Merge pull request #831 from kernelkit/boot-wait-slow-devs

Wait for slow devices before continuing boot

board/common/rootfs/etc/finit.d/available/sysklogd.conf

@@ -0,0 +1,4 @@
+# Use <pid/syslogd> as barrier for other system tasks and service that
+# rely on modules, firmware, and device nodes to be ready.
+service if:udevd nowarn env:-/etc/default/sysklogd <run/udevadm:post/success> \
+	[S0123456789] syslogd -F $SYSLOGD_ARGS -- System log daemon

board/common/rootfs/usr/lib/finit/system/20-hw-wait.conf

@@ -0,0 +1,9 @@
+# Extend finit's default udevadm settle synchronization for situations
+# where device are very slow to probe (see #685)
+run nowarn if:udevd cgroup.init <service/udevd/ready> log \
+	[S] /usr/libexec/infix/hw-wait -- Probing hardware
+
+# Now that everything should be probed, do a final pass over the
+# uevent queue before starting syslogd and everything else
+run nowarn if:udevd cgroup.init :post <service/udevd/ready> log \
+	[S] udevadm settle -t 30 --

board/common/rootfs/usr/libexec/infix/hw-wait

@@ -0,0 +1,41 @@
+#!/bin/sh
+# (Ab)use the kernel's device link subsystem to detect consumer side
+# devices that may be very slow to probe (looking at you, mv88e6xxx!).
+
+ident=$(basename "$0")
+
+report()
+{
+    if [ -r "/tmp/$ident" ]; then
+	logger -k -p "user.$1" -t "$ident" "Waited for slow devices:"
+	sort "/tmp/$ident" | uniq -c | logger -k -p "user.$1" -t "$ident"
+    fi
+
+    rm -f "/tmp/$ident"
+}
+
+for _ in $(seq 50); do
+    again=
+
+    for dl in /sys/class/devlink/*; do
+    	[ -r "$dl/status" ] || continue
+
+    	status=$(cat "$dl/status")
+	if [ "$status" = "consumer probing" ]; then
+	    basename "$(readlink "$dl/consumer")" >>"/tmp/$ident"
+	    again=yes
+	fi
+
+    done
+
+    if [ -z "$again" ]; then
+	report notice
+	exit 0
+    fi
+
+    sleep .2
+done
+
+logger -k -p user.error -t "$ident" "Timeout waiting for devices to come online"
+report error
+exit 1

board/common/rootfs/usr/libexec/infix/init.d/20-nameif

@@ -14,24 +14,6 @@ for file in /sys/firmware/qemu_fw_cfg/by_name/opt/mactab/raw /etc/mactab; do
 	fi
 done
 
-# Sometimes the sysfs is not populated when the switch driver is loaded, with the result
-# that the DSA interface was not found (no /dsa/tagging entry in sysfs. See issue #685.
-#
-# This mitigates that problem by waiting for sysfs to come up if a DSA switch is found
-if [ -n "$(devlink -j dev info | jq -r '.info.[].driver' | grep -q mv88e6085)" ]; then
-	timeout=50
-	while [ -z "$(ls /sys/class/net/*/dsa/tagging)" ]; do
-		timeout=$((timeout-1))
-		if [ $timeout -eq 0 ]; then
-			logger -k -p user.emerg -t "$ident" "Failed to find DSA interface"
-			exit 1
-		fi
-		sleep 0.1
-	done
-
-	logger -k -p user.notice -t "$ident" "Found DSA interface in $timeout seconds"
-fi
-
 # Find CPU interfaces used for connecting to a switch managed by DSA
 for netif in /sys/class/net/*; do
 	iface=$(basename "$netif")

doc/ChangeLog.md

@@ -4,6 +4,16 @@ Change Log
 All notable changes to the project are documented in this file.
 
 
+[v24.11.1][UNRELEASED]
+-------------------------
+
+### Fixes
+
+ - Fix #685: DSA conduit interface not always detected.  Previous
+   attempt at a fix (v24.10.2) mitigated the issue, but did not
+   completely solve it.
+
+
 [v24.11.0][] - 2024-11-20
 -------------------------