When using QGIC MSI, the MSI controller is GIC not PCIe host
driver. Only set irq chip data if PCIe host is the MSI
controller.
Change-Id: I0ba9e78127424ab3da4b1f46b3ed3fa5d65d05f6
Signed-off-by: Tony Truong <truong@codeaurora.org>
Enable secondary USB port for HDK660 platform which use
host only mode.
Remove TCSR_QUSB2_PHY_CLAMP_DIG_N_1P8 resource entry for SDM660
which is not required to set now during disconnect as writing to this
register causes USB3.0 port and USB2.0 ports working independently.
Change-Id: I936a4012c30e0cd84144b12e34fb780cada541ae
Signed-off-by: Liangliang Lu <luliang@codeaurora.org>
Mhi node needs additional information for driver
probe to get success. This generic node is not
required and hence disable unwanted dtsi node for
msm8996.
Change-Id: I4dffd3d668ac065fc800e5e7d3b52fdf2733e03f
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
Add 500ms before MDM enumeration, this delay is needed for
link establishment between switch and MDM.
Change-Id: I32c4a4f9ef68b7672465ced0401fbf6dc0db0002
Signed-off-by: Rama Krishna Phani A <rphani@codeaurora.org>
This reverts commit 45380e2241 ("msm: sde: add early
display handoff feature"). To avoid conflict when
syncing up with mainline, we revert this change.
Change-Id: I9652cd367d242c6f61ed0da3a36c55493cc7d34c
Signed-off-by: Zhiqiang Tu <ztu@codeaurora.org>
select_best_cpu() has previous CPU's cluster bias which overrides
the best_cpu with best_sibling_cpu when the power cost is same.
When the power table is configured incorrectly or static_cpu_pwr_cost/
static_cluster_pwr_cost tunables are set to a large value, the
power_cost() for all candidate CPUs can return INT_MAX. So the
stats.min_cost is never changed from it's initial value i.e INT_MAX.
In the above scenario, we find stats.best_cpu >= 0 && stats.min_cost =
stats.best_sibling_cpu_cost = INT_MAX && stats.best_sibling_cpu_cost = -1
and replace best_cpu with best_sibling_cpu i.e -1.
Change-Id: I09829e278e41daaaff959428ff50927aba29104c
Signed-off-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Re-phrase PD notification registration debug message and
change state name of SSR/PDR to avoid confusion.
CRs-Fixed: 2057789
Change-Id: I5b800982c0d8c4d5ba627e28f3c9fa8eab7ca074
Signed-off-by: Yuanyuan Liu <yuanliu@codeaurora.org>
Currently platform device is not enabled as a wakeup source, so
notifying the PM core for a wakeup event action will be no-op.
Hence enable it during driver probe.
CRs-Fixed: 2058010
Change-Id: Iefe5bcd4b498a9f02fca001ebcf8ad484f443217
Signed-off-by: Yuanyuan Liu <yuanliu@codeaurora.org>
If the TE irq is alredy requested and registred with a GPIO, then
setting it to true and not continuing with further requesting of
the same.
Change-Id: Iacd1677127c2663d88826e58f1b72704b58db939
Signed-off-by: Rashi Bindra <rbindra@codeaurora.org>
Currently USB driver tries to set power role as none and waits for
disconnect notification to set new mode. But there is a chance that it
might fail and stuck in infinite loop if other device is not dual-role
capable. Fix it by having timeout and bail out in case timeout happens.
Change-Id: I569356ee445c53950ed8fdd3f9994d4ae0e86131
Signed-off-by: Vijayavardhan Vennapusa <vvreddy@codeaurora.org>
This reverts commit ae94b477af.
This change brings side effect of smmu fault in apq8098 project. It blocks their regression test.
Change-Id: Ia479188eca74ec988026f7d42207435da2f79bfc
Signed-off-by: Guchun Chen <guchunc@codeaurora.org>
According to the PD 2.0 Spec, Section 6.2.1.3, the MessageID
field shall only be incremented upon successful transmission
(meaning GoodCRC was received). Thus, increment the counter
only when pd_phy_write() succeeds.
Change-Id: I73dce5736c473a1f79db9be868442c7941d13c44
Signed-off-by: Jack Pham <jackp@codeaurora.org>
