inline.h

/*    inline.h
 *
 *    Copyright (C) 2012 by Larry Wall and others
 *
 *    You may distribute under the terms of either the GNU General Public
 *    License or the Artistic License, as specified in the README file.
 *
 *    This file contains tables and code adapted from
 *    https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which requires this
 *    copyright notice:

Copyright (c) 2008-2009 Bjoern Hoehrmann <bjoern@hoehrmann.de>

Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "Software"), to deal in
the Software without restriction, including without limitation the rights to
use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
of the Software, and to permit persons to whom the Software is furnished to do
so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

 *
 * This file is a home for static inline functions that cannot go in other
 * header files, because they depend on proto.h (included after most other
 * headers) or struct definitions.
 *
 * Each section names the header file that the functions "belong" to.
 */

/* ------------------------------- av.h ------------------------------- */

/*
=for apidoc_section $AV
=for apidoc av_count
Returns the number of elements in the array C<av>.  This is the true length of
the array, including any undefined elements.  It is always the same as
S<C<av_top_index(av) + 1>>.

=cut
*/
PERL_STATIC_INLINE Size_t
Perl_av_count(pTHX_ AV *av)
{
    PERL_ARGS_ASSERT_AV_COUNT;
    assert(SvTYPE(av) == SVt_PVAV);

    return AvFILL(av) + 1;
}

/* ------------------------------- av.c ------------------------------- */

/*
=for apidoc av_store_simple

This is a cut-down version of av_store that assumes that the array is
very straightforward - no magic, not readonly, and AvREAL - and that
C<key> is not negative. This function MUST NOT be used in situations
where any of those assumptions may not hold.

Stores an SV in an array.  The array index is specified as C<key>. It
can be dereferenced to get the C<SV*> that was stored there (= C<val>)).

Note that the caller is responsible for suitably incrementing the reference
count of C<val> before the call.

Approximate Perl equivalent: C<splice(@myarray, $key, 1, $val)>.

=cut
*/

PERL_STATIC_INLINE SV**
Perl_av_store_simple(pTHX_ AV *av, SSize_t key, SV *val)
{
    SV** ary;

    PERL_ARGS_ASSERT_AV_STORE_SIMPLE;
    assert(SvTYPE(av) == SVt_PVAV);
    assert(!SvMAGICAL(av));
    assert(!SvREADONLY(av));
    assert(AvREAL(av));
    assert(key > -1);

    ary = AvARRAY(av);

    if (AvFILLp(av) < key) {
        if (key > AvMAX(av)) {
            av_extend(av,key);
            ary = AvARRAY(av);
        }
        AvFILLp(av) = key;
    } else
        SvREFCNT_dec(ary[key]);

    ary[key] = val;
    return &ary[key];
}

/*
=for apidoc av_fetch_simple

This is a cut-down version of av_fetch that assumes that the array is
very straightforward - no magic, not readonly, and AvREAL - and that
C<key> is not negative. This function MUST NOT be used in situations
where any of those assumptions may not hold.

Returns the SV at the specified index in the array.  The C<key> is the
index.  If lval is true, you are guaranteed to get a real SV back (in case
it wasn't real before), which you can then modify.  Check that the return
value is non-null before dereferencing it to a C<SV*>.

The rough perl equivalent is C<$myarray[$key]>.

=cut
*/

PERL_STATIC_INLINE SV**
Perl_av_fetch_simple(pTHX_ AV *av, SSize_t key, I32 lval)
{
    PERL_ARGS_ASSERT_AV_FETCH_SIMPLE;
    assert(SvTYPE(av) == SVt_PVAV);
    assert(!SvMAGICAL(av));
    assert(!SvREADONLY(av));
    assert(AvREAL(av));
    assert(key > -1);

    if ( (key > AvFILLp(av)) || !AvARRAY(av)[key]) {
        return lval ? av_store_simple(av,key,newSV(0)) : NULL;
    } else {
        return &AvARRAY(av)[key];
    }
}

/* ------------------------------- cv.h ------------------------------- */

/*
=for apidoc_section $CV
=for apidoc CvGV
Returns the GV associated with the CV C<sv>, reifying it if necessary.

=cut
*/
PERL_STATIC_INLINE GV *
Perl_CvGV(pTHX_ CV *sv)
{
    PERL_ARGS_ASSERT_CVGV;

    return CvNAMED(sv)
        ? Perl_cvgv_from_hek(aTHX_ sv)
        : ((XPVCV*)MUTABLE_PTR(SvANY(sv)))->xcv_gv_u.xcv_gv;
}

PERL_STATIC_INLINE I32 *
Perl_CvDEPTH(const CV * const sv)
{
    PERL_ARGS_ASSERT_CVDEPTH;
    assert(SvTYPE(sv) == SVt_PVCV || SvTYPE(sv) == SVt_PVFM);

    return &((XPVCV*)SvANY(sv))->xcv_depth;
}

/*
 CvPROTO returns the prototype as stored, which is not necessarily what
 the interpreter should be using. Specifically, the interpreter assumes
 that spaces have been stripped, which has been the case if the prototype
 was added by toke.c, but is generally not the case if it was added elsewhere.
 Since we can't enforce the spacelessness at assignment time, this routine
 provides a temporary copy at parse time with spaces removed.
 I<orig> is the start of the original buffer, I<len> is the length of the
 prototype and will be updated when this returns.
 */

#ifdef PERL_CORE
PERL_STATIC_INLINE char *
S_strip_spaces(pTHX_ const char * orig, STRLEN * const len)
{
    SV * tmpsv;
    char * tmps;
    tmpsv = newSVpvn_flags(orig, *len, SVs_TEMP);
    tmps = SvPVX(tmpsv);
    while ((*len)--) {
        if (!isSPACE(*orig))
            *tmps++ = *orig;
        orig++;
    }
    *tmps = '\0';
    *len = tmps - SvPVX(tmpsv);
                return SvPVX(tmpsv);
}
#endif

/* ------------------------------- mg.h ------------------------------- */

#if defined(PERL_CORE) || defined(PERL_EXT)
/* assumes get-magic and stringification have already occurred */
PERL_STATIC_INLINE STRLEN
S_MgBYTEPOS(pTHX_ MAGIC *mg, SV *sv, const char *s, STRLEN len)
{
    assert(mg->mg_type == PERL_MAGIC_regex_global);
    assert(mg->mg_len != -1);
    if (mg->mg_flags & MGf_BYTES || !DO_UTF8(sv))
        return (STRLEN)mg->mg_len;
    else {
        const STRLEN pos = (STRLEN)mg->mg_len;
        /* Without this check, we may read past the end of the buffer: */
        if (pos > sv_or_pv_len_utf8(sv, s, len)) return len+1;
        return sv_or_pv_pos_u2b(sv, s, pos, NULL);
    }
}
#endif

/* ------------------------------- pad.h ------------------------------ */

#if defined(PERL_IN_PAD_C) || defined(PERL_IN_OP_C)
PERL_STATIC_INLINE bool
S_PadnameIN_SCOPE(const PADNAME * const pn, const U32 seq)
{
    PERL_ARGS_ASSERT_PADNAMEIN_SCOPE;

    /* is seq within the range _LOW to _HIGH ?
     * This is complicated by the fact that PL_cop_seqmax
     * may have wrapped around at some point */
    if (COP_SEQ_RANGE_LOW(pn) == PERL_PADSEQ_INTRO)
        return FALSE; /* not yet introduced */

    if (COP_SEQ_RANGE_HIGH(pn) == PERL_PADSEQ_INTRO) {
    /* in compiling scope */
        if (
            (seq >  COP_SEQ_RANGE_LOW(pn))
            ? (seq - COP_SEQ_RANGE_LOW(pn) < (U32_MAX >> 1))
            : (COP_SEQ_RANGE_LOW(pn) - seq > (U32_MAX >> 1))
        )
            return TRUE;
    }
    else if (
        (COP_SEQ_RANGE_LOW(pn) > COP_SEQ_RANGE_HIGH(pn))
        ?
            (  seq >  COP_SEQ_RANGE_LOW(pn)
            || seq <= COP_SEQ_RANGE_HIGH(pn))

        :    (  seq >  COP_SEQ_RANGE_LOW(pn)
             && seq <= COP_SEQ_RANGE_HIGH(pn))
    )
        return TRUE;
    return FALSE;
}
#endif

/* ------------------------------- pp.h ------------------------------- */

PERL_STATIC_INLINE I32
Perl_TOPMARK(pTHX)
{
    DEBUG_s(DEBUG_v(PerlIO_printf(Perl_debug_log,
                                 "MARK top  %p %" IVdf "\n",
                                  PL_markstack_ptr,
                                  (IV)*PL_markstack_ptr)));
    return *PL_markstack_ptr;
}

PERL_STATIC_INLINE I32
Perl_POPMARK(pTHX)
{
    DEBUG_s(DEBUG_v(PerlIO_printf(Perl_debug_log,
                                 "MARK pop  %p %" IVdf "\n",
                                  (PL_markstack_ptr-1),
                                  (IV)*(PL_markstack_ptr-1))));
    assert((PL_markstack_ptr > PL_markstack) || !"MARK underflow");
    return *PL_markstack_ptr--;
}

/* ----------------------------- regexp.h ----------------------------- */

PERL_STATIC_INLINE struct regexp *
Perl_ReANY(const REGEXP * const re)
{
    XPV* const p = (XPV*)SvANY(re);

    PERL_ARGS_ASSERT_REANY;
    assert(isREGEXP(re));

    return SvTYPE(re) == SVt_PVLV ? p->xpv_len_u.xpvlenu_rx
                                   : (struct regexp *)p;
}

/* ------------------------------- sv.h ------------------------------- */

PERL_STATIC_INLINE bool
Perl_SvTRUE(pTHX_ SV *sv)
{
    PERL_ARGS_ASSERT_SVTRUE;

    if (UNLIKELY(sv == NULL))
        return FALSE;
    SvGETMAGIC(sv);
    return SvTRUE_nomg_NN(sv);
}

PERL_STATIC_INLINE bool
Perl_SvTRUE_nomg(pTHX_ SV *sv)
{
    PERL_ARGS_ASSERT_SVTRUE_NOMG;

    if (UNLIKELY(sv == NULL))
        return FALSE;
    return SvTRUE_nomg_NN(sv);
}

PERL_STATIC_INLINE bool
Perl_SvTRUE_NN(pTHX_ SV *sv)
{
    PERL_ARGS_ASSERT_SVTRUE_NN;

    SvGETMAGIC(sv);
    return SvTRUE_nomg_NN(sv);
}

PERL_STATIC_INLINE bool
Perl_SvTRUE_common(pTHX_ SV * sv, const bool sv_2bool_is_fallback)
{
    PERL_ARGS_ASSERT_SVTRUE_COMMON;

    if (UNLIKELY(SvIMMORTAL_INTERP(sv)))
        return SvIMMORTAL_TRUE(sv);

    if (! SvOK(sv))
        return FALSE;

    if (SvPOK(sv))
        return SvPVXtrue(sv);

    if (SvIOK(sv))
        return SvIVX(sv) != 0; /* casts to bool */

    if (SvROK(sv) && !(SvOBJECT(SvRV(sv)) && HvAMAGIC(SvSTASH(SvRV(sv)))))
        return TRUE;

    if (sv_2bool_is_fallback)
        return sv_2bool_nomg(sv);

    return isGV_with_GP(sv);
}


PERL_STATIC_INLINE SV *
Perl_SvREFCNT_inc(SV *sv)
{
    if (LIKELY(sv != NULL))
        SvREFCNT(sv)++;
    return sv;
}
PERL_STATIC_INLINE SV *
Perl_SvREFCNT_inc_NN(SV *sv)
{
    PERL_ARGS_ASSERT_SVREFCNT_INC_NN;

    SvREFCNT(sv)++;
    return sv;
}
PERL_STATIC_INLINE void
Perl_SvREFCNT_inc_void(SV *sv)
{
    if (LIKELY(sv != NULL))
        SvREFCNT(sv)++;
}
PERL_STATIC_INLINE void
Perl_SvREFCNT_dec(pTHX_ SV *sv)
{
    if (LIKELY(sv != NULL)) {
        U32 rc = SvREFCNT(sv);
        if (LIKELY(rc > 1))
            SvREFCNT(sv) = rc - 1;
        else
            Perl_sv_free2(aTHX_ sv, rc);
    }
}

PERL_STATIC_INLINE void
Perl_SvREFCNT_dec_NN(pTHX_ SV *sv)
{
    U32 rc = SvREFCNT(sv);

    PERL_ARGS_ASSERT_SVREFCNT_DEC_NN;

    if (LIKELY(rc > 1))
        SvREFCNT(sv) = rc - 1;
    else
        Perl_sv_free2(aTHX_ sv, rc);
}

PERL_STATIC_INLINE void
Perl_SvAMAGIC_on(SV *sv)
{
    PERL_ARGS_ASSERT_SVAMAGIC_ON;
    assert(SvROK(sv));

    if (SvOBJECT(SvRV(sv))) HvAMAGIC_on(SvSTASH(SvRV(sv)));
}
PERL_STATIC_INLINE void
Perl_SvAMAGIC_off(SV *sv)
{
    PERL_ARGS_ASSERT_SVAMAGIC_OFF;

    if (SvROK(sv) && SvOBJECT(SvRV(sv)))
        HvAMAGIC_off(SvSTASH(SvRV(sv)));
}

PERL_STATIC_INLINE U32
Perl_SvPADSTALE_on(SV *sv)
{
    assert(!(SvFLAGS(sv) & SVs_PADTMP));
    return SvFLAGS(sv) |= SVs_PADSTALE;
}
PERL_STATIC_INLINE U32
Perl_SvPADSTALE_off(SV *sv)
{
    assert(!(SvFLAGS(sv) & SVs_PADTMP));
    return SvFLAGS(sv) &= ~SVs_PADSTALE;
}
#if defined(PERL_CORE) || defined (PERL_EXT)
PERL_STATIC_INLINE STRLEN
S_sv_or_pv_pos_u2b(pTHX_ SV *sv, const char *pv, STRLEN pos, STRLEN *lenp)
{
    PERL_ARGS_ASSERT_SV_OR_PV_POS_U2B;
    if (SvGAMAGIC(sv)) {
        U8 *hopped = utf8_hop((U8 *)pv, pos);
        if (lenp) *lenp = (STRLEN)(utf8_hop(hopped, *lenp) - hopped);
        return (STRLEN)(hopped - (U8 *)pv);
    }
    return sv_pos_u2b_flags(sv,pos,lenp,SV_CONST_RETURN);
}
#endif

/* ------------------------------- utf8.h ------------------------------- */

/*
=for apidoc_section $unicode
*/

PERL_STATIC_INLINE void
Perl_append_utf8_from_native_byte(const U8 byte, U8** dest)
{
    /* Takes an input 'byte' (Latin1 or EBCDIC) and appends it to the UTF-8
     * encoded string at '*dest', updating '*dest' to include it */

    PERL_ARGS_ASSERT_APPEND_UTF8_FROM_NATIVE_BYTE;

    if (NATIVE_BYTE_IS_INVARIANT(byte))
        *((*dest)++) = byte;
    else {
        *((*dest)++) = UTF8_EIGHT_BIT_HI(byte);
        *((*dest)++) = UTF8_EIGHT_BIT_LO(byte);
    }
}

/*
=for apidoc is_utf8_invariant_string

Returns TRUE if the first C<len> bytes of the string C<s> are the same
regardless of the UTF-8 encoding of the string (or UTF-EBCDIC encoding on
EBCDIC machines); otherwise it returns FALSE.  That is, it returns TRUE if they
are UTF-8 invariant.  On ASCII-ish machines, all the ASCII characters and only
the ASCII characters fit this definition.  On EBCDIC machines, the ASCII-range
characters are invariant, but so also are the C1 controls.

If C<len> is 0, it will be calculated using C<strlen(s)>, (which means if you
use this option, that C<s> can't have embedded C<NUL> characters and has to
have a terminating C<NUL> byte).

See also
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.

=cut

*/

#define is_utf8_invariant_string(s, len)                                    \
                                is_utf8_invariant_string_loc(s, len, NULL)

/*
=for apidoc is_utf8_invariant_string_loc

Like C<L</is_utf8_invariant_string>> but upon failure, stores the location of
the first UTF-8 variant character in the C<ep> pointer; if all characters are
UTF-8 invariant, this function does not change the contents of C<*ep>.

=cut

*/

PERL_STATIC_INLINE bool
Perl_is_utf8_invariant_string_loc(const U8* const s, STRLEN len, const U8 ** ep)
{
    const U8* send;
    const U8* x = s;

    PERL_ARGS_ASSERT_IS_UTF8_INVARIANT_STRING_LOC;

    if (len == 0) {
        len = strlen((const char *)s);
    }

    send = s + len;

/* This looks like 0x010101... */
#  define PERL_COUNT_MULTIPLIER   (~ (UINTMAX_C(0)) / 0xFF)

/* This looks like 0x808080... */
#  define PERL_VARIANTS_WORD_MASK (PERL_COUNT_MULTIPLIER * 0x80)
#  define PERL_WORDSIZE            sizeof(PERL_UINTMAX_T)
#  define PERL_WORD_BOUNDARY_MASK (PERL_WORDSIZE - 1)

/* Evaluates to 0 if 'x' is at a word boundary; otherwise evaluates to 1, by
 * or'ing together the lowest bits of 'x'.  Hopefully the final term gets
 * optimized out completely on a 32-bit system, and its mask gets optimized out
 * on a 64-bit system */
#  define PERL_IS_SUBWORD_ADDR(x) (1 & (       PTR2nat(x)                     \
                                      |   (  PTR2nat(x) >> 1)                 \
                                      | ( ( (PTR2nat(x)                       \
                                           & PERL_WORD_BOUNDARY_MASK) >> 2))))

#ifndef EBCDIC

    /* Do the word-at-a-time iff there is at least one usable full word.  That
     * means that after advancing to a word boundary, there still is at least a
     * full word left.  The number of bytes needed to advance is 'wordsize -
     * offset' unless offset is 0. */
    if ((STRLEN) (send - x) >= PERL_WORDSIZE

                            /* This term is wordsize if subword; 0 if not */
                          + PERL_WORDSIZE * PERL_IS_SUBWORD_ADDR(x)

                            /* 'offset' */
                          - (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK))
    {

        /* Process per-byte until reach word boundary.  XXX This loop could be
         * eliminated if we knew that this platform had fast unaligned reads */
        while (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK) {
            if (! UTF8_IS_INVARIANT(*x)) {
                if (ep) {
                    *ep = x;
                }

                return FALSE;
            }
            x++;
        }

        /* Here, we know we have at least one full word to process.  Process
         * per-word as long as we have at least a full word left */
        do {
            if ((* (PERL_UINTMAX_T *) x) & PERL_VARIANTS_WORD_MASK)  {

                /* Found a variant.  Just return if caller doesn't want its
                 * exact position */
                if (! ep) {
                    return FALSE;
                }

#  if   BYTEORDER == 0x1234 || BYTEORDER == 0x12345678    \
     || BYTEORDER == 0x4321 || BYTEORDER == 0x87654321

                *ep = x + variant_byte_number(* (PERL_UINTMAX_T *) x);
                assert(*ep >= s && *ep < send);

                return FALSE;

#  else   /* If weird byte order, drop into next loop to do byte-at-a-time
           checks. */

                break;
#  endif
            }

            x += PERL_WORDSIZE;

        } while (x + PERL_WORDSIZE <= send);
    }

#endif      /* End of ! EBCDIC */

    /* Process per-byte */
    while (x < send) {
        if (! UTF8_IS_INVARIANT(*x)) {
            if (ep) {
                *ep = x;
            }

            return FALSE;
        }

        x++;
    }

    return TRUE;
}

/* See if the platform has builtins for finding the most/least significant bit,
 * and which one is right for using on 32 and 64 bit operands */
#if (__has_builtin(__builtin_clz) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == INTSIZE
#    define PERL_CLZ_32 __builtin_clz
#  endif
#  if defined(U64TYPE) && U64SIZE == INTSIZE
#    define PERL_CLZ_64 __builtin_clz
#  endif
#endif
#if (__has_builtin(__builtin_ctz) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == INTSIZE
#    define PERL_CTZ_32 __builtin_ctz
#  endif
#  if defined(U64TYPE) && U64SIZE == INTSIZE
#    define PERL_CTZ_64 __builtin_ctz
#  endif
#endif

#if (__has_builtin(__builtin_clzl) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == LONGSIZE && ! defined(PERL_CLZ_32)
#    define PERL_CLZ_32 __builtin_clzl
#  endif
#  if defined(U64TYPE) && U64SIZE == LONGSIZE && ! defined(PERL_CLZ_64)
#    define PERL_CLZ_64 __builtin_clzl
#  endif
#endif
#if (__has_builtin(__builtin_ctzl) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == LONGSIZE && ! defined(PERL_CTZ_32)
#    define PERL_CTZ_32 __builtin_ctzl
#  endif
#  if defined(U64TYPE) && U64SIZE == LONGSIZE && ! defined(PERL_CTZ_64)
#    define PERL_CTZ_64 __builtin_ctzl
#  endif
#endif

#if (__has_builtin(__builtin_clzll) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == LONGLONGSIZE && ! defined(PERL_CLZ_32)
#    define PERL_CLZ_32 __builtin_clzll
#  endif
#  if defined(U64TYPE) && U64SIZE == LONGLONGSIZE && ! defined(PERL_CLZ_64)
#    define PERL_CLZ_64 __builtin_clzll
#  endif
#endif
#if (__has_builtin(__builtin_ctzll) || PERL_GCC_VERSION_GE(3,4,0))
#  if U32SIZE == LONGLONGSIZE && ! defined(PERL_CTZ_32)
#    define PERL_CTZ_32 __builtin_ctzll
#  endif
#  if defined(U64TYPE) && U64SIZE == LONGLONGSIZE && ! defined(PERL_CTZ_64)
#    define PERL_CTZ_64 __builtin_ctzll
#  endif
#endif

#if defined(_MSC_VER) && _MSC_VER >= 1400
#  include <intrin.h>
#  pragma intrinsic(_BitScanForward)
#  pragma intrinsic(_BitScanReverse)
#  ifdef _WIN64
#    pragma intrinsic(_BitScanForward64)
#    pragma intrinsic(_BitScanReverse64)
#  endif
#endif

/* The reason there are not checks to see if ffs() and ffsl() are available for
 * determining the lsb, is because these don't improve on the deBruijn method
 * fallback, which is just a branchless integer multiply, array element
 * retrieval, and shift.  The others, even if the function call overhead is
 * optimized out, have to cope with the possibility of the input being all
 * zeroes, and almost certainly will have conditionals for this eventuality.
 * khw, at the time of this commit, looked at the source for both gcc and clang
 * to verify this.  (gcc used a method inferior to deBruijn.) */

/* Below are functions to find the first, last, or only set bit in a word.  On
 * platforms with 64-bit capability, there is a pair for each operation; the
 * first taking a 64 bit operand, and the second a 32 bit one.  The logic is
 * the same in each pair, so the second is stripped of most comments. */

#ifdef U64TYPE  /* HAS_QUAD not usable outside the core */

PERL_STATIC_INLINE unsigned
Perl_lsbit_pos64(U64 word)
{
    /* Find the position (0..63) of the least significant set bit in the input
     * word */

    ASSUME(word != 0);

    /* If we can determine that the platform has a usable fast method to get
     * this info, use that */

#  if defined(PERL_CTZ_64)
#    define PERL_HAS_FAST_GET_LSB_POS64

    return (unsigned) PERL_CTZ_64(word);

#  elif U64SIZE == 8 && defined(_MSC_VER) && _MSC_VER >= 1400
#    define PERL_HAS_FAST_GET_LSB_POS64

    {
        unsigned long index;
        _BitScanForward64(&index, word);
        return (unsigned)index;
    }

#  else

    /* Here, we didn't find a fast method for finding the lsb.  Fall back to
     * making the lsb the only set bit in the word, and use our function that
     * works on words with a single bit set.
     *
     * Isolate the lsb;
     * https://stackoverflow.com/questions/757059/position-of-least-significant-bit-that-is-set
     *
     * The word will look like this, with a rightmost set bit in position 's':
     * ('x's are don't cares, and 'y's are their complements)
     *      s
     *  x..x100..00
     *  y..y011..11      Complement
     *  y..y100..00      Add 1
     *  0..0100..00      And with the original
     *
     *  (Yes, complementing and adding 1 is just taking the negative on 2's
     *  complement machines, but not on 1's complement ones, and some compilers
     *  complain about negating an unsigned.)
     */
    return single_1bit_pos64(word & (~word + 1));

#  endif

}

#  define lsbit_pos_uintmax_(word) lsbit_pos64(word)
#else   /* ! QUAD */
#  define lsbit_pos_uintmax_(word) lsbit_pos32(word)
#endif

PERL_STATIC_INLINE unsigned     /* Like above for 32 bit word */
Perl_lsbit_pos32(U32 word)
{
    /* Find the position (0..31) of the least significant set bit in the input
     * word */

    ASSUME(word != 0);

#if defined(PERL_CTZ_32)
#  define PERL_HAS_FAST_GET_LSB_POS32

    return (unsigned) PERL_CTZ_32(word);

#elif U32SIZE == 4 && defined(_MSC_VER) && _MSC_VER >= 1400
#  define PERL_HAS_FAST_GET_LSB_POS32

    {
        unsigned long index;
        _BitScanForward(&index, word);
        return (unsigned)index;
    }

#else

    return single_1bit_pos32(word & (~word + 1));

#endif

}


/* Convert the leading zeros count to the bit position of the first set bit.
 * This just subtracts from the highest position, 31 or 63.  But some compilers
 * don't optimize this optimally, and so a bit of bit twiddling encourages them
 * to do the right thing.  It turns out that subtracting a smaller non-negative
 * number 'x' from 2**n-1 for any n is the same as taking the exclusive-or of
 * the two numbers.  To see why, first note that the sum of any number, x, and
 * its complement, x', is all ones.  So all ones minus x is x'.  Then note that
 * the xor of x and all ones is x'. */
#define LZC_TO_MSBIT_POS_(size, lzc)  ((size##SIZE * CHARBITS - 1) ^ (lzc))

#ifdef U64TYPE  /* HAS_QUAD not usable outside the core */

PERL_STATIC_INLINE unsigned
Perl_msbit_pos64(U64 word)
{
    /* Find the position (0..63) of the most significant set bit in the input
     * word */

    ASSUME(word != 0);

    /* If we can determine that the platform has a usable fast method to get
     * this, use that */

#  if defined(PERL_CLZ_64)
#    define PERL_HAS_FAST_GET_MSB_POS64

    return (unsigned) LZC_TO_MSBIT_POS_(U64, PERL_CLZ_64(word));

#  elif U64SIZE == 8 && defined(_WIN64) && defined(_MSC_VER) && _MSC_VER >= 1400
#    define PERL_HAS_FAST_GET_MSB_POS64

    {
        unsigned long index;
        _BitScanReverse64(&index, word);
        return (unsigned)index;
    }

#  else

    /* Here, we didn't find a fast method for finding the msb.  Fall back to
     * making the msb the only set bit in the word, and use our function that
     * works on words with a single bit set.
     *
     * Isolate the msb; http://codeforces.com/blog/entry/10330
     *
     * Only the most significant set bit matters.  Or'ing word with its right
     * shift of 1 makes that bit and the next one to its right both 1.
     * Repeating that with the right shift of 2 makes for 4 1-bits in a row.
     * ...  We end with the msb and all to the right being 1. */
    word |= (word >>  1);
    word |= (word >>  2);
    word |= (word >>  4);
    word |= (word >>  8);
    word |= (word >> 16);
    word |= (word >> 32);

    /* Then subtracting the right shift by 1 clears all but the left-most of
     * the 1 bits, which is our desired result */
    word -= (word >> 1);

    /* Now we have a single bit set */
    return single_1bit_pos64(word);

#  endif

}

#  define msbit_pos_uintmax_(word) msbit_pos64(word)
#else   /* ! QUAD */
#  define msbit_pos_uintmax_(word) msbit_pos32(word)
#endif

PERL_STATIC_INLINE unsigned
Perl_msbit_pos32(U32 word)
{
    /* Find the position (0..31) of the most significant set bit in the input
     * word */

    ASSUME(word != 0);

#if defined(PERL_CLZ_32)
#  define PERL_HAS_FAST_GET_MSB_POS32

    return (unsigned) LZC_TO_MSBIT_POS_(U32, PERL_CLZ_32(word));

#elif U32SIZE == 4 && defined(_MSC_VER) && _MSC_VER >= 1400
#  define PERL_HAS_FAST_GET_MSB_POS32

    {
        unsigned long index;
        _BitScanReverse(&index, word);
        return (unsigned)index;
    }

#else

    word |= (word >>  1);
    word |= (word >>  2);
    word |= (word >>  4);
    word |= (word >>  8);
    word |= (word >> 16);
    word -= (word >> 1);
    return single_1bit_pos32(word);

#endif

}

#if UVSIZE == U64SIZE
#  define msbit_pos(word)  msbit_pos64(word)
#  define lsbit_pos(word)  lsbit_pos64(word)
#elif UVSIZE == U32SIZE
#  define msbit_pos(word)  msbit_pos32(word)
#  define lsbit_pos(word)  lsbit_pos32(word)
#endif

#ifdef U64TYPE  /* HAS_QUAD not usable outside the core */

PERL_STATIC_INLINE unsigned
Perl_single_1bit_pos64(U64 word)
{
    /* Given a 64-bit word known to contain all zero bits except one 1 bit,
     * find and return the 1's position: 0..63 */

#  ifdef PERL_CORE    /* macro not exported */
    ASSUME(isPOWER_OF_2(word));
#  else
    ASSUME(word && (word & (word-1)) == 0);
#  endif

    /* The only set bit is both the most and least significant bit.  If we have
     * a fast way of finding either one, use that.
     *
     * It may appear at first glance that those functions call this one, but
     * they don't if the corresponding #define is set */

#  ifdef PERL_HAS_FAST_GET_MSB_POS64

    return msbit_pos64(word);

#  elif defined(PERL_HAS_FAST_GET_LSB_POS64)

    return lsbit_pos64(word);

#  else

    /* The position of the only set bit in a word can be quickly calculated
     * using deBruijn sequences.  See for example
     * https://en.wikipedia.org/wiki/De_Bruijn_sequence */
    return PL_deBruijn_bitpos_tab64[(word * PERL_deBruijnMagic64_)
                                                    >> PERL_deBruijnShift64_];
#  endif

}

#endif

PERL_STATIC_INLINE unsigned
Perl_single_1bit_pos32(U32 word)
{
    /* Given a 32-bit word known to contain all zero bits except one 1 bit,
     * find and return the 1's position: 0..31 */

#ifdef PERL_CORE    /* macro not exported */
    ASSUME(isPOWER_OF_2(word));
#else
    ASSUME(word && (word & (word-1)) == 0);
#endif
#ifdef PERL_HAS_FAST_GET_MSB_POS32

    return msbit_pos32(word);

#elif defined(PERL_HAS_FAST_GET_LSB_POS32)

    return lsbit_pos32(word);

/* Unlikely, but possible for the platform to have a wider fast operation but
 * not a narrower one.  But easy enough to handle the case by widening the
 * parameter size.  (Going the other way, emulating 64 bit by two 32 bit ops
 * would be slower than the deBruijn method.) */
#elif defined(PERL_HAS_FAST_GET_MSB_POS64)

    return msbit_pos64(word);

#elif defined(PERL_HAS_FAST_GET_LSB_POS64)

    return lsbit_pos64(word);

#else

    return PL_deBruijn_bitpos_tab32[(word * PERL_deBruijnMagic32_)
                                                    >> PERL_deBruijnShift32_];
#endif

}

/*
=for apidoc valid_utf8_to_uvchr
Like C<L<perlapi/utf8_to_uvchr_buf>>, but should only be called when it is
known that the next character in the input UTF-8 string C<s> is well-formed
(I<e.g.>, it passes C<L<perlapi/isUTF8_CHAR>>.  Surrogates, non-character code
points, and non-Unicode code points are allowed.

=cut

 */

PERL_STATIC_INLINE UV
Perl_valid_utf8_to_uvchr(const U8 *s, STRLEN *retlen)
{

#ifndef EBCDIC
#  ifdef PERL_HAS_FAST_GET_MSB_POS32
#    define PERL_USE_MSB_FOR_VALID_UTF8_ msbit_pos32
#  elif defined(PERL_HAS_FAST_GET_MSB_POS64)
#    define PERL_USE_MSB_FOR_VALID_UTF8_ msbit_pos64
#  endif
#endif
#ifdef PERL_USE_MSB_FOR_VALID_UTF8_

    /* Given that *s is known to be a legal start byte, the following returns
     * its UTF8SKIP, avoiding an array lookup.  This only makes sense to do if
     * we know that the platform does clz effectively with a single machine
     * instruction; otherwise the lookup is cheaper.
     *
     * The code takes the byte, left shifts it by one, discarding the new upper
     * bit, retaining the original one.  The position of its complement's first
     * set bit, subtracted from the total bits, yields UTF8SKIP.  The shift
     * accomplishes two things.  It eliminates the discontinuity in a single
     * leading 1 is illegal.  That is 0 leading 1's mean length 1 vs 2 leading
     * 1's mean length 2.  And it keeps the input to the clz instruction from
     * ever being all 0's, which would yield undefined behavior.
     *
     * To illustrate, where 'x' is a don't care:
     *  0xxxxxxx
     *  0xxxxxx0    << 1, retaining original top bit
     *  1xxxxxx1    complement
     *  7 => 1      msb of complement => subtracted from 8
     * -------------------
     *  11110xxx
     *  1110xxx0    << 1, retaining original top bit
     *  0001xxx1    complement
     *  4 => 4      msb => subtracted from 8
     * -------------------
     *  11111111
     *  11111110    << 1, retaining original top bit
     *  00000001    complement
     *  0 => 8      msb => subtracted from 8
     */
     PERL_UINT_FAST8_T expectlen
         = CHARBITS
         - PERL_USE_MSB_FOR_VALID_UTF8_(
                                    /* Left shift 1, discarding new top bit */
                          (U8) ~(  ((*s << 1) & nBIT_MASK(CHARBITS - 1))
                                    /* Add back the original top bit */
                                  | (*s & (1 << (CHARBITS - 1)))));
#else
     PERL_UINT_FAST8_T expectlen = UTF8SKIP(s);
#endif

    /* Remove the leading bits that indicate the number of bytes, leaving just
     * the bits that are part of the value */
    UV uv = NATIVE_UTF8_TO_I8(*s) & UTF_START_MASK(expectlen);

    PERL_ARGS_ASSERT_VALID_UTF8_TO_UVCHR;
    assert(! UTF8_IS_CONTINUATION(*s));

    switch (expectlen) {
        PERL_UINT_FAST8_T countdown;

      case 8:   /* Start byte FF is special */
        expectlen = UTF8_MAXBYTES;
        /*FALLTHROUGH*/

      default: /* Don't unroll the loop for rarely encountered high code points
                */
        countdown = expectlen;
        while (countdown-- > 4) {
            s++;
            uv = UTF8_ACCUMULATE(uv, *s);
        }
        /*FALLTHROUGH*/

      case 4:
        s++;
        uv = UTF8_ACCUMULATE(uv, *s);
        /*FALLTHROUGH*/

      case 3:
        s++;
        uv = UTF8_ACCUMULATE(uv, *s);
        /*FALLTHROUGH*/

      case 2:
        s++;
        uv = UTF8_ACCUMULATE(uv, *s);

#ifndef EBCDIC

        /*FALLTHROUGH*/

        /* On ASCII platforms, UTF_START_MASK() works on all UTF-8 invariants,
         * so the call to it before this switch() already set up the length 1
         * case, so no further action should be done. */
      case 1:
        break;

#else
        break;  /* EBCDIC: For case 2: */

        /* There are length 1 characters on EBCDIC platforms for which
         * UTF_START_MASK() doesn't work properly.  Instead we have to handle
         * that case specially.
         *
         * An invariant is trivially returned; throw away our earlier
         * calculation */
      case 1:
        if (retlen) {
            *retlen = 1;
        }
        return *(s - 1);

#endif

    }

    if (retlen) {
        *retlen = expectlen;
    }

    return UNI_TO_NATIVE(uv);
}

#ifndef EBCDIC

PERL_STATIC_INLINE unsigned int
Perl_variant_byte_number(PERL_UINTMAX_T word)
{
    /* This returns the position in a word (0..7) of the first variant byte in
     * it.  This is a helper function.  Note that there are no branches */

    /* Get just the msb bits of each byte */
    word &= PERL_VARIANTS_WORD_MASK;

    /* This should only be called if we know there is a variant byte in the
     * word */
    assert(word);

#  if BYTEORDER == 0x1234 || BYTEORDER == 0x12345678

    /* Bytes are stored like
     *  Byte8 ... Byte2 Byte1
     *  63..56...15...8 7...0
     * so getting the lsb of the whole modified word is getting the msb of the
     * first byte that has its msb set */
    word = lsbit_pos_uintmax_(word);

    /* Here, word contains the position 7,15,23,...55,63 of that bit.  Convert
     * to 0..7 */
    return (unsigned int) ((word + 1) >> 3) - 1;

#  elif BYTEORDER == 0x4321 || BYTEORDER == 0x87654321

    /* Bytes are stored like
     *  Byte1 Byte2  ... Byte8
     * 63..56 55..47 ... 7...0
     * so getting the msb of the whole modified word is getting the msb of the
     * first byte that has its msb set */
    word = msbit_pos_uintmax_(word);

    /* Here, word contains the position 63,55,...,23,15,7 of that bit.  Convert
     * to 0..7 */
    word = ((word + 1) >> 3) - 1;

    /* And invert the result because of the reversed byte order on this
     * platform */
    word = CHARBITS - word - 1;

    return (unsigned int) word;

#  else
#    error Unexpected byte order
#  endif

}

#endif
#if defined(PERL_CORE) || defined(PERL_EXT)

/*
=for apidoc variant_under_utf8_count

This function looks at the sequence of bytes between C<s> and C<e>, which are
assumed to be encoded in ASCII/Latin1, and returns how many of them would
change should the string be translated into UTF-8.  Due to the nature of UTF-8,
each of these would occupy two bytes instead of the single one in the input
string.  Thus, this function returns the precise number of bytes the string
would expand by when translated to UTF-8.

Unlike most of the other functions that have C<utf8> in their name, the input
to this function is NOT a UTF-8-encoded string.  The function name is slightly
I<odd> to emphasize this.

This function is internal to Perl because khw thinks that any XS code that
would want this is probably operating too close to the internals.  Presenting a
valid use case could change that.

See also
C<L<perlapi/is_utf8_invariant_string>>
and
C<L<perlapi/is_utf8_invariant_string_loc>>,

=cut

*/

PERL_STATIC_INLINE Size_t
S_variant_under_utf8_count(const U8* const s, const U8* const e)
{
    const U8* x = s;
    Size_t count = 0;

    PERL_ARGS_ASSERT_VARIANT_UNDER_UTF8_COUNT;

#  ifndef EBCDIC

    /* Test if the string is long enough to use word-at-a-time.  (Logic is the
     * same as for is_utf8_invariant_string()) */
    if ((STRLEN) (e - x) >= PERL_WORDSIZE
                          + PERL_WORDSIZE * PERL_IS_SUBWORD_ADDR(x)
                          - (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK))
    {

        /* Process per-byte until reach word boundary.  XXX This loop could be
         * eliminated if we knew that this platform had fast unaligned reads */
        while (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK) {
            count += ! UTF8_IS_INVARIANT(*x++);
        }

        /* Process per-word as long as we have at least a full word left */
        do {    /* Commit 03c1e4ab1d6ee9062fb3f94b0ba31db6698724b1 contains an
                   explanation of how this works */
            PERL_UINTMAX_T increment
                = ((((* (PERL_UINTMAX_T *) x) & PERL_VARIANTS_WORD_MASK) >> 7)
                      * PERL_COUNT_MULTIPLIER)
                    >> ((PERL_WORDSIZE - 1) * CHARBITS);
            count += (Size_t) increment;
            x += PERL_WORDSIZE;
        } while (x + PERL_WORDSIZE <= e);
    }

#  endif

    /* Process per-byte */
    while (x < e) {
        if (! UTF8_IS_INVARIANT(*x)) {
            count++;
        }

        x++;
    }

    return count;
}

#endif

#ifndef PERL_IN_REGEXEC_C   /* Keep  these around for that file */
#  undef PERL_WORDSIZE
#  undef PERL_COUNT_MULTIPLIER
#  undef PERL_WORD_BOUNDARY_MASK
#  undef PERL_VARIANTS_WORD_MASK
#endif

/*
=for apidoc is_utf8_string

Returns TRUE if the first C<len> bytes of string C<s> form a valid
Perl-extended-UTF-8 string; returns FALSE otherwise.  If C<len> is 0, it will
be calculated using C<strlen(s)> (which means if you use this option, that C<s>
can't have embedded C<NUL> characters and has to have a terminating C<NUL>
byte).  Note that all characters being ASCII constitute 'a valid UTF-8 string'.

This function considers Perl's extended UTF-8 to be valid.  That means that
code points above Unicode, surrogates, and non-character code points are
considered valid by this function.  Use C<L</is_strict_utf8_string>>,
C<L</is_c9strict_utf8_string>>, or C<L</is_utf8_string_flags>> to restrict what
code points are considered valid.

See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,

=cut
*/

#define is_utf8_string(s, len)  is_utf8_string_loclen(s, len, NULL, NULL)

#if defined(PERL_CORE) || defined (PERL_EXT)

/*
=for apidoc is_utf8_non_invariant_string

Returns TRUE if L<perlapi/is_utf8_invariant_string> returns FALSE for the first
C<len> bytes of the string C<s>, but they are, nonetheless, legal Perl-extended
UTF-8; otherwise returns FALSE.

A TRUE return means that at least one code point represented by the sequence
either is a wide character not representable as a single byte, or the
representation differs depending on whether the sequence is encoded in UTF-8 or
not.

See also
C<L<perlapi/is_utf8_invariant_string>>,
C<L<perlapi/is_utf8_string>>

=cut

This is commonly used to determine if a SV's UTF-8 flag should be turned on.
It generally needn't be if its string is entirely UTF-8 invariant, and it
shouldn't be if it otherwise contains invalid UTF-8.

It is an internal function because khw thinks that XS code shouldn't be working
at this low a level.  A valid use case could change that.

*/

PERL_STATIC_INLINE bool
Perl_is_utf8_non_invariant_string(const U8* const s, STRLEN len)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_UTF8_NON_INVARIANT_STRING;

    if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
        return FALSE;
    }

    return is_utf8_string(first_variant, len - (first_variant - s));
}

#endif

/*
=for apidoc is_strict_utf8_string

Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8-encoded string that is fully interchangeable by any application using
Unicode rules; otherwise it returns FALSE.  If C<len> is 0, it will be
calculated using C<strlen(s)> (which means if you use this option, that C<s>
can't have embedded C<NUL> characters and has to have a terminating C<NUL>
byte).  Note that all characters being ASCII constitute 'a valid UTF-8 string'.

This function returns FALSE for strings containing any
code points above the Unicode max of 0x10FFFF, surrogate code points, or
non-character code points.

See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.

=cut
*/

#define is_strict_utf8_string(s, len)  is_strict_utf8_string_loclen(s, len, NULL, NULL)

/*
=for apidoc is_c9strict_utf8_string

Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8-encoded string that conforms to
L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>;
otherwise it returns FALSE.  If C<len> is 0, it will be calculated using
C<strlen(s)> (which means if you use this option, that C<s> can't have embedded
C<NUL> characters and has to have a terminating C<NUL> byte).  Note that all
characters being ASCII constitute 'a valid UTF-8 string'.

This function returns FALSE for strings containing any code points above the
Unicode max of 0x10FFFF or surrogate code points, but accepts non-character
code points per
L<Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.

See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.

=cut
*/

#define is_c9strict_utf8_string(s, len)  is_c9strict_utf8_string_loclen(s, len, NULL, 0)

/*
=for apidoc is_utf8_string_flags

Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8 string, subject to the restrictions imposed by C<flags>;
returns FALSE otherwise.  If C<len> is 0, it will be calculated
using C<strlen(s)> (which means if you use this option, that C<s> can't have
embedded C<NUL> characters and has to have a terminating C<NUL> byte).  Note
that all characters being ASCII constitute 'a valid UTF-8 string'.

If C<flags> is 0, this gives the same results as C<L</is_utf8_string>>; if
C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
as C<L</is_strict_utf8_string>>; and if C<flags> is
C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives the same results as
C<L</is_c9strict_utf8_string>>.  Otherwise C<flags> may be any
combination of the C<UTF8_DISALLOW_I<foo>> flags understood by
C<L</utf8n_to_uvchr>>, with the same meanings.

See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_utf8_string_flags(const U8 *s, STRLEN len, const U32 flags)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_UTF8_STRING_FLAGS;
    assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
                          |UTF8_DISALLOW_PERL_EXTENDED)));

    if (len == 0) {
        len = strlen((const char *)s);
    }

    if (flags == 0) {
        return is_utf8_string(s, len);
    }

    if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
                                        == UTF8_DISALLOW_ILLEGAL_INTERCHANGE)
    {
        return is_strict_utf8_string(s, len);
    }

    if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
                                       == UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE)
    {
        return is_c9strict_utf8_string(s, len);
    }

    if (! is_utf8_invariant_string_loc(s, len, &first_variant)) {
        const U8* const send = s + len;
        const U8* x = first_variant;

        while (x < send) {
            STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
            if (UNLIKELY(! cur_len)) {
                return FALSE;
            }
            x += cur_len;
        }
    }

    return TRUE;
}

/*

=for apidoc is_utf8_string_loc

Like C<L</is_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.

See also C<L</is_utf8_string_loclen>>.

=cut
*/

#define is_utf8_string_loc(s, len, ep)  is_utf8_string_loclen(s, len, ep, 0)

/*

=for apidoc is_utf8_string_loclen

Like C<L</is_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.

See also C<L</is_utf8_string_loc>>.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN;

    if (len == 0) {
        len = strlen((const char *) s);
    }

    if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
        if (el)
            *el = len;

        if (ep) {
            *ep = s + len;
        }

        return TRUE;
    }

    {
        const U8* const send = s + len;
        const U8* x = first_variant;
        STRLEN outlen = first_variant - s;

        while (x < send) {
            const STRLEN cur_len = isUTF8_CHAR(x, send);
            if (UNLIKELY(! cur_len)) {
                break;
            }
            x += cur_len;
            outlen++;
        }

        if (el)
            *el = outlen;

        if (ep) {
            *ep = x;
        }

        return (x == send);
    }
}

/* The perl core arranges to never call the DFA below without there being at
 * least one byte available to look at.  This allows the DFA to use a do {}
 * while loop which means that calling it with a UTF-8 invariant has a single
 * conditional, same as the calling code checking for invariance ahead of time.
 * And having the calling code remove that conditional speeds up by that
 * conditional, the case where it wasn't invariant.  So there's no reason to
 * check before caling this.
 *
 * But we don't know this for non-core calls, so have to retain the check for
 * them. */
#ifdef PERL_CORE
#  define PERL_NON_CORE_CHECK_EMPTY(s,e)  assert((e) > (s))
#else
#  define PERL_NON_CORE_CHECK_EMPTY(s,e)  if ((e) <= (s)) return FALSE
#endif

/*
 * DFA for checking input is valid UTF-8 syntax.
 *
 * This uses adaptations of the table and algorithm given in
 * https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
 * documentation of the original version.  A copyright notice for the original
 * version is given at the beginning of this file.  The Perl adapations are
 * documented at the definition of PL_extended_utf8_dfa_tab[].
 *
 * This dfa is fast.  There are three exit conditions:
 *  1) a well-formed code point, acceptable to the table
 *  2) the beginning bytes of an incomplete character, whose completion might
 *     or might not be acceptable
 *  3) unacceptable to the table.  Some of the adaptations have certain,
 *     hopefully less likely to occur, legal inputs be unacceptable to the
 *     table, so these must be sorted out afterwards.
 *
 * This macro is a complete implementation of the code executing the DFA.  It
 * is passed the input sequence bounds and the table to use, and what to do
 * for each of the exit conditions.  There are three canned actions, likely to
 * be the ones you want:
 *      DFA_RETURN_SUCCESS_
 *      DFA_RETURN_FAILURE_
 *      DFA_GOTO_TEASE_APART_FF_
 *
 * You pass a parameter giving the action to take for each of the three
 * possible exit conditions:
 *
 * 'accept_action'  This is executed when the DFA accepts the input.
 *                  DFA_RETURN_SUCCESS_ is the most likely candidate.
 * 'reject_action'  This is executed when the DFA rejects the input.
 *                  DFA_RETURN_FAILURE_ is a candidate, or 'goto label' where
 *                  you have written code to distinguish the rejecting state
 *                  results.  Because it happens in several places, and
 *                  involves #ifdefs, the special action
 *                  DFA_GOTO_TEASE_APART_FF_ is what you want with
 *                  PL_extended_utf8_dfa_tab.  On platforms without
 *                  EXTRA_LONG_UTF8, there is no need to tease anything apart,
 *                  so this evaluates to DFA_RETURN_FAILURE_; otherwise you
 *                  need to have a label 'tease_apart_FF' that it will transfer
 *                  to.
 * 'incomplete_char_action'  This is executed when the DFA ran off the end
 *                  before accepting or rejecting the input.
 *                  DFA_RETURN_FAILURE_ is the likely action, but you could
 *                  have a 'goto', or NOOP.  In the latter case the DFA drops
 *                  off the end, and you place your code to handle this case
 *                  immediately after it.
 */

#define DFA_RETURN_SUCCESS_      return s - s0
#define DFA_RETURN_FAILURE_      return 0
#ifdef HAS_EXTRA_LONG_UTF8
#  define DFA_TEASE_APART_FF_  goto tease_apart_FF
#else
#  define DFA_TEASE_APART_FF_  DFA_RETURN_FAILURE_
#endif

#define PERL_IS_UTF8_CHAR_DFA(s0, e, dfa_tab,                               \
                              accept_action,                                \
                              reject_action,                                \
                              incomplete_char_action)                       \
    STMT_START {                                                            \
        const U8 * s = s0;                                                  \
        UV state = 0;                                                       \
                                                                            \
        PERL_NON_CORE_CHECK_EMPTY(s,e);                                     \
                                                                            \
        do {                                                                \
            state = dfa_tab[256 + state + dfa_tab[*s]];                     \
            s++;                                                            \
                                                                            \
            if (state == 0) {   /* Accepting state */                       \
                accept_action;                                              \
            }                                                               \
                                                                            \
            if (UNLIKELY(state == 1)) { /* Rejecting state */               \
                reject_action;                                              \
            }                                                               \
        } while (s < e);                                                    \
                                                                            \
        /* Here, dropped out of loop before end-of-char */                  \
        incomplete_char_action;                                             \
    } STMT_END


/*

=for apidoc isUTF8_CHAR

Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
that represents some code point; otherwise it evaluates to 0.  If non-zero, the
value gives how many bytes starting at C<s> comprise the code point's
representation.  Any bytes remaining before C<e>, but beyond the ones needed to
form the first code point in C<s>, are not examined.

The code point can be any that will fit in an IV on this machine, using Perl's
extension to official UTF-8 to represent those higher than the Unicode maximum
of 0x10FFFF.  That means that this macro is used to efficiently decide if the
next few bytes in C<s> is legal UTF-8 for a single character.

Use C<L</isSTRICT_UTF8_CHAR>> to restrict the acceptable code points to those
defined by Unicode to be fully interchangeable across applications;
C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
code points; and C<L</isUTF8_CHAR_flags>> for a more customized definition.

Use C<L</is_utf8_string>>, C<L</is_utf8_string_loc>>, and
C<L</is_utf8_string_loclen>> to check entire strings.

Note also that a UTF-8 "invariant" character (i.e. ASCII on non-EBCDIC
machines) is a valid UTF-8 character.

=cut

This uses an adaptation of the table and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version.  A copyright notice for the original
version is given at the beginning of this file.  The Perl adapation is
documented at the definition of PL_extended_utf8_dfa_tab[].
*/

PERL_STATIC_INLINE Size_t
Perl_isUTF8_CHAR(const U8 * const s0, const U8 * const e)
{
    PERL_ARGS_ASSERT_ISUTF8_CHAR;

    PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
                          DFA_RETURN_SUCCESS_,
                          DFA_TEASE_APART_FF_,
                          DFA_RETURN_FAILURE_);

    /* Here, we didn't return success, but dropped out of the loop.  In the
     * case of PL_extended_utf8_dfa_tab, this means the input is either
     * malformed, or the start byte was FF on a platform that the dfa doesn't
     * handle FF's.  Call a helper function. */

#ifdef HAS_EXTRA_LONG_UTF8

  tease_apart_FF:

    /* In the case of PL_extended_utf8_dfa_tab, getting here means the input is
     * either malformed, or was for the largest possible start byte, which we
     * now check, not inline */
    if (*s0 != I8_TO_NATIVE_UTF8(0xFF)) {
        return 0;
    }

    return is_utf8_FF_helper_(s0, e,
                              FALSE /* require full, not partial char */
                             );
#endif

}

/*

=for apidoc isSTRICT_UTF8_CHAR

Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
Unicode code point completely acceptable for open interchange between all
applications; otherwise it evaluates to 0.  If non-zero, the value gives how
many bytes starting at C<s> comprise the code point's representation.  Any
bytes remaining before C<e>, but beyond the ones needed to form the first code
point in C<s>, are not examined.

The largest acceptable code point is the Unicode maximum 0x10FFFF, and must not
be a surrogate nor a non-character code point.  Thus this excludes any code
point from Perl's extended UTF-8.

This is used to efficiently decide if the next few bytes in C<s> is
legal Unicode-acceptable UTF-8 for a single character.

Use C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
code points; C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8;
and C<L</isUTF8_CHAR_flags>> for a more customized definition.

Use C<L</is_strict_utf8_string>>, C<L</is_strict_utf8_string_loc>>, and
C<L</is_strict_utf8_string_loclen>> to check entire strings.

=cut

This uses an adaptation of the tables and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version.  A copyright notice for the original
version is given at the beginning of this file.  The Perl adapation is
documented at the definition of strict_extended_utf8_dfa_tab[].

*/

PERL_STATIC_INLINE Size_t
Perl_isSTRICT_UTF8_CHAR(const U8 * const s0, const U8 * const e)
{
    PERL_ARGS_ASSERT_ISSTRICT_UTF8_CHAR;

    PERL_IS_UTF8_CHAR_DFA(s0, e, PL_strict_utf8_dfa_tab,
                          DFA_RETURN_SUCCESS_,
                          goto check_hanguls,
                          DFA_RETURN_FAILURE_);
  check_hanguls:

    /* Here, we didn't return success, but dropped out of the loop.  In the
     * case of PL_strict_utf8_dfa_tab, this means the input is either
     * malformed, or was for certain Hanguls; handle them specially */

    /* The dfa above drops out for incomplete or illegal inputs, and certain
     * legal Hanguls; check and return accordingly */
    return is_HANGUL_ED_utf8_safe(s0, e);
}

/*

=for apidoc isC9_STRICT_UTF8_CHAR

Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
Unicode non-surrogate code point; otherwise it evaluates to 0.  If non-zero,
the value gives how many bytes starting at C<s> comprise the code point's
representation.  Any bytes remaining before C<e>, but beyond the ones needed to
form the first code point in C<s>, are not examined.

The largest acceptable code point is the Unicode maximum 0x10FFFF.  This
differs from C<L</isSTRICT_UTF8_CHAR>> only in that it accepts non-character
code points.  This corresponds to
L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.
which said that non-character code points are merely discouraged rather than
completely forbidden in open interchange.  See
L<perlunicode/Noncharacter code points>.

Use C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8; and
C<L</isUTF8_CHAR_flags>> for a more customized definition.

Use C<L</is_c9strict_utf8_string>>, C<L</is_c9strict_utf8_string_loc>>, and
C<L</is_c9strict_utf8_string_loclen>> to check entire strings.

=cut

This uses an adaptation of the tables and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version.  A copyright notice for the original
version is given at the beginning of this file.  The Perl adapation is
documented at the definition of PL_c9_utf8_dfa_tab[].

*/

PERL_STATIC_INLINE Size_t
Perl_isC9_STRICT_UTF8_CHAR(const U8 * const s0, const U8 * const e)
{
    PERL_ARGS_ASSERT_ISC9_STRICT_UTF8_CHAR;

    PERL_IS_UTF8_CHAR_DFA(s0, e, PL_c9_utf8_dfa_tab,
                          DFA_RETURN_SUCCESS_,
                          DFA_RETURN_FAILURE_,
                          DFA_RETURN_FAILURE_);
}

/*

=for apidoc is_strict_utf8_string_loc

Like C<L</is_strict_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.

See also C<L</is_strict_utf8_string_loclen>>.

=cut
*/

#define is_strict_utf8_string_loc(s, len, ep)                               \
                                is_strict_utf8_string_loclen(s, len, ep, 0)

/*

=for apidoc is_strict_utf8_string_loclen

Like C<L</is_strict_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.

See also C<L</is_strict_utf8_string_loc>>.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_STRICT_UTF8_STRING_LOCLEN;

    if (len == 0) {
        len = strlen((const char *) s);
    }

    if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
        if (el)
            *el = len;

        if (ep) {
            *ep = s + len;
        }

        return TRUE;
    }

    {
        const U8* const send = s + len;
        const U8* x = first_variant;
        STRLEN outlen = first_variant - s;

        while (x < send) {
            const STRLEN cur_len = isSTRICT_UTF8_CHAR(x, send);
            if (UNLIKELY(! cur_len)) {
                break;
            }
            x += cur_len;
            outlen++;
        }

        if (el)
            *el = outlen;

        if (ep) {
            *ep = x;
        }

        return (x == send);
    }
}

/*

=for apidoc is_c9strict_utf8_string_loc

Like C<L</is_c9strict_utf8_string>> but stores the location of the failure (in
the case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.

See also C<L</is_c9strict_utf8_string_loclen>>.

=cut
*/

#define is_c9strict_utf8_string_loc(s, len, ep)	                            \
                            is_c9strict_utf8_string_loclen(s, len, ep, 0)

/*

=for apidoc is_c9strict_utf8_string_loclen

Like C<L</is_c9strict_utf8_string>> but stores the location of the failure (in
the case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8 encoded
characters in the C<el> pointer.

See also C<L</is_c9strict_utf8_string_loc>>.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_c9strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_C9STRICT_UTF8_STRING_LOCLEN;

    if (len == 0) {
        len = strlen((const char *) s);
    }

    if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
        if (el)
            *el = len;

        if (ep) {
            *ep = s + len;
        }

        return TRUE;
    }

    {
        const U8* const send = s + len;
        const U8* x = first_variant;
        STRLEN outlen = first_variant - s;

        while (x < send) {
            const STRLEN cur_len = isC9_STRICT_UTF8_CHAR(x, send);
            if (UNLIKELY(! cur_len)) {
                break;
            }
            x += cur_len;
            outlen++;
        }

        if (el)
            *el = outlen;

        if (ep) {
            *ep = x;
        }

        return (x == send);
    }
}

/*

=for apidoc is_utf8_string_loc_flags

Like C<L</is_utf8_string_flags>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.

See also C<L</is_utf8_string_loclen_flags>>.

=cut
*/

#define is_utf8_string_loc_flags(s, len, ep, flags)                         \
                        is_utf8_string_loclen_flags(s, len, ep, 0, flags)


/* The above 3 actual functions could have been moved into the more general one
 * just below, and made #defines that call it with the right 'flags'.  They are
 * currently kept separate to increase their chances of getting inlined */

/*

=for apidoc is_utf8_string_loclen_flags

Like C<L</is_utf8_string_flags>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.

See also C<L</is_utf8_string_loc_flags>>.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_utf8_string_loclen_flags(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el, const U32 flags)
{
    const U8 * first_variant;

    PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN_FLAGS;
    assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
                          |UTF8_DISALLOW_PERL_EXTENDED)));

    if (len == 0) {
        len = strlen((const char *) s);
    }

    if (flags == 0) {
        return is_utf8_string_loclen(s, len, ep, el);
    }

    if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
                                        == UTF8_DISALLOW_ILLEGAL_INTERCHANGE)
    {
        return is_strict_utf8_string_loclen(s, len, ep, el);
    }

    if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
                                    == UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE)
    {
        return is_c9strict_utf8_string_loclen(s, len, ep, el);
    }

    if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
        if (el)
            *el = len;

        if (ep) {
            *ep = s + len;
        }

        return TRUE;
    }

    {
        const U8* send = s + len;
        const U8* x = first_variant;
        STRLEN outlen = first_variant - s;

        while (x < send) {
            const STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
            if (UNLIKELY(! cur_len)) {
                break;
            }
            x += cur_len;
            outlen++;
        }

        if (el)
            *el = outlen;

        if (ep) {
            *ep = x;
        }

        return (x == send);
    }
}

/*
=for apidoc utf8_distance

Returns the number of UTF-8 characters between the UTF-8 pointers C<a>
and C<b>.

WARNING: use only if you *know* that the pointers point inside the
same UTF-8 buffer.

=cut
*/

PERL_STATIC_INLINE IV
Perl_utf8_distance(pTHX_ const U8 *a, const U8 *b)
{
    PERL_ARGS_ASSERT_UTF8_DISTANCE;

    return (a < b) ? -1 * (IV) utf8_length(a, b) : (IV) utf8_length(b, a);
}

/*
=for apidoc utf8_hop

Return the UTF-8 pointer C<s> displaced by C<off> characters, either
forward or backward.

WARNING: do not use the following unless you *know* C<off> is within
the UTF-8 data pointed to by C<s> *and* that on entry C<s> is aligned
on the first byte of character or just after the last byte of a character.

=cut
*/

PERL_STATIC_INLINE U8 *
Perl_utf8_hop(const U8 *s, SSize_t off)
{
    PERL_ARGS_ASSERT_UTF8_HOP;

    /* Note: cannot use UTF8_IS_...() too eagerly here since e.g
     * the bitops (especially ~) can create illegal UTF-8.
     * In other words: in Perl UTF-8 is not just for Unicode. */

    if (off >= 0) {
        while (off--)
            s += UTF8SKIP(s);
    }
    else {
        while (off++) {
            s--;
            while (UTF8_IS_CONTINUATION(*s))
                s--;
        }
    }
    GCC_DIAG_IGNORE(-Wcast-qual)
    return (U8 *)s;
    GCC_DIAG_RESTORE
}

/*
=for apidoc utf8_hop_forward

Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
forward.

C<off> must be non-negative.

C<s> must be before or equal to C<end>.

When moving forward it will not move beyond C<end>.

Will not exceed this limit even if the string is not valid "UTF-8".

=cut
*/

PERL_STATIC_INLINE U8 *
Perl_utf8_hop_forward(const U8 *s, SSize_t off, const U8 *end)
{
    PERL_ARGS_ASSERT_UTF8_HOP_FORWARD;

    /* Note: cannot use UTF8_IS_...() too eagerly here since e.g
     * the bitops (especially ~) can create illegal UTF-8.
     * In other words: in Perl UTF-8 is not just for Unicode. */

    assert(s <= end);
    assert(off >= 0);

    while (off--) {
        STRLEN skip = UTF8SKIP(s);
        if ((STRLEN)(end - s) <= skip) {
            GCC_DIAG_IGNORE(-Wcast-qual)
            return (U8 *)end;
            GCC_DIAG_RESTORE
        }
        s += skip;
    }

    GCC_DIAG_IGNORE(-Wcast-qual)
    return (U8 *)s;
    GCC_DIAG_RESTORE
}

/*
=for apidoc utf8_hop_back

Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
backward.

C<off> must be non-positive.

C<s> must be after or equal to C<start>.

When moving backward it will not move before C<start>.

Will not exceed this limit even if the string is not valid "UTF-8".

=cut
*/

PERL_STATIC_INLINE U8 *
Perl_utf8_hop_back(const U8 *s, SSize_t off, const U8 *start)
{
    PERL_ARGS_ASSERT_UTF8_HOP_BACK;

    /* Note: cannot use UTF8_IS_...() too eagerly here since e.g
     * the bitops (especially ~) can create illegal UTF-8.
     * In other words: in Perl UTF-8 is not just for Unicode. */

    assert(start <= s);
    assert(off <= 0);

    while (off++ && s > start) {
        do {
            s--;
        } while (UTF8_IS_CONTINUATION(*s) && s > start);
    }

    GCC_DIAG_IGNORE(-Wcast-qual)
    return (U8 *)s;
    GCC_DIAG_RESTORE
}

/*
=for apidoc utf8_hop_safe

Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
either forward or backward.

When moving backward it will not move before C<start>.

When moving forward it will not move beyond C<end>.

Will not exceed those limits even if the string is not valid "UTF-8".

=cut
*/

PERL_STATIC_INLINE U8 *
Perl_utf8_hop_safe(const U8 *s, SSize_t off, const U8 *start, const U8 *end)
{
    PERL_ARGS_ASSERT_UTF8_HOP_SAFE;

    /* Note: cannot use UTF8_IS_...() too eagerly here since e.g
     * the bitops (especially ~) can create illegal UTF-8.
     * In other words: in Perl UTF-8 is not just for Unicode. */

    assert(start <= s && s <= end);

    if (off >= 0) {
        return utf8_hop_forward(s, off, end);
    }
    else {
        return utf8_hop_back(s, off, start);
    }
}

/*

=for apidoc isUTF8_CHAR_flags

Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
that represents some code point, subject to the restrictions given by C<flags>;
otherwise it evaluates to 0.  If non-zero, the value gives how many bytes
starting at C<s> comprise the code point's representation.  Any bytes remaining
before C<e>, but beyond the ones needed to form the first code point in C<s>,
are not examined.

If C<flags> is 0, this gives the same results as C<L</isUTF8_CHAR>>;
if C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
as C<L</isSTRICT_UTF8_CHAR>>;
and if C<flags> is C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives
the same results as C<L</isC9_STRICT_UTF8_CHAR>>.
Otherwise C<flags> may be any combination of the C<UTF8_DISALLOW_I<foo>> flags
understood by C<L</utf8n_to_uvchr>>, with the same meanings.

The three alternative macros are for the most commonly needed validations; they
are likely to run somewhat faster than this more general one, as they can be
inlined into your code.

Use L</is_utf8_string_flags>, L</is_utf8_string_loc_flags>, and
L</is_utf8_string_loclen_flags> to check entire strings.

=cut
*/

PERL_STATIC_INLINE STRLEN
Perl_isUTF8_CHAR_flags(const U8 * const s0, const U8 * const e, const U32 flags)
{
    PERL_ARGS_ASSERT_ISUTF8_CHAR_FLAGS;
    assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
                          |UTF8_DISALLOW_PERL_EXTENDED)));

    PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
                          goto check_success,
                          DFA_TEASE_APART_FF_,
                          DFA_RETURN_FAILURE_);

  check_success:

    return is_utf8_char_helper_(s0, e, flags);

#ifdef HAS_EXTRA_LONG_UTF8

  tease_apart_FF:

    /* In the case of PL_extended_utf8_dfa_tab, getting here means the input is
     * either malformed, or was for the largest possible start byte, which
     * indicates perl extended UTF-8, well above the Unicode maximum */
    if (   *s0 != I8_TO_NATIVE_UTF8(0xFF)
        || (flags & (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_PERL_EXTENDED)))
    {
        return 0;
    }

    /* Otherwise examine the sequence not inline */
    return is_utf8_FF_helper_(s0, e,
                              FALSE /* require full, not partial char */
                             );
#endif

}

/*

=for apidoc is_utf8_valid_partial_char

Returns 0 if the sequence of bytes starting at C<s> and looking no further than
S<C<e - 1>> is the UTF-8 encoding, as extended by Perl, for one or more code
points.  Otherwise, it returns 1 if there exists at least one non-empty
sequence of bytes that when appended to sequence C<s>, starting at position
C<e> causes the entire sequence to be the well-formed UTF-8 of some code point;
otherwise returns 0.

In other words this returns TRUE if C<s> points to a partial UTF-8-encoded code
point.

This is useful when a fixed-length buffer is being tested for being well-formed
UTF-8, but the final few bytes in it don't comprise a full character; that is,
it is split somewhere in the middle of the final code point's UTF-8
representation.  (Presumably when the buffer is refreshed with the next chunk
of data, the new first bytes will complete the partial code point.)   This
function is used to verify that the final bytes in the current buffer are in
fact the legal beginning of some code point, so that if they aren't, the
failure can be signalled without having to wait for the next read.

=cut
*/
#define is_utf8_valid_partial_char(s, e)                                    \
                                is_utf8_valid_partial_char_flags(s, e, 0)

/*

=for apidoc is_utf8_valid_partial_char_flags

Like C<L</is_utf8_valid_partial_char>>, it returns a boolean giving whether
or not the input is a valid UTF-8 encoded partial character, but it takes an
extra parameter, C<flags>, which can further restrict which code points are
considered valid.

If C<flags> is 0, this behaves identically to
C<L</is_utf8_valid_partial_char>>.  Otherwise C<flags> can be any combination
of the C<UTF8_DISALLOW_I<foo>> flags accepted by C<L</utf8n_to_uvchr>>.  If
there is any sequence of bytes that can complete the input partial character in
such a way that a non-prohibited character is formed, the function returns
TRUE; otherwise FALSE.  Non character code points cannot be determined based on
partial character input.  But many  of the other possible excluded types can be
determined from just the first one or two bytes.

=cut
 */

PERL_STATIC_INLINE bool
Perl_is_utf8_valid_partial_char_flags(const U8 * const s0, const U8 * const e, const U32 flags)
{
    PERL_ARGS_ASSERT_IS_UTF8_VALID_PARTIAL_CHAR_FLAGS;
    assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
                          |UTF8_DISALLOW_PERL_EXTENDED)));

    PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
                          DFA_RETURN_FAILURE_,
                          DFA_TEASE_APART_FF_,
                          NOOP);

    /* The NOOP above causes the DFA to drop down here iff the input was a
     * partial character.  flags=0 => can return TRUE immediately; otherwise we
     * need to check (not inline) if the partial character is the beginning of
     * a disallowed one */
    if (flags == 0) {
        return TRUE;
    }

    return cBOOL(is_utf8_char_helper_(s0, e, flags));

#ifdef HAS_EXTRA_LONG_UTF8

  tease_apart_FF:

    /* Getting here means the input is either malformed, or, in the case of
     * PL_extended_utf8_dfa_tab, was for the largest possible start byte.  The
     * latter case has to be extended UTF-8, so can fail immediately if that is
     * forbidden */

    if (   *s0 != I8_TO_NATIVE_UTF8(0xFF)
        || (flags & (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_PERL_EXTENDED)))
    {
        return 0;
    }

    return is_utf8_FF_helper_(s0, e,
                              TRUE /* Require to be a partial character */
                             );
#endif

}

/*

=for apidoc is_utf8_fixed_width_buf_flags

Returns TRUE if the fixed-width buffer starting at C<s> with length C<len>
is entirely valid UTF-8, subject to the restrictions given by C<flags>;
otherwise it returns FALSE.

If C<flags> is 0, any well-formed UTF-8, as extended by Perl, is accepted
without restriction.  If the final few bytes of the buffer do not form a
complete code point, this will return TRUE anyway, provided that
C<L</is_utf8_valid_partial_char_flags>> returns TRUE for them.

If C<flags> in non-zero, it can be any combination of the
C<UTF8_DISALLOW_I<foo>> flags accepted by C<L</utf8n_to_uvchr>>, and with the
same meanings.

This function differs from C<L</is_utf8_string_flags>> only in that the latter
returns FALSE if the final few bytes of the string don't form a complete code
point.

=cut
 */
#define is_utf8_fixed_width_buf_flags(s, len, flags)                        \
                is_utf8_fixed_width_buf_loclen_flags(s, len, 0, 0, flags)

/*

=for apidoc is_utf8_fixed_width_buf_loc_flags

Like C<L</is_utf8_fixed_width_buf_flags>> but stores the location of the
failure in the C<ep> pointer.  If the function returns TRUE, C<*ep> will point
to the beginning of any partial character at the end of the buffer; if there is
no partial character C<*ep> will contain C<s>+C<len>.

See also C<L</is_utf8_fixed_width_buf_loclen_flags>>.

=cut
*/

#define is_utf8_fixed_width_buf_loc_flags(s, len, loc, flags)               \
                is_utf8_fixed_width_buf_loclen_flags(s, len, loc, 0, flags)

/*

=for apidoc is_utf8_fixed_width_buf_loclen_flags

Like C<L</is_utf8_fixed_width_buf_loc_flags>> but stores the number of
complete, valid characters found in the C<el> pointer.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_utf8_fixed_width_buf_loclen_flags(const U8 * const s,
                                       STRLEN len,
                                       const U8 **ep,
                                       STRLEN *el,
                                       const U32 flags)
{
    const U8 * maybe_partial;

    PERL_ARGS_ASSERT_IS_UTF8_FIXED_WIDTH_BUF_LOCLEN_FLAGS;

    if (! ep) {
        ep  = &maybe_partial;
    }

    /* If it's entirely valid, return that; otherwise see if the only error is
     * that the final few bytes are for a partial character */
    return    is_utf8_string_loclen_flags(s, len, ep, el, flags)
           || is_utf8_valid_partial_char_flags(*ep, s + len, flags);
}

PERL_STATIC_INLINE UV
Perl_utf8n_to_uvchr_msgs(const U8 *s,
                      STRLEN curlen,
                      STRLEN *retlen,
                      const U32 flags,
                      U32 * errors,
                      AV ** msgs)
{
    /* This is the inlined portion of utf8n_to_uvchr_msgs.  It handles the
     * simple cases, and, if necessary calls a helper function to deal with the
     * more complex ones.  Almost all well-formed non-problematic code points
     * are considered simple, so that it's unlikely that the helper function
     * will need to be called.
     *
     * This is an adaptation of the tables and algorithm given in
     * https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides
     * comprehensive documentation of the original version.  A copyright notice
     * for the original version is given at the beginning of this file.  The
     * Perl adapation is documented at the definition of PL_strict_utf8_dfa_tab[].
     */

    const U8 * const s0 = s;
    const U8 * send = s0 + curlen;
    UV uv = 0;      /* The 0 silences some stupid compilers */
    UV state = 0;

    PERL_ARGS_ASSERT_UTF8N_TO_UVCHR_MSGS;

    /* This dfa is fast.  If it accepts the input, it was for a well-formed,
     * non-problematic code point, which can be returned immediately.
     * Otherwise we call a helper function to figure out the more complicated
     * cases. */

    while (s < send && LIKELY(state != 1)) {
        UV type = PL_strict_utf8_dfa_tab[*s];

        uv = (state == 0)
             ?  ((0xff >> type) & NATIVE_UTF8_TO_I8(*s))
             : UTF8_ACCUMULATE(uv, *s);
        state = PL_strict_utf8_dfa_tab[256 + state + type];

        if (state != 0) {
            s++;
            continue;
        }

        if (retlen) {
            *retlen = s - s0 + 1;
        }
        if (errors) {
            *errors = 0;
        }
        if (msgs) {
            *msgs = NULL;
        }

        return UNI_TO_NATIVE(uv);
    }

    /* Here is potentially problematic.  Use the full mechanism */
    return _utf8n_to_uvchr_msgs_helper(s0, curlen, retlen, flags, errors, msgs);
}

PERL_STATIC_INLINE UV
Perl_utf8_to_uvchr_buf_helper(pTHX_ const U8 *s, const U8 *send, STRLEN *retlen)
{
    PERL_ARGS_ASSERT_UTF8_TO_UVCHR_BUF_HELPER;

    assert(s < send);

    if (! ckWARN_d(WARN_UTF8)) {

        /* EMPTY is not really allowed, and asserts on debugging builds.  But
         * on non-debugging we have to deal with it, and this causes it to
         * return the REPLACEMENT CHARACTER, as the documentation indicates */
        return utf8n_to_uvchr(s, send - s, retlen,
                              (UTF8_ALLOW_ANY | UTF8_ALLOW_EMPTY));
    }
    else {
        UV ret = utf8n_to_uvchr(s, send - s, retlen, 0);
        if (retlen && ret == 0 && *s != '\0') {
            *retlen = (STRLEN) -1;
        }

        return ret;
    }
}

/* ------------------------------- perl.h ----------------------------- */

/*
=for apidoc_section $utility

=for apidoc is_safe_syscall

Test that the given C<pv> (with length C<len>) doesn't contain any internal
C<NUL> characters.
If it does, set C<errno> to C<ENOENT>, optionally warn using the C<syscalls>
category, and return FALSE.

Return TRUE if the name is safe.

C<what> and C<op_name> are used in any warning.

Used by the C<IS_SAFE_SYSCALL()> macro.

=cut
*/

PERL_STATIC_INLINE bool
Perl_is_safe_syscall(pTHX_ const char *pv, STRLEN len, const char *what, const char *op_name)
{
    /* While the Windows CE API provides only UCS-16 (or UTF-16) APIs
     * perl itself uses xce*() functions which accept 8-bit strings.
     */

    PERL_ARGS_ASSERT_IS_SAFE_SYSCALL;

    if (len > 1) {
        char *null_at;
        if (UNLIKELY((null_at = (char *)memchr(pv, 0, len-1)) != NULL)) {
                SETERRNO(ENOENT, LIB_INVARG);
                Perl_ck_warner(aTHX_ packWARN(WARN_SYSCALLS),
                                   "Invalid \\0 character in %s for %s: %s\\0%s",
                                   what, op_name, pv, null_at+1);
                return FALSE;
        }
    }

    return TRUE;
}

/*

Return true if the supplied filename has a newline character
immediately before the first (hopefully only) NUL.

My original look at this incorrectly used the len from SvPV(), but
that's incorrect, since we allow for a NUL in pv[len-1].

So instead, strlen() and work from there.

This allow for the user reading a filename, forgetting to chomp it,
then calling:

  open my $foo, "$file\0";

*/

#ifdef PERL_CORE

PERL_STATIC_INLINE bool
S_should_warn_nl(const char *pv)
{
    STRLEN len;

    PERL_ARGS_ASSERT_SHOULD_WARN_NL;

    len = strlen(pv);

    return len > 0 && pv[len-1] == '\n';
}

#endif

#if defined(PERL_IN_PP_C) || defined(PERL_IN_PP_HOT_C)

PERL_STATIC_INLINE bool
S_lossless_NV_to_IV(const NV nv, IV *ivp)
{
    /* This function determines if the input NV 'nv' may be converted without
     * loss of data to an IV.  If not, it returns FALSE taking no other action.
     * But if it is possible, it does the conversion, returning TRUE, and
     * storing the converted result in '*ivp' */

    PERL_ARGS_ASSERT_LOSSLESS_NV_TO_IV;

#  if defined(NAN_COMPARE_BROKEN) && defined(Perl_isnan)
    /* Normally any comparison with a NaN returns false; if we can't rely
     * on that behaviour, check explicitly */
    if (UNLIKELY(Perl_isnan(nv))) {
        return FALSE;
    }
#  endif

    /* Written this way so that with an always-false NaN comparison we
     * return false */
    if (!(LIKELY(nv >= (NV) IV_MIN) && LIKELY(nv < IV_MAX_P1))) {
        return FALSE;
    }

    if ((IV) nv != nv) {
        return FALSE;
    }

    *ivp = (IV) nv;
    return TRUE;
}

#endif

/* ------------------ pp.c, regcomp.c, toke.c, universal.c ------------ */

#if defined(PERL_IN_PP_C) || defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_TOKE_C) || defined(PERL_IN_UNIVERSAL_C)

#define MAX_CHARSET_NAME_LENGTH 2

PERL_STATIC_INLINE const char *
S_get_regex_charset_name(const U32 flags, STRLEN* const lenp)
{
    PERL_ARGS_ASSERT_GET_REGEX_CHARSET_NAME;

    /* Returns a string that corresponds to the name of the regex character set
     * given by 'flags', and *lenp is set the length of that string, which
     * cannot exceed MAX_CHARSET_NAME_LENGTH characters */

    *lenp = 1;
    switch (get_regex_charset(flags)) {
        case REGEX_DEPENDS_CHARSET: return DEPENDS_PAT_MODS;
        case REGEX_LOCALE_CHARSET:  return LOCALE_PAT_MODS;
        case REGEX_UNICODE_CHARSET: return UNICODE_PAT_MODS;
        case REGEX_ASCII_RESTRICTED_CHARSET: return ASCII_RESTRICT_PAT_MODS;
        case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
            *lenp = 2;
            return ASCII_MORE_RESTRICT_PAT_MODS;
    }
    /* The NOT_REACHED; hides an assert() which has a rather complex
     * definition in perl.h. */
    NOT_REACHED; /* NOTREACHED */
    return "?";	    /* Unknown */
}

#endif

/*

Return false if any get magic is on the SV other than taint magic.

*/

PERL_STATIC_INLINE bool
Perl_sv_only_taint_gmagic(SV *sv)
{
    MAGIC *mg = SvMAGIC(sv);

    PERL_ARGS_ASSERT_SV_ONLY_TAINT_GMAGIC;

    while (mg) {
        if (mg->mg_type != PERL_MAGIC_taint
            && !(mg->mg_flags & MGf_GSKIP)
            && mg->mg_virtual->svt_get) {
            return FALSE;
        }
        mg = mg->mg_moremagic;
    }

    return TRUE;
}

/* ------------------ cop.h ------------------------------------------- */

/* implement GIMME_V() macro */

PERL_STATIC_INLINE U8
Perl_gimme_V(pTHX)
{
    I32 cxix;
    U8  gimme = (PL_op->op_flags & OPf_WANT);

    if (gimme)
        return gimme;
    cxix = PL_curstackinfo->si_cxsubix;
    if (cxix < 0)
        return PL_curstackinfo->si_type == PERLSI_SORT ? G_SCALAR: G_VOID;
    assert(cxstack[cxix].blk_gimme & G_WANT);
    return (cxstack[cxix].blk_gimme & G_WANT);
}


/* Enter a block. Push a new base context and return its address. */

PERL_STATIC_INLINE PERL_CONTEXT *
Perl_cx_pushblock(pTHX_ U8 type, U8 gimme, SV** sp, I32 saveix)
{
    PERL_CONTEXT * cx;

    PERL_ARGS_ASSERT_CX_PUSHBLOCK;

    CXINC;
    cx = CX_CUR();
    cx->cx_type        = type;
    cx->blk_gimme      = gimme;
    cx->blk_oldsaveix  = saveix;
    cx->blk_oldsp      = (I32)(sp - PL_stack_base);
    cx->blk_oldcop     = PL_curcop;
    cx->blk_oldmarksp  = (I32)(PL_markstack_ptr - PL_markstack);
    cx->blk_oldscopesp = PL_scopestack_ix;
    cx->blk_oldpm      = PL_curpm;
    cx->blk_old_tmpsfloor = PL_tmps_floor;

    PL_tmps_floor        = PL_tmps_ix;
    CX_DEBUG(cx, "PUSH");
    return cx;
}


/* Exit a block (RETURN and LAST). */

PERL_STATIC_INLINE void
Perl_cx_popblock(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_POPBLOCK;

    CX_DEBUG(cx, "POP");
    /* these 3 are common to cx_popblock and cx_topblock */
    PL_markstack_ptr = PL_markstack + cx->blk_oldmarksp;
    PL_scopestack_ix = cx->blk_oldscopesp;
    PL_curpm         = cx->blk_oldpm;

    /* LEAVE_SCOPE() should have made this true. /(?{})/ cheats
     * and leaves a CX entry lying around for repeated use, so
     * skip for multicall */                  \
    assert(   (CxTYPE(cx) == CXt_SUB && CxMULTICALL(cx))
            || PL_savestack_ix == cx->blk_oldsaveix);
    PL_curcop     = cx->blk_oldcop;
    PL_tmps_floor = cx->blk_old_tmpsfloor;
}

/* Continue a block elsewhere (e.g. NEXT, REDO, GOTO).
 * Whereas cx_popblock() restores the state to the point just before
 * cx_pushblock() was called,  cx_topblock() restores it to the point just
 * *after* cx_pushblock() was called. */

PERL_STATIC_INLINE void
Perl_cx_topblock(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_TOPBLOCK;

    CX_DEBUG(cx, "TOP");
    /* these 3 are common to cx_popblock and cx_topblock */
    PL_markstack_ptr = PL_markstack + cx->blk_oldmarksp;
    PL_scopestack_ix = cx->blk_oldscopesp;
    PL_curpm         = cx->blk_oldpm;

    PL_stack_sp      = PL_stack_base + cx->blk_oldsp;
}


PERL_STATIC_INLINE void
Perl_cx_pushsub(pTHX_ PERL_CONTEXT *cx, CV *cv, OP *retop, bool hasargs)
{
    U8 phlags = CX_PUSHSUB_GET_LVALUE_MASK(Perl_was_lvalue_sub);

    PERL_ARGS_ASSERT_CX_PUSHSUB;

    PERL_DTRACE_PROBE_ENTRY(cv);
    cx->blk_sub.old_cxsubix     = PL_curstackinfo->si_cxsubix;
    PL_curstackinfo->si_cxsubix = cx - PL_curstackinfo->si_cxstack;
    cx->blk_sub.cv = cv;
    cx->blk_sub.olddepth = CvDEPTH(cv);
    cx->blk_sub.prevcomppad = PL_comppad;
    cx->cx_type |= (hasargs) ? CXp_HASARGS : 0;
    cx->blk_sub.retop = retop;
    SvREFCNT_inc_simple_void_NN(cv);
    cx->blk_u16 = PL_op->op_private & (phlags|OPpDEREF);
}


/* subsets of cx_popsub() */

PERL_STATIC_INLINE void
Perl_cx_popsub_common(pTHX_ PERL_CONTEXT *cx)
{
    CV *cv;

    PERL_ARGS_ASSERT_CX_POPSUB_COMMON;
    assert(CxTYPE(cx) == CXt_SUB);

    PL_comppad = cx->blk_sub.prevcomppad;
    PL_curpad = LIKELY(PL_comppad) ? AvARRAY(PL_comppad) : NULL;
    cv = cx->blk_sub.cv;
    CvDEPTH(cv) = cx->blk_sub.olddepth;
    cx->blk_sub.cv = NULL;
    SvREFCNT_dec(cv);
    PL_curstackinfo->si_cxsubix = cx->blk_sub.old_cxsubix;
}


/* handle the @_ part of leaving a sub */

PERL_STATIC_INLINE void
Perl_cx_popsub_args(pTHX_ PERL_CONTEXT *cx)
{
    AV *av;

    PERL_ARGS_ASSERT_CX_POPSUB_ARGS;
    assert(CxTYPE(cx) == CXt_SUB);
    assert(AvARRAY(MUTABLE_AV(
        PadlistARRAY(CvPADLIST(cx->blk_sub.cv))[
                CvDEPTH(cx->blk_sub.cv)])) == PL_curpad);

    CX_POP_SAVEARRAY(cx);
    av = MUTABLE_AV(PAD_SVl(0));
    if (UNLIKELY(AvREAL(av)))
        /* abandon @_ if it got reified */
        clear_defarray(av, 0);
    else {
        CLEAR_ARGARRAY(av);
    }
}


PERL_STATIC_INLINE void
Perl_cx_popsub(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_POPSUB;
    assert(CxTYPE(cx) == CXt_SUB);

    PERL_DTRACE_PROBE_RETURN(cx->blk_sub.cv);

    if (CxHASARGS(cx))
        cx_popsub_args(cx);
    cx_popsub_common(cx);
}


PERL_STATIC_INLINE void
Perl_cx_pushformat(pTHX_ PERL_CONTEXT *cx, CV *cv, OP *retop, GV *gv)
{
    PERL_ARGS_ASSERT_CX_PUSHFORMAT;

    cx->blk_format.old_cxsubix = PL_curstackinfo->si_cxsubix;
    PL_curstackinfo->si_cxsubix= cx - PL_curstackinfo->si_cxstack;
    cx->blk_format.cv          = cv;
    cx->blk_format.retop       = retop;
    cx->blk_format.gv          = gv;
    cx->blk_format.dfoutgv     = PL_defoutgv;
    cx->blk_format.prevcomppad = PL_comppad;
    cx->blk_u16                = 0;

    SvREFCNT_inc_simple_void_NN(cv);
    CvDEPTH(cv)++;
    SvREFCNT_inc_void(cx->blk_format.dfoutgv);
}


PERL_STATIC_INLINE void
Perl_cx_popformat(pTHX_ PERL_CONTEXT *cx)
{
    CV *cv;
    GV *dfout;

    PERL_ARGS_ASSERT_CX_POPFORMAT;
    assert(CxTYPE(cx) == CXt_FORMAT);

    dfout = cx->blk_format.dfoutgv;
    setdefout(dfout);
    cx->blk_format.dfoutgv = NULL;
    SvREFCNT_dec_NN(dfout);

    PL_comppad = cx->blk_format.prevcomppad;
    PL_curpad = LIKELY(PL_comppad) ? AvARRAY(PL_comppad) : NULL;
    cv = cx->blk_format.cv;
    cx->blk_format.cv = NULL;
    --CvDEPTH(cv);
    SvREFCNT_dec_NN(cv);
    PL_curstackinfo->si_cxsubix = cx->blk_format.old_cxsubix;
}


PERL_STATIC_INLINE void
Perl_push_evalortry_common(pTHX_ PERL_CONTEXT *cx, OP *retop, SV *namesv)
{
    cx->blk_eval.retop         = retop;
    cx->blk_eval.old_namesv    = namesv;
    cx->blk_eval.old_eval_root = PL_eval_root;
    cx->blk_eval.cur_text      = PL_parser ? PL_parser->linestr : NULL;
    cx->blk_eval.cv            = NULL; /* later set by doeval_compile() */
    cx->blk_eval.cur_top_env   = PL_top_env;

    assert(!(PL_in_eval     & ~ 0x3F));
    assert(!(PL_op->op_type & ~0x1FF));
    cx->blk_u16 = (PL_in_eval & 0x3F) | ((U16)PL_op->op_type << 7);
}

PERL_STATIC_INLINE void
Perl_cx_pusheval(pTHX_ PERL_CONTEXT *cx, OP *retop, SV *namesv)
{
    PERL_ARGS_ASSERT_CX_PUSHEVAL;

    Perl_push_evalortry_common(aTHX_ cx, retop, namesv);

    cx->blk_eval.old_cxsubix    = PL_curstackinfo->si_cxsubix;
    PL_curstackinfo->si_cxsubix = cx - PL_curstackinfo->si_cxstack;
}

PERL_STATIC_INLINE void
Perl_cx_pushtry(pTHX_ PERL_CONTEXT *cx, OP *retop)
{
    PERL_ARGS_ASSERT_CX_PUSHTRY;

    Perl_push_evalortry_common(aTHX_ cx, retop, NULL);

    /* Don't actually change it, just store the current value so it's restored
     * by the common popeval */
    cx->blk_eval.old_cxsubix = PL_curstackinfo->si_cxsubix;
}


PERL_STATIC_INLINE void
Perl_cx_popeval(pTHX_ PERL_CONTEXT *cx)
{
    SV *sv;

    PERL_ARGS_ASSERT_CX_POPEVAL;
    assert(CxTYPE(cx) == CXt_EVAL);

    PL_in_eval = CxOLD_IN_EVAL(cx);
    assert(!(PL_in_eval & 0xc0));
    PL_eval_root = cx->blk_eval.old_eval_root;
    sv = cx->blk_eval.cur_text;
    if (sv && CxEVAL_TXT_REFCNTED(cx)) {
        cx->blk_eval.cur_text = NULL;
        SvREFCNT_dec_NN(sv);
    }

    sv = cx->blk_eval.old_namesv;
    if (sv) {
        cx->blk_eval.old_namesv = NULL;
        SvREFCNT_dec_NN(sv);
    }
    PL_curstackinfo->si_cxsubix = cx->blk_eval.old_cxsubix;
}


/* push a plain loop, i.e.
 *     { block }
 *     while (cond) { block }
 *     for (init;cond;continue) { block }
 * This loop can be last/redo'ed etc.
 */

PERL_STATIC_INLINE void
Perl_cx_pushloop_plain(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_PUSHLOOP_PLAIN;
    cx->blk_loop.my_op = cLOOP;
}


/* push a true for loop, i.e.
 *     for var (list) { block }
 */

PERL_STATIC_INLINE void
Perl_cx_pushloop_for(pTHX_ PERL_CONTEXT *cx, void *itervarp, SV* itersave)
{
    PERL_ARGS_ASSERT_CX_PUSHLOOP_FOR;

    /* this one line is common with cx_pushloop_plain */
    cx->blk_loop.my_op = cLOOP;

    cx->blk_loop.itervar_u.svp = (SV**)itervarp;
    cx->blk_loop.itersave      = itersave;
#ifdef USE_ITHREADS
    cx->blk_loop.oldcomppad = PL_comppad;
#endif
}


/* pop all loop types, including plain */

PERL_STATIC_INLINE void
Perl_cx_poploop(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_POPLOOP;

    assert(CxTYPE_is_LOOP(cx));
    if (  CxTYPE(cx) == CXt_LOOP_ARY
       || CxTYPE(cx) == CXt_LOOP_LAZYSV)
    {
        /* Free ary or cur. This assumes that state_u.ary.ary
         * aligns with state_u.lazysv.cur. See cx_dup() */
        SV *sv = cx->blk_loop.state_u.lazysv.cur;
        cx->blk_loop.state_u.lazysv.cur = NULL;
        SvREFCNT_dec_NN(sv);
        if (CxTYPE(cx) == CXt_LOOP_LAZYSV) {
            sv = cx->blk_loop.state_u.lazysv.end;
            cx->blk_loop.state_u.lazysv.end = NULL;
            SvREFCNT_dec_NN(sv);
        }
    }
    if (cx->cx_type & (CXp_FOR_PAD|CXp_FOR_GV)) {
        SV *cursv;
        SV **svp = (cx)->blk_loop.itervar_u.svp;
        if ((cx->cx_type & CXp_FOR_GV))
            svp = &GvSV((GV*)svp);
        cursv = *svp;
        *svp = cx->blk_loop.itersave;
        cx->blk_loop.itersave = NULL;
        SvREFCNT_dec(cursv);
    }
}


PERL_STATIC_INLINE void
Perl_cx_pushwhen(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_PUSHWHEN;

    cx->blk_givwhen.leave_op = cLOGOP->op_other;
}


PERL_STATIC_INLINE void
Perl_cx_popwhen(pTHX_ PERL_CONTEXT *cx)
{
    PERL_ARGS_ASSERT_CX_POPWHEN;
    assert(CxTYPE(cx) == CXt_WHEN);

    PERL_UNUSED_ARG(cx);
    PERL_UNUSED_CONTEXT;
    /* currently NOOP */
}


PERL_STATIC_INLINE void
Perl_cx_pushgiven(pTHX_ PERL_CONTEXT *cx, SV *orig_defsv)
{
    PERL_ARGS_ASSERT_CX_PUSHGIVEN;

    cx->blk_givwhen.leave_op = cLOGOP->op_other;
    cx->blk_givwhen.defsv_save = orig_defsv;
}


PERL_STATIC_INLINE void
Perl_cx_popgiven(pTHX_ PERL_CONTEXT *cx)
{
    SV *sv;

    PERL_ARGS_ASSERT_CX_POPGIVEN;
    assert(CxTYPE(cx) == CXt_GIVEN);

    sv = GvSV(PL_defgv);
    GvSV(PL_defgv) = cx->blk_givwhen.defsv_save;
    cx->blk_givwhen.defsv_save = NULL;
    SvREFCNT_dec(sv);
}

/* ------------------ util.h ------------------------------------------- */

/*
=for apidoc_section $string

=for apidoc foldEQ

Returns true if the leading C<len> bytes of the strings C<s1> and C<s2> are the
same
case-insensitively; false otherwise.  Uppercase and lowercase ASCII range bytes
match themselves and their opposite case counterparts.  Non-cased and non-ASCII
range bytes match only themselves.

=cut
*/

PERL_STATIC_INLINE I32
Perl_foldEQ(const char *s1, const char *s2, I32 len)
{
    const U8 *a = (const U8 *)s1;
    const U8 *b = (const U8 *)s2;

    PERL_ARGS_ASSERT_FOLDEQ;

    assert(len >= 0);

    while (len--) {
        if (*a != *b && *a != PL_fold[*b])
            return 0;
        a++,b++;
    }
    return 1;
}

PERL_STATIC_INLINE I32
Perl_foldEQ_latin1(const char *s1, const char *s2, I32 len)
{
    /* Compare non-UTF-8 using Unicode (Latin1) semantics.  Works on all folds
     * representable without UTF-8, except for LATIN_SMALL_LETTER_SHARP_S, and
     * does not check for this.  Nor does it check that the strings each have
     * at least 'len' characters. */

    const U8 *a = (const U8 *)s1;
    const U8 *b = (const U8 *)s2;

    PERL_ARGS_ASSERT_FOLDEQ_LATIN1;

    assert(len >= 0);

    while (len--) {
        if (*a != *b && *a != PL_fold_latin1[*b]) {
            return 0;
        }
        a++, b++;
    }
    return 1;
}

/*
=for apidoc_section $locale
=for apidoc foldEQ_locale

Returns true if the leading C<len> bytes of the strings C<s1> and C<s2> are the
same case-insensitively in the current locale; false otherwise.

=cut
*/

PERL_STATIC_INLINE I32
Perl_foldEQ_locale(const char *s1, const char *s2, I32 len)
{
    const U8 *a = (const U8 *)s1;
    const U8 *b = (const U8 *)s2;

    PERL_ARGS_ASSERT_FOLDEQ_LOCALE;

    assert(len >= 0);

    while (len--) {
        if (*a != *b && *a != PL_fold_locale[*b])
            return 0;
        a++,b++;
    }
    return 1;
}

/*
=for apidoc_section $string
=for apidoc my_strnlen

The C library C<strnlen> if available, or a Perl implementation of it.

C<my_strnlen()> computes the length of the string, up to C<maxlen>
characters.  It will never attempt to address more than C<maxlen>
characters, making it suitable for use with strings that are not
guaranteed to be NUL-terminated.

=cut

Description stolen from http://man.openbsd.org/strnlen.3,
implementation stolen from PostgreSQL.
*/
#ifndef HAS_STRNLEN

PERL_STATIC_INLINE Size_t
Perl_my_strnlen(const char *str, Size_t maxlen)
{
    const char *end = (char *) memchr(str, '\0', maxlen);

    PERL_ARGS_ASSERT_MY_STRNLEN;

    if (end == NULL) return maxlen;
    return end - str;
}

#endif

#if ! defined (HAS_MEMRCHR) && (defined(PERL_CORE) || defined(PERL_EXT))

PERL_STATIC_INLINE void *
S_my_memrchr(const char * s, const char c, const STRLEN len)
{
    /* memrchr(), since many platforms lack it */

    const char * t = s + len - 1;

    PERL_ARGS_ASSERT_MY_MEMRCHR;

    while (t >= s) {
        if (*t == c) {
            return (void *) t;
        }
        t--;
    }

    return NULL;
}

#endif

PERL_STATIC_INLINE char *
Perl_mortal_getenv(const char * str)
{
    /* This implements a (mostly) thread-safe, sequential-call-safe getenv().
     *
     * It's (mostly) thread-safe because it uses a mutex to prevent other
     * threads (that look at this mutex) from destroying the result before this
     * routine has a chance to copy the result to a place that won't be
     * destroyed before the caller gets a chance to handle it.  That place is a
     * mortal SV.  khw chose this over SAVEFREEPV because he is under the
     * impression that the SV will hang around longer under more circumstances
     *
     * The reason it isn't completely thread-safe is that other code could
     * simply not pay attention to the mutex.  All of the Perl core uses the
     * mutex, but it is possible for code from, say XS, to not use this mutex,
     * defeating the safety.
     *
     * getenv() returns, in some implementations, a pointer to a spot in the
     * **environ array, which could be invalidated at any time by this or
     * another thread changing the environment.  Other implementations copy the
     * **environ value to a static buffer, returning a pointer to that.  That
     * buffer might or might not be invalidated by a getenv() call in another
     * thread.  If it does get zapped, we need an exclusive lock.  Otherwise,
     * many getenv() calls can safely be running simultaneously, so a
     * many-reader (but no simultaneous writers) lock is ok.  There is a
     * Configure probe to see if another thread destroys the buffer, and the
     * mutex is defined accordingly.
     *
     * But in all cases, using the mutex prevents these problems, as long as
     * all code uses the same mutex..
     *
     * A complication is that this can be called during phases where the
     * mortalization process isn't available.  These are in interpreter
     * destruction or early in construction.  khw believes that at these times
     * there shouldn't be anything else going on, so plain getenv is safe AS
     * LONG AS the caller acts on the return before calling it again. */

    char * ret;
    dTHX;

    PERL_ARGS_ASSERT_MORTAL_GETENV;

    /* Can't mortalize without stacks.  khw believes that no other threads
     * should be running, so no need to lock things, and this may be during a
     * phase when locking isn't even available */
    if (UNLIKELY(PL_scopestack_ix == 0)) {
        return getenv(str);
    }

#ifdef PERL_MEM_LOG

    /* A major complication arises under PERL_MEM_LOG.  When that is active,
     * every memory allocation may result in logging, depending on the value of
     * ENV{PERL_MEM_LOG} at the moment.  That means, as we create the SV for
     * saving ENV{foo}'s value (but before saving it), the logging code will
     * call us recursively to find out what ENV{PERL_MEM_LOG} is.  Without some
     * care that could lead to: 1) infinite recursion; or 2) deadlock (trying to
     * lock a boolean mutex recursively); 3) destroying the getenv() static
     * buffer; or 4) destroying the temporary created by this for the copy
     * causes a log entry to be made which could cause a new temporary to be
     * created, which will need to be destroyed at some point, leading to an
     * infinite loop.
     *
     * The solution adopted here (after some gnashing of teeth) is to detect
     * the recursive calls and calls from the logger, and treat them specially.
     * Let's say we want to do getenv("foo").  We first find
     * getenv(PERL_MEM_LOG) and save it to a fixed-length per-interpreter
     * variable, so no temporary is required.  Then we do getenv(foo}, and in
     * the process of creating a temporary to save it, this function will be
     * called recursively to do a getenv(PERL_MEM_LOG).  On the recursed call,
     * we detect that it is such a call and return our saved value instead of
     * locking and doing a new getenv().  This solves all of problems 1), 2),
     * and 3).  Because all the getenv()s are done while the mutex is locked,
     * the state cannot have changed.  To solve 4), we don't create a temporary
     * when this is called from the logging code.  That code disposes of the
     * return value while the mutex is still locked.
     *
     * The value of getenv(PERL_MEM_LOG) can be anything, but only initial
     * digits and 3 particular letters are significant; the rest are ignored by
     * the memory logging code.  Thus the per-interpreter variable only needs
     * to be large enough to save the significant information, the size of
     * which is known at compile time.  The first byte is extra, reserved for
     * flags for our use.  To protect against overflowing, only the reserved
     * byte, as many digits as don't overflow, and the three letters are
     * stored.
     *
     * The reserved byte has two bits:
     *      0x1 if set indicates that if we get here, it is a recursive call of
     *          getenv()
     *      0x2 if set indicates that the call is from the logging code.
     *
     * If the flag indicates this is a recursive call, just return the stored
     * value of PL_mem_log;  An empty value gets turned into NULL. */
    if (strEQ(str, "PERL_MEM_LOG") && PL_mem_log[0] & 0x1) {
        if (PL_mem_log[1] == '\0') {
            return NULL;
        } else {
            return PL_mem_log + 1;
        }
    }

#endif

    GETENV_LOCK;

#ifdef PERL_MEM_LOG

    /* Here we are in a critical section.  As explained above, we do our own
     * getenv(PERL_MEM_LOG), saving the result safely. */
    ret = getenv("PERL_MEM_LOG");
    if (ret == NULL) {  /* No logging active */

        /* Return that immediately if called from the logging code */
        if (PL_mem_log[0] & 0x2) {
            GETENV_UNLOCK;
            return NULL;
        }

        PL_mem_log[1] = '\0';
    }
    else {
        char *mem_log_meat = PL_mem_log + 1;    /* first byte reserved */

        /* There is nothing to prevent the value of PERL_MEM_LOG from being an
         * extremely long string.  But we want only a few characters from it.
         * PL_mem_log has been made large enough to hold just the ones we need.
         * First the file descriptor. */
        if (isDIGIT(*ret)) {
            const char * s = ret;
            if (UNLIKELY(*s == '0')) {

                /* Reduce multiple leading zeros to a single one.  This is to
                 * allow the caller to change what to do with leading zeros. */
                *mem_log_meat++ = '0';
                s++;
                while (*s == '0') {
                    s++;
                }
            }

            /* If the input overflows, copy just enough for the result to also
             * overflow, plus 1 to make sure */
            while (isDIGIT(*s) && s < ret + TYPE_DIGITS(UV) + 1) {
                *mem_log_meat++ = *s++;
            }
        }

        /* Then each of the three significant characters */
        if (strchr(ret, 'm')) {
            *mem_log_meat++ = 'm';
        }
        if (strchr(ret, 's')) {
            *mem_log_meat++ = 's';
        }
        if (strchr(ret, 't')) {
            *mem_log_meat++ = 't';
        }
        *mem_log_meat = '\0';

        assert(mem_log_meat < PL_mem_log + sizeof(PL_mem_log));
    }

    /* If we are being called from the logger, it only needs the significant
     * portion of PERL_MEM_LOG, and doesn't need a safe copy */
    if (PL_mem_log[0] & 0x2) {
        assert(strEQ(str, "PERL_MEM_LOG"));
        GETENV_UNLOCK;
        return PL_mem_log + 1;
    }

    /* Here is a generic getenv().  This could be a getenv("PERL_MEM_LOG") that
     * is coming from other than the logging code, so it should be treated the
     * same as any other getenv(), returning the full value, not just the
     * significant part, and having its value saved.  Set the flag that
     * indicates any call to this routine will be a recursion from here */
    PL_mem_log[0] = 0x1;

#endif

    /* Now get the value of the real desired variable, and save a copy */
    ret = getenv(str);

    if (ret != NULL) {
        ret = SvPVX(sv_2mortal(newSVpv(ret, 0)));
    }

    GETENV_UNLOCK;

#ifdef PERL_MEM_LOG

    /* Clear the buffer */
    Zero(PL_mem_log, sizeof(PL_mem_log), char);

#endif

    return ret;
}

/*
 * ex: set ts=8 sts=4 sw=4 et:
 */