000 (16:52 +0800)]
Graphics: Add a path to allocate memory without mapping into GPU
BZ: 17008
For 9134 issue, the GPU address space is used out. add a path to allocate
memory without mapping into GPU to refine the usage of the GPU address
space.
Change-Id: Ibeffde0907238f7e232224d2f59b7db3051ccfb9
Signed-off-by: gwang23 <gang.a.wang@intel.com>
Reviewed-on: http://android.intel.com:8080/22084
Reviewed-by: Xu, Randy <randy.xu@intel.com>
Tested-by: Xu, Randy <randy.xu@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Signed-off-by: Tong Bo <box.tong@intel.com>
Reviewed-on: http://android.intel.com:8080/27740
Shijie Zhang [Fri, 23 Dec 2011 11:04:41 +0000 (19:04 +0800)]
GPIO key: enable camera button functional on ICS
BZ: 18147
Camera gpio_key should be registered to make camera button functional
Change-Id: I00eb503b36200350819aaf418e56bab83d7aab68
Signed-off-by: Shijie Zhang <shijie.zhang@intel.com>
Reviewed-on: http://android.intel.com:8080/29781
Reviewed-by: Yang, Bin <bin.yang@intel.com>
Reviewed-by: Du, Alek <alek.du@intel.com>
Tested-by: Wang, Zhifeng <zhifeng.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Omair Mohammed Abdullah [Tue, 8 Nov 2011 10:54:46 +0000 (16:24 +0530)]
[REBASED on main (r3branch)from r2-stable branch] audio: sst: Fix for errors revealed by klocwork
BZ: 17975
Fixes errors in driver revealed by klocwork and other static analysis tools
Change-Id: I661df04fd1fae138a6c21842ff5699bc28d046bf
old-Change-Id: Iea1f15842e6c7f39a90a7c038e4d9817131f29a6
Signed-off-by: Omair Mohammed Abdullah <omair.m.abdullah@intel.com>
Reviewed-on: http://android.intel.com:8080/29537
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Omair Mohammed Abdullah [Mon, 21 Nov 2011 10:17:03 +0000 (15:47 +0530)]
[REBASED on main r3(stable)from r2-stable] audio: sst: make sst driver mad workqueue single threaded
BZ: 17975
Driver IPC handling had various issues
Makes the sst driver mad workqueue single threaded. Multiple workqueue tasks in
a single threaded workqueue will not run in parallel.
FW state should be set as soon as possible while freeing stream so that later
messages are ignored.
Drop IPC uses ctrl_blk not data_blk, so setting the ctrl_blk to on.
Adds locking to checking and changing of stream states in free_stream and
drop_stream to prevent a race condition.
Added missing block condition to firmware context save.
Change-Id: I117e413b593094a629bb3e29ab9aa75c4f4fe233
old-Change-Id: Ia4e4dbd33b32882ada369ae5a9ce9cce5f6900f6
Signed-off-by: Omair Mohammed Abdullah <omair.m.abdullah@intel.com>
Reviewed-on: http://android.intel.com:8080/29524
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ramesh Babu K V [Tue, 13 Dec 2011 06:54:48 +0000 (12:24 +0530)]
[REBASED on main (r3branch)from r2-stable branch]audio:ensure msic pll enabled during vaud power up
BZ: 17975
Race condition might occur when disabling the vibra and enabling the
audio path. If vibra is already active and playback starts, msic pll
will be enabled immediately, otherwise it will be enabled during
vaud power on sequence. In corner cases, vaud will be switched off
after pll is enabled and there after pll is not enabled. This causes
no audio on the sinks.
Enable the msic pll while powering up the vaud rail, irrespective
of it's previous state. This makes sure that pll is enabled all
the time while powering up the vaud for audio stream use cases.
Change-Id: I9964c009b4aff6a455aa2d49afdb4e1c044d4aa5
Old-Change-Id: I02e0aacca19d2b0dee5d1d8f528cdcd9ba651575
Signed-off-by: Ramesh Babu K V <ramesh.babu@intel.com>
Reviewed-on: http://android.intel.com:8080/29541
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ramesh Babu K V [Sat, 10 Dec 2011 08:49:13 +0000 (14:19 +0530)]
[REBASED on main (r3branch)from r2-stable branch]audio:configure pcm1 to slave during hw_free
BZ: 17975
MSIC PCM1 must be configured to slave and short/long sync
mode in order to tristate the PCM1 bus.
Configure the PCM1 to slave mode before disabling
the PCM1.
Change-Id: I572805e07ae14a62de31b533d68b4258ac18094e
Old-Change-Id: Ia80fedc45c44d983335f4887af1480cc9405fbea
Signed-off-by: Ramesh Babu K V <ramesh.babu@intel.com>
Reviewed-on: http://android.intel.com:8080/29539
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ramesh Babu K V [Mon, 21 Nov 2011 15:10:16 +0000 (20:40 +0530)]
[REBASED on main (r3branch)from r2-stable branch]ASoC:dapm-Add DAPM stream completion event
BZ: 17975
This adds a callback function to be called at the completion of a
DAPM stream event.
This can be used by codec to perform operation based on
DAPM graphs after completion of stream events.
This work is adopted from the upstream asoc kernel.
rebasing it to r3 -- Main branch.
Change-Id: I6b89f5a55f04a508ae27131dfe0357c4134ad11c
Old-Change-Id: I138dc0cb1ab90495c96312a7e54f234a48320c9f
Reviewed-on: http://android.intel.com:8080/29544
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Vaibhav Agarwal [Wed, 14 Dec 2011 08:50:26 +0000 (14:20 +0530)]
Audio: Modify IRQ_BASE start address for msic_audio to OC_AUDIO
BZ: 16836
With K3.0 migration, IRQ_BASE start address was modified to
ACC_DET(0x00F). However, actual start address is OC_AUDIO(0x00E)
Due to this,incorrect intr_status values are read during jack detection.
Thus,incorrect events are reported.
The solution is to modify IRQ_BASE start address for msic_audio to
OC_AUDIO(0x00E)
Change-Id: Icab57906c6bcdce1d6a105f6ee52e367c70076c8
Signed-off-by: Vaibhav Agarwal <vaibhav.agarwal@intel.com>
Reviewed-on: http://android.intel.com:8080/27313
Reviewed-by: Koul, Vinod <vinod.koul@intel.com>
Reviewed-by: Babu, Ramesh <ramesh.babu@intel.com>
Reviewed-by: Gross, Mark <mark.gross@intel.com>
Reviewed-by: Abdullah, Omair M <omair.m.abdullah@intel.com>
Reviewed-by: Gupta, ArvindX K <arvindx.k.gupta@intel.com>
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ramesh Babu K V [Mon, 21 Nov 2011 15:12:14 +0000 (20:42 +0530)]
[REBASED on main (r3branch)from r2-stable branch]audio:disable the clocks in stream event callback
BZ: 17975
MSIC PLL and external PLLIN clock sources are disabled
during vaud bias off event. However vaud bias will continue
to be active when headset is connected. This will make the
PLLs to active even during no audio playback.
Disable the PLLs after DAPM power down sequence completes.
This will save some mW of power.
Signed-off-by: Ramesh Babu K V <ramesh.babu@intel.com>
rebase the patch from main (r2-stable) to r3 main.
Change-Id: I1f81ee72ea95699d47425e47102ab5e5daadd3e5
Old-Change-Id: Ie6cab0806e588174515879ef24274854d192e59d
Reviewed-on: http://android.intel.com:8080/29546
Reviewed-by: M, Arulselvan <arulselvan.m@intel.com>
Tested-by: M, Arulselvan <arulselvan.m@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Li, Ning [Thu, 22 Dec 2011 06:46:45 +0000 (14:46 +0800)]
[PORT FROM R2] Add keypad driver patches
BZ: 18179
This BZ includes the following BZ:
4915, 13482
Signed-off-by: Li, Ning <ning.li@intel.com>
Change-Id: Ief9354e35afa55991a0fba63f6f9edbac7778bed
Reviewed-on: http://android.intel.com:8080/29697
Reviewed-by: Li, Ning <ning.li@intel.com>
Reviewed-by: Yang, Bin <bin.yang@intel.com>
Reviewed-by: Du, Alek <alek.du@intel.com>
Tested-by: Wang, Zhifeng <zhifeng.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Li, Ning [Thu, 22 Dec 2011 04:53:55 +0000 (12:53 +0800)]
[PORT FROM R2] Add GPIO controller driver patch
BZ: 18157
This BZ includes BZ:
4270
Signed-off-by: Li, Ning <ning.li@intel.com>
Change-Id: I0c0b8d0f0f02f106ab83344eb23eecccabd328cf
Reviewed-on: http://android.intel.com:8080/29689
Reviewed-by: Li, Ning <ning.li@intel.com>
Reviewed-by: Yang, Bin <bin.yang@intel.com>
Reviewed-by: Du, Alek <alek.du@intel.com>
Tested-by: Wang, Zhifeng <zhifeng.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Zhang Dongxing [Wed, 16 Nov 2011 06:08:23 +0000 (14:08 +0800)]
[PORT FROM R2]POWERDK: Powerdk upgrade for enhancements for android wuwatch
BZ: 17764
Powerdk upgrade for enancements for android wuwatch version 2.2.2
release.
Signed-off-by: Zhang Dongxing <dongxing.zhang@intel.com>
Change-Id: I5a64d6068a28e70808f1c6b6cbef31ad73b640e7
Org-Change-Id: I5c1851bc119079e8b48cdaa6a3585c6d967b28b6
Reviewed-on: http://android.intel.com:8080/28038
Reviewed-by: Zhang, Dongxing <dongxing.zhang@intel.com>
Reviewed-by: Guerard, Christophe <christophe.guerard@intel.com>
Tested-by: Zhang, Dongxing <dongxing.zhang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
jzha144 [Mon, 12 Dec 2011 05:08:27 +0000 (13:08 +0800)]
OPROFILE: open oprofile
BZ: 17027
open oprofile
Change-Id: I376797121afcb332a8910f7fbe51f570128fc6ae
Signed-off-by: jzha144 <jun.zhang@intel.com>
Reviewed-on: http://android.intel.com:8080/26972
Reviewed-by: Guerard, Christophe <christophe.guerard@intel.com>
Tested-by: Zhang, Dongxing <dongxing.zhang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Teemu Tuominen [Tue, 20 Dec 2011 14:51:26 +0000 (16:51 +0200)]
atomisp: hmm: prevent might_sleep's in atomisp_isr stack
BZ: 18086
This patch removes mutexes from hmm_bo->status queries and makes
hmm_store to use kmap_atomic when in interrupt context. Potential sleeps
were revealed by R3 kernel having the following R2 patches applied.
86d19d0 atomisp: (re)start ISP and handle frame buffers from ISR
b318087 atomisp: hmm: hmm_bo cannot use mutex anymore
Change-Id: I252f9b0e2ea2158429955a6c8e0aa3b1929346fb
Signed-off-by: Teemu Tuominen <teemux.tuominen@intel.com>
Reviewed-on: http://android.intel.com:8080/29552
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Durgadoss R [Wed, 30 Nov 2011 09:08:55 +0000 (14:38 +0530)]
EM/THERMAL: Port Platform Thermal Driver to K3.0
BZ: 15135
This patch ports the intel_mid_thermal driver to K3.0.
This driver was already upstreamed, but without the latest
updates. So, this patch updates the code that is already available
in K3.0 and enables the driver in Kconfig. The new driver depends
on the MSIC_GPADC driver. The updated code has been tested on
Medfield PR2 platforms. Some major updates are:
* Use of GPADC driver to manipulate the ADC channels.
* Reduce the precision loss in linear interpolation
* Use ADC to Temp conversion table
* Needs IRQ initialization in intel_msic.c to load
Change-Id: I506254d12e4edf169e426ac8b02acf9281799bd0
Signed-off-by: Durgadoss R <durgadoss.r@intel.com>
Reviewed-on: http://android.intel.com:8080/26761
Reviewed-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Reviewed-by: Mitra, DebalinaX <debalinax.mitra@intel.com>
Tested-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ivan Micovic [Wed, 14 Dec 2011 17:06:36 +0000 (18:06 +0100)]
[PORT FROM R2] atomisp: ANR threshold set to correct value of 20
BZ: 18089
After testing ANR threshold is decided to be 20
Change-Id: Ie3f16f579c7b8ddb56c9528dba71ed4358057688
Orig-Change-Id: Ied61d89d618cd433f7259c382926c450b1431b45
Reviewed-on: http://android.intel.com:8080/28067
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Sutic, Aleksandar <aleksandar.sutic@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-on: http://android.intel.com:8080/29601
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Micovic, Ivan <ivan.micovic@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Jozef Kruger [Thu, 15 Dec 2011 19:07:06 +0000 (11:07 -0800)]
[PORT FROM R2] atomisp: optimize DVS
BZ: 18089
This patch optimizes the DVS code.
Change-Id: I4e630e5687e47237edb94989d3887417f1f745a2
Orig-Change-Id: Ie3efeda5d26d4a6c4783d7d343607083d0b18bdd
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/27836
Reviewed-by: Bhat, Sheethal <sheethal.bhat@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-on: http://android.intel.com:8080/29600
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Lokesh Gupta [Thu, 15 Dec 2011 15:56:51 +0000 (16:56 +0100)]
[PORT FROM R2] atomisp: css/fw release from baseline of
12dec2011
BZ: 18089
This release provides the following fixes:
- Add cache control to acceleration API
- Fix XNR bug
Change-Id: I0185d5624362618ea019f267aa6e7100c9b12341
Orig-Change-Id: I7fd0c5b40a4ab97a475a5e8ee93417333761dcb9
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/27653
Reviewed-by: buildbot <buildbot@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-on: http://android.intel.com:8080/29599
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Tao Jing [Mon, 5 Dec 2011 08:15:05 +0000 (16:15 +0800)]
[PORT FROM R2] atomisp: restore default cc and ctc table
BZ: 18089
the default cc and ctc table should be used when
macc effect is enabled.
Change-Id: I87c040e043e160dcedd50b2c32dc6d8e6cce3f9d
Orig-Change-Id: Ieeafa2aea0a458683913062eaef0426eb5e462bc
Signed-off-by: Tao Jing <jing.tao@intel.com>
Reviewed-on: http://android.intel.com:8080/27282
Reviewed-by: buildbot <buildbot@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Gerasimow, Timo <timo.gerasimow@intel.com>
Reviewed-by: Xia, Yu Y <yu.y.xia@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-on: http://android.intel.com:8080/29598
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Yoshio Wada [Tue, 13 Dec 2011 08:37:47 +0000 (09:37 +0100)]
[PORT FROM R2] atomisp: mt9e013: fix the apparent different FOV for 720p
BZ: 18089
sensor settings for 720p was wrong.
output size is smaller than reqiurement.
This is a quick and dirty hack that adds special case for
this particular problem in the resolution selection function.
Change-Id: I540f965e5b0301ed809fbcb0a5e334bad8472c6e
Orig-Change-Id: Icfc7061585a02fbf77f4be92c3a77a0f9d64c3ba
Signed-off-by: Lasse Kantola <lasse.kantola@intel.com>
Reviewed-on: http://android.intel.com:8080/27151
Reviewed-by: buildbot <buildbot@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Von Oerthel, Jurgen <jurgen.von.oerthel@intel.com>
Reviewed-by: Kaikumaa, Timo <timo.kaikumaa@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-on: http://android.intel.com:8080/29597
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Huang Shenbo [Tue, 13 Dec 2011 07:52:15 +0000 (15:52 +0800)]
[PORT FROM R2] mt9m114: add awb status checking for 3a stabilization
BZ: 18089
Also check if awb is steady during 3a stabiliztion checking.
Change-Id: Ifcd0718318d49c167644965036a6c472082452d1
Orig-Change-Id: I8820274891e58a0fb211ebc54950dc9ae96d6c6c
Signed-off-by: Huang Shenbo <shenbo.huang@intel.com>
Reviewed-on: http://android.intel.com:8080/27130
Reviewed-by: buildbot <buildbot@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Tao, Jing <jing.tao@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-on: http://android.intel.com:8080/29596
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Tao Jing [Mon, 19 Dec 2011 07:06:44 +0000 (15:06 +0800)]
[PORT FROM R2] mrst.c: not power opt fail when GP_CAMERA_0_POWER_DOWN not available
BZ: 18089
On PR3, the GP_CORE_013 is not a connected pin but it is connected to
CAM_STBY_CTRL ON PR2. And it has been requested to remove it from PR3
SFI table as this GPIO Should not modified bu kernel from its original
state(GPIO input 20k Pull dow)
so this patch rework the mt9e013_power_ctrl: ipc_msic_vprog programming
is down whatever result of camera_sensor_gpio(GP_CAMERA_0_POWER_DOWN).
And calls to gpio_set_value() should depend on result of camera_sensor_gpio()
Change-Id: I9a44bdaf5a60d836ddeeac39175930b8cbc97f3f
Orig-Change-Id: I92ae6bdb9655b63233f10a4ad9aa5b216d484637
Signed-off-by: Tao Jing <jing.tao@intel.com>
Reviewed-on: http://android.intel.com:8080/29595
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Illyas Mansoor [Wed, 14 Dec 2011 06:40:00 +0000 (12:10 +0530)]
[PORT FROM R2] mid_pmu: check lss_08 audio driver to be in d0i0 for lpmp3 trigger
BZ: 17345
Current code implicitly assumes that LSS_08 is in D0i0 since S0ix command was
not possible, but it could so happen that LSS_08 or LSS_09 could be in D0i0 and
we could potentially enter LPMP3 mode.
This is incorrect, we need to make sure we enter LPMP3 IFF LSS_08 is in D0i0.
Fixed this by checking for D0i0 in target lpmp3 check.
Regarding LSS 09 this is audio DMA driver this is okay to be in D0i3, for LPMP3
to happen.
Change-Id: Ia5d3b2ff6ccf10f97021afd07f525834f5546495
Orig-Change-Id: I095569fe0c75021a824e94c0ed0b67b38f4b58ef
Signed-off-by: Illyas Mansoor <illyas.mansoor@intel.com>
Reviewed-on: http://android.intel.com:8080/29310
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Dyut Kumar Sil [Mon, 28 Nov 2011 10:46:23 +0000 (16:16 +0530)]
[PORT FROM R2] mid_pmu: create a sysfs entry to set all the south complex devices in D0i0
BZ: 16961
This patch create a sysfs entry /sys/module/mid_pmu/parameters/pmu_driver_state
Doing, echo "off" > /sys/module/mid_pmu/parameters/pmu_driver_state will
set all the south complex devices in D0i0 state and any further request
to set any south complex device in some other low power state will
be ignored.
Removed the code corresponding to echo "on" path as we are always supposed
to do reboot after issuing echo "off".
Change-Id: I2f75b9fbb042218105aa58c5d709af4c8f0d6754
Orig-Change-Id: I94f1c85ee6cd44b74c455d0dd5a7b35e1e86b512
signed-off-by: Dyut Kumar Sil
Reviewed-on: http://android.intel.com:8080/29309
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: Mansoor, Illyas <illyas.mansoor@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Ramachandra Sudarshan N [Fri, 2 Dec 2011 08:38:06 +0000 (14:08 +0530)]
[PORT FROM R2] mid_pmu: serialize access to 0xD0 and 0xD4 regs of root device (bus 0, slot 0)
BZ: 6332
Apart from mid_pmu there are other drivers which read/write
0xD0 and 0xD4 registers in config space of pci root device hence
there is a potential for race condition. This patch resolves the
issue by creating an API which serializes access to the above registers.
other drivers like coretemp.c, atomisp are accessing 0xD0 and 0xD4
registers. mid_pmu uses it to map north complex PM registers, which will be
in turn used in pmu_nc_set_power_state. where as coretemp.c is using it to
read thermal sensor registers. hence used for various purpose in
different drivers
once this patch gets merged the plan is to ask individual driver
owners to use this API instead of directly accessing the above registers.
Change-Id: I57a220c79283c20ded693cf6c5c94720bf428360
Orig-Change-Id: Ib31ae2cf591749b9693dbaf16b1fee3c4120bf66
Signed-off-by: Sudarshan Ramachandra <sudarshan.n.ramachandra@intel.com>
Reviewed-on: http://android.intel.com:8080/29308
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: Mansoor, Illyas <illyas.mansoor@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Illyas Mansoor [Fri, 2 Dec 2011 19:41:25 +0000 (01:11 +0530)]
[PORT FROM R2] i2c-designware-pci: increase runtime_pm autosuspend delay to 50msec
BZ: 16187
i2c-designware has runtime autosuspend delay set
at 5msec which causes lot of suspend/resume trash.
With pmu_sss_states we can clearly see the trash
for I2C0 and I2C2 devices with repeated touching device
we see a count of around 1400 for I2C0.
Changed the autosuspend delay to 50msecs this reduced
the trashing considerably to around 29 for I2C0 and 3 for
I2C2
Change-Id: Ia0f6f908ef1c2a87672f10fd62549b5036b77d41
Orig-Change-Id: Icc7018da6542750739bb932851aecdd3596561b1
Signed-off-by: Illyas Mansoor <illyas.mansoor@intel.com>
Reviewed-on: http://android.intel.com:8080/29307
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Jouni Hogander [Fri, 4 Nov 2011 07:52:09 +0000 (09:52 +0200)]
[PORT FROM R2] mid_pmu: Use timeout when waiting for interactive command to complete
BZ: 13779
Currently possible hang exists in mid_pmu code wait for interactive
command complete handling. In some error cases interrupt indicating
command completion doesn't occure and this causes hang without any
error messages.
We don't want to continue when hang happens because it's fatal
error. Instead of quiet hang dump out some useful information and make
it clear where hang happened.
Change-Id: I3d9356221ae7186a59578de7039ab84735178a31
Orig-Change-Id: I31c6ce97a22e6673aff892d3f4b65d8a57fd98b4
Signed-off-by: Jouni Hogander <jouni.hogander@intel.com>
Reviewed-on: http://android.intel.com:8080/29306
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: Mansoor, Illyas <illyas.mansoor@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Illyas Mansoor [Fri, 4 Nov 2011 08:10:30 +0000 (13:40 +0530)]
[PORT FROM R2] intel_idle: reduce s0i3 exit latency number
BZ: 13767
Current code has S0i3 exit latency set to 5000usecs
This is much higher than the measured latency which
is around 2800usecs
s0i3_entry_latency = 894.375usecs
s0i3_exit_latency = 1882usecs
Need to reduce this to 2800usecs
Change-Id: I51d9a111c427483a115833aa19a536c44cca563b
Orig-Change-Id: If5d0f0de1a8cb66617fb0423190fe51188bdf422
Signed-off-by: Illyas Mansoor <illyas.mansoor@intel.com>
Reviewed-on: http://android.intel.com:8080/29305
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Illyas Mansoor [Fri, 11 Nov 2011 13:13:45 +0000 (18:43 +0530)]
[PORT FROM R2] intel_idle: re-introduce c6 to c4 demotion
BZ: 14243
In patch #19777 BZ 7240 we reasoned that using set_s0ix_complete implementation
will effectively remove the need for C6 to C4 demotion.
But based on BZ 12588 analysis and conclusion that C6 to C4 offload is still
required in some cases even thou set_s0ix_complete() is called suggests that
we need to get back C6 to C4 demotion logic and also get back the fix for LPMP3
BZ 5212 power regression. Also in case we demote we dont have to increment/decrement
C6 counter.
This patch fixes this scenario:
1. s0i3 command issued in MID_PMU
2. c6 offload bit not set
3. c4 demoted but need_reschdule was set so no mwait executed.
4. idle: c6 latency found so soc_s0ix_idle get called with eax==C6_HINT
5. in this case mfld_s0ix_enter not called
6. need_reschudle now not true so mwait(c6) issued.
Hence will have to get back the c6 to c4 demotion logic.
Change-Id: Ib35e2756a3d99e30418869b84f8befb7b8faa98e
Orig-Change-Id: Ic7740686c45679438154b456a06f99f947577eed
Signed-off-by: Illyas Mansoor <illyas.mansoor@intel.com>
Reviewed-on: http://android.intel.com:8080/29304
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Yanmin Zhang [Fri, 4 Nov 2011 07:54:41 +0000 (15:54 +0800)]
[PORT FROM R2] mid_pmu: soc_s0ix_idle shouldn't try C6 when cpu offload bit not set
BZ:12588
1) Sometimes, mfld_s0ix_enter prints error cpu offload bit not set. Later on,
soc_s0ix_idle shouldn't try C6. We demote it to C4.
2) mfld_s3_enter should call pmu_set_s0ix_complete when mfld_s0ix_enter fails.
Change-Id: Iec1f85f69a29d658463f0901d8c3b2b0a32bf5b5
Orig-Change-Id: I639cb30b02754fb9ab4f2915ef468778597c2a72
Signed-off-by: Yanmin Zhang <yanmin.zhang@intel.com>
Reviewed-on: http://android.intel.com:8080/29303
Reviewed-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: Mansoor, Illyas <illyas.mansoor@intel.com>
Tested-by: Martin, LoicX <loicx.martin@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Durgadoss R [Thu, 24 Nov 2011 09:18:13 +0000 (14:48 +0530)]
EM/THERMAL: Enable Platform Thermal Driver in K3.0
BZ: 15135
This patch enables the Medfield thermal driver on
k3.0. Since this driver depends on GPADC driver and
Thermal Frameowrk, this patch enables them also.
Change-Id: I9f8e832df03d3fdc420df2d5ce826d4b2e31d3d8
Signed-off-by: Durgadoss R <durgadoss.r@intel.com>
Reviewed-on: http://android.intel.com:8080/26758
Reviewed-by: Gross, Mark <mark.gross@intel.com>
Tested-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Reviewed-by: Mitra, DebalinaX <debalinax.mitra@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Durgadoss R [Tue, 6 Dec 2011 16:08:45 +0000 (21:38 +0530)]
EM/THERMAL: Initialize irq for msic_thermal in intel_msic.c
BZ: 15135
This patch adds the irq entry necessary for platform
thermal driver. Since there is no IRQ for this driver,
0xff is assigned. This is due to change in mrst.c from
K2.6.35 to K3.0. Without this change the intel_mid_thermal.c
driver will not load in K3.0.
Change-Id: Idd9a88c026a95623cfe7de79b722b3138fca2b71
[Ported to R3]Change-Id: Ifdceafacf40fa9a795bd8c52172985071b578d29
Signed-off-by: Durgadoss R <durgadoss.r@intel.com>
Reviewed-on: http://android.intel.com:8080/29249
Reviewed-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Tested-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Durgadoss R [Mon, 5 Dec 2011 08:53:29 +0000 (14:23 +0530)]
EM/THERMAL: Port coretemp driver to k3.0
BZ: 16290
This patch ports the coretemp driver to k3.0
This driver was completely re-written between
2.6.35 and now, to add Pkgtemp support. Hence,
there are many changes.
Change-Id: I28ec8f23dff977d405d1afcbc84914581e6f9219
[Ported to R3]Change-Id: I96f096b51ead78f05aafb4e2752a3cd367475108
Signed-off-by: Durgadoss R <durgadoss.r@intel.com>
Reviewed-on: http://android.intel.com:8080/29232
Reviewed-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Tested-by: Gugwad, SantoshX <santoshx.gugwad@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
santosh [Tue, 20 Dec 2011 04:20:20 +0000 (09:50 +0530)]
EM/THERMAL: Enable Coretemp driver in ICS/k3.0
BZ: 16290
This patch enables the coretemp driver in ICS/k3.0.
Change-Id: I5cebe1e3edf904e3ef48d3c3f8f677066b1f44af
Signed-off-by: santosh <santoshx.gugwad@intel.com>
Reviewed-on: http://android.intel.com:8080/29388
Reviewed-by: R, Durgadoss <durgadoss.r@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
David Cohen [Wed, 30 Nov 2011 14:31:35 +0000 (16:31 +0200)]
[PORT FROM R2] atomisp: hmm: use vmalloc() to allocate large memory
BZ: 17919
When large memory area is required on alloc_[user/private]_pages()
functions, kmalloc() may get in trouble if free memory is low and
fragmented. vmalloc() is able to request fragmented memory and make it
virtually contiguous.
As vmalloc'ed memory size is PAGE_SIZE aligned, this patch defined a
threshold:
- If requested mem size is > PAGE_SIZE, vmalloc() is used.
- If mem size is <= PAGE_SIZE, kmalloc() is used instead.
Change-Id: I5b634529de7366642ab1f3090b731930bcfb945d
Orig-Change-Id: I4b1b832b7552603b981ad1017ddf72caddf7ea25
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Tested-by: Ziyu Jiang <ziyux.jiang@intel.com>
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/27155
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/29283
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
ziyux.jiang [Tue, 13 Dec 2011 16:12:07 +0000 (00:12 +0800)]
[PORT FROM R2] atomisp: fix alloc pages block issue in drivers
BZ: 17919
when there is not enough memory in kernel, allocation pages may fails
and then sleep, so system will block in kernel and can not exit. To
avoid this case, we modify allocation pages flag to
GFP_NOWAIT(not will sleep when allocation) | __GFP_NOWARN on allocating
private pages, and modify memory allocation mechanism.
Change-Id: Ia60ee672920ecc6d5ef308903a8b51b241c10625
Orig-Change-Id: I7e8dd2f6fcacadb307eac925028d27934197c67c
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/27152
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/29282
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Jozef Kruger [Tue, 13 Dec 2011 20:09:48 +0000 (12:09 -0800)]
[PORT FROM R2] atomisp: use cached buffers for DIS statistics
BZ: 17919
This patch enables cached buffers for DIS statistics instead of
uncached ones. This speeds up the reading of DIS statistics a lot.
Change-Id: Ic70ba8d9c0dc10a66d1ac15529253af01bbcbecd
Orig-Change-Id: Iaa54af9e38e364a267d124b84d38c6772a8855c3
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/27232
Reviewed-by: Bhat, Sheethal <sheethal.bhat@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Le Thenaff, Erwan <erwan.le.thenaff@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/29281
Reviewed-by: Lampila, KalleX <kallex.lampila@intel.com>
Leo Yan [Mon, 19 Dec 2011 07:23:10 +0000 (15:23 +0800)]
ltr502als.c: Proximity status check after mode switching in early suspend
BZ: 18010
If proximity status is changed during mode switch, there
may be no interrupt occurred, so we need to check the data
register after mode switching in the early suspend handler.
Change-Id: If0e9948568ce8320ae2efbee3fb4cd0343395b2d
Signed-off-by: Leo Yan <leo.yan@intel.com>
Reviewed-on: http://android.intel.com:8080/28059
Reviewed-by: Liu, Hong <hong.liu@intel.com>
Reviewed-by: Du, Alek <alek.du@intel.com>
Tested-by: Wang, Zhifeng <zhifeng.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Bin Gao [Fri, 16 Dec 2011 10:37:11 +0000 (02:37 -0800)]
usb: disable SRAM usage temporarily for CLV
BZ: 17299
HACK: Enabling SRAM usage in USB OTG driver would make USB OTG unstable during
large file transfer. For now, let's disable SRAM for Cloverview until this
issue is fully root caused and resolved.
Change-Id: Iccd5200b6a6b62798b5b3be970bca37aac4b6169
Signed-off-by: Prabu Renganathan <prabu.renganathan@intel.com>
Signed-off-by: Bin Gao <bin.gao@intel.com>
Reviewed-on: http://android.intel.com:8080/27226
Reviewed-by: Zhuang, Jin Can <jin.can.zhuang@intel.com>
Reviewed-by: Meng, Zhe <zhe.meng@intel.com>
Tested-by: Meng, Zhe <zhe.meng@intel.com>
Reviewed-by: Tang, Richard <richard.tang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Bin Gao [Thu, 15 Dec 2011 18:31:04 +0000 (10:31 -0800)]
usb: add usb OTG CLV support
BZ: 17298
For Clovertrail, USB PHY is removed from MSIC and a standalone ULPI PHY
chip is used instead. Corresponding otg driver change is as following:
1. Get rid of any access to MSIC
2. Control VBUS by PHY
3. Enable wake up detection by ULPI registers rather than MSIC registers
4. Control the PHY chip by two GPIO pins (reset, power on/off, etc.)
5. Disable ACA detection for clovertrail in driver, will do it later.
6. Disable usb charger detection. A new charger detection implementation
is needed per the change of PMIC. Will do it later.
Change-Id: Ie44d82544e1aebf398010d595da2ad4e3d8ade75
Signed-off-by: Bin Gao <bin.gao@intel.com@intel.com>
Signed-off-by: Prabu Renganathan <prabu.renganathan@intel.com>
Reviewed-on: http://android.intel.com:8080/27225
Reviewed-by: Zhuang, Jin Can <jin.can.zhuang@intel.com>
Reviewed-by: Gao, Bin <bin.gao@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Meng, Zhe <zhe.meng@intel.com>
Tested-by: Meng, Zhe <zhe.meng@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Fabien Barthes [Wed, 14 Dec 2011 17:38:13 +0000 (18:38 +0100)]
[PORT FROM R2] [Workaround][Audio] Reduce pmdown_time to 1 second
BZ: 14981
ASoc core powering down delay is reduced to 1 sec. The switching time
was measured in the driver to be around 110ms for switching b/w HS
and IHF, so 1 second delay is sufficient.
Change-Id: Id877db188f7fb03223ee3a0c7e6f69834b02527b
Old-Change-Id: Ib4528a9af4b5d0b6fa37c847d0ec484ec5e1e11d
Signed-off-by: Fabien Barthes <fabienx.barthes@intel.com>
Reviewed-on: http://android.intel.com:8080/29238
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Bruce Beare [Wed, 14 Dec 2011 22:27:27 +0000 (14:27 -0800)]
screen: Set correct physical display dimensions for PR3
BZ: 16083
Physical dimensions were set to the CDK values. They are changed
to the correct values for the PR3.
Change-Id: I0ff78e53de7e86ad00a46e087b71545d1bb8ec21
Author: Mark Horn <mark.d.horn@intel.com>
Signed-off-by: Bruce Beare <bruce.j.beare@intel.com>
Reviewed-on: http://android.intel.com:8080/27366
Reviewed-by: Horn, Mark D <mark.d.horn@intel.com>
Reviewed-by: Gross, Mark <mark.gross@intel.com>
Tested-by: Li, XiaojingX <xiaojingx.li@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Olivier Stoltz Douchet [Thu, 15 Dec 2011 17:45:33 +0000 (18:45 +0100)]
hsi_ffl_tty.c: Adding a WAKE post-boot handshake
BZ: 17580
This patch is adding an ACWAKE post modem boot handshake during the TTY
activate callback to ensure that the modem has boot-up prior to have the TTY
interface up and running.
This handshake is based on the fact that the 6260 modem is asserting its CAWAKE
line after it has boot whenever the CAWAKE line is asserted. This patch is
simply asserting the ACWAKE line and waiting for the CAWAKE line at the very
end of the activate TTY callback function. This way the modem is always up and
ready when the TTY interface is active, shall it be at boot time or further to
a reset.
Change-Id: I67385f754cc845997145deca70fc6035880c2fb2
Signed-off-by: Olivier Stoltz Douchet <olivierx.stoltz-douchet@intel.com>
Reviewed-on: http://android.intel.com:8080/27670
Reviewed-by: Gross, Mark <mark.gross@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Akue, LoicX <loicx.akue@intel.com>
Reviewed-by: Morel, YannickX <yannickx.morel@intel.com>
Reviewed-by: Predon, Frederic <frederic.predon@intel.com>
Reviewed-by: Lucas, GuillaumeX <guillaumex.lucas@intel.com>
Reviewed-by: Robert, Denis <denis.robert@intel.com>
Tested-by: Robert, Denis <denis.robert@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Alek Du [Mon, 19 Dec 2011 05:56:16 +0000 (05:56 +0000)]
[PORT FROM R2] hsu driver sync with R2
BZ: 17765
R2 and R3 HSU driver diffs a little bit. This patch
will sync up them.
Change-Id: I7491b917a192dac497deee6af9c177ff8833e243
Signed-off-by: Alek Du <alek.du@intel.com>
Reviewed-on: http://android.intel.com:8080/27899
Tested-by: Wang, Zhifeng <zhifeng.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
David Cohen [Fri, 25 Nov 2011 13:32:20 +0000 (15:32 +0200)]
[PORT FROM R2] atomisp: (re)start ISP and handle frame buffers from ISR
BZ: 17272
Atomisp driver needs to manage ISP and frame buffers in low latency way
to allow camera to work with high fps. This patch makes ISR to (re)start
ISP whenever it's possible and handle frame buffers.
Change-Id: I01b098937b9174a48bd97af216cb7b3c01b24c60
Orig-Change-Id: Id2bd9a7aa5dd4948350d442d72b9c1ca16cb0bb6
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/26723
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28058
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
David Cohen [Mon, 12 Dec 2011 19:41:20 +0000 (21:41 +0200)]
[PORT FROM R2] atomisp: hmm: hmm_bo cannot use mutex anymore
BZ: 17272
hmm_bo interface can now be called from interrupt context. This patch
replaces mutex by spinlock to avoid issues in atomic context.
Change-Id: I20cecd2c6c5cb61179ae8757bea37b97f204ff8b
Orig-Change-Id: I0edc7d054af8c2b96c5c9b67a4edccab49af489f
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/27075
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Le Thenaff, Erwan <erwan.le.thenaff@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28057
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
David Cohen [Thu, 8 Dec 2011 09:36:24 +0000 (11:36 +0200)]
[PORT FROM R2] atomisp: create proper interface for buffer management
BZ: 17272
Atomisp does buffer management inside its workqueue.
As ISR will need to do same task, this patch adds a proper interface for
it and remove embedded code from workqueue.
Change-Id: Ic37aa53e3504922dc63155dc736cc04cf437386e
Orig-Change-Id: I2b5e5fa0b199d37ff16d931752d6ba7bd38a424d
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/26722
Reviewed-by: Le Thenaff, Erwan <erwan.le.thenaff@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28056
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
David Cohen [Wed, 7 Dec 2011 16:37:27 +0000 (18:37 +0200)]
[PORT FROM R2] atomisp: move watchdog away from workqueue
BZ: 17272
Atomisp has an embedded watchdog inside main workqueue.
As ISR is going to handle ISP/buffers, watchdog cannot rely anymore on
workqueue to handle it. This patch creates a proper interface for the
software watchdog and makes it to be handled from ISR, which is the
proper place to track if ISP is alive.
Change-Id: I90c42fb292a9bab9ad036d3dfbd1acbf39adad79
Orig-Change-Id: I551876cc49ef539e5fac871e6ca1a2336aada60d
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/26721
Reviewed-by: Le Thenaff, Erwan <erwan.le.thenaff@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28055
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Huang Shenbo [Fri, 2 Dec 2011 04:11:11 +0000 (12:11 +0800)]
[PORT FROM R2] mt9m114: add 480P settings for soc sensor
BZ: 17272
Need to add 480P settings as required by video app.
This patch adds the 480P settings in soc driver.
Change-Id: I2057b3318c75a129427f9e2978b9d6d9a65b990a
Orig-Change-Id: If8ac3b2da3070cab87978e04acc4cf9485691abf
Signed-off-by: Huang Shenbo <shenbo.huang@intel.com>
Reviewed-on: http://android.intel.com:8080/26680
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28054
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Vinod Govindapillai [Fri, 25 Nov 2011 13:21:52 +0000 (15:21 +0200)]
[PORT FROM R2] atomisp: Fix for Unload firware causing ISP Timeout
BZ: 17272
When a firmware is marked to be unloaded, it is unloaded during
the execution of WQ. WQ call be triggered for the ISP intermediate
stages well.
For a safe unloading of accelerations, the best place would
be after the FRAME DONE interrupt. Once the FRAM DONE
interrupt is received, it is guranteed that not more firmware
are executing. So moving the unlading part after the FRAME DONE
interrupt. In this way it is safer even in case of multiple
acceleration firmwares as well.
Change-Id: I371cd1fabc301fab8071b4840c0983a7db128133
Orig-Change-Id: I0760af3d1522a6cf9b8f43b4cdba6ea72e00d543
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-on: http://android.intel.com:8080/26679
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28053
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Hu Gang [Thu, 8 Dec 2011 08:50:06 +0000 (16:50 +0800)]
[PORT FROM R2] Revert "atomisp: fix alloc pages block issue in driver."
BZ: 17272
Issues:
Phone reset in low memory condition when doing burst capture.
Solution:
Suspect the patch in memory allocation in atomisp driver.
So revert this patch first.
Change-Id: Id7a985e86481a1de01a4b7e3360a61dfc342f64c
Orig-Change-Id: I8d49403963a8c2c7e9c097b60f439f3458bf2756
Signed-off-by: Hu Gang <gang.a.hu@intel.com>
Reviewed-on: http://android.intel.com:8080/26678
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28051
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
ziyux.jiang [Wed, 30 Nov 2011 14:30:53 +0000 (22:30 +0800)]
[PORT FROM R2] LM3554: limit lm3554 flash's current.
BZ: 17272
to prevent battery over-current, we set peak current limit
to be 1000mA.
Change-Id: I67032019b1f6ae8de030faba8ef4c351cb2cc8d8
Orig-Change-Id: I6292cd6d20558fd5589b8c11d2448c46d3e18b0e
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/25677
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28050
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
ziyux.jiang [Tue, 29 Nov 2011 11:26:53 +0000 (19:26 +0800)]
[PORT FROM R2] atomisp: fix alloc pages block issue in driver.
BZ: 17272
when there are not enough memory in kernel, allocation pages maybe fails and
then sleep, so system will block in kernel and can not exit. to avoid this
case, we modify allocation pages flag to GFP_NOWAIT(not will sleep when
allocation). and modify allocation mechanism.
Change-Id: If5362583197a650bcb6d05b1a811d2a18967468c
Orig-Change-Id: I7e687b1a9107ebeedb71f6ce84a25b19169520af
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/25504
Reviewed-by: Zheng, ZixiangX <zixiangx.zheng@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28049
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Robert Jong-A-Lock [Tue, 6 Dec 2011 00:03:08 +0000 (01:03 +0100)]
[PORT FROM R2] atomisp: mt9e013: updated sensor settings (4)
BZ: 17272
removed maximised VBI for wide preview. caused timeout in css
for yet unexplained reason.
Change-Id: I9f2cbd235632b4a9d00a421d3876a7807569e440
Orig-Change-Id: Ia7a02855706f64657e701048d955fd660edf148b
Reviewed-on: http://android.intel.com:8080/26146
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Le Thenaff, Erwan <erwan.le.thenaff@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28048
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Jozef Kruger [Wed, 30 Nov 2011 20:23:47 +0000 (12:23 -0800)]
[PORT FROM R2] atomisp: remove unused types from atomisp.h
BZ: 17272
This patch removes types from atomisp.h that are not used in atomisp.h itself.
The use of the atomisp_frame structure is replaced by v4l2_framebuffer which was
introduced in v4l2 for exactly this use case. Also, since this is a v4l2 driver,
we should use v4l2 constructs for frames, not introduce our own.
This makes the atomisp interface smaller and easier to document.
Change-Id: I819ba895ec9042145252eb0e869bd7109b3b5ae8
Orig-Change-Id: I2ffd39c3566bf86148dbdaaafe372ae15df1f345
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/25736
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28047
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Michie Yamauchi [Thu, 24 Nov 2011 19:11:08 +0000 (20:11 +0100)]
[PORT FROM R2] atomisp: Gamma correction parameters (K1,K2)
BZ: 17272
atomisp_cmd.c: The atomisp_gamma_correction function is added
to update Gamma correction parameters(K1,K2).
sh_css_params.c: Gamma correction parameters(K1,K2) are passed to ISP.
Change-Id: I94649f83cdc795aa66f6201d02d9f1cb2c6e0be8
Orig-Change-Id: I2de4963a22be9ce6be9f42d65d869e1b62385c5c
Signed-off-by: Michie Yamauchi <michie.yamauchi@intel.com>
Reviewed-on: http://android.intel.com:8080/25245
Reviewed-by: Gupta, Lokesh <lokesh.gupta@intel.com>
Reviewed-by: Tuppurainen, Miikka <miikka.tuppurainen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28046
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Robert Jong-A-Lock [Wed, 30 Nov 2011 15:41:34 +0000 (16:41 +0100)]
[PORT FROM R2] atomisp: mt9e013: updated sensor settings (3)
BZ: 17272
Maximised VBI settings for VGA, QVGA and QCIF video modes
Change-Id: Ifa7bb3b0d5d109cf58a476afb81f9ec7898812ac
Orig-Change-Id: Id13f23945a97ce71ba04275a0674e60ba982c1df
Reviewed-on: http://android.intel.com:8080/25715
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28044
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Robert Jong-A-Lock [Wed, 30 Nov 2011 11:24:16 +0000 (12:24 +0100)]
[PORT FROM R2] atomisp: mt9e013: updated sensor settings (2)
BZ: 17272
fix FoV issue with QVGA and QCIF. QVGA now has a FoV of 78% and
QCIF 82% horizontally. VGA, QVGA and QCIF do not use skipping
anymore, only binning and sensor scaling.
Change-Id: I109e0cf7839734ce005ee17861ff498fbd9f2897
Orig-Change-Id: Iecf6b2610fc82d2bddb3c1995340343cfab14913
Reviewed-on: http://android.intel.com:8080/25697
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28043
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Robert Jong-A-Lock [Tue, 29 Nov 2011 22:43:31 +0000 (23:43 +0100)]
[PORT FROM R2] atomisp: mt9e013: updated sensor settings (1)
BZ: 17272
according to aptina, skipping shoudl not be used. it causes color
differences between modes that use it and modes that do not. This
patch should remove the color differenc ebetween 4:3 preview and
4:3 still modes.
8M resolution now supports 15fps.
Change-Id: I7dbc9874ac008e207d49184bdba0161dc73d7a65
Orig-Change-Id: I4c751838cbc0070537ea2dd2168ef28567efa7d4
Reviewed-on: http://android.intel.com:8080/25638
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Kantola, Lasse <lasse.kantola@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28042
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Vitali Samurov [Mon, 31 Oct 2011 11:57:42 +0000 (13:57 +0200)]
[PORT FROM R2] The slew rate of VC motor current rising time has changed.
BZ: 17272
Currently VC motor has '0' in both configuration registers: 'slew_rate' and 'vcm_step_time'.
Therefore a current in the VC motor rises very fast (instantaneous) and it causes a big
mechanical impact to the lens unit, audible as a loud click.
This patch makes VC current rising slope slightly inclined and it makes a lens movement to be
more smooth. Patch should not affect to the speed of AF, because a duration of single Af step
remains the same.
Change-Id: Ibda9f7af5cdd74287bb25ddb83c53fa4dbaf1248
Orig-Change-Id: Ie35b30cafa9848a7175798a3832bd06be511aec8
Signed-off-by: Vitali Samurov <vitali.samurov@intel.com>
Reviewed-on: http://android.intel.com:8080/22776
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28041
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Lokesh Gupta [Wed, 30 Nov 2011 09:20:44 +0000 (10:20 +0100)]
[PORT FROM R2] atomisp: CSS release Synergy BL 29nov2011
BZ: 17272
This is the CSS release from Synergy baseline of 29nov2011.
It fixes the following issues:
- Standalone acceleration firmware has no support
to get its completion interrupt translated
- ISP parameter for Gamma correction (K1,K2)
should be tunable from ISP parameter file
- ISP BNR/CNR
- MACC table index order difference between ISP
and USER
- MACC table segment selection problem
Change-Id: I2b29127df3f98865703bf0c35a054f565357143f
Orig-Change-Id: I2eb9ad577e6949ceb1eebbf0666d8450b4375a1c
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/25729
Reviewed-by: Yamauchi, Michie <michie.yamauchi@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28039
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Jozef Kruger [Tue, 29 Nov 2011 18:40:57 +0000 (10:40 -0800)]
[PORT FROM R2] mt93013: remove compiler warnings
BZ: 17272
This patch removes a compiler warning from the mt93013 driver.
Change-Id: Icd21d170ff37d9d167c2dc078ae08507158c9740
Orig-Change-Id: I4fddefa3d2e759f9b204f0fc68805b58cc64b6de
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/25630
Reviewed-by: Jong-a-lock, Robert <robert.jong-a-lock@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Tested-by: Ilvonen, Ville <ville.ilvonen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28037
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Jozef Kruger [Tue, 29 Nov 2011 18:46:35 +0000 (10:46 -0800)]
[PORT FROM R2] mt9m114: remove compiler warnings
BZ: 17272
This patch removes compiler warnings from teh mt9m114 code.
Change-Id: Iba54e92b3441720749a70fd3f6f9d871bf509875
Orig-Change-Id: I3a735ea1b4001a32465fcd1cc13f6feb4493443e
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/25631
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28036
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Huang Shenbo [Fri, 25 Nov 2011 06:26:17 +0000 (14:26 +0800)]
[PORT FROM R2] mt9m114: add speedup ae/awb settings and change loading sequence
BZ: 17272
1. Add speedup ae/awb settings to shorten the ae/awb
initialization time leading to less s2s time.
2. Change setting loaded sequence per vendor suggestion.
3. Add QCIF resolution settings.
Change-Id: I7c9009a5701e1d2f5149b0c584d5e2ff398ecba7
Orig-Change-Id: I4904679567f5acdf0ba4cb52b93094f63c875fac
Signed-off-by: Huang Shenbo <shenbo.huang@intel.com>
Reviewed-on: http://android.intel.com:8080/25283
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28035
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tao Jing [Tue, 29 Nov 2011 07:13:33 +0000 (15:13 +0800)]
[PORT FROM R2] atomisp: fix error put_page when get_user_pages failed
BZ: 17272
when get_user_pages failed with -EFAULT(-14), put_page would be run
for bo->pages[i], which causes kernel hang.
This patch fix this issue.
Change-Id: I83c2d3c79168d4153733eae5cc83ba1ddfd60402
Orig-Change-Id: I1e2908929b9372e704f24cde99f4ebeba6820b22
Signed-off-by: Tao Jing <jing.tao@intel.com>
Reviewed-on: http://android.intel.com:8080/25514
Reviewed-by: Jiang, ZiyuX <ziyux.jiang@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28034
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Andy Shevchenko [Fri, 25 Nov 2011 09:11:25 +0000 (11:11 +0200)]
[PORT FROM R2] atomisp: clean up media entities at exit
BZ: 17272
There are two places where we have memory leaks (found by kmemleak). This
change eliminates both of them.
Change-Id: If61ca8f85caf7a77407417f747ce0c781253caea
Orig-Change-Id: I8eb84e5261b2790561aca774681f77dddd2e7697
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25442
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28033
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Andy Shevchenko [Fri, 25 Nov 2011 09:11:24 +0000 (11:11 +0200)]
[PORT FROM R2] atomisp: remove useless assignment in ->remove()
BZ: 17272
There is no need to do assignment in structure that will be freed anyway.
Change-Id: I92a5a4b2442995537790664732b1891339ed4ba9
Orig-Change-Id: I93430f6d8b8bfd7fdcf5fb47da74729d26733e67
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25441
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28032
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Andy Shevchenko [Fri, 25 Nov 2011 09:11:23 +0000 (11:11 +0200)]
[PORT FROM R2] atomisp: remove unnecessary assignment
BZ: 17272
There is unnecessary to assign the vfl_type in the atomisp_video_init() because
the video_register_device() takes the type as mandatory argument.
Change-Id: Ib076eaedb0171b04b8a19928fd530a730c3693d1
Orig-Change-Id: Ieb1df952b1c4af028e3286d24d06179da98619bb
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25440
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28031
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Andy Shevchenko [Fri, 25 Nov 2011 09:11:22 +0000 (11:11 +0200)]
[PORT FROM R2] atomisp: fix release function
BZ: 17272
'rmmod atomisp' will generate an oops on the current kernel.
The struct atomisp_video_pipe contains the struct video_device in-place. During
initialization of the pipes, each of them takes the video_device_release() as a
->release() callback. Due to this the video_unregister_device() will call the
kfree() on non-allocated memory.
In our case we should use the video_device_release_empty() instead of the
video_de3vice_release().
Change-Id: I9bd12d0b087267b3ee242ea60cd62c37a5639be4
Orig-Change-Id: Ib243c5c48b7c085d11c8ef44bb5c4a33bad1374f
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25439
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28030
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Andy Shevchenko [Fri, 25 Nov 2011 09:11:21 +0000 (11:11 +0200)]
[PORT FROM R2] atomisp: free file_input resources on exit
BZ: 17272
atomisp_unregister_entities() is dedicated to unregister entire set of
registered entities. However, there is file_dev entity which is not
unregistered by mentioned function. This patch adds absent line.
Change-Id: Iae86db4b0db87fea28f1df0ae54eac6da6ac5e2f
Orig-Change-Id: I81b906dabb5f46f4b36d78b68fdc4ddcbe36d8ad
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25438
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28029
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Teemu Tuominen [Wed, 14 Dec 2011 13:28:39 +0000 (15:28 +0200)]
atomisp: porting driver R2 fix-patches
BZ: 17272
Forward ported patch from r2-stable starts to threat warnings as errors
f271944ad8bcad5c4cbbd5b3a99f50f452da56b8 atomisp: remove compiler warnings
Cannot enable -Werror while tng disabled, see known issue:
27390f90d76fcce1cc59eb64b080e43237079163 atomisp: porting driver
Change-Id: I89c11f7c0b6918d356a367960b46094d96a6ad12
Signed-off-by: Teemu Tuominen <teemux.tuominen@intel.com>
Reviewed-on: http://android.intel.com:8080/28028
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Jozef Kruger [Tue, 29 Nov 2011 18:34:10 +0000 (10:34 -0800)]
[PORT FROM R2] atomisp: remove compiler warnings
BZ: 17272
This patch removes compiler warnings from the atomisp code.
Change-Id: I1ffa0f17a824207082f3da8b530f2da53aea4cc5
Orig-Change-Id: Ie60b94261d21337fe9922d7b7d5ad30609dce6d8
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/25628
Reviewed-by: Gupta, Lokesh <lokesh.gupta@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28027
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
David Cohen [Thu, 24 Nov 2011 14:48:51 +0000 (16:48 +0200)]
[PORT FROM R2] atomisp: mt9e013: fail OTP checksum if only 0's are found
BZ: 17272
If OTP data has only 0's checksum will pass, but data is invalid. This
patch makes it fail.
Change-Id: If461bb9b44736bfdcea96318c47bec185955e1db
Orig-Change-Id: I4a5cf7276ad44600fffbfce72e4cd50eb2d7c582
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25232
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Gerasimow, Timo <timo.gerasimow@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28026
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
David Cohen [Wed, 23 Nov 2011 06:32:16 +0000 (08:32 +0200)]
[PORT FROM R2] atomisp: mt9e013: add OTP checksum fail message
BZ: 17272
This patch adds OTP checksum fail message.
Change-Id: Icf21fa50859a391f2853c0abd5b6e9b9d6dc1b73
Orig-Change-Id: I3824d81b5fbf501f57b4380cd8281acd6e3d6cab
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/25231
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28025
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Lokesh Gupta [Tue, 22 Nov 2011 13:11:14 +0000 (14:11 +0100)]
[PORT FROM R2] atomisp: CSS/FW release Synergy BL
20111120
BZ: 17272
This release fixes the following issues:
- Images captured with night mode are abnormal
- CNR part of BZ14470 is fixed (BZ still open for BNR bug)
Change-Id: I1c62642b1ac2a1cb5736f496e0a41871a5cf6d65
Orig-Change-Id: I48e96ce83a86110d184da1cb355857017e5be53c
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/24895
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28024
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
David Cohen [Tue, 22 Nov 2011 18:20:17 +0000 (20:20 +0200)]
[PORT FROM R2] atomisp: mt9e013: store OTP data in sensor's private data
BZ: 17272
When sensor initialized, this patch reads OTP data and stores it in
sensor's private data. This way, when userspace requests OTP data driver
doesn't need to access sensor anymore.
Change-Id: I5071e5b893ff57b0073c3284ee447a4aa9f5c142
Orig-Change-Id: Ide7e328a3889c2719b45622ef612c8ddd6cb29cb
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/24938
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28023
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Kalle Lampila [Fri, 16 Dec 2011 12:52:15 +0000 (14:52 +0200)]
atomisp: mt9e013: make sure that mutex is init before it is used
BZ: 17272
Moved mutex_init before v4l2_i2c_subdev_init() call.
Change-Id: I522100e1ce30a6ae890386e7ee41a97af54eed8c
Signed-off-by: Kalle Lampila <kallex.lampila@intel.com>
Reviewed-on: http://android.intel.com:8080/28022
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
David Cohen [Tue, 22 Nov 2011 16:39:39 +0000 (18:39 +0200)]
[PORT FROM R2] atomisp: mt9e013: add lock protection to sensor's ioctl
BZ: 17272
Currently mt9e013 sensor lacks lock protection for concurrent ioctl
calls which may lead to racing condition.
This patch adds missing locks.
Change-Id: Ia0d3ca5156491a0cda169bedc7fe8731fffda5cc
Orig-Change-Id: I90f3cdca7c5b0cb7cf524244e8a4bf806d187223
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/24937
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28021
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
David Cohen [Tue, 22 Nov 2011 16:34:13 +0000 (18:34 +0200)]
[PORT FROM R2] atomisp: mt9e013: add missing timeout check when requesting OTP data
BZ: 17272
This patch adds missing timeout check when requesting OTP data. Without
this check, timeout may be confused with bad data.
Change-Id: Ic2971ba74cbcc69454302770966e15947b055ada
Orig-Change-Id: I1b562e8e799ff88fa17e614aceb7f6911d003d85
Signed-off-by: David Cohen <david.a.cohen@intel.com>
Reviewed-on: http://android.intel.com:8080/24936
Reviewed-by: Koskinen, Ilkka <ilkka.koskinen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28020
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Vinod Govindapillai [Wed, 23 Nov 2011 18:00:32 +0000 (20:00 +0200)]
[PORT FROM R2] atomisp: Validate the acceleration fw pointer from user space
BZ: 17272
Adds a check in atomisp_acc_fw_alloc if the acceleration firmware
passed is not NULL and the size of the firmware is not zero. If
either the fw is NULL or the size is 0, it returns an error.
Change-Id: I008448bc532c2d4e092bc58076755fda9c00d8ef
Orig-Change-Id: I9faad107fae526f125aadeedcbfe19e3285d8097
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-on: http://android.intel.com:8080/25125
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28016
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Jozef Kruger [Tue, 15 Nov 2011 23:15:42 +0000 (15:15 -0800)]
[PORT FROM R2] atomisp: add CSS support for atomic ISP restart
BZ: 17272
This is a rework of patch 22895. The difference with that patch is that
I removed the moving of the css 3a statistics reading. This seems to cause
a strange kernel crash which seems only indirectly related to this patch.
This patch now includes:
1. Double buffering of 3A statistics inside CSS layer.
2. Functions to determine whether the next ISP binary can be
started from an atomic context (the ISR).
Change-Id: Ic9a7a809c707d2e22ab07c4d373eea872c6132be
Orig-Change-Id: I6e633badc4f309cd8ea36548852b6fd8330daec3
Signed-off-by: Jozef Kruger <jozef.kruger@intel.com>
Reviewed-on: http://android.intel.com:8080/24166
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28015
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Lokesh Gupta [Tue, 15 Nov 2011 14:24:37 +0000 (15:24 +0100)]
[PORT FROM R2] atomisp: Provide a debug trace infra for CSS using ftrace
BZ: 17272
ftrace is the preferred way to dump debug traces in the kernel
because it is much more efficient than kernel printk. This patch
allows the use of ftrace to dump debug information from CSS. A new
API has also been added to CSS to set the verbosity level of the
traces.
Change-Id: I1edf61ca38d9931430dcaca3e1dc9e9d09585fd4
Orig-Change-Id: I8698a4c277052e95dc52c488e2d1d0f56e0080c8
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/24621
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28014
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Lokesh Gupta [Sat, 19 Nov 2011 13:27:13 +0000 (14:27 +0100)]
[PORT FROM R2] atomisp: Camera crash when switch resolution after color correction
BZ: 17272
When selecting color correction (sky | grass), and then taking
successive pictures at different resolutions caused the camera to
crash. This patch fixes the bug.
Change-Id: I3b2f738081049e3d1412b622d9caf1f9fd47c142
Orig-Change-Id: I49adf7c246ec328d0b5e8aa4cf517b8dae6e9e90
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/24617
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Gerasimow, Timo <timo.gerasimow@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28013
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
ziyux.jiang [Thu, 17 Nov 2011 13:59:57 +0000 (21:59 +0800)]
[PORT FROM R2] atomisp: camera burst capture speed stability issue.
BZ: 17272
previous yuv-ds algorithm is that the yuv-ds is enabled when
sensor.width > out.width*(1+10%) or sensor.height > out.height*(1+10%).
if that, 6M will be cropped first and then did yuv-ds.
so now we change yuv-ds algorithm to be that the yuv-ds is enabled
when sensor.width > out.width*(1+10%) and sensor.height > out.height*(1+10%).
if that, 6M only will be cropped and no yuv-ds is needed.
Change-Id: If0fbec87b03c8f600df1c6c58ac040af1d51cd10
Orig-Change-Id: I72026f545cf0feee958d357dc59ba4b38f760207
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/24347
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Tao, Jing <jing.tao@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28012
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Lokesh Gupta [Fri, 11 Nov 2011 10:30:35 +0000 (11:30 +0100)]
[PORT FROM R2] atomisp: CSS tried to free memory that was already freed
BZ: 17272
Some pointers within the CSS code were not being initialized
and reset correctly, due to which memory allocation/free-up
caused warnings. This patch fixes these cases. A new function
is added to properly uninitialize the sh_css_sp module.
Change-Id: I885a7aeb2b77197562bc468a673c2653cc30f3fc
Orig-Change-Id: I98370340392d3b7aca2f5c63403d3fbfe17164bd
Signed-off-by: Lokesh Gupta <lokesh.gupta@intel.com>
Reviewed-on: http://android.intel.com:8080/23911
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Toivonen, Tuukka <tuukka.toivonen@intel.com>
Reviewed-by: Von Oerthel, Jurgen <jurgen.von.oerthel@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28011
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
ziyux.jiang [Tue, 15 Nov 2011 17:47:08 +0000 (01:47 +0800)]
[PORT FROM R2] atomisp: Fix related failure check.
BZ: 17272
when frame allocation or others fail, we should return related error messages
instead of doing nothing.
Change-Id: I99ad6d5b721e5dcba7132cec3d68866c3e042717
Orig-Change-Id: I3199519ef7a0b6ca1ad01f6184671e43a30a06f6
Signed-off-by: ziyux.jiang <ziyux.jiang@intel.com>
Reviewed-on: http://android.intel.com:8080/24108
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Reviewed-by: Hu, Gang A <gang.a.hu@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Zhang, Xiaolin <xiaolin.zhang@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28010
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Daewoo Kim [Tue, 15 Nov 2011 02:24:23 +0000 (03:24 +0100)]
[PORT FROM R2] atomisp: implement chroma enhancement
BZ: 17272
Implement chroma enhancement
Change-Id: If76501d47d1faf1787f9d69c26438ec3d3afbfc0
Orig-Change-Id: I7aaf6a71e90f342d2d6de7eafe9e90260475364f
Signed-off-by: Daewoo Kim <daewoo.kim@intel.com>
Reviewed-on: http://android.intel.com:8080/24078
Reviewed-by: Gupta, Lokesh <lokesh.gupta@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Wang, Wen W <wen.w.wang@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28009
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
Yoshio Wada [Wed, 16 Nov 2011 09:43:11 +0000 (10:43 +0100)]
[PORT FROM R2] atomisp: shading table interpolation is not correct when output is larger than input
BZ: 17272
Change-Id: Icdac356e75c47e89103e6d364c4b5e731c100c0f
Orig-Change-Id: I5aee14fe00ca3784269e6c8ae70152e05a67cf2c
Signed-off-by: Yoshio Wada <yoshio.wada@intel.com>
Reviewed-on: http://android.intel.com:8080/24213
Reviewed-by: Kantola, Lasse <lasse.kantola@intel.com>
Reviewed-by: Finnila, Tapio <tapio.finnila@intel.com>
Reviewed-by: Kruger, Jozef <jozef.kruger@intel.com>
Reviewed-by: Cohen, David A <david.a.cohen@intel.com>
Tested-by: Koski, Anttu <anttu.koski@intel.com>
Reviewed-by: buildbot <buildbot@intel.com>
Tested-by: buildbot <buildbot@intel.com>
Reviewed-on: http://android.intel.com:8080/28008
Reviewed-by: Tuominen, TeemuX <teemux.tuominen@intel.com>
Reviewed-by: Koski, Anttu <anttu.koski@intel.com>
projects / kernel / kernel-mfld-blackbay.git / log