From: geoff@infradead.org (Geoff Levand)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 2/7] arm64: Convert hcalls to use ISS field
Date: Fri, 03 Oct 2014 14:56:27 -0700 [thread overview]
Message-ID: <1412373387.32313.41.camel@smoke> (raw)
In-Reply-To: <20141003105106.GB26643@leverpostej>
Hi Mark,
On Fri, 2014-10-03 at 11:51 +0100, Mark Rutland wrote:
> On Thu, Sep 25, 2014 at 01:23:26AM +0100, Geoff Levand wrote:
> index 7a5df52..99c319c 100644
> > --- a/arch/arm64/include/asm/virt.h
> > +++ b/arch/arm64/include/asm/virt.h
> > @@ -21,6 +21,26 @@
> > #define BOOT_CPU_MODE_EL1 (0xe11)
> > #define BOOT_CPU_MODE_EL2 (0xe12)
> >
> > +/*
> > + * HVC_GET_VECTORS - Return the value of the vbar_el2 register.
> > + */
> > +
> > +#define HVC_GET_VECTORS 1
> > +
> > +/*
> > + * HVC_SET_VECTORS - Set the value of the vbar_el2 register.
> > + *
> > + * @x0: Physical address of the new vector table.
> > + */
> > +
> > +#define HVC_SET_VECTORS 2
> > +
> > +/*
> > + * HVC_CALL_HYP - Execute a hyp routine.
> > + */
> > +
> > +#define HVC_CALL_HYP 3
> > +
> > #ifndef __ASSEMBLY__
> >
> > /*
> > diff --git a/arch/arm64/kernel/hyp-stub.S b/arch/arm64/kernel/hyp-stub.S
> > index a272f33..8b39522 100644
> > --- a/arch/arm64/kernel/hyp-stub.S
> > +++ b/arch/arm64/kernel/hyp-stub.S
> > @@ -53,14 +53,22 @@ ENDPROC(__hyp_stub_vectors)
> > .align 11
> >
> > el1_sync:
> > - mrs x1, esr_el2
> > - lsr x1, x1, #26
> > - cmp x1, #0x16
> > - b.ne 2f // Not an HVC trap
> > - cbz x0, 1f
> > - msr vbar_el2, x0 // Set vbar_el2
> > + mrs x18, esr_el2
> > + lsr x17, x18, #26 // x17=EC
>
> The first patch enabled kvm_arm.h macros to be used from asm. Can't we
> use them here? If we don't include kvm_arm.h in the hyp stub, maybe we
> should (ideally we'd factor common stuff out, but I can see that getting
> messy fast).
I didn't want to tackle the cleanup of the headers in this patch
series, and I thought it odd to include a kvm header in this core
arm64 file, but maybe it is better to include kvm_arm.h and use the
ESR_EL2_ macros so the values are at least easier to understand and
keep track of.
> Assuming we can use them:
>
> lsr x17, x18, #ESR_EL2_EC_SHIFT
>
> > + and x18, x18, #0xffffff // x18=ISS
>
> and x18, x18, ESR_EL2_ISS
>
> > +
> > + cmp x17, #0x16
>
> cmp x17, #ESR_EL2_EC_HVC64
>
> > + b.ne 2f // Not an HVC trap
> > +
> > + cmp x18, #HVC_GET_VECTORS
> > + b.ne 1f
> > + mrs x0, vbar_el2
> > b 2f
> > -1: mrs x0, vbar_el2 // Return vbar_el2
> > +
> > +1: cmp x18, #HVC_SET_VECTORS
> > + b.ne 2f
> > + msr vbar_el2, x0
> > +
> > 2: eret
> > ENDPROC(el1_sync)
> >
> > @@ -100,11 +108,12 @@ ENDPROC(\label)
> > * initialisation entry point.
> > */
> >
> > -ENTRY(__hyp_get_vectors)
> > - mov x0, xzr
> > - // fall through
> > ENTRY(__hyp_set_vectors)
> > - hvc #0
> > + hvc #HVC_SET_VECTORS
> > ret
> > -ENDPROC(__hyp_get_vectors)
> > ENDPROC(__hyp_set_vectors)
> > +
> > +ENTRY(__hyp_get_vectors)
> > + hvc #HVC_GET_VECTORS
> > + ret
> > +ENDPROC(__hyp_get_vectors)
> > diff --git a/arch/arm64/kvm/hyp.S b/arch/arm64/kvm/hyp.S
> > index b72aa9f..9607f15 100644
> > --- a/arch/arm64/kvm/hyp.S
> > +++ b/arch/arm64/kvm/hyp.S
> > @@ -26,6 +26,7 @@
> > #include <asm/kvm_asm.h>
> > #include <asm/kvm_arm.h>
> > #include <asm/kvm_mmu.h>
> > +#include <asm/virt.h>
>
> As far as I can see, virt.h only defines BOOT_CPU_MODE_EL{1,2}, which we
> don't use below. So I don't think we need this include.
We need asm/virt.h for the new HVC_ macros added to it by this
patch (above)...
> Otherwise this looks fine to me.
>
> Mark.
>
> >
> > #define CPU_GP_REG_OFFSET(x) (CPU_GP_REGS + x)
> > #define CPU_XREG_OFFSET(x) CPU_GP_REG_OFFSET(CPU_USER_PT_REGS + 8*x)
> > @@ -1105,12 +1106,9 @@ __hyp_panic_str:
> > * in Hyp mode (see init_hyp_mode in arch/arm/kvm/arm.c). Return values are
> > * passed in r0 and r1.
> > *
> > - * A function pointer with a value of 0 has a special meaning, and is
> > - * used to implement __hyp_get_vectors in the same way as in
> > - * arch/arm64/kernel/hyp_stub.S.
> > */
> > ENTRY(kvm_call_hyp)
> > - hvc #0
> > + hvc #HVC_CALL_HYP
> > ret
> > ENDPROC(kvm_call_hyp)
next prev parent reply other threads:[~2014-10-03 21:56 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-25 0:23 [PATCH 0/7] arm64 kexec kernel patches V3 Geoff Levand
2014-09-25 0:23 ` [PATCH 3/7] arm64: Add new hcall HVC_CALL_FUNC Geoff Levand
2014-09-25 0:23 ` [PATCH 4/7] arm64: Add EL2 switch to soft_restart Geoff Levand
2014-09-25 0:23 ` [PATCH 2/7] arm64: Convert hcalls to use ISS field Geoff Levand
2014-10-03 10:51 ` Mark Rutland
2014-10-03 21:56 ` Geoff Levand [this message]
2014-10-06 10:13 ` Mark Rutland
2014-09-25 0:23 ` [PATCH 1/7] arm64/kvm: Fix assembler compatibility of macros Geoff Levand
2014-10-03 10:26 ` Mark Rutland
2014-10-03 22:27 ` Geoff Levand
2014-10-06 10:10 ` Mark Rutland
2014-09-25 0:23 ` [PATCH 7/7] arm64/kexec: Enable kexec in the arm64 defconfig Geoff Levand
2014-09-25 0:23 ` [PATCH 5/7] arm64: Move proc-macros.S to include/asm Geoff Levand
2014-09-25 0:23 ` [PATCH 6/7] arm64/kexec: Add core kexec support Geoff Levand
2014-09-25 18:28 ` Vivek Goyal
2014-09-25 19:02 ` Geoff Levand
2014-09-25 19:08 ` Vivek Goyal
2014-09-30 18:18 ` Vivek Goyal
2014-09-30 19:54 ` Geoff Levand
2014-10-01 14:56 ` Vivek Goyal
2014-10-03 18:35 ` Geoff Levand
2014-10-07 13:44 ` Vivek Goyal
2014-10-07 18:42 ` Geoff Levand
2014-10-07 18:45 ` Vivek Goyal
2014-10-07 20:12 ` Geoff Levand
2014-10-07 20:22 ` Vivek Goyal
2014-10-09 22:26 ` Geoff Levand
2014-10-23 23:08 ` Geoff Levand
2014-10-08 9:28 ` Mark Rutland
2014-10-07 18:48 ` Vivek Goyal
2014-10-01 16:16 ` Mark Rutland
2014-10-01 17:36 ` Vivek Goyal
2014-10-01 17:56 ` Mark Rutland
2014-10-01 17:47 ` Vivek Goyal
2014-10-01 18:03 ` Mark Rutland
2014-10-01 18:09 ` Vivek Goyal
2014-10-01 18:19 ` Mark Rutland
2014-10-01 18:31 ` Vivek Goyal
2014-10-01 19:22 ` Vivek Goyal
2014-10-02 10:26 ` Mark Rutland
2014-10-02 13:54 ` Vivek Goyal
2014-10-02 16:53 ` Mark Rutland
2014-09-30 23:59 ` [PATCH V2 " Geoff Levand
2014-09-30 20:29 ` [PATCH 0/7] arm64 kexec kernel patches V3 Vivek Goyal
2014-09-30 21:27 ` Geoff Levand
2014-10-02 19:08 ` Vivek Goyal
2014-10-02 22:59 ` Geoff Levand
2014-10-07 13:43 ` Vivek Goyal
2014-10-07 14:06 ` Mark Rutland
2014-10-01 15:19 ` Vivek Goyal
2014-10-03 21:16 ` Geoff Levand
2014-10-07 13:40 ` Vivek Goyal
2014-10-07 18:29 ` Geoff Levand
2014-10-08 9:42 ` Mark Rutland
2014-10-09 3:21 ` Dave Young
2014-10-09 10:09 ` Mark Rutland
2014-10-09 10:19 ` Ard Biesheuvel
2014-10-10 5:30 ` Dave Young
2014-10-07 15:22 ` Mark Rutland
2014-10-07 18:28 ` Geoff Levand
2014-10-07 18:09 ` Vivek Goyal
2014-10-07 20:07 ` Geoff Levand
2014-10-08 9:52 ` Leif Lindholm
2014-10-08 10:07 ` Mark Rutland
2014-10-09 9:22 ` Will Deacon
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=1412373387.32313.41.camel@smoke \
--to=geoff@infradead.org \
--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).