lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201123143510.GR3021@hirez.programming.kicks-ass.net>
Date:   Mon, 23 Nov 2020 15:35:10 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Rafael J. Wysocki" <rafael@...nel.org>
Cc:     Viresh Kumar <viresh.kumar@...aro.org>,
        Ingo Molnar <mingo@...nel.org>,
        the arch/x86 maintainers <x86@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Will Deacon <will@...nel.org>, svens@...ux.ibm.com,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] intel_idle: Fix intel_idle() vs tracing

On Mon, Nov 23, 2020 at 02:54:47PM +0100, Rafael J. Wysocki wrote:

> intel_idle_init_cstates_acpi() needs to be updated too as it doesn't
> pick up the flags automatically.

Ooh, it has two different state init routines :-/, sorry I missed that.
See below.

> It looks like CPUIDLE_FLAG_RCU_IDLE needs to be copied too.

I might need more clue again; processor_idle() needs this because it
calls into ACPI (acpi_idle_enter_bm()) for that BM crud. I didn't find
anything like that here.


---
Subject: intel_idle: Fix intel_idle() vs tracing
From: Peter Zijlstra <peterz@...radead.org>
Date: Fri Nov 20 11:28:35 CET 2020

cpuidle->enter() callbacks should not call into tracing because RCU
has already been disabled. Instead of doing the broadcast thing
itself, simply advertise to the cpuidle core that those states stop
the timer.

Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
 drivers/idle/intel_idle.c |   40 +++++++++++++++++++++++-----------------
 1 file changed, 23 insertions(+), 17 deletions(-)

--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -126,26 +126,9 @@ static __cpuidle int intel_idle(struct c
 	struct cpuidle_state *state = &drv->states[index];
 	unsigned long eax = flg2MWAIT(state->flags);
 	unsigned long ecx = 1; /* break on interrupt flag */
-	bool tick;
-
-	if (!static_cpu_has(X86_FEATURE_ARAT)) {
-		/*
-		 * Switch over to one-shot tick broadcast if the target C-state
-		 * is deeper than C1.
-		 */
-		if ((eax >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK) {
-			tick = true;
-			tick_broadcast_enter();
-		} else {
-			tick = false;
-		}
-	}
 
 	mwait_idle_with_hints(eax, ecx);
 
-	if (!static_cpu_has(X86_FEATURE_ARAT) && tick)
-		tick_broadcast_exit();
-
 	return index;
 }
 
@@ -1227,6 +1210,23 @@ static bool __init intel_idle_acpi_cst_e
 	return false;
 }
 
+static bool __init intel_idle_state_needs_timer_stop(struct cpuidle_state *state)
+{
+	unsigned long eax = flg2MWAIT(state->flags);
+
+	if (boot_cpu_has(X86_FEATURE_ARAT))
+		return false;
+
+	/*
+	 * Switch over to one-shot tick broadcast if the target C-state
+	 * is deeper than C1.
+	 */
+	if ((eax >> MWAIT_SUBSTATE_SIZE) & MWAIT_CSTATE_MASK)
+		return true;
+
+	return false;
+}
+
 static void __init intel_idle_init_cstates_acpi(struct cpuidle_driver *drv)
 {
 	int cstate, limit = min_t(int, CPUIDLE_STATE_MAX, acpi_state_table.count);
@@ -1269,6 +1269,9 @@ static void __init intel_idle_init_cstat
 		if (disabled_states_mask & BIT(cstate))
 			state->flags |= CPUIDLE_FLAG_OFF;
 
+		if (intel_idle_state_needs_timer_stop(state))
+			state->flags |= CPUIDLE_FLAG_TIMER_STOP;
+
 		state->enter = intel_idle;
 		state->enter_s2idle = intel_idle_s2idle;
 	}
@@ -1507,6 +1510,9 @@ static void __init intel_idle_init_cstat
 		     !(cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_ALWAYS_ENABLE)))
 			drv->states[drv->state_count].flags |= CPUIDLE_FLAG_OFF;
 
+		if (intel_idle_state_needs_timer_stop(&drv->states[drv->state_count]))
+			drv->states[drv->state_count].flags |= CPUIDLE_FLAG_TIMER_STOP;
+
 		drv->state_count++;
 	}
 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ