Merge 4.9.153 into android-4.9

Changes in 4.9.153
	r8169: Add support for new Realtek Ethernet
	ipv6: Consider sk_bound_dev_if when binding a socket to a v4 mapped address
	ipv6: Take rcu_read_lock in __inet6_bind for mapped addresses
	platform/x86: asus-wmi: Tell the EC the OS will handle the display off hotkey
	e1000e: allow non-monotonic SYSTIM readings
	writeback: don't decrement wb->refcnt if !wb->bdi
	serial: set suppress_bind_attrs flag only if builtin
	ALSA: oxfw: add support for APOGEE duet FireWire
	MIPS: SiByte: Enable swiotlb for SWARM, LittleSur and BigSur
	arm64: perf: set suppress_bind_attrs flag to true
	selinux: always allow mounting submounts
	rxe: IB_WR_REG_MR does not capture MR's iova field
	jffs2: Fix use of uninitialized delayed_work, lockdep breakage
	pstore/ram: Do not treat empty buffers as valid
	powerpc/xmon: Fix invocation inside lock region
	powerpc/pseries/cpuidle: Fix preempt warning
	media: firewire: Fix app_info parameter type in avc_ca{,_app}_info
	net: call sk_dst_reset when set SO_DONTROUTE
	scsi: target: use consistent left-aligned ASCII INQUIRY data
	clk: imx6q: reset exclusive gates on init
	kconfig: fix file name and line number of warn_ignored_character()
	kconfig: fix memory leak when EOF is encountered in quotation
	mmc: atmel-mci: do not assume idle after atmci_request_end
	tty/serial: do not free trasnmit buffer page under port lock
	perf intel-pt: Fix error with config term "pt=0"
	perf svghelper: Fix unchecked usage of strncpy()
	perf parse-events: Fix unchecked usage of strncpy()
	dm kcopyd: Fix bug causing workqueue stalls
	tools lib subcmd: Don't add the kernel sources to the include path
	dm snapshot: Fix excessive memory usage and workqueue stalls
	ALSA: bebob: fix model-id of unit for Apogee Ensemble
	sysfs: Disable lockdep for driver bind/unbind files
	scsi: smartpqi: correct lun reset issues
	scsi: megaraid: fix out-of-bound array accesses
	ocfs2: fix panic due to unrecovered local alloc
	mm/page-writeback.c: don't break integrity writeback on ->writepage() error
	mm, proc: be more verbose about unstable VMA flags in /proc/<pid>/smaps
	ipmi:ssif: Fix handling of multi-part return messages
	locking/qspinlock: Pull in asm/byteorder.h to ensure correct endianness
	Linux 4.9.153

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

1 files changed, 5 insertions, 1 deletions

diff --git a/drivers/clk/imx/clk-imx6q.c b/drivers/clk/imx/clk-imx6q.c
index 93a19667003d..14682df5d312 100644
--- a/drivers/clk/imx/clk-imx6q.c
+++ b/drivers/clk/imx/clk-imx6q.c
@@ -258,8 +258,12 @@ static void __init imx6q_clocks_init(struct device_node *ccm_node)
 	 * lvds1_gate and lvds2_gate are pseudo-gates.  Both can be
 	 * independently configured as clock inputs or outputs.  We treat
 	 * the "output_enable" bit as a gate, even though it's really just
-	 * enabling clock output.
+	 * enabling clock output. Initially the gate bits are cleared, as
+	 * otherwise the exclusive configuration gets locked in the setup done
+	 * by software running before the clock driver, with no way to change
+	 * it.
 	 */
+	writel(readl(base + 0x160) & ~0x3c00, base + 0x160);
 	clk[IMX6QDL_CLK_LVDS1_GATE] = imx_clk_gate_exclusive("lvds1_gate", "lvds1_sel", base + 0x160, 10, BIT(12));
 	clk[IMX6QDL_CLK_LVDS2_GATE] = imx_clk_gate_exclusive("lvds2_gate", "lvds2_sel", base + 0x160, 11, BIT(13));