qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 07/36] target-arm: Clean up trap/undef handling of SRS
Date: Thu, 18 Feb 2016 14:34:39 +0000	[thread overview]
Message-ID: <1455806108-6961-8-git-send-email-peter.maydell@linaro.org> (raw)
In-Reply-To: <1455806108-6961-1-git-send-email-peter.maydell@linaro.org>

The SRS instruction is:
 * UNDEFINED in Hyp mode
 * UNPREDICTABLE in User or System mode
 * UNPREDICTABLE if the specified mode isn't accessible
 * trapped to EL3 if EL3 is AArch64 and we are at Secure EL1

Clean up the code to handle all these cases cleanly, including
picking UNDEF as our choice of UNPREDICTABLE behaviour rather
blindly trusting the mode field passed in the instruction.
As part of this, move the check for IS_USER into gen_srs()
itself rather than having it done by the caller.

The exception is that we don't UNDEF for calls from System
mode, which need a runtime check. This will be dealt with in
the following commits.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Sergey Fedorov <serge.fdrv@gmail.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
---
 target-arm/translate.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 61 insertions(+), 5 deletions(-)

diff --git a/target-arm/translate.c b/target-arm/translate.c
index cf3dc33..7bceb05 100644
--- a/target-arm/translate.c
+++ b/target-arm/translate.c
@@ -7578,8 +7578,67 @@ static void gen_srs(DisasContext *s,
                     uint32_t mode, uint32_t amode, bool writeback)
 {
     int32_t offset;
-    TCGv_i32 addr = tcg_temp_new_i32();
-    TCGv_i32 tmp = tcg_const_i32(mode);
+    TCGv_i32 addr, tmp;
+    bool undef = false;
+
+    /* SRS is:
+     * - trapped to EL3 if EL3 is AArch64 and we are at Secure EL1
+     * - UNDEFINED in Hyp mode
+     * - UNPREDICTABLE in User or System mode
+     * - UNPREDICTABLE if the specified mode is:
+     * -- not implemented
+     * -- not a valid mode number
+     * -- a mode that's at a higher exception level
+     * -- Monitor, if we are Non-secure
+     * For the UNPREDICTABLE cases we choose to UNDEF, except that for
+     * "current mode is System" we will write a garbage SPSR.
+     * (This is because we don't have access to our current mode here
+     * and would have to do a runtime check to UNDEF for System.)
+     */
+    if (s->current_el == 1 && !s->ns) {
+        gen_exception_insn(s, 4, EXCP_UDEF, syn_uncategorized(), 3);
+        return;
+    }
+
+    if (s->current_el == 0 || s->current_el == 2) {
+        undef = true;
+    }
+
+    switch (mode) {
+    case ARM_CPU_MODE_USR:
+    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:
+        break;
+    case ARM_CPU_MODE_HYP:
+        if (s->current_el == 1 || !arm_dc_feature(s, ARM_FEATURE_EL2)) {
+            undef = true;
+        }
+        break;
+    case ARM_CPU_MODE_MON:
+        /* No need to check specifically for "are we non-secure" because
+         * we've already made EL0 UNDEF and handled the trap for S-EL1;
+         * so if this isn't EL3 then we must be non-secure.
+         */
+        if (s->current_el != 3) {
+            undef = true;
+        }
+        break;
+    default:
+        undef = true;
+    }
+
+    if (undef) {
+        gen_exception_insn(s, 4, EXCP_UDEF, syn_uncategorized(),
+                           default_exception_el(s));
+        return;
+    }
+
+    addr = tcg_temp_new_i32();
+    tmp = tcg_const_i32(mode);
     gen_helper_get_r13_banked(addr, cpu_env, tmp);
     tcg_temp_free_i32(tmp);
     switch (amode) {
@@ -7739,9 +7798,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
             }
         } else if ((insn & 0x0e5fffe0) == 0x084d0500) {
             /* srs */
-            if (IS_USER(s)) {
-                goto illegal_op;
-            }
             ARCH(6);
             gen_srs(s, (insn & 0x1f), (insn >> 23) & 3, insn & (1 << 21));
             return;
-- 
1.9.1

  parent reply	other threads:[~2016-02-18 14:35 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-18 14:34 [Qemu-devel] [PULL 00/36] target-arm queue Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 01/36] target-arm: correct CNTFRQ access rights Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 02/36] target-arm: Fix handling of SCR.SMD Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 03/36] target-arm: Implement MDCR_EL3.TDOSA and MDCR_EL2.TDOSA traps Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 04/36] target-arm: Implement MDCR_EL2.TDRA traps Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 05/36] target-arm: Implement MDCR_EL3.TDA and MDCR_EL2.TDA traps Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 06/36] target-arm: Report correct syndrome for FPEXC32_EL2 traps Peter Maydell
2016-02-18 14:34 ` Peter Maydell [this message]
2016-02-18 14:34 ` [Qemu-devel] [PULL 08/36] target-arm: Move get/set_r13_banked() to op_helper.c Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 09/36] target-arm: Move bank_number() into internals.h Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 10/36] target-arm: Combine user-only and softmmu get/set_r13_banked() Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 11/36] target-arm: UNDEF in the UNPREDICTABLE SRS-from-System case Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 12/36] target-arm: Add the pmceid0 and pmceid1 registers Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 13/36] target-arm: Add the pmovsclr_el0 and pmintenclr_el1 registers Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 14/36] target-arm: Add PMUSERENR_EL0 register Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 15/36] ARM: PL061: Clear PL061 device state after reset Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 16/36] ARM: PL061: Cleaning field of PL061 device state Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 17/36] hw/sd/sdhci.c: Remove x-drive property Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 18/36] hw/sd/sd.c: QOMify Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 19/36] hw/sd/sd.c: Convert sd_reset() function into Device reset method Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 20/36] hw/sd: Add QOM bus which SD cards plug in to Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 21/36] hw/sd/sdhci.c: Update to use SDBus APIs Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 22/36] sdhci_sysbus: Create SD card device in users, not the device itself Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 23/36] hw/sd/pxa2xx_mmci: convert to SysBusDevice object Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 24/36] hw/sd/pxa2xx_mmci: Update to use new SDBus APIs Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 25/36] hw/sd/pxa2xx_mmci: Convert to VMStateDescription Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 26/36] hw/sd/pxa2xx_mmci: Add reset function Peter Maydell
2016-02-18 14:34 ` [Qemu-devel] [PULL 27/36] hw/sd: implement CMD23 (SET_BLOCK_COUNT) for MMC compatibility Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 28/36] hw/sd: model a power-up delay, as a workaround for an EDK2 bug Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 29/36] hw/sd: use guest error logging rather than fprintf to stderr Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 30/36] hw/timer: QOM'ify arm_timer (pass 1) Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 31/36] hw/timer: QOM'ify arm_timer (pass 2) Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 32/36] hw/timer: QOM'ify exynos4210_mct Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 33/36] hw/timer: QOM'ify exynos4210_pwm Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 34/36] hw/timer: QOM'ify exynos4210_rtc Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 35/36] hw/timer: QOM'ify pl031 Peter Maydell
2016-02-18 14:35 ` [Qemu-devel] [PULL 36/36] hw/timer: QOM'ify pxa2xx_timer Peter Maydell
2016-02-18 15:19 ` [Qemu-devel] [PULL 00/36] target-arm queue 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=1455806108-6961-8-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 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).