Update LuaJIT to LuaJIT/LuaJIT@d0e8893

CMakeLists.txt

@@ -167,7 +167,7 @@ endif()
 
 set(MEGA_ZLIB_VER "1.2.12")
 set(MEGA_LUA51_VER "5.1.5")
-set(MEGA_LUAJIT_VER "2.1.0-7306ba7")
+set(MEGA_LUAJIT_VER "2.1.0-d0e8893")
 set(MEGA_LIBOGG_VER "1.3.2")
 set(MEGA_LIBVORBIS_VER "1.3.5")
 set(MEGA_LIBTHEORA_VER "1.1.1")

libs/LuaJIT/doc/contact.html

@@ -50,8 +50,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>
@@ -86,6 +84,13 @@ xD("fyZKB8xv\"FJytmz8.KAB0u52D")
 </p>
 </noscript>
 
+<p><i>
+Note: I cannot reply to GMail, Google Workplace, Outlook or Office365
+mail addresses, since they prefer to mindlessly filter out mails sent
+from small domains using independent mail servers, such as mine. If you
+don't like that, please complain to Google or Microsoft, not me.
+</i></p>
+
 <h2>Copyright</h2>
 <p>
 All documentation is

libs/LuaJIT/doc/ext_buffer.html

@@ -61,8 +61,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_c_api.html

@@ -50,8 +50,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_ffi.html

@@ -50,8 +50,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_ffi_api.html

@@ -55,8 +55,6 @@ td.abiparam { font-weight: bold; width: 6em; }
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_ffi_semantics.html

@@ -55,8 +55,6 @@ td.convop { font-style: italic; width: 40%; }
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_ffi_tutorial.html

@@ -57,8 +57,6 @@ td.idiomlua b { font-weight: normal; color: #2142bf; }
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_jit.html

@@ -50,8 +50,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/ext_profiler.html

@@ -50,8 +50,6 @@
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/extensions.html

@@ -67,8 +67,6 @@ td.excinterop {
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/faq.html

@@ -53,8 +53,6 @@ dd { margin-left: 1.5em; }
 </li><li>
 <a class="current" href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>
@@ -65,8 +63,6 @@ dd { margin-left: 1.5em; }
 <ul style="padding: 0;">
 <li>The <a href="https://luajit.org/list.html"><span class="ext">&raquo;</span>&nbsp;LuaJIT mailing list</a> focuses on topics
 related to LuaJIT.</li>
-<li>The <a href="http://wiki.luajit.org/"><span class="ext">&raquo;</span>&nbsp;LuaJIT wiki</a> gathers community
-resources about LuaJIT.</li>
 <li>News about Lua itself can be found at the
 <a href="https://www.lua.org/lua-l.html"><span class="ext">&raquo;</span>&nbsp;Lua mailing list</a>.
 The mailing list archives are worth checking out for older postings

libs/LuaJIT/doc/install.html

@@ -73,8 +73,6 @@ td.compatx {
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/luajit.html

@@ -135,8 +135,6 @@ table.feature small {
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/doc/running.html

@@ -72,8 +72,6 @@ td.param_default {
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>
@@ -222,6 +220,12 @@ mix the three forms, but note that setting an optimization level
 overrides all earlier flags.
 </p>
 <p>
+Note that <tt>-Ofma</tt> is not enabled by default at any level,
+because it affects floating-point result accuracy. Only enable this,
+if you fully understand the trade-offs of FMA for performance (higher),
+determinism (lower) and numerical accuracy (higher).
+</p>
+<p>
 Here are the available flags and at what optimization levels they
 are enabled:
 </p>
@@ -253,6 +257,8 @@ are enabled:
 <td class="flag_name">sink</td><td class="flag_level">&nbsp;</td><td class="flag_level">&nbsp;</td><td class="flag_level">&bull;</td><td class="flag_desc">Allocation/Store Sinking</td></tr>
 <tr class="even">
 <td class="flag_name">fuse</td><td class="flag_level">&nbsp;</td><td class="flag_level">&nbsp;</td><td class="flag_level">&bull;</td><td class="flag_desc">Fusion of operands into instructions</td></tr>
+<tr class="odd">
+<td class="flag_name">fma </td><td class="flag_level">&nbsp;</td><td class="flag_level">&nbsp;</td><td class="flag_level">&nbsp;</td><td class="flag_desc">Fused multiply-add</td></tr>
 </table>
 <p>
 Here are the parameters and their default settings:

libs/LuaJIT/doc/status.html

@@ -53,8 +53,6 @@ ul li { padding-bottom: 0.3em; }
 </li><li>
 <a href="faq.html">FAQ</a>
 </li><li>
-<a href="http://wiki.luajit.org/">Wiki <span class="ext">&raquo;</span></a>
-</li><li>
 <a href="https://luajit.org/list.html">Mailing List <span class="ext">&raquo;</span></a>
 </li></ul>
 </div>

libs/LuaJIT/src/lib_base.c

@@ -303,7 +303,7 @@ LJLIB_ASM(tonumber)		LJLIB_REC(.)
 	while (lj_char_isspace((unsigned char)(*ep))) ep++;
 	if (*ep == '\0') {
 	  if (LJ_DUALNUM && LJ_LIKELY(ul < 0x80000000u+neg)) {
-	    if (neg) ul = (unsigned long)-(long)ul;
+	    if (neg) ul = ~ul+1u;
 	    setintV(L->base-1-LJ_FR2, (int32_t)ul);
 	  } else {
 	    lua_Number n = (lua_Number)ul;

libs/LuaJIT/src/lib_bit.c

@@ -155,7 +155,8 @@ LJLIB_CF(bit_tohex)		LJLIB_REC(.)
 #endif
   SBuf *sb = lj_buf_tmp_(L);
   SFormat sf = (STRFMT_UINT|STRFMT_T_HEX);
-  if (n < 0) { n = -n; sf |= STRFMT_F_UPPER; }
+  if (n < 0) { n = (int32_t)(~(uint32_t)n+1u); sf |= STRFMT_F_UPPER; }
+  if ((uint32_t)n > 254) n = 254;
   sf |= ((SFormat)((n+1)&255) << STRFMT_SH_PREC);
 #if LJ_HASFFI
   if (n < 16) b &= ((uint64_t)1 << 4*n)-1;

libs/LuaJIT/src/lib_buffer.c

@@ -128,7 +128,7 @@ LJLIB_CF(buffer_method_put)		LJLIB_REC(.)
       lj_strfmt_putfnum((SBuf *)sbx, STRFMT_G14, numV(o));
     } else if (tvisbuf(o)) {
       SBufExt *sbx2 = bufV(o);
-      if (sbx2 == sbx) lj_err_arg(L, arg+1, LJ_ERR_BUFFER_SELF);
+      if (sbx2 == sbx) lj_err_arg(L, (int)(arg+1), LJ_ERR_BUFFER_SELF);
       lj_buf_putmem((SBuf *)sbx, sbx2->r, sbufxlen(sbx2));
     } else if (!mo && !tvisnil(mo = lj_meta_lookup(L, o, MM_tostring))) {
       /* Call __tostring metamethod inline. */
@@ -140,7 +140,7 @@ LJLIB_CF(buffer_method_put)		LJLIB_REC(.)
       L->top = L->base + narg;
       goto retry;  /* Retry with the result. */
     } else {
-      lj_err_argtype(L, arg+1, "string/number/__tostring");
+      lj_err_argtype(L, (int)(arg+1), "string/number/__tostring");
     }
     /* Probably not useful to inline other __tostring MMs, e.g. FFI numbers. */
   }
@@ -169,7 +169,7 @@ LJLIB_CF(buffer_method_get)		LJLIB_REC(.)
   for (arg = 1; arg < narg; arg++) {
     TValue *o = &L->base[arg];
     MSize n = tvisnil(o) ? LJ_MAX_BUF :
-	      (MSize) lj_lib_checkintrange(L, arg+1, 0, LJ_MAX_BUF);
+	      (MSize) lj_lib_checkintrange(L, (int)(arg+1), 0, LJ_MAX_BUF);
     MSize len = sbufxlen(sbx);
     if (n > len) n = len;
     setstrV(L, o, lj_str_new(L, sbx->r, n));
@@ -177,7 +177,7 @@ LJLIB_CF(buffer_method_get)		LJLIB_REC(.)
   }
   if (sbx->r == sbx->w && !sbufiscow(sbx)) sbx->r = sbx->w = sbx->b;
   lj_gc_check(L);
-  return narg-1;
+  return (int)(narg-1);
 }
 
 #if LJ_HASFFI

libs/LuaJIT/src/lj_arch.h

@@ -93,6 +93,9 @@
 #elif defined(__CYGWIN__)
 #define LJ_TARGET_CYGWIN	1
 #define LUAJIT_OS	LUAJIT_OS_POSIX
+#elif defined(__QNX__)
+#define LJ_TARGET_QNX		1
+#define LUAJIT_OS	LUAJIT_OS_POSIX
 #else
 #define LUAJIT_OS	LUAJIT_OS_OTHER
 #endif
@@ -463,11 +466,17 @@
 #endif
 #endif
 #elif !LJ_TARGET_PS3
+#if __clang__
+#if ((__clang_major__ < 3) || ((__clang_major__ == 3) && __clang_minor__ < 5))
+#error "Need at least Clang 3.5 or newer"
+#endif
+#else
 #if (__GNUC__ < 4) || ((__GNUC__ == 4) && __GNUC_MINOR__ < 3)
 #error "Need at least GCC 4.3 or newer"
 #endif
 #endif
 #endif
+#endif
 
 /* Check target-specific constraints. */
 #ifndef _BUILDVM_H

libs/LuaJIT/src/lj_asm_arm.h

@@ -313,7 +313,11 @@ static void asm_fusexref(ASMState *as, ARMIns ai, Reg rd, IRRef ref,
 }
 
 #if !LJ_SOFTFP
-/* Fuse to multiply-add/sub instruction. */
+/*
+** Fuse to multiply-add/sub instruction.
+** VMLA rounds twice (UMA, not FMA) -- no need to check for JIT_F_OPT_FMA.
+** VFMA needs VFPv4, which is uncommon on the remaining ARM32 targets.
+*/
 static int asm_fusemadd(ASMState *as, IRIns *ir, ARMIns ai, ARMIns air)
 {
   IRRef lref = ir->op1, rref = ir->op2;

libs/LuaJIT/src/lj_asm_arm64.h

@@ -337,7 +337,8 @@ static int asm_fusemadd(ASMState *as, IRIns *ir, A64Ins ai, A64Ins air)
 {
   IRRef lref = ir->op1, rref = ir->op2;
   IRIns *irm;
-  if (lref != rref &&
+  if ((as->flags & JIT_F_OPT_FMA) &&
+      lref != rref &&
       ((mayfuse(as, lref) && (irm = IR(lref), irm->o == IR_MUL) &&
        ra_noreg(irm->r)) ||
        (mayfuse(as, rref) && (irm = IR(rref), irm->o == IR_MUL) &&
@@ -1201,7 +1202,7 @@ dotypecheck:
       tmp = ra_scratch(as, allow);
       rset_clear(allow, tmp);
     }
-    if (irt_isnum(t) && !(ir->op2 & IRSLOAD_CONVERT))
+    if (ra_hasreg(dest) && tmp != dest)
       emit_dn(as, A64I_FMOV_D_R, (dest & 31), tmp);
     /* Need type check, even if the load result is unused. */
     asm_guardcc(as, irt_isnum(t) ? CC_LS : CC_NE);

libs/LuaJIT/src/lj_asm_mips.h

@@ -1894,7 +1894,7 @@ static void asm_arithov(ASMState *as, IRIns *ir)
   lj_assertA(!irt_is64(ir->t), "bad usage");
   if (irref_isk(ir->op2)) {
     int k = IR(ir->op2)->i;
-    if (ir->o == IR_SUBOV) k = -k;
+    if (ir->o == IR_SUBOV) k = (int)(~(unsigned int)k+1u);
     if (checki16(k)) {  /* (dest < left) == (k >= 0 ? 1 : 0) */
       left = ra_alloc1(as, ir->op1, RSET_GPR);
       asm_guard(as, k >= 0 ? MIPSI_BNE : MIPSI_BEQ, RID_TMP, RID_ZERO);

libs/LuaJIT/src/lj_asm_ppc.h

@@ -235,7 +235,8 @@ static int asm_fusemadd(ASMState *as, IRIns *ir, PPCIns pi, PPCIns pir)
 {
   IRRef lref = ir->op1, rref = ir->op2;
   IRIns *irm;
-  if (lref != rref &&
+  if ((as->flags & JIT_F_OPT_FMA) &&
+      lref != rref &&
       ((mayfuse(as, lref) && (irm = IR(lref), irm->o == IR_MUL) &&
 	ra_noreg(irm->r)) ||
        (mayfuse(as, rref) && (irm = IR(rref), irm->o == IR_MUL) &&

libs/LuaJIT/src/lj_carith.c

@@ -207,7 +207,7 @@ static int carith_int64(lua_State *L, CTState *cts, CDArith *ca, MMS mm)
       else
 	*up = lj_carith_powu64(u0, u1);
       break;
-    case MM_unm: *up = (uint64_t)-(int64_t)u0; break;
+    case MM_unm: *up = ~u0+1u; break;
     default:
       lj_assertL(0, "bad metamethod %d", mm);
       break;

libs/LuaJIT/src/lj_cparse.c

@@ -488,7 +488,7 @@ static void cp_expr_prefix(CPState *cp, CPValue *k)
   } else if (cp_opt(cp, '+')) {
     cp_expr_unary(cp, k);  /* Nothing to do (well, integer promotion). */
   } else if (cp_opt(cp, '-')) {
-    cp_expr_unary(cp, k); k->i32 = -k->i32;
+    cp_expr_unary(cp, k); k->i32 = (int32_t)(~(uint32_t)k->i32+1);
   } else if (cp_opt(cp, '~')) {
     cp_expr_unary(cp, k); k->i32 = ~k->i32;
   } else if (cp_opt(cp, '!')) {

libs/LuaJIT/src/lj_crecord.c

@@ -1504,9 +1504,13 @@ void LJ_FASTCALL recff_cdata_arith(jit_State *J, RecordFFData *rd)
 	if (ctype_isenum(ct->info)) ct = ctype_child(cts, ct);
 	goto ok;
       } else if (ctype_isfunc(ct->info)) {
+	CTypeID id0 = i ? ctype_typeid(cts, s[0]) : 0;
 	tr = emitir(IRT(IR_FLOAD, IRT_PTR), tr, IRFL_CDATA_PTR);
 	ct = ctype_get(cts,
 	  lj_ctype_intern(cts, CTINFO(CT_PTR, CTALIGN_PTR|id), CTSIZE_PTR));
+	if (i) {
+	  s[0] = ctype_get(cts, id0);  /* cts->tab may have been reallocated. */
+	}
 	goto ok;
       } else {
 	tr = emitir(IRT(IR_ADD, IRT_PTR), tr, lj_ir_kintp(J, sizeof(GCcdata)));
@@ -1875,7 +1879,8 @@ TRef recff_bit64_tohex(jit_State *J, RecordFFData *rd, TRef hdr)
   } else {
     n = id ? 16 : 8;
   }
-  if (n < 0) { n = -n; sf |= STRFMT_F_UPPER; }
+  if (n < 0) { n = (int32_t)(~n+1u); sf |= STRFMT_F_UPPER; }
+  if ((uint32_t)n > 254) n = 254;
   sf |= ((SFormat)((n+1)&255) << STRFMT_SH_PREC);
   if (id) {
     tr = crec_ct_tv(J, ctype_get(cts, id), 0, J->base[0], &rd->argv[0]);

libs/LuaJIT/src/lj_ctype.c

@@ -191,8 +191,20 @@ CTypeID lj_ctype_intern(CTState *cts, CTInfo info, CTSize size)
   }
   id = cts->top;
   if (LJ_UNLIKELY(id >= cts->sizetab)) {
+#ifdef LUAJIT_CTYPE_CHECK_ANCHOR
+    CType *ct;
+#endif
     if (id >= CTID_MAX) lj_err_msg(cts->L, LJ_ERR_TABOV);
+#ifdef LUAJIT_CTYPE_CHECK_ANCHOR
+    ct = lj_mem_newvec(cts->L, id+1, CType);
+    memcpy(ct, cts->tab, id*sizeof(CType));
+    memset(cts->tab, 0, id*sizeof(CType));
+    lj_mem_freevec(cts->g, cts->tab, cts->sizetab, CType);
+    cts->tab = ct;
+    cts->sizetab = id+1;
+#else
     lj_mem_growvec(cts->L, cts->tab, cts->sizetab, CTID_MAX, CType);
+#endif
   }
   cts->top = id+1;
   cts->tab[id].info = info;
@@ -570,7 +582,7 @@ GCstr *lj_ctype_repr_int64(lua_State *L, uint64_t n, int isunsigned)
   if (isunsigned) {
     *--p = 'U';
   } else if ((int64_t)n < 0) {
-    n = (uint64_t)-(int64_t)n;
+    n = ~n+1u;
     sign = 1;
   }
   do { *--p = (char)('0' + n % 10); } while (n /= 10);

libs/LuaJIT/src/lj_debug.c

@@ -101,9 +101,12 @@ static BCPos debug_framepc(lua_State *L, GCfunc *fn, cTValue *nextframe)
   pos = proto_bcpos(pt, ins) - 1;
 #if LJ_HASJIT
   if (pos > pt->sizebc) {  /* Undo the effects of lj_trace_exit for JLOOP. */
-    GCtrace *T = (GCtrace *)((char *)(ins-1) - offsetof(GCtrace, startins));
+    if (bc_isret(bc_op(ins[-1]))) {
-    lj_assertL(bc_isret(bc_op(ins[-1])), "return bytecode expected");
+      GCtrace *T = (GCtrace *)((char *)(ins-1) - offsetof(GCtrace, startins));
-    pos = proto_bcpos(pt, mref(T->startpc, const BCIns));
+      pos = proto_bcpos(pt, mref(T->startpc, const BCIns));
+    } else {
+      pos = NO_BCPOS;  /* Punt in case of stack overflow for stitched trace. */
+    }
   }
 #endif
   return pos;

libs/LuaJIT/src/lj_emit_arm.h

@@ -157,7 +157,7 @@ static int emit_kdelta2(ASMState *as, Reg rd, int32_t i)
       if (other) {
 	int32_t delta = i - other;
 	uint32_t sh, inv = 0, k2, k;
-	if (delta < 0) { delta = -delta; inv = ARMI_ADD^ARMI_SUB; }
+	if (delta < 0) { delta = (int32_t)(~(uint32_t)delta+1u); inv = ARMI_ADD^ARMI_SUB; }
 	sh = lj_ffs(delta) & ~1;
 	k2 = emit_isk12(0, delta & (255 << sh));
 	k = emit_isk12(0, delta & ~(255 << sh));

libs/LuaJIT/src/lj_emit_arm64.h

@@ -27,8 +27,8 @@ static uint64_t get_k64val(ASMState *as, IRRef ref)
 /* Encode constant in K12 format for data processing instructions. */
 static uint32_t emit_isk12(int64_t n)
 {
-  uint64_t k = (n < 0) ? -n : n;
+  uint64_t k = n < 0 ? ~(uint64_t)n+1u : (uint64_t)n;
-  uint32_t m = (n < 0) ? 0x40000000 : 0;
+  uint32_t m = n < 0 ? 0x40000000 : 0;
   if (k < 0x1000) {
     return A64I_K12|m|A64F_U12(k);
   } else if ((k & 0xfff000) == k) {
@@ -177,7 +177,7 @@ static int emit_kdelta(ASMState *as, Reg rd, uint64_t k, int lim)
 	emit_dm(as, A64I_MOVx, rd, r);
 	return 1;
       } else {
-	uint32_t k12 = emit_isk12(delta < 0 ? -delta : delta);
+	uint32_t k12 = emit_isk12(delta < 0 ? (int64_t)(~(uint64_t)delta+1u) : delta);
 	if (k12) {
 	  emit_dn(as, (delta < 0 ? A64I_SUBx : A64I_ADDx)^k12, rd, r);
 	  return 1;
@@ -417,7 +417,8 @@ static void emit_addptr(ASMState *as, Reg r, int32_t ofs)
 {
   if (ofs)
     emit_opk(as, ofs < 0 ? A64I_SUBx : A64I_ADDx, r, r,
-		 ofs < 0 ? -ofs : ofs, rset_exclude(RSET_GPR, r));
+		 ofs < 0 ? (int32_t)(~(uint32_t)ofs+1u) : ofs,
+		 rset_exclude(RSET_GPR, r));
 }
 
 #define emit_spsub(as, ofs)	emit_addptr(as, RID_SP, -(ofs))

libs/LuaJIT/src/lj_err.c

@@ -329,12 +329,12 @@ static void err_unwind_win_jit(global_State *g, int errcode)
   memset(&hist, 0, sizeof(hist));
   RtlCaptureContext(&ctx);
   while (1) {
-    uintptr_t frame, base, addr = ctx.CONTEXT_REG_PC;
+    DWORD64 frame, base, addr = ctx.CONTEXT_REG_PC;
     void *hdata;
     PRUNTIME_FUNCTION func = RtlLookupFunctionEntry(addr, &base, &hist);
     if (!func) {  /* Found frame without .pdata: must be JIT-compiled code. */
       ExitNo exitno;
-      uintptr_t stub = lj_trace_unwind(G2J(g), addr - sizeof(MCode), &exitno);
+      uintptr_t stub = lj_trace_unwind(G2J(g), (uintptr_t)(addr - sizeof(MCode)), &exitno);
       if (stub) {  /* Jump to side exit to unwind the trace. */
 	ctx.CONTEXT_REG_PC = stub;
 	G2J(g)->exitcode = errcode;
@@ -777,6 +777,7 @@ LJ_NOINLINE void lj_err_mem(lua_State *L)
 {
   if (L->status == LUA_ERRERR+1)  /* Don't touch the stack during lua_open. */
     lj_vm_unwind_c(L->cframe, LUA_ERRMEM);
+  if (curr_funcisL(L)) L->top = curr_topL(L);
   setstrV(L, L->top++, lj_err_str(L, LJ_ERR_ERRMEM));
   lj_err_throw(L, LUA_ERRMEM);
 }

libs/LuaJIT/src/lj_ffrecord.c

@@ -1117,7 +1117,7 @@ static LJ_AINLINE TRef recff_sbufx_len(jit_State *J, TRef trr, TRef trw)
 }
 
 /* Emit typecheck for string buffer. */
-static TRef recff_sbufx_check(jit_State *J, RecordFFData *rd, int arg)
+static TRef recff_sbufx_check(jit_State *J, RecordFFData *rd, ptrdiff_t arg)
 {
   TRef trtype, ud = J->base[arg];
   if (!tvisbuf(&rd->argv[arg])) lj_trace_err(J, LJ_TRERR_BADTYPE);
@@ -1135,7 +1135,7 @@ static TRef recff_sbufx_write(jit_State *J, TRef ud)
 }
 
 /* Check for integer in range for the buffer API. */
-static TRef recff_sbufx_checkint(jit_State *J, RecordFFData *rd, int arg)
+static TRef recff_sbufx_checkint(jit_State *J, RecordFFData *rd, ptrdiff_t arg)
 {
   TRef tr = J->base[arg];
   TRef trlim = lj_ir_kint(J, LJ_MAX_BUF);

libs/LuaJIT/src/lj_jit.h

@@ -87,10 +87,11 @@
 #define JIT_F_OPT_ABC		(JIT_F_OPT << 7)
 #define JIT_F_OPT_SINK		(JIT_F_OPT << 8)
 #define JIT_F_OPT_FUSE		(JIT_F_OPT << 9)
+#define JIT_F_OPT_FMA		(JIT_F_OPT << 10)
 
 /* Optimizations names for -O. Must match the order above. */
 #define JIT_F_OPTSTRING	\
-  "\4fold\3cse\3dce\3fwd\3dse\6narrow\4loop\3abc\4sink\4fuse"
+  "\4fold\3cse\3dce\3fwd\3dse\6narrow\4loop\3abc\4sink\4fuse\3fma"
 
 /* Optimization levels set a fixed combination of flags. */
 #define JIT_F_OPT_0	0
@@ -99,6 +100,7 @@
 #define JIT_F_OPT_3	(JIT_F_OPT_2|\
   JIT_F_OPT_FWD|JIT_F_OPT_DSE|JIT_F_OPT_ABC|JIT_F_OPT_SINK|JIT_F_OPT_FUSE)
 #define JIT_F_OPT_DEFAULT	JIT_F_OPT_3
+/* Note: FMA is not set by default. */
 
 /* -- JIT engine parameters ----------------------------------------------- */
 

libs/LuaJIT/src/lj_obj.h

@@ -413,7 +413,7 @@ typedef struct GCproto {
 #define PROTO_UV_IMMUTABLE	0x4000	/* Immutable upvalue. */
 
 #define proto_kgc(pt, idx) \
-  check_exp((uintptr_t)(intptr_t)(idx) >= (uintptr_t)-(intptr_t)(pt)->sizekgc, \
+  check_exp((uintptr_t)(intptr_t)(idx) >= ~(uintptr_t)(pt)->sizekgc+1u, \
 	    gcref(mref((pt)->k, GCRef)[(idx)]))
 #define proto_knumtv(pt, idx) \
   check_exp((uintptr_t)(idx) < (pt)->sizekn, &mref((pt)->k, TValue)[(idx)])

libs/LuaJIT/src/lj_opt_fold.c

@@ -267,7 +267,7 @@ static int32_t kfold_intop(int32_t k1, int32_t k2, IROp op)
   case IR_SUB: k1 -= k2; break;
   case IR_MUL: k1 *= k2; break;
   case IR_MOD: k1 = lj_vm_modi(k1, k2); break;
-  case IR_NEG: k1 = -k1; break;
+  case IR_NEG: k1 = (int32_t)(~(uint32_t)k1+1u); break;
   case IR_BAND: k1 &= k2; break;
   case IR_BOR: k1 |= k2; break;
   case IR_BXOR: k1 ^= k2; break;
@@ -1366,7 +1366,7 @@ LJFOLDF(simplify_intsub_k)
   if (fright->i == 0)  /* i - 0 ==> i */
     return LEFTFOLD;
   fins->o = IR_ADD;  /* i - k ==> i + (-k) */
-  fins->op2 = (IRRef1)lj_ir_kint(J, -fright->i);  /* Overflow for -2^31 ok. */
+  fins->op2 = (IRRef1)lj_ir_kint(J, (int32_t)(~(uint32_t)fright->i+1u));  /* Overflow for -2^31 ok. */
   return RETRYFOLD;
 }
 
@@ -1397,7 +1397,7 @@ LJFOLDF(simplify_intsub_k64)
   if (k == 0)  /* i - 0 ==> i */
     return LEFTFOLD;
   fins->o = IR_ADD;  /* i - k ==> i + (-k) */
-  fins->op2 = (IRRef1)lj_ir_kint64(J, (uint64_t)-(int64_t)k);
+  fins->op2 = (IRRef1)lj_ir_kint64(J, ~k+1u);
   return RETRYFOLD;
 }
 

libs/LuaJIT/src/lj_parse.c

@@ -964,22 +964,22 @@ static void bcemit_unop(FuncState *fs, BCOp op, ExpDesc *e)
 #if LJ_HASFFI
       if (e->k == VKCDATA) {  /* Fold in-place since cdata is not interned. */
 	GCcdata *cd = cdataV(&e->u.nval);
-	int64_t *p = (int64_t *)cdataptr(cd);
+	uint64_t *p = (uint64_t *)cdataptr(cd);
 	if (cd->ctypeid == CTID_COMPLEX_DOUBLE)
-	  p[1] ^= (int64_t)U64x(80000000,00000000);
+	  p[1] ^= U64x(80000000,00000000);
 	else
-	  *p = -*p;
+	  *p = ~*p+1u;
 	return;
       } else
 #endif
       if (expr_isnumk(e) && !expr_numiszero(e)) {  /* Avoid folding to -0. */
 	TValue *o = expr_numtv(e);
 	if (tvisint(o)) {
-	  int32_t k = intV(o);
+	  int32_t k = intV(o), negk = (int32_t)(~(uint32_t)k+1u);
-	  if (k == -k)
+	  if (k == negk)
 	    setnumV(o, -(lua_Number)k);
 	  else
-	    setintV(o, -k);
+	    setintV(o, negk);
 	  return;
 	} else {
 	  o->u64 ^= U64x(80000000,00000000);

libs/LuaJIT/src/lj_prng.c

@@ -125,7 +125,7 @@ static PRGR libfunc_rgr;
 #if __MAC_OS_X_VERSION_MIN_REQUIRED >= 101200
 #define LJ_TARGET_HAS_GETENTROPY	1
 #endif
-#elif (LJ_TARGET_BSD && !defined(__NetBSD__)) || LJ_TARGET_SOLARIS || LJ_TARGET_CYGWIN
+#elif (LJ_TARGET_BSD && !defined(__NetBSD__)) || LJ_TARGET_SOLARIS || LJ_TARGET_CYGWIN || LJ_TARGET_QNX
 #define LJ_TARGET_HAS_GETENTROPY	1
 #endif
 

libs/LuaJIT/src/lj_profile.c

@@ -185,7 +185,11 @@ static void profile_timer_start(ProfileState *ps)
   tm.it_value.tv_sec = tm.it_interval.tv_sec = interval / 1000;
   tm.it_value.tv_usec = tm.it_interval.tv_usec = (interval % 1000) * 1000;
   setitimer(ITIMER_PROF, &tm, NULL);
+#if LJ_TARGET_QNX
+  sa.sa_flags = 0;
+#else
   sa.sa_flags = SA_RESTART;
+#endif
   sa.sa_handler = profile_signal;
   sigemptyset(&sa.sa_mask);
   sigaction(SIGPROF, &sa, &ps->oldsa);

libs/LuaJIT/src/lj_record.c

@@ -1945,6 +1945,8 @@ static void rec_varg(jit_State *J, BCReg dst, ptrdiff_t nresults)
     } else if (dst + nresults > J->maxslot) {
       J->maxslot = dst + (BCReg)nresults;
     }
+    if (J->baseslot + J->maxslot >= LJ_MAX_JSLOTS)
+      lj_trace_err(J, LJ_TRERR_STACKOV);
     for (i = 0; i < nresults; i++)
       J->base[dst+i] = i < nvararg ? getslot(J, i - nvararg - 1 - LJ_FR2) : TREF_NIL;
   } else {  /* Unknown number of varargs passed to trace. */
@@ -1964,7 +1966,7 @@ static void rec_varg(jit_State *J, BCReg dst, ptrdiff_t nresults)
 	vbase = emitir(IRT(IR_ADD, IRT_PGC), vbase, lj_ir_kint(J, frofs-8*(1+LJ_FR2)));
 	for (i = 0; i < nload; i++) {
 	  IRType t = itype2irt(&J->L->base[i-1-LJ_FR2-nvararg]);
-	  J->base[dst+i] = lj_record_vload(J, vbase, i, t);
+	  J->base[dst+i] = lj_record_vload(J, vbase, (MSize)i, t);
 	}
       } else {
 	emitir(IRTGI(IR_LE), fr, lj_ir_kint(J, frofs));
@@ -2022,8 +2024,6 @@ static void rec_varg(jit_State *J, BCReg dst, ptrdiff_t nresults)
       lj_trace_err_info(J, LJ_TRERR_NYIBC);
     }
   }
-  if (J->baseslot + J->maxslot >= LJ_MAX_JSLOTS)
-    lj_trace_err(J, LJ_TRERR_STACKOV);
 }
 
 /* -- Record allocations -------------------------------------------------- */
@@ -2572,7 +2572,8 @@ void lj_record_ins(jit_State *J)
     break;
   case BC_JLOOP:
     rec_loop_jit(J, rc, rec_loop(J, ra,
-				 !bc_isret(bc_op(traceref(J, rc)->startins))));
+				 !bc_isret(bc_op(traceref(J, rc)->startins)) &&
+				 bc_op(traceref(J, rc)->startins) != BC_ITERN));
     break;
 
   case BC_IFORL:

libs/LuaJIT/src/lj_strfmt.c

@@ -102,7 +102,7 @@ retlit:
 char * LJ_FASTCALL lj_strfmt_wint(char *p, int32_t k)
 {
   uint32_t u = (uint32_t)k;
-  if (k < 0) { u = (uint32_t)-k; *p++ = '-'; }
+  if (k < 0) { u = ~u+1u; *p++ = '-'; }
   if (u < 10000) {
     if (u < 10) goto dig1;
     if (u < 100) goto dig2;
@@ -287,7 +287,7 @@ SBuf *lj_strfmt_putfxint(SBuf *sb, SFormat sf, uint64_t k)
   /* Figure out signed prefixes. */
   if (STRFMT_TYPE(sf) == STRFMT_INT) {
     if ((int64_t)k < 0) {
-      k = (uint64_t)-(int64_t)k;
+      k = ~k+1u;
       prefix = 256 + '-';
     } else if ((sf & STRFMT_F_PLUS)) {
       prefix = 256 + '+';

libs/LuaJIT/src/lj_strscan.c

@@ -124,19 +124,19 @@ static StrScanFmt strscan_hex(const uint8_t *p, TValue *o,
   case STRSCAN_INT:
     if (!(opt & STRSCAN_OPT_TONUM) && x < 0x80000000u+neg &&
 	!(x == 0 && neg)) {
-      o->i = neg ? -(int32_t)x : (int32_t)x;
+      o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
       return STRSCAN_INT;  /* Fast path for 32 bit integers. */
     }
     if (!(opt & STRSCAN_OPT_C)) { fmt = STRSCAN_NUM; break; }
     /* fallthrough */
   case STRSCAN_U32:
     if (dig > 8) return STRSCAN_ERROR;
-    o->i = neg ? -(int32_t)x : (int32_t)x;
+    o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
     return STRSCAN_U32;
   case STRSCAN_I64:
   case STRSCAN_U64:
     if (dig > 16) return STRSCAN_ERROR;
-    o->u64 = neg ? (uint64_t)-(int64_t)x : x;
+    o->u64 = neg ? ~x+1u : x;
     return fmt;
   default:
     break;
@@ -168,12 +168,12 @@ static StrScanFmt strscan_oct(const uint8_t *p, TValue *o,
     /* fallthrough */
   case STRSCAN_U32:
     if ((x >> 32)) return STRSCAN_ERROR;
-    o->i = neg ? -(int32_t)x : (int32_t)x;
+    o->i = neg ? (int32_t)(~(uint32_t)x+1u) : (int32_t)x;
     break;
   default:
   case STRSCAN_I64:
   case STRSCAN_U64:
-    o->u64 = neg ? (uint64_t)-(int64_t)x : x;
+    o->u64 = neg ? ~x+1u : x;
     break;
   }
   return fmt;
@@ -229,18 +229,18 @@ static StrScanFmt strscan_dec(const uint8_t *p, TValue *o,
       switch (fmt) {
       case STRSCAN_INT:
 	if (!(opt & STRSCAN_OPT_TONUM) && x < 0x80000000u+neg) {
-	  o->i = neg ? -(int32_t)x : (int32_t)x;
+	  o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
 	  return STRSCAN_INT;  /* Fast path for 32 bit integers. */
 	}
 	if (!(opt & STRSCAN_OPT_C)) { fmt = STRSCAN_NUM; goto plainnumber; }
 	/* fallthrough */
       case STRSCAN_U32:
 	if ((x >> 32) != 0) return STRSCAN_ERROR;
-	o->i = neg ? -(int32_t)x : (int32_t)x;
+	o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
 	return STRSCAN_U32;
       case STRSCAN_I64:
       case STRSCAN_U64:
-	o->u64 = neg ? (uint64_t)-(int64_t)x : x;
+	o->u64 = neg ? ~x+1u : x;
 	return fmt;
       default:
       plainnumber:  /* Fast path for plain numbers < 2^63. */
@@ -348,18 +348,18 @@ static StrScanFmt strscan_bin(const uint8_t *p, TValue *o,
   switch (fmt) {
   case STRSCAN_INT:
     if (!(opt & STRSCAN_OPT_TONUM) && x < 0x80000000u+neg) {
-      o->i = neg ? -(int32_t)x : (int32_t)x;
+      o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
       return STRSCAN_INT;  /* Fast path for 32 bit integers. */
     }
     if (!(opt & STRSCAN_OPT_C)) { fmt = STRSCAN_NUM; break; }
     /* fallthrough */
   case STRSCAN_U32:
     if (dig > 32) return STRSCAN_ERROR;
-    o->i = neg ? -(int32_t)x : (int32_t)x;
+    o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
     return STRSCAN_U32;
   case STRSCAN_I64:
   case STRSCAN_U64:
-    o->u64 = neg ? (uint64_t)-(int64_t)x : x;
+    o->u64 = neg ? ~x+1u : x;
     return fmt;
   default:
     break;
@@ -468,7 +468,7 @@ StrScanFmt lj_strscan_scan(const uint8_t *p, MSize len, TValue *o,
 	if (xx >= STRSCAN_MAXEXP) return STRSCAN_ERROR;
 	p++;
       }
-      ex += negx ? -(int32_t)xx : (int32_t)xx;
+      ex += negx ? (int32_t)(~xx+1u) : (int32_t)xx;
     }
 
     /* Parse suffix. */
@@ -507,7 +507,7 @@ StrScanFmt lj_strscan_scan(const uint8_t *p, MSize len, TValue *o,
 	o->n = -0.0;
 	return STRSCAN_NUM;
       } else {
-	o->i = neg ? -(int32_t)x : (int32_t)x;
+	o->i = neg ? (int32_t)(~x+1u) : (int32_t)x;
 	return STRSCAN_INT;
       }
     }

libs/LuaJIT/src/lj_vmmath.c

@@ -36,6 +36,17 @@ LJ_FUNCA double lj_wrap_fmod(double x, double y) { return fmod(x, y); }
 
 /* -- Helper functions ---------------------------------------------------- */
 
+/* Required to prevent the C compiler from applying FMA optimizations.
+**
+** Yes, there's -ffp-contract and the FP_CONTRACT pragma ... in theory.
+** But the current state of C compilers is a mess in this regard.
+** Also, this function is not performance sensitive at all.
+*/
+LJ_NOINLINE static double lj_vm_floormul(double x, double y)
+{
+  return lj_vm_floor(x / y) * y;
+}
+
 double lj_vm_foldarith(double x, double y, int op)
 {
   switch (op) {
@@ -43,7 +54,7 @@ double lj_vm_foldarith(double x, double y, int op)
   case IR_SUB - IR_ADD: return x-y; break;
   case IR_MUL - IR_ADD: return x*y; break;
   case IR_DIV - IR_ADD: return x/y; break;
-  case IR_MOD - IR_ADD: return x-lj_vm_floor(x/y)*y; break;
+  case IR_MOD - IR_ADD: return x-lj_vm_floormul(x, y); break;
   case IR_POW - IR_ADD: return pow(x, y); break;
   case IR_NEG - IR_ADD: return -x; break;
   case IR_ABS - IR_ADD: return fabs(x); break;
@@ -64,11 +75,11 @@ int32_t LJ_FASTCALL lj_vm_modi(int32_t a, int32_t b)
   uint32_t y, ua, ub;
   /* This must be checked before using this function. */
   lj_assertX(b != 0, "modulo with zero divisor");
-  ua = a < 0 ? (uint32_t)-a : (uint32_t)a;
+  ua = a < 0 ? ~(uint32_t)a+1u : (uint32_t)a;
-  ub = b < 0 ? (uint32_t)-b : (uint32_t)b;
+  ub = b < 0 ? ~(uint32_t)b+1u : (uint32_t)b;
   y = ua % ub;
   if (y != 0 && (a^b) < 0) y = y - ub;
-  if (((int32_t)y^b) < 0) y = (uint32_t)-(int32_t)y;
+  if (((int32_t)y^b) < 0) y = ~y+1u;
   return (int32_t)y;
 }
 #endif

libs/LuaJIT/src/nxbuild.bat

@@ -121,6 +121,7 @@ goto :BUILD
 @set TARGETLIB=libluajit_%TARGETLIB_SUFFIX%.a
 :BUILD
 del %TARGETLIB%
+@set LJCOMPILE=%LJCOMPILE% -fPIC
 @if "%1" neq "noamalg" goto :AMALG
 for %%f in (lj_*.c lib_*.c) do (
   %LJCOMPILE% %%f

libs/LuaJIT/src/ps4build.bat

@@ -51,7 +51,7 @@ if exist minilua.exe.manifest^
 minilua %DASM% -LN %DASMFLAGS% -o host\buildvm_arch.h %DASC%
 @if errorlevel 1 goto :BAD
 
-%LJCOMPILE% /I "." /I %DASMDIR% %GC64% -DLUAJIT_TARGET=LUAJIT_ARCH_X64 -DLUAJIT_OS=LUAJIT_OS_OTHER -DLUAJIT_DISABLE_JIT -DLUAJIT_DISABLE_FFI -DLUAJIT_NO_UNWIND host\buildvm*.c
+%LJCOMPILE% /I "." /I %DASMDIR% %GC64% -DLUAJIT_TARGET=LUAJIT_ARCH_X64 -DLUAJIT_OS=LUAJIT_OS_OTHER -DLUAJIT_DISABLE_JIT -DLUAJIT_DISABLE_FFI -DLUAJIT_USE_SYSMALLOC -DLUAJIT_NO_UNWIND host\buildvm*.c
 @if errorlevel 1 goto :BAD
 %LJLINK% /out:buildvm.exe buildvm*.obj
 @if errorlevel 1 goto :BAD
@@ -78,7 +78,7 @@ buildvm -m folddef -o lj_folddef.h lj_opt_fold.c
 @set LJLIB="%SCE_ORBIS_SDK_DIR%\host_tools\bin\orbis-ar" rcus
 @set INCLUDE=""
 
-orbis-as -o lj_vm.o lj_vm.s
+"%SCE_ORBIS_SDK_DIR%\host_tools\bin\orbis-as" -o lj_vm.o lj_vm.s
 
 @if "%1" neq "debug" goto :NODEBUG
 @shift

libs/LuaJIT/src/ps5build.bat

@@ -74,11 +74,11 @@ buildvm -m folddef -o lj_folddef.h lj_opt_fold.c
 @if errorlevel 1 goto :BAD
 
 @rem ---- Cross compiler ----
-@set LJCOMPILE="%SCE_PROSPERO_SDK_DIR%\host_tools\bin\prospero-clang" -c -Wall -DLUAJIT_DISABLE_FFI %GC64%
+@set LJCOMPILE="%SCE_PROSPERO_SDK_DIR%\host_tools\bin\prospero-clang" -c -Wall -DLUAJIT_DISABLE_FFI -DLUAJIT_USE_SYSMALLOC %GC64%
 @set LJLIB="%SCE_PROSPERO_SDK_DIR%\host_tools\bin\prospero-llvm-ar" rcus
 @set INCLUDE=""
 
-%SCE_PROSPERO_SDK_DIR%\host_tools\bin\prospero-llvm-as -o lj_vm.o lj_vm.s
+"%SCE_PROSPERO_SDK_DIR%\host_tools\bin\prospero-clang" -c -o lj_vm.o lj_vm.s
 
 @if "%1" neq "debug" goto :NODEBUG
 @shift

libs/LuaJIT/src/vm_arm64.dasc

@@ -2636,7 +2636,9 @@ static void build_ins(BuildCtx *ctx, BCOp op, int defop)
     |.macro ins_arithmod, res, reg1, reg2
     |  fdiv d2, reg1, reg2
     |  frintm d2, d2
-    |  fmsub res, d2, reg2, reg1
+    |  // Cannot use fmsub, because FMA is not enabled by default.
+    |  fmul d2, d2, reg2
+    |  fsub res, reg1, d2
     |.endmacro
     |
     |.macro ins_arithdn, intins, fpins