GitHub/LineageOS/android_kernel_motorola_exynos9610.git
5 years ago[RAMEN9610-12270] media: vb2: fix for wrong GFP mask to sg_alloc_table_from_pages
Janghyuck Kim [Fri, 15 Feb 2019 08:03:41 +0000 (17:03 +0900)]
[RAMEN9610-12270] media: vb2: fix for wrong GFP mask to sg_alloc_table_from_pages

sg_alloc_table_from_pages() would be failed if free memory is not
enough. GFP mask is changed to GFP_KERNEL to reclaim memory in this
situation.

Change-Id: I654975ce7903f1545e10ac1bc2f905f1caff8268
Signed-off-by: Janghyuck Kim <janghyuck.kim@samsung.com>
5 years ago[RAMEN9610-12270] media: smfc: force unshareable for HWFC buffer
Cho KyongHo [Mon, 18 Feb 2019 04:19:34 +0000 (13:19 +0900)]
[RAMEN9610-12270] media: smfc: force unshareable for HWFC buffer

Camera driver has no chance to maintain its destination buffer before
reading the buffer by SMFC if HWFC is enabled because SMFC starts
reading the buffer before MSCS finishes writing to the buffer.
Therefore, we should make SMFC read the buffer with the unshareable
attribute.

Change-Id: I9e9e326ecaa742fba1995a1bfc3f2ff60e68ef37
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270] media/vb2-dma-sg: add IOMMU_UNCACHED memops flag
Cho KyongHo [Mon, 18 Feb 2019 04:10:45 +0000 (13:10 +0900)]
[RAMEN9610-12270] media/vb2-dma-sg: add IOMMU_UNCACHED memops flag

VB2_DMA_SG_MEMFLAG_IOMMU_UNCACHED is introduced to prevent configuring
IOMMU_CACHE to iovmm_map() and ion_iovmm_map() which force shareable
mapping to the page table of the client device.
It is required by some shareable devices(devices with I/O Cache
coherency) which does not want to create shareable mappings for some
buffers.

Change-Id: Ic0c380a35cb6044d8d92500716081d45535965fe
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270] media: vb2: deliver alloc/map context to memops
Cho KyongHo [Mon, 18 Feb 2019 03:57:02 +0000 (12:57 +0900)]
[RAMEN9610-12270] media: vb2: deliver alloc/map context to memops

vb2 memops never know about the context in which it is working because
vb2 does not deliver information about the current working context.
However, vb2 memops sometimes needs to know the current context to
improve the performance or manage memory access correctly. Instead of
passing such information from vb2-core, it had better generated from
the client drivers because the information might be driver-specific
or memops-specific.
Let's deliver the information with flags. The flags should be memops-
specific and the client drivers can configure the flags in a callback,
memops.mem_flags(). It is called before .alloc() for mmap,
.get_userptr() for userptr and .map_dmabuf() for dmabuf.
The flags are defined by and specific to memops implementations.

Change-Id: I0b68df4bed659fb99e94755bf9333594b68c0b3b
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270] android: ion: do not force IOMMU_CACHE
Cho KyongHo [Mon, 18 Feb 2019 02:32:11 +0000 (11:32 +0900)]
[RAMEN9610-12270] android: ion: do not force IOMMU_CACHE

Some I/O coherent DMA may not want to force IOMMU_CACHE. Let's leave
its decision to the clients.

Change-Id: I574e55cfbf79a17b55ea71ba8753cfe799f7b5db
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270] iommu/exynos: do not force IOMMU_CACHE
Cho KyongHo [Mon, 18 Feb 2019 02:28:45 +0000 (11:28 +0900)]
[RAMEN9610-12270] iommu/exynos: do not force IOMMU_CACHE

IOMMU_CACHE of Exynos IOMMU means shareable in the page tables of IOMMU.
Forcing shareable to all I/O coherent DMAs may obstruct changes to
shareability property of DMAs of the client drivers.

Change-Id: Ibb5e0f5f5a0e0ef6942b2d7901014e6b94e430b1
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270] g2d: add missing IOMMU_CACHE to iommu mapping
Cho KyongHo [Mon, 18 Feb 2019 02:24:59 +0000 (11:24 +0900)]
[RAMEN9610-12270] g2d: add missing IOMMU_CACHE to iommu mapping

IOMMU_CACHE should be specified to iovmm_map() if G2D supports IO Cache
Coherency but it was missing because ION and IOMMU drivers adds
IOMMU_CACHE if their clients has DMA attributes of IO cache coherency.
But it is not correct to delegate configuration to IOMMU_CACHE to ION
and IOMMU drivers because G2D driver does not have any room to change
such configurations.

Change-Id: I6c20727c913af52781230669621bf8cf12faafdc
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12270][COMMON] media: vb2: modify to support offset in dmabuf
Seungchul Kim [Thu, 27 Sep 2018 05:03:43 +0000 (14:03 +0900)]
[RAMEN9610-12270][COMMON] media: vb2: modify to support offset in dmabuf

The offset was added to v4l2_buffer,
but vb2 does not support offset.
So vb2 is modified to support offset calculation.

Change-Id: Iee6e4b1146043eae00450556ba4686ba412aec0d
Signed-off-by: Seungchul Kim <sc377.kim@samsung.com>
5 years ago[RAMEN9610-12270] media: vb2: avoid accessing buffer from dmabuf_container
hyesoo.yu [Mon, 13 Aug 2018 09:42:52 +0000 (18:42 +0900)]
[RAMEN9610-12270] media: vb2: avoid accessing buffer from dmabuf_container

dmabuf_container is a dmabuf but it just has metadata of buffers that
it is containing. Trying to access buffer information through the
result of dma_buf_map_attachment() should be avoided.

Change-Id: I55c2e9a11ce4932d98cc0e72ab25e85f27a9f670
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12270] videobuf2: sync the buffer with as much as payload
hyesoo.yu [Fri, 31 Aug 2018 01:19:55 +0000 (10:19 +0900)]
[RAMEN9610-12270] videobuf2: sync the buffer with as much as payload

Change-Id: Ibda4b01e19a26e3ad771ee45bfa688ec394c55fe
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12270] media: videobuf2-core: don't call memop 'finish' when queueing
Hans Verkuil [Mon, 21 May 2018 12:43:02 +0000 (08:43 -0400)]
[RAMEN9610-12270] media: videobuf2-core: don't call memop 'finish' when queueing

[ Upstream commit 90b2da89a083e1395cb322521a42397c49ae4500 ]

When a buffer is queued or requeued in vb2_buffer_done, then don't
call the finish memop. In this case the buffer is only returned to vb2,
not to userspace.

Calling 'finish' here will cause an unbalance when the queue is
canceled, since the core will call the same memop again.

Change-Id: I64c1727150b2ffc37b41324cde0c8d92fedab228
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
5 years ago[RAMEN9610-12266][MODAP-18597[9610] wlbt: do not unmap A-Box memory from MMU when...
Ivan Priest [Fri, 15 Feb 2019 11:22:54 +0000 (11:22 +0000)]
[RAMEN9610-12266][MODAP-18597[9610] wlbt: do not unmap A-Box memory from MMU when BT restarts

The A-Box memory was being unmapped from the MMU when BT restarted,
this caused a kernel panic if A-Box was still using the memory.
Once the A-Box memory is mapped, do not unmap it.

Change-Id: I1bcd78aa1adbac12e87b8c8ad3c4afe1b2fd9314
SCSC-Bug-Id: SSB-49425
Signed-off-by: Ivan Priest <i.priest@samsung.com>
5 years ago[RAMEN9610-12262][COMMON] usb: add __nocfi to avoid cfi checking
Kisang Lee [Mon, 18 Feb 2019 04:45:18 +0000 (13:45 +0900)]
[RAMEN9610-12262][COMMON] usb: add __nocfi to avoid cfi checking

Change-Id: I8795e8c2f8828bb71d85a1b48516f037d77bdfaf
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-12260][COMMON] fbdev: dpu: PHY status check code is added
hwangjae lee [Mon, 18 Feb 2019 04:15:47 +0000 (13:15 +0900)]
[RAMEN9610-12260][COMMON] fbdev: dpu: PHY status check code is added

Change-Id: Iab233973f1822972e5632a4c3e39acaf3b4b491f
Signed-off-by: hwangjae lee <hj-yo.lee@samsung.com>
5 years ago[RAMEN9610-12257][COMMON][MODAP-18889] Change alloc type GFP_KERNEL -> GFP_ATOMIC
Jiyoung Jeong [Mon, 18 Feb 2019 04:23:06 +0000 (13:23 +0900)]
[RAMEN9610-12257][COMMON][MODAP-18889] Change alloc type GFP_KERNEL -> GFP_ATOMIC

Change-Id: I7ff09d8a728c02e5687c448f380302d073d41538
Signed-off-by: Jiyoung Jeong <ji_0.jeong@samsung.com>
5 years ago[RAMEN9610-12248] Subject: binder: fix race between munmap() and direct reclaim
Munjoo Kim [Thu, 14 Feb 2019 11:02:14 +0000 (20:02 +0900)]
[RAMEN9610-12248] Subject: binder: fix race between munmap() and direct reclaim

An munmap() on a binder device causes binder_vma_close() to be called
which clears the alloc->vma pointer.

If direct reclaim causes binder_alloc_free_page() to be called, there
is a race where alloc->vma is read into a local vma pointer and then
used later after the mm->mmap_sem is acquired. This can result in
calling zap_page_range() with an invalid vma which manifests as a
use-after-free in zap_page_range().

The fix is to check alloc->vma after acquiring the mmap_sem (which we
were acquiring anyway) and bail out of binder_alloc_free_page() if it
has changed to NULL.

Change-Id: I39c05c5446643aa7affb398904a6d3a1c35a9ea7
Signed-off-by: Todd Kjos <tkjos@google.com>
5 years ago[RAMEN9610-12196] crypto: check the validation of crypto pointer
Boojin Kim [Fri, 15 Feb 2019 04:51:42 +0000 (13:51 +0900)]
[RAMEN9610-12196] crypto: check the validation of crypto pointer

This patch checkes the validation of crypto pointer before puting it.

Change-Id: If016ae6108cb59d3cdf9c010340f022845a744a0
Signed-off-by: Boojin Kim <boojin.kim@samsung.com>
5 years ago[RAMEN9610-12194][9610] thermal: exynos_tmu: Add cpu max frequency limit control...
Chungwoo Park [Mon, 21 Jan 2019 10:18:39 +0000 (19:18 +0900)]
[RAMEN9610-12194][9610] thermal: exynos_tmu: Add cpu max frequency limit control code.

This patch added thermal driver control to cpu max freqency through temperature.
First if ect has limited_frequency value, exynos-tmu operate limit max frequency to pm qos.
If temperature higher than limited_threshold, driver set max qos to limited_frequency.
And if temperature is lower than limited_threshold_release,
driver release max qos value.

Change-Id: Iaccd295818eeceae035631a6dc7eebdd38044e40
Signed-off-by: Chungwoo Park <cww.park@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: include <slub.h>
Cho KyongHo [Fri, 26 Oct 2018 06:27:37 +0000 (15:27 +0900)]
[RAMEN9610-12171] android: ion: include <slub.h>

<linux/slub.h> is needed by kmalloc_array().

Change-Id: I7237cd0a3ddd53fa8419f265c459e6afc3c3c487
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] dma-buf: add dma_buf_[un]map_attacment_area
hyesoo.yu [Fri, 31 Aug 2018 01:15:20 +0000 (10:15 +0900)]
[RAMEN9610-12171] dma-buf: add dma_buf_[un]map_attacment_area

dma-buf supports the dma_buf_[un]map_attachment_area
to inform exporter of the payload area to use
buffer synchronization or mapping for DMA.

Change-Id: I690a6f9695df77a650eadceb38a8280b1a569408
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] dma-buf: fix gcc build error due to type
hyesoo.yu [Wed, 5 Sep 2018 00:31:06 +0000 (09:31 +0900)]
[RAMEN9610-12171] dma-buf: fix gcc build error due to type

Change-Id: I33c7cf960b02c9d1c43a438671fb038da59076a7
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] mm/hpa: fix possible inifinite page isolation
Cho KyongHo [Thu, 7 Feb 2019 07:52:12 +0000 (16:52 +0900)]
[RAMEN9610-12171] mm/hpa: fix possible inifinite page isolation

If HPA starts isolating pages in a page block that is currently
isolated in another thread, the page block possibly remains isolated
infinitely because HPA decides the target migratetype of the page block
from the current migratetype of it. Once it gets migratetype of a page
block as MT_ISOLATE and passes it to alloc_contig_range(), the final
migratetype of the page block becomes MT_ISOLATE which is not expected.

See the following race condition described by Jaewon Kim.
CPU0                              CPU1

get_pageblock_migratetype
 -> get MT_NON_ISOLATE as previous MT
alloc_contig_range_fast
 __alloc_contig_range
  start_isolate_page_range
   -> set to MT_ISOALTE
                                  get_pageblock_migratetype
                                   -> get MT_ISOLATE as previous MT
                                   alloc_contig_range_fast
                                    __alloc_contig_range
                                     start_isolate_page_range
                                      -> set to MT_ISOALTE
  undo_isolate_page_range to the
   -> recover to MT_NON_ISOLATE
                                     undo_isolate_page_range
                                      -> recover to MT_ISOLATE

He observed a page block is remained isolated for a log time.

Change-Id: I933b5c1a281ee32f400d63913c52301bb01cb550
Reported-by: Jaewon Kim <jaewon31.kim@samsung.com>
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] arm64: dts: limit HPA of Exynos9610 to lower 4GB
Cho KyongHo [Mon, 31 Dec 2018 04:30:35 +0000 (13:30 +0900)]
[RAMEN9610-12171] arm64: dts: limit HPA of Exynos9610 to lower 4GB

PPMPU of Exynos9610 is capable of protecting 4GB.

Change-Id: If2ffa0d3f4fb52da1fe9f3b7a42ddbee30c9689b
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: remove trailing whitespaces
Cho KyongHo [Thu, 25 Oct 2018 07:51:50 +0000 (16:51 +0900)]
[RAMEN9610-12171] android: ion: remove trailing whitespaces

Since 'commit 739bf5308f1d ("ion: remove recursive lock")', the list
of devices which have mapping of a buffer in the list of buffers shown
by /sys/kernel/debug/ion/buffers and oom killer notifier is removed
because it causes deadlock. But the title of the removed display item
and the trailing whitespaces at each line are still alive.

Change-Id: Ib6cb01306b97849ac6eea7198f0f6c725faa415b
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: cacheflush on entire proteted area
Cho KyongHo [Tue, 15 Jan 2019 12:28:21 +0000 (21:28 +0900)]
[RAMEN9610-12171] android: ion: cacheflush on entire proteted area

Pages from the CMA heap tends to be cached in the CPU caches with dirty
state since they are served as anon and page cache pages for the
userspace.
Flushing caches on buffer allocation is intended for prenventing
corruption from writing back to DRAM from the dirth cache lines while
updating the buffer from DMA. However, cache flush should be performed
on the entire allocated area if the buffer is to be protected from
non-secure access to prevent the dirty write-back to the protected area.
It is also applied to the carveout heap because allocation from a
carveout heap by an anonymous user is not prohibited.

Change-Id: Ieb1ae2efacefe1d23f03a4cc72615d45b4309eb7
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] ion: fix stack-out-of-bounds
hyesoo.yu [Tue, 11 Dec 2018 08:53:54 +0000 (17:53 +0900)]
[RAMEN9610-12171] ion: fix stack-out-of-bounds

The align variable is 4bytes but casted with 8bytes,
so it makes stack-out-of-bounds problem.

Change-Id: I458b42cc953c18a6f28bdba20c262fe5eaeb5a60
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: fix assigning long to int
Cho KyongHo [Fri, 23 Nov 2018 08:29:38 +0000 (17:29 +0900)]
[RAMEN9610-12171] android: ion: fix assigning long to int

This fixes SEC_INVALID_ASSIGNMENT_WITH_UNSIGNED_VALUES

Change-Id: I87124a4421b76cfb0e4060bb33c273f3cc879398
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: fix assigning shifted int to size_t
Cho KyongHo [Fri, 23 Nov 2018 08:21:22 +0000 (17:21 +0900)]
[RAMEN9610-12171] android: ion: fix assigning shifted int to size_t

This fixes SEC_BITSHIFT_MIGHT_OVERFLOW.

Change-Id: If39bb9e4da6c9c6acdba25ec817933204c0a6801
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: fix missing mutex_unlock()
Cho KyongHo [Fri, 23 Nov 2018 08:10:07 +0000 (17:10 +0900)]
[RAMEN9610-12171] android: ion: fix missing mutex_unlock()

mutex_unlock() is missing in exceptional case in __ion_iovmm_unmap().

Change-Id: Ic3379cacd396e867e44127d85b12040c7817c4e3
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] ion: add ion_iovmm_[un]map_attr
hyesoo.yu [Wed, 7 Nov 2018 09:45:31 +0000 (18:45 +0900)]
[RAMEN9610-12171] ion: add ion_iovmm_[un]map_attr

ion_iovmm_[un]map supports to map the device virtual
address depending the attribute of the buffer such as
secure. On the other hand, ion_iovmm_[un]map_attr supports
to map using the requested attribute by parameter
rather than attribute of buffer.

Change-Id: I022089fcfc7d1a8e1a9893fe5972bbf1c190c35b
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] mm: hpa: change allocate logic from buddy
hyesoo.yu [Wed, 11 Oct 2017 06:56:23 +0000 (15:56 +0900)]
[RAMEN9610-12171] mm: hpa: change allocate logic from buddy

The search order is from smallest order with the same
migrate type for reducing the memory fragment and allocate
only 64KB from buddylist and then expand the buddy pages
if the allocated page's order is bigger than requested
order to avoid allocation failure.

Change-Id: If1b224a7034830615dd59f1c92e603bab1158590
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: never free pages if unprotect fails
Cho KyongHo [Tue, 30 Oct 2018 07:37:49 +0000 (16:37 +0900)]
[RAMEN9610-12171] android: ion: never free pages if unprotect fails

If releasing H/W protection to a buffer fails, the buffer might not
be unusable in Linux forever becuase we do not have an idea to
determine if the buffer is accessible in Linux. So, we should mark
that buffer unusable with holding allocated buffer.

Change-Id: I7443506d3092adfd5538383faed7fc80f3bef8df
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: do not free secure addr on smc error
Cho KyongHo [Tue, 30 Oct 2018 07:31:54 +0000 (16:31 +0900)]
[RAMEN9610-12171] android: ion: do not free secure addr on smc error

ion_secure_unprotect() first calls smc to release any protection to
the given buffer and frees secure address for H/Ws that needs to access
that buffer. However, if releasing protection fails, the given buffer
may not be usable in Linux forever because the buffer might be
protected by H/W.

Change-Id: I91fb007e3df9330240d8900b5dc2da88e9564f8b
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: make ion_secure_iova_free() static
Cho KyongHo [Tue, 30 Oct 2018 07:23:10 +0000 (16:23 +0900)]
[RAMEN9610-12171] android: ion: make ion_secure_iova_free() static

ion_secure_iova_free() is never used outside of ion_buffer_protect.c.

Change-Id: I0d2fa1797c7bc7a93c151182e1ca782a8d149fbb
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] ion: remove unnecessary lock
hyesoo.yu [Tue, 16 Oct 2018 08:45:20 +0000 (17:45 +0900)]
[RAMEN9610-12171] ion: remove unnecessary lock

Change-Id: I7acae56998ca14c44c5fde913c60312ec515dde4
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] ion: remove recursive lock
hyesoo.yu [Wed, 17 Oct 2018 01:24:25 +0000 (10:24 +0900)]
[RAMEN9610-12171] ion: remove recursive lock

The oom notifiers is called from alloc_pages when the
memory exhausts. If the ion tries to memory allocation
with acquiring lock, and it happens out-of-memory, and
oom notifiers tries to the same lock, it makes recursive
lock.

Currently, ion_device->buffer_lock, ion_device->lock and
ion_heap->free_lock does not try to allocate memory but
buffer->lock tries to allocate memory, so we remove that
on oom notifier.

If that lock is removed, buffer->iovas and buffer->vaddr,
buffer->kmap_cnt can't be accessed because that requires
the lock to access without any synchronization problem.

Change-Id: Ie85038640d4bd271b28c993017fac865843ed83d
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] android: ion: add 'ion,hpa_limit' to dts
Cho KyongHo [Mon, 15 Oct 2018 12:21:25 +0000 (21:21 +0900)]
[RAMEN9610-12171] android: ion: add 'ion,hpa_limit' to dts

'ion,hpa_limit' property in 'ion-hpa-heap' node specifies the largest
address plus one that HPA can allocate. Therefore ion_hpa_heap should
not allocate pages from the physical address specified by
'ion,hpa_limit' to the end of possible physical address
(0xFFFF_FFFF_FFFF_FFFF).

Change-Id: I0c2cfdca1adbbf1c6602ebb3b68cc0cd5036854e
Signed-off-by: Cho KyongHo <pullip.cho@samsung.com>
5 years ago[RAMEN9610-12171] ion: add heap debug node
hyesoo.yu [Tue, 21 Aug 2018 04:46:05 +0000 (13:46 +0900)]
[RAMEN9610-12171] ion: add heap debug node

This debug node show allocation status for heaps to
allocate contiguous memory to know memory fragmentation,
memory leak, or memory staticstics.

Change-Id: I99a74a695587ffea75dcd543b9d59fa94d757ead
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] ion: support dma_buf_[un]map_attachment_area
hyesoo.yu [Fri, 31 Aug 2018 01:18:41 +0000 (10:18 +0900)]
[RAMEN9610-12171] ion: support dma_buf_[un]map_attachment_area

ion support to synchronize the buffer with as much
as the payload.

Change-Id: Ib7fb028d5372d528ae448215ff73704c84befa67
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] ion: protect buffer id by buffer_lock of device
hyesoo.yu [Thu, 13 Sep 2018 08:56:58 +0000 (17:56 +0900)]
[RAMEN9610-12171] ion: protect buffer id by buffer_lock of device

The exynos_ion_alloc_fixup function set the id
for the buffer.

The buffer id must be unique. The ida library
does its own locking without special requirement
to support unique id.
However it is possible not to set the unique id
when the returned id is error by the ida full or
other reason, and the id is set as MAX_BUFFER_ID
and updates last_buffer_id as zero because that
critical region is not protected by lock.

Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
Change-Id: Idf737b68995278a382c8db5708e13223f46dcf9f

5 years ago[RAMEN9610-12171] ion: allocate zero for system heap id
hyesoo.yu [Thu, 13 Sep 2018 02:13:31 +0000 (11:13 +0900)]
[RAMEN9610-12171] ion: allocate zero for system heap id

Traditionally, the system heap is was always zero.
The heap id is created according to the registered order
after kernel 4.12, but the system heap is registered first
to provide compatibility with users to use the system heap.

Change-Id: I46f203a533f6c29b62ced6b49c0a5a5a98e2a914
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] ion: fix gen pool management leak for error
hyesoo.yu [Mon, 10 Sep 2018 09:55:04 +0000 (18:55 +0900)]
[RAMEN9610-12171] ion: fix gen pool management leak for error

Change-Id: I09f6536f9f3e91e3d3598bcfbbbdd929a0558306
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171][COMMON] ion: add event log on debugfs
hyesoo.yu [Fri, 1 Jun 2018 09:36:58 +0000 (18:36 +0900)]
[RAMEN9610-12171][COMMON] ion: add event log on debugfs

timestamp        type     heap            buffer_id  size       time   remarks
------------------------------------------------------------------------
[000113.120867]    alloc  ion_system_heap     1     32400        178   noncached
[000113.121074]     mmap  ion_system_heap     1     32400        521
[000113.121153]     free  ion_system_heap     1     32400      31555
[000113.526191]    alloc  ion_system_heap     2     32400        177   cached
[000113.526398]     mmap  ion_system_heap     2     32400        472

show the event for ion buffers when alloc, mmap,
free, cache maintenance, system mmu mapping with timestamp
until 1024 record.

Change-Id: Ia0a4c69e7509c2d4ae035ca8fa8ab90354708d9c
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12171] ion: attach prefix [Exynos][ION] to all error logs
hyesoo.yu [Mon, 6 Aug 2018 08:01:32 +0000 (17:01 +0900)]
[RAMEN9610-12171] ion: attach prefix [Exynos][ION] to all error logs

Introducing four printk helper for ION driver
- perr: print logs in error level.
- perrfn: print logs with current function name.
- perrdev: print logs with devce name.
- perrfndev: print logs with current function name and device name.

All above macros attaches newline after the log.

Change-Id: Iabd441ff7b5b808baa832ec6ecf826b0e93e9a43
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-12143][9610] drivers: charger: enable irqs and add debug log
Suji Lee [Thu, 14 Feb 2019 05:29:26 +0000 (14:29 +0900)]
[RAMEN9610-12143][9610] drivers: charger: enable irqs and add debug log

unmask bat, det_bat, ivr, micd irq and add the registers in test_read

Change-Id: I1fba531a2249a4aa35e013849601d12f84bb491e
Signed-off-by: Suji Lee <suji0908.lee@samsung.com>
5 years ago[RAMEN9610-12132] vipx: add logs for debugging
Sanghwa Park [Wed, 13 Feb 2019 06:26:51 +0000 (15:26 +0900)]
[RAMEN9610-12132] vipx: add logs for debugging

Change-Id: I2ad36a3049c7e3a300987284c0c0e0ced44de4b9
Signed-off-by: Sanghwa Park <senius.park@samsung.com>
5 years ago[RAMEN9610-12132] vipx: add suspend/resume interface
Sanghwa Park [Wed, 13 Feb 2019 06:26:07 +0000 (15:26 +0900)]
[RAMEN9610-12132] vipx: add suspend/resume interface

Change-Id: I52ce913fbac7ec3ceda1e426e33643eaef88603d
Signed-off-by: Sanghwa Park <senius.park@samsung.com>
5 years ago[RAMEN9610-12132] vipx: add API to release kernel binary
Sanghwa Park [Wed, 13 Feb 2019 05:43:43 +0000 (14:43 +0900)]
[RAMEN9610-12132] vipx: add API to release kernel binary

Change-Id: Id55c6e9dc5d3356a5b70bb7f28dfa18e1bc6fd58
Signed-off-by: Sanghwa Park <senius.park@samsung.com>
5 years ago[RAMEN9610-12105][COMMON] lib: Kconfig: Enable freq logging when MINIMIZE_MODE
Changki Kim [Wed, 13 Feb 2019 04:47:33 +0000 (13:47 +0900)]
[RAMEN9610-12105][COMMON] lib: Kconfig: Enable freq logging when MINIMIZE_MODE

Change-Id: I243195a5c40e8d007c859fe794dac817e287657a
Signed-off-by: Changki Kim <changki.kim@samsung.com>
5 years ago[RAMEN9610-12100][COMMON] usb: gadget: remove unnecessary delay in functionfs bind
Kisang Lee [Wed, 13 Feb 2019 04:10:21 +0000 (13:10 +0900)]
[RAMEN9610-12100][COMMON] usb: gadget: remove unnecessary delay in functionfs bind

Change-Id: I8bb6d6c26e505abbc8fe1fc493983d71410459cf
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-12059] ccic : sync s2mu106 ccic code
Junhan Bae [Tue, 12 Feb 2019 04:53:33 +0000 (13:53 +0900)]
[RAMEN9610-12059] ccic : sync s2mu106 ccic code

Change-Id: I3fc4fee2eccd286d6dad4eab33746c2a30ba7261
Signed-off-by: Junhan Bae <junhan84.bae@samsung.com>
5 years ago[RAMEN9610-11863][9610] drivers: fuelgauge: Maintain UI SOC by difference of SOC_R...
Keunho Hwang [Fri, 1 Feb 2019 09:09:55 +0000 (18:09 +0900)]
[RAMEN9610-11863][9610] drivers: fuelgauge: Maintain UI SOC by difference of SOC_R & raw SOC

Change-Id: I1237c6cd785a2ee7ba4d24b10aeaa5d0ea96f22b
Signed-off-by: Keunho Hwang <keunho.hwang@samsung.com>
5 years ago[RAMEN9610-12054][COMMON] rtc: support s2mpu09 RTC alarm1 function
Jieun Yi [Mon, 11 Feb 2019 06:14:56 +0000 (15:14 +0900)]
[RAMEN9610-12054][COMMON] rtc: support s2mpu09 RTC alarm1 function

Alarm1 function allows the device to wake up from the power-off state.

Change-Id: I1dc648fba2e7639b1305466a7323d7552d16b002
Signed-off-by: Jieun Yi <jieun.yi@samsung.com>
5 years ago[RAMEN9610-12050][common][9610] wlbt : enable NAPI & RPS with update host version...
Youngsoo [Tue, 12 Feb 2019 05:02:08 +0000 (14:02 +0900)]
[RAMEN9610-12050][common][9610] wlbt : enable NAPI & RPS with update host version 6.62.1

1. [9610] wlbt: delibrately panic kernel on fw panic
2. [9610] wlbt: [DO NOT MERGE UPSTREAM] enable NAPI
3. [9610] wlbt: Prevent issues in slsi_acs_init
4. [9610] wlbt: make mutex acquire safe in HIP
5. [9610] wlbt: update debug messages for hip4_mbulk_to_skb()
6. wlbt: Destroy Bluez workqueue on close
7. [9610][7885][7872][7570] wlbt: Passing Correct Request Id in CFG80211 ...
8. [9610] wlbt: Relax allocation in process context
9. [9610] wlbt: NAPI: fix broken HIP4 watchdog
10. [9610] wlbt: increment scsc_release to 6.62.1

Change-Id: I7acbea3ed623ab01df599781ed279bd3824f715b

5 years ago[RAMEN9610-12039][9610] drivers, dts: add battery driver. DT for ERD.
Keunho Hwang [Mon, 11 Feb 2019 12:47:00 +0000 (21:47 +0900)]
[RAMEN9610-12039][9610] drivers, dts: add battery driver. DT for ERD.

Add new battery data DT, battery driver for S2MU004
base ERD

Change-Id: I3c03e6f47b57bf0761ebf374a6fb80cc463b06de
Signed-off-by: Keunho Hwang <keunho.hwang@samsung.com>
5 years ago[RAMEN9610-12039][9610] drivers, include: sync power_supply
Keunho Hwang [Mon, 11 Feb 2019 06:32:13 +0000 (15:32 +0900)]
[RAMEN9610-12039][9610] drivers, include: sync power_supply

For the code sync with S2MU106, update power_supply.h

Change-Id: I90069b6c818413e4ab65871eb9d99176255bc883
Signed-off-by: Keunho Hwang <keunho.hwang@samsung.com>
5 years ago[RAMEN9610-12039][9610] drivers: power: add s2mcs02 charger for code sync
Keunho Hwang [Mon, 11 Feb 2019 06:18:10 +0000 (15:18 +0900)]
[RAMEN9610-12039][9610] drivers: power: add s2mcs02 charger for code sync

Add s2mcs02 charger driver for code sync.

Change-Id: Ifb977b368d96262f85666eabbb401f83e6a54b7c
Signed-off-by: Keunho Hwang <keunho.hwang@samsung.com>
5 years ago[RAMEN9610-12039][9610] drivers: power, led: add s2mu106 drivers for code sync
Keunho Hwang [Mon, 11 Feb 2019 12:37:36 +0000 (21:37 +0900)]
[RAMEN9610-12039][9610] drivers: power, led: add s2mu106 drivers for code sync

Add S2MU106 charger, fuelgauge, powermeter, FLED,
RGB LED, haptic driver for code sync.

Change-Id: Ib0dec6feed3aff93dfedbe8c15466218bab47ae5
Signed-off-by: Keunho Hwang <keunho.hwang@samsung.com>
5 years ago[RAMEN9610-12036][9610] media:radio: Change RDS RT validate condition
Yunsu Kim [Tue, 12 Feb 2019 01:58:07 +0000 (10:58 +0900)]
[RAMEN9610-12036][9610] media:radio: Change RDS RT validate condition

Change-Id: I3403c862ccac343d36fb0d8cc7aa09f86e8990e6
Signed-off-by: Yunsu Kim <iamyunsu.kim@samsung.com>
5 years ago[RAMEN9610-11989][COMMON] muic: integrate MU004 and MU106
Kim Taejeong [Fri, 8 Feb 2019 05:31:00 +0000 (14:31 +0900)]
[RAMEN9610-11989][COMMON] muic: integrate MU004 and MU106

Change-Id: Iadc8acec977cd1675c03e628f1d70cdb9e02b5d4
Signed-off-by: Kim Taejeong <tj.kim@samsung.com>
5 years ago[RAMEN9610-11989][COMMON] mfd: integrate MU004 and MU106
Kim Taejeong [Mon, 11 Feb 2019 07:07:19 +0000 (16:07 +0900)]
[RAMEN9610-11989][COMMON] mfd: integrate MU004 and MU106

Change-Id: I8ec36e5eaa46df2b9a96e7f21eba925eb31cbbfc
Signed-off-by: Kim Taejeong <tj.kim@samsung.com>
5 years ago[RAMEN9610-11980][COMMON] usb: gadget: add vendor specific command in mass storage
Kisang Lee [Thu, 27 Dec 2018 11:03:37 +0000 (20:03 +0900)]
[RAMEN9610-11980][COMMON] usb: gadget: add vendor specific command in mass storage

Change-Id: If8b4e8a89bbff020beee88618693514b61cffd22
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-11980][COMMON] arm64: config: enable mass_storage for charging mode
Kisang Lee [Thu, 27 Dec 2018 11:08:22 +0000 (20:08 +0900)]
[RAMEN9610-11980][COMMON] arm64: config: enable mass_storage for charging mode

Change-Id: I6ff2e94732e526e9543ff9be13b1ec9e7a55f04e
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-11930] fbdev: dpu20: added frame rate information in psr_info
ChiHun Won [Thu, 7 Feb 2019 07:52:33 +0000 (16:52 +0900)]
[RAMEN9610-11930] fbdev: dpu20: added frame rate information in psr_info

Change-Id: I12af987261cc24bbe5d87f7f75da4a6ed87f6efb
Signed-off-by: ChiHun Won <chihun.won@samsung.com>
5 years ago[RAMEN9610-11923][COMMON] ASoC: abox: request dram during firmware execution
Gyeongtaek Lee [Thu, 22 Nov 2018 08:06:01 +0000 (17:06 +0900)]
[RAMEN9610-11923][COMMON] ASoC: abox: request dram during firmware execution

Firmware can access DRAM anytime but current source code
only request DRAM during DMA operation.

Change-Id: Ie3e8697c863281967aedeb645f886c747ed21e6a
Signed-off-by: Gyeongtaek Lee <gt82.lee@samsung.com>
5 years ago[RAMEN9610-11799][9610] media: radio: sync samsung/exynos9609
Yunsu Kim [Sat, 2 Feb 2019 10:34:20 +0000 (19:34 +0900)]
[RAMEN9610-11799][9610] media: radio: sync samsung/exynos9609

Change-Id: I33fed3e018a9a1d9db5c58b77cbc036ed5dc9a94
Signed-off-by: Yunsu Kim <iamyunsu.kim@samsung.com>
5 years ago[RAMEN9610-11423][COMMON][MODAP-16389] gnss_if: fix wake_lock for iod
Soungkwan Kimn [Wed, 2 Jan 2019 07:44:24 +0000 (16:44 +0900)]
[RAMEN9610-11423][COMMON][MODAP-16389] gnss_if: fix wake_lock for iod

Change-Id: I48e72faa8f3a440364e067f5ce86fea314379c21
Signed-off-by: Soungkwan Kimn <sk.kimn@samsung.com>
5 years ago[RAMEN9610-11423][COMMON][MODAP-16389] gnss_if: add sysfs for debug
Soungkwan Kimn [Wed, 2 Jan 2019 11:26:58 +0000 (20:26 +0900)]
[RAMEN9610-11423][COMMON][MODAP-16389] gnss_if: add sysfs for debug

Change-Id: I595973afea20ace6773d5caff433e75086418b0c
Signed-off-by: Soungkwan Kimn <sk.kimn@samsung.com>
5 years ago[RAMEN9610-11413][COMMON][MODAP-16546] drivers: modem_if: Open embms channel for...
Jiyoung Jeong [Wed, 23 Jan 2019 11:25:24 +0000 (20:25 +0900)]
[RAMEN9610-11413][COMMON][MODAP-16546] drivers: modem_if:  Open embms channel for network

Change-Id: I423b4314885af52e42570c961b360289e5b2d725
Signed-off-by: Jiyoung Jeong <ji_0.jeong@samsung.com>
5 years ago[RAMEN9610-11292][9610] media: radio: Remove RDS debugging logs, Increase FM_RDS_MEM_...
Yunsu Kim [Wed, 23 Jan 2019 08:02:49 +0000 (17:02 +0900)]
[RAMEN9610-11292][9610] media: radio: Remove RDS debugging logs, Increase FM_RDS_MEM_SIZE_PARSER size to 2000

Change-Id: If0ff07353429238e6cd72360ba38d2ce0b994da1
Signed-off-by: Yunsu Kim <iamyunsu.kim@samsung.com>
5 years ago[RAMEN9610-11757] irq/work: Improve the flag definitions
Bartosz Golaszewski [Fri, 5 Jan 2018 04:19:56 +0000 (05:19 +0100)]
[RAMEN9610-11757] irq/work: Improve the flag definitions

IRQ_WORK_FLAGS is defined simply to 3UL. This is confusing as it
says nothing about its purpose. Define IRQ_WORK_FLAGS as a bitwise
OR of IRQ_WORK_PENDING and IRQ_WORK_BUSY and change its name to
IRQ_WORK_CLAIMED.

While we're at it: use the BIT() macro for all flags.

Change-Id: Ie68d7f4a642fa160600bc329c8a7b580e9627a93
Signed-off-by: Bartosz Golaszewski <brgl@bgdev.pl>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1515125996-21564-1-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
5 years ago[RAMEN9610-11757] irq/work: Use llist_for_each_entry_safe
Thomas Gleixner [Sun, 12 Nov 2017 12:02:51 +0000 (13:02 +0100)]
[RAMEN9610-11757] irq/work: Use llist_for_each_entry_safe

The llist_for_each_entry() loop in irq_work_run_list() is unsafe because
once the works PENDING bit is cleared it can be requeued on another CPU.

Use llist_for_each_entry_safe() instead.

Change-Id: I8680657f89c008f879b4e88e3499a7e44f2978a1
Fixes: 16c0890dc66d ("irq/work: Don't reinvent the wheel but use existing llist API")
Reported-by:Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <frederic@kernel.org>
Cc: Byungchul Park <byungchul.park@lge.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Petri Latvala <petri.latvala@intel.com>
Link: http://lkml.kernel.org/r/151027307351.14762.4611888896020658384@mail.alporthouse.com
5 years ago[RAMEN9610-11757] irq/work: Don't reinvent the wheel but use existing llist API
Byungchul Park [Tue, 31 Oct 2017 01:46:54 +0000 (02:46 +0100)]
[RAMEN9610-11757] irq/work: Don't reinvent the wheel but use existing llist API

Use the proper llist APIs instead of open-coded variants of them.

Change-Id: I66dd8fc945a5e2cebca80388d2887f7d6ef71988
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1509414414-14987-1-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
5 years ago[RAMEN9610-11757] irq_work: Map irq_work_on_queue() to irq_work_on() in !SMP
Paul E. McKenney [Fri, 18 Aug 2017 17:59:16 +0000 (10:59 -0700)]
[RAMEN9610-11757] irq_work: Map irq_work_on_queue() to irq_work_on() in !SMP

Commit 478850160636 ("irq_work: Implement remote queueing") provides
irq_work_on_queue() only for SMP builds.  However, providing it simplifies
code that submits irq_work to lists of CPUs, eliminating the !SMP special
cases.  This commit therefore maps irq_work_on_queue() to irq_work_on()
in !SMP builds, but validating the specified CPU.

Change-Id: I08c088556952a06471567c86c1993d41d971749a
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
5 years ago[RAMEN9610-11731][common][9610] wlbt: Regulatory Rules should not be updated based...
Mohit Ghuley [Wed, 30 Jan 2019 10:11:02 +0000 (15:41 +0530)]
[RAMEN9610-11731][common][9610] wlbt: Regulatory Rules should not be updated based on beacon hints.

CFG80211 Regulatory Rules should not be
updated based on beacon hints.

Correction for Config Name in Kconfig performed as well.

Change-Id: Ib30ffbf7459c6fb6b92dc8578b2cb3dbd4b1356b
SCSC-Bug-Id: SSB-47118
Signed-off-by: Mohit Ghuley <mohit.ghuley@samsung.com>
5 years ago[RAMEN9610-11727][HACK] g2d: lock memory frequency as 845Mhz
Hyesoo Yu [Wed, 30 Jan 2019 08:03:47 +0000 (17:03 +0900)]
[RAMEN9610-11727][HACK] g2d: lock memory frequency as 845Mhz

Change-Id: Id5f3f6c1f95f3ca39ab6fa0b63777f0384fc93b9
Signed-off-by: Hyesoo Yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-11727] g2d: remove use-after-free of context for performance
hyesoo.yu [Thu, 20 Dec 2018 05:29:09 +0000 (14:29 +0900)]
[RAMEN9610-11727] g2d: remove use-after-free of context for performance

delayed work should not be called after device close.
However cancel_delayed_work() will cancel future invocations,
it does not guarantee it to be finished before cancel_delayed_work()
returns if delayed work is currently being processed.

Naive way to fix would be to change cancel_delayed_work into
cancle_delayed_work_sync. However it create a deadlock because
delayed work calls g2d_put_performance, which will wait on
g2d_dev->lock_qos. That mutex was already locked in
previously g2d_put_performance, which is waiting for
delayed work to finish.

So, use flush_delayed_work in g2d_release instead of
cancel_flush_work. Then, we don't request g2d_put_performance
in g2d_release because the request of performance always
have the delayed work to release the lock, so that must be
called before g2d context release.

Change-Id: I0c47ced4ca10c68f382c3f0a368f1bf16b7cd4b8
Signed-off-by: hyesoo.yu <hyesoo.yu@samsung.com>
5 years ago[RAMEN9610-11639][COMMON] scsi: ufs: Print available lane for infomation
JaeHun Jung [Tue, 23 Oct 2018 08:39:32 +0000 (17:39 +0900)]
[RAMEN9610-11639][COMMON] scsi: ufs: Print available lane for infomation

Change-Id: Icbc16b176d9966dc94613fbed6e1908b0b7c198d
Signed-off-by: JaeHun Jung <jh0801.jung@samsung.com>
5 years ago[RAMEN9610-11637][COMMON] usb: host: re-initilize runtime pm
Kisang Lee [Tue, 29 Jan 2019 07:16:10 +0000 (16:16 +0900)]
[RAMEN9610-11637][COMMON] usb: host: re-initilize runtime pm

Change-Id: Ia630f71b0d1171e8dbeb73782ec0618d47a2bba4
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-11637][COMMON] usb: host: allocate dma memory in booting time
taehyun.cho [Wed, 31 Oct 2018 08:15:22 +0000 (17:15 +0900)]
[RAMEN9610-11637][COMMON] usb: host: allocate dma memory in booting time

Change-Id: If6dd884c7a8bedbddf81a97dcb48a94e18c634f8
Signed-off-by: taehyun.cho <taehyun.cho@samsung.com>
5 years ago[RAMEN9610-11637][COMMON] usb: phy: add USB3.0 isolation feature
Kisang Lee [Tue, 29 Jan 2019 07:08:17 +0000 (16:08 +0900)]
[RAMEN9610-11637][COMMON] usb: phy: add USB3.0 isolation feature

Change-Id: I9332363df4ceea63c60f2d9511fda670921e68ed
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-11636][COMMON] mmc: dw_mmc: sdcard power off when to card detect
dukhyun kwon [Thu, 27 Dec 2018 05:59:20 +0000 (14:59 +0900)]
[RAMEN9610-11636][COMMON] mmc: dw_mmc: sdcard power off when to card detect

Change-Id: I9a53d9e51347d947d66d8648bc7f21d672d3e83c
Signed-off-by: dukhyun kwon <d_hyun.kwon@samsung.com>
5 years ago[RAMEN9610-11636][9610] arm64: dts: enable skip init no card tray for sdcard
dukhyun kwon [Thu, 27 Dec 2018 10:31:53 +0000 (19:31 +0900)]
[RAMEN9610-11636][9610] arm64: dts: enable skip init no card tray for sdcard

Change-Id: Id81a4c82446f555013968952d80cde51621a39e2
Signed-off-by: dukhyun kwon <d_hyun.kwon@samsung.com>
5 years ago[RAMEN9610-11636][COMMON] mmc: dw_mmc: skip init when to not card tray.
dukhyun kwon [Thu, 27 Dec 2018 10:30:22 +0000 (19:30 +0900)]
[RAMEN9610-11636][COMMON] mmc: dw_mmc: skip init when to not card tray.

Change-Id: I95cb1bb52c1937c7528b3367e308f791cea46663
Signed-off-by: dukhyun kwon <d_hyun.kwon@samsung.com>
5 years ago[RAMEN9610-11578] arm64: dts: chub reserved memory size is modified
Boojin Kim [Fri, 28 Dec 2018 06:12:58 +0000 (15:12 +0900)]
[RAMEN9610-11578] arm64: dts: chub reserved memory size is modified

This patch reduces chub reserved memory from 512KB to 320KB

Change-Id: I464a9fb5342209453fc90155fee8b896e80e38f9
Signed-off-by: Boojin Kim <boojin.kim@samsung.com>
5 years ago[RAMEN9610-11577][COMMON] lib: dss: change dss virt-addr base
Donghyeok Choe [Thu, 1 Nov 2018 01:21:24 +0000 (10:21 +0900)]
[RAMEN9610-11577][COMMON] lib: dss: change dss virt-addr base

RAMEN/A9 assign a new reserved memory to the camera
library. It was created because there was a request
to match Makalu by moving virt-addr base of DSS area.

Change-Id: I295110f600385377275561c0b70ea8f5e2ce170f
Signed-off-by: Donghyeok Choe <d7271.choe@samsung.com>
5 years ago[RAMEN9610-11577][COMMON] soc: samsung: change prototype chipid_board_id_show
Sangwook Ju [Wed, 19 Dec 2018 09:53:18 +0000 (18:53 +0900)]
[RAMEN9610-11577][COMMON] soc: samsung: change prototype chipid_board_id_show

Change-Id: Id45c26d4fa9d93be7eebc1ec05e7c17f763c208d
Signed-off-by: Sangwook Ju <sw.ju@samsung.com>
5 years ago[RAMEN9610-11577][COMMON] soc: exynos-chipid: Add Board revision sysfs node
Jaehyoung Choi [Mon, 15 Oct 2018 10:41:29 +0000 (19:41 +0900)]
[RAMEN9610-11577][COMMON] soc: exynos-chipid: Add Board revision sysfs node

Change-Id: I407e2fbdeb876584b278a966ba74be6292683f01
Signed-off-by: Jaehyoung Choi <jkkkkk.choi@samsung.com>
5 years ago[RAMEN9610-11575] change iic div calculation formula for frquency not up to the set...
ts_songwei1113_tmp [Mon, 5 Nov 2018 09:45:42 +0000 (17:45 +0800)]
[RAMEN9610-11575] change iic div calculation formula for frquency not up to the set value

Change-Id: Ic6514913c0cb9062ef8daaffb433601f6dad6bb3

5 years ago[RAMEN9610-11559][COMMON] spi: s3c64xx: Fix runtime PM feature for SPI driver
Kyungwoo Kang [Tue, 28 Aug 2018 11:09:56 +0000 (20:09 +0900)]
[RAMEN9610-11559][COMMON] spi: s3c64xx: Fix runtime PM feature for SPI driver

Add protection code for runtime pm at SPI driver.
Now spi driver waits for runtime_pm until recovers from
suspended mode.

Change-Id: I2402e4f13e92e3d941f0b2e91716558ad7ecc8d0
Signed-off-by: Kyungwoo Kang <kwoo.kang@samsung.com>
5 years ago[RAMEN9610-11554][9610] soc: samsung: cal-if: Added cal_print_wakeup_reason.
Jeonghoon Jang [Tue, 15 Jan 2019 07:42:00 +0000 (16:42 +0900)]
[RAMEN9610-11554][9610] soc: samsung: cal-if: Added cal_print_wakeup_reason.

Change-Id: I27f38d8896d07e7c368d35b77da7ae94fae51e1c
Signed-off-by: Jeonghoon Jang <jnghn.jang@samsung.com>
5 years ago[RAMEN9610-11479][9610] vipx: support ofi framework
Sanghwa Park [Mon, 28 Jan 2019 05:50:25 +0000 (14:50 +0900)]
[RAMEN9610-11479][9610] vipx: support ofi framework

  - add ofi framework support
  - remove reserved memory
  - enhance debug feture

Change-Id: I2b20c2872e264ef12ad264b23c1dc7027cb62ad8
Signed-off-by: Sanghwa Park <senius.park@samsung.com>
5 years ago[RAMEN9610-11497][COMMON] arm64: dts: Add dbg-uart-ch property to serial_0 node
myung-su.cha [Mon, 28 Jan 2019 13:23:02 +0000 (22:23 +0900)]
[RAMEN9610-11497][COMMON] arm64: dts: Add dbg-uart-ch property to serial_0 node

dbg-uart-ch property is used for lenovo uart console resume latency
issue in user binary.

Change-Id: I3060a96df912bc86bedee43b35fd126bd52600cf
Signed-off-by: myung-su.cha <myung-su.cha@samsung.com>
5 years ago[RAMEN9610-11497][COMMMON] serial: samsung: Set default Baud rate as 115200
myung-su.cha [Mon, 28 Jan 2019 12:59:09 +0000 (21:59 +0900)]
[RAMEN9610-11497][COMMMON] serial: samsung: Set default Baud rate as 115200

Abnormal resume latency issue in user binary is because Baud
rate is set 9600. To resolve this issue, this patch makes Baud
rate as 115200 when requested Baud rate is 0.

Change-Id: I71b54f2adc63f96032b9c628a5afb7e5e5377fc8
Signed-off-by: myung-su.cha <myung-su.cha@samsung.com>
5 years ago[RAMEN9610-11632][9610] fimc-is2: add function for bts_update for thermal
Wooyeon Kim [Tue, 29 Jan 2019 08:21:01 +0000 (17:21 +0900)]
[RAMEN9610-11632][9610] fimc-is2: add function for bts_update for thermal

 - update bts function for thermal

Change-Id: Ic7ea5c425a9a839bf4db2921d7d1458a50243b29
Signed-off-by: Wooyeon Kim <wooy88.kim@samsung.com>
5 years ago[RAMEN9610-11632][COMMON] fimc-is2: add function for bts_update_scen when throttling...
Wooyeon Kim [Mon, 28 Jan 2019 10:54:32 +0000 (19:54 +0900)]
[RAMEN9610-11632][COMMON] fimc-is2: add function for bts_update_scen when throttling state

 - when throttling state, need to adjust camera device's mo

Change-Id: Ibb4a463788e3cd41f217d1490af97cba7e5df3d5
Signed-off-by: Wooyeon Kim <wooy88.kim@samsung.com>
5 years ago[RAMEN9610-11632][9610] bts: add scenario for thermal throttling
Taekki Kim [Mon, 28 Jan 2019 09:57:51 +0000 (18:57 +0900)]
[RAMEN9610-11632][9610] bts: add scenario for thermal throttling

This patch adds scenario for thermal throttling.
The bandwidth of ISP blocks limited when thermal throttling.

Change-Id: If36c08fbbf8a74b144e28896b44e91b8b055bc3a
Signed-off-by: Taekki Kim <taekki.kim@samsung.com>
5 years ago[RAMEN9610-11469][COMMON] usb: dwc3: change workqueue for connection stability
Kisang Lee [Fri, 25 Jan 2019 04:02:10 +0000 (13:02 +0900)]
[RAMEN9610-11469][COMMON] usb: dwc3: change workqueue for connection stability

Change-Id: I8ec23741edd5c5a367e7996caf049814c389aaed
Signed-off-by: Kisang Lee <kisang80.lee@samsung.com>
5 years ago[RAMEN9610-11232] cpufreq: acme: sync cur freq with resume freq when trying to suspend
Choonghoon Park [Wed, 16 Jan 2019 06:57:16 +0000 (15:57 +0900)]
[RAMEN9610-11232] cpufreq: acme: sync cur freq with resume freq when trying to suspend

Change-Id: I517c2c6116f8ceca91ba760743349e9af7f1eaf0
Signed-off-by: Choonghoon Park <choong.park@samsung.com>
Signed-off-by: Hyunju Kang <hjtop.kang@gmail.com>
5 years ago[RAMEN9610-11445][9610] wlbt : update MAC addr scheme to avoid merge conflict
wangxuan [Tue, 23 Oct 2018 04:05:47 +0000 (12:05 +0800)]
[RAMEN9610-11445][9610] wlbt : update MAC addr scheme to avoid merge conflict

Read wifi/bt addr from utag

Change-Id: I34ceefe178bdd0c3285f1f1cc0b504991ee65fbb