summary refs log tree commit diff stats
path: root/target/arm
diff options
context:
space:
mode:
Diffstat (limited to 'target/arm')
-rw-r--r--target/arm/arm-qmp-cmds.c2
-rw-r--r--target/arm/cpu.c2
-rw-r--r--target/arm/cpu.h4
-rw-r--r--target/arm/cpu64.c38
-rw-r--r--target/arm/helper.c285
-rw-r--r--target/arm/kvm.c2
-rw-r--r--target/arm/tcg/arith_helper.c296
-rw-r--r--target/arm/tcg/meson.build1
-rw-r--r--target/arm/tcg/op_addsub.c.inc (renamed from target/arm/op_addsub.h)0
9 files changed, 333 insertions, 297 deletions
diff --git a/target/arm/arm-qmp-cmds.c b/target/arm/arm-qmp-cmds.c
index 3cc8cc738b..33cea080d1 100644
--- a/target/arm/arm-qmp-cmds.c
+++ b/target/arm/arm-qmp-cmds.c
@@ -94,7 +94,7 @@ static const char *cpu_model_advertised_features[] = {
     "sve640", "sve768", "sve896", "sve1024", "sve1152", "sve1280",
     "sve1408", "sve1536", "sve1664", "sve1792", "sve1920", "sve2048",
     "kvm-no-adjvtime", "kvm-steal-time",
-    "pauth", "pauth-impdef", "pauth-qarma3",
+    "pauth", "pauth-impdef", "pauth-qarma3", "pauth-qarma5",
     NULL
 };
 
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index dcedadc89e..dc0231233a 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -2653,6 +2653,8 @@ static const Property arm_cpu_properties[] = {
     DEFINE_PROP_INT32("core-count", ARMCPU, core_count, -1),
     /* True to default to the backward-compat old CNTFRQ rather than 1Ghz */
     DEFINE_PROP_BOOL("backcompat-cntfrq", ARMCPU, backcompat_cntfrq, false),
+    DEFINE_PROP_BOOL("backcompat-pauth-default-use-qarma5", ARMCPU,
+                      backcompat_pauth_default_use_qarma5, false),
 };
 
 static const gchar *arm_gdb_arch_name(CPUState *cs)
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 12b8466542..9a6e8e589c 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -972,6 +972,9 @@ struct ArchCPU {
     /* QOM property to indicate we should use the back-compat CNTFRQ default */
     bool backcompat_cntfrq;
 
+    /* QOM property to indicate we should use the back-compat QARMA5 default */
+    bool backcompat_pauth_default_use_qarma5;
+
     /* Specify the number of cores in this CPU cluster. Used for the L2CTLR
      * register.
      */
@@ -1062,6 +1065,7 @@ struct ArchCPU {
     bool prop_pauth;
     bool prop_pauth_impdef;
     bool prop_pauth_qarma3;
+    bool prop_pauth_qarma5;
     bool prop_lpa2;
 
     /* DCZ blocksize, in log_2(words), ie low 4 bits of DCZID_EL0 */
diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
index dca83e4518..8188ede5cc 100644
--- a/target/arm/cpu64.c
+++ b/target/arm/cpu64.c
@@ -520,25 +520,40 @@ void arm_cpu_pauth_finalize(ARMCPU *cpu, Error **errp)
         }
 
         if (cpu->prop_pauth) {
-            if (cpu->prop_pauth_impdef && cpu->prop_pauth_qarma3) {
+            if ((cpu->prop_pauth_impdef && cpu->prop_pauth_qarma3) ||
+                (cpu->prop_pauth_impdef && cpu->prop_pauth_qarma5) ||
+                (cpu->prop_pauth_qarma3 && cpu->prop_pauth_qarma5)) {
                 error_setg(errp,
-                           "cannot enable both pauth-impdef and pauth-qarma3");
+                           "cannot enable pauth-impdef, pauth-qarma3 and "
+                           "pauth-qarma5 at the same time");
                 return;
             }
 
-            if (cpu->prop_pauth_impdef) {
-                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, API, features);
-                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, GPI, 1);
+            bool use_default = !cpu->prop_pauth_qarma5 &&
+                               !cpu->prop_pauth_qarma3 &&
+                               !cpu->prop_pauth_impdef;
+
+            if (cpu->prop_pauth_qarma5 ||
+                (use_default &&
+                 cpu->backcompat_pauth_default_use_qarma5)) {
+                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, APA, features);
+                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, GPA, 1);
             } else if (cpu->prop_pauth_qarma3) {
                 isar2 = FIELD_DP64(isar2, ID_AA64ISAR2, APA3, features);
                 isar2 = FIELD_DP64(isar2, ID_AA64ISAR2, GPA3, 1);
+            } else if (cpu->prop_pauth_impdef ||
+                       (use_default &&
+                        !cpu->backcompat_pauth_default_use_qarma5)) {
+                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, API, features);
+                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, GPI, 1);
             } else {
-                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, APA, features);
-                isar1 = FIELD_DP64(isar1, ID_AA64ISAR1, GPA, 1);
+                g_assert_not_reached();
             }
-        } else if (cpu->prop_pauth_impdef || cpu->prop_pauth_qarma3) {
-            error_setg(errp, "cannot enable pauth-impdef or "
-                       "pauth-qarma3 without pauth");
+        } else if (cpu->prop_pauth_impdef ||
+                   cpu->prop_pauth_qarma3 ||
+                   cpu->prop_pauth_qarma5) {
+            error_setg(errp, "cannot enable pauth-impdef, pauth-qarma3 or "
+                       "pauth-qarma5 without pauth");
             error_append_hint(errp, "Add pauth=on to the CPU property list.\n");
         }
     }
@@ -553,6 +568,8 @@ static const Property arm_cpu_pauth_impdef_property =
     DEFINE_PROP_BOOL("pauth-impdef", ARMCPU, prop_pauth_impdef, false);
 static const Property arm_cpu_pauth_qarma3_property =
     DEFINE_PROP_BOOL("pauth-qarma3", ARMCPU, prop_pauth_qarma3, false);
+static Property arm_cpu_pauth_qarma5_property =
+    DEFINE_PROP_BOOL("pauth-qarma5", ARMCPU, prop_pauth_qarma5, false);
 
 void aarch64_add_pauth_properties(Object *obj)
 {
@@ -573,6 +590,7 @@ void aarch64_add_pauth_properties(Object *obj)
     } else {
         qdev_property_add_static(DEVICE(obj), &arm_cpu_pauth_impdef_property);
         qdev_property_add_static(DEVICE(obj), &arm_cpu_pauth_qarma3_property);
+        qdev_property_add_static(DEVICE(obj), &arm_cpu_pauth_qarma5_property);
     }
 }
 
diff --git a/target/arm/helper.c b/target/arm/helper.c
index 5b595f951b..6399767851 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -17,11 +17,9 @@
 #include "qemu/main-loop.h"
 #include "qemu/timer.h"
 #include "qemu/bitops.h"
-#include "qemu/crc32c.h"
 #include "qemu/qemu-print.h"
 #include "exec/exec-all.h"
 #include "exec/translation-block.h"
-#include <zlib.h> /* for crc32 */
 #include "hw/irq.h"
 #include "system/cpu-timers.h"
 #include "system/kvm.h"
@@ -10984,289 +10982,6 @@ ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
     };
 }
 
-/*
- * Note that signed overflow is undefined in C.  The following routines are
- * careful to use unsigned types where modulo arithmetic is required.
- * Failure to do so _will_ break on newer gcc.
- */
-
-/* Signed saturating arithmetic.  */
-
-/* Perform 16-bit signed saturating addition.  */
-static inline uint16_t add16_sat(uint16_t a, uint16_t b)
-{
-    uint16_t res;
-
-    res = a + b;
-    if (((res ^ a) & 0x8000) && !((a ^ b) & 0x8000)) {
-        if (a & 0x8000) {
-            res = 0x8000;
-        } else {
-            res = 0x7fff;
-        }
-    }
-    return res;
-}
-
-/* Perform 8-bit signed saturating addition.  */
-static inline uint8_t add8_sat(uint8_t a, uint8_t b)
-{
-    uint8_t res;
-
-    res = a + b;
-    if (((res ^ a) & 0x80) && !((a ^ b) & 0x80)) {
-        if (a & 0x80) {
-            res = 0x80;
-        } else {
-            res = 0x7f;
-        }
-    }
-    return res;
-}
-
-/* Perform 16-bit signed saturating subtraction.  */
-static inline uint16_t sub16_sat(uint16_t a, uint16_t b)
-{
-    uint16_t res;
-
-    res = a - b;
-    if (((res ^ a) & 0x8000) && ((a ^ b) & 0x8000)) {
-        if (a & 0x8000) {
-            res = 0x8000;
-        } else {
-            res = 0x7fff;
-        }
-    }
-    return res;
-}
-
-/* Perform 8-bit signed saturating subtraction.  */
-static inline uint8_t sub8_sat(uint8_t a, uint8_t b)
-{
-    uint8_t res;
-
-    res = a - b;
-    if (((res ^ a) & 0x80) && ((a ^ b) & 0x80)) {
-        if (a & 0x80) {
-            res = 0x80;
-        } else {
-            res = 0x7f;
-        }
-    }
-    return res;
-}
-
-#define ADD16(a, b, n) RESULT(add16_sat(a, b), n, 16);
-#define SUB16(a, b, n) RESULT(sub16_sat(a, b), n, 16);
-#define ADD8(a, b, n)  RESULT(add8_sat(a, b), n, 8);
-#define SUB8(a, b, n)  RESULT(sub8_sat(a, b), n, 8);
-#define PFX q
-
-#include "op_addsub.h"
-
-/* Unsigned saturating arithmetic.  */
-static inline uint16_t add16_usat(uint16_t a, uint16_t b)
-{
-    uint16_t res;
-    res = a + b;
-    if (res < a) {
-        res = 0xffff;
-    }
-    return res;
-}
-
-static inline uint16_t sub16_usat(uint16_t a, uint16_t b)
-{
-    if (a > b) {
-        return a - b;
-    } else {
-        return 0;
-    }
-}
-
-static inline uint8_t add8_usat(uint8_t a, uint8_t b)
-{
-    uint8_t res;
-    res = a + b;
-    if (res < a) {
-        res = 0xff;
-    }
-    return res;
-}
-
-static inline uint8_t sub8_usat(uint8_t a, uint8_t b)
-{
-    if (a > b) {
-        return a - b;
-    } else {
-        return 0;
-    }
-}
-
-#define ADD16(a, b, n) RESULT(add16_usat(a, b), n, 16);
-#define SUB16(a, b, n) RESULT(sub16_usat(a, b), n, 16);
-#define ADD8(a, b, n)  RESULT(add8_usat(a, b), n, 8);
-#define SUB8(a, b, n)  RESULT(sub8_usat(a, b), n, 8);
-#define PFX uq
-
-#include "op_addsub.h"
-
-/* Signed modulo arithmetic.  */
-#define SARITH16(a, b, n, op) do { \
-    int32_t sum; \
-    sum = (int32_t)(int16_t)(a) op (int32_t)(int16_t)(b); \
-    RESULT(sum, n, 16); \
-    if (sum >= 0) \
-        ge |= 3 << (n * 2); \
-    } while (0)
-
-#define SARITH8(a, b, n, op) do { \
-    int32_t sum; \
-    sum = (int32_t)(int8_t)(a) op (int32_t)(int8_t)(b); \
-    RESULT(sum, n, 8); \
-    if (sum >= 0) \
-        ge |= 1 << n; \
-    } while (0)
-
-
-#define ADD16(a, b, n) SARITH16(a, b, n, +)
-#define SUB16(a, b, n) SARITH16(a, b, n, -)
-#define ADD8(a, b, n)  SARITH8(a, b, n, +)
-#define SUB8(a, b, n)  SARITH8(a, b, n, -)
-#define PFX s
-#define ARITH_GE
-
-#include "op_addsub.h"
-
-/* Unsigned modulo arithmetic.  */
-#define ADD16(a, b, n) do { \
-    uint32_t sum; \
-    sum = (uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b); \
-    RESULT(sum, n, 16); \
-    if ((sum >> 16) == 1) \
-        ge |= 3 << (n * 2); \
-    } while (0)
-
-#define ADD8(a, b, n) do { \
-    uint32_t sum; \
-    sum = (uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b); \
-    RESULT(sum, n, 8); \
-    if ((sum >> 8) == 1) \
-        ge |= 1 << n; \
-    } while (0)
-
-#define SUB16(a, b, n) do { \
-    uint32_t sum; \
-    sum = (uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b); \
-    RESULT(sum, n, 16); \
-    if ((sum >> 16) == 0) \
-        ge |= 3 << (n * 2); \
-    } while (0)
-
-#define SUB8(a, b, n) do { \
-    uint32_t sum; \
-    sum = (uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b); \
-    RESULT(sum, n, 8); \
-    if ((sum >> 8) == 0) \
-        ge |= 1 << n; \
-    } while (0)
-
-#define PFX u
-#define ARITH_GE
-
-#include "op_addsub.h"
-
-/* Halved signed arithmetic.  */
-#define ADD16(a, b, n) \
-  RESULT(((int32_t)(int16_t)(a) + (int32_t)(int16_t)(b)) >> 1, n, 16)
-#define SUB16(a, b, n) \
-  RESULT(((int32_t)(int16_t)(a) - (int32_t)(int16_t)(b)) >> 1, n, 16)
-#define ADD8(a, b, n) \
-  RESULT(((int32_t)(int8_t)(a) + (int32_t)(int8_t)(b)) >> 1, n, 8)
-#define SUB8(a, b, n) \
-  RESULT(((int32_t)(int8_t)(a) - (int32_t)(int8_t)(b)) >> 1, n, 8)
-#define PFX sh
-
-#include "op_addsub.h"
-
-/* Halved unsigned arithmetic.  */
-#define ADD16(a, b, n) \
-  RESULT(((uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b)) >> 1, n, 16)
-#define SUB16(a, b, n) \
-  RESULT(((uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b)) >> 1, n, 16)
-#define ADD8(a, b, n) \
-  RESULT(((uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b)) >> 1, n, 8)
-#define SUB8(a, b, n) \
-  RESULT(((uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b)) >> 1, n, 8)
-#define PFX uh
-
-#include "op_addsub.h"
-
-static inline uint8_t do_usad(uint8_t a, uint8_t b)
-{
-    if (a > b) {
-        return a - b;
-    } else {
-        return b - a;
-    }
-}
-
-/* Unsigned sum of absolute byte differences.  */
-uint32_t HELPER(usad8)(uint32_t a, uint32_t b)
-{
-    uint32_t sum;
-    sum = do_usad(a, b);
-    sum += do_usad(a >> 8, b >> 8);
-    sum += do_usad(a >> 16, b >> 16);
-    sum += do_usad(a >> 24, b >> 24);
-    return sum;
-}
-
-/* For ARMv6 SEL instruction.  */
-uint32_t HELPER(sel_flags)(uint32_t flags, uint32_t a, uint32_t b)
-{
-    uint32_t mask;
-
-    mask = 0;
-    if (flags & 1) {
-        mask |= 0xff;
-    }
-    if (flags & 2) {
-        mask |= 0xff00;
-    }
-    if (flags & 4) {
-        mask |= 0xff0000;
-    }
-    if (flags & 8) {
-        mask |= 0xff000000;
-    }
-    return (a & mask) | (b & ~mask);
-}
-
-/*
- * CRC helpers.
- * The upper bytes of val (above the number specified by 'bytes') must have
- * been zeroed out by the caller.
- */
-uint32_t HELPER(crc32)(uint32_t acc, uint32_t val, uint32_t bytes)
-{
-    uint8_t buf[4];
-
-    stl_le_p(buf, val);
-
-    /* zlib crc32 converts the accumulator and output to one's complement.  */
-    return crc32(acc ^ 0xffffffff, buf, bytes) ^ 0xffffffff;
-}
-
-uint32_t HELPER(crc32c)(uint32_t acc, uint32_t val, uint32_t bytes)
-{
-    uint8_t buf[4];
-
-    stl_le_p(buf, val);
-
-    /* Linux crc32c converts the output to one's complement.  */
-    return crc32c(acc, buf, bytes) ^ 0xffffffff;
-}
 
 /*
  * Return the exception level to which FP-disabled exceptions should
diff --git a/target/arm/kvm.c b/target/arm/kvm.c
index a9444a2c7a..da30bdbb23 100644
--- a/target/arm/kvm.c
+++ b/target/arm/kvm.c
@@ -2387,7 +2387,7 @@ void kvm_arch_on_sigbus_vcpu(CPUState *c, int code, void *addr)
              */
             if (code == BUS_MCEERR_AR) {
                 kvm_cpu_synchronize_state(c);
-                if (!acpi_ghes_record_errors(ACPI_HEST_SRC_ID_SEA, paddr)) {
+                if (!acpi_ghes_memory_errors(ACPI_HEST_SRC_ID_SEA, paddr)) {
                     kvm_inject_arm_sea(c);
                 } else {
                     error_report("failed to record the error");
diff --git a/target/arm/tcg/arith_helper.c b/target/arm/tcg/arith_helper.c
new file mode 100644
index 0000000000..9a555c7966
--- /dev/null
+++ b/target/arm/tcg/arith_helper.c
@@ -0,0 +1,296 @@
+/*
+ * ARM generic helpers for various arithmetical operations.
+ *
+ * This code is licensed under the GNU GPL v2 or later.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "exec/helper-proto.h"
+#include "qemu/crc32c.h"
+#include <zlib.h> /* for crc32 */
+
+/*
+ * Note that signed overflow is undefined in C.  The following routines are
+ * careful to use unsigned types where modulo arithmetic is required.
+ * Failure to do so _will_ break on newer gcc.
+ */
+
+/* Signed saturating arithmetic.  */
+
+/* Perform 16-bit signed saturating addition.  */
+static inline uint16_t add16_sat(uint16_t a, uint16_t b)
+{
+    uint16_t res;
+
+    res = a + b;
+    if (((res ^ a) & 0x8000) && !((a ^ b) & 0x8000)) {
+        if (a & 0x8000) {
+            res = 0x8000;
+        } else {
+            res = 0x7fff;
+        }
+    }
+    return res;
+}
+
+/* Perform 8-bit signed saturating addition.  */
+static inline uint8_t add8_sat(uint8_t a, uint8_t b)
+{
+    uint8_t res;
+
+    res = a + b;
+    if (((res ^ a) & 0x80) && !((a ^ b) & 0x80)) {
+        if (a & 0x80) {
+            res = 0x80;
+        } else {
+            res = 0x7f;
+        }
+    }
+    return res;
+}
+
+/* Perform 16-bit signed saturating subtraction.  */
+static inline uint16_t sub16_sat(uint16_t a, uint16_t b)
+{
+    uint16_t res;
+
+    res = a - b;
+    if (((res ^ a) & 0x8000) && ((a ^ b) & 0x8000)) {
+        if (a & 0x8000) {
+            res = 0x8000;
+        } else {
+            res = 0x7fff;
+        }
+    }
+    return res;
+}
+
+/* Perform 8-bit signed saturating subtraction.  */
+static inline uint8_t sub8_sat(uint8_t a, uint8_t b)
+{
+    uint8_t res;
+
+    res = a - b;
+    if (((res ^ a) & 0x80) && ((a ^ b) & 0x80)) {
+        if (a & 0x80) {
+            res = 0x80;
+        } else {
+            res = 0x7f;
+        }
+    }
+    return res;
+}
+
+#define ADD16(a, b, n) RESULT(add16_sat(a, b), n, 16);
+#define SUB16(a, b, n) RESULT(sub16_sat(a, b), n, 16);
+#define ADD8(a, b, n)  RESULT(add8_sat(a, b), n, 8);
+#define SUB8(a, b, n)  RESULT(sub8_sat(a, b), n, 8);
+#define PFX q
+
+#include "op_addsub.c.inc"
+
+/* Unsigned saturating arithmetic.  */
+static inline uint16_t add16_usat(uint16_t a, uint16_t b)
+{
+    uint16_t res;
+    res = a + b;
+    if (res < a) {
+        res = 0xffff;
+    }
+    return res;
+}
+
+static inline uint16_t sub16_usat(uint16_t a, uint16_t b)
+{
+    if (a > b) {
+        return a - b;
+    } else {
+        return 0;
+    }
+}
+
+static inline uint8_t add8_usat(uint8_t a, uint8_t b)
+{
+    uint8_t res;
+    res = a + b;
+    if (res < a) {
+        res = 0xff;
+    }
+    return res;
+}
+
+static inline uint8_t sub8_usat(uint8_t a, uint8_t b)
+{
+    if (a > b) {
+        return a - b;
+    } else {
+        return 0;
+    }
+}
+
+#define ADD16(a, b, n) RESULT(add16_usat(a, b), n, 16);
+#define SUB16(a, b, n) RESULT(sub16_usat(a, b), n, 16);
+#define ADD8(a, b, n)  RESULT(add8_usat(a, b), n, 8);
+#define SUB8(a, b, n)  RESULT(sub8_usat(a, b), n, 8);
+#define PFX uq
+
+#include "op_addsub.c.inc"
+
+/* Signed modulo arithmetic.  */
+#define SARITH16(a, b, n, op) do { \
+    int32_t sum; \
+    sum = (int32_t)(int16_t)(a) op (int32_t)(int16_t)(b); \
+    RESULT(sum, n, 16); \
+    if (sum >= 0) \
+        ge |= 3 << (n * 2); \
+    } while (0)
+
+#define SARITH8(a, b, n, op) do { \
+    int32_t sum; \
+    sum = (int32_t)(int8_t)(a) op (int32_t)(int8_t)(b); \
+    RESULT(sum, n, 8); \
+    if (sum >= 0) \
+        ge |= 1 << n; \
+    } while (0)
+
+
+#define ADD16(a, b, n) SARITH16(a, b, n, +)
+#define SUB16(a, b, n) SARITH16(a, b, n, -)
+#define ADD8(a, b, n)  SARITH8(a, b, n, +)
+#define SUB8(a, b, n)  SARITH8(a, b, n, -)
+#define PFX s
+#define ARITH_GE
+
+#include "op_addsub.c.inc"
+
+/* Unsigned modulo arithmetic.  */
+#define ADD16(a, b, n) do { \
+    uint32_t sum; \
+    sum = (uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b); \
+    RESULT(sum, n, 16); \
+    if ((sum >> 16) == 1) \
+        ge |= 3 << (n * 2); \
+    } while (0)
+
+#define ADD8(a, b, n) do { \
+    uint32_t sum; \
+    sum = (uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b); \
+    RESULT(sum, n, 8); \
+    if ((sum >> 8) == 1) \
+        ge |= 1 << n; \
+    } while (0)
+
+#define SUB16(a, b, n) do { \
+    uint32_t sum; \
+    sum = (uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b); \
+    RESULT(sum, n, 16); \
+    if ((sum >> 16) == 0) \
+        ge |= 3 << (n * 2); \
+    } while (0)
+
+#define SUB8(a, b, n) do { \
+    uint32_t sum; \
+    sum = (uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b); \
+    RESULT(sum, n, 8); \
+    if ((sum >> 8) == 0) \
+        ge |= 1 << n; \
+    } while (0)
+
+#define PFX u
+#define ARITH_GE
+
+#include "op_addsub.c.inc"
+
+/* Halved signed arithmetic.  */
+#define ADD16(a, b, n) \
+  RESULT(((int32_t)(int16_t)(a) + (int32_t)(int16_t)(b)) >> 1, n, 16)
+#define SUB16(a, b, n) \
+  RESULT(((int32_t)(int16_t)(a) - (int32_t)(int16_t)(b)) >> 1, n, 16)
+#define ADD8(a, b, n) \
+  RESULT(((int32_t)(int8_t)(a) + (int32_t)(int8_t)(b)) >> 1, n, 8)
+#define SUB8(a, b, n) \
+  RESULT(((int32_t)(int8_t)(a) - (int32_t)(int8_t)(b)) >> 1, n, 8)
+#define PFX sh
+
+#include "op_addsub.c.inc"
+
+/* Halved unsigned arithmetic.  */
+#define ADD16(a, b, n) \
+  RESULT(((uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b)) >> 1, n, 16)
+#define SUB16(a, b, n) \
+  RESULT(((uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b)) >> 1, n, 16)
+#define ADD8(a, b, n) \
+  RESULT(((uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b)) >> 1, n, 8)
+#define SUB8(a, b, n) \
+  RESULT(((uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b)) >> 1, n, 8)
+#define PFX uh
+
+#include "op_addsub.c.inc"
+
+static inline uint8_t do_usad(uint8_t a, uint8_t b)
+{
+    if (a > b) {
+        return a - b;
+    } else {
+        return b - a;
+    }
+}
+
+/* Unsigned sum of absolute byte differences.  */
+uint32_t HELPER(usad8)(uint32_t a, uint32_t b)
+{
+    uint32_t sum;
+    sum = do_usad(a, b);
+    sum += do_usad(a >> 8, b >> 8);
+    sum += do_usad(a >> 16, b >> 16);
+    sum += do_usad(a >> 24, b >> 24);
+    return sum;
+}
+
+/* For ARMv6 SEL instruction.  */
+uint32_t HELPER(sel_flags)(uint32_t flags, uint32_t a, uint32_t b)
+{
+    uint32_t mask;
+
+    mask = 0;
+    if (flags & 1) {
+        mask |= 0xff;
+    }
+    if (flags & 2) {
+        mask |= 0xff00;
+    }
+    if (flags & 4) {
+        mask |= 0xff0000;
+    }
+    if (flags & 8) {
+        mask |= 0xff000000;
+    }
+    return (a & mask) | (b & ~mask);
+}
+
+/*
+ * CRC helpers.
+ * The upper bytes of val (above the number specified by 'bytes') must have
+ * been zeroed out by the caller.
+ */
+uint32_t HELPER(crc32)(uint32_t acc, uint32_t val, uint32_t bytes)
+{
+    uint8_t buf[4];
+
+    stl_le_p(buf, val);
+
+    /* zlib crc32 converts the accumulator and output to one's complement.  */
+    return crc32(acc ^ 0xffffffff, buf, bytes) ^ 0xffffffff;
+}
+
+uint32_t HELPER(crc32c)(uint32_t acc, uint32_t val, uint32_t bytes)
+{
+    uint8_t buf[4];
+
+    stl_le_p(buf, val);
+
+    /* Linux crc32c converts the output to one's complement.  */
+    return crc32c(acc, buf, bytes) ^ 0xffffffff;
+}
diff --git a/target/arm/tcg/meson.build b/target/arm/tcg/meson.build
index 09238989c5..1f9077c372 100644
--- a/target/arm/tcg/meson.build
+++ b/target/arm/tcg/meson.build
@@ -40,6 +40,7 @@ arm_ss.add(files(
   'tlb_helper.c',
   'vec_helper.c',
   'tlb-insns.c',
+  'arith_helper.c',
 ))
 
 arm_ss.add(when: 'TARGET_AARCH64', if_true: files(
diff --git a/target/arm/op_addsub.h b/target/arm/tcg/op_addsub.c.inc
index ca4a1893c3..ca4a1893c3 100644
--- a/target/arm/op_addsub.h
+++ b/target/arm/tcg/op_addsub.c.inc