public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Marc Zyngier <marc.zyngier@arm.com>
To: Christoffer Dall <cdall@linaro.org>,
	kvmarm@lists.cs.columbia.edu,
	linux-arm-kernel@lists.infradead.org
Cc: kvm@vger.kernel.org, Will Deacon <will.deacon@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>
Subject: Re: [PATCH v3 02/20] arm64: Use physical counter for in-kernel reads
Date: Mon, 9 Oct 2017 17:10:40 +0100	[thread overview]
Message-ID: <0f84a7fb-14cd-d172-6b90-eefb05a87aea@arm.com> (raw)
In-Reply-To: <20170923004207.22356-3-cdall@linaro.org>

On 23/09/17 01:41, Christoffer Dall wrote:
> Using the physical counter allows KVM to retain the offset between the
> virtual and physical counter as long as it is actively running a VCPU.
> 
> As soon as a VCPU is released, another thread is scheduled or we start
> running userspace applications, we reset the offset to 0, so that
> userspace accessing the virtual timer can still read the cirtual counter

s/cirtual/virtual/

> and get the same view of time as the kernel.
> 
> This opens up potential improvements for KVM performance.
> 
> VHE kernels or kernels continuing to use the virtual timer are
> unaffected.
> 
> Signed-off-by: Christoffer Dall <cdall@linaro.org>
> ---
>  arch/arm64/include/asm/arch_timer.h  | 9 ++++-----
>  drivers/clocksource/arm_arch_timer.c | 3 +--
>  2 files changed, 5 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/arch_timer.h b/arch/arm64/include/asm/arch_timer.h
> index a652ce0..1859a1c 100644
> --- a/arch/arm64/include/asm/arch_timer.h
> +++ b/arch/arm64/include/asm/arch_timer.h
> @@ -148,11 +148,10 @@ static inline void arch_timer_set_cntkctl(u32 cntkctl)
>  
>  static inline u64 arch_counter_get_cntpct(void)
>  {
> -	/*
> -	 * AArch64 kernel and user space mandate the use of CNTVCT.
> -	 */
> -	BUG();
> -	return 0;
> +	u64 cval;
> +	isb();
> +	asm volatile("mrs %0, cntpct_el0" : "=r" (cval));
> +	return cval;

This would be just fine if we were blessed with quality HW. This is 
unfortunately not the case, and we need a staggering amount of crap to 
deal with timer errata.

I suggest you replace this with the (fully untested) following:

diff --git a/arch/arm64/include/asm/arch_timer.h b/arch/arm64/include/asm/arch_timer.h
index a652ce0a5cb2..04275de614db 100644
--- a/arch/arm64/include/asm/arch_timer.h
+++ b/arch/arm64/include/asm/arch_timer.h
@@ -52,6 +52,7 @@ struct arch_timer_erratum_workaround {
 	const char *desc;
 	u32 (*read_cntp_tval_el0)(void);
 	u32 (*read_cntv_tval_el0)(void);
+	u64 (*read_cntpct_el0)(void);
 	u64 (*read_cntvct_el0)(void);
 	int (*set_next_event_phys)(unsigned long, struct clock_event_device *);
 	int (*set_next_event_virt)(unsigned long, struct clock_event_device *);
@@ -148,11 +149,8 @@ static inline void arch_timer_set_cntkctl(u32 cntkctl)
 
 static inline u64 arch_counter_get_cntpct(void)
 {
-	/*
-	 * AArch64 kernel and user space mandate the use of CNTVCT.
-	 */
-	BUG();
-	return 0;
+	isb();
+	return arch_timer_reg_read_stable(cntpct_el0);
 }
 
 static inline u64 arch_counter_get_cntvct(void)
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index fd4b7f684bd0..5b41a96fa8dd 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -217,6 +217,11 @@ static u32 notrace fsl_a008585_read_cntv_tval_el0(void)
 	return __fsl_a008585_read_reg(cntv_tval_el0);
 }
 
+static u64 notrace fsl_a008585_read_cntpct_el0(void)
+{
+	return __fsl_a008585_read_reg(cntpct_el0);
+}
+
 static u64 notrace fsl_a008585_read_cntvct_el0(void)
 {
 	return __fsl_a008585_read_reg(cntvct_el0);
@@ -258,6 +263,11 @@ static u32 notrace hisi_161010101_read_cntv_tval_el0(void)
 	return __hisi_161010101_read_reg(cntv_tval_el0);
 }
 
+static u64 notrace hisi_161010101_read_cntpct_el0(void)
+{
+	return __hisi_161010101_read_reg(cntpct_el0);
+}
+
 static u64 notrace hisi_161010101_read_cntvct_el0(void)
 {
 	return __hisi_161010101_read_reg(cntvct_el0);
@@ -296,6 +306,15 @@ static u64 notrace arm64_858921_read_cntvct_el0(void)
 	new = read_sysreg(cntvct_el0);
 	return (((old ^ new) >> 32) & 1) ? old : new;
 }
+
+static u64 notrace arm64_858921_read_cntpct_el0(void)
+{
+	u64 old, new;
+
+	old = read_sysreg(cntpct_el0);
+	new = read_sysreg(cntpct_el0);
+	return (((old ^ new) >> 32) & 1) ? old : new;
+}
 #endif
 
 #ifdef CONFIG_ARM_ARCH_TIMER_OOL_WORKAROUND
@@ -310,7 +329,7 @@ static void erratum_set_next_event_tval_generic(const int access, unsigned long
 						struct clock_event_device *clk)
 {
 	unsigned long ctrl;
-	u64 cval = evt + arch_counter_get_cntvct();
+	u64 cval = evt + arch_timer_read_counter();
 
 	ctrl = arch_timer_reg_read(access, ARCH_TIMER_REG_CTRL, clk);
 	ctrl |= ARCH_TIMER_CTRL_ENABLE;
@@ -346,6 +365,7 @@ static const struct arch_timer_erratum_workaround ool_workarounds[] = {
 		.desc = "Freescale erratum a005858",
 		.read_cntp_tval_el0 = fsl_a008585_read_cntp_tval_el0,
 		.read_cntv_tval_el0 = fsl_a008585_read_cntv_tval_el0,
+		.read_cntpct_el0 = fsl_a008585_read_cntpct_el0,
 		.read_cntvct_el0 = fsl_a008585_read_cntvct_el0,
 		.set_next_event_phys = erratum_set_next_event_tval_phys,
 		.set_next_event_virt = erratum_set_next_event_tval_virt,
@@ -358,6 +378,7 @@ static const struct arch_timer_erratum_workaround ool_workarounds[] = {
 		.desc = "HiSilicon erratum 161010101",
 		.read_cntp_tval_el0 = hisi_161010101_read_cntp_tval_el0,
 		.read_cntv_tval_el0 = hisi_161010101_read_cntv_tval_el0,
+		.read_cntpct_el0 = hisi_161010101_read_cntpct_el0,
 		.read_cntvct_el0 = hisi_161010101_read_cntvct_el0,
 		.set_next_event_phys = erratum_set_next_event_tval_phys,
 		.set_next_event_virt = erratum_set_next_event_tval_virt,
@@ -368,6 +389,7 @@ static const struct arch_timer_erratum_workaround ool_workarounds[] = {
 		.desc = "HiSilicon erratum 161010101",
 		.read_cntp_tval_el0 = hisi_161010101_read_cntp_tval_el0,
 		.read_cntv_tval_el0 = hisi_161010101_read_cntv_tval_el0,
+		.read_cntpct_el0 = hisi_161010101_read_cntpct_el0,
 		.read_cntvct_el0 = hisi_161010101_read_cntvct_el0,
 		.set_next_event_phys = erratum_set_next_event_tval_phys,
 		.set_next_event_virt = erratum_set_next_event_tval_virt,
@@ -378,6 +400,7 @@ static const struct arch_timer_erratum_workaround ool_workarounds[] = {
 		.match_type = ate_match_local_cap_id,
 		.id = (void *)ARM64_WORKAROUND_858921,
 		.desc = "ARM erratum 858921",
+		.read_cntpct_el0 = arm64_858921_read_cntpct_el0,
 		.read_cntvct_el0 = arm64_858921_read_cntvct_el0,
 	},
 #endif

which ensures (at least in theory) that we do the right thing on our buggy
hardware...

>  }
>  
>  static inline u64 arch_counter_get_cntvct(void)
> diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
> index fd4b7f6..9b3322a 100644
> --- a/drivers/clocksource/arm_arch_timer.c
> +++ b/drivers/clocksource/arm_arch_timer.c
> @@ -890,8 +890,7 @@ static void __init arch_counter_register(unsigned type)
>  
>  	/* Register the CP15 based counter if we have one */
>  	if (type & ARCH_TIMER_TYPE_CP15) {
> -		if (IS_ENABLED(CONFIG_ARM64) ||
> -		    arch_timer_uses_ppi == ARCH_TIMER_VIRT_PPI)
> +		if (arch_timer_uses_ppi == ARCH_TIMER_VIRT_PPI)
>  			arch_timer_read_counter = arch_counter_get_cntvct;
>  		else
>  			arch_timer_read_counter = arch_counter_get_cntpct;
> 

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...

  reply	other threads:[~2017-10-09 16:10 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-23  0:41 [PATCH v3 00/20] KVM: arm/arm64: Optimize arch timer register handling Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 01/20] irqchip/gic: Deal with broken firmware exposing only 4kB of GICv2 CPU interface Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 02/20] arm64: Use physical counter for in-kernel reads Christoffer Dall
2017-10-09 16:10   ` Marc Zyngier [this message]
2017-10-17 15:33   ` Will Deacon
2017-10-18 10:00     ` Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 03/20] arm64: Use the physical counter when available for read_cycles Christoffer Dall
2017-10-09 16:21   ` Marc Zyngier
2017-10-18 11:34     ` Christoffer Dall
2017-10-18 15:52       ` Marc Zyngier
2017-09-23  0:41 ` [PATCH v3 04/20] KVM: arm/arm64: Guard kvm_vgic_map_is_active against !vgic_initialized Christoffer Dall
2017-10-09 16:22   ` Marc Zyngier
2017-09-23  0:41 ` [PATCH v3 05/20] KVM: arm/arm64: Support calling vgic_update_irq_pending from irq context Christoffer Dall
2017-10-09 16:37   ` Marc Zyngier
2017-10-18 11:54     ` Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 06/20] KVM: arm/arm64: Check that system supports split eoi/deactivate Christoffer Dall
2017-10-09 16:47   ` Marc Zyngier
2017-10-18 13:41     ` Christoffer Dall
2017-10-18 16:03       ` Marc Zyngier
2017-10-18 19:16         ` Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 07/20] KVM: arm/arm64: Make timer_arm and timer_disarm helpers more generic Christoffer Dall
2017-10-09 17:05   ` Marc Zyngier
2017-10-18 16:47     ` Christoffer Dall
2017-10-18 16:53       ` Marc Zyngier
2017-09-23  0:41 ` [PATCH v3 08/20] KVM: arm/arm64: Rename soft timer to bg_timer Christoffer Dall
2017-10-09 17:06   ` Marc Zyngier
2017-09-23  0:41 ` [PATCH v3 09/20] KVM: arm/arm64: Use separate timer for phys timer emulation Christoffer Dall
2017-10-09 17:23   ` Marc Zyngier
2017-10-19  7:38     ` Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 10/20] KVM: arm/arm64: Move timer/vgic flush/sync under disabled irq Christoffer Dall
2017-10-09 17:34   ` Marc Zyngier
2017-09-23  0:41 ` [PATCH v3 11/20] KVM: arm/arm64: Move timer save/restore out of the hyp code Christoffer Dall
2017-10-09 17:47   ` Marc Zyngier
2017-10-19  7:46     ` Christoffer Dall
2017-09-23  0:41 ` [PATCH v3 12/20] genirq: Document vcpu_info usage for percpu_devid interrupts Christoffer Dall
2017-10-09 17:48   ` Marc Zyngier
2017-09-23  0:42 ` [PATCH v3 13/20] KVM: arm/arm64: Set VCPU affinity for virt timer irq Christoffer Dall
2017-10-09 17:52   ` Marc Zyngier
2017-09-23  0:42 ` [PATCH v3 14/20] KVM: arm/arm64: Avoid timer save/restore in vcpu entry/exit Christoffer Dall
2017-10-10  8:47   ` Marc Zyngier
2017-10-19  8:15     ` Christoffer Dall
2017-09-23  0:42 ` [PATCH v3 15/20] KVM: arm/arm64: Support EL1 phys timer register access in set/get reg Christoffer Dall
2017-10-10  9:10   ` Marc Zyngier
2017-10-19  8:32     ` Christoffer Dall
2017-09-23  0:42 ` [PATCH v3 16/20] KVM: arm/arm64: Use kvm_arm_timer_set/get_reg for guest register traps Christoffer Dall
2017-10-10  9:12   ` Marc Zyngier
2017-09-23  0:42 ` [PATCH v3 17/20] KVM: arm/arm64: Move phys_timer_emulate function Christoffer Dall
2017-10-10  9:21   ` Marc Zyngier
2017-09-23  0:42 ` [PATCH v3 18/20] KVM: arm/arm64: Avoid phys timer emulation in vcpu entry/exit Christoffer Dall
2017-10-10  9:45   ` Marc Zyngier
2017-10-19  8:44     ` Christoffer Dall
2017-09-23  0:42 ` [PATCH v3 19/20] KVM: arm/arm64: Get rid of kvm_timer_flush_hwstate Christoffer Dall
2017-10-10  9:46   ` Marc Zyngier
2017-09-23  0:42 ` [PATCH v3 20/20] KVM: arm/arm64: Rework kvm_timer_should_fire Christoffer Dall
2017-10-10  9:59   ` Marc Zyngier

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0f84a7fb-14cd-d172-6b90-eefb05a87aea@arm.com \
    --to=marc.zyngier@arm.com \
    --cc=catalin.marinas@arm.com \
    --cc=cdall@linaro.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=will.deacon@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox