Diagnostic query: Unexpected kernel panic on macOS Sequoia 15.1 after waking (1794694837)

[Owen Parker]

Following the recent point release update, my system has encountered intermittent sleep/wake panics. The panic log cites IOPlatformPluginFamily and AppleThunderboltNHI assertions.

I have already tested clearing NVRAM, booted in Safe Mode to isolate third-party kexts, and disabled Power Nap. Has anyone isolated whether this is linked to external DisplayPort docks or an APFS snapshot indexing bottleneck?

[Hannah Brooks]

The AppleThunderboltNHI assertion during wake strongly correlates with mds_stores holding an I/O lock on the APFS snapshot volume while the Thunderbolt controller attempts to re-enumerate the DisplayPort alt-mode path; verify via log show --predicate 'process == "mds_stores" && eventMessage contains "snapshot"' --last 1h immediately post-panic. If the dock presents an NVMe target, disable Spotlight indexing on that volume (mdutil -i off /Volumes/DockVolume) to eliminate the metadata contention window. A clean install migrating only user data via Migration Assistant (excluding System/Applications) resolves the IOPlatformPluginFamily plugin mismatch often introduced by in-place 15.0→15.1 upgrades.

MacBook Air M2 (16GB / 512GB) | iPad mini 7

Portland, OR | "Export solved community threads to PDF while they load."

[Liam O’Connor]

The NHI panic frequently originates from the Thunderbolt NHI firmware failing to relinquish the PCIe link ownership to the AppleThunderboltDPAdapter driver when a paired AirPods handoff or HomeKit Bluetooth LE scan triggers a PME (Power Management Event) storm during dark wake. Disable “Allow Handoff between this Mac and your iCloud devices” and toggle Bluetooth off via blueutil -p 0 in a launchd SleepServices script (/Library/LaunchDaemons/com.local.sleepwake.plist) to isolate the radio coexistence interrupt. If stable, update the dock’s PD controller firmware (CalDigit/OWC/CalDigit utilities) as the root cause is typically a stale Thunderbolt NVM firmware violating the Sequoia 15.1 ACPI _DSM contract for D3cold transitions.

iMac 24″ M3 (16GB) | HomePod mini Stereo Pair

Dublin, IE | "Clean Wi-Fi multicast fixes half of AirPlay drops."

[Rachel Kim]

WindowServer spikes on wake are a symptom, not the cause—the panic backtrace IOPlatformPluginFamily::getPlatform() -> AppleThunderboltNHI::interruptOccurred() indicates the display driver (AppleMuxControl/AGDC) is blocked waiting for the NHI to signal link training completion on the DisplayPort MST hub. Strip all login items (osascript -e 'tell application "System Events" to get the name of every login item') and disable WindowServer hardware acceleration for Safari (defaults write com.apple.Safari WebKitEnableGPUAcceleration -bool false) to reduce the GPU reset pressure during the critical 2-second wake window. If the panic ceases, the dock’s MST hub firmware is violating the kIOFBConnectFlags handshake; a Thunderbolt 4 certified cable (active 0.8m/40Gbps) often resolves the signal integrity failure triggering the NHI watchdog.

MacBook Pro 14″ M4 (32GB) | iPhone 15

Vancouver, BC | "Read crash logs bottom-up from the faulting thread."

[Author]

Posts