All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 01/27] target-arm: Fix "no 64-bit EL2" assumption in arm_excp_unmasked()
Date: Tue, 27 Oct 2015 14:33:03 +0000	[thread overview]
Message-ID: <1445956409-1818-2-git-send-email-peter.maydell@linaro.org> (raw)
In-Reply-To: <1445956409-1818-1-git-send-email-peter.maydell@linaro.org>

The code in arm_excp_unmasked() suppresses the ability of PSTATE.AIF
to mask exceptions from a lower EL targeting EL2 or EL3 if the
CPU is 64-bit. This is correct for a target of EL3, but not correct
for targeting EL2. Further, we go to some effort to calculate
scr and hcr values which are not used at all for the 64-bit CPU
case.

Rearrange the code to correctly implement the 64-bit CPU logic
and keep the hcr/scr calculations in the 32-bit CPU codepath.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1444327729-4120-1-git-send-email-peter.maydell@linaro.org
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Tested-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
---
 target-arm/cpu.h | 82 +++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 52 insertions(+), 30 deletions(-)

diff --git a/target-arm/cpu.h b/target-arm/cpu.h
index 3daa7f5..a271205 100644
--- a/target-arm/cpu.h
+++ b/target-arm/cpu.h
@@ -1525,8 +1525,6 @@ static inline bool arm_excp_unmasked(CPUState *cs, unsigned int excp_idx,
     CPUARMState *env = cs->env_ptr;
     unsigned int cur_el = arm_current_el(env);
     bool secure = arm_is_secure(env);
-    bool scr;
-    bool hcr;
     bool pstate_unmasked;
     int8_t unmasked = 0;
 
@@ -1540,31 +1538,10 @@ static inline bool arm_excp_unmasked(CPUState *cs, unsigned int excp_idx,
 
     switch (excp_idx) {
     case EXCP_FIQ:
-        /* If FIQs are routed to EL3 or EL2 then there are cases where we
-         * override the CPSR.F in determining if the exception is masked or
-         * not.  If neither of these are set then we fall back to the CPSR.F
-         * setting otherwise we further assess the state below.
-         */
-        hcr = (env->cp15.hcr_el2 & HCR_FMO);
-        scr = (env->cp15.scr_el3 & SCR_FIQ);
-
-        /* When EL3 is 32-bit, the SCR.FW bit controls whether the CPSR.F bit
-         * masks FIQ interrupts when taken in non-secure state.  If SCR.FW is
-         * set then FIQs can be masked by CPSR.F when non-secure but only
-         * when FIQs are only routed to EL3.
-         */
-        scr = scr && !((env->cp15.scr_el3 & SCR_FW) && !hcr);
         pstate_unmasked = !(env->daif & PSTATE_F);
         break;
 
     case EXCP_IRQ:
-        /* When EL3 execution state is 32-bit, if HCR.IMO is set then we may
-         * override the CPSR.I masking when in non-secure state.  The SCR.IRQ
-         * setting has already been taken into consideration when setting the
-         * target EL, so it does not have a further affect here.
-         */
-        hcr = (env->cp15.hcr_el2 & HCR_IMO);
-        scr = false;
         pstate_unmasked = !(env->daif & PSTATE_I);
         break;
 
@@ -1589,13 +1566,58 @@ static inline bool arm_excp_unmasked(CPUState *cs, unsigned int excp_idx,
      * interrupt.
      */
     if ((target_el > cur_el) && (target_el != 1)) {
-        /* ARM_FEATURE_AARCH64 enabled means the highest EL is AArch64.
-         * This code currently assumes that EL2 is not implemented
-         * (and so that highest EL will be 3 and the target_el also 3).
-         */
-        if (arm_feature(env, ARM_FEATURE_AARCH64) ||
-            ((scr || hcr) && (!secure))) {
-            unmasked = 1;
+        /* Exceptions targeting a higher EL may not be maskable */
+        if (arm_feature(env, ARM_FEATURE_AARCH64)) {
+            /* 64-bit masking rules are simple: exceptions to EL3
+             * can't be masked, and exceptions to EL2 can only be
+             * masked from Secure state. The HCR and SCR settings
+             * don't affect the masking logic, only the interrupt routing.
+             */
+            if (target_el == 3 || !secure) {
+                unmasked = 1;
+            }
+        } else {
+            /* The old 32-bit-only environment has a more complicated
+             * masking setup. HCR and SCR bits not only affect interrupt
+             * routing but also change the behaviour of masking.
+             */
+            bool hcr, scr;
+
+            switch (excp_idx) {
+            case EXCP_FIQ:
+                /* If FIQs are routed to EL3 or EL2 then there are cases where
+                 * we override the CPSR.F in determining if the exception is
+                 * masked or not. If neither of these are set then we fall back
+                 * to the CPSR.F setting otherwise we further assess the state
+                 * below.
+                 */
+                hcr = (env->cp15.hcr_el2 & HCR_FMO);
+                scr = (env->cp15.scr_el3 & SCR_FIQ);
+
+                /* When EL3 is 32-bit, the SCR.FW bit controls whether the
+                 * CPSR.F bit masks FIQ interrupts when taken in non-secure
+                 * state. If SCR.FW is set then FIQs can be masked by CPSR.F
+                 * when non-secure but only when FIQs are only routed to EL3.
+                 */
+                scr = scr && !((env->cp15.scr_el3 & SCR_FW) && !hcr);
+                break;
+            case EXCP_IRQ:
+                /* When EL3 execution state is 32-bit, if HCR.IMO is set then
+                 * we may override the CPSR.I masking when in non-secure state.
+                 * The SCR.IRQ setting has already been taken into consideration
+                 * when setting the target EL, so it does not have a further
+                 * affect here.
+                 */
+                hcr = (env->cp15.hcr_el2 & HCR_IMO);
+                scr = false;
+                break;
+            default:
+                g_assert_not_reached();
+            }
+
+            if ((scr || hcr) && !secure) {
+                unmasked = 1;
+            }
         }
     }
 
-- 
1.9.1

  reply	other threads:[~2015-10-27 14:33 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-27 14:33 [Qemu-devel] [PULL 00/27] target-arm queue Peter Maydell
2015-10-27 14:33 ` Peter Maydell [this message]
2015-10-27 14:33 ` [Qemu-devel] [PULL 02/27] target-arm/translate.c: Handle non-executable page-straddling Thumb insns Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 03/27] target-arm: Add support for SPSR_(ABT|UND|IRQ|FIQ) Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 04/27] arm_gic_kvm: Disable live migration if not supported Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 05/27] hw/arm/virt: don't use a15memmap directly Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 06/27] i.MX: Standardize i.MX serial debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 07/27] i.MX: Standardize i.MX GPIO debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 08/27] i.MX: Standardize i.MX I2C debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 09/27] i.MX: Standardize i.MX AVIC debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 10/27] i.MX: Standardize i.MX CCM debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 11/27] i.MX: Standardize i.MX FEC debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 12/27] i.MX: Standardize i.MX EPIT debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 13/27] i.MX: Standardize i.MX GPT debug Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 14/27] target-arm: Add HPFAR_EL2 Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 15/27] target-arm: lpae: Make t0sz and t1sz signed integers Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 16/27] target-arm: lpae: Move declaration of t0sz and t1sz Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 17/27] target-arm: Add support for AArch32 S2 negative t0sz Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 18/27] target-arm: lpae: Replace tsz with computed inputsize Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 19/27] target-arm: lpae: Rename granule_sz to stride Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 20/27] target-arm: Add computation of starting level for S2 PTW Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 21/27] target-arm: Add support for S2 page-table protection bits Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 22/27] target-arm: Avoid inline for get_phys_addr Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 23/27] target-arm: Add ARMMMUFaultInfo Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 24/27] target-arm: Add S2 translation to 64bit S1 PTWs Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 25/27] target-arm: Add S2 translation to 32bit " Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 26/27] target-arm: Route S2 MMU faults to EL2 Peter Maydell
2015-10-27 14:33 ` [Qemu-devel] [PULL 27/27] target-arm: Add support for S1 + S2 MMU translations Peter Maydell
2015-10-27 15:57 ` [Qemu-devel] [PULL 00/27] target-arm queue Peter Maydell
2015-10-27 16:00   ` Peter Maydell

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=1445956409-1818-2-git-send-email-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.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 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.