public inbox for openembedded-core@lists.openembedded.org
 help / color / mirror / Atom feed
From: "Mittal, Anuj" <anuj.mittal@intel.com>
To: "Qi.Chen@windriver.com" <Qi.Chen@windriver.com>,
	"openembedded-core@lists.openembedded.org"
	<openembedded-core@lists.openembedded.org>
Subject: Re: [OE-core][hardknott][PATCH V3] busybox: backport patches to fix CVEs
Date: Thu, 30 Dec 2021 15:03:50 +0000	[thread overview]
Message-ID: <b57878c065e1c9cc29d590f90a2b76604472539c.camel@intel.com> (raw)
In-Reply-To: <20211229014449.119943-1-Qi.Chen@windriver.com>

On Tue, 2021-12-28 at 17:44 -0800, Chen Qi wrote:
> Backport patches to fix the following CVEs:
> CVE-2021-42373
> CVE-2021-42378
> CVE-2021-42379
> CVE-2021-42380
> CVE-2021-42381
> CVE-2021-42382
> CVE-2021-42383
> CVE-2021-42384
> CVE-2021-42385
> CVE-2021-42386

This is causing these ptests to fail:

'busybox': ['awk_break', 'awk_continue'],

Please see:

qemux86-64:
https://autobuilder.yoctoproject.org/typhoon/#/builders/81/builds/2947

qemuarm64:
https://autobuilder.yoctoproject.org/typhoon/#/builders/82/builds/2674

Thanks,

Anuj

> 
> Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
> ---
>  .../busybox/busybox/0001-awk-fix-CVEs.patch   | 3142
> +++++++++++++++++
>  .../0002-man-fix-segfault-in-man-1.patch      |   30 +
>  meta/recipes-core/busybox/busybox_1.33.2.bb   |    2 +
>  3 files changed, 3174 insertions(+)
>  create mode 100644 meta/recipes-core/busybox/busybox/0001-awk-fix-
> CVEs.patch
>  create mode 100644 meta/recipes-core/busybox/busybox/0002-man-fix-
> segfault-in-man-1.patch
> 
> diff --git a/meta/recipes-core/busybox/busybox/0001-awk-fix-
> CVEs.patch b/meta/recipes-core/busybox/busybox/0001-awk-fix-
> CVEs.patch
> new file mode 100644
> index 0000000000..9b7079cbe9
> --- /dev/null
> +++ b/meta/recipes-core/busybox/busybox/0001-awk-fix-CVEs.patch
> @@ -0,0 +1,3142 @@
> +From 0b313943ee5c6611d32c55c3d16084d04a4c2a6d Mon Sep 17 00:00:00
> 2001
> +From: Chen Qi <Qi.Chen@windriver.com>
> +Date: Mon, 27 Dec 2021 21:25:33 -0800
> +Subject: [PATCH] awk: fix CVEs
> +
> +The awk CVEs is hard to be separated, thus we use the following
> method
> +to format the current patch.
> +git format-patch --stdout 1_33_1..1_34_1 -- editors/awk.c | git am -
> 3
> +
> +CVE: CVE-2021-42378
> +CVE: CVE-2021-42379
> +CVE: CVE-2021-42380
> +CVE: CVE-2021-42381
> +CVE: CVE-2021-42382
> +CVE: CVE-2021-42383
> +CVE: CVE-2021-42384
> +CVE: CVE-2021-42385
> +CVE: CVE-2021-42386
> +
> +Upstream-Status: Backport
> +
> +Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
> +---
> + editors/awk.c | 2060 +++++++++++++++++++++++++++++-----------------
> ---
> + 1 file changed, 1204 insertions(+), 856 deletions(-)
> +
> +diff --git a/editors/awk.c b/editors/awk.c
> +index 2c15f9e4e..f6314ac72 100644
> +--- a/editors/awk.c
> ++++ b/editors/awk.c
> +@@ -66,6 +66,8 @@
> + #endif
> + #ifndef debug_printf_parse
> + # define debug_printf_parse(...) (fprintf(stderr, __VA_ARGS__))
> ++#else
> ++# define debug_parse_print_tc(...) ((void)0)
> + #endif
> + 
> + 
> +@@ -91,7 +93,6 @@ enum {
> + };
> + 
> + #define       MAXVARFMT       240
> +-#define       MINNVBLOCK      64
> + 
> + /* variable flags */
> + #define       VF_NUMBER       0x0001  /* 1 = primary type is number
> */
> +@@ -101,7 +102,7 @@ enum {
> + #define       VF_USER         0x0200  /* 1 = user input (may be
> numeric string) */
> + #define       VF_SPECIAL      0x0400  /* 1 = requires extra
> handling when changed */
> + #define       VF_WALK         0x0800  /* 1 = variable has alloc'd
> x.walker list */
> +-#define       VF_FSTR         0x1000  /* 1 = var::string points to
> fstring buffer */
> ++#define       VF_FSTR         0x1000  /* 1 = don't free()
> var::string (not malloced, or is owned by something else) */
> + #define       VF_CHILD        0x2000  /* 1 = function arg; x.parent
> points to source */
> + #define       VF_DIRTY        0x4000  /* 1 = variable was set
> explicitly */
> + 
> +@@ -118,8 +119,8 @@ typedef struct walker_list {
> + /* Variable */
> + typedef struct var_s {
> +       unsigned type;            /* flags */
> +-      double number;
> +       char *string;
> ++      double number;
> +       union {
> +               int aidx;               /* func arg idx (for
> compilation stage) */
> +               struct xhash_s *array;  /* array ptr */
> +@@ -138,6 +139,7 @@ typedef struct chain_s {
> + /* Function */
> + typedef struct func_s {
> +       unsigned nargs;
> ++      smallint defined;
> +       struct chain_s body;
> + } func;
> + 
> +@@ -177,7 +179,7 @@ typedef struct node_s {
> +               struct node_s *n;
> +               var *v;
> +               int aidx;
> +-              char *new_progname;
> ++              const char *new_progname;
> +               regex_t *re;
> +       } l;
> +       union {
> +@@ -190,91 +192,120 @@ typedef struct node_s {
> +       } a;
> + } node;
> + 
> +-/* Block of temporary variables */
> +-typedef struct nvblock_s {
> +-      int size;
> +-      var *pos;
> +-      struct nvblock_s *prev;
> +-      struct nvblock_s *next;
> +-      var nv[];
> +-} nvblock;
> +-
> + typedef struct tsplitter_s {
> +       node n;
> +       regex_t re[2];
> + } tsplitter;
> + 
> + /* simple token classes */
> +-/* Order and hex values are very important!!!  See next_token() */
> +-#define       TC_SEQSTART     (1 << 0)                /* ( */
> +-#define       TC_SEQTERM      (1 << 1)                /* ) */
> +-#define       TC_REGEXP       (1 << 2)                /* /.../ */
> +-#define       TC_OUTRDR       (1 << 3)                /* | > >> */
> +-#define       TC_UOPPOST      (1 << 4)                /* unary
> postfix operator */
> +-#define       TC_UOPPRE1      (1 << 5)                /* unary
> prefix operator */
> +-#define       TC_BINOPX       (1 << 6)                /* two-opnd
> operator */
> +-#define       TC_IN           (1 << 7)
> +-#define       TC_COMMA        (1 << 8)
> +-#define       TC_PIPE         (1 << 9)                /* input
> redirection pipe */
> +-#define       TC_UOPPRE2      (1 << 10)               /* unary
> prefix operator */
> +-#define       TC_ARRTERM      (1 << 11)               /* ] */
> +-#define       TC_GRPSTART     (1 << 12)               /* { */
> +-#define       TC_GRPTERM      (1 << 13)               /* } */
> +-#define       TC_SEMICOL      (1 << 14)
> +-#define       TC_NEWLINE      (1 << 15)
> +-#define       TC_STATX        (1 << 16)               /* ctl
> statement (for, next...) */
> +-#define       TC_WHILE        (1 << 17)
> +-#define       TC_ELSE         (1 << 18)
> +-#define       TC_BUILTIN      (1 << 19)
> ++/* order and hex values are very important!!!  See next_token() */
> ++#define TC_LPAREN       (1 << 0)        /* ( */
> ++#define TC_RPAREN       (1 << 1)        /* ) */
> ++#define TC_REGEXP       (1 << 2)        /* /.../ */
> ++#define TC_OUTRDR       (1 << 3)        /* | > >> */
> ++#define TC_UOPPOST      (1 << 4)        /* unary postfix operator
> ++ -- */
> ++#define TC_UOPPRE1      (1 << 5)        /* unary prefix operator ++
> -- $ */
> ++#define TC_BINOPX       (1 << 6)        /* two-opnd operator */
> ++#define TC_IN           (1 << 7)        /* 'in' */
> ++#define TC_COMMA        (1 << 8)        /* , */
> ++#define TC_PIPE         (1 << 9)        /* input redirection pipe |
> */
> ++#define TC_UOPPRE2      (1 << 10)       /* unary prefix operator +
> - ! */
> ++#define TC_ARRTERM      (1 << 11)       /* ] */
> ++#define TC_LBRACE       (1 << 12)       /* { */
> ++#define TC_RBRACE       (1 << 13)       /* } */
> ++#define TC_SEMICOL      (1 << 14)       /* ; */
> ++#define TC_NEWLINE      (1 << 15)
> ++#define TC_STATX        (1 << 16)       /* ctl statement (for,
> next...) */
> ++#define TC_WHILE        (1 << 17)       /* 'while' */
> ++#define TC_ELSE         (1 << 18)       /* 'else' */
> ++#define TC_BUILTIN      (1 << 19)
> + /* This costs ~50 bytes of code.
> +  * A separate class to support deprecated "length" form. If we
> don't need that
> +  * (i.e. if we demand that only "length()" with () is valid), then
> TC_LENGTH
> +  * can be merged with TC_BUILTIN:
> +  */
> +-#define       TC_LENGTH       (1 << 20)
> +-#define       TC_GETLINE      (1 << 21)
> +-#define       TC_FUNCDECL     (1 << 22)               /* 'function'
> 'func' */
> +-#define       TC_BEGIN        (1 << 23)
> +-#define       TC_END          (1 << 24)
> +-#define       TC_EOF          (1 << 25)
> +-#define       TC_VARIABLE     (1 << 26)
> +-#define       TC_ARRAY        (1 << 27)
> +-#define       TC_FUNCTION     (1 << 28)
> +-#define       TC_STRING       (1 << 29)
> +-#define       TC_NUMBER       (1 << 30)
> +-
> +-#define       TC_UOPPRE  (TC_UOPPRE1 | TC_UOPPRE2)
> +-
> +-/* combined token classes */
> +-#define       TC_BINOP   (TC_BINOPX | TC_COMMA | TC_PIPE | TC_IN)
> +-//#define     TC_UNARYOP (TC_UOPPRE | TC_UOPPOST)
> +-#define       TC_OPERAND (TC_VARIABLE | TC_ARRAY | TC_FUNCTION \
> +-                   | TC_BUILTIN | TC_LENGTH | TC_GETLINE \
> +-                   | TC_SEQSTART | TC_STRING | TC_NUMBER)
> +-
> +-#define       TC_STATEMNT (TC_STATX | TC_WHILE)
> +-#define       TC_OPTERM  (TC_SEMICOL | TC_NEWLINE)
> ++#define TC_LENGTH       (1 << 20)       /* 'length' */
> ++#define TC_GETLINE      (1 << 21)       /* 'getline' */
> ++#define TC_FUNCDECL     (1 << 22)       /* 'function' 'func' */
> ++#define TC_BEGIN        (1 << 23)       /* 'BEGIN' */
> ++#define TC_END          (1 << 24)       /* 'END' */
> ++#define TC_EOF          (1 << 25)
> ++#define TC_VARIABLE     (1 << 26)       /* name */
> ++#define TC_ARRAY        (1 << 27)       /* name[ */
> ++#define TC_FUNCTION     (1 << 28)       /* name( */
> ++#define TC_STRING       (1 << 29)       /* "..." */
> ++#define TC_NUMBER       (1 << 30)
> ++
> ++#ifndef debug_parse_print_tc
> ++static void debug_parse_print_tc(uint32_t n)
> ++{
> ++      if (n & TC_LPAREN  ) debug_printf_parse(" LPAREN"  );
> ++      if (n & TC_RPAREN  ) debug_printf_parse(" RPAREN"  );
> ++      if (n & TC_REGEXP  ) debug_printf_parse(" REGEXP"  );
> ++      if (n & TC_OUTRDR  ) debug_printf_parse(" OUTRDR"  );
> ++      if (n & TC_UOPPOST ) debug_printf_parse(" UOPPOST" );
> ++      if (n & TC_UOPPRE1 ) debug_printf_parse(" UOPPRE1" );
> ++      if (n & TC_BINOPX  ) debug_printf_parse(" BINOPX"  );
> ++      if (n & TC_IN      ) debug_printf_parse(" IN"      );
> ++      if (n & TC_COMMA   ) debug_printf_parse(" COMMA"   );
> ++      if (n & TC_PIPE    ) debug_printf_parse(" PIPE"    );
> ++      if (n & TC_UOPPRE2 ) debug_printf_parse(" UOPPRE2" );
> ++      if (n & TC_ARRTERM ) debug_printf_parse(" ARRTERM" );
> ++      if (n & TC_LBRACE  ) debug_printf_parse(" LBRACE"  );
> ++      if (n & TC_RBRACE  ) debug_printf_parse(" RBRACE"  );
> ++      if (n & TC_SEMICOL ) debug_printf_parse(" SEMICOL" );
> ++      if (n & TC_NEWLINE ) debug_printf_parse(" NEWLINE" );
> ++      if (n & TC_STATX   ) debug_printf_parse(" STATX"   );
> ++      if (n & TC_WHILE   ) debug_printf_parse(" WHILE"   );
> ++      if (n & TC_ELSE    ) debug_printf_parse(" ELSE"    );
> ++      if (n & TC_BUILTIN ) debug_printf_parse(" BUILTIN" );
> ++      if (n & TC_LENGTH  ) debug_printf_parse(" LENGTH"  );
> ++      if (n & TC_GETLINE ) debug_printf_parse(" GETLINE" );
> ++      if (n & TC_FUNCDECL) debug_printf_parse(" FUNCDECL");
> ++      if (n & TC_BEGIN   ) debug_printf_parse(" BEGIN"   );
> ++      if (n & TC_END     ) debug_printf_parse(" END"     );
> ++      if (n & TC_EOF     ) debug_printf_parse(" EOF"     );
> ++      if (n & TC_VARIABLE) debug_printf_parse(" VARIABLE");
> ++      if (n & TC_ARRAY   ) debug_printf_parse(" ARRAY"   );
> ++      if (n & TC_FUNCTION) debug_printf_parse(" FUNCTION");
> ++      if (n & TC_STRING  ) debug_printf_parse(" STRING"  );
> ++      if (n & TC_NUMBER  ) debug_printf_parse(" NUMBER"  );
> ++}
> ++#endif
> ++
> ++/* combined token classes ("token [class] sets") */
> ++#define       TS_UOPPRE   (TC_UOPPRE1 | TC_UOPPRE2)
> ++
> ++#define       TS_BINOP    (TC_BINOPX | TC_COMMA | TC_PIPE | TC_IN)
> ++//#define TS_UNARYOP (TS_UOPPRE | TC_UOPPOST)
> ++#define       TS_OPERAND  (TC_VARIABLE | TC_ARRAY | TC_FUNCTION \
> ++                    | TC_BUILTIN | TC_LENGTH | TC_GETLINE \
> ++                    | TC_LPAREN | TC_STRING | TC_NUMBER)
> ++
> ++#define       TS_LVALUE   (TC_VARIABLE | TC_ARRAY)
> ++#define       TS_STATEMNT (TC_STATX | TC_WHILE)
> + 
> + /* word tokens, cannot mean something else if not expected */
> +-#define       TC_WORD    (TC_IN | TC_STATEMNT | TC_ELSE \
> +-                   | TC_BUILTIN | TC_LENGTH | TC_GETLINE \
> +-                   | TC_FUNCDECL | TC_BEGIN | TC_END)
> ++#define       TS_WORD     (TC_IN | TS_STATEMNT | TC_ELSE \
> ++                    | TC_BUILTIN | TC_LENGTH | TC_GETLINE \
> ++                    | TC_FUNCDECL | TC_BEGIN | TC_END)
> + 
> + /* discard newlines after these */
> +-#define       TC_NOTERM  (TC_COMMA | TC_GRPSTART | TC_GRPTERM \
> +-                   | TC_BINOP | TC_OPTERM)
> ++#define       TS_NOTERM   (TS_BINOP | TC_COMMA | TC_LBRACE |
> TC_RBRACE \
> ++                    | TC_SEMICOL | TC_NEWLINE)
> + 
> + /* what can expression begin with */
> +-#define       TC_OPSEQ   (TC_OPERAND | TC_UOPPRE | TC_REGEXP)
> ++#define       TS_OPSEQ    (TS_OPERAND | TS_UOPPRE | TC_REGEXP)
> + /* what can group begin with */
> +-#define       TC_GRPSEQ  (TC_OPSEQ | TC_OPTERM | TC_STATEMNT |
> TC_GRPSTART)
> ++#define       TS_GRPSEQ   (TS_OPSEQ | TS_STATEMNT \
> ++                    | TC_SEMICOL | TC_NEWLINE | TC_LBRACE)
> + 
> +-/* if previous token class is CONCAT1 and next is CONCAT2,
> concatenation */
> ++/* if previous token class is CONCAT_L and next is CONCAT_R,
> concatenation */
> + /* operator is inserted between them */
> +-#define       TC_CONCAT1 (TC_VARIABLE | TC_ARRTERM | TC_SEQTERM \
> ++#define       TS_CONCAT_L (TC_VARIABLE | TC_ARRTERM | TC_RPAREN \
> +                    | TC_STRING | TC_NUMBER | TC_UOPPOST \
> +                    | TC_LENGTH)
> +-#define       TC_CONCAT2 (TC_OPERAND | TC_UOPPRE)
> ++#define       TS_CONCAT_R (TS_OPERAND | TS_UOPPRE)
> + 
> + #define       OF_RES1     0x010000
> + #define       OF_RES2     0x020000
> +@@ -284,13 +315,12 @@ typedef struct tsplitter_s {
> + #define       OF_CHECKED  0x200000
> + #define       OF_REQUIRED 0x400000
> + 
> +-
> + /* combined operator flags */
> + #define       xx      0
> + #define       xV      OF_RES2
> + #define       xS      (OF_RES2 | OF_STR2)
> + #define       Vx      OF_RES1
> +-#define       Rx      (OF_RES1 | OF_NUM1 | OF_REQUIRED)
> ++#define       Rx      OF_REQUIRED
> + #define       VV      (OF_RES1 | OF_RES2)
> + #define       Nx      (OF_RES1 | OF_NUM1)
> + #define       NV      (OF_RES1 | OF_NUM1 | OF_RES2)
> +@@ -302,8 +332,7 @@ typedef struct tsplitter_s {
> + #define       OPNMASK   0x007F
> + 
> + /* operator priority is a highest byte (even: r->l, odd: l->r
> grouping)
> +- * For builtins it has different meaning: n n s3 s2 s1 v3 v2 v1,
> +- * n - min. number of args, vN - resolve Nth arg to var, sN -
> resolve to string
> ++ * (for builtins it has different meaning)
> +  */
> + #undef P
> + #undef PRIMASK
> +@@ -313,10 +342,8 @@ typedef struct tsplitter_s {
> + #define PRIMASK2  0x7E000000
> + 
> + /* Operation classes */
> +-
> + #define       SHIFT_TIL_THIS  0x0600
> + #define       RECUR_FROM_THIS 0x1000
> +-
> + enum {
> +       OC_DELETE = 0x0100,     OC_EXEC = 0x0200,       OC_NEWSOURCE
> = 0x0300,
> +       OC_PRINT = 0x0400,      OC_PRINTF = 0x0500,     OC_WALKINIT =
> 0x0600,
> +@@ -358,8 +385,8 @@ enum {
> + #define NTCC    '\377'
> + 
> + static const char tokenlist[] ALIGN1 =
> +-      "\1("         NTC                                   /*
> TC_SEQSTART */
> +-      "\1)"         NTC                                   /*
> TC_SEQTERM */
> ++      "\1("         NTC                                   /*
> TC_LPAREN */
> ++      "\1)"         NTC                                   /*
> TC_RPAREN */
> +       "\1/"         NTC                                   /*
> TC_REGEXP */
> +       "\2>>"        "\1>"         "\1|"       NTC         /*
> TC_OUTRDR */
> +       "\2++"        "\2--"        NTC                     /*
> TC_UOPPOST */
> +@@ -376,8 +403,8 @@ static const char tokenlist[] ALIGN1 =
> +       "\1|"         NTC                                   /*
> TC_PIPE */
> +       "\1+"         "\1-"         "\1!"       NTC         /*
> TC_UOPPRE2 */
> +       "\1]"         NTC                                   /*
> TC_ARRTERM */
> +-      "\1{"         NTC                                   /*
> TC_GRPSTART */
> +-      "\1}"         NTC                                   /*
> TC_GRPTERM */
> ++      "\1{"         NTC                                   /*
> TC_LBRACE */
> ++      "\1}"         NTC                                   /*
> TC_RBRACE */
> +       "\1;"         NTC                                   /*
> TC_SEMICOL */
> +       "\1\n"        NTC                                   /*
> TC_NEWLINE */
> +       "\2if"        "\2do"        "\3for"     "\5break"   /*
> TC_STATX */
> +@@ -391,7 +418,7 @@ static const char tokenlist[] ALIGN1 =
> +       "\5close"     "\6system"    "\6fflush"  "\5atan2"
> +       "\3cos"       "\3exp"       "\3int"     "\3log"
> +       "\4rand"      "\3sin"       "\4sqrt"    "\5srand"
> +-      "\6gensub"    "\4gsub"      "\5index"   /* "\6length" was
> here */
> ++      "\6gensub"    "\4gsub"      "\5index"   /* "\6length" was
> here */
> +       "\5match"     "\5split"     "\7sprintf" "\3sub"
> +       "\6substr"    "\7systime"   "\10strftime" "\6mktime"
> +       "\7tolower"   "\7toupper"   NTC
> +@@ -403,25 +430,32 @@ static const char tokenlist[] ALIGN1 =
> +       /* compiler adds trailing "\0" */
> +       ;
> + 
> +-#define OC_B  OC_BUILTIN
> +-
> + static const uint32_t tokeninfo[] ALIGN4 = {
> +       0,
> +       0,
> +-      OC_REGEXP,
> ++#define TI_REGEXP OC_REGEXP
> ++      TI_REGEXP,
> +       xS|'a',                  xS|'w',                  xS|'|',
> +       OC_UNARY|xV|P(9)|'p',    OC_UNARY|xV|P(9)|'m',
> +-      OC_UNARY|xV|P(9)|'P',    OC_UNARY|xV|P(9)|'M',   
> OC_FIELD|xV|P(5),
> ++#define TI_PREINC (OC_UNARY|xV|P(9)|'P')
> ++#define TI_PREDEC (OC_UNARY|xV|P(9)|'M')
> ++      TI_PREINC,               TI_PREDEC,              
> OC_FIELD|xV|P(5),
> +       OC_COMPARE|VV|P(39)|5,   OC_MOVE|VV|P(74),       
> OC_REPLACE|NV|P(74)|'+', OC_REPLACE|NV|P(74)|'-',
> +       OC_REPLACE|NV|P(74)|'*', OC_REPLACE|NV|P(74)|'/',
> OC_REPLACE|NV|P(74)|'%', OC_REPLACE|NV|P(74)|'&',
> +       OC_BINARY|NV|P(29)|'+',  OC_BINARY|NV|P(29)|'-', 
> OC_REPLACE|NV|P(74)|'&', OC_BINARY|NV|P(15)|'&',
> +       OC_BINARY|NV|P(25)|'/',  OC_BINARY|NV|P(25)|'%', 
> OC_BINARY|NV|P(15)|'&',  OC_BINARY|NV|P(25)|'*',
> +       OC_COMPARE|VV|P(39)|4,   OC_COMPARE|VV|P(39)|3,  
> OC_COMPARE|VV|P(39)|0,   OC_COMPARE|VV|P(39)|1,
> +-      OC_COMPARE|VV|P(39)|2,   OC_MATCH|Sx|P(45)|'!',  
> OC_MATCH|Sx|P(45)|'~',   OC_LAND|Vx|P(55),
> +-      OC_LOR|Vx|P(59),         OC_TERNARY|Vx|P(64)|'?',
> OC_COLON|xx|P(67)|':',
> +-      OC_IN|SV|P(49), /* TC_IN */
> +-      OC_COMMA|SS|P(80),
> +-      OC_PGETLINE|SV|P(37),
> ++#define TI_LESS     (OC_COMPARE|VV|P(39)|2)
> ++      TI_LESS,                 OC_MATCH|Sx|P(45)|'!',  
> OC_MATCH|Sx|P(45)|'~',   OC_LAND|Vx|P(55),
> ++#define TI_TERNARY  (OC_TERNARY|Vx|P(64)|'?')
> ++#define TI_COLON    (OC_COLON|xx|P(67)|':')
> ++      OC_LOR|Vx|P(59),         TI_TERNARY,              TI_COLON,
> ++#define TI_IN       (OC_IN|SV|P(49))
> ++      TI_IN,
> ++#define TI_COMMA    (OC_COMMA|SS|P(80))
> ++      TI_COMMA,
> ++#define TI_PGETLINE (OC_PGETLINE|SV|P(37))
> ++      TI_PGETLINE,
> +       OC_UNARY|xV|P(19)|'+',   OC_UNARY|xV|P(19)|'-',  
> OC_UNARY|xV|P(19)|'!',
> +       0, /* ] */
> +       0,
> +@@ -434,20 +468,45 @@ static const uint32_t tokeninfo[] ALIGN4 = {
> +       OC_RETURN|Vx, OC_EXIT|Nx,
> +       ST_WHILE,
> +       0, /* else */
> +-      OC_B|B_an|P(0x83), OC_B|B_co|P(0x41), OC_B|B_ls|P(0x83),
> OC_B|B_or|P(0x83),
> +-      OC_B|B_rs|P(0x83), OC_B|B_xo|P(0x83),
> +-      OC_FBLTIN|Sx|F_cl, OC_FBLTIN|Sx|F_sy, OC_FBLTIN|Sx|F_ff,
> OC_B|B_a2|P(0x83),
> +-      OC_FBLTIN|Nx|F_co, OC_FBLTIN|Nx|F_ex, OC_FBLTIN|Nx|F_in,
> OC_FBLTIN|Nx|F_lg,
> +-      OC_FBLTIN|F_rn,    OC_FBLTIN|Nx|F_si, OC_FBLTIN|Nx|F_sq,
> OC_FBLTIN|Nx|F_sr,
> +-      OC_B|B_ge|P(0xd6), OC_B|B_gs|P(0xb6), OC_B|B_ix|P(0x9b), /*
> OC_FBLTIN|Sx|F_le, was here */
> +-      OC_B|B_ma|P(0x89), OC_B|B_sp|P(0x8b), OC_SPRINTF,       
> OC_B|B_su|P(0xb6),
> +-      OC_B|B_ss|P(0x8f), OC_FBLTIN|F_ti,    OC_B|B_ti|P(0x0b),
> OC_B|B_mt|P(0x0b),
> +-      OC_B|B_lo|P(0x49), OC_B|B_up|P(0x49),
> +-      OC_FBLTIN|Sx|F_le, /* TC_LENGTH */
> +-      OC_GETLINE|SV|P(0),
> +-      0,                 0,
> +-      0,
> +-      0 /* TC_END */
> ++// OC_B's are builtins with enforced minimum number of arguments
> (two upper bits).
> ++//  Highest byte bit pattern: nn s3s2s1 v3v2v1
> ++//  nn - min. number of args, sN - resolve Nth arg to string, vN -
> resolve to var
> ++// OC_F's are builtins with zero or one argument.
> ++//  |Rx| enforces that arg is present for: system, close, cos, sin,
> exp, int, log, sqrt
> ++//  Check for no args is present in builtins' code (not in this
> table): rand, systime
> ++//  Have one _optional_ arg: fflush, srand, length
> ++#define OC_B   OC_BUILTIN
> ++#define OC_F   OC_FBLTIN
> ++#define A1     P(0x40) /*one arg*/
> ++#define A2     P(0x80) /*two args*/
> ++#define A3     P(0xc0) /*three args*/
> ++#define __v    P(1)
> ++#define _vv    P(3)
> ++#define __s__v P(9)
> ++#define __s_vv P(0x0b)
> ++#define __svvv P(0x0f)
> ++#define _ss_vv P(0x1b)
> ++#define _s_vv_ P(0x16)
> ++#define ss_vv_ P(0x36)
> ++      OC_B|B_an|_vv|A2,   OC_B|B_co|__v|A1,   OC_B|B_ls|_vv|A2,  
> OC_B|B_or|_vv|A2,   // and    compl   lshift   or
> ++      OC_B|B_rs|_vv|A2,  
> OC_B|B_xo|_vv|A2,                                           // rshift
> xor
> ++      OC_F|F_cl|Sx|Rx,    OC_F|F_sy|Sx|Rx,    OC_F|F_ff|Sx,      
> OC_B|B_a2|_vv|A2,   // close  system  fflush   atan2
> ++      OC_F|F_co|Nx|Rx,    OC_F|F_ex|Nx|Rx,    OC_F|F_in|Nx|Rx,   
> OC_F|F_lg|Nx|Rx,    // cos    exp     int      log
> ++      OC_F|F_rn,          OC_F|F_si|Nx|Rx,    OC_F|F_sq|Nx|Rx,   
> OC_F|F_sr|Nx,       // rand   sin     sqrt     srand
> ++      OC_B|B_ge|_s_vv_|A3,OC_B|B_gs|ss_vv_|A2,OC_B|B_ix|_ss_vv|A2, 
>                    // gensub gsub    index  /*length was here*/
> ++      OC_B|B_ma|__s__v|A2,OC_B|B_sp|__s_vv|A2,OC_SPRINTF,        
> OC_B|B_su|ss_vv_|A2,// match  split   sprintf  sub
> ++      OC_B|B_ss|__svvv|A2,OC_F|F_ti,          OC_B|B_ti|__s_vv,  
> OC_B|B_mt|__s_vv,   // substr systime strftime mktime
> ++      OC_B|B_lo|__s__v|A1,OC_B|B_up|__s__v|A1,                     
>                    // tolower toupper
> ++      OC_F|F_le|Sx,   // length
> ++      OC_GETLINE|SV,  // getline
> ++      0, 0, // func function
> ++      0, // BEGIN
> ++      0  // END
> ++#undef A1
> ++#undef A2
> ++#undef A3
> ++#undef OC_B
> ++#undef OC_F
> + };
> + 
> + /* internal variable names and their initial values       */
> +@@ -488,21 +547,29 @@ struct globals {
> +       chain *seq;
> +       node *break_ptr, *continue_ptr;
> +       rstream *iF;
> +-      xhash *vhash, *ahash, *fdhash, *fnhash;
> ++      xhash *ahash;  /* argument names, used only while parsing
> function bodies */
> ++      xhash *fnhash; /* function names, used only in parsing stage
> */
> ++      xhash *vhash;  /* variables and arrays */
> ++      //xhash *fdhash; /* file objects, used only in execution
> stage */
> ++      //we are reusing ahash as fdhash, via define (see later)
> +       const char *g_progname;
> +       int g_lineno;
> +       int nfields;
> +       int maxfields; /* used in fsrealloc() only */
> +       var *Fields;
> +-      nvblock *g_cb;
> +       char *g_pos;
> +-      char *g_buf;
> ++      char g_saved_ch;
> +       smallint icase;
> +       smallint exiting;
> +       smallint nextrec;
> +       smallint nextfile;
> +       smallint is_f0_split;
> +       smallint t_rollback;
> ++
> ++      /* former statics from various functions */
> ++      smallint next_token__concat_inserted;
> ++      uint32_t next_token__save_tclass;
> ++      uint32_t next_token__save_info;
> + };
> + struct globals2 {
> +       uint32_t t_info; /* often used */
> +@@ -515,32 +582,35 @@ struct globals2 {
> +       /* former statics from various functions */
> +       char *split_f0__fstrings;
> + 
> +-      uint32_t next_token__save_tclass;
> +-      uint32_t next_token__save_info;
> +-      uint32_t next_token__ltclass;
> +-      smallint next_token__concat_inserted;
> +-
> +-      smallint next_input_file__files_happen;
> +       rstream next_input_file__rsm;
> ++      smallint next_input_file__files_happen;
> ++
> ++      smalluint exitcode;
> + 
> +-      var *evaluate__fnargs;
> +       unsigned evaluate__seed;
> ++      var *evaluate__fnargs;
> +       regex_t evaluate__sreg;
> + 
> +-      var ptest__v;
> ++      var ptest__tmpvar;
> ++      var awk_printf__tmpvar;
> ++      var as_regex__tmpvar;
> ++      var exit__tmpvar;
> ++      var main__tmpvar;
> + 
> +       tsplitter exec_builtin__tspl;
> + 
> +       /* biggest and least used members go last */
> +       tsplitter fsplitter, rsplitter;
> ++
> ++      char g_buf[MAXVARFMT + 1];
> + };
> + #define G1 (ptr_to_globals[-1])
> + #define G (*(struct globals2 *)ptr_to_globals)
> + /* For debug. nm --size-sort awk.o | grep -vi ' [tr] ' */
> +-/*char G1size[sizeof(G1)]; - 0x74 */
> +-/*char Gsize[sizeof(G)]; - 0x1c4 */
> ++//char G1size[sizeof(G1)]; // 0x70
> ++//char Gsize[sizeof(G)]; // 0x2f8
> + /* Trying to keep most of members accessible with short offsets: */
> +-/*char Gofs_seed[offsetof(struct globals2, evaluate__seed)]; - 0x90
> */
> ++//char Gofs_seed[offsetof(struct globals2, evaluate__seed)]; //
> 0x7c
> + #define t_double     (G1.t_double    )
> + #define beginseq     (G1.beginseq    )
> + #define mainseq      (G1.mainseq     )
> +@@ -549,18 +619,20 @@ struct globals2 {
> + #define break_ptr    (G1.break_ptr   )
> + #define continue_ptr (G1.continue_ptr)
> + #define iF           (G1.iF          )
> +-#define vhash        (G1.vhash       )
> + #define ahash        (G1.ahash       )
> +-#define fdhash       (G1.fdhash      )
> + #define fnhash       (G1.fnhash      )
> ++#define vhash        (G1.vhash       )
> ++#define fdhash       ahash
> ++//^^^^^^^^^^^^^^^^^^ ahash is cleared after every function parsing,
> ++// and ends up empty after parsing phase. Thus, we can simply reuse
> it
> ++// for fdhash in execution stage.
> + #define g_progname   (G1.g_progname  )
> + #define g_lineno     (G1.g_lineno    )
> + #define nfields      (G1.nfields     )
> + #define maxfields    (G1.maxfields   )
> + #define Fields       (G1.Fields      )
> +-#define g_cb         (G1.g_cb        )
> + #define g_pos        (G1.g_pos       )
> +-#define g_buf        (G1.g_buf       )
> ++#define g_saved_ch   (G1.g_saved_ch  )
> + #define icase        (G1.icase       )
> + #define exiting      (G1.exiting     )
> + #define nextrec      (G1.nextrec     )
> +@@ -574,25 +646,13 @@ struct globals2 {
> + #define intvar       (G.intvar      )
> + #define fsplitter    (G.fsplitter   )
> + #define rsplitter    (G.rsplitter   )
> ++#define g_buf        (G.g_buf       )
> + #define INIT_G() do { \
> +       SET_PTR_TO_GLOBALS((char*)xzalloc(sizeof(G1)+sizeof(G)) +
> sizeof(G1)); \
> +-      G.next_token__ltclass = TC_OPTERM; \
> ++      t_tclass = TC_NEWLINE; \
> +       G.evaluate__seed = 1; \
> + } while (0)
> + 
> +-
> +-/* function prototypes */
> +-static void handle_special(var *);
> +-static node *parse_expr(uint32_t);
> +-static void chain_group(void);
> +-static var *evaluate(node *, var *);
> +-static rstream *next_input_file(void);
> +-static int fmt_num(char *, int, const char *, double, int);
> +-static int awk_exit(int) NORETURN;
> +-
> +-/* ---- error handling ---- */
> +-
> +-static const char EMSG_INTERNAL_ERROR[] ALIGN1 = "Internal error";
> + static const char EMSG_UNEXP_EOS[] ALIGN1 = "Unexpected end of
> string";
> + static const char EMSG_UNEXP_TOKEN[] ALIGN1 = "Unexpected token";
> + static const char EMSG_DIV_BY_ZERO[] ALIGN1 = "Division by zero";
> +@@ -604,10 +664,7 @@ static const char EMSG_UNDEF_FUNC[] ALIGN1 =
> "Call to undefined function";
> + static const char EMSG_NO_MATH[] ALIGN1 = "Math support is not
> compiled in";
> + static const char EMSG_NEGATIVE_FIELD[] ALIGN1 = "Access to
> negative field";
> + 
> +-static void zero_out_var(var *vp)
> +-{
> +-      memset(vp, 0, sizeof(*vp));
> +-}
> ++static int awk_exit(void) NORETURN;
> + 
> + static void syntax_error(const char *message) NORETURN;
> + static void syntax_error(const char *message)
> +@@ -638,12 +695,40 @@ static xhash *hash_init(void)
> +       return newhash;
> + }
> + 
> ++static void hash_clear(xhash *hash)
> ++{
> ++      unsigned i;
> ++      hash_item *hi, *thi;
> ++
> ++      for (i = 0; i < hash->csize; i++) {
> ++              hi = hash->items[i];
> ++              while (hi) {
> ++                      thi = hi;
> ++                      hi = hi->next;
> ++//FIXME: this assumes that it's a hash of *variables*:
> ++                      free(thi->data.v.string);
> ++                      free(thi);
> ++              }
> ++              hash->items[i] = NULL;
> ++      }
> ++      hash->glen = hash->nel = 0;
> ++}
> ++
> ++#if 0 //UNUSED
> ++static void hash_free(xhash *hash)
> ++{
> ++      hash_clear(hash);
> ++      free(hash->items);
> ++      free(hash);
> ++}
> ++#endif
> ++
> + /* find item in hash, return ptr to data, NULL if not found */
> +-static void *hash_search(xhash *hash, const char *name)
> ++static NOINLINE void *hash_search3(xhash *hash, const char *name,
> unsigned idx)
> + {
> +       hash_item *hi;
> + 
> +-      hi = hash->items[hashidx(name) % hash->csize];
> ++      hi = hash->items[idx % hash->csize];
> +       while (hi) {
> +               if (strcmp(hi->name, name) == 0)
> +                       return &hi->data;
> +@@ -652,6 +737,11 @@ static void *hash_search(xhash *hash, const
> char *name)
> +       return NULL;
> + }
> + 
> ++static void *hash_search(xhash *hash, const char *name)
> ++{
> ++      return hash_search3(hash, name, hashidx(name));
> ++}
> ++
> + /* grow hash if it becomes too big */
> + static void hash_rebuild(xhash *hash)
> + {
> +@@ -687,16 +777,17 @@ static void *hash_find(xhash *hash, const char
> *name)
> +       unsigned idx;
> +       int l;
> + 
> +-      hi = hash_search(hash, name);
> ++      idx = hashidx(name);
> ++      hi = hash_search3(hash, name, idx);
> +       if (!hi) {
> +-              if (++hash->nel / hash->csize > 10)
> ++              if (++hash->nel > hash->csize * 8)
> +                       hash_rebuild(hash);
> + 
> +               l = strlen(name) + 1;
> +               hi = xzalloc(sizeof(*hi) + l);
> +               strcpy(hi->name, name);
> + 
> +-              idx = hashidx(name) % hash->csize;
> ++              idx = idx % hash->csize;
> +               hi->next = hash->items[idx];
> +               hash->items[idx] = hi;
> +               hash->glen += l;
> +@@ -731,7 +822,7 @@ static void hash_remove(xhash *hash, const char
> *name)
> + 
> + static char *skip_spaces(char *p)
> + {
> +-      while (1) {
> ++      for (;;) {
> +               if (*p == '\\' && p[1] == '\n') {
> +                       p++;
> +                       t_lineno++;
> +@@ -747,8 +838,10 @@ static char *skip_spaces(char *p)
> + static char *nextword(char **s)
> + {
> +       char *p = *s;
> +-      while (*(*s)++ != '\0')
> ++      char *q = p;
> ++      while (*q++ != '\0')
> +               continue;
> ++      *s = q;
> +       return p;
> + }
> + 
> +@@ -811,10 +904,27 @@ static double my_strtod(char **pp)
> + 
> + /* -------- working with variables (set/get/copy/etc) -------- */
> + 
> +-static xhash *iamarray(var *v)
> ++static void fmt_num(const char *format, double n)
> + {
> +-      var *a = v;
> ++      if (n == (long long)n) {
> ++              snprintf(g_buf, MAXVARFMT, "%lld", (long long)n);
> ++      } else {
> ++              const char *s = format;
> ++              char c;
> ++
> ++              do { c = *s; } while (c && *++s);
> ++              if (strchr("diouxX", c)) {
> ++                      snprintf(g_buf, MAXVARFMT, format, (int)n);
> ++              } else if (strchr("eEfFgGaA", c)) {
> ++                      snprintf(g_buf, MAXVARFMT, format, n);
> ++              } else {
> ++                      syntax_error(EMSG_INV_FMT);
> ++              }
> ++      }
> ++}
> + 
> ++static xhash *iamarray(var *a)
> ++{
> +       while (a->type & VF_CHILD)
> +               a = a->x.parent;
> + 
> +@@ -825,23 +935,7 @@ static xhash *iamarray(var *v)
> +       return a->x.array;
> + }
> + 
> +-static void clear_array(xhash *array)
> +-{
> +-      unsigned i;
> +-      hash_item *hi, *thi;
> +-
> +-      for (i = 0; i < array->csize; i++) {
> +-              hi = array->items[i];
> +-              while (hi) {
> +-                      thi = hi;
> +-                      hi = hi->next;
> +-                      free(thi->data.v.string);
> +-                      free(thi);
> +-              }
> +-              array->items[i] = NULL;
> +-      }
> +-      array->glen = array->nel = 0;
> +-}
> ++#define clear_array(array) hash_clear(array)
> + 
> + /* clear a variable */
> + static var *clrvar(var *v)
> +@@ -855,6 +949,8 @@ static var *clrvar(var *v)
> +       return v;
> + }
> + 
> ++static void handle_special(var *);
> ++
> + /* assign string value to variable */
> + static var *setvar_p(var *v, char *value)
> + {
> +@@ -901,7 +997,7 @@ static const char *getvar_s(var *v)
> + {
> +       /* if v is numeric and has no cached string, convert it to
> string */
> +       if ((v->type & (VF_NUMBER | VF_CACHED)) == VF_NUMBER) {
> +-              fmt_num(g_buf, MAXVARFMT, getvar_s(intvar[CONVFMT]),
> v->number, TRUE);
> ++              fmt_num(getvar_s(intvar[CONVFMT]), v->number);
> +               v->string = xstrdup(g_buf);
> +               v->type |= VF_CACHED;
> +       }
> +@@ -920,6 +1016,7 @@ static double getvar_i(var *v)
> +                       v->number = my_strtod(&s);
> +                       debug_printf_eval("%f (s:'%s')\n", v->number,
> s);
> +                       if (v->type & VF_USER) {
> ++//TODO: skip_spaces() also skips backslash+newline, is it intended
> here?
> +                               s = skip_spaces(s);
> +                               if (*s != '\0')
> +                                       v->type &= ~VF_USER;
> +@@ -981,94 +1078,28 @@ static int istrue(var *v)
> +       return (v->string && v->string[0]);
> + }
> + 
> +-/* temporary variables allocator. Last allocated should be first
> freed */
> +-static var *nvalloc(int n)
> +-{
> +-      nvblock *pb = NULL;
> +-      var *v, *r;
> +-      int size;
> +-
> +-      while (g_cb) {
> +-              pb = g_cb;
> +-              if ((g_cb->pos - g_cb->nv) + n <= g_cb->size)
> +-                      break;
> +-              g_cb = g_cb->next;
> +-      }
> +-
> +-      if (!g_cb) {
> +-              size = (n <= MINNVBLOCK) ? MINNVBLOCK : n;
> +-              g_cb = xzalloc(sizeof(nvblock) + size * sizeof(var));
> +-              g_cb->size = size;
> +-              g_cb->pos = g_cb->nv;
> +-              g_cb->prev = pb;
> +-              /*g_cb->next = NULL; - xzalloc did it */
> +-              if (pb)
> +-                      pb->next = g_cb;
> +-      }
> +-
> +-      v = r = g_cb->pos;
> +-      g_cb->pos += n;
> +-
> +-      while (v < g_cb->pos) {
> +-              v->type = 0;
> +-              v->string = NULL;
> +-              v++;
> +-      }
> +-
> +-      return r;
> +-}
> +-
> +-static void nvfree(var *v)
> +-{
> +-      var *p;
> +-
> +-      if (v < g_cb->nv || v >= g_cb->pos)
> +-              syntax_error(EMSG_INTERNAL_ERROR);
> +-
> +-      for (p = v; p < g_cb->pos; p++) {
> +-              if ((p->type & (VF_ARRAY | VF_CHILD)) == VF_ARRAY) {
> +-                      clear_array(iamarray(p));
> +-                      free(p->x.array->items);
> +-                      free(p->x.array);
> +-              }
> +-              if (p->type & VF_WALK) {
> +-                      walker_list *n;
> +-                      walker_list *w = p->x.walker;
> +-                      debug_printf_walker("nvfree: freeing walker
> @%p\n", &p->x.walker);
> +-                      p->x.walker = NULL;
> +-                      while (w) {
> +-                              n = w->prev;
> +-                              debug_printf_walker(" free(%p)\n",
> w);
> +-                              free(w);
> +-                              w = n;
> +-                      }
> +-              }
> +-              clrvar(p);
> +-      }
> +-
> +-      g_cb->pos = v;
> +-      while (g_cb->prev && g_cb->pos == g_cb->nv) {
> +-              g_cb = g_cb->prev;
> +-      }
> +-}
> +-
> + /* ------- awk program text parsing ------- */
> + 
> +-/* Parse next token pointed by global pos, place results into
> global ttt.
> +- * If token isn't expected, give away. Return token class
> ++/* Parse next token pointed by global pos, place results into
> global t_XYZ variables.
> ++ * If token isn't expected, print error message and die.
> ++ * Return token class (also store it in t_tclass).
> +  */
> + static uint32_t next_token(uint32_t expected)
> + {
> +-#define concat_inserted (G.next_token__concat_inserted)
> +-#define save_tclass     (G.next_token__save_tclass)
> +-#define save_info       (G.next_token__save_info)
> +-/* Initialized to TC_OPTERM: */
> +-#define ltclass         (G.next_token__ltclass)
> ++#define concat_inserted (G1.next_token__concat_inserted)
> ++#define save_tclass     (G1.next_token__save_tclass)
> ++#define save_info       (G1.next_token__save_info)
> + 
> +-      char *p, *s;
> ++      char *p;
> +       const char *tl;
> +-      uint32_t tc;
> +       const uint32_t *ti;
> ++      uint32_t tc, last_token_class;
> ++
> ++      last_token_class = t_tclass; /* t_tclass is initialized to
> TC_NEWLINE */
> ++
> ++      debug_printf_parse("%s() expected(%x):", __func__, expected);
> ++      debug_parse_print_tc(expected);
> ++      debug_printf_parse("\n");
> + 
> +       if (t_rollback) {
> +               debug_printf_parse("%s: using rolled-back token\n",
> __func__);
> +@@ -1080,6 +1111,10 @@ static uint32_t next_token(uint32_t expected)
> +               t_info = save_info;
> +       } else {
> +               p = g_pos;
> ++              if (g_saved_ch != '\0') {
> ++                      *p = g_saved_ch;
> ++                      g_saved_ch = '\0';
> ++              }
> +  readnext:
> +               p = skip_spaces(p);
> +               g_lineno = t_lineno;
> +@@ -1087,15 +1122,12 @@ static uint32_t next_token(uint32_t
> expected)
> +                       while (*p != '\n' && *p != '\0')
> +                               p++;
> + 
> +-              if (*p == '\n')
> +-                      t_lineno++;
> +-
> +               if (*p == '\0') {
> +                       tc = TC_EOF;
> +                       debug_printf_parse("%s: token found:
> TC_EOF\n", __func__);
> +               } else if (*p == '\"') {
> +                       /* it's a string */
> +-                      t_string = s = ++p;
> ++                      char *s = t_string = ++p;
> +                       while (*p != '\"') {
> +                               char *pp;
> +                               if (*p == '\0' || *p == '\n')
> +@@ -1110,7 +1142,7 @@ static uint32_t next_token(uint32_t expected)
> +                       debug_printf_parse("%s: token found:'%s'
> TC_STRING\n", __func__, t_string);
> +               } else if ((expected & TC_REGEXP) && *p == '/') {
> +                       /* it's regexp */
> +-                      t_string = s = ++p;
> ++                      char *s = t_string = ++p;
> +                       while (*p != '/') {
> +                               if (*p == '\0' || *p == '\n')
> +                                       syntax_error(EMSG_UNEXP_EOS);
> +@@ -1141,6 +1173,11 @@ static uint32_t next_token(uint32_t expected)
> +                       tc = TC_NUMBER;
> +                       debug_printf_parse("%s: token found:%f
> TC_NUMBER\n", __func__, t_double);
> +               } else {
> ++                      char *end_of_name;
> ++
> ++                      if (*p == '\n')
> ++                              t_lineno++;
> ++
> +                       /* search for something known */
> +                       tl = tokenlist;
> +                       tc = 0x00000001;
> +@@ -1155,9 +1192,9 @@ static uint32_t next_token(uint32_t expected)
> +                                * token matches,
> +                                * and it's not a longer word,
> +                                */
> +-                              if ((tc & (expected | TC_WORD |
> TC_NEWLINE))
> ++                              if ((tc & (expected | TS_WORD |
> TC_NEWLINE))
> +                                && strncmp(p, tl, l) == 0
> +-                               && !((tc & TC_WORD) &&
> isalnum_(p[l]))
> ++                               && !((tc & TS_WORD) &&
> isalnum_(p[l]))
> +                               ) {
> +                                       /* then this is what we are
> looking for */
> +                                       t_info = *ti;
> +@@ -1174,67 +1211,94 @@ static uint32_t next_token(uint32_t
> expected)
> +                       if (!isalnum_(*p))
> +                               syntax_error(EMSG_UNEXP_TOKEN); /* no
> */
> +                       /* yes */
> +-                      t_string = --p;
> +-                      while (isalnum_(*++p)) {
> +-                              p[-1] = *p;
> +-                      }
> +-                      p[-1] = '\0';
> +-                      tc = TC_VARIABLE;
> +-                      /* also consume whitespace between
> functionname and bracket */
> +-                      if (!(expected & TC_VARIABLE) || (expected &
> TC_ARRAY))
> ++                      t_string = p;
> ++                      while (isalnum_(*p))
> ++                              p++;
> ++                      end_of_name = p;
> ++
> ++                      if (last_token_class == TC_FUNCDECL)
> ++                              /* eat space in "function FUNC (...)
> {...}" declaration */
> +                               p = skip_spaces(p);
> ++                      else if (expected & TC_ARRAY) {
> ++                              /* eat space between array name and [
> */
> ++                              char *s = skip_spaces(p);
> ++                              if (*s == '[') /* array ref, not just
> a name? */
> ++                                      p = s;
> ++                      }
> ++                      /* else: do NOT consume whitespace after
> variable name!
> ++                       * gawk allows definition "function FUNC (p)
> {...}" - note space,
> ++                       * but disallows the call "FUNC (p)" because
> it isn't one -
> ++                       * expression "v (a)" should NOT be parsed as
> TC_FUNCTION:
> ++                       * it is a valid concatenation if "v" is a
> variable,
> ++                       * not a function name (and type of name is
> not known at parse time).
> ++                       */
> ++
> +                       if (*p == '(') {
> ++                              p++;
> +                               tc = TC_FUNCTION;
> +                               debug_printf_parse("%s: token
> found:'%s' TC_FUNCTION\n", __func__, t_string);
> ++                      } else if (*p == '[') {
> ++                              p++;
> ++                              tc = TC_ARRAY;
> ++                              debug_printf_parse("%s: token
> found:'%s' TC_ARRAY\n", __func__, t_string);
> +                       } else {
> +-                              if (*p == '[') {
> +-                                      p++;
> +-                                      tc = TC_ARRAY;
> +-                                      debug_printf_parse("%s: token
> found:'%s' TC_ARRAY\n", __func__, t_string);
> +-                              } else
> +-                                      debug_printf_parse("%s: token
> found:'%s' TC_VARIABLE\n", __func__, t_string);
> ++                              tc = TC_VARIABLE;
> ++                              debug_printf_parse("%s: token
> found:'%s' TC_VARIABLE\n", __func__, t_string);
> ++                              if (end_of_name == p) {
> ++                                      /* there is no space for
> trailing NUL in t_string!
> ++                                       * We need to save the char
> we are going to NUL.
> ++                                       * (we'll use it in future
> call to next_token())
> ++                                       */
> ++                                      g_saved_ch = *end_of_name;
> ++// especially pathological example is V="abc"; V.2 - it's V
> concatenated to .2
> ++// (it evaluates to "abc0.2"). Because of this case, we can't
> simply cache
> ++// '.' and analyze it later: we also have to *store it back* in
> next
> ++// next_token(), in order to give my_strtod() the undamaged ".2"
> string.
> ++                              }
> +                       }
> ++                      *end_of_name = '\0'; /* terminate t_string */
> +               }
> +  token_found:
> +               g_pos = p;
> + 
> +               /* skipping newlines in some cases */
> +-              if ((ltclass & TC_NOTERM) && (tc & TC_NEWLINE))
> ++              if ((last_token_class & TS_NOTERM) && (tc &
> TC_NEWLINE))
> +                       goto readnext;
> + 
> +               /* insert concatenation operator when needed */
> +-              debug_printf_parse("%s: %x %x %x concat_inserted?\n",
> __func__,
> +-                      (ltclass & TC_CONCAT1), (tc & TC_CONCAT2),
> (expected & TC_BINOP));
> +-              if ((ltclass & TC_CONCAT1) && (tc & TC_CONCAT2) &&
> (expected & TC_BINOP)
> +-               && !(ltclass == TC_LENGTH && tc == TC_SEQSTART) /*
> but not for "length(..." */
> ++              debug_printf_parse("%s: concat_inserted if all
> nonzero: %x %x %x %x\n", __func__,
> ++                      (last_token_class & TS_CONCAT_L), (tc &
> TS_CONCAT_R), (expected & TS_BINOP),
> ++                      !(last_token_class == TC_LENGTH && tc ==
> TC_LPAREN));
> ++              if ((last_token_class & TS_CONCAT_L) && (tc &
> TS_CONCAT_R) && (expected & TS_BINOP)
> ++               && !(last_token_class == TC_LENGTH && tc ==
> TC_LPAREN) /* but not for "length(..." */
> +               ) {
> +                       concat_inserted = TRUE;
> +                       save_tclass = tc;
> +                       save_info = t_info;
> +-                      tc = TC_BINOP;
> ++                      tc = TC_BINOPX;
> +                       t_info = OC_CONCAT | SS | P(35);
> +               }
> + 
> +-              debug_printf_parse("%s: t_tclass=tc=%x\n", __func__,
> t_tclass);
> +               t_tclass = tc;
> ++              debug_printf_parse("%s: t_tclass=tc=%x\n", __func__,
> tc);
> +       }
> +-      ltclass = t_tclass;
> +-
> +       /* Are we ready for this? */
> +-      if (!(ltclass & expected)) {
> +-              syntax_error((ltclass & (TC_NEWLINE | TC_EOF)) ?
> ++      if (!(t_tclass & expected)) {
> ++              syntax_error((last_token_class & (TC_NEWLINE |
> TC_EOF)) ?
> +                               EMSG_UNEXP_EOS : EMSG_UNEXP_TOKEN);
> +       }
> + 
> +-      debug_printf_parse("%s: returning, ltclass:%x t_double:%f\n",
> __func__, ltclass, t_double);
> +-      return ltclass;
> ++      debug_printf_parse("%s: returning, t_double:%f t_tclass:",
> __func__, t_double);
> ++      debug_parse_print_tc(t_tclass);
> ++      debug_printf_parse("\n");
> ++
> ++      return t_tclass;
> + #undef concat_inserted
> + #undef save_tclass
> + #undef save_info
> +-#undef ltclass
> + }
> + 
> +-static void rollback_token(void)
> ++static ALWAYS_INLINE void rollback_token(void)
> + {
> +       t_rollback = TRUE;
> + }
> +@@ -1251,169 +1315,188 @@ static node *new_node(uint32_t info)
> + 
> + static void mk_re_node(const char *s, node *n, regex_t *re)
> + {
> +-      n->info = OC_REGEXP;
> ++      n->info = TI_REGEXP;
> +       n->l.re = re;
> +       n->r.ire = re + 1;
> +       xregcomp(re, s, REG_EXTENDED);
> +       xregcomp(re + 1, s, REG_EXTENDED | REG_ICASE);
> + }
> + 
> +-static node *condition(void)
> ++static node *parse_expr(uint32_t);
> ++
> ++static node *parse_lrparen_list(void)
> + {
> +-      next_token(TC_SEQSTART);
> +-      return parse_expr(TC_SEQTERM);
> ++      next_token(TC_LPAREN);
> ++      return parse_expr(TC_RPAREN);
> + }
> + 
> + /* parse expression terminated by given argument, return ptr
> +  * to built subtree. Terminator is eaten by parse_expr */
> +-static node *parse_expr(uint32_t iexp)
> ++static node *parse_expr(uint32_t term_tc)
> + {
> +       node sn;
> +       node *cn = &sn;
> +       node *vn, *glptr;
> +-      uint32_t tc, xtc;
> ++      uint32_t tc, expected_tc;
> +       var *v;
> + 
> +-      debug_printf_parse("%s(%x)\n", __func__, iexp);
> ++      debug_printf_parse("%s() term_tc(%x):", __func__, term_tc);
> ++      debug_parse_print_tc(term_tc);
> ++      debug_printf_parse("\n");
> + 
> +       sn.info = PRIMASK;
> +       sn.r.n = sn.a.n = glptr = NULL;
> +-      xtc = TC_OPERAND | TC_UOPPRE | TC_REGEXP | iexp;
> ++      expected_tc = TS_OPERAND | TS_UOPPRE | TC_REGEXP | term_tc;
> + 
> +-      while (!((tc = next_token(xtc)) & iexp)) {
> ++      while (!((tc = next_token(expected_tc)) & term_tc)) {
> + 
> +-              if (glptr && (t_info == (OC_COMPARE | VV | P(39) |
> 2))) {
> ++              if (glptr && (t_info == TI_LESS)) {
> +                       /* input redirection (<) attached to glptr
> node */
> +                       debug_printf_parse("%s: input redir\n",
> __func__);
> +                       cn = glptr->l.n = new_node(OC_CONCAT | SS |
> P(37));
> +                       cn->a.n = glptr;
> +-                      xtc = TC_OPERAND | TC_UOPPRE;
> ++                      expected_tc = TS_OPERAND | TS_UOPPRE;
> +                       glptr = NULL;
> +-
> +-              } else if (tc & (TC_BINOP | TC_UOPPOST)) {
> +-                      debug_printf_parse("%s: TC_BINOP | TC_UOPPOST
> tc:%x\n", __func__, tc);
> ++                      continue;
> ++              }
> ++              if (tc & (TS_BINOP | TC_UOPPOST)) {
> ++                      debug_printf_parse("%s: TS_BINOP | TC_UOPPOST
> tc:%x\n", __func__, tc);
> +                       /* for binary and postfix-unary operators,
> jump back over
> +                        * previous operators with higher priority */
> +                       vn = cn;
> +                       while (((t_info & PRIMASK) > (vn->a.n->info &
> PRIMASK2))
> +-                          || ((t_info == vn->info) && ((t_info &
> OPCLSMASK) == OC_COLON))
> ++                          || ((t_info == vn->info) && t_info ==
> TI_COLON)
> +                       ) {
> +                               vn = vn->a.n;
> +                               if (!vn->a.n)
> syntax_error(EMSG_UNEXP_TOKEN);
> +                       }
> +-                      if ((t_info & OPCLSMASK) == OC_TERNARY)
> ++                      if (t_info == TI_TERNARY)
> ++//TODO: why?
> +                               t_info += P(6);
> +                       cn = vn->a.n->r.n = new_node(t_info);
> +                       cn->a.n = vn->a.n;
> +-                      if (tc & TC_BINOP) {
> ++                      if (tc & TS_BINOP) {
> +                               cn->l.n = vn;
> +-                              xtc = TC_OPERAND | TC_UOPPRE |
> TC_REGEXP;
> +-                              if ((t_info & OPCLSMASK) ==
> OC_PGETLINE) {
> ++//FIXME: this is the place to detect and reject assignments to non-
> lvalues.
> ++//Currently we allow "assignments" to consts and temporaries,
> nonsense like this:
> ++// awk 'BEGIN { "qwe" = 1 }'
> ++// awk 'BEGIN { 7 *= 7 }'
> ++// awk 'BEGIN { length("qwe") = 1 }'
> ++// awk 'BEGIN { (1+1) += 3 }'
> ++                              expected_tc = TS_OPERAND | TS_UOPPRE
> | TC_REGEXP;
> ++                              if (t_info == TI_PGETLINE) {
> +                                       /* it's a pipe */
> +                                       next_token(TC_GETLINE);
> +                                       /* give maximum priority to
> this pipe */
> +                                       cn->info &= ~PRIMASK;
> +-                                      xtc = TC_OPERAND | TC_UOPPRE
> | TC_BINOP | iexp;
> ++                                      expected_tc = TS_OPERAND |
> TS_UOPPRE | TS_BINOP | term_tc;
> +                               }
> +                       } else {
> +                               cn->r.n = vn;
> +-                              xtc = TC_OPERAND | TC_UOPPRE |
> TC_BINOP | iexp;
> ++                              expected_tc = TS_OPERAND | TS_UOPPRE
> | TS_BINOP | term_tc;
> +                       }
> +                       vn->a.n = cn;
> ++                      continue;
> ++              }
> + 
> +-              } else {
> +-                      debug_printf_parse("%s: other\n", __func__);
> +-                      /* for operands and prefix-unary operators,
> attach them
> +-                       * to last node */
> +-                      vn = cn;
> +-                      cn = vn->r.n = new_node(t_info);
> +-                      cn->a.n = vn;
> +-                      xtc = TC_OPERAND | TC_UOPPRE | TC_REGEXP;
> +-                      if (tc & (TC_OPERAND | TC_REGEXP)) {
> +-                              debug_printf_parse("%s: TC_OPERAND |
> TC_REGEXP\n", __func__);
> +-                              xtc = TC_UOPPRE | TC_UOPPOST |
> TC_BINOP | TC_OPERAND | iexp;
> +-                              /* one should be very careful with
> switch on tclass -
> +-                               * only simple tclasses should be
> used! */
> +-                              switch (tc) {
> +-                              case TC_VARIABLE:
> +-                              case TC_ARRAY:
> +-                                      debug_printf_parse("%s:
> TC_VARIABLE | TC_ARRAY\n", __func__);
> +-                                      cn->info = OC_VAR;
> +-                                      v = hash_search(ahash,
> t_string);
> +-                                      if (v != NULL) {
> +-                                              cn->info = OC_FNARG;
> +-                                              cn->l.aidx = v-
> >x.aidx;
> +-                                      } else {
> +-                                              cn->l.v =
> newvar(t_string);
> +-                                      }
> +-                                      if (tc & TC_ARRAY) {
> +-                                              cn->info |= xS;
> +-                                              cn->r.n =
> parse_expr(TC_ARRTERM);
> +-                                      }
> +-                                      break;
> ++              debug_printf_parse("%s: other, t_info:%x\n",
> __func__, t_info);
> ++              /* for operands and prefix-unary operators, attach
> them
> ++               * to last node */
> ++              vn = cn;
> ++              cn = vn->r.n = new_node(t_info);
> ++              cn->a.n = vn;
> + 
> +-                              case TC_NUMBER:
> +-                              case TC_STRING:
> +-                                      debug_printf_parse("%s:
> TC_NUMBER | TC_STRING\n", __func__);
> +-                                      cn->info = OC_VAR;
> +-                                      v = cn->l.v =
> xzalloc(sizeof(var));
> +-                                      if (tc & TC_NUMBER)
> +-                                              setvar_i(v,
> t_double);
> +-                                      else {
> +-                                              setvar_s(v,
> t_string);
> +-                                              xtc &= ~TC_UOPPOST;
> /* "str"++ is not allowed */
> +-                                      }
> +-                                      break;
> ++              expected_tc = TS_OPERAND | TS_UOPPRE | TC_REGEXP;
> ++              if (t_info == TI_PREINC || t_info == TI_PREDEC)
> ++                      expected_tc = TS_LVALUE | TC_UOPPRE1;
> + 
> +-                              case TC_REGEXP:
> +-                                      debug_printf_parse("%s:
> TC_REGEXP\n", __func__);
> +-                                      mk_re_node(t_string, cn,
> xzalloc(sizeof(regex_t)*2));
> +-                                      break;
> ++              if (!(tc & (TS_OPERAND | TC_REGEXP)))
> ++                      continue;
> + 
> +-                              case TC_FUNCTION:
> +-                                      debug_printf_parse("%s:
> TC_FUNCTION\n", __func__);
> +-                                      cn->info = OC_FUNC;
> +-                                      cn->r.f = newfunc(t_string);
> +-                                      cn->l.n = condition();
> +-                                      break;
> ++              debug_printf_parse("%s: TS_OPERAND | TC_REGEXP\n",
> __func__);
> ++              expected_tc = TS_UOPPRE | TC_UOPPOST | TS_BINOP |
> TS_OPERAND | term_tc;
> ++              /* one should be very careful with switch on tclass -
> ++               * only simple tclasses should be used (TC_xyz, not
> TS_xyz) */
> ++              switch (tc) {
> ++              case TC_VARIABLE:
> ++              case TC_ARRAY:
> ++                      debug_printf_parse("%s: TC_VARIABLE |
> TC_ARRAY\n", __func__);
> ++                      cn->info = OC_VAR;
> ++                      v = hash_search(ahash, t_string);
> ++                      if (v != NULL) {
> ++                              cn->info = OC_FNARG;
> ++                              cn->l.aidx = v->x.aidx;
> ++                      } else {
> ++                              cn->l.v = newvar(t_string);
> ++                      }
> ++                      if (tc & TC_ARRAY) {
> ++                              cn->info |= xS;
> ++                              cn->r.n = parse_expr(TC_ARRTERM);
> ++                      }
> ++                      break;
> + 
> +-                              case TC_SEQSTART:
> +-                                      debug_printf_parse("%s:
> TC_SEQSTART\n", __func__);
> +-                                      cn = vn->r.n =
> parse_expr(TC_SEQTERM);
> +-                                      if (!cn)
> +-                                              syntax_error("Empty
> sequence");
> +-                                      cn->a.n = vn;
> +-                                      break;
> ++              case TC_NUMBER:
> ++              case TC_STRING:
> ++                      debug_printf_parse("%s: TC_NUMBER |
> TC_STRING\n", __func__);
> ++                      cn->info = OC_VAR;
> ++                      v = cn->l.v = xzalloc(sizeof(var));
> ++                      if (tc & TC_NUMBER)
> ++                              setvar_i(v, t_double);
> ++                      else {
> ++                              setvar_s(v, t_string);
> ++                              expected_tc &= ~TC_UOPPOST; /*
> "str"++ is not allowed */
> ++                      }
> ++                      break;
> + 
> +-                              case TC_GETLINE:
> +-                                      debug_printf_parse("%s:
> TC_GETLINE\n", __func__);
> +-                                      glptr = cn;
> +-                                      xtc = TC_OPERAND | TC_UOPPRE
> | TC_BINOP | iexp;
> +-                                      break;
> ++              case TC_REGEXP:
> ++                      debug_printf_parse("%s: TC_REGEXP\n",
> __func__);
> ++                      mk_re_node(t_string, cn,
> xzalloc(sizeof(regex_t)*2));
> ++                      break;
> + 
> +-                              case TC_BUILTIN:
> +-                                      debug_printf_parse("%s:
> TC_BUILTIN\n", __func__);
> +-                                      cn->l.n = condition();
> +-                                      break;
> ++              case TC_FUNCTION:
> ++                      debug_printf_parse("%s: TC_FUNCTION\n",
> __func__);
> ++                      cn->info = OC_FUNC;
> ++                      cn->r.f = newfunc(t_string);
> ++                      cn->l.n = parse_expr(TC_RPAREN);
> ++                      break;
> + 
> +-                              case TC_LENGTH:
> +-                                      debug_printf_parse("%s:
> TC_LENGTH\n", __func__);
> +-                                      next_token(TC_SEQSTART /*
> length(...) */
> +-                                              | TC_OPTERM    /*
> length; (or newline)*/
> +-                                              | TC_GRPTERM   /*
> length } */
> +-                                              | TC_BINOPX    /*
> length <op> NUM */
> +-                                              | TC_COMMA     /*
> print length, 1 */
> +-                                      );
> +-                                      rollback_token();
> +-                                      if (t_tclass & TC_SEQSTART) {
> +-                                              /* It was a "("
> token. Handle just like TC_BUILTIN */
> +-                                              cn->l.n =
> condition();
> +-                                      }
> +-                                      break;
> +-                              }
> ++              case TC_LPAREN:
> ++                      debug_printf_parse("%s: TC_LPAREN\n",
> __func__);
> ++                      cn = vn->r.n = parse_expr(TC_RPAREN);
> ++                      if (!cn)
> ++                              syntax_error("Empty sequence");
> ++                      cn->a.n = vn;
> ++                      break;
> ++
> ++              case TC_GETLINE:
> ++                      debug_printf_parse("%s: TC_GETLINE\n",
> __func__);
> ++                      glptr = cn;
> ++                      expected_tc = TS_OPERAND | TS_UOPPRE |
> TS_BINOP | term_tc;
> ++                      break;
> ++
> ++              case TC_BUILTIN:
> ++                      debug_printf_parse("%s: TC_BUILTIN\n",
> __func__);
> ++                      cn->l.n = parse_lrparen_list();
> ++                      break;
> ++
> ++              case TC_LENGTH:
> ++                      debug_printf_parse("%s: TC_LENGTH\n",
> __func__);
> ++                      tc = next_token(TC_LPAREN /* length(...) */
> ++                              | TC_SEMICOL   /* length; */
> ++                              | TC_NEWLINE   /* length<newline> */
> ++                              | TC_RBRACE    /* length } */
> ++                              | TC_BINOPX    /* length <op> NUM */
> ++                              | TC_COMMA     /* print length, 1 */
> ++                      );
> ++                      if (tc != TC_LPAREN)
> ++                              rollback_token();
> ++                      else {
> ++                              /* It was a "(" token. Handle just
> like TC_BUILTIN */
> ++                              cn->l.n = parse_expr(TC_RPAREN);
> +                       }
> ++                      break;
> +               }
> +-      }
> ++      } /* while() */
> + 
> +       debug_printf_parse("%s() returns %p\n", __func__, sn.r.n);
> +       return sn.r.n;
> +@@ -1430,7 +1513,7 @@ static node *chain_node(uint32_t info)
> +       if (seq->programname != g_progname) {
> +               seq->programname = g_progname;
> +               n = chain_node(OC_NEWSOURCE);
> +-              n->l.new_progname = xstrdup(g_progname);
> ++              n->l.new_progname = g_progname;
> +       }
> + 
> +       n = seq->last;
> +@@ -1446,14 +1529,16 @@ static void chain_expr(uint32_t info)
> + 
> +       n = chain_node(info);
> + 
> +-      n->l.n = parse_expr(TC_OPTERM | TC_GRPTERM);
> ++      n->l.n = parse_expr(TC_SEMICOL | TC_NEWLINE | TC_RBRACE);
> +       if ((info & OF_REQUIRED) && !n->l.n)
> +               syntax_error(EMSG_TOO_FEW_ARGS);
> + 
> +-      if (t_tclass & TC_GRPTERM)
> ++      if (t_tclass & TC_RBRACE)
> +               rollback_token();
> + }
> + 
> ++static void chain_group(void);
> ++
> + static node *chain_loop(node *nn)
> + {
> +       node *n, *n2, *save_brk, *save_cont;
> +@@ -1477,207 +1562,284 @@ static node *chain_loop(node *nn)
> +       return n;
> + }
> + 
> ++static void chain_until_rbrace(void)
> ++{
> ++      uint32_t tc;
> ++      while ((tc = next_token(TS_GRPSEQ | TC_RBRACE)) != TC_RBRACE)
> {
> ++              debug_printf_parse("%s: !TC_RBRACE\n", __func__);
> ++              if (tc == TC_NEWLINE)
> ++                      continue;
> ++              rollback_token();
> ++              chain_group();
> ++      }
> ++      debug_printf_parse("%s: TC_RBRACE\n", __func__);
> ++}
> ++
> + /* parse group and attach it to chain */
> + static void chain_group(void)
> + {
> +-      uint32_t c;
> ++      uint32_t tc;
> +       node *n, *n2, *n3;
> + 
> +       do {
> +-              c = next_token(TC_GRPSEQ);
> +-      } while (c & TC_NEWLINE);
> +-
> +-      if (c & TC_GRPSTART) {
> +-              debug_printf_parse("%s: TC_GRPSTART\n", __func__);
> +-              while (next_token(TC_GRPSEQ | TC_GRPTERM) !=
> TC_GRPTERM) {
> +-                      debug_printf_parse("%s: !TC_GRPTERM\n",
> __func__);
> +-                      if (t_tclass & TC_NEWLINE)
> +-                              continue;
> +-                      rollback_token();
> +-                      chain_group();
> +-              }
> +-              debug_printf_parse("%s: TC_GRPTERM\n", __func__);
> +-      } else if (c & (TC_OPSEQ | TC_OPTERM)) {
> +-              debug_printf_parse("%s: TC_OPSEQ | TC_OPTERM\n",
> __func__);
> ++              tc = next_token(TS_GRPSEQ);
> ++      } while (tc == TC_NEWLINE);
> ++
> ++      if (tc == TC_LBRACE) {
> ++              debug_printf_parse("%s: TC_LBRACE\n", __func__);
> ++              chain_until_rbrace();
> ++              return;
> ++      }
> ++      if (tc & (TS_OPSEQ | TC_SEMICOL)) {
> ++              debug_printf_parse("%s: TS_OPSEQ | TC_SEMICOL\n",
> __func__);
> +               rollback_token();
> +               chain_expr(OC_EXEC | Vx);
> +-      } else {
> +-              /* TC_STATEMNT */
> +-              debug_printf_parse("%s: TC_STATEMNT(?)\n", __func__);
> +-              switch (t_info & OPCLSMASK) {
> +-              case ST_IF:
> +-                      debug_printf_parse("%s: ST_IF\n", __func__);
> +-                      n = chain_node(OC_BR | Vx);
> +-                      n->l.n = condition();
> ++              return;
> ++      }
> ++
> ++      /* TS_STATEMNT */
> ++      debug_printf_parse("%s: TS_STATEMNT(?)\n", __func__);
> ++      switch (t_info & OPCLSMASK) {
> ++      case ST_IF:
> ++              debug_printf_parse("%s: ST_IF\n", __func__);
> ++              n = chain_node(OC_BR | Vx);
> ++              n->l.n = parse_lrparen_list();
> ++              chain_group();
> ++              n2 = chain_node(OC_EXEC);
> ++              n->r.n = seq->last;
> ++              if (next_token(TS_GRPSEQ | TC_RBRACE | TC_ELSE) ==
> TC_ELSE) {
> +                       chain_group();
> +-                      n2 = chain_node(OC_EXEC);
> +-                      n->r.n = seq->last;
> +-                      if (next_token(TC_GRPSEQ | TC_GRPTERM |
> TC_ELSE) == TC_ELSE) {
> +-                              chain_group();
> +-                              n2->a.n = seq->last;
> +-                      } else {
> +-                              rollback_token();
> +-                      }
> +-                      break;
> ++                      n2->a.n = seq->last;
> ++              } else {
> ++                      rollback_token();
> ++              }
> ++              break;
> + 
> +-              case ST_WHILE:
> +-                      debug_printf_parse("%s: ST_WHILE\n",
> __func__);
> +-                      n2 = condition();
> +-                      n = chain_loop(NULL);
> +-                      n->l.n = n2;
> +-                      break;
> ++      case ST_WHILE:
> ++              debug_printf_parse("%s: ST_WHILE\n", __func__);
> ++              n2 = parse_lrparen_list();
> ++              n = chain_loop(NULL);
> ++              n->l.n = n2;
> ++              break;
> + 
> +-              case ST_DO:
> +-                      debug_printf_parse("%s: ST_DO\n", __func__);
> +-                      n2 = chain_node(OC_EXEC);
> +-                      n = chain_loop(NULL);
> +-                      n2->a.n = n->a.n;
> +-                      next_token(TC_WHILE);
> +-                      n->l.n = condition();
> +-                      break;
> ++      case ST_DO:
> ++              debug_printf_parse("%s: ST_DO\n", __func__);
> ++              n2 = chain_node(OC_EXEC);
> ++              n = chain_loop(NULL);
> ++              n2->a.n = n->a.n;
> ++              next_token(TC_WHILE);
> ++              n->l.n = parse_lrparen_list();
> ++              break;
> + 
> +-              case ST_FOR:
> +-                      debug_printf_parse("%s: ST_FOR\n", __func__);
> +-                      next_token(TC_SEQSTART);
> +-                      n2 = parse_expr(TC_SEMICOL | TC_SEQTERM);
> +-                      if (t_tclass & TC_SEQTERM) {    /* for-in */
> +-                              if (!n2 || (n2->info & OPCLSMASK) !=
> OC_IN)
> +-
>                                       syntax_error(EMSG_UNEXP_TOKEN);
> +-                              n = chain_node(OC_WALKINIT | VV);
> +-                              n->l.n = n2->l.n;
> +-                              n->r.n = n2->r.n;
> +-                              n = chain_loop(NULL);
> +-                              n->info = OC_WALKNEXT | Vx;
> +-                              n->l.n = n2->l.n;
> +-                      } else {                        /* for (;;)
> */
> +-                              n = chain_node(OC_EXEC | Vx);
> +-                              n->l.n = n2;
> +-                              n2 = parse_expr(TC_SEMICOL);
> +-                              n3 = parse_expr(TC_SEQTERM);
> +-                              n = chain_loop(n3);
> +-                              n->l.n = n2;
> +-                              if (!n2)
> +-                                      n->info = OC_EXEC;
> +-                      }
> +-                      break;
> ++      case ST_FOR:
> ++              debug_printf_parse("%s: ST_FOR\n", __func__);
> ++              next_token(TC_LPAREN);
> ++              n2 = parse_expr(TC_SEMICOL | TC_RPAREN);
> ++              if (t_tclass & TC_RPAREN) {     /* for (I in ARRAY)
> */
> ++                      if (!n2 || n2->info != TI_IN)
> ++                              syntax_error(EMSG_UNEXP_TOKEN);
> ++                      n = chain_node(OC_WALKINIT | VV);
> ++                      n->l.n = n2->l.n;
> ++                      n->r.n = n2->r.n;
> ++                      n = chain_loop(NULL);
> ++                      n->info = OC_WALKNEXT | Vx;
> ++                      n->l.n = n2->l.n;
> ++              } else {                        /* for (;;) */
> ++                      n = chain_node(OC_EXEC | Vx);
> ++                      n->l.n = n2;
> ++                      n2 = parse_expr(TC_SEMICOL);
> ++                      n3 = parse_expr(TC_RPAREN);
> ++                      n = chain_loop(n3);
> ++                      n->l.n = n2;
> ++                      if (!n2)
> ++                              n->info = OC_EXEC;
> ++              }
> ++              break;
> + 
> +-              case OC_PRINT:
> +-              case OC_PRINTF:
> +-                      debug_printf_parse("%s: OC_PRINT[F]\n",
> __func__);
> +-                      n = chain_node(t_info);
> +-                      n->l.n = parse_expr(TC_OPTERM | TC_OUTRDR |
> TC_GRPTERM);
> +-                      if (t_tclass & TC_OUTRDR) {
> +-                              n->info |= t_info;
> +-                              n->r.n = parse_expr(TC_OPTERM |
> TC_GRPTERM);
> +-                      }
> +-                      if (t_tclass & TC_GRPTERM)
> +-                              rollback_token();
> +-                      break;
> ++      case OC_PRINT:
> ++      case OC_PRINTF:
> ++              debug_printf_parse("%s: OC_PRINT[F]\n", __func__);
> ++              n = chain_node(t_info);
> ++              n->l.n = parse_expr(TC_SEMICOL | TC_NEWLINE |
> TC_OUTRDR | TC_RBRACE);
> ++              if (t_tclass & TC_OUTRDR) {
> ++                      n->info |= t_info;
> ++                      n->r.n = parse_expr(TC_SEMICOL | TC_NEWLINE |
> TC_RBRACE);
> ++              }
> ++              if (t_tclass & TC_RBRACE)
> ++                      rollback_token();
> ++              break;
> + 
> +-              case OC_BREAK:
> +-                      debug_printf_parse("%s: OC_BREAK\n",
> __func__);
> +-                      n = chain_node(OC_EXEC);
> +-                      n->a.n = break_ptr;
> +-                      chain_expr(t_info);
> +-                      break;
> ++      case OC_BREAK:
> ++              debug_printf_parse("%s: OC_BREAK\n", __func__);
> ++              n = chain_node(OC_EXEC);
> ++              if (!break_ptr)
> ++                      syntax_error("'break' not in a loop");
> ++              n->a.n = break_ptr;
> ++              chain_expr(t_info);
> ++              break;
> + 
> +-              case OC_CONTINUE:
> +-                      debug_printf_parse("%s: OC_CONTINUE\n",
> __func__);
> +-                      n = chain_node(OC_EXEC);
> +-                      n->a.n = continue_ptr;
> +-                      chain_expr(t_info);
> +-                      break;
> ++      case OC_CONTINUE:
> ++              debug_printf_parse("%s: OC_CONTINUE\n", __func__);
> ++              n = chain_node(OC_EXEC);
> ++              if (!continue_ptr)
> ++                      syntax_error("'continue' not in a loop");
> ++              n->a.n = continue_ptr;
> ++              chain_expr(t_info);
> ++              break;
> + 
> +-              /* delete, next, nextfile, return, exit */
> +-              default:
> +-                      debug_printf_parse("%s: default\n",
> __func__);
> +-                      chain_expr(t_info);
> +-              }
> ++      /* delete, next, nextfile, return, exit */
> ++      default:
> ++              debug_printf_parse("%s: default\n", __func__);
> ++              chain_expr(t_info);
> +       }
> + }
> + 
> + static void parse_program(char *p)
> + {
> +-      uint32_t tclass;
> +-      node *cn;
> +-      func *f;
> +-      var *v;
> ++      debug_printf_parse("%s()\n", __func__);
> + 
> +       g_pos = p;
> +       t_lineno = 1;
> +-      while ((tclass = next_token(TC_EOF | TC_OPSEQ | TC_GRPSTART |
> +-                      TC_OPTERM | TC_BEGIN | TC_END | TC_FUNCDECL))
> != TC_EOF) {
> ++      for (;;) {
> ++              uint32_t tclass;
> + 
> +-              if (tclass & TC_OPTERM) {
> +-                      debug_printf_parse("%s: TC_OPTERM\n",
> __func__);
> ++              tclass = next_token(TS_OPSEQ | TC_LBRACE | TC_BEGIN |
> TC_END | TC_FUNCDECL
> ++                      | TC_EOF | TC_NEWLINE /* but not TC_SEMICOL
> */);
> ++ got_tok:
> ++              if (tclass == TC_EOF) {
> ++                      debug_printf_parse("%s: TC_EOF\n", __func__);
> ++                      break;
> ++              }
> ++              if (tclass == TC_NEWLINE) {
> ++                      debug_printf_parse("%s: TC_NEWLINE\n",
> __func__);
> +                       continue;
> +               }
> +-
> +-              seq = &mainseq;
> +-              if (tclass & TC_BEGIN) {
> ++              if (tclass == TC_BEGIN) {
> +                       debug_printf_parse("%s: TC_BEGIN\n",
> __func__);
> +                       seq = &beginseq;
> +-                      chain_group();
> +-              } else if (tclass & TC_END) {
> ++                      /* ensure there is no newline between BEGIN
> and { */
> ++                      next_token(TC_LBRACE);
> ++                      chain_until_rbrace();
> ++                      goto next_tok;
> ++              }
> ++              if (tclass == TC_END) {
> +                       debug_printf_parse("%s: TC_END\n", __func__);
> +                       seq = &endseq;
> +-                      chain_group();
> +-              } else if (tclass & TC_FUNCDECL) {
> ++                      /* ensure there is no newline between END and
> { */
> ++                      next_token(TC_LBRACE);
> ++                      chain_until_rbrace();
> ++                      goto next_tok;
> ++              }
> ++              if (tclass == TC_FUNCDECL) {
> ++                      func *f;
> ++
> +                       debug_printf_parse("%s: TC_FUNCDECL\n",
> __func__);
> +                       next_token(TC_FUNCTION);
> +-                      g_pos++;
> +                       f = newfunc(t_string);
> +-                      f->body.first = NULL;
> +-                      f->nargs = 0;
> +-                      /* Match func arg list: a comma sep list of
> >= 0 args, and a close paren */
> +-                      while (next_token(TC_VARIABLE | TC_SEQTERM |
> TC_COMMA)) {
> +-                              /* Either an empty arg list, or
> trailing comma from prev iter
> +-                               * must be followed by an arg */
> +-                              if (f->nargs == 0 && t_tclass ==
> TC_SEQTERM)
> +-                                      break;
> +-
> +-                              /* TC_SEQSTART/TC_COMMA must be
> followed by TC_VARIABLE */
> +-                              if (t_tclass != TC_VARIABLE)
> ++                      if (f->defined)
> ++                              syntax_error("Duplicate function");
> ++                      f->defined = 1;
> ++                      //f->body.first = NULL; - already is
> ++                      //f->nargs = 0; - already is
> ++                      /* func arg list: comma sep list of args, and
> a close paren */
> ++                      for (;;) {
> ++                              var *v;
> ++                              if (next_token(TC_VARIABLE |
> TC_RPAREN) == TC_RPAREN) {
> ++                                      if (f->nargs == 0)
> ++                                              break; /* func() is
> ok */
> ++                                      /* func(a,) is not ok */
> +                                       syntax_error(EMSG_UNEXP_TOKEN
> );
> +-
> ++                              }
> +                               v = findvar(ahash, t_string);
> +                               v->x.aidx = f->nargs++;
> +-
> +                               /* Arg followed either by end of arg
> list or 1 comma */
> +-                              if (next_token(TC_COMMA | TC_SEQTERM)
> & TC_SEQTERM)
> ++                              if (next_token(TC_COMMA | TC_RPAREN)
> == TC_RPAREN)
> +                                       break;
> +-                              if (t_tclass != TC_COMMA)
> +-
>                                       syntax_error(EMSG_UNEXP_TOKEN);
> ++                              /* it was a comma, we ate it */
> +                       }
> +                       seq = &f->body;
> +-                      chain_group();
> +-                      clear_array(ahash);
> +-              } else if (tclass & TC_OPSEQ) {
> +-                      debug_printf_parse("%s: TC_OPSEQ\n",
> __func__);
> ++                      /* ensure there is { after "func F(...)" -
> but newlines are allowed */
> ++                      while (next_token(TC_LBRACE | TC_NEWLINE) ==
> TC_NEWLINE)
> ++                              continue;
> ++                      chain_until_rbrace();
> ++                      hash_clear(ahash);
> ++                      goto next_tok;
> ++              }
> ++              seq = &mainseq;
> ++              if (tclass & TS_OPSEQ) {
> ++                      node *cn;
> ++
> ++                      debug_printf_parse("%s: TS_OPSEQ\n",
> __func__);
> +                       rollback_token();
> +                       cn = chain_node(OC_TEST);
> +-                      cn->l.n = parse_expr(TC_OPTERM | TC_EOF |
> TC_GRPSTART);
> +-                      if (t_tclass & TC_GRPSTART) {
> +-                              debug_printf_parse("%s:
> TC_GRPSTART\n", __func__);
> +-                              rollback_token();
> +-                              chain_group();
> ++                      cn->l.n = parse_expr(TC_SEMICOL | TC_NEWLINE
> | TC_EOF | TC_LBRACE);
> ++                      if (t_tclass == TC_LBRACE) {
> ++                              debug_printf_parse("%s: TC_LBRACE\n",
> __func__);
> ++                              chain_until_rbrace();
> +                       } else {
> +-                              debug_printf_parse("%s:
> !TC_GRPSTART\n", __func__);
> ++                              /* no action, assume default "{ print
> }" */
> ++                              debug_printf_parse("%s:
> !TC_LBRACE\n", __func__);
> +                               chain_node(OC_PRINT);
> +                       }
> +                       cn->r.n = mainseq.last;
> +-              } else /* if (tclass & TC_GRPSTART) */ {
> +-                      debug_printf_parse("%s: TC_GRPSTART(?)\n",
> __func__);
> +-                      rollback_token();
> +-                      chain_group();
> ++                      goto next_tok;
> +               }
> +-      }
> +-      debug_printf_parse("%s: TC_EOF\n", __func__);
> ++              /* tclass == TC_LBRACE */
> ++              debug_printf_parse("%s: TC_LBRACE(?)\n", __func__);
> ++              chain_until_rbrace();
> ++ next_tok:
> ++              /* Same as next_token() at the top of the loop, +
> TC_SEMICOL */
> ++              tclass = next_token(TS_OPSEQ | TC_LBRACE | TC_BEGIN |
> TC_END | TC_FUNCDECL
> ++                      | TC_EOF | TC_NEWLINE | TC_SEMICOL);
> ++              /* gawk allows many newlines, but does not allow more
> than one semicolon:
> ++               *  BEGIN {...}<newline>;<newline>;
> ++               * would complain "each rule must have a pattern or
> an action part".
> ++               * Same message for
> ++               *  ; BEGIN {...}
> ++               */
> ++              if (tclass != TC_SEMICOL)
> ++                      goto got_tok; /* use this token */
> ++              /* else: loop back - ate the semicolon, get and use
> _next_ token */
> ++      } /* for (;;) */
> + }
> + 
> +-
> + /* -------- program execution part -------- */
> + 
> ++/* temporary variables allocator */
> ++static var *nvalloc(int sz)
> ++{
> ++      return xzalloc(sz * sizeof(var));
> ++}
> ++
> ++static void nvfree(var *v, int sz)
> ++{
> ++      var *p = v;
> ++
> ++      while (--sz >= 0) {
> ++              if ((p->type & (VF_ARRAY | VF_CHILD)) == VF_ARRAY) {
> ++                      clear_array(iamarray(p));
> ++                      free(p->x.array->items);
> ++                      free(p->x.array);
> ++              }
> ++              if (p->type & VF_WALK) {
> ++                      walker_list *n;
> ++                      walker_list *w = p->x.walker;
> ++                      debug_printf_walker("nvfree: freeing walker
> @%p\n", &p->x.walker);
> ++                      p->x.walker = NULL;
> ++                      while (w) {
> ++                              n = w->prev;
> ++                              debug_printf_walker(" free(%p)\n",
> w);
> ++                              free(w);
> ++                              w = n;
> ++                      }
> ++              }
> ++              clrvar(p);
> ++              p++;
> ++      }
> ++
> ++      free(v);
> ++}
> ++
> + static node *mk_splitter(const char *s, tsplitter *spl)
> + {
> +       regex_t *re, *ire;
> +@@ -1686,7 +1848,7 @@ static node *mk_splitter(const char *s,
> tsplitter *spl)
> +       re = &spl->re[0];
> +       ire = &spl->re[1];
> +       n = &spl->n;
> +-      if ((n->info & OPCLSMASK) == OC_REGEXP) {
> ++      if (n->info == TI_REGEXP) {
> +               regfree(re);
> +               regfree(ire); // TODO: nuke ire, use re+1?
> +       }
> +@@ -1699,21 +1861,28 @@ static node *mk_splitter(const char *s,
> tsplitter *spl)
> +       return n;
> + }
> + 
> +-/* use node as a regular expression. Supplied with node ptr and
> regex_t
> ++static var *evaluate(node *, var *);
> ++
> ++/* Use node as a regular expression. Supplied with node ptr and
> regex_t
> +  * storage space. Return ptr to regex (if result points to preg, it
> should
> +- * be later regfree'd manually
> ++ * be later regfree'd manually).
> +  */
> + static regex_t *as_regex(node *op, regex_t *preg)
> + {
> +       int cflags;
> +-      var *v;
> +       const char *s;
> + 
> +-      if ((op->info & OPCLSMASK) == OC_REGEXP) {
> ++      if (op->info == TI_REGEXP) {
> +               return icase ? op->r.ire : op->l.re;
> +       }
> +-      v = nvalloc(1);
> +-      s = getvar_s(evaluate(op, v));
> ++
> ++      //tmpvar = nvalloc(1);
> ++#define TMPVAR (&G.as_regex__tmpvar)
> ++      // We use a single "static" tmpvar (instead of on-stack or
> malloced one)
> ++      // to decrease memory consumption in deeply-recursive awk
> programs.
> ++      // The rule to work safely is to never call evaluate() while
> our static
> ++      // TMPVAR's value is still needed.
> ++      s = getvar_s(evaluate(op, TMPVAR));
> + 
> +       cflags = icase ? REG_EXTENDED | REG_ICASE : REG_EXTENDED;
> +       /* Testcase where REG_EXTENDED fails (unpaired '{'):
> +@@ -1725,7 +1894,8 @@ static regex_t *as_regex(node *op, regex_t
> *preg)
> +               cflags &= ~REG_EXTENDED;
> +               xregcomp(preg, s, cflags);
> +       }
> +-      nvfree(v);
> ++      //nvfree(tmpvar, 1);
> ++#undef TMPVAR
> +       return preg;
> + }
> + 
> +@@ -1745,12 +1915,22 @@ static char* qrealloc(char *b, int n, int
> *size)
> + /* resize field storage space */
> + static void fsrealloc(int size)
> + {
> +-      int i;
> ++      int i, newsize;
> + 
> +       if (size >= maxfields) {
> ++              /* Sanity cap, easier than catering for overflows */
> ++              if (size > 0xffffff)
> ++                      bb_die_memory_exhausted();
> ++
> +               i = maxfields;
> +               maxfields = size + 16;
> +-              Fields = xrealloc(Fields, maxfields *
> sizeof(Fields[0]));
> ++
> ++              newsize = maxfields * sizeof(Fields[0]);
> ++              debug_printf_eval("fsrealloc: xrealloc(%p, %u)\n",
> Fields, newsize);
> ++              Fields = xrealloc(Fields, newsize);
> ++              debug_printf_eval("fsrealloc: Fields=%p..%p\n",
> Fields, (char*)Fields + newsize - 1);
> ++              /* ^^^ did Fields[] move? debug aid for L.v getting
> "upstaged" by R.v in evaluate() */
> ++
> +               for (; i < maxfields; i++) {
> +                       Fields[i].type = VF_SPECIAL;
> +                       Fields[i].string = NULL;
> +@@ -1802,13 +1982,13 @@ static int awk_split(const char *s, node
> *spl, char **slist)
> +               c[2] = '\n';
> + 
> +       n = 0;
> +-      if ((spl->info & OPCLSMASK) == OC_REGEXP) {  /* regex split
> */
> ++      if (spl->info == TI_REGEXP) {  /* regex split */
> +               if (!*s)
> +                       return n; /* "": zero fields */
> +               n++; /* at least one field will be there */
> +               do {
> +                       int l;
> +-                      regmatch_t pmatch[2]; // TODO: why [2]? [1]
> is enough...
> ++                      regmatch_t pmatch[1];
> + 
> +                       l = strcspn(s, c+2); /* len till next NUL or
> \n */
> +                       if (regexec1_nonempty(icase ? spl->r.ire :
> spl->l.re, s, pmatch) == 0
> +@@ -1969,7 +2149,7 @@ static node *nextarg(node **pn)
> +       node *n;
> + 
> +       n = *pn;
> +-      if (n && (n->info & OPCLSMASK) == OC_COMMA) {
> ++      if (n && n->info == TI_COMMA) {
> +               *pn = n->r.n;
> +               n = n->l.n;
> +       } else {
> +@@ -2000,8 +2180,7 @@ static void hashwalk_init(var *v, xhash
> *array)
> +       for (i = 0; i < array->csize; i++) {
> +               hi = array->items[i];
> +               while (hi) {
> +-                      strcpy(w->end, hi->name);
> +-                      nextword(&w->end);
> ++                      w->end = stpcpy(w->end, hi->name) + 1;
> +                       hi = hi->next;
> +               }
> +       }
> +@@ -2027,15 +2206,18 @@ static int hashwalk_next(var *v)
> + /* evaluate node, return 1 when result is true, 0 otherwise */
> + static int ptest(node *pattern)
> + {
> +-      /* ptest__v is "static": to save stack space? */
> +-      return istrue(evaluate(pattern, &G.ptest__v));
> ++      // We use a single "static" tmpvar (instead of on-stack or
> malloced one)
> ++      // to decrease memory consumption in deeply-recursive awk
> programs.
> ++      // The rule to work safely is to never call evaluate() while
> our static
> ++      // TMPVAR's value is still needed.
> ++      return istrue(evaluate(pattern, &G.ptest__tmpvar));
> + }
> + 
> + /* read next record from stream rsm into a variable v */
> + static int awk_getline(rstream *rsm, var *v)
> + {
> +       char *b;
> +-      regmatch_t pmatch[2]; // TODO: why [2]? [1] is enough...
> ++      regmatch_t pmatch[1];
> +       int size, a, p, pp = 0;
> +       int fd, so, eo, r, rp;
> +       char c, *m, *s;
> +@@ -2061,7 +2243,7 @@ static int awk_getline(rstream *rsm, var *v)
> +               so = eo = p;
> +               r = 1;
> +               if (p > 0) {
> +-                      if ((rsplitter.n.info & OPCLSMASK) ==
> OC_REGEXP) {
> ++                      if (rsplitter.n.info == TI_REGEXP) {
> +                               if (regexec(icase ? rsplitter.n.r.ire
> : rsplitter.n.l.re,
> +                                                       b, 1, pmatch,
> 0) == 0) {
> +                                       so = pmatch[0].rm_so;
> +@@ -2133,82 +2315,126 @@ static int awk_getline(rstream *rsm, var
> *v)
> +       return r;
> + }
> + 
> +-static int fmt_num(char *b, int size, const char *format, double n,
> int int_as_int)
> +-{
> +-      int r = 0;
> +-      char c;
> +-      const char *s = format;
> +-
> +-      if (int_as_int && n == (long long)n) {
> +-              r = snprintf(b, size, "%lld", (long long)n);
> +-      } else {
> +-              do { c = *s; } while (c && *++s);
> +-              if (strchr("diouxX", c)) {
> +-                      r = snprintf(b, size, format, (int)n);
> +-              } else if (strchr("eEfgG", c)) {
> +-                      r = snprintf(b, size, format, n);
> +-              } else {
> +-                      syntax_error(EMSG_INV_FMT);
> +-              }
> +-      }
> +-      return r;
> +-}
> +-
> + /* formatted output into an allocated buffer, return ptr to buffer
> */
> +-static char *awk_printf(node *n)
> ++#if !ENABLE_FEATURE_AWK_GNU_EXTENSIONS
> ++# define awk_printf(a, b) awk_printf(a)
> ++#endif
> ++static char *awk_printf(node *n, size_t *len)
> + {
> +-      char *b = NULL;
> +-      char *fmt, *s, *f;
> +-      const char *s1;
> +-      int i, j, incr, bsize;
> +-      char c, c1;
> +-      var *v, *arg;
> +-
> +-      v = nvalloc(1);
> +-      fmt = f = xstrdup(getvar_s(evaluate(nextarg(&n), v)));
> +-
> ++      char *b;
> ++      char *fmt, *f;
> ++      size_t i;
> ++
> ++      //tmpvar = nvalloc(1);
> ++#define TMPVAR (&G.awk_printf__tmpvar)
> ++      // We use a single "static" tmpvar (instead of on-stack or
> malloced one)
> ++      // to decrease memory consumption in deeply-recursive awk
> programs.
> ++      // The rule to work safely is to never call evaluate() while
> our static
> ++      // TMPVAR's value is still needed.
> ++      fmt = f = xstrdup(getvar_s(evaluate(nextarg(&n), TMPVAR)));
> ++      // ^^^^^^^^^ here we immediately strdup() the value, so the
> later call
> ++      // to evaluate() potentially recursing into another
> awk_printf() can't
> ++      // mangle the value.
> ++
> ++      b = NULL;
> +       i = 0;
> +-      while (*f) {
> ++      while (1) { /* "print one format spec" loop */
> ++              char *s;
> ++              char c;
> ++              char sv;
> ++              var *arg;
> ++              size_t slen;
> ++
> ++              /* Find end of the next format spec, or end of line
> */
> +               s = f;
> +-              while (*f && (*f != '%' || *++f == '%'))
> +-                      f++;
> +-              while (*f && !isalpha(*f)) {
> +-                      if (*f == '*')
> +-                              syntax_error("%*x formats are not
> supported");
> ++              while (1) {
> ++                      c = *f;
> ++                      if (!c) /* no percent chars found at all */
> ++                              goto nul;
> +                       f++;
> ++                      if (c == '%')
> ++                              break;
> +               }
> +-
> +-              incr = (f - s) + MAXVARFMT;
> +-              b = qrealloc(b, incr + i, &bsize);
> ++              /* we are past % in "....%..." */
> +               c = *f;
> +-              if (c != '\0')
> ++              if (!c) /* "....%" */
> ++                      goto nul;
> ++              if (c == '%') { /* "....%%...." */
> ++                      slen = f - s;
> ++                      s = xstrndup(s, slen);
> +                       f++;
> +-              c1 = *f;
> ++                      goto append; /* print "....%" part verbatim
> */
> ++              }
> ++              while (1) {
> ++                      if (isalpha(c))
> ++                              break;
> ++                      if (c == '*')
> ++                              syntax_error("%*x formats are not
> supported");
> ++                      c = *++f;
> ++                      if (!c) { /* "....%...." and no letter found
> after % */
> ++                              /* Example: awk 'BEGIN { printf
> "^^^%^^^\n"; }' */
> ++ nul:
> ++                              slen = f - s;
> ++                              goto tail; /* print remaining string,
> exit loop */
> ++                      }
> ++              }
> ++              /* we are at A in "....%...A..." */
> ++
> ++              arg = evaluate(nextarg(&n), TMPVAR);
> ++
> ++              /* Result can be arbitrarily long. Example:
> ++               *  printf "%99999s", "BOOM"
> ++               */
> ++              sv = *++f;
> +               *f = '\0';
> +-              arg = evaluate(nextarg(&n), v);
> +-
> +-              j = i;
> +-              if (c == 'c' || !c) {
> +-                      i += sprintf(b+i, s, is_numeric(arg) ?
> +-                                      (char)getvar_i(arg) :
> *getvar_s(arg));
> +-              } else if (c == 's') {
> +-                      s1 = getvar_s(arg);
> +-                      b = qrealloc(b, incr+i+strlen(s1), &bsize);
> +-                      i += sprintf(b+i, s, s1);
> ++              if (c == 'c') {
> ++                      char cc = is_numeric(arg) ? getvar_i(arg) :
> *getvar_s(arg);
> ++                      char *r = xasprintf(s, cc ? cc : '^' /* else
> strlen will be wrong */);
> ++                      slen = strlen(r);
> ++                      if (cc == '\0') /* if cc is NUL, re-format
> the string with it */
> ++                              sprintf(r, s, cc);
> ++                      s = r;
> +               } else {
> +-                      i += fmt_num(b+i, incr, s, getvar_i(arg),
> FALSE);
> ++                      if (c == 's') {
> ++                              s = xasprintf(s, getvar_s(arg));
> ++                      } else {
> ++                              double d = getvar_i(arg);
> ++                              if (strchr("diouxX", c)) {
> ++//TODO: make it wider here (%x -> %llx etc)?
> ++                                      s = xasprintf(s, (int)d);
> ++                              } else if (strchr("eEfFgGaA", c)) {
> ++                                      s = xasprintf(s, d);
> ++                              } else {
> ++//TODO: GNU Awk 5.0.1: printf "%W" prints "%W", does not error out
> ++                                      syntax_error(EMSG_INV_FMT);
> ++                              }
> ++                      }
> ++                      slen = strlen(s);
> +               }
> +-              *f = c1;
> +-
> +-              /* if there was an error while sprintf, return value
> is negative */
> +-              if (i < j)
> +-                      i = j;
> ++              *f = sv;
> ++ append:
> ++              if (i == 0) {
> ++                      b = s;
> ++                      i = slen;
> ++                      continue;
> ++              }
> ++ tail:
> ++              b = xrealloc(b, i + slen + 1);
> ++              strcpy(b + i, s);
> ++              i += slen;
> ++              if (!c) /* s is NOT allocated and this is the last
> part of string? */
> ++                      break;
> ++              free(s);
> +       }
> + 
> +       free(fmt);
> +-      nvfree(v);
> +-      b = xrealloc(b, i + 1);
> +-      b[i] = '\0';
> ++      //nvfree(tmpvar, 1);
> ++#undef TMPVAR
> ++
> ++#if ENABLE_FEATURE_AWK_GNU_EXTENSIONS
> ++      if (len)
> ++              *len = i;
> ++#endif
> +       return b;
> + }
> + 
> +@@ -2338,33 +2564,59 @@ static NOINLINE int do_mktime(const char
> *ds)
> +       return mktime(&then);
> + }
> + 
> ++/* Reduce stack usage in exec_builtin() by keeping match() code
> separate */
> ++static NOINLINE var *do_match(node *an1, const char *as0)
> ++{
> ++      regmatch_t pmatch[1];
> ++      regex_t sreg, *re;
> ++      int n, start, len;
> ++
> ++      re = as_regex(an1, &sreg);
> ++      n = regexec(re, as0, 1, pmatch, 0);
> ++      if (re == &sreg)
> ++              regfree(re);
> ++      start = 0;
> ++      len = -1;
> ++      if (n == 0) {
> ++              start = pmatch[0].rm_so + 1;
> ++              len = pmatch[0].rm_eo - pmatch[0].rm_so;
> ++      }
> ++      setvar_i(newvar("RLENGTH"), len);
> ++      return setvar_i(newvar("RSTART"), start);
> ++}
> ++
> ++/* Reduce stack usage in evaluate() by keeping builtins' code
> separate */
> + static NOINLINE var *exec_builtin(node *op, var *res)
> + {
> + #define tspl (G.exec_builtin__tspl)
> + 
> +-      var *tv;
> ++      var *tmpvars;
> +       node *an[4];
> +       var *av[4];
> +       const char *as[4];
> +-      regmatch_t pmatch[2];
> +-      regex_t sreg, *re;
> +       node *spl;
> +       uint32_t isr, info;
> +       int nargs;
> +       time_t tt;
> +       int i, l, ll, n;
> + 
> +-      tv = nvalloc(4);
> ++      tmpvars = nvalloc(4);
> ++#define TMPVAR0 (tmpvars)
> ++#define TMPVAR1 (tmpvars + 1)
> ++#define TMPVAR2 (tmpvars + 2)
> ++#define TMPVAR3 (tmpvars + 3)
> ++#define TMPVAR(i) (tmpvars + (i))
> +       isr = info = op->info;
> +       op = op->l.n;
> + 
> +       av[2] = av[3] = NULL;
> +       for (i = 0; i < 4 && op; i++) {
> +               an[i] = nextarg(&op);
> +-              if (isr & 0x09000000)
> +-                      av[i] = evaluate(an[i], &tv[i]);
> +-              if (isr & 0x08000000)
> +-                      as[i] = getvar_s(av[i]);
> ++              if (isr & 0x09000000) {
> ++                      av[i] = evaluate(an[i], TMPVAR(i));
> ++                      if (isr & 0x08000000)
> ++                              as[i] = getvar_s(av[i]);
> ++              }
> +               isr >>= 1;
> +       }
> + 
> +@@ -2386,8 +2638,8 @@ static NOINLINE var *exec_builtin(node *op,
> var *res)
> +               char *s, *s1;
> + 
> +               if (nargs > 2) {
> +-                      spl = (an[2]->info & OPCLSMASK) == OC_REGEXP
> ?
> +-                              an[2] :
> mk_splitter(getvar_s(evaluate(an[2], &tv[2])), &tspl);
> ++                      spl = (an[2]->info == TI_REGEXP) ? an[2]
> ++                              :
> mk_splitter(getvar_s(evaluate(an[2], TMPVAR2)), &tspl);
> +               } else {
> +                       spl = &fsplitter.n;
> +               }
> +@@ -2501,20 +2753,7 @@ static NOINLINE var *exec_builtin(node *op,
> var *res)
> +               break;
> + 
> +       case B_ma:
> +-              re = as_regex(an[1], &sreg);
> +-              n = regexec(re, as[0], 1, pmatch, 0);
> +-              if (n == 0) {
> +-                      pmatch[0].rm_so++;
> +-                      pmatch[0].rm_eo++;
> +-              } else {
> +-                      pmatch[0].rm_so = 0;
> +-                      pmatch[0].rm_eo = -1;
> +-              }
> +-              setvar_i(newvar("RSTART"), pmatch[0].rm_so);
> +-              setvar_i(newvar("RLENGTH"), pmatch[0].rm_eo -
> pmatch[0].rm_so);
> +-              setvar_i(res, pmatch[0].rm_so);
> +-              if (re == &sreg)
> +-                      regfree(re);
> ++              res = do_match(an[1], as[0]);
> +               break;
> + 
> +       case B_ge:
> +@@ -2530,14 +2769,79 @@ static NOINLINE var *exec_builtin(node *op,
> var *res)
> +               break;
> +       }
> + 
> +-      nvfree(tv);
> ++      nvfree(tmpvars, 4);
> ++#undef TMPVAR0
> ++#undef TMPVAR1
> ++#undef TMPVAR2
> ++#undef TMPVAR3
> ++#undef TMPVAR
> ++
> +       return res;
> + #undef tspl
> + }
> + 
> ++/* if expr looks like "var=value", perform assignment and return 1,
> ++ * otherwise return 0 */
> ++static int is_assignment(const char *expr)
> ++{
> ++      char *exprc, *val;
> ++
> ++      val = (char*)endofname(expr);
> ++      if (val == (char*)expr || *val != '=') {
> ++              return FALSE;
> ++      }
> ++
> ++      exprc = xstrdup(expr);
> ++      val = exprc + (val - expr);
> ++      *val++ = '\0';
> ++
> ++      unescape_string_in_place(val);
> ++      setvar_u(newvar(exprc), val);
> ++      free(exprc);
> ++      return TRUE;
> ++}
> ++
> ++/* switch to next input file */
> ++static rstream *next_input_file(void)
> ++{
> ++#define rsm          (G.next_input_file__rsm)
> ++#define files_happen (G.next_input_file__files_happen)
> ++
> ++      const char *fname, *ind;
> ++
> ++      if (rsm.F)
> ++              fclose(rsm.F);
> ++      rsm.F = NULL;
> ++      rsm.pos = rsm.adv = 0;
> ++
> ++      for (;;) {
> ++              if (getvar_i(intvar[ARGIND])+1 >=
> getvar_i(intvar[ARGC])) {
> ++                      if (files_happen)
> ++                              return NULL;
> ++                      fname = "-";
> ++                      rsm.F = stdin;
> ++                      break;
> ++              }
> ++              ind = getvar_s(incvar(intvar[ARGIND]));
> ++              fname = getvar_s(findvar(iamarray(intvar[ARGV]),
> ind));
> ++              if (fname && *fname && !is_assignment(fname)) {
> ++                      rsm.F = xfopen_stdin(fname);
> ++                      break;
> ++              }
> ++      }
> ++
> ++      files_happen = TRUE;
> ++      setvar_s(intvar[FILENAME], fname);
> ++      return &rsm;
> ++#undef rsm
> ++#undef files_happen
> ++}
> ++
> + /*
> +  * Evaluate node - the heart of the program. Supplied with subtree
> +- * and place where to store result. returns ptr to result.
> ++ * and "res" variable to assign the result to if we evaluate an
> expression.
> ++ * If node refers to e.g. a variable or a field, no assignment
> happens.
> ++ * Return ptr to the result (which may or may not be the "res"
> variable!)
> +  */
> + #define XC(n) ((n) >> 8)
> + 
> +@@ -2549,14 +2853,16 @@ static var *evaluate(node *op, var *res)
> + #define seed   (G.evaluate__seed)
> + #define sreg   (G.evaluate__sreg)
> + 
> +-      var *v1;
> ++      var *tmpvars;
> + 
> +       if (!op)
> +               return setvar_s(res, NULL);
> + 
> +       debug_printf_eval("entered %s()\n", __func__);
> + 
> +-      v1 = nvalloc(2);
> ++      tmpvars = nvalloc(2);
> ++#define TMPVAR0 (tmpvars)
> ++#define TMPVAR1 (tmpvars + 1)
> + 
> +       while (op) {
> +               struct {
> +@@ -2578,48 +2884,35 @@ static var *evaluate(node *op, var *res)
> +               op1 = op->l.n;
> +               debug_printf_eval("opinfo:%08x opn:%08x\n", opinfo,
> opn);
> + 
> +-              /* "delete" is special:
> +-               * "delete array[var--]" must evaluate index expr
> only once,
> +-               * must not evaluate it in "execute inevitable
> things" part.
> +-               */
> +-              if (XC(opinfo & OPCLSMASK) == XC(OC_DELETE)) {
> +-                      uint32_t info = op1->info & OPCLSMASK;
> +-                      var *v;
> +-
> +-                      debug_printf_eval("DELETE\n");
> +-                      if (info == OC_VAR) {
> +-                              v = op1->l.v;
> +-                      } else if (info == OC_FNARG) {
> +-                              v = &fnargs[op1->l.aidx];
> +-                      } else {
> +-                              syntax_error(EMSG_NOT_ARRAY);
> ++              /* execute inevitable things */
> ++              if (opinfo & OF_RES1) {
> ++                      if ((opinfo & OF_REQUIRED) && !op1)
> ++                              syntax_error(EMSG_TOO_FEW_ARGS);
> ++                      L.v = evaluate(op1, TMPVAR0);
> ++                      if (opinfo & OF_STR1) {
> ++                              L.s = getvar_s(L.v);
> ++                              debug_printf_eval("L.s:'%s'\n", L.s);
> +                       }
> +-                      if (op1->r.n) { /* array ref? */
> +-                              const char *s;
> +-                              s = getvar_s(evaluate(op1->r.n, v1));
> +-                              hash_remove(iamarray(v), s);
> +-                      } else {
> +-                              clear_array(iamarray(v));
> ++                      if (opinfo & OF_NUM1) {
> ++                              L_d = getvar_i(L.v);
> ++                              debug_printf_eval("L_d:%f\n", L_d);
> +                       }
> +-                      goto next;
> +               }
> +-
> +-              /* execute inevitable things */
> +-              if (opinfo & OF_RES1)
> +-                      L.v = evaluate(op1, v1);
> +-              if (opinfo & OF_RES2)
> +-                      R.v = evaluate(op->r.n, v1+1);
> +-              if (opinfo & OF_STR1) {
> +-                      L.s = getvar_s(L.v);
> +-                      debug_printf_eval("L.s:'%s'\n", L.s);
> +-              }
> +-              if (opinfo & OF_STR2) {
> +-                      R.s = getvar_s(R.v);
> +-                      debug_printf_eval("R.s:'%s'\n", R.s);
> +-              }
> +-              if (opinfo & OF_NUM1) {
> +-                      L_d = getvar_i(L.v);
> +-                      debug_printf_eval("L_d:%f\n", L_d);
> ++              /* NB: Must get string/numeric values of L (done
> above)
> ++               * _before_ evaluate()'ing R.v: if both L and R are
> $NNNs,
> ++               * and right one is large, then L.v points to
> Fields[NNN1],
> ++               * second evaluate() reallocates and moves (!)
> Fields[],
> ++               * R.v points to Fields[NNN2] but L.v now points to
> freed mem!
> ++               * (Seen trying to evaluate "$444 $44444")
> ++               */
> ++              if (opinfo & OF_RES2) {
> ++                      R.v = evaluate(op->r.n, TMPVAR1);
> ++                      //TODO: L.v may be invalid now, set L.v to
> NULL to catch bugs?
> ++                      //L.v = NULL;
> ++                      if (opinfo & OF_STR2) {
> ++                              R.s = getvar_s(R.v);
> ++                              debug_printf_eval("R.s:'%s'\n", R.s);
> ++                      }
> +               }
> + 
> +               debug_printf_eval("switch(0x%x)\n", XC(opinfo &
> OPCLSMASK));
> +@@ -2629,7 +2922,8 @@ static var *evaluate(node *op, var *res)
> + 
> +               /* test pattern */
> +               case XC( OC_TEST ):
> +-                      if ((op1->info & OPCLSMASK) == OC_COMMA) {
> ++                      debug_printf_eval("TEST\n");
> ++                      if (op1->info == TI_COMMA) {
> +                               /* it's range pattern */
> +                               if ((opinfo & OF_CHECKED) ||
> ptest(op1->l.n)) {
> +                                       op->info |= OF_CHECKED;
> +@@ -2646,25 +2940,32 @@ static var *evaluate(node *op, var *res)
> + 
> +               /* just evaluate an expression, also used as
> unconditional jump */
> +               case XC( OC_EXEC ):
> ++                      debug_printf_eval("EXEC\n");
> +                       break;
> + 
> +               /* branch, used in if-else and various loops */
> +               case XC( OC_BR ):
> ++                      debug_printf_eval("BR\n");
> +                       op = istrue(L.v) ? op->a.n : op->r.n;
> +                       break;
> + 
> +               /* initialize for-in loop */
> +               case XC( OC_WALKINIT ):
> ++                      debug_printf_eval("WALKINIT\n");
> +                       hashwalk_init(L.v, iamarray(R.v));
> +                       break;
> + 
> +               /* get next array item */
> +               case XC( OC_WALKNEXT ):
> ++                      debug_printf_eval("WALKNEXT\n");
> +                       op = hashwalk_next(L.v) ? op->a.n : op->r.n;
> +                       break;
> + 
> +               case XC( OC_PRINT ):
> +-              case XC( OC_PRINTF ): {
> ++                      debug_printf_eval("PRINT /\n");
> ++              case XC( OC_PRINTF ):
> ++                      debug_printf_eval("PRINTF\n");
> ++              {
> +                       FILE *F = stdout;
> + 
> +                       if (op->r.n) {
> +@@ -2682,55 +2983,94 @@ static var *evaluate(node *op, var *res)
> +                               F = rsm->F;
> +                       }
> + 
> ++                      /* Can't just check 'opinfo == OC_PRINT'
> here, parser ORs
> ++                       * additional bits to opinfos of print/printf
> with redirects
> ++                       */
> +                       if ((opinfo & OPCLSMASK) == OC_PRINT) {
> +                               if (!op1) {
> +                                       fputs(getvar_s(intvar[F0]),
> F);
> +                               } else {
> +-                                      while (op1) {
> +-                                              var *v =
> evaluate(nextarg(&op1), v1);
> ++                                      for (;;) {
> ++                                              var *v =
> evaluate(nextarg(&op1), TMPVAR0);
> +                                               if (v->type &
> VF_NUMBER) {
> +-
>                                                       fmt_num(g_buf,
> MAXVARFMT, getvar_s(intvar[OFMT]),
> +-
>                                                                      
>  getvar_i(v), TRUE);
> ++                                                      fmt_num(getva
> r_s(intvar[OFMT]),
> ++                                                                   
>    getvar_i(v));
> +                                                       fputs(g_buf,
> F);
> +                                               } else {
> +                                                       fputs(getvar_
> s(v), F);
> +                                               }
> +-
> +-                                              if (op1)
> +-
>                                                       fputs(getvar_s(
> intvar[OFS]), F);
> ++                                              if (!op1)
> ++                                                      break;
> ++                                              fputs(getvar_s(intvar
> [OFS]), F);
> +                                       }
> +                               }
> +                               fputs(getvar_s(intvar[ORS]), F);
> +-
> +-                      } else {        /* OC_PRINTF */
> +-                              char *s = awk_printf(op1);
> ++                      } else {        /* PRINTF */
> ++                              IF_FEATURE_AWK_GNU_EXTENSIONS(size_t
> len;)
> ++                              char *s = awk_printf(op1, &len);
> ++#if ENABLE_FEATURE_AWK_GNU_EXTENSIONS
> ++                              fwrite(s, len, 1, F);
> ++#else
> +                               fputs(s, F);
> ++#endif
> +                               free(s);
> +                       }
> +                       fflush(F);
> +                       break;
> +               }
> + 
> +-              /* case XC( OC_DELETE ): - moved to happen before arg
> evaluation */
> ++              case XC( OC_DELETE ):
> ++                      debug_printf_eval("DELETE\n");
> ++              {
> ++                      /* "delete" is special:
> ++                       * "delete array[var--]" must evaluate index
> expr only once.
> ++                       */
> ++                      uint32_t info = op1->info & OPCLSMASK;
> ++                      var *v;
> ++
> ++                      if (info == OC_VAR) {
> ++                              v = op1->l.v;
> ++                      } else if (info == OC_FNARG) {
> ++                              v = &fnargs[op1->l.aidx];
> ++                      } else {
> ++                              syntax_error(EMSG_NOT_ARRAY);
> ++                      }
> ++                      if (op1->r.n) { /* array ref? */
> ++                              const char *s;
> ++                              s = getvar_s(evaluate(op1->r.n,
> TMPVAR0));
> ++                              hash_remove(iamarray(v), s);
> ++                      } else {
> ++                              clear_array(iamarray(v));
> ++                      }
> ++                      break;
> ++              }
> + 
> +               case XC( OC_NEWSOURCE ):
> ++                      debug_printf_eval("NEWSOURCE\n");
> +                       g_progname = op->l.new_progname;
> +                       break;
> + 
> +               case XC( OC_RETURN ):
> ++                      debug_printf_eval("RETURN\n");
> +                       copyvar(res, L.v);
> +                       break;
> + 
> +               case XC( OC_NEXTFILE ):
> ++                      debug_printf_eval("NEXTFILE\n");
> +                       nextfile = TRUE;
> +               case XC( OC_NEXT ):
> ++                      debug_printf_eval("NEXT\n");
> +                       nextrec = TRUE;
> +               case XC( OC_DONE ):
> ++                      debug_printf_eval("DONE\n");
> +                       clrvar(res);
> +                       break;
> + 
> +               case XC( OC_EXIT ):
> +-                      awk_exit(L_d);
> ++                      debug_printf_eval("EXIT\n");
> ++                      if (op1)
> ++                              G.exitcode = (int)L_d;
> ++                      awk_exit();
> + 
> +               /* -- recursive node type -- */
> + 
> +@@ -2749,15 +3089,18 @@ static var *evaluate(node *op, var *res)
> +                       break;
> + 
> +               case XC( OC_IN ):
> ++                      debug_printf_eval("IN\n");
> +                       setvar_i(res, hash_search(iamarray(R.v), L.s)
> ? 1 : 0);
> +                       break;
> + 
> +               case XC( OC_REGEXP ):
> ++                      debug_printf_eval("REGEXP\n");
> +                       op1 = op;
> +                       L.s = getvar_s(intvar[F0]);
> +                       goto re_cont;
> + 
> +               case XC( OC_MATCH ):
> ++                      debug_printf_eval("MATCH\n");
> +                       op1 = op->r.n;
> +  re_cont:
> +                       {
> +@@ -2772,61 +3115,80 @@ static var *evaluate(node *op, var *res)
> +               case XC( OC_MOVE ):
> +                       debug_printf_eval("MOVE\n");
> +                       /* if source is a temporary string, jusk
> relink it to dest */
> +-//Disabled: if R.v is numeric but happens to have cached R.v-
> >string,
> +-//then L.v ends up being a string, which is wrong
> +-//                    if (R.v == v1+1 && R.v->string) {
> +-//                            res = setvar_p(L.v, R.v->string);
> +-//                            R.v->string = NULL;
> +-//                    } else {
> ++                      if (R.v == TMPVAR1
> ++                       && !(R.v->type & VF_NUMBER)
> ++                              /* Why check !NUMBER? if R.v is a
> number but has cached R.v->string,
> ++                               * L.v ends up a string, which is
> wrong */
> ++                       /*&& R.v->string - always not NULL (right?)
> */
> ++                      ) {
> ++                              res = setvar_p(L.v, R.v->string); /*
> avoids strdup */
> ++                              R.v->string = NULL;
> ++                      } else {
> +                               res = copyvar(L.v, R.v);
> +-//                    }
> ++                      }
> +                       break;
> + 
> +               case XC( OC_TERNARY ):
> +-                      if ((op->r.n->info & OPCLSMASK) != OC_COLON)
> ++                      debug_printf_eval("TERNARY\n");
> ++                      if (op->r.n->info != TI_COLON)
> +                               syntax_error(EMSG_POSSIBLE_ERROR);
> +                       res = evaluate(istrue(L.v) ? op->r.n->l.n :
> op->r.n->r.n, res);
> +                       break;
> + 
> +               case XC( OC_FUNC ): {
> +-                      var *vbeg, *v;
> ++                      var *argvars, *sv_fnargs;
> +                       const char *sv_progname;
> ++                      int nargs, i;
> + 
> +-                      /* The body might be empty, still has to eval
> the args */
> +-                      if (!op->r.n->info && !op->r.f->body.first)
> ++                      debug_printf_eval("FUNC\n");
> ++
> ++                      if (!op->r.f->defined)
> +                               syntax_error(EMSG_UNDEF_FUNC);
> + 
> +-                      vbeg = v = nvalloc(op->r.f->nargs + 1);
> ++                      /* The body might be empty, still has to eval
> the args */
> ++                      nargs = op->r.f->nargs;
> ++                      argvars = nvalloc(nargs);
> ++                      i = 0;
> +                       while (op1) {
> +-                              var *arg = evaluate(nextarg(&op1),
> v1);
> +-                              copyvar(v, arg);
> +-                              v->type |= VF_CHILD;
> +-                              v->x.parent = arg;
> +-                              if (++v - vbeg >= op->r.f->nargs)
> +-                                      break;
> ++                              var *arg = evaluate(nextarg(&op1),
> TMPVAR0);
> ++                              if (i == nargs) {
> ++                                      /* call with more arguments
> than function takes.
> ++                                       * (gawk warns: "warning:
> function 'f' called with more arguments than declared").
> ++                                       * They are still evaluated,
> but discarded: */
> ++                                      clrvar(arg);
> ++                                      continue;
> ++                              }
> ++                              copyvar(&argvars[i], arg);
> ++                              argvars[i].type |= VF_CHILD;
> ++                              argvars[i].x.parent = arg;
> ++                              i++;
> +                       }
> + 
> +-                      v = fnargs;
> +-                      fnargs = vbeg;
> ++                      sv_fnargs = fnargs;
> +                       sv_progname = g_progname;
> + 
> ++                      fnargs = argvars;
> +                       res = evaluate(op->r.f->body.first, res);
> ++                      nvfree(argvars, nargs);
> + 
> +                       g_progname = sv_progname;
> +-                      nvfree(fnargs);
> +-                      fnargs = v;
> ++                      fnargs = sv_fnargs;
> + 
> +                       break;
> +               }
> + 
> +               case XC( OC_GETLINE ):
> +-              case XC( OC_PGETLINE ): {
> ++                      debug_printf_eval("GETLINE /\n");
> ++              case XC( OC_PGETLINE ):
> ++                      debug_printf_eval("PGETLINE\n");
> ++              {
> +                       rstream *rsm;
> +                       int i;
> + 
> +                       if (op1) {
> +                               rsm = newfile(L.s);
> +                               if (!rsm->F) {
> ++                                      /* NB: can't use "opinfo ==
> TI_PGETLINE", would break "cmd" | getline */
> +                                       if ((opinfo & OPCLSMASK) ==
> OC_PGETLINE) {
> +                                               rsm->F = popen(L.s,
> "r");
> +                                               rsm->is_pipe = TRUE;
> +@@ -2861,16 +3223,34 @@ static var *evaluate(node *op, var *res)
> +               /* simple builtins */
> +               case XC( OC_FBLTIN ): {
> +                       double R_d = R_d; /* for compiler */
> ++                      debug_printf_eval("FBLTIN\n");
> ++
> ++                      if (op1 && op1->info == TI_COMMA)
> ++                              /* Simple builtins take one arg
> maximum */
> ++                              syntax_error("Too many arguments");
> + 
> +                       switch (opn) {
> +                       case F_in:
> +                               R_d = (long long)L_d;
> +                               break;
> + 
> +-                      case F_rn:
> +-                              R_d = (double)rand() /
> (double)RAND_MAX;
> ++                      case F_rn: /*rand*/
> ++                              if (op1)
> ++                                      syntax_error("Too many
> arguments");
> ++                      {
> ++#if RAND_MAX >= 0x7fffffff
> ++                              uint32_t u = ((uint32_t)rand() << 16)
> ^ rand();
> ++                              uint64_t v = ((uint64_t)rand() << 32)
> | u;
> ++                              /* the above shift+or is optimized
> out on 32-bit arches */
> ++# if RAND_MAX > 0x7fffffff
> ++                              v &= 0x7fffffffffffffffULL;
> ++# endif
> ++                              R_d = (double)v /
> 0x8000000000000000ULL;
> ++#else
> ++# error Not implemented for this value of RAND_MAX
> ++#endif
> +                               break;
> +-
> ++                      }
> +                       case F_co:
> +                               if (ENABLE_FEATURE_AWK_LIBM) {
> +                                       R_d = cos(L_d);
> +@@ -2910,7 +3290,9 @@ static var *evaluate(node *op, var *res)
> +                               srand(seed);
> +                               break;
> + 
> +-                      case F_ti:
> ++                      case F_ti: /*systime*/
> ++                              if (op1)
> ++                                      syntax_error("Too many
> arguments");
> +                               R_d = time(NULL);
> +                               break;
> + 
> +@@ -2949,7 +3331,7 @@ static var *evaluate(node *op, var *res)
> +                               rstream *rsm;
> +                               int err = 0;
> +                               rsm = (rstream *)hash_search(fdhash,
> L.s);
> +-                              debug_printf_eval("OC_FBLTIN F_cl
> rsm:%p\n", rsm);
> ++                              debug_printf_eval("OC_FBLTIN close:
> op1:%p s:'%s' rsm:%p\n", op1, L.s, rsm);
> +                               if (rsm) {
> +                                       debug_printf_eval("OC_FBLTIN
> F_cl "
> +                                               "rsm->is_pipe:%d, -
> >F:%p\n",
> +@@ -2960,6 +3342,11 @@ static var *evaluate(node *op, var *res)
> +                                        */
> +                                       if (rsm->F)
> +                                               err = rsm->is_pipe ?
> pclose(rsm->F) : fclose(rsm->F);
> ++//TODO: fix this case:
> ++// $ awk 'BEGIN { print close(""); print ERRNO }'
> ++// -1
> ++// close of redirection that was never opened
> ++// (we print 0, 0)
> +                                       free(rsm->buffer);
> +                                       hash_remove(fdhash, L.s);
> +                               }
> +@@ -2974,14 +3361,18 @@ static var *evaluate(node *op, var *res)
> +               }
> + 
> +               case XC( OC_BUILTIN ):
> ++                      debug_printf_eval("BUILTIN\n");
> +                       res = exec_builtin(op, res);
> +                       break;
> + 
> +               case XC( OC_SPRINTF ):
> +-                      setvar_p(res, awk_printf(op1));
> ++                      debug_printf_eval("SPRINTF\n");
> ++                      setvar_p(res, awk_printf(op1, NULL));
> +                       break;
> + 
> +-              case XC( OC_UNARY ): {
> ++              case XC( OC_UNARY ):
> ++                      debug_printf_eval("UNARY\n");
> ++              {
> +                       double Ld, R_d;
> + 
> +                       Ld = R_d = getvar_i(R.v);
> +@@ -3011,7 +3402,9 @@ static var *evaluate(node *op, var *res)
> +                       break;
> +               }
> + 
> +-              case XC( OC_FIELD ): {
> ++              case XC( OC_FIELD ):
> ++                      debug_printf_eval("FIELD\n");
> ++              {
> +                       int i = (int)getvar_i(R.v);
> +                       if (i < 0)
> +                               syntax_error(EMSG_NEGATIVE_FIELD);
> +@@ -3028,26 +3421,33 @@ static var *evaluate(node *op, var *res)
> + 
> +               /* concatenation (" ") and index joining (",") */
> +               case XC( OC_CONCAT ):
> ++                      debug_printf_eval("CONCAT /\n");
> +               case XC( OC_COMMA ): {
> +                       const char *sep = "";
> +-                      if ((opinfo & OPCLSMASK) == OC_COMMA)
> ++                      debug_printf_eval("COMMA\n");
> ++                      if (opinfo == TI_COMMA)
> +                               sep = getvar_s(intvar[SUBSEP]);
> +                       setvar_p(res, xasprintf("%s%s%s", L.s, sep,
> R.s));
> +                       break;
> +               }
> + 
> +               case XC( OC_LAND ):
> ++                      debug_printf_eval("LAND\n");
> +                       setvar_i(res, istrue(L.v) ? ptest(op->r.n) :
> 0);
> +                       break;
> + 
> +               case XC( OC_LOR ):
> ++                      debug_printf_eval("LOR\n");
> +                       setvar_i(res, istrue(L.v) ? 1 : ptest(op-
> >r.n));
> +                       break;
> + 
> +               case XC( OC_BINARY ):
> +-              case XC( OC_REPLACE ): {
> ++                      debug_printf_eval("BINARY /\n");
> ++              case XC( OC_REPLACE ):
> ++                      debug_printf_eval("REPLACE\n");
> ++              {
> +                       double R_d = getvar_i(R.v);
> +-                      debug_printf_eval("BINARY/REPLACE: R_d:%f
> opn:%c\n", R_d, opn);
> ++                      debug_printf_eval("R_d:%f opn:%c\n", R_d,
> opn);
> +                       switch (opn) {
> +                       case '+':
> +                               L_d += R_d;
> +@@ -3083,6 +3483,7 @@ static var *evaluate(node *op, var *res)
> +               case XC( OC_COMPARE ): {
> +                       int i = i; /* for compiler */
> +                       double Ld;
> ++                      debug_printf_eval("COMPARE\n");
> + 
> +                       if (is_numeric(L.v) && is_numeric(R.v)) {
> +                               Ld = getvar_i(L.v) - getvar_i(R.v);
> +@@ -3109,7 +3510,7 @@ static var *evaluate(node *op, var *res)
> +               default:
> +                       syntax_error(EMSG_POSSIBLE_ERROR);
> +               } /* switch */
> +- next:
> ++
> +               if ((opinfo & OPCLSMASK) <= SHIFT_TIL_THIS)
> +                       op = op->a.n;
> +               if ((opinfo & OPCLSMASK) >= RECUR_FROM_THIS)
> +@@ -3118,7 +3519,10 @@ static var *evaluate(node *op, var *res)
> +                       break;
> +       } /* while (op) */
> + 
> +-      nvfree(v1);
> ++      nvfree(tmpvars, 2);
> ++#undef TMPVAR0
> ++#undef TMPVAR1
> ++
> +       debug_printf_eval("returning from %s(): %p\n", __func__,
> res);
> +       return res;
> + #undef fnargs
> +@@ -3126,25 +3530,21 @@ static var *evaluate(node *op, var *res)
> + #undef sreg
> + }
> + 
> +-
> + /* -------- main & co. -------- */
> + 
> +-static int awk_exit(int r)
> ++static int awk_exit(void)
> + {
> +-      var tv;
> +       unsigned i;
> +-      hash_item *hi;
> +-
> +-      zero_out_var(&tv);
> + 
> +       if (!exiting) {
> +               exiting = TRUE;
> +               nextrec = FALSE;
> +-              evaluate(endseq.first, &tv);
> ++              evaluate(endseq.first, &G.exit__tmpvar);
> +       }
> + 
> +       /* waiting for children */
> +       for (i = 0; i < fdhash->csize; i++) {
> ++              hash_item *hi;
> +               hi = fdhash->items[i];
> +               while (hi) {
> +                       if (hi->data.rs.F && hi->data.rs.is_pipe)
> +@@ -3153,65 +3553,7 @@ static int awk_exit(int r)
> +               }
> +       }
> + 
> +-      exit(r);
> +-}
> +-
> +-/* if expr looks like "var=value", perform assignment and return 1,
> +- * otherwise return 0 */
> +-static int is_assignment(const char *expr)
> +-{
> +-      char *exprc, *val;
> +-
> +-      if (!isalnum_(*expr) || (val = strchr(expr, '=')) == NULL) {
> +-              return FALSE;
> +-      }
> +-
> +-      exprc = xstrdup(expr);
> +-      val = exprc + (val - expr);
> +-      *val++ = '\0';
> +-
> +-      unescape_string_in_place(val);
> +-      setvar_u(newvar(exprc), val);
> +-      free(exprc);
> +-      return TRUE;
> +-}
> +-
> +-/* switch to next input file */
> +-static rstream *next_input_file(void)
> +-{
> +-#define rsm          (G.next_input_file__rsm)
> +-#define files_happen (G.next_input_file__files_happen)
> +-
> +-      FILE *F;
> +-      const char *fname, *ind;
> +-
> +-      if (rsm.F)
> +-              fclose(rsm.F);
> +-      rsm.F = NULL;
> +-      rsm.pos = rsm.adv = 0;
> +-
> +-      for (;;) {
> +-              if (getvar_i(intvar[ARGIND])+1 >=
> getvar_i(intvar[ARGC])) {
> +-                      if (files_happen)
> +-                              return NULL;
> +-                      fname = "-";
> +-                      F = stdin;
> +-                      break;
> +-              }
> +-              ind = getvar_s(incvar(intvar[ARGIND]));
> +-              fname = getvar_s(findvar(iamarray(intvar[ARGV]),
> ind));
> +-              if (fname && *fname && !is_assignment(fname)) {
> +-                      F = xfopen_stdin(fname);
> +-                      break;
> +-              }
> +-      }
> +-
> +-      files_happen = TRUE;
> +-      setvar_s(intvar[FILENAME], fname);
> +-      rsm.F = F;
> +-      return &rsm;
> +-#undef rsm
> +-#undef files_happen
> ++      exit(G.exitcode);
> + }
> + 
> + int awk_main(int argc, char **argv) MAIN_EXTERNALLY_VISIBLE;
> +@@ -3224,12 +3566,7 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> + #if ENABLE_FEATURE_AWK_GNU_EXTENSIONS
> +       llist_t *list_e = NULL;
> + #endif
> +-      int i, j;
> +-      var *v;
> +-      var tv;
> +-      char **envp;
> +-      char *vnames = (char *)vNames; /* cheat */
> +-      char *vvalues = (char *)vValues;
> ++      int i;
> + 
> +       INIT_G();
> + 
> +@@ -3238,48 +3575,43 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +       if (ENABLE_LOCALE_SUPPORT)
> +               setlocale(LC_NUMERIC, "C");
> + 
> +-      zero_out_var(&tv);
> +-
> +-      /* allocate global buffer */
> +-      g_buf = xmalloc(MAXVARFMT + 1);
> +-
> +-      vhash = hash_init();
> +-      ahash = hash_init();
> +-      fdhash = hash_init();
> +-      fnhash = hash_init();
> +-
> +       /* initialize variables */
> +-      for (i = 0; *vnames; i++) {
> +-              intvar[i] = v = newvar(nextword(&vnames));
> +-              if (*vvalues != '\377')
> +-                      setvar_s(v, nextword(&vvalues));
> +-              else
> +-                      setvar_i(v, 0);
> +-
> +-              if (*vnames == '*') {
> +-                      v->type |= VF_SPECIAL;
> +-                      vnames++;
> ++      vhash = hash_init();
> ++      {
> ++              char *vnames = (char *)vNames; /* cheat */
> ++              char *vvalues = (char *)vValues;
> ++              for (i = 0; *vnames; i++) {
> ++                      var *v;
> ++                      intvar[i] = v = newvar(nextword(&vnames));
> ++                      if (*vvalues != '\377')
> ++                              setvar_s(v, nextword(&vvalues));
> ++                      else
> ++                              setvar_i(v, 0);
> ++
> ++                      if (*vnames == '*') {
> ++                              v->type |= VF_SPECIAL;
> ++                              vnames++;
> ++                      }
> +               }
> +       }
> + 
> +       handle_special(intvar[FS]);
> +       handle_special(intvar[RS]);
> + 
> +-      newfile("/dev/stdin")->F = stdin;
> +-      newfile("/dev/stdout")->F = stdout;
> +-      newfile("/dev/stderr")->F = stderr;
> +-
> +       /* Huh, people report that sometimes environ is NULL. Oh
> well. */
> +-      if (environ) for (envp = environ; *envp; envp++) {
> +-              /* environ is writable, thus we don't strdup it
> needlessly */
> +-              char *s = *envp;
> +-              char *s1 = strchr(s, '=');
> +-              if (s1) {
> +-                      *s1 = '\0';
> +-                      /* Both findvar and setvar_u take const char*
> +-                       * as 2nd arg -> environment is not trashed
> */
> +-                      setvar_u(findvar(iamarray(intvar[ENVIRON]),
> s), s1 + 1);
> +-                      *s1 = '=';
> ++      if (environ) {
> ++              char **envp;
> ++              for (envp = environ; *envp; envp++) {
> ++                      /* environ is writable, thus we don't strdup
> it needlessly */
> ++                      char *s = *envp;
> ++                      char *s1 = strchr(s, '=');
> ++                      if (s1) {
> ++                              *s1 = '\0';
> ++                              /* Both findvar and setvar_u take
> const char*
> ++                               * as 2nd arg -> environment is not
> trashed */
> ++                              setvar_u(findvar(iamarray(intvar[ENVI
> RON]), s), s1 + 1);
> ++                              *s1 = '=';
> ++                      }
> +               }
> +       }
> +       opt = getopt32(argv, OPTSTR_AWK, &opt_F, &list_v, &list_f,
> IF_FEATURE_AWK_GNU_EXTENSIONS(&list_e,) NULL);
> +@@ -3295,20 +3627,19 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +               if (!is_assignment(llist_pop(&list_v)))
> +                       bb_show_usage();
> +       }
> ++
> ++      /* Parse all supplied programs */
> ++      fnhash = hash_init();
> ++      ahash = hash_init();
> +       while (list_f) {
> +-              char *s = NULL;
> +-              FILE *from_file;
> ++              int fd;
> ++              char *s;
> + 
> +               g_progname = llist_pop(&list_f);
> +-              from_file = xfopen_stdin(g_progname);
> +-              /* one byte is reserved for some trick in next_token
> */
> +-              for (i = j = 1; j > 0; i += j) {
> +-                      s = xrealloc(s, i + 4096);
> +-                      j = fread(s + i, 1, 4094, from_file);
> +-              }
> +-              s[i] = '\0';
> +-              fclose(from_file);
> +-              parse_program(s + 1);
> ++              fd = xopen_stdin(g_progname);
> ++              s = xmalloc_read(fd, NULL); /* it's NUL-terminated */
> ++              close(fd);
> ++              parse_program(s);
> +               free(s);
> +       }
> +       g_progname = "cmd. line";
> +@@ -3317,11 +3648,23 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +               parse_program(llist_pop(&list_e));
> +       }
> + #endif
> ++//FIXME: preserve order of -e and -f
> ++//TODO: implement -i LIBRARY and -E FILE too, they are easy-ish
> +       if (!(opt & (OPT_f | OPT_e))) {
> +               if (!*argv)
> +                       bb_show_usage();
> +               parse_program(*argv++);
> +       }
> ++      /* Free unused parse structures */
> ++      //hash_free(fnhash); // ~250 bytes when empty, used only for
> function names
> ++      //^^^^^^^^^^^^^^^^^ does not work, hash_clear() inside SEGVs
> ++      // (IOW: hash_clear() assumes it's a hash of variables.
> fnhash is not).
> ++      free(fnhash->items);
> ++      free(fnhash);
> ++      fnhash = NULL; // debug
> ++      //hash_free(ahash); // empty after parsing, will reuse as
> fdhash instead of freeing
> ++
> ++      /* Parsing done, on to executing */
> + 
> +       /* fill in ARGV array */
> +       setari_u(intvar[ARGV], 0, "awk");
> +@@ -3330,9 +3673,14 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +               setari_u(intvar[ARGV], ++i, *argv++);
> +       setvar_i(intvar[ARGC], i + 1);
> + 
> +-      evaluate(beginseq.first, &tv);
> ++      //fdhash = ahash; // done via define
> ++      newfile("/dev/stdin")->F = stdin;
> ++      newfile("/dev/stdout")->F = stdout;
> ++      newfile("/dev/stderr")->F = stderr;
> ++
> ++      evaluate(beginseq.first, &G.main__tmpvar);
> +       if (!mainseq.first && !endseq.first)
> +-              awk_exit(EXIT_SUCCESS);
> ++              awk_exit();
> + 
> +       /* input file could already be opened in BEGIN block */
> +       if (!iF)
> +@@ -3347,7 +3695,7 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +                       nextrec = FALSE;
> +                       incvar(intvar[NR]);
> +                       incvar(intvar[FNR]);
> +-                      evaluate(mainseq.first, &tv);
> ++                      evaluate(mainseq.first, &G.main__tmpvar);
> + 
> +                       if (nextfile)
> +                               break;
> +@@ -3359,6 +3707,6 @@ int awk_main(int argc UNUSED_PARAM, char
> **argv)
> +               iF = next_input_file();
> +       }
> + 
> +-      awk_exit(EXIT_SUCCESS);
> ++      awk_exit();
> +       /*return 0;*/
> + }
> +-- 
> +2.33.0
> +
> diff --git a/meta/recipes-core/busybox/busybox/0002-man-fix-segfault-
> in-man-1.patch b/meta/recipes-core/busybox/busybox/0002-man-fix-
> segfault-in-man-1.patch
> new file mode 100644
> index 0000000000..4a930b7b6f
> --- /dev/null
> +++ b/meta/recipes-core/busybox/busybox/0002-man-fix-segfault-in-man-
> 1.patch
> @@ -0,0 +1,30 @@
> +From 4975cace9bf96bfde174f8bb5cc4068d2ea294d4 Mon Sep 17 00:00:00
> 2001
> +From: Denys Vlasenko <vda.linux@googlemail.com>
> +Date: Tue, 15 Jun 2021 14:47:46 +0200
> +Subject: [PATCH] man: fix segfault in "man 1"
> +
> +function                                             old     new  
> delta
> +man_main                                             942     954    
> +12
> +
> +Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
> +
> +Upstream-Status: Backport [4d4fc5ca5ee4f]
> +CVE: CVE-2021-42373
> +Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
> +---
> + miscutils/man.c | 2 +-
> + 1 file changed, 1 insertion(+), 1 deletion(-)
> +
> +diff --git a/miscutils/man.c b/miscutils/man.c
> +index 722f6641e..d319e8bba 100644
> +--- a/miscutils/man.c
> ++++ b/miscutils/man.c
> +@@ -324,7 +324,7 @@ int man_main(int argc UNUSED_PARAM, char **argv)
> + 
> +       /* is 1st ARG a SECTION? */
> +       sec_list = conf_sec_list;
> +-      if (is_section_name(conf_sec_list, *argv)) {
> ++      if (is_section_name(conf_sec_list, *argv) && argv[1]) {
> +               /* yes */
> +               sec_list = *argv++;
> +       }
> diff --git a/meta/recipes-core/busybox/busybox_1.33.2.bb
> b/meta/recipes-core/busybox/busybox_1.33.2.bb
> index 44c83ab83c..4a0d3b4556 100644
> --- a/meta/recipes-core/busybox/busybox_1.33.2.bb
> +++ b/meta/recipes-core/busybox/busybox_1.33.2.bb
> @@ -48,6 +48,8 @@ SRC_URI =
> "https://busybox.net/downloads/busybox-${PV}.tar.bz2;name=tarball \
>            
> file://0001-sysctl-ignore-EIO-of-stable_secret-below-proc-sys-ne.patch
>  \
>            
> file://0001-gen_build_files-Use-C-locale-when-calling-sed-on-glo.patch
>  \
>             file://0001-mktemp-add-tmpdir-option.patch \
> +           file://0001-awk-fix-CVEs.patch \
> +           file://0002-man-fix-segfault-in-man-1.patch \
>             "
>  SRC_URI_append_libc-musl = " file://musl.cfg "
>  
> 
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#160046):
> https://lists.openembedded.org/g/openembedded-core/message/160046
> Mute This Topic: https://lists.openembedded.org/mt/88007703/3616702
> Group Owner: openembedded-core+owner@lists.openembedded.org
> Unsubscribe:
> https://lists.openembedded.org/g/openembedded-core/unsub [
> anuj.mittal@intel.com]
> -=-=-=-=-=-=-=-=-=-=-=-
> 


  reply	other threads:[~2021-12-30 15:04 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-29  1:44 [OE-core][hardknott][PATCH V3] busybox: backport patches to fix CVEs Chen Qi
2021-12-30 15:03 ` Mittal, Anuj [this message]
2021-12-31  2:06   ` ChenQi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b57878c065e1c9cc29d590f90a2b76604472539c.camel@intel.com \
    --to=anuj.mittal@intel.com \
    --cc=Qi.Chen@windriver.com \
    --cc=openembedded-core@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox