linux-kernel - Re: Loongson (and other $ARCHs?) idle VS timer enqueue

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: <20230422113811.GB1215777@hirez.programming.kicks-ass.net>
Date:   Sat, 22 Apr 2023 13:38:11 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Frederic Weisbecker <frederic@...nel.org>
Cc:     Huacai Chen <chenhuacai@...nel.org>,
        WANG Xuerui <kernel@...0n.name>,
        Thomas Gleixner <tglx@...utronix.de>,
        "Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
        Anna-Maria Behnsen <anna-maria@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: Loongson (and other $ARCHs?) idle VS timer enqueue

On Sat, Apr 22, 2023 at 10:08:14AM +0200, Peter Zijlstra wrote:
> On Fri, Apr 21, 2023 at 06:47:29PM +0200, Frederic Weisbecker wrote:
> 
> > > *HOWEVER*
> > > 
> > > intel_idle_irq() is affected -- except that we only (normally) use that
> > > for very shallow idle states and it won't interact with NOHZ (because we
> > > only disable the tick for deep idle states).
> > 
> > Well I don't know, that doesn't look comfortable... :)
> > 
> > Also why does it need to enable IRQs if ecx=1 ?
> 
> Supposedly this is some interrupt latency hack. See commit:
> 
>   c227233ad64c ("intel_idle: enable interrupts before C1 on Xeons")

Something like so perhaps...

diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h
index 778df05f8539..07a4072c43de 100644
--- a/arch/x86/include/asm/mwait.h
+++ b/arch/x86/include/asm/mwait.h
@@ -115,8 +115,14 @@ static __always_inline void mwait_idle_with_hints(unsigned long eax, unsigned lo
 		}
 
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
-		if (!need_resched())
-			__mwait(eax, ecx);
+		if (!need_resched()) {
+			if (ecx & 1) {
+				__mwait(eax, ecx);
+			} else {
+				__sti_mwait(eax, ecx);
+				raw_local_irq_disable();
+			}
+		}
 	}
 	current_clr_polling();
 }
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 938c17f25d94..4a823bd0f5e0 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -130,11 +130,12 @@ static unsigned int mwait_substates __initdata;
 #define MWAIT2flg(eax) ((eax & 0xFF) << 24)
 
 static __always_inline int __intel_idle(struct cpuidle_device *dev,
-					struct cpuidle_driver *drv, int index)
+					struct cpuidle_driver *drv,
+					int index, bool irqoff)
 {
 	struct cpuidle_state *state = &drv->states[index];
 	unsigned long eax = flg2MWAIT(state->flags);
-	unsigned long ecx = 1; /* break on interrupt flag */
+	unsigned long ecx = 1*irqoff; /* break on interrupt flag */
 
 	mwait_idle_with_hints(eax, ecx);
 
@@ -158,19 +159,13 @@ static __always_inline int __intel_idle(struct cpuidle_device *dev,
 static __cpuidle int intel_idle(struct cpuidle_device *dev,
 				struct cpuidle_driver *drv, int index)
 {
-	return __intel_idle(dev, drv, index);
+	return __intel_idle(dev, drv, index, true);
 }
 
 static __cpuidle int intel_idle_irq(struct cpuidle_device *dev,
 				    struct cpuidle_driver *drv, int index)
 {
-	int ret;
-
-	raw_local_irq_enable();
-	ret = __intel_idle(dev, drv, index);
-	raw_local_irq_disable();
-
-	return ret;
+	return __intel_idle(dev, drv, index, false);
 }
 
 static __cpuidle int intel_idle_ibrs(struct cpuidle_device *dev,
@@ -183,7 +178,7 @@ static __cpuidle int intel_idle_ibrs(struct cpuidle_device *dev,
 	if (smt_active)
 		native_wrmsrl(MSR_IA32_SPEC_CTRL, 0);
 
-	ret = __intel_idle(dev, drv, index);
+	ret = __intel_idle(dev, drv, index, true);
 
 	if (smt_active)
 		native_wrmsrl(MSR_IA32_SPEC_CTRL, spec_ctrl);
@@ -195,7 +190,7 @@ static __cpuidle int intel_idle_xstate(struct cpuidle_device *dev,
 				       struct cpuidle_driver *drv, int index)
 {
 	fpu_idle_fpregs();
-	return __intel_idle(dev, drv, index);
+	return __intel_idle(dev, drv, index, true);
 }
 
 /**