[PULL 00/39] target-arm queue

[PULL 00/39] target-arm queue

[Peter Maydell]

Most of this is the Neon decodetree patches, followed by Edgar's versal cleanups.

thanks
-- PMM


The following changes since commit 2ef486e76d64436be90f7359a3071fb2a56ce835:

  Merge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging (2020-05-03 14:12:56 +0100)

are available in the Git repository at:

  https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200504

for you to fetch changes up to 9aefc6cf9b73f66062d2f914a0136756e7a28211:

  target/arm: Move gen_ function typedefs to translate.h (2020-05-04 12:59:26 +0100)

----------------------------------------------------------------
target-arm queue:
 * Start of conversion of Neon insns to decodetree
 * versal board: support SD and RTC
 * Implement ARMv8.2-TTS2UXN
 * Make VQDMULL undefined when U=1
 * Some minor code cleanups

----------------------------------------------------------------
Edgar E. Iglesias (11):
      hw/arm: versal: Remove inclusion of arm_gicv3_common.h
      hw/arm: versal: Move misplaced comment
      hw/arm: versal-virt: Fix typo xlnx-ve -> xlnx-versal
      hw/arm: versal: Embed the UARTs into the SoC type
      hw/arm: versal: Embed the GEMs into the SoC type
      hw/arm: versal: Embed the ADMAs into the SoC type
      hw/arm: versal: Embed the APUs into the SoC type
      hw/arm: versal: Add support for SD
      hw/arm: versal: Add support for the RTC
      hw/arm: versal-virt: Add support for SD
      hw/arm: versal-virt: Add support for the RTC

Fredrik Strupe (1):
      target/arm: Make VQDMULL undefined when U=1

Peter Maydell (25):
      target/arm: Don't use a TLB for ARMMMUIdx_Stage2
      target/arm: Use enum constant in get_phys_addr_lpae() call
      target/arm: Add new 's1_is_el0' argument to get_phys_addr_lpae()
      target/arm: Implement ARMv8.2-TTS2UXN
      target/arm: Use correct variable for setting 'max' cpu's ID_AA64DFR0
      target/arm/translate-vfp.inc.c: Remove duplicate simd_r32 check
      target/arm: Don't allow Thumb Neon insns without FEATURE_NEON
      target/arm: Add stubs for AArch32 Neon decodetree
      target/arm: Convert VCMLA (vector) to decodetree
      target/arm: Convert VCADD (vector) to decodetree
      target/arm: Convert V[US]DOT (vector) to decodetree
      target/arm: Convert VFM[AS]L (vector) to decodetree
      target/arm: Convert VCMLA (scalar) to decodetree
      target/arm: Convert V[US]DOT (scalar) to decodetree
      target/arm: Convert VFM[AS]L (scalar) to decodetree
      target/arm: Convert Neon load/store multiple structures to decodetree
      target/arm: Convert Neon 'load single structure to all lanes' to decodetree
      target/arm: Convert Neon 'load/store single structure' to decodetree
      target/arm: Convert Neon 3-reg-same VADD/VSUB to decodetree
      target/arm: Convert Neon 3-reg-same logic ops to decodetree
      target/arm: Convert Neon 3-reg-same VMAX/VMIN to decodetree
      target/arm: Convert Neon 3-reg-same comparisons to decodetree
      target/arm: Convert Neon 3-reg-same VQADD/VQSUB to decodetree
      target/arm: Convert Neon 3-reg-same VMUL, VMLA, VMLS, VSHL to decodetree
      target/arm: Move gen_ function typedefs to translate.h

Philippe Mathieu-Daudé (2):
      hw/arm/mps2-tz: Use TYPE_IOTKIT instead of hardcoded string
      target/arm: Use uint64_t for midr field in CPU state struct

 include/hw/arm/xlnx-versal.h    |  31 +-
 target/arm/cpu-param.h          |   2 +-
 target/arm/cpu.h                |  38 ++-
 target/arm/translate-a64.h      |   9 -
 target/arm/translate.h          |  26 ++
 target/arm/neon-dp.decode       |  86 +++++
 target/arm/neon-ls.decode       |  52 +++
 target/arm/neon-shared.decode   |  66 ++++
 hw/arm/mps2-tz.c                |   2 +-
 hw/arm/xlnx-versal-virt.c       |  74 ++++-
 hw/arm/xlnx-versal.c            | 115 +++++--
 target/arm/cpu.c                |   3 +-
 target/arm/cpu64.c              |   8 +-
 target/arm/helper.c             | 183 ++++------
 target/arm/translate-a64.c      |  17 -
 target/arm/translate-neon.inc.c | 714 +++++++++++++++++++++++++++++++++++++++
 target/arm/translate-vfp.inc.c  |   6 -
 target/arm/translate.c          | 716 +++-------------------------------------
 target/arm/Makefile.objs        |  18 +
 19 files changed, 1302 insertions(+), 864 deletions(-)
 create mode 100644 target/arm/neon-dp.decode
 create mode 100644 target/arm/neon-ls.decode
 create mode 100644 target/arm/neon-shared.decode
 create mode 100644 target/arm/translate-neon.inc.c

[PULL 01/39] target/arm: Make VQDMULL undefined when U=1

[Peter Maydell]

From: Fredrik Strupe <fredrik@strupe.net>

According to Arm ARM, VQDMULL is only valid when U=0, while having
U=1 is unallocated.

Signed-off-by: Fredrik Strupe <fredrik@strupe.net>
Fixes: 695272dcb976 ("target-arm: Handle UNDEF cases for Neon 3-regs-different-widths")
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/translate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/arm/translate.c b/target/arm/translate.c
index d4ad2028f12..e3fc7924428 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -6016,7 +6016,7 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                     {0, 0, 0, 0}, /* VMLSL */
                     {0, 0, 0, 9}, /* VQDMLSL */
                     {0, 0, 0, 0}, /* Integer VMULL */
-                    {0, 0, 0, 1}, /* VQDMULL */
+                    {0, 0, 0, 9}, /* VQDMULL */
                     {0, 0, 0, 0xa}, /* Polynomial VMULL */
                     {0, 0, 0, 7}, /* Reserved: always UNDEF */
                 };
-- 
2.20.1

[PULL 02/39] hw/arm/mps2-tz: Use TYPE_IOTKIT instead of hardcoded string

[Peter Maydell]

From: Philippe Mathieu-Daudé <f4bug@amsat.org>

By using the TYPE_* definitions for devices, we can:
 - quickly find where devices are used with 'git-grep'
 - easily rename a device (one-line change).

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200428154650.21991-1-f4bug@amsat.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/mps2-tz.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/arm/mps2-tz.c b/hw/arm/mps2-tz.c
index a8dea7dde1c..2c43041564c 100644
--- a/hw/arm/mps2-tz.c
+++ b/hw/arm/mps2-tz.c
@@ -395,7 +395,7 @@ static void mps2tz_common_init(MachineState *machine)
         exit(EXIT_FAILURE);
     }
 
-    sysbus_init_child_obj(OBJECT(machine), "iotkit", &mms->iotkit,
+    sysbus_init_child_obj(OBJECT(machine), TYPE_IOTKIT, &mms->iotkit,
                           sizeof(mms->iotkit), mmc->armsse_type);
     iotkitdev = DEVICE(&mms->iotkit);
     object_property_set_link(OBJECT(&mms->iotkit), OBJECT(system_memory),
-- 
2.20.1

[PULL 03/39] target/arm: Don't use a TLB for ARMMMUIdx_Stage2

[Peter Maydell]

We define ARMMMUIdx_Stage2 as being an MMU index which uses a QEMU
TLB.  However we never actually use the TLB -- all stage 2 lookups
are done by direct calls to get_phys_addr_lpae() followed by a
physical address load via address_space_ld*().

Remove Stage2 from the list of ARM MMU indexes which correspond to
real core MMU indexes, and instead put it in the set of "NOTLB" ARM
MMU indexes.

This allows us to drop NB_MMU_MODES to 11.  It also means we can
safely add support for the ARMv8.3-TTS2UXN extension, which adds
permission bits to the stage 2 descriptors which define execute
permission separatel for EL0 and EL1; supporting that while keeping
Stage2 in a QEMU TLB would require us to use separate TLBs for
"Stage2 for an EL0 access" and "Stage2 for an EL1 access", which is a
lot of extra complication given we aren't even using the QEMU TLB.

In the process of updating the comment on our MMU index use,
fix a couple of other minor errors:
 * NS EL2 EL2&0 was missing from the list in the comment
 * some text hadn't been updated from when we bumped NB_MMU_MODES
   above 8

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200330210400.11724-2-peter.maydell@linaro.org
---
 target/arm/cpu-param.h |   2 +-
 target/arm/cpu.h       |  21 +++++---
 target/arm/helper.c    | 112 ++++-------------------------------------
 3 files changed, 27 insertions(+), 108 deletions(-)

diff --git a/target/arm/cpu-param.h b/target/arm/cpu-param.h
index d593b60b28d..6321385b469 100644
--- a/target/arm/cpu-param.h
+++ b/target/arm/cpu-param.h
@@ -29,6 +29,6 @@
 # define TARGET_PAGE_BITS_MIN  10
 #endif
 
-#define NB_MMU_MODES 12
+#define NB_MMU_MODES 11
 
 #endif
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 8b9f2961ba0..fe03a74bf08 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -2801,6 +2801,9 @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync);
  *     handling via the TLB. The only way to do a stage 1 translation without
  *     the immediate stage 2 translation is via the ATS or AT system insns,
  *     which can be slow-pathed and always do a page table walk.
+ *     The only use of stage 2 translations is either as part of an s1+2
+ *     lookup or when loading the descriptors during a stage 1 page table walk,
+ *     and in both those cases we don't use the TLB.
  *  4. we can also safely fold together the "32 bit EL3" and "64 bit EL3"
  *     translation regimes, because they map reasonably well to each other
  *     and they can't both be active at the same time.
@@ -2816,15 +2819,15 @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync);
  * NS EL1 EL1&0 stage 1+2 (aka NS PL1)
  * NS EL1 EL1&0 stage 1+2 +PAN
  * NS EL0 EL2&0
+ * NS EL2 EL2&0
  * NS EL2 EL2&0 +PAN
  * NS EL2 (aka NS PL2)
  * S EL0 EL1&0 (aka S PL0)
  * S EL1 EL1&0 (not used if EL3 is 32 bit)
  * S EL1 EL1&0 +PAN
  * S EL3 (aka S PL1)
- * NS EL1&0 stage 2
  *
- * for a total of 12 different mmu_idx.
+ * for a total of 11 different mmu_idx.
  *
  * R profile CPUs have an MPU, but can use the same set of MMU indexes
  * as A profile. They only need to distinguish NS EL0 and NS EL1 (and
@@ -2846,7 +2849,8 @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync);
  * are not quite the same -- different CPU types (most notably M profile
  * vs A/R profile) would like to use MMU indexes with different semantics,
  * but since we don't ever need to use all of those in a single CPU we
- * can avoid setting NB_MMU_MODES to more than 8. The lower bits of
+ * can avoid having to set NB_MMU_MODES to "total number of A profile MMU
+ * modes + total number of M profile MMU modes". The lower bits of
  * ARMMMUIdx are the core TLB mmu index, and the higher bits are always
  * the same for any particular CPU.
  * Variables of type ARMMUIdx are always full values, and the core
@@ -2894,8 +2898,6 @@ typedef enum ARMMMUIdx {
     ARMMMUIdx_SE10_1_PAN = 9 | ARM_MMU_IDX_A,
     ARMMMUIdx_SE3        = 10 | ARM_MMU_IDX_A,
 
-    ARMMMUIdx_Stage2     = 11 | ARM_MMU_IDX_A,
-
     /*
      * These are not allocated TLBs and are used only for AT system
      * instructions or for the first stage of an S12 page table walk.
@@ -2903,6 +2905,14 @@ typedef enum ARMMMUIdx {
     ARMMMUIdx_Stage1_E0 = 0 | ARM_MMU_IDX_NOTLB,
     ARMMMUIdx_Stage1_E1 = 1 | ARM_MMU_IDX_NOTLB,
     ARMMMUIdx_Stage1_E1_PAN = 2 | ARM_MMU_IDX_NOTLB,
+    /*
+     * Not allocated a TLB: used only for second stage of an S12 page
+     * table walk, or for descriptor loads during first stage of an S1
+     * page table walk. Note that if we ever want to have a TLB for this
+     * then various TLB flush insns which currently are no-ops or flush
+     * only stage 1 MMU indexes will need to change to flush stage 2.
+     */
+    ARMMMUIdx_Stage2     = 3 | ARM_MMU_IDX_NOTLB,
 
     /*
      * M-profile.
@@ -2936,7 +2946,6 @@ typedef enum ARMMMUIdxBit {
     TO_CORE_BIT(SE10_1),
     TO_CORE_BIT(SE10_1_PAN),
     TO_CORE_BIT(SE3),
-    TO_CORE_BIT(Stage2),
 
     TO_CORE_BIT(MUser),
     TO_CORE_BIT(MPriv),
diff --git a/target/arm/helper.c b/target/arm/helper.c
index dfefb9b3d9b..f785e012cde 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -814,8 +814,7 @@ static void tlbiall_nsnh_write(CPUARMState *env, const ARMCPRegInfo *ri,
     tlb_flush_by_mmuidx(cs,
                         ARMMMUIdxBit_E10_1 |
                         ARMMMUIdxBit_E10_1_PAN |
-                        ARMMMUIdxBit_E10_0 |
-                        ARMMMUIdxBit_Stage2);
+                        ARMMMUIdxBit_E10_0);
 }
 
 static void tlbiall_nsnh_is_write(CPUARMState *env, const ARMCPRegInfo *ri,
@@ -826,46 +825,9 @@ static void tlbiall_nsnh_is_write(CPUARMState *env, const ARMCPRegInfo *ri,
     tlb_flush_by_mmuidx_all_cpus_synced(cs,
                                         ARMMMUIdxBit_E10_1 |
                                         ARMMMUIdxBit_E10_1_PAN |
-                                        ARMMMUIdxBit_E10_0 |
-                                        ARMMMUIdxBit_Stage2);
+                                        ARMMMUIdxBit_E10_0);
 }
 
-static void tlbiipas2_write(CPUARMState *env, const ARMCPRegInfo *ri,
-                            uint64_t value)
-{
-    /* Invalidate by IPA. This has to invalidate any structures that
-     * contain only stage 2 translation information, but does not need
-     * to apply to structures that contain combined stage 1 and stage 2
-     * translation information.
-     * This must NOP if EL2 isn't implemented or SCR_EL3.NS is zero.
-     */
-    CPUState *cs = env_cpu(env);
-    uint64_t pageaddr;
-
-    if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) {
-        return;
-    }
-
-    pageaddr = sextract64(value << 12, 0, 40);
-
-    tlb_flush_page_by_mmuidx(cs, pageaddr, ARMMMUIdxBit_Stage2);
-}
-
-static void tlbiipas2_is_write(CPUARMState *env, const ARMCPRegInfo *ri,
-                               uint64_t value)
-{
-    CPUState *cs = env_cpu(env);
-    uint64_t pageaddr;
-
-    if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) {
-        return;
-    }
-
-    pageaddr = sextract64(value << 12, 0, 40);
-
-    tlb_flush_page_by_mmuidx_all_cpus_synced(cs, pageaddr,
-                                             ARMMMUIdxBit_Stage2);
-}
 
 static void tlbiall_hyp_write(CPUARMState *env, const ARMCPRegInfo *ri,
                               uint64_t value)
@@ -4055,8 +4017,7 @@ static void vttbr_write(CPUARMState *env, const ARMCPRegInfo *ri,
         tlb_flush_by_mmuidx(cs,
                             ARMMMUIdxBit_E10_1 |
                             ARMMMUIdxBit_E10_1_PAN |
-                            ARMMMUIdxBit_E10_0 |
-                            ARMMMUIdxBit_Stage2);
+                            ARMMMUIdxBit_E10_0);
         raw_write(env, ri, value);
     }
 }
@@ -4538,11 +4499,6 @@ static int alle1_tlbmask(CPUARMState *env)
         return ARMMMUIdxBit_SE10_1 |
                ARMMMUIdxBit_SE10_1_PAN |
                ARMMMUIdxBit_SE10_0;
-    } else if (arm_feature(env, ARM_FEATURE_EL2)) {
-        return ARMMMUIdxBit_E10_1 |
-               ARMMMUIdxBit_E10_1_PAN |
-               ARMMMUIdxBit_E10_0 |
-               ARMMMUIdxBit_Stage2;
     } else {
         return ARMMMUIdxBit_E10_1 |
                ARMMMUIdxBit_E10_1_PAN |
@@ -4689,44 +4645,6 @@ static void tlbi_aa64_vae3is_write(CPUARMState *env, const ARMCPRegInfo *ri,
                                              ARMMMUIdxBit_SE3);
 }
 
-static void tlbi_aa64_ipas2e1_write(CPUARMState *env, const ARMCPRegInfo *ri,
-                                    uint64_t value)
-{
-    /* Invalidate by IPA. This has to invalidate any structures that
-     * contain only stage 2 translation information, but does not need
-     * to apply to structures that contain combined stage 1 and stage 2
-     * translation information.
-     * This must NOP if EL2 isn't implemented or SCR_EL3.NS is zero.
-     */
-    ARMCPU *cpu = env_archcpu(env);
-    CPUState *cs = CPU(cpu);
-    uint64_t pageaddr;
-
-    if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) {
-        return;
-    }
-
-    pageaddr = sextract64(value << 12, 0, 48);
-
-    tlb_flush_page_by_mmuidx(cs, pageaddr, ARMMMUIdxBit_Stage2);
-}
-
-static void tlbi_aa64_ipas2e1is_write(CPUARMState *env, const ARMCPRegInfo *ri,
-                                      uint64_t value)
-{
-    CPUState *cs = env_cpu(env);
-    uint64_t pageaddr;
-
-    if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) {
-        return;
-    }
-
-    pageaddr = sextract64(value << 12, 0, 48);
-
-    tlb_flush_page_by_mmuidx_all_cpus_synced(cs, pageaddr,
-                                             ARMMMUIdxBit_Stage2);
-}
-
 static CPAccessResult aa64_zva_access(CPUARMState *env, const ARMCPRegInfo *ri,
                                       bool isread)
 {
@@ -4965,12 +4883,10 @@ static const ARMCPRegInfo v8_cp_reginfo[] = {
       .writefn = tlbi_aa64_vae1_write },
     { .name = "TLBI_IPAS2E1IS", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 1,
-      .access = PL2_W, .type = ARM_CP_NO_RAW,
-      .writefn = tlbi_aa64_ipas2e1is_write },
+      .access = PL2_W, .type = ARM_CP_NOP },
     { .name = "TLBI_IPAS2LE1IS", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 5,
-      .access = PL2_W, .type = ARM_CP_NO_RAW,
-      .writefn = tlbi_aa64_ipas2e1is_write },
+      .access = PL2_W, .type = ARM_CP_NOP },
     { .name = "TLBI_ALLE1IS", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 4,
       .access = PL2_W, .type = ARM_CP_NO_RAW,
@@ -4981,12 +4897,10 @@ static const ARMCPRegInfo v8_cp_reginfo[] = {
       .writefn = tlbi_aa64_alle1is_write },
     { .name = "TLBI_IPAS2E1", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 1,
-      .access = PL2_W, .type = ARM_CP_NO_RAW,
-      .writefn = tlbi_aa64_ipas2e1_write },
+      .access = PL2_W, .type = ARM_CP_NOP },
     { .name = "TLBI_IPAS2LE1", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 5,
-      .access = PL2_W, .type = ARM_CP_NO_RAW,
-      .writefn = tlbi_aa64_ipas2e1_write },
+      .access = PL2_W, .type = ARM_CP_NOP },
     { .name = "TLBI_ALLE1", .state = ARM_CP_STATE_AA64,
       .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 4,
       .access = PL2_W, .type = ARM_CP_NO_RAW,
@@ -5067,20 +4981,16 @@ static const ARMCPRegInfo v8_cp_reginfo[] = {
       .writefn = tlbimva_hyp_is_write },
     { .name = "TLBIIPAS2",
       .cp = 15, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 1,
-      .type = ARM_CP_NO_RAW, .access = PL2_W,
-      .writefn = tlbiipas2_write },
+      .type = ARM_CP_NOP, .access = PL2_W },
     { .name = "TLBIIPAS2IS",
       .cp = 15, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 1,
-      .type = ARM_CP_NO_RAW, .access = PL2_W,
-      .writefn = tlbiipas2_is_write },
+      .type = ARM_CP_NOP, .access = PL2_W },
     { .name = "TLBIIPAS2L",
       .cp = 15, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 5,
-      .type = ARM_CP_NO_RAW, .access = PL2_W,
-      .writefn = tlbiipas2_write },
+      .type = ARM_CP_NOP, .access = PL2_W },
     { .name = "TLBIIPAS2LIS",
       .cp = 15, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 5,
-      .type = ARM_CP_NO_RAW, .access = PL2_W,
-      .writefn = tlbiipas2_is_write },
+      .type = ARM_CP_NOP, .access = PL2_W },
     /* 32 bit cache operations */
     { .name = "ICIALLUIS", .cp = 15, .opc1 = 0, .crn = 7, .crm = 1, .opc2 = 0,
       .type = ARM_CP_NOP, .access = PL1_W, .accessfn = aa64_cacheop_pou_access },
-- 
2.20.1

[PULL 04/39] target/arm: Use enum constant in get_phys_addr_lpae() call

[Peter Maydell]

The access_type argument to get_phys_addr_lpae() is an MMUAccessType;
use the enum constant MMU_DATA_LOAD rather than a literal 0 when we
call it in S1_ptw_translate().

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200330210400.11724-3-peter.maydell@linaro.org
---
 target/arm/helper.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index f785e012cde..f17841ec24e 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -10052,8 +10052,9 @@ static hwaddr S1_ptw_translate(CPUARMState *env, ARMMMUIdx mmu_idx,
             pcacheattrs = &cacheattrs;
         }
 
-        ret = get_phys_addr_lpae(env, addr, 0, ARMMMUIdx_Stage2, &s2pa,
-                                 &txattrs, &s2prot, &s2size, fi, pcacheattrs);
+        ret = get_phys_addr_lpae(env, addr, MMU_DATA_LOAD, ARMMMUIdx_Stage2,
+                                 &s2pa, &txattrs, &s2prot, &s2size, fi,
+                                 pcacheattrs);
         if (ret) {
             assert(fi->type != ARMFault_None);
             fi->s2addr = addr;
-- 
2.20.1

[PULL 05/39] target/arm: Add new 's1_is_el0' argument to get_phys_addr_lpae()

[Peter Maydell]

For ARMv8.2-TTS2UXN, the stage 2 page table walk wants to know
whether the stage 1 access is for EL0 or not, because whether
exec permission is given can depend on whether this is an EL0
or EL1 access. Add a new argument to get_phys_addr_lpae() so
the call sites can pass this information in.

Since get_phys_addr_lpae() doesn't already have a doc comment,
add one so we have a place to put the documentation of the
semantics of the new s1_is_el0 argument.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200330210400.11724-4-peter.maydell@linaro.org
---
 target/arm/helper.c | 29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)

diff --git a/target/arm/helper.c b/target/arm/helper.c
index f17841ec24e..b26f5801943 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -41,6 +41,7 @@
 
 static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address,
                                MMUAccessType access_type, ARMMMUIdx mmu_idx,
+                               bool s1_is_el0,
                                hwaddr *phys_ptr, MemTxAttrs *txattrs, int *prot,
                                target_ulong *page_size_ptr,
                                ARMMMUFaultInfo *fi, ARMCacheAttrs *cacheattrs);
@@ -10053,6 +10054,7 @@ static hwaddr S1_ptw_translate(CPUARMState *env, ARMMMUIdx mmu_idx,
         }
 
         ret = get_phys_addr_lpae(env, addr, MMU_DATA_LOAD, ARMMMUIdx_Stage2,
+                                 false,
                                  &s2pa, &txattrs, &s2prot, &s2size, fi,
                                  pcacheattrs);
         if (ret) {
@@ -10655,8 +10657,32 @@ static ARMVAParameters aa32_va_parameters(CPUARMState *env, uint32_t va,
     };
 }
 
+/**
+ * get_phys_addr_lpae: perform one stage of page table walk, LPAE format
+ *
+ * Returns false if the translation was successful. Otherwise, phys_ptr, attrs,
+ * prot and page_size may not be filled in, and the populated fsr value provides
+ * information on why the translation aborted, in the format of a long-format
+ * DFSR/IFSR fault register, with the following caveats:
+ *  * the WnR bit is never set (the caller must do this).
+ *
+ * @env: CPUARMState
+ * @address: virtual address to get physical address for
+ * @access_type: MMU_DATA_LOAD, MMU_DATA_STORE or MMU_INST_FETCH
+ * @mmu_idx: MMU index indicating required translation regime
+ * @s1_is_el0: if @mmu_idx is ARMMMUIdx_Stage2 (so this is a stage 2 page table
+ *             walk), must be true if this is stage 2 of a stage 1+2 walk for an
+ *             EL0 access). If @mmu_idx is anything else, @s1_is_el0 is ignored.
+ * @phys_ptr: set to the physical address corresponding to the virtual address
+ * @attrs: set to the memory transaction attributes to use
+ * @prot: set to the permissions for the page containing phys_ptr
+ * @page_size_ptr: set to the size of the page containing phys_ptr
+ * @fi: set to fault info if the translation fails
+ * @cacheattrs: (if non-NULL) set to the cacheability/shareability attributes
+ */
 static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address,
                                MMUAccessType access_type, ARMMMUIdx mmu_idx,
+                               bool s1_is_el0,
                                hwaddr *phys_ptr, MemTxAttrs *txattrs, int *prot,
                                target_ulong *page_size_ptr,
                                ARMMMUFaultInfo *fi, ARMCacheAttrs *cacheattrs)
@@ -11748,6 +11774,7 @@ bool get_phys_addr(CPUARMState *env, target_ulong address,
 
             /* S1 is done. Now do S2 translation.  */
             ret = get_phys_addr_lpae(env, ipa, access_type, ARMMMUIdx_Stage2,
+                                     mmu_idx == ARMMMUIdx_E10_0,
                                      phys_ptr, attrs, &s2_prot,
                                      page_size, fi,
                                      cacheattrs != NULL ? &cacheattrs2 : NULL);
@@ -11872,7 +11899,7 @@ bool get_phys_addr(CPUARMState *env, target_ulong address,
     }
 
     if (regime_using_lpae_format(env, mmu_idx)) {
-        return get_phys_addr_lpae(env, address, access_type, mmu_idx,
+        return get_phys_addr_lpae(env, address, access_type, mmu_idx, false,
                                   phys_ptr, attrs, prot, page_size,
                                   fi, cacheattrs);
     } else if (regime_sctlr(env, mmu_idx) & SCTLR_XP) {
-- 
2.20.1

[PULL 06/39] target/arm: Implement ARMv8.2-TTS2UXN

[Peter Maydell]

The ARMv8.2-TTS2UXN feature extends the XN field in stage 2
translation table descriptors from just bit [54] to bits [54:53],
allowing stage 2 to control execution permissions separately for EL0
and EL1. Implement the new semantics of the XN field and enable
the feature for our 'max' CPU.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200330210400.11724-5-peter.maydell@linaro.org
---
 target/arm/cpu.h    | 15 +++++++++++++++
 target/arm/cpu.c    |  1 +
 target/arm/cpu64.c  |  2 ++
 target/arm/helper.c | 37 +++++++++++++++++++++++++++++++------
 4 files changed, 49 insertions(+), 6 deletions(-)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index fe03a74bf08..9aae324d0f6 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -3610,6 +3610,11 @@ static inline bool isar_feature_aa32_ccidx(const ARMISARegisters *id)
     return FIELD_EX32(id->id_mmfr4, ID_MMFR4, CCIDX) != 0;
 }
 
+static inline bool isar_feature_aa32_tts2uxn(const ARMISARegisters *id)
+{
+    return FIELD_EX32(id->id_mmfr4, ID_MMFR4, XNX) != 0;
+}
+
 /*
  * 64-bit feature tests via id registers.
  */
@@ -3822,6 +3827,11 @@ static inline bool isar_feature_aa64_ccidx(const ARMISARegisters *id)
     return FIELD_EX64(id->id_aa64mmfr2, ID_AA64MMFR2, CCIDX) != 0;
 }
 
+static inline bool isar_feature_aa64_tts2uxn(const ARMISARegisters *id)
+{
+    return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, XNX) != 0;
+}
+
 /*
  * Feature tests for "does this exist in either 32-bit or 64-bit?"
  */
@@ -3850,6 +3860,11 @@ static inline bool isar_feature_any_ccidx(const ARMISARegisters *id)
     return isar_feature_aa64_ccidx(id) || isar_feature_aa32_ccidx(id);
 }
 
+static inline bool isar_feature_any_tts2uxn(const ARMISARegisters *id)
+{
+    return isar_feature_aa64_tts2uxn(id) || isar_feature_aa32_tts2uxn(id);
+}
+
 /*
  * Forward to the above feature tests given an ARMCPU pointer.
  */
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index 141d947775a..f588344df83 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -2686,6 +2686,7 @@ static void arm_max_initfn(Object *obj)
             t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */
             t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */
             t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */
+            t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */
             cpu->isar.id_mmfr4 = t;
         }
 #endif
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index 74afc28d537..e232c0ea12c 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -673,6 +673,7 @@ static void aarch64_max_initfn(Object *obj)
         t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1);
         t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* ATS1E1 */
         t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* VMID16 */
+        t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* TTS2UXN */
         cpu->isar.id_aa64mmfr1 = t;
 
         t = cpu->isar.id_aa64mmfr2;
@@ -706,6 +707,7 @@ static void aarch64_max_initfn(Object *obj)
         u = FIELD_DP32(u, ID_MMFR4, HPDS, 1); /* AA32HPD */
         u = FIELD_DP32(u, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */
         u = FIELD_DP32(u, ID_MMFR4, CNP, 1); /* TTCNP */
+        u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */
         cpu->isar.id_mmfr4 = u;
 
         u = cpu->isar.id_aa64dfr0;
diff --git a/target/arm/helper.c b/target/arm/helper.c
index b26f5801943..a94f6507950 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -9908,9 +9908,10 @@ simple_ap_to_rw_prot(CPUARMState *env, ARMMMUIdx mmu_idx, int ap)
  *
  * @env:     CPUARMState
  * @s2ap:    The 2-bit stage2 access permissions (S2AP)
- * @xn:      XN (execute-never) bit
+ * @xn:      XN (execute-never) bits
+ * @s1_is_el0: true if this is S2 of an S1+2 walk for EL0
  */
-static int get_S2prot(CPUARMState *env, int s2ap, int xn)
+static int get_S2prot(CPUARMState *env, int s2ap, int xn, bool s1_is_el0)
 {
     int prot = 0;
 
@@ -9920,9 +9921,32 @@ static int get_S2prot(CPUARMState *env, int s2ap, int xn)
     if (s2ap & 2) {
         prot |= PAGE_WRITE;
     }
-    if (!xn) {
-        if (arm_el_is_aa64(env, 2) || prot & PAGE_READ) {
+
+    if (cpu_isar_feature(any_tts2uxn, env_archcpu(env))) {
+        switch (xn) {
+        case 0:
             prot |= PAGE_EXEC;
+            break;
+        case 1:
+            if (s1_is_el0) {
+                prot |= PAGE_EXEC;
+            }
+            break;
+        case 2:
+            break;
+        case 3:
+            if (!s1_is_el0) {
+                prot |= PAGE_EXEC;
+            }
+            break;
+        default:
+            g_assert_not_reached();
+        }
+    } else {
+        if (!extract32(xn, 1, 1)) {
+            if (arm_el_is_aa64(env, 2) || prot & PAGE_READ) {
+                prot |= PAGE_EXEC;
+            }
         }
     }
     return prot;
@@ -10901,13 +10925,14 @@ static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address,
     }
 
     ap = extract32(attrs, 4, 2);
-    xn = extract32(attrs, 12, 1);
 
     if (mmu_idx == ARMMMUIdx_Stage2) {
         ns = true;
-        *prot = get_S2prot(env, ap, xn);
+        xn = extract32(attrs, 11, 2);
+        *prot = get_S2prot(env, ap, xn, s1_is_el0);
     } else {
         ns = extract32(attrs, 3, 1);
+        xn = extract32(attrs, 12, 1);
         pxn = extract32(attrs, 11, 1);
         *prot = get_S1prot(env, mmu_idx, aarch64, ap, ns, xn, pxn);
     }
-- 
2.20.1

[PULL 07/39] target/arm: Use correct variable for setting 'max' cpu's ID_AA64DFR0

[Peter Maydell]

In aarch64_max_initfn() we update both 32-bit and 64-bit ID
registers.  The intended pattern is that for 64-bit ID registers we
use FIELD_DP64 and the uint64_t 't' register, while 32-bit ID
registers use FIELD_DP32 and the uint32_t 'u' register.  For
ID_AA64DFR0 we accidentally used 'u', meaning that the top 32 bits of
this 64-bit ID register would end up always zero.  Luckily at the
moment that's what they should be anyway, so this bug has no visible
effects.

Use the right-sized variable.

Fixes: 3bec78447a958d481991
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200423110915.10527-1-peter.maydell@linaro.org
---
 target/arm/cpu64.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index e232c0ea12c..9bdf75b1abb 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -710,9 +710,9 @@ static void aarch64_max_initfn(Object *obj)
         u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */
         cpu->isar.id_mmfr4 = u;
 
-        u = cpu->isar.id_aa64dfr0;
-        u = FIELD_DP64(u, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */
-        cpu->isar.id_aa64dfr0 = u;
+        t = cpu->isar.id_aa64dfr0;
+        t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */
+        cpu->isar.id_aa64dfr0 = t;
 
         u = cpu->isar.id_dfr0;
         u = FIELD_DP32(u, ID_DFR0, PERFMON, 5); /* v8.4-PMU */
-- 
2.20.1

[PULL 08/39] target/arm: Use uint64_t for midr field in CPU state struct

[Peter Maydell]

From: Philippe Mathieu-Daudé <f4bug@amsat.org>

MIDR_EL1 is a 64-bit system register with the top 32-bit being RES0.
Represent it in QEMU's ARMCPU struct with a uint64_t, not a
uint32_t.

This fixes an error when compiling with -Werror=conversion
because we were manipulating the register value using a
local uint64_t variable:

  target/arm/cpu64.c: In function ‘aarch64_max_initfn’:
  target/arm/cpu64.c:628:21: error: conversion from ‘uint64_t’ {aka ‘long unsigned int’} to ‘uint32_t’ {aka ‘unsigned int’} may change value [-Werror=conversion]
    628 |         cpu->midr = t;
        |                     ^

and future-proofs us against a possible future architecture
change using some of the top 32 bits.

Suggested-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Message-id: 20200428172634.29707-1-f4bug@amsat.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target/arm/cpu.h | 2 +-
 target/arm/cpu.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 9aae324d0f6..8608da6b6fc 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -894,7 +894,7 @@ struct ARMCPU {
         uint64_t id_aa64dfr0;
         uint64_t id_aa64dfr1;
     } isar;
-    uint32_t midr;
+    uint64_t midr;
     uint32_t revidr;
     uint32_t reset_fpsid;
     uint32_t ctr;
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index f588344df83..5d64adfe76e 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -2755,7 +2755,7 @@ static const ARMCPUInfo arm_cpus[] = {
 static Property arm_cpu_properties[] = {
     DEFINE_PROP_BOOL("start-powered-off", ARMCPU, start_powered_off, false),
     DEFINE_PROP_UINT32("psci-conduit", ARMCPU, psci_conduit, 0),
-    DEFINE_PROP_UINT32("midr", ARMCPU, midr, 0),
+    DEFINE_PROP_UINT64("midr", ARMCPU, midr, 0),
     DEFINE_PROP_UINT64("mp-affinity", ARMCPU,
                         mp_affinity, ARM64_AFFINITY_INVALID),
     DEFINE_PROP_INT32("node-id", ARMCPU, node_id, CPU_UNSET_NUMA_NODE_ID),
-- 
2.20.1

[PULL 09/39] hw/arm: versal: Remove inclusion of arm_gicv3_common.h

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Remove inclusion of arm_gicv3_common.h, this already gets
included via xlnx-versal.h.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-2-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/xlnx-versal.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index 94460f2343b..c73b2fe7555 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -20,7 +20,6 @@
 #include "hw/arm/boot.h"
 #include "kvm_arm.h"
 #include "hw/misc/unimp.h"
-#include "hw/intc/arm_gicv3_common.h"
 #include "hw/arm/xlnx-versal.h"
 #include "hw/char/pl011.h"
 
-- 
2.20.1

[PULL 10/39] hw/arm: versal: Move misplaced comment

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Move misplaced comment.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-3-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/xlnx-versal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index c73b2fe7555..cc696e44c06 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -36,7 +36,6 @@ static void versal_create_apu_cpus(Versal *s)
 
         obj = object_new(XLNX_VERSAL_ACPU_TYPE);
         if (!obj) {
-            /* Secondary CPUs start in PSCI powered-down state */
             error_report("Unable to create apu.cpu[%d] of type %s",
                          i, XLNX_VERSAL_ACPU_TYPE);
             exit(EXIT_FAILURE);
@@ -49,6 +48,7 @@ static void versal_create_apu_cpus(Versal *s)
         object_property_set_int(obj, s->cfg.psci_conduit,
                                 "psci-conduit", &error_abort);
         if (i) {
+            /* Secondary CPUs start in PSCI powered-down state */
             object_property_set_bool(obj, true,
                                      "start-powered-off", &error_abort);
         }
-- 
2.20.1

[PULL 11/39] hw/arm: versal-virt: Fix typo xlnx-ve -> xlnx-versal

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Fix typo xlnx-ve -> xlnx-versal.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-4-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/xlnx-versal-virt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c
index 878a2751400..8a608074d15 100644
--- a/hw/arm/xlnx-versal-virt.c
+++ b/hw/arm/xlnx-versal-virt.c
@@ -440,7 +440,7 @@ static void versal_virt_init(MachineState *machine)
         psci_conduit = QEMU_PSCI_CONDUIT_SMC;
     }
 
-    sysbus_init_child_obj(OBJECT(machine), "xlnx-ve", &s->soc,
+    sysbus_init_child_obj(OBJECT(machine), "xlnx-versal", &s->soc,
                           sizeof(s->soc), TYPE_XLNX_VERSAL);
     object_property_set_link(OBJECT(&s->soc), OBJECT(machine->ram),
                              "ddr", &error_abort);
-- 
2.20.1

[PULL 12/39] hw/arm: versal: Embed the UARTs into the SoC type

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Embed the UARTs into the SoC type.

Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-5-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h |  3 ++-
 hw/arm/xlnx-versal.c         | 12 ++++++------
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index 6c0a692b2fd..a3dfd064b3c 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -15,6 +15,7 @@
 #include "hw/sysbus.h"
 #include "hw/arm/boot.h"
 #include "hw/intc/arm_gicv3.h"
+#include "hw/char/pl011.h"
 
 #define TYPE_XLNX_VERSAL "xlnx-versal"
 #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL)
@@ -49,7 +50,7 @@ typedef struct Versal {
         MemoryRegion mr_ocm;
 
         struct {
-            SysBusDevice *uart[XLNX_VERSAL_NR_UARTS];
+            PL011State uart[XLNX_VERSAL_NR_UARTS];
             SysBusDevice *gem[XLNX_VERSAL_NR_GEMS];
             SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS];
         } iou;
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index cc696e44c06..dbde03b7e69 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -21,7 +21,6 @@
 #include "kvm_arm.h"
 #include "hw/misc/unimp.h"
 #include "hw/arm/xlnx-versal.h"
-#include "hw/char/pl011.h"
 
 #define XLNX_VERSAL_ACPU_TYPE ARM_CPU_TYPE_NAME("cortex-a72")
 #define GEM_REVISION        0x40070106
@@ -144,16 +143,17 @@ static void versal_create_uarts(Versal *s, qemu_irq *pic)
         DeviceState *dev;
         MemoryRegion *mr;
 
-        dev = qdev_create(NULL, TYPE_PL011);
-        s->lpd.iou.uart[i] = SYS_BUS_DEVICE(dev);
+        sysbus_init_child_obj(OBJECT(s), name,
+                              &s->lpd.iou.uart[i], sizeof(s->lpd.iou.uart[i]),
+                              TYPE_PL011);
+        dev = DEVICE(&s->lpd.iou.uart[i]);
         qdev_prop_set_chr(dev, "chardev", serial_hd(i));
-        object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal);
         qdev_init_nofail(dev);
 
-        mr = sysbus_mmio_get_region(s->lpd.iou.uart[i], 0);
+        mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
         memory_region_add_subregion(&s->mr_ps, addrs[i], mr);
 
-        sysbus_connect_irq(s->lpd.iou.uart[i], 0, pic[irqs[i]]);
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irqs[i]]);
         g_free(name);
     }
 }
-- 
2.20.1

[PULL 13/39] hw/arm: versal: Embed the GEMs into the SoC type

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Embed the GEMs into the SoC type.

Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-6-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h |  3 ++-
 hw/arm/xlnx-versal.c         | 15 ++++++++-------
 2 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index a3dfd064b3c..01da736a5b2 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -16,6 +16,7 @@
 #include "hw/arm/boot.h"
 #include "hw/intc/arm_gicv3.h"
 #include "hw/char/pl011.h"
+#include "hw/net/cadence_gem.h"
 
 #define TYPE_XLNX_VERSAL "xlnx-versal"
 #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL)
@@ -51,7 +52,7 @@ typedef struct Versal {
 
         struct {
             PL011State uart[XLNX_VERSAL_NR_UARTS];
-            SysBusDevice *gem[XLNX_VERSAL_NR_GEMS];
+            CadenceGEMState gem[XLNX_VERSAL_NR_GEMS];
             SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS];
         } iou;
     } lpd;
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index dbde03b7e69..e424aa789e3 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -170,25 +170,26 @@ static void versal_create_gems(Versal *s, qemu_irq *pic)
         DeviceState *dev;
         MemoryRegion *mr;
 
-        dev = qdev_create(NULL, "cadence_gem");
-        s->lpd.iou.gem[i] = SYS_BUS_DEVICE(dev);
-        object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal);
+        sysbus_init_child_obj(OBJECT(s), name,
+                              &s->lpd.iou.gem[i], sizeof(s->lpd.iou.gem[i]),
+                              TYPE_CADENCE_GEM);
+        dev = DEVICE(&s->lpd.iou.gem[i]);
         if (nd->used) {
             qemu_check_nic_model(nd, "cadence_gem");
             qdev_set_nic_properties(dev, nd);
         }
-        object_property_set_int(OBJECT(s->lpd.iou.gem[i]),
+        object_property_set_int(OBJECT(dev),
                                 2, "num-priority-queues",
                                 &error_abort);
-        object_property_set_link(OBJECT(s->lpd.iou.gem[i]),
+        object_property_set_link(OBJECT(dev),
                                  OBJECT(&s->mr_ps), "dma",
                                  &error_abort);
         qdev_init_nofail(dev);
 
-        mr = sysbus_mmio_get_region(s->lpd.iou.gem[i], 0);
+        mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
         memory_region_add_subregion(&s->mr_ps, addrs[i], mr);
 
-        sysbus_connect_irq(s->lpd.iou.gem[i], 0, pic[irqs[i]]);
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irqs[i]]);
         g_free(name);
     }
 }
-- 
2.20.1

[PULL 14/39] hw/arm: versal: Embed the ADMAs into the SoC type

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Embed the ADMAs into the SoC type.

Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-7-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h |  3 ++-
 hw/arm/xlnx-versal.c         | 14 +++++++-------
 2 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index 01da736a5b2..94b7826fd49 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -16,6 +16,7 @@
 #include "hw/arm/boot.h"
 #include "hw/intc/arm_gicv3.h"
 #include "hw/char/pl011.h"
+#include "hw/dma/xlnx-zdma.h"
 #include "hw/net/cadence_gem.h"
 
 #define TYPE_XLNX_VERSAL "xlnx-versal"
@@ -53,7 +54,7 @@ typedef struct Versal {
         struct {
             PL011State uart[XLNX_VERSAL_NR_UARTS];
             CadenceGEMState gem[XLNX_VERSAL_NR_GEMS];
-            SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS];
+            XlnxZDMA adma[XLNX_VERSAL_NR_ADMAS];
         } iou;
     } lpd;
 
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index e424aa789e3..ebd2dc51beb 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -203,18 +203,18 @@ static void versal_create_admas(Versal *s, qemu_irq *pic)
         DeviceState *dev;
         MemoryRegion *mr;
 
-        dev = qdev_create(NULL, "xlnx.zdma");
-        s->lpd.iou.adma[i] = SYS_BUS_DEVICE(dev);
-        object_property_set_int(OBJECT(s->lpd.iou.adma[i]), 128, "bus-width",
-                                &error_abort);
-        object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal);
+        sysbus_init_child_obj(OBJECT(s), name,
+                              &s->lpd.iou.adma[i], sizeof(s->lpd.iou.adma[i]),
+                              TYPE_XLNX_ZDMA);
+        dev = DEVICE(&s->lpd.iou.adma[i]);
+        object_property_set_int(OBJECT(dev), 128, "bus-width", &error_abort);
         qdev_init_nofail(dev);
 
-        mr = sysbus_mmio_get_region(s->lpd.iou.adma[i], 0);
+        mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
         memory_region_add_subregion(&s->mr_ps,
                                     MM_ADMA_CH0 + i * MM_ADMA_CH0_SIZE, mr);
 
-        sysbus_connect_irq(s->lpd.iou.adma[i], 0, pic[VERSAL_ADMA_IRQ_0 + i]);
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[VERSAL_ADMA_IRQ_0 + i]);
         g_free(name);
     }
 }
-- 
2.20.1

[PULL 15/39] hw/arm: versal: Embed the APUs into the SoC type

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Embed the APUs into the SoC type.

Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-8-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h |  2 +-
 hw/arm/xlnx-versal-virt.c    |  4 ++--
 hw/arm/xlnx-versal.c         | 19 +++++--------------
 3 files changed, 8 insertions(+), 17 deletions(-)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index 94b7826fd49..426b66449de 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -36,7 +36,7 @@ typedef struct Versal {
     struct {
         struct {
             MemoryRegion mr;
-            ARMCPU *cpu[XLNX_VERSAL_NR_ACPUS];
+            ARMCPU cpu[XLNX_VERSAL_NR_ACPUS];
             GICv3State gic;
         } apu;
     } fpd;
diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c
index 8a608074d15..d7be1ad4942 100644
--- a/hw/arm/xlnx-versal-virt.c
+++ b/hw/arm/xlnx-versal-virt.c
@@ -469,9 +469,9 @@ static void versal_virt_init(MachineState *machine)
     s->binfo.get_dtb = versal_virt_get_dtb;
     s->binfo.modify_dtb = versal_virt_modify_dtb;
     if (machine->kernel_filename) {
-        arm_load_kernel(s->soc.fpd.apu.cpu[0], machine, &s->binfo);
+        arm_load_kernel(&s->soc.fpd.apu.cpu[0], machine, &s->binfo);
     } else {
-        AddressSpace *as = arm_boot_address_space(s->soc.fpd.apu.cpu[0],
+        AddressSpace *as = arm_boot_address_space(&s->soc.fpd.apu.cpu[0],
                                                   &s->binfo);
         /* Some boot-loaders (e.g u-boot) don't like blobs at address 0 (NULL).
          * Offset things by 4K.  */
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index ebd2dc51beb..c8a296e2e05 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -31,19 +31,11 @@ static void versal_create_apu_cpus(Versal *s)
 
     for (i = 0; i < ARRAY_SIZE(s->fpd.apu.cpu); i++) {
         Object *obj;
-        char *name;
-
-        obj = object_new(XLNX_VERSAL_ACPU_TYPE);
-        if (!obj) {
-            error_report("Unable to create apu.cpu[%d] of type %s",
-                         i, XLNX_VERSAL_ACPU_TYPE);
-            exit(EXIT_FAILURE);
-        }
-
-        name = g_strdup_printf("apu-cpu[%d]", i);
-        object_property_add_child(OBJECT(s), name, obj, &error_fatal);
-        g_free(name);
 
+        object_initialize_child(OBJECT(s), "apu-cpu[*]",
+                                &s->fpd.apu.cpu[i], sizeof(s->fpd.apu.cpu[i]),
+                                XLNX_VERSAL_ACPU_TYPE, &error_abort, NULL);
+        obj = OBJECT(&s->fpd.apu.cpu[i]);
         object_property_set_int(obj, s->cfg.psci_conduit,
                                 "psci-conduit", &error_abort);
         if (i) {
@@ -57,7 +49,6 @@ static void versal_create_apu_cpus(Versal *s)
         object_property_set_link(obj, OBJECT(&s->fpd.apu.mr), "memory",
                                  &error_abort);
         object_property_set_bool(obj, true, "realized", &error_fatal);
-        s->fpd.apu.cpu[i] = ARM_CPU(obj);
     }
 }
 
@@ -95,7 +86,7 @@ static void versal_create_apu_gic(Versal *s, qemu_irq *pic)
     }
 
     for (i = 0; i < nr_apu_cpus; i++) {
-        DeviceState *cpudev = DEVICE(s->fpd.apu.cpu[i]);
+        DeviceState *cpudev = DEVICE(&s->fpd.apu.cpu[i]);
         int ppibase = XLNX_VERSAL_NR_IRQS + i * GIC_INTERNAL + GIC_NR_SGIS;
         qemu_irq maint_irq;
         int ti;
-- 
2.20.1

[PULL 16/39] hw/arm: versal: Add support for SD

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Add support for SD.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-9-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h | 12 ++++++++++++
 hw/arm/xlnx-versal.c         | 31 +++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index 426b66449de..e11693e29d3 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -14,6 +14,7 @@
 
 #include "hw/sysbus.h"
 #include "hw/arm/boot.h"
+#include "hw/sd/sdhci.h"
 #include "hw/intc/arm_gicv3.h"
 #include "hw/char/pl011.h"
 #include "hw/dma/xlnx-zdma.h"
@@ -26,6 +27,7 @@
 #define XLNX_VERSAL_NR_UARTS   2
 #define XLNX_VERSAL_NR_GEMS    2
 #define XLNX_VERSAL_NR_ADMAS   8
+#define XLNX_VERSAL_NR_SDS     2
 #define XLNX_VERSAL_NR_IRQS    192
 
 typedef struct Versal {
@@ -58,6 +60,13 @@ typedef struct Versal {
         } iou;
     } lpd;
 
+    /* The Platform Management Controller subsystem.  */
+    struct {
+        struct {
+            SDHCIState sd[XLNX_VERSAL_NR_SDS];
+        } iou;
+    } pmc;
+
     struct {
         MemoryRegion *mr_ddr;
         uint32_t psci_conduit;
@@ -80,6 +89,7 @@ typedef struct Versal {
 #define VERSAL_GEM1_IRQ_0          58
 #define VERSAL_GEM1_WAKE_IRQ_0     59
 #define VERSAL_ADMA_IRQ_0          60
+#define VERSAL_SD0_IRQ_0           126
 
 /* Architecturally reserved IRQs suitable for virtualization.  */
 #define VERSAL_RSVD_IRQ_FIRST 111
@@ -129,6 +139,8 @@ typedef struct Versal {
 #define MM_FPD_CRF                  0xfd1a0000U
 #define MM_FPD_CRF_SIZE             0x140000
 
+#define MM_PMC_SD0                  0xf1040000U
+#define MM_PMC_SD0_SIZE             0x10000
 #define MM_PMC_CRP                  0xf1260000U
 #define MM_PMC_CRP_SIZE             0x10000
 #endif
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index c8a296e2e05..e263bdf77a9 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -210,6 +210,36 @@ static void versal_create_admas(Versal *s, qemu_irq *pic)
     }
 }
 
+#define SDHCI_CAPABILITIES  0x280737ec6481 /* Same as on ZynqMP.  */
+static void versal_create_sds(Versal *s, qemu_irq *pic)
+{
+    int i;
+
+    for (i = 0; i < ARRAY_SIZE(s->pmc.iou.sd); i++) {
+        DeviceState *dev;
+        MemoryRegion *mr;
+
+        sysbus_init_child_obj(OBJECT(s), "sd[*]",
+                              &s->pmc.iou.sd[i], sizeof(s->pmc.iou.sd[i]),
+                              TYPE_SYSBUS_SDHCI);
+        dev = DEVICE(&s->pmc.iou.sd[i]);
+
+        object_property_set_uint(OBJECT(dev),
+                                 3, "sd-spec-version", &error_fatal);
+        object_property_set_uint(OBJECT(dev), SDHCI_CAPABILITIES, "capareg",
+                                 &error_fatal);
+        object_property_set_uint(OBJECT(dev), UHS_I, "uhs", &error_fatal);
+        qdev_init_nofail(dev);
+
+        mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
+        memory_region_add_subregion(&s->mr_ps,
+                                    MM_PMC_SD0 + i * MM_PMC_SD0_SIZE, mr);
+
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0,
+                           pic[VERSAL_SD0_IRQ_0 + i * 2]);
+    }
+}
+
 /* This takes the board allocated linear DDR memory and creates aliases
  * for each split DDR range/aperture on the Versal address map.
  */
@@ -292,6 +322,7 @@ static void versal_realize(DeviceState *dev, Error **errp)
     versal_create_uarts(s, pic);
     versal_create_gems(s, pic);
     versal_create_admas(s, pic);
+    versal_create_sds(s, pic);
     versal_map_ddr(s);
     versal_unimp(s);
 
-- 
2.20.1

[PULL 17/39] hw/arm: versal: Add support for the RTC

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

hw/arm: versal: Add support for the RTC.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-10-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 include/hw/arm/xlnx-versal.h |  8 ++++++++
 hw/arm/xlnx-versal.c         | 21 +++++++++++++++++++++
 2 files changed, 29 insertions(+)

diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h
index e11693e29d3..9c9f47ba9da 100644
--- a/include/hw/arm/xlnx-versal.h
+++ b/include/hw/arm/xlnx-versal.h
@@ -19,6 +19,7 @@
 #include "hw/char/pl011.h"
 #include "hw/dma/xlnx-zdma.h"
 #include "hw/net/cadence_gem.h"
+#include "hw/rtc/xlnx-zynqmp-rtc.h"
 
 #define TYPE_XLNX_VERSAL "xlnx-versal"
 #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL)
@@ -65,6 +66,8 @@ typedef struct Versal {
         struct {
             SDHCIState sd[XLNX_VERSAL_NR_SDS];
         } iou;
+
+        XlnxZynqMPRTC rtc;
     } pmc;
 
     struct {
@@ -89,7 +92,10 @@ typedef struct Versal {
 #define VERSAL_GEM1_IRQ_0          58
 #define VERSAL_GEM1_WAKE_IRQ_0     59
 #define VERSAL_ADMA_IRQ_0          60
+#define VERSAL_RTC_APB_ERR_IRQ     121
 #define VERSAL_SD0_IRQ_0           126
+#define VERSAL_RTC_ALARM_IRQ       142
+#define VERSAL_RTC_SECONDS_IRQ     143
 
 /* Architecturally reserved IRQs suitable for virtualization.  */
 #define VERSAL_RSVD_IRQ_FIRST 111
@@ -143,4 +149,6 @@ typedef struct Versal {
 #define MM_PMC_SD0_SIZE             0x10000
 #define MM_PMC_CRP                  0xf1260000U
 #define MM_PMC_CRP_SIZE             0x10000
+#define MM_PMC_RTC                  0xf12a0000
+#define MM_PMC_RTC_SIZE             0x10000
 #endif
diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c
index e263bdf77a9..321171bccef 100644
--- a/hw/arm/xlnx-versal.c
+++ b/hw/arm/xlnx-versal.c
@@ -240,6 +240,26 @@ static void versal_create_sds(Versal *s, qemu_irq *pic)
     }
 }
 
+static void versal_create_rtc(Versal *s, qemu_irq *pic)
+{
+    SysBusDevice *sbd;
+    MemoryRegion *mr;
+
+    sysbus_init_child_obj(OBJECT(s), "rtc", &s->pmc.rtc, sizeof(s->pmc.rtc),
+                          TYPE_XLNX_ZYNQMP_RTC);
+    sbd = SYS_BUS_DEVICE(&s->pmc.rtc);
+    qdev_init_nofail(DEVICE(sbd));
+
+    mr = sysbus_mmio_get_region(sbd, 0);
+    memory_region_add_subregion(&s->mr_ps, MM_PMC_RTC, mr);
+
+    /*
+     * TODO: Connect the ALARM and SECONDS interrupts once our RTC model
+     * supports them.
+     */
+    sysbus_connect_irq(sbd, 1, pic[VERSAL_RTC_APB_ERR_IRQ]);
+}
+
 /* This takes the board allocated linear DDR memory and creates aliases
  * for each split DDR range/aperture on the Versal address map.
  */
@@ -323,6 +343,7 @@ static void versal_realize(DeviceState *dev, Error **errp)
     versal_create_gems(s, pic);
     versal_create_admas(s, pic);
     versal_create_sds(s, pic);
+    versal_create_rtc(s, pic);
     versal_map_ddr(s);
     versal_unimp(s);
 
-- 
2.20.1

[PULL 18/39] hw/arm: versal-virt: Add support for SD

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Add support for SD.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-11-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/xlnx-versal-virt.c | 46 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c
index d7be1ad4942..0afee48672a 100644
--- a/hw/arm/xlnx-versal-virt.c
+++ b/hw/arm/xlnx-versal-virt.c
@@ -20,6 +20,7 @@
 #include "hw/arm/sysbus-fdt.h"
 #include "hw/arm/fdt.h"
 #include "cpu.h"
+#include "hw/qdev-properties.h"
 #include "hw/arm/xlnx-versal.h"
 
 #define TYPE_XLNX_VERSAL_VIRT_MACHINE MACHINE_TYPE_NAME("xlnx-versal-virt")
@@ -256,6 +257,32 @@ static void fdt_add_zdma_nodes(VersalVirt *s)
     }
 }
 
+static void fdt_add_sd_nodes(VersalVirt *s)
+{
+    const char clocknames[] = "clk_xin\0clk_ahb";
+    const char compat[] = "arasan,sdhci-8.9a";
+    int i;
+
+    for (i = ARRAY_SIZE(s->soc.pmc.iou.sd) - 1; i >= 0; i--) {
+        uint64_t addr = MM_PMC_SD0 + MM_PMC_SD0_SIZE * i;
+        char *name = g_strdup_printf("/sdhci@%" PRIx64, addr);
+
+        qemu_fdt_add_subnode(s->fdt, name);
+
+        qemu_fdt_setprop_cells(s->fdt, name, "clocks",
+                               s->phandle.clk_25Mhz, s->phandle.clk_25Mhz);
+        qemu_fdt_setprop(s->fdt, name, "clock-names",
+                         clocknames, sizeof(clocknames));
+        qemu_fdt_setprop_cells(s->fdt, name, "interrupts",
+                               GIC_FDT_IRQ_TYPE_SPI, VERSAL_SD0_IRQ_0 + i * 2,
+                               GIC_FDT_IRQ_FLAGS_LEVEL_HI);
+        qemu_fdt_setprop_sized_cells(s->fdt, name, "reg",
+                                     2, addr, 2, MM_PMC_SD0_SIZE);
+        qemu_fdt_setprop(s->fdt, name, "compatible", compat, sizeof(compat));
+        g_free(name);
+    }
+}
+
 static void fdt_nop_memory_nodes(void *fdt, Error **errp)
 {
     Error *err = NULL;
@@ -411,10 +438,23 @@ static void create_virtio_regions(VersalVirt *s)
     }
 }
 
+static void sd_plugin_card(SDHCIState *sd, DriveInfo *di)
+{
+    BlockBackend *blk = di ? blk_by_legacy_dinfo(di) : NULL;
+    DeviceState *card;
+
+    card = qdev_create(qdev_get_child_bus(DEVICE(sd), "sd-bus"), TYPE_SD_CARD);
+    object_property_add_child(OBJECT(sd), "card[*]", OBJECT(card),
+                              &error_fatal);
+    qdev_prop_set_drive(card, "drive", blk, &error_fatal);
+    object_property_set_bool(OBJECT(card), true, "realized", &error_fatal);
+}
+
 static void versal_virt_init(MachineState *machine)
 {
     VersalVirt *s = XLNX_VERSAL_VIRT_MACHINE(machine);
     int psci_conduit = QEMU_PSCI_CONDUIT_DISABLED;
+    int i;
 
     /*
      * If the user provides an Operating System to be loaded, we expect them
@@ -455,6 +495,7 @@ static void versal_virt_init(MachineState *machine)
     fdt_add_gic_nodes(s);
     fdt_add_timer_nodes(s);
     fdt_add_zdma_nodes(s);
+    fdt_add_sd_nodes(s);
     fdt_add_cpu_nodes(s, psci_conduit);
     fdt_add_clk_node(s, "/clk125", 125000000, s->phandle.clk_125Mhz);
     fdt_add_clk_node(s, "/clk25", 25000000, s->phandle.clk_25Mhz);
@@ -464,6 +505,11 @@ static void versal_virt_init(MachineState *machine)
     memory_region_add_subregion_overlap(get_system_memory(),
                                         0, &s->soc.fpd.apu.mr, 0);
 
+    /* Plugin SD cards.  */
+    for (i = 0; i < ARRAY_SIZE(s->soc.pmc.iou.sd); i++) {
+        sd_plugin_card(&s->soc.pmc.iou.sd[i], drive_get_next(IF_SD));
+    }
+
     s->binfo.ram_size = machine->ram_size;
     s->binfo.loader_start = 0x0;
     s->binfo.get_dtb = versal_virt_get_dtb;
-- 
2.20.1

[PULL 19/39] hw/arm: versal-virt: Add support for the RTC

[Peter Maydell]

From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com>

Add support for the RTC.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Message-id: 20200427181649.26851-12-edgar.iglesias@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/xlnx-versal-virt.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c
index 0afee48672a..7e749e19267 100644
--- a/hw/arm/xlnx-versal-virt.c
+++ b/hw/arm/xlnx-versal-virt.c
@@ -283,6 +283,27 @@ static void fdt_add_sd_nodes(VersalVirt *s)
     }
 }
 
+static void fdt_add_rtc_node(VersalVirt *s)
+{
+    const char compat[] = "xlnx,zynqmp-rtc";
+    const char interrupt_names[] = "alarm\0sec";
+    char *name = g_strdup_printf("/rtc@%x", MM_PMC_RTC);
+
+    qemu_fdt_add_subnode(s->fdt, name);
+
+    qemu_fdt_setprop_cells(s->fdt, name, "interrupts",
+                           GIC_FDT_IRQ_TYPE_SPI, VERSAL_RTC_ALARM_IRQ,
+                           GIC_FDT_IRQ_FLAGS_LEVEL_HI,
+                           GIC_FDT_IRQ_TYPE_SPI, VERSAL_RTC_SECONDS_IRQ,
+                           GIC_FDT_IRQ_FLAGS_LEVEL_HI);
+    qemu_fdt_setprop(s->fdt, name, "interrupt-names",
+                     interrupt_names, sizeof(interrupt_names));
+    qemu_fdt_setprop_sized_cells(s->fdt, name, "reg",
+                                 2, MM_PMC_RTC, 2, MM_PMC_RTC_SIZE);
+    qemu_fdt_setprop(s->fdt, name, "compatible", compat, sizeof(compat));
+    g_free(name);
+}
+
 static void fdt_nop_memory_nodes(void *fdt, Error **errp)
 {
     Error *err = NULL;
@@ -496,6 +517,7 @@ static void versal_virt_init(MachineState *machine)
     fdt_add_timer_nodes(s);
     fdt_add_zdma_nodes(s);
     fdt_add_sd_nodes(s);
+    fdt_add_rtc_node(s);
     fdt_add_cpu_nodes(s, psci_conduit);
     fdt_add_clk_node(s, "/clk125", 125000000, s->phandle.clk_125Mhz);
     fdt_add_clk_node(s, "/clk25", 25000000, s->phandle.clk_25Mhz);
-- 
2.20.1

[PULL 20/39] target/arm/translate-vfp.inc.c: Remove duplicate simd_r32 check

[Peter Maydell]

Somewhere along theline we accidentally added a duplicate
"using D16-D31 when they don't exist" check to do_vfm_dp()
(probably an artifact of a patchseries rebase). Remove it.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200430181003.21682-2-peter.maydell@linaro.org
---
 target/arm/translate-vfp.inc.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/target/arm/translate-vfp.inc.c b/target/arm/translate-vfp.inc.c
index b087bbd812e..e1a90175983 100644
--- a/target/arm/translate-vfp.inc.c
+++ b/target/arm/translate-vfp.inc.c
@@ -1872,12 +1872,6 @@ static bool do_vfm_dp(DisasContext *s, arg_VFMA_dp *a, bool neg_n, bool neg_d)
         return false;
     }
 
-    /* UNDEF accesses to D16-D31 if they don't exist. */
-    if (!dc_isar_feature(aa32_simd_r32, s) &&
-        ((a->vd | a->vn | a->vm) & 0x10)) {
-        return false;
-    }
-
     if (!vfp_access_check(s)) {
         return true;
     }
-- 
2.20.1

[PULL 21/39] target/arm: Don't allow Thumb Neon insns without FEATURE_NEON

[Peter Maydell]

We were accidentally permitting decode of Thumb Neon insns even if
the CPU didn't have the FEATURE_NEON bit set, because the feature
check was being done before the call to disas_neon_data_insn() and
disas_neon_ls_insn() in the Arm decoder but was omitted from the
Thumb decoder.  Push the feature bit check down into the called
functions so it is done for both Arm and Thumb encodings.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200430181003.21682-3-peter.maydell@linaro.org
---
 target/arm/translate.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/target/arm/translate.c b/target/arm/translate.c
index e3fc7924428..4cf5267be00 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -3258,6 +3258,10 @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
     TCGv_i32 tmp2;
     TCGv_i64 tmp64;
 
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return 1;
+    }
+
     /* FIXME: this access check should not take precedence over UNDEF
      * for invalid encodings; we will generate incorrect syndrome information
      * for attempts to execute invalid vfp/neon encodings with FP disabled.
@@ -5002,6 +5006,10 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
     TCGv_ptr ptr1, ptr2, ptr3;
     TCGv_i64 tmp64;
 
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return 1;
+    }
+
     /* FIXME: this access check should not take precedence over UNDEF
      * for invalid encodings; we will generate incorrect syndrome information
      * for attempts to execute invalid vfp/neon encodings with FP disabled.
@@ -10948,10 +10956,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
 
         if (((insn >> 25) & 7) == 1) {
             /* NEON Data processing.  */
-            if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
-                goto illegal_op;
-            }
-
             if (disas_neon_data_insn(s, insn)) {
                 goto illegal_op;
             }
@@ -10959,10 +10963,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
         }
         if ((insn & 0x0f100000) == 0x04000000) {
             /* NEON load/store.  */
-            if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
-                goto illegal_op;
-            }
-
             if (disas_neon_ls_insn(s, insn)) {
                 goto illegal_op;
             }
-- 
2.20.1

[PULL 22/39] target/arm: Add stubs for AArch32 Neon decodetree

[Peter Maydell]

Add the infrastructure for building and invoking a decodetree decoder
for the AArch32 Neon encodings.  At the moment the new decoder covers
nothing, so we always fall back to the existing hand-written decode.

We follow the same pattern we did for the VFP decodetree conversion
(commit 78e138bc1f672c145ef6ace74617d and following): code that deals
with Neon will be moving gradually out to translate-neon.vfp.inc,
which we #include into translate.c.

In order to share the decode files between A32 and T32, we
split Neon into 3 parts:
 * data-processing
 * load-store
 * 'shared' encodings

The first two groups of instructions have similar but not identical
A32 and T32 encodings, so we need to manually transform the T32
encoding into the A32 one before calling the decoder; the third group
covers the Neon instructions which are identical in A32 and T32.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-4-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       | 29 ++++++++++++++++++++++++++
 target/arm/neon-ls.decode       | 29 ++++++++++++++++++++++++++
 target/arm/neon-shared.decode   | 27 +++++++++++++++++++++++++
 target/arm/translate-neon.inc.c | 32 +++++++++++++++++++++++++++++
 target/arm/translate.c          | 36 +++++++++++++++++++++++++++++++--
 target/arm/Makefile.objs        | 18 +++++++++++++++++
 6 files changed, 169 insertions(+), 2 deletions(-)
 create mode 100644 target/arm/neon-dp.decode
 create mode 100644 target/arm/neon-ls.decode
 create mode 100644 target/arm/neon-shared.decode
 create mode 100644 target/arm/translate-neon.inc.c

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
new file mode 100644
index 00000000000..c89a1a58591
--- /dev/null
+++ b/target/arm/neon-dp.decode
@@ -0,0 +1,29 @@
+# AArch32 Neon data-processing instruction descriptions
+#
+#  Copyright (c) 2020 Linaro, Ltd
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2 of the License, or (at your option) any later version.
+#
+# This library 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
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, see <http://www.gnu.org/licenses/>.
+
+#
+# This file is processed by scripts/decodetree.py
+#
+
+# Encodings for Neon data processing instructions where the T32 encoding
+# is a simple transformation of the A32 encoding.
+# More specifically, this file covers instructions where the A32 encoding is
+#   0b1111_001p_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq
+# and the T32 encoding is
+#   0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq
+# This file works on the A32 encoding only; calling code for T32 has to
+# transform the insn into the A32 version first.
diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode
new file mode 100644
index 00000000000..2b16c9256df
--- /dev/null
+++ b/target/arm/neon-ls.decode
@@ -0,0 +1,29 @@
+# AArch32 Neon load/store instruction descriptions
+#
+#  Copyright (c) 2020 Linaro, Ltd
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2 of the License, or (at your option) any later version.
+#
+# This library 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
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, see <http://www.gnu.org/licenses/>.
+
+#
+# This file is processed by scripts/decodetree.py
+#
+
+# Encodings for Neon load/store instructions where the T32 encoding
+# is a simple transformation of the A32 encoding.
+# More specifically, this file covers instructions where the A32 encoding is
+#   0b1111_0100_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx
+# and the T32 encoding is
+#   0b1111_1001_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx
+# This file works on the A32 encoding only; calling code for T32 has to
+# transform the insn into the A32 version first.
diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
new file mode 100644
index 00000000000..3aea7c5e188
--- /dev/null
+++ b/target/arm/neon-shared.decode
@@ -0,0 +1,27 @@
+# AArch32 Neon instruction descriptions
+#
+#  Copyright (c) 2020 Linaro, Ltd
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2 of the License, or (at your option) any later version.
+#
+# This library 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
+# Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public
+# License along with this library; if not, see <http://www.gnu.org/licenses/>.
+
+#
+# This file is processed by scripts/decodetree.py
+#
+
+# Encodings for Neon instructions whose encoding is the same for
+# both A32 and T32.
+
+# More specifically, this covers:
+# 2reg scalar ext: 0b1111_1110_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx
+# 3same ext:       0b1111_110x_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
new file mode 100644
index 00000000000..a33e81ba3ab
--- /dev/null
+++ b/target/arm/translate-neon.inc.c
@@ -0,0 +1,32 @@
+/*
+ *  ARM translation: AArch32 Neon instructions
+ *
+ *  Copyright (c) 2003 Fabrice Bellard
+ *  Copyright (c) 2005-2007 CodeSourcery
+ *  Copyright (c) 2007 OpenedHand, Ltd.
+ *  Copyright (c) 2020 Linaro, Ltd.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * This file is intended to be included from translate.c; it uses
+ * some macros and definitions provided by that file.
+ * It might be possible to convert it to a standalone .c file eventually.
+ */
+
+/* Include the generated Neon decoder */
+#include "decode-neon-dp.inc.c"
+#include "decode-neon-ls.inc.c"
+#include "decode-neon-shared.inc.c"
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 4cf5267be00..5a82a56e8ed 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -1313,8 +1313,9 @@ static TCGv_ptr vfp_reg_ptr(bool dp, int reg)
 
 #define ARM_CP_RW_BIT   (1 << 20)
 
-/* Include the VFP decoder */
+/* Include the VFP and Neon decoders */
 #include "translate-vfp.inc.c"
+#include "translate-neon.inc.c"
 
 static inline void iwmmxt_load_reg(TCGv_i64 var, int reg)
 {
@@ -10949,7 +10950,10 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
         /* Unconditional instructions.  */
         /* TODO: Perhaps merge these into one decodetree output file.  */
         if (disas_a32_uncond(s, insn) ||
-            disas_vfp_uncond(s, insn)) {
+            disas_vfp_uncond(s, insn) ||
+            disas_neon_dp(s, insn) ||
+            disas_neon_ls(s, insn) ||
+            disas_neon_shared(s, insn)) {
             return;
         }
         /* fall back to legacy decoder */
@@ -11102,6 +11106,33 @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn)
         ARCH(6T2);
     }
 
+    if ((insn & 0xef000000) == 0xef000000) {
+        /*
+         * T32 encodings 0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq
+         * transform into
+         * A32 encodings 0b1111_001p_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq
+         */
+        uint32_t a32_insn = (insn & 0xe2ffffff) |
+            ((insn & (1 << 28)) >> 4) | (1 << 28);
+
+        if (disas_neon_dp(s, a32_insn)) {
+            return;
+        }
+    }
+
+    if ((insn & 0xff100000) == 0xf9000000) {
+        /*
+         * T32 encodings 0b1111_1001_ppp0_qqqq_qqqq_qqqq_qqqq_qqqq
+         * transform into
+         * A32 encodings 0b1111_0100_ppp0_qqqq_qqqq_qqqq_qqqq_qqqq
+         */
+        uint32_t a32_insn = (insn & 0x00ffffff) | 0xf4000000;
+
+        if (disas_neon_ls(s, a32_insn)) {
+            return;
+        }
+    }
+
     /*
      * TODO: Perhaps merge these into one decodetree output file.
      * Note disas_vfp is written for a32 with cond field in the
@@ -11109,6 +11140,7 @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn)
      */
     if (disas_t32(s, insn) ||
         disas_vfp_uncond(s, insn) ||
+        disas_neon_shared(s, insn) ||
         ((insn >> 28) == 0xe && disas_vfp(s, insn))) {
         return;
     }
diff --git a/target/arm/Makefile.objs b/target/arm/Makefile.objs
index cf26c16f5f6..775b3e24f22 100644
--- a/target/arm/Makefile.objs
+++ b/target/arm/Makefile.objs
@@ -18,6 +18,21 @@ target/arm/decode-sve.inc.c: $(SRC_PATH)/target/arm/sve.decode $(DECODETREE)
 	  $(PYTHON) $(DECODETREE) --decode disas_sve -o $@ $<,\
 	  "GEN", $(TARGET_DIR)$@)
 
+target/arm/decode-neon-shared.inc.c: $(SRC_PATH)/target/arm/neon-shared.decode $(DECODETREE)
+	$(call quiet-command,\
+	  $(PYTHON) $(DECODETREE) --static-decode disas_neon_shared -o $@ $<,\
+	  "GEN", $(TARGET_DIR)$@)
+
+target/arm/decode-neon-dp.inc.c: $(SRC_PATH)/target/arm/neon-dp.decode $(DECODETREE)
+	$(call quiet-command,\
+	  $(PYTHON) $(DECODETREE) --static-decode disas_neon_dp -o $@ $<,\
+	  "GEN", $(TARGET_DIR)$@)
+
+target/arm/decode-neon-ls.inc.c: $(SRC_PATH)/target/arm/neon-ls.decode $(DECODETREE)
+	$(call quiet-command,\
+	  $(PYTHON) $(DECODETREE) --static-decode disas_neon_ls -o $@ $<,\
+	  "GEN", $(TARGET_DIR)$@)
+
 target/arm/decode-vfp.inc.c: $(SRC_PATH)/target/arm/vfp.decode $(DECODETREE)
 	$(call quiet-command,\
 	  $(PYTHON) $(DECODETREE) --static-decode disas_vfp -o $@ $<,\
@@ -49,6 +64,9 @@ target/arm/decode-t16.inc.c: $(SRC_PATH)/target/arm/t16.decode $(DECODETREE)
 	  "GEN", $(TARGET_DIR)$@)
 
 target/arm/translate-sve.o: target/arm/decode-sve.inc.c
+target/arm/translate.o: target/arm/decode-neon-shared.inc.c
+target/arm/translate.o: target/arm/decode-neon-dp.inc.c
+target/arm/translate.o: target/arm/decode-neon-ls.inc.c
 target/arm/translate.o: target/arm/decode-vfp.inc.c
 target/arm/translate.o: target/arm/decode-vfp-uncond.inc.c
 target/arm/translate.o: target/arm/decode-a32.inc.c
-- 
2.20.1

[PULL 23/39] target/arm: Convert VCMLA (vector) to decodetree

[Peter Maydell]

Convert the VCMLA (vector) insns in the 3same extension group to
decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-5-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   | 11 ++++++++++
 target/arm/translate-neon.inc.c | 37 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 11 +---------
 3 files changed, 49 insertions(+), 10 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index 3aea7c5e188..d1d707a56d5 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -25,3 +25,14 @@
 # More specifically, this covers:
 # 2reg scalar ext: 0b1111_1110_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx
 # 3same ext:       0b1111_110x_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx
+
+# VFP/Neon register fields; same as vfp.decode
+%vm_dp  5:1 0:4
+%vm_sp  0:4 5:1
+%vn_dp  7:1 16:4
+%vn_sp  16:4 7:1
+%vd_dp  22:1 12:4
+%vd_sp  12:4 22:1
+
+VCMLA          1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index a33e81ba3ab..0baae1338a3 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -30,3 +30,40 @@
 #include "decode-neon-dp.inc.c"
 #include "decode-neon-ls.inc.c"
 #include "decode-neon-shared.inc.c"
+
+static bool trans_VCMLA(DisasContext *s, arg_VCMLA *a)
+{
+    int opr_sz;
+    TCGv_ptr fpst;
+    gen_helper_gvec_3_ptr *fn_gvec_ptr;
+
+    if (!dc_isar_feature(aa32_vcma, s)
+        || (!a->size && !dc_isar_feature(aa32_fp16_arith, s))) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vn | a->vm | a->vd) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    opr_sz = (1 + a->q) * 8;
+    fpst = get_fpstatus_ptr(1);
+    fn_gvec_ptr = a->size ? gen_helper_gvec_fcmlas : gen_helper_gvec_fcmlah;
+    tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(1, a->vn),
+                       vfp_reg_offset(1, a->vm),
+                       fpst, opr_sz, opr_sz, a->rot,
+                       fn_gvec_ptr);
+    tcg_temp_free_ptr(fpst);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 5a82a56e8ed..ae6799c6ae0 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7048,16 +7048,7 @@ static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn)
     bool is_long = false, q = extract32(insn, 6, 1);
     bool ptr_is_env = false;
 
-    if ((insn & 0xfe200f10) == 0xfc200800) {
-        /* VCMLA -- 1111 110R R.1S .... .... 1000 ...0 .... */
-        int size = extract32(insn, 20, 1);
-        data = extract32(insn, 23, 2); /* rot */
-        if (!dc_isar_feature(aa32_vcma, s)
-            || (!size && !dc_isar_feature(aa32_fp16_arith, s))) {
-            return 1;
-        }
-        fn_gvec_ptr = size ? gen_helper_gvec_fcmlas : gen_helper_gvec_fcmlah;
-    } else if ((insn & 0xfea00f10) == 0xfc800800) {
+    if ((insn & 0xfea00f10) == 0xfc800800) {
         /* VCADD -- 1111 110R 1.0S .... .... 1000 ...0 .... */
         int size = extract32(insn, 20, 1);
         data = extract32(insn, 24, 1); /* rot */
-- 
2.20.1

[PULL 24/39] target/arm: Convert VCADD (vector) to decodetree

[Peter Maydell]

Convert the VCADD (vector) insns to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-6-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |  3 +++
 target/arm/translate-neon.inc.c | 37 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 11 +---------
 3 files changed, 41 insertions(+), 10 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index d1d707a56d5..ed65dae1809 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -36,3 +36,6 @@
 
 VCMLA          1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp
+
+VCADD          1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 0baae1338a3..28011e88d9e 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -67,3 +67,40 @@ static bool trans_VCMLA(DisasContext *s, arg_VCMLA *a)
     tcg_temp_free_ptr(fpst);
     return true;
 }
+
+static bool trans_VCADD(DisasContext *s, arg_VCADD *a)
+{
+    int opr_sz;
+    TCGv_ptr fpst;
+    gen_helper_gvec_3_ptr *fn_gvec_ptr;
+
+    if (!dc_isar_feature(aa32_vcma, s)
+        || (!a->size && !dc_isar_feature(aa32_fp16_arith, s))) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vn | a->vm | a->vd) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    opr_sz = (1 + a->q) * 8;
+    fpst = get_fpstatus_ptr(1);
+    fn_gvec_ptr = a->size ? gen_helper_gvec_fcadds : gen_helper_gvec_fcaddh;
+    tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(1, a->vn),
+                       vfp_reg_offset(1, a->vm),
+                       fpst, opr_sz, opr_sz, a->rot,
+                       fn_gvec_ptr);
+    tcg_temp_free_ptr(fpst);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index ae6799c6ae0..993bead82ff 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7048,16 +7048,7 @@ static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn)
     bool is_long = false, q = extract32(insn, 6, 1);
     bool ptr_is_env = false;
 
-    if ((insn & 0xfea00f10) == 0xfc800800) {
-        /* VCADD -- 1111 110R 1.0S .... .... 1000 ...0 .... */
-        int size = extract32(insn, 20, 1);
-        data = extract32(insn, 24, 1); /* rot */
-        if (!dc_isar_feature(aa32_vcma, s)
-            || (!size && !dc_isar_feature(aa32_fp16_arith, s))) {
-            return 1;
-        }
-        fn_gvec_ptr = size ? gen_helper_gvec_fcadds : gen_helper_gvec_fcaddh;
-    } else if ((insn & 0xfeb00f00) == 0xfc200d00) {
+    if ((insn & 0xfeb00f00) == 0xfc200d00) {
         /* V[US]DOT -- 1111 1100 0.10 .... .... 1101 .Q.U .... */
         bool u = extract32(insn, 4, 1);
         if (!dc_isar_feature(aa32_dp, s)) {
-- 
2.20.1

[PULL 25/39] target/arm: Convert V[US]DOT (vector) to decodetree

[Peter Maydell]

Convert the V[US]DOT (vector) insns to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-7-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |  4 ++++
 target/arm/translate-neon.inc.c | 32 ++++++++++++++++++++++++++++++++
 target/arm/translate.c          |  9 +--------
 3 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index ed65dae1809..c9c641905d3 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -39,3 +39,7 @@ VCMLA          1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \
 
 VCADD          1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp
+
+# VUDOT and VSDOT
+VDOT           1111 110 00 . 10 .... .... 1101 . q:1 . u:1 .... \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 28011e88d9e..6537506c5b6 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -104,3 +104,35 @@ static bool trans_VCADD(DisasContext *s, arg_VCADD *a)
     tcg_temp_free_ptr(fpst);
     return true;
 }
+
+static bool trans_VDOT(DisasContext *s, arg_VDOT *a)
+{
+    int opr_sz;
+    gen_helper_gvec_3 *fn_gvec;
+
+    if (!dc_isar_feature(aa32_dp, s)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vn | a->vm | a->vd) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    opr_sz = (1 + a->q) * 8;
+    fn_gvec = a->u ? gen_helper_gvec_udot_b : gen_helper_gvec_sdot_b;
+    tcg_gen_gvec_3_ool(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(1, a->vn),
+                       vfp_reg_offset(1, a->vm),
+                       opr_sz, opr_sz, 0, fn_gvec);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 993bead82ff..7d3aea8c98a 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7048,14 +7048,7 @@ static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn)
     bool is_long = false, q = extract32(insn, 6, 1);
     bool ptr_is_env = false;
 
-    if ((insn & 0xfeb00f00) == 0xfc200d00) {
-        /* V[US]DOT -- 1111 1100 0.10 .... .... 1101 .Q.U .... */
-        bool u = extract32(insn, 4, 1);
-        if (!dc_isar_feature(aa32_dp, s)) {
-            return 1;
-        }
-        fn_gvec = u ? gen_helper_gvec_udot_b : gen_helper_gvec_sdot_b;
-    } else if ((insn & 0xff300f10) == 0xfc200810) {
+    if ((insn & 0xff300f10) == 0xfc200810) {
         /* VFM[AS]L -- 1111 1100 S.10 .... .... 1000 .Q.1 .... */
         int is_s = extract32(insn, 23, 1);
         if (!dc_isar_feature(aa32_fhm, s)) {
-- 
2.20.1

[PULL 26/39] target/arm: Convert VFM[AS]L (vector) to decodetree

[Peter Maydell]

Convert the VFM[AS]L (vector) insns to decodetree.  This is the last
insn in the legacy decoder for the 3same_ext group, so we can
delete the legacy decoder function for the group entirely.

Note that in disas_thumb2_insn() the parts of this encoding space
where the decodetree decoder returns false will correctly be directed
to illegal_op by the "(insn & (1 << 28))" check so they won't fall
into disas_coproc_insn() by mistake.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-8-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |  6 +++
 target/arm/translate-neon.inc.c | 31 +++++++++++
 target/arm/translate.c          | 92 +--------------------------------
 3 files changed, 38 insertions(+), 91 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index c9c641905d3..90cd5c871e2 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -43,3 +43,9 @@ VCADD          1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \
 # VUDOT and VSDOT
 VDOT           1111 110 00 . 10 .... .... 1101 . q:1 . u:1 .... \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp
+
+# VFM[AS]L
+VFML           1111 110 0 s:1 . 10 .... .... 1000 . 0 . 1 .... \
+               vm=%vm_sp vn=%vn_sp vd=%vd_dp q=0
+VFML           1111 110 0 s:1 . 10 .... .... 1000 . 1 . 1 .... \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp q=1
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 6537506c5b6..6c58abc54b5 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -136,3 +136,34 @@ static bool trans_VDOT(DisasContext *s, arg_VDOT *a)
                        opr_sz, opr_sz, 0, fn_gvec);
     return true;
 }
+
+static bool trans_VFML(DisasContext *s, arg_VFML *a)
+{
+    int opr_sz;
+
+    if (!dc_isar_feature(aa32_fhm, s)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        (a->vd & 0x10)) {
+        return false;
+    }
+
+    if (a->vd & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    opr_sz = (1 + a->q) * 8;
+    tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(a->q, a->vn),
+                       vfp_reg_offset(a->q, a->vm),
+                       cpu_env, opr_sz, opr_sz, a->s, /* is_2 == 0 */
+                       gen_helper_gvec_fmlal_a32);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 7d3aea8c98a..79cd9138fe7 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7032,84 +7032,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
     return 0;
 }
 
-/* Advanced SIMD three registers of the same length extension.
- *  31           25    23  22    20   16   12  11   10   9    8        3     0
- * +---------------+-----+---+-----+----+----+---+----+---+----+---------+----+
- * | 1 1 1 1 1 1 0 | op1 | D | op2 | Vn | Vd | 1 | o3 | 0 | o4 | N Q M U | Vm |
- * +---------------+-----+---+-----+----+----+---+----+---+----+---------+----+
- */
-static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn)
-{
-    gen_helper_gvec_3 *fn_gvec = NULL;
-    gen_helper_gvec_3_ptr *fn_gvec_ptr = NULL;
-    int rd, rn, rm, opr_sz;
-    int data = 0;
-    int off_rn, off_rm;
-    bool is_long = false, q = extract32(insn, 6, 1);
-    bool ptr_is_env = false;
-
-    if ((insn & 0xff300f10) == 0xfc200810) {
-        /* VFM[AS]L -- 1111 1100 S.10 .... .... 1000 .Q.1 .... */
-        int is_s = extract32(insn, 23, 1);
-        if (!dc_isar_feature(aa32_fhm, s)) {
-            return 1;
-        }
-        is_long = true;
-        data = is_s; /* is_2 == 0 */
-        fn_gvec_ptr = gen_helper_gvec_fmlal_a32;
-        ptr_is_env = true;
-    } else {
-        return 1;
-    }
-
-    VFP_DREG_D(rd, insn);
-    if (rd & q) {
-        return 1;
-    }
-    if (q || !is_long) {
-        VFP_DREG_N(rn, insn);
-        VFP_DREG_M(rm, insn);
-        if ((rn | rm) & q & !is_long) {
-            return 1;
-        }
-        off_rn = vfp_reg_offset(1, rn);
-        off_rm = vfp_reg_offset(1, rm);
-    } else {
-        rn = VFP_SREG_N(insn);
-        rm = VFP_SREG_M(insn);
-        off_rn = vfp_reg_offset(0, rn);
-        off_rm = vfp_reg_offset(0, rm);
-    }
-
-    if (s->fp_excp_el) {
-        gen_exception_insn(s, s->pc_curr, EXCP_UDEF,
-                           syn_simd_access_trap(1, 0xe, false), s->fp_excp_el);
-        return 0;
-    }
-    if (!s->vfp_enabled) {
-        return 1;
-    }
-
-    opr_sz = (1 + q) * 8;
-    if (fn_gvec_ptr) {
-        TCGv_ptr ptr;
-        if (ptr_is_env) {
-            ptr = cpu_env;
-        } else {
-            ptr = get_fpstatus_ptr(1);
-        }
-        tcg_gen_gvec_3_ptr(vfp_reg_offset(1, rd), off_rn, off_rm, ptr,
-                           opr_sz, opr_sz, data, fn_gvec_ptr);
-        if (!ptr_is_env) {
-            tcg_temp_free_ptr(ptr);
-        }
-    } else {
-        tcg_gen_gvec_3_ool(vfp_reg_offset(1, rd), off_rn, off_rm,
-                           opr_sz, opr_sz, data, fn_gvec);
-    }
-    return 0;
-}
-
 /* Advanced SIMD two registers and a scalar extension.
  *  31             24   23  22   20   16   12  11   10   9    8        3     0
  * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+
@@ -10956,12 +10878,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
                     }
                 }
             }
-        } else if ((insn & 0x0e000a00) == 0x0c000800
-                   && arm_dc_feature(s, ARM_FEATURE_V8)) {
-            if (disas_neon_insn_3same_ext(s, insn)) {
-                goto illegal_op;
-            }
-            return;
         } else if ((insn & 0x0f000a00) == 0x0e000800
                    && arm_dc_feature(s, ARM_FEATURE_V8)) {
             if (disas_neon_insn_2reg_scalar_ext(s, insn)) {
@@ -11145,15 +11061,9 @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn)
             }
             break;
         }
-        if ((insn & 0xfe000a00) == 0xfc000800
+        if ((insn & 0xff000a00) == 0xfe000800
             && arm_dc_feature(s, ARM_FEATURE_V8)) {
             /* The Thumb2 and ARM encodings are identical.  */
-            if (disas_neon_insn_3same_ext(s, insn)) {
-                goto illegal_op;
-            }
-        } else if ((insn & 0xff000a00) == 0xfe000800
-                   && arm_dc_feature(s, ARM_FEATURE_V8)) {
-            /* The Thumb2 and ARM encodings are identical.  */
             if (disas_neon_insn_2reg_scalar_ext(s, insn)) {
                 goto illegal_op;
             }
-- 
2.20.1

[PULL 27/39] target/arm: Convert VCMLA (scalar) to decodetree

[Peter Maydell]

Convert VCMLA (scalar) in the 2reg-scalar-ext group to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-9-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |  5 +++++
 target/arm/translate-neon.inc.c | 40 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 26 +--------------------
 3 files changed, 46 insertions(+), 25 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index 90cd5c871e2..c11d755ed14 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -49,3 +49,8 @@ VFML           1111 110 0 s:1 . 10 .... .... 1000 . 0 . 1 .... \
                vm=%vm_sp vn=%vn_sp vd=%vd_dp q=0
 VFML           1111 110 0 s:1 . 10 .... .... 1000 . 1 . 1 .... \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp q=1
+
+VCMLA_scalar   1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \
+               vn=%vn_dp vd=%vd_dp size=0
+VCMLA_scalar   1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp size=1 index=0
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 6c58abc54b5..92eccbf8236 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -167,3 +167,43 @@ static bool trans_VFML(DisasContext *s, arg_VFML *a)
                        gen_helper_gvec_fmlal_a32);
     return true;
 }
+
+static bool trans_VCMLA_scalar(DisasContext *s, arg_VCMLA_scalar *a)
+{
+    gen_helper_gvec_3_ptr *fn_gvec_ptr;
+    int opr_sz;
+    TCGv_ptr fpst;
+
+    if (!dc_isar_feature(aa32_vcma, s)) {
+        return false;
+    }
+    if (a->size == 0 && !dc_isar_feature(aa32_fp16_arith, s)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vd | a->vn) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    fn_gvec_ptr = (a->size ? gen_helper_gvec_fcmlas_idx
+                   : gen_helper_gvec_fcmlah_idx);
+    opr_sz = (1 + a->q) * 8;
+    fpst = get_fpstatus_ptr(1);
+    tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(1, a->vn),
+                       vfp_reg_offset(1, a->vm),
+                       fpst, opr_sz, opr_sz,
+                       (a->index << 2) | a->rot, fn_gvec_ptr);
+    tcg_temp_free_ptr(fpst);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 79cd9138fe7..4cb8c6d55b1 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7049,31 +7049,7 @@ static int disas_neon_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn)
     bool is_long = false, q = extract32(insn, 6, 1);
     bool ptr_is_env = false;
 
-    if ((insn & 0xff000f10) == 0xfe000800) {
-        /* VCMLA (indexed) -- 1111 1110 S.RR .... .... 1000 ...0 .... */
-        int rot = extract32(insn, 20, 2);
-        int size = extract32(insn, 23, 1);
-        int index;
-
-        if (!dc_isar_feature(aa32_vcma, s)) {
-            return 1;
-        }
-        if (size == 0) {
-            if (!dc_isar_feature(aa32_fp16_arith, s)) {
-                return 1;
-            }
-            /* For fp16, rm is just Vm, and index is M.  */
-            rm = extract32(insn, 0, 4);
-            index = extract32(insn, 5, 1);
-        } else {
-            /* For fp32, rm is the usual M:Vm, and index is 0.  */
-            VFP_DREG_M(rm, insn);
-            index = 0;
-        }
-        data = (index << 2) | rot;
-        fn_gvec_ptr = (size ? gen_helper_gvec_fcmlas_idx
-                       : gen_helper_gvec_fcmlah_idx);
-    } else if ((insn & 0xffb00f00) == 0xfe200d00) {
+    if ((insn & 0xffb00f00) == 0xfe200d00) {
         /* V[US]DOT -- 1111 1110 0.10 .... .... 1101 .Q.U .... */
         int u = extract32(insn, 4, 1);
 
-- 
2.20.1

[PULL 28/39] target/arm: Convert V[US]DOT (scalar) to decodetree

[Peter Maydell]

Convert the V[US]DOT (scalar) insns in the 2reg-scalar-ext group
to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-10-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |  3 +++
 target/arm/translate-neon.inc.c | 35 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 13 +-----------
 3 files changed, 39 insertions(+), 12 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index c11d755ed14..63a46c63c07 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -54,3 +54,6 @@ VCMLA_scalar   1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \
                vn=%vn_dp vd=%vd_dp size=0
 VCMLA_scalar   1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp size=1 index=0
+
+VDOT_scalar    1111 1110 0 . 10 .... .... 1101 . q:1 index:1 u:1 rm:4 \
+               vm=%vm_dp vn=%vn_dp vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 92eccbf8236..7cc6ccb0697 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -207,3 +207,38 @@ static bool trans_VCMLA_scalar(DisasContext *s, arg_VCMLA_scalar *a)
     tcg_temp_free_ptr(fpst);
     return true;
 }
+
+static bool trans_VDOT_scalar(DisasContext *s, arg_VDOT_scalar *a)
+{
+    gen_helper_gvec_3 *fn_gvec;
+    int opr_sz;
+    TCGv_ptr fpst;
+
+    if (!dc_isar_feature(aa32_dp, s)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vd | a->vn) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    fn_gvec = a->u ? gen_helper_gvec_udot_idx_b : gen_helper_gvec_sdot_idx_b;
+    opr_sz = (1 + a->q) * 8;
+    fpst = get_fpstatus_ptr(1);
+    tcg_gen_gvec_3_ool(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(1, a->vn),
+                       vfp_reg_offset(1, a->rm),
+                       opr_sz, opr_sz, a->index, fn_gvec);
+    tcg_temp_free_ptr(fpst);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 4cb8c6d55b1..8574d0964f7 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -7049,18 +7049,7 @@ static int disas_neon_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn)
     bool is_long = false, q = extract32(insn, 6, 1);
     bool ptr_is_env = false;
 
-    if ((insn & 0xffb00f00) == 0xfe200d00) {
-        /* V[US]DOT -- 1111 1110 0.10 .... .... 1101 .Q.U .... */
-        int u = extract32(insn, 4, 1);
-
-        if (!dc_isar_feature(aa32_dp, s)) {
-            return 1;
-        }
-        fn_gvec = u ? gen_helper_gvec_udot_idx_b : gen_helper_gvec_sdot_idx_b;
-        /* rm is just Vm, and index is M.  */
-        data = extract32(insn, 5, 1); /* index */
-        rm = extract32(insn, 0, 4);
-    } else if ((insn & 0xffa00f10) == 0xfe000810) {
+    if ((insn & 0xffa00f10) == 0xfe000810) {
         /* VFM[AS]L -- 1111 1110 0.0S .... .... 1000 .Q.1 .... */
         int is_s = extract32(insn, 20, 1);
         int vm20 = extract32(insn, 0, 3);
-- 
2.20.1

[PULL 29/39] target/arm: Convert VFM[AS]L (scalar) to decodetree

[Peter Maydell]

Convert the VFM[AS]L (scalar) insns in the 2reg-scalar-ext group
to decodetree. These are the last ones in the group so we can remove
all the legacy decode for the group.

Note that in disas_thumb2_insn() the parts of this encoding space
where the decodetree decoder returns false will correctly be directed
to illegal_op by the "(insn & (1 << 28))" check so they won't fall
into disas_coproc_insn() by mistake.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-11-peter.maydell@linaro.org
---
 target/arm/neon-shared.decode   |   7 +++
 target/arm/translate-neon.inc.c |  32 ++++++++++
 target/arm/translate.c          | 107 +-------------------------------
 3 files changed, 40 insertions(+), 106 deletions(-)

diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode
index 63a46c63c07..f297ba8cdfc 100644
--- a/target/arm/neon-shared.decode
+++ b/target/arm/neon-shared.decode
@@ -57,3 +57,10 @@ VCMLA_scalar   1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \
 
 VDOT_scalar    1111 1110 0 . 10 .... .... 1101 . q:1 index:1 u:1 rm:4 \
                vm=%vm_dp vn=%vn_dp vd=%vd_dp
+
+%vfml_scalar_q0_rm 0:3 5:1
+%vfml_scalar_q1_index 5:1 3:1
+VFML_scalar    1111 1110 0 . 0 s:1 .... .... 1000 . 0 . 1 index:1 ... \
+               rm=%vfml_scalar_q0_rm vn=%vn_sp vd=%vd_dp q=0
+VFML_scalar    1111 1110 0 . 0 s:1 .... .... 1000 . 1 . 1 . rm:3 \
+               index=%vfml_scalar_q1_index vn=%vn_dp vd=%vd_dp q=1
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 7cc6ccb0697..b06542b8b83 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -242,3 +242,35 @@ static bool trans_VDOT_scalar(DisasContext *s, arg_VDOT_scalar *a)
     tcg_temp_free_ptr(fpst);
     return true;
 }
+
+static bool trans_VFML_scalar(DisasContext *s, arg_VFML_scalar *a)
+{
+    int opr_sz;
+
+    if (!dc_isar_feature(aa32_fhm, s)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd & 0x10) || (a->q && (a->vn & 0x10)))) {
+        return false;
+    }
+
+    if (a->vd & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    opr_sz = (1 + a->q) * 8;
+    tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd),
+                       vfp_reg_offset(a->q, a->vn),
+                       vfp_reg_offset(a->q, a->rm),
+                       cpu_env, opr_sz, opr_sz,
+                       (a->index << 2) | a->s, /* is_2 == 0 */
+                       gen_helper_gvec_fmlal_idx_a32);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 8574d0964f7..e269642a480 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -2610,8 +2610,6 @@ static int disas_dsp_insn(DisasContext *s, uint32_t insn)
 }
 
 #define VFP_REG_SHR(x, n) (((n) > 0) ? (x) >> (n) : (x) << -(n))
-#define VFP_SREG(insn, bigbit, smallbit) \
-  ((VFP_REG_SHR(insn, bigbit - 1) & 0x1e) | (((insn) >> (smallbit)) & 1))
 #define VFP_DREG(reg, insn, bigbit, smallbit) do { \
     if (dc_isar_feature(aa32_simd_r32, s)) { \
         reg = (((insn) >> (bigbit)) & 0x0f) \
@@ -2622,11 +2620,8 @@ static int disas_dsp_insn(DisasContext *s, uint32_t insn)
         reg = ((insn) >> (bigbit)) & 0x0f; \
     }} while (0)
 
-#define VFP_SREG_D(insn) VFP_SREG(insn, 12, 22)
 #define VFP_DREG_D(reg, insn) VFP_DREG(reg, insn, 12, 22)
-#define VFP_SREG_N(insn) VFP_SREG(insn, 16,  7)
 #define VFP_DREG_N(reg, insn) VFP_DREG(reg, insn, 16,  7)
-#define VFP_SREG_M(insn) VFP_SREG(insn,  0,  5)
 #define VFP_DREG_M(reg, insn) VFP_DREG(reg, insn,  0,  5)
 
 static void gen_neon_dup_low16(TCGv_i32 var)
@@ -7032,94 +7027,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
     return 0;
 }
 
-/* Advanced SIMD two registers and a scalar extension.
- *  31             24   23  22   20   16   12  11   10   9    8        3     0
- * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+
- * | 1 1 1 1 1 1 1 0 | o1 | D | o2 | Vn | Vd | 1 | o3 | 0 | o4 | N Q M U | Vm |
- * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+
- *
- */
-
-static int disas_neon_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn)
-{
-    gen_helper_gvec_3 *fn_gvec = NULL;
-    gen_helper_gvec_3_ptr *fn_gvec_ptr = NULL;
-    int rd, rn, rm, opr_sz, data;
-    int off_rn, off_rm;
-    bool is_long = false, q = extract32(insn, 6, 1);
-    bool ptr_is_env = false;
-
-    if ((insn & 0xffa00f10) == 0xfe000810) {
-        /* VFM[AS]L -- 1111 1110 0.0S .... .... 1000 .Q.1 .... */
-        int is_s = extract32(insn, 20, 1);
-        int vm20 = extract32(insn, 0, 3);
-        int vm3 = extract32(insn, 3, 1);
-        int m = extract32(insn, 5, 1);
-        int index;
-
-        if (!dc_isar_feature(aa32_fhm, s)) {
-            return 1;
-        }
-        if (q) {
-            rm = vm20;
-            index = m * 2 + vm3;
-        } else {
-            rm = vm20 * 2 + m;
-            index = vm3;
-        }
-        is_long = true;
-        data = (index << 2) | is_s; /* is_2 == 0 */
-        fn_gvec_ptr = gen_helper_gvec_fmlal_idx_a32;
-        ptr_is_env = true;
-    } else {
-        return 1;
-    }
-
-    VFP_DREG_D(rd, insn);
-    if (rd & q) {
-        return 1;
-    }
-    if (q || !is_long) {
-        VFP_DREG_N(rn, insn);
-        if (rn & q & !is_long) {
-            return 1;
-        }
-        off_rn = vfp_reg_offset(1, rn);
-        off_rm = vfp_reg_offset(1, rm);
-    } else {
-        rn = VFP_SREG_N(insn);
-        off_rn = vfp_reg_offset(0, rn);
-        off_rm = vfp_reg_offset(0, rm);
-    }
-    if (s->fp_excp_el) {
-        gen_exception_insn(s, s->pc_curr, EXCP_UDEF,
-                           syn_simd_access_trap(1, 0xe, false), s->fp_excp_el);
-        return 0;
-    }
-    if (!s->vfp_enabled) {
-        return 1;
-    }
-
-    opr_sz = (1 + q) * 8;
-    if (fn_gvec_ptr) {
-        TCGv_ptr ptr;
-        if (ptr_is_env) {
-            ptr = cpu_env;
-        } else {
-            ptr = get_fpstatus_ptr(1);
-        }
-        tcg_gen_gvec_3_ptr(vfp_reg_offset(1, rd), off_rn, off_rm, ptr,
-                           opr_sz, opr_sz, data, fn_gvec_ptr);
-        if (!ptr_is_env) {
-            tcg_temp_free_ptr(ptr);
-        }
-    } else {
-        tcg_gen_gvec_3_ool(vfp_reg_offset(1, rd), off_rn, off_rm,
-                           opr_sz, opr_sz, data, fn_gvec);
-    }
-    return 0;
-}
-
 static int disas_coproc_insn(DisasContext *s, uint32_t insn)
 {
     int cpnum, is64, crn, crm, opc1, opc2, isread, rt, rt2;
@@ -10843,12 +10750,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
                     }
                 }
             }
-        } else if ((insn & 0x0f000a00) == 0x0e000800
-                   && arm_dc_feature(s, ARM_FEATURE_V8)) {
-            if (disas_neon_insn_2reg_scalar_ext(s, insn)) {
-                goto illegal_op;
-            }
-            return;
         }
         goto illegal_op;
     }
@@ -11026,13 +10927,7 @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn)
             }
             break;
         }
-        if ((insn & 0xff000a00) == 0xfe000800
-            && arm_dc_feature(s, ARM_FEATURE_V8)) {
-            /* The Thumb2 and ARM encodings are identical.  */
-            if (disas_neon_insn_2reg_scalar_ext(s, insn)) {
-                goto illegal_op;
-            }
-        } else if (((insn >> 24) & 3) == 3) {
+        if (((insn >> 24) & 3) == 3) {
             /* Translate into the equivalent ARM encoding.  */
             insn = (insn & 0xe2ffffff) | ((insn & (1 << 28)) >> 4) | (1 << 28);
             if (disas_neon_data_insn(s, insn)) {
-- 
2.20.1

[PULL 30/39] target/arm: Convert Neon load/store multiple structures to decodetree

[Peter Maydell]

Convert the Neon "load/store multiple structures" insns to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-12-peter.maydell@linaro.org
---
 target/arm/neon-ls.decode       |   7 ++
 target/arm/translate-neon.inc.c | 124 ++++++++++++++++++++++++++++++++
 target/arm/translate.c          |  91 +----------------------
 3 files changed, 133 insertions(+), 89 deletions(-)

diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode
index 2b16c9256df..dd03d5a37bd 100644
--- a/target/arm/neon-ls.decode
+++ b/target/arm/neon-ls.decode
@@ -27,3 +27,10 @@
 #   0b1111_1001_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx
 # This file works on the A32 encoding only; calling code for T32 has to
 # transform the insn into the A32 version first.
+
+%vd_dp  22:1 12:4
+
+# Neon load/store multiple structures
+
+VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \
+               vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index b06542b8b83..966c0d92012 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -274,3 +274,127 @@ static bool trans_VFML_scalar(DisasContext *s, arg_VFML_scalar *a)
                        gen_helper_gvec_fmlal_idx_a32);
     return true;
 }
+
+static struct {
+    int nregs;
+    int interleave;
+    int spacing;
+} const neon_ls_element_type[11] = {
+    {1, 4, 1},
+    {1, 4, 2},
+    {4, 1, 1},
+    {2, 2, 2},
+    {1, 3, 1},
+    {1, 3, 2},
+    {3, 1, 1},
+    {1, 1, 1},
+    {1, 2, 1},
+    {1, 2, 2},
+    {2, 1, 1}
+};
+
+static void gen_neon_ldst_base_update(DisasContext *s, int rm, int rn,
+                                      int stride)
+{
+    if (rm != 15) {
+        TCGv_i32 base;
+
+        base = load_reg(s, rn);
+        if (rm == 13) {
+            tcg_gen_addi_i32(base, base, stride);
+        } else {
+            TCGv_i32 index;
+            index = load_reg(s, rm);
+            tcg_gen_add_i32(base, base, index);
+            tcg_temp_free_i32(index);
+        }
+        store_reg(s, rn, base);
+    }
+}
+
+static bool trans_VLDST_multiple(DisasContext *s, arg_VLDST_multiple *a)
+{
+    /* Neon load/store multiple structures */
+    int nregs, interleave, spacing, reg, n;
+    MemOp endian = s->be_data;
+    int mmu_idx = get_mem_index(s);
+    int size = a->size;
+    TCGv_i64 tmp64;
+    TCGv_i32 addr, tmp;
+
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist */
+    if (!dc_isar_feature(aa32_simd_r32, s) && (a->vd & 0x10)) {
+        return false;
+    }
+    if (a->itype > 10) {
+        return false;
+    }
+    /* Catch UNDEF cases for bad values of align field */
+    switch (a->itype & 0xc) {
+    case 4:
+        if (a->align >= 2) {
+            return false;
+        }
+        break;
+    case 8:
+        if (a->align == 3) {
+            return false;
+        }
+        break;
+    default:
+        break;
+    }
+    nregs = neon_ls_element_type[a->itype].nregs;
+    interleave = neon_ls_element_type[a->itype].interleave;
+    spacing = neon_ls_element_type[a->itype].spacing;
+    if (size == 3 && (interleave | spacing) != 1) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    /* For our purposes, bytes are always little-endian.  */
+    if (size == 0) {
+        endian = MO_LE;
+    }
+    /*
+     * Consecutive little-endian elements from a single register
+     * can be promoted to a larger little-endian operation.
+     */
+    if (interleave == 1 && endian == MO_LE) {
+        size = 3;
+    }
+    tmp64 = tcg_temp_new_i64();
+    addr = tcg_temp_new_i32();
+    tmp = tcg_const_i32(1 << size);
+    load_reg_var(s, addr, a->rn);
+    for (reg = 0; reg < nregs; reg++) {
+        for (n = 0; n < 8 >> size; n++) {
+            int xs;
+            for (xs = 0; xs < interleave; xs++) {
+                int tt = a->vd + reg + spacing * xs;
+
+                if (a->l) {
+                    gen_aa32_ld_i64(s, tmp64, addr, mmu_idx, endian | size);
+                    neon_store_element64(tt, n, size, tmp64);
+                } else {
+                    neon_load_element64(tmp64, tt, n, size);
+                    gen_aa32_st_i64(s, tmp64, addr, mmu_idx, endian | size);
+                }
+                tcg_gen_add_i32(addr, addr, tmp);
+            }
+        }
+    }
+    tcg_temp_free_i32(addr);
+    tcg_temp_free_i32(tmp);
+    tcg_temp_free_i64(tmp64);
+
+    gen_neon_ldst_base_update(s, a->rm, a->rn, nregs * interleave * 8);
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index e269642a480..be56cbb0614 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -3214,45 +3214,19 @@ static void gen_neon_trn_u16(TCGv_i32 t0, TCGv_i32 t1)
 }
 
 
-static struct {
-    int nregs;
-    int interleave;
-    int spacing;
-} const neon_ls_element_type[11] = {
-    {1, 4, 1},
-    {1, 4, 2},
-    {4, 1, 1},
-    {2, 2, 2},
-    {1, 3, 1},
-    {1, 3, 2},
-    {3, 1, 1},
-    {1, 1, 1},
-    {1, 2, 1},
-    {1, 2, 2},
-    {2, 1, 1}
-};
-
 /* Translate a NEON load/store element instruction.  Return nonzero if the
    instruction is invalid.  */
 static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
 {
     int rd, rn, rm;
-    int op;
     int nregs;
-    int interleave;
-    int spacing;
     int stride;
     int size;
     int reg;
     int load;
-    int n;
     int vec_size;
-    int mmu_idx;
-    MemOp endian;
     TCGv_i32 addr;
     TCGv_i32 tmp;
-    TCGv_i32 tmp2;
-    TCGv_i64 tmp64;
 
     if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
         return 1;
@@ -3274,70 +3248,9 @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
     rn = (insn >> 16) & 0xf;
     rm = insn & 0xf;
     load = (insn & (1 << 21)) != 0;
-    endian = s->be_data;
-    mmu_idx = get_mem_index(s);
     if ((insn & (1 << 23)) == 0) {
-        /* Load store all elements.  */
-        op = (insn >> 8) & 0xf;
-        size = (insn >> 6) & 3;
-        if (op > 10)
-            return 1;
-        /* Catch UNDEF cases for bad values of align field */
-        switch (op & 0xc) {
-        case 4:
-            if (((insn >> 5) & 1) == 1) {
-                return 1;
-            }
-            break;
-        case 8:
-            if (((insn >> 4) & 3) == 3) {
-                return 1;
-            }
-            break;
-        default:
-            break;
-        }
-        nregs = neon_ls_element_type[op].nregs;
-        interleave = neon_ls_element_type[op].interleave;
-        spacing = neon_ls_element_type[op].spacing;
-        if (size == 3 && (interleave | spacing) != 1) {
-            return 1;
-        }
-        /* For our purposes, bytes are always little-endian.  */
-        if (size == 0) {
-            endian = MO_LE;
-        }
-        /* Consecutive little-endian elements from a single register
-         * can be promoted to a larger little-endian operation.
-         */
-        if (interleave == 1 && endian == MO_LE) {
-            size = 3;
-        }
-        tmp64 = tcg_temp_new_i64();
-        addr = tcg_temp_new_i32();
-        tmp2 = tcg_const_i32(1 << size);
-        load_reg_var(s, addr, rn);
-        for (reg = 0; reg < nregs; reg++) {
-            for (n = 0; n < 8 >> size; n++) {
-                int xs;
-                for (xs = 0; xs < interleave; xs++) {
-                    int tt = rd + reg + spacing * xs;
-
-                    if (load) {
-                        gen_aa32_ld_i64(s, tmp64, addr, mmu_idx, endian | size);
-                        neon_store_element64(tt, n, size, tmp64);
-                    } else {
-                        neon_load_element64(tmp64, tt, n, size);
-                        gen_aa32_st_i64(s, tmp64, addr, mmu_idx, endian | size);
-                    }
-                    tcg_gen_add_i32(addr, addr, tmp2);
-                }
-            }
-        }
-        tcg_temp_free_i32(addr);
-        tcg_temp_free_i32(tmp2);
-        tcg_temp_free_i64(tmp64);
-        stride = nregs * interleave * 8;
+        /* Load store all elements -- handled already by decodetree */
+        return 1;
     } else {
         size = (insn >> 10) & 3;
         if (size == 3) {
-- 
2.20.1

[PULL 31/39] target/arm: Convert Neon 'load single structure to all lanes' to decodetree

[Peter Maydell]

Convert the Neon "load single structure to all lanes" insns to
decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-13-peter.maydell@linaro.org
---
 target/arm/neon-ls.decode       |  5 +++
 target/arm/translate-neon.inc.c | 73 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 55 +------------------------
 3 files changed, 80 insertions(+), 53 deletions(-)

diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode
index dd03d5a37bd..f0ab6d2c987 100644
--- a/target/arm/neon-ls.decode
+++ b/target/arm/neon-ls.decode
@@ -34,3 +34,8 @@
 
 VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \
                vd=%vd_dp
+
+# Neon load single element to all lanes
+
+VLD_all_lanes  1111 0100 1 . 1 0 rn:4 .... 11 n:2 size:2 t:1 a:1 rm:4 \
+               vd=%vd_dp
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 966c0d92012..e60e9559bad 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -398,3 +398,76 @@ static bool trans_VLDST_multiple(DisasContext *s, arg_VLDST_multiple *a)
     gen_neon_ldst_base_update(s, a->rm, a->rn, nregs * interleave * 8);
     return true;
 }
+
+static bool trans_VLD_all_lanes(DisasContext *s, arg_VLD_all_lanes *a)
+{
+    /* Neon load single structure to all lanes */
+    int reg, stride, vec_size;
+    int vd = a->vd;
+    int size = a->size;
+    int nregs = a->n + 1;
+    TCGv_i32 addr, tmp;
+
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist */
+    if (!dc_isar_feature(aa32_simd_r32, s) && (a->vd & 0x10)) {
+        return false;
+    }
+
+    if (size == 3) {
+        if (nregs != 4 || a->a == 0) {
+            return false;
+        }
+        /* For VLD4 size == 3 a == 1 means 32 bits at 16 byte alignment */
+        size = 2;
+    }
+    if (nregs == 1 && a->a == 1 && size == 0) {
+        return false;
+    }
+    if (nregs == 3 && a->a == 1) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    /*
+     * VLD1 to all lanes: T bit indicates how many Dregs to write.
+     * VLD2/3/4 to all lanes: T bit indicates register stride.
+     */
+    stride = a->t ? 2 : 1;
+    vec_size = nregs == 1 ? stride * 8 : 8;
+
+    tmp = tcg_temp_new_i32();
+    addr = tcg_temp_new_i32();
+    load_reg_var(s, addr, a->rn);
+    for (reg = 0; reg < nregs; reg++) {
+        gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s),
+                        s->be_data | size);
+        if ((vd & 1) && vec_size == 16) {
+            /*
+             * We cannot write 16 bytes at once because the
+             * destination is unaligned.
+             */
+            tcg_gen_gvec_dup_i32(size, neon_reg_offset(vd, 0),
+                                 8, 8, tmp);
+            tcg_gen_gvec_mov(0, neon_reg_offset(vd + 1, 0),
+                             neon_reg_offset(vd, 0), 8, 8);
+        } else {
+            tcg_gen_gvec_dup_i32(size, neon_reg_offset(vd, 0),
+                                 vec_size, vec_size, tmp);
+        }
+        tcg_gen_addi_i32(addr, addr, 1 << size);
+        vd += stride;
+    }
+    tcg_temp_free_i32(tmp);
+    tcg_temp_free_i32(addr);
+
+    gen_neon_ldst_base_update(s, a->rm, a->rn, (1 << size) * nregs);
+
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index be56cbb0614..7099274c92a 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -3224,7 +3224,6 @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
     int size;
     int reg;
     int load;
-    int vec_size;
     TCGv_i32 addr;
     TCGv_i32 tmp;
 
@@ -3254,58 +3253,8 @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
     } else {
         size = (insn >> 10) & 3;
         if (size == 3) {
-            /* Load single element to all lanes.  */
-            int a = (insn >> 4) & 1;
-            if (!load) {
-                return 1;
-            }
-            size = (insn >> 6) & 3;
-            nregs = ((insn >> 8) & 3) + 1;
-
-            if (size == 3) {
-                if (nregs != 4 || a == 0) {
-                    return 1;
-                }
-                /* For VLD4 size==3 a == 1 means 32 bits at 16 byte alignment */
-                size = 2;
-            }
-            if (nregs == 1 && a == 1 && size == 0) {
-                return 1;
-            }
-            if (nregs == 3 && a == 1) {
-                return 1;
-            }
-            addr = tcg_temp_new_i32();
-            load_reg_var(s, addr, rn);
-
-            /* VLD1 to all lanes: bit 5 indicates how many Dregs to write.
-             * VLD2/3/4 to all lanes: bit 5 indicates register stride.
-             */
-            stride = (insn & (1 << 5)) ? 2 : 1;
-            vec_size = nregs == 1 ? stride * 8 : 8;
-
-            tmp = tcg_temp_new_i32();
-            for (reg = 0; reg < nregs; reg++) {
-                gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s),
-                                s->be_data | size);
-                if ((rd & 1) && vec_size == 16) {
-                    /* We cannot write 16 bytes at once because the
-                     * destination is unaligned.
-                     */
-                    tcg_gen_gvec_dup_i32(size, neon_reg_offset(rd, 0),
-                                         8, 8, tmp);
-                    tcg_gen_gvec_mov(0, neon_reg_offset(rd + 1, 0),
-                                     neon_reg_offset(rd, 0), 8, 8);
-                } else {
-                    tcg_gen_gvec_dup_i32(size, neon_reg_offset(rd, 0),
-                                         vec_size, vec_size, tmp);
-                }
-                tcg_gen_addi_i32(addr, addr, 1 << size);
-                rd += stride;
-            }
-            tcg_temp_free_i32(tmp);
-            tcg_temp_free_i32(addr);
-            stride = (1 << size) * nregs;
+            /* Load single element to all lanes -- handled by decodetree  */
+            return 1;
         } else {
             /* Single element.  */
             int idx = (insn >> 4) & 0xf;
-- 
2.20.1

[PULL 32/39] target/arm: Convert Neon 'load/store single structure' to decodetree

[Peter Maydell]

Convert the Neon "load/store single structure to one lane" insns to
decodetree.

As this is the last set of insns in the neon load/store group,
we can remove the whole disas_neon_ls_insn() function.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-14-peter.maydell@linaro.org
---
 target/arm/neon-ls.decode       |  11 +++
 target/arm/translate-neon.inc.c |  89 +++++++++++++++++++
 target/arm/translate.c          | 147 --------------------------------
 3 files changed, 100 insertions(+), 147 deletions(-)

diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode
index f0ab6d2c987..c7b03a72e8d 100644
--- a/target/arm/neon-ls.decode
+++ b/target/arm/neon-ls.decode
@@ -39,3 +39,14 @@ VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \
 
 VLD_all_lanes  1111 0100 1 . 1 0 rn:4 .... 11 n:2 size:2 t:1 a:1 rm:4 \
                vd=%vd_dp
+
+# Neon load/store single structure to one lane
+%imm1_5_p1 5:1 !function=plus1
+%imm1_6_p1 6:1 !function=plus1
+
+VLDST_single   1111 0100 1 . l:1 0 rn:4 .... 00 n:2 reg_idx:3 align:1 rm:4 \
+               vd=%vd_dp size=0 stride=1
+VLDST_single   1111 0100 1 . l:1 0 rn:4 .... 01 n:2 reg_idx:2 align:2 rm:4 \
+               vd=%vd_dp size=1 stride=%imm1_5_p1
+VLDST_single   1111 0100 1 . l:1 0 rn:4 .... 10 n:2 reg_idx:1 align:3 rm:4 \
+               vd=%vd_dp size=2 stride=%imm1_6_p1
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index e60e9559bad..c881d1cf607 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -26,6 +26,11 @@
  * It might be possible to convert it to a standalone .c file eventually.
  */
 
+static inline int plus1(DisasContext *s, int x)
+{
+    return x + 1;
+}
+
 /* Include the generated Neon decoder */
 #include "decode-neon-dp.inc.c"
 #include "decode-neon-ls.inc.c"
@@ -471,3 +476,87 @@ static bool trans_VLD_all_lanes(DisasContext *s, arg_VLD_all_lanes *a)
 
     return true;
 }
+
+static bool trans_VLDST_single(DisasContext *s, arg_VLDST_single *a)
+{
+    /* Neon load/store single structure to one lane */
+    int reg;
+    int nregs = a->n + 1;
+    int vd = a->vd;
+    TCGv_i32 addr, tmp;
+
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist */
+    if (!dc_isar_feature(aa32_simd_r32, s) && (a->vd & 0x10)) {
+        return false;
+    }
+
+    /* Catch the UNDEF cases. This is unavoidably a bit messy. */
+    switch (nregs) {
+    case 1:
+        if (((a->align & (1 << a->size)) != 0) ||
+            (a->size == 2 && ((a->align & 3) == 1 || (a->align & 3) == 2))) {
+            return false;
+        }
+        break;
+    case 3:
+        if ((a->align & 1) != 0) {
+            return false;
+        }
+        /* fall through */
+    case 2:
+        if (a->size == 2 && (a->align & 2) != 0) {
+            return false;
+        }
+        break;
+    case 4:
+        if ((a->size == 2) && ((a->align & 3) == 3)) {
+            return false;
+        }
+        break;
+    default:
+        abort();
+    }
+    if ((vd + a->stride * (nregs - 1)) > 31) {
+        /*
+         * Attempts to write off the end of the register file are
+         * UNPREDICTABLE; we choose to UNDEF because otherwise we would
+         * access off the end of the array that holds the register data.
+         */
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    tmp = tcg_temp_new_i32();
+    addr = tcg_temp_new_i32();
+    load_reg_var(s, addr, a->rn);
+    /*
+     * TODO: if we implemented alignment exceptions, we should check
+     * addr against the alignment encoded in a->align here.
+     */
+    for (reg = 0; reg < nregs; reg++) {
+        if (a->l) {
+            gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s),
+                            s->be_data | a->size);
+            neon_store_element(vd, a->reg_idx, a->size, tmp);
+        } else { /* Store */
+            neon_load_element(tmp, vd, a->reg_idx, a->size);
+            gen_aa32_st_i32(s, tmp, addr, get_mem_index(s),
+                            s->be_data | a->size);
+        }
+        vd += a->stride;
+        tcg_gen_addi_i32(addr, addr, 1 << a->size);
+    }
+    tcg_temp_free_i32(addr);
+    tcg_temp_free_i32(tmp);
+
+    gen_neon_ldst_base_update(s, a->rm, a->rn, (1 << a->size) * nregs);
+
+    return true;
+}
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 7099274c92a..613be39ef36 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -3213,140 +3213,6 @@ static void gen_neon_trn_u16(TCGv_i32 t0, TCGv_i32 t1)
     tcg_temp_free_i32(rd);
 }
 
-
-/* Translate a NEON load/store element instruction.  Return nonzero if the
-   instruction is invalid.  */
-static int disas_neon_ls_insn(DisasContext *s, uint32_t insn)
-{
-    int rd, rn, rm;
-    int nregs;
-    int stride;
-    int size;
-    int reg;
-    int load;
-    TCGv_i32 addr;
-    TCGv_i32 tmp;
-
-    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
-        return 1;
-    }
-
-    /* FIXME: this access check should not take precedence over UNDEF
-     * for invalid encodings; we will generate incorrect syndrome information
-     * for attempts to execute invalid vfp/neon encodings with FP disabled.
-     */
-    if (s->fp_excp_el) {
-        gen_exception_insn(s, s->pc_curr, EXCP_UDEF,
-                           syn_simd_access_trap(1, 0xe, false), s->fp_excp_el);
-        return 0;
-    }
-
-    if (!s->vfp_enabled)
-      return 1;
-    VFP_DREG_D(rd, insn);
-    rn = (insn >> 16) & 0xf;
-    rm = insn & 0xf;
-    load = (insn & (1 << 21)) != 0;
-    if ((insn & (1 << 23)) == 0) {
-        /* Load store all elements -- handled already by decodetree */
-        return 1;
-    } else {
-        size = (insn >> 10) & 3;
-        if (size == 3) {
-            /* Load single element to all lanes -- handled by decodetree  */
-            return 1;
-        } else {
-            /* Single element.  */
-            int idx = (insn >> 4) & 0xf;
-            int reg_idx;
-            switch (size) {
-            case 0:
-                reg_idx = (insn >> 5) & 7;
-                stride = 1;
-                break;
-            case 1:
-                reg_idx = (insn >> 6) & 3;
-                stride = (insn & (1 << 5)) ? 2 : 1;
-                break;
-            case 2:
-                reg_idx = (insn >> 7) & 1;
-                stride = (insn & (1 << 6)) ? 2 : 1;
-                break;
-            default:
-                abort();
-            }
-            nregs = ((insn >> 8) & 3) + 1;
-            /* Catch the UNDEF cases. This is unavoidably a bit messy. */
-            switch (nregs) {
-            case 1:
-                if (((idx & (1 << size)) != 0) ||
-                    (size == 2 && ((idx & 3) == 1 || (idx & 3) == 2))) {
-                    return 1;
-                }
-                break;
-            case 3:
-                if ((idx & 1) != 0) {
-                    return 1;
-                }
-                /* fall through */
-            case 2:
-                if (size == 2 && (idx & 2) != 0) {
-                    return 1;
-                }
-                break;
-            case 4:
-                if ((size == 2) && ((idx & 3) == 3)) {
-                    return 1;
-                }
-                break;
-            default:
-                abort();
-            }
-            if ((rd + stride * (nregs - 1)) > 31) {
-                /* Attempts to write off the end of the register file
-                 * are UNPREDICTABLE; we choose to UNDEF because otherwise
-                 * the neon_load_reg() would write off the end of the array.
-                 */
-                return 1;
-            }
-            tmp = tcg_temp_new_i32();
-            addr = tcg_temp_new_i32();
-            load_reg_var(s, addr, rn);
-            for (reg = 0; reg < nregs; reg++) {
-                if (load) {
-                    gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s),
-                                    s->be_data | size);
-                    neon_store_element(rd, reg_idx, size, tmp);
-                } else { /* Store */
-                    neon_load_element(tmp, rd, reg_idx, size);
-                    gen_aa32_st_i32(s, tmp, addr, get_mem_index(s),
-                                    s->be_data | size);
-                }
-                rd += stride;
-                tcg_gen_addi_i32(addr, addr, 1 << size);
-            }
-            tcg_temp_free_i32(addr);
-            tcg_temp_free_i32(tmp);
-            stride = nregs * (1 << size);
-        }
-    }
-    if (rm != 15) {
-        TCGv_i32 base;
-
-        base = load_reg(s, rn);
-        if (rm == 13) {
-            tcg_gen_addi_i32(base, base, stride);
-        } else {
-            TCGv_i32 index;
-            index = load_reg(s, rm);
-            tcg_gen_add_i32(base, base, index);
-            tcg_temp_free_i32(index);
-        }
-        store_reg(s, rn, base);
-    }
-    return 0;
-}
-
 static inline void gen_neon_narrow(int size, TCGv_i32 dest, TCGv_i64 src)
 {
     switch (size) {
@@ -10596,13 +10462,6 @@ static void disas_arm_insn(DisasContext *s, unsigned int insn)
             }
             return;
         }
-        if ((insn & 0x0f100000) == 0x04000000) {
-            /* NEON load/store.  */
-            if (disas_neon_ls_insn(s, insn)) {
-                goto illegal_op;
-            }
-            return;
-        }
         if ((insn & 0x0e000f00) == 0x0c000100) {
             if (arm_dc_feature(s, ARM_FEATURE_IWMMXT)) {
                 /* iWMMXt register transfer.  */
@@ -10807,12 +10666,6 @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn)
         }
         break;
     case 12:
-        if ((insn & 0x01100000) == 0x01000000) {
-            if (disas_neon_ls_insn(s, insn)) {
-                goto illegal_op;
-            }
-            break;
-        }
         goto illegal_op;
     default:
     illegal_op:
-- 
2.20.1

[PULL 33/39] target/arm: Convert Neon 3-reg-same VADD/VSUB to decodetree

[Peter Maydell]

Convert the Neon 3-reg-same VADD and VSUB insns to decodetree.

Note that we don't need the neon_3r_sizes[op] check here because all
size values are OK for VADD and VSUB; we'll add this when we convert
the first insn that has size restrictions.

For this we need one of the GVecGen*Fn typedefs currently in
translate-a64.h; move them all to translate.h as a block so they
are visible to the 32-bit decoder.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-15-peter.maydell@linaro.org
---
 target/arm/translate-a64.h      |  9 --------
 target/arm/translate.h          |  9 ++++++++
 target/arm/neon-dp.decode       | 17 +++++++++++++++
 target/arm/translate-neon.inc.c | 38 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 14 ++++--------
 5 files changed, 68 insertions(+), 19 deletions(-)

diff --git a/target/arm/translate-a64.h b/target/arm/translate-a64.h
index 4c2c91ae1b2..f02fbb63a4a 100644
--- a/target/arm/translate-a64.h
+++ b/target/arm/translate-a64.h
@@ -115,13 +115,4 @@ static inline int vec_full_reg_size(DisasContext *s)
 
 bool disas_sve(DisasContext *, uint32_t);
 
-/* Note that the gvec expanders operate on offsets + sizes.  */
-typedef void GVecGen2Fn(unsigned, uint32_t, uint32_t, uint32_t, uint32_t);
-typedef void GVecGen2iFn(unsigned, uint32_t, uint32_t, int64_t,
-                         uint32_t, uint32_t);
-typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t,
-                        uint32_t, uint32_t, uint32_t);
-typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t,
-                        uint32_t, uint32_t, uint32_t);
-
 #endif /* TARGET_ARM_TRANSLATE_A64_H */
diff --git a/target/arm/translate.h b/target/arm/translate.h
index 98b319f3f69..95b43e7ab65 100644
--- a/target/arm/translate.h
+++ b/target/arm/translate.h
@@ -305,4 +305,13 @@ void gen_sshl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b);
 #define dc_isar_feature(name, ctx) \
     ({ DisasContext *ctx_ = (ctx); isar_feature_##name(ctx_->isar); })
 
+/* Note that the gvec expanders operate on offsets + sizes.  */
+typedef void GVecGen2Fn(unsigned, uint32_t, uint32_t, uint32_t, uint32_t);
+typedef void GVecGen2iFn(unsigned, uint32_t, uint32_t, int64_t,
+                         uint32_t, uint32_t);
+typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t,
+                        uint32_t, uint32_t, uint32_t);
+typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t,
+                        uint32_t, uint32_t, uint32_t);
+
 #endif /* TARGET_ARM_TRANSLATE_H */
diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index c89a1a58591..a61b1e88476 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -18,6 +18,10 @@
 #
 # This file is processed by scripts/decodetree.py
 #
+# VFP/Neon register fields; same as vfp.decode
+%vm_dp  5:1 0:4
+%vn_dp  7:1 16:4
+%vd_dp  22:1 12:4
 
 # Encodings for Neon data processing instructions where the T32 encoding
 # is a simple transformation of the A32 encoding.
@@ -27,3 +31,16 @@
 #   0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq
 # This file works on the A32 encoding only; calling code for T32 has to
 # transform the insn into the A32 version first.
+
+######################################################################
+# 3-reg-same grouping:
+# 1111 001 U 0 D sz:2 Vn:4 Vd:4 opc:4 N Q M op Vm:4
+######################################################################
+
+&3same vm vn vd q size
+
+@3same           .... ... . . . size:2 .... .... .... . q:1 . . .... \
+                 &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp
+
+VADD_3s          1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same
+VSUB_3s          1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index c881d1cf607..bd9e697b3e2 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -560,3 +560,41 @@ static bool trans_VLDST_single(DisasContext *s, arg_VLDST_single *a)
 
     return true;
 }
+
+static bool do_3same(DisasContext *s, arg_3same *a, GVecGen3Fn fn)
+{
+    int vec_size = a->q ? 16 : 8;
+    int rd_ofs = neon_reg_offset(a->vd, 0);
+    int rn_ofs = neon_reg_offset(a->vn, 0);
+    int rm_ofs = neon_reg_offset(a->vm, 0);
+
+    if (!arm_dc_feature(s, ARM_FEATURE_NEON)) {
+        return false;
+    }
+
+    /* UNDEF accesses to D16-D31 if they don't exist. */
+    if (!dc_isar_feature(aa32_simd_r32, s) &&
+        ((a->vd | a->vn | a->vm) & 0x10)) {
+        return false;
+    }
+
+    if ((a->vn | a->vm | a->vd) & a->q) {
+        return false;
+    }
+
+    if (!vfp_access_check(s)) {
+        return true;
+    }
+
+    fn(a->size, rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size);
+    return true;
+}
+
+#define DO_3SAME(INSN, FUNC)                                            \
+    static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a)        \
+    {                                                                   \
+        return do_3same(s, a, FUNC);                                    \
+    }
+
+DO_3SAME(VADD, tcg_gen_gvec_add)
+DO_3SAME(VSUB, tcg_gen_gvec_sub)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 613be39ef36..061bc7c31c9 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4885,16 +4885,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             }
             return 0;
 
-        case NEON_3R_VADD_VSUB:
-            if (u) {
-                tcg_gen_gvec_sub(size, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-            } else {
-                tcg_gen_gvec_add(size, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-            }
-            return 0;
-
         case NEON_3R_VQADD:
             tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc),
                            rn_ofs, rm_ofs, vec_size, vec_size,
@@ -4970,6 +4960,10 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size,
                            u ? &ushl_op[size] : &sshl_op[size]);
             return 0;
+
+        case NEON_3R_VADD_VSUB:
+            /* Already handled by decodetree */
+            return 1;
         }
 
         if (size == 3) {
-- 
2.20.1

[PULL 34/39] target/arm: Convert Neon 3-reg-same logic ops to decodetree

[Peter Maydell]

Convert the Neon logic ops in the 3-reg-same grouping to decodetree.
Note that for the logic ops the 'size' field forms part of their
decode and the actual operations are always bitwise.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-16-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       | 12 +++++++++++
 target/arm/translate-neon.inc.c | 19 +++++++++++++++++
 target/arm/translate.c          | 38 +--------------------------------
 3 files changed, 32 insertions(+), 37 deletions(-)

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index a61b1e88476..f62dbaa72d5 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -42,5 +42,17 @@
 @3same           .... ... . . . size:2 .... .... .... . q:1 . . .... \
                  &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp
 
+@3same_logic     .... ... . . . .. .... .... .... . q:1 .. .... \
+                 &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0
+
+VAND_3s          1111 001 0 0 . 00 .... .... 0001 ... 1 .... @3same_logic
+VBIC_3s          1111 001 0 0 . 01 .... .... 0001 ... 1 .... @3same_logic
+VORR_3s          1111 001 0 0 . 10 .... .... 0001 ... 1 .... @3same_logic
+VORN_3s          1111 001 0 0 . 11 .... .... 0001 ... 1 .... @3same_logic
+VEOR_3s          1111 001 1 0 . 00 .... .... 0001 ... 1 .... @3same_logic
+VBSL_3s          1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic
+VBIT_3s          1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic
+VBIF_3s          1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic
+
 VADD_3s          1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same
 VSUB_3s          1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index bd9e697b3e2..507f0abe801 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -598,3 +598,22 @@ static bool do_3same(DisasContext *s, arg_3same *a, GVecGen3Fn fn)
 
 DO_3SAME(VADD, tcg_gen_gvec_add)
 DO_3SAME(VSUB, tcg_gen_gvec_sub)
+DO_3SAME(VAND, tcg_gen_gvec_and)
+DO_3SAME(VBIC, tcg_gen_gvec_andc)
+DO_3SAME(VORR, tcg_gen_gvec_or)
+DO_3SAME(VORN, tcg_gen_gvec_orc)
+DO_3SAME(VEOR, tcg_gen_gvec_xor)
+
+/* These insns are all gvec_bitsel but with the inputs in various orders. */
+#define DO_3SAME_BITSEL(INSN, O1, O2, O3)                               \
+    static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
+                                uint32_t rn_ofs, uint32_t rm_ofs,       \
+                                uint32_t oprsz, uint32_t maxsz)         \
+    {                                                                   \
+        tcg_gen_gvec_bitsel(vece, rd_ofs, O1, O2, O3, oprsz, maxsz);    \
+    }                                                                   \
+    DO_3SAME(INSN, gen_##INSN##_3s)
+
+DO_3SAME_BITSEL(VBSL, rd_ofs, rn_ofs, rm_ofs)
+DO_3SAME_BITSEL(VBIT, rm_ofs, rn_ofs, rd_ofs)
+DO_3SAME_BITSEL(VBIF, rm_ofs, rd_ofs, rn_ofs)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 061bc7c31c9..9affa92cbe7 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4848,43 +4848,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             }
             return 1;
 
-        case NEON_3R_LOGIC: /* Logic ops.  */
-            switch ((u << 2) | size) {
-            case 0: /* VAND */
-                tcg_gen_gvec_and(0, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-                break;
-            case 1: /* VBIC */
-                tcg_gen_gvec_andc(0, rd_ofs, rn_ofs, rm_ofs,
-                                  vec_size, vec_size);
-                break;
-            case 2: /* VORR */
-                tcg_gen_gvec_or(0, rd_ofs, rn_ofs, rm_ofs,
-                                vec_size, vec_size);
-                break;
-            case 3: /* VORN */
-                tcg_gen_gvec_orc(0, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-                break;
-            case 4: /* VEOR */
-                tcg_gen_gvec_xor(0, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-                break;
-            case 5: /* VBSL */
-                tcg_gen_gvec_bitsel(MO_8, rd_ofs, rd_ofs, rn_ofs, rm_ofs,
-                                    vec_size, vec_size);
-                break;
-            case 6: /* VBIT */
-                tcg_gen_gvec_bitsel(MO_8, rd_ofs, rm_ofs, rn_ofs, rd_ofs,
-                                    vec_size, vec_size);
-                break;
-            case 7: /* VBIF */
-                tcg_gen_gvec_bitsel(MO_8, rd_ofs, rm_ofs, rd_ofs, rn_ofs,
-                                    vec_size, vec_size);
-                break;
-            }
-            return 0;
-
         case NEON_3R_VQADD:
             tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc),
                            rn_ofs, rm_ofs, vec_size, vec_size,
@@ -4962,6 +4925,7 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             return 0;
 
         case NEON_3R_VADD_VSUB:
+        case NEON_3R_LOGIC:
             /* Already handled by decodetree */
             return 1;
         }
-- 
2.20.1

[PULL 35/39] target/arm: Convert Neon 3-reg-same VMAX/VMIN to decodetree

[Peter Maydell]

Convert the Neon 3-reg-same VMAX and VMIN insns to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-17-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       |  5 +++++
 target/arm/translate-neon.inc.c | 14 ++++++++++++++
 target/arm/translate.c          | 21 ++-------------------
 3 files changed, 21 insertions(+), 19 deletions(-)

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index f62dbaa72d5..b721d39c7ba 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -54,5 +54,10 @@ VBSL_3s          1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic
 VBIT_3s          1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic
 VBIF_3s          1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic
 
+VMAX_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same
+VMAX_U_3s        1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same
+VMIN_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same
+VMIN_U_3s        1111 001 1 0 . .. .... .... 0110 . . . 1 .... @3same
+
 VADD_3s          1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same
 VSUB_3s          1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 507f0abe801..ab1740201c4 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -617,3 +617,17 @@ DO_3SAME(VEOR, tcg_gen_gvec_xor)
 DO_3SAME_BITSEL(VBSL, rd_ofs, rn_ofs, rm_ofs)
 DO_3SAME_BITSEL(VBIT, rm_ofs, rn_ofs, rd_ofs)
 DO_3SAME_BITSEL(VBIF, rm_ofs, rd_ofs, rn_ofs)
+
+#define DO_3SAME_NO_SZ_3(INSN, FUNC)                                    \
+    static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a)        \
+    {                                                                   \
+        if (a->size == 3) {                                             \
+            return false;                                               \
+        }                                                               \
+        return do_3same(s, a, FUNC);                                    \
+    }
+
+DO_3SAME_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax)
+DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax)
+DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin)
+DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 9affa92cbe7..2f054cfa783 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4899,25 +4899,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                              rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size);
             return 0;
 
-        case NEON_3R_VMAX:
-            if (u) {
-                tcg_gen_gvec_umax(size, rd_ofs, rn_ofs, rm_ofs,
-                                  vec_size, vec_size);
-            } else {
-                tcg_gen_gvec_smax(size, rd_ofs, rn_ofs, rm_ofs,
-                                  vec_size, vec_size);
-            }
-            return 0;
-        case NEON_3R_VMIN:
-            if (u) {
-                tcg_gen_gvec_umin(size, rd_ofs, rn_ofs, rm_ofs,
-                                  vec_size, vec_size);
-            } else {
-                tcg_gen_gvec_smin(size, rd_ofs, rn_ofs, rm_ofs,
-                                  vec_size, vec_size);
-            }
-            return 0;
-
         case NEON_3R_VSHL:
             /* Note the operation is vshl vd,vm,vn */
             tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size,
@@ -4926,6 +4907,8 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
 
         case NEON_3R_VADD_VSUB:
         case NEON_3R_LOGIC:
+        case NEON_3R_VMAX:
+        case NEON_3R_VMIN:
             /* Already handled by decodetree */
             return 1;
         }
-- 
2.20.1

[PULL 36/39] target/arm: Convert Neon 3-reg-same comparisons to decodetree

[Peter Maydell]

Convert the Neon comparison ops in the 3-reg-same grouping
to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-18-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       |  8 ++++++++
 target/arm/translate-neon.inc.c | 22 ++++++++++++++++++++++
 target/arm/translate.c          | 23 +++--------------------
 3 files changed, 33 insertions(+), 20 deletions(-)

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index b721d39c7ba..b89ea6819a9 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -54,6 +54,11 @@ VBSL_3s          1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic
 VBIT_3s          1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic
 VBIF_3s          1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic
 
+VCGT_S_3s        1111 001 0 0 . .. .... .... 0011 . . . 0 .... @3same
+VCGT_U_3s        1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same
+VCGE_S_3s        1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same
+VCGE_U_3s        1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same
+
 VMAX_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same
 VMAX_U_3s        1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same
 VMIN_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same
@@ -61,3 +66,6 @@ VMIN_U_3s        1111 001 1 0 . .. .... .... 0110 . . . 1 .... @3same
 
 VADD_3s          1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same
 VSUB_3s          1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
+
+VTST_3s          1111 001 0 0 . .. .... .... 1000 . . . 1 .... @3same
+VCEQ_3s          1111 001 1 0 . .. .... .... 1000 . . . 1 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index ab1740201c4..952e4456f5e 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -631,3 +631,25 @@ DO_3SAME_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax)
 DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax)
 DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin)
 DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin)
+
+#define DO_3SAME_CMP(INSN, COND)                                        \
+    static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
+                                uint32_t rn_ofs, uint32_t rm_ofs,       \
+                                uint32_t oprsz, uint32_t maxsz)         \
+    {                                                                   \
+        tcg_gen_gvec_cmp(COND, vece, rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz); \
+    }                                                                   \
+    DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s)
+
+DO_3SAME_CMP(VCGT_S, TCG_COND_GT)
+DO_3SAME_CMP(VCGT_U, TCG_COND_GTU)
+DO_3SAME_CMP(VCGE_S, TCG_COND_GE)
+DO_3SAME_CMP(VCGE_U, TCG_COND_GEU)
+DO_3SAME_CMP(VCEQ, TCG_COND_EQ)
+
+static void gen_VTST_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs,
+                         uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz)
+{
+    tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &cmtst_op[vece]);
+}
+DO_3SAME_NO_SZ_3(VTST, gen_VTST_3s)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 2f054cfa783..0e6ecc0969a 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4879,26 +4879,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
                            u ? &mls_op[size] : &mla_op[size]);
             return 0;
 
-        case NEON_3R_VTST_VCEQ:
-            if (u) { /* VCEQ */
-                tcg_gen_gvec_cmp(TCG_COND_EQ, size, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-            } else { /* VTST */
-                tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs,
-                               vec_size, vec_size, &cmtst_op[size]);
-            }
-            return 0;
-
-        case NEON_3R_VCGT:
-            tcg_gen_gvec_cmp(u ? TCG_COND_GTU : TCG_COND_GT, size,
-                             rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size);
-            return 0;
-
-        case NEON_3R_VCGE:
-            tcg_gen_gvec_cmp(u ? TCG_COND_GEU : TCG_COND_GE, size,
-                             rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size);
-            return 0;
-
         case NEON_3R_VSHL:
             /* Note the operation is vshl vd,vm,vn */
             tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size,
@@ -4909,6 +4889,9 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
         case NEON_3R_LOGIC:
         case NEON_3R_VMAX:
         case NEON_3R_VMIN:
+        case NEON_3R_VTST_VCEQ:
+        case NEON_3R_VCGT:
+        case NEON_3R_VCGE:
             /* Already handled by decodetree */
             return 1;
         }
-- 
2.20.1

[PULL 37/39] target/arm: Convert Neon 3-reg-same VQADD/VQSUB to decodetree

[Peter Maydell]

Convert the Neon VQADD/VQSUB insns in the 3-reg-same grouping
to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-19-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       |  6 ++++++
 target/arm/translate-neon.inc.c | 15 +++++++++++++++
 target/arm/translate.c          | 14 ++------------
 3 files changed, 23 insertions(+), 12 deletions(-)

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index b89ea6819a9..ab59b349aaa 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -42,6 +42,9 @@
 @3same           .... ... . . . size:2 .... .... .... . q:1 . . .... \
                  &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp
 
+VQADD_S_3s       1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same
+VQADD_U_3s       1111 001 1 0 . .. .... .... 0000 . . . 1 .... @3same
+
 @3same_logic     .... ... . . . .. .... .... .... . q:1 .. .... \
                  &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0
 
@@ -54,6 +57,9 @@ VBSL_3s          1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic
 VBIT_3s          1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic
 VBIF_3s          1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic
 
+VQSUB_S_3s       1111 001 0 0 . .. .... .... 0010 . . . 1 .... @3same
+VQSUB_U_3s       1111 001 1 0 . .. .... .... 0010 . . . 1 .... @3same
+
 VCGT_S_3s        1111 001 0 0 . .. .... .... 0011 . . . 0 .... @3same
 VCGT_U_3s        1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same
 VCGE_S_3s        1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 952e4456f5e..854ab70cd79 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -653,3 +653,18 @@ static void gen_VTST_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs,
     tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &cmtst_op[vece]);
 }
 DO_3SAME_NO_SZ_3(VTST, gen_VTST_3s)
+
+#define DO_3SAME_GVEC4(INSN, OPARRAY)                                   \
+    static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
+                                uint32_t rn_ofs, uint32_t rm_ofs,       \
+                                uint32_t oprsz, uint32_t maxsz)         \
+    {                                                                   \
+        tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc),           \
+                       rn_ofs, rm_ofs, oprsz, maxsz, &OPARRAY[vece]);   \
+    }                                                                   \
+    DO_3SAME(INSN, gen_##INSN##_3s)
+
+DO_3SAME_GVEC4(VQADD_S, sqadd_op)
+DO_3SAME_GVEC4(VQADD_U, uqadd_op)
+DO_3SAME_GVEC4(VQSUB_S, sqsub_op)
+DO_3SAME_GVEC4(VQSUB_U, uqsub_op)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 0e6ecc0969a..13ce1a5fc1d 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4848,18 +4848,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             }
             return 1;
 
-        case NEON_3R_VQADD:
-            tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc),
-                           rn_ofs, rm_ofs, vec_size, vec_size,
-                           (u ? uqadd_op : sqadd_op) + size);
-            return 0;
-
-        case NEON_3R_VQSUB:
-            tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc),
-                           rn_ofs, rm_ofs, vec_size, vec_size,
-                           (u ? uqsub_op : sqsub_op) + size);
-            return 0;
-
         case NEON_3R_VMUL: /* VMUL */
             if (u) {
                 /* Polynomial case allows only P8.  */
@@ -4892,6 +4880,8 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
         case NEON_3R_VTST_VCEQ:
         case NEON_3R_VCGT:
         case NEON_3R_VCGE:
+        case NEON_3R_VQADD:
+        case NEON_3R_VQSUB:
             /* Already handled by decodetree */
             return 1;
         }
-- 
2.20.1

[PULL 38/39] target/arm: Convert Neon 3-reg-same VMUL, VMLA, VMLS, VSHL to decodetree

[Peter Maydell]

Convert the Neon VMUL, VMLA, VMLS and VSHL insns in the
3-reg-same grouping to decodetree.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-20-peter.maydell@linaro.org
---
 target/arm/neon-dp.decode       |  9 +++++++
 target/arm/translate-neon.inc.c | 44 +++++++++++++++++++++++++++++++++
 target/arm/translate.c          | 28 +++------------------
 3 files changed, 56 insertions(+), 25 deletions(-)

diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode
index ab59b349aaa..ec3a92fe753 100644
--- a/target/arm/neon-dp.decode
+++ b/target/arm/neon-dp.decode
@@ -65,6 +65,9 @@ VCGT_U_3s        1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same
 VCGE_S_3s        1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same
 VCGE_U_3s        1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same
 
+VSHL_S_3s        1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same
+VSHL_U_3s        1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same
+
 VMAX_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same
 VMAX_U_3s        1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same
 VMIN_S_3s        1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same
@@ -75,3 +78,9 @@ VSUB_3s          1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same
 
 VTST_3s          1111 001 0 0 . .. .... .... 1000 . . . 1 .... @3same
 VCEQ_3s          1111 001 1 0 . .. .... .... 1000 . . . 1 .... @3same
+
+VMLA_3s          1111 001 0 0 . .. .... .... 1001 . . . 0 .... @3same
+VMLS_3s          1111 001 1 0 . .. .... .... 1001 . . . 0 .... @3same
+
+VMUL_3s          1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same
+VMUL_p_3s        1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same
diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c
index 854ab70cd79..50b77b6d714 100644
--- a/target/arm/translate-neon.inc.c
+++ b/target/arm/translate-neon.inc.c
@@ -631,6 +631,7 @@ DO_3SAME_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax)
 DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax)
 DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin)
 DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin)
+DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul)
 
 #define DO_3SAME_CMP(INSN, COND)                                        \
     static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
@@ -668,3 +669,46 @@ DO_3SAME_GVEC4(VQADD_S, sqadd_op)
 DO_3SAME_GVEC4(VQADD_U, uqadd_op)
 DO_3SAME_GVEC4(VQSUB_S, sqsub_op)
 DO_3SAME_GVEC4(VQSUB_U, uqsub_op)
+
+static void gen_VMUL_p_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs,
+                           uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz)
+{
+    tcg_gen_gvec_3_ool(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz,
+                       0, gen_helper_gvec_pmul_b);
+}
+
+static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a)
+{
+    if (a->size != 0) {
+        return false;
+    }
+    return do_3same(s, a, gen_VMUL_p_3s);
+}
+
+#define DO_3SAME_GVEC3_NO_SZ_3(INSN, OPARRAY)                           \
+    static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
+                                uint32_t rn_ofs, uint32_t rm_ofs,       \
+                                uint32_t oprsz, uint32_t maxsz)         \
+    {                                                                   \
+        tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs,                          \
+                       oprsz, maxsz, &OPARRAY[vece]);                   \
+    }                                                                   \
+    DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s)
+
+
+DO_3SAME_GVEC3_NO_SZ_3(VMLA, mla_op)
+DO_3SAME_GVEC3_NO_SZ_3(VMLS, mls_op)
+
+#define DO_3SAME_GVEC3_SHIFT(INSN, OPARRAY)                             \
+    static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs,         \
+                                uint32_t rn_ofs, uint32_t rm_ofs,       \
+                                uint32_t oprsz, uint32_t maxsz)         \
+    {                                                                   \
+        /* Note the operation is vshl vd,vm,vn */                       \
+        tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs,                          \
+                       oprsz, maxsz, &OPARRAY[vece]);                   \
+    }                                                                   \
+    DO_3SAME(INSN, gen_##INSN##_3s)
+
+DO_3SAME_GVEC3_SHIFT(VSHL_S, sshl_op)
+DO_3SAME_GVEC3_SHIFT(VSHL_U, ushl_op)
diff --git a/target/arm/translate.c b/target/arm/translate.c
index 13ce1a5fc1d..025747c0bd4 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -4848,31 +4848,6 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
             }
             return 1;
 
-        case NEON_3R_VMUL: /* VMUL */
-            if (u) {
-                /* Polynomial case allows only P8.  */
-                if (size != 0) {
-                    return 1;
-                }
-                tcg_gen_gvec_3_ool(rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size,
-                                   0, gen_helper_gvec_pmul_b);
-            } else {
-                tcg_gen_gvec_mul(size, rd_ofs, rn_ofs, rm_ofs,
-                                 vec_size, vec_size);
-            }
-            return 0;
-
-        case NEON_3R_VML: /* VMLA, VMLS */
-            tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size,
-                           u ? &mls_op[size] : &mla_op[size]);
-            return 0;
-
-        case NEON_3R_VSHL:
-            /* Note the operation is vshl vd,vm,vn */
-            tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size,
-                           u ? &ushl_op[size] : &sshl_op[size]);
-            return 0;
-
         case NEON_3R_VADD_VSUB:
         case NEON_3R_LOGIC:
         case NEON_3R_VMAX:
@@ -4882,6 +4857,9 @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn)
         case NEON_3R_VCGE:
         case NEON_3R_VQADD:
         case NEON_3R_VQSUB:
+        case NEON_3R_VMUL:
+        case NEON_3R_VML:
+        case NEON_3R_VSHL:
             /* Already handled by decodetree */
             return 1;
         }
-- 
2.20.1

[PULL 39/39] target/arm: Move gen_ function typedefs to translate.h

[Peter Maydell]

We're going to want at least some of the NeonGen* typedefs
for the refactored 32-bit Neon decoder, so move them all
to translate.h since it makes more sense to keep them in
one group.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200430181003.21682-23-peter.maydell@linaro.org
---
 target/arm/translate.h     | 17 +++++++++++++++++
 target/arm/translate-a64.c | 17 -----------------
 2 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/target/arm/translate.h b/target/arm/translate.h
index 95b43e7ab65..cb7925ea461 100644
--- a/target/arm/translate.h
+++ b/target/arm/translate.h
@@ -314,4 +314,21 @@ typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t,
 typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t,
                         uint32_t, uint32_t, uint32_t);
 
+/* Function prototype for gen_ functions for calling Neon helpers */
+typedef void NeonGenOneOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32);
+typedef void NeonGenTwoOpFn(TCGv_i32, TCGv_i32, TCGv_i32);
+typedef void NeonGenTwoOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32, TCGv_i32);
+typedef void NeonGenTwo64OpFn(TCGv_i64, TCGv_i64, TCGv_i64);
+typedef void NeonGenTwo64OpEnvFn(TCGv_i64, TCGv_ptr, TCGv_i64, TCGv_i64);
+typedef void NeonGenNarrowFn(TCGv_i32, TCGv_i64);
+typedef void NeonGenNarrowEnvFn(TCGv_i32, TCGv_ptr, TCGv_i64);
+typedef void NeonGenWidenFn(TCGv_i64, TCGv_i32);
+typedef void NeonGenTwoSingleOPFn(TCGv_i32, TCGv_i32, TCGv_i32, TCGv_ptr);
+typedef void NeonGenTwoDoubleOPFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_ptr);
+typedef void NeonGenOneOpFn(TCGv_i64, TCGv_i64);
+typedef void CryptoTwoOpFn(TCGv_ptr, TCGv_ptr);
+typedef void CryptoThreeOpIntFn(TCGv_ptr, TCGv_ptr, TCGv_i32);
+typedef void CryptoThreeOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr);
+typedef void AtomicThreeOpFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGArg, MemOp);
+
 #endif /* TARGET_ARM_TRANSLATE_H */
diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
index efb1c4adc4e..a896f9c4b83 100644
--- a/target/arm/translate-a64.c
+++ b/target/arm/translate-a64.c
@@ -70,23 +70,6 @@ typedef struct AArch64DecodeTable {
     AArch64DecodeFn *disas_fn;
 } AArch64DecodeTable;
 
-/* Function prototype for gen_ functions for calling Neon helpers */
-typedef void NeonGenOneOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32);
-typedef void NeonGenTwoOpFn(TCGv_i32, TCGv_i32, TCGv_i32);
-typedef void NeonGenTwoOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32, TCGv_i32);
-typedef void NeonGenTwo64OpFn(TCGv_i64, TCGv_i64, TCGv_i64);
-typedef void NeonGenTwo64OpEnvFn(TCGv_i64, TCGv_ptr, TCGv_i64, TCGv_i64);
-typedef void NeonGenNarrowFn(TCGv_i32, TCGv_i64);
-typedef void NeonGenNarrowEnvFn(TCGv_i32, TCGv_ptr, TCGv_i64);
-typedef void NeonGenWidenFn(TCGv_i64, TCGv_i32);
-typedef void NeonGenTwoSingleOPFn(TCGv_i32, TCGv_i32, TCGv_i32, TCGv_ptr);
-typedef void NeonGenTwoDoubleOPFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_ptr);
-typedef void NeonGenOneOpFn(TCGv_i64, TCGv_i64);
-typedef void CryptoTwoOpFn(TCGv_ptr, TCGv_ptr);
-typedef void CryptoThreeOpIntFn(TCGv_ptr, TCGv_ptr, TCGv_i32);
-typedef void CryptoThreeOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr);
-typedef void AtomicThreeOpFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGArg, MemOp);
-
 /* initialize TCG globals.  */
 void a64_translate_init(void)
 {
-- 
2.20.1

Re: [PULL 00/39] target-arm queue

[Peter Maydell]

On Mon, 4 May 2020 at 13:33, Peter Maydell <peter.maydell@linaro.org> wrote:
>
> Most of this is the Neon decodetree patches, followed by Edgar's versal cleanups.
>
> thanks
> -- PMM
>
>
> The following changes since commit 2ef486e76d64436be90f7359a3071fb2a56ce835:
>
>   Merge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging (2020-05-03 14:12:56 +0100)
>
> are available in the Git repository at:
>
>   https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200504
>
> for you to fetch changes up to 9aefc6cf9b73f66062d2f914a0136756e7a28211:
>
>   target/arm: Move gen_ function typedefs to translate.h (2020-05-04 12:59:26 +0100)
>
> ----------------------------------------------------------------
> target-arm queue:
>  * Start of conversion of Neon insns to decodetree
>  * versal board: support SD and RTC
>  * Implement ARMv8.2-TTS2UXN
>  * Make VQDMULL undefined when U=1
>  * Some minor code cleanups


Applied, thanks.

Please update the changelog at https://wiki.qemu.org/ChangeLog/5.1
for any user-visible changes.

-- PMM

Re: [PULL 00/39] target-arm queue

[no-reply]

Patchew URL: https://patchew.org/QEMU/20200504123309.3808-1-peter.maydell@linaro.org/



Hi,

This series failed build test on FreeBSD host. Please find the details below.

=== TEST SCRIPT BEGIN ===
#!/bin/bash
# Testing script will be invoked under the git checkout with
# HEAD pointing to a commit that has the patches applied on top of "base"
# branch
if qemu-system-x86_64 --help >/dev/null 2>&1; then
  QEMU=qemu-system-x86_64
elif /usr/libexec/qemu-kvm --help >/dev/null 2>&1; then
  QEMU=/usr/libexec/qemu-kvm
else
  exit 1
fi
make vm-build-freebsd J=21 QEMU=$QEMU
exit 0
=== TEST SCRIPT END ===




The full log is available at
http://patchew.org/logs/20200504123309.3808-1-peter.maydell@linaro.org/testing.FreeBSD/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to patchew-devel@redhat.com

Re: [PULL 00/39] target-arm queue

[no-reply]

Patchew URL: https://patchew.org/QEMU/20200504123309.3808-1-peter.maydell@linaro.org/



Hi,

This series failed the docker-mingw@fedora build test. Please find the testing commands and
their output below. If you have Docker installed, you can probably reproduce it
locally.

=== TEST SCRIPT BEGIN ===
#! /bin/bash
export ARCH=x86_64
make docker-image-fedora V=1 NETWORK=1
time make docker-test-mingw@fedora J=14 NETWORK=1
=== TEST SCRIPT END ===




The full log is available at
http://patchew.org/logs/20200504123309.3808-1-peter.maydell@linaro.org/testing.docker-mingw@fedora/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to patchew-devel@redhat.com