Merge branch 's390-reorg' of git://repo.or.cz/qemu/rth

* 's390-reorg' of git://repo.or.cz/qemu/rth: (149 commits) target-s390: Claim maintainership target-s390: Use noreturn for exception and load_psw target-s390: Use TCG_CALL_NO_WG for misc helpers target-s390: Use TCG_CALL_NO_WG for integer helpers target-s390: Use TCG_CALL_NO_WG for floating-point helpers target-s390: Use TCG_CALL_NO_WG for memory helpers target-s390: Perform COMPARE AND SWAP inline target-s390: Optimize get_address target-s390: Optimize ADDC/SUBB target-s390: Optimize ADDU/SUBU CC testing target-s390: Tidy comparisons target-s390: Optmize emitting discards target-s390: Optimize XC target-s390: Fix cpu_clone_regs target-s390: Implement LOAD/SET FP AND SIGNAL target-s390: Implement SET ROUNDING MODE target-s390: Use uint64_to_float128 target-s390: Implement LCDFR target-s390: Check insn operand specifications target-s390: Implement CPSDR ...
author: Blue Swirl <blauwirbel@gmail.com> 2013-01-12 12:46:57 +0000
committer: Blue Swirl <blauwirbel@gmail.com> 2013-01-12 12:46:57 +0000
commit: 837d1f978224f7e7b020c71ffb10b291952cc596 (patch)
tree: f1e4e0b95ad756a48976b6d6889b41e44552454a
parent: a6fc23e5dc76660792c8363d058adf22caa82945 (diff)
parent: 2b35e93fbfc52c2be6cf85e5a54a3707cdabd914 (diff)
download: qemu-837d1f978224f7e7b020c71ffb10b291952cc596.tar.gz
16 files changed, 6308 insertions, 5937 deletions
diff --git a/MAINTAINERS b/MAINTAINERS
index 2991e1d2b3..35c260d549 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -98,6 +98,7 @@ S: Maintained
 F: target-ppc/
 
 S390
+M: Richard Henderson <rth@twiddle.net>
 M: Alexander Graf <agraf@suse.de>
 S: Maintained
 F: target-s390x/
diff --git a/disas/s390.c b/disas/s390.c
index 0859dfa19f..25499ba419 100644
--- a/disas/s390.c
+++ b/disas/s390.c
@@ -589,6 +589,16 @@ static const struct s390_operand s390_operands[] =
   { 4, 32, S390_OPERAND_CCODE },
 #define I8_32 46                  /* 8 bit signed value starting at 32 */
   { 8, 32, S390_OPERAND_SIGNED },
+#define U8_24 47                  /* 8 bit unsigned value starting at 24 */
+  { 8, 24, 0 },
+#define U8_32 48                  /* 8 bit unsigned value starting at 32 */
+  { 8, 32, 0 },
+#define I16_32 49
+  { 16, 32, S390_OPERAND_SIGNED },
+#define M4_16 50                  /* 4-bit condition-code starting at 12 */
+  { 4, 16, S390_OPERAND_CCODE },
+#define I8_16 51
+  { 8, 16, S390_OPERAND_SIGNED },
 /* QEMU-END */
 };
 
@@ -663,7 +673,9 @@ static const struct s390_operand s390_operands[] =
    This is just a workaround for existing code e.g. glibc.  */
 #define INSTR_RRE_RR_OPT 4, { R_24,RO_28,0,0,0,0 }             /* efpc, sfpc */
 #define INSTR_RRF_F0FF   4, { F_16,F_24,F_28,0,0,0 }           /* e.g. madbr */
-#define INSTR_RRF_F0FF2  4, { F_24,F_16,F_28,0,0,0 }           /* e.g. cpsdr */
+/* QEMU-MOD */
+#define INSTR_RRF_F0FF2  4, { F_24,F_28,F_16,0,0,0 }           /* e.g. cpsdr */
+/* QEMU-END */
 #define INSTR_RRF_F0FR   4, { F_24,F_16,R_28,0,0,0 }           /* e.g. iedtr */
 #define INSTR_RRF_FUFF   4, { F_24,F_16,F_28,U4_20,0,0 }       /* e.g. didbr */
 #define INSTR_RRF_RURR   4, { R_24,R_28,R_16,U4_20,0,0 }       /* e.g. .insn */
@@ -801,11 +813,35 @@ static const struct s390_operand s390_operands[] =
 #define MASK_SSF_RRDRD   { 0xff, 0x0f, 0x00, 0x00, 0x00, 0x00 }
 
 /* QEMU-ADD: */
-#define INSTR_RIE_MRRP   6, { M4_32,R_8,R_12,J16_16,0,0 }	/* e.g. crj */
+#define INSTR_RIE_MRRP   6, { M4_32, R_8, R_12, J16_16, 0, 0 } /* e.g. crj */
 #define MASK_RIE_MRRP    { 0xff, 0x00, 0x00, 0x00, 0x0f, 0xff }
 
-#define INSTR_RIE_MRIP   6, { M4_12,R_8,I8_32,J16_16,0,0 }      /* e.g. cij */
+#define INSTR_RIE_MRIP   6, { M4_12, R_8, I8_32, J16_16, 0, 0 } /* e.g. cij */
 #define MASK_RIE_MRIP    { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+
+#define INSTR_RIE_RRIII  6, { R_8, R_12, U8_16, U8_24, U8_32, 0 } /* risbg */
+#define MASK_RIE_RRIII   { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+#define INSTR_RIE_MRI    6, { M4_32, R_8, I16_16, 0, 0, 0 }    /* e.g. cit */
+#define MASK_RIE_MRI     { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+#define INSTR_RIE_MRU    6, { M4_32, R_8, U16_16, 0, 0, 0 }    /* e.g. clfit */
+#define MASK_RIE_MRU     { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+#define INSTR_RIE_RRI    6, { R_8, R_12, I16_16, 0, 0, 0 }
+#define MASK_RIE_RRI     { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+
+#define INSTR_RXY_URRD   6, { U8_8, D20_20, X_12, B_16, 0, 0 }
+#define MASK_RXY_URRD    { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+
+#define INSTR_SIL_DRI    6, { D_20, B_16, I16_32, 0, 0, 0 }
+#define MASK_SIL_DRI     { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 }
+
+#define INSTR_RSY_MRRD   6, { M4_12, R_8, D20_20, B_16, 0, 0 }
+#define MASK_SRY_MRRD    { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
+
+#define INSTR_RRF_MRR    6, { M4_16, R_24, R_28, 0, 0, 0 }
+#define MASK_RRF_MRR     { 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 }
+
+#define INSTR_SIY_DRI    6, { D20_20, B_16, I8_16, 0, 0, 0 }
+#define MASK_SIY_DRI     { 0xff, 0x00, 0x00, 0x00, 0x00, 0xff }
 /* QEMU-END */
 
 /* The opcode formats table (blueprints for .insn pseudo mnemonic).  */
@@ -926,6 +962,30 @@ static const struct s390_opcode s390_opcodes[] =
   { "ldeb", OP48(0xed0000000004LL), MASK_RXE_FRRD, INSTR_RXE_FRRD, 3, 0},
   { "brxlg", OP48(0xec0000000045LL), MASK_RIE_RRP, INSTR_RIE_RRP, 2, 2},
   { "brxhg", OP48(0xec0000000044LL), MASK_RIE_RRP, INSTR_RIE_RRP, 2, 2},
+/* QEMU-ADD: */
+  { "crj",   OP48(0xec0000000076LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
+  { "cgrj",  OP48(0xec0000000064LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
+  { "clrj",  OP48(0xec0000000077LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
+  { "clgrj", OP48(0xec0000000065LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
+  { "cij",   OP48(0xec000000007eLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
+  { "cgij",  OP48(0xec000000007cLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
+  { "clij",  OP48(0xec000000007fLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
+  { "clgij", OP48(0xec000000007dLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
+  { "risbg", OP48(0xec0000000055LL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "risbhg", OP48(0xec000000005dLL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "risblg", OP48(0xec0000000051LL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "rnsbg", OP48(0xec0000000054LL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "rosbg", OP48(0xec0000000056LL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "rxsbg", OP48(0xec0000000057LL), MASK_RIE_RRIII, INSTR_RIE_RRIII, 3, 6},
+  { "cit", OP48(0xec0000000072LL), MASK_RIE_MRI, INSTR_RIE_MRI, 3, 6},
+  { "cgit", OP48(0xec0000000070LL), MASK_RIE_MRI, INSTR_RIE_MRI, 3, 6},
+  { "clfit", OP48(0xec0000000073LL), MASK_RIE_MRU, INSTR_RIE_MRU, 3, 6},
+  { "clgit", OP48(0xec0000000071LL), MASK_RIE_MRU, INSTR_RIE_MRU, 3, 6},
+  { "ahik", OP48(0xec00000000d8LL), MASK_RIE_RRI, INSTR_RIE_RRI, 3, 6},
+  { "aghik", OP48(0xec00000000d9LL), MASK_RIE_RRI, INSTR_RIE_RRI, 3, 6},
+  { "alhsik", OP48(0xec00000000daLL), MASK_RIE_RRI, INSTR_RIE_RRI, 3, 6},
+  { "alghsik", OP48(0xec00000000dbLL), MASK_RIE_RRI, INSTR_RIE_RRI, 3, 6},
+/* QEMU-END */
   { "tp", OP48(0xeb00000000c0LL), MASK_RSL_R0RD, INSTR_RSL_R0RD, 3, 0},
   { "stamy", OP48(0xeb000000009bLL), MASK_RSY_AARD, INSTR_RSY_AARD, 2, 3},
   { "lamy", OP48(0xeb000000009aLL), MASK_RSY_AARD, INSTR_RSY_AARD, 2, 3},
@@ -985,6 +1045,20 @@ static const struct s390_opcode s390_opcodes[] =
   { "srag", OP48(0xeb000000000aLL), MASK_RSE_RRRD, INSTR_RSE_RRRD, 2, 2},
   { "lmg", OP48(0xeb0000000004LL), MASK_RSY_RRRD, INSTR_RSY_RRRD, 2, 3},
   { "lmg", OP48(0xeb0000000004LL), MASK_RSE_RRRD, INSTR_RSE_RRRD, 2, 2},
+/* QEMU-ADD: */
+  { "loc", OP48(0xeb00000000f2LL), MASK_SRY_MRRD, INSTR_RSY_MRRD, 3, 6},
+  { "locg", OP48(0xeb00000000e2LL), MASK_SRY_MRRD, INSTR_RSY_MRRD, 3, 6},
+  { "stoc", OP48(0xeb00000000f3LL), MASK_SRY_MRRD, INSTR_RSY_MRRD, 3, 6},
+  { "stocg", OP48(0xeb00000000e3LL), MASK_SRY_MRRD, INSTR_RSY_MRRD, 3, 6},
+  { "srak", OP48(0xeb00000000dcLL), MASK_RSY_RRRD, INSTR_RSY_RRRD, 3, 6},
+  { "slak", OP48(0xeb00000000ddLL), MASK_RSY_RRRD, INSTR_RSY_RRRD, 3, 6},
+  { "srlk", OP48(0xeb00000000deLL), MASK_RSY_RRRD, INSTR_RSY_RRRD, 3, 6},
+  { "sllk", OP48(0xeb00000000dfLL), MASK_RSY_RRRD, INSTR_RSY_RRRD, 3, 6},
+  { "asi", OP48(0xeb000000006aLL), MASK_SIY_DRI, INSTR_SIY_DRI, 3, 6},
+  { "alsi", OP48(0xeb000000006eLL), MASK_SIY_DRI, INSTR_SIY_DRI, 3, 6},
+  { "agsi", OP48(0xeb000000007aLL), MASK_SIY_DRI, INSTR_SIY_DRI, 3, 6},
+  { "algsi", OP48(0xeb000000007eLL), MASK_SIY_DRI, INSTR_SIY_DRI, 3, 6},
+/* QEMU-END */
   { "unpka", OP8(0xeaLL), MASK_SS_L0RDRD, INSTR_SS_L0RDRD, 3, 0},
   { "pka", OP8(0xe9LL), MASK_SS_L2RDRD, INSTR_SS_L2RDRD, 3, 0},
   { "mvcin", OP8(0xe8LL), MASK_SS_L0RDRD, INSTR_SS_L0RDRD, 3, 0},
@@ -993,6 +1067,17 @@ static const struct s390_opcode s390_opcodes[] =
   { "tprot", OP16(0xe501LL), MASK_SSE_RDRD, INSTR_SSE_RDRD, 3, 0},
   { "strag", OP48(0xe50000000002LL), MASK_SSE_RDRD, INSTR_SSE_RDRD, 2, 2},
   { "lasp", OP16(0xe500LL), MASK_SSE_RDRD, INSTR_SSE_RDRD, 3, 0},
+/* QEMU-ADD: */
+  { "mvhhi", OP16(0xe544LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "mvghi", OP16(0xe548LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "mvhi", OP16(0xe54cLL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "chhsi", OP16(0xe554LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "clhhsi", OP16(0xe555LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "cghsi", OP16(0xe558LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "clghsi", OP16(0xe559LL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "chsi", OP16(0xe55cLL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+  { "clfhsi", OP16(0xe55dLL), MASK_SIL_DRI, INSTR_SIL_DRI, 3, 6},
+/* QEMU-END */
   { "slb", OP48(0xe30000000099LL), MASK_RXY_RRRD, INSTR_RXY_RRRD, 3, 3},
   { "slb", OP48(0xe30000000099LL), MASK_RXE_RRRD, INSTR_RXE_RRRD, 3, 2},
   { "alc", OP48(0xe30000000098LL), MASK_RXY_RRRD, INSTR_RXY_RRRD, 3, 3},
@@ -1116,6 +1201,9 @@ static const struct s390_opcode s390_opcodes[] =
   { "lrag", OP48(0xe30000000003LL), MASK_RXY_RRRD, INSTR_RXY_RRRD, 2, 3},
   { "lrag", OP48(0xe30000000003LL), MASK_RXE_RRRD, INSTR_RXE_RRRD, 2, 2},
   { "ltg", OP48(0xe30000000002LL), MASK_RXY_RRRD, INSTR_RXY_RRRD, 2, 4},
+/* QEMU-ADD: */
+  { "pfd", OP48(0xe30000000036LL), MASK_RXY_URRD, INSTR_RXY_URRD, 3, 6},
+/* QEMU-END */
   { "unpku", OP8(0xe2LL), MASK_SS_L0RDRD, INSTR_SS_L0RDRD, 3, 0},
   { "pku", OP8(0xe1LL), MASK_SS_L0RDRD, INSTR_SS_L0RDRD, 3, 0},
   { "edmk", OP8(0xdfLL), MASK_SS_L0RDRD, INSTR_SS_L0RDRD, 3, 0},
@@ -1135,6 +1223,32 @@ static const struct s390_opcode s390_opcodes[] =
   { "csst", OP16(0xc802LL), MASK_SSF_RRDRD, INSTR_SSF_RRDRD, 2, 5},
   { "ectg", OP16(0xc801LL), MASK_SSF_RRDRD, INSTR_SSF_RRDRD, 2, 5},
   { "mvcos", OP16(0xc800LL), MASK_SSF_RRDRD, INSTR_SSF_RRDRD, 2, 4},
+/* QEMU-ADD: */
+  { "exrl", OP16(0xc600ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "pfdrl", OP16(0xc602ll), MASK_RIL_UP, INSTR_RIL_UP, 3, 6},
+  { "cghrl", OP16(0xc604ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "chrl", OP16(0xc605ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "clghrl", OP16(0xc606ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "clhrl", OP16(0xc607ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "cgrl", OP16(0xc608ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "clgrl", OP16(0xc60all), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "cgfrl", OP16(0xc60cll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "crl", OP16(0xc60dll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "clgfrl", OP16(0xc60ell), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "clrl", OP16(0xc60fll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+
+  { "llhrl", OP16(0xc400ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "lghrl", OP16(0xc404ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "lhrl", OP16(0xc405ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "llghrl", OP16(0xc406ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "sthrl", OP16(0xc407ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "lgrl", OP16(0xc408ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "stgrl", OP16(0xc40bll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "lgfrl", OP16(0xc40cll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "lrl", OP16(0xc40dll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "llgfrl", OP16(0xc40ell), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+  { "strl", OP16(0xc40fll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
+/* QEMU-END */
   { "clfi", OP16(0xc20fLL), MASK_RIL_RU, INSTR_RIL_RU, 2, 4},
   { "clgfi", OP16(0xc20eLL), MASK_RIL_RU, INSTR_RIL_RU, 2, 4},
   { "cfi", OP16(0xc20dLL), MASK_RIL_RI, INSTR_RIL_RI, 2, 4},
@@ -1265,6 +1379,29 @@ static const struct s390_opcode s390_opcodes[] =
   { "ltgr", OP16(0xb902LL), MASK_RRE_RR, INSTR_RRE_RR, 2, 2},
   { "lngr", OP16(0xb901LL), MASK_RRE_RR, INSTR_RRE_RR, 2, 2},
   { "lpgr", OP16(0xb900LL), MASK_RRE_RR, INSTR_RRE_RR, 2, 2},
+/* QEMU-ADD: */
+  { "crt", OP16(0xb972LL), MASK_RRF_M0RR, INSTR_RRF_M0RR, 3, 6},
+  { "cgrt", OP16(0xb960LL), MASK_RRF_M0RR, INSTR_RRF_M0RR, 3, 6},
+  { "clrt", OP16(0xb973LL), MASK_RRF_M0RR, INSTR_RRF_M0RR, 3, 6},
+  { "clgrt", OP16(0xb961LL), MASK_RRF_M0RR, INSTR_RRF_M0RR, 3, 6},
+  { "locr", OP16(0xb9f2LL), MASK_RRF_MRR, INSTR_RRF_MRR, 3, 6},
+  { "locgr", OP16(0xb9e2LL), MASK_RRF_MRR, INSTR_RRF_MRR, 3, 6},
+  { "popcnt", OP16(0xb9e1LL), MASK_RRE_RR, INSTR_RRE_RR, 3, 6},
+  { "ngrk", OP16(0xb9e4LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "ogrk", OP16(0xb9e6LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "xgrk", OP16(0xb9e7LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "agrk", OP16(0xb9e8LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "sgrk", OP16(0xb9e9LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "algrk", OP16(0xb9eaLL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "slgrk", OP16(0xb9ebLL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "nrk", OP16(0xb9f4LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "ork", OP16(0xb9f6LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "xrk", OP16(0xb9f7LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "ark", OP16(0xb9f8LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "srk", OP16(0xb9f9LL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "alrk", OP16(0xb9faLL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+  { "slrk", OP16(0xb9fbLL), MASK_RRF_R0RR, INSTR_RRF_R0RR, 3, 6},
+/* QEMU-END */
   { "lctl", OP8(0xb7LL), MASK_RS_CCRD, INSTR_RS_CCRD, 3, 0},
   { "stctl", OP8(0xb6LL), MASK_RS_CCRD, INSTR_RS_CCRD, 3, 0},
   { "rrxtr", OP16(0xb3ffLL), MASK_RRF_FFFU, INSTR_RRF_FFFU, 2, 5},
@@ -1426,6 +1563,20 @@ static const struct s390_opcode s390_opcodes[] =
   { "ltebr", OP16(0xb302LL), MASK_RRE_FF, INSTR_RRE_FF, 3, 0},
   { "lnebr", OP16(0xb301LL), MASK_RRE_FF, INSTR_RRE_FF, 3, 0},
   { "lpebr", OP16(0xb300LL), MASK_RRE_FF, INSTR_RRE_FF, 3, 0},
+/* QEMU-ADD: */
+  { "clfebr", OP16(0xb39cLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "clfdbr", OP16(0xb39dLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "clfxbr", OP16(0xb39eLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "clgebr", OP16(0xb3acLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "clgdbr", OP16(0xb3adLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "clgxbr", OP16(0xb3aeLL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "celfbr", OP16(0xb390LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "cdlfbr", OP16(0xb391LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "cxlfbr", OP16(0xb392LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "celgbr", OP16(0xb3a0LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "cdlgbr", OP16(0xb3a1LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+  { "cxlgbr", OP16(0xb3a2LL), MASK_RRF_UUFF, INSTR_RRF_UUFF, 3, 6},
+/* QEMU-END */
   { "trap4", OP16(0xb2ffLL), MASK_S_RD, INSTR_S_RD, 3, 0},
   { "lfas", OP16(0xb2bdLL), MASK_S_RD, INSTR_S_RD, 2, 5},
   { "srnmt", OP16(0xb2b9LL), MASK_S_RD, INSTR_S_RD, 2, 5},
@@ -1774,22 +1925,6 @@ static const struct s390_opcode s390_opcodes[] =
   { "sckpf", OP16(0x0107LL), MASK_E, INSTR_E, 3, 0},
   { "upt", OP16(0x0102LL), MASK_E, INSTR_E, 3, 0},
   { "pr", OP16(0x0101LL), MASK_E, INSTR_E, 3, 0},
-
-/* QEMU-ADD: */
-  { "crj",   OP48(0xec0000000076LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
-  { "cgrj",  OP48(0xec0000000064LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
-  { "clrj",  OP48(0xec0000000077LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
-  { "clgrj", OP48(0xec0000000065LL), MASK_RIE_MRRP, INSTR_RIE_MRRP, 3, 6},
-
-  { "cij",   OP48(0xec000000007eLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
-  { "cgij",  OP48(0xec000000007cLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
-  { "clij",  OP48(0xec000000007fLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
-  { "clgij", OP48(0xec000000007dLL), MASK_RIE_MRIP, INSTR_RIE_MRIP, 3, 6},
-
-  { "lrl",   OP16(0xc40dll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
-  { "lgrl",  OP16(0xc408ll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
-  { "lgfrl", OP16(0xc40cll), MASK_RIL_RP, INSTR_RIL_RP, 3, 6},
-/* QEMU-END */
 };
 
 static const int s390_num_opcodes =
diff --git a/gdbstub.c b/gdbstub.c
index a8dd437ec0..e62dc798c3 100644
--- a/gdbstub.c
+++ b/gdbstub.c
@@ -40,6 +40,7 @@
 #include "cpu.h"
 #include "qemu/sockets.h"
 #include "sysemu/kvm.h"
+#include "qemu/bitops.h"
 
 #ifndef TARGET_CPU_MEMORY_RW_DEBUG
 static inline int target_memory_rw_debug(CPUArchState *env, target_ulong addr,
@@ -1535,27 +1536,34 @@ static int cpu_gdb_write_register(CPUAlphaState *env, uint8_t *mem_buf, int n)
 }
 #elif defined (TARGET_S390X)
 
-#define NUM_CORE_REGS S390_NUM_TOTAL_REGS
+#define NUM_CORE_REGS  S390_NUM_REGS
 
 static int cpu_gdb_read_register(CPUS390XState *env, uint8_t *mem_buf, int n)
 {
+    uint64_t val;
+    int cc_op;
+
     switch (n) {
-        case S390_PSWM_REGNUM: GET_REGL(env->psw.mask); break;
-        case S390_PSWA_REGNUM: GET_REGL(env->psw.addr); break;
-        case S390_R0_REGNUM ... S390_R15_REGNUM:
-            GET_REGL(env->regs[n-S390_R0_REGNUM]); break;
-        case S390_A0_REGNUM ... S390_A15_REGNUM:
-            GET_REG32(env->aregs[n-S390_A0_REGNUM]); break;
-        case S390_FPC_REGNUM: GET_REG32(env->fpc); break;
-        case S390_F0_REGNUM ... S390_F15_REGNUM:
-            /* XXX */
-            break;
-        case S390_PC_REGNUM: GET_REGL(env->psw.addr); break;
-        case S390_CC_REGNUM:
-            env->cc_op = calc_cc(env, env->cc_op, env->cc_src, env->cc_dst,
-                                 env->cc_vr);
-            GET_REG32(env->cc_op);
-            break;
+    case S390_PSWM_REGNUM:
+        cc_op = calc_cc(env, env->cc_op, env->cc_src, env->cc_dst, env->cc_vr);
+        val = deposit64(env->psw.mask, 44, 2, cc_op);
+        GET_REGL(val);
+        break;
+    case S390_PSWA_REGNUM:
+        GET_REGL(env->psw.addr);
+        break;
+    case S390_R0_REGNUM ... S390_R15_REGNUM:
+        GET_REGL(env->regs[n-S390_R0_REGNUM]);
+        break;
+    case S390_A0_REGNUM ... S390_A15_REGNUM:
+        GET_REG32(env->aregs[n-S390_A0_REGNUM]);
+        break;
+    case S390_FPC_REGNUM:
+        GET_REG32(env->fpc);
+        break;
+    case S390_F0_REGNUM ... S390_F15_REGNUM:
+        GET_REG64(env->fregs[n-S390_F0_REGNUM].ll);
+        break;
     }
 
     return 0;
@@ -1570,20 +1578,30 @@ static int cpu_gdb_write_register(CPUS390XState *env, uint8_t *mem_buf, int n)
     tmp32 = ldl_p(mem_buf);
 
     switch (n) {
-        case S390_PSWM_REGNUM: env->psw.mask = tmpl; break;
-        case S390_PSWA_REGNUM: env->psw.addr = tmpl; break;
-        case S390_R0_REGNUM ... S390_R15_REGNUM:
-            env->regs[n-S390_R0_REGNUM] = tmpl; break;
-        case S390_A0_REGNUM ... S390_A15_REGNUM:
-            env->aregs[n-S390_A0_REGNUM] = tmp32; r=4; break;
-        case S390_FPC_REGNUM: env->fpc = tmp32; r=4; break;
-        case S390_F0_REGNUM ... S390_F15_REGNUM:
-            /* XXX */
-            break;
-        case S390_PC_REGNUM: env->psw.addr = tmpl; break;
-        case S390_CC_REGNUM: env->cc_op = tmp32; r=4; break;
+    case S390_PSWM_REGNUM:
+        env->psw.mask = tmpl;
+        env->cc_op = extract64(tmpl, 44, 2);
+        break;
+    case S390_PSWA_REGNUM:
+        env->psw.addr = tmpl;
+        break;
+    case S390_R0_REGNUM ... S390_R15_REGNUM:
+        env->regs[n-S390_R0_REGNUM] = tmpl;
+        break;
+    case S390_A0_REGNUM ... S390_A15_REGNUM:
+        env->aregs[n-S390_A0_REGNUM] = tmp32;
+        r = 4;
+        break;
+    case S390_FPC_REGNUM:
+        env->fpc = tmp32;
+        r = 4;
+        break;
+    case S390_F0_REGNUM ... S390_F15_REGNUM:
+        env->fregs[n-S390_F0_REGNUM].ll = tmpl;
+        break;
+    default:
+        return 0;
     }
-
     return r;
 }
 #elif defined (TARGET_LM32)
diff --git a/linux-user/main.c b/linux-user/main.c
index 9ade1bfabd..0181bc2112 100644
--- a/linux-user/main.c
+++ b/linux-user/main.c
@@ -2938,71 +2938,115 @@ void cpu_loop(CPUAlphaState *env)
 #ifdef TARGET_S390X
 void cpu_loop(CPUS390XState *env)
 {
-    int trapnr;
+    int trapnr, n, sig;
     target_siginfo_t info;
+    target_ulong addr;
 
     while (1) {
-        trapnr = cpu_s390x_exec (env);
-
+        trapnr = cpu_s390x_exec(env);
         switch (trapnr) {
         case EXCP_INTERRUPT:
-            /* just indicate that signals should be handled asap */
+            /* Just indicate that signals should be handled asap.  */
             break;
-        case EXCP_DEBUG:
-            {
-                int sig;
 
-                sig = gdb_handlesig (env, TARGET_SIGTRAP);
-                if (sig) {
-                    info.si_signo = sig;
-                    info.si_errno = 0;
-                    info.si_code = TARGET_TRAP_BRKPT;
-                    queue_signal(env, info.si_signo, &info);
-                }
+        case EXCP_SVC:
+            n = env->int_svc_code;
+            if (!n) {
+                /* syscalls > 255 */
+                n = env->regs[1];
             }
+            env->psw.addr += env->int_svc_ilen;
+            env->regs[2] = do_syscall(env, n, env->regs[2], env->regs[3],
+                                      env->regs[4], env->regs[5],
+                                      env->regs[6], env->regs[7], 0, 0);
             break;
-        case EXCP_SVC:
-            {
-                int n = env->int_svc_code;
-                if (!n) {
-                    /* syscalls > 255 */
-                    n = env->regs[1];
-                }
-                env->psw.addr += env->int_svc_ilc;
-                env->regs[2] = do_syscall(env, n,
-                           env->regs[2],
-                           env->regs[3],
-                           env->regs[4],
-                           env->regs[5],
-                           env->regs[6],
-                           env->regs[7],
-                           0, 0);
+
+        case EXCP_DEBUG:
+            sig = gdb_handlesig(env, TARGET_SIGTRAP);
+            if (sig) {
+                n = TARGET_TRAP_BRKPT;
+                goto do_signal_pc;
             }
             break;
-        case EXCP_ADDR:
-            {
-                info.si_signo = SIGSEGV;
-                info.si_errno = 0;
+        case EXCP_PGM:
+            n = env->int_pgm_code;
+            switch (n) {
+            case PGM_OPERATION:
+            case PGM_PRIVILEGED:
+                sig = SIGILL;
+                n = TARGET_ILL_ILLOPC;
+                goto do_signal_pc;
+            case PGM_PROTECTION:
+            case PGM_ADDRESSING:
+                sig = SIGSEGV;
                 /* XXX: check env->error_code */
-                info.si_code = TARGET_SEGV_MAPERR;
-                info._sifields._sigfault._addr = env->__excp_addr;
-                queue_signal(env, info.si_signo, &info);
+                n = TARGET_SEGV_MAPERR;
+                addr = env->__excp_addr;
+                goto do_signal;
+            case PGM_EXECUTE:
+            case PGM_SPECIFICATION:
+            case PGM_SPECIAL_OP:
+            case PGM_OPERAND:
+            do_sigill_opn:
+                sig = SIGILL;
+                n = TARGET_ILL_ILLOPN;
+                goto do_signal_pc;
+
+            case PGM_FIXPT_OVERFLOW:
+                sig = SIGFPE;
+                n = TARGET_FPE_INTOVF;
+                goto do_signal_pc;
+            case PGM_FIXPT_DIVIDE:
+                sig = SIGFPE;
+                n = TARGET_FPE_INTDIV;
+                goto do_signal_pc;
+
+            case PGM_DATA:
+                n = (env->fpc >> 8) & 0xff;
+                if (n == 0xff) {
+                    /* compare-and-trap */
+                    goto do_sigill_opn;
+                } else {
+                    /* An IEEE exception, simulated or otherwise.  */
+                    if (n & 0x80) {
+                        n = TARGET_FPE_FLTINV;
+                    } else if (n & 0x40) {
+                        n = TARGET_FPE_FLTDIV;
+                    } else if (n & 0x20) {
+                        n = TARGET_FPE_FLTOVF;
+                    } else if (n & 0x10) {
+                        n = TARGET_FPE_FLTUND;
+                    } else if (n & 0x08) {
+                        n = TARGET_FPE_FLTRES;
+                    } else {
+                        /* ??? Quantum exception; BFP, DFP error.  */
+                        goto do_sigill_opn;
+                    }
+                    sig = SIGFPE;
+                    goto do_signal_pc;
+                }
+
+            default:
+                fprintf(stderr, "Unhandled program exception: %#x\n", n);
+                cpu_dump_state(env, stderr, fprintf, 0);
+                exit(1);
             }
             break;
-        case EXCP_SPEC:
-            {
-                fprintf(stderr,"specification exception insn 0x%08x%04x\n", ldl(env->psw.addr), lduw(env->psw.addr + 4));
-                info.si_signo = SIGILL;
-                info.si_errno = 0;
-                info.si_code = TARGET_ILL_ILLOPC;
-                info._sifields._sigfault._addr = env->__excp_addr;
-                queue_signal(env, info.si_signo, &info);
-            }
+
+        do_signal_pc:
+            addr = env->psw.addr;
+        do_signal:
+            info.si_signo = sig;
+            info.si_errno = 0;
+            info.si_code = n;
+            info._sifields._sigfault._addr = addr;
+            queue_signal(env, info.si_signo, &info);
             break;
+
         default:
-            printf ("Unhandled trap: 0x%x\n", trapnr);
+            fprintf(stderr, "Unhandled trap: 0x%x\n", trapnr);
             cpu_dump_state(env, stderr, fprintf, 0);
-            exit (1);
+            exit(1);
         }
         process_pending_signals (env);
     }
diff --git a/linux-user/s390x/syscall.h b/linux-user/s390x/syscall.h
index c2ea151ea5..e4603b79c3 100644
--- a/linux-user/s390x/syscall.h
+++ b/linux-user/s390x/syscall.h
@@ -16,7 +16,7 @@ struct target_pt_regs {
     target_psw_t psw;
     abi_ulong gprs[TARGET_NUM_GPRS];
     abi_ulong orig_gpr2;
-    unsigned short ilc;
+    unsigned short ilen;
     unsigned short trap;
 };
 
diff --git a/target-s390x/cc_helper.c b/target-s390x/cc_helper.c
index 19ef145da9..a6d60bf885 100644
--- a/target-s390x/cc_helper.c
+++ b/target-s390x/cc_helper.c
@@ -20,6 +20,7 @@
 
 #include "cpu.h"
 #include "helper.h"
+#include "qemu/host-utils.h"
 
 /* #define DEBUG_HELPER */
 #ifdef DEBUG_HELPER
@@ -28,8 +29,7 @@
 #define HELPER_LOG(x...)
 #endif
 
-static inline uint32_t cc_calc_ltgt_32(CPUS390XState *env, int32_t src,
-                                       int32_t dst)
+static uint32_t cc_calc_ltgt_32(int32_t src, int32_t dst)
 {
     if (src == dst) {
         return 0;
@@ -40,13 +40,12 @@ static inline uint32_t cc_calc_ltgt_32(CPUS390XState *env, int32_t src,
     }
 }
 
-static inline uint32_t cc_calc_ltgt0_32(CPUS390XState *env, int32_t dst)
+static uint32_t cc_calc_ltgt0_32(int32_t dst)
 {
-    return cc_calc_ltgt_32(env, dst, 0);
+    return cc_calc_ltgt_32(dst, 0);
 }
 
-static inline uint32_t cc_calc_ltgt_64(CPUS390XState *env, int64_t src,
-                                       int64_t dst)
+static uint32_t cc_calc_ltgt_64(int64_t src, int64_t dst)
 {
     if (src == dst) {
         return 0;
@@ -57,13 +56,12 @@ static inline uint32_t cc_calc_ltgt_64(CPUS390XState *env, int64_t src,
     }
 }
 
-static inline uint32_t cc_calc_ltgt0_64(CPUS390XState *env, int64_t dst)
+static uint32_t cc_calc_ltgt0_64(int64_t dst)
 {
-    return cc_calc_ltgt_64(env, dst, 0);
+    return cc_calc_ltgt_64(dst, 0);
 }
 
-static inline uint32_t cc_calc_ltugtu_32(CPUS390XState *env, uint32_t src,
-                                         uint32_t dst)
+static uint32_t cc_calc_ltugtu_32(uint32_t src, uint32_t dst)
 {
     if (src == dst) {
         return 0;
@@ -74,8 +72,7 @@ static inline uint32_t cc_calc_ltugtu_32(CPUS390XState *env, uint32_t src,
     }
 }
 
-static inline uint32_t cc_calc_ltugtu_64(CPUS390XState *env, uint64_t src,
-                                         uint64_t dst)
+static uint32_t cc_calc_ltugtu_64(uint64_t src, uint64_t dst)
 {
     if (src == dst) {
         return 0;
@@ -86,13 +83,11 @@ static inline uint32_t cc_calc_ltugtu_64(CPUS390XState *env, uint64_t src,
     }
 }
 
-static inline uint32_t cc_calc_tm_32(CPUS390XState *env, uint32_t val,
-                                     uint32_t mask)
+static uint32_t cc_calc_tm_32(uint32_t val, uint32_t mask)
 {
-    uint16_t r = val & mask;
+    uint32_t r = val & mask;
 
-    HELPER_LOG("%s: val 0x%x mask 0x%x\n", __func__, val, mask);
-    if (r == 0 || mask == 0) {
+    if (r == 0) {
         return 0;
     } else if (r == mask) {
         return 3;
@@ -101,23 +96,17 @@ static inline uint32_t cc_calc_tm_32(CPUS390XState *env, uint32_t val,
     }
 }
 
-/* set condition code for test under mask */
-static inline uint32_t cc_calc_tm_64(CPUS390XState *env, uint64_t val,
-                                     uint32_t mask)
+static uint32_t cc_calc_tm_64(uint64_t val, uint64_t mask)
 {
-    uint16_t r = val & mask;
+    uint64_t r = val & mask;
 
-    HELPER_LOG("%s: val 0x%lx mask 0x%x r 0x%x\n", __func__, val, mask, r);
-    if (r == 0 || mask == 0) {
+    if (r == 0) {
         return 0;
     } else if (r == mask) {
         return 3;
     } else {
-        while (!(mask & 0x8000)) {
-            mask <<= 1;
-            val <<= 1;
-        }
-        if (val & 0x8000) {
+        int top = clz64(mask);
+        if ((int64_t)(val << top) < 0) {
             return 2;
         } else {
             return 1;
@@ -125,13 +114,12 @@ static inline uint32_t cc_calc_tm_64(CPUS390XState *env, uint64_t val,
     }
 }
 
-static inline uint32_t cc_calc_nz(CPUS390XState *env, uint64_t dst)
+static uint32_t cc_calc_nz(uint64_t dst)
 {
     return !!dst;
 }
 
-static inline uint32_t cc_calc_add_64(CPUS390XState *env, int64_t a1,
-                                      int64_t a2, int64_t ar)
+static uint32_t cc_calc_add_64(int64_t a1, int64_t a2, int64_t ar)
 {
     if ((a1 > 0 && a2 > 0 && ar < 0) || (a1 < 0 && a2 < 0 && ar > 0)) {
         return 3; /* overflow */
@@ -146,26 +134,22 @@ static inline uint32_t cc_calc_add_64(CPUS390XState *env, int64_t a1,
     }
 }
 
-static inline uint32_t cc_calc_addu_64(CPUS390XState *env, uint64_t a1,
-                                       uint64_t a2, uint64_t ar)
+static uint32_t cc_calc_addu_64(uint64_t a1, uint64_t a2, uint64_t ar)
 {
-    if (ar == 0) {
-        if (a1) {
-            return 2;
-        } else {
-            return 0;
-        }
-    } else {
-        if (ar < a1 || ar < a2) {
-            return 3;
-        } else {
-            return 1;
-        }
-    }
+    return (ar != 0) + 2 * (ar < a1);
+}
+
+static uint32_t cc_calc_addc_64(uint64_t a1, uint64_t a2, uint64_t ar)
+{
+    /* Recover a2 + carry_in.  */
+    uint64_t a2c = ar - a1;
+    /* Check for a2+carry_in overflow, then a1+a2c overflow.  */
+    int carry_out = (a2c < a2) || (ar < a1);
+
+    return (ar != 0) + 2 * carry_out;
 }
 
-static inline uint32_t cc_calc_sub_64(CPUS390XState *env, int64_t a1,
-                                      int64_t a2, int64_t ar)
+static uint32_t cc_calc_sub_64(int64_t a1, int64_t a2, int64_t ar)
 {
     if ((a1 > 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
         return 3; /* overflow */
@@ -180,8 +164,7 @@ static inline uint32_t cc_calc_sub_64(CPUS390XState *env, int64_t a1,
     }
 }
 
-static inline uint32_t cc_calc_subu_64(CPUS390XState *env, uint64_t a1,
-                                       uint64_t a2, uint64_t ar)
+static uint32_t cc_calc_subu_64(uint64_t a1, uint64_t a2, uint64_t ar)
 {
     if (ar == 0) {
         return 2;
@@ -194,7 +177,25 @@ static inline uint32_t cc_calc_subu_64(CPUS390XState *env, uint64_t a1,
     }
 }
 
-static inline uint32_t cc_calc_abs_64(CPUS390XState *env, int64_t dst)
+static uint32_t cc_calc_subb_64(uint64_t a1, uint64_t a2, uint64_t ar)
+{
+    /* We had borrow-in if normal subtraction isn't equal.  */
+    int borrow_in = ar - (a1 - a2);
+    int borrow_out;
+
+    /* If a2 was ULONG_MAX, and borrow_in, then a2 is logically 65 bits,
+       and we must have had borrow out.  */
+    if (borrow_in && a2 == (uint64_t)-1) {
+        borrow_out = 1;
+    } else {
+        a2 += borrow_in;
+        borrow_out = (a2 > a1);
+    }
+
+    return (ar != 0) + 2 * !borrow_out;
+}
+
+static uint32_t cc_calc_abs_64(int64_t dst)
 {
     if ((uint64_t)dst == 0x8000000000000000ULL) {
         return 3;
@@ -205,12 +206,12 @@ static inline uint32_t cc_calc_abs_64(CPUS390XState *env, int64_t dst)
     }
 }
 
-static inline uint32_t cc_calc_nabs_64(CPUS390XState *env, int64_t dst)
+static uint32_t cc_calc_nabs_64(int64_t dst)
 {
     return !!dst;
 }
 
-static inline uint32_t cc_calc_comp_64(CPUS390XState *env, int64_t dst)
+static uint32_t cc_calc_comp_64(int64_t dst)
 {
     if ((uint64_t)dst == 0x8000000000000000ULL) {
         return 3;
@@ -224,8 +225,7 @@ static inline uint32_t cc_calc_comp_64(CPUS390XState *env, int64_t dst)
 }
 
 
-static inline uint32_t cc_calc_add_32(CPUS390XState *env, int32_t a1,
-                                      int32_t a2, int32_t ar)
+static uint32_t cc_calc_add_32(int32_t a1, int32_t a2, int32_t ar)
 {
     if ((a1 > 0 && a2 > 0 && ar < 0) || (a1 < 0 && a2 < 0 && ar > 0)) {
         return 3; /* overflow */
@@ -240,26 +240,22 @@ static inline uint32_t cc_calc_add_32(CPUS390XState *env, int32_t a1,
     }
 }
 
-static inline uint32_t cc_calc_addu_32(CPUS390XState *env, uint32_t a1,
-                                       uint32_t a2, uint32_t ar)
+static uint32_t cc_calc_addu_32(uint32_t a1, uint32_t a2, uint32_t ar)
 {
-    if (ar == 0) {
-        if (a1) {
-            return 2;
-        } else {
-            return 0;
-        }
-    } else {
-        if (ar < a1 || ar < a2) {
-            return 3;
-        } else {
-            return 1;
-        }
-    }
+    return (ar != 0) + 2 * (ar < a1);
+}
+
+static uint32_t cc_calc_addc_32(uint32_t a1, uint32_t a2, uint32_t ar)
+{
+    /* Recover a2 + carry_in.  */
+    uint32_t a2c = ar - a1;
+    /* Check for a2+carry_in overflow, then a1+a2c overflow.  */
+    int carry_out = (a2c < a2) || (ar < a1);
+
+    return (ar != 0) + 2 * carry_out;
 }
 
-static inline uint32_t cc_calc_sub_32(CPUS390XState *env, int32_t a1,
-                                      int32_t a2, int32_t ar)
+static uint32_t cc_calc_sub_32(int32_t a1, int32_t a2, int32_t ar)
 {
     if ((a1 > 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
         return 3; /* overflow */
@@ -274,8 +270,7 @@ static inline uint32_t cc_calc_sub_32(CPUS390XState *env, int32_t a1,
     }
 }
 
-static inline uint32_t cc_calc_subu_32(CPUS390XState *env, uint32_t a1,
-                                       uint32_t a2, uint32_t ar)
+static uint32_t cc_calc_subu_32(uint32_t a1, uint32_t a2, uint32_t ar)
 {
     if (ar == 0) {
         return 2;
@@ -288,7 +283,25 @@ static inline uint32_t cc_calc_subu_32(CPUS390XState *env, uint32_t a1,
     }
 }
 
-static inline uint32_t cc_calc_abs_32(CPUS390XState *env, int32_t dst)
+static uint32_t cc_calc_subb_32(uint32_t a1, uint32_t a2, uint32_t ar)
+{
+    /* We had borrow-in if normal subtraction isn't equal.  */
+    int borrow_in = ar - (a1 - a2);
+    int borrow_out;
+
+    /* If a2 was UINT_MAX, and borrow_in, then a2 is logically 65 bits,
+       and we must have had borrow out.  */
+    if (borrow_in && a2 == (uint32_t)-1) {
+        borrow_out = 1;
+    } else {
+        a2 += borrow_in;
+        borrow_out = (a2 > a1);
+    }
+
+    return (ar != 0) + 2 * !borrow_out;
+}
+
+static uint32_t cc_calc_abs_32(int32_t dst)
 {
     if ((uint32_t)dst == 0x80000000UL) {
         return 3;
@@ -299,12 +312,12 @@ static inline uint32_t cc_calc_abs_32(CPUS390XState *env, int32_t dst)
     }
 }
 
-static inline uint32_t cc_calc_nabs_32(CPUS390XState *env, int32_t dst)
+static uint32_t cc_calc_nabs_32(int32_t dst)
 {
     return !!dst;
 }
 
-static inline uint32_t cc_calc_comp_32(CPUS390XState *env, int32_t dst)
+static uint32_t cc_calc_comp_32(int32_t dst)
 {
     if ((uint32_t)dst == 0x80000000UL) {
         return 3;
@@ -318,69 +331,80 @@ static inline uint32_t cc_calc_comp_32(CPUS390XState *env, int32_t dst)
 }
 
 /* calculate condition code for insert character under mask insn */
-static inline uint32_t cc_calc_icm_32(CPUS390XState *env, uint32_t mask,
-                                      uint32_t val)
+static uint32_t cc_calc_icm(uint64_t mask, uint64_t val)
 {
-    uint32_t cc;
-
-    HELPER_LOG("%s: mask 0x%x val %d\n", __func__, mask, val);
-    if (mask == 0xf) {
-        if (!val) {
-            return 0;
-        } else if (val & 0x80000000) {
+    if ((val & mask) == 0) {
+        return 0;
+    } else {
+        int top = clz64(mask);
+        if ((int64_t)(val << top) < 0) {
             return 1;
         } else {
             return 2;
         }
     }
+}
+
+static uint32_t cc_calc_sla_32(uint32_t src, int shift)
+{
+    uint32_t mask = ((1U << shift) - 1U) << (32 - shift);
+    uint32_t sign = 1U << 31;
+    uint32_t match;
+    int32_t r;
 
-    if (!val || !mask) {
-        cc = 0;
+    /* Check if the sign bit stays the same.  */
+    if (src & sign) {
+        match = mask;
     } else {
-        while (mask != 1) {
-            mask >>= 1;
-            val >>= 8;
-        }
-        if (val & 0x80) {
-            cc = 1;
-        } else {
-            cc = 2;
-        }
+        match = 0;
     }
-    return cc;
+    if ((src & mask) != match) {
+        /* Overflow.  */
+        return 3;
+    }
+
+    r = ((src << shift) & ~sign) | (src & sign);
+    if (r == 0) {
+        return 0;
+    } else if (r < 0) {
+        return 1;
+    }
+    return 2;
 }
 
-static inline uint32_t cc_calc_slag(CPUS390XState *env, uint64_t src,
-                                    uint64_t shift)
+static uint32_t cc_calc_sla_64(uint64_t src, int shift)
 {
     uint64_t mask = ((1ULL << shift) - 1ULL) << (64 - shift);
-    uint64_t match, r;
+    uint64_t sign = 1ULL << 63;
+    uint64_t match;
+    int64_t r;
 
-    /* check if the sign bit stays the same */
-    if (src & (1ULL << 63)) {
+    /* Check if the sign bit stays the same.  */
+    if (src & sign) {
         match = mask;
     } else {
         match = 0;
     }
-
     if ((src & mask) != match) {
-        /* overflow */
+        /* Overflow.  */
         return 3;
     }
 
-    r = ((src << shift) & ((1ULL << 63) - 1)) | (src & (1ULL << 63));
-
-    if ((int64_t)r == 0) {
+    r = ((src << shift) & ~sign) | (src & sign);
+    if (r == 0) {
         return 0;
-    } else if ((int64_t)r < 0) {
+    } else if (r < 0) {
         return 1;
     }
-
     return 2;
 }
 
+static uint32_t cc_calc_flogr(uint64_t dst)
+{
+    return dst ? 2 : 0;
+}
 
-static inline uint32_t do_calc_cc(CPUS390XState *env, uint32_t cc_op,
+static uint32_t do_calc_cc(CPUS390XState *env, uint32_t cc_op,
                                   uint64_t src, uint64_t dst, uint64_t vr)
 {
     uint32_t r = 0;
@@ -394,95 +418,110 @@ static inline uint32_t do_calc_cc(CPUS390XState *env, uint32_t cc_op,
         r = cc_op;
         break;
     case CC_OP_LTGT0_32:
-        r = cc_calc_ltgt0_32(env, dst);
+        r = cc_calc_ltgt0_32(dst);
         break;
     case CC_OP_LTGT0_64:
-        r =  cc_calc_ltgt0_64(env, dst);
+        r =  cc_calc_ltgt0_64(dst);
         break;
     case CC_OP_LTGT_32:
-        r =  cc_calc_ltgt_32(env, src, dst);
+        r =  cc_calc_ltgt_32(src, dst);
         break;
     case CC_OP_LTGT_64:
-        r =  cc_calc_ltgt_64(env, src, dst);
+        r =  cc_calc_ltgt_64(src, dst);
         break;
     case CC_OP_LTUGTU_32:
-        r =  cc_calc_ltugtu_32(env, src, dst);
+        r =  cc_calc_ltugtu_32(src, dst);
         break;
     case CC_OP_LTUGTU_64:
-        r =  cc_calc_ltugtu_64(env, src, dst);
+        r =  cc_calc_ltugtu_64(src, dst);
         break;
     case CC_OP_TM_32:
-        r =  cc_calc_tm_32(env, src, dst);
+        r =  cc_calc_tm_32(src, dst);
         break;
     case CC_OP_TM_64:
-        r =  cc_calc_tm_64(env, src, dst);
+        r =  cc_calc_tm_64(src, dst);
         break;
     case CC_OP_NZ:
-        r =  cc_calc_nz(env, dst);
+        r =  cc_calc_nz(dst);
         break;
     case CC_OP_ADD_64:
-        r =  cc_calc_add_64(env, src, dst, vr);
+        r =  cc_calc_add_64(src, dst, vr);
         break;
     case CC_OP_ADDU_64:
-        r =  cc_calc_addu_64(env, src, dst, vr);
+        r =  cc_calc_addu_64(src, dst, vr);
+        break;
+    case CC_OP_ADDC_64:
+        r =  cc_calc_addc_64(src, dst, vr);
         break;
     case CC_OP_SUB_64:
-        r =  cc_calc_sub_64(env, src, dst, vr);
+        r =  cc_calc_sub_64(src, dst, vr);
         break;
     case CC_OP_SUBU_64:
-        r =  cc_calc_subu_64(env, src, dst, vr);
+        r =  cc_calc_subu_64(src, dst, vr);
+        break;
+    case CC_OP_SUBB_64:
+        r =  cc_calc_subb_64(src, dst, vr);
         break;
     case CC_OP_ABS_64:
-        r =  cc_calc_abs_64(env, dst);
+        r =  cc_calc_abs_64(dst);
         break;
     case CC_OP_NABS_64:
-        r =  cc_calc_nabs_64(env, dst);
+        r =  cc_calc_nabs_64(dst);
         break;
     case CC_OP_COMP_64:
-        r =  cc_calc_comp_64(env, dst);
+        r =  cc_calc_comp_64(dst);
         break;
 
     case CC_OP_ADD_32:
-        r =  cc_calc_add_32(env, src, dst, vr);
+        r =  cc_calc_add_32(src, dst, vr);
         break;
     case CC_OP_ADDU_32:
-        r =  cc_calc_addu_32(env, src, dst, vr);
+        r =  cc_calc_addu_32(src, dst, vr);
+        break;
+    case CC_OP_ADDC_32:
+        r =  cc_calc_addc_32(src, dst, vr);
         break;
     case CC_OP_SUB_32:
-        r =  cc_calc_sub_32(env, src, dst, vr);
+        r =  cc_calc_sub_32(src, dst, vr);
         break;
     case CC_OP_SUBU_32:
-        r =  cc_calc_subu_32(env, src, dst, vr);
+        r =  cc_calc_subu_32(src, dst, vr);
+        break;
+    case CC_OP_SUBB_32:
+        r =  cc_calc_subb_32(src, dst, vr);
         break;
     case CC_OP_ABS_32:
-        r =  cc_calc_abs_64(env, dst);
+        r =  cc_calc_abs_32(dst);
         break;
     case CC_OP_NABS_32:
-        r =  cc_calc_nabs_64(env, dst);
+        r =  cc_calc_nabs_32(dst);
         break;
     case CC_OP_COMP_32:
-        r =  cc_calc_comp_32(env, dst);
+        r =  cc_calc_comp_32(dst);
         break;
 
     case CC_OP_ICM:
-        r =  cc_calc_icm_32(env, src, dst);
+        r =  cc_calc_icm(src, dst);
         break;
-    case CC_OP_SLAG:
-        r =  cc_calc_slag(env, src, dst);
+    case CC_OP_SLA_32:
+        r =  cc_calc_sla_32(src, dst);
         break;
-
-    case CC_OP_LTGT_F32:
-        r = set_cc_f32(env, src, dst);
+    case CC_OP_SLA_64:
+        r =  cc_calc_sla_64(src, dst);
         break;
-    case CC_OP_LTGT_F64:
-        r = set_cc_f64(env, src, dst);
+    case CC_OP_FLOGR:
+        r = cc_calc_flogr(dst);
         break;
+
     case CC_OP_NZ_F32:
         r = set_cc_nz_f32(dst);
         break;
     case CC_OP_NZ_F64:
         r = set_cc_nz_f64(dst);
         break;
+    case CC_OP_NZ_F128:
+        r = set_cc_nz_f128(make_float128(src, dst));
+        break;
 
     default:
         cpu_abort(env, "Unknown CC operation: %s\n", cc_name(cc_op));
@@ -505,18 +544,6 @@ uint32_t HELPER(calc_cc)(CPUS390XState *env, uint32_t cc_op, uint64_t src,
     return do_calc_cc(env, cc_op, src, dst, vr);
 }
 
-/* insert psw mask and condition code into r1 */
-void HELPER(ipm)(CPUS390XState *env, uint32_t cc, uint32_t r1)
-{
-    uint64_t r = env->regs[r1];
-
-    r &= 0xffffffff00ffffffULL;
-    r |= (cc << 28) | ((env->psw.mask >> 40) & 0xf);
-    env->regs[r1] = r;
-    HELPER_LOG("%s: cc %d psw.mask 0x%lx r1 0x%lx\n", __func__,
-               cc, env->psw.mask, r);
-}
-
 #ifndef CONFIG_USER_ONLY
 void HELPER(load_psw)(CPUS390XState *env, uint64_t mask, uint64_t addr)
 {
diff --git a/target-s390x/cpu.h b/target-s390x/cpu.h
index cd565c91de..bc3fab226b 100644
--- a/target-s390x/cpu.h
+++ b/target-s390x/cpu.h
@@ -60,17 +60,20 @@ typedef struct ExtQueue {
 } ExtQueue;
 
 typedef struct CPUS390XState {
-    uint64_t regs[16];	/* GP registers */
+    uint64_t regs[16];     /* GP registers */
+    CPU_DoubleU fregs[16]; /* FP registers */
+    uint32_t aregs[16];    /* access registers */
 
-    uint32_t aregs[16];	/* access registers */
+    uint32_t fpc;          /* floating-point control register */
+    uint32_t cc_op;
 
-    uint32_t fpc;	/* floating-point control register */
-    CPU_DoubleU fregs[16]; /* FP registers */
     float_status fpu_status; /* passed to softfloat lib */
 
+    /* The low part of a 128-bit return, or remainder of a divide.  */
+    uint64_t retxl;
+
     PSW psw;
 
-    uint32_t cc_op;
     uint64_t cc_src;
     uint64_t cc_dst;
     uint64_t cc_vr;
@@ -79,15 +82,15 @@ typedef struct CPUS390XState {
     uint64_t psa;
 
     uint32_t int_pgm_code;
-    uint32_t int_pgm_ilc;
+    uint32_t int_pgm_ilen;
 
     uint32_t int_svc_code;
-    uint32_t int_svc_ilc;
+    uint32_t int_svc_ilen;
 
     uint64_t cregs[16]; /* control registers */
 
-    int pending_int;
     ExtQueue ext_queue[MAX_EXT_QUEUE];
+    int pending_int;
 
     int ext_index;
 
@@ -113,7 +116,7 @@ static inline void cpu_clone_regs(CPUS390XState *env, target_ulong newsp)
     if (newsp) {
         env->regs[15] = newsp;
     }
-    env->regs[0] = 0;
+    env->regs[2] = 0;
 }
 #endif
 
@@ -253,25 +256,31 @@ static inline void cpu_get_tb_cpu_state(CPUS390XState* env, target_ulong *pc,
              ((env->psw.mask & PSW_MASK_32) ? FLAG_MASK_32 : 0);
 }
 
-static inline int get_ilc(uint8_t opc)
+/* While the PoO talks about ILC (a number between 1-3) what is actually
+   stored in LowCore is shifted left one bit (an even between 2-6).  As
+   this is the actual length of the insn and therefore more useful, that
+   is what we want to pass around and manipulate.  To make sure that we
+   have applied this distinction universally, rename the "ILC" to "ILEN".  */
+static inline int get_ilen(uint8_t opc)
 {
     switch (opc >> 6) {
     case 0:
-        return 1;
+        return 2;
     case 1:
     case 2:
-        return 2;
-    case 3:
-        return 3;
+        return 4;
+    default:
+        return 6;
     }
-
-    return 0;
 }
 
-#define ILC_LATER       0x20
-#define ILC_LATER_INC   0x21
-#define ILC_LATER_INC_2 0x22
-
+#ifndef CONFIG_USER_ONLY
+/* In several cases of runtime exceptions, we havn't recorded the true
+   instruction length.  Use these codes when raising exceptions in order
+   to re-compute the length by examining the insn in memory.  */
+#define ILEN_LATER       0x20
+#define ILEN_LATER_INC   0x21
+#endif
 
 S390CPU *cpu_s390x_init(const char *cpu_model);
 void s390x_translate_init(void);
@@ -352,21 +361,10 @@ static inline void cpu_set_tls(CPUS390XState *env, target_ulong newtls)
 
 #include "exec/exec-all.h"
 
-#ifdef CONFIG_USER_ONLY
-
-#define EXCP_OPEX 1 /* operation exception (sigill) */
-#define EXCP_SVC 2 /* supervisor call (syscall) */
-#define EXCP_ADDR 5 /* addressing exception */
-#define EXCP_SPEC 6 /* specification exception */
-
-#else
-
 #define EXCP_EXT 1 /* external interrupt */
 #define EXCP_SVC 2 /* supervisor call (syscall) */
 #define EXCP_PGM 3 /* program interruption */
 
-#endif /* CONFIG_USER_ONLY */
-
 #define INTERRUPT_EXT        (1 << 0)
 #define INTERRUPT_TOD        (1 << 1)
 #define INTERRUPT_CPUTIMER   (1 << 2)
@@ -430,79 +428,6 @@ static inline void cpu_set_tls(CPUS390XState *env, target_ulong newtls)
 /* Total.  */
 #define S390_NUM_REGS 51
 
-/* Pseudo registers -- PC and condition code.  */
-#define S390_PC_REGNUM S390_NUM_REGS
-#define S390_CC_REGNUM (S390_NUM_REGS+1)
-#define S390_NUM_PSEUDO_REGS 2
-#define S390_NUM_TOTAL_REGS (S390_NUM_REGS+2)
-
-
-
-/* Program Status Word.  */
-#define S390_PSWM_REGNUM 0
-#define S390_PSWA_REGNUM 1
-/* General Purpose Registers.  */
-#define S390_R0_REGNUM 2
-#define S390_R1_REGNUM 3
-#define S390_R2_REGNUM 4
-#define S390_R3_REGNUM 5
-#define S390_R4_REGNUM 6
-#define S390_R5_REGNUM 7
-#define S390_R6_REGNUM 8
-#define S390_R7_REGNUM 9
-#define S390_R8_REGNUM 10
-#define S390_R9_REGNUM 11
-#define S390_R10_REGNUM 12
-#define S390_R11_REGNUM 13
-#define S390_R12_REGNUM 14
-#define S390_R13_REGNUM 15
-#define S390_R14_REGNUM 16
-#define S390_R15_REGNUM 17
-/* Access Registers.  */
-#define S390_A0_REGNUM 18
-#define S390_A1_REGNUM 19
-#define S390_A2_REGNUM 20
-#define S390_A3_REGNUM 21
-#define S390_A4_REGNUM 22
-#define S390_A5_REGNUM 23
-#define S390_A6_REGNUM 24
-#define S390_A7_REGNUM 25
-#define S390_A8_REGNUM 26
-#define S390_A9_REGNUM 27
-#define S390_A10_REGNUM 28
-#define S390_A11_REGNUM 29
-#define S390_A12_REGNUM 30
-#define S390_A13_REGNUM 31
-#define S390_A14_REGNUM 32
-#define S390_A15_REGNUM 33
-/* Floating Point Control Word.  */
-#define S390_FPC_REGNUM 34
-/* Floating Point Registers.  */
-#define S390_F0_REGNUM 35
-#define S390_F1_REGNUM 36
-#define S390_F2_REGNUM 37
-#define S390_F3_REGNUM 38
-#define S390_F4_REGNUM 39
-#define S390_F5_REGNUM 40
-#define S390_F6_REGNUM 41
-#define S390_F7_REGNUM 42
-#define S390_F8_REGNUM 43
-#define S390_F9_REGNUM 44
-#define S390_F10_REGNUM 45
-#define S390_F11_REGNUM 46
-#define S390_F12_REGNUM 47
-#define S390_F13_REGNUM 48
-#define S390_F14_REGNUM 49
-#define S390_F15_REGNUM 50
-/* Total.  */
-#define S390_NUM_REGS 51
-
-/* Pseudo registers -- PC and condition code.  */
-#define S390_PC_REGNUM S390_NUM_REGS
-#define S390_CC_REGNUM (S390_NUM_REGS+1)
-#define S390_NUM_PSEUDO_REGS 2
-#define S390_NUM_TOTAL_REGS (S390_NUM_REGS+2)
-
 /* CC optimization */
 
 enum cc_op {
@@ -524,15 +449,19 @@ enum cc_op {
 
     CC_OP_ADD_64,               /* overflow on add (64bit) */
     CC_OP_ADDU_64,              /* overflow on unsigned add (64bit) */
+    CC_OP_ADDC_64,              /* overflow on unsigned add-carry (64bit) */
     CC_OP_SUB_64,               /* overflow on subtraction (64bit) */
     CC_OP_SUBU_64,              /* overflow on unsigned subtraction (64bit) */
+    CC_OP_SUBB_64,              /* overflow on unsigned sub-borrow (64bit) */
     CC_OP_ABS_64,               /* sign eval on abs (64bit) */
     CC_OP_NABS_64,              /* sign eval on nabs (64bit) */
 
     CC_OP_ADD_32,               /* overflow on add (32bit) */
     CC_OP_ADDU_32,              /* overflow on unsigned add (32bit) */
+    CC_OP_ADDC_32,              /* overflow on unsigned add-carry (32bit) */
     CC_OP_SUB_32,               /* overflow on subtraction (32bit) */
     CC_OP_SUBU_32,              /* overflow on unsigned subtraction (32bit) */
+    CC_OP_SUBB_32,              /* overflow on unsigned sub-borrow (32bit) */
     CC_OP_ABS_32,               /* sign eval on abs (64bit) */
     CC_OP_NABS_32,              /* sign eval on nabs (64bit) */
 
@@ -542,14 +471,14 @@ enum cc_op {
     CC_OP_TM_32,                /* test under mask (32bit) */
     CC_OP_TM_64,                /* test under mask (64bit) */
 
-    CC_OP_LTGT_F32,             /* FP compare (32bit) */
-    CC_OP_LTGT_F64,             /* FP compare (64bit) */
-
     CC_OP_NZ_F32,               /* FP dst != 0 (32bit) */
     CC_OP_NZ_F64,               /* FP dst != 0 (64bit) */
+    CC_OP_NZ_F128,              /* FP dst != 0 (128bit) */
 
     CC_OP_ICM,                  /* insert characters under mask */
-    CC_OP_SLAG,                 /* Calculate shift left signed */
+    CC_OP_SLA_32,               /* Calculate shift left signed (32bit) */
+    CC_OP_SLA_64,               /* Calculate shift left signed (64bit) */
+    CC_OP_FLOGR,                /* find leftmost one */
     CC_OP_MAX
 };
 
@@ -569,26 +498,31 @@ static const char *cc_names[] = {
     [CC_OP_LTGT0_64]  = "CC_OP_LTGT0_64",
     [CC_OP_ADD_64]    = "CC_OP_ADD_64",
     [CC_OP_ADDU_64]   = "CC_OP_ADDU_64",
+    [CC_OP_ADDC_64]   = "CC_OP_ADDC_64",
     [CC_OP_SUB_64]    = "CC_OP_SUB_64",
     [CC_OP_SUBU_64]   = "CC_OP_SUBU_64",
+    [CC_OP_SUBB_64]   = "CC_OP_SUBB_64",
     [CC_OP_ABS_64]    = "CC_OP_ABS_64",
     [CC_OP_NABS_64]   = "CC_OP_NABS_64",
     [CC_OP_ADD_32]    = "CC_OP_ADD_32",
     [CC_OP_ADDU_32]   = "CC_OP_ADDU_32",
+    [CC_OP_ADDC_32]   = "CC_OP_ADDC_32",
     [CC_OP_SUB_32]    = "CC_OP_SUB_32",
     [CC_OP_SUBU_32]   = "CC_OP_SUBU_32",
+    [CC_OP_SUBB_32]   = "CC_OP_SUBB_32",
     [CC_OP_ABS_32]    = "CC_OP_ABS_32",
     [CC_OP_NABS_32]   = "CC_OP_NABS_32",
     [CC_OP_COMP_32]   = "CC_OP_COMP_32",
     [CC_OP_COMP_64]   = "CC_OP_COMP_64",
     [CC_OP_TM_32]     = "CC_OP_TM_32",
     [CC_OP_TM_64]     = "CC_OP_TM_64",
-    [CC_OP_LTGT_F32]  = "CC_OP_LTGT_F32",
-    [CC_OP_LTGT_F64]  = "CC_OP_LTGT_F64",
     [CC_OP_NZ_F32]    = "CC_OP_NZ_F32",
     [CC_OP_NZ_F64]    = "CC_OP_NZ_F64",
+    [CC_OP_NZ_F128]   = "CC_OP_NZ_F128",
     [CC_OP_ICM]       = "CC_OP_ICM",
-    [CC_OP_SLAG]      = "CC_OP_SLAG",
+    [CC_OP_SLA_32]    = "CC_OP_SLA_32",
+    [CC_OP_SLA_64]    = "CC_OP_SLA_64",
+    [CC_OP_FLOGR]     = "CC_OP_FLOGR",
 };
 
 static inline const char *cc_name(int cc_op)
@@ -605,9 +539,9 @@ typedef struct LowCore
     uint32_t        ext_params;               /* 0x080 */
     uint16_t        cpu_addr;                 /* 0x084 */
     uint16_t        ext_int_code;             /* 0x086 */
-    uint16_t        svc_ilc;                  /* 0x088 */
+    uint16_t        svc_ilen;                 /* 0x088 */
     uint16_t        svc_code;                 /* 0x08a */
-    uint16_t        pgm_ilc;                  /* 0x08c */
+    uint16_t        pgm_ilen;                 /* 0x08c */
     uint16_t        pgm_code;                 /* 0x08e */
     uint32_t        data_exc_code;            /* 0x090 */
     uint16_t        mon_class_num;            /* 0x094 */
@@ -991,12 +925,13 @@ static inline void cpu_pc_from_tb(CPUS390XState *env, TranslationBlock* tb)
 }
 
 /* fpu_helper.c */
-uint32_t set_cc_f32(CPUS390XState *env, float32 v1, float32 v2);
-uint32_t set_cc_f64(CPUS390XState *env, float64 v1, float64 v2);
 uint32_t set_cc_nz_f32(float32 v);
 uint32_t set_cc_nz_f64(float64 v);
+uint32_t set_cc_nz_f128(float128 v);
 
 /* misc_helper.c */
-void program_interrupt(CPUS390XState *env, uint32_t code, int ilc);
+void program_interrupt(CPUS390XState *env, uint32_t code, int ilen);
+void QEMU_NORETURN runtime_exception(CPUS390XState *env, int excp,
+                                     uintptr_t retaddr);
 
 #endif
diff --git a/target-s390x/fpu_helper.c b/target-s390x/fpu_helper.c
index 173f820428..94375b6a63 100644
--- a/target-s390x/fpu_helper.c
+++ b/target-s390x/fpu_helper.c
@@ -32,6 +32,52 @@
 #define HELPER_LOG(x...)
 #endif
 
+#define RET128(F) (env->retxl = F.low, F.high)
+
+#define convert_bit(mask, from, to) \
+    (to < from                      \
+     ? (mask / (from / to)) & to    \
+     : (mask & from) * (to / from))
+
+static void ieee_exception(CPUS390XState *env, uint32_t dxc, uintptr_t retaddr)
+{
+    /* Install the DXC code.  */
+    env->fpc = (env->fpc & ~0xff00) | (dxc << 8);
+    /* Trap.  */
+    runtime_exception(env, PGM_DATA, retaddr);
+}
+
+/* Should be called after any operation that may raise IEEE exceptions.  */
+static void handle_exceptions(CPUS390XState *env, uintptr_t retaddr)
+{
+    unsigned s390_exc, qemu_exc;
+
+    /* Get the exceptions raised by the current operation.  Reset the
+       fpu_status contents so that the next operation has a clean slate.  */
+    qemu_exc = env->fpu_status.float_exception_flags;
+    if (qemu_exc == 0) {
+        return;
+    }
+    env->fpu_status.float_exception_flags = 0;
+
+    /* Convert softfloat exception bits to s390 exception bits.  */
+    s390_exc = 0;
+    s390_exc |= convert_bit(qemu_exc, float_flag_invalid, 0x80);
+    s390_exc |= convert_bit(qemu_exc, float_flag_divbyzero, 0x40);
+    s390_exc |= convert_bit(qemu_exc, float_flag_overflow, 0x20);
+    s390_exc |= convert_bit(qemu_exc, float_flag_underflow, 0x10);
+    s390_exc |= convert_bit(qemu_exc, float_flag_inexact, 0x08);
+
+    /* Install the exceptions that we raised.  */
+    env->fpc |= s390_exc << 16;
+
+    /* Send signals for enabled exceptions.  */
+    s390_exc &= env->fpc >> 24;
+    if (s390_exc) {
+        ieee_exception(env, s390_exc, retaddr);
+    }
+}
+
 static inline int float_comp_to_cc(CPUS390XState *env, int float_compare)
 {
     switch (float_compare) {
@@ -48,19 +94,6 @@ static inline int float_comp_to_cc(CPUS390XState *env, int float_compare)
     }
 }
 
-/* condition codes for binary FP ops */
-uint32_t set_cc_f32(CPUS390XState *env, float32 v1, float32 v2)
-{
-    return float_comp_to_cc(env, float32_compare_quiet(v1, v2,
-                                                       &env->fpu_status));
-}
-
-uint32_t set_cc_f64(CPUS390XState *env, float64 v1, float64 v2)
-{
-    return float_comp_to_cc(env, float64_compare_quiet(v1, v2,
-                                                       &env->fpu_status));
-}
-
 /* condition codes for unary FP ops */
 uint32_t set_cc_nz_f32(float32 v)
 {
@@ -88,7 +121,7 @@ uint32_t set_cc_nz_f64(float64 v)
     }
 }
 
-static uint32_t set_cc_nz_f128(float128 v)
+uint32_t set_cc_nz_f128(float128 v)
 {
     if (float128_is_any_nan(v)) {
         return 3;
@@ -101,433 +134,211 @@ static uint32_t set_cc_nz_f128(float128 v)
     }
 }
 
-/* convert 32-bit int to 64-bit float */
-void HELPER(cdfbr)(CPUS390XState *env, uint32_t f1, int32_t v2)
+/* 32-bit FP addition */
+uint64_t HELPER(aeb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    HELPER_LOG("%s: converting %d to f%d\n", __func__, v2, f1);
-    env->fregs[f1].d = int32_to_float64(v2, &env->fpu_status);
+    float32 ret = float32_add(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 32-bit int to 128-bit float */
-void HELPER(cxfbr)(CPUS390XState *env, uint32_t f1, int32_t v2)
+/* 64-bit FP addition */
+uint64_t HELPER(adb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    CPU_QuadU v1;
-
-    v1.q = int32_to_float128(v2, &env->fpu_status);
-    env->fregs[f1].ll = v1.ll.upper;
-    env->fregs[f1 + 2].ll = v1.ll.lower;
+    float64 ret = float64_add(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 64-bit int to 32-bit float */
-void HELPER(cegbr)(CPUS390XState *env, uint32_t f1, int64_t v2)
+/* 128-bit FP addition */
+uint64_t HELPER(axb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                     uint64_t bh, uint64_t bl)
 {
-    HELPER_LOG("%s: converting %ld to f%d\n", __func__, v2, f1);
-    env->fregs[f1].l.upper = int64_to_float32(v2, &env->fpu_status);
+    float128 ret = float128_add(make_float128(ah, al),
+                                make_float128(bh, bl),
+                                &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* convert 64-bit int to 64-bit float */
-void HELPER(cdgbr)(CPUS390XState *env, uint32_t f1, int64_t v2)
+/* 32-bit FP subtraction */
+uint64_t HELPER(seb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    HELPER_LOG("%s: converting %ld to f%d\n", __func__, v2, f1);
-    env->fregs[f1].d = int64_to_float64(v2, &env->fpu_status);
+    float32 ret = float32_sub(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 64-bit int to 128-bit float */
-void HELPER(cxgbr)(CPUS390XState *env, uint32_t f1, int64_t v2)
+/* 64-bit FP subtraction */
+uint64_t HELPER(sdb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    CPU_QuadU x1;
-
-    x1.q = int64_to_float128(v2, &env->fpu_status);
-    HELPER_LOG("%s: converted %ld to 0x%lx and 0x%lx\n", __func__, v2,
-               x1.ll.upper, x1.ll.lower);
-    env->fregs[f1].ll = x1.ll.upper;
-    env->fregs[f1 + 2].ll = x1.ll.lower;
+    float64 ret = float64_sub(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 32-bit int to 32-bit float */
-void HELPER(cefbr)(CPUS390XState *env, uint32_t f1, int32_t v2)
+/* 128-bit FP subtraction */
+uint64_t HELPER(sxb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                     uint64_t bh, uint64_t bl)
 {
-    env->fregs[f1].l.upper = int32_to_float32(v2, &env->fpu_status);
-    HELPER_LOG("%s: converting %d to 0x%d in f%d\n", __func__, v2,
-               env->fregs[f1].l.upper, f1);
+    float128 ret = float128_sub(make_float128(ah, al),
+                                make_float128(bh, bl),
+                                &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* 32-bit FP addition RR */
-uint32_t HELPER(aebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 32-bit FP division */
+uint64_t HELPER(deb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    env->fregs[f1].l.upper = float32_add(env->fregs[f1].l.upper,
-                                         env->fregs[f2].l.upper,
-                                         &env->fpu_status);
-    HELPER_LOG("%s: adding 0x%d resulting in 0x%d in f%d\n", __func__,
-               env->fregs[f2].l.upper, env->fregs[f1].l.upper, f1);
-
-    return set_cc_nz_f32(env->fregs[f1].l.upper);
+    float32 ret = float32_div(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP addition RR */
-uint32_t HELPER(adbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 64-bit FP division */
+uint64_t HELPER(ddb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    env->fregs[f1].d = float64_add(env->fregs[f1].d, env->fregs[f2].d,
-                                   &env->fpu_status);
-    HELPER_LOG("%s: adding 0x%ld resulting in 0x%ld in f%d\n", __func__,
-               env->fregs[f2].d, env->fregs[f1].d, f1);
-
-    return set_cc_nz_f64(env->fregs[f1].d);
+    float64 ret = float64_div(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 32-bit FP subtraction RR */
-uint32_t HELPER(sebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 128-bit FP division */
+uint64_t HELPER(dxb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                     uint64_t bh, uint64_t bl)
 {
-    env->fregs[f1].l.upper = float32_sub(env->fregs[f1].l.upper,
-                                         env->fregs[f2].l.upper,
-                                         &env->fpu_status);
-    HELPER_LOG("%s: adding 0x%d resulting in 0x%d in f%d\n", __func__,
-               env->fregs[f2].l.upper, env->fregs[f1].l.upper, f1);
-
-    return set_cc_nz_f32(env->fregs[f1].l.upper);
+    float128 ret = float128_div(make_float128(ah, al),
+                                make_float128(bh, bl),
+                                &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* 64-bit FP subtraction RR */
-uint32_t HELPER(sdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 32-bit FP multiplication */
+uint64_t HELPER(meeb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    env->fregs[f1].d = float64_sub(env->fregs[f1].d, env->fregs[f2].d,
-                                   &env->fpu_status);
-    HELPER_LOG("%s: subtracting 0x%ld resulting in 0x%ld in f%d\n",
-               __func__, env->fregs[f2].d, env->fregs[f1].d, f1);
-
-    return set_cc_nz_f64(env->fregs[f1].d);
+    float32 ret = float32_mul(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 32-bit FP division RR */
-void HELPER(debr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 64-bit FP multiplication */
+uint64_t HELPER(mdb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    env->fregs[f1].l.upper = float32_div(env->fregs[f1].l.upper,
-                                         env->fregs[f2].l.upper,
-                                         &env->fpu_status);
+    float64 ret = float64_mul(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 128-bit FP division RR */
-void HELPER(dxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 64/32-bit FP multiplication */
+uint64_t HELPER(mdeb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-    CPU_QuadU res;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    res.q = float128_div(v1.q, v2.q, &env->fpu_status);
-    env->fregs[f1].ll = res.ll.upper;
-    env->fregs[f1 + 2].ll = res.ll.lower;
+    float64 ret = float32_to_float64(f2, &env->fpu_status);
+    ret = float64_mul(f1, ret, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP multiplication RR */
-void HELPER(mdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 128-bit FP multiplication */
+uint64_t HELPER(mxb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                     uint64_t bh, uint64_t bl)
 {
-    env->fregs[f1].d = float64_mul(env->fregs[f1].d, env->fregs[f2].d,
-                                   &env->fpu_status);
+    float128 ret = float128_mul(make_float128(ah, al),
+                                make_float128(bh, bl),
+                                &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* 128-bit FP multiplication RR */
-void HELPER(mxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* 128/64-bit FP multiplication */
+uint64_t HELPER(mxdb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                      uint64_t f2)
 {
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-    CPU_QuadU res;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    res.q = float128_mul(v1.q, v2.q, &env->fpu_status);
-    env->fregs[f1].ll = res.ll.upper;
-    env->fregs[f1 + 2].ll = res.ll.lower;
+    float128 ret = float64_to_float128(f2, &env->fpu_status);
+    ret = float128_mul(make_float128(ah, al), ret, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
 /* convert 32-bit float to 64-bit float */
-void HELPER(ldebr)(CPUS390XState *env, uint32_t r1, uint32_t r2)
+uint64_t HELPER(ldeb)(CPUS390XState *env, uint64_t f2)
 {
-    env->fregs[r1].d = float32_to_float64(env->fregs[r2].l.upper,
-                                          &env->fpu_status);
+    float64 ret = float32_to_float64(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
 /* convert 128-bit float to 64-bit float */
-void HELPER(ldxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+uint64_t HELPER(ldxb)(CPUS390XState *env, uint64_t ah, uint64_t al)
 {
-    CPU_QuadU x2;
-
-    x2.ll.upper = env->fregs[f2].ll;
-    x2.ll.lower = env->fregs[f2 + 2].ll;
-    env->fregs[f1].d = float128_to_float64(x2.q, &env->fpu_status);
-    HELPER_LOG("%s: to 0x%ld\n", __func__, env->fregs[f1].d);
+    float64 ret = float128_to_float64(make_float128(ah, al), &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
 /* convert 64-bit float to 128-bit float */
-void HELPER(lxdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU res;
-
-    res.q = float64_to_float128(env->fregs[f2].d, &env->fpu_status);
-    env->fregs[f1].ll = res.ll.upper;
-    env->fregs[f1 + 2].ll = res.ll.lower;
-}
-
-/* convert 64-bit float to 32-bit float */
-void HELPER(ledbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    float64 d2 = env->fregs[f2].d;
-
-    env->fregs[f1].l.upper = float64_to_float32(d2, &env->fpu_status);
-}
-
-/* convert 128-bit float to 32-bit float */
-void HELPER(lexbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU x2;
-
-    x2.ll.upper = env->fregs[f2].ll;
-    x2.ll.lower = env->fregs[f2 + 2].ll;
-    env->fregs[f1].l.upper = float128_to_float32(x2.q, &env->fpu_status);
-    HELPER_LOG("%s: to 0x%d\n", __func__, env->fregs[f1].l.upper);
-}
-
-/* absolute value of 32-bit float */
-uint32_t HELPER(lpebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    float32 v1;
-    float32 v2 = env->fregs[f2].d;
-
-    v1 = float32_abs(v2);
-    env->fregs[f1].d = v1;
-    return set_cc_nz_f32(v1);
-}
-
-/* absolute value of 64-bit float */
-uint32_t HELPER(lpdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    float64 v1;
-    float64 v2 = env->fregs[f2].d;
-
-    v1 = float64_abs(v2);
-    env->fregs[f1].d = v1;
-    return set_cc_nz_f64(v1);
-}
-
-/* absolute value of 128-bit float */
-uint32_t HELPER(lpxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    v1.q = float128_abs(v2.q);
-    env->fregs[f1].ll = v1.ll.upper;
-    env->fregs[f1 + 2].ll = v1.ll.lower;
-    return set_cc_nz_f128(v1.q);
-}
-
-/* load and test 64-bit float */
-uint32_t HELPER(ltdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    env->fregs[f1].d = env->fregs[f2].d;
-    return set_cc_nz_f64(env->fregs[f1].d);
-}
-
-/* load and test 32-bit float */
-uint32_t HELPER(ltebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    env->fregs[f1].l.upper = env->fregs[f2].l.upper;
-    return set_cc_nz_f32(env->fregs[f1].l.upper);
-}
-
-/* load and test 128-bit float */
-uint32_t HELPER(ltxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU x;
-
-    x.ll.upper = env->fregs[f2].ll;
-    x.ll.lower = env->fregs[f2 + 2].ll;
-    env->fregs[f1].ll = x.ll.upper;
-    env->fregs[f1 + 2].ll = x.ll.lower;
-    return set_cc_nz_f128(x.q);
-}
-
-/* load complement of 32-bit float */
-uint32_t HELPER(lcebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    env->fregs[f1].l.upper = float32_chs(env->fregs[f2].l.upper);
-
-    return set_cc_nz_f32(env->fregs[f1].l.upper);
-}
-
-/* load complement of 64-bit float */
-uint32_t HELPER(lcdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    env->fregs[f1].d = float64_chs(env->fregs[f2].d);
-
-    return set_cc_nz_f64(env->fregs[f1].d);
-}
-
-/* load complement of 128-bit float */
-uint32_t HELPER(lcxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU x1, x2;
-
-    x2.ll.upper = env->fregs[f2].ll;
-    x2.ll.lower = env->fregs[f2 + 2].ll;
-    x1.q = float128_chs(x2.q);
-    env->fregs[f1].ll = x1.ll.upper;
-    env->fregs[f1 + 2].ll = x1.ll.lower;
-    return set_cc_nz_f128(x1.q);
-}
-
-/* 32-bit FP addition RM */
-void HELPER(aeb)(CPUS390XState *env, uint32_t f1, uint32_t val)
-{
-    float32 v1 = env->fregs[f1].l.upper;
-    CPU_FloatU v2;
-
-    v2.l = val;
-    HELPER_LOG("%s: adding 0x%d from f%d and 0x%d\n", __func__,
-               v1, f1, v2.f);
-    env->fregs[f1].l.upper = float32_add(v1, v2.f, &env->fpu_status);
-}
-
-/* 32-bit FP division RM */
-void HELPER(deb)(CPUS390XState *env, uint32_t f1, uint32_t val)
-{
-    float32 v1 = env->fregs[f1].l.upper;
-    CPU_FloatU v2;
-
-    v2.l = val;
-    HELPER_LOG("%s: dividing 0x%d from f%d by 0x%d\n", __func__,
-               v1, f1, v2.f);
-    env->fregs[f1].l.upper = float32_div(v1, v2.f, &env->fpu_status);
-}
-
-/* 32-bit FP multiplication RM */
-void HELPER(meeb)(CPUS390XState *env, uint32_t f1, uint32_t val)
+uint64_t HELPER(lxdb)(CPUS390XState *env, uint64_t f2)
 {
-    float32 v1 = env->fregs[f1].l.upper;
-    CPU_FloatU v2;
-
-    v2.l = val;
-    HELPER_LOG("%s: multiplying 0x%d from f%d and 0x%d\n", __func__,
-               v1, f1, v2.f);
-    env->fregs[f1].l.upper = float32_mul(v1, v2.f, &env->fpu_status);
+    float128 ret = float64_to_float128(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* 32-bit FP compare RR */
-uint32_t HELPER(cebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* convert 32-bit float to 128-bit float */
+uint64_t HELPER(lxeb)(CPUS390XState *env, uint64_t f2)
 {
-    float32 v1 = env->fregs[f1].l.upper;
-    float32 v2 = env->fregs[f2].l.upper;
-
-    HELPER_LOG("%s: comparing 0x%d from f%d and 0x%d\n", __func__,
-               v1, f1, v2);
-    return set_cc_f32(env, v1, v2);
+    float128 ret = float32_to_float128(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* 64-bit FP compare RR */
-uint32_t HELPER(cdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    float64 v1 = env->fregs[f1].d;
-    float64 v2 = env->fregs[f2].d;
-
-    HELPER_LOG("%s: comparing 0x%ld from f%d and 0x%ld\n", __func__,
-               v1, f1, v2);
-    return set_cc_f64(env, v1, v2);
-}
-
-/* 128-bit FP compare RR */
-uint32_t HELPER(cxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
-{
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-
-    return float_comp_to_cc(env, float128_compare_quiet(v1.q, v2.q,
-                                                   &env->fpu_status));
-}
-
-/* 64-bit FP compare RM */
-uint32_t HELPER(cdb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* convert 64-bit float to 32-bit float */
+uint64_t HELPER(ledb)(CPUS390XState *env, uint64_t f2)
 {
-    float64 v1 = env->fregs[f1].d;
-    CPU_DoubleU v2;
-
-    v2.ll = cpu_ldq_data(env, a2);
-    HELPER_LOG("%s: comparing 0x%ld from f%d and 0x%lx\n", __func__, v1,
-               f1, v2.d);
-    return set_cc_f64(env, v1, v2.d);
+    float32 ret = float64_to_float32(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP addition RM */
-uint32_t HELPER(adb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* convert 128-bit float to 32-bit float */
+uint64_t HELPER(lexb)(CPUS390XState *env, uint64_t ah, uint64_t al)
 {
-    float64 v1 = env->fregs[f1].d;
-    CPU_DoubleU v2;
-
-    v2.ll = cpu_ldq_data(env, a2);
-    HELPER_LOG("%s: adding 0x%lx from f%d and 0x%lx\n", __func__,
-               v1, f1, v2.d);
-    env->fregs[f1].d = v1 = float64_add(v1, v2.d, &env->fpu_status);
-    return set_cc_nz_f64(v1);
+    float32 ret = float128_to_float32(make_float128(ah, al), &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 32-bit FP subtraction RM */
-void HELPER(seb)(CPUS390XState *env, uint32_t f1, uint32_t val)
+/* 32-bit FP compare */
+uint32_t HELPER(ceb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    float32 v1 = env->fregs[f1].l.upper;
-    CPU_FloatU v2;
-
-    v2.l = val;
-    env->fregs[f1].l.upper = float32_sub(v1, v2.f, &env->fpu_status);
+    int cmp = float32_compare_quiet(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return float_comp_to_cc(env, cmp);
 }
 
-/* 64-bit FP subtraction RM */
-uint32_t HELPER(sdb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* 64-bit FP compare */
+uint32_t HELPER(cdb)(CPUS390XState *env, uint64_t f1, uint64_t f2)
 {
-    float64 v1 = env->fregs[f1].d;
-    CPU_DoubleU v2;
-
-    v2.ll = cpu_ldq_data(env, a2);
-    env->fregs[f1].d = v1 = float64_sub(v1, v2.d, &env->fpu_status);
-    return set_cc_nz_f64(v1);
+    int cmp = float64_compare_quiet(f1, f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return float_comp_to_cc(env, cmp);
 }
 
-/* 64-bit FP multiplication RM */
-void HELPER(mdb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* 128-bit FP compare */
+uint32_t HELPER(cxb)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                     uint64_t bh, uint64_t bl)
 {
-    float64 v1 = env->fregs[f1].d;
-    CPU_DoubleU v2;
-
-    v2.ll = cpu_ldq_data(env, a2);
-    HELPER_LOG("%s: multiplying 0x%lx from f%d and 0x%ld\n", __func__,
-               v1, f1, v2.d);
-    env->fregs[f1].d = float64_mul(v1, v2.d, &env->fpu_status);
+    int cmp = float128_compare_quiet(make_float128(ah, al),
+                                     make_float128(bh, bl),
+                                     &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return float_comp_to_cc(env, cmp);
 }
 
-/* 64-bit FP division RM */
-void HELPER(ddb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
-{
-    float64 v1 = env->fregs[f1].d;
-    CPU_DoubleU v2;
-
-    v2.ll = cpu_ldq_data(env, a2);
-    HELPER_LOG("%s: dividing 0x%lx from f%d by 0x%ld\n", __func__,
-               v1, f1, v2.d);
-    env->fregs[f1].d = float64_div(v1, v2.d, &env->fpu_status);
-}
-
-static void set_round_mode(CPUS390XState *env, int m3)
+static int swap_round_mode(CPUS390XState *env, int m3)
 {
+    int ret = env->fpu_status.float_rounding_mode;
     switch (m3) {
     case 0:
         /* current mode */
@@ -551,232 +362,242 @@ static void set_round_mode(CPUS390XState *env, int m3)
         set_float_rounding_mode(float_round_down, &env->fpu_status);
         break;
     }
+    return ret;
 }
 
-/* convert 32-bit float to 64-bit int */
-uint32_t HELPER(cgebr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit int to 32-bit float */
+uint64_t HELPER(cegb)(CPUS390XState *env, int64_t v2, uint32_t m3)
 {
-    float32 v2 = env->fregs[f2].l.upper;
-
-    set_round_mode(env, m3);
-    env->regs[r1] = float32_to_int64(v2, &env->fpu_status);
-    return set_cc_nz_f32(v2);
+    int hold = swap_round_mode(env, m3);
+    float32 ret = int64_to_float32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 64-bit float to 64-bit int */
-uint32_t HELPER(cgdbr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit int to 64-bit float */
+uint64_t HELPER(cdgb)(CPUS390XState *env, int64_t v2, uint32_t m3)
 {
-    float64 v2 = env->fregs[f2].d;
-
-    set_round_mode(env, m3);
-    env->regs[r1] = float64_to_int64(v2, &env->fpu_status);
-    return set_cc_nz_f64(v2);
+    int hold = swap_round_mode(env, m3);
+    float64 ret = int64_to_float64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 128-bit float to 64-bit int */
-uint32_t HELPER(cgxbr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit int to 128-bit float */
+uint64_t HELPER(cxgb)(CPUS390XState *env, int64_t v2, uint32_t m3)
 {
-    CPU_QuadU v2;
-
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    set_round_mode(env, m3);
-    env->regs[r1] = float128_to_int64(v2.q, &env->fpu_status);
-    if (float128_is_any_nan(v2.q)) {
-        return 3;
-    } else if (float128_is_zero(v2.q)) {
-        return 0;
-    } else if (float128_is_neg(v2.q)) {
-        return 1;
-    } else {
-        return 2;
-    }
+    int hold = swap_round_mode(env, m3);
+    float128 ret = int64_to_float128(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* convert 32-bit float to 32-bit int */
-uint32_t HELPER(cfebr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit uint to 32-bit float */
+uint64_t HELPER(celgb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    float32 v2 = env->fregs[f2].l.upper;
-
-    set_round_mode(env, m3);
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
-        float32_to_int32(v2, &env->fpu_status);
-    return set_cc_nz_f32(v2);
+    int hold = swap_round_mode(env, m3);
+    float32 ret = uint64_to_float32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 64-bit float to 32-bit int */
-uint32_t HELPER(cfdbr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit uint to 64-bit float */
+uint64_t HELPER(cdlgb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    float64 v2 = env->fregs[f2].d;
-
-    set_round_mode(env, m3);
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
-        float64_to_int32(v2, &env->fpu_status);
-    return set_cc_nz_f64(v2);
+    int hold = swap_round_mode(env, m3);
+    float64 ret = uint64_to_float64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 128-bit float to 32-bit int */
-uint32_t HELPER(cfxbr)(CPUS390XState *env, uint32_t r1, uint32_t f2,
-                       uint32_t m3)
+/* convert 64-bit uint to 128-bit float */
+uint64_t HELPER(cxlgb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    CPU_QuadU v2;
-
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
-        float128_to_int32(v2.q, &env->fpu_status);
-    return set_cc_nz_f128(v2.q);
+    int hold = swap_round_mode(env, m3);
+    float128 ret = uint64_to_float128(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
 }
 
-/* load 32-bit FP zero */
-void HELPER(lzer)(CPUS390XState *env, uint32_t f1)
+/* convert 32-bit float to 64-bit int */
+uint64_t HELPER(cgeb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    env->fregs[f1].l.upper = float32_zero;
+    int hold = swap_round_mode(env, m3);
+    int64_t ret = float32_to_int64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* load 64-bit FP zero */
-void HELPER(lzdr)(CPUS390XState *env, uint32_t f1)
+/* convert 64-bit float to 64-bit int */
+uint64_t HELPER(cgdb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    env->fregs[f1].d = float64_zero;
+    int hold = swap_round_mode(env, m3);
+    int64_t ret = float64_to_int64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* load 128-bit FP zero */
-void HELPER(lzxr)(CPUS390XState *env, uint32_t f1)
+/* convert 128-bit float to 64-bit int */
+uint64_t HELPER(cgxb)(CPUS390XState *env, uint64_t h, uint64_t l, uint32_t m3)
 {
-    CPU_QuadU x;
-
-    x.q = float64_to_float128(float64_zero, &env->fpu_status);
-    env->fregs[f1].ll = x.ll.upper;
-    env->fregs[f1 + 1].ll = x.ll.lower;
+    int hold = swap_round_mode(env, m3);
+    float128 v2 = make_float128(h, l);
+    int64_t ret = float128_to_int64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 128-bit FP subtraction RR */
-uint32_t HELPER(sxbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* convert 32-bit float to 32-bit int */
+uint64_t HELPER(cfeb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-    CPU_QuadU res;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    res.q = float128_sub(v1.q, v2.q, &env->fpu_status);
-    env->fregs[f1].ll = res.ll.upper;
-    env->fregs[f1 + 2].ll = res.ll.lower;
-    return set_cc_nz_f128(res.q);
+    int hold = swap_round_mode(env, m3);
+    int32_t ret = float32_to_int32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 128-bit FP addition RR */
-uint32_t HELPER(axbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* convert 64-bit float to 32-bit int */
+uint64_t HELPER(cfdb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    CPU_QuadU v1;
-    CPU_QuadU v2;
-    CPU_QuadU res;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
-    v2.ll.upper = env->fregs[f2].ll;
-    v2.ll.lower = env->fregs[f2 + 2].ll;
-    res.q = float128_add(v1.q, v2.q, &env->fpu_status);
-    env->fregs[f1].ll = res.ll.upper;
-    env->fregs[f1 + 2].ll = res.ll.lower;
-    return set_cc_nz_f128(res.q);
+    int hold = swap_round_mode(env, m3);
+    int32_t ret = float64_to_int32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 32-bit FP multiplication RR */
-void HELPER(meebr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* convert 128-bit float to 32-bit int */
+uint64_t HELPER(cfxb)(CPUS390XState *env, uint64_t h, uint64_t l, uint32_t m3)
 {
-    env->fregs[f1].l.upper = float32_mul(env->fregs[f1].l.upper,
-                                         env->fregs[f2].l.upper,
-                                         &env->fpu_status);
+    int hold = swap_round_mode(env, m3);
+    float128 v2 = make_float128(h, l);
+    int32_t ret = float128_to_int32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP division RR */
-void HELPER(ddbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* convert 32-bit float to 64-bit uint */
+uint64_t HELPER(clgeb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    env->fregs[f1].d = float64_div(env->fregs[f1].d, env->fregs[f2].d,
-                                   &env->fpu_status);
+    int hold = swap_round_mode(env, m3);
+    uint64_t ret;
+    v2 = float32_to_float64(v2, &env->fpu_status);
+    ret = float64_to_uint64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP multiply and add RM */
-void HELPER(madb)(CPUS390XState *env, uint32_t f1, uint64_t a2, uint32_t f3)
+/* convert 64-bit float to 64-bit uint */
+uint64_t HELPER(clgdb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    CPU_DoubleU v2;
+    int hold = swap_round_mode(env, m3);
+    uint64_t ret = float64_to_uint64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
+}
 
-    HELPER_LOG("%s: f1 %d a2 0x%lx f3 %d\n", __func__, f1, a2, f3);
-    v2.ll = cpu_ldq_data(env, a2);
-    env->fregs[f1].d = float64_add(env->fregs[f1].d,
-                                   float64_mul(v2.d, env->fregs[f3].d,
-                                               &env->fpu_status),
-                                   &env->fpu_status);
+/* convert 128-bit float to 64-bit uint */
+uint64_t HELPER(clgxb)(CPUS390XState *env, uint64_t h, uint64_t l, uint32_t m3)
+{
+    int hold = swap_round_mode(env, m3);
+    float128 v2 = make_float128(h, l);
+    /* ??? Not 100% correct.  */
+    uint64_t ret = float128_to_int64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP multiply and add RR */
-void HELPER(madbr)(CPUS390XState *env, uint32_t f1, uint32_t f3, uint32_t f2)
+/* convert 32-bit float to 32-bit uint */
+uint64_t HELPER(clfeb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    HELPER_LOG("%s: f1 %d f2 %d f3 %d\n", __func__, f1, f2, f3);
-    env->fregs[f1].d = float64_add(float64_mul(env->fregs[f2].d,
-                                               env->fregs[f3].d,
-                                               &env->fpu_status),
-                                   env->fregs[f1].d, &env->fpu_status);
+    int hold = swap_round_mode(env, m3);
+    uint32_t ret = float32_to_uint32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 64-bit FP multiply and subtract RR */
-void HELPER(msdbr)(CPUS390XState *env, uint32_t f1, uint32_t f3, uint32_t f2)
+/* convert 64-bit float to 32-bit uint */
+uint64_t HELPER(clfdb)(CPUS390XState *env, uint64_t v2, uint32_t m3)
 {
-    HELPER_LOG("%s: f1 %d f2 %d f3 %d\n", __func__, f1, f2, f3);
-    env->fregs[f1].d = float64_sub(float64_mul(env->fregs[f2].d,
-                                               env->fregs[f3].d,
-                                               &env->fpu_status),
-                                   env->fregs[f1].d, &env->fpu_status);
+    int hold = swap_round_mode(env, m3);
+    uint32_t ret = float64_to_uint32(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* 32-bit FP multiply and add RR */
-void HELPER(maebr)(CPUS390XState *env, uint32_t f1, uint32_t f3, uint32_t f2)
+/* convert 128-bit float to 32-bit uint */
+uint64_t HELPER(clfxb)(CPUS390XState *env, uint64_t h, uint64_t l, uint32_t m3)
 {
-    env->fregs[f1].l.upper = float32_add(env->fregs[f1].l.upper,
-                                         float32_mul(env->fregs[f2].l.upper,
-                                                     env->fregs[f3].l.upper,
-                                                     &env->fpu_status),
-                                         &env->fpu_status);
+    int hold = swap_round_mode(env, m3);
+    float128 v2 = make_float128(h, l);
+    /* Not 100% correct.  */
+    uint32_t ret = float128_to_int64(v2, &env->fpu_status);
+    set_float_rounding_mode(hold, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 32-bit float to 64-bit float */
-void HELPER(ldeb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* 32-bit FP multiply and add */
+uint64_t HELPER(maeb)(CPUS390XState *env, uint64_t f1,
+                      uint64_t f2, uint64_t f3)
 {
-    uint32_t v2;
+    float32 ret = float32_muladd(f2, f3, f1, 0, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
+}
 
-    v2 = cpu_ldl_data(env, a2);
-    env->fregs[f1].d = float32_to_float64(v2,
-                                          &env->fpu_status);
+/* 64-bit FP multiply and add */
+uint64_t HELPER(madb)(CPUS390XState *env, uint64_t f1,
+                      uint64_t f2, uint64_t f3)
+{
+    float64 ret = float64_muladd(f2, f3, f1, 0, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
-/* convert 64-bit float to 128-bit float */
-void HELPER(lxdb)(CPUS390XState *env, uint32_t f1, uint64_t a2)
+/* 32-bit FP multiply and subtract */
+uint64_t HELPER(mseb)(CPUS390XState *env, uint64_t f1,
+                      uint64_t f2, uint64_t f3)
 {
-    CPU_DoubleU v2;
-    CPU_QuadU v1;
+    float32 ret = float32_muladd(f2, f3, f1, float_muladd_negate_c,
+                                 &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
+}
 
-    v2.ll = cpu_ldq_data(env, a2);
-    v1.q = float64_to_float128(v2.d, &env->fpu_status);
-    env->fregs[f1].ll = v1.ll.upper;
-    env->fregs[f1 + 2].ll = v1.ll.lower;
+/* 64-bit FP multiply and subtract */
+uint64_t HELPER(msdb)(CPUS390XState *env, uint64_t f1,
+                      uint64_t f2, uint64_t f3)
+{
+    float64 ret = float64_muladd(f2, f3, f1, float_muladd_negate_c,
+                                 &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
 }
 
 /* test data class 32-bit */
-uint32_t HELPER(tceb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
+uint32_t HELPER(tceb)(uint64_t f1, uint64_t m2)
 {
-    float32 v1 = env->fregs[f1].l.upper;
+    float32 v1 = f1;
     int neg = float32_is_neg(v1);
     uint32_t cc = 0;
 
-    HELPER_LOG("%s: v1 0x%lx m2 0x%lx neg %d\n", __func__, (long)v1, m2, neg);
     if ((float32_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
         (float32_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
         (float32_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
@@ -786,19 +607,16 @@ uint32_t HELPER(tceb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
         /* assume normalized number */
         cc = 1;
     }
-
     /* FIXME: denormalized? */
     return cc;
 }
 
 /* test data class 64-bit */
-uint32_t HELPER(tcdb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
+uint32_t HELPER(tcdb)(uint64_t v1, uint64_t m2)
 {
-    float64 v1 = env->fregs[f1].d;
     int neg = float64_is_neg(v1);
     uint32_t cc = 0;
 
-    HELPER_LOG("%s: v1 0x%lx m2 0x%lx neg %d\n", __func__, v1, m2, neg);
     if ((float64_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
         (float64_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
         (float64_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
@@ -813,20 +631,16 @@ uint32_t HELPER(tcdb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
 }
 
 /* test data class 128-bit */
-uint32_t HELPER(tcxb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
+uint32_t HELPER(tcxb)(uint64_t ah, uint64_t al, uint64_t m2)
 {
-    CPU_QuadU v1;
+    float128 v1 = make_float128(ah, al);
+    int neg = float128_is_neg(v1);
     uint32_t cc = 0;
-    int neg;
-
-    v1.ll.upper = env->fregs[f1].ll;
-    v1.ll.lower = env->fregs[f1 + 2].ll;
 
-    neg = float128_is_neg(v1.q);
-    if ((float128_is_zero(v1.q) && (m2 & (1 << (11-neg)))) ||
-        (float128_is_infinity(v1.q) && (m2 & (1 << (5-neg)))) ||
-        (float128_is_any_nan(v1.q) && (m2 & (1 << (3-neg)))) ||
-        (float128_is_signaling_nan(v1.q) && (m2 & (1 << (1-neg))))) {
+    if ((float128_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
+        (float128_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
+        (float128_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
+        (float128_is_signaling_nan(v1) && (m2 & (1 << (1-neg))))) {
         cc = 1;
     } else if (m2 & (1 << (9-neg))) {
         /* assume normalized number */
@@ -836,8 +650,64 @@ uint32_t HELPER(tcxb)(CPUS390XState *env, uint32_t f1, uint64_t m2)
     return cc;
 }
 
-/* square root 64-bit RR */
-void HELPER(sqdbr)(CPUS390XState *env, uint32_t f1, uint32_t f2)
+/* square root 32-bit */
+uint64_t HELPER(sqeb)(CPUS390XState *env, uint64_t f2)
+{
+    float32 ret = float32_sqrt(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
+}
+
+/* square root 64-bit */
+uint64_t HELPER(sqdb)(CPUS390XState *env, uint64_t f2)
+{
+    float64 ret = float64_sqrt(f2, &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return ret;
+}
+
+/* square root 128-bit */
+uint64_t HELPER(sqxb)(CPUS390XState *env, uint64_t ah, uint64_t al)
+{
+    float128 ret = float128_sqrt(make_float128(ah, al), &env->fpu_status);
+    handle_exceptions(env, GETPC());
+    return RET128(ret);
+}
+
+static const int fpc_to_rnd[4] = {
+    float_round_nearest_even,
+    float_round_to_zero,
+    float_round_up,
+    float_round_down
+};
+
+/* set fpc */
+void HELPER(sfpc)(CPUS390XState *env, uint64_t fpc)
+{
+    /* Install everything in the main FPC.  */
+    env->fpc = fpc;
+
+    /* Install the rounding mode in the shadow fpu_status.  */
+    set_float_rounding_mode(fpc_to_rnd[fpc & 3], &env->fpu_status);
+}
+
+/* set fpc and signal */
+void HELPER(sfas)(CPUS390XState *env, uint64_t val)
 {
-    env->fregs[f1].d = float64_sqrt(env->fregs[f2].d, &env->fpu_status);
+    uint32_t signalling = env->fpc;
+    uint32_t source = val;
+    uint32_t s390_exc;
+
+    /* The contents of the source operand are placed in the FPC register;
+       then the flags in the FPC register are set to the logical OR of the
+       signalling flags and the source flags.  */
+    env->fpc = source | (signalling & 0x00ff0000);
+    set_float_rounding_mode(fpc_to_rnd[source & 3], &env->fpu_status);
+
+    /* If any signalling flag is 1 and the corresponding source mask
+       is also 1, a simulated-iee-exception trap occurs.  */
+    s390_exc = (signalling >> 16) & (source >> 24);
+    if (s390_exc) {
+        ieee_exception(env, s390_exc | 3, GETPC());
+    }
 }
diff --git a/target-s390x/helper.c b/target-s390x/helper.c
index 42e06eb85e..9a132e6d2c 100644
--- a/target-s390x/helper.c
+++ b/target-s390x/helper.c
@@ -99,10 +99,10 @@ void do_interrupt(CPUS390XState *env)
 int cpu_s390x_handle_mmu_fault(CPUS390XState *env, target_ulong address,
                                int rw, int mmu_idx)
 {
-    /* fprintf(stderr, "%s: address 0x%lx rw %d mmu_idx %d\n",
-       __func__, address, rw, mmu_idx); */
-    env->exception_index = EXCP_ADDR;
-    /* FIXME: find out how this works on a real machine */
+    env->exception_index = EXCP_PGM;
+    env->int_pgm_code = PGM_ADDRESSING;
+    /* On real machines this value is dropped into LowMem.  Since this
+       is userland, simply put this someplace that cpu_loop can find it.  */
     env->__excp_addr = address;
     return 1;
 }
@@ -111,11 +111,11 @@ int cpu_s390x_handle_mmu_fault(CPUS390XState *env, target_ulong address,
 
 /* Ensure to exit the TB after this call! */
 static void trigger_pgm_exception(CPUS390XState *env, uint32_t code,
-                                  uint32_t ilc)
+                                  uint32_t ilen)
 {
     env->exception_index = EXCP_PGM;
     env->int_pgm_code = code;
-    env->int_pgm_ilc = ilc;
+    env->int_pgm_ilen = ilen;
 }
 
 static int trans_bits(CPUS390XState *env, uint64_t mode)
@@ -143,30 +143,30 @@ static int trans_bits(CPUS390XState *env, uint64_t mode)
 static void trigger_prot_fault(CPUS390XState *env, target_ulong vaddr,
                                uint64_t mode)
 {
-    int ilc = ILC_LATER_INC_2;
+    int ilen = ILEN_LATER_INC;
     int bits = trans_bits(env, mode) | 4;
 
     DPRINTF("%s: vaddr=%016" PRIx64 " bits=%d\n", __func__, vaddr, bits);
 
     stq_phys(env->psa + offsetof(LowCore, trans_exc_code), vaddr | bits);
-    trigger_pgm_exception(env, PGM_PROTECTION, ilc);
+    trigger_pgm_exception(env, PGM_PROTECTION, ilen);
 }
 
 static void trigger_page_fault(CPUS390XState *env, target_ulong vaddr,
                                uint32_t type, uint64_t asc, int rw)
 {
-    int ilc = ILC_LATER;
+    int ilen = ILEN_LATER;
     int bits = trans_bits(env, asc);
 
+    /* Code accesses have an undefined ilc.  */
     if (rw == 2) {
-        /* code has is undefined ilc */
-        ilc = 2;
+        ilen = 2;
     }
 
     DPRINTF("%s: vaddr=%016" PRIx64 " bits=%d\n", __func__, vaddr, bits);
 
     stq_phys(env->psa + offsetof(LowCore, trans_exc_code), vaddr | bits);
-    trigger_pgm_exception(env, type, ilc);
+    trigger_pgm_exception(env, type, ilen);
 }
 
 static int mmu_translate_asce(CPUS390XState *env, target_ulong vaddr,
@@ -406,7 +406,7 @@ int cpu_s390x_handle_mmu_fault(CPUS390XState *env, target_ulong orig_vaddr,
     if (raddr > (ram_size + virtio_size)) {
         DPRINTF("%s: aaddr %" PRIx64 " > ram_size %" PRIx64 "\n", __func__,
                 (uint64_t)aaddr, (uint64_t)ram_size);
-        trigger_pgm_exception(env, PGM_ADDRESSING, ILC_LATER);
+        trigger_pgm_exception(env, PGM_ADDRESSING, ILEN_LATER);
         return 1;
     }
 
@@ -454,18 +454,19 @@ void load_psw(CPUS390XState *env, uint64_t mask, uint64_t addr)
 
     env->psw.addr = addr;
     env->psw.mask = mask;
-    env->cc_op = (mask >> 13) & 3;
+    env->cc_op = (mask >> 44) & 3;
 }
 
 static uint64_t get_psw_mask(CPUS390XState *env)
 {
-    uint64_t r = env->psw.mask;
+    uint64_t r;
 
     env->cc_op = calc_cc(env, env->cc_op, env->cc_src, env->cc_dst, env->cc_vr);
 
-    r &= ~(3ULL << 13);
+    r = env->psw.mask;
+    r &= ~PSW_MASK_CC;
     assert(!(env->cc_op & ~3));
-    r |= env->cc_op << 13;
+    r |= (uint64_t)env->cc_op << 44;
 
     return r;
 }
@@ -479,9 +480,9 @@ static void do_svc_interrupt(CPUS390XState *env)
     lowcore = cpu_physical_memory_map(env->psa, &len, 1);
 
     lowcore->svc_code = cpu_to_be16(env->int_svc_code);
-    lowcore->svc_ilc = cpu_to_be16(env->int_svc_ilc);
+    lowcore->svc_ilen = cpu_to_be16(env->int_svc_ilen);
     lowcore->svc_old_psw.mask = cpu_to_be64(get_psw_mask(env));
-    lowcore->svc_old_psw.addr = cpu_to_be64(env->psw.addr + (env->int_svc_ilc));
+    lowcore->svc_old_psw.addr = cpu_to_be64(env->psw.addr + env->int_svc_ilen);
     mask = be64_to_cpu(lowcore->svc_new_psw.mask);
     addr = be64_to_cpu(lowcore->svc_new_psw.addr);
 
@@ -495,28 +496,26 @@ static void do_program_interrupt(CPUS390XState *env)
     uint64_t mask, addr;
     LowCore *lowcore;
     hwaddr len = TARGET_PAGE_SIZE;
-    int ilc = env->int_pgm_ilc;
+    int ilen = env->int_pgm_ilen;
 
-    switch (ilc) {
-    case ILC_LATER:
-        ilc = get_ilc(cpu_ldub_code(env, env->psw.addr));
-        break;
-    case ILC_LATER_INC:
-        ilc = get_ilc(cpu_ldub_code(env, env->psw.addr));
-        env->psw.addr += ilc * 2;
+    switch (ilen) {
+    case ILEN_LATER:
+        ilen = get_ilen(cpu_ldub_code(env, env->psw.addr));
         break;
-    case ILC_LATER_INC_2:
-        ilc = get_ilc(cpu_ldub_code(env, env->psw.addr)) * 2;
-        env->psw.addr += ilc;
+    case ILEN_LATER_INC:
+        ilen = get_ilen(cpu_ldub_code(env, env->psw.addr));
+        env->psw.addr += ilen;
         break;
+    default:
+        assert(ilen == 2 || ilen == 4 || ilen == 6);
     }
 
-    qemu_log_mask(CPU_LOG_INT, "%s: code=0x%x ilc=%d\n",
-                  __func__, env->int_pgm_code, ilc);
+    qemu_log_mask(CPU_LOG_INT, "%s: code=0x%x ilen=%d\n",
+                  __func__, env->int_pgm_code, ilen);
 
     lowcore = cpu_physical_memory_map(env->psa, &len, 1);
 
-    lowcore->pgm_ilc = cpu_to_be16(ilc);
+    lowcore->pgm_ilen = cpu_to_be16(ilen);
     lowcore->pgm_code = cpu_to_be16(env->int_pgm_code);
     lowcore->program_old_psw.mask = cpu_to_be64(get_psw_mask(env));
     lowcore->program_old_psw.addr = cpu_to_be64(env->psw.addr);
@@ -526,7 +525,7 @@ static void do_program_interrupt(CPUS390XState *env)
     cpu_physical_memory_unmap(lowcore, len, 1, len);
 
     DPRINTF("%s: %x %x %" PRIx64 " %" PRIx64 "\n", __func__,
-            env->int_pgm_code, ilc, env->psw.mask,
+            env->int_pgm_code, ilen, env->psw.mask,
             env->psw.addr);
 
     load_psw(env, mask, addr);
diff --git a/target-s390x/helper.h b/target-s390x/helper.h
index c4926c52ad..dd90d93bee 100644
--- a/target-s390x/helper.h
+++ b/target-s390x/helper.h
@@ -1,152 +1,120 @@
 #include "exec/def-helper.h"
 
-DEF_HELPER_2(exception, void, env, i32)
-DEF_HELPER_4(nc, i32, env, i32, i64, i64)
-DEF_HELPER_4(oc, i32, env, i32, i64, i64)
-DEF_HELPER_4(xc, i32, env, i32, i64, i64)
-DEF_HELPER_4(mvc, void, env, i32, i64, i64)
-DEF_HELPER_4(clc, i32, env, i32, i64, i64)
+DEF_HELPER_2(exception, noreturn, env, i32)
+DEF_HELPER_FLAGS_4(nc, TCG_CALL_NO_WG, i32, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(oc, TCG_CALL_NO_WG, i32, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(xc, TCG_CALL_NO_WG, i32, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(mvc, TCG_CALL_NO_WG, void, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(clc, TCG_CALL_NO_WG, i32, env, i32, i64, i64)
 DEF_HELPER_3(mvcl, i32, env, i32, i32)
-DEF_HELPER_FLAGS_1(set_cc_comp_s32, TCG_CALL_NO_RWG_SE, i32, s32)
-DEF_HELPER_FLAGS_1(set_cc_comp_s64, TCG_CALL_NO_RWG_SE, i32, s64)
-DEF_HELPER_FLAGS_2(set_cc_icm, TCG_CALL_NO_RWG_SE, i32, i32, i32)
-DEF_HELPER_4(clm, i32, env, i32, i32, i64)
-DEF_HELPER_4(stcm, void, env, i32, i32, i64)
-DEF_HELPER_3(mlg, void, env, i32, i64)
-DEF_HELPER_3(dlg, void, env, i32, i64)
-DEF_HELPER_FLAGS_3(set_cc_add64, TCG_CALL_NO_RWG_SE, i32, s64, s64, s64)
-DEF_HELPER_FLAGS_3(set_cc_addu64, TCG_CALL_NO_RWG_SE, i32, i64, i64, i64)
-DEF_HELPER_FLAGS_3(set_cc_add32, TCG_CALL_NO_RWG_SE, i32, s32, s32, s32)
-DEF_HELPER_FLAGS_3(set_cc_addu32, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32)
-DEF_HELPER_FLAGS_3(set_cc_sub64, TCG_CALL_NO_RWG_SE, i32, s64, s64, s64)
-DEF_HELPER_FLAGS_3(set_cc_subu64, TCG_CALL_NO_RWG_SE, i32, i64, i64, i64)
-DEF_HELPER_FLAGS_3(set_cc_sub32, TCG_CALL_NO_RWG_SE, i32, s32, s32, s32)
-DEF_HELPER_FLAGS_3(set_cc_subu32, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32)
-DEF_HELPER_4(srst, i32, env, i32, i32, i32)
-DEF_HELPER_4(clst, i32, env, i32, i32, i32)
+DEF_HELPER_FLAGS_4(clm, TCG_CALL_NO_WG, i32, env, i32, i32, i64)
+DEF_HELPER_FLAGS_3(mul128, TCG_CALL_NO_RWG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(divs32, TCG_CALL_NO_WG, s64, env, s64, s64)
+DEF_HELPER_FLAGS_3(divu32, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(divs64, TCG_CALL_NO_WG, s64, env, s64, s64)
+DEF_HELPER_FLAGS_4(divu64, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_4(srst, i64, env, i64, i64, i64)
+DEF_HELPER_4(clst, i64, env, i64, i64, i64)
 DEF_HELPER_4(mvpg, void, env, i64, i64, i64)
-DEF_HELPER_4(mvst, void, env, i32, i32, i32)
-DEF_HELPER_4(csg, i32, env, i32, i64, i32)
-DEF_HELPER_4(cdsg, i32, env, i32, i64, i32)
-DEF_HELPER_4(cs, i32, env, i32, i64, i32)
+DEF_HELPER_4(mvst, i64, env, i64, i64, i64)
 DEF_HELPER_5(ex, i32, env, i32, i64, i64, i64)
 DEF_HELPER_FLAGS_1(abs_i32, TCG_CALL_NO_RWG_SE, i32, s32)
 DEF_HELPER_FLAGS_1(nabs_i32, TCG_CALL_NO_RWG_SE, s32, s32)
 DEF_HELPER_FLAGS_1(abs_i64, TCG_CALL_NO_RWG_SE, i64, s64)
 DEF_HELPER_FLAGS_1(nabs_i64, TCG_CALL_NO_RWG_SE, s64, s64)
-DEF_HELPER_4(stcmh, void, env, i32, i64, i32)
-DEF_HELPER_4(icmh, i32, env, i32, i64, i32)
-DEF_HELPER_3(ipm, void, env, i32, i32)
-DEF_HELPER_FLAGS_3(addc_u32, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32)
-DEF_HELPER_FLAGS_3(set_cc_addc_u64, TCG_CALL_NO_RWG_SE, i32, i64, i64, i64)
-DEF_HELPER_4(stam, void, env, i32, i64, i32)
-DEF_HELPER_4(lam, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_4(stam, TCG_CALL_NO_WG, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_4(lam, TCG_CALL_NO_WG, void, env, i32, i64, i32)
 DEF_HELPER_4(mvcle, i32, env, i32, i64, i32)
 DEF_HELPER_4(clcle, i32, env, i32, i64, i32)
-DEF_HELPER_4(slb, i32, env, i32, i32, i32)
-DEF_HELPER_5(slbg, i32, env, i32, i32, i64, i64)
-DEF_HELPER_3(cefbr, void, env, i32, s32)
-DEF_HELPER_3(cdfbr, void, env, i32, s32)
-DEF_HELPER_3(cxfbr, void, env, i32, s32)
-DEF_HELPER_3(cegbr, void, env, i32, s64)
-DEF_HELPER_3(cdgbr, void, env, i32, s64)
-DEF_HELPER_3(cxgbr, void, env, i32, s64)
-DEF_HELPER_3(adbr, i32, env, i32, i32)
-DEF_HELPER_3(aebr, i32, env, i32, i32)
-DEF_HELPER_3(sebr, i32, env, i32, i32)
-DEF_HELPER_3(sdbr, i32, env, i32, i32)
-DEF_HELPER_3(debr, void, env, i32, i32)
-DEF_HELPER_3(dxbr, void, env, i32, i32)
-DEF_HELPER_3(mdbr, void, env, i32, i32)
-DEF_HELPER_3(mxbr, void, env, i32, i32)
-DEF_HELPER_3(ldebr, void, env, i32, i32)
-DEF_HELPER_3(ldxbr, void, env, i32, i32)
-DEF_HELPER_3(lxdbr, void, env, i32, i32)
-DEF_HELPER_3(ledbr, void, env, i32, i32)
-DEF_HELPER_3(lexbr, void, env, i32, i32)
-DEF_HELPER_3(lpebr, i32, env, i32, i32)
-DEF_HELPER_3(lpdbr, i32, env, i32, i32)
-DEF_HELPER_3(lpxbr, i32, env, i32, i32)
-DEF_HELPER_3(ltebr, i32, env, i32, i32)
-DEF_HELPER_3(ltdbr, i32, env, i32, i32)
-DEF_HELPER_3(ltxbr, i32, env, i32, i32)
-DEF_HELPER_3(lcebr, i32, env, i32, i32)
-DEF_HELPER_3(lcdbr, i32, env, i32, i32)
-DEF_HELPER_3(lcxbr, i32, env, i32, i32)
-DEF_HELPER_3(aeb, void, env, i32, i32)
-DEF_HELPER_3(deb, void, env, i32, i32)
-DEF_HELPER_3(meeb, void, env, i32, i32)
-DEF_HELPER_3(cdb, i32, env, i32, i64)
-DEF_HELPER_3(adb, i32, env, i32, i64)
-DEF_HELPER_3(seb, void, env, i32, i32)
-DEF_HELPER_3(sdb, i32, env, i32, i64)
-DEF_HELPER_3(mdb, void, env, i32, i64)
-DEF_HELPER_3(ddb, void, env, i32, i64)
-DEF_HELPER_FLAGS_3(cebr, TCG_CALL_NO_SE, i32, env, i32, i32)
-DEF_HELPER_FLAGS_3(cdbr, TCG_CALL_NO_SE, i32, env, i32, i32)
-DEF_HELPER_FLAGS_3(cxbr, TCG_CALL_NO_SE, i32, env, i32, i32)
-DEF_HELPER_4(cgebr, i32, env, i32, i32, i32)
-DEF_HELPER_4(cgdbr, i32, env, i32, i32, i32)
-DEF_HELPER_4(cgxbr, i32, env, i32, i32, i32)
-DEF_HELPER_2(lzer, void, env, i32)
-DEF_HELPER_2(lzdr, void, env, i32)
-DEF_HELPER_2(lzxr, void, env, i32)
-DEF_HELPER_4(cfebr, i32, env, i32, i32, i32)
-DEF_HELPER_4(cfdbr, i32, env, i32, i32, i32)
-DEF_HELPER_4(cfxbr, i32, env, i32, i32, i32)
-DEF_HELPER_3(axbr, i32, env, i32, i32)
-DEF_HELPER_3(sxbr, i32, env, i32, i32)
-DEF_HELPER_3(meebr, void, env, i32, i32)
-DEF_HELPER_3(ddbr, void, env, i32, i32)
-DEF_HELPER_4(madb, void, env, i32, i64, i32)
-DEF_HELPER_4(maebr, void, env, i32, i32, i32)
-DEF_HELPER_4(madbr, void, env, i32, i32, i32)
-DEF_HELPER_4(msdbr, void, env, i32, i32, i32)
-DEF_HELPER_3(ldeb, void, env, i32, i64)
-DEF_HELPER_3(lxdb, void, env, i32, i64)
-DEF_HELPER_FLAGS_3(tceb, TCG_CALL_NO_SE, i32, env, i32, i64)
-DEF_HELPER_FLAGS_3(tcdb, TCG_CALL_NO_SE, i32, env, i32, i64)
-DEF_HELPER_FLAGS_3(tcxb, TCG_CALL_NO_SE, i32, env, i32, i64)
-DEF_HELPER_3(flogr, i32, env, i32, i64)
-DEF_HELPER_3(sqdbr, void, env, i32, i32)
+DEF_HELPER_3(cegb, i64, env, s64, i32)
+DEF_HELPER_3(cdgb, i64, env, s64, i32)
+DEF_HELPER_3(cxgb, i64, env, s64, i32)
+DEF_HELPER_3(celgb, i64, env, i64, i32)
+DEF_HELPER_3(cdlgb, i64, env, i64, i32)
+DEF_HELPER_3(cxlgb, i64, env, i64, i32)
+DEF_HELPER_FLAGS_3(aeb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(adb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_5(axb, TCG_CALL_NO_WG, i64, env, i64, i64, i64, i64)
+DEF_HELPER_FLAGS_3(seb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(sdb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_5(sxb, TCG_CALL_NO_WG, i64, env, i64, i64, i64, i64)
+DEF_HELPER_FLAGS_3(deb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(ddb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_5(dxb, TCG_CALL_NO_WG, i64, env, i64, i64, i64, i64)
+DEF_HELPER_FLAGS_3(meeb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(mdeb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(mdb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_5(mxb, TCG_CALL_NO_WG, i64, env, i64, i64, i64, i64)
+DEF_HELPER_FLAGS_4(mxdb, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_2(ldeb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_3(ldxb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_2(lxdb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_2(lxeb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_2(ledb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_3(lexb, TCG_CALL_NO_WG, i64, env, i64, i64)
+DEF_HELPER_FLAGS_3(ceb, TCG_CALL_NO_WG_SE, i32, env, i64, i64)
+DEF_HELPER_FLAGS_3(cdb, TCG_CALL_NO_WG_SE, i32, env, i64, i64)
+DEF_HELPER_FLAGS_5(cxb, TCG_CALL_NO_WG_SE, i32, env, i64, i64, i64, i64)
+DEF_HELPER_FLAGS_3(cgeb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_3(cgdb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_4(cgxb, TCG_CALL_NO_WG, i64, env, i64, i64, i32)
+DEF_HELPER_FLAGS_3(cfeb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_3(cfdb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_4(cfxb, TCG_CALL_NO_WG, i64, env, i64, i64, i32)
+DEF_HELPER_FLAGS_3(clgeb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_3(clgdb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_4(clgxb, TCG_CALL_NO_WG, i64, env, i64, i64, i32)
+DEF_HELPER_FLAGS_3(clfeb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_3(clfdb, TCG_CALL_NO_WG, i64, env, i64, i32)
+DEF_HELPER_FLAGS_4(clfxb, TCG_CALL_NO_WG, i64, env, i64, i64, i32)
+DEF_HELPER_FLAGS_4(maeb, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(madb, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(mseb, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(msdb, TCG_CALL_NO_WG, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_2(tceb, TCG_CALL_NO_RWG_SE, i32, i64, i64)
+DEF_HELPER_FLAGS_2(tcdb, TCG_CALL_NO_RWG_SE, i32, i64, i64)
+DEF_HELPER_FLAGS_3(tcxb, TCG_CALL_NO_RWG_SE, i32, i64, i64, i64)
+DEF_HELPER_FLAGS_1(clz, TCG_CALL_NO_RWG_SE, i64, i64)
+DEF_HELPER_FLAGS_2(sqeb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_2(sqdb, TCG_CALL_NO_WG, i64, env, i64)
+DEF_HELPER_FLAGS_3(sqxb, TCG_CALL_NO_WG, i64, env, i64, i64)
 DEF_HELPER_FLAGS_1(cvd, TCG_CALL_NO_RWG_SE, i64, s32)
-DEF_HELPER_4(unpk, void, env, i32, i64, i64)
-DEF_HELPER_4(tr, void, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(unpk, TCG_CALL_NO_WG, void, env, i32, i64, i64)
+DEF_HELPER_FLAGS_4(tr, TCG_CALL_NO_WG, void, env, i32, i64, i64)
+DEF_HELPER_4(cksm, i64, env, i64, i64, i64)
+DEF_HELPER_FLAGS_5(calc_cc, TCG_CALL_NO_RWG_SE, i32, env, i32, i64, i64, i64)
+DEF_HELPER_FLAGS_2(sfpc, TCG_CALL_NO_RWG, void, env, i64)
+DEF_HELPER_FLAGS_2(sfas, TCG_CALL_NO_WG, void, env, i64)
+DEF_HELPER_FLAGS_1(popcnt, TCG_CALL_NO_RWG_SE, i64, i64)
 
-DEF_HELPER_3(servc, i32, env, i32, i64)
+#ifndef CONFIG_USER_ONLY
+DEF_HELPER_3(servc, i32, env, i64, i64)
 DEF_HELPER_4(diag, i64, env, i32, i64, i64)
-DEF_HELPER_3(load_psw, void, env, i64, i64)
-DEF_HELPER_1(program_interrupt, void, i32)
-DEF_HELPER_FLAGS_2(stidp, TCG_CALL_NO_RWG, void, env, i64)
+DEF_HELPER_3(load_psw, noreturn, env, i64, i64)
 DEF_HELPER_FLAGS_2(spx, TCG_CALL_NO_RWG, void, env, i64)
-DEF_HELPER_FLAGS_1(sck, TCG_CALL_NO_RWG, i32, i64)
-DEF_HELPER_2(stck, i32, env, i64)
-DEF_HELPER_2(stcke, i32, env, i64)
+DEF_HELPER_FLAGS_1(stck, TCG_CALL_NO_RWG_SE, i64, env)
 DEF_HELPER_FLAGS_2(sckc, TCG_CALL_NO_RWG, void, env, i64)
-DEF_HELPER_FLAGS_2(stckc, TCG_CALL_NO_RWG, void, env, i64)
+DEF_HELPER_FLAGS_1(stckc, TCG_CALL_NO_RWG, i64, env)
 DEF_HELPER_FLAGS_2(spt, TCG_CALL_NO_RWG, void, env, i64)
-DEF_HELPER_FLAGS_2(stpt, TCG_CALL_NO_RWG, void, env, i64)
-DEF_HELPER_4(stsi, i32, env, i64, i32, i32)
-DEF_HELPER_4(lctl, void, env, i32, i64, i32)
-DEF_HELPER_4(lctlg, void, env, i32, i64, i32)
-DEF_HELPER_4(stctl, void, env, i32, i64, i32)
-DEF_HELPER_4(stctg, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_1(stpt, TCG_CALL_NO_RWG, i64, env)
+DEF_HELPER_4(stsi, i32, env, i64, i64, i64)
+DEF_HELPER_FLAGS_4(lctl, TCG_CALL_NO_WG, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_4(lctlg, TCG_CALL_NO_WG, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_4(stctl, TCG_CALL_NO_WG, void, env, i32, i64, i32)
+DEF_HELPER_FLAGS_4(stctg, TCG_CALL_NO_WG, void, env, i32, i64, i32)
 DEF_HELPER_FLAGS_2(tprot, TCG_CALL_NO_RWG, i32, i64, i64)
 DEF_HELPER_FLAGS_2(iske, TCG_CALL_NO_RWG_SE, i64, env, i64)
-DEF_HELPER_FLAGS_3(sske, TCG_CALL_NO_RWG, void, env, i32, i64)
-DEF_HELPER_FLAGS_3(rrbe, TCG_CALL_NO_RWG, i32, env, i32, i64)
-DEF_HELPER_3(csp, i32, env, i32, i32)
+DEF_HELPER_FLAGS_3(sske, TCG_CALL_NO_RWG, void, env, i64, i64)
+DEF_HELPER_FLAGS_2(rrbe, TCG_CALL_NO_RWG, i32, env, i64)
+DEF_HELPER_3(csp, i32, env, i32, i64)
 DEF_HELPER_4(mvcs, i32, env, i64, i64, i64)
 DEF_HELPER_4(mvcp, i32, env, i64, i64, i64)
 DEF_HELPER_4(sigp, i32, env, i64, i32, i64)
-DEF_HELPER_2(sacf, void, env, i64)
+DEF_HELPER_FLAGS_2(sacf, TCG_CALL_NO_WG, void, env, i64)
 DEF_HELPER_FLAGS_3(ipte, TCG_CALL_NO_RWG, void, env, i64, i64)
 DEF_HELPER_FLAGS_1(ptlb, TCG_CALL_NO_RWG, void, env)
-DEF_HELPER_3(lra, i32, env, i64, i32)
-DEF_HELPER_3(stura, void, env, i64, i32)
-DEF_HELPER_3(cksm, void, env, i32, i32)
-
-DEF_HELPER_FLAGS_5(calc_cc, TCG_CALL_NO_RWG_SE,
-                   i32, env, i32, i64, i64, i64)
+DEF_HELPER_2(lra, i64, env, i64)
+DEF_HELPER_FLAGS_3(stura, TCG_CALL_NO_WG, void, env, i64, i64)
+#endif
 
 #include "exec/def-helper.h"
diff --git a/target-s390x/insn-data.def b/target-s390x/insn-data.def
new file mode 100644
index 0000000000..b42ebb6a1a
--- /dev/null
+++ b/target-s390x/insn-data.def
@@ -0,0 +1,813 @@
+/* ADD */
+    C(0x1a00, AR,      RR_a,  Z,   r1, r2, new, r1_32, add, adds32)
+    C(0xb9f8, ARK,     RRF_a, DO,  r2, r3, new, r1_32, add, adds32)
+    C(0x5a00, A,       RX_a,  Z,   r1, m2_32s, new, r1_32, add, adds32)
+    C(0xe35a, AY,      RXY_a, LD,  r1, m2_32s, new, r1_32, add, adds32)
+    C(0xb908, AGR,     RRE,   Z,   r1, r2, r1, 0, add, adds64)
+    C(0xb918, AGFR,    RRE,   Z,   r1, r2_32s, r1, 0, add, adds64)
+    C(0xb9e8, AGRK,    RRF_a, DO,  r2, r3, r1, 0, add, adds64)
+    C(0xe308, AG,      RXY_a, Z,   r1, m2_64, r1, 0, add, adds64)
+    C(0xe318, AGF,     RXY_a, Z,   r1, m2_32s, r1, 0, add, adds64)
+    C(0xb30a, AEBR,    RRE,   Z,   e1, e2, new, e1, aeb, f32)
+    C(0xb31a, ADBR,    RRE,   Z,   f1_o, f2_o, f1, 0, adb, f64)
+    C(0xb34a, AXBR,    RRE,   Z,   0, x2_o, x1, 0, axb, f128)
+    C(0xed0a, AEB,     RXE,   Z,   e1, m2_32u, new, e1, aeb, f32)
+    C(0xed1a, ADB,     RXE,   Z,   f1_o, m2_64, f1, 0, adb, f64)
+/* ADD IMMEDIATE */
+    C(0xc209, AFI,     RIL_a, EI,  r1, i2, new, r1_32, add, adds32)
+    C(0xeb6a, ASI,     SIY,   GIE, m1_32s, i2, new, m1_32, add, adds32)
+    C(0xecd8, AHIK,    RIE_d, DO,  r3, i2, new, r1_32, add, adds32)
+    C(0xc208, AGFI,    RIL_a, EI,  r1, i2, r1, 0, add, adds64)
+    C(0xeb7a, AGSI,    SIY,   GIE, m1_64, i2, new, m1_64, add, adds64)
+    C(0xecd9, AGHIK,   RIE_d, DO,  r3, i2, r1, 0, add, adds64)
+/* ADD HALFWORD */
+    C(0x4a00, AH,      RX_a,  Z,   r1, m2_16s, new, r1_32, add, adds32)
+    C(0xe37a, AHY,     RXY_a, LD,  r1, m2_16s, new, r1_32, add, adds32)
+/* ADD HALFWORD IMMEDIATE */
+    C(0xa70a, AHI,     RI_a,  Z,   r1, i2, new, r1_32, add, adds32)
+    C(0xa70b, AGHI,    RI_a,  Z,   r1, i2, r1, 0, add, adds64)
+
+/* ADD LOGICAL */
+    C(0x1e00, ALR,     RR_a,  Z,   r1, r2, new, r1_32, add, addu32)
+    C(0xb9fa, ALRK,    RRF_a, DO,  r2, r3, new, r1_32, add, addu32)
+    C(0x5e00, AL,      RX_a,  Z,   r1, m2_32u, new, r1_32, add, addu32)
+    C(0xe35e, ALY,     RXY_a, LD,  r1, m2_32u, new, r1_32, add, addu32)
+    C(0xb90a, ALGR,    RRE,   Z,   r1, r2, r1, 0, add, addu64)
+    C(0xb91a, ALGFR,   RRE,   Z,   r1, r2_32u, r1, 0, add, addu64)
+    C(0xb9ea, ALGRK,   RRF_a, DO,  r2, r3, r1, 0, add, addu64)
+    C(0xe30a, ALG,     RXY_a, Z,   r1, m2_64, r1, 0, add, addu64)
+    C(0xe31a, ALGF,    RXY_a, Z,   r1, m2_32u, r1, 0, add, addu64)
+/* ADD LOGICAL IMMEDIATE */
+    C(0xc20b, ALFI,    RIL_a, EI,  r1, i2_32u, new, r1_32, add, addu32)
+    C(0xc20a, ALGFI,   RIL_a, EI,  r1, i2_32u, r1, 0, add, addu64)
+/* ADD LOGICAL WITH SIGNED IMMEDIATE */
+    C(0xeb6e, ALSI,    SIY,   GIE, m1_32u, i2, new, m1_32, add, addu32)
+    C(0xecda, ALHSIK,  RIE_d, DO,  r3, i2, new, r1_32, add, addu32)
+    C(0xeb7e, ALGSI,   SIY,   GIE, m1_64, i2, new, m1_64, add, addu64)
+    C(0xecdb, ALGHSIK, RIE_d, DO,  r3, i2, r1, 0, add, addu64)
+/* ADD LOGICAL WITH CARRY */
+    C(0xb998, ALCR,    RRE,   Z,   r1, r2, new, r1_32, addc, addc32)
+    C(0xb988, ALCGR,   RRE,   Z,   r1, r2, r1, 0, addc, addc64)
+    C(0xe398, ALC,     RXY_a, Z,   r1, m2_32u, new, r1_32, addc, addc32)
+    C(0xe388, ALCG,    RXY_a, Z,   r1, m2_64, r1, 0, addc, addc64)
+
+/* AND */
+    C(0x1400, NR,      RR_a,  Z,   r1, r2, new, r1_32, and, nz32)
+    C(0xb9f4, NRK,     RRF_a, DO,  r2, r3, new, r1_32, and, nz32)
+    C(0x5400, N,       RX_a,  Z,   r1, m2_32s, new, r1_32, and, nz32)
+    C(0xe354, NY,      RXY_a, LD,  r1, m2_32s, new, r1_32, and, nz32)
+    C(0xb980, NGR,     RRE,   Z,   r1, r2, r1, 0, and, nz64)
+    C(0xb9e4, NGRK,    RRF_a, DO,  r2, r3, r1, 0, and, nz64)
+    C(0xe380, NG,      RXY_a, Z,   r1, m2_64, r1, 0, and, nz64)
+    C(0xd400, NC,      SS_a,  Z,   la1, a2, 0, 0, nc, 0)
+/* AND IMMEDIATE */
+    D(0xc00a, NIHF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, andi, 0, 0x2020)
+    D(0xc00b, NILF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, andi, 0, 0x2000)
+    D(0xa504, NIHH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, andi, 0, 0x1030)
+    D(0xa505, NIHL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, andi, 0, 0x1020)
+    D(0xa506, NILH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, andi, 0, 0x1010)
+    D(0xa507, NILL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, andi, 0, 0x1000)
+    C(0x9400, NI,      SI,    Z,   m1_8u, i2_8u, new, m1_8, and, nz64)
+    C(0xeb54, NIY,     SIY,   LD,  m1_8u, i2_8u, new, m1_8, and, nz64)
+
+/* BRANCH AND SAVE */
+    C(0x0d00, BASR,    RR_a,  Z,   0, r2_nz, r1, 0, bas, 0)
+    C(0x4d00, BAS,     RX_a,  Z,   0, a2, r1, 0, bas, 0)
+/* BRANCH RELATIVE AND SAVE */
+    C(0xa705, BRAS,    RI_b,  Z,   0, 0, r1, 0, basi, 0)
+    C(0xc005, BRASL,   RIL_b, Z,   0, 0, r1, 0, basi, 0)
+/* BRANCH ON CONDITION */
+    C(0x0700, BCR,     RR_b,  Z,   0, r2_nz, 0, 0, bc, 0)
+    C(0x4700, BC,      RX_b,  Z,   0, a2, 0, 0, bc, 0)
+/* BRANCH RELATIVE ON CONDITION */
+    C(0xa704, BRC,     RI_c,  Z,   0, 0, 0, 0, bc, 0)
+    C(0xc004, BRCL,    RIL_c, Z,   0, 0, 0, 0, bc, 0)
+/* BRANCH ON COUNT */
+    C(0x0600, BCTR,    RR_a,  Z,   0, r2_nz, 0, 0, bct32, 0)
+    C(0xb946, BCTGR,   RRE,   Z,   0, r2_nz, 0, 0, bct64, 0)
+    C(0x4600, BCT,     RX_a,  Z,   0, a2, 0, 0, bct32, 0)
+    C(0xe346, BCTG,    RXY_a, Z,   0, a2, 0, 0, bct64, 0)
+/* BRANCH RELATIVE ON COUNT */
+    C(0xa706, BRCT,    RI_b,  Z,   0, 0, 0, 0, bct32, 0)
+    C(0xa707, BRCTG,   RI_b,  Z,   0, 0, 0, 0, bct64, 0)
+/* BRANCH ON INDEX */
+    D(0x8600, BXH,     RS_a,  Z,   0, a2, 0, 0, bx32, 0, 0)
+    D(0x8700, BXLE,    RS_a,  Z,   0, a2, 0, 0, bx32, 0, 1)
+    D(0xeb44, BXHG,    RSY_a, Z,   0, a2, 0, 0, bx64, 0, 0)
+    D(0xeb45, BXLEG,   RSY_a, Z,   0, a2, 0, 0, bx64, 0, 1)
+/* BRANCH RELATIVE ON INDEX */
+    D(0x8400, BRXH,    RSI,   Z,   0, 0, 0, 0, bx32, 0, 0)
+    D(0x8500, BRXLE,   RSI,   Z,   0, 0, 0, 0, bx32, 0, 1)
+    D(0xec44, BRXHG,   RIE_e, Z,   0, 0, 0, 0, bx64, 0, 0)
+    D(0xec45, BRXHLE,  RIE_e, Z,   0, 0, 0, 0, bx64, 0, 1)
+
+/* CHECKSUM */
+    C(0xb241, CKSM,    RRE,   Z,   r1_o, ra2, new, r1_32, cksm, 0)
+
+/* COPY SIGN */
+    C(0xb372, CPSDR,   RRF_b, FPSSH, f3_o, f2_o, f1, 0, cps, 0)
+
+/* COMPARE */
+    C(0x1900, CR,      RR_a,  Z,   r1_o, r2_o, 0, 0, 0, cmps32)
+    C(0x5900, C,       RX_a,  Z,   r1_o, m2_32s, 0, 0, 0, cmps32)
+    C(0xe359, CY,      RXY_a, LD,  r1_o, m2_32s, 0, 0, 0, cmps32)
+    C(0xb920, CGR,     RRE,   Z,   r1_o, r2_o, 0, 0, 0, cmps64)
+    C(0xb930, CGFR,    RRE,   Z,   r1_o, r2_32s, 0, 0, 0, cmps64)
+    C(0xe320, CG,      RXY_a, Z,   r1_o, m2_64, 0, 0, 0, cmps64)
+    C(0xe330, CGF,     RXY_a, Z,   r1_o, m2_32s, 0, 0, 0, cmps64)
+    C(0xb309, CEBR,    RRE,   Z,   e1, e2, 0, 0, ceb, 0)
+    C(0xb319, CDBR,    RRE,   Z,   f1_o, f2_o, 0, 0, cdb, 0)
+    C(0xb349, CXBR,    RRE,   Z,   x1_o, x2_o, 0, 0, cxb, 0)
+    C(0xed09, CEB,     RXE,   Z,   e1, m2_32u, 0, 0, ceb, 0)
+    C(0xed19, CDB,     RXE,   Z,   f1_o, m2_64, 0, 0, cdb, 0)
+/* COMPARE IMMEDIATE */
+    C(0xc20d, CFI,     RIL_a, EI,  r1, i2, 0, 0, 0, cmps32)
+    C(0xc20c, CGFI,    RIL_a, EI,  r1, i2, 0, 0, 0, cmps64)
+/* COMPARE RELATIVE LONG */
+    C(0xc60d, CRL,     RIL_b, GIE, r1, mri2_32s, 0, 0, 0, cmps32)
+    C(0xc608, CGRL,    RIL_b, GIE, r1, mri2_64, 0, 0, 0, cmps64)
+    C(0xc60c, CGFRL,   RIL_b, GIE, r1, mri2_32s, 0, 0, 0, cmps64)
+/* COMPARE HALFWORD */
+    C(0x4900, CH,      RX_a,  Z,   r1_o, m2_16s, 0, 0, 0, cmps32)
+    C(0xe379, CHY,     RXY_a, LD,  r1_o, m2_16s, 0, 0, 0, cmps32)
+    C(0xe334, CGH,     RXY_a, GIE, r1_o, m2_16s, 0, 0, 0, cmps64)
+/* COMPARE HALFWORD IMMEDIATE */
+    C(0xa70e, CHI,     RI_a,  Z,   r1_o, i2, 0, 0, 0, cmps32)
+    C(0xa70f, CGHI,    RI_a,  Z,   r1_o, i2, 0, 0, 0, cmps64)
+    C(0xe554, CHHSI,   SIL,   GIE, m1_16s, i2, 0, 0, 0, cmps64)
+    C(0xe55c, CHSI,    SIL,   GIE, m1_32s, i2, 0, 0, 0, cmps64)
+    C(0xe558, CGHSI,   SIL,   GIE, m1_64, i2, 0, 0, 0, cmps64)
+/* COMPARE HALFWORD RELATIVE LONG */
+    C(0xc605, CHRL,    RIL_a, GIE, r1_o, mri2_32s, 0, 0, 0, cmps32)
+    C(0xc604, CGHRL,   RIL_a, GIE, r1_o, mri2_64, 0, 0, 0, cmps64)
+
+/* COMPARE LOGICAL */
+    C(0x1500, CLR,     RR_a,  Z,   r1, r2, 0, 0, 0, cmpu32)
+    C(0x5500, CL,      RX_a,  Z,   r1, m2_32s, 0, 0, 0, cmpu32)
+    C(0xe355, CLY,     RXY_a, LD,  r1, m2_32s, 0, 0, 0, cmpu32)
+    C(0xb921, CLGR,    RRE,   Z,   r1, r2, 0, 0, 0, cmpu64)
+    C(0xb931, CLGFR,   RRE,   Z,   r1, r2_32u, 0, 0, 0, cmpu64)
+    C(0xe321, CLG,     RXY_a, Z,   r1, m2_64, 0, 0, 0, cmpu64)
+    C(0xe331, CLGF,    RXY_a, Z,   r1, m2_32u, 0, 0, 0, cmpu64)
+    C(0xd500, CLC,     SS_a,  Z,   la1, a2, 0, 0, clc, 0)
+/* COMPARE LOGICAL IMMEDIATE */
+    C(0xc20f, CLFI,    RIL_a, EI,  r1, i2, 0, 0, 0, cmpu32)
+    C(0xc20e, CLGFI,   RIL_a, EI,  r1, i2_32u, 0, 0, 0, cmpu64)
+    C(0x9500, CLI,     SI,    Z,   m1_8u, i2_8u, 0, 0, 0, cmpu64)
+    C(0xeb55, CLIY,    SIY,   LD,  m1_8u, i2_8u, 0, 0, 0, cmpu64)
+    C(0xe555, CLHHSI,  SIL,   GIE, m1_16u, i2_16u, 0, 0, 0, cmpu64)
+    C(0xe55d, CLFHSI,  SIL,   GIE, m1_32u, i2_16u, 0, 0, 0, cmpu64)
+    C(0xe559, CLGHSI,  SIL,   GIE, m1_64, i2_16u, 0, 0, 0, cmpu64)
+/* COMPARE LOGICAL RELATIVE LONG */
+    C(0xc60f, CLRL,    RIL_b, GIE, r1_o, mri2_32u, 0, 0, 0, cmpu32)
+    C(0xc60a, CLGRL,   RIL_b, GIE, r1_o, mri2_64, 0, 0, 0, cmpu64)
+    C(0xc60e, CLGFRL,  RIL_b, GIE, r1_o, mri2_32u, 0, 0, 0, cmpu64)
+    C(0xc607, CLHRL,   RIL_b, GIE, r1_o, mri2_16u, 0, 0, 0, cmpu32)
+    C(0xc606, CLGHRL,  RIL_b, GIE, r1_o, mri2_16u, 0, 0, 0, cmpu64)
+/* COMPARE LOGICAL LONG EXTENDED */
+    C(0xa900, CLCLE,   RS_a,  Z,   0, a2, 0, 0, clcle, 0)
+/* COMPARE LOGICAL CHARACTERS UNDER MASK */
+    C(0xbd00, CLM,     RS_b,  Z,   r1_o, a2, 0, 0, clm, 0)
+    C(0xeb21, CLMY,    RSY_b, LD,  r1_o, a2, 0, 0, clm, 0)
+    C(0xeb20, CLMH,    RSY_b, Z,   r1_sr32, a2, 0, 0, clm, 0)
+/* COMPARE LOGICAL STRING */
+    C(0xb25d, CLST,    RRE,   Z,   r1_o, r2_o, 0, 0, clst, 0)
+
+/* COMPARE AND BRANCH */
+    D(0xecf6, CRB,     RRS,   GIE, r1_32s, r2_32s, 0, 0, cj, 0, 0)
+    D(0xece4, CGRB,    RRS,   GIE, r1_o, r2_o, 0, 0, cj, 0, 0)
+    D(0xec76, CRJ,     RIE_b, GIE, r1_32s, r2_32s, 0, 0, cj, 0, 0)
+    D(0xec64, CGRJ,    RIE_b, GIE, r1_o, r2_o, 0, 0, cj, 0, 0)
+    D(0xecfe, CIB,     RIS,   GIE, r1_32s, i2, 0, 0, cj, 0, 0)
+    D(0xecfc, CGIB,    RIS,   GIE, r1_o, i2, 0, 0, cj, 0, 0)
+    D(0xec7e, CIJ,     RIE_c, GIE, r1_32s, i2, 0, 0, cj, 0, 0)
+    D(0xec7c, CGIJ,    RIE_c, GIE, r1_o, i2, 0, 0, cj, 0, 0)
+/* COMPARE LOGICAL AND BRANCH */
+    D(0xecf7, CLRB,    RRS,   GIE, r1_32u, r2_32u, 0, 0, cj, 0, 1)
+    D(0xece5, CLGRB,   RRS,   GIE, r1_o, r2_o, 0, 0, cj, 0, 1)
+    D(0xec77, CLRJ,    RIE_b, GIE, r1_32u, r2_32u, 0, 0, cj, 0, 1)
+    D(0xec65, CLGRJ,   RIE_b, GIE, r1_o, r2_o, 0, 0, cj, 0, 1)
+    D(0xecff, CLIB,    RIS,   GIE, r1_32u, i2_8u, 0, 0, cj, 0, 1)
+    D(0xecfd, CLGIB,   RIS,   GIE, r1_o, i2_8u, 0, 0, cj, 0, 1)
+    D(0xec7f, CLIJ,    RIE_c, GIE, r1_32u, i2_8u, 0, 0, cj, 0, 1)
+    D(0xec7d, CLGIJ,   RIE_c, GIE, r1_o, i2_8u, 0, 0, cj, 0, 1)
+
+/* COMPARE AND SWAP */
+    D(0xba00, CS,      RS_a,  Z,   r3_32u, r1_32u, new, r1_32, cs, 0, 0)
+    D(0xeb14, CSY,     RSY_a, LD,  r3_32u, r1_32u, new, r1_32, cs, 0, 0)
+    D(0xeb30, CSG,     RSY_a, Z,   r3_o, r1_o, new, r1, cs, 0, 1)
+/* COMPARE DOUBLE AND SWAP */
+    D(0xbb00, CDS,     RS_a,  Z,   r3_D32, r1_D32, new, r1_D32, cs, 0, 1)
+    D(0xeb31, CDSY,    RSY_a, LD,  r3_D32, r1_D32, new, r1_D32, cs, 0, 1)
+    C(0xeb3e, CDSG,    RSY_a, Z,   0, 0, 0, 0, cdsg, 0)
+
+/* COMPARE AND TRAP */
+    D(0xb972, CRT,     RRF_c, GIE, r1_32s, r2_32s, 0, 0, ct, 0, 0)
+    D(0xb960, CGRT,    RRF_c, GIE, r1_o, r2_o, 0, 0, ct, 0, 0)
+    D(0xec72, CIT,     RIE_a, GIE, r1_32s, i2, 0, 0, ct, 0, 0)
+    D(0xec70, CGIT,    RIE_a, GIE, r1_o, i2, 0, 0, ct, 0, 0)
+/* COMPARE LOGICAL AND TRAP */
+    D(0xb973, CLRT,    RRF_c, GIE, r1_32u, r2_32u, 0, 0, ct, 0, 1)
+    D(0xb961, CLGRT,   RRF_c, GIE, r1_o, r2_o, 0, 0, ct, 0, 1)
+    D(0xec73, CLFIT,   RIE_a, GIE, r1_32u, i2_32u, 0, 0, ct, 0, 1)
+    D(0xec71, CLGIT,   RIE_a, GIE, r1_o, i2_32u, 0, 0, ct, 0, 0)
+
+/* CONVERT TO DECIMAL */
+    C(0x4e00, CVD,     RX_a,  Z,   r1_o, a2, 0, 0, cvd, 0)
+    C(0xe326, CVDY,    RXY_a, LD,  r1_o, a2, 0, 0, cvd, 0)
+/* CONVERT TO FIXED */
+    C(0xb398, CFEBR,   RRF_e, Z,   0, e2, new, r1_32, cfeb, 0)
+    C(0xb399, CFDBR,   RRF_e, Z,   0, f2_o, new, r1_32, cfdb, 0)
+    C(0xb39a, CFXBR,   RRF_e, Z,   0, x2_o, new, r1_32, cfxb, 0)
+    C(0xb3a8, CGEBR,   RRF_e, Z,   0, e2, r1, 0, cgeb, 0)
+    C(0xb3a9, CGDBR,   RRF_e, Z,   0, f2_o, r1, 0, cgdb, 0)
+    C(0xb3aa, CGXBR,   RRF_e, Z,   0, x2_o, r1, 0, cgxb, 0)
+/* CONVERT FROM FIXED */
+    C(0xb394, CEFBR,   RRF_e, Z,   0, r2_32s, new, e1, cegb, 0)
+    C(0xb395, CDFBR,   RRF_e, Z,   0, r2_32s, f1, 0, cdgb, 0)
+    C(0xb396, CXFBR,   RRF_e, Z,   0, r2_32s, x1, 0, cxgb, 0)
+    C(0xb3a4, CEGBR,   RRF_e, Z,   0, r2_o, new, e1, cegb, 0)
+    C(0xb3a5, CDGBR,   RRF_e, Z,   0, r2_o, f1, 0, cdgb, 0)
+    C(0xb3a6, CXGBR,   RRF_e, Z,   0, r2_o, x1, 0, cxgb, 0)
+/* CONVERT TO LOGICAL */
+    C(0xb39c, CLFEBR,  RRF_e, FPE, 0, e2, new, r1_32, clfeb, 0)
+    C(0xb39d, CLFDBR,  RRF_e, FPE, 0, f2_o, new, r1_32, clfdb, 0)
+    C(0xb39e, CLFXBR,  RRF_e, FPE, 0, x2_o, new, r1_32, clfxb, 0)
+    C(0xb3ac, CLGEBR,  RRF_e, FPE, 0, e2, r1, 0, clgeb, 0)
+    C(0xb3ad, CLGDBR,  RRF_e, FPE, 0, f2_o, r1, 0, clgdb, 0)
+    C(0xb3ae, CLGXBR,  RRF_e, FPE, 0, x2_o, r1, 0, clgxb, 0)
+/* CONVERT FROM LOGICAL */
+    C(0xb390, CELFBR,  RRF_e, FPE, 0, r2_32u, new, e1, celgb, 0)
+    C(0xb391, CDLFBR,  RRF_e, FPE, 0, r2_32u, f1, 0, cdlgb, 0)
+    C(0xb392, CXLFBR,  RRF_e, FPE, 0, r2_32u, x1, 0, cxlgb, 0)
+    C(0xb3a0, CELGBR,  RRF_e, FPE, 0, r2_o, new, e1, celgb, 0)
+    C(0xb3a1, CDLGBR,  RRF_e, FPE, 0, r2_o, f1, 0, cdlgb, 0)
+    C(0xb3a2, CXLGBR,  RRF_e, FPE, 0, r2_o, x1, 0, cxlgb, 0)
+
+/* DIVIDE */
+    C(0x1d00, DR,      RR_a,  Z,   r1_D32, r2_32s, new_P, r1_P32, divs32, 0)
+    C(0x5d00, D,       RX_a,  Z,   r1_D32, m2_32s, new_P, r1_P32, divs32, 0)
+    C(0xb30d, DEBR,    RRE,   Z,   e1, e2, new, e1, deb, 0)
+    C(0xb31d, DDBR,    RRE,   Z,   f1_o, f2_o, f1, 0, ddb, 0)
+    C(0xb34d, DXBR,    RRE,   Z,   0, x2_o, x1, 0, dxb, 0)
+    C(0xed0d, DEB,     RXE,   Z,   e1, m2_32u, new, e1, deb, 0)
+    C(0xed1d, DDB,     RXE,   Z,   f1_o, m2_64, f1, 0, ddb, 0)
+/* DIVIDE LOGICAL */
+    C(0xb997, DLR,     RRE,   Z,   r1_D32, r2_32u, new_P, r1_P32, divu32, 0)
+    C(0xe397, DL,      RXY_a, Z,   r1_D32, m2_32u, new_P, r1_P32, divu32, 0)
+    C(0xb987, DLGR,    RRE,   Z,   0, r2_o, r1_P, 0, divu64, 0)
+    C(0xe387, DLG,     RXY_a, Z,   0, m2_64, r1_P, 0, divu64, 0)
+/* DIVIDE SINGLE */
+    C(0xb90d, DSGR,    RRE,   Z,   r1p1, r2, r1_P, 0, divs64, 0)
+    C(0xb91d, DSGFR,   RRE,   Z,   r1p1, r2_32s, r1_P, 0, divs64, 0)
+    C(0xe30d, DSG,     RXY_a, Z,   r1p1, m2_64, r1_P, 0, divs64, 0)
+    C(0xe31d, DSGF,    RXY_a, Z,   r1p1, m2_32s, r1_P, 0, divs64, 0)
+
+/* EXCLUSIVE OR */
+    C(0x1700, XR,      RR_a,  Z,   r1, r2, new, r1_32, xor, nz32)
+    C(0xb9f7, XRK,     RRF_a, DO,  r2, r3, new, r1_32, xor, nz32)
+    C(0x5700, X,       RX_a,  Z,   r1, m2_32s, new, r1_32, xor, nz32)
+    C(0xe357, XY,      RXY_a, LD,  r1, m2_32s, new, r1_32, xor, nz32)
+    C(0xb982, XGR,     RRE,   Z,   r1, r2, r1, 0, xor, nz64)
+    C(0xb9e7, XGRK,    RRF_a, DO,  r2, r3, r1, 0, xor, nz64)
+    C(0xe382, XG,      RXY_a, Z,   r1, m2_64, r1, 0, xor, nz64)
+    C(0xd700, XC,      SS_a,  Z,   0, 0, 0, 0, xc, 0)
+/* EXCLUSIVE OR IMMEDIATE */
+    D(0xc006, XIHF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, xori, 0, 0x2020)
+    D(0xc007, XILF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, xori, 0, 0x2000)
+    C(0x9700, XI,      SI,    Z,   m1_8u, i2_8u, new, m1_8, xor, nz64)
+    C(0xeb57, XIY,     SIY,   LD,  m1_8u, i2_8u, new, m1_8, xor, nz64)
+
+/* EXECUTE */
+    C(0x4400, EX,      RX_a,  Z,   r1_o, a2, 0, 0, ex, 0)
+/* EXECUTE RELATIVE LONG */
+    C(0xc600, EXRL,    RIL_b, EE,  r1_o, ri2, 0, 0, ex, 0)
+
+/* EXTRACT ACCESS */
+    C(0xb24f, EAR,     RRE,   Z,   0, 0, new, r1_32, ear, 0)
+/* EXTRACT FPC */
+    C(0xb38c, EFPC,    RRE,   Z,   0, 0, new, r1_32, efpc, 0)
+
+/* FIND LEFTMOST ONE */
+    C(0xb983, FLOGR,   RRE,   EI,  0, r2_o, r1_P, 0, flogr, 0)
+
+/* INSERT CHARACTER */
+    C(0x4300, IC,      RX_a,  Z,   0, m2_8u, 0, r1_8, mov2, 0)
+    C(0xe373, ICY,     RXY_a, LD,  0, m2_8u, 0, r1_8, mov2, 0)
+/* INSERT CHARACTERS UNDER MASK */
+    D(0xbf00, ICM,     RS_b,  Z,   0, a2, r1, 0, icm, 0, 0)
+    D(0xeb81, ICMY,    RSY_b, LD,  0, a2, r1, 0, icm, 0, 0)
+    D(0xeb80, ICMH,    RSY_b, Z,   0, a2, r1, 0, icm, 0, 32)
+/* INSERT IMMEDIATE */
+    D(0xc008, IIHF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, insi, 0, 0x2020)
+    D(0xc009, IILF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, insi, 0, 0x2000)
+    D(0xa500, IIHH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, insi, 0, 0x1030)
+    D(0xa501, IIHL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, insi, 0, 0x1020)
+    D(0xa502, IILH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, insi, 0, 0x1010)
+    D(0xa503, IILL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, insi, 0, 0x1000)
+/* INSERT PROGRAM MASK */
+    C(0xb222, IPM,     RRE,   Z,   0, 0, r1, 0, ipm, 0)
+
+/* LOAD */
+    C(0x1800, LR,      RR_a,  Z,   0, r2_o, 0, cond_r1r2_32, mov2, 0)
+    C(0x5800, L,       RX_a,  Z,   0, a2, new, r1_32, ld32s, 0)
+    C(0xe358, LY,      RXY_a, Z,   0, a2, new, r1_32, ld32s, 0)
+    C(0xb904, LGR,     RRE,   Z,   0, r2_o, 0, r1, mov2, 0)
+    C(0xb914, LGFR,    RRE,   Z,   0, r2_32s, 0, r1, mov2, 0)
+    C(0xe304, LG,      RXY_a, Z,   0, a2, r1, 0, ld64, 0)
+    C(0xe314, LGF,     RXY_a, Z,   0, a2, r1, 0, ld32s, 0)
+    C(0x2800, LDR,     RR_a,  Z,   0, f2_o, 0, f1, mov2, 0)
+    C(0x6800, LD,      RX_a,  Z,   0, m2_64, 0, f1, mov2, 0)
+    C(0xed65, LDY,     RXY_a, LD,  0, m2_64, 0, f1, mov2, 0)
+    C(0x3800, LER,     RR_a,  Z,   0, e2, 0, cond_e1e2, mov2, 0)
+    C(0x7800, LE,      RX_a,  Z,   0, m2_32u, 0, e1, mov2, 0)
+    C(0xed64, LEY,     RXY_a, LD,  0, m2_32u, 0, e1, mov2, 0)
+    C(0xb365, LXR,     RRE,   Z,   0, x2_o, 0, x1, movx, 0)
+/* LOAD IMMEDIATE */
+    C(0xc001, LGFI,    RIL_a, EI,  0, i2, 0, r1, mov2, 0)
+/* LOAD RELATIVE LONG */
+    C(0xc40d, LRL,     RIL_b, GIE, 0, ri2, new, r1_32, ld32s, 0)
+    C(0xc408, LGRL,    RIL_b, GIE, 0, ri2, r1, 0, ld64, 0)
+    C(0xc40c, LGFRL,   RIL_b, GIE, 0, ri2, r1, 0, ld32s, 0)
+/* LOAD ADDRESS */
+    C(0x4100, LA,      RX_a,  Z,   0, a2, 0, r1, mov2, 0)
+    C(0xe371, LAY,     RXY_a, LD,  0, a2, 0, r1, mov2, 0)
+/* LOAD ADDRESS RELATIVE LONG */
+    C(0xc000, LARL,    RIL_b, Z,   0, ri2, 0, r1, mov2, 0)
+/* LOAD AND TEST */
+    C(0x1200, LTR,     RR_a,  Z,   0, r2_o, 0, cond_r1r2_32, mov2, s32)
+    C(0xb902, LTGR,    RRE,   Z,   0, r2_o, 0, r1, mov2, s64)
+    C(0xb912, LTGFR,   RRE,   Z,   0, r2_32s, 0, r1, mov2, s64)
+    C(0xe312, LT,      RXY_a, EI,  0, a2, new, r1_32, ld32s, s64)
+    C(0xe302, LTG,     RXY_a, EI,  0, a2, r1, 0, ld64, s64)
+    C(0xe332, LTGF,    RXY_a, GIE, 0, a2, r1, 0, ld32s, s64)
+    C(0xb302, LTEBR,   RRE,   Z,   0, e2, 0, cond_e1e2, mov2, f32)
+    C(0xb312, LTDBR,   RRE,   Z,   0, f2_o, 0, f1, mov2, f64)
+    C(0xb342, LTXBR,   RRE,   Z,   0, x2_o, 0, x1, movx, f128)
+/* LOAD BYTE */
+    C(0xb926, LBR,     RRE,   EI,  0, r2_8s, 0, r1_32, mov2, 0)
+    C(0xb906, LGBR,    RRE,   EI,  0, r2_8s, 0, r1, mov2, 0)
+    C(0xe376, LB,      RXY_a, LD,  0, a2, new, r1_32, ld8s, 0)
+    C(0xe377, LGB,     RXY_a, LD,  0, a2, r1, 0, ld8s, 0)
+/* LOAD COMPLEMENT */
+    C(0x1300, LCR,     RR_a,  Z,   0, r2, new, r1_32, neg, neg32)
+    C(0xb903, LCGR,    RRE,   Z,   0, r2, r1, 0, neg, neg64)
+    C(0xb913, LCGFR,   RRE,   Z,   0, r2_32s, r1, 0, neg, neg64)
+    C(0xb303, LCEBR,   RRE,   Z,   0, e2, new, e1, negf32, f32)
+    C(0xb313, LCDBR,   RRE,   Z,   0, f2_o, f1, 0, negf64, f64)
+    C(0xb343, LCXBR,   RRE,   Z,   0, x2_o, x1, 0, negf128, f128)
+    C(0xb373, LCDFR,   RRE,   FPSSH, 0, f2_o, f1, 0, negf64, 0)
+/* LOAD HALFWORD */
+    C(0xb927, LHR,     RRE,   EI,  0, r2_16s, 0, r1_32, mov2, 0)
+    C(0xb907, LGHR,    RRE,   EI,  0, r2_16s, 0, r1, mov2, 0)
+    C(0x4800, LH,      RX_a,  Z,   0, a2, new, r1_32, ld16s, 0)
+    C(0xe378, LHY,     RXY_a, LD,  0, a2, new, r1_32, ld16s, 0)
+    C(0xe315, LGH,     RXY_a, Z,   0, a2, r1, 0, ld16s, 0)
+/* LOAD HALFWORD IMMEDIATE */
+    C(0xa708, LHI,     RI_a,  Z,   0, i2, 0, r1_32, mov2, 0)
+    C(0xa709, LGHI,    RI_a,  Z,   0, i2, 0, r1, mov2, 0)
+/* LOAD HALFWORD RELATIVE LONG */
+    C(0xc405, LHRL,    RIL_b, GIE, 0, ri2, new, r1_32, ld16s, 0)
+    C(0xc404, LGHRL,   RIL_b, GIE, 0, ri2, r1, 0, ld16s, 0)
+/* LOAD LOGICAL */
+    C(0xb916, LLGFR,   RRE,   Z,   0, r2_32u, 0, r1, mov2, 0)
+    C(0xe316, LLGF,    RXY_a, Z,   0, a2, r1, 0, ld32u, 0)
+/* LOAD LOGICAL RELATIVE LONG */
+    C(0xc40e, LLGFRL,  RIL_b, GIE, 0, ri2, r1, 0, ld32u, 0)
+/* LOAD LOGICAL CHARACTER */
+    C(0xb994, LLCR,    RRE,   EI,  0, r2_8u, 0, r1_32, mov2, 0)
+    C(0xb984, LLGCR,   RRE,   EI,  0, r2_8u, 0, r1, mov2, 0)
+    C(0xe394, LLC,     RXY_a, EI,  0, a2, new, r1_32, ld8u, 0)
+    C(0xe390, LLGC,    RXY_a, Z,   0, a2, r1, 0, ld8u, 0)
+/* LOAD LOGICAL HALFWORD */
+    C(0xb995, LLHR,    RRE,   EI,  0, r2_16u, 0, r1_32, mov2, 0)
+    C(0xb985, LLGHR,   RRE,   EI,  0, r2_16u, 0, r1, mov2, 0)
+    C(0xe395, LLH,     RXY_a, EI,  0, a2, new, r1_32, ld16u, 0)
+    C(0xe391, LLGH,    RXY_a, Z,   0, a2, r1, 0, ld16u, 0)
+/* LOAD LOGICAL HALFWORD RELATIVE LONG */
+    C(0xc402, LLHRL,   RIL_b, GIE, 0, ri2, new, r1_32, ld16u, 0)
+    C(0xc406, LLGHRL,  RIL_b, GIE, 0, ri2, r1, 0, ld16u, 0)
+/* LOAD LOGICAL IMMEDATE */
+    D(0xc00e, LLIHF,   RIL_a, EI, 0, i2_32u_shl, 0, r1, mov2, 0, 32)
+    D(0xc00f, LLILF,   RIL_a, EI, 0, i2_32u_shl, 0, r1, mov2, 0, 0)
+    D(0xa50c, LLIHH,   RI_a,  Z,  0, i2_16u_shl, 0, r1, mov2, 0, 48)
+    D(0xa50d, LLIHL,   RI_a,  Z,  0, i2_16u_shl, 0, r1, mov2, 0, 32)
+    D(0xa50e, LLILH,   RI_a,  Z,  0, i2_16u_shl, 0, r1, mov2, 0, 16)
+    D(0xa50f, LLILL,   RI_a,  Z,  0, i2_16u_shl, 0, r1, mov2, 0, 0)
+/* LOAD LOGICAL THIRTY ONE BITS */
+    C(0xb917, LLGTR,   RRE,   Z,  0, r2_o, r1, 0, llgt, 0)
+    C(0xe317, LLGT,    RXY_a, Z,  0, m2_32u, r1, 0, llgt, 0)
+/* LOAD FPR FROM GR */
+    C(0xb3c1, LDGR,    RRE,   FPRGR, 0, r2_o, 0, f1, mov2, 0)
+/* LOAD GR FROM FPR */
+    C(0xb3cd, LGDR,    RRE,   FPRGR, 0, f2_o, 0, r1, mov2, 0)
+/* LOAD NEGATIVE */
+    C(0x1100, LNR,     RR_a,  Z,   0, r2_32s, new, r1_32, nabs, nabs32)
+    C(0xb901, LNGR,    RRE,   Z,   0, r2, r1, 0, nabs, nabs64)
+    C(0xb911, LNGFR,   RRE,   Z,   0, r2_32s, r1, 0, nabs, nabs64)
+    C(0xb301, LNEBR,   RRE,   Z,   0, e2, new, e1, nabsf32, f32)
+    C(0xb311, LNDBR,   RRE,   Z,   0, f2_o, f1, 0, nabsf64, f64)
+    C(0xb341, LNXBR,   RRE,   Z,   0, x2_o, x1, 0, nabsf128, f128)
+/* LOAD ON CONDITION */
+    C(0xb9f2, LOCR,    RRF_c, LOC, r1, r2, new, r1_32, loc, 0)
+    C(0xb9e2, LOCGR,   RRF_c, LOC, r1, r2, r1, 0, loc, 0)
+    C(0xebf2, LOC,     RSY_b, LOC, r1, m2_32u, new, r1_32, loc, 0)
+    C(0xebe2, LOCG,    RSY_b, LOC, r1, m2_64, r1, 0, loc, 0)
+/* LOAD POSITIVE */
+    C(0x1000, LPR,     RR_a,  Z,   0, r2_32s, new, r1_32, abs, abs32)
+    C(0xb900, LPGR,    RRE,   Z,   0, r2, r1, 0, abs, abs64)
+    C(0xb910, LPGFR,   RRE,   Z,   0, r2_32s, r1, 0, abs, abs64)
+    C(0xb300, LPEBR,   RRE,   Z,   0, e2, new, e1, absf32, f32)
+    C(0xb310, LPDBR,   RRE,   Z,   0, f2_o, f1, 0, absf64, f64)
+    C(0xb340, LPXBR,   RRE,   Z,   0, x2_o, x1, 0, absf128, f128)
+/* LOAD REVERSED */
+    C(0xb91f, LRVR,    RRE,   Z,   0, r2_32u, new, r1_32, rev32, 0)
+    C(0xb90f, LRVGR,   RRE,   Z,   0, r2_o, r1, 0, rev64, 0)
+    C(0xe31f, LRVH,    RXY_a, Z,   0, m2_16u, new, r1_16, rev16, 0)
+    C(0xe31e, LRV,     RXY_a, Z,   0, m2_32u, new, r1_32, rev32, 0)
+    C(0xe30f, LRVG,    RXY_a, Z,   0, m2_64, r1, 0, rev64, 0)
+/* LOAD ZERO */
+    C(0xb374, LZER,    RRE,   Z,   0, 0, 0, e1, zero, 0)
+    C(0xb375, LZDR,    RRE,   Z,   0, 0, 0, f1, zero, 0)
+    C(0xb376, LZXR,    RRE,   Z,   0, 0, 0, x1, zero2, 0)
+
+/* LOAD FPC */
+    C(0xb29d, LFPC,    S,     Z,   0, m2_32u, 0, 0, sfpc, 0)
+/* LOAD FPC AND SIGNAL */
+    C(0xb2bd, LFAS,    S,     IEEEE_SIM, 0, m2_32u, 0, 0, sfas, 0)
+
+/* LOAD LENGTHENED */
+    C(0xb304, LDEBR,   RRE,   Z,   0, e2, f1, 0, ldeb, 0)
+    C(0xb305, LXDBR,   RRE,   Z,   0, f2_o, x1, 0, lxdb, 0)
+    C(0xb306, LXEBR,   RRE,   Z,   0, e2, x1, 0, lxeb, 0)
+    C(0xed04, LDEB,    RXE,   Z,   0, m2_32u, f1, 0, ldeb, 0)
+    C(0xed05, LXDB,    RXE,   Z,   0, m2_64, x1, 0, lxdb, 0)
+    C(0xed06, LXEB,    RXE,   Z,   0, m2_32u, x1, 0, lxeb, 0)
+/* LOAD ROUNDED */
+    C(0xb344, LEDBR,   RRE,   Z,   0, f2_o, new, e1, ledb, 0)
+    C(0xb345, LDXBR,   RRE,   Z,   0, x2_o, f1, 0, ldxb, 0)
+    C(0xb346, LEXBR,   RRE,   Z,   0, x2_o, new, e1, lexb, 0)
+
+/* LOAD MULTIPLE */
+    C(0x9800, LM,      RS_a,  Z,   0, a2, 0, 0, lm32, 0)
+    C(0xeb98, LMY,     RSY_a, LD,  0, a2, 0, 0, lm32, 0)
+    C(0xeb04, LMG,     RSY_a, Z,   0, a2, 0, 0, lm64, 0)
+/* LOAD MULTIPLE HIGH */
+    C(0xeb96, LMH,     RSY_a, Z,   0, a2, 0, 0, lmh, 0)
+/* LOAD ACCESS MULTIPLE */
+    C(0x9a00, LAM,     RS_a,  Z,   0, a2, 0, 0, lam, 0)
+    C(0xeb9a, LAMY,    RSY_a, LD,  0, a2, 0, 0, lam, 0)
+
+/* MOVE */
+    C(0xd200, MVC,     SS_a,  Z,   la1, a2, 0, 0, mvc, 0)
+    C(0xe544, MVHHI,   SIL,   GIE, la1, i2, 0, m1_16, mov2, 0)
+    C(0xe54c, MVHI,    SIL,   GIE, la1, i2, 0, m1_32, mov2, 0)
+    C(0xe548, MVGHI,   SIL,   GIE, la1, i2, 0, m1_64, mov2, 0)
+    C(0x9200, MVI,     SI,    Z,   la1, i2, 0, m1_8, mov2, 0)
+    C(0xeb52, MVIY,    SIY,   LD,  la1, i2, 0, m1_8, mov2, 0)
+/* MOVE LONG */
+    C(0x0e00, MVCL,    RR_a,  Z,   0, 0, 0, 0, mvcl, 0)
+/* MOVE LONG EXTENDED */
+    C(0xa800, MVCLE,   RS_a,  Z,   0, a2, 0, 0, mvcle, 0)
+/* MOVE PAGE */
+    C(0xb254, MVPG,    RRE,   Z,   r1_o, r2_o, 0, 0, mvpg, 0)
+/* MOVE STRING */
+    C(0xb255, MVST,    RRE,   Z,   r1_o, r2_o, 0, 0, mvst, 0)
+
+/* MULTIPLY */
+    C(0x1c00, MR,      RR_a,  Z,   r1p1_32s, r2_32s, new, r1_D32, mul, 0)
+    C(0x5c00, M,       RX_a,  Z,   r1p1_32s, m2_32s, new, r1_D32, mul, 0)
+    C(0xe35c, MFY,     RXY_a, GIE, r1p1_32s, m2_32s, new, r1_D32, mul, 0)
+    C(0xb317, MEEBR,   RRE,   Z,   e1, e2, new, e1, meeb, 0)
+    C(0xb31c, MDBR,    RRE,   Z,   f1_o, f2_o, f1, 0, mdb, 0)
+    C(0xb34c, MXBR,    RRE,   Z,   0, x2_o, x1, 0, mxb, 0)
+    C(0xb30c, MDEBR,   RRE,   Z,   f1_o, e2, f1, 0, mdeb, 0)
+    C(0xb307, MXDBR,   RRE,   Z,   0, f2_o, x1, 0, mxdb, 0)
+    C(0xed17, MEEB,    RXE,   Z,   e1, m2_32u, new, e1, meeb, 0)
+    C(0xed1c, MDB,     RXE,   Z,   f1_o, m2_64, f1, 0, mdb, 0)
+    C(0xed0c, MDEB,    RXE,   Z,   f1_o, m2_32u, f1, 0, mdeb, 0)
+    C(0xed07, MXDB,    RXE,   Z,   0, m2_64, x1, 0, mxdb, 0)
+/* MULTIPLY HALFWORD */
+    C(0x4c00, MH,      RX_a,  Z,   r1_o, m2_16s, new, r1_32, mul, 0)
+    C(0xe37c, MHY,     RXY_a, GIE, r1_o, m2_16s, new, r1_32, mul, 0)
+/* MULTIPLY HALFWORD IMMEDIATE */
+    C(0xa70c, MHI,     RI_a,  Z,   r1_o, i2, new, r1_32, mul, 0)
+    C(0xa70d, MGHI,    RI_a,  Z,   r1_o, i2, r1, 0, mul, 0)
+/* MULTIPLY LOGICAL */
+    C(0xb996, MLR,     RRE,   Z,   r1p1_32u, r2_32u, new, r1_D32, mul, 0)
+    C(0xe396, ML,      RXY_a, Z,   r1p1_32u, m2_32u, new, r1_D32, mul, 0)
+    C(0xb986, MLGR,    RRE,   Z,   r1p1, r2_o, r1_P, 0, mul128, 0)
+    C(0xe386, MLG,     RXY_a, Z,   r1p1, m2_64, r1_P, 0, mul128, 0)
+/* MULTIPLY SINGLE */
+    C(0xb252, MSR,     RRE,   Z,   r1_o, r2_o, new, r1_32, mul, 0)
+    C(0x7100, MS,      RX_a,  Z,   r1_o, m2_32s, new, r1_32, mul, 0)
+    C(0xe351, MSY,     RXY_a, LD,  r1_o, m2_32s, new, r1_32, mul, 0)
+    C(0xb90c, MSGR,    RRE,   Z,   r1_o, r2_o, r1, 0, mul, 0)
+    C(0xb91c, MSGFR,   RRE,   Z,   r1_o, r2_32s, r1, 0, mul, 0)
+    C(0xe30c, MSG,     RXY_a, Z,   r1_o, m2_64, r1, 0, mul, 0)
+    C(0xe31c, MSGF,    RXY_a, Z,   r1_o, m2_32s, r1, 0, mul, 0)
+/* MULTIPLY SINGLE IMMEDIATE */
+    C(0xc201, MSFI,    RIL_a, GIE, r1_o, i2, new, r1_32, mul, 0)
+    C(0xc200, MSGFI,   RIL_a, GIE, r1_o, i2, r1, 0, mul, 0)
+
+/* MULTIPLY AND ADD */
+    C(0xb30e, MAEBR,   RRD,   Z,   e1, e2, new, e1, maeb, 0)
+    C(0xb31e, MADBR,   RRD,   Z,   f1_o, f2_o, f1, 0, madb, 0)
+    C(0xed0e, MAEB,    RXF,   Z,   e1, m2_32u, new, e1, maeb, 0)
+    C(0xed1e, MADB,    RXF,   Z,   f1_o, m2_64, f1, 0, madb, 0)
+/* MULTIPLY AND SUBTRACT */
+    C(0xb30f, MSEBR,   RRD,   Z,   e1, e2, new, e1, mseb, 0)
+    C(0xb31f, MSDBR,   RRD,   Z,   f1_o, f2_o, f1, 0, msdb, 0)
+    C(0xed0f, MSEB,    RXF,   Z,   e1, m2_32u, new, e1, mseb, 0)
+    C(0xed1f, MSDB,    RXF,   Z,   f1_o, m2_64, f1, 0, msdb, 0)
+
+/* OR */
+    C(0x1600, OR,      RR_a,  Z,   r1, r2, new, r1_32, or, nz32)
+    C(0xb9f6, ORK,     RRF_a, DO,  r2, r3, new, r1_32, or, nz32)
+    C(0x5600, O,       RX_a,  Z,   r1, m2_32s, new, r1_32, or, nz32)
+    C(0xe356, OY,      RXY_a, LD,  r1, m2_32s, new, r1_32, or, nz32)
+    C(0xb981, OGR,     RRE,   Z,   r1, r2, r1, 0, or, nz64)
+    C(0xb9e6, OGRK,    RRF_a, DO,  r2, r3, r1, 0, or, nz64)
+    C(0xe381, OG,      RXY_a, Z,   r1, m2_64, r1, 0, or, nz64)
+    C(0xd600, OC,      SS_a,  Z,   la1, a2, 0, 0, oc, 0)
+/* OR IMMEDIATE */
+    D(0xc00c, OIHF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, ori, 0, 0x2020)
+    D(0xc00d, OILF,    RIL_a, EI,  r1_o, i2_32u, r1, 0, ori, 0, 0x2000)
+    D(0xa508, OIHH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, ori, 0, 0x1030)
+    D(0xa509, OIHL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, ori, 0, 0x1020)
+    D(0xa50a, OILH,    RI_a,  Z,   r1_o, i2_16u, r1, 0, ori, 0, 0x1010)
+    D(0xa50b, OILL,    RI_a,  Z,   r1_o, i2_16u, r1, 0, ori, 0, 0x1000)
+    C(0x9600, OI,      SI,    Z,   m1_8u, i2_8u, new, m1_8, or, nz64)
+    C(0xeb56, OIY,     SIY,   LD,  m1_8u, i2_8u, new, m1_8, or, nz64)
+
+/* PREFETCH */
+    /* Implemented as nops of course.  */
+    C(0xe336, PFD,     RXY_b, GIE, 0, 0, 0, 0, 0, 0)
+    C(0xc602, PFDRL,   RIL_c, GIE, 0, 0, 0, 0, 0, 0)
+
+/* POPULATION COUNT */
+    C(0xb9e1, POPCNT,  RRE,   PC,  0, r2_o, r1, 0, popcnt, nz64)
+
+/* ROTATE LEFT SINGLE LOGICAL */
+    C(0xeb1d, RLL,     RSY_a, Z,   r3_o, sh32, new, r1_32, rll32, 0)
+    C(0xeb1c, RLLG,    RSY_a, Z,   r3_o, sh64, r1, 0, rll64, 0)
+
+/* ROTATE THEN INSERT SELECTED BITS */
+    C(0xec55, RISBG,   RIE_f, GIE, 0, r2, r1, 0, risbg, s64)
+    C(0xec5d, RISBHG,  RIE_f, GIE, 0, r2, r1, 0, risbg, 0)
+    C(0xec51, RISBLG,  RIE_f, GIE, 0, r2, r1, 0, risbg, 0)
+/* ROTATE_THEN <OP> SELECTED BITS */
+    C(0xec54, RNSBG,   RIE_f, GIE, 0, r2, r1, 0, rosbg, 0)
+    C(0xec56, ROSBG,   RIE_f, GIE, 0, r2, r1, 0, rosbg, 0)
+    C(0xec57, RXSBG,   RIE_f, GIE, 0, r2, r1, 0, rosbg, 0)
+
+/* SEARCH STRING */
+    C(0xb25e, SRST,    RRE,   Z,   r1_o, r2_o, 0, 0, srst, 0)
+
+/* SET ACCESS */
+    C(0xb24e, SAR,     RRE,   Z,   0, r2_o, 0, 0, sar, 0)
+/* SET FPC */
+    C(0xb384, SFPC,    RRE,   Z,   0, r1_o, 0, 0, sfpc, 0)
+/* SET FPC AND SIGNAL */
+    C(0xb385, SFASR,   RRE,   IEEEE_SIM, 0, r1_o, 0, 0, sfas, 0)
+/* SET BFP ROUNDING MODE */
+    C(0xb299, SRNM,    S,     Z,   0, 0, 0, 0, srnm, 0)
+    C(0xb2b8, SRNMB,   S,     FPE, 0, 0, 0, 0, srnm, 0)
+/* SET DFP ROUNDING MODE */
+    C(0xb2b9, SRNMT,   S,     DFP, 0, 0, 0, 0, srnm, 0)
+
+/* SHIFT LEFT SINGLE */
+    D(0x8b00, SLA,     RS_a,  Z,   r1, sh32, new, r1_32, sla, 0, 31)
+    D(0xebdd, SLAK,    RSY_a, DO,  r3, sh32, new, r1_32, sla, 0, 31)
+    D(0xeb0b, SLAG,    RSY_a, Z,   r3, sh64, r1, 0, sla, 0, 63)
+/* SHIFT LEFT SINGLE LOGICAL */
+    C(0x8900, SLL,     RS_a,  Z,   r1_o, sh32, new, r1_32, sll, 0)
+    C(0xebdf, SLLK,    RSY_a, DO,  r3_o, sh32, new, r1_32, sll, 0)
+    C(0xeb0d, SLLG,    RSY_a, Z,   r3_o, sh64, r1, 0, sll, 0)
+/* SHIFT RIGHT SINGLE */
+    C(0x8a00, SRA,     RS_a,  Z,   r1_32s, sh32, new, r1_32, sra, s32)
+    C(0xebdc, SRAK,    RSY_a, DO,  r3_32s, sh32, new, r1_32, sra, s32)
+    C(0xeb0a, SRAG,    RSY_a, Z,   r3_o, sh64, r1, 0, sra, s64)
+/* SHIFT RIGHT SINGLE LOGICAL */
+    C(0x8800, SRL,     RS_a,  Z,   r1_32u, sh32, new, r1_32, srl, 0)
+    C(0xebde, SRLK,    RSY_a, DO,  r3_32u, sh32, new, r1_32, srl, 0)
+    C(0xeb0c, SRLG,    RSY_a, Z,   r3_o, sh64, r1, 0, srl, 0)
+/* SHIFT LEFT DOUBLE */
+    D(0x8f00, SLDA,    RS_a,  Z,   r1_D32, sh64, new, r1_D32, sla, 0, 31)
+/* SHIFT LEFT DOUBLE LOGICAL */
+    C(0x8d00, SLDL,    RS_a,  Z,   r1_D32, sh64, new, r1_D32, sll, 0)
+/* SHIFT RIGHT DOUBLE */
+    C(0x8e00, SRDA,    RS_a,  Z,   r1_D32, sh64, new, r1_D32, sra, s64)
+/* SHIFT RIGHT DOUBLE LOGICAL */
+    C(0x8c00, SRDL,    RS_a,  Z,   r1_D32, sh64, new, r1_D32, srl, 0)
+
+/* SQUARE ROOT */
+    C(0xb314, SQEBR,   RRE,   Z,   0, e2, new, e1, sqeb, 0)
+    C(0xb315, SQDBR,   RRE,   Z,   0, f2_o, f1, 0, sqdb, 0)
+    C(0xb316, SQXBR,   RRE,   Z,   0, x2_o, x1, 0, sqxb, 0)
+    C(0xed14, SQEB,    RXE,   Z,   0, m2_32u, new, e1, sqeb, 0)
+    C(0xed15, SQDB,    RXE,   Z,   0, m2_64, f1, 0, sqdb, 0)
+
+/* STORE */
+    C(0x5000, ST,      RX_a,  Z,   r1_o, a2, 0, 0, st32, 0)
+    C(0xe350, STY,     RXY_a, LD,  r1_o, a2, 0, 0, st32, 0)
+    C(0xe324, STG,     RXY_a, Z,   r1_o, a2, 0, 0, st64, 0)
+    C(0x6000, STD,     RX_a,  Z,   f1_o, a2, 0, 0, st64, 0)
+    C(0xed67, STDY,    RXY_a, LD,  f1_o, a2, 0, 0, st64, 0)
+    C(0x7000, STE,     RX_a,  Z,   e1, a2, 0, 0, st32, 0)
+    C(0xed66, STEY,    RXY_a, LD,  e1, a2, 0, 0, st32, 0)
+/* STORE RELATIVE LONG */
+    C(0xc40f, STRL,    RIL_b, GIE, r1_o, ri2, 0, 0, st32, 0)
+    C(0xc40b, STGRL,   RIL_b, GIE, r1_o, ri2, 0, 0, st64, 0)
+/* STORE CHARACTER */
+    C(0x4200, STC,     RX_a,  Z,   r1_o, a2, 0, 0, st8, 0)
+    C(0xe372, STCY,    RXY_a, LD,  r1_o, a2, 0, 0, st8, 0)
+/* STORE CHARACTERS UNDER MASK */
+    D(0xbe00, STCM,    RS_b,  Z,   r1_o, a2, 0, 0, stcm, 0, 0)
+    D(0xeb2d, STCMY,   RSY_b, LD,  r1_o, a2, 0, 0, stcm, 0, 0)
+    D(0xeb2c, STCMH,   RSY_b, LD,  r1_o, a2, 0, 0, stcm, 0, 32)
+/* STORE HALFWORD */
+    C(0x4000, STH,     RX_a,  Z,   r1_o, a2, 0, 0, st16, 0)
+    C(0xe370, STHY,    RXY_a, LD,  r1_o, a2, 0, 0, st16, 0)
+/* STORE HALFWORD RELATIVE LONG */
+    C(0xc407, STHRL,   RIL_b, GIE, r1_o, ri2, 0, 0, st16, 0)
+/* STORE ON CONDITION */
+    D(0xebf3, STOC,    RSY_b, LOC, 0, 0, 0, 0, soc, 0, 0)
+    D(0xebe3, STOCG,   RSY_b, LOC, 0, 0, 0, 0, soc, 0, 1)
+/* STORE REVERSED */
+    C(0xe33f, STRVH,   RXY_a, Z,   la2, r1_16u, new, m1_16, rev16, 0)
+    C(0xe33e, STRV,    RXY_a, Z,   la2, r1_32u, new, m1_32, rev32, 0)
+    C(0xe32f, STRVG,   RXY_a, Z,   la2, r1_o, new, m1_64, rev64, 0)
+
+/* STORE FPC */
+    C(0xb29c, STFPC,   S,     Z,   0, a2, new, m2_32, efpc, 0)
+
+/* STORE MULTIPLE */
+    D(0x9000, STM,     RS_a,  Z,   0, a2, 0, 0, stm, 0, 4)
+    D(0xeb90, STMY,    RSY_a, LD,  0, a2, 0, 0, stm, 0, 4)
+    D(0xeb24, STMG,    RSY_a, Z,   0, a2, 0, 0, stm, 0, 8)
+/* STORE MULTIPLE HIGH */
+    C(0xeb26, STMH,    RSY_a, Z,   0, a2, 0, 0, stmh, 0)
+/* STORE ACCESS MULTIPLE */
+    C(0x9b00, STAM,    RS_a,  Z,   0, a2, 0, 0, stam, 0)
+    C(0xeb9b, STAMY,   RSY_a, LD,  0, a2, 0, 0, stam, 0)
+
+/* SUBTRACT */
+    C(0x1b00, SR,      RR_a,  Z,   r1, r2, new, r1_32, sub, subs32)
+    C(0xb9f9, SRK,     RRF_a, DO,  r2, r3, new, r1_32, sub, subs32)
+    C(0x5b00, S,       RX_a,  Z,   r1, m2_32s, new, r1_32, sub, subs32)
+    C(0xe35b, SY,      RXY_a, LD,  r1, m2_32s, new, r1_32, sub, subs32)
+    C(0xb909, SGR,     RRE,   Z,   r1, r2, r1, 0, sub, subs64)
+    C(0xb919, SGFR,    RRE,   Z,   r1, r2_32s, r1, 0, sub, subs64)
+    C(0xb9e9, SGRK,    RRF_a, DO,  r2, r3, r1, 0, sub, subs64)
+    C(0xe309, SG,      RXY_a, Z,   r1, m2_64, r1, 0, sub, subs64)
+    C(0xe319, SGF,     RXY_a, Z,   r1, m2_32s, r1, 0, sub, subs64)
+    C(0xb30b, SEBR,    RRE,   Z,   e1, e2, new, e1, seb, f32)
+    C(0xb31b, SDBR,    RRE,   Z,   f1_o, f2_o, f1, 0, sdb, f64)
+    C(0xb34b, SXBR,    RRE,   Z,   0, x2_o, x1, 0, sxb, f128)
+    C(0xed0b, SEB,     RXE,   Z,   e1, m2_32u, new, e1, seb, f32)
+    C(0xed1b, SDB,     RXE,   Z,   f1_o, m2_64, f1, 0, sdb, f64)
+/* SUBTRACT HALFWORD */
+    C(0x4b00, SH,      RX_a,  Z,   r1, m2_16s, new, r1_32, sub, subs32)
+    C(0xe37b, SHY,     RXY_a, LD,  r1, m2_16s, new, r1_32, sub, subs32)
+/* SUBTRACT LOGICAL */
+    C(0x1f00, SLR,     RR_a,  Z,   r1, r2, new, r1_32, sub, subu32)
+    C(0xb9fb, SLRK,    RRF_a, DO,  r2, r3, new, r1_32, sub, subu32)
+    C(0x5f00, SL,      RX_a,  Z,   r1, m2_32u, new, r1_32, sub, subu32)
+    C(0xe35f, SLY,     RXY_a, LD,  r1, m2_32u, new, r1_32, sub, subu32)
+    C(0xb90b, SLGR,    RRE,   Z,   r1, r2, r1, 0, sub, subu64)
+    C(0xb91b, SLGFR,   RRE,   Z,   r1, r2_32u, r1, 0, sub, subu64)
+    C(0xb9eb, SLGRK,   RRF_a, DO,  r2, r3, r1, 0, sub, subu64)
+    C(0xe30b, SLG,     RXY_a, Z,   r1, m2_64, r1, 0, sub, subu64)
+    C(0xe31b, SLGF,    RXY_a, Z,   r1, m2_32u, r1, 0, sub, subu64)
+/* SUBTRACT LOGICAL IMMEDIATE */
+    C(0xc205, SLFI,    RIL_a, EI,  r1, i2_32u, new, r1_32, sub, subu32)
+    C(0xc204, SLGFI,   RIL_a, EI,  r1, i2_32u, r1, 0, sub, subu64)
+/* SUBTRACT LOGICAL WITH BORROW */
+    C(0xb999, SLBR,    RRE,   Z,   r1, r2, new, r1_32, subb, subb32)
+    C(0xb989, SLBGR,   RRE,   Z,   r1, r2, r1, 0, subb, subb64)
+    C(0xe399, SLB,     RXY_a, Z,   r1, m2_32u, new, r1_32, subb, subb32)
+    C(0xe389, SLBG,    RXY_a, Z,   r1, m2_64, r1, 0, subb, subb64)
+
+/* SUPERVISOR CALL */
+    C(0x0a00, SVC,     I,     Z,   0, 0, 0, 0, svc, 0)
+
+/* TEST DATA CLASS */
+    C(0xed10, TCEB,    RXE,   Z,   e1, a2, 0, 0, tceb, 0)
+    C(0xed11, TCDB,    RXE,   Z,   f1_o, a2, 0, 0, tcdb, 0)
+    C(0xed12, TCXB,    RXE,   Z,   x1_o, a2, 0, 0, tcxb, 0)
+
+/* TEST UNDER MASK */
+    C(0x9100, TM,      SI,    Z,   m1_8u, i2_8u, 0, 0, 0, tm32)
+    C(0xeb51, TMY,     SIY,   LD,  m1_8u, i2_8u, 0, 0, 0, tm32)
+    D(0xa702, TMHH,    RI_a,  Z,   r1_o, i2_16u_shl, 0, 0, 0, tm64, 48)
+    D(0xa703, TMHL,    RI_a,  Z,   r1_o, i2_16u_shl, 0, 0, 0, tm64, 32)
+    D(0xa700, TMLH,    RI_a,  Z,   r1_o, i2_16u_shl, 0, 0, 0, tm64, 16)
+    D(0xa701, TMLL,    RI_a,  Z,   r1_o, i2_16u_shl, 0, 0, 0, tm64, 0)
+
+/* TRANSLATE */
+    C(0xdc00, TR,      SS_a,  Z,   la1, a2, 0, 0, tr, 0)
+
+/* UNPACK */
+    /* Really format SS_b, but we pack both lengths into one argument
+       for the helper call, so we might as well leave one 8-bit field.  */
+    C(0xf300, UNPK,    SS_a,  Z,   la1, a2, 0, 0, unpk, 0)
+
+#ifndef CONFIG_USER_ONLY
+/* COMPARE AND SWAP AND PURGE */
+    C(0xb250, CSP,     RRE,   Z,   0, ra2, 0, 0, csp, 0)
+/* DIAGNOSE (KVM hypercall) */
+    C(0x8300, DIAG,    RX_a,  Z,   0, 0, 0, 0, diag, 0)
+/* INSERT STORAGE KEY EXTENDED */
+    C(0xb229, ISKE,    RRE,   Z,   0, r2_o, new, r1_8, iske, 0)
+/* INVALIDATE PAGE TABLE ENTRY */
+    C(0xb221, IPTE,    RRF_a, Z,   r1_o, r2_o, 0, 0, ipte, 0)
+/* LOAD CONTROL */
+    C(0xb700, LCTL,    RS_a,  Z,   0, a2, 0, 0, lctl, 0)
+    C(0xeb2f, LCTLG,   RSY_a, Z,   0, a2, 0, 0, lctlg, 0)
+/* LOAD PSW */
+    C(0x8200, LPSW,    S,     Z,   0, a2, 0, 0, lpsw, 0)
+/* LOAD PSW EXTENDED */
+    C(0xb2b2, LPSWE,   S,     Z,   0, a2, 0, 0, lpswe, 0)
+/* LOAD REAL ADDRESS */
+    C(0xb100, LRA,     RX_a,  Z,   0, a2, r1, 0, lra, 0)
+    C(0xe313, LRAY,    RXY_a, LD,  0, a2, r1, 0, lra, 0)
+    C(0xe303, LRAG,    RXY_a, Z,   0, a2, r1, 0, lra, 0)
+/* MOVE TO PRIMARY */
+    C(0xda00, MVCP,    SS_d,  Z,   la1, a2, 0, 0, mvcp, 0)
+/* MOVE TO SECONDARY */
+    C(0xdb00, MVCS,    SS_d,  Z,   la1, a2, 0, 0, mvcs, 0)
+/* PURGE TLB */
+    C(0xb20d, PTLB,    S,     Z,   0, 0, 0, 0, ptlb, 0)
+/* RESET REFERENCE BIT EXTENDED */
+    C(0xb22a, RRBE,    RRE,   Z,   0, r2_o, 0, 0, rrbe, 0)
+/* SERVICE CALL LOGICAL PROCESSOR (PV hypercall) */
+    C(0xb220, SERVC,   RRE,   Z,   r1_o, r2_o, 0, 0, servc, 0)
+/* SET ADDRESSING MODE */
+    /* We only do 64-bit, so accept this as a no-op.
+       Let SAM24 and SAM31 signal illegal instruction.  */
+    C(0x010e, SAM64,   E,     Z,   0, 0, 0, 0, 0, 0)
+/* SET ADDRESS SPACE CONTROL FAST */
+    C(0xb279, SACF,    S,     Z,   0, a2, 0, 0, sacf, 0)
+/* SET CLOCK */
+    /* ??? Not implemented - is it necessary? */
+    C(0xb204, SCK,     S,     Z,   0, 0, 0, 0, 0, 0)
+/* SET CLOCK COMPARATOR */
+    C(0xb206, SCKC,    S,     Z,   0, m2_64, 0, 0, sckc, 0)
+/* SET CPU TIMER */
+    C(0xb208, SPT,     S,     Z,   0, m2_64, 0, 0, spt, 0)
+/* SET PREFIX */
+    C(0xb210, SPX,     S,     Z,   0, m2_32u, 0, 0, spx, 0)
+/* SET PSW KEY FROM ADDRESS */
+    C(0xb20a, SPKA,    S,     Z,   0, a2, 0, 0, spka, 0)
+/* SET STORAGE KEY EXTENDED */
+    C(0xb22b, SSKE,    RRF_c, Z,   r1_o, r2_o, 0, 0, sske, 0)
+/* SET SYSTEM MASK */
+    C(0x8000, SSM,     S,     Z,   0, m2_8u, 0, 0, ssm, 0)
+/* SIGNAL PROCESSOR */
+    C(0xae00, SIGP,    RS_a,  Z,   r3_o, a2, 0, 0, sigp, 0)
+/* STORE CLOCK */
+    C(0xb205, STCK,    S,     Z,   la2, 0, new, m1_64, stck, 0)
+    C(0xb27c, STCKF,   S,     Z,   la2, 0, new, m1_64, stck, 0)
+/* STORE CLOCK EXTENDED */
+    C(0xb278, STCKE,   S,     Z,   0, a2, 0, 0, stcke, 0)
+/* STORE CLOCK COMPARATOR */
+    C(0xb207, STCKC,   S,     Z,   la2, 0, new, m1_64, stckc, 0)
+/* STORE CONTROL */
+    C(0xb600, STCTL,   RS_a,  Z,   0, a2, 0, 0, stctl, 0)
+    C(0xeb25, STCTG,   RSY_a, Z,   0, a2, 0, 0, stctg, 0)
+/* STORE CPU ADDRESS */
+    C(0xb212, STAP,    S,     Z,   la2, 0, new, m1_16, stap, 0)
+/* STORE CPU ID */
+    C(0xb202, STIDP,   S,     Z,   la2, 0, new, m1_64, stidp, 0)
+/* STORE CPU TIMER */
+    C(0xb209, STPT,    S,     Z,   la2, 0, new, m1_64, stpt, 0)
+/* STORE FACILITY LIST */
+    C(0xb2b1, STFL,    S,     Z,   0, 0, 0, 0, stfl, 0)
+/* STORE PREFIX */
+    C(0xb211, STPX,    S,     Z,   la2, 0, new, m1_32, stpx, 0)
+/* STORE SYSTEM INFORMATION */
+    C(0xb27d, STSI,    S,     Z,   0, a2, 0, 0, stsi, 0)
+/* STORE THEN AND SYSTEM MASK */
+    C(0xac00, STNSM,   SI,    Z,   la1, 0, 0, 0, stnosm, 0)
+/* STORE THEN OR SYSTEM MASK */
+    C(0xad00, STOSM,   SI,    Z,   la1, 0, 0, 0, stnosm, 0)
+/* STORE USING REAL ADDRESS */
+    C(0xb246, STURA,   RRE,   Z,   r1_o, r2_o, 0, 0, stura, 0)
+/* TEST PROTECTION */
+    C(0xe501, TPROT,   SSE,   Z,   la1, a2, 0, 0, tprot, 0)
+
+/* I/O Instructions.  For each we simply indicate non-operation.  */
+    C(0xb276, XSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb230, CSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb231, HSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb232, MSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb23b, RCHP,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb238, RSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb233, SSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb234, STSCH,   S,     Z,   0, 0, 0, 0, subchannel, 0)
+    C(0xb235, TSCH,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+    /* ??? Not listed in PoO ninth edition, but there's a linux driver that
+       uses it: "A CHSC subchannel is usually present on LPAR only."  */
+    C(0xb25f, CHSC,    S,     Z,   0, 0, 0, 0, subchannel, 0)
+#endif /* CONFIG_USER_ONLY */
diff --git a/target-s390x/insn-format.def b/target-s390x/insn-format.def
new file mode 100644
index 0000000000..0e898b90bd
--- /dev/null
+++ b/target-s390x/insn-format.def
@@ -0,0 +1,55 @@
+/* Description of s390 insn formats.  */
+/* NAME   F1,          F2... */
+F0(E)
+F1(I,     I(1, 8, 8))
+F2(RI_a,  R(1, 8),     I(2,16,16))
+F2(RI_b,  R(1, 8),     I(2,16,16))
+F2(RI_c,  M(1, 8),     I(2,16,16))
+F3(RIE_a, R(1, 8),     I(2,16,16),  M(3,32))
+F4(RIE_b, R(1, 8),     R(2,12),     M(3,32),   I(4,16,16))
+F4(RIE_c, R(1, 8),     I(2,32, 8),  M(3,12),   I(4,16,16))
+F3(RIE_d, R(1, 8),     I(2,16,16),  R(3,12))
+F3(RIE_e, R(1, 8),     I(2,16,16),  R(3,12))
+F5(RIE_f, R(1, 8),     R(2,12),     I(3,16,8), I(4,24,8),  I(5,32,8))
+F2(RIL_a, R(1, 8),     I(2,16,32))
+F2(RIL_b, R(1, 8),     I(2,16,32))
+F2(RIL_c, M(1, 8),     I(2,16,32))
+F4(RIS,   R(1, 8),     I(2,32, 8),  M(3,12),   BD(4,16,20))
+/* ??? The PoO does not call out subtypes _a and _b for RR, as it does
+   for e.g. RX.  Our checking requires this for e.g. BCR.  */
+F2(RR_a,  R(1, 8),     R(2,12))
+F2(RR_b,  M(1, 8),     R(2,12))
+F2(RRE,   R(1,24),     R(2,28))
+F3(RRD,   R(1,16),     R(2,28),     R(3,24))
+F4(RRF_a, R(1,24),     R(2,28),     R(3,16),   M(4,20))
+F4(RRF_b, R(1,24),     R(2,28),     R(3,16),   M(4,20))
+F4(RRF_c, R(1,24),     R(2,28),     M(3,16),   M(4,20))
+F4(RRF_d, R(1,24),     R(2,28),     M(3,16),   M(4,20))
+F4(RRF_e, R(1,24),     R(2,28),     M(3,16),   M(4,20))
+F4(RRS,   R(1, 8),     R(2,12),     M(3,32),   BD(4,16,20))
+F3(RS_a,  R(1, 8),     BD(2,16,20), R(3,12))
+F3(RS_b,  R(1, 8),     BD(2,16,20), M(3,12))
+F3(RSI,   R(1, 8),     I(2,16,16),  R(3,12))
+F2(RSL,   L(1, 8, 4),  BD(1,16,20))
+F3(RSY_a, R(1, 8),     BDL(2),      R(3,12))
+F3(RSY_b, R(1, 8),     BDL(2),      M(3,12))
+F2(RX_a,  R(1, 8),     BXD(2))
+F2(RX_b,  M(1, 8),     BXD(2))
+F2(RXE,   R(1, 8),     BXD(2))
+F3(RXF,   R(1,32),     BXD(2),      R(3, 8))
+F2(RXY_a, R(1, 8),     BXDL(2))
+F2(RXY_b, M(1, 8),     BXDL(2))
+F1(S,     BD(2,16,20))
+F2(SI,    BD(1,16,20), I(2,8,8))
+F2(SIL,   BD(1,16,20), I(2,32,16))
+F2(SIY,   BDL(1),      I(2, 8, 8))
+F3(SS_a,  L(1, 8, 8),  BD(1,16,20), BD(2,32,36))
+F4(SS_b,  L(1, 8, 4),  BD(1,16,20), L(2,12,4),   BD(2,32,36))
+F4(SS_c,  L(1, 8, 4),  BD(1,16,20), BD(2,32,36), I(3,12, 4))
+/* ??? Odd man out.  The L1 field here is really a register, but the
+   easy way to compress the fields has R1 and B1 overlap.  */
+F4(SS_d,  L(1, 8, 4),  BD(1,16,20), BD(2,32,36), R(3,12))
+F4(SS_e,  R(1, 8),     BD(2,16,20), R(3,12),     BD(4,32,36))
+F3(SS_f,  BD(1,16,20), L(2,8,8),    BD(2,32,36))
+F2(SSE,   BD(1,16,20), BD(2,32,36))
+F3(SSF,   BD(1,16,20), BD(2,32,36), R(3,8))
diff --git a/target-s390x/int_helper.c b/target-s390x/int_helper.c
index b683709860..685830124f 100644
--- a/target-s390x/int_helper.c
+++ b/target-s390x/int_helper.c
@@ -30,46 +30,97 @@
 #endif
 
 /* 64/64 -> 128 unsigned multiplication */
-void HELPER(mlg)(CPUS390XState *env, uint32_t r1, uint64_t v2)
+uint64_t HELPER(mul128)(CPUS390XState *env, uint64_t v1, uint64_t v2)
 {
-#if HOST_LONG_BITS == 64 && defined(__GNUC__)
-    /* assuming 64-bit hosts have __uint128_t */
-    __uint128_t res = (__uint128_t)env->regs[r1 + 1];
+    uint64_t reth;
+    mulu64(&env->retxl, &reth, v1, v2);
+    return reth;
+}
 
-    res *= (__uint128_t)v2;
-    env->regs[r1] = (uint64_t)(res >> 64);
-    env->regs[r1 + 1] = (uint64_t)res;
-#else
-    mulu64(&env->regs[r1 + 1], &env->regs[r1], env->regs[r1 + 1], v2);
-#endif
+/* 64/32 -> 32 signed division */
+int64_t HELPER(divs32)(CPUS390XState *env, int64_t a, int64_t b64)
+{
+    int32_t ret, b = b64;
+    int64_t q;
+
+    if (b == 0) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+
+    ret = q = a / b;
+    env->retxl = a % b;
+
+    /* Catch non-representable quotient.  */
+    if (ret != q) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+
+    return ret;
 }
 
-/* 128 -> 64/64 unsigned division */
-void HELPER(dlg)(CPUS390XState *env, uint32_t r1, uint64_t v2)
+/* 64/32 -> 32 unsigned division */
+uint64_t HELPER(divu32)(CPUS390XState *env, uint64_t a, uint64_t b64)
+{
+    uint32_t ret, b = b64;
+    uint64_t q;
+
+    if (b == 0) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+
+    ret = q = a / b;
+    env->retxl = a % b;
+
+    /* Catch non-representable quotient.  */
+    if (ret != q) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+
+    return ret;
+}
+
+/* 64/64 -> 64 signed division */
+int64_t HELPER(divs64)(CPUS390XState *env, int64_t a, int64_t b)
 {
-    uint64_t divisor = v2;
+    /* Catch divide by zero, and non-representable quotient (MIN / -1).  */
+    if (b == 0 || (b == -1 && a == (1ll << 63))) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+    env->retxl = a % b;
+    return a / b;
+}
 
-    if (!env->regs[r1]) {
+/* 128 -> 64/64 unsigned division */
+uint64_t HELPER(divu64)(CPUS390XState *env, uint64_t ah, uint64_t al,
+                        uint64_t b)
+{
+    uint64_t ret;
+    /* Signal divide by zero.  */
+    if (b == 0) {
+        runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+    }
+    if (ah == 0) {
         /* 64 -> 64/64 case */
-        env->regs[r1] = env->regs[r1 + 1] % divisor;
-        env->regs[r1 + 1] = env->regs[r1 + 1] / divisor;
-        return;
+        env->retxl = al % b;
+        ret = al / b;
     } else {
+        /* ??? Move i386 idivq helper to host-utils.  */
 #if HOST_LONG_BITS == 64 && defined(__GNUC__)
         /* assuming 64-bit hosts have __uint128_t */
-        __uint128_t dividend = (((__uint128_t)env->regs[r1]) << 64) |
-            (env->regs[r1 + 1]);
-        __uint128_t quotient = dividend / divisor;
-        __uint128_t remainder = dividend % divisor;
-
-        env->regs[r1 + 1] = quotient;
-        env->regs[r1] = remainder;
+        __uint128_t a = ((__uint128_t)ah << 64) | al;
+        __uint128_t q = a / b;
+        env->retxl = a % b;
+        ret = q;
+        if (ret != q) {
+            runtime_exception(env, PGM_FIXPT_DIVIDE, GETPC());
+        }
 #else
         /* 32-bit hosts would need special wrapper functionality - just abort if
            we encounter such a case; it's very unlikely anyways. */
         cpu_abort(env, "128 -> 64/64 division not implemented\n");
 #endif
     }
+    return ret;
 }
 
 /* absolute value 32-bit */
@@ -114,69 +165,10 @@ int64_t HELPER(nabs_i64)(int64_t val)
     }
 }
 
-/* add with carry 32-bit unsigned */
-uint32_t HELPER(addc_u32)(uint32_t cc, uint32_t v1, uint32_t v2)
+/* count leading zeros, for find leftmost one */
+uint64_t HELPER(clz)(uint64_t v)
 {
-    uint32_t res;
-
-    res = v1 + v2;
-    if (cc & 2) {
-        res++;
-    }
-
-    return res;
-}
-
-/* subtract unsigned v2 from v1 with borrow */
-uint32_t HELPER(slb)(CPUS390XState *env, uint32_t cc, uint32_t r1, uint32_t v2)
-{
-    uint32_t v1 = env->regs[r1];
-    uint32_t res = v1 + (~v2) + (cc >> 1);
-
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) | res;
-    if (cc & 2) {
-        /* borrow */
-        return v1 ? 1 : 0;
-    } else {
-        return v1 ? 3 : 2;
-    }
-}
-
-/* subtract unsigned v2 from v1 with borrow */
-uint32_t HELPER(slbg)(CPUS390XState *env, uint32_t cc, uint32_t r1,
-                      uint64_t v1, uint64_t v2)
-{
-    uint64_t res = v1 + (~v2) + (cc >> 1);
-
-    env->regs[r1] = res;
-    if (cc & 2) {
-        /* borrow */
-        return v1 ? 1 : 0;
-    } else {
-        return v1 ? 3 : 2;
-    }
-}
-
-/* find leftmost one */
-uint32_t HELPER(flogr)(CPUS390XState *env, uint32_t r1, uint64_t v2)
-{
-    uint64_t res = 0;
-    uint64_t ov2 = v2;
-
-    while (!(v2 & 0x8000000000000000ULL) && v2) {
-        v2 <<= 1;
-        res++;
-    }
-
-    if (!v2) {
-        env->regs[r1] = 64;
-        env->regs[r1 + 1] = 0;
-        return 0;
-    } else {
-        env->regs[r1] = res;
-        env->regs[r1 + 1] = ov2 & ~(0x8000000000000000ULL >> res);
-        return 2;
-    }
+    return clz64(v);
 }
 
 uint64_t HELPER(cvd)(int32_t bin)
@@ -199,3 +191,15 @@ uint64_t HELPER(cvd)(int32_t bin)
 
     return dec;
 }
+
+uint64_t HELPER(popcnt)(uint64_t r2)
+{
+    uint64_t ret = 0;
+    int i;
+
+    for (i = 0; i < 64; i += 8) {
+        uint64_t t = ctpop32((r2 >> i) & 0xff);
+        ret |= t << i;
+    }
+    return ret;
+}
diff --git a/target-s390x/mem_helper.c b/target-s390x/mem_helper.c
index bed21e6e1c..372334b3c8 100644
--- a/target-s390x/mem_helper.c
+++ b/target-s390x/mem_helper.c
@@ -304,214 +304,142 @@ uint32_t HELPER(clm)(CPUS390XState *env, uint32_t r1, uint32_t mask,
     return cc;
 }
 
-/* store character under mask */
-void HELPER(stcm)(CPUS390XState *env, uint32_t r1, uint32_t mask,
-                  uint64_t addr)
+static inline uint64_t fix_address(CPUS390XState *env, uint64_t a)
 {
-    uint8_t r;
-
-    HELPER_LOG("%s: r1 0x%x mask 0x%x addr 0x%lx\n", __func__, r1, mask,
-               addr);
-    while (mask) {
-        if (mask & 8) {
-            r = (r1 & 0xff000000UL) >> 24;
-            cpu_stb_data(env, addr, r);
-            HELPER_LOG("mask 0x%x %02x (0x%lx) ", mask, r, addr);
-            addr++;
-        }
-        mask = (mask << 1) & 0xf;
-        r1 <<= 8;
+    /* 31-Bit mode */
+    if (!(env->psw.mask & PSW_MASK_64)) {
+        a &= 0x7fffffff;
     }
-    HELPER_LOG("\n");
+    return a;
 }
 
 static inline uint64_t get_address(CPUS390XState *env, int x2, int b2, int d2)
 {
     uint64_t r = d2;
-
     if (x2) {
         r += env->regs[x2];
     }
-
     if (b2) {
         r += env->regs[b2];
     }
-
-    /* 31-Bit mode */
-    if (!(env->psw.mask & PSW_MASK_64)) {
-        r &= 0x7fffffff;
-    }
-
-    return r;
+    return fix_address(env, r);
 }
 
 static inline uint64_t get_address_31fix(CPUS390XState *env, int reg)
 {
-    uint64_t r = env->regs[reg];
-
-    /* 31-Bit mode */
-    if (!(env->psw.mask & PSW_MASK_64)) {
-        r &= 0x7fffffff;
-    }
-
-    return r;
+    return fix_address(env, env->regs[reg]);
 }
 
 /* search string (c is byte to search, r2 is string, r1 end of string) */
-uint32_t HELPER(srst)(CPUS390XState *env, uint32_t c, uint32_t r1, uint32_t r2)
+uint64_t HELPER(srst)(CPUS390XState *env, uint64_t r0, uint64_t end,
+                      uint64_t str)
 {
-    uint64_t i;
-    uint32_t cc = 2;
-    uint64_t str = get_address_31fix(env, r2);
-    uint64_t end = get_address_31fix(env, r1);
-
-    HELPER_LOG("%s: c %d *r1 0x%" PRIx64 " *r2 0x%" PRIx64 "\n", __func__,
-               c, env->regs[r1], env->regs[r2]);
-
-    for (i = str; i != end; i++) {
-        if (cpu_ldub_data(env, i) == c) {
-            env->regs[r1] = i;
-            cc = 1;
-            break;
+    uint32_t len;
+    uint8_t v, c = r0;
+
+    str = fix_address(env, str);
+    end = fix_address(env, end);
+
+    /* Assume for now that R2 is unmodified.  */
+    env->retxl = str;
+
+    /* Lest we fail to service interrupts in a timely manner, limit the
+       amount of work we're willing to do.  For now, lets cap at 8k.  */
+    for (len = 0; len < 0x2000; ++len) {
+        if (str + len == end) {
+            /* Character not found.  R1 & R2 are unmodified.  */
+            env->cc_op = 2;
+            return end;
+        }
+        v = cpu_ldub_data(env, str + len);
+        if (v == c) {
+            /* Character found.  Set R1 to the location; R2 is unmodified.  */
+            env->cc_op = 1;
+            return str + len;
         }
     }
 
-    return cc;
+    /* CPU-determined bytes processed.  Advance R2 to next byte to process.  */
+    env->retxl = str + len;
+    env->cc_op = 3;
+    return end;
 }
 
 /* unsigned string compare (c is string terminator) */
-uint32_t HELPER(clst)(CPUS390XState *env, uint32_t c, uint32_t r1, uint32_t r2)
+uint64_t HELPER(clst)(CPUS390XState *env, uint64_t c, uint64_t s1, uint64_t s2)
 {
-    uint64_t s1 = get_address_31fix(env, r1);
-    uint64_t s2 = get_address_31fix(env, r2);
-    uint8_t v1, v2;
-    uint32_t cc;
+    uint32_t len;
 
     c = c & 0xff;
-#ifdef CONFIG_USER_ONLY
-    if (!c) {
-        HELPER_LOG("%s: comparing '%s' and '%s'\n",
-                   __func__, (char *)g2h(s1), (char *)g2h(s2));
-    }
-#endif
-    for (;;) {
-        v1 = cpu_ldub_data(env, s1);
-        v2 = cpu_ldub_data(env, s2);
-        if ((v1 == c || v2 == c) || (v1 != v2)) {
-            break;
+    s1 = fix_address(env, s1);
+    s2 = fix_address(env, s2);
+
+    /* Lest we fail to service interrupts in a timely manner, limit the
+       amount of work we're willing to do.  For now, lets cap at 8k.  */
+    for (len = 0; len < 0x2000; ++len) {
+        uint8_t v1 = cpu_ldub_data(env, s1 + len);
+        uint8_t v2 = cpu_ldub_data(env, s2 + len);
+        if (v1 == v2) {
+            if (v1 == c) {
+                /* Equal.  CC=0, and don't advance the registers.  */
+                env->cc_op = 0;
+                env->retxl = s2;
+                return s1;
+            }
+        } else {
+            /* Unequal.  CC={1,2}, and advance the registers.  Note that
+               the terminator need not be zero, but the string that contains
+               the terminator is by definition "low".  */
+            env->cc_op = (v1 == c ? 1 : v2 == c ? 2 : v1 < v2 ? 1 : 2);
+            env->retxl = s2 + len;
+            return s1 + len;
         }
-        s1++;
-        s2++;
     }
 
-    if (v1 == v2) {
-        cc = 0;
-    } else {
-        cc = (v1 < v2) ? 1 : 2;
-        /* FIXME: 31-bit mode! */
-        env->regs[r1] = s1;
-        env->regs[r2] = s2;
-    }
-    return cc;
+    /* CPU-determined bytes equal; advance the registers.  */
+    env->cc_op = 3;
+    env->retxl = s2 + len;
+    return s1 + len;
 }
 
 /* move page */
 void HELPER(mvpg)(CPUS390XState *env, uint64_t r0, uint64_t r1, uint64_t r2)
 {
     /* XXX missing r0 handling */
+    env->cc_op = 0;
 #ifdef CONFIG_USER_ONLY
-    int i;
-
-    for (i = 0; i < TARGET_PAGE_SIZE; i++) {
-        cpu_stb_data(env, r1 + i, cpu_ldub_data(env, r2 + i));
-    }
+    memmove(g2h(r1), g2h(r2), TARGET_PAGE_SIZE);
 #else
     mvc_fast_memmove(env, TARGET_PAGE_SIZE, r1, r2);
 #endif
 }
 
 /* string copy (c is string terminator) */
-void HELPER(mvst)(CPUS390XState *env, uint32_t c, uint32_t r1, uint32_t r2)
+uint64_t HELPER(mvst)(CPUS390XState *env, uint64_t c, uint64_t d, uint64_t s)
 {
-    uint64_t dest = get_address_31fix(env, r1);
-    uint64_t src = get_address_31fix(env, r2);
-    uint8_t v;
+    uint32_t len;
 
     c = c & 0xff;
-#ifdef CONFIG_USER_ONLY
-    if (!c) {
-        HELPER_LOG("%s: copy '%s' to 0x%lx\n", __func__, (char *)g2h(src),
-                   dest);
-    }
-#endif
-    for (;;) {
-        v = cpu_ldub_data(env, src);
-        cpu_stb_data(env, dest, v);
+    d = fix_address(env, d);
+    s = fix_address(env, s);
+
+    /* Lest we fail to service interrupts in a timely manner, limit the
+       amount of work we're willing to do.  For now, lets cap at 8k.  */
+    for (len = 0; len < 0x2000; ++len) {
+        uint8_t v = cpu_ldub_data(env, s + len);
+        cpu_stb_data(env, d + len, v);
         if (v == c) {
-            break;
+            /* Complete.  Set CC=1 and advance R1.  */
+            env->cc_op = 1;
+            env->retxl = s;
+            return d + len;
         }
-        src++;
-        dest++;
-    }
-    env->regs[r1] = dest; /* FIXME: 31-bit mode! */
-}
-
-/* compare and swap 64-bit */
-uint32_t HELPER(csg)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
-{
-    /* FIXME: locking? */
-    uint32_t cc;
-    uint64_t v2 = cpu_ldq_data(env, a2);
-
-    if (env->regs[r1] == v2) {
-        cc = 0;
-        cpu_stq_data(env, a2, env->regs[r3]);
-    } else {
-        cc = 1;
-        env->regs[r1] = v2;
-    }
-    return cc;
-}
-
-/* compare double and swap 64-bit */
-uint32_t HELPER(cdsg)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
-{
-    /* FIXME: locking? */
-    uint32_t cc;
-    uint64_t v2_hi = cpu_ldq_data(env, a2);
-    uint64_t v2_lo = cpu_ldq_data(env, a2 + 8);
-    uint64_t v1_hi = env->regs[r1];
-    uint64_t v1_lo = env->regs[r1 + 1];
-
-    if ((v1_hi == v2_hi) && (v1_lo == v2_lo)) {
-        cc = 0;
-        cpu_stq_data(env, a2, env->regs[r3]);
-        cpu_stq_data(env, a2 + 8, env->regs[r3 + 1]);
-    } else {
-        cc = 1;
-        env->regs[r1] = v2_hi;
-        env->regs[r1 + 1] = v2_lo;
     }
 
-    return cc;
-}
-
-/* compare and swap 32-bit */
-uint32_t HELPER(cs)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
-{
-    /* FIXME: locking? */
-    uint32_t cc;
-    uint32_t v2 = cpu_ldl_data(env, a2);
-
-    HELPER_LOG("%s: r1 %d a2 0x%lx r3 %d\n", __func__, r1, a2, r3);
-    if (((uint32_t)env->regs[r1]) == v2) {
-        cc = 0;
-        cpu_stl_data(env, a2, (uint32_t)env->regs[r3]);
-    } else {
-        cc = 1;
-        env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) | v2;
-    }
-    return cc;
+    /* Incomplete.  Set CC=3 and signal to advance R1 and R2.  */
+    env->cc_op = 3;
+    env->retxl = s + len;
+    return d + len;
 }
 
 static uint32_t helper_icm(CPUS390XState *env, uint32_t r1, uint64_t address,
@@ -594,7 +522,7 @@ uint32_t HELPER(ex)(CPUS390XState *env, uint32_t cc, uint64_t v1,
         HELPER_LOG("%s: svc %ld via execute\n", __func__, (insn | v1) & 0xff);
         env->psw.addr = ret - 4;
         env->int_svc_code = (insn | v1) & 0xff;
-        env->int_svc_ilc = 4;
+        env->int_svc_ilen = 4;
         helper_exception(env, EXCP_SVC);
     } else if ((insn & 0xff00) == 0xbf00) {
         uint32_t insn2, r1, r3, b2, d2;
@@ -613,55 +541,6 @@ uint32_t HELPER(ex)(CPUS390XState *env, uint32_t cc, uint64_t v1,
     return cc;
 }
 
-/* store character under mask high operates on the upper half of r1 */
-void HELPER(stcmh)(CPUS390XState *env, uint32_t r1, uint64_t address,
-                   uint32_t mask)
-{
-    int pos = 56; /* top of the upper half of r1 */
-
-    while (mask) {
-        if (mask & 8) {
-            cpu_stb_data(env, address, (env->regs[r1] >> pos) & 0xff);
-            address++;
-        }
-        mask = (mask << 1) & 0xf;
-        pos -= 8;
-    }
-}
-
-/* insert character under mask high; same as icm, but operates on the
-   upper half of r1 */
-uint32_t HELPER(icmh)(CPUS390XState *env, uint32_t r1, uint64_t address,
-                      uint32_t mask)
-{
-    int pos = 56; /* top of the upper half of r1 */
-    uint64_t rmask = 0xff00000000000000ULL;
-    uint8_t val = 0;
-    int ccd = 0;
-    uint32_t cc = 0;
-
-    while (mask) {
-        if (mask & 8) {
-            env->regs[r1] &= ~rmask;
-            val = cpu_ldub_data(env, address);
-            if ((val & 0x80) && !ccd) {
-                cc = 1;
-            }
-            ccd = 1;
-            if (val && cc == 0) {
-                cc = 2;
-            }
-            env->regs[r1] |= (uint64_t)val << pos;
-            address++;
-        }
-        mask = (mask << 1) & 0xf;
-        pos -= 8;
-        rmask >>= 8;
-    }
-
-    return cc;
-}
-
 /* load access registers r1 to r3 from memory at a2 */
 void HELPER(lam)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3)
 {
@@ -822,42 +701,49 @@ uint32_t HELPER(clcle)(CPUS390XState *env, uint32_t r1, uint64_t a2,
 }
 
 /* checksum */
-void HELPER(cksm)(CPUS390XState *env, uint32_t r1, uint32_t r2)
+uint64_t HELPER(cksm)(CPUS390XState *env, uint64_t r1,
+                      uint64_t src, uint64_t src_len)
 {
-    uint64_t src = get_address_31fix(env, r2);
-    uint64_t src_len = env->regs[(r2 + 1) & 15];
-    uint64_t cksm = (uint32_t)env->regs[r1];
+    uint64_t max_len, len;
+    uint64_t cksm = (uint32_t)r1;
 
-    while (src_len >= 4) {
-        cksm += cpu_ldl_data(env, src);
+    /* Lest we fail to service interrupts in a timely manner, limit the
+       amount of work we're willing to do.  For now, lets cap at 8k.  */
+    max_len = (src_len > 0x2000 ? 0x2000 : src_len);
 
-        /* move to next word */
-        src_len -= 4;
-        src += 4;
+    /* Process full words as available.  */
+    for (len = 0; len + 4 <= max_len; len += 4, src += 4) {
+        cksm += (uint32_t)cpu_ldl_data(env, src);
     }
 
-    switch (src_len) {
-    case 0:
-        break;
+    switch (max_len - len) {
     case 1:
         cksm += cpu_ldub_data(env, src) << 24;
+        len += 1;
         break;
     case 2:
         cksm += cpu_lduw_data(env, src) << 16;
+        len += 2;
         break;
     case 3:
         cksm += cpu_lduw_data(env, src) << 16;
         cksm += cpu_ldub_data(env, src + 2) << 8;
+        len += 3;
         break;
     }
 
-    /* indicate we've processed everything */
-    env->regs[r2] = src + src_len;
-    env->regs[(r2 + 1) & 15] = 0;
+    /* Fold the carry from the checksum.  Note that we can see carry-out
+       during folding more than once (but probably not more than twice).  */
+    while (cksm > 0xffffffffull) {
+        cksm = (uint32_t)cksm + (cksm >> 32);
+    }
+
+    /* Indicate whether or not we've processed everything.  */
+    env->cc_op = (len == src_len ? 0 : 3);
 
-    /* store result */
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
-        ((uint32_t)cksm + (cksm >> 32));
+    /* Return both cksm and processed length.  */
+    env->retxl = cksm;
+    return len;
 }
 
 void HELPER(unpk)(CPUS390XState *env, uint32_t len, uint64_t dest,
@@ -1007,7 +893,7 @@ uint64_t HELPER(iske)(CPUS390XState *env, uint64_t r2)
 }
 
 /* set storage key extended */
-void HELPER(sske)(CPUS390XState *env, uint32_t r1, uint64_t r2)
+void HELPER(sske)(CPUS390XState *env, uint64_t r1, uint64_t r2)
 {
     uint64_t addr = get_address(env, 0, 0, r2);
 
@@ -1019,7 +905,7 @@ void HELPER(sske)(CPUS390XState *env, uint32_t r1, uint64_t r2)
 }
 
 /* reset reference bit extended */
-uint32_t HELPER(rrbe)(CPUS390XState *env, uint32_t r1, uint64_t r2)
+uint32_t HELPER(rrbe)(CPUS390XState *env, uint64_t r2)
 {
     uint8_t re;
     uint8_t key;
@@ -1045,16 +931,16 @@ uint32_t HELPER(rrbe)(CPUS390XState *env, uint32_t r1, uint64_t r2)
 }
 
 /* compare and swap and purge */
-uint32_t HELPER(csp)(CPUS390XState *env, uint32_t r1, uint32_t r2)
+uint32_t HELPER(csp)(CPUS390XState *env, uint32_t r1, uint64_t r2)
 {
     uint32_t cc;
     uint32_t o1 = env->regs[r1];
-    uint64_t a2 = get_address_31fix(env, r2) & ~3ULL;
+    uint64_t a2 = r2 & ~3ULL;
     uint32_t o2 = cpu_ldl_data(env, a2);
 
     if (o1 == o2) {
         cpu_stl_data(env, a2, env->regs[(r1 + 1) & 15]);
-        if (env->regs[r2] & 0x3) {
+        if (r2 & 0x3) {
             /* flush TLB / ALB */
             tlb_flush(env, 1);
         }
@@ -1154,13 +1040,13 @@ void HELPER(ptlb)(CPUS390XState *env)
 }
 
 /* store using real address */
-void HELPER(stura)(CPUS390XState *env, uint64_t addr, uint32_t v1)
+void HELPER(stura)(CPUS390XState *env, uint64_t addr, uint64_t v1)
 {
-    stw_phys(get_address(env, 0, 0, addr), v1);
+    stw_phys(get_address(env, 0, 0, addr), (uint32_t)v1);
 }
 
 /* load real address */
-uint32_t HELPER(lra)(CPUS390XState *env, uint64_t addr, uint32_t r1)
+uint64_t HELPER(lra)(CPUS390XState *env, uint64_t addr)
 {
     uint32_t cc = 0;
     int old_exc = env->exception_index;
@@ -1184,14 +1070,7 @@ uint32_t HELPER(lra)(CPUS390XState *env, uint64_t addr, uint32_t r1)
     }
     env->exception_index = old_exc;
 
-    if (!(env->psw.mask & PSW_MASK_64)) {
-        env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
-            (ret & 0xffffffffULL);
-    } else {
-        env->regs[r1] = ret;
-    }
-
-    return cc;
+    env->cc_op = cc;
+    return ret;
 }
-
 #endif
diff --git a/target-s390x/misc_helper.c b/target-s390x/misc_helper.c
index 70f9739685..3015bfe3f2 100644
--- a/target-s390x/misc_helper.c
+++ b/target-s390x/misc_helper.c
@@ -41,7 +41,27 @@
 #define HELPER_LOG(x...)
 #endif
 
-/* raise an exception */
+/* Raise an exception dynamically from a helper function.  */
+void QEMU_NORETURN runtime_exception(CPUS390XState *env, int excp,
+                                     uintptr_t retaddr)
+{
+    int t;
+
+    env->exception_index = EXCP_PGM;
+    env->int_pgm_code = excp;
+
+    /* Use the (ultimate) callers address to find the insn that trapped.  */
+    cpu_restore_state(env, retaddr);
+
+    /* Advance past the insn.  */
+    t = cpu_ldub_code(env, env->psw.addr);
+    env->int_pgm_ilen = t = get_ilen(t);
+    env->psw.addr += 2 * t;
+
+    cpu_loop_exit(env);
+}
+
+/* Raise an exception statically from a TB.  */
 void HELPER(exception)(CPUS390XState *env, uint32_t excp)
 {
     HELPER_LOG("%s: exception %d\n", __func__, excp);
@@ -50,7 +70,7 @@ void HELPER(exception)(CPUS390XState *env, uint32_t excp)
 }
 
 #ifndef CONFIG_USER_ONLY
-void program_interrupt(CPUS390XState *env, uint32_t code, int ilc)
+void program_interrupt(CPUS390XState *env, uint32_t code, int ilen)
 {
     qemu_log_mask(CPU_LOG_INT, "program interrupt at %#" PRIx64 "\n",
                   env->psw.addr);
@@ -61,18 +81,16 @@ void program_interrupt(CPUS390XState *env, uint32_t code, int ilc)
 #endif
     } else {
         env->int_pgm_code = code;
-        env->int_pgm_ilc = ilc;
+        env->int_pgm_ilen = ilen;
         env->exception_index = EXCP_PGM;
         cpu_loop_exit(env);
     }
 }
 
 /* SCLP service call */
-uint32_t HELPER(servc)(CPUS390XState *env, uint32_t r1, uint64_t r2)
+uint32_t HELPER(servc)(CPUS390XState *env, uint64_t r1, uint64_t r2)
 {
-    int r;
-
-    r = sclp_service_call(r1, r2);
+    int r = sclp_service_call(r1, r2);
     if (r < 0) {
         program_interrupt(env, -r, 4);
         return 0;
@@ -105,38 +123,22 @@ uint64_t HELPER(diag)(CPUS390XState *env, uint32_t num, uint64_t mem,
     }
 
     if (r) {
-        program_interrupt(env, PGM_OPERATION, ILC_LATER_INC);
+        program_interrupt(env, PGM_OPERATION, ILEN_LATER_INC);
     }
 
     return r;
 }
 
-/* Store CPU ID */
-void HELPER(stidp)(CPUS390XState *env, uint64_t a1)
-{
-    cpu_stq_data(env, a1, env->cpu_num);
-}
-
 /* Set Prefix */
 void HELPER(spx)(CPUS390XState *env, uint64_t a1)
 {
-    uint32_t prefix;
-
-    prefix = cpu_ldl_data(env, a1);
-    env->psa = prefix & 0xfffff000;
+    uint32_t prefix = a1 & 0x7fffe000;
+    env->psa = prefix;
     qemu_log("prefix: %#x\n", prefix);
     tlb_flush_page(env, 0);
     tlb_flush_page(env, TARGET_PAGE_SIZE);
 }
 
-/* Set Clock */
-uint32_t HELPER(sck)(uint64_t a1)
-{
-    /* XXX not implemented - is it necessary? */
-
-    return 0;
-}
-
 static inline uint64_t clock_value(CPUS390XState *env)
 {
     uint64_t time;
@@ -148,32 +150,14 @@ static inline uint64_t clock_value(CPUS390XState *env)
 }
 
 /* Store Clock */
-uint32_t HELPER(stck)(CPUS390XState *env, uint64_t a1)
-{
-    cpu_stq_data(env, a1, clock_value(env));
-
-    return 0;
-}
-
-/* Store Clock Extended */
-uint32_t HELPER(stcke)(CPUS390XState *env, uint64_t a1)
+uint64_t HELPER(stck)(CPUS390XState *env)
 {
-    cpu_stb_data(env, a1, 0);
-    /* basically the same value as stck */
-    cpu_stq_data(env, a1 + 1, clock_value(env) | env->cpu_num);
-    /* more fine grained than stck */
-    cpu_stq_data(env, a1 + 9, 0);
-    /* XXX programmable fields */
-    cpu_stw_data(env, a1 + 17, 0);
-
-    return 0;
+    return clock_value(env);
 }
 
 /* Set Clock Comparator */
-void HELPER(sckc)(CPUS390XState *env, uint64_t a1)
+void HELPER(sckc)(CPUS390XState *env, uint64_t time)
 {
-    uint64_t time = cpu_ldq_data(env, a1);
-
     if (time == -1ULL) {
         return;
     }
@@ -187,17 +171,15 @@ void HELPER(sckc)(CPUS390XState *env, uint64_t a1)
 }
 
 /* Store Clock Comparator */
-void HELPER(stckc)(CPUS390XState *env, uint64_t a1)
+uint64_t HELPER(stckc)(CPUS390XState *env)
 {
     /* XXX implement */
-    cpu_stq_data(env, a1, 0);
+    return 0;
 }
 
 /* Set CPU Timer */
-void HELPER(spt)(CPUS390XState *env, uint64_t a1)
+void HELPER(spt)(CPUS390XState *env, uint64_t time)
 {
-    uint64_t time = cpu_ldq_data(env, a1);
-
     if (time == -1ULL) {
         return;
     }
@@ -209,15 +191,15 @@ void HELPER(spt)(CPUS390XState *env, uint64_t a1)
 }
 
 /* Store CPU Timer */
-void HELPER(stpt)(CPUS390XState *env, uint64_t a1)
+uint64_t HELPER(stpt)(CPUS390XState *env)
 {
     /* XXX implement */
-    cpu_stq_data(env, a1, 0);
+    return 0;
 }
 
 /* Store System Information */
-uint32_t HELPER(stsi)(CPUS390XState *env, uint64_t a0, uint32_t r0,
-                      uint32_t r1)
+uint32_t HELPER(stsi)(CPUS390XState *env, uint64_t a0,
+                      uint64_t r0, uint64_t r1)
 {
     int cc = 0;
     int sel1, sel2;
diff --git a/target-s390x/translate.c b/target-s390x/translate.c
index 9e34741311..a57296c64f 100644
--- a/target-s390x/translate.c
+++ b/target-s390x/translate.c
@@ -18,7 +18,6 @@
  * License along with this library; if not, see <http://www.gnu.org/licenses/>.
  */
 
-/* #define DEBUG_ILLEGAL_INSTRUCTIONS */
 /* #define DEBUG_INLINE_BRANCHES */
 #define S390X_DEBUG_DISAS
 /* #define S390X_DEBUG_DISAS_VERBOSE */
@@ -33,6 +32,7 @@
 #include "disas/disas.h"
 #include "tcg-op.h"
 #include "qemu/log.h"
+#include "qemu/host-utils.h"
 
 /* global register indexes */
 static TCGv_ptr cpu_env;
@@ -42,29 +42,41 @@ static TCGv_ptr cpu_env;
 #define GEN_HELPER 1
 #include "helper.h"
 
+
+/* Information that (most) every instruction needs to manipulate.  */
 typedef struct DisasContext DisasContext;
+typedef struct DisasInsn DisasInsn;
+typedef struct DisasFields DisasFields;
+
 struct DisasContext {
-    uint64_t pc;
-    int is_jmp;
-    enum cc_op cc_op;
     struct TranslationBlock *tb;
+    const DisasInsn *insn;
+    DisasFields *fields;
+    uint64_t pc, next_pc;
+    enum cc_op cc_op;
+    bool singlestep_enabled;
 };
 
-#define DISAS_EXCP 4
+/* Information carried about a condition to be evaluated.  */
+typedef struct {
+    TCGCond cond:8;
+    bool is_64;
+    bool g1;
+    bool g2;
+    union {
+        struct { TCGv_i64 a, b; } s64;
+        struct { TCGv_i32 a, b; } s32;
+    } u;
+} DisasCompare;
 
-static void gen_op_calc_cc(DisasContext *s);
+#define DISAS_EXCP 4
 
 #ifdef DEBUG_INLINE_BRANCHES
 static uint64_t inline_branch_hit[CC_OP_MAX];
 static uint64_t inline_branch_miss[CC_OP_MAX];
 #endif
 
-static inline void debug_insn(uint64_t insn)
-{
-    LOG_DISAS("insn: 0x%" PRIx64 "\n", insn);
-}
-
-static inline uint64_t pc_to_link_info(DisasContext *s, uint64_t pc)
+static uint64_t pc_to_link_info(DisasContext *s, uint64_t pc)
 {
     if (!(s->tb->flags & FLAG_MASK_64)) {
         if (s->tb->flags & FLAG_MASK_32) {
@@ -97,7 +109,7 @@ void cpu_dump_state(CPUS390XState *env, FILE *f, fprintf_function cpu_fprintf,
     }
 
     for (i = 0; i < 16; i++) {
-        cpu_fprintf(f, "F%02d=%016" PRIx64, i, *(uint64_t *)&env->fregs[i]);
+        cpu_fprintf(f, "F%02d=%016" PRIx64, i, env->fregs[i].ll);
         if ((i % 4) == 3) {
             cpu_fprintf(f, "\n");
         } else {
@@ -134,21 +146,22 @@ static TCGv_i64 cc_src;
 static TCGv_i64 cc_dst;
 static TCGv_i64 cc_vr;
 
-static char cpu_reg_names[10*3 + 6*4];
+static char cpu_reg_names[32][4];
 static TCGv_i64 regs[16];
+static TCGv_i64 fregs[16];
 
 static uint8_t gen_opc_cc_op[OPC_BUF_SIZE];
 
 void s390x_translate_init(void)
 {
     int i;
-    size_t cpu_reg_names_size = sizeof(cpu_reg_names);
-    char *p;
 
     cpu_env = tcg_global_reg_new_ptr(TCG_AREG0, "env");
-    psw_addr = tcg_global_mem_new_i64(TCG_AREG0, offsetof(CPUS390XState, psw.addr),
+    psw_addr = tcg_global_mem_new_i64(TCG_AREG0,
+                                      offsetof(CPUS390XState, psw.addr),
                                       "psw_addr");
-    psw_mask = tcg_global_mem_new_i64(TCG_AREG0, offsetof(CPUS390XState, psw.mask),
+    psw_mask = tcg_global_mem_new_i64(TCG_AREG0,
+                                      offsetof(CPUS390XState, psw.mask),
                                       "psw_mask");
 
     cc_op = tcg_global_mem_new_i32(TCG_AREG0, offsetof(CPUS390XState, cc_op),
@@ -160,116 +173,87 @@ void s390x_translate_init(void)
     cc_vr = tcg_global_mem_new_i64(TCG_AREG0, offsetof(CPUS390XState, cc_vr),
                                    "cc_vr");
 
-    p = cpu_reg_names;
     for (i = 0; i < 16; i++) {
-        snprintf(p, cpu_reg_names_size, "r%d", i);
+        snprintf(cpu_reg_names[i], sizeof(cpu_reg_names[0]), "r%d", i);
         regs[i] = tcg_global_mem_new(TCG_AREG0,
-                                     offsetof(CPUS390XState, regs[i]), p);
-        p += (i < 10) ? 3 : 4;
-        cpu_reg_names_size -= (i < 10) ? 3 : 4;
+                                     offsetof(CPUS390XState, regs[i]),
+                                     cpu_reg_names[i]);
     }
-}
 
-static inline TCGv_i64 load_reg(int reg)
-{
-    TCGv_i64 r = tcg_temp_new_i64();
-    tcg_gen_mov_i64(r, regs[reg]);
-    return r;
-}
-
-static inline TCGv_i64 load_freg(int reg)
-{
-    TCGv_i64 r = tcg_temp_new_i64();
-    tcg_gen_ld_i64(r, cpu_env, offsetof(CPUS390XState, fregs[reg].d));
-    return r;
-}
+    for (i = 0; i < 16; i++) {
+        snprintf(cpu_reg_names[i + 16], sizeof(cpu_reg_names[0]), "f%d", i);
+        fregs[i] = tcg_global_mem_new(TCG_AREG0,
+                                      offsetof(CPUS390XState, fregs[i].d),
+                                      cpu_reg_names[i + 16]);
+    }
 
-static inline TCGv_i32 load_freg32(int reg)
-{
-    TCGv_i32 r = tcg_temp_new_i32();
-    tcg_gen_ld_i32(r, cpu_env, offsetof(CPUS390XState, fregs[reg].l.upper));
-    return r;
+    /* register helpers */
+#define GEN_HELPER 2
+#include "helper.h"
 }
 
-static inline TCGv_i32 load_reg32(int reg)
+static TCGv_i64 load_reg(int reg)
 {
-    TCGv_i32 r = tcg_temp_new_i32();
-    tcg_gen_trunc_i64_i32(r, regs[reg]);
+    TCGv_i64 r = tcg_temp_new_i64();
+    tcg_gen_mov_i64(r, regs[reg]);
     return r;
 }
 
-static inline TCGv_i64 load_reg32_i64(int reg)
+static TCGv_i64 load_freg32_i64(int reg)
 {
     TCGv_i64 r = tcg_temp_new_i64();
-    tcg_gen_ext32s_i64(r, regs[reg]);
+    tcg_gen_shri_i64(r, fregs[reg], 32);
     return r;
 }
 
-static inline void store_reg(int reg, TCGv_i64 v)
+static void store_reg(int reg, TCGv_i64 v)
 {
     tcg_gen_mov_i64(regs[reg], v);
 }
 
-static inline void store_freg(int reg, TCGv_i64 v)
+static void store_freg(int reg, TCGv_i64 v)
 {
-    tcg_gen_st_i64(v, cpu_env, offsetof(CPUS390XState, fregs[reg].d));
+    tcg_gen_mov_i64(fregs[reg], v);
 }
 
-static inline void store_reg32(int reg, TCGv_i32 v)
+static void store_reg32_i64(int reg, TCGv_i64 v)
 {
-#if HOST_LONG_BITS == 32
-    tcg_gen_mov_i32(TCGV_LOW(regs[reg]), v);
-#else
-    TCGv_i64 tmp = tcg_temp_new_i64();
-    tcg_gen_extu_i32_i64(tmp, v);
     /* 32 bit register writes keep the upper half */
-    tcg_gen_deposit_i64(regs[reg], regs[reg], tmp, 0, 32);
-    tcg_temp_free_i64(tmp);
-#endif
-}
-
-static inline void store_reg32_i64(int reg, TCGv_i64 v)
-{
-    /* 32 bit register writes keep the upper half */
-#if HOST_LONG_BITS == 32
-    tcg_gen_mov_i32(TCGV_LOW(regs[reg]), TCGV_LOW(v));
-#else
     tcg_gen_deposit_i64(regs[reg], regs[reg], v, 0, 32);
-#endif
 }
 
-static inline void store_reg16(int reg, TCGv_i32 v)
+static void store_reg32h_i64(int reg, TCGv_i64 v)
 {
-    TCGv_i64 tmp = tcg_temp_new_i64();
-    tcg_gen_extu_i32_i64(tmp, v);
-    /* 16 bit register writes keep the upper bytes */
-    tcg_gen_deposit_i64(regs[reg], regs[reg], tmp, 0, 16);
-    tcg_temp_free_i64(tmp);
+    tcg_gen_deposit_i64(regs[reg], regs[reg], v, 32, 32);
 }
 
-static inline void store_reg8(int reg, TCGv_i64 v)
+static void store_freg32_i64(int reg, TCGv_i64 v)
 {
-    /* 8 bit register writes keep the upper bytes */
-    tcg_gen_deposit_i64(regs[reg], regs[reg], v, 0, 8);
+    tcg_gen_deposit_i64(fregs[reg], fregs[reg], v, 32, 32);
 }
 
-static inline void store_freg32(int reg, TCGv_i32 v)
+static void return_low128(TCGv_i64 dest)
 {
-    tcg_gen_st_i32(v, cpu_env, offsetof(CPUS390XState, fregs[reg].l.upper));
+    tcg_gen_ld_i64(dest, cpu_env, offsetof(CPUS390XState, retxl));
 }
 
-static inline void update_psw_addr(DisasContext *s)
+static void update_psw_addr(DisasContext *s)
 {
     /* psw.addr */
     tcg_gen_movi_i64(psw_addr, s->pc);
 }
 
-static inline void potential_page_fault(DisasContext *s)
+static void update_cc_op(DisasContext *s)
+{
+    if (s->cc_op != CC_OP_DYNAMIC && s->cc_op != CC_OP_STATIC) {
+        tcg_gen_movi_i32(cc_op, s->cc_op);
+    }
+}
+
+static void potential_page_fault(DisasContext *s)
 {
-#ifndef CONFIG_USER_ONLY
     update_psw_addr(s);
-    gen_op_calc_cc(s);
-#endif
+    update_cc_op(s);
 }
 
 static inline uint64_t ld_code2(CPUS390XState *env, uint64_t pc)
@@ -279,18 +263,15 @@ static inline uint64_t ld_code2(CPUS390XState *env, uint64_t pc)
 
 static inline uint64_t ld_code4(CPUS390XState *env, uint64_t pc)
 {
-    return (uint64_t)cpu_ldl_code(env, pc);
+    return (uint64_t)(uint32_t)cpu_ldl_code(env, pc);
 }
 
 static inline uint64_t ld_code6(CPUS390XState *env, uint64_t pc)
 {
-    uint64_t opc;
-    opc = (uint64_t)cpu_lduw_code(env, pc) << 32;
-    opc |= (uint64_t)(uint32_t)cpu_ldl_code(env, pc + 2);
-    return opc;
+    return (ld_code2(env, pc) << 32) | ld_code4(env, pc + 2);
 }
 
-static inline int get_mem_index(DisasContext *s)
+static int get_mem_index(DisasContext *s)
 {
     switch (s->tb->flags & FLAG_MASK_ASC) {
     case PSW_ASC_PRIMARY >> 32:
@@ -305,179 +286,114 @@ static inline int get_mem_index(DisasContext *s)
     }
 }
 
-static inline void gen_debug(DisasContext *s)
+static void gen_exception(int excp)
 {
-    TCGv_i32 tmp = tcg_const_i32(EXCP_DEBUG);
-    update_psw_addr(s);
-    gen_op_calc_cc(s);
+    TCGv_i32 tmp = tcg_const_i32(excp);
     gen_helper_exception(cpu_env, tmp);
     tcg_temp_free_i32(tmp);
-    s->is_jmp = DISAS_EXCP;
-}
-
-#ifdef CONFIG_USER_ONLY
-
-static void gen_illegal_opcode(CPUS390XState *env, DisasContext *s, int ilc)
-{
-    TCGv_i32 tmp = tcg_const_i32(EXCP_SPEC);
-    update_psw_addr(s);
-    gen_op_calc_cc(s);
-    gen_helper_exception(cpu_env, tmp);
-    tcg_temp_free_i32(tmp);
-    s->is_jmp = DISAS_EXCP;
-}
-
-#else /* CONFIG_USER_ONLY */
-
-static void debug_print_inst(CPUS390XState *env, DisasContext *s, int ilc)
-{
-#ifdef DEBUG_ILLEGAL_INSTRUCTIONS
-    uint64_t inst = 0;
-
-    switch (ilc & 3) {
-    case 1:
-        inst = ld_code2(env, s->pc);
-        break;
-    case 2:
-        inst = ld_code4(env, s->pc);
-        break;
-    case 3:
-        inst = ld_code6(env, s->pc);
-        break;
-    }
-
-    fprintf(stderr, "Illegal instruction [%d at %016" PRIx64 "]: 0x%016"
-            PRIx64 "\n", ilc, s->pc, inst);
-#endif
 }
 
-static void gen_program_exception(CPUS390XState *env, DisasContext *s, int ilc,
-                                  int code)
+static void gen_program_exception(DisasContext *s, int code)
 {
     TCGv_i32 tmp;
 
-    debug_print_inst(env, s, ilc);
-
-    /* remember what pgm exeption this was */
+    /* Remember what pgm exeption this was.  */
     tmp = tcg_const_i32(code);
     tcg_gen_st_i32(tmp, cpu_env, offsetof(CPUS390XState, int_pgm_code));
     tcg_temp_free_i32(tmp);
 
-    tmp = tcg_const_i32(ilc);
-    tcg_gen_st_i32(tmp, cpu_env, offsetof(CPUS390XState, int_pgm_ilc));
+    tmp = tcg_const_i32(s->next_pc - s->pc);
+    tcg_gen_st_i32(tmp, cpu_env, offsetof(CPUS390XState, int_pgm_ilen));
     tcg_temp_free_i32(tmp);
 
-    /* advance past instruction */
-    s->pc += (ilc * 2);
+    /* Advance past instruction.  */
+    s->pc = s->next_pc;
     update_psw_addr(s);
 
-    /* save off cc */
-    gen_op_calc_cc(s);
-
-    /* trigger exception */
-    tmp = tcg_const_i32(EXCP_PGM);
-    gen_helper_exception(cpu_env, tmp);
-    tcg_temp_free_i32(tmp);
+    /* Save off cc.  */
+    update_cc_op(s);
 
-    /* end TB here */
-    s->is_jmp = DISAS_EXCP;
-}
-
-
-static void gen_illegal_opcode(CPUS390XState *env, DisasContext *s, int ilc)
-{
-    gen_program_exception(env, s, ilc, PGM_SPECIFICATION);
+    /* Trigger exception.  */
+    gen_exception(EXCP_PGM);
 }
 
-static void gen_privileged_exception(CPUS390XState *env, DisasContext *s,
-                                     int ilc)
+static inline void gen_illegal_opcode(DisasContext *s)
 {
-    gen_program_exception(env, s, ilc, PGM_PRIVILEGED);
+    gen_program_exception(s, PGM_SPECIFICATION);
 }
 
-static void check_privileged(CPUS390XState *env, DisasContext *s, int ilc)
+static inline void check_privileged(DisasContext *s)
 {
     if (s->tb->flags & (PSW_MASK_PSTATE >> 32)) {
-        gen_privileged_exception(env, s, ilc);
+        gen_program_exception(s, PGM_PRIVILEGED);
     }
 }
 
-#endif /* CONFIG_USER_ONLY */
-
 static TCGv_i64 get_address(DisasContext *s, int x2, int b2, int d2)
 {
-    TCGv_i64 tmp;
+    TCGv_i64 tmp = tcg_temp_new_i64();
+    bool need_31 = !(s->tb->flags & FLAG_MASK_64);
 
-    /* 31-bitify the immediate part; register contents are dealt with below */
-    if (!(s->tb->flags & FLAG_MASK_64)) {
-        d2 &= 0x7fffffffUL;
-    }
+    /* Note that d2 is limited to 20 bits, signed.  If we crop negative
+       displacements early we create larger immedate addends.  */
 
-    if (x2) {
-        if (d2) {
-            tmp = tcg_const_i64(d2);
-            tcg_gen_add_i64(tmp, tmp, regs[x2]);
-        } else {
-            tmp = load_reg(x2);
-        }
-        if (b2) {
-            tcg_gen_add_i64(tmp, tmp, regs[b2]);
-        }
+    /* Note that addi optimizes the imm==0 case.  */
+    if (b2 && x2) {
+        tcg_gen_add_i64(tmp, regs[b2], regs[x2]);
+        tcg_gen_addi_i64(tmp, tmp, d2);
     } else if (b2) {
-        if (d2) {
-            tmp = tcg_const_i64(d2);
-            tcg_gen_add_i64(tmp, tmp, regs[b2]);
-        } else {
-            tmp = load_reg(b2);
-        }
+        tcg_gen_addi_i64(tmp, regs[b2], d2);
+    } else if (x2) {
+        tcg_gen_addi_i64(tmp, regs[x2], d2);
     } else {
-        tmp = tcg_const_i64(d2);
+        if (need_31) {
+            d2 &= 0x7fffffff;
+            need_31 = false;
+        }
+        tcg_gen_movi_i64(tmp, d2);
     }
-
-    /* 31-bit mode mask if there are values loaded from registers */
-    if (!(s->tb->flags & FLAG_MASK_64) && (x2 || b2)) {
-        tcg_gen_andi_i64(tmp, tmp, 0x7fffffffUL);
+    if (need_31) {
+        tcg_gen_andi_i64(tmp, tmp, 0x7fffffff);
     }
 
     return tmp;
 }
 
-static void gen_op_movi_cc(DisasContext *s, uint32_t val)
+static inline bool live_cc_data(DisasContext *s)
 {
-    s->cc_op = CC_OP_CONST0 + val;
+    return (s->cc_op != CC_OP_DYNAMIC
+            && s->cc_op != CC_OP_STATIC
+            && s->cc_op > 3);
 }
 
-static void gen_op_update1_cc_i64(DisasContext *s, enum cc_op op, TCGv_i64 dst)
+static inline void gen_op_movi_cc(DisasContext *s, uint32_t val)
 {
-    tcg_gen_discard_i64(cc_src);
-    tcg_gen_mov_i64(cc_dst, dst);
-    tcg_gen_discard_i64(cc_vr);
-    s->cc_op = op;
+    if (live_cc_data(s)) {
+        tcg_gen_discard_i64(cc_src);
+        tcg_gen_discard_i64(cc_dst);
+        tcg_gen_discard_i64(cc_vr);
+    }
+    s->cc_op = CC_OP_CONST0 + val;
 }
 
-static void gen_op_update1_cc_i32(DisasContext *s, enum cc_op op, TCGv_i32 dst)
+static void gen_op_update1_cc_i64(DisasContext *s, enum cc_op op, TCGv_i64 dst)
 {
-    tcg_gen_discard_i64(cc_src);
-    tcg_gen_extu_i32_i64(cc_dst, dst);
-    tcg_gen_discard_i64(cc_vr);
+    if (live_cc_data(s)) {
+        tcg_gen_discard_i64(cc_src);
+        tcg_gen_discard_i64(cc_vr);
+    }
+    tcg_gen_mov_i64(cc_dst, dst);
     s->cc_op = op;
 }
 
 static void gen_op_update2_cc_i64(DisasContext *s, enum cc_op op, TCGv_i64 src,
                                   TCGv_i64 dst)
 {
+    if (live_cc_data(s)) {
+        tcg_gen_discard_i64(cc_vr);
+    }
     tcg_gen_mov_i64(cc_src, src);
     tcg_gen_mov_i64(cc_dst, dst);
-    tcg_gen_discard_i64(cc_vr);
-    s->cc_op = op;
-}
-
-static void gen_op_update2_cc_i32(DisasContext *s, enum cc_op op, TCGv_i32 src,
-                                  TCGv_i32 dst)
-{
-    tcg_gen_extu_i32_i64(cc_src, src);
-    tcg_gen_extu_i32_i64(cc_dst, dst);
-    tcg_gen_discard_i64(cc_vr);
     s->cc_op = op;
 }
 
@@ -490,214 +406,71 @@ static void gen_op_update3_cc_i64(DisasContext *s, enum cc_op op, TCGv_i64 src,
     s->cc_op = op;
 }
 
-static void gen_op_update3_cc_i32(DisasContext *s, enum cc_op op, TCGv_i32 src,
-                                  TCGv_i32 dst, TCGv_i32 vr)
-{
-    tcg_gen_extu_i32_i64(cc_src, src);
-    tcg_gen_extu_i32_i64(cc_dst, dst);
-    tcg_gen_extu_i32_i64(cc_vr, vr);
-    s->cc_op = op;
-}
-
-static inline void set_cc_nz_u32(DisasContext *s, TCGv_i32 val)
-{
-    gen_op_update1_cc_i32(s, CC_OP_NZ, val);
-}
-
-static inline void set_cc_nz_u64(DisasContext *s, TCGv_i64 val)
+static void set_cc_nz_u64(DisasContext *s, TCGv_i64 val)
 {
     gen_op_update1_cc_i64(s, CC_OP_NZ, val);
 }
 
-static inline void cmp_32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2,
-                          enum cc_op cond)
-{
-    gen_op_update2_cc_i32(s, cond, v1, v2);
-}
-
-static inline void cmp_64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2,
-                          enum cc_op cond)
-{
-    gen_op_update2_cc_i64(s, cond, v1, v2);
-}
-
-static inline void cmp_s32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2)
-{
-    cmp_32(s, v1, v2, CC_OP_LTGT_32);
-}
-
-static inline void cmp_u32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2)
-{
-    cmp_32(s, v1, v2, CC_OP_LTUGTU_32);
-}
-
-static inline void cmp_s32c(DisasContext *s, TCGv_i32 v1, int32_t v2)
-{
-    /* XXX optimize for the constant? put it in s? */
-    TCGv_i32 tmp = tcg_const_i32(v2);
-    cmp_32(s, v1, tmp, CC_OP_LTGT_32);
-    tcg_temp_free_i32(tmp);
-}
-
-static inline void cmp_u32c(DisasContext *s, TCGv_i32 v1, uint32_t v2)
-{
-    TCGv_i32 tmp = tcg_const_i32(v2);
-    cmp_32(s, v1, tmp, CC_OP_LTUGTU_32);
-    tcg_temp_free_i32(tmp);
-}
-
-static inline void cmp_s64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2)
-{
-    cmp_64(s, v1, v2, CC_OP_LTGT_64);
-}
-
-static inline void cmp_u64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2)
-{
-    cmp_64(s, v1, v2, CC_OP_LTUGTU_64);
-}
-
-static inline void cmp_s64c(DisasContext *s, TCGv_i64 v1, int64_t v2)
-{
-    TCGv_i64 tmp = tcg_const_i64(v2);
-    cmp_s64(s, v1, tmp);
-    tcg_temp_free_i64(tmp);
-}
-
-static inline void cmp_u64c(DisasContext *s, TCGv_i64 v1, uint64_t v2)
-{
-    TCGv_i64 tmp = tcg_const_i64(v2);
-    cmp_u64(s, v1, tmp);
-    tcg_temp_free_i64(tmp);
-}
-
-static inline void set_cc_s32(DisasContext *s, TCGv_i32 val)
-{
-    gen_op_update1_cc_i32(s, CC_OP_LTGT0_32, val);
-}
-
-static inline void set_cc_s64(DisasContext *s, TCGv_i64 val)
-{
-    gen_op_update1_cc_i64(s, CC_OP_LTGT0_64, val);
-}
-
-static void set_cc_add64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2, TCGv_i64 vr)
-{
-    gen_op_update3_cc_i64(s, CC_OP_ADD_64, v1, v2, vr);
-}
-
-static void set_cc_addu64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2,
-                          TCGv_i64 vr)
-{
-    gen_op_update3_cc_i64(s, CC_OP_ADDU_64, v1, v2, vr);
-}
-
-static void set_cc_sub64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2, TCGv_i64 vr)
-{
-    gen_op_update3_cc_i64(s, CC_OP_SUB_64, v1, v2, vr);
-}
-
-static void set_cc_subu64(DisasContext *s, TCGv_i64 v1, TCGv_i64 v2,
-                          TCGv_i64 vr)
-{
-    gen_op_update3_cc_i64(s, CC_OP_SUBU_64, v1, v2, vr);
-}
-
-static void set_cc_abs64(DisasContext *s, TCGv_i64 v1)
-{
-    gen_op_update1_cc_i64(s, CC_OP_ABS_64, v1);
-}
-
-static void set_cc_nabs64(DisasContext *s, TCGv_i64 v1)
-{
-    gen_op_update1_cc_i64(s, CC_OP_NABS_64, v1);
-}
-
-static void set_cc_add32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2, TCGv_i32 vr)
-{
-    gen_op_update3_cc_i32(s, CC_OP_ADD_32, v1, v2, vr);
-}
-
-static void set_cc_addu32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2,
-                          TCGv_i32 vr)
+static void gen_set_cc_nz_f32(DisasContext *s, TCGv_i64 val)
 {
-    gen_op_update3_cc_i32(s, CC_OP_ADDU_32, v1, v2, vr);
+    gen_op_update1_cc_i64(s, CC_OP_NZ_F32, val);
 }
 
-static void set_cc_sub32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2, TCGv_i32 vr)
+static void gen_set_cc_nz_f64(DisasContext *s, TCGv_i64 val)
 {
-    gen_op_update3_cc_i32(s, CC_OP_SUB_32, v1, v2, vr);
+    gen_op_update1_cc_i64(s, CC_OP_NZ_F64, val);
 }
 
-static void set_cc_subu32(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2,
-                          TCGv_i32 vr)
+static void gen_set_cc_nz_f128(DisasContext *s, TCGv_i64 vh, TCGv_i64 vl)
 {
-    gen_op_update3_cc_i32(s, CC_OP_SUBU_32, v1, v2, vr);
-}
-
-static void set_cc_abs32(DisasContext *s, TCGv_i32 v1)
-{
-    gen_op_update1_cc_i32(s, CC_OP_ABS_32, v1);
-}
-
-static void set_cc_nabs32(DisasContext *s, TCGv_i32 v1)
-{
-    gen_op_update1_cc_i32(s, CC_OP_NABS_32, v1);
-}
-
-static void set_cc_comp32(DisasContext *s, TCGv_i32 v1)
-{
-    gen_op_update1_cc_i32(s, CC_OP_COMP_32, v1);
-}
-
-static void set_cc_comp64(DisasContext *s, TCGv_i64 v1)
-{
-    gen_op_update1_cc_i64(s, CC_OP_COMP_64, v1);
-}
-
-static void set_cc_icm(DisasContext *s, TCGv_i32 v1, TCGv_i32 v2)
-{
-    gen_op_update2_cc_i32(s, CC_OP_ICM, v1, v2);
-}
-
-static void set_cc_cmp_f32_i64(DisasContext *s, TCGv_i32 v1, TCGv_i64 v2)
-{
-    tcg_gen_extu_i32_i64(cc_src, v1);
-    tcg_gen_mov_i64(cc_dst, v2);
-    tcg_gen_discard_i64(cc_vr);
-    s->cc_op = CC_OP_LTGT_F32;
-}
-
-static void gen_set_cc_nz_f32(DisasContext *s, TCGv_i32 v1)
-{
-    gen_op_update1_cc_i32(s, CC_OP_NZ_F32, v1);
+    gen_op_update2_cc_i64(s, CC_OP_NZ_F128, vh, vl);
 }
 
 /* CC value is in env->cc_op */
-static inline void set_cc_static(DisasContext *s)
+static void set_cc_static(DisasContext *s)
 {
-    tcg_gen_discard_i64(cc_src);
-    tcg_gen_discard_i64(cc_dst);
-    tcg_gen_discard_i64(cc_vr);
-    s->cc_op = CC_OP_STATIC;
-}
-
-static inline void gen_op_set_cc_op(DisasContext *s)
-{
-    if (s->cc_op != CC_OP_DYNAMIC && s->cc_op != CC_OP_STATIC) {
-        tcg_gen_movi_i32(cc_op, s->cc_op);
+    if (live_cc_data(s)) {
+        tcg_gen_discard_i64(cc_src);
+        tcg_gen_discard_i64(cc_dst);
+        tcg_gen_discard_i64(cc_vr);
     }
-}
-
-static inline void gen_update_cc_op(DisasContext *s)
-{
-    gen_op_set_cc_op(s);
+    s->cc_op = CC_OP_STATIC;
 }
 
 /* calculates cc into cc_op */
 static void gen_op_calc_cc(DisasContext *s)
 {
-    TCGv_i32 local_cc_op = tcg_const_i32(s->cc_op);
-    TCGv_i64 dummy = tcg_const_i64(0);
+    TCGv_i32 local_cc_op;
+    TCGv_i64 dummy;
+
+    TCGV_UNUSED_I32(local_cc_op);
+    TCGV_UNUSED_I64(dummy);
+    switch (s->cc_op) {
+    default:
+        dummy = tcg_const_i64(0);
+        /* FALLTHRU */
+    case CC_OP_ADD_64:
+    case CC_OP_ADDU_64:
+    case CC_OP_ADDC_64:
+    case CC_OP_SUB_64:
+    case CC_OP_SUBU_64:
+    case CC_OP_SUBB_64:
+    case CC_OP_ADD_32:
+    case CC_OP_ADDU_32:
+    case CC_OP_ADDC_32:
+    case CC_OP_SUB_32:
+    case CC_OP_SUBU_32:
+    case CC_OP_SUBB_32:
+        local_cc_op = tcg_const_i32(s->cc_op);
+        break;
+    case CC_OP_CONST0:
+    case CC_OP_CONST1:
+    case CC_OP_CONST2:
+    case CC_OP_CONST3:
+    case CC_OP_STATIC:
+    case CC_OP_DYNAMIC:
+        break;
+    }
 
     switch (s->cc_op) {
     case CC_OP_CONST0:
@@ -721,6 +494,7 @@ static void gen_op_calc_cc(DisasContext *s)
     case CC_OP_COMP_64:
     case CC_OP_NZ_F32:
     case CC_OP_NZ_F64:
+    case CC_OP_FLOGR:
         /* 1 argument */
         gen_helper_calc_cc(cc_op, cpu_env, local_cc_op, dummy, cc_dst, dummy);
         break;
@@ -731,20 +505,24 @@ static void gen_op_calc_cc(DisasContext *s)
     case CC_OP_LTUGTU_64:
     case CC_OP_TM_32:
     case CC_OP_TM_64:
-    case CC_OP_LTGT_F32:
-    case CC_OP_LTGT_F64:
-    case CC_OP_SLAG:
+    case CC_OP_SLA_32:
+    case CC_OP_SLA_64:
+    case CC_OP_NZ_F128:
         /* 2 arguments */
         gen_helper_calc_cc(cc_op, cpu_env, local_cc_op, cc_src, cc_dst, dummy);
         break;
     case CC_OP_ADD_64:
     case CC_OP_ADDU_64:
+    case CC_OP_ADDC_64:
     case CC_OP_SUB_64:
     case CC_OP_SUBU_64:
+    case CC_OP_SUBB_64:
     case CC_OP_ADD_32:
     case CC_OP_ADDU_32:
+    case CC_OP_ADDC_32:
     case CC_OP_SUB_32:
     case CC_OP_SUBU_32:
+    case CC_OP_SUBB_32:
         /* 3 arguments */
         gen_helper_calc_cc(cc_op, cpu_env, local_cc_op, cc_src, cc_dst, cc_vr);
         break;
@@ -756,4358 +534,4204 @@ static void gen_op_calc_cc(DisasContext *s)
         tcg_abort();
     }
 
-    tcg_temp_free_i32(local_cc_op);
+    if (!TCGV_IS_UNUSED_I32(local_cc_op)) {
+        tcg_temp_free_i32(local_cc_op);
+    }
+    if (!TCGV_IS_UNUSED_I64(dummy)) {
+        tcg_temp_free_i64(dummy);
+    }
 
     /* We now have cc in cc_op as constant */
     set_cc_static(s);
 }
 
-static inline void decode_rr(DisasContext *s, uint64_t insn, int *r1, int *r2)
-{
-    debug_insn(insn);
-
-    *r1 = (insn >> 4) & 0xf;
-    *r2 = insn & 0xf;
-}
-
-static inline TCGv_i64 decode_rx(DisasContext *s, uint64_t insn, int *r1,
-                                 int *x2, int *b2, int *d2)
+static int use_goto_tb(DisasContext *s, uint64_t dest)
 {
-    debug_insn(insn);
-
-    *r1 = (insn >> 20) & 0xf;
-    *x2 = (insn >> 16) & 0xf;
-    *b2 = (insn >> 12) & 0xf;
-    *d2 = insn & 0xfff;
-
-    return get_address(s, *x2, *b2, *d2);
-}
-
-static inline void decode_rs(DisasContext *s, uint64_t insn, int *r1, int *r3,
-                             int *b2, int *d2)
-{
-    debug_insn(insn);
-
-    *r1 = (insn >> 20) & 0xf;
-    /* aka m3 */
-    *r3 = (insn >> 16) & 0xf;
-    *b2 = (insn >> 12) & 0xf;
-    *d2 = insn & 0xfff;
-}
-
-static inline TCGv_i64 decode_si(DisasContext *s, uint64_t insn, int *i2,
-                                 int *b1, int *d1)
-{
-    debug_insn(insn);
-
-    *i2 = (insn >> 16) & 0xff;
-    *b1 = (insn >> 12) & 0xf;
-    *d1 = insn & 0xfff;
-
-    return get_address(s, 0, *b1, *d1);
-}
-
-static inline void gen_goto_tb(DisasContext *s, int tb_num, target_ulong pc)
-{
-    TranslationBlock *tb;
-
-    gen_update_cc_op(s);
-
-    tb = s->tb;
     /* NOTE: we handle the case where the TB spans two pages here */
-    if ((pc & TARGET_PAGE_MASK) == (tb->pc & TARGET_PAGE_MASK) ||
-        (pc & TARGET_PAGE_MASK) == ((s->pc - 1) & TARGET_PAGE_MASK))  {
-        /* jump to same page: we can use a direct jump */
-        tcg_gen_goto_tb(tb_num);
-        tcg_gen_movi_i64(psw_addr, pc);
-        tcg_gen_exit_tb((tcg_target_long)tb + tb_num);
-    } else {
-        /* jump to another page: currently not optimized */
-        tcg_gen_movi_i64(psw_addr, pc);
-        tcg_gen_exit_tb(0);
-    }
+    return (((dest & TARGET_PAGE_MASK) == (s->tb->pc & TARGET_PAGE_MASK)
+             || (dest & TARGET_PAGE_MASK) == ((s->pc - 1) & TARGET_PAGE_MASK))
+            && !s->singlestep_enabled
+            && !(s->tb->cflags & CF_LAST_IO));
 }
 
-static inline void account_noninline_branch(DisasContext *s, int cc_op)
+static void account_noninline_branch(DisasContext *s, int cc_op)
 {
 #ifdef DEBUG_INLINE_BRANCHES
     inline_branch_miss[cc_op]++;
 #endif
 }
 
-static inline void account_inline_branch(DisasContext *s)
+static void account_inline_branch(DisasContext *s, int cc_op)
 {
 #ifdef DEBUG_INLINE_BRANCHES
-    inline_branch_hit[s->cc_op]++;
+    inline_branch_hit[cc_op]++;
 #endif
 }
 
-static void gen_jcc(DisasContext *s, uint32_t mask, int skip)
+/* Table of mask values to comparison codes, given a comparison as input.
+   For such, CC=3 should not be possible.  */
+static const TCGCond ltgt_cond[16] = {
+    TCG_COND_NEVER,  TCG_COND_NEVER,     /*    |    |    | x */
+    TCG_COND_GT,     TCG_COND_GT,        /*    |    | GT | x */
+    TCG_COND_LT,     TCG_COND_LT,        /*    | LT |    | x */
+    TCG_COND_NE,     TCG_COND_NE,        /*    | LT | GT | x */
+    TCG_COND_EQ,     TCG_COND_EQ,        /* EQ |    |    | x */
+    TCG_COND_GE,     TCG_COND_GE,        /* EQ |    | GT | x */
+    TCG_COND_LE,     TCG_COND_LE,        /* EQ | LT |    | x */
+    TCG_COND_ALWAYS, TCG_COND_ALWAYS,    /* EQ | LT | GT | x */
+};
+
+/* Table of mask values to comparison codes, given a logic op as input.
+   For such, only CC=0 and CC=1 should be possible.  */
+static const TCGCond nz_cond[16] = {
+    TCG_COND_NEVER, TCG_COND_NEVER,      /*    |    | x | x */
+    TCG_COND_NEVER, TCG_COND_NEVER,
+    TCG_COND_NE, TCG_COND_NE,            /*    | NE | x | x */
+    TCG_COND_NE, TCG_COND_NE,
+    TCG_COND_EQ, TCG_COND_EQ,            /* EQ |    | x | x */
+    TCG_COND_EQ, TCG_COND_EQ,
+    TCG_COND_ALWAYS, TCG_COND_ALWAYS,    /* EQ | NE | x | x */
+    TCG_COND_ALWAYS, TCG_COND_ALWAYS,
+};
+
+/* Interpret MASK in terms of S->CC_OP, and fill in C with all the
+   details required to generate a TCG comparison.  */
+static void disas_jcc(DisasContext *s, DisasCompare *c, uint32_t mask)
 {
-    TCGv_i32 tmp, tmp2, r;
-    TCGv_i64 tmp64;
-    int old_cc_op;
+    TCGCond cond;
+    enum cc_op old_cc_op = s->cc_op;
+
+    if (mask == 15 || mask == 0) {
+        c->cond = (mask ? TCG_COND_ALWAYS : TCG_COND_NEVER);
+        c->u.s32.a = cc_op;
+        c->u.s32.b = cc_op;
+        c->g1 = c->g2 = true;
+        c->is_64 = false;
+        return;
+    }
 
-    switch (s->cc_op) {
+    /* Find the TCG condition for the mask + cc op.  */
+    switch (old_cc_op) {
     case CC_OP_LTGT0_32:
-        tmp = tcg_temp_new_i32();
-        tcg_gen_trunc_i64_i32(tmp, cc_dst);
-        switch (mask) {
-        case 0x8 | 0x4: /* dst <= 0 */
-            tcg_gen_brcondi_i32(TCG_COND_GT, tmp, 0, skip);
-            break;
-        case 0x8 | 0x2: /* dst >= 0 */
-            tcg_gen_brcondi_i32(TCG_COND_LT, tmp, 0, skip);
-            break;
-        case 0x8: /* dst == 0 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, tmp, 0, skip);
-            break;
-        case 0x7: /* dst != 0 */
-        case 0x6: /* dst != 0 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, tmp, 0, skip);
-            break;
-        case 0x4: /* dst < 0 */
-            tcg_gen_brcondi_i32(TCG_COND_GE, tmp, 0, skip);
-            break;
-        case 0x2: /* dst > 0 */
-            tcg_gen_brcondi_i32(TCG_COND_LE, tmp, 0, skip);
-            break;
-        default:
-            tcg_temp_free_i32(tmp);
-            goto do_dynamic;
-        }
-        account_inline_branch(s);
-        tcg_temp_free_i32(tmp);
-        break;
     case CC_OP_LTGT0_64:
-        switch (mask) {
-        case 0x8 | 0x4: /* dst <= 0 */
-            tcg_gen_brcondi_i64(TCG_COND_GT, cc_dst, 0, skip);
-            break;
-        case 0x8 | 0x2: /* dst >= 0 */
-            tcg_gen_brcondi_i64(TCG_COND_LT, cc_dst, 0, skip);
-            break;
-        case 0x8: /* dst == 0 */
-            tcg_gen_brcondi_i64(TCG_COND_NE, cc_dst, 0, skip);
-            break;
-        case 0x7: /* dst != 0 */
-        case 0x6: /* dst != 0 */
-            tcg_gen_brcondi_i64(TCG_COND_EQ, cc_dst, 0, skip);
-            break;
-        case 0x4: /* dst < 0 */
-            tcg_gen_brcondi_i64(TCG_COND_GE, cc_dst, 0, skip);
-            break;
-        case 0x2: /* dst > 0 */
-            tcg_gen_brcondi_i64(TCG_COND_LE, cc_dst, 0, skip);
-            break;
-        default:
+    case CC_OP_LTGT_32:
+    case CC_OP_LTGT_64:
+        cond = ltgt_cond[mask];
+        if (cond == TCG_COND_NEVER) {
             goto do_dynamic;
         }
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_LTGT_32:
-        tmp = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i32();
-        tcg_gen_trunc_i64_i32(tmp, cc_src);
-        tcg_gen_trunc_i64_i32(tmp2, cc_dst);
-        switch (mask) {
-        case 0x8 | 0x4: /* src <= dst */
-            tcg_gen_brcond_i32(TCG_COND_GT, tmp, tmp2, skip);
-            break;
-        case 0x8 | 0x2: /* src >= dst */
-            tcg_gen_brcond_i32(TCG_COND_LT, tmp, tmp2, skip);
-            break;
-        case 0x8: /* src == dst */
-            tcg_gen_brcond_i32(TCG_COND_NE, tmp, tmp2, skip);
-            break;
-        case 0x7: /* src != dst */
-        case 0x6: /* src != dst */
-            tcg_gen_brcond_i32(TCG_COND_EQ, tmp, tmp2, skip);
-            break;
-        case 0x4: /* src < dst */
-            tcg_gen_brcond_i32(TCG_COND_GE, tmp, tmp2, skip);
-            break;
-        case 0x2: /* src > dst */
-            tcg_gen_brcond_i32(TCG_COND_LE, tmp, tmp2, skip);
-            break;
-        default:
-            tcg_temp_free_i32(tmp);
-            tcg_temp_free_i32(tmp2);
+
+    case CC_OP_LTUGTU_32:
+    case CC_OP_LTUGTU_64:
+        cond = tcg_unsigned_cond(ltgt_cond[mask]);
+        if (cond == TCG_COND_NEVER) {
             goto do_dynamic;
         }
-        account_inline_branch(s);
-        tcg_temp_free_i32(tmp);
-        tcg_temp_free_i32(tmp2);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_LTGT_64:
-        switch (mask) {
-        case 0x8 | 0x4: /* src <= dst */
-            tcg_gen_brcond_i64(TCG_COND_GT, cc_src, cc_dst, skip);
-            break;
-        case 0x8 | 0x2: /* src >= dst */
-            tcg_gen_brcond_i64(TCG_COND_LT, cc_src, cc_dst, skip);
-            break;
-        case 0x8: /* src == dst */
-            tcg_gen_brcond_i64(TCG_COND_NE, cc_src, cc_dst, skip);
-            break;
-        case 0x7: /* src != dst */
-        case 0x6: /* src != dst */
-            tcg_gen_brcond_i64(TCG_COND_EQ, cc_src, cc_dst, skip);
-            break;
-        case 0x4: /* src < dst */
-            tcg_gen_brcond_i64(TCG_COND_GE, cc_src, cc_dst, skip);
-            break;
-        case 0x2: /* src > dst */
-            tcg_gen_brcond_i64(TCG_COND_LE, cc_src, cc_dst, skip);
-            break;
-        default:
+
+    case CC_OP_NZ:
+        cond = nz_cond[mask];
+        if (cond == TCG_COND_NEVER) {
             goto do_dynamic;
         }
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_LTUGTU_32:
-        tmp = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i32();
-        tcg_gen_trunc_i64_i32(tmp, cc_src);
-        tcg_gen_trunc_i64_i32(tmp2, cc_dst);
+
+    case CC_OP_TM_32:
+    case CC_OP_TM_64:
         switch (mask) {
-        case 0x8 | 0x4: /* src <= dst */
-            tcg_gen_brcond_i32(TCG_COND_GTU, tmp, tmp2, skip);
-            break;
-        case 0x8 | 0x2: /* src >= dst */
-            tcg_gen_brcond_i32(TCG_COND_LTU, tmp, tmp2, skip);
-            break;
-        case 0x8: /* src == dst */
-            tcg_gen_brcond_i32(TCG_COND_NE, tmp, tmp2, skip);
-            break;
-        case 0x7: /* src != dst */
-        case 0x6: /* src != dst */
-            tcg_gen_brcond_i32(TCG_COND_EQ, tmp, tmp2, skip);
+        case 8:
+            cond = TCG_COND_EQ;
             break;
-        case 0x4: /* src < dst */
-            tcg_gen_brcond_i32(TCG_COND_GEU, tmp, tmp2, skip);
-            break;
-        case 0x2: /* src > dst */
-            tcg_gen_brcond_i32(TCG_COND_LEU, tmp, tmp2, skip);
+        case 4 | 2 | 1:
+            cond = TCG_COND_NE;
             break;
         default:
-            tcg_temp_free_i32(tmp);
-            tcg_temp_free_i32(tmp2);
             goto do_dynamic;
         }
-        account_inline_branch(s);
-        tcg_temp_free_i32(tmp);
-        tcg_temp_free_i32(tmp2);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_LTUGTU_64:
+
+    case CC_OP_ICM:
         switch (mask) {
-        case 0x8 | 0x4: /* src <= dst */
-            tcg_gen_brcond_i64(TCG_COND_GTU, cc_src, cc_dst, skip);
+        case 8:
+            cond = TCG_COND_EQ;
             break;
-        case 0x8 | 0x2: /* src >= dst */
-            tcg_gen_brcond_i64(TCG_COND_LTU, cc_src, cc_dst, skip);
-            break;
-        case 0x8: /* src == dst */
-            tcg_gen_brcond_i64(TCG_COND_NE, cc_src, cc_dst, skip);
-            break;
-        case 0x7: /* src != dst */
-        case 0x6: /* src != dst */
-            tcg_gen_brcond_i64(TCG_COND_EQ, cc_src, cc_dst, skip);
-            break;
-        case 0x4: /* src < dst */
-            tcg_gen_brcond_i64(TCG_COND_GEU, cc_src, cc_dst, skip);
-            break;
-        case 0x2: /* src > dst */
-            tcg_gen_brcond_i64(TCG_COND_LEU, cc_src, cc_dst, skip);
+        case 4 | 2 | 1:
+        case 4 | 2:
+            cond = TCG_COND_NE;
             break;
         default:
             goto do_dynamic;
         }
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_NZ:
-        switch (mask) {
-        /* dst == 0 || dst != 0 */
-        case 0x8 | 0x4:
-        case 0x8 | 0x4 | 0x2:
-        case 0x8 | 0x4 | 0x2 | 0x1:
-        case 0x8 | 0x4 | 0x1:
-            break;
-        /* dst == 0 */
-        case 0x8:
-        case 0x8 | 0x2:
-        case 0x8 | 0x2 | 0x1:
-        case 0x8 | 0x1:
-            tcg_gen_brcondi_i64(TCG_COND_NE, cc_dst, 0, skip);
+
+    case CC_OP_FLOGR:
+        switch (mask & 0xa) {
+        case 8: /* src == 0 -> no one bit found */
+            cond = TCG_COND_EQ;
             break;
-        /* dst != 0 */
-        case 0x4:
-        case 0x4 | 0x2:
-        case 0x4 | 0x2 | 0x1:
-        case 0x4 | 0x1:
-            tcg_gen_brcondi_i64(TCG_COND_EQ, cc_dst, 0, skip);
+        case 2: /* src != 0 -> one bit found */
+            cond = TCG_COND_NE;
             break;
         default:
             goto do_dynamic;
         }
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_TM_32:
-        tmp = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i32();
 
-        tcg_gen_trunc_i64_i32(tmp, cc_src);
-        tcg_gen_trunc_i64_i32(tmp2, cc_dst);
-        tcg_gen_and_i32(tmp, tmp, tmp2);
+    case CC_OP_ADDU_32:
+    case CC_OP_ADDU_64:
         switch (mask) {
-        case 0x8: /* val & mask == 0 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, tmp, 0, skip);
+        case 8 | 2: /* vr == 0 */
+            cond = TCG_COND_EQ;
+            break;
+        case 4 | 1: /* vr != 0 */
+            cond = TCG_COND_NE;
             break;
-        case 0x4 | 0x2 | 0x1: /* val & mask != 0 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, tmp, 0, skip);
+        case 8 | 4: /* no carry -> vr >= src */
+            cond = TCG_COND_GEU;
+            break;
+        case 2 | 1: /* carry -> vr < src */
+            cond = TCG_COND_LTU;
             break;
         default:
-            tcg_temp_free_i32(tmp);
-            tcg_temp_free_i32(tmp2);
             goto do_dynamic;
         }
-        tcg_temp_free_i32(tmp);
-        tcg_temp_free_i32(tmp2);
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_TM_64:
-        tmp64 = tcg_temp_new_i64();
 
-        tcg_gen_and_i64(tmp64, cc_src, cc_dst);
-        switch (mask) {
-        case 0x8: /* val & mask == 0 */
-            tcg_gen_brcondi_i64(TCG_COND_NE, tmp64, 0, skip);
+    case CC_OP_SUBU_32:
+    case CC_OP_SUBU_64:
+        /* Note that CC=0 is impossible; treat it as dont-care.  */
+        switch (mask & 7) {
+        case 2: /* zero -> op1 == op2 */
+            cond = TCG_COND_EQ;
             break;
-        case 0x4 | 0x2 | 0x1: /* val & mask != 0 */
-            tcg_gen_brcondi_i64(TCG_COND_EQ, tmp64, 0, skip);
+        case 4 | 1: /* !zero -> op1 != op2 */
+            cond = TCG_COND_NE;
             break;
-        default:
-            tcg_temp_free_i64(tmp64);
-            goto do_dynamic;
-        }
-        tcg_temp_free_i64(tmp64);
-        account_inline_branch(s);
-        break;
-    case CC_OP_ICM:
-        switch (mask) {
-        case 0x8: /* val == 0 */
-            tcg_gen_brcondi_i64(TCG_COND_NE, cc_dst, 0, skip);
+        case 4: /* borrow (!carry) -> op1 < op2 */
+            cond = TCG_COND_LTU;
             break;
-        case 0x4 | 0x2 | 0x1: /* val != 0 */
-        case 0x4 | 0x2: /* val != 0 */
-            tcg_gen_brcondi_i64(TCG_COND_EQ, cc_dst, 0, skip);
+        case 2 | 1: /* !borrow (carry) -> op1 >= op2 */
+            cond = TCG_COND_GEU;
             break;
         default:
             goto do_dynamic;
         }
-        account_inline_branch(s);
+        account_inline_branch(s, old_cc_op);
         break;
-    case CC_OP_STATIC:
-        old_cc_op = s->cc_op;
-        goto do_dynamic_nocccalc;
-    case CC_OP_DYNAMIC:
+
     default:
-do_dynamic:
-        old_cc_op = s->cc_op;
-        /* calculate cc value */
+    do_dynamic:
+        /* Calculate cc value.  */
         gen_op_calc_cc(s);
+        /* FALLTHRU */
 
-do_dynamic_nocccalc:
-        /* jump based on cc */
+    case CC_OP_STATIC:
+        /* Jump based on CC.  We'll load up the real cond below;
+           the assignment here merely avoids a compiler warning.  */
         account_noninline_branch(s, old_cc_op);
+        old_cc_op = CC_OP_STATIC;
+        cond = TCG_COND_NEVER;
+        break;
+    }
+
+    /* Load up the arguments of the comparison.  */
+    c->is_64 = true;
+    c->g1 = c->g2 = false;
+    switch (old_cc_op) {
+    case CC_OP_LTGT0_32:
+        c->is_64 = false;
+        c->u.s32.a = tcg_temp_new_i32();
+        tcg_gen_trunc_i64_i32(c->u.s32.a, cc_dst);
+        c->u.s32.b = tcg_const_i32(0);
+        break;
+    case CC_OP_LTGT_32:
+    case CC_OP_LTUGTU_32:
+    case CC_OP_SUBU_32:
+        c->is_64 = false;
+        c->u.s32.a = tcg_temp_new_i32();
+        tcg_gen_trunc_i64_i32(c->u.s32.a, cc_src);
+        c->u.s32.b = tcg_temp_new_i32();
+        tcg_gen_trunc_i64_i32(c->u.s32.b, cc_dst);
+        break;
+
+    case CC_OP_LTGT0_64:
+    case CC_OP_NZ:
+    case CC_OP_FLOGR:
+        c->u.s64.a = cc_dst;
+        c->u.s64.b = tcg_const_i64(0);
+        c->g1 = true;
+        break;
+    case CC_OP_LTGT_64:
+    case CC_OP_LTUGTU_64:
+    case CC_OP_SUBU_64:
+        c->u.s64.a = cc_src;
+        c->u.s64.b = cc_dst;
+        c->g1 = c->g2 = true;
+        break;
 
+    case CC_OP_TM_32:
+    case CC_OP_TM_64:
+    case CC_OP_ICM:
+        c->u.s64.a = tcg_temp_new_i64();
+        c->u.s64.b = tcg_const_i64(0);
+        tcg_gen_and_i64(c->u.s64.a, cc_src, cc_dst);
+        break;
+
+    case CC_OP_ADDU_32:
+        c->is_64 = false;
+        c->u.s32.a = tcg_temp_new_i32();
+        c->u.s32.b = tcg_temp_new_i32();
+        tcg_gen_trunc_i64_i32(c->u.s32.a, cc_vr);
+        if (cond == TCG_COND_EQ || cond == TCG_COND_NE) {
+            tcg_gen_movi_i32(c->u.s32.b, 0);
+        } else {
+            tcg_gen_trunc_i64_i32(c->u.s32.b, cc_src);
+        }
+        break;
+
+    case CC_OP_ADDU_64:
+        c->u.s64.a = cc_vr;
+        c->g1 = true;
+        if (cond == TCG_COND_EQ || cond == TCG_COND_NE) {
+            c->u.s64.b = tcg_const_i64(0);
+        } else {
+            c->u.s64.b = cc_src;
+            c->g2 = true;
+        }
+        break;
+
+    case CC_OP_STATIC:
+        c->is_64 = false;
+        c->u.s32.a = cc_op;
+        c->g1 = true;
         switch (mask) {
-        case 0x8 | 0x4 | 0x2 | 0x1:
-            /* always true */
-            break;
         case 0x8 | 0x4 | 0x2: /* cc != 3 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, cc_op, 3, skip);
+            cond = TCG_COND_NE;
+            c->u.s32.b = tcg_const_i32(3);
             break;
         case 0x8 | 0x4 | 0x1: /* cc != 2 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, cc_op, 2, skip);
+            cond = TCG_COND_NE;
+            c->u.s32.b = tcg_const_i32(2);
             break;
         case 0x8 | 0x2 | 0x1: /* cc != 1 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, cc_op, 1, skip);
+            cond = TCG_COND_NE;
+            c->u.s32.b = tcg_const_i32(1);
             break;
-        case 0x8 | 0x2: /* cc == 0 || cc == 2 */
-            tmp = tcg_temp_new_i32();
-            tcg_gen_andi_i32(tmp, cc_op, 1);
-            tcg_gen_brcondi_i32(TCG_COND_NE, tmp, 0, skip);
-            tcg_temp_free_i32(tmp);
+        case 0x8 | 0x2: /* cc == 0 || cc == 2 => (cc & 1) == 0 */
+            cond = TCG_COND_EQ;
+            c->g1 = false;
+            c->u.s32.a = tcg_temp_new_i32();
+            c->u.s32.b = tcg_const_i32(0);
+            tcg_gen_andi_i32(c->u.s32.a, cc_op, 1);
             break;
         case 0x8 | 0x4: /* cc < 2 */
-            tcg_gen_brcondi_i32(TCG_COND_GEU, cc_op, 2, skip);
+            cond = TCG_COND_LTU;
+            c->u.s32.b = tcg_const_i32(2);
             break;
         case 0x8: /* cc == 0 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, cc_op, 0, skip);
+            cond = TCG_COND_EQ;
+            c->u.s32.b = tcg_const_i32(0);
             break;
         case 0x4 | 0x2 | 0x1: /* cc != 0 */
-            tcg_gen_brcondi_i32(TCG_COND_EQ, cc_op, 0, skip);
+            cond = TCG_COND_NE;
+            c->u.s32.b = tcg_const_i32(0);
             break;
-        case 0x4 | 0x1: /* cc == 1 || cc == 3 */
-            tmp = tcg_temp_new_i32();
-            tcg_gen_andi_i32(tmp, cc_op, 1);
-            tcg_gen_brcondi_i32(TCG_COND_EQ, tmp, 0, skip);
-            tcg_temp_free_i32(tmp);
+        case 0x4 | 0x1: /* cc == 1 || cc == 3 => (cc & 1) != 0 */
+            cond = TCG_COND_NE;
+            c->g1 = false;
+            c->u.s32.a = tcg_temp_new_i32();
+            c->u.s32.b = tcg_const_i32(0);
+            tcg_gen_andi_i32(c->u.s32.a, cc_op, 1);
             break;
         case 0x4: /* cc == 1 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, cc_op, 1, skip);
+            cond = TCG_COND_EQ;
+            c->u.s32.b = tcg_const_i32(1);
             break;
         case 0x2 | 0x1: /* cc > 1 */
-            tcg_gen_brcondi_i32(TCG_COND_LEU, cc_op, 1, skip);
+            cond = TCG_COND_GTU;
+            c->u.s32.b = tcg_const_i32(1);
             break;
         case 0x2: /* cc == 2 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, cc_op, 2, skip);
+            cond = TCG_COND_EQ;
+            c->u.s32.b = tcg_const_i32(2);
             break;
         case 0x1: /* cc == 3 */
-            tcg_gen_brcondi_i32(TCG_COND_NE, cc_op, 3, skip);
+            cond = TCG_COND_EQ;
+            c->u.s32.b = tcg_const_i32(3);
             break;
-        default: /* cc is masked by something else */
-            tmp = tcg_const_i32(3);
-            /* 3 - cc */
-            tcg_gen_sub_i32(tmp, tmp, cc_op);
-            tmp2 = tcg_const_i32(1);
-            /* 1 << (3 - cc) */
-            tcg_gen_shl_i32(tmp2, tmp2, tmp);
-            r = tcg_const_i32(mask);
-            /* mask & (1 << (3 - cc)) */
-            tcg_gen_and_i32(r, r, tmp2);
-            tcg_temp_free_i32(tmp);
-            tcg_temp_free_i32(tmp2);
-
-            tcg_gen_brcondi_i32(TCG_COND_EQ, r, 0, skip);
-            tcg_temp_free_i32(r);
+        default:
+            /* CC is masked by something else: (8 >> cc) & mask.  */
+            cond = TCG_COND_NE;
+            c->g1 = false;
+            c->u.s32.a = tcg_const_i32(8);
+            c->u.s32.b = tcg_const_i32(0);
+            tcg_gen_shr_i32(c->u.s32.a, c->u.s32.a, cc_op);
+            tcg_gen_andi_i32(c->u.s32.a, c->u.s32.a, mask);
             break;
         }
         break;
+
+    default:
+        abort();
     }
+    c->cond = cond;
 }
 
-static void gen_bcr(DisasContext *s, uint32_t mask, TCGv_i64 target,
-                    uint64_t offset)
+static void free_compare(DisasCompare *c)
 {
-    int skip;
-
-    if (mask == 0xf) {
-        /* unconditional */
-        tcg_gen_mov_i64(psw_addr, target);
-        tcg_gen_exit_tb(0);
-    } else if (mask == 0) {
-        /* ignore cc and never match */
-        gen_goto_tb(s, 0, offset + 2);
-    } else {
-        TCGv_i64 new_addr = tcg_temp_local_new_i64();
-
-        tcg_gen_mov_i64(new_addr, target);
-        skip = gen_new_label();
-        gen_jcc(s, mask, skip);
-        tcg_gen_mov_i64(psw_addr, new_addr);
-        tcg_temp_free_i64(new_addr);
-        tcg_gen_exit_tb(0);
-        gen_set_label(skip);
-        tcg_temp_free_i64(new_addr);
-        gen_goto_tb(s, 1, offset + 2);
+    if (!c->g1) {
+        if (c->is_64) {
+            tcg_temp_free_i64(c->u.s64.a);
+        } else {
+            tcg_temp_free_i32(c->u.s32.a);
+        }
+    }
+    if (!c->g2) {
+        if (c->is_64) {
+            tcg_temp_free_i64(c->u.s64.b);
+        } else {
+            tcg_temp_free_i32(c->u.s32.b);
+        }
     }
 }
 
-static void gen_brc(uint32_t mask, DisasContext *s, int32_t offset)
+/* ====================================================================== */
+/* Define the insn format enumeration.  */
+#define F0(N)                         FMT_##N,
+#define F1(N, X1)                     F0(N)
+#define F2(N, X1, X2)                 F0(N)
+#define F3(N, X1, X2, X3)             F0(N)
+#define F4(N, X1, X2, X3, X4)         F0(N)
+#define F5(N, X1, X2, X3, X4, X5)     F0(N)
+
+typedef enum {
+#include "insn-format.def"
+} DisasFormat;
+
+#undef F0
+#undef F1
+#undef F2
+#undef F3
+#undef F4
+#undef F5
+
+/* Define a structure to hold the decoded fields.  We'll store each inside
+   an array indexed by an enum.  In order to conserve memory, we'll arrange
+   for fields that do not exist at the same time to overlap, thus the "C"
+   for compact.  For checking purposes there is an "O" for original index
+   as well that will be applied to availability bitmaps.  */
+
+enum DisasFieldIndexO {
+    FLD_O_r1,
+    FLD_O_r2,
+    FLD_O_r3,
+    FLD_O_m1,
+    FLD_O_m3,
+    FLD_O_m4,
+    FLD_O_b1,
+    FLD_O_b2,
+    FLD_O_b4,
+    FLD_O_d1,
+    FLD_O_d2,
+    FLD_O_d4,
+    FLD_O_x2,
+    FLD_O_l1,
+    FLD_O_l2,
+    FLD_O_i1,
+    FLD_O_i2,
+    FLD_O_i3,
+    FLD_O_i4,
+    FLD_O_i5
+};
+
+enum DisasFieldIndexC {
+    FLD_C_r1 = 0,
+    FLD_C_m1 = 0,
+    FLD_C_b1 = 0,
+    FLD_C_i1 = 0,
+
+    FLD_C_r2 = 1,
+    FLD_C_b2 = 1,
+    FLD_C_i2 = 1,
+
+    FLD_C_r3 = 2,
+    FLD_C_m3 = 2,
+    FLD_C_i3 = 2,
+
+    FLD_C_m4 = 3,
+    FLD_C_b4 = 3,
+    FLD_C_i4 = 3,
+    FLD_C_l1 = 3,
+
+    FLD_C_i5 = 4,
+    FLD_C_d1 = 4,
+
+    FLD_C_d2 = 5,
+
+    FLD_C_d4 = 6,
+    FLD_C_x2 = 6,
+    FLD_C_l2 = 6,
+
+    NUM_C_FIELD = 7
+};
+
+struct DisasFields {
+    unsigned op:8;
+    unsigned op2:8;
+    unsigned presentC:16;
+    unsigned int presentO;
+    int c[NUM_C_FIELD];
+};
+
+/* This is the way fields are to be accessed out of DisasFields.  */
+#define have_field(S, F)  have_field1((S), FLD_O_##F)
+#define get_field(S, F)   get_field1((S), FLD_O_##F, FLD_C_##F)
+
+static bool have_field1(const DisasFields *f, enum DisasFieldIndexO c)
+{
+    return (f->presentO >> c) & 1;
+}
+
+static int get_field1(const DisasFields *f, enum DisasFieldIndexO o,
+                      enum DisasFieldIndexC c)
+{
+    assert(have_field1(f, o));
+    return f->c[c];
+}
+
+/* Describe the layout of each field in each format.  */
+typedef struct DisasField {
+    unsigned int beg:8;
+    unsigned int size:8;
+    unsigned int type:2;
+    unsigned int indexC:6;
+    enum DisasFieldIndexO indexO:8;
+} DisasField;
+
+typedef struct DisasFormatInfo {
+    DisasField op[NUM_C_FIELD];
+} DisasFormatInfo;
+
+#define R(N, B)       {  B,  4, 0, FLD_C_r##N, FLD_O_r##N }
+#define M(N, B)       {  B,  4, 0, FLD_C_m##N, FLD_O_m##N }
+#define BD(N, BB, BD) { BB,  4, 0, FLD_C_b##N, FLD_O_b##N }, \
+                      { BD, 12, 0, FLD_C_d##N, FLD_O_d##N }
+#define BXD(N)        { 16,  4, 0, FLD_C_b##N, FLD_O_b##N }, \
+                      { 12,  4, 0, FLD_C_x##N, FLD_O_x##N }, \
+                      { 20, 12, 0, FLD_C_d##N, FLD_O_d##N }
+#define BDL(N)        { 16,  4, 0, FLD_C_b##N, FLD_O_b##N }, \
+                      { 20, 20, 2, FLD_C_d##N, FLD_O_d##N }
+#define BXDL(N)       { 16,  4, 0, FLD_C_b##N, FLD_O_b##N }, \
+                      { 12,  4, 0, FLD_C_x##N, FLD_O_x##N }, \
+                      { 20, 20, 2, FLD_C_d##N, FLD_O_d##N }
+#define I(N, B, S)    {  B,  S, 1, FLD_C_i##N, FLD_O_i##N }
+#define L(N, B, S)    {  B,  S, 0, FLD_C_l##N, FLD_O_l##N }
+
+#define F0(N)                     { { } },
+#define F1(N, X1)                 { { X1 } },
+#define F2(N, X1, X2)             { { X1, X2 } },
+#define F3(N, X1, X2, X3)         { { X1, X2, X3 } },
+#define F4(N, X1, X2, X3, X4)     { { X1, X2, X3, X4 } },
+#define F5(N, X1, X2, X3, X4, X5) { { X1, X2, X3, X4, X5 } },
+
+static const DisasFormatInfo format_info[] = {
+#include "insn-format.def"
+};
+
+#undef F0
+#undef F1
+#undef F2
+#undef F3
+#undef F4
+#undef F5
+#undef R
+#undef M
+#undef BD
+#undef BXD
+#undef BDL
+#undef BXDL
+#undef I
+#undef L
+
+/* Generally, we'll extract operands into this structures, operate upon
+   them, and store them back.  See the "in1", "in2", "prep", "wout" sets
+   of routines below for more details.  */
+typedef struct {
+    bool g_out, g_out2, g_in1, g_in2;
+    TCGv_i64 out, out2, in1, in2;
+    TCGv_i64 addr1;
+} DisasOps;
+
+/* Instructions can place constraints on their operands, raising specification
+   exceptions if they are violated.  To make this easy to automate, each "in1",
+   "in2", "prep", "wout" helper will have a SPEC_<name> define that equals one
+   of the following, or 0.  To make this easy to document, we'll put the
+   SPEC_<name> defines next to <name>.  */
+
+#define SPEC_r1_even    1
+#define SPEC_r2_even    2
+#define SPEC_r3_even    4
+#define SPEC_r1_f128    8
+#define SPEC_r2_f128    16
+
+/* Return values from translate_one, indicating the state of the TB.  */
+typedef enum {
+    /* Continue the TB.  */
+    NO_EXIT,
+    /* We have emitted one or more goto_tb.  No fixup required.  */
+    EXIT_GOTO_TB,
+    /* We are not using a goto_tb (for whatever reason), but have updated
+       the PC (for whatever reason), so there's no need to do it again on
+       exiting the TB.  */
+    EXIT_PC_UPDATED,
+    /* We are exiting the TB, but have neither emitted a goto_tb, nor
+       updated the PC for the next instruction to be executed.  */
+    EXIT_PC_STALE,
+    /* We are ending the TB with a noreturn function call, e.g. longjmp.
+       No following code will be executed.  */
+    EXIT_NORETURN,
+} ExitStatus;
+
+typedef enum DisasFacility {
+    FAC_Z,                  /* zarch (default) */
+    FAC_CASS,               /* compare and swap and store */
+    FAC_CASS2,              /* compare and swap and store 2*/
+    FAC_DFP,                /* decimal floating point */
+    FAC_DFPR,               /* decimal floating point rounding */
+    FAC_DO,                 /* distinct operands */
+    FAC_EE,                 /* execute extensions */
+    FAC_EI,                 /* extended immediate */
+    FAC_FPE,                /* floating point extension */
+    FAC_FPSSH,              /* floating point support sign handling */
+    FAC_FPRGR,              /* FPR-GR transfer */
+    FAC_GIE,                /* general instructions extension */
+    FAC_HFP_MA,             /* HFP multiply-and-add/subtract */
+    FAC_HW,                 /* high-word */
+    FAC_IEEEE_SIM,          /* IEEE exception sumilation */
+    FAC_LOC,                /* load/store on condition */
+    FAC_LD,                 /* long displacement */
+    FAC_PC,                 /* population count */
+    FAC_SCF,                /* store clock fast */
+    FAC_SFLE,               /* store facility list extended */
+} DisasFacility;
+
+struct DisasInsn {
+    unsigned opc:16;
+    DisasFormat fmt:8;
+    DisasFacility fac:8;
+    unsigned spec:8;
+
+    const char *name;
+
+    void (*help_in1)(DisasContext *, DisasFields *, DisasOps *);
+    void (*help_in2)(DisasContext *, DisasFields *, DisasOps *);
+    void (*help_prep)(DisasContext *, DisasFields *, DisasOps *);
+    void (*help_wout)(DisasContext *, DisasFields *, DisasOps *);
+    void (*help_cout)(DisasContext *, DisasOps *);
+    ExitStatus (*help_op)(DisasContext *, DisasOps *);
+
+    uint64_t data;
+};
+
+/* ====================================================================== */
+/* Miscelaneous helpers, used by several operations.  */
+
+static void help_l2_shift(DisasContext *s, DisasFields *f,
+                          DisasOps *o, int mask)
 {
-    int skip;
+    int b2 = get_field(f, b2);
+    int d2 = get_field(f, d2);
 
-    if (mask == 0xf) {
-        /* unconditional */
-        gen_goto_tb(s, 0, s->pc + offset);
-    } else if (mask == 0) {
-        /* ignore cc and never match */
-        gen_goto_tb(s, 0, s->pc + 4);
+    if (b2 == 0) {
+        o->in2 = tcg_const_i64(d2 & mask);
     } else {
-        skip = gen_new_label();
-        gen_jcc(s, mask, skip);
-        gen_goto_tb(s, 0, s->pc + offset);
-        gen_set_label(skip);
-        gen_goto_tb(s, 1, s->pc + 4);
+        o->in2 = get_address(s, 0, b2, d2);
+        tcg_gen_andi_i64(o->in2, o->in2, mask);
     }
-    s->is_jmp = DISAS_TB_JUMP;
 }
 
-static void gen_op_mvc(DisasContext *s, int l, TCGv_i64 s1, TCGv_i64 s2)
+static ExitStatus help_goto_direct(DisasContext *s, uint64_t dest)
 {
-    TCGv_i64 tmp, tmp2;
-    int i;
-    int l_memset = gen_new_label();
-    int l_out = gen_new_label();
-    TCGv_i64 dest = tcg_temp_local_new_i64();
-    TCGv_i64 src = tcg_temp_local_new_i64();
-    TCGv_i32 vl;
+    if (dest == s->next_pc) {
+        return NO_EXIT;
+    }
+    if (use_goto_tb(s, dest)) {
+        update_cc_op(s);
+        tcg_gen_goto_tb(0);
+        tcg_gen_movi_i64(psw_addr, dest);
+        tcg_gen_exit_tb((tcg_target_long)s->tb);
+        return EXIT_GOTO_TB;
+    } else {
+        tcg_gen_movi_i64(psw_addr, dest);
+        return EXIT_PC_UPDATED;
+    }
+}
 
-    /* Find out if we should use the inline version of mvc */
-    switch (l) {
-    case 0:
-    case 1:
-    case 2:
-    case 3:
-    case 4:
-    case 5:
-    case 6:
-    case 7:
-    case 11:
-    case 15:
-        /* use inline */
-        break;
-    default:
-        /* Fall back to helper */
-        vl = tcg_const_i32(l);
-        potential_page_fault(s);
-        gen_helper_mvc(cpu_env, vl, s1, s2);
-        tcg_temp_free_i32(vl);
-        return;
+static ExitStatus help_branch(DisasContext *s, DisasCompare *c,
+                              bool is_imm, int imm, TCGv_i64 cdest)
+{
+    ExitStatus ret;
+    uint64_t dest = s->pc + 2 * imm;
+    int lab;
+
+    /* Take care of the special cases first.  */
+    if (c->cond == TCG_COND_NEVER) {
+        ret = NO_EXIT;
+        goto egress;
+    }
+    if (is_imm) {
+        if (dest == s->next_pc) {
+            /* Branch to next.  */
+            ret = NO_EXIT;
+            goto egress;
+        }
+        if (c->cond == TCG_COND_ALWAYS) {
+            ret = help_goto_direct(s, dest);
+            goto egress;
+        }
+    } else {
+        if (TCGV_IS_UNUSED_I64(cdest)) {
+            /* E.g. bcr %r0 -> no branch.  */
+            ret = NO_EXIT;
+            goto egress;
+        }
+        if (c->cond == TCG_COND_ALWAYS) {
+            tcg_gen_mov_i64(psw_addr, cdest);
+            ret = EXIT_PC_UPDATED;
+            goto egress;
+        }
     }
 
-    tcg_gen_mov_i64(dest, s1);
-    tcg_gen_mov_i64(src, s2);
+    if (use_goto_tb(s, s->next_pc)) {
+        if (is_imm && use_goto_tb(s, dest)) {
+            /* Both exits can use goto_tb.  */
+            update_cc_op(s);
 
-    if (!(s->tb->flags & FLAG_MASK_64)) {
-        /* XXX what if we overflow while moving? */
-        tcg_gen_andi_i64(dest, dest, 0x7fffffffUL);
-        tcg_gen_andi_i64(src, src, 0x7fffffffUL);
+            lab = gen_new_label();
+            if (c->is_64) {
+                tcg_gen_brcond_i64(c->cond, c->u.s64.a, c->u.s64.b, lab);
+            } else {
+                tcg_gen_brcond_i32(c->cond, c->u.s32.a, c->u.s32.b, lab);
+            }
+
+            /* Branch not taken.  */
+            tcg_gen_goto_tb(0);
+            tcg_gen_movi_i64(psw_addr, s->next_pc);
+            tcg_gen_exit_tb((tcg_target_long)s->tb + 0);
+
+            /* Branch taken.  */
+            gen_set_label(lab);
+            tcg_gen_goto_tb(1);
+            tcg_gen_movi_i64(psw_addr, dest);
+            tcg_gen_exit_tb((tcg_target_long)s->tb + 1);
+
+            ret = EXIT_GOTO_TB;
+        } else {
+            /* Fallthru can use goto_tb, but taken branch cannot.  */
+            /* Store taken branch destination before the brcond.  This
+               avoids having to allocate a new local temp to hold it.
+               We'll overwrite this in the not taken case anyway.  */
+            if (!is_imm) {
+                tcg_gen_mov_i64(psw_addr, cdest);
+            }
+
+            lab = gen_new_label();
+            if (c->is_64) {
+                tcg_gen_brcond_i64(c->cond, c->u.s64.a, c->u.s64.b, lab);
+            } else {
+                tcg_gen_brcond_i32(c->cond, c->u.s32.a, c->u.s32.b, lab);
+            }
+
+            /* Branch not taken.  */
+            update_cc_op(s);
+            tcg_gen_goto_tb(0);
+            tcg_gen_movi_i64(psw_addr, s->next_pc);
+            tcg_gen_exit_tb((tcg_target_long)s->tb + 0);
+
+            gen_set_label(lab);
+            if (is_imm) {
+                tcg_gen_movi_i64(psw_addr, dest);
+            }
+            ret = EXIT_PC_UPDATED;
+        }
+    } else {
+        /* Fallthru cannot use goto_tb.  This by itself is vanishingly rare.
+           Most commonly we're single-stepping or some other condition that
+           disables all use of goto_tb.  Just update the PC and exit.  */
+
+        TCGv_i64 next = tcg_const_i64(s->next_pc);
+        if (is_imm) {
+            cdest = tcg_const_i64(dest);
+        }
+
+        if (c->is_64) {
+            tcg_gen_movcond_i64(c->cond, psw_addr, c->u.s64.a, c->u.s64.b,
+                                cdest, next);
+        } else {
+            TCGv_i32 t0 = tcg_temp_new_i32();
+            TCGv_i64 t1 = tcg_temp_new_i64();
+            TCGv_i64 z = tcg_const_i64(0);
+            tcg_gen_setcond_i32(c->cond, t0, c->u.s32.a, c->u.s32.b);
+            tcg_gen_extu_i32_i64(t1, t0);
+            tcg_temp_free_i32(t0);
+            tcg_gen_movcond_i64(TCG_COND_NE, psw_addr, t1, z, cdest, next);
+            tcg_temp_free_i64(t1);
+            tcg_temp_free_i64(z);
+        }
+
+        if (is_imm) {
+            tcg_temp_free_i64(cdest);
+        }
+        tcg_temp_free_i64(next);
+
+        ret = EXIT_PC_UPDATED;
     }
 
-    tmp = tcg_temp_new_i64();
-    tcg_gen_addi_i64(tmp, src, 1);
-    tcg_gen_brcond_i64(TCG_COND_EQ, dest, tmp, l_memset);
-    tcg_temp_free_i64(tmp);
+ egress:
+    free_compare(c);
+    return ret;
+}
 
-    switch (l) {
-    case 0:
-        tmp = tcg_temp_new_i64();
+/* ====================================================================== */
+/* The operations.  These perform the bulk of the work for any insn,
+   usually after the operands have been loaded and output initialized.  */
 
-        tcg_gen_qemu_ld8u(tmp, src, get_mem_index(s));
-        tcg_gen_qemu_st8(tmp, dest, get_mem_index(s));
+static ExitStatus op_abs(DisasContext *s, DisasOps *o)
+{
+    gen_helper_abs_i64(o->out, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        break;
-    case 1:
-        tmp = tcg_temp_new_i64();
+static ExitStatus op_absf32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_andi_i64(o->out, o->in2, 0x7fffffffull);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld16u(tmp, src, get_mem_index(s));
-        tcg_gen_qemu_st16(tmp, dest, get_mem_index(s));
+static ExitStatus op_absf64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_andi_i64(o->out, o->in2, 0x7fffffffffffffffull);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        break;
-    case 3:
-        tmp = tcg_temp_new_i64();
+static ExitStatus op_absf128(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_andi_i64(o->out, o->in1, 0x7fffffffffffffffull);
+    tcg_gen_mov_i64(o->out2, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld32u(tmp, src, get_mem_index(s));
-        tcg_gen_qemu_st32(tmp, dest, get_mem_index(s));
+static ExitStatus op_add(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_add_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        break;
-    case 4:
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
-
-        tcg_gen_qemu_ld32u(tmp, src, get_mem_index(s));
-        tcg_gen_addi_i64(src, src, 4);
-        tcg_gen_qemu_ld8u(tmp2, src, get_mem_index(s));
-        tcg_gen_qemu_st32(tmp, dest, get_mem_index(s));
-        tcg_gen_addi_i64(dest, dest, 4);
-        tcg_gen_qemu_st8(tmp2, dest, get_mem_index(s));
-
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 7:
-        tmp = tcg_temp_new_i64();
+static ExitStatus op_addc(DisasContext *s, DisasOps *o)
+{
+    DisasCompare cmp;
+    TCGv_i64 carry;
+
+    tcg_gen_add_i64(o->out, o->in1, o->in2);
+
+    /* The carry flag is the msb of CC, therefore the branch mask that would
+       create that comparison is 3.  Feeding the generated comparison to
+       setcond produces the carry flag that we desire.  */
+    disas_jcc(s, &cmp, 3);
+    carry = tcg_temp_new_i64();
+    if (cmp.is_64) {
+        tcg_gen_setcond_i64(cmp.cond, carry, cmp.u.s64.a, cmp.u.s64.b);
+    } else {
+        TCGv_i32 t = tcg_temp_new_i32();
+        tcg_gen_setcond_i32(cmp.cond, t, cmp.u.s32.a, cmp.u.s32.b);
+        tcg_gen_extu_i32_i64(carry, t);
+        tcg_temp_free_i32(t);
+    }
+    free_compare(&cmp);
 
-        tcg_gen_qemu_ld64(tmp, src, get_mem_index(s));
-        tcg_gen_qemu_st64(tmp, dest, get_mem_index(s));
+    tcg_gen_add_i64(o->out, o->out, carry);
+    tcg_temp_free_i64(carry);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        break;
-    default:
-        /* The inline version can become too big for too uneven numbers, only
-           use it on known good lengths */
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_const_i64(8);
-        for (i = 0; (i + 7) <= l; i += 8) {
-            tcg_gen_qemu_ld64(tmp, src, get_mem_index(s));
-            tcg_gen_qemu_st64(tmp, dest, get_mem_index(s));
-
-            tcg_gen_add_i64(src, src, tmp2);
-            tcg_gen_add_i64(dest, dest, tmp2);
-        }
+static ExitStatus op_aeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_aeb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp2);
-        tmp2 = tcg_const_i64(1);
+static ExitStatus op_adb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_adb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        for (; i <= l; i++) {
-            tcg_gen_qemu_ld8u(tmp, src, get_mem_index(s));
-            tcg_gen_qemu_st8(tmp, dest, get_mem_index(s));
+static ExitStatus op_axb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_axb(o->out, cpu_env, o->out, o->out2, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
 
-            tcg_gen_add_i64(src, src, tmp2);
-            tcg_gen_add_i64(dest, dest, tmp2);
-        }
+static ExitStatus op_and(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_and_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp);
-        break;
+static ExitStatus op_andi(DisasContext *s, DisasOps *o)
+{
+    int shift = s->insn->data & 0xff;
+    int size = s->insn->data >> 8;
+    uint64_t mask = ((1ull << size) - 1) << shift;
+
+    assert(!o->g_in2);
+    tcg_gen_shli_i64(o->in2, o->in2, shift);
+    tcg_gen_ori_i64(o->in2, o->in2, ~mask);
+    tcg_gen_and_i64(o->out, o->in1, o->in2);
+
+    /* Produce the CC from only the bits manipulated.  */
+    tcg_gen_andi_i64(cc_dst, o->out, mask);
+    set_cc_nz_u64(s, cc_dst);
+    return NO_EXIT;
+}
+
+static ExitStatus op_bas(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_movi_i64(o->out, pc_to_link_info(s, s->next_pc));
+    if (!TCGV_IS_UNUSED_I64(o->in2)) {
+        tcg_gen_mov_i64(psw_addr, o->in2);
+        return EXIT_PC_UPDATED;
+    } else {
+        return NO_EXIT;
     }
+}
 
-    tcg_gen_br(l_out);
+static ExitStatus op_basi(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_movi_i64(o->out, pc_to_link_info(s, s->next_pc));
+    return help_goto_direct(s, s->pc + 2 * get_field(s->fields, i2));
+}
 
-    gen_set_label(l_memset);
-    /* memset case (dest == (src + 1)) */
+static ExitStatus op_bc(DisasContext *s, DisasOps *o)
+{
+    int m1 = get_field(s->fields, m1);
+    bool is_imm = have_field(s->fields, i2);
+    int imm = is_imm ? get_field(s->fields, i2) : 0;
+    DisasCompare c;
+
+    disas_jcc(s, &c, m1);
+    return help_branch(s, &c, is_imm, imm, o->in2);
+}
+
+static ExitStatus op_bct32(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    bool is_imm = have_field(s->fields, i2);
+    int imm = is_imm ? get_field(s->fields, i2) : 0;
+    DisasCompare c;
+    TCGv_i64 t;
+
+    c.cond = TCG_COND_NE;
+    c.is_64 = false;
+    c.g1 = false;
+    c.g2 = false;
+
+    t = tcg_temp_new_i64();
+    tcg_gen_subi_i64(t, regs[r1], 1);
+    store_reg32_i64(r1, t);
+    c.u.s32.a = tcg_temp_new_i32();
+    c.u.s32.b = tcg_const_i32(0);
+    tcg_gen_trunc_i64_i32(c.u.s32.a, t);
+    tcg_temp_free_i64(t);
+
+    return help_branch(s, &c, is_imm, imm, o->in2);
+}
+
+static ExitStatus op_bct64(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    bool is_imm = have_field(s->fields, i2);
+    int imm = is_imm ? get_field(s->fields, i2) : 0;
+    DisasCompare c;
+
+    c.cond = TCG_COND_NE;
+    c.is_64 = true;
+    c.g1 = true;
+    c.g2 = false;
 
-    tmp = tcg_temp_new_i64();
-    tmp2 = tcg_temp_new_i64();
-    /* fill tmp with the byte */
-    tcg_gen_qemu_ld8u(tmp, src, get_mem_index(s));
-    tcg_gen_shli_i64(tmp2, tmp, 8);
-    tcg_gen_or_i64(tmp, tmp, tmp2);
-    tcg_gen_shli_i64(tmp2, tmp, 16);
-    tcg_gen_or_i64(tmp, tmp, tmp2);
-    tcg_gen_shli_i64(tmp2, tmp, 32);
-    tcg_gen_or_i64(tmp, tmp, tmp2);
-    tcg_temp_free_i64(tmp2);
+    tcg_gen_subi_i64(regs[r1], regs[r1], 1);
+    c.u.s64.a = regs[r1];
+    c.u.s64.b = tcg_const_i64(0);
 
-    tmp2 = tcg_const_i64(8);
+    return help_branch(s, &c, is_imm, imm, o->in2);
+}
 
-    for (i = 0; (i + 7) <= l; i += 8) {
-        tcg_gen_qemu_st64(tmp, dest, get_mem_index(s));
-        tcg_gen_addi_i64(dest, dest, 8);
+static ExitStatus op_bx32(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    bool is_imm = have_field(s->fields, i2);
+    int imm = is_imm ? get_field(s->fields, i2) : 0;
+    DisasCompare c;
+    TCGv_i64 t;
+
+    c.cond = (s->insn->data ? TCG_COND_LE : TCG_COND_GT);
+    c.is_64 = false;
+    c.g1 = false;
+    c.g2 = false;
+
+    t = tcg_temp_new_i64();
+    tcg_gen_add_i64(t, regs[r1], regs[r3]);
+    c.u.s32.a = tcg_temp_new_i32();
+    c.u.s32.b = tcg_temp_new_i32();
+    tcg_gen_trunc_i64_i32(c.u.s32.a, t);
+    tcg_gen_trunc_i64_i32(c.u.s32.b, regs[r3 | 1]);
+    store_reg32_i64(r1, t);
+    tcg_temp_free_i64(t);
+
+    return help_branch(s, &c, is_imm, imm, o->in2);
+}
+
+static ExitStatus op_bx64(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    bool is_imm = have_field(s->fields, i2);
+    int imm = is_imm ? get_field(s->fields, i2) : 0;
+    DisasCompare c;
+
+    c.cond = (s->insn->data ? TCG_COND_LE : TCG_COND_GT);
+    c.is_64 = true;
+
+    if (r1 == (r3 | 1)) {
+        c.u.s64.b = load_reg(r3 | 1);
+        c.g2 = false;
+    } else {
+        c.u.s64.b = regs[r3 | 1];
+        c.g2 = true;
     }
 
-    tcg_temp_free_i64(tmp2);
-    tmp2 = tcg_const_i64(1);
+    tcg_gen_add_i64(regs[r1], regs[r1], regs[r3]);
+    c.u.s64.a = regs[r1];
+    c.g1 = true;
+
+    return help_branch(s, &c, is_imm, imm, o->in2);
+}
 
-    for (; i <= l; i++) {
-        tcg_gen_qemu_st8(tmp, dest, get_mem_index(s));
-        tcg_gen_addi_i64(dest, dest, 1);
+static ExitStatus op_cj(DisasContext *s, DisasOps *o)
+{
+    int imm, m3 = get_field(s->fields, m3);
+    bool is_imm;
+    DisasCompare c;
+
+    c.cond = ltgt_cond[m3];
+    if (s->insn->data) {
+        c.cond = tcg_unsigned_cond(c.cond);
     }
+    c.is_64 = c.g1 = c.g2 = true;
+    c.u.s64.a = o->in1;
+    c.u.s64.b = o->in2;
 
-    tcg_temp_free_i64(tmp2);
-    tcg_temp_free_i64(tmp);
+    is_imm = have_field(s->fields, i4);
+    if (is_imm) {
+        imm = get_field(s->fields, i4);
+    } else {
+        imm = 0;
+        o->out = get_address(s, 0, get_field(s->fields, b4),
+                             get_field(s->fields, d4));
+    }
 
-    gen_set_label(l_out);
+    return help_branch(s, &c, is_imm, imm, o->out);
+}
 
-    tcg_temp_free(dest);
-    tcg_temp_free(src);
+static ExitStatus op_ceb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_ceb(cc_op, cpu_env, o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
 }
 
-static void gen_op_clc(DisasContext *s, int l, TCGv_i64 s1, TCGv_i64 s2)
+static ExitStatus op_cdb(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp;
-    TCGv_i64 tmp2;
-    TCGv_i32 vl;
+    gen_helper_cdb(cc_op, cpu_env, o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
 
-    /* check for simple 32bit or 64bit match */
-    switch (l) {
-    case 0:
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
+static ExitStatus op_cxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_cxb(cc_op, cpu_env, o->out, o->out2, o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld8u(tmp, s1, get_mem_index(s));
-        tcg_gen_qemu_ld8u(tmp2, s2, get_mem_index(s));
-        cmp_u64(s, tmp, tmp2);
+static ExitStatus op_cfeb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cfeb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f32(s, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        return;
-    case 1:
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
+static ExitStatus op_cfdb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cfdb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f64(s, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld16u(tmp, s1, get_mem_index(s));
-        tcg_gen_qemu_ld16u(tmp2, s2, get_mem_index(s));
-        cmp_u64(s, tmp, tmp2);
+static ExitStatus op_cfxb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cfxb(o->out, cpu_env, o->in1, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f128(s, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        return;
-    case 3:
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
+static ExitStatus op_cgeb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cgeb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f32(s, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld32u(tmp, s1, get_mem_index(s));
-        tcg_gen_qemu_ld32u(tmp2, s2, get_mem_index(s));
-        cmp_u64(s, tmp, tmp2);
+static ExitStatus op_cgdb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cgdb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f64(s, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        return;
-    case 7:
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
+static ExitStatus op_cgxb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cgxb(o->out, cpu_env, o->in1, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f128(s, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_gen_qemu_ld64(tmp, s1, get_mem_index(s));
-        tcg_gen_qemu_ld64(tmp2, s2, get_mem_index(s));
-        cmp_u64(s, tmp, tmp2);
+static ExitStatus op_clfeb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clfeb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f32(s, o->in2);
+    return NO_EXIT;
+}
 
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        return;
-    }
+static ExitStatus op_clfdb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clfdb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f64(s, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clfxb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clfxb(o->out, cpu_env, o->in1, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f128(s, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clgeb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clgeb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f32(s, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clgdb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clgdb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f64(s, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clgxb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_clgxb(o->out, cpu_env, o->in1, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    gen_set_cc_nz_f128(s, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cegb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cegb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cdgb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cdgb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cxgb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cxgb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_celgb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_celgb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cdlgb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cdlgb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cxlgb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    gen_helper_cxlgb(o->out, cpu_env, o->in2, m3);
+    tcg_temp_free_i32(m3);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cksm(DisasContext *s, DisasOps *o)
+{
+    int r2 = get_field(s->fields, r2);
+    TCGv_i64 len = tcg_temp_new_i64();
 
     potential_page_fault(s);
-    vl = tcg_const_i32(l);
-    gen_helper_clc(cc_op, cpu_env, vl, s1, s2);
-    tcg_temp_free_i32(vl);
+    gen_helper_cksm(len, cpu_env, o->in1, o->in2, regs[r2 + 1]);
     set_cc_static(s);
+    return_low128(o->out);
+
+    tcg_gen_add_i64(regs[r2], regs[r2], len);
+    tcg_gen_sub_i64(regs[r2 + 1], regs[r2 + 1], len);
+    tcg_temp_free_i64(len);
+
+    return NO_EXIT;
 }
 
-static void disas_e3(CPUS390XState *env, DisasContext* s, int op, int r1,
-                     int x2, int b2, int d2)
+static ExitStatus op_clc(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 addr, tmp, tmp2, tmp3, tmp4;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
+    int l = get_field(s->fields, l1);
+    TCGv_i32 vl;
 
-    LOG_DISAS("disas_e3: op 0x%x r1 %d x2 %d b2 %d d2 %d\n",
-              op, r1, x2, b2, d2);
-    addr = get_address(s, x2, b2, d2);
-    switch (op) {
-    case 0x2: /* LTG R1,D2(X2,B2) [RXY] */
-    case 0x4: /* lg r1,d2(x2,b2) */
-        tcg_gen_qemu_ld64(regs[r1], addr, get_mem_index(s));
-        if (op == 0x2) {
-            set_cc_s64(s, regs[r1]);
-        }
-        break;
-    case 0x12: /* LT R1,D2(X2,B2) [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        store_reg32(r1, tmp32_1);
-        set_cc_s32(s, tmp32_1);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xc: /* MSG      R1,D2(X2,B2)     [RXY] */
-    case 0x1c: /* MSGF     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        if (op == 0xc) {
-            tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        } else {
-            tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        }
-        tcg_gen_mul_i64(regs[r1], regs[r1], tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0xd: /* DSG      R1,D2(X2,B2)     [RXY] */
-    case 0x1d: /* DSGF      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        if (op == 0x1d) {
-            tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        } else {
-            tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        }
-        tmp4 = load_reg(r1 + 1);
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_div_i64(tmp3, tmp4, tmp2);
-        store_reg(r1 + 1, tmp3);
-        tcg_gen_rem_i64(tmp3, tmp4, tmp2);
-        store_reg(r1, tmp3);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i64(tmp4);
-        break;
-    case 0x8: /* AG      R1,D2(X2,B2)     [RXY] */
-    case 0xa: /* ALG      R1,D2(X2,B2)     [RXY] */
-    case 0x18: /* AGF       R1,D2(X2,B2)     [RXY] */
-    case 0x1a: /* ALGF      R1,D2(X2,B2)     [RXY] */
-        if (op == 0x1a) {
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        } else if (op == 0x18) {
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        } else {
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        }
-        tmp4 = load_reg(r1);
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_add_i64(tmp3, tmp4, tmp2);
-        store_reg(r1, tmp3);
-        switch (op) {
-        case 0x8:
-        case 0x18:
-            set_cc_add64(s, tmp4, tmp2, tmp3);
-            break;
-        case 0xa:
-        case 0x1a:
-            set_cc_addu64(s, tmp4, tmp2, tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i64(tmp4);
-        break;
-    case 0x9: /* SG      R1,D2(X2,B2)     [RXY] */
-    case 0xb: /* SLG      R1,D2(X2,B2)     [RXY] */
-    case 0x19: /* SGF      R1,D2(X2,B2)     [RXY] */
-    case 0x1b: /* SLGF     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        if (op == 0x19) {
-            tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        } else if (op == 0x1b) {
-            tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        } else {
-            tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        }
-        tmp4 = load_reg(r1);
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_sub_i64(tmp3, tmp4, tmp2);
-        store_reg(r1, tmp3);
-        switch (op) {
-        case 0x9:
-        case 0x19:
-            set_cc_sub64(s, tmp4, tmp2, tmp3);
-            break;
-        case 0xb:
-        case 0x1b:
-            set_cc_subu64(s, tmp4, tmp2, tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i64(tmp4);
-        break;
-    case 0xf: /* LRVG     R1,D2(X2,B2)     [RXE] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        tcg_gen_bswap64_i64(tmp2, tmp2);
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x14: /* LGF      R1,D2(X2,B2)     [RXY] */
-    case 0x16: /* LLGF      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        if (op == 0x14) {
-            tcg_gen_ext32s_i64(tmp2, tmp2);
-        }
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x15: /* LGH     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld16s(tmp2, addr, get_mem_index(s));
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x17: /* LLGT      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_andi_i64(tmp2, tmp2, 0x7fffffffULL);
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x1e: /* LRV R1,D2(X2,B2) [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_gen_bswap32_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x1f: /* LRVH R1,D2(X2,B2) [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld16u(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_gen_bswap16_i32(tmp32_1, tmp32_1);
-        store_reg16(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x20: /* CG      R1,D2(X2,B2)     [RXY] */
-    case 0x21: /* CLG      R1,D2(X2,B2) */
-    case 0x30: /* CGF       R1,D2(X2,B2)     [RXY] */
-    case 0x31: /* CLGF      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        switch (op) {
-        case 0x20:
-        case 0x21:
-            tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-            break;
-        case 0x30:
-            tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-            break;
-        case 0x31:
-            tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-            break;
-        default:
-            tcg_abort();
-        }
-        switch (op) {
-        case 0x20:
-        case 0x30:
-            cmp_s64(s, regs[r1], tmp2);
-            break;
-        case 0x21:
-        case 0x31:
-            cmp_u64(s, regs[r1], tmp2);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x24: /* stg r1, d2(x2,b2) */
-        tcg_gen_qemu_st64(regs[r1], addr, get_mem_index(s));
-        break;
-    case 0x3e: /* STRV R1,D2(X2,B2) [RXY] */
-        tmp32_1 = load_reg32(r1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_bswap32_i32(tmp32_1, tmp32_1);
-        tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_gen_qemu_st32(tmp2, addr, get_mem_index(s));
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x50: /* STY  R1,D2(X2,B2) [RXY] */
-        tmp32_1 = load_reg32(r1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_gen_qemu_st32(tmp2, addr, get_mem_index(s));
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x57: /* XY R1,D2(X2,B2) [RXY] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_gen_xor_i32(tmp32_2, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_2);
-        set_cc_nz_u32(s, tmp32_2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x58: /* LY R1,D2(X2,B2) [RXY] */
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp3, addr, get_mem_index(s));
-        store_reg32_i64(r1, tmp3);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x5a: /* AY R1,D2(X2,B2) [RXY] */
-    case 0x5b: /* SY R1,D2(X2,B2) [RXY] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp32_3 = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32s(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        tcg_temp_free_i64(tmp2);
-        switch (op) {
-        case 0x5a:
-            tcg_gen_add_i32(tmp32_3, tmp32_1, tmp32_2);
-            break;
-        case 0x5b:
-            tcg_gen_sub_i32(tmp32_3, tmp32_1, tmp32_2);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg32(r1, tmp32_3);
-        switch (op) {
-        case 0x5a:
-            set_cc_add32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x5b:
-            set_cc_sub32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x71: /* LAY R1,D2(X2,B2) [RXY] */
-        store_reg(r1, addr);
-        break;
-    case 0x72: /* STCY R1,D2(X2,B2) [RXY] */
-        tmp32_1 = load_reg32(r1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_ext_i32_i64(tmp2, tmp32_1);
-        tcg_gen_qemu_st8(tmp2, addr, get_mem_index(s));
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x73: /* ICY R1,D2(X2,B2) [RXY] */
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8u(tmp3, addr, get_mem_index(s));
-        store_reg8(r1, tmp3);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x76: /* LB R1,D2(X2,B2) [RXY] */
-    case 0x77: /* LGB R1,D2(X2,B2) [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8s(tmp2, addr, get_mem_index(s));
-        switch (op) {
-        case 0x76:
-            tcg_gen_ext8s_i64(tmp2, tmp2);
-            store_reg32_i64(r1, tmp2);
-            break;
-        case 0x77:
-            tcg_gen_ext8s_i64(tmp2, tmp2);
-            store_reg(r1, tmp2);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x78: /* LHY R1,D2(X2,B2) [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld16s(tmp2, addr, get_mem_index(s));
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x80: /* NG      R1,D2(X2,B2)     [RXY] */
-    case 0x81: /* OG      R1,D2(X2,B2)     [RXY] */
-    case 0x82: /* XG      R1,D2(X2,B2)     [RXY] */
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld64(tmp3, addr, get_mem_index(s));
-        switch (op) {
-        case 0x80:
-            tcg_gen_and_i64(regs[r1], regs[r1], tmp3);
-            break;
-        case 0x81:
-            tcg_gen_or_i64(regs[r1], regs[r1], tmp3);
-            break;
-        case 0x82:
-            tcg_gen_xor_i64(regs[r1], regs[r1], tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        set_cc_nz_u64(s, regs[r1]);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x86: /* MLG      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_const_i32(r1);
-        tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        gen_helper_mlg(cpu_env, tmp32_1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x87: /* DLG      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_const_i32(r1);
-        tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        gen_helper_dlg(cpu_env, tmp32_1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x88: /* ALCG      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        /* XXX possible optimization point */
-        gen_op_calc_cc(s);
-        tcg_gen_extu_i32_i64(tmp3, cc_op);
-        tcg_gen_shri_i64(tmp3, tmp3, 1);
-        tcg_gen_andi_i64(tmp3, tmp3, 1);
-        tcg_gen_add_i64(tmp3, tmp2, tmp3);
-        tcg_gen_add_i64(tmp3, regs[r1], tmp3);
-        store_reg(r1, tmp3);
-        set_cc_addu64(s, regs[r1], tmp2, tmp3);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x89: /* SLBG      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_const_i32(r1);
-        tcg_gen_qemu_ld64(tmp2, addr, get_mem_index(s));
-        /* XXX possible optimization point */
-        gen_op_calc_cc(s);
-        gen_helper_slbg(cc_op, cpu_env, cc_op, tmp32_1, regs[r1], tmp2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x90: /* LLGC      R1,D2(X2,B2)     [RXY] */
-        tcg_gen_qemu_ld8u(regs[r1], addr, get_mem_index(s));
-        break;
-    case 0x91: /* LLGH      R1,D2(X2,B2)     [RXY] */
-        tcg_gen_qemu_ld16u(regs[r1], addr, get_mem_index(s));
-        break;
-    case 0x94: /* LLC     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8u(tmp2, addr, get_mem_index(s));
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x95: /* LLH     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld16u(tmp2, addr, get_mem_index(s));
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x96: /* ML      R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32u_i64(tmp3, tmp3);
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_mul_i64(tmp2, tmp2, tmp3);
-        store_reg32_i64((r1 + 1) & 15, tmp2);
-        tcg_gen_shri_i64(tmp2, tmp2, 32);
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+    switch (l + 1) {
+    case 1:
+        tcg_gen_qemu_ld8u(cc_src, o->addr1, get_mem_index(s));
+        tcg_gen_qemu_ld8u(cc_dst, o->in2, get_mem_index(s));
         break;
-    case 0x97: /* DL     R1,D2(X2,B2)     [RXY] */
-        /* reg(r1) = reg(r1, r1+1) % ld32(addr) */
-        /* reg(r1+1) = reg(r1, r1+1) / ld32(addr) */
-        tmp = load_reg(r1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32u_i64(tmp2, tmp2);
-        tcg_gen_ext32u_i64(tmp3, tmp3);
-        tcg_gen_shli_i64(tmp, tmp, 32);
-        tcg_gen_or_i64(tmp, tmp, tmp3);
-
-        tcg_gen_rem_i64(tmp3, tmp, tmp2);
-        tcg_gen_div_i64(tmp, tmp, tmp2);
-        store_reg32_i64((r1 + 1) & 15, tmp);
-        store_reg32_i64(r1, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+    case 2:
+        tcg_gen_qemu_ld16u(cc_src, o->addr1, get_mem_index(s));
+        tcg_gen_qemu_ld16u(cc_dst, o->in2, get_mem_index(s));
         break;
-    case 0x98: /* ALC     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp32_3 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        /* XXX possible optimization point */
-        gen_op_calc_cc(s);
-        gen_helper_addc_u32(tmp32_3, cc_op, tmp32_1, tmp32_2);
-        set_cc_addu32(s, tmp32_1, tmp32_2, tmp32_3);
-        store_reg32(r1, tmp32_3);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
+    case 4:
+        tcg_gen_qemu_ld32u(cc_src, o->addr1, get_mem_index(s));
+        tcg_gen_qemu_ld32u(cc_dst, o->in2, get_mem_index(s));
         break;
-    case 0x99: /* SLB     R1,D2(X2,B2)     [RXY] */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        /* XXX possible optimization point */
-        gen_op_calc_cc(s);
-        gen_helper_slb(cc_op, cpu_env, cc_op, tmp32_1, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
+    case 8:
+        tcg_gen_qemu_ld64(cc_src, o->addr1, get_mem_index(s));
+        tcg_gen_qemu_ld64(cc_dst, o->in2, get_mem_index(s));
         break;
     default:
-        LOG_DISAS("illegal e3 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 3);
-        break;
+        potential_page_fault(s);
+        vl = tcg_const_i32(l);
+        gen_helper_clc(cc_op, cpu_env, vl, o->addr1, o->in2);
+        tcg_temp_free_i32(vl);
+        set_cc_static(s);
+        return NO_EXIT;
     }
+    gen_op_update2_cc_i64(s, CC_OP_LTUGTU_64, cc_src, cc_dst);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clcle(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    potential_page_fault(s);
+    gen_helper_clcle(cc_op, cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clm(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 m3 = tcg_const_i32(get_field(s->fields, m3));
+    TCGv_i32 t1 = tcg_temp_new_i32();
+    tcg_gen_trunc_i64_i32(t1, o->in1);
+    potential_page_fault(s);
+    gen_helper_clm(cc_op, cpu_env, t1, m3, o->in2);
+    set_cc_static(s);
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(m3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_clst(DisasContext *s, DisasOps *o)
+{
+    potential_page_fault(s);
+    gen_helper_clst(o->in1, cpu_env, regs[0], o->in1, o->in2);
+    set_cc_static(s);
+    return_low128(o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cps(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 t = tcg_temp_new_i64();
+    tcg_gen_andi_i64(t, o->in1, 0x8000000000000000ull);
+    tcg_gen_andi_i64(o->out, o->in2, 0x7fffffffffffffffull);
+    tcg_gen_or_i64(o->out, o->out, t);
+    tcg_temp_free_i64(t);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cs(DisasContext *s, DisasOps *o)
+{
+    /* FIXME: needs an atomic solution for CONFIG_USER_ONLY.  */
+    int d2 = get_field(s->fields, d2);
+    int b2 = get_field(s->fields, b2);
+    int is_64 = s->insn->data;
+    TCGv_i64 addr, mem, cc, z;
+
+    /* Note that in1 = R3 (new value) and
+       in2 = (zero-extended) R1 (expected value).  */
+
+    /* Load the memory into the (temporary) output.  While the PoO only talks
+       about moving the memory to R1 on inequality, if we include equality it
+       means that R1 is equal to the memory in all conditions.  */
+    addr = get_address(s, 0, b2, d2);
+    if (is_64) {
+        tcg_gen_qemu_ld64(o->out, addr, get_mem_index(s));
+    } else {
+        tcg_gen_qemu_ld32u(o->out, addr, get_mem_index(s));
+    }
+
+    /* Are the memory and expected values (un)equal?  Note that this setcond
+       produces the output CC value, thus the NE sense of the test.  */
+    cc = tcg_temp_new_i64();
+    tcg_gen_setcond_i64(TCG_COND_NE, cc, o->in2, o->out);
+
+    /* If the memory and expected values are equal (CC==0), copy R3 to MEM.
+       Recall that we are allowed to unconditionally issue the store (and
+       thus any possible write trap), so (re-)store the original contents
+       of MEM in case of inequality.  */
+    z = tcg_const_i64(0);
+    mem = tcg_temp_new_i64();
+    tcg_gen_movcond_i64(TCG_COND_EQ, mem, cc, z, o->in1, o->out);
+    if (is_64) {
+        tcg_gen_qemu_st64(mem, addr, get_mem_index(s));
+    } else {
+        tcg_gen_qemu_st32(mem, addr, get_mem_index(s));
+    }
+    tcg_temp_free_i64(z);
+    tcg_temp_free_i64(mem);
     tcg_temp_free_i64(addr);
+
+    /* Store CC back to cc_op.  Wait until after the store so that any
+       exception gets the old cc_op value.  */
+    tcg_gen_trunc_i64_i32(cc_op, cc);
+    tcg_temp_free_i64(cc);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_cdsg(DisasContext *s, DisasOps *o)
+{
+    /* FIXME: needs an atomic solution for CONFIG_USER_ONLY.  */
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    int d2 = get_field(s->fields, d2);
+    int b2 = get_field(s->fields, b2);
+    TCGv_i64 addrh, addrl, memh, meml, outh, outl, cc, z;
+
+    /* Note that R1:R1+1 = expected value and R3:R3+1 = new value.  */
+
+    addrh = get_address(s, 0, b2, d2);
+    addrl = get_address(s, 0, b2, d2 + 8);
+    outh = tcg_temp_new_i64();
+    outl = tcg_temp_new_i64();
+
+    tcg_gen_qemu_ld64(outh, addrh, get_mem_index(s));
+    tcg_gen_qemu_ld64(outl, addrl, get_mem_index(s));
+
+    /* Fold the double-word compare with arithmetic.  */
+    cc = tcg_temp_new_i64();
+    z = tcg_temp_new_i64();
+    tcg_gen_xor_i64(cc, outh, regs[r1]);
+    tcg_gen_xor_i64(z, outl, regs[r1 + 1]);
+    tcg_gen_or_i64(cc, cc, z);
+    tcg_gen_movi_i64(z, 0);
+    tcg_gen_setcond_i64(TCG_COND_NE, cc, cc, z);
+
+    memh = tcg_temp_new_i64();
+    meml = tcg_temp_new_i64();
+    tcg_gen_movcond_i64(TCG_COND_EQ, memh, cc, z, regs[r3], outh);
+    tcg_gen_movcond_i64(TCG_COND_EQ, meml, cc, z, regs[r3 + 1], outl);
+    tcg_temp_free_i64(z);
+
+    tcg_gen_qemu_st64(memh, addrh, get_mem_index(s));
+    tcg_gen_qemu_st64(meml, addrl, get_mem_index(s));
+    tcg_temp_free_i64(memh);
+    tcg_temp_free_i64(meml);
+    tcg_temp_free_i64(addrh);
+    tcg_temp_free_i64(addrl);
+
+    /* Save back state now that we've passed all exceptions.  */
+    tcg_gen_mov_i64(regs[r1], outh);
+    tcg_gen_mov_i64(regs[r1 + 1], outl);
+    tcg_gen_trunc_i64_i32(cc_op, cc);
+    tcg_temp_free_i64(outh);
+    tcg_temp_free_i64(outl);
+    tcg_temp_free_i64(cc);
+    set_cc_static(s);
+    return NO_EXIT;
 }
 
 #ifndef CONFIG_USER_ONLY
-static void disas_e5(CPUS390XState *env, DisasContext* s, uint64_t insn)
+static ExitStatus op_csp(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2;
-    int op = (insn >> 32) & 0xff;
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    check_privileged(s);
+    gen_helper_csp(cc_op, cpu_env, r1, o->in2);
+    tcg_temp_free_i32(r1);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+#endif
 
-    tmp = get_address(s, 0, (insn >> 28) & 0xf, (insn >> 16) & 0xfff);
-    tmp2 = get_address(s, 0, (insn >> 12) & 0xf, insn & 0xfff);
+static ExitStatus op_cvd(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 t1 = tcg_temp_new_i64();
+    TCGv_i32 t2 = tcg_temp_new_i32();
+    tcg_gen_trunc_i64_i32(t2, o->in1);
+    gen_helper_cvd(t1, t2);
+    tcg_temp_free_i32(t2);
+    tcg_gen_qemu_st64(t1, o->in2, get_mem_index(s));
+    tcg_temp_free_i64(t1);
+    return NO_EXIT;
+}
 
-    LOG_DISAS("disas_e5: insn %" PRIx64 "\n", insn);
-    switch (op) {
-    case 0x01: /* TPROT    D1(B1),D2(B2)  [SSE] */
-        /* Test Protection */
-        potential_page_fault(s);
-        gen_helper_tprot(cc_op, tmp, tmp2);
-        set_cc_static(s);
-        break;
-    default:
-        LOG_DISAS("illegal e5 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 3);
-        break;
+static ExitStatus op_ct(DisasContext *s, DisasOps *o)
+{
+    int m3 = get_field(s->fields, m3);
+    int lab = gen_new_label();
+    TCGv_i32 t;
+    TCGCond c;
+
+    c = tcg_invert_cond(ltgt_cond[m3]);
+    if (s->insn->data) {
+        c = tcg_unsigned_cond(c);
     }
+    tcg_gen_brcond_i64(c, o->in1, o->in2, lab);
 
-    tcg_temp_free_i64(tmp);
-    tcg_temp_free_i64(tmp2);
+    /* Set DXC to 0xff.  */
+    t = tcg_temp_new_i32();
+    tcg_gen_ld_i32(t, cpu_env, offsetof(CPUS390XState, fpc));
+    tcg_gen_ori_i32(t, t, 0xff00);
+    tcg_gen_st_i32(t, cpu_env, offsetof(CPUS390XState, fpc));
+    tcg_temp_free_i32(t);
+
+    /* Trap.  */
+    gen_program_exception(s, PGM_DATA);
+
+    gen_set_label(lab);
+    return NO_EXIT;
+}
+
+#ifndef CONFIG_USER_ONLY
+static ExitStatus op_diag(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 tmp;
+
+    check_privileged(s);
+    potential_page_fault(s);
+
+    /* We pretend the format is RX_a so that D2 is the field we want.  */
+    tmp = tcg_const_i32(get_field(s->fields, d2) & 0xfff);
+    gen_helper_diag(regs[2], cpu_env, tmp, regs[2], regs[1]);
+    tcg_temp_free_i32(tmp);
+    return NO_EXIT;
 }
 #endif
 
-static void disas_eb(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int r3, int b2, int d2)
+static ExitStatus op_divs32(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2, tmp3, tmp4;
-    TCGv_i32 tmp32_1, tmp32_2;
-    int i, stm_len;
-    int ilc = 3;
+    gen_helper_divs32(o->out2, cpu_env, o->in1, o->in2);
+    return_low128(o->out);
+    return NO_EXIT;
+}
 
-    LOG_DISAS("disas_eb: op 0x%x r1 %d r3 %d b2 %d d2 0x%x\n",
-              op, r1, r3, b2, d2);
-    switch (op) {
-    case 0xc: /* SRLG     R1,R3,D2(B2)     [RSY] */
-    case 0xd: /* SLLG     R1,R3,D2(B2)     [RSY] */
-    case 0xa: /* SRAG     R1,R3,D2(B2)     [RSY] */
-    case 0xb: /* SLAG     R1,R3,D2(B2)     [RSY] */
-    case 0x1c: /* RLLG     R1,R3,D2(B2)     [RSY] */
-        if (b2) {
-            tmp = get_address(s, 0, b2, d2);
-            tcg_gen_andi_i64(tmp, tmp, 0x3f);
-        } else {
-            tmp = tcg_const_i64(d2 & 0x3f);
-        }
-        switch (op) {
-        case 0xc:
-            tcg_gen_shr_i64(regs[r1], regs[r3], tmp);
-            break;
-        case 0xd:
-            tcg_gen_shl_i64(regs[r1], regs[r3], tmp);
-            break;
-        case 0xa:
-            tcg_gen_sar_i64(regs[r1], regs[r3], tmp);
-            break;
-        case 0xb:
-            tmp2 = tcg_temp_new_i64();
-            tmp3 = tcg_temp_new_i64();
-            gen_op_update2_cc_i64(s, CC_OP_SLAG, regs[r3], tmp);
-            tcg_gen_shl_i64(tmp2, regs[r3], tmp);
-            /* override sign bit with source sign */
-            tcg_gen_andi_i64(tmp2, tmp2, ~0x8000000000000000ULL);
-            tcg_gen_andi_i64(tmp3, regs[r3], 0x8000000000000000ULL);
-            tcg_gen_or_i64(regs[r1], tmp2, tmp3);
-            tcg_temp_free_i64(tmp2);
-            tcg_temp_free_i64(tmp3);
-            break;
-        case 0x1c:
-            tcg_gen_rotl_i64(regs[r1], regs[r3], tmp);
-            break;
-        default:
-            tcg_abort();
-            break;
-        }
-        if (op == 0xa) {
-            set_cc_s64(s, regs[r1]);
-        }
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x1d: /* RLL    R1,R3,D2(B2)        [RSY] */
-        if (b2) {
-            tmp = get_address(s, 0, b2, d2);
-            tcg_gen_andi_i64(tmp, tmp, 0x3f);
-        } else {
-            tmp = tcg_const_i64(d2 & 0x3f);
-        }
-        tmp32_1 = tcg_temp_new_i32();
-        tmp32_2 = load_reg32(r3);
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        switch (op) {
-        case 0x1d:
-            tcg_gen_rotl_i32(tmp32_1, tmp32_2, tmp32_1);
-            break;
-        default:
-            tcg_abort();
-            break;
-        }
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
+static ExitStatus op_divu32(DisasContext *s, DisasOps *o)
+{
+    gen_helper_divu32(o->out2, cpu_env, o->in1, o->in2);
+    return_low128(o->out);
+    return NO_EXIT;
+}
+
+static ExitStatus op_divs64(DisasContext *s, DisasOps *o)
+{
+    gen_helper_divs64(o->out2, cpu_env, o->in1, o->in2);
+    return_low128(o->out);
+    return NO_EXIT;
+}
+
+static ExitStatus op_divu64(DisasContext *s, DisasOps *o)
+{
+    gen_helper_divu64(o->out2, cpu_env, o->out, o->out2, o->in2);
+    return_low128(o->out);
+    return NO_EXIT;
+}
+
+static ExitStatus op_deb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_deb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ddb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_ddb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_dxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_dxb(o->out, cpu_env, o->out, o->out2, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ear(DisasContext *s, DisasOps *o)
+{
+    int r2 = get_field(s->fields, r2);
+    tcg_gen_ld32u_i64(o->out, cpu_env, offsetof(CPUS390XState, aregs[r2]));
+    return NO_EXIT;
+}
+
+static ExitStatus op_efpc(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_ld32u_i64(o->out, cpu_env, offsetof(CPUS390XState, fpc));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ex(DisasContext *s, DisasOps *o)
+{
+    /* ??? Perhaps a better way to implement EXECUTE is to set a bit in
+       tb->flags, (ab)use the tb->cs_base field as the address of
+       the template in memory, and grab 8 bits of tb->flags/cflags for
+       the contents of the register.  We would then recognize all this
+       in gen_intermediate_code_internal, generating code for exactly
+       one instruction.  This new TB then gets executed normally.
+
+       On the other hand, this seems to be mostly used for modifying
+       MVC inside of memcpy, which needs a helper call anyway.  So
+       perhaps this doesn't bear thinking about any further.  */
+
+    TCGv_i64 tmp;
+
+    update_psw_addr(s);
+    update_cc_op(s);
+
+    tmp = tcg_const_i64(s->next_pc);
+    gen_helper_ex(cc_op, cpu_env, cc_op, o->in1, o->in2, tmp);
+    tcg_temp_free_i64(tmp);
+
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_flogr(DisasContext *s, DisasOps *o)
+{
+    /* We'll use the original input for cc computation, since we get to
+       compare that against 0, which ought to be better than comparing
+       the real output against 64.  It also lets cc_dst be a convenient
+       temporary during our computation.  */
+    gen_op_update1_cc_i64(s, CC_OP_FLOGR, o->in2);
+
+    /* R1 = IN ? CLZ(IN) : 64.  */
+    gen_helper_clz(o->out, o->in2);
+
+    /* R1+1 = IN & ~(found bit).  Note that we may attempt to shift this
+       value by 64, which is undefined.  But since the shift is 64 iff the
+       input is zero, we still get the correct result after and'ing.  */
+    tcg_gen_movi_i64(o->out2, 0x8000000000000000ull);
+    tcg_gen_shr_i64(o->out2, o->out2, o->out);
+    tcg_gen_andc_i64(o->out2, cc_dst, o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_icm(DisasContext *s, DisasOps *o)
+{
+    int m3 = get_field(s->fields, m3);
+    int pos, len, base = s->insn->data;
+    TCGv_i64 tmp = tcg_temp_new_i64();
+    uint64_t ccm;
+
+    switch (m3) {
+    case 0xf:
+        /* Effectively a 32-bit load.  */
+        tcg_gen_qemu_ld32u(tmp, o->in2, get_mem_index(s));
+        len = 32;
+        goto one_insert;
+
+    case 0xc:
+    case 0x6:
+    case 0x3:
+        /* Effectively a 16-bit load.  */
+        tcg_gen_qemu_ld16u(tmp, o->in2, get_mem_index(s));
+        len = 16;
+        goto one_insert;
+
+    case 0x8:
+    case 0x4:
+    case 0x2:
+    case 0x1:
+        /* Effectively an 8-bit load.  */
+        tcg_gen_qemu_ld8u(tmp, o->in2, get_mem_index(s));
+        len = 8;
+        goto one_insert;
+
+    one_insert:
+        pos = base + ctz32(m3) * 8;
+        tcg_gen_deposit_i64(o->out, o->out, tmp, pos, len);
+        ccm = ((1ull << len) - 1) << pos;
         break;
-    case 0x4:  /* LMG      R1,R3,D2(B2)     [RSE] */
-    case 0x24: /* STMG     R1,R3,D2(B2)     [RSE] */
-        stm_len = 8;
-        goto do_mh;
-    case 0x26: /* STMH     R1,R3,D2(B2)     [RSE] */
-    case 0x96: /* LMH      R1,R3,D2(B2)     [RSE] */
-        stm_len = 4;
-do_mh:
-        /* Apparently, unrolling lmg/stmg of any size gains performance -
-           even for very long ones... */
-        tmp = get_address(s, 0, b2, d2);
-        tmp3 = tcg_const_i64(stm_len);
-        tmp4 = tcg_const_i64(op == 0x26 ? 32 : 4);
-        for (i = r1;; i = (i + 1) % 16) {
-            switch (op) {
-            case 0x4:
-                tcg_gen_qemu_ld64(regs[i], tmp, get_mem_index(s));
-                break;
-            case 0x96:
-                tmp2 = tcg_temp_new_i64();
-#if HOST_LONG_BITS == 32
-                tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-                tcg_gen_trunc_i64_i32(TCGV_HIGH(regs[i]), tmp2);
-#else
-                tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-                tcg_gen_shl_i64(tmp2, tmp2, tmp4);
-                tcg_gen_ext32u_i64(regs[i], regs[i]);
-                tcg_gen_or_i64(regs[i], regs[i], tmp2);
-#endif
-                tcg_temp_free_i64(tmp2);
-                break;
-            case 0x24:
-                tcg_gen_qemu_st64(regs[i], tmp, get_mem_index(s));
-                break;
-            case 0x26:
-                tmp2 = tcg_temp_new_i64();
-                tcg_gen_shr_i64(tmp2, regs[i], tmp4);
-                tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-                tcg_temp_free_i64(tmp2);
-                break;
-            default:
-                tcg_abort();
-            }
-            if (i == r3) {
-                break;
+
+    default:
+        /* This is going to be a sequence of loads and inserts.  */
+        pos = base + 32 - 8;
+        ccm = 0;
+        while (m3) {
+            if (m3 & 0x8) {
+                tcg_gen_qemu_ld8u(tmp, o->in2, get_mem_index(s));
+                tcg_gen_addi_i64(o->in2, o->in2, 1);
+                tcg_gen_deposit_i64(o->out, o->out, tmp, pos, 8);
+                ccm |= 0xff << pos;
             }
-            tcg_gen_add_i64(tmp, tmp, tmp3);
+            m3 = (m3 << 1) & 0xf;
+            pos -= 8;
         }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i64(tmp4);
-        break;
-    case 0x2c: /* STCMH R1,M3,D2(B2) [RSY] */
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_stcmh(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
         break;
+    }
+
+    tcg_gen_movi_i64(tmp, ccm);
+    gen_op_update2_cc_i64(s, CC_OP_ICM, tmp, o->out);
+    tcg_temp_free_i64(tmp);
+    return NO_EXIT;
+}
+
+static ExitStatus op_insi(DisasContext *s, DisasOps *o)
+{
+    int shift = s->insn->data & 0xff;
+    int size = s->insn->data >> 8;
+    tcg_gen_deposit_i64(o->out, o->in1, o->in2, shift, size);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ipm(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 t1;
+
+    gen_op_calc_cc(s);
+    tcg_gen_andi_i64(o->out, o->out, ~0xff000000ull);
+
+    t1 = tcg_temp_new_i64();
+    tcg_gen_shli_i64(t1, psw_mask, 20);
+    tcg_gen_shri_i64(t1, t1, 36);
+    tcg_gen_or_i64(o->out, o->out, t1);
+
+    tcg_gen_extu_i32_i64(t1, cc_op);
+    tcg_gen_shli_i64(t1, t1, 28);
+    tcg_gen_or_i64(o->out, o->out, t1);
+    tcg_temp_free_i64(t1);
+    return NO_EXIT;
+}
+
 #ifndef CONFIG_USER_ONLY
-    case 0x2f: /* LCTLG     R1,R3,D2(B2)     [RSE] */
-        /* Load Control */
-        check_privileged(env, s, ilc);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_lctlg(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x25: /* STCTG     R1,R3,D2(B2)     [RSE] */
-        /* Store Control */
-        check_privileged(env, s, ilc);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_stctg(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
+static ExitStatus op_ipte(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_ipte(cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_iske(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_iske(o->out, cpu_env, o->in2);
+    return NO_EXIT;
+}
 #endif
-    case 0x30: /* CSG     R1,R3,D2(B2)     [RSY] */
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        /* XXX rewrite in tcg */
-        gen_helper_csg(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x3e: /* CDSG R1,R3,D2(B2) [RSY] */
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        /* XXX rewrite in tcg */
-        gen_helper_cdsg(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x51: /* TMY D1(B1),I2 [SIY] */
-        tmp = get_address(s, 0, b2, d2); /* SIY -> this is the destination */
-        tmp2 = tcg_const_i64((r1 << 4) | r3);
-        tcg_gen_qemu_ld8u(tmp, tmp, get_mem_index(s));
-        /* yes, this is a 32 bit operation with 64 bit tcg registers, because
-           that incurs less conversions */
-        cmp_64(s, tmp, tmp2, CC_OP_TM_32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x52: /* MVIY D1(B1),I2 [SIY] */
-        tmp = get_address(s, 0, b2, d2); /* SIY -> this is the destination */
-        tmp2 = tcg_const_i64((r1 << 4) | r3);
-        tcg_gen_qemu_st8(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x55: /* CLIY D1(B1),I2 [SIY] */
-        tmp3 = get_address(s, 0, b2, d2); /* SIY -> this is the 1st operand */
-        tmp = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld8u(tmp, tmp3, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        cmp_u32c(s, tmp32_1, (r1 << 4) | r3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x80: /* ICMH      R1,M3,D2(B2)     [RSY] */
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        /* XXX split CC calculation out */
-        gen_helper_icmh(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    default:
-        LOG_DISAS("illegal eb operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, ilc);
-        break;
-    }
+
+static ExitStatus op_ldeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_ldeb(o->out, cpu_env, o->in2);
+    return NO_EXIT;
 }
 
-static void disas_ed(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int x2, int b2, int d2, int r1b)
+static ExitStatus op_ledb(DisasContext *s, DisasOps *o)
 {
-    TCGv_i32 tmp_r1, tmp32;
-    TCGv_i64 addr, tmp;
-    addr = get_address(s, x2, b2, d2);
-    tmp_r1 = tcg_const_i32(r1);
-    switch (op) {
-    case 0x4: /* LDEB R1,D2(X2,B2) [RXE] */
-        potential_page_fault(s);
-        gen_helper_ldeb(cpu_env, tmp_r1, addr);
-        break;
-    case 0x5: /* LXDB R1,D2(X2,B2) [RXE] */
-        potential_page_fault(s);
-        gen_helper_lxdb(cpu_env, tmp_r1, addr);
-        break;
-    case 0x9: /* CEB    R1,D2(X2,B2)       [RXE] */
-        tmp = tcg_temp_new_i64();
-        tmp32 = load_freg32(r1);
-        tcg_gen_qemu_ld32u(tmp, addr, get_mem_index(s));
-        set_cc_cmp_f32_i64(s, tmp32, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32);
-        break;
-    case 0xa: /* AEB    R1,D2(X2,B2)       [RXE] */
-        tmp = tcg_temp_new_i64();
-        tmp32 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        gen_helper_aeb(cpu_env, tmp_r1, tmp32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32);
-
-        tmp32 = load_freg32(r1);
-        gen_set_cc_nz_f32(s, tmp32);
-        tcg_temp_free_i32(tmp32);
-        break;
-    case 0xb: /* SEB    R1,D2(X2,B2)       [RXE] */
-        tmp = tcg_temp_new_i64();
-        tmp32 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        gen_helper_seb(cpu_env, tmp_r1, tmp32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32);
-
-        tmp32 = load_freg32(r1);
-        gen_set_cc_nz_f32(s, tmp32);
-        tcg_temp_free_i32(tmp32);
-        break;
-    case 0xd: /* DEB    R1,D2(X2,B2)       [RXE] */
-        tmp = tcg_temp_new_i64();
-        tmp32 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        gen_helper_deb(cpu_env, tmp_r1, tmp32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32);
-        break;
-    case 0x10: /* TCEB   R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_tceb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x11: /* TCDB   R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_tcdb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x12: /* TCXB   R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_tcxb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x17: /* MEEB   R1,D2(X2,B2)       [RXE] */
-        tmp = tcg_temp_new_i64();
-        tmp32 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp, addr, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        gen_helper_meeb(cpu_env, tmp_r1, tmp32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32);
-        break;
-    case 0x19: /* CDB    R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_cdb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x1a: /* ADB    R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_adb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x1b: /* SDB    R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_sdb(cc_op, cpu_env, tmp_r1, addr);
-        set_cc_static(s);
-        break;
-    case 0x1c: /* MDB    R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_mdb(cpu_env, tmp_r1, addr);
-        break;
-    case 0x1d: /* DDB    R1,D2(X2,B2)       [RXE] */
-        potential_page_fault(s);
-        gen_helper_ddb(cpu_env, tmp_r1, addr);
-        break;
-    case 0x1e: /* MADB  R1,R3,D2(X2,B2) [RXF] */
-        /* for RXF insns, r1 is R3 and r1b is R1 */
-        tmp32 = tcg_const_i32(r1b);
-        potential_page_fault(s);
-        gen_helper_madb(cpu_env, tmp32, addr, tmp_r1);
-        tcg_temp_free_i32(tmp32);
-        break;
-    default:
-        LOG_DISAS("illegal ed operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 3);
-        return;
+    gen_helper_ledb(o->out, cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ldxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_ldxb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lexb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_lexb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lxdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_lxdb(o->out, cpu_env, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lxeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_lxeb(o->out, cpu_env, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_llgt(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_andi_i64(o->out, o->in2, 0x7fffffff);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld8s(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld8s(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld8u(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld8u(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld16s(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld16s(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld16u(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld16u(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld32s(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld32s(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld32u(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld32u(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_ld64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_ld64(o->out, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_loc(DisasContext *s, DisasOps *o)
+{
+    DisasCompare c;
+
+    disas_jcc(s, &c, get_field(s->fields, m3));
+
+    if (c.is_64) {
+        tcg_gen_movcond_i64(c.cond, o->out, c.u.s64.a, c.u.s64.b,
+                            o->in2, o->in1);
+        free_compare(&c);
+    } else {
+        TCGv_i32 t32 = tcg_temp_new_i32();
+        TCGv_i64 t, z;
+
+        tcg_gen_setcond_i32(c.cond, t32, c.u.s32.a, c.u.s32.b);
+        free_compare(&c);
+
+        t = tcg_temp_new_i64();
+        tcg_gen_extu_i32_i64(t, t32);
+        tcg_temp_free_i32(t32);
+
+        z = tcg_const_i64(0);
+        tcg_gen_movcond_i64(TCG_COND_NE, o->out, t, z, o->in2, o->in1);
+        tcg_temp_free_i64(t);
+        tcg_temp_free_i64(z);
     }
-    tcg_temp_free_i32(tmp_r1);
-    tcg_temp_free_i64(addr);
+
+    return NO_EXIT;
 }
 
-static void disas_a5(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int i2)
+#ifndef CONFIG_USER_ONLY
+static ExitStatus op_lctl(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2;
-    TCGv_i32 tmp32;
-    LOG_DISAS("disas_a5: op 0x%x r1 %d i2 0x%x\n", op, r1, i2);
-    switch (op) {
-    case 0x0: /* IIHH     R1,I2     [RI] */
-        tmp = tcg_const_i64(i2);
-        tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 48, 16);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x1: /* IIHL     R1,I2     [RI] */
-        tmp = tcg_const_i64(i2);
-        tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 32, 16);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x2: /* IILH     R1,I2     [RI] */
-        tmp = tcg_const_i64(i2);
-        tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 16, 16);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x3: /* IILL     R1,I2     [RI] */
-        tmp = tcg_const_i64(i2);
-        tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 0, 16);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x4: /* NIHH     R1,I2     [RI] */
-    case 0x8: /* OIHH     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tmp32 = tcg_temp_new_i32();
-        switch (op) {
-        case 0x4:
-            tmp2 = tcg_const_i64((((uint64_t)i2) << 48)
-                               | 0x0000ffffffffffffULL);
-            tcg_gen_and_i64(tmp, tmp, tmp2);
-            break;
-        case 0x8:
-            tmp2 = tcg_const_i64(((uint64_t)i2) << 48);
-            tcg_gen_or_i64(tmp, tmp, tmp2);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg(r1, tmp);
-        tcg_gen_shri_i64(tmp2, tmp, 48);
-        tcg_gen_trunc_i64_i32(tmp32, tmp2);
-        set_cc_nz_u32(s, tmp32);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x5: /* NIHL     R1,I2     [RI] */
-    case 0x9: /* OIHL     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tmp32 = tcg_temp_new_i32();
-        switch (op) {
-        case 0x5:
-            tmp2 = tcg_const_i64((((uint64_t)i2) << 32)
-                               | 0xffff0000ffffffffULL);
-            tcg_gen_and_i64(tmp, tmp, tmp2);
-            break;
-        case 0x9:
-            tmp2 = tcg_const_i64(((uint64_t)i2) << 32);
-            tcg_gen_or_i64(tmp, tmp, tmp2);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg(r1, tmp);
-        tcg_gen_shri_i64(tmp2, tmp, 32);
-        tcg_gen_trunc_i64_i32(tmp32, tmp2);
-        tcg_gen_andi_i32(tmp32, tmp32, 0xffff);
-        set_cc_nz_u32(s, tmp32);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x6: /* NILH     R1,I2     [RI] */
-    case 0xa: /* OILH     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tmp32 = tcg_temp_new_i32();
-        switch (op) {
-        case 0x6:
-            tmp2 = tcg_const_i64((((uint64_t)i2) << 16)
-                               | 0xffffffff0000ffffULL);
-            tcg_gen_and_i64(tmp, tmp, tmp2);
-            break;
-        case 0xa:
-            tmp2 = tcg_const_i64(((uint64_t)i2) << 16);
-            tcg_gen_or_i64(tmp, tmp, tmp2);
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_lctl(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lctlg(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_lctlg(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+static ExitStatus op_lra(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_lra(o->out, cpu_env, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lpsw(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 t1, t2;
+
+    check_privileged(s);
+
+    t1 = tcg_temp_new_i64();
+    t2 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld32u(t1, o->in2, get_mem_index(s));
+    tcg_gen_addi_i64(o->in2, o->in2, 4);
+    tcg_gen_qemu_ld32u(t2, o->in2, get_mem_index(s));
+    /* Convert the 32-bit PSW_MASK into the 64-bit PSW_MASK.  */
+    tcg_gen_shli_i64(t1, t1, 32);
+    gen_helper_load_psw(cpu_env, t1, t2);
+    tcg_temp_free_i64(t1);
+    tcg_temp_free_i64(t2);
+    return EXIT_NORETURN;
+}
+
+static ExitStatus op_lpswe(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 t1, t2;
+
+    check_privileged(s);
+
+    t1 = tcg_temp_new_i64();
+    t2 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld64(t1, o->in2, get_mem_index(s));
+    tcg_gen_addi_i64(o->in2, o->in2, 8);
+    tcg_gen_qemu_ld64(t2, o->in2, get_mem_index(s));
+    gen_helper_load_psw(cpu_env, t1, t2);
+    tcg_temp_free_i64(t1);
+    tcg_temp_free_i64(t2);
+    return EXIT_NORETURN;
+}
+#endif
+
+static ExitStatus op_lam(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    potential_page_fault(s);
+    gen_helper_lam(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lm32(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    TCGv_i64 t = tcg_temp_new_i64();
+    TCGv_i64 t4 = tcg_const_i64(4);
+
+    while (1) {
+        tcg_gen_qemu_ld32u(t, o->in2, get_mem_index(s));
+        store_reg32_i64(r1, t);
+        if (r1 == r3) {
             break;
-        default:
-            tcg_abort();
         }
-        store_reg(r1, tmp);
-        tcg_gen_shri_i64(tmp, tmp, 16);
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        tcg_gen_andi_i32(tmp32, tmp32, 0xffff);
-        set_cc_nz_u32(s, tmp32);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x7: /* NILL     R1,I2     [RI] */
-    case 0xb: /* OILL     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tmp32 = tcg_temp_new_i32();
-        switch (op) {
-        case 0x7:
-            tmp2 = tcg_const_i64(i2 | 0xffffffffffff0000ULL);
-            tcg_gen_and_i64(tmp, tmp, tmp2);
-            break;
-        case 0xb:
-            tmp2 = tcg_const_i64(i2);
-            tcg_gen_or_i64(tmp, tmp, tmp2);
+        tcg_gen_add_i64(o->in2, o->in2, t4);
+        r1 = (r1 + 1) & 15;
+    }
+
+    tcg_temp_free_i64(t);
+    tcg_temp_free_i64(t4);
+    return NO_EXIT;
+}
+
+static ExitStatus op_lmh(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    TCGv_i64 t = tcg_temp_new_i64();
+    TCGv_i64 t4 = tcg_const_i64(4);
+
+    while (1) {
+        tcg_gen_qemu_ld32u(t, o->in2, get_mem_index(s));
+        store_reg32h_i64(r1, t);
+        if (r1 == r3) {
             break;
-        default:
-            tcg_abort();
         }
-        store_reg(r1, tmp);
-        tcg_gen_trunc_i64_i32(tmp32, tmp);
-        tcg_gen_andi_i32(tmp32, tmp32, 0xffff);
-        set_cc_nz_u32(s, tmp32);        /* signedness should not matter here */
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xc: /* LLIHH     R1,I2     [RI] */
-        tmp = tcg_const_i64( ((uint64_t)i2) << 48 );
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xd: /* LLIHL     R1,I2     [RI] */
-        tmp = tcg_const_i64( ((uint64_t)i2) << 32 );
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xe: /* LLILH     R1,I2     [RI] */
-        tmp = tcg_const_i64( ((uint64_t)i2) << 16 );
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xf: /* LLILL     R1,I2     [RI] */
-        tmp = tcg_const_i64(i2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    default:
-        LOG_DISAS("illegal a5 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 2);
-        return;
+        tcg_gen_add_i64(o->in2, o->in2, t4);
+        r1 = (r1 + 1) & 15;
     }
+
+    tcg_temp_free_i64(t);
+    tcg_temp_free_i64(t4);
+    return NO_EXIT;
 }
 
-static void disas_a7(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int i2)
+static ExitStatus op_lm64(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
-    int l1;
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    TCGv_i64 t8 = tcg_const_i64(8);
 
-    LOG_DISAS("disas_a7: op 0x%x r1 %d i2 0x%x\n", op, r1, i2);
-    switch (op) {
-    case 0x0: /* TMLH or TMH     R1,I2     [RI] */
-    case 0x1: /* TMLL or TML     R1,I2     [RI] */
-    case 0x2: /* TMHH     R1,I2     [RI] */
-    case 0x3: /* TMHL     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tmp2 = tcg_const_i64((uint16_t)i2);
-        switch (op) {
-        case 0x0:
-            tcg_gen_shri_i64(tmp, tmp, 16);
-            break;
-        case 0x1:
-            break;
-        case 0x2:
-            tcg_gen_shri_i64(tmp, tmp, 48);
-            break;
-        case 0x3:
-            tcg_gen_shri_i64(tmp, tmp, 32);
+    while (1) {
+        tcg_gen_qemu_ld64(regs[r1], o->in2, get_mem_index(s));
+        if (r1 == r3) {
             break;
         }
-        tcg_gen_andi_i64(tmp, tmp, 0xffff);
-        cmp_64(s, tmp, tmp2, CC_OP_TM_64);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x4: /* brc m1, i2 */
-        gen_brc(r1, s, i2 * 2LL);
-        return;
-    case 0x5: /* BRAS     R1,I2     [RI] */
-        tmp = tcg_const_i64(pc_to_link_info(s, s->pc + 4));
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        gen_goto_tb(s, 0, s->pc + i2 * 2LL);
-        s->is_jmp = DISAS_TB_JUMP;
-        break;
-    case 0x6: /* BRCT     R1,I2     [RI] */
-        tmp32_1 = load_reg32(r1);
-        tcg_gen_subi_i32(tmp32_1, tmp32_1, 1);
-        store_reg32(r1, tmp32_1);
-        gen_update_cc_op(s);
-        l1 = gen_new_label();
-        tcg_gen_brcondi_i32(TCG_COND_EQ, tmp32_1, 0, l1);
-        gen_goto_tb(s, 0, s->pc + (i2 * 2LL));
-        gen_set_label(l1);
-        gen_goto_tb(s, 1, s->pc + 4);
-        s->is_jmp = DISAS_TB_JUMP;
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x7: /* BRCTG     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tcg_gen_subi_i64(tmp, tmp, 1);
-        store_reg(r1, tmp);
-        gen_update_cc_op(s);
-        l1 = gen_new_label();
-        tcg_gen_brcondi_i64(TCG_COND_EQ, tmp, 0, l1);
-        gen_goto_tb(s, 0, s->pc + (i2 * 2LL));
-        gen_set_label(l1);
-        gen_goto_tb(s, 1, s->pc + 4);
-        s->is_jmp = DISAS_TB_JUMP;
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x8: /* lhi r1, i2 */
-        tmp32_1 = tcg_const_i32(i2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x9: /* lghi r1, i2 */
-        tmp = tcg_const_i64(i2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xa: /* AHI     R1,I2     [RI] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp32_3 = tcg_const_i32(i2);
+        tcg_gen_add_i64(o->in2, o->in2, t8);
+        r1 = (r1 + 1) & 15;
+    }
 
-        if (i2 < 0) {
-            tcg_gen_subi_i32(tmp32_2, tmp32_1, -i2);
-        } else {
-            tcg_gen_add_i32(tmp32_2, tmp32_1, tmp32_3);
-        }
+    tcg_temp_free_i64(t8);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mov2(DisasContext *s, DisasOps *o)
+{
+    o->out = o->in2;
+    o->g_out = o->g_in2;
+    TCGV_UNUSED_I64(o->in2);
+    o->g_in2 = false;
+    return NO_EXIT;
+}
+
+static ExitStatus op_movx(DisasContext *s, DisasOps *o)
+{
+    o->out = o->in1;
+    o->out2 = o->in2;
+    o->g_out = o->g_in1;
+    o->g_out2 = o->g_in2;
+    TCGV_UNUSED_I64(o->in1);
+    TCGV_UNUSED_I64(o->in2);
+    o->g_in1 = o->g_in2 = false;
+    return NO_EXIT;
+}
 
-        store_reg32(r1, tmp32_2);
-        set_cc_add32(s, tmp32_1, tmp32_3, tmp32_2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
+static ExitStatus op_mvc(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 l = tcg_const_i32(get_field(s->fields, l1));
+    potential_page_fault(s);
+    gen_helper_mvc(cpu_env, l, o->addr1, o->in2);
+    tcg_temp_free_i32(l);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mvcl(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r2 = tcg_const_i32(get_field(s->fields, r2));
+    potential_page_fault(s);
+    gen_helper_mvcl(cc_op, cpu_env, r1, r2);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mvcle(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    potential_page_fault(s);
+    gen_helper_mvcle(cc_op, cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+#ifndef CONFIG_USER_ONLY
+static ExitStatus op_mvcp(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, l1);
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_mvcp(cc_op, cpu_env, regs[r1], o->addr1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mvcs(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, l1);
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_mvcs(cc_op, cpu_env, regs[r1], o->addr1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+#endif
+
+static ExitStatus op_mvpg(DisasContext *s, DisasOps *o)
+{
+    potential_page_fault(s);
+    gen_helper_mvpg(cpu_env, regs[0], o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mvst(DisasContext *s, DisasOps *o)
+{
+    potential_page_fault(s);
+    gen_helper_mvst(o->in1, cpu_env, regs[0], o->in1, o->in2);
+    set_cc_static(s);
+    return_low128(o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mul(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_mul_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mul128(DisasContext *s, DisasOps *o)
+{
+    gen_helper_mul128(o->out, cpu_env, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_meeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_meeb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mdeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_mdeb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_mdb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_mxb(o->out, cpu_env, o->out, o->out2, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mxdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_mxdb(o->out, cpu_env, o->out, o->out2, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_maeb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 r3 = load_freg32_i64(get_field(s->fields, r3));
+    gen_helper_maeb(o->out, cpu_env, o->in1, o->in2, r3);
+    tcg_temp_free_i64(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_madb(DisasContext *s, DisasOps *o)
+{
+    int r3 = get_field(s->fields, r3);
+    gen_helper_madb(o->out, cpu_env, o->in1, o->in2, fregs[r3]);
+    return NO_EXIT;
+}
+
+static ExitStatus op_mseb(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 r3 = load_freg32_i64(get_field(s->fields, r3));
+    gen_helper_mseb(o->out, cpu_env, o->in1, o->in2, r3);
+    tcg_temp_free_i64(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_msdb(DisasContext *s, DisasOps *o)
+{
+    int r3 = get_field(s->fields, r3);
+    gen_helper_msdb(o->out, cpu_env, o->in1, o->in2, fregs[r3]);
+    return NO_EXIT;
+}
+
+static ExitStatus op_nabs(DisasContext *s, DisasOps *o)
+{
+    gen_helper_nabs_i64(o->out, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_nabsf32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_ori_i64(o->out, o->in2, 0x80000000ull);
+    return NO_EXIT;
+}
+
+static ExitStatus op_nabsf64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_ori_i64(o->out, o->in2, 0x8000000000000000ull);
+    return NO_EXIT;
+}
+
+static ExitStatus op_nabsf128(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_ori_i64(o->out, o->in1, 0x8000000000000000ull);
+    tcg_gen_mov_i64(o->out2, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_nc(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 l = tcg_const_i32(get_field(s->fields, l1));
+    potential_page_fault(s);
+    gen_helper_nc(cc_op, cpu_env, l, o->addr1, o->in2);
+    tcg_temp_free_i32(l);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_neg(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_neg_i64(o->out, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_negf32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_xori_i64(o->out, o->in2, 0x80000000ull);
+    return NO_EXIT;
+}
+
+static ExitStatus op_negf64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_xori_i64(o->out, o->in2, 0x8000000000000000ull);
+    return NO_EXIT;
+}
+
+static ExitStatus op_negf128(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_xori_i64(o->out, o->in1, 0x8000000000000000ull);
+    tcg_gen_mov_i64(o->out2, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_oc(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 l = tcg_const_i32(get_field(s->fields, l1));
+    potential_page_fault(s);
+    gen_helper_oc(cc_op, cpu_env, l, o->addr1, o->in2);
+    tcg_temp_free_i32(l);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_or(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_or_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ori(DisasContext *s, DisasOps *o)
+{
+    int shift = s->insn->data & 0xff;
+    int size = s->insn->data >> 8;
+    uint64_t mask = ((1ull << size) - 1) << shift;
+
+    assert(!o->g_in2);
+    tcg_gen_shli_i64(o->in2, o->in2, shift);
+    tcg_gen_or_i64(o->out, o->in1, o->in2);
+
+    /* Produce the CC from only the bits manipulated.  */
+    tcg_gen_andi_i64(cc_dst, o->out, mask);
+    set_cc_nz_u64(s, cc_dst);
+    return NO_EXIT;
+}
+
+static ExitStatus op_popcnt(DisasContext *s, DisasOps *o)
+{
+    gen_helper_popcnt(o->out, o->in2);
+    return NO_EXIT;
+}
+
+#ifndef CONFIG_USER_ONLY
+static ExitStatus op_ptlb(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_ptlb(cpu_env);
+    return NO_EXIT;
+}
+#endif
+
+static ExitStatus op_risbg(DisasContext *s, DisasOps *o)
+{
+    int i3 = get_field(s->fields, i3);
+    int i4 = get_field(s->fields, i4);
+    int i5 = get_field(s->fields, i5);
+    int do_zero = i4 & 0x80;
+    uint64_t mask, imask, pmask;
+    int pos, len, rot;
+
+    /* Adjust the arguments for the specific insn.  */
+    switch (s->fields->op2) {
+    case 0x55: /* risbg */
+        i3 &= 63;
+        i4 &= 63;
+        pmask = ~0;
+        break;
+    case 0x5d: /* risbhg */
+        i3 &= 31;
+        i4 &= 31;
+        pmask = 0xffffffff00000000ull;
+        break;
+    case 0x51: /* risblg */
+        i3 &= 31;
+        i4 &= 31;
+        pmask = 0x00000000ffffffffull;
         break;
-    case 0xb: /* aghi r1, i2 */
-        tmp = load_reg(r1);
-        tmp2 = tcg_const_i64(i2);
+    default:
+        abort();
+    }
 
-        if (i2 < 0) {
-            tcg_gen_subi_i64(regs[r1], tmp, -i2);
+    /* MASK is the set of bits to be inserted from R2.
+       Take care for I3/I4 wraparound.  */
+    mask = pmask >> i3;
+    if (i3 <= i4) {
+        mask ^= pmask >> i4 >> 1;
+    } else {
+        mask |= ~(pmask >> i4 >> 1);
+    }
+    mask &= pmask;
+
+    /* IMASK is the set of bits to be kept from R1.  In the case of the high/low
+       insns, we need to keep the other half of the register.  */
+    imask = ~mask | ~pmask;
+    if (do_zero) {
+        if (s->fields->op2 == 0x55) {
+            imask = 0;
         } else {
-            tcg_gen_add_i64(regs[r1], tmp, tmp2);
+            imask = ~pmask;
         }
-        set_cc_add64(s, tmp, tmp2, regs[r1]);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0xc: /* MHI     R1,I2     [RI] */
-        tmp32_1 = load_reg32(r1);
-        tcg_gen_muli_i32(tmp32_1, tmp32_1, i2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xd: /* MGHI     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        tcg_gen_muli_i64(tmp, tmp, i2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
+    }
+
+    /* In some cases we can implement this with deposit, which can be more
+       efficient on some hosts.  */
+    if (~mask == imask && i3 <= i4) {
+        if (s->fields->op2 == 0x5d) {
+            i3 += 32, i4 += 32;
+        }
+        /* Note that we rotate the bits to be inserted to the lsb, not to
+           the position as described in the PoO.  */
+        len = i4 - i3 + 1;
+        pos = 63 - i4;
+        rot = (i5 - pos) & 63;
+    } else {
+        pos = len = -1;
+        rot = i5 & 63;
+    }
+
+    /* Rotate the input as necessary.  */
+    tcg_gen_rotli_i64(o->in2, o->in2, rot);
+
+    /* Insert the selected bits into the output.  */
+    if (pos >= 0) {
+        tcg_gen_deposit_i64(o->out, o->out, o->in2, pos, len);
+    } else if (imask == 0) {
+        tcg_gen_andi_i64(o->out, o->in2, mask);
+    } else {
+        tcg_gen_andi_i64(o->in2, o->in2, mask);
+        tcg_gen_andi_i64(o->out, o->out, imask);
+        tcg_gen_or_i64(o->out, o->out, o->in2);
+    }
+    return NO_EXIT;
+}
+
+static ExitStatus op_rosbg(DisasContext *s, DisasOps *o)
+{
+    int i3 = get_field(s->fields, i3);
+    int i4 = get_field(s->fields, i4);
+    int i5 = get_field(s->fields, i5);
+    uint64_t mask;
+
+    /* If this is a test-only form, arrange to discard the result.  */
+    if (i3 & 0x80) {
+        o->out = tcg_temp_new_i64();
+        o->g_out = false;
+    }
+
+    i3 &= 63;
+    i4 &= 63;
+    i5 &= 63;
+
+    /* MASK is the set of bits to be operated on from R2.
+       Take care for I3/I4 wraparound.  */
+    mask = ~0ull >> i3;
+    if (i3 <= i4) {
+        mask ^= ~0ull >> i4 >> 1;
+    } else {
+        mask |= ~(~0ull >> i4 >> 1);
+    }
+
+    /* Rotate the input as necessary.  */
+    tcg_gen_rotli_i64(o->in2, o->in2, i5);
+
+    /* Operate.  */
+    switch (s->fields->op2) {
+    case 0x55: /* AND */
+        tcg_gen_ori_i64(o->in2, o->in2, ~mask);
+        tcg_gen_and_i64(o->out, o->out, o->in2);
         break;
-    case 0xe: /* CHI     R1,I2     [RI] */
-        tmp32_1 = load_reg32(r1);
-        cmp_s32c(s, tmp32_1, i2);
-        tcg_temp_free_i32(tmp32_1);
+    case 0x56: /* OR */
+        tcg_gen_andi_i64(o->in2, o->in2, mask);
+        tcg_gen_or_i64(o->out, o->out, o->in2);
         break;
-    case 0xf: /* CGHI     R1,I2     [RI] */
-        tmp = load_reg(r1);
-        cmp_s64c(s, tmp, i2);
-        tcg_temp_free_i64(tmp);
+    case 0x57: /* XOR */
+        tcg_gen_andi_i64(o->in2, o->in2, mask);
+        tcg_gen_xor_i64(o->out, o->out, o->in2);
         break;
     default:
-        LOG_DISAS("illegal a7 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 2);
-        return;
+        abort();
     }
+
+    /* Set the CC.  */
+    tcg_gen_andi_i64(cc_dst, o->out, mask);
+    set_cc_nz_u64(s, cc_dst);
+    return NO_EXIT;
+}
+
+static ExitStatus op_rev16(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_bswap16_i64(o->out, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_rev32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_bswap32_i64(o->out, o->in2);
+    return NO_EXIT;
 }
 
-static void disas_b2(CPUS390XState *env, DisasContext *s, int op,
-                     uint32_t insn)
+static ExitStatus op_rev64(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2, tmp3;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
-    int r1, r2;
-    int ilc = 2;
+    tcg_gen_bswap64_i64(o->out, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_rll32(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 t1 = tcg_temp_new_i32();
+    TCGv_i32 t2 = tcg_temp_new_i32();
+    TCGv_i32 to = tcg_temp_new_i32();
+    tcg_gen_trunc_i64_i32(t1, o->in1);
+    tcg_gen_trunc_i64_i32(t2, o->in2);
+    tcg_gen_rotl_i32(to, t1, t2);
+    tcg_gen_extu_i32_i64(o->out, to);
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t2);
+    tcg_temp_free_i32(to);
+    return NO_EXIT;
+}
+
+static ExitStatus op_rll64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_rotl_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
 #ifndef CONFIG_USER_ONLY
-    int r3, d2, b2;
+static ExitStatus op_rrbe(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_rrbe(cc_op, cpu_env, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sacf(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_sacf(cpu_env, o->in2);
+    /* Addressing mode has changed, so end the block.  */
+    return EXIT_PC_STALE;
+}
 #endif
 
-    r1 = (insn >> 4) & 0xf;
-    r2 = insn & 0xf;
+static ExitStatus op_sar(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    tcg_gen_st32_i64(o->in2, cpu_env, offsetof(CPUS390XState, aregs[r1]));
+    return NO_EXIT;
+}
 
-    LOG_DISAS("disas_b2: op 0x%x r1 %d r2 %d\n", op, r1, r2);
+static ExitStatus op_seb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_seb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
 
-    switch (op) {
-    case 0x22: /* IPM    R1               [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        gen_op_calc_cc(s);
-        gen_helper_ipm(cpu_env, cc_op, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x41: /* CKSM    R1,R2     [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        potential_page_fault(s);
-        gen_helper_cksm(cpu_env, tmp32_1, tmp32_2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        gen_op_movi_cc(s, 0);
-        break;
-    case 0x4e: /* SAR     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_st_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, aregs[r1]));
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x4f: /* EAR     R1,R2     [RRE] */
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_ld_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, aregs[r2]));
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x52: /* MSR     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r2);
-        tcg_gen_mul_i32(tmp32_1, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x54: /* MVPG     R1,R2     [RRE] */
-        tmp = load_reg(0);
-        tmp2 = load_reg(r1);
-        tmp3 = load_reg(r2);
-        potential_page_fault(s);
-        gen_helper_mvpg(cpu_env, tmp, tmp2, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        /* XXX check CCO bit and set CC accordingly */
-        gen_op_movi_cc(s, 0);
-        break;
-    case 0x55: /* MVST     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(0);
-        tmp32_2 = tcg_const_i32(r1);
-        tmp32_3 = tcg_const_i32(r2);
-        potential_page_fault(s);
-        gen_helper_mvst(cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        gen_op_movi_cc(s, 1);
-        break;
-    case 0x5d: /* CLST     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(0);
-        tmp32_2 = tcg_const_i32(r1);
-        tmp32_3 = tcg_const_i32(r2);
-        potential_page_fault(s);
-        gen_helper_clst(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x5e: /* SRST     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(0);
-        tmp32_2 = tcg_const_i32(r1);
-        tmp32_3 = tcg_const_i32(r2);
-        potential_page_fault(s);
-        gen_helper_srst(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
+static ExitStatus op_sdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sdb(o->out, cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sxb(o->out, cpu_env, o->out, o->out2, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sqeb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sqeb(o->out, cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sqdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sqdb(o->out, cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sqxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sqxb(o->out, cpu_env, o->in1, o->in2);
+    return_low128(o->out2);
+    return NO_EXIT;
+}
 
 #ifndef CONFIG_USER_ONLY
-    case 0x02: /* STIDP     D2(B2)     [S] */
-        /* Store CPU ID */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_stidp(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x04: /* SCK       D2(B2)     [S] */
-        /* Set Clock */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_sck(cc_op, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x05: /* STCK     D2(B2)     [S] */
-        /* Store Clock */
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_stck(cc_op, cpu_env, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x06: /* SCKC     D2(B2)     [S] */
-        /* Set Clock Comparator */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_sckc(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x07: /* STCKC    D2(B2)     [S] */
-        /* Store Clock Comparator */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_stckc(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x08: /* SPT      D2(B2)     [S] */
-        /* Set CPU Timer */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_spt(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x09: /* STPT     D2(B2)     [S] */
-        /* Store CPU Timer */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_stpt(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x0a: /* SPKA     D2(B2)     [S] */
-        /* Set PSW Key from Address */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_andi_i64(tmp2, psw_mask, ~PSW_MASK_KEY);
-        tcg_gen_shli_i64(tmp, tmp, PSW_SHIFT_KEY - 4);
-        tcg_gen_or_i64(psw_mask, tmp2, tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x0d: /* PTLB                [S] */
-        /* Purge TLB */
-        check_privileged(env, s, ilc);
-        gen_helper_ptlb(cpu_env);
-        break;
-    case 0x10: /* SPX      D2(B2)     [S] */
-        /* Set Prefix Register */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_spx(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x11: /* STPX     D2(B2)     [S] */
-        /* Store Prefix */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_ld_i64(tmp2, cpu_env, offsetof(CPUS390XState, psa));
-        tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x12: /* STAP     D2(B2)     [S] */
-        /* Store CPU Address */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_ld_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, cpu_num));
-        tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-        tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x21: /* IPTE     R1,R2      [RRE] */
-        /* Invalidate PTE */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        gen_helper_ipte(cpu_env, tmp, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x29: /* ISKE     R1,R2      [RRE] */
-        /* Insert Storage Key Extended */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp = load_reg(r2);
-        tmp2 = tcg_temp_new_i64();
-        gen_helper_iske(tmp2, cpu_env, tmp);
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x2a: /* RRBE     R1,R2      [RRE] */
-        /* Set Storage Key Extended */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp32_1 = load_reg32(r1);
-        tmp = load_reg(r2);
-        gen_helper_rrbe(cc_op, cpu_env, tmp32_1, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x2b: /* SSKE     R1,R2      [RRE] */
-        /* Set Storage Key Extended */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp32_1 = load_reg32(r1);
-        tmp = load_reg(r2);
-        gen_helper_sske(cpu_env, tmp32_1, tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x34: /* STCH ? */
-        /* Store Subchannel */
-        check_privileged(env, s, ilc);
-        gen_op_movi_cc(s, 3);
-        break;
-    case 0x46: /* STURA    R1,R2      [RRE] */
-        /* Store Using Real Address */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp32_1 = load_reg32(r1);
-        tmp = load_reg(r2);
-        potential_page_fault(s);
-        gen_helper_stura(cpu_env, tmp, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x50: /* CSP      R1,R2      [RRE] */
-        /* Compare And Swap And Purge */
-        check_privileged(env, s, ilc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        gen_helper_csp(cc_op, cpu_env, tmp32_1, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x5f: /* CHSC ? */
-        /* Channel Subsystem Call */
-        check_privileged(env, s, ilc);
-        gen_op_movi_cc(s, 3);
-        break;
-    case 0x78: /* STCKE    D2(B2)     [S] */
-        /* Store Clock Extended */
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_stcke(cc_op, cpu_env, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x79: /* SACF    D2(B2)     [S] */
-        /* Store Clock Extended */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        potential_page_fault(s);
-        gen_helper_sacf(cpu_env, tmp);
-        tcg_temp_free_i64(tmp);
-        /* addressing mode has changed, so end the block */
-        s->pc += ilc * 2;
-        update_psw_addr(s);
-        s->is_jmp = DISAS_EXCP;
-        break;
-    case 0x7d: /* STSI     D2,(B2)     [S] */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = load_reg32(0);
-        tmp32_2 = load_reg32(1);
-        potential_page_fault(s);
-        gen_helper_stsi(cc_op, cpu_env, tmp, tmp32_1, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x9d: /* LFPC      D2(B2)   [S] */
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        tcg_gen_st_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, fpc));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xb1: /* STFL     D2(B2)     [S] */
-        /* Store Facility List (CPU features) at 200 */
-        check_privileged(env, s, ilc);
-        tmp2 = tcg_const_i64(0xc0000000);
-        tmp = tcg_const_i64(200);
-        tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xb2: /* LPSWE    D2(B2)     [S] */
-        /* Load PSW Extended */
-        check_privileged(env, s, ilc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld64(tmp2, tmp, get_mem_index(s));
-        tcg_gen_addi_i64(tmp, tmp, 8);
-        tcg_gen_qemu_ld64(tmp3, tmp, get_mem_index(s));
-        gen_helper_load_psw(cpu_env, tmp2, tmp3);
-        /* we need to keep cc_op intact */
-        s->is_jmp = DISAS_JUMP;
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x20: /* SERVC     R1,R2     [RRE] */
-        /* SCLP Service call (PV hypercall) */
-        check_privileged(env, s, ilc);
-        potential_page_fault(s);
-        tmp32_1 = load_reg32(r2);
-        tmp = load_reg(r1);
-        gen_helper_servc(cc_op, cpu_env, tmp32_1, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
+static ExitStatus op_servc(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_servc(cc_op, cpu_env, o->in2, o->in1);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sigp(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_sigp(cc_op, cpu_env, o->in2, r1, o->in1);
+    tcg_temp_free_i32(r1);
+    return NO_EXIT;
+}
 #endif
-    default:
-        LOG_DISAS("illegal b2 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, ilc);
-        break;
+
+static ExitStatus op_soc(DisasContext *s, DisasOps *o)
+{
+    DisasCompare c;
+    TCGv_i64 a;
+    int lab, r1;
+
+    disas_jcc(s, &c, get_field(s->fields, m3));
+
+    lab = gen_new_label();
+    if (c.is_64) {
+        tcg_gen_brcond_i64(c.cond, c.u.s64.a, c.u.s64.b, lab);
+    } else {
+        tcg_gen_brcond_i32(c.cond, c.u.s32.a, c.u.s32.b, lab);
     }
+    free_compare(&c);
+
+    r1 = get_field(s->fields, r1);
+    a = get_address(s, 0, get_field(s->fields, b2), get_field(s->fields, d2));
+    if (s->insn->data) {
+        tcg_gen_qemu_st64(regs[r1], a, get_mem_index(s));
+    } else {
+        tcg_gen_qemu_st32(regs[r1], a, get_mem_index(s));
+    }
+    tcg_temp_free_i64(a);
+
+    gen_set_label(lab);
+    return NO_EXIT;
 }
 
-static void disas_b3(CPUS390XState *env, DisasContext *s, int op, int m3,
-                     int r1, int r2)
+static ExitStatus op_sla(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
-    LOG_DISAS("disas_b3: op 0x%x m3 0x%x r1 %d r2 %d\n", op, m3, r1, r2);
-#define FP_HELPER(i) \
-    tmp32_1 = tcg_const_i32(r1); \
-    tmp32_2 = tcg_const_i32(r2); \
-    gen_helper_ ## i(cpu_env, tmp32_1, tmp32_2); \
-    tcg_temp_free_i32(tmp32_1); \
-    tcg_temp_free_i32(tmp32_2);
-
-#define FP_HELPER_CC(i) \
-    tmp32_1 = tcg_const_i32(r1); \
-    tmp32_2 = tcg_const_i32(r2); \
-    gen_helper_ ## i(cc_op, cpu_env, tmp32_1, tmp32_2); \
-    set_cc_static(s); \
-    tcg_temp_free_i32(tmp32_1); \
-    tcg_temp_free_i32(tmp32_2);
+    uint64_t sign = 1ull << s->insn->data;
+    enum cc_op cco = s->insn->data == 31 ? CC_OP_SLA_32 : CC_OP_SLA_64;
+    gen_op_update2_cc_i64(s, cco, o->in1, o->in2);
+    tcg_gen_shl_i64(o->out, o->in1, o->in2);
+    /* The arithmetic left shift is curious in that it does not affect
+       the sign bit.  Copy that over from the source unchanged.  */
+    tcg_gen_andi_i64(o->out, o->out, ~sign);
+    tcg_gen_andi_i64(o->in1, o->in1, sign);
+    tcg_gen_or_i64(o->out, o->out, o->in1);
+    return NO_EXIT;
+}
 
-    switch (op) {
-    case 0x0: /* LPEBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lpebr);
-        break;
-    case 0x2: /* LTEBR       R1,R2             [RRE] */
-        FP_HELPER_CC(ltebr);
-        break;
-    case 0x3: /* LCEBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lcebr);
-        break;
-    case 0x4: /* LDEBR       R1,R2             [RRE] */
-        FP_HELPER(ldebr);
-        break;
-    case 0x5: /* LXDBR       R1,R2             [RRE] */
-        FP_HELPER(lxdbr);
-        break;
-    case 0x9: /* CEBR        R1,R2             [RRE] */
-        FP_HELPER_CC(cebr);
-        break;
-    case 0xa: /* AEBR        R1,R2             [RRE] */
-        FP_HELPER_CC(aebr);
-        break;
-    case 0xb: /* SEBR        R1,R2             [RRE] */
-        FP_HELPER_CC(sebr);
-        break;
-    case 0xd: /* DEBR        R1,R2             [RRE] */
-        FP_HELPER(debr);
-        break;
-    case 0x10: /* LPDBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lpdbr);
-        break;
-    case 0x12: /* LTDBR       R1,R2             [RRE] */
-        FP_HELPER_CC(ltdbr);
-        break;
-    case 0x13: /* LCDBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lcdbr);
-        break;
-    case 0x15: /* SQBDR       R1,R2             [RRE] */
-        FP_HELPER(sqdbr);
-        break;
-    case 0x17: /* MEEBR       R1,R2             [RRE] */
-        FP_HELPER(meebr);
-        break;
-    case 0x19: /* CDBR        R1,R2             [RRE] */
-        FP_HELPER_CC(cdbr);
-        break;
-    case 0x1a: /* ADBR        R1,R2             [RRE] */
-        FP_HELPER_CC(adbr);
-        break;
-    case 0x1b: /* SDBR        R1,R2             [RRE] */
-        FP_HELPER_CC(sdbr);
-        break;
-    case 0x1c: /* MDBR        R1,R2             [RRE] */
-        FP_HELPER(mdbr);
-        break;
-    case 0x1d: /* DDBR        R1,R2             [RRE] */
-        FP_HELPER(ddbr);
-        break;
-    case 0xe: /* MAEBR  R1,R3,R2 [RRF] */
-    case 0x1e: /* MADBR R1,R3,R2 [RRF] */
-    case 0x1f: /* MSDBR R1,R3,R2 [RRF] */
-        /* for RRF insns, m3 is R1, r1 is R3, and r2 is R2 */
-        tmp32_1 = tcg_const_i32(m3);
-        tmp32_2 = tcg_const_i32(r2);
-        tmp32_3 = tcg_const_i32(r1);
-        switch (op) {
-        case 0xe:
-            gen_helper_maebr(cpu_env, tmp32_1, tmp32_3, tmp32_2);
-            break;
-        case 0x1e:
-            gen_helper_madbr(cpu_env, tmp32_1, tmp32_3, tmp32_2);
-            break;
-        case 0x1f:
-            gen_helper_msdbr(cpu_env, tmp32_1, tmp32_3, tmp32_2);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x40: /* LPXBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lpxbr);
-        break;
-    case 0x42: /* LTXBR       R1,R2             [RRE] */
-        FP_HELPER_CC(ltxbr);
-        break;
-    case 0x43: /* LCXBR       R1,R2             [RRE] */
-        FP_HELPER_CC(lcxbr);
-        break;
-    case 0x44: /* LEDBR       R1,R2             [RRE] */
-        FP_HELPER(ledbr);
-        break;
-    case 0x45: /* LDXBR       R1,R2             [RRE] */
-        FP_HELPER(ldxbr);
-        break;
-    case 0x46: /* LEXBR       R1,R2             [RRE] */
-        FP_HELPER(lexbr);
-        break;
-    case 0x49: /* CXBR        R1,R2             [RRE] */
-        FP_HELPER_CC(cxbr);
-        break;
-    case 0x4a: /* AXBR        R1,R2             [RRE] */
-        FP_HELPER_CC(axbr);
-        break;
-    case 0x4b: /* SXBR        R1,R2             [RRE] */
-        FP_HELPER_CC(sxbr);
-        break;
-    case 0x4c: /* MXBR        R1,R2             [RRE] */
-        FP_HELPER(mxbr);
-        break;
-    case 0x4d: /* DXBR        R1,R2             [RRE] */
-        FP_HELPER(dxbr);
-        break;
-    case 0x65: /* LXR         R1,R2             [RRE] */
-        tmp = load_freg(r2);
-        store_freg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        tmp = load_freg(r2 + 2);
-        store_freg(r1 + 2, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x74: /* LZER        R1                [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        gen_helper_lzer(cpu_env, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x75: /* LZDR        R1                [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        gen_helper_lzdr(cpu_env, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x76: /* LZXR        R1                [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        gen_helper_lzxr(cpu_env, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x84: /* SFPC        R1                [RRE] */
-        tmp32_1 = load_reg32(r1);
-        tcg_gen_st_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, fpc));
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x8c: /* EFPC        R1                [RRE] */
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_ld_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, fpc));
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x94: /* CEFBR       R1,R2             [RRE] */
-    case 0x95: /* CDFBR       R1,R2             [RRE] */
-    case 0x96: /* CXFBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = load_reg32(r2);
-        switch (op) {
-        case 0x94:
-            gen_helper_cefbr(cpu_env, tmp32_1, tmp32_2);
-            break;
-        case 0x95:
-            gen_helper_cdfbr(cpu_env, tmp32_1, tmp32_2);
-            break;
-        case 0x96:
-            gen_helper_cxfbr(cpu_env, tmp32_1, tmp32_2);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x98: /* CFEBR       R1,R2             [RRE] */
-    case 0x99: /* CFDBR              R1,R2             [RRE] */
-    case 0x9a: /* CFXBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        tmp32_3 = tcg_const_i32(m3);
-        switch (op) {
-        case 0x98:
-            gen_helper_cfebr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x99:
-            gen_helper_cfdbr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x9a:
-            gen_helper_cfxbr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        default:
-            tcg_abort();
-        }
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0xa4: /* CEGBR       R1,R2             [RRE] */
-    case 0xa5: /* CDGBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp = load_reg(r2);
-        switch (op) {
-        case 0xa4:
-            gen_helper_cegbr(cpu_env, tmp32_1, tmp);
-            break;
-        case 0xa5:
-            gen_helper_cdgbr(cpu_env, tmp32_1, tmp);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xa6: /* CXGBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp = load_reg(r2);
-        gen_helper_cxgbr(cpu_env, tmp32_1, tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xa8: /* CGEBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        tmp32_3 = tcg_const_i32(m3);
-        gen_helper_cgebr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0xa9: /* CGDBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        tmp32_3 = tcg_const_i32(m3);
-        gen_helper_cgdbr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
+static ExitStatus op_sll(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_shl_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sra(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_sar_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_srl(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_shr_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sfpc(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sfpc(cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sfas(DisasContext *s, DisasOps *o)
+{
+    gen_helper_sfas(cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_srnm(DisasContext *s, DisasOps *o)
+{
+    int b2 = get_field(s->fields, b2);
+    int d2 = get_field(s->fields, d2);
+    TCGv_i64 t1 = tcg_temp_new_i64();
+    TCGv_i64 t2 = tcg_temp_new_i64();
+    int mask, pos, len;
+
+    switch (s->fields->op2) {
+    case 0x99: /* SRNM */
+        pos = 0, len = 2;
         break;
-    case 0xaa: /* CGXBR       R1,R2             [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        tmp32_3 = tcg_const_i32(m3);
-        gen_helper_cgxbr(cc_op, cpu_env, tmp32_1, tmp32_2, tmp32_3);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
+    case 0xb8: /* SRNMB */
+        pos = 0, len = 3;
         break;
+    case 0xb9: /* SRNMT */
+        pos = 4, len = 3;
     default:
-        LOG_DISAS("illegal b3 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 2);
-        break;
+        tcg_abort();
     }
+    mask = (1 << len) - 1;
 
-#undef FP_HELPER_CC
-#undef FP_HELPER
+    /* Insert the value into the appropriate field of the FPC.  */
+    if (b2 == 0) {
+        tcg_gen_movi_i64(t1, d2 & mask);
+    } else {
+        tcg_gen_addi_i64(t1, regs[b2], d2);
+        tcg_gen_andi_i64(t1, t1, mask);
+    }
+    tcg_gen_ld32u_i64(t2, cpu_env, offsetof(CPUS390XState, fpc));
+    tcg_gen_deposit_i64(t2, t2, t1, pos, len);
+    tcg_temp_free_i64(t1);
+
+    /* Then install the new FPC to set the rounding mode in fpu_status.  */
+    gen_helper_sfpc(cpu_env, t2);
+    tcg_temp_free_i64(t2);
+    return NO_EXIT;
 }
 
-static void disas_b9(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int r2)
+#ifndef CONFIG_USER_ONLY
+static ExitStatus op_spka(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2, tmp3;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
+    check_privileged(s);
+    tcg_gen_shri_i64(o->in2, o->in2, 4);
+    tcg_gen_deposit_i64(psw_mask, psw_mask, o->in2, PSW_SHIFT_KEY - 4, 4);
+    return NO_EXIT;
+}
 
-    LOG_DISAS("disas_b9: op 0x%x r1 %d r2 %d\n", op, r1, r2);
-    switch (op) {
-    case 0x0: /* LPGR     R1,R2     [RRE] */
-    case 0x1: /* LNGR     R1,R2     [RRE] */
-    case 0x2: /* LTGR R1,R2 [RRE] */
-    case 0x3: /* LCGR     R1,R2     [RRE] */
-    case 0x10: /* LPGFR R1,R2 [RRE] */
-    case 0x11: /* LNFGR     R1,R2     [RRE] */
-    case 0x12: /* LTGFR R1,R2 [RRE] */
-    case 0x13: /* LCGFR    R1,R2     [RRE] */
-        if (op & 0x10) {
-            tmp = load_reg32_i64(r2);
-        } else {
-            tmp = load_reg(r2);
-        }
-        switch (op & 0xf) {
-        case 0x0: /* LP?GR */
-            set_cc_abs64(s, tmp);
-            gen_helper_abs_i64(tmp, tmp);
-            store_reg(r1, tmp);
-            break;
-        case 0x1: /* LN?GR */
-            set_cc_nabs64(s, tmp);
-            gen_helper_nabs_i64(tmp, tmp);
-            store_reg(r1, tmp);
-            break;
-        case 0x2: /* LT?GR */
-            if (r1 != r2) {
-                store_reg(r1, tmp);
-            }
-            set_cc_s64(s, tmp);
-            break;
-        case 0x3: /* LC?GR */
-            tcg_gen_neg_i64(regs[r1], tmp);
-            set_cc_comp64(s, regs[r1]);
-            break;
-        }
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x4: /* LGR R1,R2 [RRE] */
-        store_reg(r1, regs[r2]);
-        break;
-    case 0x6: /* LGBR R1,R2 [RRE] */
-        tmp2 = load_reg(r2);
-        tcg_gen_ext8s_i64(tmp2, tmp2);
-        store_reg(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x8: /* AGR     R1,R2     [RRE] */
-    case 0xa: /* ALGR     R1,R2     [RRE] */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_add_i64(tmp3, tmp, tmp2);
-        store_reg(r1, tmp3);
-        switch (op) {
-        case 0x8:
-            set_cc_add64(s, tmp, tmp2, tmp3);
-            break;
-        case 0xa:
-            set_cc_addu64(s, tmp, tmp2, tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+static ExitStatus op_sske(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_sske(cpu_env, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_ssm(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    tcg_gen_deposit_i64(psw_mask, psw_mask, o->in2, 56, 8);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stap(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    /* ??? Surely cpu address != cpu number.  In any case the previous
+       version of this stored more than the required half-word, so it
+       is unlikely this has ever been tested.  */
+    tcg_gen_ld32u_i64(o->out, cpu_env, offsetof(CPUS390XState, cpu_num));
+    return NO_EXIT;
+}
+
+static ExitStatus op_stck(DisasContext *s, DisasOps *o)
+{
+    gen_helper_stck(o->out, cpu_env);
+    /* ??? We don't implement clock states.  */
+    gen_op_movi_cc(s, 0);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stcke(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 c1 = tcg_temp_new_i64();
+    TCGv_i64 c2 = tcg_temp_new_i64();
+    gen_helper_stck(c1, cpu_env);
+    /* Shift the 64-bit value into its place as a zero-extended
+       104-bit value.  Note that "bit positions 64-103 are always
+       non-zero so that they compare differently to STCK"; we set
+       the least significant bit to 1.  */
+    tcg_gen_shli_i64(c2, c1, 56);
+    tcg_gen_shri_i64(c1, c1, 8);
+    tcg_gen_ori_i64(c2, c2, 0x10000);
+    tcg_gen_qemu_st64(c1, o->in2, get_mem_index(s));
+    tcg_gen_addi_i64(o->in2, o->in2, 8);
+    tcg_gen_qemu_st64(c2, o->in2, get_mem_index(s));
+    tcg_temp_free_i64(c1);
+    tcg_temp_free_i64(c2);
+    /* ??? We don't implement clock states.  */
+    gen_op_movi_cc(s, 0);
+    return NO_EXIT;
+}
+
+static ExitStatus op_sckc(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_sckc(cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stckc(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_stckc(o->out, cpu_env);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stctg(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_stctg(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stctl(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_stctl(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stidp(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    tcg_gen_ld32u_i64(o->out, cpu_env, offsetof(CPUS390XState, cpu_num));
+    return NO_EXIT;
+}
+
+static ExitStatus op_spt(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_spt(cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stfl(DisasContext *s, DisasOps *o)
+{
+    TCGv_i64 f, a;
+    /* We really ought to have more complete indication of facilities
+       that we implement.  Address this when STFLE is implemented.  */
+    check_privileged(s);
+    f = tcg_const_i64(0xc0000000);
+    a = tcg_const_i64(200);
+    tcg_gen_qemu_st32(f, a, get_mem_index(s));
+    tcg_temp_free_i64(f);
+    tcg_temp_free_i64(a);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stpt(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_stpt(o->out, cpu_env);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stsi(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_stsi(cc_op, cpu_env, o->in2, regs[0], regs[1]);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_spx(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    gen_helper_spx(cpu_env, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_subchannel(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    /* Not operational.  */
+    gen_op_movi_cc(s, 3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stpx(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    tcg_gen_ld_i64(o->out, cpu_env, offsetof(CPUS390XState, psa));
+    tcg_gen_andi_i64(o->out, o->out, 0x7fffe000);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stnosm(DisasContext *s, DisasOps *o)
+{
+    uint64_t i2 = get_field(s->fields, i2);
+    TCGv_i64 t;
+
+    check_privileged(s);
+
+    /* It is important to do what the instruction name says: STORE THEN.
+       If we let the output hook perform the store then if we fault and
+       restart, we'll have the wrong SYSTEM MASK in place.  */
+    t = tcg_temp_new_i64();
+    tcg_gen_shri_i64(t, psw_mask, 56);
+    tcg_gen_qemu_st8(t, o->addr1, get_mem_index(s));
+    tcg_temp_free_i64(t);
+
+    if (s->fields->op == 0xac) {
+        tcg_gen_andi_i64(psw_mask, psw_mask,
+                         (i2 << 56) | 0x00ffffffffffffffull);
+    } else {
+        tcg_gen_ori_i64(psw_mask, psw_mask, i2 << 56);
+    }
+    return NO_EXIT;
+}
+
+static ExitStatus op_stura(DisasContext *s, DisasOps *o)
+{
+    check_privileged(s);
+    potential_page_fault(s);
+    gen_helper_stura(cpu_env, o->in2, o->in1);
+    return NO_EXIT;
+}
+#endif
+
+static ExitStatus op_st8(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_st8(o->in1, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_st16(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_st16(o->in1, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_st32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_st32(o->in1, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_st64(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_qemu_st64(o->in1, o->in2, get_mem_index(s));
+    return NO_EXIT;
+}
+
+static ExitStatus op_stam(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 r1 = tcg_const_i32(get_field(s->fields, r1));
+    TCGv_i32 r3 = tcg_const_i32(get_field(s->fields, r3));
+    potential_page_fault(s);
+    gen_helper_stam(cpu_env, r1, o->in2, r3);
+    tcg_temp_free_i32(r1);
+    tcg_temp_free_i32(r3);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stcm(DisasContext *s, DisasOps *o)
+{
+    int m3 = get_field(s->fields, m3);
+    int pos, base = s->insn->data;
+    TCGv_i64 tmp = tcg_temp_new_i64();
+
+    pos = base + ctz32(m3) * 8;
+    switch (m3) {
+    case 0xf:
+        /* Effectively a 32-bit store.  */
+        tcg_gen_shri_i64(tmp, o->in1, pos);
+        tcg_gen_qemu_st32(tmp, o->in2, get_mem_index(s));
         break;
-    case 0x9: /* SGR     R1,R2     [RRE] */
-    case 0xb: /* SLGR     R1,R2     [RRE] */
-    case 0x1b: /* SLGFR     R1,R2     [RRE] */
-    case 0x19: /* SGFR     R1,R2     [RRE] */
-        tmp = load_reg(r1);
-        switch (op) {
-        case 0x1b:
-            tmp32_1 = load_reg32(r2);
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-            tcg_temp_free_i32(tmp32_1);
-            break;
-        case 0x19:
-            tmp32_1 = load_reg32(r2);
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_ext_i32_i64(tmp2, tmp32_1);
-            tcg_temp_free_i32(tmp32_1);
-            break;
-        default:
-            tmp2 = load_reg(r2);
-            break;
-        }
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_sub_i64(tmp3, tmp, tmp2);
-        store_reg(r1, tmp3);
-        switch (op) {
-        case 0x9:
-        case 0x19:
-            set_cc_sub64(s, tmp, tmp2, tmp3);
-            break;
-        case 0xb:
-        case 0x1b:
-            set_cc_subu64(s, tmp, tmp2, tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+
+    case 0xc:
+    case 0x6:
+    case 0x3:
+        /* Effectively a 16-bit store.  */
+        tcg_gen_shri_i64(tmp, o->in1, pos);
+        tcg_gen_qemu_st16(tmp, o->in2, get_mem_index(s));
         break;
-    case 0xc: /* MSGR      R1,R2     [RRE] */
-    case 0x1c: /* MSGFR      R1,R2     [RRE] */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        if (op == 0x1c) {
-            tcg_gen_ext32s_i64(tmp2, tmp2);
-        }
-        tcg_gen_mul_i64(tmp, tmp, tmp2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
+
+    case 0x8:
+    case 0x4:
+    case 0x2:
+    case 0x1:
+        /* Effectively an 8-bit store.  */
+        tcg_gen_shri_i64(tmp, o->in1, pos);
+        tcg_gen_qemu_st8(tmp, o->in2, get_mem_index(s));
         break;
-    case 0xd: /* DSGR      R1,R2     [RRE] */
-    case 0x1d: /* DSGFR      R1,R2     [RRE] */
-        tmp = load_reg(r1 + 1);
-        if (op == 0xd) {
-            tmp2 = load_reg(r2);
-        } else {
-            tmp32_1 = load_reg32(r2);
-            tmp2 = tcg_temp_new_i64();
-            tcg_gen_ext_i32_i64(tmp2, tmp32_1);
-            tcg_temp_free_i32(tmp32_1);
+
+    default:
+        /* This is going to be a sequence of shifts and stores.  */
+        pos = base + 32 - 8;
+        while (m3) {
+            if (m3 & 0x8) {
+                tcg_gen_shri_i64(tmp, o->in1, pos);
+                tcg_gen_qemu_st8(tmp, o->in2, get_mem_index(s));
+                tcg_gen_addi_i64(o->in2, o->in2, 1);
+            }
+            m3 = (m3 << 1) & 0xf;
+            pos -= 8;
         }
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_div_i64(tmp3, tmp, tmp2);
-        store_reg(r1 + 1, tmp3);
-        tcg_gen_rem_i64(tmp3, tmp, tmp2);
-        store_reg(r1, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
         break;
-    case 0x14: /* LGFR     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tmp = tcg_temp_new_i64();
-        tcg_gen_ext_i32_i64(tmp, tmp32_1);
-        store_reg(r1, tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x16: /* LLGFR      R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tmp = tcg_temp_new_i64();
-        tcg_gen_extu_i32_i64(tmp, tmp32_1);
-        store_reg(r1, tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x17: /* LLGTR      R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tmp = tcg_temp_new_i64();
-        tcg_gen_andi_i32(tmp32_1, tmp32_1, 0x7fffffffUL);
-        tcg_gen_extu_i32_i64(tmp, tmp32_1);
-        store_reg(r1, tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x18: /* AGFR     R1,R2     [RRE] */
-    case 0x1a: /* ALGFR     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tmp2 = tcg_temp_new_i64();
-        if (op == 0x18) {
-            tcg_gen_ext_i32_i64(tmp2, tmp32_1);
-        } else {
-            tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-        }
-        tcg_temp_free_i32(tmp32_1);
-        tmp = load_reg(r1);
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_add_i64(tmp3, tmp, tmp2);
-        store_reg(r1, tmp3);
-        if (op == 0x18) {
-            set_cc_add64(s, tmp, tmp2, tmp3);
+    }
+    tcg_temp_free_i64(tmp);
+    return NO_EXIT;
+}
+
+static ExitStatus op_stm(DisasContext *s, DisasOps *o)
+{
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    int size = s->insn->data;
+    TCGv_i64 tsize = tcg_const_i64(size);
+
+    while (1) {
+        if (size == 8) {
+            tcg_gen_qemu_st64(regs[r1], o->in2, get_mem_index(s));
         } else {
-            set_cc_addu64(s, tmp, tmp2, tmp3);
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x0f: /* LRVGR    R1,R2     [RRE] */
-        tcg_gen_bswap64_i64(regs[r1], regs[r2]);
-        break;
-    case 0x1f: /* LRVR     R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_bswap32_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x20: /* CGR      R1,R2     [RRE] */
-    case 0x30: /* CGFR     R1,R2     [RRE] */
-        tmp2 = load_reg(r2);
-        if (op == 0x30) {
-            tcg_gen_ext32s_i64(tmp2, tmp2);
+            tcg_gen_qemu_st32(regs[r1], o->in2, get_mem_index(s));
         }
-        tmp = load_reg(r1);
-        cmp_s64(s, tmp, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x21: /* CLGR     R1,R2     [RRE] */
-    case 0x31: /* CLGFR    R1,R2     [RRE] */
-        tmp2 = load_reg(r2);
-        if (op == 0x31) {
-            tcg_gen_ext32u_i64(tmp2, tmp2);
-        }
-        tmp = load_reg(r1);
-        cmp_u64(s, tmp, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x26: /* LBR R1,R2 [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_ext8s_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x27: /* LHR R1,R2 [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_ext16s_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x80: /* NGR R1,R2 [RRE] */
-    case 0x81: /* OGR R1,R2 [RRE] */
-    case 0x82: /* XGR R1,R2 [RRE] */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        switch (op) {
-        case 0x80:
-            tcg_gen_and_i64(tmp, tmp, tmp2);
-            break;
-        case 0x81:
-            tcg_gen_or_i64(tmp, tmp, tmp2);
-            break;
-        case 0x82:
-            tcg_gen_xor_i64(tmp, tmp, tmp2);
+        if (r1 == r3) {
             break;
-        default:
-            tcg_abort();
         }
-        store_reg(r1, tmp);
-        set_cc_nz_u64(s, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x83: /* FLOGR R1,R2 [RRE] */
-        tmp = load_reg(r2);
-        tmp32_1 = tcg_const_i32(r1);
-        gen_helper_flogr(cc_op, cpu_env, tmp32_1, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x84: /* LLGCR R1,R2 [RRE] */
-        tmp = load_reg(r2);
-        tcg_gen_andi_i64(tmp, tmp, 0xff);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x85: /* LLGHR R1,R2 [RRE] */
-        tmp = load_reg(r2);
-        tcg_gen_andi_i64(tmp, tmp, 0xffff);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x87: /* DLGR      R1,R2     [RRE] */
-        tmp32_1 = tcg_const_i32(r1);
-        tmp = load_reg(r2);
-        gen_helper_dlg(cpu_env, tmp32_1, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x88: /* ALCGR     R1,R2     [RRE] */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        tmp3 = tcg_temp_new_i64();
-        gen_op_calc_cc(s);
-        tcg_gen_extu_i32_i64(tmp3, cc_op);
-        tcg_gen_shri_i64(tmp3, tmp3, 1);
-        tcg_gen_andi_i64(tmp3, tmp3, 1);
-        tcg_gen_add_i64(tmp3, tmp2, tmp3);
-        tcg_gen_add_i64(tmp3, tmp, tmp3);
-        store_reg(r1, tmp3);
-        set_cc_addu64(s, tmp, tmp2, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x89: /* SLBGR   R1,R2     [RRE] */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        tmp32_1 = tcg_const_i32(r1);
-        gen_op_calc_cc(s);
-        gen_helper_slbg(cc_op, cpu_env, cc_op, tmp32_1, tmp, tmp2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x94: /* LLCR R1,R2 [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_andi_i32(tmp32_1, tmp32_1, 0xff);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x95: /* LLHR R1,R2 [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_andi_i32(tmp32_1, tmp32_1, 0xffff);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x96: /* MLR     R1,R2     [RRE] */
-        /* reg(r1, r1+1) = reg(r1+1) * reg(r2) */
-        tmp2 = load_reg(r2);
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32u_i64(tmp2, tmp2);
-        tcg_gen_ext32u_i64(tmp3, tmp3);
-        tcg_gen_mul_i64(tmp2, tmp2, tmp3);
-        store_reg32_i64((r1 + 1) & 15, tmp2);
-        tcg_gen_shri_i64(tmp2, tmp2, 32);
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x97: /* DLR     R1,R2     [RRE] */
-        /* reg(r1) = reg(r1, r1+1) % reg(r2) */
-        /* reg(r1+1) = reg(r1, r1+1) / reg(r2) */
-        tmp = load_reg(r1);
-        tmp2 = load_reg(r2);
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32u_i64(tmp2, tmp2);
-        tcg_gen_ext32u_i64(tmp3, tmp3);
-        tcg_gen_shli_i64(tmp, tmp, 32);
-        tcg_gen_or_i64(tmp, tmp, tmp3);
-
-        tcg_gen_rem_i64(tmp3, tmp, tmp2);
-        tcg_gen_div_i64(tmp, tmp, tmp2);
-        store_reg32_i64((r1 + 1) & 15, tmp);
-        store_reg32_i64(r1, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x98: /* ALCR    R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r2);
-        tmp32_3 = tcg_temp_new_i32();
-        /* XXX possible optimization point */
-        gen_op_calc_cc(s);
-        gen_helper_addc_u32(tmp32_3, cc_op, tmp32_1, tmp32_2);
-        set_cc_addu32(s, tmp32_1, tmp32_2, tmp32_3);
-        store_reg32(r1, tmp32_3);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x99: /* SLBR    R1,R2     [RRE] */
-        tmp32_1 = load_reg32(r2);
-        tmp32_2 = tcg_const_i32(r1);
-        gen_op_calc_cc(s);
-        gen_helper_slb(cc_op, cpu_env, cc_op, tmp32_2, tmp32_1);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    default:
-        LOG_DISAS("illegal b9 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 2);
-        break;
+        tcg_gen_add_i64(o->in2, o->in2, tsize);
+        r1 = (r1 + 1) & 15;
     }
+
+    tcg_temp_free_i64(tsize);
+    return NO_EXIT;
 }
 
-static void disas_c0(CPUS390XState *env, DisasContext *s, int op, int r1, int i2)
+static ExitStatus op_stmh(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp;
-    TCGv_i32 tmp32_1, tmp32_2;
-    uint64_t target = s->pc + i2 * 2LL;
-    int l1;
-
-    LOG_DISAS("disas_c0: op 0x%x r1 %d i2 %d\n", op, r1, i2);
+    int r1 = get_field(s->fields, r1);
+    int r3 = get_field(s->fields, r3);
+    TCGv_i64 t = tcg_temp_new_i64();
+    TCGv_i64 t4 = tcg_const_i64(4);
+    TCGv_i64 t32 = tcg_const_i64(32);
 
-    switch (op) {
-    case 0: /* larl r1, i2 */
-        tmp = tcg_const_i64(target);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x1: /* LGFI R1,I2 [RIL] */
-        tmp = tcg_const_i64((int64_t)i2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x4: /* BRCL     M1,I2     [RIL] */
-        /* m1 & (1 << (3 - cc)) */
-        tmp32_1 = tcg_const_i32(3);
-        tmp32_2 = tcg_const_i32(1);
-        gen_op_calc_cc(s);
-        tcg_gen_sub_i32(tmp32_1, tmp32_1, cc_op);
-        tcg_gen_shl_i32(tmp32_2, tmp32_2, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tmp32_1 = tcg_const_i32(r1); /* m1 == r1 */
-        tcg_gen_and_i32(tmp32_1, tmp32_1, tmp32_2);
-        l1 = gen_new_label();
-        tcg_gen_brcondi_i32(TCG_COND_EQ, tmp32_1, 0, l1);
-        gen_goto_tb(s, 0, target);
-        gen_set_label(l1);
-        gen_goto_tb(s, 1, s->pc + 6);
-        s->is_jmp = DISAS_TB_JUMP;
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x5: /* brasl r1, i2 */
-        tmp = tcg_const_i64(pc_to_link_info(s, s->pc + 6));
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        gen_goto_tb(s, 0, target);
-        s->is_jmp = DISAS_TB_JUMP;
-        break;
-    case 0x7: /* XILF R1,I2 [RIL] */
-    case 0xb: /* NILF R1,I2 [RIL] */
-    case 0xd: /* OILF R1,I2 [RIL] */
-        tmp32_1 = load_reg32(r1);
-        switch (op) {
-        case 0x7:
-            tcg_gen_xori_i32(tmp32_1, tmp32_1, (uint32_t)i2);
-            break;
-        case 0xb:
-            tcg_gen_andi_i32(tmp32_1, tmp32_1, (uint32_t)i2);
+    while (1) {
+        tcg_gen_shl_i64(t, regs[r1], t32);
+        tcg_gen_qemu_st32(t, o->in2, get_mem_index(s));
+        if (r1 == r3) {
             break;
-        case 0xd:
-            tcg_gen_ori_i32(tmp32_1, tmp32_1, (uint32_t)i2);
-            break;
-        default:
-            tcg_abort();
         }
-        store_reg32(r1, tmp32_1);
-        set_cc_nz_u32(s, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x9: /* IILF R1,I2 [RIL] */
-        tmp32_1 = tcg_const_i32((uint32_t)i2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xa: /* NIHF R1,I2 [RIL] */
-        tmp = load_reg(r1);
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_andi_i64(tmp, tmp, (((uint64_t)((uint32_t)i2)) << 32)
-                                   | 0xffffffffULL);
-        store_reg(r1, tmp);
-        tcg_gen_shri_i64(tmp, tmp, 32);
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        set_cc_nz_u32(s, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xe: /* LLIHF R1,I2 [RIL] */
-        tmp = tcg_const_i64(((uint64_t)(uint32_t)i2) << 32);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xf: /* LLILF R1,I2 [RIL] */
-        tmp = tcg_const_i64((uint32_t)i2);
-        store_reg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    default:
-        LOG_DISAS("illegal c0 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 3);
-        break;
+        tcg_gen_add_i64(o->in2, o->in2, t4);
+        r1 = (r1 + 1) & 15;
     }
+
+    tcg_temp_free_i64(t);
+    tcg_temp_free_i64(t4);
+    tcg_temp_free_i64(t32);
+    return NO_EXIT;
 }
 
-static void disas_c2(CPUS390XState *env, DisasContext *s, int op, int r1,
-                     int i2)
+static ExitStatus op_srst(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2, tmp3;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3;
+    potential_page_fault(s);
+    gen_helper_srst(o->in1, cpu_env, regs[0], o->in1, o->in2);
+    set_cc_static(s);
+    return_low128(o->in2);
+    return NO_EXIT;
+}
 
-    switch (op) {
-    case 0x4: /* SLGFI R1,I2 [RIL] */
-    case 0xa: /* ALGFI R1,I2 [RIL] */
-        tmp = load_reg(r1);
-        tmp2 = tcg_const_i64((uint64_t)(uint32_t)i2);
-        tmp3 = tcg_temp_new_i64();
-        switch (op) {
-        case 0x4:
-            tcg_gen_sub_i64(tmp3, tmp, tmp2);
-            set_cc_subu64(s, tmp, tmp2, tmp3);
-            break;
-        case 0xa:
-            tcg_gen_add_i64(tmp3, tmp, tmp2);
-            set_cc_addu64(s, tmp, tmp2, tmp3);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg(r1, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x5: /* SLFI R1,I2 [RIL] */
-    case 0xb: /* ALFI R1,I2 [RIL] */
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_const_i32(i2);
-        tmp32_3 = tcg_temp_new_i32();
-        switch (op) {
-        case 0x5:
-            tcg_gen_sub_i32(tmp32_3, tmp32_1, tmp32_2);
-            set_cc_subu32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0xb:
-            tcg_gen_add_i32(tmp32_3, tmp32_1, tmp32_2);
-            set_cc_addu32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg32(r1, tmp32_3);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0xc: /* CGFI R1,I2 [RIL] */
-        tmp = load_reg(r1);
-        cmp_s64c(s, tmp, (int64_t)i2);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xe: /* CLGFI R1,I2 [RIL] */
-        tmp = load_reg(r1);
-        cmp_u64c(s, tmp, (uint64_t)(uint32_t)i2);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xd: /* CFI R1,I2 [RIL] */
-        tmp32_1 = load_reg32(r1);
-        cmp_s32c(s, tmp32_1, i2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xf: /* CLFI R1,I2 [RIL] */
-        tmp32_1 = load_reg32(r1);
-        cmp_u32c(s, tmp32_1, i2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    default:
-        LOG_DISAS("illegal c2 operation 0x%x\n", op);
-        gen_illegal_opcode(env, s, 3);
-        break;
-    }
+static ExitStatus op_sub(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_sub_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
 }
 
-static void gen_and_or_xor_i32(int opc, TCGv_i32 tmp, TCGv_i32 tmp2)
+static ExitStatus op_subb(DisasContext *s, DisasOps *o)
 {
-    switch (opc & 0xf) {
-    case 0x4:
-        tcg_gen_and_i32(tmp, tmp, tmp2);
-        break;
-    case 0x6:
-        tcg_gen_or_i32(tmp, tmp, tmp2);
-        break;
-    case 0x7:
-        tcg_gen_xor_i32(tmp, tmp, tmp2);
-        break;
-    default:
-        tcg_abort();
+    DisasCompare cmp;
+    TCGv_i64 borrow;
+
+    tcg_gen_sub_i64(o->out, o->in1, o->in2);
+
+    /* The !borrow flag is the msb of CC.  Since we want the inverse of
+       that, we ask for a comparison of CC=0 | CC=1 -> mask of 8 | 4.  */
+    disas_jcc(s, &cmp, 8 | 4);
+    borrow = tcg_temp_new_i64();
+    if (cmp.is_64) {
+        tcg_gen_setcond_i64(cmp.cond, borrow, cmp.u.s64.a, cmp.u.s64.b);
+    } else {
+        TCGv_i32 t = tcg_temp_new_i32();
+        tcg_gen_setcond_i32(cmp.cond, t, cmp.u.s32.a, cmp.u.s32.b);
+        tcg_gen_extu_i32_i64(borrow, t);
+        tcg_temp_free_i32(t);
     }
+    free_compare(&cmp);
+
+    tcg_gen_sub_i64(o->out, o->out, borrow);
+    tcg_temp_free_i64(borrow);
+    return NO_EXIT;
 }
 
-static void disas_s390_insn(CPUS390XState *env, DisasContext *s)
+static ExitStatus op_svc(DisasContext *s, DisasOps *o)
 {
-    TCGv_i64 tmp, tmp2, tmp3, tmp4;
-    TCGv_i32 tmp32_1, tmp32_2, tmp32_3, tmp32_4;
-    unsigned char opc;
-    uint64_t insn;
-    int op, r1, r2, r3, d1, d2, x2, b1, b2, i, i2, r1b;
-    TCGv_i32 vl;
-    int ilc;
-    int l1;
+    TCGv_i32 t;
 
-    opc = cpu_ldub_code(env, s->pc);
-    LOG_DISAS("opc 0x%x\n", opc);
+    update_psw_addr(s);
+    update_cc_op(s);
 
-    ilc = get_ilc(opc);
+    t = tcg_const_i32(get_field(s->fields, i1) & 0xff);
+    tcg_gen_st_i32(t, cpu_env, offsetof(CPUS390XState, int_svc_code));
+    tcg_temp_free_i32(t);
+
+    t = tcg_const_i32(s->next_pc - s->pc);
+    tcg_gen_st_i32(t, cpu_env, offsetof(CPUS390XState, int_svc_ilen));
+    tcg_temp_free_i32(t);
+
+    gen_exception(EXCP_SVC);
+    return EXIT_NORETURN;
+}
+
+static ExitStatus op_tceb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_tceb(cc_op, o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_tcdb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_tcdb(cc_op, o->in1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_tcxb(DisasContext *s, DisasOps *o)
+{
+    gen_helper_tcxb(cc_op, o->out, o->out2, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
 
-    switch (opc) {
 #ifndef CONFIG_USER_ONLY
-    case 0x01: /* SAM */
-        insn = ld_code2(env, s->pc);
-        /* set addressing mode, but we only do 64bit anyways */
-        break;
+static ExitStatus op_tprot(DisasContext *s, DisasOps *o)
+{
+    potential_page_fault(s);
+    gen_helper_tprot(cc_op, o->addr1, o->in2);
+    set_cc_static(s);
+    return NO_EXIT;
+}
 #endif
-    case 0x6: /* BCTR     R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r1);
-        tcg_gen_subi_i32(tmp32_1, tmp32_1, 1);
-        store_reg32(r1, tmp32_1);
-
-        if (r2) {
-            gen_update_cc_op(s);
-            l1 = gen_new_label();
-            tcg_gen_brcondi_i32(TCG_COND_NE, tmp32_1, 0, l1);
-
-            /* not taking the branch, jump to after the instruction */
-            gen_goto_tb(s, 0, s->pc + 2);
-            gen_set_label(l1);
-
-            /* take the branch, move R2 into psw.addr */
-            tmp32_1 = load_reg32(r2);
-            tmp = tcg_temp_new_i64();
-            tcg_gen_extu_i32_i64(tmp, tmp32_1);
-            tcg_gen_mov_i64(psw_addr, tmp);
-            s->is_jmp = DISAS_JUMP;
-            tcg_temp_free_i32(tmp32_1);
-            tcg_temp_free_i64(tmp);
-        }
-        break;
-    case 0x7: /* BCR    M1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        if (r2) {
-            tmp = load_reg(r2);
-            gen_bcr(s, r1, tmp, s->pc);
-            tcg_temp_free_i64(tmp);
-            s->is_jmp = DISAS_TB_JUMP;
-        } else {
-            /* XXX: "serialization and checkpoint-synchronization function"? */
-        }
-        break;
-    case 0xa: /* SVC    I         [RR] */
-        insn = ld_code2(env, s->pc);
-        debug_insn(insn);
-        i = insn & 0xff;
-        update_psw_addr(s);
-        gen_op_calc_cc(s);
-        tmp32_1 = tcg_const_i32(i);
-        tmp32_2 = tcg_const_i32(ilc * 2);
-        tmp32_3 = tcg_const_i32(EXCP_SVC);
-        tcg_gen_st_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, int_svc_code));
-        tcg_gen_st_i32(tmp32_2, cpu_env, offsetof(CPUS390XState, int_svc_ilc));
-        gen_helper_exception(cpu_env, tmp32_3);
-        s->is_jmp = DISAS_EXCP;
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0xd: /* BASR   R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp = tcg_const_i64(pc_to_link_info(s, s->pc + 2));
-        store_reg(r1, tmp);
-        if (r2) {
-            tmp2 = load_reg(r2);
-            tcg_gen_mov_i64(psw_addr, tmp2);
-            tcg_temp_free_i64(tmp2);
-            s->is_jmp = DISAS_JUMP;
-        }
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0xe: /* MVCL   R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r2);
-        potential_page_fault(s);
-        gen_helper_mvcl(cc_op, cpu_env, tmp32_1, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x10: /* LPR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r2);
-        set_cc_abs32(s, tmp32_1);
-        gen_helper_abs_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x11: /* LNR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r2);
-        set_cc_nabs32(s, tmp32_1);
-        gen_helper_nabs_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x12: /* LTR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r2);
-        if (r1 != r2) {
-            store_reg32(r1, tmp32_1);
-        }
-        set_cc_s32(s, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x13: /* LCR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r2);
-        tcg_gen_neg_i32(tmp32_1, tmp32_1);
-        store_reg32(r1, tmp32_1);
-        set_cc_comp32(s, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x14: /* NR     R1,R2     [RR] */
-    case 0x16: /* OR     R1,R2     [RR] */
-    case 0x17: /* XR     R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_2 = load_reg32(r2);
-        tmp32_1 = load_reg32(r1);
-        gen_and_or_xor_i32(opc, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_1);
-        set_cc_nz_u32(s, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x18: /* LR     R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x15: /* CLR    R1,R2     [RR] */
-    case 0x19: /* CR     R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r2);
-        if (opc == 0x15) {
-            cmp_u32(s, tmp32_1, tmp32_2);
-        } else {
-            cmp_s32(s, tmp32_1, tmp32_2);
+
+static ExitStatus op_tr(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 l = tcg_const_i32(get_field(s->fields, l1));
+    potential_page_fault(s);
+    gen_helper_tr(cpu_env, l, o->addr1, o->in2);
+    tcg_temp_free_i32(l);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_unpk(DisasContext *s, DisasOps *o)
+{
+    TCGv_i32 l = tcg_const_i32(get_field(s->fields, l1));
+    potential_page_fault(s);
+    gen_helper_unpk(cpu_env, l, o->addr1, o->in2);
+    tcg_temp_free_i32(l);
+    return NO_EXIT;
+}
+
+static ExitStatus op_xc(DisasContext *s, DisasOps *o)
+{
+    int d1 = get_field(s->fields, d1);
+    int d2 = get_field(s->fields, d2);
+    int b1 = get_field(s->fields, b1);
+    int b2 = get_field(s->fields, b2);
+    int l = get_field(s->fields, l1);
+    TCGv_i32 t32;
+
+    o->addr1 = get_address(s, 0, b1, d1);
+
+    /* If the addresses are identical, this is a store/memset of zero.  */
+    if (b1 == b2 && d1 == d2 && (l + 1) <= 32) {
+        o->in2 = tcg_const_i64(0);
+
+        l++;
+        while (l >= 8) {
+            tcg_gen_qemu_st64(o->in2, o->addr1, get_mem_index(s));
+            l -= 8;
+            if (l > 0) {
+                tcg_gen_addi_i64(o->addr1, o->addr1, 8);
+            }
         }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x1a: /* AR     R1,R2     [RR] */
-    case 0x1e: /* ALR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r2);
-        tmp32_3 = tcg_temp_new_i32();
-        tcg_gen_add_i32(tmp32_3, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_3);
-        if (opc == 0x1a) {
-            set_cc_add32(s, tmp32_1, tmp32_2, tmp32_3);
-        } else {
-            set_cc_addu32(s, tmp32_1, tmp32_2, tmp32_3);
+        if (l >= 4) {
+            tcg_gen_qemu_st32(o->in2, o->addr1, get_mem_index(s));
+            l -= 4;
+            if (l > 0) {
+                tcg_gen_addi_i64(o->addr1, o->addr1, 4);
+            }
         }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x1b: /* SR     R1,R2     [RR] */
-    case 0x1f: /* SLR    R1,R2     [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r2);
-        tmp32_3 = tcg_temp_new_i32();
-        tcg_gen_sub_i32(tmp32_3, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_3);
-        if (opc == 0x1b) {
-            set_cc_sub32(s, tmp32_1, tmp32_2, tmp32_3);
-        } else {
-            set_cc_subu32(s, tmp32_1, tmp32_2, tmp32_3);
+        if (l >= 2) {
+            tcg_gen_qemu_st16(o->in2, o->addr1, get_mem_index(s));
+            l -= 2;
+            if (l > 0) {
+                tcg_gen_addi_i64(o->addr1, o->addr1, 2);
+            }
         }
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x1c: /* MR     R1,R2     [RR] */
-        /* reg(r1, r1+1) = reg(r1+1) * reg(r2) */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp2 = load_reg(r2);
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32s_i64(tmp2, tmp2);
-        tcg_gen_ext32s_i64(tmp3, tmp3);
-        tcg_gen_mul_i64(tmp2, tmp2, tmp3);
-        store_reg32_i64((r1 + 1) & 15, tmp2);
-        tcg_gen_shri_i64(tmp2, tmp2, 32);
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x1d: /* DR     R1,R2               [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r1 + 1);
-        tmp32_3 = load_reg32(r2);
-
-        tmp = tcg_temp_new_i64(); /* dividend */
-        tmp2 = tcg_temp_new_i64(); /* divisor */
-        tmp3 = tcg_temp_new_i64();
-
-        /* dividend is r(r1 << 32) | r(r1 + 1) */
-        tcg_gen_extu_i32_i64(tmp, tmp32_1);
-        tcg_gen_extu_i32_i64(tmp2, tmp32_2);
-        tcg_gen_shli_i64(tmp, tmp, 32);
-        tcg_gen_or_i64(tmp, tmp, tmp2);
-
-        /* divisor is r(r2) */
-        tcg_gen_ext_i32_i64(tmp2, tmp32_3);
-
-        tcg_gen_div_i64(tmp3, tmp, tmp2);
-        tcg_gen_rem_i64(tmp, tmp, tmp2);
-
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp3);
-
-        store_reg32(r1, tmp32_1); /* remainder */
-        store_reg32(r1 + 1, tmp32_2); /* quotient */
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x28: /* LDR    R1,R2               [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp = load_freg(r2);
-        store_freg(r1, tmp);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x38: /* LER    R1,R2               [RR] */
-        insn = ld_code2(env, s->pc);
-        decode_rr(s, insn, &r1, &r2);
-        tmp32_1 = load_freg32(r2);
-        store_freg32(r1, tmp32_1);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x40: /* STH    R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = load_reg(r1);
-        tcg_gen_qemu_st16(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x41:        /* la */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        store_reg(r1, tmp); /* FIXME: 31/24-bit addressing */
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x42: /* STC    R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = load_reg(r1);
-        tcg_gen_qemu_st8(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x43: /* IC     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8u(tmp2, tmp, get_mem_index(s));
-        store_reg8(r1, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x44: /* EX     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = load_reg(r1);
-        tmp3 = tcg_const_i64(s->pc + 4);
-        update_psw_addr(s);
-        gen_op_calc_cc(s);
-        gen_helper_ex(cc_op, cpu_env, cc_op, tmp2, tmp, tmp3);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x46: /* BCT    R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tcg_temp_free_i64(tmp);
-
-        tmp32_1 = load_reg32(r1);
-        tcg_gen_subi_i32(tmp32_1, tmp32_1, 1);
-        store_reg32(r1, tmp32_1);
-
-        gen_update_cc_op(s);
-        l1 = gen_new_label();
-        tcg_gen_brcondi_i32(TCG_COND_NE, tmp32_1, 0, l1);
-
-        /* not taking the branch, jump to after the instruction */
-        gen_goto_tb(s, 0, s->pc + 4);
-        gen_set_label(l1);
-
-        /* take the branch, move R2 into psw.addr */
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tcg_gen_mov_i64(psw_addr, tmp);
-        s->is_jmp = DISAS_JUMP;
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp);
-        break;
-    case 0x47: /* BC     M1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        gen_bcr(s, r1, tmp, s->pc + 4);
-        tcg_temp_free_i64(tmp);
-        s->is_jmp = DISAS_TB_JUMP;
-        break;
-    case 0x48: /* LH     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld16s(tmp2, tmp, get_mem_index(s));
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x49: /* CH     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld16s(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        cmp_s32(s, tmp32_1, tmp32_2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x4a: /* AH     R1,D2(X2,B2)     [RX] */
-    case 0x4b: /* SH     R1,D2(X2,B2)     [RX] */
-    case 0x4c: /* MH     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp32_3 = tcg_temp_new_i32();
-
-        tcg_gen_qemu_ld16s(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        switch (opc) {
-        case 0x4a:
-            tcg_gen_add_i32(tmp32_3, tmp32_1, tmp32_2);
-            set_cc_add32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x4b:
-            tcg_gen_sub_i32(tmp32_3, tmp32_1, tmp32_2);
-            set_cc_sub32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x4c:
-            tcg_gen_mul_i32(tmp32_3, tmp32_1, tmp32_2);
-            break;
-        default:
-            tcg_abort();
+        if (l) {
+            tcg_gen_qemu_st8(o->in2, o->addr1, get_mem_index(s));
         }
-        store_reg32(r1, tmp32_3);
+        gen_op_movi_cc(s, 0);
+        return NO_EXIT;
+    }
 
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x4d: /* BAS    R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_const_i64(pc_to_link_info(s, s->pc + 4));
-        store_reg(r1, tmp2);
-        tcg_gen_mov_i64(psw_addr, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        s->is_jmp = DISAS_JUMP;
-        break;
-    case 0x4e: /* CVD    R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_trunc_i64_i32(tmp32_1, regs[r1]);
-        gen_helper_cvd(tmp2, tmp32_1);
-        tcg_gen_qemu_st64(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x50: /* st r1, d2(x2, b2) */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = load_reg(r1);
-        tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x55: /* CL     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tmp32_2 = load_reg32(r1);
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        cmp_u32(s, tmp32_2, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x54: /* N      R1,D2(X2,B2)     [RX] */
-    case 0x56: /* O      R1,D2(X2,B2)     [RX] */
-    case 0x57: /* X      R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        gen_and_or_xor_i32(opc, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_1);
-        set_cc_nz_u32(s, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x58: /* l r1, d2(x2, b2) */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0x59: /* C      R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tmp32_2 = load_reg32(r1);
-        tcg_gen_qemu_ld32s(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        cmp_s32(s, tmp32_2, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x5a: /* A      R1,D2(X2,B2)     [RX] */
-    case 0x5b: /* S      R1,D2(X2,B2)     [RX] */
-    case 0x5e: /* AL     R1,D2(X2,B2)     [RX] */
-    case 0x5f: /* SL     R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tmp32_3 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32s(tmp, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp);
-        switch (opc) {
-        case 0x5a:
-        case 0x5e:
-            tcg_gen_add_i32(tmp32_3, tmp32_1, tmp32_2);
-            break;
-        case 0x5b:
-        case 0x5f:
-            tcg_gen_sub_i32(tmp32_3, tmp32_1, tmp32_2);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg32(r1, tmp32_3);
-        switch (opc) {
-        case 0x5a:
-            set_cc_add32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x5e:
-            set_cc_addu32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x5b:
-            set_cc_sub32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        case 0x5f:
-            set_cc_subu32(s, tmp32_1, tmp32_2, tmp32_3);
-            break;
-        default:
-            tcg_abort();
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i32(tmp32_3);
-        break;
-    case 0x5c: /* M      R1,D2(X2,B2)        [RX] */
-        /* reg(r1, r1+1) = reg(r1+1) * *(s32*)addr */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32s(tmp2, tmp, get_mem_index(s));
-        tmp3 = load_reg((r1 + 1) & 15);
-        tcg_gen_ext32s_i64(tmp2, tmp2);
-        tcg_gen_ext32s_i64(tmp3, tmp3);
-        tcg_gen_mul_i64(tmp2, tmp2, tmp3);
-        store_reg32_i64((r1 + 1) & 15, tmp2);
-        tcg_gen_shri_i64(tmp2, tmp2, 32);
-        store_reg32_i64(r1, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x5d: /* D      R1,D2(X2,B2)        [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp3 = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r1 + 1);
-
-        tmp = tcg_temp_new_i64();
-        tmp2 = tcg_temp_new_i64();
-
-        /* dividend is r(r1 << 32) | r(r1 + 1) */
-        tcg_gen_extu_i32_i64(tmp, tmp32_1);
-        tcg_gen_extu_i32_i64(tmp2, tmp32_2);
-        tcg_gen_shli_i64(tmp, tmp, 32);
-        tcg_gen_or_i64(tmp, tmp, tmp2);
-
-        /* divisor is in memory */
-        tcg_gen_qemu_ld32s(tmp2, tmp3, get_mem_index(s));
-
-        /* XXX divisor == 0 -> FixP divide exception */
-
-        tcg_gen_div_i64(tmp3, tmp, tmp2);
-        tcg_gen_rem_i64(tmp, tmp, tmp2);
-
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp3);
-
-        store_reg32(r1, tmp32_1); /* remainder */
-        store_reg32(r1 + 1, tmp32_2); /* quotient */
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-    case 0x60: /* STD    R1,D2(X2,B2)        [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = load_freg(r1);
-        tcg_gen_qemu_st64(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x68: /* LD    R1,D2(X2,B2)        [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld64(tmp2, tmp, get_mem_index(s));
-        store_freg(r1, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
+    /* But in general we'll defer to a helper.  */
+    o->in2 = get_address(s, 0, b2, d2);
+    t32 = tcg_const_i32(l);
+    potential_page_fault(s);
+    gen_helper_xc(cc_op, cpu_env, t32, o->addr1, o->in2);
+    tcg_temp_free_i32(t32);
+    set_cc_static(s);
+    return NO_EXIT;
+}
+
+static ExitStatus op_xor(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_xor_i64(o->out, o->in1, o->in2);
+    return NO_EXIT;
+}
+
+static ExitStatus op_xori(DisasContext *s, DisasOps *o)
+{
+    int shift = s->insn->data & 0xff;
+    int size = s->insn->data >> 8;
+    uint64_t mask = ((1ull << size) - 1) << shift;
+
+    assert(!o->g_in2);
+    tcg_gen_shli_i64(o->in2, o->in2, shift);
+    tcg_gen_xor_i64(o->out, o->in1, o->in2);
+
+    /* Produce the CC from only the bits manipulated.  */
+    tcg_gen_andi_i64(cc_dst, o->out, mask);
+    set_cc_nz_u64(s, cc_dst);
+    return NO_EXIT;
+}
+
+static ExitStatus op_zero(DisasContext *s, DisasOps *o)
+{
+    o->out = tcg_const_i64(0);
+    return NO_EXIT;
+}
+
+static ExitStatus op_zero2(DisasContext *s, DisasOps *o)
+{
+    o->out = tcg_const_i64(0);
+    o->out2 = o->out;
+    o->g_out2 = true;
+    return NO_EXIT;
+}
+
+/* ====================================================================== */
+/* The "Cc OUTput" generators.  Given the generated output (and in some cases
+   the original inputs), update the various cc data structures in order to
+   be able to compute the new condition code.  */
+
+static void cout_abs32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_ABS_32, o->out);
+}
+
+static void cout_abs64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_ABS_64, o->out);
+}
+
+static void cout_adds32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADD_32, o->in1, o->in2, o->out);
+}
+
+static void cout_adds64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADD_64, o->in1, o->in2, o->out);
+}
+
+static void cout_addu32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADDU_32, o->in1, o->in2, o->out);
+}
+
+static void cout_addu64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADDU_64, o->in1, o->in2, o->out);
+}
+
+static void cout_addc32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADDC_32, o->in1, o->in2, o->out);
+}
+
+static void cout_addc64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_ADDC_64, o->in1, o->in2, o->out);
+}
+
+static void cout_cmps32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_LTGT_32, o->in1, o->in2);
+}
+
+static void cout_cmps64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_LTGT_64, o->in1, o->in2);
+}
+
+static void cout_cmpu32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_LTUGTU_32, o->in1, o->in2);
+}
+
+static void cout_cmpu64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_LTUGTU_64, o->in1, o->in2);
+}
+
+static void cout_f32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_NZ_F32, o->out);
+}
+
+static void cout_f64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_NZ_F64, o->out);
+}
+
+static void cout_f128(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_NZ_F128, o->out, o->out2);
+}
+
+static void cout_nabs32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_NABS_32, o->out);
+}
+
+static void cout_nabs64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_NABS_64, o->out);
+}
+
+static void cout_neg32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_COMP_32, o->out);
+}
+
+static void cout_neg64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_COMP_64, o->out);
+}
+
+static void cout_nz32(DisasContext *s, DisasOps *o)
+{
+    tcg_gen_ext32u_i64(cc_dst, o->out);
+    gen_op_update1_cc_i64(s, CC_OP_NZ, cc_dst);
+}
+
+static void cout_nz64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_NZ, o->out);
+}
+
+static void cout_s32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_LTGT0_32, o->out);
+}
+
+static void cout_s64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update1_cc_i64(s, CC_OP_LTGT0_64, o->out);
+}
+
+static void cout_subs32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUB_32, o->in1, o->in2, o->out);
+}
+
+static void cout_subs64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUB_64, o->in1, o->in2, o->out);
+}
+
+static void cout_subu32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUBU_32, o->in1, o->in2, o->out);
+}
+
+static void cout_subu64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUBU_64, o->in1, o->in2, o->out);
+}
+
+static void cout_subb32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUBB_32, o->in1, o->in2, o->out);
+}
+
+static void cout_subb64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update3_cc_i64(s, CC_OP_SUBB_64, o->in1, o->in2, o->out);
+}
+
+static void cout_tm32(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_TM_32, o->in1, o->in2);
+}
+
+static void cout_tm64(DisasContext *s, DisasOps *o)
+{
+    gen_op_update2_cc_i64(s, CC_OP_TM_64, o->in1, o->in2);
+}
+
+/* ====================================================================== */
+/* The "PREPeration" generators.  These initialize the DisasOps.OUT fields
+   with the TCG register to which we will write.  Used in combination with
+   the "wout" generators, in some cases we need a new temporary, and in
+   some cases we can write to a TCG global.  */
+
+static void prep_new(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->out = tcg_temp_new_i64();
+}
+#define SPEC_prep_new 0
+
+static void prep_new_P(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->out = tcg_temp_new_i64();
+    o->out2 = tcg_temp_new_i64();
+}
+#define SPEC_prep_new_P 0
+
+static void prep_r1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->out = regs[get_field(f, r1)];
+    o->g_out = true;
+}
+#define SPEC_prep_r1 0
+
+static void prep_r1_P(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    o->out = regs[r1];
+    o->out2 = regs[r1 + 1];
+    o->g_out = o->g_out2 = true;
+}
+#define SPEC_prep_r1_P SPEC_r1_even
+
+static void prep_f1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->out = fregs[get_field(f, r1)];
+    o->g_out = true;
+}
+#define SPEC_prep_f1 0
+
+static void prep_x1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    o->out = fregs[r1];
+    o->out2 = fregs[r1 + 2];
+    o->g_out = o->g_out2 = true;
+}
+#define SPEC_prep_x1 SPEC_r1_f128
+
+/* ====================================================================== */
+/* The "Write OUTput" generators.  These generally perform some non-trivial
+   copy of data to TCG globals, or to main memory.  The trivial cases are
+   generally handled by having a "prep" generator install the TCG global
+   as the destination of the operation.  */
+
+static void wout_r1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    store_reg(get_field(f, r1), o->out);
+}
+#define SPEC_wout_r1 0
+
+static void wout_r1_8(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    tcg_gen_deposit_i64(regs[r1], regs[r1], o->out, 0, 8);
+}
+#define SPEC_wout_r1_8 0
+
+static void wout_r1_16(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    tcg_gen_deposit_i64(regs[r1], regs[r1], o->out, 0, 16);
+}
+#define SPEC_wout_r1_16 0
+
+static void wout_r1_32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    store_reg32_i64(get_field(f, r1), o->out);
+}
+#define SPEC_wout_r1_32 0
+
+static void wout_r1_P32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    store_reg32_i64(r1, o->out);
+    store_reg32_i64(r1 + 1, o->out2);
+}
+#define SPEC_wout_r1_P32 SPEC_r1_even
+
+static void wout_r1_D32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    store_reg32_i64(r1 + 1, o->out);
+    tcg_gen_shri_i64(o->out, o->out, 32);
+    store_reg32_i64(r1, o->out);
+}
+#define SPEC_wout_r1_D32 SPEC_r1_even
+
+static void wout_e1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    store_freg32_i64(get_field(f, r1), o->out);
+}
+#define SPEC_wout_e1 0
+
+static void wout_f1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    store_freg(get_field(f, r1), o->out);
+}
+#define SPEC_wout_f1 0
+
+static void wout_x1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int f1 = get_field(s->fields, r1);
+    store_freg(f1, o->out);
+    store_freg(f1 + 2, o->out2);
+}
+#define SPEC_wout_x1 SPEC_r1_f128
+
+static void wout_cond_r1r2_32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    if (get_field(f, r1) != get_field(f, r2)) {
+        store_reg32_i64(get_field(f, r1), o->out);
+    }
+}
+#define SPEC_wout_cond_r1r2_32 0
+
+static void wout_cond_e1e2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    if (get_field(f, r1) != get_field(f, r2)) {
+        store_freg32_i64(get_field(f, r1), o->out);
+    }
+}
+#define SPEC_wout_cond_e1e2 0
+
+static void wout_m1_8(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    tcg_gen_qemu_st8(o->out, o->addr1, get_mem_index(s));
+}
+#define SPEC_wout_m1_8 0
+
+static void wout_m1_16(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    tcg_gen_qemu_st16(o->out, o->addr1, get_mem_index(s));
+}
+#define SPEC_wout_m1_16 0
+
+static void wout_m1_32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    tcg_gen_qemu_st32(o->out, o->addr1, get_mem_index(s));
+}
+#define SPEC_wout_m1_32 0
+
+static void wout_m1_64(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    tcg_gen_qemu_st64(o->out, o->addr1, get_mem_index(s));
+}
+#define SPEC_wout_m1_64 0
+
+static void wout_m2_32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    tcg_gen_qemu_st32(o->out, o->in2, get_mem_index(s));
+}
+#define SPEC_wout_m2_32 0
+
+/* ====================================================================== */
+/* The "INput 1" generators.  These load the first operand to an insn.  */
+
+static void in1_r1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = load_reg(get_field(f, r1));
+}
+#define SPEC_in1_r1 0
+
+static void in1_r1_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = regs[get_field(f, r1)];
+    o->g_in1 = true;
+}
+#define SPEC_in1_r1_o 0
+
+static void in1_r1_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32s_i64(o->in1, regs[get_field(f, r1)]);
+}
+#define SPEC_in1_r1_32s 0
+
+static void in1_r1_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32u_i64(o->in1, regs[get_field(f, r1)]);
+}
+#define SPEC_in1_r1_32u 0
+
+static void in1_r1_sr32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_shri_i64(o->in1, regs[get_field(f, r1)], 32);
+}
+#define SPEC_in1_r1_sr32 0
+
+static void in1_r1p1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = load_reg(get_field(f, r1) + 1);
+}
+#define SPEC_in1_r1p1 SPEC_r1_even
+
+static void in1_r1p1_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32s_i64(o->in1, regs[get_field(f, r1) + 1]);
+}
+#define SPEC_in1_r1p1_32s SPEC_r1_even
+
+static void in1_r1p1_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32u_i64(o->in1, regs[get_field(f, r1) + 1]);
+}
+#define SPEC_in1_r1p1_32u SPEC_r1_even
+
+static void in1_r1_D32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_concat32_i64(o->in1, regs[r1 + 1], regs[r1]);
+}
+#define SPEC_in1_r1_D32 SPEC_r1_even
+
+static void in1_r2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = load_reg(get_field(f, r2));
+}
+#define SPEC_in1_r2 0
+
+static void in1_r3(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = load_reg(get_field(f, r3));
+}
+#define SPEC_in1_r3 0
+
+static void in1_r3_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = regs[get_field(f, r3)];
+    o->g_in1 = true;
+}
+#define SPEC_in1_r3_o 0
+
+static void in1_r3_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32s_i64(o->in1, regs[get_field(f, r3)]);
+}
+#define SPEC_in1_r3_32s 0
+
+static void in1_r3_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_ext32u_i64(o->in1, regs[get_field(f, r3)]);
+}
+#define SPEC_in1_r3_32u 0
+
+static void in1_r3_D32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r3 = get_field(f, r3);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_concat32_i64(o->in1, regs[r3 + 1], regs[r3]);
+}
+#define SPEC_in1_r3_D32 SPEC_r3_even
+
+static void in1_e1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = load_freg32_i64(get_field(f, r1));
+}
+#define SPEC_in1_e1 0
+
+static void in1_f1_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = fregs[get_field(f, r1)];
+    o->g_in1 = true;
+}
+#define SPEC_in1_f1_o 0
+
+static void in1_x1_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    o->out = fregs[r1];
+    o->out2 = fregs[r1 + 2];
+    o->g_out = o->g_out2 = true;
+}
+#define SPEC_in1_x1_o SPEC_r1_f128
+
+static void in1_f3_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in1 = fregs[get_field(f, r3)];
+    o->g_in1 = true;
+}
+#define SPEC_in1_f3_o 0
+
+static void in1_la1(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->addr1 = get_address(s, 0, get_field(f, b1), get_field(f, d1));
+}
+#define SPEC_in1_la1 0
+
+static void in1_la2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int x2 = have_field(f, x2) ? get_field(f, x2) : 0;
+    o->addr1 = get_address(s, x2, get_field(f, b2), get_field(f, d2));
+}
+#define SPEC_in1_la2 0
+
+static void in1_m1_8u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld8u(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_8u 0
+
+static void in1_m1_16s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld16s(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_16s 0
+
+static void in1_m1_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld16u(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_16u 0
+
+static void in1_m1_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld32s(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_32s 0
+
+static void in1_m1_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld32u(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_32u 0
+
+static void in1_m1_64(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in1_la1(s, f, o);
+    o->in1 = tcg_temp_new_i64();
+    tcg_gen_qemu_ld64(o->in1, o->addr1, get_mem_index(s));
+}
+#define SPEC_in1_m1_64 0
+
+/* ====================================================================== */
+/* The "INput 2" generators.  These load the second operand to an insn.  */
+
+static void in2_r1_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = regs[get_field(f, r1)];
+    o->g_in2 = true;
+}
+#define SPEC_in2_r1_o 0
+
+static void in2_r1_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext16u_i64(o->in2, regs[get_field(f, r1)]);
+}
+#define SPEC_in2_r1_16u 0
+
+static void in2_r1_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext32u_i64(o->in2, regs[get_field(f, r1)]);
+}
+#define SPEC_in2_r1_32u 0
+
+static void in2_r1_D32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r1 = get_field(f, r1);
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_concat32_i64(o->in2, regs[r1 + 1], regs[r1]);
+}
+#define SPEC_in2_r1_D32 SPEC_r1_even
+
+static void in2_r2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = load_reg(get_field(f, r2));
+}
+#define SPEC_in2_r2 0
+
+static void in2_r2_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = regs[get_field(f, r2)];
+    o->g_in2 = true;
+}
+#define SPEC_in2_r2_o 0
+
+static void in2_r2_nz(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r2 = get_field(f, r2);
+    if (r2 != 0) {
+        o->in2 = load_reg(r2);
+    }
+}
+#define SPEC_in2_r2_nz 0
+
+static void in2_r2_8s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext8s_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_8s 0
+
+static void in2_r2_8u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext8u_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_8u 0
+
+static void in2_r2_16s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext16s_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_16s 0
+
+static void in2_r2_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext16u_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_16u 0
+
+static void in2_r3(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = load_reg(get_field(f, r3));
+}
+#define SPEC_in2_r3 0
+
+static void in2_r2_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext32s_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_32s 0
+
+static void in2_r2_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_temp_new_i64();
+    tcg_gen_ext32u_i64(o->in2, regs[get_field(f, r2)]);
+}
+#define SPEC_in2_r2_32u 0
+
+static void in2_e2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = load_freg32_i64(get_field(f, r2));
+}
+#define SPEC_in2_e2 0
+
+static void in2_f2_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = fregs[get_field(f, r2)];
+    o->g_in2 = true;
+}
+#define SPEC_in2_f2_o 0
+
+static void in2_x2_o(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int r2 = get_field(f, r2);
+    o->in1 = fregs[r2];
+    o->in2 = fregs[r2 + 2];
+    o->g_in1 = o->g_in2 = true;
+}
+#define SPEC_in2_x2_o SPEC_r2_f128
+
+static void in2_ra2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = get_address(s, 0, get_field(f, r2), 0);
+}
+#define SPEC_in2_ra2 0
+
+static void in2_a2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    int x2 = have_field(f, x2) ? get_field(f, x2) : 0;
+    o->in2 = get_address(s, x2, get_field(f, b2), get_field(f, d2));
+}
+#define SPEC_in2_a2 0
+
+static void in2_ri2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_const_i64(s->pc + (int64_t)get_field(f, i2) * 2);
+}
+#define SPEC_in2_ri2 0
+
+static void in2_sh32(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    help_l2_shift(s, f, o, 31);
+}
+#define SPEC_in2_sh32 0
+
+static void in2_sh64(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    help_l2_shift(s, f, o, 63);
+}
+#define SPEC_in2_sh64 0
+
+static void in2_m2_8u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld8u(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_8u 0
+
+static void in2_m2_16s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld16s(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_16s 0
+
+static void in2_m2_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld16u(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_16u 0
+
+static void in2_m2_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld32s(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_32s 0
+
+static void in2_m2_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld32u(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_32u 0
+
+static void in2_m2_64(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_a2(s, f, o);
+    tcg_gen_qemu_ld64(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_m2_64 0
+
+static void in2_mri2_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_ri2(s, f, o);
+    tcg_gen_qemu_ld16u(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_mri2_16u 0
+
+static void in2_mri2_32s(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_ri2(s, f, o);
+    tcg_gen_qemu_ld32s(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_mri2_32s 0
+
+static void in2_mri2_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_ri2(s, f, o);
+    tcg_gen_qemu_ld32u(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_mri2_32u 0
+
+static void in2_mri2_64(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    in2_ri2(s, f, o);
+    tcg_gen_qemu_ld64(o->in2, o->in2, get_mem_index(s));
+}
+#define SPEC_in2_mri2_64 0
+
+static void in2_i2(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_const_i64(get_field(f, i2));
+}
+#define SPEC_in2_i2 0
+
+static void in2_i2_8u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_const_i64((uint8_t)get_field(f, i2));
+}
+#define SPEC_in2_i2_8u 0
+
+static void in2_i2_16u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_const_i64((uint16_t)get_field(f, i2));
+}
+#define SPEC_in2_i2_16u 0
+
+static void in2_i2_32u(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    o->in2 = tcg_const_i64((uint32_t)get_field(f, i2));
+}
+#define SPEC_in2_i2_32u 0
+
+static void in2_i2_16u_shl(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    uint64_t i2 = (uint16_t)get_field(f, i2);
+    o->in2 = tcg_const_i64(i2 << s->insn->data);
+}
+#define SPEC_in2_i2_16u_shl 0
+
+static void in2_i2_32u_shl(DisasContext *s, DisasFields *f, DisasOps *o)
+{
+    uint64_t i2 = (uint32_t)get_field(f, i2);
+    o->in2 = tcg_const_i64(i2 << s->insn->data);
+}
+#define SPEC_in2_i2_32u_shl 0
+
+/* ====================================================================== */
+
+/* Find opc within the table of insns.  This is formulated as a switch
+   statement so that (1) we get compile-time notice of cut-paste errors
+   for duplicated opcodes, and (2) the compiler generates the binary
+   search tree, rather than us having to post-process the table.  */
+
+#define C(OPC, NM, FT, FC, I1, I2, P, W, OP, CC) \
+    D(OPC, NM, FT, FC, I1, I2, P, W, OP, CC, 0)
+
+#define D(OPC, NM, FT, FC, I1, I2, P, W, OP, CC, D) insn_ ## NM,
+
+enum DisasInsnEnum {
+#include "insn-data.def"
+};
+
+#undef D
+#define D(OPC, NM, FT, FC, I1, I2, P, W, OP, CC, D) {                       \
+    .opc = OPC,                                                             \
+    .fmt = FMT_##FT,                                                        \
+    .fac = FAC_##FC,                                                        \
+    .spec = SPEC_in1_##I1 | SPEC_in2_##I2 | SPEC_prep_##P | SPEC_wout_##W,  \
+    .name = #NM,                                                            \
+    .help_in1 = in1_##I1,                                                   \
+    .help_in2 = in2_##I2,                                                   \
+    .help_prep = prep_##P,                                                  \
+    .help_wout = wout_##W,                                                  \
+    .help_cout = cout_##CC,                                                 \
+    .help_op = op_##OP,                                                     \
+    .data = D                                                               \
+ },
+
+/* Allow 0 to be used for NULL in the table below.  */
+#define in1_0  NULL
+#define in2_0  NULL
+#define prep_0  NULL
+#define wout_0  NULL
+#define cout_0  NULL
+#define op_0  NULL
+
+#define SPEC_in1_0 0
+#define SPEC_in2_0 0
+#define SPEC_prep_0 0
+#define SPEC_wout_0 0
+
+static const DisasInsn insn_info[] = {
+#include "insn-data.def"
+};
+
+#undef D
+#define D(OPC, NM, FT, FC, I1, I2, P, W, OP, CC, D) \
+    case OPC: return &insn_info[insn_ ## NM];
+
+static const DisasInsn *lookup_opc(uint16_t opc)
+{
+    switch (opc) {
+#include "insn-data.def"
+    default:
+        return NULL;
+    }
+}
+
+#undef D
+#undef C
+
+/* Extract a field from the insn.  The INSN should be left-aligned in
+   the uint64_t so that we can more easily utilize the big-bit-endian
+   definitions we extract from the Principals of Operation.  */
+
+static void extract_field(DisasFields *o, const DisasField *f, uint64_t insn)
+{
+    uint32_t r, m;
+
+    if (f->size == 0) {
+        return;
+    }
+
+    /* Zero extract the field from the insn.  */
+    r = (insn << f->beg) >> (64 - f->size);
+
+    /* Sign-extend, or un-swap the field as necessary.  */
+    switch (f->type) {
+    case 0: /* unsigned */
         break;
-    case 0x70: /* STE R1,D2(X2,B2) [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_freg32(r1);
-        tcg_gen_extu_i32_i64(tmp2, tmp32_1);
-        tcg_gen_qemu_st32(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
+    case 1: /* signed */
+        assert(f->size <= 32);
+        m = 1u << (f->size - 1);
+        r = (r ^ m) - m;
         break;
-    case 0x71: /* MS      R1,D2(X2,B2)     [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32s(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        tcg_gen_mul_i32(tmp32_1, tmp32_1, tmp32_2);
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
+    case 2: /* dl+dh split, signed 20 bit. */
+        r = ((int8_t)r << 12) | (r >> 8);
         break;
-    case 0x78: /* LE     R1,D2(X2,B2)        [RX] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = tcg_temp_new_i32();
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp2);
-        store_freg32(r1, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
+    default:
+        abort();
+    }
+
+    /* Validate that the "compressed" encoding we selected above is valid.
+       I.e. we havn't make two different original fields overlap.  */
+    assert(((o->presentC >> f->indexC) & 1) == 0);
+    o->presentC |= 1 << f->indexC;
+    o->presentO |= 1 << f->indexO;
+
+    o->c[f->indexC] = r;
+}
+
+/* Lookup the insn at the current PC, extracting the operands into O and
+   returning the info struct for the insn.  Returns NULL for invalid insn.  */
+
+static const DisasInsn *extract_insn(CPUS390XState *env, DisasContext *s,
+                                     DisasFields *f)
+{
+    uint64_t insn, pc = s->pc;
+    int op, op2, ilen;
+    const DisasInsn *info;
+
+    insn = ld_code2(env, pc);
+    op = (insn >> 8) & 0xff;
+    ilen = get_ilen(op);
+    s->next_pc = s->pc + ilen;
+
+    switch (ilen) {
+    case 2:
+        insn = insn << 48;
         break;
-#ifndef CONFIG_USER_ONLY
-    case 0x80: /* SSM      D2(B2)       [S] */
-        /* Set System Mask */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_andi_i64(tmp3, psw_mask, ~0xff00000000000000ULL);
-        tcg_gen_qemu_ld8u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_shli_i64(tmp2, tmp2, 56);
-        tcg_gen_or_i64(psw_mask, tmp3, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+    case 4:
+        insn = ld_code4(env, pc) << 32;
         break;
-    case 0x82: /* LPSW     D2(B2)       [S] */
-        /* Load PSW */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-        tcg_gen_addi_i64(tmp, tmp, 4);
-        tcg_gen_qemu_ld32u(tmp3, tmp, get_mem_index(s));
-        gen_helper_load_psw(cpu_env, tmp2, tmp3);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        /* we need to keep cc_op intact */
-        s->is_jmp = DISAS_JUMP;
+    case 6:
+        insn = (insn << 48) | (ld_code4(env, pc + 2) << 16);
         break;
-    case 0x83: /* DIAG     R1,R3,D2     [RS] */
-        /* Diagnose call (KVM hypercall) */
-        check_privileged(env, s, ilc);
-        potential_page_fault(s);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp32_1 = tcg_const_i32(insn & 0xfff);
-        tmp2 = load_reg(2);
-        tmp3 = load_reg(1);
-        gen_helper_diag(tmp2, cpu_env, tmp32_1, tmp2, tmp3);
-        store_reg(2, tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
+    default:
+        abort();
+    }
+
+    /* We can't actually determine the insn format until we've looked up
+       the full insn opcode.  Which we can't do without locating the
+       secondary opcode.  Assume by default that OP2 is at bit 40; for
+       those smaller insns that don't actually have a secondary opcode
+       this will correctly result in OP2 = 0. */
+    switch (op) {
+    case 0x01: /* E */
+    case 0x80: /* S */
+    case 0x82: /* S */
+    case 0x93: /* S */
+    case 0xb2: /* S, RRF, RRE */
+    case 0xb3: /* RRE, RRD, RRF */
+    case 0xb9: /* RRE, RRF */
+    case 0xe5: /* SSE, SIL */
+        op2 = (insn << 8) >> 56;
+        break;
+    case 0xa5: /* RI */
+    case 0xa7: /* RI */
+    case 0xc0: /* RIL */
+    case 0xc2: /* RIL */
+    case 0xc4: /* RIL */
+    case 0xc6: /* RIL */
+    case 0xc8: /* SSF */
+    case 0xcc: /* RIL */
+        op2 = (insn << 12) >> 60;
+        break;
+    case 0xd0 ... 0xdf: /* SS */
+    case 0xe1: /* SS */
+    case 0xe2: /* SS */
+    case 0xe8: /* SS */
+    case 0xe9: /* SS */
+    case 0xea: /* SS */
+    case 0xee ... 0xf3: /* SS */
+    case 0xf8 ... 0xfd: /* SS */
+        op2 = 0;
         break;
-#endif
-    case 0x88: /* SRL    R1,D2(B2)        [RS] */
-    case 0x89: /* SLL    R1,D2(B2)        [RS] */
-    case 0x8a: /* SRA    R1,D2(B2)        [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_temp_new_i32();
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp);
-        tcg_gen_andi_i32(tmp32_2, tmp32_2, 0x3f);
-        switch (opc) {
-        case 0x88:
-            tcg_gen_shr_i32(tmp32_1, tmp32_1, tmp32_2);
-            break;
-        case 0x89:
-            tcg_gen_shl_i32(tmp32_1, tmp32_1, tmp32_2);
-            break;
-        case 0x8a:
-            tcg_gen_sar_i32(tmp32_1, tmp32_1, tmp32_2);
-            set_cc_s32(s, tmp32_1);
-            break;
-        default:
-            tcg_abort();
-        }
-        store_reg32(r1, tmp32_1);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
+    default:
+        op2 = (insn << 40) >> 56;
         break;
-    case 0x8c: /* SRDL   R1,D2(B2)        [RS] */
-    case 0x8d: /* SLDL   R1,D2(B2)        [RS] */
-    case 0x8e: /* SRDA   R1,D2(B2)        [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2); /* shift */
-        tmp2 = tcg_temp_new_i64();
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = load_reg32(r1 + 1);
-        tcg_gen_concat_i32_i64(tmp2, tmp32_2, tmp32_1); /* operand */
-        switch (opc) {
-        case 0x8c:
-            tcg_gen_shr_i64(tmp2, tmp2, tmp);
-            break;
-        case 0x8d:
-            tcg_gen_shl_i64(tmp2, tmp2, tmp);
-            break;
-        case 0x8e:
-            tcg_gen_sar_i64(tmp2, tmp2, tmp);
-            set_cc_s64(s, tmp2);
-            break;
+    }
+
+    memset(f, 0, sizeof(*f));
+    f->op = op;
+    f->op2 = op2;
+
+    /* Lookup the instruction.  */
+    info = lookup_opc(op << 8 | op2);
+
+    /* If we found it, extract the operands.  */
+    if (info != NULL) {
+        DisasFormat fmt = info->fmt;
+        int i;
+
+        for (i = 0; i < NUM_C_FIELD; ++i) {
+            extract_field(f, &format_info[fmt].op[i], insn);
         }
-        tcg_gen_shri_i64(tmp, tmp2, 32);
-        tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-        store_reg32(r1, tmp32_1);
-        tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-        store_reg32(r1 + 1, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x98: /* LM     R1,R3,D2(B2)     [RS] */
-    case 0x90: /* STM    R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = tcg_temp_new_i64();
-        tmp3 = tcg_const_i64(4);
-        tmp4 = tcg_const_i64(0xffffffff00000000ULL);
-        for (i = r1;; i = (i + 1) % 16) {
-            if (opc == 0x98) {
-                tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-                tcg_gen_and_i64(regs[i], regs[i], tmp4);
-                tcg_gen_or_i64(regs[i], regs[i], tmp2);
-            } else {
-                tcg_gen_qemu_st32(regs[i], tmp, get_mem_index(s));
-            }
-            if (i == r3) {
-                break;
+    }
+    return info;
+}
+
+static ExitStatus translate_one(CPUS390XState *env, DisasContext *s)
+{
+    const DisasInsn *insn;
+    ExitStatus ret = NO_EXIT;
+    DisasFields f;
+    DisasOps o;
+
+    /* Search for the insn in the table.  */
+    insn = extract_insn(env, s, &f);
+
+    /* Not found means unimplemented/illegal opcode.  */
+    if (insn == NULL) {
+        qemu_log_mask(LOG_UNIMP, "unimplemented opcode 0x%02x%02x\n",
+                      f.op, f.op2);
+        gen_illegal_opcode(s);
+        return EXIT_NORETURN;
+    }
+
+    /* Check for insn specification exceptions.  */
+    if (insn->spec) {
+        int spec = insn->spec, excp = 0, r;
+
+        if (spec & SPEC_r1_even) {
+            r = get_field(&f, r1);
+            if (r & 1) {
+                excp = PGM_SPECIFICATION;
             }
-            tcg_gen_add_i64(tmp, tmp, tmp3);
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        tcg_temp_free_i64(tmp4);
-        break;
-    case 0x91: /* TM     D1(B1),I2        [SI] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_si(s, insn, &i2, &b1, &d1);
-        tmp2 = tcg_const_i64(i2);
-        tcg_gen_qemu_ld8u(tmp, tmp, get_mem_index(s));
-        cmp_64(s, tmp, tmp2, CC_OP_TM_32);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x92: /* MVI    D1(B1),I2        [SI] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_si(s, insn, &i2, &b1, &d1);
-        tmp2 = tcg_const_i64(i2);
-        tcg_gen_qemu_st8(tmp2, tmp, get_mem_index(s));
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x94: /* NI     D1(B1),I2        [SI] */
-    case 0x96: /* OI     D1(B1),I2        [SI] */
-    case 0x97: /* XI     D1(B1),I2        [SI] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_si(s, insn, &i2, &b1, &d1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8u(tmp2, tmp, get_mem_index(s));
-        switch (opc) {
-        case 0x94:
-            tcg_gen_andi_i64(tmp2, tmp2, i2);
-            break;
-        case 0x96:
-            tcg_gen_ori_i64(tmp2, tmp2, i2);
-            break;
-        case 0x97:
-            tcg_gen_xori_i64(tmp2, tmp2, i2);
-            break;
-        default:
-            tcg_abort();
         }
-        tcg_gen_qemu_st8(tmp2, tmp, get_mem_index(s));
-        set_cc_nz_u64(s, tmp2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x95: /* CLI    D1(B1),I2        [SI] */
-        insn = ld_code4(env, s->pc);
-        tmp = decode_si(s, insn, &i2, &b1, &d1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_qemu_ld8u(tmp2, tmp, get_mem_index(s));
-        cmp_u64c(s, tmp2, i2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0x9a: /* LAM      R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_lam(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0x9b: /* STAM     R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_stam(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xa5:
-        insn = ld_code4(env, s->pc);
-        r1 = (insn >> 20) & 0xf;
-        op = (insn >> 16) & 0xf;
-        i2 = insn & 0xffff;
-        disas_a5(env, s, op, r1, i2);
-        break;
-    case 0xa7:
-        insn = ld_code4(env, s->pc);
-        r1 = (insn >> 20) & 0xf;
-        op = (insn >> 16) & 0xf;
-        i2 = (short)insn;
-        disas_a7(env, s, op, r1, i2);
-        break;
-    case 0xa8: /* MVCLE   R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_mvcle(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xa9: /* CLCLE   R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_clcle(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-#ifndef CONFIG_USER_ONLY
-    case 0xac: /* STNSM   D1(B1),I2     [SI] */
-    case 0xad: /* STOSM   D1(B1),I2     [SI] */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        tmp = decode_si(s, insn, &i2, &b1, &d1);
-        tmp2 = tcg_temp_new_i64();
-        tcg_gen_shri_i64(tmp2, psw_mask, 56);
-        tcg_gen_qemu_st8(tmp2, tmp, get_mem_index(s));
-        if (opc == 0xac) {
-            tcg_gen_andi_i64(psw_mask, psw_mask,
-                    ((uint64_t)i2 << 56) | 0x00ffffffffffffffULL);
-        } else {
-            tcg_gen_ori_i64(psw_mask, psw_mask, (uint64_t)i2 << 56);
-        }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-    case 0xae: /* SIGP   R1,R3,D2(B2)     [RS] */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp2 = load_reg(r3);
-        tmp32_1 = tcg_const_i32(r1);
-        potential_page_fault(s);
-        gen_helper_sigp(cc_op, cpu_env, tmp, tmp32_1, tmp2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-    case 0xb1: /* LRA    R1,D2(X2, B2)     [RX] */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        tmp = decode_rx(s, insn, &r1, &x2, &b2, &d2);
-        tmp32_1 = tcg_const_i32(r1);
-        potential_page_fault(s);
-        gen_helper_lra(cc_op, cpu_env, tmp, tmp32_1);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        break;
-#endif
-    case 0xb2:
-        insn = ld_code4(env, s->pc);
-        op = (insn >> 16) & 0xff;
-        switch (op) {
-        case 0x9c: /* STFPC    D2(B2) [S] */
-            d2 = insn & 0xfff;
-            b2 = (insn >> 12) & 0xf;
-            tmp32_1 = tcg_temp_new_i32();
-            tmp = tcg_temp_new_i64();
-            tmp2 = get_address(s, 0, b2, d2);
-            tcg_gen_ld_i32(tmp32_1, cpu_env, offsetof(CPUS390XState, fpc));
-            tcg_gen_extu_i32_i64(tmp, tmp32_1);
-            tcg_gen_qemu_st32(tmp, tmp2, get_mem_index(s));
-            tcg_temp_free_i32(tmp32_1);
-            tcg_temp_free_i64(tmp);
-            tcg_temp_free_i64(tmp2);
-            break;
-        default:
-            disas_b2(env, s, op, insn);
-            break;
+        if (spec & SPEC_r2_even) {
+            r = get_field(&f, r2);
+            if (r & 1) {
+                excp = PGM_SPECIFICATION;
+            }
         }
-        break;
-    case 0xb3:
-        insn = ld_code4(env, s->pc);
-        op = (insn >> 16) & 0xff;
-        r3 = (insn >> 12) & 0xf; /* aka m3 */
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        disas_b3(env, s, op, r3, r1, r2);
-        break;
-#ifndef CONFIG_USER_ONLY
-    case 0xb6: /* STCTL     R1,R3,D2(B2)     [RS] */
-        /* Store Control */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_stctl(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xb7: /* LCTL      R1,R3,D2(B2)     [RS] */
-        /* Load Control */
-        check_privileged(env, s, ilc);
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_lctl(cpu_env, tmp32_1, tmp, tmp32_2);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-#endif
-    case 0xb9:
-        insn = ld_code4(env, s->pc);
-        r1 = (insn >> 4) & 0xf;
-        r2 = insn & 0xf;
-        op = (insn >> 16) & 0xff;
-        disas_b9(env, s, op, r1, r2);
-        break;
-    case 0xba: /* CS     R1,R3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = tcg_const_i32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_cs(cc_op, cpu_env, tmp32_1, tmp, tmp32_2);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xbd: /* CLM    R1,M3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_clm(cc_op, cpu_env, tmp32_1, tmp32_2, tmp);
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xbe: /* STCM R1,M3,D2(B2) [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        tmp = get_address(s, 0, b2, d2);
-        tmp32_1 = load_reg32(r1);
-        tmp32_2 = tcg_const_i32(r3);
-        potential_page_fault(s);
-        gen_helper_stcm(cpu_env, tmp32_1, tmp32_2, tmp);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i32(tmp32_1);
-        tcg_temp_free_i32(tmp32_2);
-        break;
-    case 0xbf: /* ICM    R1,M3,D2(B2)     [RS] */
-        insn = ld_code4(env, s->pc);
-        decode_rs(s, insn, &r1, &r3, &b2, &d2);
-        if (r3 == 15) {
-            /* effectively a 32-bit load */
-            tmp = get_address(s, 0, b2, d2);
-            tmp32_1 = tcg_temp_new_i32();
-            tmp32_2 = tcg_const_i32(r3);
-            tcg_gen_qemu_ld32u(tmp, tmp, get_mem_index(s));
-            store_reg32_i64(r1, tmp);
-            tcg_gen_trunc_i64_i32(tmp32_1, tmp);
-            set_cc_icm(s, tmp32_2, tmp32_1);
-            tcg_temp_free_i64(tmp);
-            tcg_temp_free_i32(tmp32_1);
-            tcg_temp_free_i32(tmp32_2);
-        } else if (r3) {
-            uint32_t mask = 0x00ffffffUL;
-            uint32_t shift = 24;
-            int m3 = r3;
-            tmp = get_address(s, 0, b2, d2);
-            tmp2 = tcg_temp_new_i64();
-            tmp32_1 = load_reg32(r1);
-            tmp32_2 = tcg_temp_new_i32();
-            tmp32_3 = tcg_const_i32(r3);
-            tmp32_4 = tcg_const_i32(0);
-            while (m3) {
-                if (m3 & 8) {
-                    tcg_gen_qemu_ld8u(tmp2, tmp, get_mem_index(s));
-                    tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
-                    if (shift) {
-                        tcg_gen_shli_i32(tmp32_2, tmp32_2, shift);
-                    }
-                    tcg_gen_andi_i32(tmp32_1, tmp32_1, mask);
-                    tcg_gen_or_i32(tmp32_1, tmp32_1, tmp32_2);
-                    tcg_gen_or_i32(tmp32_4, tmp32_4, tmp32_2);
-                    tcg_gen_addi_i64(tmp, tmp, 1);
-                }
-                m3 = (m3 << 1) & 0xf;
-                mask = (mask >> 8) | 0xff000000UL;
-                shift -= 8;
+        if (spec & SPEC_r3_even) {
+            r = get_field(&f, r3);
+            if (r & 1) {
+                excp = PGM_SPECIFICATION;
             }
-            store_reg32(r1, tmp32_1);
-            set_cc_icm(s, tmp32_3, tmp32_4);
-            tcg_temp_free_i64(tmp);
-            tcg_temp_free_i64(tmp2);
-            tcg_temp_free_i32(tmp32_1);
-            tcg_temp_free_i32(tmp32_2);
-            tcg_temp_free_i32(tmp32_3);
-            tcg_temp_free_i32(tmp32_4);
-        } else {
-            /* i.e. env->cc = 0 */
-            gen_op_movi_cc(s, 0);
         }
-        break;
-    case 0xc0:
-    case 0xc2:
-        insn = ld_code6(env, s->pc);
-        r1 = (insn >> 36) & 0xf;
-        op = (insn >> 32) & 0xf;
-        i2 = (int)insn;
-        switch (opc) {
-        case 0xc0:
-            disas_c0(env, s, op, r1, i2);
-            break;
-        case 0xc2:
-            disas_c2(env, s, op, r1, i2);
-            break;
-        default:
-            tcg_abort();
+        if (spec & SPEC_r1_f128) {
+            r = get_field(&f, r1);
+            if (r > 13) {
+                excp = PGM_SPECIFICATION;
+            }
         }
-        break;
-    case 0xd2: /* MVC    D1(L,B1),D2(B2)         [SS] */
-    case 0xd4: /* NC     D1(L,B1),D2(B2)         [SS] */
-    case 0xd5: /* CLC    D1(L,B1),D2(B2)         [SS] */
-    case 0xd6: /* OC     D1(L,B1),D2(B2)         [SS] */
-    case 0xd7: /* XC     D1(L,B1),D2(B2)         [SS] */
-    case 0xdc: /* TR     D1(L,B1),D2(B2)         [SS] */
-    case 0xf3: /* UNPK   D1(L1,B1),D2(L2,B2)     [SS] */
-        insn = ld_code6(env, s->pc);
-        vl = tcg_const_i32((insn >> 32) & 0xff);
-        b1 = (insn >> 28) & 0xf;
-        b2 = (insn >> 12) & 0xf;
-        d1 = (insn >> 16) & 0xfff;
-        d2 = insn & 0xfff;
-        tmp = get_address(s, 0, b1, d1);
-        tmp2 = get_address(s, 0, b2, d2);
-        switch (opc) {
-        case 0xd2:
-            gen_op_mvc(s, (insn >> 32) & 0xff, tmp, tmp2);
-            break;
-        case 0xd4:
-            potential_page_fault(s);
-            gen_helper_nc(cc_op, cpu_env, vl, tmp, tmp2);
-            set_cc_static(s);
-            break;
-        case 0xd5:
-            gen_op_clc(s, (insn >> 32) & 0xff, tmp, tmp2);
-            break;
-        case 0xd6:
-            potential_page_fault(s);
-            gen_helper_oc(cc_op, cpu_env, vl, tmp, tmp2);
-            set_cc_static(s);
-            break;
-        case 0xd7:
-            potential_page_fault(s);
-            gen_helper_xc(cc_op, cpu_env, vl, tmp, tmp2);
-            set_cc_static(s);
-            break;
-        case 0xdc:
-            potential_page_fault(s);
-            gen_helper_tr(cpu_env, vl, tmp, tmp2);
-            set_cc_static(s);
-            break;
-        case 0xf3:
-            potential_page_fault(s);
-            gen_helper_unpk(cpu_env, vl, tmp, tmp2);
-            break;
-        default:
-            tcg_abort();
+        if (spec & SPEC_r2_f128) {
+            r = get_field(&f, r2);
+            if (r > 13) {
+                excp = PGM_SPECIFICATION;
+            }
         }
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        break;
-#ifndef CONFIG_USER_ONLY
-    case 0xda: /* MVCP     D1(R1,B1),D2(B2),R3   [SS] */
-    case 0xdb: /* MVCS     D1(R1,B1),D2(B2),R3   [SS] */
-        check_privileged(env, s, ilc);
-        potential_page_fault(s);
-        insn = ld_code6(env, s->pc);
-        r1 = (insn >> 36) & 0xf;
-        r3 = (insn >> 32) & 0xf;
-        b1 = (insn >> 28) & 0xf;
-        d1 = (insn >> 16) & 0xfff;
-        b2 = (insn >> 12) & 0xf;
-        d2 = insn & 0xfff;
-        tmp = load_reg(r1);
-        /* XXX key in r3 */
-        tmp2 = get_address(s, 0, b1, d1);
-        tmp3 = get_address(s, 0, b2, d2);
-        if (opc == 0xda) {
-            gen_helper_mvcp(cc_op, cpu_env, tmp, tmp2, tmp3);
-        } else {
-            gen_helper_mvcs(cc_op, cpu_env, tmp, tmp2, tmp3);
+        if (excp) {
+            gen_program_exception(s, excp);
+            return EXIT_NORETURN;
         }
-        set_cc_static(s);
-        tcg_temp_free_i64(tmp);
-        tcg_temp_free_i64(tmp2);
-        tcg_temp_free_i64(tmp3);
-        break;
-#endif
-    case 0xe3:
-        insn = ld_code6(env, s->pc);
-        debug_insn(insn);
-        op = insn & 0xff;
-        r1 = (insn >> 36) & 0xf;
-        x2 = (insn >> 32) & 0xf;
-        b2 = (insn >> 28) & 0xf;
-        d2 = ((int)((((insn >> 16) & 0xfff)
-           | ((insn << 4) & 0xff000)) << 12)) >> 12;
-        disas_e3(env, s, op,  r1, x2, b2, d2 );
-        break;
-#ifndef CONFIG_USER_ONLY
-    case 0xe5:
-        /* Test Protection */
-        check_privileged(env, s, ilc);
-        insn = ld_code6(env, s->pc);
-        debug_insn(insn);
-        disas_e5(env, s, insn);
-        break;
-#endif
-    case 0xeb:
-        insn = ld_code6(env, s->pc);
-        debug_insn(insn);
-        op = insn & 0xff;
-        r1 = (insn >> 36) & 0xf;
-        r3 = (insn >> 32) & 0xf;
-        b2 = (insn >> 28) & 0xf;
-        d2 = ((int)((((insn >> 16) & 0xfff)
-           | ((insn << 4) & 0xff000)) << 12)) >> 12;
-        disas_eb(env, s, op, r1, r3, b2, d2);
-        break;
-    case 0xed:
-        insn = ld_code6(env, s->pc);
-        debug_insn(insn);
-        op = insn & 0xff;
-        r1 = (insn >> 36) & 0xf;
-        x2 = (insn >> 32) & 0xf;
-        b2 = (insn >> 28) & 0xf;
-        d2 = (short)((insn >> 16) & 0xfff);
-        r1b = (insn >> 12) & 0xf;
-        disas_ed(env, s, op, r1, x2, b2, d2, r1b);
-        break;
-    default:
-        qemu_log_mask(LOG_UNIMP, "unimplemented opcode 0x%x\n", opc);
-        gen_illegal_opcode(env, s, ilc);
-        break;
     }
 
-    /* Instruction length is encoded in the opcode */
-    s->pc += (ilc * 2);
+    /* Set up the strutures we use to communicate with the helpers. */
+    s->insn = insn;
+    s->fields = &f;
+    o.g_out = o.g_out2 = o.g_in1 = o.g_in2 = false;
+    TCGV_UNUSED_I64(o.out);
+    TCGV_UNUSED_I64(o.out2);
+    TCGV_UNUSED_I64(o.in1);
+    TCGV_UNUSED_I64(o.in2);
+    TCGV_UNUSED_I64(o.addr1);
+
+    /* Implement the instruction.  */
+    if (insn->help_in1) {
+        insn->help_in1(s, &f, &o);
+    }
+    if (insn->help_in2) {
+        insn->help_in2(s, &f, &o);
+    }
+    if (insn->help_prep) {
+        insn->help_prep(s, &f, &o);
+    }
+    if (insn->help_op) {
+        ret = insn->help_op(s, &o);
+    }
+    if (insn->help_wout) {
+        insn->help_wout(s, &f, &o);
+    }
+    if (insn->help_cout) {
+        insn->help_cout(s, &o);
+    }
+
+    /* Free any temporaries created by the helpers.  */
+    if (!TCGV_IS_UNUSED_I64(o.out) && !o.g_out) {
+        tcg_temp_free_i64(o.out);
+    }
+    if (!TCGV_IS_UNUSED_I64(o.out2) && !o.g_out2) {
+        tcg_temp_free_i64(o.out2);
+    }
+    if (!TCGV_IS_UNUSED_I64(o.in1) && !o.g_in1) {
+        tcg_temp_free_i64(o.in1);
+    }
+    if (!TCGV_IS_UNUSED_I64(o.in2) && !o.g_in2) {
+        tcg_temp_free_i64(o.in2);
+    }
+    if (!TCGV_IS_UNUSED_I64(o.addr1)) {
+        tcg_temp_free_i64(o.addr1);
+    }
+
+    /* Advance to the next instruction.  */
+    s->pc = s->next_pc;
+    return ret;
 }
 
 static inline void gen_intermediate_code_internal(CPUS390XState *env,
@@ -5121,6 +4745,8 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
     int j, lj = -1;
     int num_insns, max_insns;
     CPUBreakpoint *bp;
+    ExitStatus status;
+    bool do_debug;
 
     pc_start = tb->pc;
 
@@ -5129,10 +4755,10 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
         pc_start &= 0x7fffffff;
     }
 
-    dc.pc = pc_start;
-    dc.is_jmp = DISAS_NEXT;
     dc.tb = tb;
+    dc.pc = pc_start;
     dc.cc_op = CC_OP_DYNAMIC;
+    do_debug = dc.singlestep_enabled = env->singlestep_enabled;
 
     gen_opc_end = tcg_ctx.gen_opc_buf + OPC_MAX_SIZE;
 
@@ -5147,14 +4773,6 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
     gen_icount_start();
 
     do {
-        if (unlikely(!QTAILQ_EMPTY(&env->breakpoints))) {
-            QTAILQ_FOREACH(bp, &env->breakpoints, entry) {
-                if (bp->pc == dc.pc) {
-                    gen_debug(&dc);
-                    break;
-                }
-            }
-        }
         if (search_pc) {
             j = tcg_ctx.gen_opc_ptr - tcg_ctx.gen_opc_buf;
             if (lj < j) {
@@ -5168,7 +4786,7 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
             tcg_ctx.gen_opc_instr_start[lj] = 1;
             tcg_ctx.gen_opc_icount[lj] = num_insns;
         }
-        if (num_insns + 1 == max_insns && (tb->cflags & CF_LAST_IO)) {
+        if (++num_insns == max_insns && (tb->cflags & CF_LAST_IO)) {
             gen_io_start();
         }
 
@@ -5176,36 +4794,58 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
             tcg_gen_debug_insn_start(dc.pc);
         }
 
-        disas_s390_insn(env, &dc);
-
-        num_insns++;
-        if (env->singlestep_enabled) {
-            gen_debug(&dc);
+        status = NO_EXIT;
+        if (unlikely(!QTAILQ_EMPTY(&env->breakpoints))) {
+            QTAILQ_FOREACH(bp, &env->breakpoints, entry) {
+                if (bp->pc == dc.pc) {
+                    status = EXIT_PC_STALE;
+                    do_debug = true;
+                    break;
+                }
+            }
+        }
+        if (status == NO_EXIT) {
+            status = translate_one(env, &dc);
         }
-    } while (!dc.is_jmp && tcg_ctx.gen_opc_ptr < gen_opc_end
-             && dc.pc < next_page_start
-             && num_insns < max_insns && !env->singlestep_enabled
-             && !singlestep);
-
-    if (!dc.is_jmp) {
-        update_psw_addr(&dc);
-    }
 
-    if (singlestep && dc.cc_op != CC_OP_DYNAMIC) {
-        gen_op_calc_cc(&dc);
-    } else {
-        /* next TB starts off with CC_OP_DYNAMIC, so make sure the cc op type
-           is in env */
-        gen_op_set_cc_op(&dc);
-    }
+        /* If we reach a page boundary, are single stepping,
+           or exhaust instruction count, stop generation.  */
+        if (status == NO_EXIT
+            && (dc.pc >= next_page_start
+                || tcg_ctx.gen_opc_ptr >= gen_opc_end
+                || num_insns >= max_insns
+                || singlestep
+                || env->singlestep_enabled)) {
+            status = EXIT_PC_STALE;
+        }
+    } while (status == NO_EXIT);
 
     if (tb->cflags & CF_LAST_IO) {
         gen_io_end();
     }
-    /* Generate the return instruction */
-    if (dc.is_jmp != DISAS_TB_JUMP) {
-        tcg_gen_exit_tb(0);
+
+    switch (status) {
+    case EXIT_GOTO_TB:
+    case EXIT_NORETURN:
+        break;
+    case EXIT_PC_STALE:
+        update_psw_addr(&dc);
+        /* FALLTHRU */
+    case EXIT_PC_UPDATED:
+        /* Next TB starts off with CC_OP_DYNAMIC, so make sure the
+           cc op type is in env */
+        update_cc_op(&dc);
+        /* Exit the TB, either by raising a debug exception or by return.  */
+        if (do_debug) {
+            gen_exception(EXCP_DEBUG);
+        } else {
+            tcg_gen_exit_tb(0);
+        }
+        break;
+    default:
+        abort();
     }
+
     gen_icount_end(tb, num_insns);
     *tcg_ctx.gen_opc_ptr = INDEX_op_end;
     if (search_pc) {
@@ -5218,6 +4858,7 @@ static inline void gen_intermediate_code_internal(CPUS390XState *env,
         tb->size = dc.pc - pc_start;
         tb->icount = num_insns;
     }
+
 #if defined(S390X_DEBUG_DISAS)
     if (qemu_loglevel_mask(CPU_LOG_TB_IN_ASM)) {
         qemu_log("IN: %s\n", lookup_symbol(pc_start));
author	Blue Swirl <blauwirbel@gmail.com>	2013-01-12 12:46:57 +0000
committer	Blue Swirl <blauwirbel@gmail.com>	2013-01-12 12:46:57 +0000
commit	837d1f978224f7e7b020c71ffb10b291952cc596 (patch)
tree	f1e4e0b95ad756a48976b6d6889b41e44552454a
parent	a6fc23e5dc76660792c8363d058adf22caa82945 (diff)
parent	2b35e93fbfc52c2be6cf85e5a54a3707cdabd914 (diff)
download	qemu-837d1f978224f7e7b020c71ffb10b291952cc596.tar.gz