From: marc.zyngier@arm.com (Marc Zyngier)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v3 25/32] arm64: KVM: 32bit conditional execution emulation
Date: Wed, 24 Apr 2013 14:13:37 +0100 [thread overview]
Message-ID: <5177DA81.3020400@arm.com> (raw)
In-Reply-To: <20130423230057.GJ20569@gmail.com>
On 24/04/13 00:00, Christoffer Dall wrote:
> On Mon, Apr 08, 2013 at 05:17:27PM +0100, Marc Zyngier wrote:
>> As conditional instructions can trap on AArch32, add the thinest
>> possible emulation layer to keep 32bit guests happy.
>>
>> Reviewed-by: Christopher Covington <cov@codeaurora.org>
>> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>> arch/arm64/include/asm/kvm_emulate.h | 13 ++-
>> arch/arm64/kvm/Makefile | 2 +-
>> arch/arm64/kvm/emulate.c | 154 +++++++++++++++++++++++++++++++++++
>> 3 files changed, 166 insertions(+), 3 deletions(-)
>> create mode 100644 arch/arm64/kvm/emulate.c
>>
>> diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
>> index 37a6567..8d4ab33 100644
>> --- a/arch/arm64/include/asm/kvm_emulate.h
>> +++ b/arch/arm64/include/asm/kvm_emulate.h
>> @@ -31,6 +31,9 @@
>> unsigned long *vcpu_reg32(const struct kvm_vcpu *vcpu, u8 reg_num);
>> unsigned long *vcpu_spsr32(const struct kvm_vcpu *vcpu);
>>
>> +bool kvm_condition_valid32(const struct kvm_vcpu *vcpu);
>> +void kvm_skip_instr32(struct kvm_vcpu *vcpu, bool is_wide_instr);
>> +
>> void kvm_inject_undefined(struct kvm_vcpu *vcpu);
>> void kvm_inject_dabt(struct kvm_vcpu *vcpu, unsigned long addr);
>> void kvm_inject_pabt(struct kvm_vcpu *vcpu, unsigned long addr);
>> @@ -57,12 +60,18 @@ static inline bool vcpu_mode_is_32bit(const struct kvm_vcpu *vcpu)
>>
>> static inline bool kvm_condition_valid(const struct kvm_vcpu *vcpu)
>> {
>> - return true; /* No conditionals on arm64 */
>> + if (vcpu_mode_is_32bit(vcpu))
>> + return kvm_condition_valid32(vcpu);
>> +
>> + return true;
>> }
>>
>> static inline void kvm_skip_instr(struct kvm_vcpu *vcpu, bool is_wide_instr)
>> {
>> - *vcpu_pc(vcpu) += 4;
>> + if (vcpu_mode_is_32bit(vcpu))
>> + kvm_skip_instr32(vcpu, is_wide_instr);
>> + else
>> + *vcpu_pc(vcpu) += 4;
>> }
>>
>> static inline void vcpu_set_thumb(struct kvm_vcpu *vcpu)
>> diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile
>> index 1668448..88c6639 100644
>> --- a/arch/arm64/kvm/Makefile
>> +++ b/arch/arm64/kvm/Makefile
>> @@ -11,7 +11,7 @@ obj-$(CONFIG_KVM_ARM_HOST) += kvm.o
>> kvm-$(CONFIG_KVM_ARM_HOST) += $(addprefix ../../../virt/kvm/, kvm_main.o coalesced_mmio.o)
>> kvm-$(CONFIG_KVM_ARM_HOST) += $(addprefix ../../../arch/arm/kvm/, arm.o mmu.o mmio.o psci.o perf.o)
>>
>> -kvm-$(CONFIG_KVM_ARM_HOST) += inject_fault.o regmap.o
>> +kvm-$(CONFIG_KVM_ARM_HOST) += emulate.o inject_fault.o regmap.o
>> kvm-$(CONFIG_KVM_ARM_HOST) += hyp.o hyp-init.o handle_exit.o
>> kvm-$(CONFIG_KVM_ARM_HOST) += guest.o reset.o sys_regs.o sys_regs_generic_v8.o
>>
>> diff --git a/arch/arm64/kvm/emulate.c b/arch/arm64/kvm/emulate.c
>> new file mode 100644
>> index 0000000..01d4713
>> --- /dev/null
>> +++ b/arch/arm64/kvm/emulate.c
>> @@ -0,0 +1,154 @@
>> +/*
>> + * (not much of an) Emulation layer for 32bit guests.
>> + *
>> + * Copyright (C) 2012 - Virtual Open Systems and Columbia University
>> + * Author: Christoffer Dall <c.dall@virtualopensystems.com>
>
> don't you want to add yourself here?
There's really nothing I actually wrote here. Most of it is your code,
and the rest comes from other parts of arch/arm.
I guess I could always mention myself for the ESR_EL2_* macros... ;-)
>> + *
>> + * This program is free software: you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License version 2 as
>> + * published by the Free Software Foundation.
>> + *
>> + * This program is distributed in the hope that it will be useful,
>> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
>> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
>> + * GNU General Public License for more details.
>> + *
>> + * You should have received a copy of the GNU General Public License
>> + * along with this program. If not, see <http://www.gnu.org/licenses/>.
>> + */
>> +
>> +#include <linux/kvm_host.h>
>> +#include <asm/kvm_emulate.h>
>> +
>> +/*
>> + * stolen from arch/arm/kernel/opcodes.c
>> + *
>> + * condition code lookup table
>> + * index into the table is test code: EQ, NE, ... LT, GT, AL, NV
>> + *
>> + * bit position in short is condition code: NZCV
>> + */
>> +static const unsigned short cc_map[16] = {
>> + 0xF0F0, /* EQ == Z set */
>> + 0x0F0F, /* NE */
>> + 0xCCCC, /* CS == C set */
>> + 0x3333, /* CC */
>> + 0xFF00, /* MI == N set */
>> + 0x00FF, /* PL */
>> + 0xAAAA, /* VS == V set */
>> + 0x5555, /* VC */
>> + 0x0C0C, /* HI == C set && Z clear */
>> + 0xF3F3, /* LS == C clear || Z set */
>> + 0xAA55, /* GE == (N==V) */
>> + 0x55AA, /* LT == (N!=V) */
>> + 0x0A05, /* GT == (!Z && (N==V)) */
>> + 0xF5FA, /* LE == (Z || (N!=V)) */
>> + 0xFFFF, /* AL always */
>> + 0 /* NV */
>> +};
>> +
>> +static int kvm_vcpu_get_condition(const struct kvm_vcpu *vcpu)
>> +{
>> + u32 esr = kvm_vcpu_get_hsr(vcpu);
>> +
>> + if (esr & ESR_EL2_CV)
>> + return (esr & ESR_EL2_COND) >> ESR_EL2_COND_SHIFT;
>> +
>> + return -1;
>> +}
>> +
>> +/*
>> + * Check if a trapped instruction should have been executed or not.
>> + */
>> +bool kvm_condition_valid32(const struct kvm_vcpu *vcpu)
>> +{
>> + unsigned long cpsr;
>> + u32 cpsr_cond;
>> + int cond;
>> +
>> + /* Top two bits non-zero? Unconditional. */
>> + if (kvm_vcpu_get_hsr(vcpu) >> 30)
>> + return true;
>> +
>> + /* Is condition field valid? */
>> + cond = kvm_vcpu_get_condition(vcpu);
>> + if (cond == 0xE)
>> + return true;
>> +
>> + cpsr = *vcpu_cpsr(vcpu);
>> +
>> + if (cond < 0) {
>> + /* This can happen in Thumb mode: examine IT state. */
>> + unsigned long it;
>> +
>> + it = ((cpsr >> 8) & 0xFC) | ((cpsr >> 25) & 0x3);
>> +
>> + /* it == 0 => unconditional. */
>> + if (it == 0)
>> + return true;
>> +
>> + /* The cond for this insn works out as the top 4 bits. */
>> + cond = (it >> 4);
>> + }
>> +
>> + cpsr_cond = cpsr >> 28;
>> +
>> + if (!((cc_map[cond] >> cpsr_cond) & 1))
>> + return false;
>> +
>> + return true;
>> +}
>> +
>> +/**
>> + * adjust_itstate - adjust ITSTATE when emulating instructions in IT-block
>> + * @vcpu: The VCPU pointer
>> + *
>> + * When exceptions occur while instructions are executed in Thumb IF-THEN
>> + * blocks, the ITSTATE field of the CPSR is not advanved (updated), so we have
>
> s/advanved/advanced/
See? You actually hold the copyright on that spelling mistake! ;-)
>
>> + * to do this little bit of work manually. The fields map like this:
>> + *
>> + * IT[7:0] -> CPSR[26:25],CPSR[15:10]
>> + */
>> +static void kvm_adjust_itstate(struct kvm_vcpu *vcpu)
>> +{
>> + unsigned long itbits, cond;
>> + unsigned long cpsr = *vcpu_cpsr(vcpu);
>> + bool is_arm = !(cpsr & COMPAT_PSR_T_BIT);
>> +
>> + BUG_ON(is_arm && (cpsr & COMPAT_PSR_IT_MASK));
>> +
>> + if (!(cpsr & COMPAT_PSR_IT_MASK))
>> + return;
>> +
>> + cond = (cpsr & 0xe000) >> 13;
>> + itbits = (cpsr & 0x1c00) >> (10 - 2);
>> + itbits |= (cpsr & (0x3 << 25)) >> 25;
>> +
>> + /* Perform ITAdvance (see page A-52 in ARM DDI 0406C) */
>
> That's actually page A2-52 to be a little more precise...
Yup. Feel free to update the 32bit code as well... ;-)
>> + if ((itbits & 0x7) == 0)
>> + itbits = cond = 0;
>> + else
>> + itbits = (itbits << 1) & 0x1f;
>> +
>> + cpsr &= ~COMPAT_PSR_IT_MASK;
>> + cpsr |= cond << 13;
>> + cpsr |= (itbits & 0x1c) << (10 - 2);
>> + cpsr |= (itbits & 0x3) << 25;
>> + *vcpu_cpsr(vcpu) = cpsr;
>> +}
>> +
>> +/**
>> + * kvm_skip_instr - skip a trapped instruction and proceed to the next
>> + * @vcpu: The vcpu pointer
>> + */
>> +void kvm_skip_instr32(struct kvm_vcpu *vcpu, bool is_wide_instr)
>> +{
>> + bool is_thumb;
>> +
>> + is_thumb = !!(*vcpu_cpsr(vcpu) & COMPAT_PSR_T_BIT);
>
> again we don't need the '!!' when it's a bool.
I'm still not prepared to consider a bitwise and as a bool, and I'm
happy for the !! to make the conversion to a bool completely obvious.
Thanks,
M.
--
Jazz is not dead. It just smells funny...
next prev parent reply other threads:[~2013-04-24 13:13 UTC|newest]
Thread overview: 116+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-08 16:17 [PATCH v3 00/32] Port of KVM to arm64 Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 01/32] arm64: add explicit symbols to ESR_EL1 decoding Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 02/32] arm64: KVM: define HYP and Stage-2 translation page flags Marc Zyngier
2013-04-10 14:07 ` Will Deacon
2013-04-12 15:22 ` Marc Zyngier
2013-04-26 17:01 ` Catalin Marinas
2013-04-26 17:11 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 03/32] arm64: KVM: HYP mode idmap support Marc Zyngier
2013-04-23 22:57 ` Christoffer Dall
2013-04-24 9:36 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 04/32] arm64: KVM: EL2 register definitions Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 05/32] arm64: KVM: system register definitions for 64bit guests Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 06/32] arm64: KVM: Basic ESR_EL2 helpers and vcpu register access Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 07/32] arm64: KVM: fault injection into a guest Marc Zyngier
2013-04-10 16:40 ` Will Deacon
2013-04-12 15:29 ` Marc Zyngier
2013-04-23 22:57 ` Christoffer Dall
2013-04-24 10:04 ` Marc Zyngier
2013-04-24 16:46 ` Christoffer Dall
2013-04-29 16:26 ` Catalin Marinas
2013-05-07 16:29 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 08/32] arm64: KVM: architecture specific MMU backend Marc Zyngier
2013-04-23 22:58 ` Christoffer Dall
2013-04-24 11:03 ` Marc Zyngier
2013-04-24 11:10 ` Will Deacon
2013-04-24 16:50 ` Christoffer Dall
2013-04-24 16:55 ` Christoffer Dall
2013-04-25 12:59 ` Marc Zyngier
2013-04-25 15:13 ` Christoffer Dall
2013-04-29 17:35 ` Catalin Marinas
2013-04-30 10:23 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 09/32] arm64: KVM: user space interface Marc Zyngier
2013-04-10 16:45 ` Will Deacon
2013-04-12 15:31 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 10/32] arm64: KVM: system register handling Marc Zyngier
2013-04-10 17:04 ` Will Deacon
2013-04-12 15:48 ` Marc Zyngier
2013-04-23 23:01 ` Christoffer Dall
2013-04-24 13:37 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 11/32] arm64: KVM: CPU specific system registers handling Marc Zyngier
2013-04-10 17:06 ` Will Deacon
2013-04-12 16:04 ` Marc Zyngier
2013-04-23 22:59 ` Christoffer Dall
2013-04-24 9:33 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 12/32] arm64: KVM: virtual CPU reset Marc Zyngier
2013-04-10 17:07 ` Will Deacon
2013-04-12 16:04 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 13/32] arm64: KVM: kvm_arch and kvm_vcpu_arch definitions Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 14/32] arm64: KVM: MMIO access backend Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 15/32] arm64: KVM: guest one-reg interface Marc Zyngier
2013-04-10 17:13 ` Will Deacon
2013-04-12 16:35 ` Marc Zyngier
2013-04-23 22:59 ` Christoffer Dall
2013-04-24 11:27 ` Marc Zyngier
2013-04-24 17:05 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 16/32] arm64: KVM: hypervisor initialization code Marc Zyngier
2013-05-02 11:03 ` Catalin Marinas
2013-05-02 13:28 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 17/32] arm64: KVM: HYP mode world switch implementation Marc Zyngier
2013-04-23 22:59 ` Christoffer Dall
2013-04-24 11:39 ` Marc Zyngier
2013-04-24 17:08 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 18/32] arm64: KVM: Exit handling Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 19/32] arm64: KVM: Plug the VGIC Marc Zyngier
2013-04-23 23:00 ` Christoffer Dall
2013-04-24 11:43 ` Marc Zyngier
2013-05-02 14:38 ` Catalin Marinas
2013-04-08 16:17 ` [PATCH v3 20/32] arm64: KVM: Plug the arch timer Marc Zyngier
2013-04-23 23:00 ` Christoffer Dall
2013-04-24 11:43 ` Marc Zyngier
2013-05-02 15:31 ` Catalin Marinas
2013-04-08 16:17 ` [PATCH v3 21/32] arm64: KVM: PSCI implementation Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 22/32] arm64: KVM: Build system integration Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 23/32] arm64: KVM: define 32bit specific registers Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 24/32] arm64: KVM: 32bit GP register access Marc Zyngier
2013-04-23 23:00 ` Christoffer Dall
2013-04-24 13:06 ` Marc Zyngier
2013-04-24 17:09 ` Christoffer Dall
2013-05-02 16:09 ` Catalin Marinas
2013-05-07 16:28 ` Marc Zyngier
2013-05-07 16:33 ` Catalin Marinas
2013-05-11 0:36 ` Christoffer Dall
2013-05-11 7:51 ` Peter Maydell
2013-05-11 9:43 ` Catalin Marinas
2013-05-12 18:51 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 25/32] arm64: KVM: 32bit conditional execution emulation Marc Zyngier
2013-04-23 23:00 ` Christoffer Dall
2013-04-24 13:13 ` Marc Zyngier [this message]
2013-04-24 17:11 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 26/32] arm64: KVM: 32bit handling of coprocessor traps Marc Zyngier
2013-04-23 23:01 ` Christoffer Dall
2013-04-24 13:42 ` Marc Zyngier
2013-04-24 17:14 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 27/32] arm64: KVM: CPU specific 32bit coprocessor access Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 28/32] arm64: KVM: 32bit specific register world switch Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 29/32] arm64: KVM: 32bit guest fault injection Marc Zyngier
2013-04-23 23:02 ` Christoffer Dall
2013-04-24 13:46 ` Marc Zyngier
2013-04-24 17:15 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 30/32] arm64: KVM: enable initialization of a 32bit vcpu Marc Zyngier
2013-04-23 23:02 ` Christoffer Dall
2013-04-24 13:49 ` Marc Zyngier
2013-04-24 17:17 ` Christoffer Dall
2013-05-07 16:36 ` Marc Zyngier
2013-05-11 0:38 ` Christoffer Dall
2013-05-11 8:04 ` Peter Maydell
2013-05-11 16:26 ` Christoffer Dall
2013-05-11 16:31 ` Peter Maydell
2013-05-13 9:01 ` Marc Zyngier
2013-05-13 15:46 ` Christoffer Dall
2013-04-08 16:17 ` [PATCH v3 31/32] arm64: KVM: userspace API documentation Marc Zyngier
2013-04-23 23:02 ` Christoffer Dall
2013-04-24 13:52 ` Marc Zyngier
2013-04-08 16:17 ` [PATCH v3 32/32] arm64: KVM: MAINTAINERS update Marc Zyngier
2013-04-23 23:04 ` [PATCH v3 00/32] Port of KVM to arm64 Christoffer Dall
2013-05-03 13:17 ` Catalin Marinas
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=5177DA81.3020400@arm.com \
--to=marc.zyngier@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).