All of lore.kernel.org
 help / color / mirror / Atom feed
From: will.deacon@arm.com (Will Deacon)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 6/7] ARM: KVM: switch to a dual-step HYP init code
Date: Wed, 3 Apr 2013 11:07:51 +0100	[thread overview]
Message-ID: <20130403100751.GD11040@mudshark.cambridge.arm.com> (raw)
In-Reply-To: <1364909115-3810-7-git-send-email-marc.zyngier@arm.com>

On Tue, Apr 02, 2013 at 02:25:14PM +0100, Marc Zyngier wrote:
> Our HYP init code suffers from two major design issues:
> - it cannot support CPU hotplug, as we tear down the idmap very early
> - it cannot perform a TLB invalidation when switching from init to
>   runtime mappings, as pages are manipulated from PL1 exclusively

[...]

> diff --git a/arch/arm/kvm/init.S b/arch/arm/kvm/init.S
> index 35a463f..b2c6967 100644
> --- a/arch/arm/kvm/init.S
> +++ b/arch/arm/kvm/init.S
> @@ -21,6 +21,7 @@
>  #include <asm/asm-offsets.h>
>  #include <asm/kvm_asm.h>
>  #include <asm/kvm_arm.h>
> +#include <asm/kvm_mmu.h>
>  
>  /********************************************************************
>   * Hypervisor initialization
> @@ -47,6 +48,9 @@ __kvm_hyp_init:
>  	W(b)	.
>  
>  __do_hyp_init:
> +	cmp	r2, #0			@ We have a SP?
> +	bne	phase2			@ Yes, second stage init
> +
>  	@ Set the HTTBR to point to the hypervisor PGD pointer passed
>  	mcrr	p15, 4, r0, r1, c2
>  
> @@ -96,14 +100,35 @@ __do_hyp_init:
>  	orr	r0, r0, r1
>  	isb
>  	mcr	p15, 4, r0, c1, c0, 0	@ HSCR
> -	isb
>  
> -	@ Set stack pointer and return to the kernel
> +	eret
> +
> +phase2:
> +	@ Set stack pointer
>  	mov	sp, r2
>  
>  	@ Set HVBAR to point to the HYP vectors
>  	mcr	p15, 4, r3, c12, c0, 0	@ HVBAR
>  
> +	@ Jump to the trampoline page
> +	ldr	r2, =#PAGE_MASK

Shifting right by PAGE_SHIFT can avoid the load.

> +	adr	r3, target
> +	bic	r3, r3, r2
> +	ldr	r2, =#TRAMPOLINE_VA
> +	add	r3, r3, r2
> +	mov	pc, r3
> +
> +	nop

<insert dead chicken and voodoo chant here>

> +
> +target:	@ We're now in the trampoline code, switch page tables
> +	mcrr	p15, 4, r0, r1, c2
> +	isb
> +
> +	@ Invalidate the old TLBs
> +	mcr	p15, 4, r0, c8, c7, 0	@ TLBIALLH
> +	dsb
> +	isb

You don't actually need this isb (there's an eret next!).

>  	eret
>  
>  	.ltorg

Will

WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will.deacon@arm.com>
To: Marc Zyngier <Marc.Zyngier@arm.com>
Cc: "linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"kvmarm@lists.cs.columbia.edu" <kvmarm@lists.cs.columbia.edu>,
	Catalin Marinas <Catalin.Marinas@arm.com>,
	"cdall@cs.columbia.edu" <cdall@cs.columbia.edu>
Subject: Re: [PATCH 6/7] ARM: KVM: switch to a dual-step HYP init code
Date: Wed, 3 Apr 2013 11:07:51 +0100	[thread overview]
Message-ID: <20130403100751.GD11040@mudshark.cambridge.arm.com> (raw)
In-Reply-To: <1364909115-3810-7-git-send-email-marc.zyngier@arm.com>

On Tue, Apr 02, 2013 at 02:25:14PM +0100, Marc Zyngier wrote:
> Our HYP init code suffers from two major design issues:
> - it cannot support CPU hotplug, as we tear down the idmap very early
> - it cannot perform a TLB invalidation when switching from init to
>   runtime mappings, as pages are manipulated from PL1 exclusively

[...]

> diff --git a/arch/arm/kvm/init.S b/arch/arm/kvm/init.S
> index 35a463f..b2c6967 100644
> --- a/arch/arm/kvm/init.S
> +++ b/arch/arm/kvm/init.S
> @@ -21,6 +21,7 @@
>  #include <asm/asm-offsets.h>
>  #include <asm/kvm_asm.h>
>  #include <asm/kvm_arm.h>
> +#include <asm/kvm_mmu.h>
>  
>  /********************************************************************
>   * Hypervisor initialization
> @@ -47,6 +48,9 @@ __kvm_hyp_init:
>  	W(b)	.
>  
>  __do_hyp_init:
> +	cmp	r2, #0			@ We have a SP?
> +	bne	phase2			@ Yes, second stage init
> +
>  	@ Set the HTTBR to point to the hypervisor PGD pointer passed
>  	mcrr	p15, 4, r0, r1, c2
>  
> @@ -96,14 +100,35 @@ __do_hyp_init:
>  	orr	r0, r0, r1
>  	isb
>  	mcr	p15, 4, r0, c1, c0, 0	@ HSCR
> -	isb
>  
> -	@ Set stack pointer and return to the kernel
> +	eret
> +
> +phase2:
> +	@ Set stack pointer
>  	mov	sp, r2
>  
>  	@ Set HVBAR to point to the HYP vectors
>  	mcr	p15, 4, r3, c12, c0, 0	@ HVBAR
>  
> +	@ Jump to the trampoline page
> +	ldr	r2, =#PAGE_MASK

Shifting right by PAGE_SHIFT can avoid the load.

> +	adr	r3, target
> +	bic	r3, r3, r2
> +	ldr	r2, =#TRAMPOLINE_VA
> +	add	r3, r3, r2
> +	mov	pc, r3
> +
> +	nop

<insert dead chicken and voodoo chant here>

> +
> +target:	@ We're now in the trampoline code, switch page tables
> +	mcrr	p15, 4, r0, r1, c2
> +	isb
> +
> +	@ Invalidate the old TLBs
> +	mcr	p15, 4, r0, c8, c7, 0	@ TLBIALLH
> +	dsb
> +	isb

You don't actually need this isb (there's an eret next!).

>  	eret
>  
>  	.ltorg

Will

  reply	other threads:[~2013-04-03 10:07 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-02 13:25 [PATCH 0/7] ARM: KVM: Revamping the HYP init code for fun and profit Marc Zyngier
2013-04-02 13:25 ` Marc Zyngier
2013-04-02 13:25 ` [PATCH 1/7] ARM: KVM: simplify HYP mapping population Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03 23:13   ` Christoffer Dall
2013-04-03 23:13     ` Christoffer Dall
2013-04-04 12:35     ` Marc Zyngier
2013-04-04 12:35       ` Marc Zyngier
2013-04-02 13:25 ` [PATCH 2/7] ARM: KVM: fix HYP mapping limitations around zero Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03 23:14   ` Christoffer Dall
2013-04-03 23:14     ` Christoffer Dall
2013-04-04 12:40     ` Marc Zyngier
2013-04-04 12:40       ` Marc Zyngier
2013-04-02 13:25 ` [PATCH 3/7] ARM: KVM: move to a KVM provided HYP idmap Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03  9:43   ` Will Deacon
2013-04-03  9:43     ` Will Deacon
2013-04-03  9:46     ` Marc Zyngier
2013-04-03  9:46       ` Marc Zyngier
2013-04-03 23:14   ` Christoffer Dall
2013-04-03 23:14     ` Christoffer Dall
2013-04-02 13:25 ` [PATCH 4/7] ARM: KVM: enforce page alignment for identity mapped code Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03  9:50   ` Will Deacon
2013-04-03  9:50     ` Will Deacon
2013-04-03 10:00     ` Marc Zyngier
2013-04-03 10:00       ` Marc Zyngier
2013-04-03 23:15       ` Christoffer Dall
2013-04-03 23:15         ` Christoffer Dall
2013-04-04 10:47         ` Marc Zyngier
2013-04-04 10:47           ` Marc Zyngier
2013-04-04 15:32           ` Christoffer Dall
2013-04-04 15:32             ` Christoffer Dall
2013-04-02 13:25 ` [PATCH 5/7] ARM: KVM: parametrize HYP page table freeing Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03 23:15   ` Christoffer Dall
2013-04-03 23:15     ` Christoffer Dall
2013-04-02 13:25 ` [PATCH 6/7] ARM: KVM: switch to a dual-step HYP init code Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03 10:07   ` Will Deacon [this message]
2013-04-03 10:07     ` Will Deacon
2013-04-03 10:38     ` Marc Zyngier
2013-04-03 10:38       ` Marc Zyngier
2013-04-03 23:15       ` Christoffer Dall
2013-04-03 23:15         ` Christoffer Dall
2013-04-04 11:05         ` Marc Zyngier
2013-04-04 11:05           ` Marc Zyngier
2013-04-03 23:15   ` Christoffer Dall
2013-04-03 23:15     ` Christoffer Dall
2013-04-04 12:52     ` Marc Zyngier
2013-04-04 12:52       ` Marc Zyngier
2013-04-04 22:10   ` Geoff Levand
2013-04-04 22:10     ` Geoff Levand
2013-04-05  9:08     ` Marc Zyngier
2013-04-05  9:08       ` Marc Zyngier
2013-04-05 16:46       ` Geoff Levand
2013-04-05 16:46         ` Geoff Levand
2013-04-05 16:54         ` Marc Zyngier
2013-04-05 16:54           ` Marc Zyngier
2013-04-18 15:54       ` Russell King - ARM Linux
2013-04-18 15:54         ` Russell King - ARM Linux
2013-04-18 16:01         ` Marc Zyngier
2013-04-18 16:01           ` Marc Zyngier
2013-04-02 13:25 ` [PATCH 7/7] ARM: KVM: perform HYP initilization for hotplugged CPUs Marc Zyngier
2013-04-02 13:25   ` Marc Zyngier
2013-04-03 23:16   ` Christoffer Dall
2013-04-03 23:16     ` Christoffer Dall
2013-04-03 23:18 ` [PATCH 0/7] ARM: KVM: Revamping the HYP init code for fun and profit Christoffer Dall
2013-04-03 23:18   ` Christoffer Dall

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=20130403100751.GD11040@mudshark.cambridge.arm.com \
    --to=will.deacon@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.