Message ID | 1486149538-20432-4-git-send-email-julien.grall@arm.com |
---|---|
State | Superseded |
Headers | show |
Series | None | expand |
diff --git a/xen/arch/arm/acpi/boot.c b/xen/arch/arm/acpi/boot.c index 889208a..65ef632 100644 --- a/xen/arch/arm/acpi/boot.c +++ b/xen/arch/arm/acpi/boot.c @@ -272,6 +272,11 @@ int __init acpi_boot_table_init(void) return 0; disable: + + /* Panic if the user has requested ACPI but Xen is able to initialize. */ + if ( param_acpi_force ) + panic("Unable to use ACPI"); + disable_acpi(); return error;
On ARM, when the user put 'acpi=force' Xen will use ACPI over DT. Currently, if Xen fails to initialize ACPI it will fallback on DT. This behavior makes difficult for a user to notice Xen didn't used ACPI has requested on platform where the firmware is providing both ACPI and DT. Rather than fallback on DT during a failure, panic when 'acpi=force'. Signed-off-by: Julien Grall <julien.grall@arm.com> --- I am wondering if we should do the same with acpi=on. So a user would notice directly if something went wrong with ACPI. Otherwise you would boot up to the prompt and barely notice that DT was used. --- xen/arch/arm/acpi/boot.c | 5 +++++ 1 file changed, 5 insertions(+)