From: mst@redhat.com (Michael S. Tsirkin)
To: linux-arm-kernel@lists.infradead.org
Subject: [kvmarm] [PATCH 08/29] arm64: KVM: user space interface
Date: Sun, 10 Mar 2013 11:23:26 +0200 [thread overview]
Message-ID: <20130310092326.GA8163@redhat.com> (raw)
In-Reply-To: <2d088fa3db5159ac94cb6813a5e6c03b@localhost>
On Fri, Mar 08, 2013 at 04:46:15AM +0100, Marc Zyngier wrote:
> On Thu, 7 Mar 2013 10:09:03 +0200, "Michael S. Tsirkin" <mst@redhat.com>
> wrote:
> > On Tue, Mar 05, 2013 at 03:47:24AM +0000, Marc Zyngier wrote:
> >> Provide the kvm.h file that defines the user space visible
> >> interface.
> >>
> >> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
> >> ---
> >> arch/arm64/include/uapi/asm/kvm.h | 112
> >> ++++++++++++++++++++++++++++++++++++++
> >> 1 file changed, 112 insertions(+)
> >> create mode 100644 arch/arm64/include/uapi/asm/kvm.h
> >>
> >> diff --git a/arch/arm64/include/uapi/asm/kvm.h
> >> b/arch/arm64/include/uapi/asm/kvm.h
> >> new file mode 100644
> >> index 0000000..f5525f1
> >> --- /dev/null
> >> +++ b/arch/arm64/include/uapi/asm/kvm.h
> >> @@ -0,0 +1,112 @@
> >> +/*
> >> + * Copyright (C) 2012 - ARM Ltd
> >> + * Author: Marc Zyngier <marc.zyngier@arm.com>
> >> + *
> >> + * Derived from arch/arm/include/uapi/asm/kvm.h:
> >> + * Copyright (C) 2012 - Virtual Open Systems and Columbia University
> >> + * Author: Christoffer Dall <c.dall@virtualopensystems.com>
> >> + *
> >> + * 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/>.
> >> + */
> >> +
> >> +#ifndef __ARM_KVM_H__
> >> +#define __ARM_KVM_H__
> >> +
> >> +#define KVM_SPSR_EL1 0
> >> +#define KVM_NR_SPSR 1
> >> +
> >> +#ifndef __ASSEMBLY__
> >> +#include <asm/types.h>
> >> +#include <asm/ptrace.h>
> >> +
> >> +#define __KVM_HAVE_GUEST_DEBUG
> >> +#define __KVM_HAVE_IRQ_LINE
> >> +
> >> +#define KVM_REG_SIZE(id) \
> >> + (1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
> >> +
> >> +struct kvm_regs {
> >> + struct user_pt_regs regs; /* sp = sp_el0 */
> >> +
> >> + unsigned long sp_el1;
> >> + unsigned long elr_el1;
> >> +
> >> + unsigned long spsr[KVM_NR_SPSR];
> >> +};
> >> +
> >
> > Using long in uapi is generally a mistake: with gcc it has
> > different size depending on whether you build in 64 and 32 bit mode.
> > I think it is better to make it __u64.
>
> Ah, well spotted. Definitely a mistake.
>
> >> +/* Supported Processor Types */
> >> +#define KVM_ARM_TARGET_CORTEX_A57 0
> >> +#define KVM_ARM_NUM_TARGETS 1
> >> +
> >> +/* KVM_ARM_SET_DEVICE_ADDR ioctl id encoding */
> >> +#define KVM_ARM_DEVICE_TYPE_SHIFT 0
> >> +#define KVM_ARM_DEVICE_TYPE_MASK (0xffff << KVM_ARM_DEVICE_TYPE_SHIFT)
> >> +#define KVM_ARM_DEVICE_ID_SHIFT 16
> >> +#define KVM_ARM_DEVICE_ID_MASK (0xffff << KVM_ARM_DEVICE_ID_SHIFT)
> >> +
> >> +/* Supported device IDs */
> >> +#define KVM_ARM_DEVICE_VGIC_V2 0
> >> +
> >> +/* Supported VGIC address types */
> >> +#define KVM_VGIC_V2_ADDR_TYPE_DIST 0
> >> +#define KVM_VGIC_V2_ADDR_TYPE_CPU 1
> >> +
> >> +#define KVM_VGIC_V2_DIST_SIZE 0x1000
> >> +#define KVM_VGIC_V2_CPU_SIZE 0x2000
> >> +
> >> +struct kvm_vcpu_init {
> >> + __u32 target;
> >> + __u32 features[7];
> >> +};
> >> +
> >> +struct kvm_sregs {
> >> +};
> >> +
> >> +struct kvm_fpu {
> >> +};
> >> +
> >> +struct kvm_guest_debug_arch {
> >> +};
> >> +
> >> +struct kvm_debug_exit_arch {
> >> +};
> >> +
> >
> > This is a problem too: structure alignment is different
> > in -m32 versus -m64 modes, which will affect the offset of
> > the following fields. I think it's best to add a "padding"
> > field in there, and size it to multiple of 8 bytes.
>
> Sorry, which field are you referring to?
Sorry, I was wrong here: while it seems to be compler dependent gcc
returns 0 as size of an empty structure.
> > I think the same is a good idea for other empty stuctures,
> > since otherwise the padding is implicit and
> > not initialized by gcc, and it is hard not too leak
> > info to userspace when you copy these structures out.
> >
> > And, it'll be handy if you want to extend the structures
> > down the line.
>
> Sure, what would make some sense.
>
> Thanks,
>
> M.
> --
> Fast, cheap, reliable. Pick two.
next prev parent reply other threads:[~2013-03-10 9:23 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-05 3:47 [PATCH 00/29] Port of KVM to arm64 Marc Zyngier
2013-03-05 3:47 ` [PATCH 01/29] arm64: KVM: define HYP and Stage-2 translation page flags Marc Zyngier
2013-03-05 3:47 ` [PATCH 02/29] arm64: KVM: HYP mode idmap support Marc Zyngier
2013-03-05 3:47 ` [PATCH 03/29] arm64: KVM: EL2 register definitions Marc Zyngier
2013-03-05 3:47 ` [PATCH 04/29] arm64: KVM: system register definitions for 64bit guests Marc Zyngier
2013-03-07 10:33 ` [kvmarm] " Alexander Graf
2013-03-08 3:23 ` Marc Zyngier
2013-03-12 13:20 ` Christopher Covington
2013-03-12 13:41 ` Christopher Covington
2013-03-12 13:50 ` Marc Zyngier
2013-03-05 3:47 ` [PATCH 05/29] arm64: KVM: Basic ESR_EL2 helpers and vcpu register access Marc Zyngier
2013-03-16 0:55 ` Geoff Levand
2013-03-05 3:47 ` [PATCH 06/29] arm64: KVM: fault injection into a guest Marc Zyngier
2013-03-12 13:20 ` Christopher Covington
2013-03-12 14:25 ` Marc Zyngier
2013-03-16 1:03 ` Geoff Levand
2013-03-05 3:47 ` [PATCH 07/29] arm64: KVM: architecture specific MMU backend Marc Zyngier
2013-03-05 3:47 ` [PATCH 08/29] arm64: KVM: user space interface Marc Zyngier
2013-03-07 8:09 ` Michael S. Tsirkin
2013-03-08 3:46 ` [kvmarm] " Marc Zyngier
2013-03-10 9:23 ` Michael S. Tsirkin [this message]
2013-03-05 3:47 ` [PATCH 09/29] arm64: KVM: system register handling Marc Zyngier
2013-03-07 10:30 ` [kvmarm] " Alexander Graf
2013-03-08 3:29 ` Marc Zyngier
2013-03-25 8:19 ` Marc Zyngier
2013-04-23 23:07 ` Christoffer Dall
2013-03-05 3:47 ` [PATCH 10/29] arm64: KVM: Cortex-A57 specific system registers handling Marc Zyngier
2013-03-13 18:30 ` Christopher Covington
2013-03-14 10:26 ` Marc Zyngier
2013-03-05 3:47 ` [PATCH 11/29] arm64: KVM: virtual CPU reset Marc Zyngier
2013-03-05 3:47 ` [PATCH 12/29] arm64: KVM: kvm_arch and kvm_vcpu_arch definitions Marc Zyngier
2013-03-12 17:30 ` Christopher Covington
2013-03-05 3:47 ` [PATCH 13/29] arm64: KVM: MMIO access backend Marc Zyngier
2013-03-05 3:47 ` [PATCH 14/29] arm64: KVM: guest one-reg interface Marc Zyngier
2013-03-12 17:31 ` Christopher Covington
2013-03-12 18:05 ` Marc Zyngier
2013-03-12 22:07 ` Christopher Covington
2013-03-13 7:48 ` Marc Zyngier
2013-03-13 20:34 ` Christopher Covington
2013-03-14 8:57 ` [kvmarm] " Peter Maydell
2013-03-20 20:06 ` Christopher Covington
2013-03-05 3:47 ` [PATCH 15/29] arm64: KVM: hypervisor initialization code Marc Zyngier
2013-03-05 3:47 ` [PATCH 16/29] arm64: KVM: HYP mode world switch implementation Marc Zyngier
2013-03-13 19:59 ` Christopher Covington
2013-03-20 20:04 ` Christopher Covington
2013-03-21 11:54 ` Marc Zyngier
2013-03-05 3:47 ` [PATCH 17/29] arm64: KVM: Exit handling Marc Zyngier
2013-03-05 3:47 ` [PATCH 18/29] arm64: KVM: Plug the VGIC Marc Zyngier
2013-03-05 3:47 ` [PATCH 19/29] arm64: KVM: Plug the arch timer Marc Zyngier
2013-03-05 3:47 ` [PATCH 20/29] arm64: KVM: PSCI implementation Marc Zyngier
2013-03-05 3:47 ` [PATCH 21/29] arm64: KVM: Build system integration Marc Zyngier
2013-03-05 3:47 ` [PATCH 22/29] arm64: KVM: define 32bit specific registers Marc Zyngier
2013-03-18 17:03 ` Christopher Covington
2013-03-05 3:47 ` [PATCH 23/29] arm64: KVM: 32bit GP register access Marc Zyngier
2013-03-16 0:24 ` Geoff Levand
2013-03-05 3:47 ` [PATCH 24/29] arm64: KVM: 32bit conditional execution emulation Marc Zyngier
2013-03-18 17:04 ` Christopher Covington
2013-03-05 3:47 ` [PATCH 25/29] arm64: KVM: 32bit handling of coprocessor traps Marc Zyngier
2013-03-05 3:47 ` [PATCH 26/29] arm64: KVM: 32bit coprocessor access for Cortex-A57 Marc Zyngier
2013-03-05 3:47 ` [PATCH 27/29] arm64: KVM: 32bit specific register world switch Marc Zyngier
2013-03-05 3:47 ` [PATCH 28/29] arm64: KVM: 32bit guest fault injection Marc Zyngier
2013-03-18 18:45 ` Christopher Covington
2013-03-05 3:47 ` [PATCH 29/29] arm64: KVM: enable initialization of a 32bit vcpu Marc Zyngier
2013-03-18 18:56 ` Christopher Covington
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=20130310092326.GA8163@redhat.com \
--to=mst@redhat.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).