qemu-arm.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Edgar E. Iglesias" <edgar.iglesias@gmail.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-devel@nongnu.org, patches@linaro.org, qemu-arm@nongnu.org,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Alex Bennée" <alex.bennee@linaro.org>
Subject: Re: [PATCH 6/8] target-arm: Handle exception return from AArch64 to non-EL0 AArch32
Date: Tue, 19 Jan 2016 17:47:52 +0100	[thread overview]
Message-ID: <20160119164752.GP29396@toto> (raw)
In-Reply-To: <1452796451-2946-7-git-send-email-peter.maydell@linaro.org>

On Thu, Jan 14, 2016 at 06:34:09PM +0000, Peter Maydell wrote:
> Remove the assumptions that the AArch64 exception return code was
> making about a return to AArch32 always being a return to EL0.
> This includes pulling out the illegal-SPSR checks so we can apply
> them for return to 32 bit as well as return to 64-bit.
> 
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>


> ---
>  target-arm/op_helper.c | 80 +++++++++++++++++++++++++++++++++++++-------------
>  1 file changed, 59 insertions(+), 21 deletions(-)
> 
> diff --git a/target-arm/op_helper.c b/target-arm/op_helper.c
> index e42d287..38d46d8 100644
> --- a/target-arm/op_helper.c
> +++ b/target-arm/op_helper.c
> @@ -640,12 +640,51 @@ void HELPER(pre_smc)(CPUARMState *env, uint32_t syndrome)
>      }
>  }
>  
> +static int el_from_spsr(uint32_t spsr)
> +{
> +    /* Return the exception level that this SPSR is requesting a return to,
> +     * or -1 if it is invalid (an illegal return)
> +     */
> +    if (spsr & PSTATE_nRW) {
> +        switch (spsr & CPSR_M) {
> +        case ARM_CPU_MODE_USR:
> +            return 0;
> +        case ARM_CPU_MODE_HYP:
> +            return 2;
> +        case ARM_CPU_MODE_FIQ:
> +        case ARM_CPU_MODE_IRQ:
> +        case ARM_CPU_MODE_SVC:
> +        case ARM_CPU_MODE_ABT:
> +        case ARM_CPU_MODE_UND:
> +        case ARM_CPU_MODE_SYS:
> +            return 1;
> +        case ARM_CPU_MODE_MON:
> +            /* Returning to Mon from AArch64 is never possible,
> +             * so this is an illegal return.
> +             */
> +        default:
> +            return -1;
> +        }
> +    } else {
> +        if (extract32(spsr, 1, 1)) {
> +            /* Return with reserved M[1] bit set */
> +            return -1;
> +        }
> +        if (extract32(spsr, 0, 4) == 1) {
> +            /* return to EL0 with M[0] bit set */
> +            return -1;
> +        }
> +        return extract32(spsr, 2, 2);
> +    }
> +}
> +
>  void HELPER(exception_return)(CPUARMState *env)
>  {
>      int cur_el = arm_current_el(env);
>      unsigned int spsr_idx = aarch64_banked_spsr_index(cur_el);
>      uint32_t spsr = env->banked_spsr[spsr_idx];
>      int new_el;
> +    bool return_to_aa64 = (spsr & PSTATE_nRW) == 0;
>  
>      aarch64_save_sp(env, cur_el);
>  
> @@ -662,35 +701,34 @@ void HELPER(exception_return)(CPUARMState *env)
>          spsr &= ~PSTATE_SS;
>      }
>  
> -    if (spsr & PSTATE_nRW) {
> -        /* TODO: We currently assume EL1/2/3 are running in AArch64.  */
> +    new_el = el_from_spsr(spsr);
> +    if (new_el == -1) {
> +        goto illegal_return;
> +    }
> +    if (new_el > cur_el
> +        || (new_el == 2 && !arm_feature(env, ARM_FEATURE_EL2))) {
> +        /* Disallow return to an EL which is unimplemented or higher
> +         * than the current one.
> +         */
> +        goto illegal_return;
> +    }
> +
> +    if (new_el != 0 && arm_el_is_aa64(env, new_el) != return_to_aa64) {
> +        /* Return to an EL which is configured for a different register width */
> +        goto illegal_return;
> +    }
> +
> +    if (!return_to_aa64) {
>          env->aarch64 = 0;
> -        new_el = 0;
> -        env->uncached_cpsr = 0x10;
> +        env->uncached_cpsr = spsr & CPSR_M;
>          cpsr_write(env, spsr, ~0);
>          if (!arm_singlestep_active(env)) {
>              env->uncached_cpsr &= ~PSTATE_SS;
>          }
>          aarch64_sync_64_to_32(env);
>  
> -        env->regs[15] = env->elr_el[1] & ~0x1;
> +        env->regs[15] = env->elr_el[cur_el] & ~0x1;
>      } else {
> -        new_el = extract32(spsr, 2, 2);
> -        if (new_el > cur_el
> -            || (new_el == 2 && !arm_feature(env, ARM_FEATURE_EL2))) {
> -            /* Disallow return to an EL which is unimplemented or higher
> -             * than the current one.
> -             */
> -            goto illegal_return;
> -        }
> -        if (extract32(spsr, 1, 1)) {
> -            /* Return with reserved M[1] bit set */
> -            goto illegal_return;
> -        }
> -        if (new_el == 0 && (spsr & PSTATE_SP)) {
> -            /* Return to EL0 with M[0] bit set */
> -            goto illegal_return;
> -        }
>          env->aarch64 = 1;
>          pstate_write(env, spsr);
>          if (!arm_singlestep_active(env)) {
> -- 
> 1.9.1
> 

  reply	other threads:[~2016-01-19 16:47 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-14 18:34 [PATCH 0/8] target-arm: support mixed 32/64 bit execution beyond EL0 Peter Maydell
2016-01-14 18:34 ` [PATCH 1/8] target-arm: Properly support EL2 and EL3 in arm_el_is_aa64() Peter Maydell
2016-01-15 14:38   ` Edgar E. Iglesias
2016-01-15 14:50     ` Peter Maydell
2016-01-15 15:37       ` Edgar E. Iglesias
2016-01-15 15:47         ` Peter Maydell
2016-01-15 20:37           ` Edgar E. Iglesias
2016-01-29 16:45   ` [Qemu-devel] " Sergey Fedorov
2016-01-29 16:50     ` Sergey Fedorov
2016-01-29 17:05     ` Peter Maydell
2016-01-29 17:08       ` Sergey Fedorov
2016-01-14 18:34 ` [PATCH 2/8] target-arm: Move aarch64_cpu_do_interrupt() to helper.c Peter Maydell
2016-01-15 14:39   ` Edgar E. Iglesias
2016-01-29 16:46   ` [Qemu-devel] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 3/8] target-arm: Use a single entry point for AArch64 and AArch32 exceptions Peter Maydell
2016-01-15 14:54   ` Edgar E. Iglesias
2016-01-29 16:46   ` [Qemu-arm] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 4/8] target-arm: Pull semihosting handling out to arm_cpu_do_interrupt() Peter Maydell
2016-01-29 16:46   ` [Qemu-devel] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 5/8] target-arm: Fix wrong AArch64 entry offset for EL2/EL3 target Peter Maydell
2016-01-19 16:40   ` Edgar E. Iglesias
2016-01-29 16:47   ` [Qemu-devel] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 6/8] target-arm: Handle exception return from AArch64 to non-EL0 AArch32 Peter Maydell
2016-01-19 16:47   ` Edgar E. Iglesias [this message]
2016-01-29 16:47   ` [Qemu-arm] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 7/8] target-arm: Implement remaining illegal return event checks Peter Maydell
2016-01-19 16:53   ` Edgar E. Iglesias
2016-01-19 16:58     ` Peter Maydell
2016-01-29 16:47   ` [Qemu-devel] " Sergey Fedorov
2016-01-14 18:34 ` [PATCH 8/8] target-arm: ignore ELR_ELx[1] for exception return to 32-bit ARM mode Peter Maydell
2016-01-19 16:56   ` Edgar E. Iglesias
2016-01-29 16:48   ` [Qemu-devel] [Qemu-arm] " Sergey Fedorov

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=20160119164752.GP29396@toto \
    --to=edgar.iglesias@gmail.com \
    --cc=alex.bennee@linaro.org \
    --cc=patches@linaro.org \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.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).