Disable CONFIG_UBIFS_FS_ZSTD to save vmalloc memory.
Then CONFIG_CRYPTO_ZSTD, CONFIG_ZSTD_COMPRESS and
CONFIG_ZSTD_DECOMPRESS are also disabled.
Change-Id: Ia1a8df1a4bb585aab7fd65892c3e27c6f6f53a42
Signed-off-by: Qingqing Zhou <qqzhou@codeaurora.org>
send uevent for slate dsp and bt state to userspace.
add sysfs entries to hold the states of Slate BTSS and DSPSS.
Change-Id: Iffccd90a4a1b821fc0592d272a61f15bb5c13b82
Signed-off-by: Ajit Kumar <kajit@codeaurora.org>
Signed-off-by: Praveen Koya <pkoya@codeaurora.org>
msm_dirconn_uncfg_reg() currently uses readl_/writel_relaxed() which do not
check for per tile address for gpio intr cfg. Replace to use
msm_readl/writel_intr_cfg() which takes care of per tile.
While at this also fix dir_conn_reg to use per tile gpio cfg.
Change-Id: If87e35f2943593c0723bd88a29d483fed70f7041
Signed-off-by: Maulik Shah <mkshah@codeaurora.org>
Add glink-slatecom driver which relies on slatecom as a transport
layer for sending data between various sub systems. Add
documentation file for the glink-slatecom driver.
Change-Id: If3b824c9d2efea920e52579b38e00eeff529f49f
Signed-off-by: Vivek Golani <vgolani@codeaurora.org>
Added crypto-qti-virt.c which sends crypto requests to backend(HOST)
via hab channel from frontend(GUEST) while performing hardware based
file encryption. This Makes the driver to use the hypervisor backend
for ICE virtualization instead of calling directly to TZ.
Added virtio_blk_qti_crypto.c which creates a keyslot manager device
for virtual disk to manage keyslots, which will forward the request
to crypto virtual library and eventually will program the ICE slot
in the backend(HOST).
Test:
1.Basic_SimpleEncryption
2.ModifyEnforcedFiles_FileCreationWithinEnforcedFolder
3.PIN, pattern, password
4.verified filename encryption.
Change-Id: Idc57d9958e1cecba68eecd556d18ec54fa1c02b0
Signed-off-by: Santosh Dronamraju <sdronamr@codeaurora.org>
Change the load factor from 3000 to 30000 for sdxlemur.
Change-Id: I939610150bcaddba7ff508ea130009a010cfb6c0
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
Chipsets which have PCIe Gen4 PHY and controller on MX rail
requires explicit vote for higher corners than the default.
Therefore add support to vote for and scale MX rail based on the
Gen speed which PCIe link operates at.
Change-Id: I6aa595aa6e3955ff09d8eeae604431b47a14985a
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
Driver allocates in_buf_list.addr during open_client_mhi_channels,
which is called for every CONNECT event for AT_CTRL channels.
in_buf_list.addr is not released during DISCONNECT event.
Adding a fix to release this memory during a channel DISCONNECT
event.
Change-Id: Idd7cac731137934d820e6a8a07335982575a913b
Signed-off-by: Subramanian Ananthanarayanan <skananth@codeaurora.org>
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
There is a possible race condition, if the client calls
MHI close and channel gets a start command then
mhi_dev_alloc_evt_buf_evt_req() would be called. This can lead to
tr_events and ereqs getting freed from both the contexts.
Fix is to use a mutex to prevent this from happening parallelly.
Change-Id: I9ac10211d5bd5cee96e69b7fc10955e24569cf64
Signed-off-by: Subramanian Ananthanarayanan <skananth@codeaurora.org>
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
On certain platforms host might take a long time to send M0
to device. Only modified msm files in this commit.
Change-Id: I8c5f07a7926e01afa84a983ef98721749787d2da
Signed-off-by: Siva Kumar Akkireddi <sivaa@codeaurora.org>
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
On some customer platforms, resume from hibernate results in a
cold boot of the end-point. The previous mhi device state gets
cleared and this results in mhi host issuing a reset. Clear the
reset while waiting for M0 from host to enable host to proceed
with issuing M0.
Change-Id: I18aa4e760d59e37842a9f3a6dba1478a98137bb8
Signed-off-by: Siva Kumar Akkireddi <sivaa@codeaurora.org>
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
mhi_dev_cache_host_cfg will be called during cold boot and also
when processing mhi device reset command from host. Do not re-allocate
the command, event and channel context cache memory if they have been
already allocated.
Change-Id: I55826e05b3d83c9fcf3d4b23e26b91253130671e
Signed-off-by: Siva Kumar Akkireddi <sivaa@codeaurora.org>
Signed-off-by: Gauri Joshi <gaurjosh@codeaurora.org>
Change dmesg print for ddr cap frequency driver.
Change-Id: I2d660f51596dc6c309ad932b543c7c030b1b04cb
Signed-off-by: Vinay Sudra <vsudra@codeaurora.org>
When the device operates in full speed mode, the CONNECTSPD field
vaue is 1. Using bitwise & operator on this value for super speed
check results in full speed being erroneously calculated as super
speed. Fix this by comparing the absolute value of CONNECTSPD for
super speed check.
Change-Id: I2bf497748f621b1982aecda55a9a280d93f84858
Signed-off-by: Elson Roy Serrao <eserrao@codeaurora.org>
QCOM_WATCHDOG_WAKEUP_ENABLE is hardware feature, so remove it for
sdxnighthjar.
Change-Id: I3946e9a7252a2305034a8f8559fc1d0d0294e87a
Signed-off-by: Lijuan Gao <lijuang@codeaurora.org>
With the default DPOLICY_BG discard thread is ioaware, which prevents
the discard thread from issuing the discard commands. On low RAM setups,
it is observed that these discard commands in the cache are consuming
high memory. This patch aims to relax the memory pressure on the system
due to f2fs pending discard cmds by changing the policy to DPOLICY_FORCE
based on the nm_i->ram_thresh configured.
Change-Id: I01b692df6f0ecdf92e316d3cea764d87d5f73ffe
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
On virtualization platform, cpu frequency is controlled by host.
Enable fake cpufreq driver for guest.
Change-Id: Ief774fede1c2a5bf5db62b28a53d3650c4d5854d
Signed-off-by: Zhiqiang Tu <ztu@codeaurora.org>
When creating a timeline allocate an ID but don't populate the slot
with the timeline pointer until we are done initializing the timeline.
This avoids a race if somebody tries to free the same identifier before
the create operation is complete. Also, correct kgsl_timeline_by_id
function by doing idr_find and refcount increase under same lock. This
eliminates the chance that a timeline will be destroyed while somebody
is still using it.
Change-Id: Id894baac1bbd6042f668ea4afca39fae3b5a2d86
Signed-off-by: Deepak Kumar <dkumar@codeaurora.org>
The change is for general scheduler improvement.
Change-Id: Ib5d58c23d3a67a81d682e08e58f308e5c659a7f8
Signed-off-by: Kishore Sri venkata Ganesh Bolisetty <bsrivenk@codeaurora.org>
This change will not request for gpio_free if the gpio_request
is not called for a particular gpio.
Change-Id: I39bedb3495d59334ea61953eaeb8ccf59d6633f8
Signed-off-by: Balakrishna Godavarthi <bgodavar@codeaurora.org>
Check PCIE link status before accessing MHI registers at DRV
resume to overcome NOC error.
Change-Id: I796e22797d91fba69e258addd0a73b04a7146d80
Signed-off-by: Vivek Pernamitta <vpernami@codeaurora.org>
Applications have no chance to destroy timelines if crash. This change
could clean isolated timelines.
Change-Id: If15a5b8f5f45c6c6b051fcaea81bf51c2bc7aa64
Signed-off-by: Kasin Li <donglil@codeaurora.org>
Touch uses gpio as irq in LA. As per latest change from pinctrl,
wakeup bit is disabled/enabled when disable_irq/enable_irq is called.
During transtion from LA to LE, disable_irq is called leading to
wakeup bit getting disabled and in LE Touch interrupts does not come.
So, use disable_irq_wake to disable Touch interrupts while transition
from LA to LE and use enable_irq_wake to enable back Touch interrupts
while transition from LE to LA.
Change-Id: I288bfbd26be9ff1da1b7e14104f98338eb8d8715
Signed-off-by: Ritesh Kumar <riteshk@codeaurora.org>
Select REGMAP_IRQ for wcd937x and later targets.
Change-Id: I791d26a5ed7690bbe9df638bd85fd4f7c59050f0
Signed-off-by: Rutvik Patel <prutvikr@codeaurora.org>
Do integer overflow check on the length of buffer
to be copied before copying to the buffer.
Change-Id: I425861170125fb3069739af84975c4893462624e
Acked-by: Chenna Kesava Raju <chennak@qti.qualcomm.com>
Signed-off-by: Jeya R <jeyr@codeaurora.org>`
In kgsl_ioctl_gpu_aux_command, if the number of generic objects passed is
UINT_MAX - 1 along with a syncpoint object, the count variable will be
incremented twice: once for timestamp sync object and another for aux
command syncpoint object resulting in count value as 0. Fix this by
allowing maximum of UINT_MAX - 2 generic objects.
Change-Id: Ief8c8da9b147e4b45b1f53d42473b772cedf77a8
Signed-off-by: Kamal Agrawal <kamaagra@codeaurora.org>
