Patchwork [10/25] x86: split condition code and shift templates

login
register
mail settings
Submitter Blue Swirl
Date June 9, 2012, 4:19 p.m.
Message ID <9f36e1649da6f3677ede376999ca2d80fae3092c.1339258553.git.blauwirbel@gmail.com>
Download mbox | patch
Permalink /patch/163950/
State New
Headers show

Comments

Blue Swirl - June 9, 2012, 4:19 p.m.
Move shift templates from helper_template.h to
shift_helper_template.h and the condition code helpers
to cc_helper_template.h.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
---
 target-i386/cc_helper_template.h    |  277 +++++++++++++++++++++++++++
 target-i386/helper_template.h       |  353 -----------------------------------
 target-i386/op_helper.c             |   26 +++-
 target-i386/shift_helper_template.h |  110 +++++++++++
 4 files changed, 409 insertions(+), 357 deletions(-)
 create mode 100644 target-i386/cc_helper_template.h
 delete mode 100644 target-i386/helper_template.h
 create mode 100644 target-i386/shift_helper_template.h

Patch

diff --git a/target-i386/cc_helper_template.h b/target-i386/cc_helper_template.h
new file mode 100644
index 0000000..ff22830
--- /dev/null
+++ b/target-i386/cc_helper_template.h
@@ -0,0 +1,277 @@ 
+/*
+ *  x86 condition code helpers
+ *
+ *  Copyright (c) 2008 Fabrice Bellard
+ *
+ * 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/>.
+ */
+
+#define DATA_BITS (1 << (3 + SHIFT))
+#define SIGN_MASK (((target_ulong)1) << (DATA_BITS - 1))
+
+#if DATA_BITS == 8
+#define SUFFIX b
+#define DATA_TYPE uint8_t
+#define DATA_MASK 0xff
+#elif DATA_BITS == 16
+#define SUFFIX w
+#define DATA_TYPE uint16_t
+#define DATA_MASK 0xffff
+#elif DATA_BITS == 32
+#define SUFFIX l
+#define DATA_TYPE uint32_t
+#define DATA_MASK 0xffffffff
+#elif DATA_BITS == 64
+#define SUFFIX q
+#define DATA_TYPE uint64_t
+#define DATA_MASK 0xffffffffffffffffULL
+#else
+#error unhandled operand size
+#endif
+
+/* dynamic flags computation */
+
+static int glue(compute_all_add, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_SRC;
+    src2 = CC_DST - CC_SRC;
+    cf = (DATA_TYPE)CC_DST < (DATA_TYPE)src1;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = lshift((src1 ^ src2 ^ -1) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_add, SUFFIX)(void)
+{
+    int cf;
+    target_long src1;
+
+    src1 = CC_SRC;
+    cf = (DATA_TYPE)CC_DST < (DATA_TYPE)src1;
+    return cf;
+}
+
+static int glue(compute_all_adc, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_SRC;
+    src2 = CC_DST - CC_SRC - 1;
+    cf = (DATA_TYPE)CC_DST <= (DATA_TYPE)src1;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = lshift((src1 ^ src2 ^ -1) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_adc, SUFFIX)(void)
+{
+    int cf;
+    target_long src1;
+
+    src1 = CC_SRC;
+    cf = (DATA_TYPE)CC_DST <= (DATA_TYPE)src1;
+    return cf;
+}
+
+static int glue(compute_all_sub, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_DST + CC_SRC;
+    src2 = CC_SRC;
+    cf = (DATA_TYPE)src1 < (DATA_TYPE)src2;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = lshift((src1 ^ src2) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_sub, SUFFIX)(void)
+{
+    int cf;
+    target_long src1, src2;
+
+    src1 = CC_DST + CC_SRC;
+    src2 = CC_SRC;
+    cf = (DATA_TYPE)src1 < (DATA_TYPE)src2;
+    return cf;
+}
+
+static int glue(compute_all_sbb, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_DST + CC_SRC + 1;
+    src2 = CC_SRC;
+    cf = (DATA_TYPE)src1 <= (DATA_TYPE)src2;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = lshift((src1 ^ src2) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_sbb, SUFFIX)(void)
+{
+    int cf;
+    target_long src1, src2;
+
+    src1 = CC_DST + CC_SRC + 1;
+    src2 = CC_SRC;
+    cf = (DATA_TYPE)src1 <= (DATA_TYPE)src2;
+    return cf;
+}
+
+static int glue(compute_all_logic, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+
+    cf = 0;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = 0;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = 0;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_logic, SUFFIX)(void)
+{
+    return 0;
+}
+
+static int glue(compute_all_inc, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_DST - 1;
+    src2 = 1;
+    cf = CC_SRC;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = ((CC_DST & DATA_MASK) == SIGN_MASK) << 11;
+    return cf | pf | af | zf | sf | of;
+}
+
+#if DATA_BITS == 32
+static int glue(compute_c_inc, SUFFIX)(void)
+{
+    return CC_SRC;
+}
+#endif
+
+static int glue(compute_all_dec, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+    target_long src1, src2;
+
+    src1 = CC_DST + 1;
+    src2 = 1;
+    cf = CC_SRC;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = (CC_DST ^ src1 ^ src2) & 0x10;
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = ((CC_DST & DATA_MASK) == ((target_ulong)SIGN_MASK - 1)) << 11;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_all_shl, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+
+    cf = (CC_SRC >> (DATA_BITS - 1)) & CC_C;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = 0; /* undefined */
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    /* of is defined if shift count == 1 */
+    of = lshift(CC_SRC ^ CC_DST, 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+static int glue(compute_c_shl, SUFFIX)(void)
+{
+    return (CC_SRC >> (DATA_BITS - 1)) & CC_C;
+}
+
+#if DATA_BITS == 32
+static int glue(compute_c_sar, SUFFIX)(void)
+{
+    return CC_SRC & 1;
+}
+#endif
+
+static int glue(compute_all_sar, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+
+    cf = CC_SRC & 1;
+    pf = parity_table[(uint8_t)CC_DST];
+    af = 0; /* undefined */
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    /* of is defined if shift count == 1 */
+    of = lshift(CC_SRC ^ CC_DST, 12 - DATA_BITS) & CC_O;
+    return cf | pf | af | zf | sf | of;
+}
+
+#if DATA_BITS == 32
+static int glue(compute_c_mul, SUFFIX)(void)
+{
+    int cf;
+
+    cf = (CC_SRC != 0);
+    return cf;
+}
+#endif
+
+/* NOTE: we compute the flags like the P4. On olders CPUs, only OF and
+   CF are modified and it is slower to do that. */
+static int glue(compute_all_mul, SUFFIX)(void)
+{
+    int cf, pf, af, zf, sf, of;
+
+    cf = (CC_SRC != 0);
+    pf = parity_table[(uint8_t)CC_DST];
+    af = 0; /* undefined */
+    zf = ((DATA_TYPE)CC_DST == 0) << 6;
+    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
+    of = cf << 11;
+    return cf | pf | af | zf | sf | of;
+}
+
+#undef DATA_BITS
+#undef SIGN_MASK
+#undef DATA_TYPE
+#undef DATA_MASK
+#undef SUFFIX
diff --git a/target-i386/helper_template.h b/target-i386/helper_template.h
deleted file mode 100644
index 8b2d4f0..0000000
--- a/target-i386/helper_template.h
+++ /dev/null
@@ -1,353 +0,0 @@ 
-/*
- *  i386 helpers
- *
- *  Copyright (c) 2008 Fabrice Bellard
- *
- * 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/>.
- */
-
-#define DATA_BITS (1 << (3 + SHIFT))
-#define SHIFT_MASK (DATA_BITS - 1)
-#define SIGN_MASK (((target_ulong)1) << (DATA_BITS - 1))
-#if DATA_BITS <= 32
-#define SHIFT1_MASK 0x1f
-#else
-#define SHIFT1_MASK 0x3f
-#endif
-
-#if DATA_BITS == 8
-#define SUFFIX b
-#define DATA_TYPE uint8_t
-#define DATA_STYPE int8_t
-#define DATA_MASK 0xff
-#elif DATA_BITS == 16
-#define SUFFIX w
-#define DATA_TYPE uint16_t
-#define DATA_STYPE int16_t
-#define DATA_MASK 0xffff
-#elif DATA_BITS == 32
-#define SUFFIX l
-#define DATA_TYPE uint32_t
-#define DATA_STYPE int32_t
-#define DATA_MASK 0xffffffff
-#elif DATA_BITS == 64
-#define SUFFIX q
-#define DATA_TYPE uint64_t
-#define DATA_STYPE int64_t
-#define DATA_MASK 0xffffffffffffffffULL
-#else
-#error unhandled operand size
-#endif
-
-/* dynamic flags computation */
-
-static int glue(compute_all_add, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_SRC;
-    src2 = CC_DST - CC_SRC;
-    cf = (DATA_TYPE)CC_DST < (DATA_TYPE)src1;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = lshift((src1 ^ src2 ^ -1) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_add, SUFFIX)(void)
-{
-    int cf;
-    target_long src1;
-
-    src1 = CC_SRC;
-    cf = (DATA_TYPE)CC_DST < (DATA_TYPE)src1;
-    return cf;
-}
-
-static int glue(compute_all_adc, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_SRC;
-    src2 = CC_DST - CC_SRC - 1;
-    cf = (DATA_TYPE)CC_DST <= (DATA_TYPE)src1;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = lshift((src1 ^ src2 ^ -1) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_adc, SUFFIX)(void)
-{
-    int cf;
-    target_long src1;
-
-    src1 = CC_SRC;
-    cf = (DATA_TYPE)CC_DST <= (DATA_TYPE)src1;
-    return cf;
-}
-
-static int glue(compute_all_sub, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_DST + CC_SRC;
-    src2 = CC_SRC;
-    cf = (DATA_TYPE)src1 < (DATA_TYPE)src2;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = lshift((src1 ^ src2) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_sub, SUFFIX)(void)
-{
-    int cf;
-    target_long src1, src2;
-
-    src1 = CC_DST + CC_SRC;
-    src2 = CC_SRC;
-    cf = (DATA_TYPE)src1 < (DATA_TYPE)src2;
-    return cf;
-}
-
-static int glue(compute_all_sbb, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_DST + CC_SRC + 1;
-    src2 = CC_SRC;
-    cf = (DATA_TYPE)src1 <= (DATA_TYPE)src2;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = lshift((src1 ^ src2) & (src1 ^ CC_DST), 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_sbb, SUFFIX)(void)
-{
-    int cf;
-    target_long src1, src2;
-
-    src1 = CC_DST + CC_SRC + 1;
-    src2 = CC_SRC;
-    cf = (DATA_TYPE)src1 <= (DATA_TYPE)src2;
-    return cf;
-}
-
-static int glue(compute_all_logic, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-
-    cf = 0;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = 0;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = 0;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_logic, SUFFIX)(void)
-{
-    return 0;
-}
-
-static int glue(compute_all_inc, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_DST - 1;
-    src2 = 1;
-    cf = CC_SRC;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = ((CC_DST & DATA_MASK) == SIGN_MASK) << 11;
-    return cf | pf | af | zf | sf | of;
-}
-
-#if DATA_BITS == 32
-static int glue(compute_c_inc, SUFFIX)(void)
-{
-    return CC_SRC;
-}
-#endif
-
-static int glue(compute_all_dec, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-    target_long src1, src2;
-
-    src1 = CC_DST + 1;
-    src2 = 1;
-    cf = CC_SRC;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = (CC_DST ^ src1 ^ src2) & 0x10;
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = ((CC_DST & DATA_MASK) == ((target_ulong)SIGN_MASK - 1)) << 11;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_all_shl, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-
-    cf = (CC_SRC >> (DATA_BITS - 1)) & CC_C;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = 0; /* undefined */
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    /* of is defined if shift count == 1 */
-    of = lshift(CC_SRC ^ CC_DST, 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-static int glue(compute_c_shl, SUFFIX)(void)
-{
-    return (CC_SRC >> (DATA_BITS - 1)) & CC_C;
-}
-
-#if DATA_BITS == 32
-static int glue(compute_c_sar, SUFFIX)(void)
-{
-    return CC_SRC & 1;
-}
-#endif
-
-static int glue(compute_all_sar, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-
-    cf = CC_SRC & 1;
-    pf = parity_table[(uint8_t)CC_DST];
-    af = 0; /* undefined */
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    /* of is defined if shift count == 1 */
-    of = lshift(CC_SRC ^ CC_DST, 12 - DATA_BITS) & CC_O;
-    return cf | pf | af | zf | sf | of;
-}
-
-#if DATA_BITS == 32
-static int glue(compute_c_mul, SUFFIX)(void)
-{
-    int cf;
-
-    cf = (CC_SRC != 0);
-    return cf;
-}
-#endif
-
-/* NOTE: we compute the flags like the P4. On olders CPUs, only OF and
-   CF are modified and it is slower to do that. */
-static int glue(compute_all_mul, SUFFIX)(void)
-{
-    int cf, pf, af, zf, sf, of;
-
-    cf = (CC_SRC != 0);
-    pf = parity_table[(uint8_t)CC_DST];
-    af = 0; /* undefined */
-    zf = ((DATA_TYPE)CC_DST == 0) << 6;
-    sf = lshift(CC_DST, 8 - DATA_BITS) & 0x80;
-    of = cf << 11;
-    return cf | pf | af | zf | sf | of;
-}
-
-/* shifts */
-
-target_ulong glue(helper_rcl, SUFFIX)(target_ulong t0, target_ulong t1)
-{
-    int count, eflags;
-    target_ulong src;
-    target_long res;
-
-    count = t1 & SHIFT1_MASK;
-#if DATA_BITS == 16
-    count = rclw_table[count];
-#elif DATA_BITS == 8
-    count = rclb_table[count];
-#endif
-    if (count) {
-        eflags = helper_cc_compute_all(CC_OP);
-        t0 &= DATA_MASK;
-        src = t0;
-        res = (t0 << count) | ((target_ulong)(eflags & CC_C) << (count - 1));
-        if (count > 1) {
-            res |= t0 >> (DATA_BITS + 1 - count);
-        }
-        t0 = res;
-        env->cc_tmp = (eflags & ~(CC_C | CC_O)) |
-            (lshift(src ^ t0, 11 - (DATA_BITS - 1)) & CC_O) |
-            ((src >> (DATA_BITS - count)) & CC_C);
-    } else {
-        env->cc_tmp = -1;
-    }
-    return t0;
-}
-
-target_ulong glue(helper_rcr, SUFFIX)(target_ulong t0, target_ulong t1)
-{
-    int count, eflags;
-    target_ulong src;
-    target_long res;
-
-    count = t1 & SHIFT1_MASK;
-#if DATA_BITS == 16
-    count = rclw_table[count];
-#elif DATA_BITS == 8
-    count = rclb_table[count];
-#endif
-    if (count) {
-        eflags = helper_cc_compute_all(CC_OP);
-        t0 &= DATA_MASK;
-        src = t0;
-        res = (t0 >> count) |
-            ((target_ulong)(eflags & CC_C) << (DATA_BITS - count));
-        if (count > 1) {
-            res |= t0 << (DATA_BITS + 1 - count);
-        }
-        t0 = res;
-        env->cc_tmp = (eflags & ~(CC_C | CC_O)) |
-            (lshift(src ^ t0, 11 - (DATA_BITS - 1)) & CC_O) |
-            ((src >> (count - 1)) & CC_C);
-    } else {
-        env->cc_tmp = -1;
-    }
-    return t0;
-}
-
-#undef DATA_BITS
-#undef SHIFT_MASK
-#undef SHIFT1_MASK
-#undef SIGN_MASK
-#undef DATA_TYPE
-#undef DATA_STYPE
-#undef DATA_MASK
-#undef SUFFIX
diff --git a/target-i386/op_helper.c b/target-i386/op_helper.c
index 58d2cc4..a75687e 100644
--- a/target-i386/op_helper.c
+++ b/target-i386/op_helper.c
@@ -4715,25 +4715,43 @@  void cpu_vmexit(CPUX86State *nenv, uint32_t exit_code, uint64_t exit_info_1)
 #endif
 
 #define SHIFT 0
-#include "helper_template.h"
+#include "cc_helper_template.h"
 #undef SHIFT
 
 #define SHIFT 1
-#include "helper_template.h"
+#include "cc_helper_template.h"
 #undef SHIFT
 
 #define SHIFT 2
-#include "helper_template.h"
+#include "cc_helper_template.h"
 #undef SHIFT
 
 #ifdef TARGET_X86_64
 
 #define SHIFT 3
-#include "helper_template.h"
+#include "cc_helper_template.h"
 #undef SHIFT
 
 #endif
 
+#define SHIFT 0
+#include "shift_helper_template.h"
+#undef SHIFT
+
+#define SHIFT 1
+#include "shift_helper_template.h"
+#undef SHIFT
+
+#define SHIFT 2
+#include "shift_helper_template.h"
+#undef SHIFT
+
+#ifdef TARGET_X86_64
+#define SHIFT 3
+#include "shift_helper_template.h"
+#undef SHIFT
+#endif
+
 /* temporary wrappers */
 #if defined(CONFIG_USER_ONLY)
 #define ldub_data(addr) ldub_raw(addr)
diff --git a/target-i386/shift_helper_template.h b/target-i386/shift_helper_template.h
new file mode 100644
index 0000000..239ee09
--- /dev/null
+++ b/target-i386/shift_helper_template.h
@@ -0,0 +1,110 @@ 
+/*
+ *  x86 shift helpers
+ *
+ *  Copyright (c) 2008 Fabrice Bellard
+ *
+ * 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/>.
+ */
+
+#define DATA_BITS (1 << (3 + SHIFT))
+#define SHIFT_MASK (DATA_BITS - 1)
+#if DATA_BITS <= 32
+#define SHIFT1_MASK 0x1f
+#else
+#define SHIFT1_MASK 0x3f
+#endif
+
+#if DATA_BITS == 8
+#define SUFFIX b
+#define DATA_MASK 0xff
+#elif DATA_BITS == 16
+#define SUFFIX w
+#define DATA_MASK 0xffff
+#elif DATA_BITS == 32
+#define SUFFIX l
+#define DATA_MASK 0xffffffff
+#elif DATA_BITS == 64
+#define SUFFIX q
+#define DATA_MASK 0xffffffffffffffffULL
+#else
+#error unhandled operand size
+#endif
+
+target_ulong glue(helper_rcl, SUFFIX)(target_ulong t0, target_ulong t1)
+{
+    int count, eflags;
+    target_ulong src;
+    target_long res;
+
+    count = t1 & SHIFT1_MASK;
+#if DATA_BITS == 16
+    count = rclw_table[count];
+#elif DATA_BITS == 8
+    count = rclb_table[count];
+#endif
+    if (count) {
+        eflags = helper_cc_compute_all(CC_OP);
+        t0 &= DATA_MASK;
+        src = t0;
+        res = (t0 << count) | ((target_ulong)(eflags & CC_C) << (count - 1));
+        if (count > 1) {
+            res |= t0 >> (DATA_BITS + 1 - count);
+        }
+        t0 = res;
+        env->cc_tmp = (eflags & ~(CC_C | CC_O)) |
+            (lshift(src ^ t0, 11 - (DATA_BITS - 1)) & CC_O) |
+            ((src >> (DATA_BITS - count)) & CC_C);
+    } else {
+        env->cc_tmp = -1;
+    }
+    return t0;
+}
+
+target_ulong glue(helper_rcr, SUFFIX)(target_ulong t0, target_ulong t1)
+{
+    int count, eflags;
+    target_ulong src;
+    target_long res;
+
+    count = t1 & SHIFT1_MASK;
+#if DATA_BITS == 16
+    count = rclw_table[count];
+#elif DATA_BITS == 8
+    count = rclb_table[count];
+#endif
+    if (count) {
+        eflags = helper_cc_compute_all(CC_OP);
+        t0 &= DATA_MASK;
+        src = t0;
+        res = (t0 >> count) |
+            ((target_ulong)(eflags & CC_C) << (DATA_BITS - count));
+        if (count > 1) {
+            res |= t0 << (DATA_BITS + 1 - count);
+        }
+        t0 = res;
+        env->cc_tmp = (eflags & ~(CC_C | CC_O)) |
+            (lshift(src ^ t0, 11 - (DATA_BITS - 1)) & CC_O) |
+            ((src >> (count - 1)) & CC_C);
+    } else {
+        env->cc_tmp = -1;
+    }
+    return t0;
+}
+
+#undef DATA_BITS
+#undef SHIFT_MASK
+#undef SHIFT1_MASK
+#undef DATA_TYPE
+#undef DATA_MASK
+#undef SUFFIX