https://bugzilla.kernel.org/show_bug.cgi?id=221646
--- Comment #8 from Nikhil Jeikar P M ([email protected]) --- Follow-up: additional boot-media / storage behaviour with acpi=off workaround Since the original report, I did more testing on the same system (MSI PRO X870E-S EVO WIFI / MS-7E86, BIOS 2.A30, Fedora 44, kernel 7.0.12-200.fc44.x86_64). Summary With normal ACPI enabled, early boot still hangs/fails as originally reported. With acpi=off, the installed system can boot, but live USB media and some NVMe boot paths remain unreliable. Swapping M.2 slots did not change this behaviour. Storage / boot-media observations Two NVMe drives are installed: PCI address Device Link Notes 0000:03:00.0 Micron 3400 MTFDKBA1T0TFH PCIe 4.0 x4 Can boot installed OS more reliably 0000:0a:00.0 Samsung SSD 980 1TB PCIe 3.0 x4 Fresh install / boot attempts fail more often USB live installer media also fails consistently (keyboard/mouse lost, no I/O, appears hung). Windows setup reports missing drivers for onboard devices (RTL8126 5GbE, MT7925 WiFi 7, ASM4242 USB4/TB). I initially thought this was slot- or PCIe-generation-specific, but after swapping drives between M.2 slots the problem persisted. Both NVMe controllers still report the same IRQ failures under acpi=off; NVMe may continue using poll queues, while USB/xHCI does not recover reliably during early boot/initramfs. acpi=off side effects (from attached logs / current dmesg) When booting with acpi=off: ACPI: Interpreter disabled. APIC: ACPI MADT or MP tables are not detected NR_IRQS: 524544, nr_irqs: 32, preallocated irqs: 16 About 46 devices log: can't find IRQ for PCI INT A; please try using pci=biosirq Affected devices include all xHCI controllers (10:00.0, 12:00.0, 76:00.0, 78:00.3, 78:00.4, 79:00.0), both NVMe controllers, ASMedia ASM4242 USB4 host router (77:00.0), RTL8126 (0c:00.0), MT7925 (0d:00.0), and multiple chipset PCIe switches. Thunderbolt/USB4 fails completely: thunderbolt 0000:77:00.0: device links to tunneled native ports are missing! thunderbolt 0000:77:00.0: probe with driver thunderbolt failed with error -110 USB mass-storage on the AMD 800-series xHCI controller (10:00.0, bus 1 port 5) times out during early boot: usb 1-5: device descriptor read/64, error -110 -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug. _______________________________________________ acpi-bugzilla mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
