Update "capacity-dmips-mhz" for tuna. It is used to build Energy Model
which in turn is used by EAS to take placement decisions.
Change-Id: If4c0886b8a683e63f32f700f53968a1e2dbd1e42
Signed-off-by: Ankit Sharma <quic_anshar@quicinc.com>
Add qcom,atclk-dependence property to the components that depend on
atclk.
Change-Id: Idbecebe7ea0a1010b9ce0ffc429a540aa6d8f5f1
Signed-off-by: Yuanfang Zhang <quic_yuanfang@quicinc.com>
Add thermal devicetree changes and thermal configuration
for kera based on the recommendation.
Change-Id: I8c1ff2a1da70359b5d8d84959b0ad30c6d37be93
Signed-off-by: Nitesh Kumar <quic_nitekuma@quicinc.com>
Add SPS module to device tree. SPS (Smart Peripheral System)
enables the support of all BAMs in the system which provide DMA
functionality to various peripherals for kera.
Change-Id: I606567fddd7af1f3b724d6a2ea9555d986efe866
Signed-off-by: Prasanna S <quic_prass@quicinc.com>
Add QUPv3(I2C, SPI and UART) and GPI DT nodes on kera.
Change-Id: I3d0db10cd90a59500b29aade5cb9e4017395a1a0
Signed-off-by: Prasanna S <quic_prass@quicinc.com>
Add all 3 PCIe RC configurations for sdxkova. The number of address
cells and size cells are seen as 2 in msm-imem. Based on this add the
register base addresses, MSI register addresses, Host address in ranges
as 64-bit addresses.
Update the PHY settings from latest HSR.
- For PCIe0 and PCIe1 there are no changes in PHY settings, update
the corresponding PHY versions to v1.11 and v1.12.
- For PCIe2, add one change (PCS_G3S2_PRE_GAIN) as per the PHY
version v1.5.
Change-Id: I09519045a13e96878046d905bbe6f2378578c464
Signed-off-by: Sai Chaitanya Kaveti <quic_skaveti@quicinc.com>
Add the device nodes on tuna oemvm and tuivm to enable qmsgq socket
communication over gunyah message queues.
Change-Id: I42be6dabb313e914691945521962cff53f969a02
Signed-off-by: Pranav Mahesh Phansalkar <quic_pphansal@quicinc.com>
Add qrtr gunyah tuivm node to enable communication between
PVM and TUIVM.
Change-Id: If1eebed90a484ef42803fec74629042b04d7a0db
Signed-off-by: Pranav Mahesh Phansalkar <quic_pphansal@quicinc.com>
To access gpios from apps and adsp, we have added apps and
remote flag in dtsi. To identify apps is busy or not, added
qcom-apps flag during active state and added qcom-remote flag
during suspend state, so other SS can use gpios during apps
in suspend state. This flags in pinctrl file for respective
GPIOs will help eGPIO to work for AON camera usecase.
Also, increase TRE size from 64 to 1024 for camera usecase.
Change-Id: Ibc65d9e61d55a3eaeffdf78aef3f8852687f576f
Signed-off-by: Prasanna S <quic_prass@quicinc.com>
Add base device tree support for MTP harmonium and MTP NFC
platforms for Tuna SoC.
Change-Id: I4cef3297884dff5e7edee9121496c7284551c99c
Signed-off-by: Shivendra Pratap <quic_spratap@quicinc.com>
Update memory map for kera, inline with v2.
Change-Id: I28488932a269a5c1ef6c131e9385cf7604ea723e
Signed-off-by: Vijayanand Jitta <quic_vjitta@quicinc.com>
As now interconnect changes are in place, add interconnect voting
for remote processors.
Change-Id: I8a1016f0fe760b49aa88dd51bbb00ef7fdf785fb
Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
Commit I4a4ae323ea0 ("kernel-scripts: sun: Update not_preferred
settings") has updated the default isolated CPUs for Gold
clusters.
Since Isolation CPUs is preferred CPU for irq handling and not prefered
for task handling, update the ESI affinity for CQ to address the same.
Change-Id: I6fbb274d52418e9ebb965c6414c449543796d734
Signed-off-by: Nitin Rawat <quic_nitirawa@quicinc.com>
Currently cpufreq delayed work is queued with delayed
timer as 30ms. On expiry, we monitor load and then
enable storage boost feature if load exceeds certain
predefined threshold.
Unlike pineapple, we are monitoring the load request only
on prime/large cluster and not on medium cluster.
This is causing some additional delay to reach the
threshold required to enable storage boost feature.
Benchmark tools like Antutu completes read or write IO
within 120-130ms which mean any small delay can impact
Antutu to large extent. Hence add medium clusters to perf
score similar to pineapple so that load on medium cluster
along with large cluster is considered.
This will decrease the window time to reach the threshold
to start the storage boost and hence improve Storage
benchmark performance.
Change-Id: I8563cffc4da8fa7729d38fc71c8996b20b79b1ec
Signed-off-by: Nitin Rawat <quic_nitirawa@quicinc.com>
(cherry picked from commit 167166ab42)
