Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions kernel/config/armada-kernel.config.overrides
Original file line number Diff line number Diff line change
Expand Up @@ -351,3 +351,14 @@ CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m
CONFIG_IP_ADVANCED_ROUTER=y
CONFIG_IP_MULTIPLE_TABLES=y
CONFIG_WIREGUARD=m

# ============== Suspend/resume debugging ==============
# Without these, a device that will not stay in deep suspend is undiagnosable:
# pm_wakeup_irq names the IRQ that pulled the SoC out, and pm_test exercises the
# device suspend callbacks without powering down. Found the SM8550 wake sources
# with them; keeping them on so the next regression does not need a rebuild.
CONFIG_PM_DEBUG=y
CONFIG_PM_ADVANCED_DEBUG=y
CONFIG_PM_SLEEP_DEBUG=y
# /sys/kernel/debug/irq/irqs/<n> — "wdepth" shows which IRQs are armed for wake.
CONFIG_GENERIC_IRQ_DEBUGFS=y
19 changes: 19 additions & 0 deletions kernel/dts/qcs8550-ayaneo-pocketevo.dts
Original file line number Diff line number Diff line change
Expand Up @@ -107,3 +107,22 @@
};
};
};

/*
* Drop the fan tachometer interrupt.
*
* qcs8550-ayaneo-pocket-common.dtsi gives pwm-fan an edge-falling TLMM
* interrupt on GPIO 64 for the tacho. Measured on a Pocket EVO it fires about
* 80 times a second whenever the fan spins, and the SoC leaves deep suspend
* after 2-8s; with the fan stopped the same suspend holds for 13-92s. The AYN
* Odin 3, which suspends for hours, declares no tacho interrupt at all.
*
* Overridden here rather than in the shared DTSI because that file is included
* by six other boards whose fan wiring has not been checked.
*
* Trade-off: hwmon fan1_input stops reporting RPM, matching the Odin 3.
*/
&pwm_fan {
/delete-property/ interrupts;
/delete-property/ interrupt-parent;
};
Loading