diff --git a/drivers/acpid/src/main.rs b/drivers/acpid/src/main.rs
index 0e97d13649..6c362d8512 100644
--- a/drivers/acpid/src/main.rs
+++ b/drivers/acpid/src/main.rs
@@ -295,6 +295,40 @@ fn daemon(daemon: daemon::Daemon) -> ! {
                     }
                     power_events::PowerEvent::LidChanged(open) => {
                         log::info!("acpid: lid {}", if open { "opened" } else { "closed" });
+                        // Symmetric lid-switch → s2idle wiring. Mirrors
+                        // Linux logind's `HandleLidSwitch=suspend` default
+                        // for laptops.
+                        //
+                        // Lid-closed runs `enter_s2idle()` — the AML
+                        // preparation sequence (_TTS/_PTS/_SST + wake-device
+                        // arming via _DSW/_PSW) so the system is in the
+                        // correct state when the kernel MWAIT loop (Phase
+                        // 9.1, not yet landed) actually idles the CPU.
+                        //
+                        // Lid-open runs `exit_s2idle()` — the AML wake
+                        // sequence (_SST(2) → _WAK(0) → _SST(1) + wake-device
+                        // disarming). This is the userspace-driven wake path
+                        // for the case where the kernel MWAIT loop hasn't
+                        // engaged yet (e.g. a brief lid-close/open during
+                        // active workload). When MWAIT IS engaged, the
+                        // kernel-side MWAIT-return path fires kstop
+                        // reason=2 which ALSO calls exit_s2idle() — the
+                        // double-call is safe because the AML _WAK/_SST
+                        // methods are spec-required to be idempotent in
+                        // the working state (ACPI 6.5 §3.5.3).
+                        //
+                        // External-display detection (Linux's
+                        // HandleLidSwitchDocked=ignore) is not wired yet —
+                        // every lid-close triggers s2idle. Documented as a
+                        // follow-up in local/docs/evidence/lg-gram/
+                        // ASSESSMENT-2026-07-26.md.
+                        if open {
+                            log::info!("acpid: lid opened — running s2idle wake sequence");
+                            acpi_context.exit_s2idle();
+                        } else {
+                            log::info!("acpid: lid closed — entering s2idle preparation");
+                            acpi_context.enter_s2idle();
+                        }
                     }
                     power_events::PowerEvent::Notify { device, value } => {
                         log::debug!("acpid: device notification {} = {:#x}", device, value);