summaryrefslogtreecommitdiffstats
path: root/drivers/acpi
diff options
context:
space:
mode:
authorDavid Woodhouse <dwmw@amazon.co.uk>2021-11-08 16:09:41 +0000
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>2021-12-08 16:06:10 +0100
commit74d9555580c48a04b2c3b742dfb0c80777aa0b26 (patch)
tree7fcb1aa611da4680ea812d496012f10910382a82 /drivers/acpi
parent0fcfb00b28c0b7884635dacf38e46d60bf3d4eb1 (diff)
downloadlinux-stable-74d9555580c48a04b2c3b742dfb0c80777aa0b26.tar.gz
linux-stable-74d9555580c48a04b2c3b742dfb0c80777aa0b26.tar.bz2
linux-stable-74d9555580c48a04b2c3b742dfb0c80777aa0b26.zip
PM: hibernate: Allow ACPI hardware signature to be honoured
Theoretically, when the hardware signature in FACS changes, the OS is supposed to gracefully decline to attempt to resume from S4: "If the signature has changed, OSPM will not restore the system context and can boot from scratch" In practice, Windows doesn't do this and many laptop vendors do allow the signature to change especially when docking/undocking, so it would be a bad idea to simply comply with the specification by default in the general case. However, there are use cases where we do want the compliant behaviour and we know it's safe. Specifically, when resuming virtual machines where we know the hypervisor has changed sufficiently that resume will fail. We really want to be able to *tell* the guest kernel not to try, so it boots cleanly and doesn't just crash. This patch provides a way to opt in to the spec-compliant behaviour on the command line. A follow-up patch may do this automatically for certain "known good" machines based on a DMI match, or perhaps just for all hypervisor guests since there's no good reason a hypervisor would change the hardware_signature that it exposes to guests *unless* it wants them to obey the ACPI specification. Signed-off-by: David Woodhouse <dwmw@amazon.co.uk> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Diffstat (limited to 'drivers/acpi')
-rw-r--r--drivers/acpi/sleep.c26
1 files changed, 21 insertions, 5 deletions
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index eaa47753b758..4dfbfc69db34 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -877,11 +877,11 @@ static inline void acpi_sleep_syscore_init(void) {}
#ifdef CONFIG_HIBERNATION
static unsigned long s4_hardware_signature;
static struct acpi_table_facs *facs;
-static bool nosigcheck;
+static int sigcheck = -1; /* Default behaviour is just to warn */
-void __init acpi_no_s4_hw_signature(void)
+void __init acpi_check_s4_hw_signature(int check)
{
- nosigcheck = true;
+ sigcheck = check;
}
static int acpi_hibernation_begin(pm_message_t stage)
@@ -1009,12 +1009,28 @@ static void acpi_sleep_hibernate_setup(void)
hibernation_set_ops(old_suspend_ordering ?
&acpi_hibernation_ops_old : &acpi_hibernation_ops);
sleep_states[ACPI_STATE_S4] = 1;
- if (nosigcheck)
+ if (!sigcheck)
return;
acpi_get_table(ACPI_SIG_FACS, 1, (struct acpi_table_header **)&facs);
- if (facs)
+ if (facs) {
+ /*
+ * s4_hardware_signature is the local variable which is just
+ * used to warn about mismatch after we're attempting to
+ * resume (in violation of the ACPI specification.)
+ */
s4_hardware_signature = facs->hardware_signature;
+
+ if (sigcheck > 0) {
+ /*
+ * If we're actually obeying the ACPI specification
+ * then the signature is written out as part of the
+ * swsusp header, in order to allow the boot kernel
+ * to gracefully decline to resume.
+ */
+ swsusp_hardware_signature = facs->hardware_signature;
+ }
+ }
}
#else /* !CONFIG_HIBERNATION */
static inline void acpi_sleep_hibernate_setup(void) {}