From 8f3268f4af3ff85f7d4ec2f7907598f5c190e517 Mon Sep 17 00:00:00 2001 From: Glenn Randers-Pehrson Date: Sat, 10 May 2008 08:58:19 -0500 Subject: [PATCH] Imported from pngcrush-1.6.5.tar --- ChangeLog.txt | 3 + Makefile | 18 +- adler32.c | 44 +- compress.c | 5 +- crc32.c | 41 +- deflate.c | 64 +- deflate.h | 2 +- gzio.c | 138 +- infback.c | 80 +- inffast.c | 57 +- inflate.c | 123 +- inflate.h | 5 +- inftrees.c | 44 +- minigzip.c | 322 --- png.c | 145 +- png.h | 299 ++- pngconf.h | 147 +- pngcrush.c | 388 +++- pngcrush.h | 3 - pngerror.c | 78 +- pnggccrd.c | 5389 +------------------------------------------------ pngget.c | 88 +- pngmem.c | 18 +- pngpread.c | 234 ++- pngread.c | 174 +- pngrio.c | 5 +- pngrtran.c | 149 +- pngrutil.c | 208 +- pngset.c | 225 ++- pngtest.c | 91 +- pngtrans.c | 20 +- pngvcrd.c | 3905 +---------------------------------- pngwio.c | 10 +- pngwrite.c | 103 +- pngwutil.c | 150 +- trees.c | 54 +- zconf.h | 144 +- zlib.h | 92 +- zlibdefs.h | 13 + zutil.h | 22 +- 40 files changed, 2279 insertions(+), 10821 deletions(-) delete mode 100644 minigzip.c create mode 100644 zlibdefs.h diff --git a/ChangeLog.txt b/ChangeLog.txt index 772935809..5edbd6fea 100644 --- a/ChangeLog.txt +++ b/ChangeLog.txt @@ -1,5 +1,8 @@ Change log: +Version 1.6.5 (built with libpng-1.2.29 and zlib-1.2.3.2) + Discontinued adding a new gAMA chunk when writing sRGB chunk. + Version 1.6.4 (built with libpng-1.2.9rc1 and zlib-1.2.3) Fixed bug in handling of undocumented -trns_a option (Michal Politowski). Fixed bug with "nosave" handling of unknown chunks. diff --git a/Makefile b/Makefile index ca2f6d987..23895a76c 100644 --- a/Makefile +++ b/Makefile @@ -10,21 +10,19 @@ # macros -------------------------------------------------------------------- -GAS_VERSION := $(shell as --version | grep "GNU assembler" | sed -e 's/GNU assembler //' -e 's/ .*//') - # uncomment these 2 lines only if you are using an external copy of zlib: #ZINC = ../../zlib #ZLIB = ../../zlib -CC = gcc -LD = gcc +# CC = gcc +CC = gcc-4.3.0 +LD = gcc-4.3.0 +# LD = gcc RM = rm -f #CFLAGS = -I. -O -Wall -#CFLAGS = -I. -O3 -fomit-frame-pointer -Wall -DPNG_USE_PNGGCCRD -CFLAGS = -I. -O3 -fomit-frame-pointer -Wall -DPNG_USE_PNGGCCRD \ - -DGAS_VERSION="\"${GAS_VERSION}\"" -#CFLAGS = -I${ZINC} -I. -O3 -fomit-frame-pointer -Wall -DPNG_USE_PNGGCCRD \ -# -DGAS_VERSION="\"${GAS_VERSION}\"" +#CFLAGS = -I. -O3 -fomit-frame-pointer -Wall +CFLAGS = -I. -O3 -fomit-frame-pointer -Wall +#CFLAGS = -I${ZINC} -I. -O3 -fomit-frame-pointer -Wall # [note that -Wall is a gcc-specific compilation flag ("all warnings on")] LDFLAGS = O = .o @@ -44,7 +42,7 @@ ZOBJS = adler32$(O) compress$(O) crc32$(O) deflate$(O) gzio$(O) \ OBJS = pngcrush$(O) png$(O) pngerror$(O) pngget$(O) pngmem$(O) \ pngpread$(O) pngread$(O) pngrio$(O) pngrtran$(O) pngrutil$(O) \ - pngset$(O) pngtrans$(O) pngwio$(O) pngwrite$(O) pnggccrd$(O) \ + pngset$(O) pngtrans$(O) pngwio$(O) pngwrite$(O) \ pngwtran$(O) pngwutil$(O) $(ZOBJS) EXES = $(PNGCRUSH)$(E) diff --git a/adler32.c b/adler32.c index 007ba2627..8bf7dc42e 100644 --- a/adler32.c +++ b/adler32.c @@ -1,5 +1,5 @@ /* adler32.c -- compute the Adler-32 checksum of a data stream - * Copyright (C) 1995-2004 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -8,6 +8,15 @@ #define ZLIB_INTERNAL #include "zlib.h" +#define local static + +#ifdef _LARGEFILE64_SOURCE + local uLong adler32_combine_(uLong adler1, uLong adler2, off64_t len2); +#else + local uLong adler32_combine_(uLong adler1, uLong adler2, z_off_t len2); +#endif + + #define BASE 65521UL /* largest prime smaller than 65536 */ #define NMAX 5552 /* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1 */ @@ -125,10 +134,14 @@ uLong ZEXPORT adler32(adler, buf, len) } /* ========================================================================= */ -uLong ZEXPORT adler32_combine(adler1, adler2, len2) +local uLong adler32_combine_(adler1, adler2, len2) uLong adler1; uLong adler2; +#ifdef _LARGEFILE64_SOURCE + off64_t len2; +#else z_off_t len2; +#endif { unsigned long sum1; unsigned long sum2; @@ -147,3 +160,30 @@ uLong ZEXPORT adler32_combine(adler1, adler2, len2) if (sum2 > BASE) sum2 -= BASE; return sum1 | (sum2 << 16); } + +/* ========================================================================= */ +uLong ZEXPORT adler32_combine(adler1, adler2, len2) + uLong adler1; + uLong adler2; + z_off_t len2; +{ + return adler32_combine_(adler1, adler2, len2); +} + +#ifdef _LARGEFILE64_SOURCE +uLong ZEXPORT adler32_combine64(adler1, adler2, len2) + uLong adler1; + uLong adler2; + off64_t len2; +{ + return adler32_combine_(adler1, adler2, len2); +} +#else +uLong ZEXPORT adler32_combine64(adler1, adler2, len2) + uLong adler1; + uLong adler2; + z_off_t len2; +{ + return adler32_combine_(adler1, adler2, len2); +} +#endif diff --git a/compress.c b/compress.c index df04f0148..ea4dfbe9d 100644 --- a/compress.c +++ b/compress.c @@ -1,5 +1,5 @@ /* compress.c -- compress a memory buffer - * Copyright (C) 1995-2003 Jean-loup Gailly. + * Copyright (C) 1995-2005 Jean-loup Gailly. * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -75,5 +75,6 @@ int ZEXPORT compress (dest, destLen, source, sourceLen) uLong ZEXPORT compressBound (sourceLen) uLong sourceLen; { - return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + 11; + return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + + (sourceLen >> 25) + 13; } diff --git a/crc32.c b/crc32.c index f658a9ef5..b34a510fb 100644 --- a/crc32.c +++ b/crc32.c @@ -1,5 +1,5 @@ /* crc32.c -- compute the CRC-32 of a data stream - * Copyright (C) 1995-2005 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h * * Thanks to Rodney Brown for his contribution of faster @@ -68,6 +68,12 @@ local unsigned long gf2_matrix_times OF((unsigned long *mat, unsigned long vec)); local void gf2_matrix_square OF((unsigned long *square, unsigned long *mat)); +#ifdef _LARGEFILE64_SOURCE + local uLong crc32_combine_(uLong crc1, uLong crc2, off64_t len2); +#else + local uLong crc32_combine_(uLong crc1, uLong crc2, z_off_t len2); +#endif + #ifdef DYNAMIC_CRC_TABLE @@ -367,10 +373,14 @@ local void gf2_matrix_square(square, mat) } /* ========================================================================= */ -uLong ZEXPORT crc32_combine(crc1, crc2, len2) +local uLong crc32_combine_(crc1, crc2, len2) uLong crc1; uLong crc2; +#ifdef _LARGEFILE64_SOURCE + off64_t len2; +#else z_off_t len2; +#endif { int n; unsigned long row; @@ -421,3 +431,30 @@ uLong ZEXPORT crc32_combine(crc1, crc2, len2) crc1 ^= crc2; return crc1; } + +/* ========================================================================= */ +uLong ZEXPORT crc32_combine(crc1, crc2, len2) + uLong crc1; + uLong crc2; + z_off_t len2; +{ + return crc32_combine_(crc1, crc2, len2); +} + +#ifdef _LARGEFILE64_SOURCE +uLong ZEXPORT crc32_combine64(crc1, crc2, len2) + uLong crc1; + uLong crc2; + off64_t len2; +{ + return crc32_combine_(crc1, crc2, len2); +} +#else +uLong ZEXPORT crc32_combine64(crc1, crc2, len2) + uLong crc1; + uLong crc2; + z_off_t len2; +{ + return crc32_combine_(crc1, crc2, len2); +} +#endif diff --git a/deflate.c b/deflate.c index 0e946cc42..fc41b8052 100644 --- a/deflate.c +++ b/deflate.c @@ -1,5 +1,5 @@ /* deflate.c -- compress data using the deflation algorithm - * Copyright (C) 1995-2005 Jean-loup Gailly. + * Copyright (C) 1995-2006 Jean-loup Gailly. * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -52,7 +52,7 @@ #include "deflate.h" const char deflate_copyright[] = - " deflate 1.2.3 Copyright 1995-2005 Jean-loup Gailly "; + " deflate 1.2.3.3 Copyright 1995-2006 Jean-loup Gailly "; /* If you use the zlib library in a product, an acknowledgment is welcome in the documentation of your product. If for some reason you cannot @@ -106,7 +106,7 @@ local void check_match OF((deflate_state *s, IPos start, IPos match, /* Tail of hash chains */ #ifndef TOO_FAR -# define TOO_FAR 32767 +# define TOO_FAR 4096 #endif /* Matches of length 3 are discarded if their distance exceeds TOO_FAR */ @@ -481,33 +481,65 @@ int ZEXPORT deflateTune(strm, good_length, max_lazy, nice_length, max_chain) * resulting from using fixed blocks instead of stored blocks, which deflate * can emit on compressed data for some combinations of the parameters. * - * This function could be more sophisticated to provide closer upper bounds - * for every combination of windowBits and memLevel, as well as wrap. - * But even the conservative upper bound of about 14% expansion does not - * seem onerous for output buffer allocation. + * This function could be more sophisticated to provide closer upper bounds for + * every combination of windowBits and memLevel. But even the conservative + * upper bound of about 14% expansion does not seem onerous for output buffer + * allocation. */ uLong ZEXPORT deflateBound(strm, sourceLen) z_streamp strm; uLong sourceLen; { deflate_state *s; - uLong destLen; + uLong complen, wraplen; + Bytef *str; - /* conservative upper bound */ - destLen = sourceLen + - ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 11; + /* conservative upper bound for compressed data */ + complen = sourceLen + + ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5; - /* if can't get parameters, return conservative bound */ + /* if can't get parameters, return conservative bound plus zlib wrapper */ if (strm == Z_NULL || strm->state == Z_NULL) - return destLen; + return complen + 6; + + /* compute wrapper length */ + s = strm->state; + switch (s->wrap) { + case 0: /* raw deflate */ + wraplen = 0; + break; + case 1: /* zlib wrapper */ + wraplen = 6 + (s->strstart ? 4 : 0); + break; + case 2: /* gzip wrapper */ + wraplen = 18; + if (s->gzhead != NULL) { /* user-supplied gzip header */ + if (s->gzhead->extra != NULL) + wraplen += 2 + s->gzhead->extra_len; + str = s->gzhead->name; + if (str != NULL) + do { + wraplen++; + } while (*str++); + str = s->gzhead->comment; + if (str != NULL) + do { + wraplen++; + } while (*str++); + if (s->gzhead->hcrc) + wraplen += 2; + } + break; + default: /* for compiler happiness */ + wraplen = 6; + } /* if not default parameters, return conservative bound */ - s = strm->state; if (s->w_bits != 15 || s->hash_bits != 8 + 7) - return destLen; + return complen + wraplen; /* default settings: return tight bound for that case */ - return compressBound(sourceLen); + return compressBound(sourceLen) - 6 + wraplen; } /* ========================================================================= diff --git a/deflate.h b/deflate.h index 05a5ab3a2..90077d807 100644 --- a/deflate.h +++ b/deflate.h @@ -1,5 +1,5 @@ /* deflate.h -- internal compression state - * Copyright (C) 1995-2004 Jean-loup Gailly + * Copyright (C) 1995-2005 Jean-loup Gailly * For conditions of distribution and use, see copyright notice in zlib.h */ diff --git a/gzio.c b/gzio.c index 7e90f4928..df346207f 100644 --- a/gzio.c +++ b/gzio.c @@ -1,5 +1,5 @@ /* gzio.c -- IO on .gz files - * Copyright (C) 1995-2005 Jean-loup Gailly. + * Copyright (C) 1995-2006 Jean-loup Gailly. * For conditions of distribution and use, see copyright notice in zlib.h * * Compile this file with -DNO_GZCOMPRESS to avoid the compression code. @@ -7,9 +7,17 @@ /* @(#) $Id$ */ -#include +#ifdef _LARGEFILE64_SOURCE +# ifndef _LARGEFILE_SOURCE +# define _LARGEFILE_SOURCE +# endif +# ifdef _FILE_OFFSET_BITS +# undef _FILE_OFFSET_BITS +# endif +#endif #include "zutil.h" +#include #ifdef NO_DEFLATE /* for compatibility with old definition */ # define NO_GZCOMPRESS @@ -40,6 +48,14 @@ extern voidp malloc OF((uInt size)); extern void free OF((voidpf ptr)); #endif +#ifdef NO_FSEEKO +# define FSEEK fseek +# define FTELL ftell +#else +# define FSEEK fseeko +# define FTELL ftello +#endif + #define ALLOC(size) malloc(size) #define TRYFREE(p) {if (p) free(p);} @@ -65,15 +81,27 @@ typedef struct gz_stream { char *path; /* path name for debugging only */ int transparent; /* 1 if input file is not a .gz file */ char mode; /* 'w' or 'r' */ +#ifdef _LARGEFILE64_SOURCE + off64_t start; /* start of compressed data in file (header skipped) */ + off64_t in; /* bytes into deflate or inflate */ + off64_t out; /* bytes out of deflate or inflate */ +#else z_off_t start; /* start of compressed data in file (header skipped) */ z_off_t in; /* bytes into deflate or inflate */ z_off_t out; /* bytes out of deflate or inflate */ +#endif int back; /* one character push-back */ int last; /* true if push-back is last character */ } gz_stream; -local gzFile gz_open OF((const char *path, const char *mode, int fd)); +local gzFile gz_open OF((const char *path, const char *mode, int fd, + int use64)); +#ifdef _LARGEFILE64_SOURCE +local off64_t gz_seek OF((gzFile file, off64_t offset, int whence, int use64)); +#else +local z_off_t gz_seek OF((gzFile file, z_off_t offset, int whence, int use64)); +#endif local int do_flush OF((gzFile file, int flush)); local int get_byte OF((gz_stream *s)); local void check_header OF((gz_stream *s)); @@ -90,10 +118,11 @@ local uLong getLong OF((gz_stream *s)); can be checked to distinguish the two cases (if errno is zero, the zlib error is Z_MEM_ERROR). */ -local gzFile gz_open (path, mode, fd) +local gzFile gz_open (path, mode, fd, use64) const char *path; const char *mode; int fd; + int use64; { int err; int level = Z_DEFAULT_COMPRESSION; /* compression level */ @@ -165,12 +194,7 @@ local gzFile gz_open (path, mode, fd) s->stream.next_in = s->inbuf = (Byte*)ALLOC(Z_BUFSIZE); err = inflateInit2(&(s->stream), -MAX_WBITS); - /* windowBits is passed < 0 to tell that there is no zlib header. - * Note that in this case inflate *requires* an extra "dummy" byte - * after the compressed stream in order to complete decompression and - * return Z_STREAM_END. Here the gzip CRC32 ensures that 4 bytes are - * present after the compressed stream. - */ + /* windowBits is passed < 0 to tell that there is no zlib header */ if (err != Z_OK || s->inbuf == Z_NULL) { return destroy(s), (gzFile)Z_NULL; } @@ -178,7 +202,8 @@ local gzFile gz_open (path, mode, fd) s->stream.avail_out = Z_BUFSIZE; errno = 0; - s->file = fd < 0 ? F_OPEN(path, fmode) : (FILE*)fdopen(fd, fmode); + s->file = fd < 0 ? (use64 ? F_OPEN64(path, fmode) : F_OPEN(path, fmode)) : + (FILE*)fdopen(fd, fmode); if (s->file == NULL) { return destroy(s), (gzFile)Z_NULL; @@ -187,7 +212,10 @@ local gzFile gz_open (path, mode, fd) /* Write a very simple .gz header: */ fprintf(s->file, "%c%c%c%c%c%c%c%c%c%c", gz_magic[0], gz_magic[1], - Z_DEFLATED, 0 /*flags*/, 0,0,0,0 /*time*/, 0 /*xflags*/, OS_CODE); + Z_DEFLATED, 0 /*flags*/, 0,0,0,0 /*time*/, level == 9 ? 2 : + (strategy >= Z_HUFFMAN_ONLY || + (level != Z_DEFAULT_COMPRESSION && level < 2) ? + 4 : 0) /*xflags*/, OS_CODE); s->start = 10L; /* We use 10L instead of ftell(s->file) to because ftell causes an * fflush on some systems. This version of the library doesn't use @@ -196,7 +224,7 @@ local gzFile gz_open (path, mode, fd) */ } else { check_header(s); /* skip the .gz header */ - s->start = ftell(s->file) - s->stream.avail_in; + s->start = FTELL(s->file) - s->stream.avail_in; } return (gzFile)s; @@ -209,7 +237,17 @@ gzFile ZEXPORT gzopen (path, mode) const char *path; const char *mode; { - return gz_open (path, mode, -1); + return gz_open (path, mode, -1, 0); +} + +/* =========================================================================== + Opens a gzip (.gz) file for reading or writing for 64-bit offsets +*/ +gzFile ZEXPORT gzopen64 (path, mode) + const char *path; + const char *mode; +{ + return gz_open (path, mode, -1, 1); } /* =========================================================================== @@ -225,7 +263,7 @@ gzFile ZEXPORT gzdopen (fd, mode) if (fd < 0) return (gzFile)Z_NULL; sprintf(name, "", fd); /* for debugging */ - return gz_open (name, mode, fd); + return gz_open (name, mode, fd, 0); } /* =========================================================================== @@ -256,7 +294,7 @@ int ZEXPORT gzsetparams (file, level, strategy) /* =========================================================================== Read a byte from a gz_stream; update next_in and avail_in. Return EOF for end of file. - IN assertion: the stream s has been sucessfully opened for reading. + IN assertion: the stream s has been successfully opened for reading. */ local int get_byte(s) gz_stream *s; @@ -281,7 +319,7 @@ local int get_byte(s) mode to transparent if the gzip magic header is not present; set s->err to Z_DATA_ERROR if the magic header is present but the rest of the header is incorrect. - IN assertion: the stream s has already been created sucessfully; + IN assertion: the stream s has already been created successfully; s->stream.avail_in is zero for the first time, but may be non-zero for concatenated .gz files. */ @@ -301,6 +339,7 @@ local void check_header(s) if (len) s->inbuf[0] = s->stream.next_in[0]; errno = 0; len = (uInt)fread(s->inbuf + len, 1, Z_BUFSIZE >> len, s->file); + if (len == 0) s->z_eof = 1; if (len == 0 && ferror(s->file)) s->z_err = Z_ERRNO; s->stream.avail_in += len; s->stream.next_in = s->inbuf; @@ -436,7 +475,7 @@ int ZEXPORT gzread (file, buf, len) s->stream.avail_out -= n; s->stream.avail_in -= n; } - if (s->stream.avail_out > 0) { + if (s->stream.avail_out > 0 && !feof(s->file)) { s->stream.avail_out -= (uInt)fread(next_out, 1, s->stream.avail_out, s->file); } @@ -764,10 +803,17 @@ int ZEXPORT gzflush (file, flush) SEEK_END is not implemented, returns error. In this version of the library, gzseek can be extremely slow. */ -z_off_t ZEXPORT gzseek (file, offset, whence) +#ifdef _LARGEFILE64_SOURCE +local off64_t gz_seek (file, offset, whence, use64) + gzFile file; + off64_t offset; +#else +local z_off_t gz_seek (file, offset, whence, use64) gzFile file; z_off_t offset; +#endif int whence; + int use64; { gz_stream *s = (gz_stream*)file; @@ -816,7 +862,13 @@ z_off_t ZEXPORT gzseek (file, offset, whence) s->back = EOF; s->stream.avail_in = 0; s->stream.next_in = s->inbuf; - if (fseek(s->file, offset, SEEK_SET) < 0) return -1L; +#ifdef _LARGEFILE64_SOURCE + if ((use64 ? fseeko64(s->file, offset, SEEK_SET) : + FSEEK(s->file, offset, SEEK_SET)) < 0) + return -1L; +#else + if (FSEEK(s->file, offset, SEEK_SET) < 0) return -1L; +#endif s->in = s->out = offset; return offset; @@ -851,6 +903,35 @@ z_off_t ZEXPORT gzseek (file, offset, whence) return s->out; } +/* =========================================================================== + Define external functions gzseek() and gzseek64() using local gz_seek(). +*/ +z_off_t ZEXPORT gzseek (file, offset, whence) + gzFile file; + z_off_t offset; + int whence; +{ + return (z_off_t)gz_seek(file, offset, whence, 0); +} + +#ifdef _LARGEFILE64_SOURCE +off64_t ZEXPORT gzseek64 (file, offset, whence) + gzFile file; + off64_t offset; + int whence; +{ + return gz_seek(file, offset, whence, 1); +} +#else +z_off_t ZEXPORT gzseek64 (file, offset, whence) + gzFile file; + z_off_t offset; + int whence; +{ + return gz_seek(file, offset, whence, 0); +} +#endif + /* =========================================================================== Rewinds input file. */ @@ -870,7 +951,7 @@ int ZEXPORT gzrewind (file) if (!s->transparent) (void)inflateReset(&s->stream); s->in = 0; s->out = 0; - return fseek(s->file, s->start, SEEK_SET); + return FSEEK(s->file, s->start, SEEK_SET); } /* =========================================================================== @@ -884,6 +965,19 @@ z_off_t ZEXPORT gztell (file) return gzseek(file, 0L, SEEK_CUR); } +/* =========================================================================== + 64-bit version +*/ +#ifdef _LARGEFILE64_SOURCE +off64_t ZEXPORT gztell64 (file) +#else +z_off_t ZEXPORT gztell64 (file) +#endif + gzFile file; +{ + return gzseek64(file, 0L, SEEK_CUR); +} + /* =========================================================================== Returns 1 when EOF has previously been detected reading the given input stream, otherwise zero. @@ -971,7 +1065,7 @@ int ZEXPORT gzclose (file) return destroy((gz_stream*)file); } -#ifdef STDC +#if defined(STDC) && !defined(_WIN32_WCE) # define zstrerror(errnum) strerror(errnum) #else # define zstrerror(errnum) "" diff --git a/infback.c b/infback.c index 455dbc9ee..be0b3dc89 100644 --- a/infback.c +++ b/infback.c @@ -1,5 +1,5 @@ /* infback.c -- inflate using a call-back interface - * Copyright (C) 1995-2005 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -253,7 +253,7 @@ void FAR *out_desc; unsigned bits; /* bits in bit buffer */ unsigned copy; /* number of stored or match bytes to copy */ unsigned char FAR *from; /* where to copy match bytes from */ - code this; /* current decoding table entry */ + code here; /* current decoding table entry */ code last; /* parent table entry */ unsigned len; /* length to copy for repeats, bits to drop */ int ret; /* return code */ @@ -389,19 +389,19 @@ void FAR *out_desc; state->have = 0; while (state->have < state->nlen + state->ndist) { for (;;) { - this = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if (this.val < 16) { - NEEDBITS(this.bits); - DROPBITS(this.bits); - state->lens[state->have++] = this.val; + if (here.val < 16) { + NEEDBITS(here.bits); + DROPBITS(here.bits); + state->lens[state->have++] = here.val; } else { - if (this.val == 16) { - NEEDBITS(this.bits + 2); - DROPBITS(this.bits); + if (here.val == 16) { + NEEDBITS(here.bits + 2); + DROPBITS(here.bits); if (state->have == 0) { strm->msg = (char *)"invalid bit length repeat"; state->mode = BAD; @@ -411,16 +411,16 @@ void FAR *out_desc; copy = 3 + BITS(2); DROPBITS(2); } - else if (this.val == 17) { - NEEDBITS(this.bits + 3); - DROPBITS(this.bits); + else if (here.val == 17) { + NEEDBITS(here.bits + 3); + DROPBITS(here.bits); len = 0; copy = 3 + BITS(3); DROPBITS(3); } else { - NEEDBITS(this.bits + 7); - DROPBITS(this.bits); + NEEDBITS(here.bits + 7); + DROPBITS(here.bits); len = 0; copy = 11 + BITS(7); DROPBITS(7); @@ -474,28 +474,28 @@ void FAR *out_desc; /* get a literal, length, or end-of-block code */ for (;;) { - this = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if (this.op && (this.op & 0xf0) == 0) { - last = this; + if (here.op && (here.op & 0xf0) == 0) { + last = here; for (;;) { - this = state->lencode[last.val + + here = state->lencode[last.val + (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + this.bits) <= bits) break; + if ((unsigned)(last.bits + here.bits) <= bits) break; PULLBYTE(); } DROPBITS(last.bits); } - DROPBITS(this.bits); - state->length = (unsigned)this.val; + DROPBITS(here.bits); + state->length = (unsigned)here.val; /* process literal */ - if (this.op == 0) { - Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ? + if (here.op == 0) { + Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", this.val)); + "inflate: literal 0x%02x\n", here.val)); ROOM(); *put++ = (unsigned char)(state->length); left--; @@ -504,21 +504,21 @@ void FAR *out_desc; } /* process end of block */ - if (this.op & 32) { + if (here.op & 32) { Tracevv((stderr, "inflate: end of block\n")); state->mode = TYPE; break; } /* invalid code */ - if (this.op & 64) { + if (here.op & 64) { strm->msg = (char *)"invalid literal/length code"; state->mode = BAD; break; } /* length code -- get extra bits, if any */ - state->extra = (unsigned)(this.op) & 15; + state->extra = (unsigned)(here.op) & 15; if (state->extra != 0) { NEEDBITS(state->extra); state->length += BITS(state->extra); @@ -528,30 +528,30 @@ void FAR *out_desc; /* get distance code */ for (;;) { - this = state->distcode[BITS(state->distbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->distcode[BITS(state->distbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if ((this.op & 0xf0) == 0) { - last = this; + if ((here.op & 0xf0) == 0) { + last = here; for (;;) { - this = state->distcode[last.val + + here = state->distcode[last.val + (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + this.bits) <= bits) break; + if ((unsigned)(last.bits + here.bits) <= bits) break; PULLBYTE(); } DROPBITS(last.bits); } - DROPBITS(this.bits); - if (this.op & 64) { + DROPBITS(here.bits); + if (here.op & 64) { strm->msg = (char *)"invalid distance code"; state->mode = BAD; break; } - state->offset = (unsigned)this.val; + state->offset = (unsigned)here.val; /* get distance extra bits, if any */ - state->extra = (unsigned)(this.op) & 15; + state->extra = (unsigned)(here.op) & 15; if (state->extra != 0) { NEEDBITS(state->extra); state->offset += BITS(state->extra); diff --git a/inffast.c b/inffast.c index bbee92ed1..0b919bb43 100644 --- a/inffast.c +++ b/inffast.c @@ -1,5 +1,5 @@ /* inffast.c -- fast decoding - * Copyright (C) 1995-2004 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -87,7 +87,7 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ code const FAR *dcode; /* local strm->distcode */ unsigned lmask; /* mask for first level of length codes */ unsigned dmask; /* mask for first level of distance codes */ - code this; /* retrieved table entry */ + code here; /* retrieved table entry */ unsigned op; /* code bits, operation, extra bits, or */ /* window position, window bytes to copy */ unsigned len; /* match length, unused bytes */ @@ -124,20 +124,20 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ hold += (unsigned long)(PUP(in)) << bits; bits += 8; } - this = lcode[hold & lmask]; + here = lcode[hold & lmask]; dolen: - op = (unsigned)(this.bits); + op = (unsigned)(here.bits); hold >>= op; bits -= op; - op = (unsigned)(this.op); + op = (unsigned)(here.op); if (op == 0) { /* literal */ - Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ? + Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", this.val)); - PUP(out) = (unsigned char)(this.val); + "inflate: literal 0x%02x\n", here.val)); + PUP(out) = (unsigned char)(here.val); } else if (op & 16) { /* length base */ - len = (unsigned)(this.val); + len = (unsigned)(here.val); op &= 15; /* number of extra bits */ if (op) { if (bits < op) { @@ -155,14 +155,14 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ hold += (unsigned long)(PUP(in)) << bits; bits += 8; } - this = dcode[hold & dmask]; + here = dcode[hold & dmask]; dodist: - op = (unsigned)(this.bits); + op = (unsigned)(here.bits); hold >>= op; bits -= op; - op = (unsigned)(this.op); + op = (unsigned)(here.op); if (op & 16) { /* distance base */ - dist = (unsigned)(this.val); + dist = (unsigned)(here.val); op &= 15; /* number of extra bits */ if (bits < op) { hold += (unsigned long)(PUP(in)) << bits; @@ -187,9 +187,30 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ if (dist > op) { /* see if copy from window */ op = dist - op; /* distance back in window */ if (op > whave) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; + if (state->sane) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } +#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR + if (len <= op - whave) { + do { + PUP(out) = 0; + } while (--len); + continue; + } + len -= op - whave; + do { + PUP(out) = 0; + } while (--op > whave); + if (op == 0) { + from = out - dist; + do { + PUP(out) = PUP(from); + } while (--len); + continue; + } +#endif } from = window - OFF; if (write == 0) { /* very common case */ @@ -259,7 +280,7 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ } } else if ((op & 64) == 0) { /* 2nd level distance code */ - this = dcode[this.val + (hold & ((1U << op) - 1))]; + here = dcode[here.val + (hold & ((1U << op) - 1))]; goto dodist; } else { @@ -269,7 +290,7 @@ unsigned start; /* inflate()'s starting value for strm->avail_out */ } } else if ((op & 64) == 0) { /* 2nd level length code */ - this = lcode[this.val + (hold & ((1U << op) - 1))]; + here = lcode[here.val + (hold & ((1U << op) - 1))]; goto dolen; } else if (op & 32) { /* end-of-block */ diff --git a/inflate.c b/inflate.c index 792fdee8e..d3c718c06 100644 --- a/inflate.c +++ b/inflate.c @@ -1,5 +1,5 @@ /* inflate.c -- zlib decompression - * Copyright (C) 1995-2005 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -121,6 +121,7 @@ z_streamp strm; state->hold = 0; state->bits = 0; state->lencode = state->distcode = state->next = state->codes; + state->sane = 1; Tracev((stderr, "inflate: reset\n")); return Z_OK; } @@ -564,7 +565,7 @@ int flush; unsigned in, out; /* save starting available input and output */ unsigned copy; /* number of stored or match bytes to copy */ unsigned char FAR *from; /* where to copy match bytes from */ - code this; /* current decoding table entry */ + code here; /* current decoding table entry */ code last; /* parent table entry */ unsigned len; /* length to copy for repeats, bits to drop */ int ret; /* return code */ @@ -876,19 +877,19 @@ int flush; case CODELENS: while (state->have < state->nlen + state->ndist) { for (;;) { - this = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if (this.val < 16) { - NEEDBITS(this.bits); - DROPBITS(this.bits); - state->lens[state->have++] = this.val; + if (here.val < 16) { + NEEDBITS(here.bits); + DROPBITS(here.bits); + state->lens[state->have++] = here.val; } else { - if (this.val == 16) { - NEEDBITS(this.bits + 2); - DROPBITS(this.bits); + if (here.val == 16) { + NEEDBITS(here.bits + 2); + DROPBITS(here.bits); if (state->have == 0) { strm->msg = (char *)"invalid bit length repeat"; state->mode = BAD; @@ -898,16 +899,16 @@ int flush; copy = 3 + BITS(2); DROPBITS(2); } - else if (this.val == 17) { - NEEDBITS(this.bits + 3); - DROPBITS(this.bits); + else if (here.val == 17) { + NEEDBITS(here.bits + 3); + DROPBITS(here.bits); len = 0; copy = 3 + BITS(3); DROPBITS(3); } else { - NEEDBITS(this.bits + 7); - DROPBITS(this.bits); + NEEDBITS(here.bits + 7); + DROPBITS(here.bits); len = 0; copy = 11 + BITS(7); DROPBITS(7); @@ -955,40 +956,40 @@ int flush; break; } for (;;) { - this = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->lencode[BITS(state->lenbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if (this.op && (this.op & 0xf0) == 0) { - last = this; + if (here.op && (here.op & 0xf0) == 0) { + last = here; for (;;) { - this = state->lencode[last.val + + here = state->lencode[last.val + (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + this.bits) <= bits) break; + if ((unsigned)(last.bits + here.bits) <= bits) break; PULLBYTE(); } DROPBITS(last.bits); } - DROPBITS(this.bits); - state->length = (unsigned)this.val; - if ((int)(this.op) == 0) { - Tracevv((stderr, this.val >= 0x20 && this.val < 0x7f ? + DROPBITS(here.bits); + state->length = (unsigned)here.val; + if ((int)(here.op) == 0) { + Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", this.val)); + "inflate: literal 0x%02x\n", here.val)); state->mode = LIT; break; } - if (this.op & 32) { + if (here.op & 32) { Tracevv((stderr, "inflate: end of block\n")); state->mode = TYPE; break; } - if (this.op & 64) { + if (here.op & 64) { strm->msg = (char *)"invalid literal/length code"; state->mode = BAD; break; } - state->extra = (unsigned)(this.op) & 15; + state->extra = (unsigned)(here.op) & 15; state->mode = LENEXT; case LENEXT: if (state->extra) { @@ -1000,28 +1001,28 @@ int flush; state->mode = DIST; case DIST: for (;;) { - this = state->distcode[BITS(state->distbits)]; - if ((unsigned)(this.bits) <= bits) break; + here = state->distcode[BITS(state->distbits)]; + if ((unsigned)(here.bits) <= bits) break; PULLBYTE(); } - if ((this.op & 0xf0) == 0) { - last = this; + if ((here.op & 0xf0) == 0) { + last = here; for (;;) { - this = state->distcode[last.val + + here = state->distcode[last.val + (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + this.bits) <= bits) break; + if ((unsigned)(last.bits + here.bits) <= bits) break; PULLBYTE(); } DROPBITS(last.bits); } - DROPBITS(this.bits); - if (this.op & 64) { + DROPBITS(here.bits); + if (here.op & 64) { strm->msg = (char *)"invalid distance code"; state->mode = BAD; break; } - state->offset = (unsigned)this.val; - state->extra = (unsigned)(this.op) & 15; + state->offset = (unsigned)here.val; + state->extra = (unsigned)(here.op) & 15; state->mode = DISTEXT; case DISTEXT: if (state->extra) { @@ -1036,11 +1037,6 @@ int flush; break; } #endif - if (state->offset > state->whave + out - left) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } Tracevv((stderr, "inflate: distance %u\n", state->offset)); state->mode = MATCH; case MATCH: @@ -1048,6 +1044,26 @@ int flush; copy = out - left; if (state->offset > copy) { /* copy from window */ copy = state->offset - copy; + if (copy > state->whave) { + if (state->sane) { + strm->msg = (char *)"invalid distance too far back"; + state->mode = BAD; + break; + } +#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR + Trace((stderr, "inflate.c too far\n")); + copy -= state->whave; + if (copy > state->length) copy = state->length; + if (copy > left) copy = left; + left -= copy; + state->length -= copy; + do { + *put++ = 0; + } while (--copy); + if (state->length == 0) state->mode = LEN; + break; +#endif + } if (copy > state->write) { copy -= state->write; from = state->window + (state->wsize - copy); @@ -1366,3 +1382,20 @@ z_streamp source; dest->state = (struct internal_state FAR *)copy; return Z_OK; } + +int ZEXPORT inflateUndermine(strm, subvert) +z_streamp strm; +int subvert; +{ + struct inflate_state FAR *state; + + if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR; + state = (struct inflate_state FAR *)strm->state; +#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR + state->sane = !subvert; + return Z_OK; +#else + state->sane = 1; + return Z_DATA_ERROR; +#endif +} diff --git a/inflate.h b/inflate.h index 07bd3e78a..d54a86886 100644 --- a/inflate.h +++ b/inflate.h @@ -1,5 +1,5 @@ /* inflate.h -- internal inflate state definition - * Copyright (C) 1995-2004 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -73,7 +73,7 @@ typedef enum { CHECK -> LENGTH -> DONE */ -/* state maintained between inflate() calls. Approximately 7K bytes. */ +/* state maintained between inflate() calls. Approximately 10K bytes. */ struct inflate_state { inflate_mode mode; /* current inflate mode */ int last; /* true if processing last block */ @@ -112,4 +112,5 @@ struct inflate_state { unsigned short lens[320]; /* temporary storage for code lengths */ unsigned short work[288]; /* work area for code table building */ code codes[ENOUGH]; /* space for code tables */ + int sane; /* if false, allow invalid distance too far */ }; diff --git a/inftrees.c b/inftrees.c index 8a9c13ff0..fb8d8431f 100644 --- a/inftrees.c +++ b/inftrees.c @@ -1,5 +1,5 @@ /* inftrees.c -- generate Huffman trees for efficient decoding - * Copyright (C) 1995-2005 Mark Adler + * Copyright (C) 1995-2006 Mark Adler * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -9,7 +9,7 @@ #define MAXBITS 15 const char inflate_copyright[] = - " inflate 1.2.3 Copyright 1995-2005 Mark Adler "; + " inflate 1.2.3.3 Copyright 1995-2006 Mark Adler "; /* If you use the zlib library in a product, an acknowledgment is welcome in the documentation of your product. If for some reason you cannot @@ -50,7 +50,7 @@ unsigned short FAR *work; unsigned fill; /* index for replicating entries */ unsigned low; /* low bits for current root entry */ unsigned mask; /* mask for low root bits */ - code this; /* table entry for duplication */ + code here; /* table entry for duplication */ code FAR *next; /* next available space in table */ const unsigned short FAR *base; /* base value table to use */ const unsigned short FAR *extra; /* extra bits table to use */ @@ -62,7 +62,7 @@ unsigned short FAR *work; 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; static const unsigned short lext[31] = { /* Length codes 257..285 extra */ 16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 18, 18, 18, 18, - 19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 201, 196}; + 19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 201, 203}; static const unsigned short dbase[32] = { /* Distance codes 0..29 base */ 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, @@ -115,11 +115,11 @@ unsigned short FAR *work; if (count[max] != 0) break; if (root > max) root = max; if (max == 0) { /* no symbols to code at all */ - this.op = (unsigned char)64; /* invalid code marker */ - this.bits = (unsigned char)1; - this.val = (unsigned short)0; - *(*table)++ = this; /* make a table to force an error */ - *(*table)++ = this; + here.op = (unsigned char)64; /* invalid code marker */ + here.bits = (unsigned char)1; + here.val = (unsigned short)0; + *(*table)++ = here; /* make a table to force an error */ + *(*table)++ = here; *bits = 1; return 0; /* no symbols, but wait for decoding to report error */ } @@ -215,18 +215,18 @@ unsigned short FAR *work; /* process all codes and make table entries */ for (;;) { /* create table entry */ - this.bits = (unsigned char)(len - drop); + here.bits = (unsigned char)(len - drop); if ((int)(work[sym]) < end) { - this.op = (unsigned char)0; - this.val = work[sym]; + here.op = (unsigned char)0; + here.val = work[sym]; } else if ((int)(work[sym]) > end) { - this.op = (unsigned char)(extra[work[sym]]); - this.val = base[work[sym]]; + here.op = (unsigned char)(extra[work[sym]]); + here.val = base[work[sym]]; } else { - this.op = (unsigned char)(32 + 64); /* end of block */ - this.val = 0; + here.op = (unsigned char)(32 + 64); /* end of block */ + here.val = 0; } /* replicate for those indices with low len bits equal to huff */ @@ -235,7 +235,7 @@ unsigned short FAR *work; min = fill; /* save offset to next table */ do { fill -= incr; - next[(huff >> drop) + fill] = this; + next[(huff >> drop) + fill] = here; } while (fill != 0); /* backwards increment the len-bit code huff */ @@ -295,20 +295,20 @@ unsigned short FAR *work; through high index bits. When the current sub-table is filled, the loop drops back to the root table to fill in any remaining entries there. */ - this.op = (unsigned char)64; /* invalid code marker */ - this.bits = (unsigned char)(len - drop); - this.val = (unsigned short)0; + here.op = (unsigned char)64; /* invalid code marker */ + here.bits = (unsigned char)(len - drop); + here.val = (unsigned short)0; while (huff != 0) { /* when done with sub-table, drop back to root table */ if (drop != 0 && (huff & mask) != low) { drop = 0; len = root; next = *table; - this.bits = (unsigned char)len; + here.bits = (unsigned char)len; } /* put invalid code marker in table */ - next[huff >> drop] = this; + next[huff >> drop] = here; /* backwards increment the len-bit code huff */ incr = 1U << (len - 1); diff --git a/minigzip.c b/minigzip.c deleted file mode 100644 index 4524b96a1..000000000 --- a/minigzip.c +++ /dev/null @@ -1,322 +0,0 @@ -/* minigzip.c -- simulate gzip using the zlib compression library - * Copyright (C) 1995-2005 Jean-loup Gailly. - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* - * minigzip is a minimal implementation of the gzip utility. This is - * only an example of using zlib and isn't meant to replace the - * full-featured gzip. No attempt is made to deal with file systems - * limiting names to 14 or 8+3 characters, etc... Error checking is - * very limited. So use minigzip only for testing; use gzip for the - * real thing. On MSDOS, use only on file names without extension - * or in pipe mode. - */ - -/* @(#) $Id$ */ - -#include -#include "zlib.h" - -#ifdef STDC -# include -# include -#endif - -#ifdef USE_MMAP -# include -# include -# include -#endif - -#if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__) -# include -# include -# define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY) -#else -# define SET_BINARY_MODE(file) -#endif - -#ifdef VMS -# define unlink delete -# define GZ_SUFFIX "-gz" -#endif -#ifdef RISCOS -# define unlink remove -# define GZ_SUFFIX "-gz" -# define fileno(file) file->__file -#endif -#if defined(__MWERKS__) && __dest_os != __be_os && __dest_os != __win32_os -# include /* for fileno */ -#endif - -#ifndef WIN32 /* unlink already in stdio.h for WIN32 */ - extern int unlink OF((const char *)); -#endif - -#ifndef GZ_SUFFIX -# define GZ_SUFFIX ".gz" -#endif -#define SUFFIX_LEN (sizeof(GZ_SUFFIX)-1) - -#define BUFLEN 16384 -#define MAX_NAME_LEN 1024 - -#ifdef MAXSEG_64K -# define local static - /* Needed for systems with limitation on stack size. */ -#else -# define local -#endif - -char *prog; - -void error OF((const char *msg)); -void gz_compress OF((FILE *in, gzFile out)); -#ifdef USE_MMAP -int gz_compress_mmap OF((FILE *in, gzFile out)); -#endif -void gz_uncompress OF((gzFile in, FILE *out)); -void file_compress OF((char *file, char *mode)); -void file_uncompress OF((char *file)); -int main OF((int argc, char *argv[])); - -/* =========================================================================== - * Display error message and exit - */ -void error(msg) - const char *msg; -{ - fprintf(stderr, "%s: %s\n", prog, msg); - exit(1); -} - -/* =========================================================================== - * Compress input to output then close both files. - */ - -void gz_compress(in, out) - FILE *in; - gzFile out; -{ - local char buf[BUFLEN]; - int len; - int err; - -#ifdef USE_MMAP - /* Try first compressing with mmap. If mmap fails (minigzip used in a - * pipe), use the normal fread loop. - */ - if (gz_compress_mmap(in, out) == Z_OK) return; -#endif - for (;;) { - len = (int)fread(buf, 1, sizeof(buf), in); - if (ferror(in)) { - perror("fread"); - exit(1); - } - if (len == 0) break; - - if (gzwrite(out, buf, (unsigned)len) != len) error(gzerror(out, &err)); - } - fclose(in); - if (gzclose(out) != Z_OK) error("failed gzclose"); -} - -#ifdef USE_MMAP /* MMAP version, Miguel Albrecht */ - -/* Try compressing the input file at once using mmap. Return Z_OK if - * if success, Z_ERRNO otherwise. - */ -int gz_compress_mmap(in, out) - FILE *in; - gzFile out; -{ - int len; - int err; - int ifd = fileno(in); - caddr_t buf; /* mmap'ed buffer for the entire input file */ - off_t buf_len; /* length of the input file */ - struct stat sb; - - /* Determine the size of the file, needed for mmap: */ - if (fstat(ifd, &sb) < 0) return Z_ERRNO; - buf_len = sb.st_size; - if (buf_len <= 0) return Z_ERRNO; - - /* Now do the actual mmap: */ - buf = mmap((caddr_t) 0, buf_len, PROT_READ, MAP_SHARED, ifd, (off_t)0); - if (buf == (caddr_t)(-1)) return Z_ERRNO; - - /* Compress the whole file at once: */ - len = gzwrite(out, (char *)buf, (unsigned)buf_len); - - if (len != (int)buf_len) error(gzerror(out, &err)); - - munmap(buf, buf_len); - fclose(in); - if (gzclose(out) != Z_OK) error("failed gzclose"); - return Z_OK; -} -#endif /* USE_MMAP */ - -/* =========================================================================== - * Uncompress input to output then close both files. - */ -void gz_uncompress(in, out) - gzFile in; - FILE *out; -{ - local char buf[BUFLEN]; - int len; - int err; - - for (;;) { - len = gzread(in, buf, sizeof(buf)); - if (len < 0) error (gzerror(in, &err)); - if (len == 0) break; - - if ((int)fwrite(buf, 1, (unsigned)len, out) != len) { - error("failed fwrite"); - } - } - if (fclose(out)) error("failed fclose"); - - if (gzclose(in) != Z_OK) error("failed gzclose"); -} - - -/* =========================================================================== - * Compress the given file: create a corresponding .gz file and remove the - * original. - */ -void file_compress(file, mode) - char *file; - char *mode; -{ - local char outfile[MAX_NAME_LEN]; - FILE *in; - gzFile out; - - strcpy(outfile, file); - strcat(outfile, GZ_SUFFIX); - - in = fopen(file, "rb"); - if (in == NULL) { - perror(file); - exit(1); - } - out = gzopen(outfile, mode); - if (out == NULL) { - fprintf(stderr, "%s: can't gzopen %s\n", prog, outfile); - exit(1); - } - gz_compress(in, out); - - unlink(file); -} - - -/* =========================================================================== - * Uncompress the given file and remove the original. - */ -void file_uncompress(file) - char *file; -{ - local char buf[MAX_NAME_LEN]; - char *infile, *outfile; - FILE *out; - gzFile in; - uInt len = (uInt)strlen(file); - - strcpy(buf, file); - - if (len > SUFFIX_LEN && strcmp(file+len-SUFFIX_LEN, GZ_SUFFIX) == 0) { - infile = file; - outfile = buf; - outfile[len-3] = '\0'; - } else { - outfile = file; - infile = buf; - strcat(infile, GZ_SUFFIX); - } - in = gzopen(infile, "rb"); - if (in == NULL) { - fprintf(stderr, "%s: can't gzopen %s\n", prog, infile); - exit(1); - } - out = fopen(outfile, "wb"); - if (out == NULL) { - perror(file); - exit(1); - } - - gz_uncompress(in, out); - - unlink(infile); -} - - -/* =========================================================================== - * Usage: minigzip [-d] [-f] [-h] [-r] [-1 to -9] [files...] - * -d : decompress - * -f : compress with Z_FILTERED - * -h : compress with Z_HUFFMAN_ONLY - * -r : compress with Z_RLE - * -1 to -9 : compression level - */ - -int main(argc, argv) - int argc; - char *argv[]; -{ - int uncompr = 0; - gzFile file; - char outmode[20]; - - strcpy(outmode, "wb6 "); - - prog = argv[0]; - argc--, argv++; - - while (argc > 0) { - if (strcmp(*argv, "-d") == 0) - uncompr = 1; - else if (strcmp(*argv, "-f") == 0) - outmode[3] = 'f'; - else if (strcmp(*argv, "-h") == 0) - outmode[3] = 'h'; - else if (strcmp(*argv, "-r") == 0) - outmode[3] = 'R'; - else if ((*argv)[0] == '-' && (*argv)[1] >= '1' && (*argv)[1] <= '9' && - (*argv)[2] == 0) - outmode[2] = (*argv)[1]; - else - break; - argc--, argv++; - } - if (outmode[3] == ' ') - outmode[3] = 0; - if (argc == 0) { - SET_BINARY_MODE(stdin); - SET_BINARY_MODE(stdout); - if (uncompr) { - file = gzdopen(fileno(stdin), "rb"); - if (file == NULL) error("can't gzdopen stdin"); - gz_uncompress(file, stdout); - } else { - file = gzdopen(fileno(stdout), outmode); - if (file == NULL) error("can't gzdopen stdout"); - gz_compress(stdin, file); - } - } else { - do { - if (uncompr) { - file_uncompress(*argv); - } else { - file_compress(*argv, outmode); - } - } while (argv++, --argc); - } - return 0; -} diff --git a/png.c b/png.c index 8b5db963d..63ae86e3c 100644 --- a/png.c +++ b/png.c @@ -1,9 +1,9 @@ /* png.c - location for general purpose libpng functions * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.21 October 4, 2007 * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2007 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -13,20 +13,20 @@ #include "png.h" /* Generate a compiler error if there is an old png.h in the search path. */ -typedef version_1_2_11beta4 Your_png_h_is_not_version_1_2_11beta4; +typedef version_1_2_29 Your_png_h_is_not_version_1_2_29; /* Version information for C files. This had better match the version * string defined in png.h. */ #ifdef PNG_USE_GLOBAL_ARRAYS /* png_libpng_ver was changed to a function in version 1.0.5c */ -const char png_libpng_ver[18] = PNG_LIBPNG_VER_STRING; +PNG_CONST char png_libpng_ver[18] = PNG_LIBPNG_VER_STRING; #ifdef PNG_READ_SUPPORTED /* png_sig was changed to a function in version 1.0.5c */ /* Place to hold the signature string for a PNG file. */ -const png_byte FARDATA png_sig[8] = {137, 80, 78, 71, 13, 10, 26, 10}; +PNG_CONST png_byte FARDATA png_sig[8] = {137, 80, 78, 71, 13, 10, 26, 10}; #endif /* PNG_READ_SUPPORTED */ /* Invoke global declarations for constant strings for known chunk types */ @@ -56,32 +56,27 @@ PNG_zTXt; /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* start of interlace block */ -const int FARDATA png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; +PNG_CONST int FARDATA png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; /* offset to next interlace block */ -const int FARDATA png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; +PNG_CONST int FARDATA png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; /* start of interlace block in the y direction */ -const int FARDATA png_pass_ystart[] = {0, 0, 4, 0, 2, 0, 1}; +PNG_CONST int FARDATA png_pass_ystart[] = {0, 0, 4, 0, 2, 0, 1}; /* offset to next interlace block in the y direction */ -const int FARDATA png_pass_yinc[] = {8, 8, 8, 4, 4, 2, 2}; - -/* width of interlace block (used in assembler routines only) */ -#ifdef PNG_HAVE_ASSEMBLER_COMBINE_ROW -const int FARDATA png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; -#endif +PNG_CONST int FARDATA png_pass_yinc[] = {8, 8, 8, 4, 4, 2, 2}; /* Height of interlace block. This is not currently used - if you need * it, uncomment it here and in png.h -const int FARDATA png_pass_height[] = {8, 8, 4, 4, 2, 2, 1}; +PNG_CONST int FARDATA png_pass_height[] = {8, 8, 4, 4, 2, 2, 1}; */ /* Mask to determine which pixels are valid in a pass */ -const int FARDATA png_pass_mask[] = {0x80, 0x08, 0x88, 0x22, 0xaa, 0x55, 0xff}; +PNG_CONST int FARDATA png_pass_mask[] = {0x80, 0x08, 0x88, 0x22, 0xaa, 0x55, 0xff}; /* Mask to determine which pixels to overwrite while displaying */ -const int FARDATA png_pass_dsp_mask[] +PNG_CONST int FARDATA png_pass_dsp_mask[] = {0xff, 0x0f, 0xff, 0x33, 0xff, 0x55, 0xff}; #endif /* PNG_READ_SUPPORTED */ @@ -97,6 +92,7 @@ const int FARDATA png_pass_dsp_mask[] void PNGAPI png_set_sig_bytes(png_structp png_ptr, int num_bytes) { + if(png_ptr == NULL) return; png_debug(1, "in png_set_sig_bytes\n"); if (num_bytes > 8) png_error(png_ptr, "Too many bytes for PNG signature."); @@ -133,7 +129,7 @@ png_sig_cmp(png_bytep sig, png_size_t start, png_size_t num_to_check) #if defined(PNG_1_0_X) || defined(PNG_1_2_X) /* (Obsolete) function to check signature bytes. It does not allow one * to check a partial signature. This function might be removed in the - * future - use png_sig_cmp(). Returns true (nonzero) if the file is a PNG. + * future - use png_sig_cmp(). Returns true (nonzero) if the file is PNG. */ int PNGAPI png_check_sig(png_bytep sig, int num) @@ -153,13 +149,14 @@ voidpf /* private */ png_zalloc(voidpf png_ptr, uInt items, uInt size) { png_voidp ptr; - png_structp p=png_ptr; + png_structp p=(png_structp)png_ptr; png_uint_32 save_flags=p->flags; png_uint_32 num_bytes; + if(png_ptr == NULL) return (NULL); if (items > PNG_UINT_32_MAX/size) { - png_warning (png_ptr, "Potential overflow in png_zalloc()"); + png_warning (p, "Potential overflow in png_zalloc()"); return (NULL); } num_bytes = (png_uint_32)items * size; @@ -266,6 +263,7 @@ void PNGAPI png_destroy_info_struct(png_structp png_ptr, png_infopp info_ptr_ptr) { png_infop info_ptr = NULL; + if(png_ptr == NULL) return; png_debug(1, "in png_destroy_info_struct\n"); if (info_ptr_ptr != NULL) @@ -304,6 +302,8 @@ png_info_init_3(png_infopp ptr_ptr, png_size_t png_info_struct_size) { png_infop info_ptr = *ptr_ptr; + if(info_ptr == NULL) return; + png_debug(1, "in png_info_init_3\n"); if(png_sizeof(png_info) > png_info_struct_size) @@ -485,6 +485,11 @@ if (mask & PNG_FREE_SPLT) #endif #if defined(PNG_UNKNOWN_CHUNKS_SUPPORTED) + if(png_ptr->unknown_chunk.data) + { + png_free(png_ptr, png_ptr->unknown_chunk.data); + png_ptr->unknown_chunk.data = NULL; + } #ifdef PNG_FREE_ME_SUPPORTED if ((mask & PNG_FREE_UNKN) & info_ptr->free_me) #else @@ -611,6 +616,7 @@ png_info_destroy(png_structp png_ptr, png_infop info_ptr) png_voidp PNGAPI png_get_io_ptr(png_structp png_ptr) { + if(png_ptr == NULL) return (NULL); return (png_ptr->io_ptr); } @@ -626,6 +632,7 @@ void PNGAPI png_init_io(png_structp png_ptr, png_FILE_p fp) { png_debug(1, "in png_init_io\n"); + if(png_ptr == NULL) return; png_ptr->io_ptr = (png_voidp)fp; } #endif @@ -641,6 +648,7 @@ png_convert_to_rfc1123(png_structp png_ptr, png_timep ptime) {"Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"}; + if(png_ptr == NULL) return (NULL); if (png_ptr->time_buffer == NULL) { png_ptr->time_buffer = (png_charp)png_malloc(png_ptr, (png_uint_32)(29* @@ -661,7 +669,7 @@ png_convert_to_rfc1123(png_structp png_ptr, png_timep ptime) #ifdef USE_FAR_KEYWORD { char near_time_buf[29]; - sprintf(near_time_buf, "%d %s %d %02d:%02d:%02d +0000", + png_snprintf6(near_time_buf,29,"%d %s %d %02d:%02d:%02d +0000", ptime->day % 32, short_months[(ptime->month - 1) % 12], ptime->year, ptime->hour % 24, ptime->minute % 60, ptime->second % 61); @@ -669,7 +677,7 @@ png_convert_to_rfc1123(png_structp png_ptr, png_timep ptime) 29*png_sizeof(char)); } #else - sprintf(png_ptr->time_buffer, "%d %s %d %02d:%02d:%02d +0000", + png_snprintf6(png_ptr->time_buffer,29,"%d %s %d %02d:%02d:%02d +0000", ptime->day % 32, short_months[(ptime->month - 1) % 12], ptime->year, ptime->hour % 24, ptime->minute % 60, ptime->second % 61); @@ -679,25 +687,16 @@ png_convert_to_rfc1123(png_structp png_ptr, png_timep ptime) } #endif /* PNG_TIME_RFC1123_SUPPORTED */ -#if 0 -/* Signature string for a PNG file. */ -png_bytep PNGAPI -png_sig_bytes(void) -{ - return ((png_bytep)"\211\120\116\107\015\012\032\012"); -} -#endif #endif /* defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) */ png_charp PNGAPI png_get_copyright(png_structp png_ptr) { - if (&png_ptr != NULL) /* silence compiler warning about unused png_ptr */ - return ((png_charp) "\n libpng version 1.2.11beta4 - June 7, 2006\n\ - Copyright (c) 1998-2006 Glenn Randers-Pehrson\n\ + png_ptr = png_ptr; /* silence compiler warning about unused png_ptr */ + return ((png_charp) "\n libpng version 1.2.29 - May 8, 2008\n\ + Copyright (c) 1998-2008 Glenn Randers-Pehrson\n\ Copyright (c) 1996-1997 Andreas Dilger\n\ Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc.\n"); - return ((png_charp) ""); } /* The following return the library version as a short string in the @@ -712,27 +711,28 @@ png_charp PNGAPI png_get_libpng_ver(png_structp png_ptr) { /* Version of *.c files used when building libpng */ - if (&png_ptr != NULL) /* silence compiler warning about unused png_ptr */ - return ((png_charp) PNG_LIBPNG_VER_STRING); - return ((png_charp) ""); + png_ptr = png_ptr; /* silence compiler warning about unused png_ptr */ + return ((png_charp) PNG_LIBPNG_VER_STRING); } png_charp PNGAPI png_get_header_ver(png_structp png_ptr) { /* Version of *.h files used when building libpng */ - if (&png_ptr != NULL) /* silence compiler warning about unused png_ptr */ - return ((png_charp) PNG_LIBPNG_VER_STRING); - return ((png_charp) ""); + png_ptr = png_ptr; /* silence compiler warning about unused png_ptr */ + return ((png_charp) PNG_LIBPNG_VER_STRING); } png_charp PNGAPI png_get_header_version(png_structp png_ptr) { /* Returns longer string containing both version and date */ - if (&png_ptr != NULL) /* silence compiler warning about unused png_ptr */ - return ((png_charp) PNG_HEADER_VERSION_STRING); - return ((png_charp) ""); + png_ptr = png_ptr; /* silence compiler warning about unused png_ptr */ + return ((png_charp) PNG_HEADER_VERSION_STRING +#ifndef PNG_READ_SUPPORTED + " (NO READ SUPPORT)" +#endif + "\n"); } #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) @@ -743,7 +743,7 @@ png_handle_as_unknown(png_structp png_ptr, png_bytep chunk_name) /* check chunk_name and return "keep" value if it's on the list, else 0 */ int i; png_bytep p; - if((png_ptr == NULL && chunk_name == NULL) || png_ptr->num_chunk_list<=0) + if(png_ptr == NULL || chunk_name == NULL || png_ptr->num_chunk_list<=0) return 0; p=png_ptr->chunk_list+png_ptr->num_chunk_list*5-5; for (i = png_ptr->num_chunk_list; i; i--, p-=5) @@ -757,6 +757,7 @@ png_handle_as_unknown(png_structp png_ptr, png_bytep chunk_name) int PNGAPI png_reset_zstream(png_structp png_ptr) { + if (png_ptr == NULL) return Z_STREAM_ERROR; return (inflateReset(&png_ptr->zstream)); } #endif /* defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) */ @@ -770,67 +771,17 @@ png_access_version_number(void) } -#if defined(PNG_READ_SUPPORTED) +#if defined(PNG_READ_SUPPORTED) && defined(PNG_ASSEMBLER_CODE_SUPPORTED) #if !defined(PNG_1_0_X) -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) - /* GRR: could add this: && defined(PNG_MMX_CODE_SUPPORTED) */ -/* this INTERNAL function was added to libpng 1.2.0 */ -void /* PRIVATE */ -png_init_mmx_flags (png_structp png_ptr) -{ - png_ptr->mmx_rowbytes_threshold = 0; - png_ptr->mmx_bitdepth_threshold = 0; - -# if (defined(PNG_USE_PNGVCRD) || defined(PNG_USE_PNGGCCRD)) - - png_ptr->asm_flags |= PNG_ASM_FLAG_MMX_SUPPORT_COMPILED; - - if (png_mmx_support() > 0) { - png_ptr->asm_flags |= PNG_ASM_FLAG_MMX_SUPPORT_IN_CPU -# ifdef PNG_HAVE_ASSEMBLER_COMBINE_ROW - | PNG_ASM_FLAG_MMX_READ_COMBINE_ROW -# endif -# ifdef PNG_HAVE_ASSEMBLER_READ_INTERLACE - | PNG_ASM_FLAG_MMX_READ_INTERLACE -# endif -# ifndef PNG_HAVE_ASSEMBLER_READ_FILTER_ROW - ; -# else - | PNG_ASM_FLAG_MMX_READ_FILTER_SUB - | PNG_ASM_FLAG_MMX_READ_FILTER_UP - | PNG_ASM_FLAG_MMX_READ_FILTER_AVG - | PNG_ASM_FLAG_MMX_READ_FILTER_PAETH ; - - png_ptr->mmx_rowbytes_threshold = PNG_MMX_ROWBYTES_THRESHOLD_DEFAULT; - png_ptr->mmx_bitdepth_threshold = PNG_MMX_BITDEPTH_THRESHOLD_DEFAULT; -# endif - } else { - png_ptr->asm_flags &= ~( PNG_ASM_FLAG_MMX_SUPPORT_IN_CPU - | PNG_MMX_READ_FLAGS - | PNG_MMX_WRITE_FLAGS ); - } - -# else /* !((PNGVCRD || PNGGCCRD) && PNG_ASSEMBLER_CODE_SUPPORTED)) */ - - /* clear all MMX flags; no support is compiled in */ - png_ptr->asm_flags &= ~( PNG_MMX_FLAGS ); - -# endif /* ?(PNGVCRD || PNGGCCRD) */ -} - -#endif /* !(PNG_ASSEMBLER_CODE_SUPPORTED) */ - /* this function was added to libpng 1.2.0 */ -#if !defined(PNG_USE_PNGGCCRD) && \ - !(defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_USE_PNGVCRD)) int PNGAPI png_mmx_support(void) { + /* obsolete, to be removed from libpng-1.4.0 */ return -1; } -#endif #endif /* PNG_1_0_X */ -#endif /* PNG_READ_SUPPORTED */ +#endif /* PNG_READ_SUPPORTED && PNG_ASSEMBLER_CODE_SUPPORTED */ #if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) #ifdef PNG_SIZE_T diff --git a/png.h b/png.h index 2e4fac8e7..2e194b7a3 100644 --- a/png.h +++ b/png.h @@ -1,15 +1,14 @@ - /* png.h - header file for PNG reference library * - * libpng version 1.2.11beta4 - June 7, 2006 - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * libpng version 1.2.29 - May 8, 2008 + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * * Authors and maintainers: * libpng versions 0.71, May 1995, through 0.88, January 1996: Guy Schalnat * libpng versions 0.89c, June 1996, through 0.96, May 1997: Andreas Dilger - * libpng versions 0.97, January 1998, through 1.2.11beta4 - June 7, 2006: Glenn + * libpng versions 0.97, January 1998, through 1.2.29 - May 8, 2008: Glenn * See also "Contributing Authors", below. * * Note about libpng version numbers: @@ -103,14 +102,14 @@ * 1.0.16 10 10016 10.so.0.1.0.16 * 1.2.6 13 10206 12.so.0.1.2.6 * 1.2.7beta1-2 13 10207 12.so.0.1.2.7beta1-2 - * 1.0.17rc1 10 10017 12.so.0.1.0.17rc1 + * 1.0.17rc1 10 10017 10.so.0.1.0.17rc1 * 1.2.7rc1 13 10207 12.so.0.1.2.7rc1 - * 1.0.17 10 10017 12.so.0.1.0.17 + * 1.0.17 10 10017 10.so.0.1.0.17 * 1.2.7 13 10207 12.so.0.1.2.7 * 1.2.8beta1-5 13 10208 12.so.0.1.2.8beta1-5 - * 1.0.18rc1-5 10 10018 12.so.0.1.0.18rc1-5 + * 1.0.18rc1-5 10 10018 10.so.0.1.0.18rc1-5 * 1.2.8rc1-5 13 10208 12.so.0.1.2.8rc1-5 - * 1.0.18 10 10018 12.so.0.1.0.18 + * 1.0.18 10 10018 10.so.0.1.0.18 * 1.2.8 13 10208 12.so.0.1.2.8 * 1.2.9beta1-3 13 10209 12.so.0.1.2.9beta1-3 * 1.2.9beta4-11 13 10209 12.so.0.9[.0] @@ -120,6 +119,79 @@ * 1.2.10rc1-3 13 10210 12.so.0.10[.0] * 1.2.10 13 10210 12.so.0.10[.0] * 1.2.11beta1-4 13 10211 12.so.0.11[.0] + * 1.0.19rc1-5 10 10019 10.so.0.19[.0] + * 1.2.11rc1-5 13 10211 12.so.0.11[.0] + * 1.0.19 10 10019 10.so.0.19[.0] + * 1.2.11 13 10211 12.so.0.11[.0] + * 1.0.20 10 10020 10.so.0.20[.0] + * 1.2.12 13 10212 12.so.0.12[.0] + * 1.2.13beta1 13 10213 12.so.0.13[.0] + * 1.0.21 10 10021 10.so.0.21[.0] + * 1.2.13 13 10213 12.so.0.13[.0] + * 1.2.14beta1-2 13 10214 12.so.0.14[.0] + * 1.0.22rc1 10 10022 10.so.0.22[.0] + * 1.2.14rc1 13 10214 12.so.0.14[.0] + * 1.0.22 10 10022 10.so.0.22[.0] + * 1.2.14 13 10214 12.so.0.14[.0] + * 1.2.15beta1-6 13 10215 12.so.0.15[.0] + * 1.0.23rc1-5 10 10023 10.so.0.23[.0] + * 1.2.15rc1-5 13 10215 12.so.0.15[.0] + * 1.0.23 10 10023 10.so.0.23[.0] + * 1.2.15 13 10215 12.so.0.15[.0] + * 1.2.16beta1-2 13 10216 12.so.0.16[.0] + * 1.2.16rc1 13 10216 12.so.0.16[.0] + * 1.0.24 10 10024 10.so.0.24[.0] + * 1.2.16 13 10216 12.so.0.16[.0] + * 1.2.17beta1-2 13 10217 12.so.0.17[.0] + * 1.0.25rc1 10 10025 10.so.0.25[.0] + * 1.2.17rc1-3 13 10217 12.so.0.17[.0] + * 1.0.25 10 10025 10.so.0.25[.0] + * 1.2.17 13 10217 12.so.0.17[.0] + * 1.0.26 10 10026 10.so.0.26[.0] + * 1.2.18 13 10218 12.so.0.18[.0] + * 1.2.19beta1-31 13 10219 12.so.0.19[.0] + * 1.0.27rc1-6 10 10027 10.so.0.27[.0] + * 1.2.19rc1-6 13 10219 12.so.0.19[.0] + * 1.0.27 10 10027 10.so.0.27[.0] + * 1.2.19 13 10219 12.so.0.19[.0] + * 1.2.20beta01-04 13 10220 12.so.0.20[.0] + * 1.0.28rc1-6 10 10028 10.so.0.28[.0] + * 1.2.20rc1-6 13 10220 12.so.0.20[.0] + * 1.0.28 10 10028 10.so.0.28[.0] + * 1.2.20 13 10220 12.so.0.20[.0] + * 1.2.21beta1-2 13 10221 12.so.0.21[.0] + * 1.2.21rc1-3 13 10221 12.so.0.21[.0] + * 1.0.29 10 10029 10.so.0.29[.0] + * 1.2.21 13 10221 12.so.0.21[.0] + * 1.2.22beta1-4 13 10222 12.so.0.22[.0] + * 1.0.30rc1 10 10030 10.so.0.30[.0] + * 1.2.22rc1 13 10222 12.so.0.22[.0] + * 1.0.30 10 10030 10.so.0.30[.0] + * 1.2.22 13 10222 12.so.0.22[.0] + * 1.2.23beta01-05 13 10223 12.so.0.23[.0] + * 1.2.23rc01 13 10223 12.so.0.23[.0] + * 1.2.23 13 10223 12.so.0.23[.0] + * 1.2.24beta01-02 13 10224 12.so.0.24[.0] + * 1.2.24rc01 13 10224 12.so.0.24[.0] + * 1.2.24 13 10224 12.so.0.24[.0] + * 1.2.25beta01-06 13 10225 12.so.0.25[.0] + * 1.2.25rc01-02 13 10225 12.so.0.25[.0] + * 1.0.31 10 10031 10.so.0.31[.0] + * 1.2.25 13 10225 12.so.0.25[.0] + * 1.2.26beta01-06 13 10226 12.so.0.26[.0] + * 1.2.26rc01 13 10226 12.so.0.26[.0] + * 1.2.26 13 10226 12.so.0.26[.0] + * 1.0.32 10 10032 10.so.0.32[.0] + * 1.2.27beta01-06 13 10227 12.so.0.27[.0] + * 1.2.27rc01 13 10227 12.so.0.27[.0] + * 1.0.33 10 10033 10.so.0.33[.0] + * 1.2.27 13 10227 12.so.0.27[.0] + * 1.0.34 10 10034 10.so.0.34[.0] + * 1.2.28 13 10228 12.so.0.28[.0] + * 1.2.29beta01-03 13 10229 12.so.0.29[.0] + * 1.2.29rc01 13 10229 12.so.0.29[.0] + * 1.0.35 10 10035 10.so.0.35[.0] + * 1.2.29 13 10229 12.so.0.29[.0] * * Henceforth the source version will match the shared-library major * and minor numbers; the shared-library major version number will be @@ -129,7 +201,7 @@ * to the source version x.y.z (leading zeros in y and z). Beta versions * were given the previous public release number plus a letter, until * version 1.0.6j; from then on they were given the upcoming public - * release number plus "betaNN" or "rcN". + * release number plus "betaNN" or "rcNN". * * Binary incompatibility exists only when applications make direct access * to the info_ptr or png_ptr members through png.h, and the compiled @@ -149,8 +221,8 @@ * If you modify libpng you may insert additional notices immediately following * this sentence. * - * libpng versions 1.2.6, August 15, 2004, through 1.2.11beta4, June 7, 2006, are - * Copyright (c) 2004, 2006 Glenn Randers-Pehrson, and are + * libpng versions 1.2.6, August 15, 2004, through 1.2.29, May 8, 2008, are + * Copyright (c) 2004, 2006-2008 Glenn Randers-Pehrson, and are * distributed according to the same disclaimer and license as libpng-1.2.5 * with the following individual added to the list of Contributing Authors: * @@ -261,13 +333,13 @@ * Y2K compliance in libpng: * ========================= * - * June 7, 2006 + * May 8, 2008 * * Since the PNG Development group is an ad-hoc body, we can't make * an official declaration. * * This is your unofficial assurance that libpng from version 0.71 and - * upward through 1.2.11beta4 are Y2K compliant. It is my belief that earlier + * upward through 1.2.29 are Y2K compliant. It is my belief that earlier * versions were also Y2K compliant. * * Libpng only has three year fields. One is a 2-byte unsigned integer @@ -323,9 +395,9 @@ */ /* Version information for png.h - this should match the version in png.c */ -#define PNG_LIBPNG_VER_STRING "1.2.11beta4" +#define PNG_LIBPNG_VER_STRING "1.2.29" #define PNG_HEADER_VERSION_STRING \ - " libpng version 1.2.11beta4 - June 7, 2006 (header)\n" + " libpng version 1.2.29 - May 8, 2008\n" #define PNG_LIBPNG_VER_SONUM 0 #define PNG_LIBPNG_VER_DLLNUM 13 @@ -333,11 +405,11 @@ /* These should match the first 3 components of PNG_LIBPNG_VER_STRING: */ #define PNG_LIBPNG_VER_MAJOR 1 #define PNG_LIBPNG_VER_MINOR 2 -#define PNG_LIBPNG_VER_RELEASE 11 +#define PNG_LIBPNG_VER_RELEASE 29 /* This should match the numeric part of the final component of * PNG_LIBPNG_VER_STRING, omitting any leading zero: */ -#define PNG_LIBPNG_VER_BUILD 4 +#define PNG_LIBPNG_VER_BUILD 0 /* Release Status */ #define PNG_LIBPNG_BUILD_ALPHA 1 @@ -354,14 +426,14 @@ #define PNG_LIBPNG_BUILD_SPECIAL 32 /* Cannot be OR'ed with PNG_LIBPNG_BUILD_PRIVATE */ -#define PNG_LIBPNG_BUILD_BASE_TYPE PNG_LIBPNG_BUILD_BETA +#define PNG_LIBPNG_BUILD_BASE_TYPE PNG_LIBPNG_BUILD_STABLE /* Careful here. At one time, Guy wanted to use 082, but that would be octal. * We must not include leading zeros. * Versions 0.7 through 1.0.0 were in the range 0 to 100 here (only * version 1.0.0 was mis-numbered 100 instead of 10000). From * version 1.0.1 it's xxyyzz, where x=major, y=minor, z=release */ -#define PNG_LIBPNG_VER 10211 /* 1.2.11 */ +#define PNG_LIBPNG_VER 10229 /* 1.2.29 */ #ifndef PNG_VERSION_INFO_ONLY /* include the compression library's header */ @@ -449,7 +521,7 @@ extern "C" { * the version above. */ #ifdef PNG_USE_GLOBAL_ARRAYS -PNG_EXPORT_VAR (const char) png_libpng_ver[18]; +PNG_EXPORT_VAR (PNG_CONST char) png_libpng_ver[18]; /* need room for 99.99.99beta99z */ #else #define png_libpng_ver png_get_header_ver(NULL) @@ -458,17 +530,14 @@ PNG_EXPORT_VAR (const char) png_libpng_ver[18]; #ifdef PNG_USE_GLOBAL_ARRAYS /* This was removed in version 1.0.5c */ /* Structures to facilitate easy interlacing. See png.c for more details */ -PNG_EXPORT_VAR (const int FARDATA) png_pass_start[7]; -PNG_EXPORT_VAR (const int FARDATA) png_pass_inc[7]; -PNG_EXPORT_VAR (const int FARDATA) png_pass_ystart[7]; -PNG_EXPORT_VAR (const int FARDATA) png_pass_yinc[7]; -PNG_EXPORT_VAR (const int FARDATA) png_pass_mask[7]; -PNG_EXPORT_VAR (const int FARDATA) png_pass_dsp_mask[7]; -#ifdef PNG_USE_PNGGCCRD -PNG_EXPORT_VAR (const int FARDATA) png_pass_width[7]; -#endif +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_start[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_inc[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_ystart[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_yinc[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_mask[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_dsp_mask[7]; /* This isn't currently used. If you need it, see png.c for more details. -PNG_EXPORT_VAR (const int FARDATA) png_pass_height[7]; +PNG_EXPORT_VAR (PNG_CONST int FARDATA) png_pass_height[7]; */ #endif @@ -605,9 +674,10 @@ typedef png_time FAR * FAR * png_timepp; * up private chunks for output even though the library doesn't actually * know about their semantics. */ +#define PNG_CHUNK_NAME_LENGTH 5 typedef struct png_unknown_chunk_t { - png_byte name[5]; + png_byte name[PNG_CHUNK_NAME_LENGTH]; png_byte *data; png_size_t size; @@ -1134,10 +1204,12 @@ struct png_struct_def png_uint_32 row_number; /* current row in interlace pass */ png_bytep prev_row; /* buffer to save previous (unfiltered) row */ png_bytep row_buf; /* buffer to save current (unfiltered) row */ +#ifndef PNG_NO_WRITE_FILTERING png_bytep sub_row; /* buffer to save "sub" row when filtering */ png_bytep up_row; /* buffer to save "up" row when filtering */ png_bytep avg_row; /* buffer to save "avg" row when filtering */ png_bytep paeth_row; /* buffer to save "Paeth" row when filtering */ +#endif png_row_info row_info; /* used for transformation routines */ png_uint_32 idat_size; /* current IDAT size for read */ @@ -1239,8 +1311,7 @@ struct png_struct_def png_size_t current_text_left; /* how much text left to read in input */ png_charp current_text; /* current text chunk buffer */ png_charp current_text_ptr; /* current location in current_text */ -# endif /* PNG_PROGRESSIVE_READ_SUPPORTED && PNG_TEXT_SUPPORTED */ - +# endif /* PNG_TEXT_SUPPORTED */ #endif /* PNG_PROGRESSIVE_READ_SUPPORTED */ #if defined(__TURBOC__) && !defined(_Windows) && !defined(__FLAT__) @@ -1322,16 +1393,20 @@ struct png_struct_def png_byte filter_type; #endif -#if defined(PNG_1_0_X) || (defined(PNG_DEBUG) && defined(PNG_USE_PNGGCCRD)) +#if defined(PNG_1_0_X) /* New member added in libpng-1.0.10, ifdef'ed out in 1.2.0 */ png_uint_32 row_buf_size; #endif /* New members added in libpng-1.2.0 */ -#if !defined(PNG_1_0_X) && defined(PNG_ASSEMBLER_CODE_SUPPORTED) +#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) +# if !defined(PNG_1_0_X) +# if defined(PNG_MMX_CODE_SUPPORTED) png_byte mmx_bitdepth_threshold; png_uint_32 mmx_rowbytes_threshold; +# endif png_uint_32 asm_flags; +# endif #endif /* New members added in libpng-1.0.2 but first enabled by default in 1.2.0 */ @@ -1361,13 +1436,21 @@ struct png_struct_def png_uint_32 user_height_max; #endif +/* New member added in libpng-1.0.25 and 1.2.17 */ +#if defined(PNG_UNKNOWN_CHUNKS_SUPPORTED) + /* storage for unknown chunk that the library doesn't recognize. */ + png_unknown_chunk unknown_chunk; +#endif + +/* New members added in libpng-1.2.26 */ + png_uint_32 old_big_row_buf_size, old_prev_row_size; }; /* This triggers a compiler error in png.c, if png.c and png.h * do not agree upon the version number. */ -typedef png_structp version_1_2_11beta4; +typedef png_structp version_1_2_29; typedef png_struct FAR * FAR * png_structpp; @@ -2015,6 +2098,7 @@ extern void *png_far_to_near PNGARG((png_structp png_ptr,png_voidp ptr, int check)); #endif /* USE_FAR_KEYWORD */ +#ifndef PNG_NO_ERROR_TEXT /* Fatal error in PNG image of libpng - can't continue */ extern PNG_EXPORT(void,png_error) PNGARG((png_structp png_ptr, png_const_charp error_message)); @@ -2022,14 +2106,22 @@ extern PNG_EXPORT(void,png_error) PNGARG((png_structp png_ptr, /* The same, but the chunk name is prepended to the error string. */ extern PNG_EXPORT(void,png_chunk_error) PNGARG((png_structp png_ptr, png_const_charp error_message)); +#else +/* Fatal error in PNG image of libpng - can't continue */ +extern PNG_EXPORT(void,png_err) PNGARG((png_structp png_ptr)); +#endif +#ifndef PNG_NO_WARNINGS /* Non-fatal error in libpng. Can continue, but may have a problem. */ extern PNG_EXPORT(void,png_warning) PNGARG((png_structp png_ptr, png_const_charp warning_message)); +#ifdef PNG_READ_SUPPORTED /* Non-fatal error in libpng, chunk name is prepended to message. */ extern PNG_EXPORT(void,png_chunk_warning) PNGARG((png_structp png_ptr, png_const_charp warning_message)); +#endif /* PNG_READ_SUPPORTED */ +#endif /* PNG_NO_WARNINGS */ /* The png_set_ functions are for storing values in the png_info_struct. * Similarly, the png_get_ calls are used to read values from the @@ -2363,7 +2455,7 @@ extern PNG_EXPORT(void,png_set_sCAL_s) PNGARG((png_structp png_ptr, handling or default unknown chunk handling is not desired. Any chunks not listed will be handled in the default manner. The IHDR and IEND chunks must not be listed. - keep = 0: follow default behavour + keep = 0: follow default behaviour = 1: do not keep = 2: keep only if safe-to-copy = 3: keep even if unsafe-to-copy @@ -2451,10 +2543,6 @@ extern PNG_EXPORT(void, png_write_png) PNGARG((png_structp png_ptr, #define png_debug2(l, m, p1, p2) #endif -#if 0 -extern PNG_EXPORT(png_bytep,png_sig_bytes) PNGARG((void)); -#endif - extern PNG_EXPORT(png_charp,png_get_copyright) PNGARG((png_structp png_ptr)); extern PNG_EXPORT(png_charp,png_get_header_ver) PNGARG((png_structp png_ptr)); extern PNG_EXPORT(png_charp,png_get_header_version) PNGARG((png_structp png_ptr)); @@ -2473,6 +2561,7 @@ extern PNG_EXPORT(png_uint_32,png_permit_mng_features) PNGARG((png_structp /* Added to version 1.2.0 */ #if defined(PNG_ASSEMBLER_CODE_SUPPORTED) +#if defined(PNG_MMX_CODE_SUPPORTED) #define PNG_ASM_FLAG_MMX_SUPPORT_COMPILED 0x01 /* not user-settable */ #define PNG_ASM_FLAG_MMX_SUPPORT_IN_CPU 0x02 /* not user-settable */ #define PNG_ASM_FLAG_MMX_READ_COMBINE_ROW 0x04 @@ -2498,6 +2587,7 @@ extern PNG_EXPORT(png_uint_32,png_permit_mng_features) PNGARG((png_structp #define PNG_SELECT_READ 1 #define PNG_SELECT_WRITE 2 +#endif /* PNG_MMX_CODE_SUPPORTED */ #if !defined(PNG_1_0_X) /* pngget.c */ @@ -2530,11 +2620,11 @@ extern PNG_EXPORT(void,png_set_mmx_thresholds) png_uint_32 mmx_rowbytes_threshold)); #endif /* PNG_1_0_X */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ #if !defined(PNG_1_0_X) /* png.c, pnggccrd.c, or pngvcrd.c */ extern PNG_EXPORT(int,png_mmx_support) PNGARG((void)); +#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ /* Strip the prepended error numbers ("#nnn ") from error and warning * messages before passing them to the error or warning handler. */ @@ -2642,16 +2732,21 @@ extern PNG_EXPORT(void,png_save_uint_16) * be found in the files where the functions are located. */ -#if defined(PNG_INTERNAL) -/* Various modes of operation. Note that after an init, mode is set to - * zero automatically when the structure is created. +/* Various modes of operation, that are visible to applications because + * they are used for unknown chunk location. */ #define PNG_HAVE_IHDR 0x01 #define PNG_HAVE_PLTE 0x02 #define PNG_HAVE_IDAT 0x04 #define PNG_AFTER_IDAT 0x08 /* Have complete zlib datastream */ #define PNG_HAVE_IEND 0x10 + +#if defined(PNG_INTERNAL) + +/* More modes of operation. Note that after an init, mode is set to + * zero automatically when the structure is created. + */ #define PNG_HAVE_gAMA 0x20 #define PNG_HAVE_cHRM 0x40 #define PNG_HAVE_sRGB 0x80 @@ -2771,11 +2866,8 @@ extern PNG_EXPORT(void,png_save_uint_16) #if !defined(PNG_NO_EXTERN) || defined(PNG_ALWAYS_EXTERN) /* place to hold the signature string for a PNG file. */ #ifdef PNG_USE_GLOBAL_ARRAYS - PNG_EXPORT_VAR (const png_byte FARDATA) png_sig[8]; + PNG_EXPORT_VAR (PNG_CONST png_byte FARDATA) png_sig[8]; #else -#if 0 -#define png_sig png_sig_bytes(NULL) -#endif #endif #endif /* PNG_NO_EXTERN */ @@ -2783,50 +2875,50 @@ extern PNG_EXPORT(void,png_save_uint_16) * define the name here, and add an invocation of the macro in png.c and * wherever it's needed. */ -#define PNG_IHDR const png_byte png_IHDR[5] = { 73, 72, 68, 82, '\0'} -#define PNG_IDAT const png_byte png_IDAT[5] = { 73, 68, 65, 84, '\0'} -#define PNG_IEND const png_byte png_IEND[5] = { 73, 69, 78, 68, '\0'} -#define PNG_PLTE const png_byte png_PLTE[5] = { 80, 76, 84, 69, '\0'} -#define PNG_bKGD const png_byte png_bKGD[5] = { 98, 75, 71, 68, '\0'} -#define PNG_cHRM const png_byte png_cHRM[5] = { 99, 72, 82, 77, '\0'} -#define PNG_gAMA const png_byte png_gAMA[5] = {103, 65, 77, 65, '\0'} -#define PNG_hIST const png_byte png_hIST[5] = {104, 73, 83, 84, '\0'} -#define PNG_iCCP const png_byte png_iCCP[5] = {105, 67, 67, 80, '\0'} -#define PNG_iTXt const png_byte png_iTXt[5] = {105, 84, 88, 116, '\0'} -#define PNG_oFFs const png_byte png_oFFs[5] = {111, 70, 70, 115, '\0'} -#define PNG_pCAL const png_byte png_pCAL[5] = {112, 67, 65, 76, '\0'} -#define PNG_sCAL const png_byte png_sCAL[5] = {115, 67, 65, 76, '\0'} -#define PNG_pHYs const png_byte png_pHYs[5] = {112, 72, 89, 115, '\0'} -#define PNG_sBIT const png_byte png_sBIT[5] = {115, 66, 73, 84, '\0'} -#define PNG_sPLT const png_byte png_sPLT[5] = {115, 80, 76, 84, '\0'} -#define PNG_sRGB const png_byte png_sRGB[5] = {115, 82, 71, 66, '\0'} -#define PNG_tEXt const png_byte png_tEXt[5] = {116, 69, 88, 116, '\0'} -#define PNG_tIME const png_byte png_tIME[5] = {116, 73, 77, 69, '\0'} -#define PNG_tRNS const png_byte png_tRNS[5] = {116, 82, 78, 83, '\0'} -#define PNG_zTXt const png_byte png_zTXt[5] = {122, 84, 88, 116, '\0'} +#define PNG_IHDR png_byte png_IHDR[5] = { 73, 72, 68, 82, '\0'} +#define PNG_IDAT png_byte png_IDAT[5] = { 73, 68, 65, 84, '\0'} +#define PNG_IEND png_byte png_IEND[5] = { 73, 69, 78, 68, '\0'} +#define PNG_PLTE png_byte png_PLTE[5] = { 80, 76, 84, 69, '\0'} +#define PNG_bKGD png_byte png_bKGD[5] = { 98, 75, 71, 68, '\0'} +#define PNG_cHRM png_byte png_cHRM[5] = { 99, 72, 82, 77, '\0'} +#define PNG_gAMA png_byte png_gAMA[5] = {103, 65, 77, 65, '\0'} +#define PNG_hIST png_byte png_hIST[5] = {104, 73, 83, 84, '\0'} +#define PNG_iCCP png_byte png_iCCP[5] = {105, 67, 67, 80, '\0'} +#define PNG_iTXt png_byte png_iTXt[5] = {105, 84, 88, 116, '\0'} +#define PNG_oFFs png_byte png_oFFs[5] = {111, 70, 70, 115, '\0'} +#define PNG_pCAL png_byte png_pCAL[5] = {112, 67, 65, 76, '\0'} +#define PNG_sCAL png_byte png_sCAL[5] = {115, 67, 65, 76, '\0'} +#define PNG_pHYs png_byte png_pHYs[5] = {112, 72, 89, 115, '\0'} +#define PNG_sBIT png_byte png_sBIT[5] = {115, 66, 73, 84, '\0'} +#define PNG_sPLT png_byte png_sPLT[5] = {115, 80, 76, 84, '\0'} +#define PNG_sRGB png_byte png_sRGB[5] = {115, 82, 71, 66, '\0'} +#define PNG_tEXt png_byte png_tEXt[5] = {116, 69, 88, 116, '\0'} +#define PNG_tIME png_byte png_tIME[5] = {116, 73, 77, 69, '\0'} +#define PNG_tRNS png_byte png_tRNS[5] = {116, 82, 78, 83, '\0'} +#define PNG_zTXt png_byte png_zTXt[5] = {122, 84, 88, 116, '\0'} #ifdef PNG_USE_GLOBAL_ARRAYS -PNG_EXPORT_VAR (const png_byte FARDATA) png_IHDR[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_IDAT[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_IEND[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_PLTE[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_bKGD[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_cHRM[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_gAMA[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_hIST[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_iCCP[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_iTXt[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_oFFs[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_pCAL[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_sCAL[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_pHYs[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_sBIT[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_sPLT[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_sRGB[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_tEXt[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_tIME[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_tRNS[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_zTXt[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_IHDR[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_IDAT[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_IEND[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_PLTE[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_bKGD[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_cHRM[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_gAMA[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_hIST[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_iCCP[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_iTXt[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_oFFs[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_pCAL[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_sCAL[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_pHYs[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_sBIT[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_sPLT[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_sRGB[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_tEXt[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_tIME[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_tRNS[5]; +PNG_EXPORT_VAR (png_byte FARDATA) png_zTXt[5]; #endif /* PNG_USE_GLOBAL_ARRAYS */ #if defined(PNG_1_0_X) || defined (PNG_1_2_X) @@ -3436,9 +3528,34 @@ PNG_EXTERN void png_do_write_intrapixel PNGARG((png_row_infop row_info, #endif #if defined(PNG_ASSEMBLER_CODE_SUPPORTED) +#if defined(PNG_MMX_CODE_SUPPORTED) /* png.c */ /* PRIVATE */ PNG_EXTERN void png_init_mmx_flags PNGARG((png_structp png_ptr)); #endif +#endif + +#if defined(PNG_INCH_CONVERSIONS) && defined(PNG_FLOATING_POINT_SUPPORTED) +PNG_EXTERN png_uint_32 png_get_pixels_per_inch PNGARG((png_structp png_ptr, +png_infop info_ptr)); + +PNG_EXTERN png_uint_32 png_get_x_pixels_per_inch PNGARG((png_structp png_ptr, +png_infop info_ptr)); + +PNG_EXTERN png_uint_32 png_get_y_pixels_per_inch PNGARG((png_structp png_ptr, +png_infop info_ptr)); + +PNG_EXTERN float png_get_x_offset_inches PNGARG((png_structp png_ptr, +png_infop info_ptr)); + +PNG_EXTERN float png_get_y_offset_inches PNGARG((png_structp png_ptr, +png_infop info_ptr)); + +#if defined(PNG_pHYs_SUPPORTED) +PNG_EXTERN png_uint_32 png_get_pHYs_dpi PNGARG((png_structp png_ptr, +png_infop info_ptr, png_uint_32 *res_x, png_uint_32 *res_y, int *unit_type)); +#endif /* PNG_pHYs_SUPPORTED */ +#endif /* PNG_INCH_CONVERSIONS && PNG_FLOATING_POINT_SUPPORTED */ + /* Maintainer: Put new private prototypes here ^ and in libpngpf.3 */ #endif /* PNG_INTERNAL */ diff --git a/pngconf.h b/pngconf.h index 9733a8edd..7a29d92fd 100644 --- a/pngconf.h +++ b/pngconf.h @@ -1,9 +1,9 @@ /* pngconf.h - machine configurable file for libpng * - * libpng version 1.2.11beta4 - June 7, 2006 + * libpng version 1.2.29 - May 8, 2008 * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2005 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -77,6 +77,14 @@ /* End of material added to libpng-1.2.8 */ +/* Added at libpng-1.2.19, removed at libpng-1.2.20 because it caused trouble + Restored at libpng-1.2.21 */ +#if !defined(PNG_NO_WARN_UNINITIALIZED_ROW) && \ + !defined(PNG_WARN_UNINITIALIZED_ROW) +# define PNG_WARN_UNINITIALIZED_ROW 1 +#endif +/* End of material added at libpng-1.2.19/1.2.21 */ + /* This is the size of the compression buffer, and thus the size of * an IDAT chunk. Make this whatever size you feel is best for your * machine. One of these will be allocated per png_struct. When this @@ -278,6 +286,7 @@ # define PNGARG(arglist) arglist #endif /* _NO_PROTO */ + #endif /* OF */ #endif /* PNGARG */ @@ -316,7 +325,7 @@ /* If you encounter a compiler error here, see the explanation * near the end of INSTALL. */ - __png.h__ already includes setjmp.h; + __pngconf.h__ already includes setjmp.h; __dont__ include it again.; # endif # endif /* __linux__ */ @@ -326,7 +335,9 @@ # ifdef __linux__ # ifdef PNG_SAVE_BSD_SOURCE -# define _BSD_SOURCE +# ifndef _BSD_SOURCE +# define _BSD_SOURCE +# endif # undef PNG_SAVE_BSD_SOURCE # endif # endif /* __linux__ */ @@ -596,16 +607,16 @@ #endif /* PNG_READ_TRANSFORMS_SUPPORTED */ #if !defined(PNG_NO_PROGRESSIVE_READ) && \ - !defined(PNG_PROGRESSIVE_READ_NOT_SUPPORTED) /* if you don't do progressive */ -# define PNG_PROGRESSIVE_READ_SUPPORTED /* reading. This is not talking */ -#endif /* about interlacing capability! You'll */ - /* still have interlacing unless you change the following line: */ + !defined(PNG_PROGRESSIVE_READ_SUPPORTED) /* if you don't do progressive */ +# define PNG_PROGRESSIVE_READ_SUPPORTED /* reading. This is not talking */ +#endif /* about interlacing capability! You'll */ + /* still have interlacing unless you change the following line: */ -#define PNG_READ_INTERLACING_SUPPORTED /* required for PNG-compliant decoders */ +#define PNG_READ_INTERLACING_SUPPORTED /* required in PNG-compliant decoders */ #ifndef PNG_NO_READ_COMPOSITE_NODIV # ifndef PNG_NO_READ_COMPOSITED_NODIV /* libpng-1.0.x misspelling */ -# define PNG_READ_COMPOSITE_NODIV_SUPPORTED /* well tested on Intel, SGI */ +# define PNG_READ_COMPOSITE_NODIV_SUPPORTED /* well tested on Intel, SGI */ # endif #endif @@ -723,26 +734,46 @@ #endif /* PNG_ASSEMBLER_CODE was enabled by default in version 1.2.0 - even when PNG_USE_PNGVCRD or PNG_USE_PNGGCCRD is not defined */ + * and removed from version 1.2.20. The following will be removed + * from libpng-1.4.0 +*/ + +#if defined(PNG_READ_SUPPORTED) && !defined(PNG_NO_OPTIMIZED_CODE) +# ifndef PNG_OPTIMIZED_CODE_SUPPORTED +# define PNG_OPTIMIZED_CODE_SUPPORTED +# endif +#endif + #if defined(PNG_READ_SUPPORTED) && !defined(PNG_NO_ASSEMBLER_CODE) # ifndef PNG_ASSEMBLER_CODE_SUPPORTED # define PNG_ASSEMBLER_CODE_SUPPORTED # endif -# if !defined(PNG_MMX_CODE_SUPPORTED) && !defined(PNG_NO_MMX_CODE) && \ - defined(__MMX__) + +# if defined(__GNUC__) && defined(__x86_64__) && (__GNUC__ < 4) + /* work around 64-bit gcc compiler bugs in gcc-3.x */ +# if !defined(PNG_MMX_CODE_SUPPORTED) && !defined(PNG_NO_MMX_CODE) +# define PNG_NO_MMX_CODE +# endif +# endif + +# if defined(__APPLE__) +# if !defined(PNG_MMX_CODE_SUPPORTED) && !defined(PNG_NO_MMX_CODE) +# define PNG_NO_MMX_CODE +# endif +# endif + +# if (defined(__MWERKS__) && ((__MWERKS__ < 0x0900) || macintosh)) +# if !defined(PNG_MMX_CODE_SUPPORTED) && !defined(PNG_NO_MMX_CODE) +# define PNG_NO_MMX_CODE +# endif +# endif + +# if !defined(PNG_MMX_CODE_SUPPORTED) && !defined(PNG_NO_MMX_CODE) # define PNG_MMX_CODE_SUPPORTED # endif -# if !defined(PNG_USE_PNGGCCRD) && !defined(PNG_NO_MMX_CODE) && \ - !defined(PNG_USE_PNGVCRD) && defined(__MMX__) -# define PNG_USE_PNGGCCRD -# endif -#endif -/* If you are sure that you don't need thread safety and you are compiling - with PNG_USE_PNGCCRD for an MMX application, you can define this for - faster execution. See pnggccrd.c. -#define PNG_THREAD_UNSAFE_OK -*/ +#endif +/* end of obsolete code to be removed from libpng-1.4.0 */ #if !defined(PNG_1_0_X) #if !defined(PNG_NO_USER_MEM) && !defined(PNG_USER_MEM_SUPPORTED) @@ -1264,7 +1295,8 @@ typedef z_stream FAR * png_zstreamp; * but might be required for some pre-1.0.5c applications. */ #if !defined(PNG_USE_LOCAL_ARRAYS) && !defined(PNG_USE_GLOBAL_ARRAYS) -# if defined(PNG_NO_GLOBAL_ARRAYS) || (defined(__GNUC__) && defined(PNG_DLL)) +# if defined(PNG_NO_GLOBAL_ARRAYS) || \ + (defined(__GNUC__) && defined(PNG_DLL)) || defined(_MSC_VER) # define PNG_USE_LOCAL_ARRAYS # else # define PNG_USE_GLOBAL_ARRAYS @@ -1401,8 +1433,7 @@ typedef z_stream FAR * png_zstreamp; # define NOCHECK 0 # define CVT_PTR(ptr) (png_far_to_near(png_ptr,ptr,CHECK)) # define CVT_PTR_NOCHECK(ptr) (png_far_to_near(png_ptr,ptr,NOCHECK)) -# define png_strcpy _fstrcpy -# define png_strncpy _fstrncpy /* Added to v 1.2.6 */ +# define png_snprintf _fsnprintf /* Added to v 1.2.19 */ # define png_strlen _fstrlen # define png_memcmp _fmemcmp /* SJT: added */ # define png_memcpy _fmemcpy @@ -1410,8 +1441,27 @@ typedef z_stream FAR * png_zstreamp; #else /* use the usual functions */ # define CVT_PTR(ptr) (ptr) # define CVT_PTR_NOCHECK(ptr) (ptr) -# define png_strcpy strcpy -# define png_strncpy strncpy /* Added to v 1.2.6 */ +# ifndef PNG_NO_SNPRINTF +# ifdef _MSC_VER +# define png_snprintf _snprintf /* Added to v 1.2.19 */ +# define png_snprintf2 _snprintf +# define png_snprintf6 _snprintf +# else +# define png_snprintf snprintf /* Added to v 1.2.19 */ +# define png_snprintf2 snprintf +# define png_snprintf6 snprintf +# endif +# else + /* You don't have or don't want to use snprintf(). Caution: Using + * sprintf instead of snprintf exposes your application to accidental + * or malevolent buffer overflows. If you don't have snprintf() + * as a general rule you should provide one (you can get one from + * Portable OpenSSH). */ +# define png_snprintf(s1,n,fmt,x1) sprintf(s1,fmt,x1) +# define png_snprintf2(s1,n,fmt,x1,x2) sprintf(s1,fmt,x1,x2) +# define png_snprintf6(s1,n,fmt,x1,x2,x3,x4,x5,x6) \ + sprintf(s1,fmt,x1,x2,x3,x4,x5,x6) +# endif # define png_strlen strlen # define png_memcmp memcmp /* SJT: added */ # define png_memcpy memcpy @@ -1427,47 +1477,6 @@ typedef z_stream FAR * png_zstreamp; # define PNG_ZBUF_SIZE 65536L #endif -#ifdef PNG_READ_SUPPORTED -/* Prior to libpng-1.0.9, this block was in pngasmrd.h */ -#if defined(PNG_INTERNAL) - -/* These are the default thresholds before the MMX code kicks in; if either - * rowbytes or bitdepth is below the threshold, plain C code is used. These - * can be overridden at runtime via the png_set_mmx_thresholds() call in - * libpng 1.2.0 and later. The values below were chosen by Intel. - */ - -#ifndef PNG_MMX_ROWBYTES_THRESHOLD_DEFAULT -# define PNG_MMX_ROWBYTES_THRESHOLD_DEFAULT 128 /* >= */ -#endif -#ifndef PNG_MMX_BITDEPTH_THRESHOLD_DEFAULT -# define PNG_MMX_BITDEPTH_THRESHOLD_DEFAULT 9 /* >= */ -#endif - -/* Set this in the makefile for VC++ on Pentium, not here. */ -/* Platform must be Pentium. Makefile must assemble and load pngvcrd.c . - * MMX will be detected at run time and used if present. - */ -#ifdef PNG_USE_PNGVCRD -# define PNG_HAVE_ASSEMBLER_COMBINE_ROW -# define PNG_HAVE_ASSEMBLER_READ_INTERLACE -# define PNG_HAVE_ASSEMBLER_READ_FILTER_ROW -#endif - -/* Set this in the makefile for gcc/as on Pentium, not here. */ -/* Platform must be Pentium. Makefile must assemble and load pnggccrd.c . - * MMX will be detected at run time and used if present. - */ -#ifdef PNG_USE_PNGGCCRD -# define PNG_HAVE_ASSEMBLER_COMBINE_ROW -# define PNG_HAVE_ASSEMBLER_READ_INTERLACE -# define PNG_HAVE_ASSEMBLER_READ_FILTER_ROW -#endif -/* - see pnggccrd.c for info about what is currently enabled */ - -#endif /* PNG_INTERNAL */ -#endif /* PNG_READ_SUPPORTED */ - /* Added at libpng-1.2.8 */ #endif /* PNG_VERSION_INFO_ONLY */ diff --git a/pngcrush.c b/pngcrush.c index 906b9f0ee..9c8db38ea 100644 --- a/pngcrush.c +++ b/pngcrush.c @@ -1,5 +1,5 @@ /* pngcrush.c - recompresses png files - * Copyright (C) 1998-2002,2006 Glenn Randers-Pehrson (glennrp@users.sf.net) + * Copyright (C) 1998-2002,2006-2008 Glenn Randers-Pehrson (glennrp@users.sf.net) * Copyright (C) 2005 Greg Roelofs * * The most recent version of pngcrush can be found at SourceForge in @@ -26,7 +26,7 @@ * */ -#define PNGCRUSH_VERSION "1.6.4" +#define PNGCRUSH_VERSION "1.6.5" /* #define PNGCRUSH_COUNT_COLORS @@ -38,7 +38,7 @@ * If you have modified this source, you may insert additional notices * immediately after this sentence. * - * Copyright (C) 1998-2002,2006 Glenn Randers-Pehrson (glennrp@users.sf.net) + * Copyright (C) 1998-2002,2006-2008 Glenn Randers-Pehrson (glennrp@users.sf.net) * Copyright (C) 2005 Greg Roelofs * * The pngcrush computer program is supplied "AS IS". The Author disclaims all @@ -128,13 +128,6 @@ #define PNG_iCCP const png_byte png_iCCP[5] = {105, 67, 67, 80, '\0'} #define PNG_IEND const png_byte png_IEND[5] = { 73, 69, 78, 68, '\0'} -#if 0 -PNG_EXPORT_VAR (const png_byte FARDATA) png_IHDR[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_IDAT[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_IEND[5]; -PNG_EXPORT_VAR (const png_byte FARDATA) png_iCCP[5]; -#endif - #define PNG_FLAG_CRC_CRITICAL_USE 0x0400 #define PNG_FLAG_CRC_CRITICAL_IGNORE 0x0800 #define PNG_FLAG_CRC_ANCILLARY_USE 0x0100 @@ -171,6 +164,10 @@ PNG_EXPORT_VAR (const png_byte FARDATA) png_iCCP[5]; # define PNGCRUSH_LOCO #endif +#ifndef PNG_UINT_31_MAX +#define PNG_UINT_31_MAX ((png_uint_32)0x7fffffffL) +#endif + /* These macros were renamed in libpng-1.2.6 */ #ifndef PNG_HANDLE_CHUNK_ALWAYS #define PNG_HANDLE_CHUNK_ALWAYS HANDLE_CHUNK_ALWAYS @@ -268,11 +265,14 @@ struct options_help { static PNG_CONST char *progname; static PNG_CONST char *inname = "pngtest" DOT "png"; static PNG_CONST char *outname = "pngout" DOT "png"; +static PNG_CONST char *mngname = "mngout" DOT "mng"; static PNG_CONST char *directory_name = "pngcrush" DOT "bak"; static PNG_CONST char *extension = "_C" DOT "png"; static png_uint_32 width, height; static png_uint_32 measured_idat_length; +static int found_gAMA = 0; +static int found_cHRM = 0; static int pngcrush_must_exit = 0; static int all_chunks_are_safe = 0; static int number_of_open_files; @@ -352,6 +352,9 @@ static int method = 10; static int pauses = 0; static int nosave = 0; static int nofilecheck = 0; +#ifdef PNGCRUSH_LOCO +static int new_mng = 0; +#endif static png_bytep row_buf; #ifdef PNGCRUSH_MULTIPLE_ROWS static png_bytepp row_pointers; @@ -397,11 +400,11 @@ static int num_palette; static png_byte palette_reorder[256]; #endif -static png_structp read_ptr, write_ptr; +static png_structp read_ptr, write_ptr, mng_ptr; static png_infop read_info_ptr, write_info_ptr; static png_infop end_info_ptr; static png_infop write_end_info_ptr; -static FILE *fpin, *fpout; +static FILE *fpin, *fpout, *mng_out; png_uint_32 measure_idats(FILE * fpin); #ifdef PNGCRUSH_LOCO static int do_loco = 0; @@ -1017,8 +1020,8 @@ png_voidp png_debug_malloc(png_structp png_ptr, png_uint_32 size) /* Make sure the caller isn't assuming zeroed memory. */ png_memset(pinfo->pointer, 0xdd, pinfo->size); if (verbose > 2) - fprintf(STDERR, "Pointer %x allocated %lu bytes\n", - (int) pinfo->pointer, size); + fprintf(STDERR, "Pointer %lux allocated %lu bytes\n", + (unsigned long) pinfo->pointer, (unsigned long)size); return (png_voidp) (pinfo->pointer); } } @@ -1052,13 +1055,14 @@ void png_debug_free(png_structp png_ptr, png_voidp ptr) the memory that is to be freed. */ memset(ptr, 0x55, pinfo->size); if (verbose > 2) - fprintf(STDERR, "Pointer %x freed %lu bytes\n", - (int) ptr, pinfo->size); + fprintf(STDERR, "Pointer %lux freed %lu bytes\n", + (unsigned long) ptr, (unsigned long)pinfo->size); png_free_default(png_ptr, pinfo); break; } if (pinfo->next == NULL) { - fprintf(STDERR, "Pointer %x not found\n", (int) ptr); + fprintf(STDERR, "Pointer %lux not found\n", + (unsigned long) ptr); break; } ppinfo = &pinfo->next; @@ -1309,23 +1313,23 @@ void show_result(void) fprintf(STDERR, " Overall result: no change\n"); else if (total_input_length > total_output_length) fprintf(STDERR, - " Overall result: %4.2f%% reduction, %ld bytes\n", + " Overall result: %4.2f%% reduction, %lu bytes\n", (100.0 - (100.0 * total_output_length) / total_input_length), - total_input_length - total_output_length); + (unsigned long)(total_input_length-total_output_length)); else fprintf(STDERR, - " Overall result: %4.2f%% increase, %ld bytes\n", + " Overall result: %4.2f%% increase, %lu bytes\n", -(100.0 - (100.0 * total_output_length) / total_input_length), - total_output_length - total_input_length); + (unsigned long)(total_output_length - total_input_length)); } t_stop = (TIME_T) clock(); t_misc += (t_stop - t_start); if (t_stop < t_start) { - t_misc += PNG_MAX_UINT; + t_misc += PNG_UINT_31_MAX; if (t_stop < 0) - t_misc += PNG_MAX_UINT; + t_misc += PNG_UINT_31_MAX; } t_start = t_stop; fprintf(STDERR, " CPU time used = %.3f seconds", @@ -1342,8 +1346,8 @@ void show_result(void) fprintf(STDERR, "MEMORY ERROR: %d bytes still allocated\n", current_allocation); while (pinfo != NULL) { - fprintf(STDERR, " %8lu bytes at %x\n", pinfo->size, - (int) pinfo->pointer); + fprintf(STDERR, " %8lu bytes at %lux\n", (unsigned long)pinfo->size, + (unsigned long) pinfo->pointer); free(pinfo->pointer); pinfo = pinfo->next; } @@ -1551,6 +1555,7 @@ int main(int argc, char *argv[]) #ifdef PNG_gAMA_SUPPORTED else if (!strncmp(argv[i], "-dou", 4)) { double_gamma++; + found_gAMA=1; global_things_have_changed = 1; } #endif @@ -1646,6 +1651,7 @@ int main(int argc, char *argv[]) else if (!strncmp(argv[i], "-g", 2)) { names++; BUMP_I; + found_gAMA=1; if (intent < 0) { #ifdef PNG_FIXED_POINT_SUPPORTED int c; @@ -1705,8 +1711,15 @@ int main(int argc, char *argv[]) names++; BUMP_I; max_idat_size = (png_uint_32) atoi(argv[i]); - if (max_idat_size == 0 || max_idat_size > PNG_MAX_UINT) + if (max_idat_size == 0 || max_idat_size > PNG_UINT_31_MAX) max_idat_size = PNG_ZBUF_SIZE; +#ifdef PNGCRUSH_LOCO + } else if (!strncmp(argv[i], "-mng", 4)) { + names++; + BUMP_I; + mngname = argv[i]; + new_mng++; +#endif } else if (!strncmp(argv[i], "-m", 2)) { names++; BUMP_I; @@ -1742,6 +1755,7 @@ int main(int argc, char *argv[]) else if (!strncmp(argv[i], "-rep", 4)) { names++; BUMP_I; + found_gAMA=1; { #ifdef PNG_FIXED_POINT_SUPPORTED int c; @@ -1823,7 +1837,7 @@ int main(int argc, char *argv[]) i += 2; BUMP_I; i -= 3; - if (strlen(argv[i + 2]) < 80 && strlen(argv[i + 3]) < 2048 && + if (strlen(argv[i + 2]) < 180 && strlen(argv[i + 3]) < 2048 && text_inputs < 10) { #ifdef PNG_iTXt_SUPPORTED if (!strncmp(argv[i], "-zi", 3)) { @@ -2135,8 +2149,47 @@ int main(int argc, char *argv[]) already_crushed = 0; +#ifdef PNGCRUSH_LOCO + if (new_mng) { + +#ifdef PNG_USER_MEM_SUPPORTED + mng_ptr = png_create_write_struct_2(PNG_LIBPNG_VER_STRING, + (png_voidp) NULL, (png_error_ptr) png_cexcept_error, + (png_error_ptr) NULL, (png_voidp) NULL, + (png_malloc_ptr) png_debug_malloc, + (png_free_ptr) png_debug_free); +#else + mng_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING, + (png_voidp) NULL, (png_error_ptr) png_cexcept_error, + (png_error_ptr) NULL); +#endif + if (mng_ptr == NULL) + fprintf(STDERR, "pngcrush could not create mng_ptr"); + + if ((mng_out = FOPEN(mngname, "wb")) == NULL) { + fprintf(STDERR, "Could not open output file %s\n", + mngname); + FCLOSE(fpin); + exit(1); + } + number_of_open_files++; + png_init_io(mng_ptr, mng_out); + png_set_write_fn(mng_ptr, (png_voidp) mng_out, + (png_rw_ptr) NULL, + NULL); +#endif + + } + idat_length[0] = measure_idats(fpin); +#ifdef PNGCRUSH_LOCO + if (new_mng) { + png_destroy_write_struct(&mng_ptr, NULL); + FCLOSE(mng_out); + } +#endif + FCLOSE(fpin); if (already_crushed) { @@ -2149,7 +2202,7 @@ int main(int argc, char *argv[]) fprintf(STDERR, " Recompressing %s\n", inname); fprintf(STDERR, " Total length of data found in IDAT chunks = %8lu\n", - idat_length[0]); + (unsigned long)idat_length[0]); fflush(STDERR); } @@ -2612,10 +2665,17 @@ int main(int argc, char *argv[]) if (!(int) (png_memcmp(mng_signature, png_signature, 8))) { png_byte buffer[40]; + unsigned long length; /* Skip the MHDR */ png_permit_mng_features(read_ptr, PNG_FLAG_MNG_FILTER_64); - png_default_read_data(read_ptr, buffer, 40); + png_default_read_data(read_ptr, buffer, 4); + length=buffer[3]+(buffer[2]<<8)+(buffer[1]<<16)+(buffer[0]<<24); + png_default_read_data(read_ptr, buffer, 4); + printf("Skipping %c%c%c%c chunk.\n",buffer[0],buffer[1], + buffer[2],buffer[3]); + png_default_read_data(read_ptr, buffer, length); + png_default_read_data(read_ptr, buffer, 4); input_format = 1; } else #endif @@ -2630,7 +2690,7 @@ int main(int argc, char *argv[]) png_read_info(read_ptr, read_info_ptr); - { /* GRR added for quick %-navigation (1) */ + /* { GRR added for quick %-navigation (1) */ /* Start of chunk-copying/removal code, in order: * - IHDR @@ -2673,8 +2733,9 @@ int main(int argc, char *argv[]) if (verbose > 1 && first_trial) { fprintf(STDERR, " IHDR chunk data:\n"); fprintf(STDERR, - " Width=%ld, height=%ld\n", width, - height); + " Width=%lu, height=%lu\n", + (unsigned long)width, + (unsigned long)height); fprintf(STDERR, " Bit depth =%d\n", bit_depth); fprintf(STDERR, " Color type=%d\n", @@ -2871,7 +2932,7 @@ int main(int argc, char *argv[]) png_fixed_point white_x, white_y, red_x, red_y, green_x, green_y, blue_x, blue_y; - if (png_get_cHRM_fixed + if (found_cHRM && png_get_cHRM_fixed (read_ptr, read_info_ptr, &white_x, &white_y, &red_x, &red_y, &green_x, &green_y, &blue_x, &blue_y)) { @@ -2945,10 +3006,10 @@ int main(int argc, char *argv[]) #endif } #ifdef PNG_FIXED_POINT_SUPPORTED - else if (png_get_gAMA_fixed + else if (found_gAMA && png_get_gAMA_fixed (read_ptr, read_info_ptr, &file_gamma)) #else - else if (png_get_gAMA + else if (found_gAMA && png_get_gAMA (read_ptr, read_info_ptr, &file_gamma)) #endif { @@ -3023,12 +3084,7 @@ int main(int argc, char *argv[]) intent); png_set_sRGB(write_ptr, write_info_ptr, intent); - } else if (file_gamma == 0) { - things_have_changed = 1; - png_set_sRGB_gAMA_and_cHRM(write_ptr, - write_info_ptr, - intent); - } else { + } else if (file_gamma != 0) { if (first_trial) { fprintf(STDERR, " Ignoring sRGB request; " #ifdef PNG_FIXED_POINT_SUPPORTED @@ -3036,7 +3092,8 @@ int main(int argc, char *argv[]) #else "gamma=%f" #endif - " is not approx. 0.455\n", file_gamma); + " is not approx. 0.455\n", + (unsigned long)file_gamma); } } #endif /* PNG_gAMA_SUPPORTED */ @@ -3054,9 +3111,10 @@ int main(int argc, char *argv[]) if (png_get_iCCP (read_ptr, read_info_ptr, &name, &compression_method, &profile, &proflen)) { - P1("Got iccp chunk, proflen=%lu\n", proflen); + P1("Got iCCP chunk, proflen=%lu\n", + (unsigned long)proflen); if (iccp_length) - P0("Will not replace existing iccp chunk.\n"); + P0("Will not replace existing iCCP chunk.\n"); if (keep_chunk("iCCP", argv)) png_set_iCCP(write_ptr, write_info_ptr, name, compression_method, profile, @@ -3067,7 +3125,7 @@ int main(int argc, char *argv[]) else if (iccp_length) { png_set_iCCP(write_ptr, write_info_ptr, iccp_name, 0, iccp_text, iccp_length); - P1("Wrote iccp chunk, proflen=%d\n", iccp_length); + P1("Wrote iCCP chunk, proflen=%d\n", iccp_length); } #endif @@ -3141,7 +3199,8 @@ int main(int argc, char *argv[]) res_y, unit_type); if (verbose > 0 && first_trial) fprintf(STDERR, " Added pHYs %lu %lu 1 chunk\n", - res_x, res_y); + (unsigned long)res_x, + (unsigned long)res_y); } } #endif @@ -3477,6 +3536,9 @@ int main(int argc, char *argv[]) printf(" Added a%scompressed iTXt chunk" ".\n", (added_text[0].compression == 1)? "n un" : " "); +#endif +#if 0 + printf(" key=%s.\n",(added_text[0].key)); #endif png_free(write_ptr, added_text); added_text = (png_textp) NULL; @@ -3520,11 +3582,12 @@ int main(int argc, char *argv[]) /* make an array of only those chunks we want to keep */ for (i = 0; i < num_unknowns; i++) { - P1("Handling unknown chunk %d\n", i); + P1("Handling unknown chunk %d %s\n", i, + (char *)unknowns[i].name); /* not EBCDIC-safe, but neither is keep_chunks(): */ P2(" unknown[%d] = %s (%lu bytes, location %d)\n", i, unknowns[i].name, - (png_uint_32)unknowns[i].size, + (unsigned long)unknowns[i].size, unknowns[i].location); if (keep_chunk((char *)unknowns[i].name, argv)) { memcpy(&unknowns_keep[num_unknowns_keep], @@ -3558,8 +3621,8 @@ int main(int argc, char *argv[]) } P0("unknown chunk handling done.\n"); #endif /* PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED */ - } /* GRR added for quick %-navigation (1) */ + /* } GRR added for quick %-navigation (1) */ png_read_transform_info(read_ptr, read_info_ptr); @@ -3610,8 +3673,7 @@ int main(int argc, char *argv[]) #ifdef PNGCRUSH_LOCO if (do_loco) { png_byte buffer[30]; - const png_byte png_MHDR[5] = - { 77, 72, 68, 82, '\0' }; + const png_byte png_MHDR[5] = { 77, 72, 68, 82, '\0' }; png_byte mng_signature[8] = { 138, 77, 78, 71, 13, 10, 26, 10 }; /* write the MNG 8-byte signature */ @@ -3691,7 +3753,7 @@ int main(int argc, char *argv[]) if (max_possible_size > max_idat_size) max_possible_size = max_idat_size; P2("reinitializing write zbuf to %lu.\n", - max_possible_size); + (unsigned long)max_possible_size); png_set_compression_buffer_size(write_ptr, max_possible_size); } @@ -3833,9 +3895,9 @@ int main(int argc, char *argv[]) t_stop = (TIME_T) clock(); t_misc += (t_stop - t_start); if (t_stop < t_start) { - t_misc += PNG_MAX_UINT; + t_misc += PNG_UINT_31_MAX; if (t_stop < 0) - t_misc += PNG_MAX_UINT; + t_misc += PNG_UINT_31_MAX; } t_start = t_stop; for (pass = 0; pass < num_pass; pass++) { @@ -3862,9 +3924,9 @@ int main(int argc, char *argv[]) t_stop = (TIME_T) clock(); t_decode += (t_stop - t_start); if (t_stop < t_start) { - t_decode += PNG_MAX_UINT; + t_decode += PNG_UINT_31_MAX; if (t_stop < 0) - t_decode += PNG_MAX_UINT; + t_decode += PNG_UINT_31_MAX; } t_start = t_stop; #ifdef PNGCRUSH_MULTIPLE_ROWS @@ -3876,9 +3938,9 @@ int main(int argc, char *argv[]) t_stop = (TIME_T) clock(); t_encode += (t_stop - t_start); if (t_stop < t_start) { - t_encode += PNG_MAX_UINT; + t_encode += PNG_UINT_31_MAX; if (t_stop < 0) - t_encode += PNG_MAX_UINT; + t_encode += PNG_UINT_31_MAX; } t_start = t_stop; } @@ -3889,9 +3951,9 @@ int main(int argc, char *argv[]) t_stop = (TIME_T) clock(); t_decode += (t_stop - t_start); if (t_stop < t_start) { - t_decode += PNG_MAX_UINT; + t_decode += PNG_UINT_31_MAX; if (t_stop < 0) - t_decode += PNG_MAX_UINT; + t_decode += PNG_UINT_31_MAX; } t_start = t_stop; } @@ -3928,7 +3990,7 @@ int main(int argc, char *argv[]) P1( "Reading and writing end_info data\n"); png_read_end(read_ptr, end_info_ptr); - { /* GRR: added for %-navigation (2) */ + /* { GRR: added for %-navigation (2) */ #if (defined(PNG_READ_tEXt_SUPPORTED) && defined(PNG_WRITE_tEXt_SUPPORTED)) \ || (defined(PNG_READ_zTXt_SUPPORTED) && defined(PNG_WRITE_zTXt_SUPPORTED)) @@ -4082,7 +4144,7 @@ int main(int argc, char *argv[]) } } #endif - } /* GRR: added for %-navigation (2) */ + /* } GRR: added for %-navigation (2) */ if (nosave == 0) { #if 0 /* doesn't work; compression level has to be the same as in IDAT */ @@ -4173,7 +4235,7 @@ int main(int argc, char *argv[]) if (nosave == 0) { P1( "Opening file for length measurement\n"); if ((fpin = FOPEN(outname, "rb")) == NULL) { - fprintf(STDERR, "Could not find file %s\n", outname); + fprintf(STDERR, "Could not find output file %s\n", outname); if (png_row_filters != NULL) { free(png_row_filters); png_row_filters = NULL; @@ -4191,7 +4253,7 @@ int main(int argc, char *argv[]) fprintf(STDERR, " IDAT length with method %3d (fm %d zl %d zs %d) = %8lu\n", trial, filter_type, zlib_level, z_strategy, - idat_length[trial]); + (unsigned long)idat_length[trial]); fflush(STDERR); } @@ -4303,12 +4365,13 @@ png_uint_32 measure_idats(FILE * fpin) png_set_sig_bytes(read_ptr, 0); measured_idat_length = png_measure_idat(read_ptr); - P2("measure_idats: IDAT length=%lu\n", measured_idat_length); + P2("measure_idats: IDAT length=%lu\n", + (unsigned long)measured_idat_length); P1( "Destroying data structs\n"); png_destroy_read_struct(&read_ptr, &read_info_ptr, &end_info_ptr); } Catch(msg) { - fprintf(STDERR, "\nWhile reading %s ", inname); + fprintf(STDERR, "\nWhile measuring IDATs in %s ", inname); fprintf(STDERR, "pngcrush caught libpng error:\n %s\n\n", msg); png_destroy_read_struct(&read_ptr, &read_info_ptr, &end_info_ptr); P1( "Destroyed data structs\n"); @@ -4326,6 +4389,8 @@ png_uint_32 png_measure_idat(png_structp png_ptr) /* Copyright (C) 1999-2002,2006 Glenn Randers-Pehrson (glennrp@users.sf.net) See notice in pngcrush.c for conditions of use and distribution */ png_uint_32 sum_idat_length = 0; + png_byte *bb = NULL; + png_uint_32 malloced_length=0; { png_byte png_signature[8] = { 137, 80, 78, 71, 13, 10, 26, 10 }; @@ -4338,10 +4403,50 @@ png_uint_32 png_measure_idat(png_structp png_ptr) #if defined(PNGCRUSH_LOCO) if (!(int) (png_memcmp(mng_signature, png_signature, 8))) { + const png_byte png_MHDR[5] = { 77, 72, 68, 82, '\0' }; + + int b; png_byte buffer[40]; - /* skip the MHDR */ - png_default_read_data(read_ptr, buffer, 40); + unsigned long length; + /* read the MHDR */ + png_default_read_data(read_ptr, buffer, 4); + length=buffer[3]+(buffer[2]<<8)+(buffer[1]<<16)+(buffer[0]<<24); + png_default_read_data(read_ptr, buffer, 4); + printf("Reading %c%c%c%c chunk.\n",buffer[0],buffer[1], + buffer[2],buffer[3]); + for (b=0; b<40; b++) + buffer[b]='\0'; + png_default_read_data(read_ptr, buffer, length); + if (verbose) { + printf(" width=%lu\n",(unsigned long)(buffer[3]+(buffer[2]<<8) + +(buffer[1]<<16)+(buffer[0]<<24))); + printf(" height=%lu\n",(unsigned long)(buffer[7]+(buffer[6]<<8) + +(buffer[5]<<16)+(buffer[4]<<24))); + printf(" ticksps=%lu\n",(unsigned long)(buffer[11]+ + (buffer[10]<<8)+(buffer[9]<<16)+(buffer[8]<<24))); + printf(" nomlayc=%lu\n",(unsigned long)(buffer[15]+ + (buffer[14]<<8)+(buffer[13]<<16)+(buffer[12]<<24))); + printf(" nomfram=%lu\n",(unsigned long)(buffer[19]+ + (buffer[18]<<8)+(buffer[17]<<16)+(buffer[16]<<24))); + printf(" nomplay=%lu\n",(unsigned long)(buffer[23]+ + (buffer[22]<<8)+(buffer[21]<<16)+(buffer[20]<<24))); + printf(" profile=%lu\n",(unsigned long)(buffer[27]+ + (buffer[26]<<8)+(buffer[25]<<16)+(buffer[24]<<24))); + } + + if (new_mng) { + /* write the MNG 8-byte signature */ + png_default_write_data(mng_ptr, &mng_signature[0], + (png_size_t) 8); + + /* Write a MHDR chunk */ + png_write_chunk(mng_ptr, (png_bytep) png_MHDR, + buffer, (png_size_t) 28); + } + + png_default_read_data(read_ptr, buffer, 4); input_format = 1; + } else #endif if (png_sig_cmp(png_signature, 0, 8)) { @@ -4353,10 +4458,15 @@ png_uint_32 png_measure_idat(png_structp png_ptr) } } -#ifdef PNG_CRC_WARN_USE if (fix) + { +#ifdef PNG_CRC_WARN_USE png_set_crc_action(png_ptr, PNG_CRC_WARN_USE, PNG_CRC_WARN_USE); #endif +#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR + inflateUndermine(&png_ptr->zstream, 1); +#endif + } for (;;) { #ifndef PNG_UINT_IDAT @@ -4382,6 +4492,85 @@ png_uint_32 png_measure_idat(png_structp png_ptr) png_reset_crc(png_ptr); png_crc_read(png_ptr, chunk_name, 4); + if (new_mng) { + const png_byte png_DHDR[5] = { 68, 72, 68, 82, '\0' }; + const png_byte png_DEFI[5] = { 68, 69, 70, 73, '\0' }; + const png_byte png_FRAM[5] = { 70, 82, 65, 77, '\0' }; + const png_byte png_nEED[5] = { 110, 69, 69, 68, '\0' }; + if (!png_memcmp(chunk_name, png_nEED, 4)) + { + /* Skip the nEED chunk */ + printf (" skipping MNG %c%c%c%c chunk, %lu bytes\n",chunk_name[0], + chunk_name[1],chunk_name[2],chunk_name[3],(unsigned long)length); + } + else { + /* copy the chunk. */ + printf (" reading MNG %c%c%c%c chunk, %lu bytes\n",chunk_name[0], + chunk_name[1],chunk_name[2],chunk_name[3],(unsigned long)length); + if (length > malloced_length) { + png_free(mng_ptr,bb); + printf (" png_malloc %lu bytes.\n",(unsigned long)length); + bb=png_malloc(mng_ptr, length); + malloced_length=length; + } + png_crc_read(png_ptr, bb, length); + png_write_chunk(mng_ptr, chunk_name, + bb, (png_size_t) length); + + if (!png_memcmp(chunk_name, png_DHDR, 4)) { + if (verbose > 1) { + printf(" objid=%lu\n",(unsigned long)(bb[1]+(bb[0]<<8))); + printf(" itype=%lu\n",(unsigned long)(bb[2])); + printf(" dtype=%lu\n",(unsigned long)(bb[3])); + printf(" width=%lu\n",(unsigned long)(bb[7]+(bb[6]<<8) + +(bb[5]<<16)+(bb[4]<<24))); + printf(" height=%lu\n",(unsigned long)(bb[11]+(bb[10]<<8) + +(bb[9]<<16)+(bb[8]<<24))); + printf(" xloc=%lu\n",(unsigned long)(bb[15]+(bb[14]<<8) + +(bb[13]<<16)+(bb[12]<<24))); + printf(" yloc=%lu\n",(unsigned long)(bb[19]+(bb[18]<<8) + +(bb[17]<<16)+(bb[16]<<24))); + } + } + if (!png_memcmp(chunk_name, png_DEFI, 4)) { + if (verbose > 1) { + printf(" objid=%lu\n",(unsigned long)(bb[1]+(bb[0]<<8))); + printf(" do_not_show=%lu\n",(unsigned long)(bb[2])); + printf(" concrete=%lu\n",(unsigned long)(bb[3])); + if (length > 4) { + printf(" xloc=%lu\n",(unsigned long)(bb[15]+(bb[14]<<8) + +(bb[13]<<16)+(bb[12]<<24))); + printf(" yloc=%lu\n",(unsigned long)(bb[19]+(bb[18]<<8) + +(bb[17]<<16)+(bb[16]<<24))); + if (length > 12) { + printf(" l_cb=%lu\n",(unsigned long)(bb[20]+(bb[19]<<8) + +(bb[18]<<16)+(bb[17]<<24))); + printf(" r_cb=%lu\n",(unsigned long)(bb[24]+(bb[23]<<8) + +(bb[22]<<16)+(bb[21]<<24))); + } + } + } + } + if (!png_memcmp(chunk_name, png_FRAM, 4)) { + if (verbose > 1) { + printf(" mode=%lu\n",(unsigned long)bb[0]); + if (length > 1) { + int ib; + printf(" name = "); + for (ib=0; bb[ib]; ib++) + { + printf ("%c", bb[ib]); + } + printf ("\n"); + } + } + } + length=0; + } + } + + else { + #ifdef PNG_UINT_IDAT if (png_get_uint_32(chunk_name) == PNG_UINT_IDAT) #else @@ -4395,8 +4584,8 @@ png_uint_32 png_measure_idat(png_structp png_ptr) if (verbose > 1) { chunk_name[4] = '\0'; - printf("Reading %s chunk, length = %ld.\n", chunk_name, - length); + printf("Reading %s chunk, length = %lu.\n", chunk_name, + (unsigned long)length); } #ifdef PNG_UINT_IHDR if (png_get_uint_32(chunk_name) == PNG_UINT_IHDR) @@ -4409,6 +4598,17 @@ png_uint_32 png_measure_idat(png_structp png_ptr) length -= 13; input_color_type = buffer[9]; } + +#ifdef PNG_gAMA_SUPPORTED + if (png_get_uint_32(chunk_name) == PNG_UINT_gAMA) + found_gAMA=1; +#endif + +#ifdef PNG_cHRM_SUPPORTED + if (png_get_uint_32(chunk_name) == PNG_UINT_cHRM) + found_cHRM=1; +#endif + #ifdef PNG_iCCP_SUPPORTED /* check for bad Photoshop iCCP chunk */ #ifdef PNG_UINT_iCCP @@ -4445,8 +4645,32 @@ png_uint_32 png_measure_idat(png_structp png_ptr) } #endif + + } png_crc_finish(png_ptr, length); +#ifdef PNGCRUSH_LOCO +#ifdef PNG_UINT_MEND + if (png_get_uint_32(chunk_name) == PNG_UINT_MEND) + return sum_idat_length; +#else + { + const png_byte png_MEND[5] = + { 77, 69, 78, 68, '\0' }; + if (!png_memcmp(chunk_name, png_MEND, 4)) + { + if (new_mng) + png_free(mng_ptr,bb); + return (0); + return sum_idat_length; + } + } +#endif +#endif + + +if (input_format == 0) + { #ifdef PNG_UINT_IEND if (png_get_uint_32(chunk_name) == PNG_UINT_IEND) #else @@ -4454,6 +4678,7 @@ png_uint_32 png_measure_idat(png_structp png_ptr) #endif return sum_idat_length; } + } } @@ -4556,8 +4781,15 @@ int count_colors(FILE * fpin) #if defined(PNGCRUSH_LOCO) if (!(int) (png_memcmp(mng_signature, png_signature, 8))) { png_byte buffer[40]; + unsigned long length; /* skip the MHDR */ - png_default_read_data(read_ptr, buffer, 40); + png_default_read_data(read_ptr, buffer, 4); + length=buffer[3]+(buffer[2]<<8)+(buffer[1]<<16)+(buffer[0]<<24); + png_default_read_data(read_ptr, buffer, 4); + printf("Skipping %c%c%c%c chunk.\n",buffer[0],buffer[1], + buffer[2],buffer[3]); + png_default_read_data(read_ptr, buffer, length); + png_default_read_data(read_ptr, buffer, 4); png_permit_mng_features(read_ptr, PNG_FLAG_MNG_FILTER_64); input_format = 1; @@ -4963,7 +5195,7 @@ void print_version_info(void) " | pngcrush %s\n" /* If you have modified this source, you may insert additional notices * immediately after this sentence: */ - " | Copyright (C) 1998-2002,2006 Glenn Randers-Pehrson\n" + " | Copyright (C) 1998-2002,2006-2008 Glenn Randers-Pehrson\n" " | Copyright (C) 2005 Greg Roelofs\n" " | This is a free, open-source program. Permission is irrevocably\n" " | granted to everyone to use this version of pngcrush without\n" @@ -4971,7 +5203,7 @@ void print_version_info(void) " | Executable name is %s\n" " | It was built with libpng version %s, and is\n" " | running with %s" - " | Copyright (C) 1998-2004,2006 Glenn Randers-Pehrson,\n" + " | Copyright (C) 1998-2004,2006-2008 Glenn Randers-Pehrson,\n" " | Copyright (C) 1996, 1997 Andreas Dilger,\n" " | Copyright (C) 1995, Guy Eric Schalnat, Group 42 Inc.,\n" " | and zlib version %s, Copyright (C) 1998-2002 (or later),\n" @@ -5009,7 +5241,7 @@ static const char *pngcrush_legal[] = { "", /* If you have modified this source, you may insert additional notices * immediately after this sentence: */ - "Copyright (C) 1998-2002,2006 Glenn Randers-Pehrson (glennrp@users.sf.net)", + "Copyright (C) 1998-2002,2006-2008 Glenn Randers-Pehrson (glennrp@users.sf.net)", "Copyright (C) 2005 Greg Roelofs", "", "DISCLAIMER: The pngcrush computer program is supplied \"AS IS\".", @@ -5197,6 +5429,12 @@ struct options_help pngcrush_options[] = { {0, " -max maximum_IDAT_size [default "STRNGIFY(PNG_ZBUF_SIZE)"]"}, {2, ""}, +#ifdef PNGCRUSH_LOCO + {0, " -mng (write a new MNG, do not crush embedded PNGs)"}, + {2, ""}, +#endif + + #ifdef PNGCRUSH_COUNT_COLORS {0, " -no_cc (no color counting)"}, {2, ""}, diff --git a/pngcrush.h b/pngcrush.h index 92050af7e..f0c961673 100644 --- a/pngcrush.h +++ b/pngcrush.h @@ -18,10 +18,7 @@ #define PNGCRUSH_LIBPNG_VER 10007 #endif -#if !defined(PNG_USE_PNGGCCRD) && !defined(PNG_USE_PNGVCRD) && \ - !defined(PNG_NO_ASSEMBLER_CODE) #define PNG_NO_ASSEMBLER_CODE -#endif #ifndef PNG_NO_ZALLOC_ZERO # define PNG_NO_ZALLOC_ZERO /* speeds it up a little */ diff --git a/pngerror.c b/pngerror.c index ad6ae0e82..b364fc00a 100644 --- a/pngerror.c +++ b/pngerror.c @@ -1,9 +1,9 @@ /* pngerror.c - stub functions for i/o and memory allocation * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.22 [October 13, 2007] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2007 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -20,15 +20,18 @@ static void /* PRIVATE */ png_default_error PNGARG((png_structp png_ptr, png_const_charp error_message)); +#ifndef PNG_NO_WARNINGS static void /* PRIVATE */ png_default_warning PNGARG((png_structp png_ptr, png_const_charp warning_message)); +#endif /* PNG_NO_WARNINGS */ /* This function is called whenever there is a fatal error. This function * should not be changed. If there is a need to handle errors differently, * you should supply a replacement error function and use png_set_error_fn() * to replace the error function at run-time. */ +#ifndef PNG_NO_ERROR_TEXT void PNGAPI png_error(png_structp png_ptr, png_const_charp error_message) { @@ -75,7 +78,20 @@ png_error(png_structp png_ptr, png_const_charp error_message) use the default handler, which will not return. */ png_default_error(png_ptr, error_message); } +#else +void PNGAPI +png_err(png_structp png_ptr) +{ + if (png_ptr != NULL && png_ptr->error_fn != NULL) + (*(png_ptr->error_fn))(png_ptr, '\0'); + /* If the custom handler doesn't exist, or if it returns, + use the default handler, which will not return. */ + png_default_error(png_ptr, '\0'); +} +#endif /* PNG_NO_ERROR_TEXT */ + +#ifndef PNG_NO_WARNINGS /* This function is called whenever there is a non-fatal error. This function * should not be changed. If there is a need to handle warnings differently, * you should supply a replacement warning function and use @@ -105,6 +121,8 @@ png_warning(png_structp png_ptr, png_const_charp warning_message) else png_default_warning(png_ptr, warning_message+offset); } +#endif /* PNG_NO_WARNINGS */ + /* These utilities are used internally to build an error message that relates * to the current chunk. The chunk name comes from png_ptr->chunk_name, @@ -118,6 +136,9 @@ static PNG_CONST char png_digit[16] = { 'A', 'B', 'C', 'D', 'E', 'F' }; +#define PNG_MAX_ERROR_TEXT 64 + +#if !defined(PNG_NO_WARNINGS) || !defined(PNG_NO_ERROR_TEXT) static void /* PRIVATE */ png_format_buffer(png_structp png_ptr, png_charp buffer, png_const_charp error_message) @@ -141,35 +162,47 @@ png_format_buffer(png_structp png_ptr, png_charp buffer, png_const_charp } if (error_message == NULL) - buffer[iout] = 0; + buffer[iout] = '\0'; else { buffer[iout++] = ':'; buffer[iout++] = ' '; - png_strncpy(buffer+iout, error_message, 63); - buffer[iout+63] = 0; + png_memcpy(buffer+iout, error_message, PNG_MAX_ERROR_TEXT); + buffer[iout+PNG_MAX_ERROR_TEXT-1] = '\0'; } } +#ifdef PNG_READ_SUPPORTED void PNGAPI png_chunk_error(png_structp png_ptr, png_const_charp error_message) { - char msg[18+64]; + char msg[18+PNG_MAX_ERROR_TEXT]; if (png_ptr == NULL) png_error(png_ptr, error_message); - png_format_buffer(png_ptr, msg, error_message); - png_error(png_ptr, msg); + else + { + png_format_buffer(png_ptr, msg, error_message); + png_error(png_ptr, msg); + } } +#endif /* PNG_READ_SUPPORTED */ +#endif /* !defined(PNG_NO_WARNINGS) || !defined(PNG_NO_ERROR_TEXT) */ +#ifndef PNG_NO_WARNINGS void PNGAPI png_chunk_warning(png_structp png_ptr, png_const_charp warning_message) { - char msg[18+64]; + char msg[18+PNG_MAX_ERROR_TEXT]; if (png_ptr == NULL) png_warning(png_ptr, warning_message); - png_format_buffer(png_ptr, msg, warning_message); - png_warning(png_ptr, msg); + else + { + png_format_buffer(png_ptr, msg, warning_message); + png_warning(png_ptr, msg); + } } +#endif /* PNG_NO_WARNINGS */ + /* This is the default error handling function. Note that replacements for * this function MUST NOT RETURN, or the program will likely crash. This @@ -206,27 +239,27 @@ png_default_error(png_structp png_ptr, png_const_charp error_message) #endif #ifdef PNG_SETJMP_SUPPORTED + if (png_ptr) + { # ifdef USE_FAR_KEYWORD { jmp_buf jmpbuf; - png_memcpy(jmpbuf,png_ptr->jmpbuf,png_sizeof(jmp_buf)); + png_memcpy(jmpbuf, png_ptr->jmpbuf, png_sizeof(jmp_buf)); longjmp(jmpbuf, 1); } # else longjmp(png_ptr->jmpbuf, 1); -# endif +# endif + } #else - /* make compiler happy */ ; - if (png_ptr) PNG_ABORT(); #endif #ifdef PNG_NO_CONSOLE_IO - /* make compiler happy */ ; - if (&error_message != NULL) - return; + error_message = error_message; /* make compiler happy */ #endif } +#ifndef PNG_NO_WARNINGS /* This function is called when there is a warning, but the library thinks * it can continue anyway. Replacement functions don't have to do anything * here if you don't want them to. In the default configuration, png_ptr is @@ -260,14 +293,11 @@ png_default_warning(png_structp png_ptr, png_const_charp warning_message) # endif fprintf(stderr, "libpng warning: %s\n", warning_message); #else - /* make compiler happy */ ; - if (warning_message) - return; + warning_message = warning_message; /* make compiler happy */ #endif - /* make compiler happy */ ; - if (png_ptr) - return; + png_ptr = png_ptr; /* make compiler happy */ } +#endif /* PNG_NO_WARNINGS */ /* This function is called when the application wants to use another method * of handling errors and warnings. Note that the error function MUST NOT diff --git a/pnggccrd.c b/pnggccrd.c index 8dc9c537e..e61523e77 100644 --- a/pnggccrd.c +++ b/pnggccrd.c @@ -1,5374 +1,53 @@ +/* pnggccrd.c was removed from libpng-1.2.20. */ -/* pnggccrd.c - mixed C/assembler version of utilities to read a PNG file - * - * For Intel x86 CPU (Pentium-MMX or later) and GNU C compiler. - * - * See http://www.intel.com/drg/pentiumII/appnotes/916/916.htm - * and http://www.intel.com/drg/pentiumII/appnotes/923/923.htm - * for Intel's performance analysis of the MMX vs. non-MMX code. - * - * Last changed in libpng 1.2.9 April 14, 2006 - * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson - * Copyright (c) 1998, Intel Corporation - * - * Based on MSVC code contributed by Nirav Chhatrapati, Intel Corp., 1998. - * Interface to libpng contributed by Gilles Vollant, 1999. - * GNU C port by Greg Roelofs, 1999-2001. - * - * Lines 2350-4300 converted in place with intel2gas 1.3.1: - * - * intel2gas -mdI pnggccrd.c.partially-msvc -o pnggccrd.c - * - * and then cleaned up by hand. See http://hermes.terminal.at/intel2gas/ . - * - * NOTE: A sufficiently recent version of GNU as (or as.exe under DOS/Windows) - * is required to assemble the newer MMX instructions such as movq. - * For djgpp, see - * - * ftp://ftp.simtel.net/pub/simtelnet/gnu/djgpp/v2gnu/bnu281b.zip - * - * (or a later version in the same directory). For Linux, check your - * distribution's web site(s) or try these links: - * - * http://rufus.w3.org/linux/RPM/binutils.html - * http://www.debian.org/Packages/stable/devel/binutils.html - * ftp://ftp.slackware.com/pub/linux/slackware/slackware/slakware/d1/ - * binutils.tgz - * - * For other platforms, see the main GNU site: - * - * ftp://ftp.gnu.org/pub/gnu/binutils/ - * - * Version 2.5.2l.15 is definitely too old... - */ +/* This code snippet is for use by configure's compilation test. */ -/* - * TEMPORARY PORTING NOTES AND CHANGELOG (mostly by Greg Roelofs) - * ===================================== - * - * 19991006: - * - fixed sign error in post-MMX cleanup code (16- & 32-bit cases) - * - * 19991007: - * - additional optimizations (possible or definite): - * x [DONE] write MMX code for 64-bit case (pixel_bytes == 8) [not tested] - * - write MMX code for 48-bit case (pixel_bytes == 6) - * - figure out what's up with 24-bit case (pixel_bytes == 3): - * why subtract 8 from width_mmx in the pass 4/5 case? - * (only width_mmx case) (near line 1606) - * x [DONE] replace pixel_bytes within each block with the true - * constant value (or are compilers smart enough to do that?) - * - rewrite all MMX interlacing code so it's aligned with - * the *beginning* of the row buffer, not the end. This - * would not only allow one to eliminate half of the memory - * writes for odd passes (that is, pass == odd), it may also - * eliminate some unaligned-data-access exceptions (assuming - * there's a penalty for not aligning 64-bit accesses on - * 64-bit boundaries). The only catch is that the "leftover" - * pixel(s) at the end of the row would have to be saved, - * but there are enough unused MMX registers in every case, - * so this is not a problem. A further benefit is that the - * post-MMX cleanup code (C code) in at least some of the - * cases could be done within the assembler block. - * x [DONE] the "v3 v2 v1 v0 v7 v6 v5 v4" comments are confusing, - * inconsistent, and don't match the MMX Programmer's Reference - * Manual conventions anyway. They should be changed to - * "b7 b6 b5 b4 b3 b2 b1 b0," where b0 indicates the byte that - * was lowest in memory (e.g., corresponding to a left pixel) - * and b7 is the byte that was highest (e.g., a right pixel). - * - * 19991016: - * - Brennan's Guide notwithstanding, gcc under Linux does *not* - * want globals prefixed by underscores when referencing them-- - * i.e., if the variable is const4, then refer to it as const4, - * not _const4. This seems to be a djgpp-specific requirement. - * Also, such variables apparently *must* be declared outside - * of functions; neither static nor automatic variables work if - * defined within the scope of a single function, but both - * static and truly global (multi-module) variables work fine. - * - * 19991023: - * - fixed png_combine_row() non-MMX replication bug (odd passes only?) - * - switched from string-concatenation-with-macros to cleaner method of - * renaming global variables for djgpp--i.e., always use prefixes in - * inlined assembler code (== strings) and conditionally rename the - * variables, not the other way around. Hence _const4, _mask8_0, etc. - * - * 19991024: - * - fixed mmxsupport()/png_do_read_interlace() first-row bug - * This one was severely weird: even though mmxsupport() doesn't touch - * ebx (where "row" pointer was stored), it nevertheless managed to zero - * the register (even in static/non-fPIC code--see below), which in turn - * caused png_do_read_interlace() to return prematurely on the first row of - * interlaced images (i.e., without expanding the interlaced pixels). - * Inspection of the generated assembly code didn't turn up any clues, - * although it did point at a minor optimization (i.e., get rid of - * mmx_supported_local variable and just use eax). Possibly the CPUID - * instruction is more destructive than it looks? (Not yet checked.) - * - "info gcc" was next to useless, so compared fPIC and non-fPIC assembly - * listings... Apparently register spillage has to do with ebx, since - * it's used to index the global offset table. Commenting it out of the - * input-reg lists in png_combine_row() eliminated compiler barfage, so - * ifdef'd with __PIC__ macro: if defined, use a global for unmask - * - * 19991107: - * - verified CPUID clobberage: 12-char string constant ("GenuineIntel", - * "AuthenticAMD", etc.) placed in ebx:ecx:edx. Still need to polish. - * - * 19991120: - * - made "diff" variable (now "_dif") global to simplify conversion of - * filtering routines (running out of regs, sigh). "diff" is still used - * in interlacing routines, however. - * - fixed up both versions of mmxsupport() (ORIG_THAT_USED_TO_CLOBBER_EBX - * macro determines which is used); original not yet tested. - * - * 20000213: - * - when compiling with gcc, be sure to use -fomit-frame-pointer - * - * 20000319: - * - fixed a register-name typo in png_do_read_interlace(), default (MMX) case, - * pass == 4 or 5, that caused visible corruption of interlaced images - * - * 20000623: - * - Various problems were reported with gcc 2.95.2 in the Cygwin environment, - * many of the form "forbidden register 0 (ax) was spilled for class AREG." - * This is explained at http://gcc.gnu.org/fom_serv/cache/23.html, and - * Chuck Wilson supplied a patch involving dummy output registers. See - * http://sourceforge.net/bugs/?func=detailbug&bug_id=108741&group_id=5624 - * for the original (anonymous) SourceForge bug report. - * - * 20000706: - * - Chuck Wilson passed along these remaining gcc 2.95.2 errors: - * pnggccrd.c: In function `png_combine_row': - * pnggccrd.c:525: more than 10 operands in `asm' - * pnggccrd.c:669: more than 10 operands in `asm' - * pnggccrd.c:828: more than 10 operands in `asm' - * pnggccrd.c:994: more than 10 operands in `asm' - * pnggccrd.c:1177: more than 10 operands in `asm' - * They are all the same problem and can be worked around by using the - * global _unmask variable unconditionally, not just in the -fPIC case. - * Reportedly earlier versions of gcc also have the problem with more than - * 10 operands; they just don't report it. Much strangeness ensues, etc. - * - * 20000729: - * - enabled png_read_filter_row_mmx_up() (shortest remaining unconverted - * MMX routine); began converting png_read_filter_row_mmx_sub() - * - to finish remaining sections: - * - clean up indentation and comments - * - preload local variables - * - add output and input regs (order of former determines numerical - * mapping of latter) - * - avoid all usage of ebx (including bx, bh, bl) register [20000823] - * - remove "$" from addressing of Shift and Mask variables [20000823] - * - * 20000731: - * - global union vars causing segfaults in png_read_filter_row_mmx_sub()? - * - * 20000822: - * - ARGH, stupid png_read_filter_row_mmx_sub() segfault only happens with - * shared-library (-fPIC) version! Code works just fine as part of static - * library. Damn damn damn damn damn, should have tested that sooner. - * ebx is getting clobbered again (explicitly this time); need to save it - * on stack or rewrite asm code to avoid using it altogether. Blargh! - * - * 20000823: - * - first section was trickiest; all remaining sections have ebx -> edx now. - * (-fPIC works again.) Also added missing underscores to various Shift* - * and *Mask* globals and got rid of leading "$" signs. - * - * 20000826: - * - added visual separators to help navigate microscopic printed copies - * (http://pobox.com/~newt/code/gpr-latest.zip, mode 10); started working - * on png_read_filter_row_mmx_avg() - * - * 20000828: - * - finished png_read_filter_row_mmx_avg(): only Paeth left! (930 lines...) - * What the hell, did png_read_filter_row_mmx_paeth(), too. Comments not - * cleaned up/shortened in either routine, but functionality is complete - * and seems to be working fine. - * - * 20000829: - * - ahhh, figured out last(?) bit of gcc/gas asm-fu: if register is listed - * as an input reg (with dummy output variables, etc.), then it *cannot* - * also appear in the clobber list or gcc 2.95.2 will barf. The solution - * is simple enough... - * - * 20000914: - * - bug in png_read_filter_row_mmx_avg(): 16-bit grayscale not handled - * correctly (but 48-bit RGB just fine) - * - * 20000916: - * - fixed bug in png_read_filter_row_mmx_avg(), bpp == 2 case; three errors: - * - "_ShiftBpp.use = 24;" should have been "_ShiftBpp.use = 16;" - * - "_ShiftRem.use = 40;" should have been "_ShiftRem.use = 48;" - * - "psllq _ShiftRem, %%mm2" should have been "psrlq _ShiftRem, %%mm2" - * - * 20010101: - * - added new png_init_mmx_flags() function (here only because it needs to - * call mmxsupport(), which should probably become global png_mmxsupport()); - * modified other MMX routines to run conditionally (png_ptr->asm_flags) - * - * 20010103: - * - renamed mmxsupport() to png_mmx_support(), with auto-set of mmx_supported, - * and made it public; moved png_init_mmx_flags() to png.c as internal func - * - * 20010104: - * - removed dependency on png_read_filter_row_c() (C code already duplicated - * within MMX version of png_read_filter_row()) so no longer necessary to - * compile it into pngrutil.o - * - * 20010310: - * - fixed buffer-overrun bug in png_combine_row() C code (non-MMX) - * - * 20020304: - * - eliminated incorrect use of width_mmx in pixel_bytes == 8 case - * - * 20040724: - * - more tinkering with clobber list at lines 4529 and 5033, to get - * it to compile on gcc-3.4. - * - * STILL TO DO: - * - test png_do_read_interlace() 64-bit case (pixel_bytes == 8) - * - write MMX code for 48-bit case (pixel_bytes == 6) - * - figure out what's up with 24-bit case (pixel_bytes == 3): - * why subtract 8 from width_mmx in the pass 4/5 case? - * (only width_mmx case) (near line 1606) - * - rewrite all MMX interlacing code so it's aligned with beginning - * of the row buffer, not the end (see 19991007 for details) - * x pick one version of mmxsupport() and get rid of the other - * - add error messages to any remaining bogus default cases - * - enable pixel_depth == 8 cases in png_read_filter_row()? (test speed) - * x add support for runtime enable/disable/query of various MMX routines - */ +#if (!defined _MSC_VER) && \ + defined(PNG_ASSEMBLER_CODE_SUPPORTED) && \ + defined(PNG_MMX_CODE_SUPPORTED) -#define PNG_INTERNAL -#include "png.h" +int PNGAPI png_dummy_mmx_support(void); -#if defined(PNG_USE_PNGGCCRD) - -int PNGAPI png_mmx_support(void); - -#ifdef PNG_USE_LOCAL_ARRAYS -static const int FARDATA png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; -static const int FARDATA png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; -static const int FARDATA png_pass_width[7] = {8, 4, 4, 2, 2, 1, 1}; -#endif - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) -/* djgpp, Win32, Cygwin, and OS2 add their own underscores to global variables, - * so define them without: */ -#if defined(__DJGPP__) || defined(WIN32) || defined(__CYGWIN__) || \ - defined(__OS2__) -# define _mmx_supported mmx_supported -# define _const4 const4 -# define _const6 const6 -# define _mask8_0 mask8_0 -# define _mask16_1 mask16_1 -# define _mask16_0 mask16_0 -# define _mask24_2 mask24_2 -# define _mask24_1 mask24_1 -# define _mask24_0 mask24_0 -# define _mask32_3 mask32_3 -# define _mask32_2 mask32_2 -# define _mask32_1 mask32_1 -# define _mask32_0 mask32_0 -# define _mask48_5 mask48_5 -# define _mask48_4 mask48_4 -# define _mask48_3 mask48_3 -# define _mask48_2 mask48_2 -# define _mask48_1 mask48_1 -# define _mask48_0 mask48_0 -# define _LBCarryMask LBCarryMask -# define _HBClearMask HBClearMask -# define _ActiveMask ActiveMask -# define _ActiveMask2 ActiveMask2 -# define _ActiveMaskEnd ActiveMaskEnd -# define _ShiftBpp ShiftBpp -# define _ShiftRem ShiftRem -#ifdef PNG_THREAD_UNSAFE_OK -# define _unmask unmask -# define _FullLength FullLength -# define _MMXLength MMXLength -# define _dif dif -# define _patemp patemp -# define _pbtemp pbtemp -# define _pctemp pctemp -#endif -#endif - - -/* These constants are used in the inlined MMX assembly code. - Ignore gcc's "At top level: defined but not used" warnings. */ - -/* GRR 20000706: originally _unmask was needed only when compiling with -fPIC, - * since that case uses the %ebx register for indexing the Global Offset Table - * and there were no other registers available. But gcc 2.95 and later emit - * "more than 10 operands in `asm'" errors when %ebx is used to preload unmask - * in the non-PIC case, so we'll just use the global unconditionally now. - */ -#ifdef PNG_THREAD_UNSAFE_OK -static int _unmask; -#endif - -static unsigned long long _mask8_0 = 0x0102040810204080LL; - -static unsigned long long _mask16_1 = 0x0101020204040808LL; -static unsigned long long _mask16_0 = 0x1010202040408080LL; - -static unsigned long long _mask24_2 = 0x0101010202020404LL; -static unsigned long long _mask24_1 = 0x0408080810101020LL; -static unsigned long long _mask24_0 = 0x2020404040808080LL; - -static unsigned long long _mask32_3 = 0x0101010102020202LL; -static unsigned long long _mask32_2 = 0x0404040408080808LL; -static unsigned long long _mask32_1 = 0x1010101020202020LL; -static unsigned long long _mask32_0 = 0x4040404080808080LL; - -static unsigned long long _mask48_5 = 0x0101010101010202LL; -static unsigned long long _mask48_4 = 0x0202020204040404LL; -static unsigned long long _mask48_3 = 0x0404080808080808LL; -static unsigned long long _mask48_2 = 0x1010101010102020LL; -static unsigned long long _mask48_1 = 0x2020202040404040LL; -static unsigned long long _mask48_0 = 0x4040808080808080LL; - -static unsigned long long _const4 = 0x0000000000FFFFFFLL; -//static unsigned long long _const5 = 0x000000FFFFFF0000LL; // NOT USED -static unsigned long long _const6 = 0x00000000000000FFLL; - -// These are used in the row-filter routines and should/would be local -// variables if not for gcc addressing limitations. -// WARNING: Their presence probably defeats the thread safety of libpng. - -#ifdef PNG_THREAD_UNSAFE_OK -static png_uint_32 _FullLength; -static png_uint_32 _MMXLength; -static int _dif; -static int _patemp; // temp variables for Paeth routine -static int _pbtemp; -static int _pctemp; -#endif - -void /* PRIVATE */ -png_squelch_warnings(void) -{ -#ifdef PNG_THREAD_UNSAFE_OK - _dif = _dif; - _patemp = _patemp; - _pbtemp = _pbtemp; - _pctemp = _pctemp; - _MMXLength = _MMXLength; -#endif - _const4 = _const4; - _const6 = _const6; - _mask8_0 = _mask8_0; - _mask16_1 = _mask16_1; - _mask16_0 = _mask16_0; - _mask24_2 = _mask24_2; - _mask24_1 = _mask24_1; - _mask24_0 = _mask24_0; - _mask32_3 = _mask32_3; - _mask32_2 = _mask32_2; - _mask32_1 = _mask32_1; - _mask32_0 = _mask32_0; - _mask48_5 = _mask48_5; - _mask48_4 = _mask48_4; - _mask48_3 = _mask48_3; - _mask48_2 = _mask48_2; - _mask48_1 = _mask48_1; - _mask48_0 = _mask48_0; -} -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - - -static int _mmx_supported = 2; - -/*===========================================================================*/ -/* */ -/* P N G _ C O M B I N E _ R O W */ -/* */ -/*===========================================================================*/ - -#if defined(PNG_HAVE_ASSEMBLER_COMBINE_ROW) - -#define BPP2 2 -#define BPP3 3 /* bytes per pixel (a.k.a. pixel_bytes) */ -#define BPP4 4 -#define BPP6 6 /* (defined only to help avoid cut-and-paste errors) */ -#define BPP8 8 - -/* Combines the row recently read in with the previous row. - This routine takes care of alpha and transparency if requested. - This routine also handles the two methods of progressive display - of interlaced images, depending on the mask value. - The mask value describes which pixels are to be combined with - the row. The pattern always repeats every 8 pixels, so just 8 - bits are needed. A one indicates the pixel is to be combined; a - zero indicates the pixel is to be skipped. This is in addition - to any alpha or transparency value associated with the pixel. - If you want all pixels to be combined, pass 0xff (255) in mask. */ - -/* Use this routine for the x86 platform - it uses a faster MMX routine - if the machine supports MMX. */ - -void /* PRIVATE */ -png_combine_row(png_structp png_ptr, png_bytep row, int mask) -{ - png_debug(1, "in png_combine_row (pnggccrd.c)\n"); - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) - if (_mmx_supported == 2) { -#if !defined(PNG_1_0_X) - /* this should have happened in png_init_mmx_flags() already */ - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } -#endif - - if (mask == 0xff) - { - png_debug(2,"mask == 0xff: doing single png_memcpy()\n"); - png_memcpy(row, png_ptr->row_buf + 1, - (png_size_t)PNG_ROWBYTES(png_ptr->row_info.pixel_depth,png_ptr->width)); - } - else /* (png_combine_row() is never called with mask == 0) */ - { - switch (png_ptr->row_info.pixel_depth) - { - case 1: /* png_ptr->row_info.pixel_depth */ - { - png_bytep sp; - png_bytep dp; - int s_inc, s_start, s_end; - int m; - int shift; - png_uint_32 i; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 7; - s_inc = 1; - } - else -#endif - { - s_start = 7; - s_end = 0; - s_inc = -1; - } - - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - int value; - - value = (*sp >> shift) & 0x1; - *dp &= (png_byte)((0x7f7f >> (7 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 2: /* png_ptr->row_info.pixel_depth */ - { - png_bytep sp; - png_bytep dp; - int s_start, s_end, s_inc; - int m; - int shift; - png_uint_32 i; - int value; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 6; - s_inc = 2; - } - else -#endif - { - s_start = 6; - s_end = 0; - s_inc = -2; - } - - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - value = (*sp >> shift) & 0x3; - *dp &= (png_byte)((0x3f3f >> (6 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 4: /* png_ptr->row_info.pixel_depth */ - { - png_bytep sp; - png_bytep dp; - int s_start, s_end, s_inc; - int m; - int shift; - png_uint_32 i; - int value; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 4; - s_inc = 4; - } - else -#endif - { - s_start = 4; - s_end = 0; - s_inc = -4; - } - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - value = (*sp >> shift) & 0xf; - *dp &= (png_byte)((0xf0f >> (4 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 8: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - png_uint_32 len; - int diff; - int dummy_value_a; // fix 'forbidden register spilled' error - int dummy_value_d; - int dummy_value_c; - int dummy_value_S; - int dummy_value_D; - _unmask = ~mask; // global variable for -fPIC version - srcptr = png_ptr->row_buf + 1; - dstptr = row; - len = png_ptr->width &~7; // reduce to multiple of 8 - diff = (int) (png_ptr->width & 7); // amount lost - - __asm__ __volatile__ ( - "movd _unmask, %%mm7 \n\t" // load bit pattern - "psubb %%mm6, %%mm6 \n\t" // zero mm6 - "punpcklbw %%mm7, %%mm7 \n\t" - "punpcklwd %%mm7, %%mm7 \n\t" - "punpckldq %%mm7, %%mm7 \n\t" // fill reg with 8 masks - - "movq _mask8_0, %%mm0 \n\t" - "pand %%mm7, %%mm0 \n\t" // nonzero if keep byte - "pcmpeqb %%mm6, %%mm0 \n\t" // zeros->1s, v versa - -// preload "movl len, %%ecx \n\t" // load length of line -// preload "movl srcptr, %%esi \n\t" // load source -// preload "movl dstptr, %%edi \n\t" // load dest - - "cmpl $0, %%ecx \n\t" // len == 0 ? - "je mainloop8end \n\t" - - "mainloop8: \n\t" - "movq (%%esi), %%mm4 \n\t" // *srcptr - "pand %%mm0, %%mm4 \n\t" - "movq %%mm0, %%mm6 \n\t" - "pandn (%%edi), %%mm6 \n\t" // *dstptr - "por %%mm6, %%mm4 \n\t" - "movq %%mm4, (%%edi) \n\t" - "addl $8, %%esi \n\t" // inc by 8 bytes processed - "addl $8, %%edi \n\t" - "subl $8, %%ecx \n\t" // dec by 8 pixels processed - "ja mainloop8 \n\t" - - "mainloop8end: \n\t" -// preload "movl diff, %%ecx \n\t" // (diff is in eax) - "movl %%eax, %%ecx \n\t" - "cmpl $0, %%ecx \n\t" - "jz end8 \n\t" -// preload "movl mask, %%edx \n\t" - "sall $24, %%edx \n\t" // make low byte, high byte - - "secondloop8: \n\t" - "sall %%edx \n\t" // move high bit to CF - "jnc skip8 \n\t" // if CF = 0 - "movb (%%esi), %%al \n\t" - "movb %%al, (%%edi) \n\t" - - "skip8: \n\t" - "incl %%esi \n\t" - "incl %%edi \n\t" - "decl %%ecx \n\t" - "jnz secondloop8 \n\t" - - "end8: \n\t" - "EMMS \n\t" // DONE - - : "=a" (dummy_value_a), // output regs (dummy) - "=d" (dummy_value_d), - "=c" (dummy_value_c), - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "3" (srcptr), // esi // input regs - "4" (dstptr), // edi - "0" (diff), // eax -// was (unmask) "b" RESERVED // ebx // Global Offset Table idx - "2" (len), // ecx - "1" (mask) // edx - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm4", "%mm6", "%mm7" // clobber list -#endif - ); - } - else /* mmx _not supported - Use modified C routine */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - register png_uint_32 i; - png_uint_32 initial_val = png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff /* *BPP1 */ ; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - - } /* end of else (_mmx_supported) */ - - break; - } /* end 8 bpp */ - - case 16: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - png_uint_32 len; - int diff; - int dummy_value_a; // fix 'forbidden register spilled' error - int dummy_value_d; - int dummy_value_c; - int dummy_value_S; - int dummy_value_D; - _unmask = ~mask; // global variable for -fPIC version - srcptr = png_ptr->row_buf + 1; - dstptr = row; - len = png_ptr->width &~7; // reduce to multiple of 8 - diff = (int) (png_ptr->width & 7); // amount lost // - - __asm__ __volatile__ ( - "movd _unmask, %%mm7 \n\t" // load bit pattern - "psubb %%mm6, %%mm6 \n\t" // zero mm6 - "punpcklbw %%mm7, %%mm7 \n\t" - "punpcklwd %%mm7, %%mm7 \n\t" - "punpckldq %%mm7, %%mm7 \n\t" // fill reg with 8 masks - - "movq _mask16_0, %%mm0 \n\t" - "movq _mask16_1, %%mm1 \n\t" - - "pand %%mm7, %%mm0 \n\t" - "pand %%mm7, %%mm1 \n\t" - - "pcmpeqb %%mm6, %%mm0 \n\t" - "pcmpeqb %%mm6, %%mm1 \n\t" - -// preload "movl len, %%ecx \n\t" // load length of line -// preload "movl srcptr, %%esi \n\t" // load source -// preload "movl dstptr, %%edi \n\t" // load dest - - "cmpl $0, %%ecx \n\t" - "jz mainloop16end \n\t" - - "mainloop16: \n\t" - "movq (%%esi), %%mm4 \n\t" - "pand %%mm0, %%mm4 \n\t" - "movq %%mm0, %%mm6 \n\t" - "movq (%%edi), %%mm7 \n\t" - "pandn %%mm7, %%mm6 \n\t" - "por %%mm6, %%mm4 \n\t" - "movq %%mm4, (%%edi) \n\t" - - "movq 8(%%esi), %%mm5 \n\t" - "pand %%mm1, %%mm5 \n\t" - "movq %%mm1, %%mm7 \n\t" - "movq 8(%%edi), %%mm6 \n\t" - "pandn %%mm6, %%mm7 \n\t" - "por %%mm7, %%mm5 \n\t" - "movq %%mm5, 8(%%edi) \n\t" - - "addl $16, %%esi \n\t" // inc by 16 bytes processed - "addl $16, %%edi \n\t" - "subl $8, %%ecx \n\t" // dec by 8 pixels processed - "ja mainloop16 \n\t" - - "mainloop16end: \n\t" -// preload "movl diff, %%ecx \n\t" // (diff is in eax) - "movl %%eax, %%ecx \n\t" - "cmpl $0, %%ecx \n\t" - "jz end16 \n\t" -// preload "movl mask, %%edx \n\t" - "sall $24, %%edx \n\t" // make low byte, high byte - - "secondloop16: \n\t" - "sall %%edx \n\t" // move high bit to CF - "jnc skip16 \n\t" // if CF = 0 - "movw (%%esi), %%ax \n\t" - "movw %%ax, (%%edi) \n\t" - - "skip16: \n\t" - "addl $2, %%esi \n\t" - "addl $2, %%edi \n\t" - "decl %%ecx \n\t" - "jnz secondloop16 \n\t" - - "end16: \n\t" - "EMMS \n\t" // DONE - - : "=a" (dummy_value_a), // output regs (dummy) - "=c" (dummy_value_c), - "=d" (dummy_value_d), - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (diff), // eax // input regs -// was (unmask) " " RESERVED // ebx // Global Offset Table idx - "1" (len), // ecx - "2" (mask), // edx - "3" (srcptr), // esi - "4" (dstptr) // edi - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm4" // clobber list - , "%mm5", "%mm6", "%mm7" -#endif - ); - } - else /* mmx _not supported - Use modified C routine */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - register png_uint_32 i; - png_uint_32 initial_val = BPP2 * png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = BPP2 * png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = BPP2 * png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = BPP2 * len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff*BPP2; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - } /* end of else (_mmx_supported) */ - - break; - } /* end 16 bpp */ - - case 24: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - png_uint_32 len; - int diff; - int dummy_value_a; // fix 'forbidden register spilled' error - int dummy_value_d; - int dummy_value_c; - int dummy_value_S; - int dummy_value_D; - _unmask = ~mask; // global variable for -fPIC version - srcptr = png_ptr->row_buf + 1; - dstptr = row; - len = png_ptr->width &~7; // reduce to multiple of 8 - diff = (int) (png_ptr->width & 7); // amount lost // - - __asm__ __volatile__ ( - "movd _unmask, %%mm7 \n\t" // load bit pattern - "psubb %%mm6, %%mm6 \n\t" // zero mm6 - "punpcklbw %%mm7, %%mm7 \n\t" - "punpcklwd %%mm7, %%mm7 \n\t" - "punpckldq %%mm7, %%mm7 \n\t" // fill reg with 8 masks - - "movq _mask24_0, %%mm0 \n\t" - "movq _mask24_1, %%mm1 \n\t" - "movq _mask24_2, %%mm2 \n\t" - - "pand %%mm7, %%mm0 \n\t" - "pand %%mm7, %%mm1 \n\t" - "pand %%mm7, %%mm2 \n\t" - - "pcmpeqb %%mm6, %%mm0 \n\t" - "pcmpeqb %%mm6, %%mm1 \n\t" - "pcmpeqb %%mm6, %%mm2 \n\t" - -// preload "movl len, %%ecx \n\t" // load length of line -// preload "movl srcptr, %%esi \n\t" // load source -// preload "movl dstptr, %%edi \n\t" // load dest - - "cmpl $0, %%ecx \n\t" - "jz mainloop24end \n\t" - - "mainloop24: \n\t" - "movq (%%esi), %%mm4 \n\t" - "pand %%mm0, %%mm4 \n\t" - "movq %%mm0, %%mm6 \n\t" - "movq (%%edi), %%mm7 \n\t" - "pandn %%mm7, %%mm6 \n\t" - "por %%mm6, %%mm4 \n\t" - "movq %%mm4, (%%edi) \n\t" - - "movq 8(%%esi), %%mm5 \n\t" - "pand %%mm1, %%mm5 \n\t" - "movq %%mm1, %%mm7 \n\t" - "movq 8(%%edi), %%mm6 \n\t" - "pandn %%mm6, %%mm7 \n\t" - "por %%mm7, %%mm5 \n\t" - "movq %%mm5, 8(%%edi) \n\t" - - "movq 16(%%esi), %%mm6 \n\t" - "pand %%mm2, %%mm6 \n\t" - "movq %%mm2, %%mm4 \n\t" - "movq 16(%%edi), %%mm7 \n\t" - "pandn %%mm7, %%mm4 \n\t" - "por %%mm4, %%mm6 \n\t" - "movq %%mm6, 16(%%edi) \n\t" - - "addl $24, %%esi \n\t" // inc by 24 bytes processed - "addl $24, %%edi \n\t" - "subl $8, %%ecx \n\t" // dec by 8 pixels processed - - "ja mainloop24 \n\t" - - "mainloop24end: \n\t" -// preload "movl diff, %%ecx \n\t" // (diff is in eax) - "movl %%eax, %%ecx \n\t" - "cmpl $0, %%ecx \n\t" - "jz end24 \n\t" -// preload "movl mask, %%edx \n\t" - "sall $24, %%edx \n\t" // make low byte, high byte - - "secondloop24: \n\t" - "sall %%edx \n\t" // move high bit to CF - "jnc skip24 \n\t" // if CF = 0 - "movw (%%esi), %%ax \n\t" - "movw %%ax, (%%edi) \n\t" - "xorl %%eax, %%eax \n\t" - "movb 2(%%esi), %%al \n\t" - "movb %%al, 2(%%edi) \n\t" - - "skip24: \n\t" - "addl $3, %%esi \n\t" - "addl $3, %%edi \n\t" - "decl %%ecx \n\t" - "jnz secondloop24 \n\t" - - "end24: \n\t" - "EMMS \n\t" // DONE - - : "=a" (dummy_value_a), // output regs (dummy) - "=d" (dummy_value_d), - "=c" (dummy_value_c), - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "3" (srcptr), // esi // input regs - "4" (dstptr), // edi - "0" (diff), // eax -// was (unmask) "b" RESERVED // ebx // Global Offset Table idx - "2" (len), // ecx - "1" (mask) // edx - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2" // clobber list - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - else /* mmx _not supported - Use modified C routine */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - register png_uint_32 i; - png_uint_32 initial_val = BPP3 * png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = BPP3 * png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = BPP3 * png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = BPP3 * len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff*BPP3; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - } /* end of else (_mmx_supported) */ - - break; - } /* end 24 bpp */ - - case 32: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - png_uint_32 len; - int diff; - int dummy_value_a; // fix 'forbidden register spilled' error - int dummy_value_d; - int dummy_value_c; - int dummy_value_S; - int dummy_value_D; - _unmask = ~mask; // global variable for -fPIC version - srcptr = png_ptr->row_buf + 1; - dstptr = row; - len = png_ptr->width &~7; // reduce to multiple of 8 - diff = (int) (png_ptr->width & 7); // amount lost // - - __asm__ __volatile__ ( - "movd _unmask, %%mm7 \n\t" // load bit pattern - "psubb %%mm6, %%mm6 \n\t" // zero mm6 - "punpcklbw %%mm7, %%mm7 \n\t" - "punpcklwd %%mm7, %%mm7 \n\t" - "punpckldq %%mm7, %%mm7 \n\t" // fill reg with 8 masks - - "movq _mask32_0, %%mm0 \n\t" - "movq _mask32_1, %%mm1 \n\t" - "movq _mask32_2, %%mm2 \n\t" - "movq _mask32_3, %%mm3 \n\t" - - "pand %%mm7, %%mm0 \n\t" - "pand %%mm7, %%mm1 \n\t" - "pand %%mm7, %%mm2 \n\t" - "pand %%mm7, %%mm3 \n\t" - - "pcmpeqb %%mm6, %%mm0 \n\t" - "pcmpeqb %%mm6, %%mm1 \n\t" - "pcmpeqb %%mm6, %%mm2 \n\t" - "pcmpeqb %%mm6, %%mm3 \n\t" - -// preload "movl len, %%ecx \n\t" // load length of line -// preload "movl srcptr, %%esi \n\t" // load source -// preload "movl dstptr, %%edi \n\t" // load dest - - "cmpl $0, %%ecx \n\t" // lcr - "jz mainloop32end \n\t" - - "mainloop32: \n\t" - "movq (%%esi), %%mm4 \n\t" - "pand %%mm0, %%mm4 \n\t" - "movq %%mm0, %%mm6 \n\t" - "movq (%%edi), %%mm7 \n\t" - "pandn %%mm7, %%mm6 \n\t" - "por %%mm6, %%mm4 \n\t" - "movq %%mm4, (%%edi) \n\t" - - "movq 8(%%esi), %%mm5 \n\t" - "pand %%mm1, %%mm5 \n\t" - "movq %%mm1, %%mm7 \n\t" - "movq 8(%%edi), %%mm6 \n\t" - "pandn %%mm6, %%mm7 \n\t" - "por %%mm7, %%mm5 \n\t" - "movq %%mm5, 8(%%edi) \n\t" - - "movq 16(%%esi), %%mm6 \n\t" - "pand %%mm2, %%mm6 \n\t" - "movq %%mm2, %%mm4 \n\t" - "movq 16(%%edi), %%mm7 \n\t" - "pandn %%mm7, %%mm4 \n\t" - "por %%mm4, %%mm6 \n\t" - "movq %%mm6, 16(%%edi) \n\t" - - "movq 24(%%esi), %%mm7 \n\t" - "pand %%mm3, %%mm7 \n\t" - "movq %%mm3, %%mm5 \n\t" - "movq 24(%%edi), %%mm4 \n\t" - "pandn %%mm4, %%mm5 \n\t" - "por %%mm5, %%mm7 \n\t" - "movq %%mm7, 24(%%edi) \n\t" - - "addl $32, %%esi \n\t" // inc by 32 bytes processed - "addl $32, %%edi \n\t" - "subl $8, %%ecx \n\t" // dec by 8 pixels processed - "ja mainloop32 \n\t" - - "mainloop32end: \n\t" -// preload "movl diff, %%ecx \n\t" // (diff is in eax) - "movl %%eax, %%ecx \n\t" - "cmpl $0, %%ecx \n\t" - "jz end32 \n\t" -// preload "movl mask, %%edx \n\t" - "sall $24, %%edx \n\t" // low byte => high byte - - "secondloop32: \n\t" - "sall %%edx \n\t" // move high bit to CF - "jnc skip32 \n\t" // if CF = 0 - "movl (%%esi), %%eax \n\t" - "movl %%eax, (%%edi) \n\t" - - "skip32: \n\t" - "addl $4, %%esi \n\t" - "addl $4, %%edi \n\t" - "decl %%ecx \n\t" - "jnz secondloop32 \n\t" - - "end32: \n\t" - "EMMS \n\t" // DONE - - : "=a" (dummy_value_a), // output regs (dummy) - "=d" (dummy_value_d), - "=c" (dummy_value_c), - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "3" (srcptr), // esi // input regs - "4" (dstptr), // edi - "0" (diff), // eax -// was (unmask) "b" RESERVED // ebx // Global Offset Table idx - "2" (len), // ecx - "1" (mask) // edx - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2", "%mm3" // clobber list - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - else /* mmx _not supported - Use modified C routine */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - register png_uint_32 i; - png_uint_32 initial_val = BPP4 * png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = BPP4 * png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = BPP4 * png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = BPP4 * len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff*BPP4; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - } /* end of else (_mmx_supported) */ - - break; - } /* end 32 bpp */ - - case 48: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - png_uint_32 len; - int diff; - int dummy_value_a; // fix 'forbidden register spilled' error - int dummy_value_d; - int dummy_value_c; - int dummy_value_S; - int dummy_value_D; - _unmask = ~mask; // global variable for -fPIC version - srcptr = png_ptr->row_buf + 1; - dstptr = row; - len = png_ptr->width &~7; // reduce to multiple of 8 - diff = (int) (png_ptr->width & 7); // amount lost // - - __asm__ __volatile__ ( - "movd _unmask, %%mm7 \n\t" // load bit pattern - "psubb %%mm6, %%mm6 \n\t" // zero mm6 - "punpcklbw %%mm7, %%mm7 \n\t" - "punpcklwd %%mm7, %%mm7 \n\t" - "punpckldq %%mm7, %%mm7 \n\t" // fill reg with 8 masks - - "movq _mask48_0, %%mm0 \n\t" - "movq _mask48_1, %%mm1 \n\t" - "movq _mask48_2, %%mm2 \n\t" - "movq _mask48_3, %%mm3 \n\t" - "movq _mask48_4, %%mm4 \n\t" - "movq _mask48_5, %%mm5 \n\t" - - "pand %%mm7, %%mm0 \n\t" - "pand %%mm7, %%mm1 \n\t" - "pand %%mm7, %%mm2 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pand %%mm7, %%mm4 \n\t" - "pand %%mm7, %%mm5 \n\t" - - "pcmpeqb %%mm6, %%mm0 \n\t" - "pcmpeqb %%mm6, %%mm1 \n\t" - "pcmpeqb %%mm6, %%mm2 \n\t" - "pcmpeqb %%mm6, %%mm3 \n\t" - "pcmpeqb %%mm6, %%mm4 \n\t" - "pcmpeqb %%mm6, %%mm5 \n\t" - -// preload "movl len, %%ecx \n\t" // load length of line -// preload "movl srcptr, %%esi \n\t" // load source -// preload "movl dstptr, %%edi \n\t" // load dest - - "cmpl $0, %%ecx \n\t" - "jz mainloop48end \n\t" - - "mainloop48: \n\t" - "movq (%%esi), %%mm7 \n\t" - "pand %%mm0, %%mm7 \n\t" - "movq %%mm0, %%mm6 \n\t" - "pandn (%%edi), %%mm6 \n\t" - "por %%mm6, %%mm7 \n\t" - "movq %%mm7, (%%edi) \n\t" - - "movq 8(%%esi), %%mm6 \n\t" - "pand %%mm1, %%mm6 \n\t" - "movq %%mm1, %%mm7 \n\t" - "pandn 8(%%edi), %%mm7 \n\t" - "por %%mm7, %%mm6 \n\t" - "movq %%mm6, 8(%%edi) \n\t" - - "movq 16(%%esi), %%mm6 \n\t" - "pand %%mm2, %%mm6 \n\t" - "movq %%mm2, %%mm7 \n\t" - "pandn 16(%%edi), %%mm7 \n\t" - "por %%mm7, %%mm6 \n\t" - "movq %%mm6, 16(%%edi) \n\t" - - "movq 24(%%esi), %%mm7 \n\t" - "pand %%mm3, %%mm7 \n\t" - "movq %%mm3, %%mm6 \n\t" - "pandn 24(%%edi), %%mm6 \n\t" - "por %%mm6, %%mm7 \n\t" - "movq %%mm7, 24(%%edi) \n\t" - - "movq 32(%%esi), %%mm6 \n\t" - "pand %%mm4, %%mm6 \n\t" - "movq %%mm4, %%mm7 \n\t" - "pandn 32(%%edi), %%mm7 \n\t" - "por %%mm7, %%mm6 \n\t" - "movq %%mm6, 32(%%edi) \n\t" - - "movq 40(%%esi), %%mm7 \n\t" - "pand %%mm5, %%mm7 \n\t" - "movq %%mm5, %%mm6 \n\t" - "pandn 40(%%edi), %%mm6 \n\t" - "por %%mm6, %%mm7 \n\t" - "movq %%mm7, 40(%%edi) \n\t" - - "addl $48, %%esi \n\t" // inc by 48 bytes processed - "addl $48, %%edi \n\t" - "subl $8, %%ecx \n\t" // dec by 8 pixels processed - - "ja mainloop48 \n\t" - - "mainloop48end: \n\t" -// preload "movl diff, %%ecx \n\t" // (diff is in eax) - "movl %%eax, %%ecx \n\t" - "cmpl $0, %%ecx \n\t" - "jz end48 \n\t" -// preload "movl mask, %%edx \n\t" - "sall $24, %%edx \n\t" // make low byte, high byte - - "secondloop48: \n\t" - "sall %%edx \n\t" // move high bit to CF - "jnc skip48 \n\t" // if CF = 0 - "movl (%%esi), %%eax \n\t" - "movl %%eax, (%%edi) \n\t" - - "skip48: \n\t" - "addl $4, %%esi \n\t" - "addl $4, %%edi \n\t" - "decl %%ecx \n\t" - "jnz secondloop48 \n\t" - - "end48: \n\t" - "EMMS \n\t" // DONE - - : "=a" (dummy_value_a), // output regs (dummy) - "=d" (dummy_value_d), - "=c" (dummy_value_c), - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "3" (srcptr), // esi // input regs - "4" (dstptr), // edi - "0" (diff), // eax -// was (unmask) "b" RESERVED // ebx // Global Offset Table idx - "2" (len), // ecx - "1" (mask) // edx - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2", "%mm3" // clobber list - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - else /* mmx _not supported - Use modified C routine */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - register png_uint_32 i; - png_uint_32 initial_val = BPP6 * png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = BPP6 * png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = BPP6 * png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = BPP6 * len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff*BPP6; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - } /* end of else (_mmx_supported) */ - - break; - } /* end 48 bpp */ - - case 64: /* png_ptr->row_info.pixel_depth */ - { - png_bytep srcptr; - png_bytep dstptr; - register png_uint_32 i; - png_uint_32 initial_val = BPP8 * png_pass_start[png_ptr->pass]; - /* png.c: png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; */ - register int stride = BPP8 * png_pass_inc[png_ptr->pass]; - /* png.c: png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; */ - register int rep_bytes = BPP8 * png_pass_width[png_ptr->pass]; - /* png.c: png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; */ - png_uint_32 len = png_ptr->width &~7; /* reduce to mult. of 8 */ - int diff = (int) (png_ptr->width & 7); /* amount lost */ - register png_uint_32 final_val = BPP8 * len; /* GRR bugfix */ - - srcptr = png_ptr->row_buf + 1 + initial_val; - dstptr = row + initial_val; - - for (i = initial_val; i < final_val; i += stride) - { - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - if (diff) /* number of leftover pixels: 3 for pngtest */ - { - final_val+=diff*BPP8; - for (; i < final_val; i += stride) - { - if (rep_bytes > (int)(final_val-i)) - rep_bytes = (int)(final_val-i); - png_memcpy(dstptr, srcptr, rep_bytes); - srcptr += stride; - dstptr += stride; - } - } - - break; - } /* end 64 bpp */ - - default: /* png_ptr->row_info.pixel_depth != 1,2,4,8,16,24,32,48,64 */ - { - /* this should never happen */ - png_warning(png_ptr, "Invalid row_info.pixel_depth in pnggccrd"); - break; - } - } /* end switch (png_ptr->row_info.pixel_depth) */ - - } /* end if (non-trivial mask) */ - -} /* end png_combine_row() */ - -#endif /* PNG_HAVE_ASSEMBLER_COMBINE_ROW */ - - - - -/*===========================================================================*/ -/* */ -/* P N G _ D O _ R E A D _ I N T E R L A C E */ -/* */ -/*===========================================================================*/ - -#if defined(PNG_READ_INTERLACING_SUPPORTED) -#if defined(PNG_HAVE_ASSEMBLER_READ_INTERLACE) - -/* png_do_read_interlace() is called after any 16-bit to 8-bit conversion - * has taken place. [GRR: what other steps come before and/or after?] - */ - -void /* PRIVATE */ -png_do_read_interlace(png_structp png_ptr) -{ - png_row_infop row_info = &(png_ptr->row_info); - png_bytep row = png_ptr->row_buf + 1; - int pass = png_ptr->pass; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - png_uint_32 transformations = png_ptr->transformations; -#endif - - png_debug(1, "in png_do_read_interlace (pnggccrd.c)\n"); - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) - if (_mmx_supported == 2) { -#if !defined(PNG_1_0_X) - /* this should have happened in png_init_mmx_flags() already */ - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } -#endif - - if (row != NULL && row_info != NULL) - { - png_uint_32 final_width; - - final_width = row_info->width * png_pass_inc[pass]; - - switch (row_info->pixel_depth) - { - case 1: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_byte v; - png_uint_32 i; - int j; - - sp = row + (png_size_t)((row_info->width - 1) >> 3); - dp = row + (png_size_t)((final_width - 1) >> 3); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (int)((row_info->width + 7) & 7); - dshift = (int)((final_width + 7) & 7); - s_start = 7; - s_end = 0; - s_inc = -1; - } - else -#endif - { - sshift = 7 - (int)((row_info->width + 7) & 7); - dshift = 7 - (int)((final_width + 7) & 7); - s_start = 0; - s_end = 7; - s_inc = 1; - } - - for (i = row_info->width; i; i--) - { - v = (png_byte)((*sp >> sshift) & 0x1); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0x7f7f >> (7 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - case 2: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_uint_32 i; - - sp = row + (png_size_t)((row_info->width - 1) >> 2); - dp = row + (png_size_t)((final_width - 1) >> 2); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (png_size_t)(((row_info->width + 3) & 3) << 1); - dshift = (png_size_t)(((final_width + 3) & 3) << 1); - s_start = 6; - s_end = 0; - s_inc = -2; - } - else -#endif - { - sshift = (png_size_t)((3 - ((row_info->width + 3) & 3)) << 1); - dshift = (png_size_t)((3 - ((final_width + 3) & 3)) << 1); - s_start = 0; - s_end = 6; - s_inc = 2; - } - - for (i = row_info->width; i; i--) - { - png_byte v; - int j; - - v = (png_byte)((*sp >> sshift) & 0x3); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0x3f3f >> (6 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - case 4: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_uint_32 i; - - sp = row + (png_size_t)((row_info->width - 1) >> 1); - dp = row + (png_size_t)((final_width - 1) >> 1); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (png_size_t)(((row_info->width + 1) & 1) << 2); - dshift = (png_size_t)(((final_width + 1) & 1) << 2); - s_start = 4; - s_end = 0; - s_inc = -4; - } - else -#endif - { - sshift = (png_size_t)((1 - ((row_info->width + 1) & 1)) << 2); - dshift = (png_size_t)((1 - ((final_width + 1) & 1)) << 2); - s_start = 0; - s_end = 4; - s_inc = 4; - } - - for (i = row_info->width; i; i--) - { - png_byte v; - int j; - - v = (png_byte)((*sp >> sshift) & 0xf); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0xf0f >> (4 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - /*====================================================================*/ - - default: /* 8-bit or larger (this is where the routine is modified) */ - { -#if 0 -// static unsigned long long _const4 = 0x0000000000FFFFFFLL; no good -// static unsigned long long const4 = 0x0000000000FFFFFFLL; no good -// unsigned long long _const4 = 0x0000000000FFFFFFLL; no good -// unsigned long long const4 = 0x0000000000FFFFFFLL; no good -#endif - png_bytep sptr, dp; - png_uint_32 i; - png_size_t pixel_bytes; - int width = (int)row_info->width; - - pixel_bytes = (row_info->pixel_depth >> 3); - - /* point sptr at the last pixel in the pre-expanded row: */ - sptr = row + (width - 1) * pixel_bytes; - - /* point dp at the last pixel position in the expanded row: */ - dp = row + (final_width - 1) * pixel_bytes; - - /* New code by Nirav Chhatrapati - Intel Corporation */ - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_INTERLACE) - /* && _mmx_supported */ ) -#else - if (_mmx_supported) -#endif - { - //-------------------------------------------------------------- - if (pixel_bytes == 3) - { - if (((pass == 0) || (pass == 1)) && width) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - int dummy_value_a; - - __asm__ __volatile__ ( - "subl $21, %%edi \n\t" - // (png_pass_inc[pass] - 1)*pixel_bytes - - ".loop3_pass0: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x x 2 1 0 - "pand (%3), %%mm0 \n\t" // z z z z z 2 1 0 - "movq %%mm0, %%mm1 \n\t" // z z z z z 2 1 0 - "psllq $16, %%mm0 \n\t" // z z z 2 1 0 z z - "movq %%mm0, %%mm2 \n\t" // z z z 2 1 0 z z - "psllq $24, %%mm0 \n\t" // 2 1 0 z z z z z - "psrlq $8, %%mm1 \n\t" // z z z z z z 2 1 - "por %%mm2, %%mm0 \n\t" // 2 1 0 2 1 0 z z - "por %%mm1, %%mm0 \n\t" // 2 1 0 2 1 0 2 1 - "movq %%mm0, %%mm3 \n\t" // 2 1 0 2 1 0 2 1 - "psllq $16, %%mm0 \n\t" // 0 2 1 0 2 1 z z - "movq %%mm3, %%mm4 \n\t" // 2 1 0 2 1 0 2 1 - "punpckhdq %%mm0, %%mm3 \n\t" // 0 2 1 0 2 1 0 2 - "movq %%mm4, 16(%%edi) \n\t" - "psrlq $32, %%mm0 \n\t" // z z z z 0 2 1 0 - "movq %%mm3, 8(%%edi) \n\t" - "punpckldq %%mm4, %%mm0 \n\t" // 1 0 2 1 0 2 1 0 - "subl $3, %%esi \n\t" - "movq %%mm0, (%%edi) \n\t" - "subl $24, %%edi \n\t" - "decl %%ecx \n\t" - "jnz .loop3_pass0 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D), - "=a" (dummy_value_a) - - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width), // ecx - "3" (&_const4) // %1(?) (0x0000000000FFFFFFLL) - -#if 0 /* %mm0, ..., %mm4 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2" // clobber list - , "%mm3", "%mm4" -#endif - ); - } - else if (((pass == 2) || (pass == 3)) && width) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - int dummy_value_a; - - __asm__ __volatile__ ( - "subl $9, %%edi \n\t" - // (png_pass_inc[pass] - 1)*pixel_bytes - - ".loop3_pass2: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x x 2 1 0 - "pand (%3), %%mm0 \n\t" // z z z z z 2 1 0 - "movq %%mm0, %%mm1 \n\t" // z z z z z 2 1 0 - "psllq $16, %%mm0 \n\t" // z z z 2 1 0 z z - "movq %%mm0, %%mm2 \n\t" // z z z 2 1 0 z z - "psllq $24, %%mm0 \n\t" // 2 1 0 z z z z z - "psrlq $8, %%mm1 \n\t" // z z z z z z 2 1 - "por %%mm2, %%mm0 \n\t" // 2 1 0 2 1 0 z z - "por %%mm1, %%mm0 \n\t" // 2 1 0 2 1 0 2 1 - "movq %%mm0, 4(%%edi) \n\t" - "psrlq $16, %%mm0 \n\t" // z z 2 1 0 2 1 0 - "subl $3, %%esi \n\t" - "movd %%mm0, (%%edi) \n\t" - "subl $12, %%edi \n\t" - "decl %%ecx \n\t" - "jnz .loop3_pass2 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D), - "=a" (dummy_value_a) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width), // ecx - "3" (&_const4) // (0x0000000000FFFFFFLL) - -#if 0 /* %mm0, ..., %mm2 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2" // clobber list -#endif - ); - } - else if (width) /* && ((pass == 4) || (pass == 5)) */ - { - int width_mmx = ((width >> 1) << 1) - 8; // GRR: huh? - if (width_mmx < 0) - width_mmx = 0; - width -= width_mmx; // 8 or 9 pix, 24 or 27 bytes - if (width_mmx) - { - // png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; - // sptr points at last pixel in pre-expanded row - // dp points at last pixel position in expanded row - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - int dummy_value_a; - int dummy_value_d; - - __asm__ __volatile__ ( - "subl $3, %%esi \n\t" - "subl $9, %%edi \n\t" - // (png_pass_inc[pass] + 1)*pixel_bytes - - ".loop3_pass4: \n\t" - "movq (%%esi), %%mm0 \n\t" // x x 5 4 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // x x 5 4 3 2 1 0 - "movq %%mm0, %%mm2 \n\t" // x x 5 4 3 2 1 0 - "psllq $24, %%mm0 \n\t" // 4 3 2 1 0 z z z - "pand (%3), %%mm1 \n\t" // z z z z z 2 1 0 - "psrlq $24, %%mm2 \n\t" // z z z x x 5 4 3 - "por %%mm1, %%mm0 \n\t" // 4 3 2 1 0 2 1 0 - "movq %%mm2, %%mm3 \n\t" // z z z x x 5 4 3 - "psllq $8, %%mm2 \n\t" // z z x x 5 4 3 z - "movq %%mm0, (%%edi) \n\t" - "psrlq $16, %%mm3 \n\t" // z z z z z x x 5 - "pand (%4), %%mm3 \n\t" // z z z z z z z 5 - "por %%mm3, %%mm2 \n\t" // z z x x 5 4 3 5 - "subl $6, %%esi \n\t" - "movd %%mm2, 8(%%edi) \n\t" - "subl $12, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop3_pass4 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D), - "=a" (dummy_value_a), - "=d" (dummy_value_d) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx), // ecx - "3" (&_const4), // 0x0000000000FFFFFFLL - "4" (&_const6) // 0x00000000000000FFLL - -#if 0 /* %mm0, ..., %mm3 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list - , "%mm2", "%mm3" -#endif - ); - } - - sptr -= width_mmx*3; - dp -= width_mmx*6; - for (i = width; i; i--) - { - png_byte v[8]; - int j; - - png_memcpy(v, sptr, 3); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 3); - dp -= 3; - } - sptr -= 3; - } - } - } /* end of pixel_bytes == 3 */ - - //-------------------------------------------------------------- - else if (pixel_bytes == 1) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 2) << 2); - width -= width_mmx; // 0-3 pixels => 0-3 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $3, %%esi \n\t" - "subl $31, %%edi \n\t" - - ".loop1_pass0: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // x x x x 3 2 1 0 - "punpcklbw %%mm0, %%mm0 \n\t" // 3 3 2 2 1 1 0 0 - "movq %%mm0, %%mm2 \n\t" // 3 3 2 2 1 1 0 0 - "punpcklwd %%mm0, %%mm0 \n\t" // 1 1 1 1 0 0 0 0 - "movq %%mm0, %%mm3 \n\t" // 1 1 1 1 0 0 0 0 - "punpckldq %%mm0, %%mm0 \n\t" // 0 0 0 0 0 0 0 0 - "punpckhdq %%mm3, %%mm3 \n\t" // 1 1 1 1 1 1 1 1 - "movq %%mm0, (%%edi) \n\t" - "punpckhwd %%mm2, %%mm2 \n\t" // 3 3 3 3 2 2 2 2 - "movq %%mm3, 8(%%edi) \n\t" - "movq %%mm2, %%mm4 \n\t" // 3 3 3 3 2 2 2 2 - "punpckldq %%mm2, %%mm2 \n\t" // 2 2 2 2 2 2 2 2 - "punpckhdq %%mm4, %%mm4 \n\t" // 3 3 3 3 3 3 3 3 - "movq %%mm2, 16(%%edi) \n\t" - "subl $4, %%esi \n\t" - "movq %%mm4, 24(%%edi) \n\t" - "subl $32, %%edi \n\t" - "subl $4, %%ecx \n\t" - "jnz .loop1_pass0 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, ..., %mm4 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1", "%mm2" // clobber list - , "%mm3", "%mm4" -#endif - ); - } - - sptr -= width_mmx; - dp -= width_mmx*8; - for (i = width; i; i--) - { - int j; - - /* I simplified this part in version 1.0.4e - * here and in several other instances where - * pixel_bytes == 1 -- GR-P - * - * Original code: - * - * png_byte v[8]; - * png_memcpy(v, sptr, pixel_bytes); - * for (j = 0; j < png_pass_inc[pass]; j++) - * { - * png_memcpy(dp, v, pixel_bytes); - * dp -= pixel_bytes; - * } - * sptr -= pixel_bytes; - * - * Replacement code is in the next three lines: - */ - - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - --sptr; - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 2) << 2); - width -= width_mmx; // 0-3 pixels => 0-3 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $3, %%esi \n\t" - "subl $15, %%edi \n\t" - - ".loop1_pass2: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x 3 2 1 0 - "punpcklbw %%mm0, %%mm0 \n\t" // 3 3 2 2 1 1 0 0 - "movq %%mm0, %%mm1 \n\t" // 3 3 2 2 1 1 0 0 - "punpcklwd %%mm0, %%mm0 \n\t" // 1 1 1 1 0 0 0 0 - "punpckhwd %%mm1, %%mm1 \n\t" // 3 3 3 3 2 2 2 2 - "movq %%mm0, (%%edi) \n\t" - "subl $4, %%esi \n\t" - "movq %%mm1, 8(%%edi) \n\t" - "subl $16, %%edi \n\t" - "subl $4, %%ecx \n\t" - "jnz .loop1_pass2 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= width_mmx; - dp -= width_mmx*4; - for (i = width; i; i--) - { - int j; - - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - --sptr; - } - } - else if (width) /* && ((pass == 4) || (pass == 5)) */ - { - int width_mmx = ((width >> 3) << 3); - width -= width_mmx; // 0-3 pixels => 0-3 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $7, %%esi \n\t" - "subl $15, %%edi \n\t" - - ".loop1_pass4: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // 7 6 5 4 3 2 1 0 - "punpcklbw %%mm0, %%mm0 \n\t" // 3 3 2 2 1 1 0 0 - "punpckhbw %%mm1, %%mm1 \n\t" // 7 7 6 6 5 5 4 4 - "movq %%mm1, 8(%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm0, (%%edi) \n\t" - "subl $16, %%edi \n\t" - "subl $8, %%ecx \n\t" - "jnz .loop1_pass4 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (none) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= width_mmx; - dp -= width_mmx*2; - for (i = width; i; i--) - { - int j; - - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - --sptr; - } - } - } /* end of pixel_bytes == 1 */ - - //-------------------------------------------------------------- - else if (pixel_bytes == 2) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 1) << 1); - width -= width_mmx; // 0,1 pixels => 0,2 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $2, %%esi \n\t" - "subl $30, %%edi \n\t" - - ".loop2_pass0: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x 3 2 1 0 - "punpcklwd %%mm0, %%mm0 \n\t" // 3 2 3 2 1 0 1 0 - "movq %%mm0, %%mm1 \n\t" // 3 2 3 2 1 0 1 0 - "punpckldq %%mm0, %%mm0 \n\t" // 1 0 1 0 1 0 1 0 - "punpckhdq %%mm1, %%mm1 \n\t" // 3 2 3 2 3 2 3 2 - "movq %%mm0, (%%edi) \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "movq %%mm1, 16(%%edi) \n\t" - "subl $4, %%esi \n\t" - "movq %%mm1, 24(%%edi) \n\t" - "subl $32, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop2_pass0 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*16 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; // 0,1 pixels => 0,2 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $2, %%esi \n\t" - "subl $14, %%edi \n\t" - - ".loop2_pass2: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x 3 2 1 0 - "punpcklwd %%mm0, %%mm0 \n\t" // 3 2 3 2 1 0 1 0 - "movq %%mm0, %%mm1 \n\t" // 3 2 3 2 1 0 1 0 - "punpckldq %%mm0, %%mm0 \n\t" // 1 0 1 0 1 0 1 0 - "punpckhdq %%mm1, %%mm1 \n\t" // 3 2 3 2 3 2 3 2 - "movq %%mm0, (%%edi) \n\t" - "subl $4, %%esi \n\t" - "movq %%mm1, 8(%%edi) \n\t" - "subl $16, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop2_pass2 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*8 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - else if (width) // pass == 4 or 5 - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; // 0,1 pixels => 0,2 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $2, %%esi \n\t" - "subl $6, %%edi \n\t" - - ".loop2_pass4: \n\t" - "movd (%%esi), %%mm0 \n\t" // x x x x 3 2 1 0 - "punpcklwd %%mm0, %%mm0 \n\t" // 3 2 3 2 1 0 1 0 - "subl $4, %%esi \n\t" - "movq %%mm0, (%%edi) \n\t" - "subl $8, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop2_pass4 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0" // clobber list -#endif - ); - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*4 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - } /* end of pixel_bytes == 2 */ - - //-------------------------------------------------------------- - else if (pixel_bytes == 4) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 1) << 1); - width -= width_mmx; // 0,1 pixels => 0,4 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $4, %%esi \n\t" - "subl $60, %%edi \n\t" - - ".loop4_pass0: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // 7 6 5 4 3 2 1 0 - "punpckldq %%mm0, %%mm0 \n\t" // 3 2 1 0 3 2 1 0 - "punpckhdq %%mm1, %%mm1 \n\t" // 7 6 5 4 7 6 5 4 - "movq %%mm0, (%%edi) \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "movq %%mm0, 16(%%edi) \n\t" - "movq %%mm0, 24(%%edi) \n\t" - "movq %%mm1, 32(%%edi) \n\t" - "movq %%mm1, 40(%%edi) \n\t" - "movq %%mm1, 48(%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm1, 56(%%edi) \n\t" - "subl $64, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop4_pass0 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*32 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 1) << 1); - width -= width_mmx; // 0,1 pixels => 0,4 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $4, %%esi \n\t" - "subl $28, %%edi \n\t" - - ".loop4_pass2: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // 7 6 5 4 3 2 1 0 - "punpckldq %%mm0, %%mm0 \n\t" // 3 2 1 0 3 2 1 0 - "punpckhdq %%mm1, %%mm1 \n\t" // 7 6 5 4 7 6 5 4 - "movq %%mm0, (%%edi) \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "movq %%mm1, 16(%%edi) \n\t" - "movq %%mm1, 24(%%edi) \n\t" - "subl $8, %%esi \n\t" - "subl $32, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop4_pass2 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*16 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - else if (width) // pass == 4 or 5 - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; // 0,1 pixels => 0,4 bytes - if (width_mmx) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $4, %%esi \n\t" - "subl $12, %%edi \n\t" - - ".loop4_pass4: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, %%mm1 \n\t" // 7 6 5 4 3 2 1 0 - "punpckldq %%mm0, %%mm0 \n\t" // 3 2 1 0 3 2 1 0 - "punpckhdq %%mm1, %%mm1 \n\t" // 7 6 5 4 7 6 5 4 - "movq %%mm0, (%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm1, 8(%%edi) \n\t" - "subl $16, %%edi \n\t" - "subl $2, %%ecx \n\t" - "jnz .loop4_pass4 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width_mmx) // ecx - -#if 0 /* %mm0, %mm1 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0", "%mm1" // clobber list -#endif - ); - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*8 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - } /* end of pixel_bytes == 4 */ - - //-------------------------------------------------------------- - else if (pixel_bytes == 8) - { -// GRR TEST: should work, but needs testing (special 64-bit version of rpng2?) - // GRR NOTE: no need to combine passes here! - if (((pass == 0) || (pass == 1)) && width) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - // source is 8-byte RRGGBBAA - // dest is 64-byte RRGGBBAA RRGGBBAA RRGGBBAA RRGGBBAA ... - __asm__ __volatile__ ( - "subl $56, %%edi \n\t" // start of last block - - ".loop8_pass0: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, (%%edi) \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "movq %%mm0, 16(%%edi) \n\t" - "movq %%mm0, 24(%%edi) \n\t" - "movq %%mm0, 32(%%edi) \n\t" - "movq %%mm0, 40(%%edi) \n\t" - "movq %%mm0, 48(%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm0, 56(%%edi) \n\t" - "subl $64, %%edi \n\t" - "decl %%ecx \n\t" - "jnz .loop8_pass0 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width) // ecx - -#if 0 /* %mm0 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0" // clobber list -#endif - ); - } - else if (((pass == 2) || (pass == 3)) && width) - { - // source is 8-byte RRGGBBAA - // dest is 32-byte RRGGBBAA RRGGBBAA RRGGBBAA RRGGBBAA - // (recall that expansion is _in place_: sptr and dp - // both point at locations within same row buffer) - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $24, %%edi \n\t" // start of last block - - ".loop8_pass2: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, (%%edi) \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "movq %%mm0, 16(%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm0, 24(%%edi) \n\t" - "subl $32, %%edi \n\t" - "decl %%ecx \n\t" - "jnz .loop8_pass2 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width) // ecx - -#if 0 /* %mm0 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0" // clobber list -#endif - ); - } - } - else if (width) // pass == 4 or 5 - { - // source is 8-byte RRGGBBAA - // dest is 16-byte RRGGBBAA RRGGBBAA - { - int dummy_value_c; // fix 'forbidden register spilled' - int dummy_value_S; - int dummy_value_D; - - __asm__ __volatile__ ( - "subl $8, %%edi \n\t" // start of last block - - ".loop8_pass4: \n\t" - "movq (%%esi), %%mm0 \n\t" // 7 6 5 4 3 2 1 0 - "movq %%mm0, (%%edi) \n\t" - "subl $8, %%esi \n\t" - "movq %%mm0, 8(%%edi) \n\t" - "subl $16, %%edi \n\t" - "decl %%ecx \n\t" - "jnz .loop8_pass4 \n\t" - "EMMS \n\t" // DONE - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "1" (sptr), // esi // input regs - "2" (dp), // edi - "0" (width) // ecx - -#if 0 /* %mm0 not supported by gcc 2.7.2.3 or egcs 1.1 */ - : "%mm0" // clobber list -#endif - ); - } - } - - } /* end of pixel_bytes == 8 */ - - //-------------------------------------------------------------- - else if (pixel_bytes == 6) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 6); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 6); - dp -= 6; - } - sptr -= 6; - } - } /* end of pixel_bytes == 6 */ - - //-------------------------------------------------------------- - else - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr-= pixel_bytes; - } - } - } // end of _mmx_supported ======================================== - - else /* MMX not supported: use modified C code - takes advantage - * of inlining of png_memcpy for a constant */ - /* GRR 19991007: does it? or should pixel_bytes in each - * block be replaced with immediate value (e.g., 1)? */ - /* GRR 19991017: replaced with constants in each case */ -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - if (pixel_bytes == 1) - { - for (i = width; i; i--) - { - int j; - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - --sptr; - } - } - else if (pixel_bytes == 3) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 3); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 3); - dp -= 3; - } - sptr -= 3; - } - } - else if (pixel_bytes == 2) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 2); - dp -= 2; - } - sptr -= 2; - } - } - else if (pixel_bytes == 4) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { -#ifdef PNG_DEBUG - if (dp < row || dp+3 > row+png_ptr->row_buf_size) - { - printf("dp out of bounds: row=%d, dp=%d, rp=%d\n", - row, dp, row+png_ptr->row_buf_size); - printf("row_buf=%d\n",png_ptr->row_buf_size); - } -#endif - png_memcpy(dp, v, 4); - dp -= 4; - } - sptr -= 4; - } - } - else if (pixel_bytes == 6) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 6); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 6); - dp -= 6; - } - sptr -= 6; - } - } - else if (pixel_bytes == 8) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 8); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 8); - dp -= 8; - } - sptr -= 8; - } - } - else /* GRR: should never be reached */ - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - - } /* end if (MMX not supported) */ - break; - } - } /* end switch (row_info->pixel_depth) */ - - row_info->width = final_width; - - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth,final_width); - } - -} /* end png_do_read_interlace() */ - -#endif /* PNG_HAVE_ASSEMBLER_READ_INTERLACE */ -#endif /* PNG_READ_INTERLACING_SUPPORTED */ - - - -#if defined(PNG_HAVE_ASSEMBLER_READ_FILTER_ROW) -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) - -// These variables are utilized in the functions below. They are declared -// globally here to ensure alignment on 8-byte boundaries. - -union uAll { - long long use; - double align; -} _LBCarryMask = {0x0101010101010101LL}, - _HBClearMask = {0x7f7f7f7f7f7f7f7fLL}, - _ActiveMask, _ActiveMask2, _ActiveMaskEnd, _ShiftBpp, _ShiftRem; - -#ifdef PNG_THREAD_UNSAFE_OK -//===========================================================================// -// // -// P N G _ R E A D _ F I L T E R _ R O W _ M M X _ A V G // -// // -//===========================================================================// - -// Optimized code for PNG Average filter decoder - -static void /* PRIVATE */ -png_read_filter_row_mmx_avg(png_row_infop row_info, png_bytep row, - png_bytep prev_row) -{ - int bpp; - int dummy_value_c; // fix 'forbidden register 2 (cx) was spilled' error - int dummy_value_S; - int dummy_value_D; - - bpp = (row_info->pixel_depth + 7) >> 3; // get # bytes per pixel - _FullLength = row_info->rowbytes; // # of bytes to filter - - __asm__ __volatile__ ( - // initialize address pointers and offset -#ifdef __PIC__ - "pushl %%ebx \n\t" // save index to Global Offset Table -#endif -//pre "movl row, %%edi \n\t" // edi: Avg(x) - "xorl %%ebx, %%ebx \n\t" // ebx: x - "movl %%edi, %%edx \n\t" -//pre "movl prev_row, %%esi \n\t" // esi: Prior(x) -//pre "subl bpp, %%edx \n\t" // (bpp is preloaded into ecx) - "subl %%ecx, %%edx \n\t" // edx: Raw(x-bpp) - - "xorl %%eax,%%eax \n\t" - - // Compute the Raw value for the first bpp bytes - // Raw(x) = Avg(x) + (Prior(x)/2) - "avg_rlp: \n\t" - "movb (%%esi,%%ebx,),%%al \n\t" // load al with Prior(x) - "incl %%ebx \n\t" - "shrb %%al \n\t" // divide by 2 - "addb -1(%%edi,%%ebx,),%%al \n\t" // add Avg(x); -1 to offset inc ebx -//pre "cmpl bpp, %%ebx \n\t" // (bpp is preloaded into ecx) - "cmpl %%ecx, %%ebx \n\t" - "movb %%al,-1(%%edi,%%ebx,) \n\t" // write Raw(x); -1 to offset inc ebx - "jb avg_rlp \n\t" // mov does not affect flags - - // get # of bytes to alignment - "movl %%edi, _dif \n\t" // take start of row - "addl %%ebx, _dif \n\t" // add bpp - "addl $0xf, _dif \n\t" // add 7+8 to incr past alignment bdry - "andl $0xfffffff8, _dif \n\t" // mask to alignment boundary - "subl %%edi, _dif \n\t" // subtract from start => value ebx at - "jz avg_go \n\t" // alignment - - // fix alignment - // Compute the Raw value for the bytes up to the alignment boundary - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - "xorl %%ecx, %%ecx \n\t" - - "avg_lp1: \n\t" - "xorl %%eax, %%eax \n\t" - "movb (%%esi,%%ebx,), %%cl \n\t" // load cl with Prior(x) - "movb (%%edx,%%ebx,), %%al \n\t" // load al with Raw(x-bpp) - "addw %%cx, %%ax \n\t" - "incl %%ebx \n\t" - "shrw %%ax \n\t" // divide by 2 - "addb -1(%%edi,%%ebx,), %%al \n\t" // add Avg(x); -1 to offset inc ebx - "cmpl _dif, %%ebx \n\t" // check if at alignment boundary - "movb %%al, -1(%%edi,%%ebx,) \n\t" // write Raw(x); -1 to offset inc ebx - "jb avg_lp1 \n\t" // repeat until at alignment boundary - - "avg_go: \n\t" - "movl _FullLength, %%eax \n\t" - "movl %%eax, %%ecx \n\t" - "subl %%ebx, %%eax \n\t" // subtract alignment fix - "andl $0x00000007, %%eax \n\t" // calc bytes over mult of 8 - "subl %%eax, %%ecx \n\t" // drop over bytes from original length - "movl %%ecx, _MMXLength \n\t" -#ifdef __PIC__ - "popl %%ebx \n\t" // restore index to Global Offset Table -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list -#ifndef __PIC__ - , "%ebx" -#endif - // GRR: INCLUDE "memory" as clobbered? (_dif, _MMXLength) - // (seems to work fine without...) - ); - - // now do the math for the rest of the row - switch (bpp) - { - case 3: - { - _ActiveMask.use = 0x0000000000ffffffLL; - _ShiftBpp.use = 24; // == 3 * 8 - _ShiftRem.use = 40; // == 64 - 24 - - __asm__ __volatile__ ( - // re-init address pointers and offset - "movq _ActiveMask, %%mm7 \n\t" - "movl _dif, %%ecx \n\t" // ecx: x = offset to - "movq _LBCarryMask, %%mm5 \n\t" // alignment boundary -// preload "movl row, %%edi \n\t" // edi: Avg(x) - "movq _HBClearMask, %%mm4 \n\t" -// preload "movl prev_row, %%esi \n\t" // esi: Prior(x) - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm2 \n\t" // load previous aligned 8 bytes - // (correct pos. in loop below) - "avg_3lp: \n\t" - "movq (%%edi,%%ecx,), %%mm0 \n\t" // load mm0 with Avg(x) - "movq %%mm5, %%mm3 \n\t" - "psrlq _ShiftRem, %%mm2 \n\t" // correct position Raw(x-bpp) - // data - "movq (%%esi,%%ecx,), %%mm1 \n\t" // load mm1 with Prior(x) - "movq %%mm7, %%mm6 \n\t" - "pand %%mm1, %%mm3 \n\t" // get lsb for each prev_row byte - "psrlq $1, %%mm1 \n\t" // divide prev_row bytes by 2 - "pand %%mm4, %%mm1 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm0 \n\t" // add (Prev_row/2) to Avg for - // each byte - // add 1st active group (Raw(x-bpp)/2) to average with LBCarry - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 1 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active - // byte - // add 2nd active group (Raw(x-bpp)/2) to average with _LBCarry - "psllq _ShiftBpp, %%mm6 \n\t" // shift the mm6 mask to cover - // bytes 3-5 - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active - // byte - - // add 3rd active group (Raw(x-bpp)/2) to average with _LBCarry - "psllq _ShiftBpp, %%mm6 \n\t" // shift mm6 mask to cover last - // two - // bytes - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - // Data only needs to be shifted once here to - // get the correct x-bpp offset. - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "addl $8, %%ecx \n\t" - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active - // byte - // now ready to write back to memory - "movq %%mm0, -8(%%edi,%%ecx,) \n\t" - // move updated Raw(x) to use as Raw(x-bpp) for next loop - "cmpl _MMXLength, %%ecx \n\t" - "movq %%mm0, %%mm2 \n\t" // mov updated Raw(x) to mm2 - "jb avg_3lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 3 bpp - - case 6: - case 4: - //case 7: // who wrote this? PNG doesn't support 5 or 7 bytes/pixel - //case 5: // GRR BOGUS - { - _ActiveMask.use = 0xffffffffffffffffLL; // use shift below to clear - // appropriate inactive bytes - _ShiftBpp.use = bpp << 3; - _ShiftRem.use = 64 - _ShiftBpp.use; - - __asm__ __volatile__ ( - "movq _HBClearMask, %%mm4 \n\t" - - // re-init address pointers and offset - "movl _dif, %%ecx \n\t" // ecx: x = offset to - // alignment boundary - - // load _ActiveMask and clear all bytes except for 1st active group - "movq _ActiveMask, %%mm7 \n\t" -// preload "movl row, %%edi \n\t" // edi: Avg(x) - "psrlq _ShiftRem, %%mm7 \n\t" -// preload "movl prev_row, %%esi \n\t" // esi: Prior(x) - "movq %%mm7, %%mm6 \n\t" - "movq _LBCarryMask, %%mm5 \n\t" - "psllq _ShiftBpp, %%mm6 \n\t" // create mask for 2nd active - // group - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm2 \n\t" // load previous aligned 8 bytes - // (we correct pos. in loop below) - "avg_4lp: \n\t" - "movq (%%edi,%%ecx,), %%mm0 \n\t" - "psrlq _ShiftRem, %%mm2 \n\t" // shift data to pos. correctly - "movq (%%esi,%%ecx,), %%mm1 \n\t" - // add (Prev_row/2) to average - "movq %%mm5, %%mm3 \n\t" - "pand %%mm1, %%mm3 \n\t" // get lsb for each prev_row byte - "psrlq $1, %%mm1 \n\t" // divide prev_row bytes by 2 - "pand %%mm4, %%mm1 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm0 \n\t" // add (Prev_row/2) to Avg for - // each byte - // add 1st active group (Raw(x-bpp)/2) to average with _LBCarry - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm7, %%mm2 \n\t" // leave only Active Group 1 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to Avg - // for each Active - // byte - // add 2nd active group (Raw(x-bpp)/2) to average with _LBCarry - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - "addl $8, %%ecx \n\t" - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active - // byte - "cmpl _MMXLength, %%ecx \n\t" - // now ready to write back to memory - "movq %%mm0, -8(%%edi,%%ecx,) \n\t" - // prep Raw(x-bpp) for next loop - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "jb avg_4lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 4,6 bpp - - case 2: - { - _ActiveMask.use = 0x000000000000ffffLL; - _ShiftBpp.use = 16; // == 2 * 8 - _ShiftRem.use = 48; // == 64 - 16 - - __asm__ __volatile__ ( - // load _ActiveMask - "movq _ActiveMask, %%mm7 \n\t" - // re-init address pointers and offset - "movl _dif, %%ecx \n\t" // ecx: x = offset to alignment - // boundary - "movq _LBCarryMask, %%mm5 \n\t" -// preload "movl row, %%edi \n\t" // edi: Avg(x) - "movq _HBClearMask, %%mm4 \n\t" -// preload "movl prev_row, %%esi \n\t" // esi: Prior(x) - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm2 \n\t" // load previous aligned 8 bytes - // (we correct pos. in loop below) - "avg_2lp: \n\t" - "movq (%%edi,%%ecx,), %%mm0 \n\t" - "psrlq _ShiftRem, %%mm2 \n\t" // shift data to pos. correctly - "movq (%%esi,%%ecx,), %%mm1 \n\t" // (GRR BUGFIX: was psllq) - // add (Prev_row/2) to average - "movq %%mm5, %%mm3 \n\t" - "pand %%mm1, %%mm3 \n\t" // get lsb for each prev_row byte - "psrlq $1, %%mm1 \n\t" // divide prev_row bytes by 2 - "pand %%mm4, %%mm1 \n\t" // clear invalid bit 7 of each - // byte - "movq %%mm7, %%mm6 \n\t" - "paddb %%mm1, %%mm0 \n\t" // add (Prev_row/2) to Avg for - // each byte - - // add 1st active group (Raw(x-bpp)/2) to average with _LBCarry - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid - // for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 1 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to Avg - // for each Active byte - - // add 2nd active group (Raw(x-bpp)/2) to average with _LBCarry - "psllq _ShiftBpp, %%mm6 \n\t" // shift the mm6 mask to cover - // bytes 2 & 3 - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid - // for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active byte - - // add 3rd active group (Raw(x-bpp)/2) to average with _LBCarry - "psllq _ShiftBpp, %%mm6 \n\t" // shift the mm6 mask to cover - // bytes 4 & 5 - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both lsb's were == 1 - // (only valid for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active byte - - // add 4th active group (Raw(x-bpp)/2) to average with _LBCarry - "psllq _ShiftBpp, %%mm6 \n\t" // shift the mm6 mask to cover - // bytes 6 & 7 - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "psllq _ShiftBpp, %%mm2 \n\t" // shift data to pos. correctly - "addl $8, %%ecx \n\t" - "movq %%mm3, %%mm1 \n\t" // now use mm1 for getting - // LBCarrys - "pand %%mm2, %%mm1 \n\t" // get LBCarrys for each byte - // where both - // lsb's were == 1 (only valid - // for active group) - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm2 \n\t" // add LBCarrys to (Raw(x-bpp)/2) - // for each byte - "pand %%mm6, %%mm2 \n\t" // leave only Active Group 2 - // bytes to add to Avg - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) + LBCarrys to - // Avg for each Active byte - - "cmpl _MMXLength, %%ecx \n\t" - // now ready to write back to memory - "movq %%mm0, -8(%%edi,%%ecx,) \n\t" - // prep Raw(x-bpp) for next loop - "movq %%mm0, %%mm2 \n\t" // mov updated Raws to mm2 - "jb avg_2lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 2 bpp - - case 1: - { - __asm__ __volatile__ ( - // re-init address pointers and offset -#ifdef __PIC__ - "pushl %%ebx \n\t" // save Global Offset Table index -#endif - "movl _dif, %%ebx \n\t" // ebx: x = offset to alignment - // boundary -// preload "movl row, %%edi \n\t" // edi: Avg(x) - "cmpl _FullLength, %%ebx \n\t" // test if offset at end of array - "jnb avg_1end \n\t" - // do Paeth decode for remaining bytes -// preload "movl prev_row, %%esi \n\t" // esi: Prior(x) - "movl %%edi, %%edx \n\t" -// preload "subl bpp, %%edx \n\t" // (bpp is preloaded into ecx) - "subl %%ecx, %%edx \n\t" // edx: Raw(x-bpp) - "xorl %%ecx, %%ecx \n\t" // zero ecx before using cl & cx - // in loop below - "avg_1lp: \n\t" - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - "xorl %%eax, %%eax \n\t" - "movb (%%esi,%%ebx,), %%cl \n\t" // load cl with Prior(x) - "movb (%%edx,%%ebx,), %%al \n\t" // load al with Raw(x-bpp) - "addw %%cx, %%ax \n\t" - "incl %%ebx \n\t" - "shrw %%ax \n\t" // divide by 2 - "addb -1(%%edi,%%ebx,), %%al \n\t" // add Avg(x); -1 to offset - // inc ebx - "cmpl _FullLength, %%ebx \n\t" // check if at end of array - "movb %%al, -1(%%edi,%%ebx,) \n\t" // write back Raw(x); - // mov does not affect flags; -1 to offset inc ebx - "jb avg_1lp \n\t" - - "avg_1end: \n\t" -#ifdef __PIC__ - "popl %%ebx \n\t" // Global Offset Table index -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list -#ifndef __PIC__ - , "%ebx" -#endif - ); - } - return; // end 1 bpp - - case 8: - { - __asm__ __volatile__ ( - // re-init address pointers and offset - "movl _dif, %%ecx \n\t" // ecx: x == offset to alignment - "movq _LBCarryMask, %%mm5 \n\t" // boundary -// preload "movl row, %%edi \n\t" // edi: Avg(x) - "movq _HBClearMask, %%mm4 \n\t" -// preload "movl prev_row, %%esi \n\t" // esi: Prior(x) - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm2 \n\t" // load previous aligned 8 bytes - // (NO NEED to correct pos. in loop below) - - "avg_8lp: \n\t" - "movq (%%edi,%%ecx,), %%mm0 \n\t" - "movq %%mm5, %%mm3 \n\t" - "movq (%%esi,%%ecx,), %%mm1 \n\t" - "addl $8, %%ecx \n\t" - "pand %%mm1, %%mm3 \n\t" // get lsb for each prev_row byte - "psrlq $1, %%mm1 \n\t" // divide prev_row bytes by 2 - "pand %%mm2, %%mm3 \n\t" // get LBCarrys for each byte - // where both lsb's were == 1 - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm1 \n\t" // clear invalid bit 7, each byte - "paddb %%mm3, %%mm0 \n\t" // add LBCarrys to Avg, each byte - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7, each byte - "paddb %%mm1, %%mm0 \n\t" // add (Prev_row/2) to Avg, each - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) to Avg for each - "cmpl _MMXLength, %%ecx \n\t" - "movq %%mm0, -8(%%edi,%%ecx,) \n\t" - "movq %%mm0, %%mm2 \n\t" // reuse as Raw(x-bpp) - "jb avg_8lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm5 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2" - , "%mm3", "%mm4", "%mm5" -#endif - ); - } - break; // end 8 bpp - - default: // bpp greater than 8 (!= 1,2,3,4,[5],6,[7],8) - { - -#ifdef PNG_DEBUG - // GRR: PRINT ERROR HERE: SHOULD NEVER BE REACHED - png_debug(1, - "Internal logic error in pnggccrd (png_read_filter_row_mmx_avg())\n"); -#endif - -#if 0 - __asm__ __volatile__ ( - "movq _LBCarryMask, %%mm5 \n\t" - // re-init address pointers and offset - "movl _dif, %%ebx \n\t" // ebx: x = offset to - // alignment boundary - "movl row, %%edi \n\t" // edi: Avg(x) - "movq _HBClearMask, %%mm4 \n\t" - "movl %%edi, %%edx \n\t" - "movl prev_row, %%esi \n\t" // esi: Prior(x) - "subl bpp, %%edx \n\t" // edx: Raw(x-bpp) - "avg_Alp: \n\t" - "movq (%%edi,%%ebx,), %%mm0 \n\t" - "movq %%mm5, %%mm3 \n\t" - "movq (%%esi,%%ebx,), %%mm1 \n\t" - "pand %%mm1, %%mm3 \n\t" // get lsb for each prev_row byte - "movq (%%edx,%%ebx,), %%mm2 \n\t" - "psrlq $1, %%mm1 \n\t" // divide prev_row bytes by 2 - "pand %%mm2, %%mm3 \n\t" // get LBCarrys for each byte - // where both lsb's were == 1 - "psrlq $1, %%mm2 \n\t" // divide raw bytes by 2 - "pand %%mm4, %%mm1 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm3, %%mm0 \n\t" // add LBCarrys to Avg for each - // byte - "pand %%mm4, %%mm2 \n\t" // clear invalid bit 7 of each - // byte - "paddb %%mm1, %%mm0 \n\t" // add (Prev_row/2) to Avg for - // each byte - "addl $8, %%ebx \n\t" - "paddb %%mm2, %%mm0 \n\t" // add (Raw/2) to Avg for each - // byte - "cmpl _MMXLength, %%ebx \n\t" - "movq %%mm0, -8(%%edi,%%ebx,) \n\t" - "jb avg_Alp \n\t" - - : // FIXASM: output regs/vars go here, e.g.: "=m" (memory_var) - - : // FIXASM: input regs, e.g.: "c" (count), "S" (src), "D" (dest) - - : "%ebx", "%edx", "%edi", "%esi" // CHECKASM: clobber list - ); -#endif /* 0 - NEVER REACHED */ - } - break; - - } // end switch (bpp) - - __asm__ __volatile__ ( - // MMX acceleration complete; now do clean-up - // check if any remaining bytes left to decode -#ifdef __PIC__ - "pushl %%ebx \n\t" // save index to Global Offset Table -#endif - "movl _MMXLength, %%ebx \n\t" // ebx: x == offset bytes after MMX -//pre "movl row, %%edi \n\t" // edi: Avg(x) - "cmpl _FullLength, %%ebx \n\t" // test if offset at end of array - "jnb avg_end \n\t" - - // do Avg decode for remaining bytes -//pre "movl prev_row, %%esi \n\t" // esi: Prior(x) - "movl %%edi, %%edx \n\t" -//pre "subl bpp, %%edx \n\t" // (bpp is preloaded into ecx) - "subl %%ecx, %%edx \n\t" // edx: Raw(x-bpp) - "xorl %%ecx, %%ecx \n\t" // zero ecx before using cl & cx below - - "avg_lp2: \n\t" - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - "xorl %%eax, %%eax \n\t" - "movb (%%esi,%%ebx,), %%cl \n\t" // load cl with Prior(x) - "movb (%%edx,%%ebx,), %%al \n\t" // load al with Raw(x-bpp) - "addw %%cx, %%ax \n\t" - "incl %%ebx \n\t" - "shrw %%ax \n\t" // divide by 2 - "addb -1(%%edi,%%ebx,), %%al \n\t" // add Avg(x); -1 to offset inc ebx - "cmpl _FullLength, %%ebx \n\t" // check if at end of array - "movb %%al, -1(%%edi,%%ebx,) \n\t" // write back Raw(x) [mov does not - "jb avg_lp2 \n\t" // affect flags; -1 to offset inc ebx] - - "avg_end: \n\t" - "EMMS \n\t" // end MMX; prep for poss. FP instrs. -#ifdef __PIC__ - "popl %%ebx \n\t" // restore index to Global Offset Table -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list -#ifndef __PIC__ - , "%ebx" -#endif - ); - -} /* end png_read_filter_row_mmx_avg() */ -#endif - - - -#ifdef PNG_THREAD_UNSAFE_OK -//===========================================================================// -// // -// P N G _ R E A D _ F I L T E R _ R O W _ M M X _ P A E T H // -// // -//===========================================================================// - -// Optimized code for PNG Paeth filter decoder - -static void /* PRIVATE */ -png_read_filter_row_mmx_paeth(png_row_infop row_info, png_bytep row, - png_bytep prev_row) -{ - int bpp; - int dummy_value_c; // fix 'forbidden register 2 (cx) was spilled' error - int dummy_value_S; - int dummy_value_D; - - bpp = (row_info->pixel_depth + 7) >> 3; // Get # bytes per pixel - _FullLength = row_info->rowbytes; // # of bytes to filter - - __asm__ __volatile__ ( -#ifdef __PIC__ - "pushl %%ebx \n\t" // save index to Global Offset Table -#endif - "xorl %%ebx, %%ebx \n\t" // ebx: x offset -//pre "movl row, %%edi \n\t" - "xorl %%edx, %%edx \n\t" // edx: x-bpp offset -//pre "movl prev_row, %%esi \n\t" - "xorl %%eax, %%eax \n\t" - - // Compute the Raw value for the first bpp bytes - // Note: the formula works out to be always - // Paeth(x) = Raw(x) + Prior(x) where x < bpp - "paeth_rlp: \n\t" - "movb (%%edi,%%ebx,), %%al \n\t" - "addb (%%esi,%%ebx,), %%al \n\t" - "incl %%ebx \n\t" -//pre "cmpl bpp, %%ebx \n\t" (bpp is preloaded into ecx) - "cmpl %%ecx, %%ebx \n\t" - "movb %%al, -1(%%edi,%%ebx,) \n\t" - "jb paeth_rlp \n\t" - // get # of bytes to alignment - "movl %%edi, _dif \n\t" // take start of row - "addl %%ebx, _dif \n\t" // add bpp - "xorl %%ecx, %%ecx \n\t" - "addl $0xf, _dif \n\t" // add 7 + 8 to incr past alignment - // boundary - "andl $0xfffffff8, _dif \n\t" // mask to alignment boundary - "subl %%edi, _dif \n\t" // subtract from start ==> value ebx - // at alignment - "jz paeth_go \n\t" - // fix alignment - - "paeth_lp1: \n\t" - "xorl %%eax, %%eax \n\t" - // pav = p - a = (a + b - c) - a = b - c - "movb (%%esi,%%ebx,), %%al \n\t" // load Prior(x) into al - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, _patemp \n\t" // Save pav for later use - "xorl %%eax, %%eax \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movb (%%edi,%%edx,), %%al \n\t" // load Raw(x-bpp) into al - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, %%ecx \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "addl _patemp, %%eax \n\t" // pcv = pav + pbv - // pc = abs(pcv) - "testl $0x80000000, %%eax \n\t" - "jz paeth_pca \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_pca: \n\t" - "movl %%eax, _pctemp \n\t" // save pc for later use - // pb = abs(pbv) - "testl $0x80000000, %%ecx \n\t" - "jz paeth_pba \n\t" - "negl %%ecx \n\t" // reverse sign of neg values - - "paeth_pba: \n\t" - "movl %%ecx, _pbtemp \n\t" // save pb for later use - // pa = abs(pav) - "movl _patemp, %%eax \n\t" - "testl $0x80000000, %%eax \n\t" - "jz paeth_paa \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_paa: \n\t" - "movl %%eax, _patemp \n\t" // save pa for later use - // test if pa <= pb - "cmpl %%ecx, %%eax \n\t" - "jna paeth_abb \n\t" - // pa > pb; now test if pb <= pc - "cmpl _pctemp, %%ecx \n\t" - "jna paeth_bbc \n\t" - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_paeth \n\t" - - "paeth_bbc: \n\t" - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - "movb (%%esi,%%ebx,), %%cl \n\t" // load Prior(x) into cl - "jmp paeth_paeth \n\t" - - "paeth_abb: \n\t" - // pa <= pb; now test if pa <= pc - "cmpl _pctemp, %%eax \n\t" - "jna paeth_abc \n\t" - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_paeth \n\t" - - "paeth_abc: \n\t" - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - "movb (%%edi,%%edx,), %%cl \n\t" // load Raw(x-bpp) into cl - - "paeth_paeth: \n\t" - "incl %%ebx \n\t" - "incl %%edx \n\t" - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - "addb %%cl, -1(%%edi,%%ebx,) \n\t" - "cmpl _dif, %%ebx \n\t" - "jb paeth_lp1 \n\t" - - "paeth_go: \n\t" - "movl _FullLength, %%ecx \n\t" - "movl %%ecx, %%eax \n\t" - "subl %%ebx, %%eax \n\t" // subtract alignment fix - "andl $0x00000007, %%eax \n\t" // calc bytes over mult of 8 - "subl %%eax, %%ecx \n\t" // drop over bytes from original length - "movl %%ecx, _MMXLength \n\t" -#ifdef __PIC__ - "popl %%ebx \n\t" // restore index to Global Offset Table -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list -#ifndef __PIC__ - , "%ebx" -#endif - ); - - // now do the math for the rest of the row - switch (bpp) - { - case 3: - { - _ActiveMask.use = 0x0000000000ffffffLL; - _ActiveMaskEnd.use = 0xffff000000000000LL; - _ShiftBpp.use = 24; // == bpp(3) * 8 - _ShiftRem.use = 40; // == 64 - 24 - - __asm__ __volatile__ ( - "movl _dif, %%ecx \n\t" -// preload "movl row, %%edi \n\t" -// preload "movl prev_row, %%esi \n\t" - "pxor %%mm0, %%mm0 \n\t" - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" - "paeth_3lp: \n\t" - "psrlq _ShiftRem, %%mm1 \n\t" // shift last 3 bytes to 1st - // 3 bytes - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "punpcklbw %%mm0, %%mm1 \n\t" // unpack High bytes of a - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // prep c=Prior(x-bpp) bytes - "punpcklbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - "psrlq _ShiftRem, %%mm3 \n\t" // shift last 3 bytes to 1st - // 3 bytes - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "punpcklbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "packuswb %%mm1, %%mm7 \n\t" - "movq (%%esi,%%ecx,), %%mm3 \n\t" // load c=Prior(x-bpp) - "pand _ActiveMask, %%mm7 \n\t" - "movq %%mm3, %%mm2 \n\t" // load b=Prior(x) step 1 - "paddb (%%edi,%%ecx,), %%mm7 \n\t" // add Paeth predictor with Raw(x) - "punpcklbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - "movq %%mm7, (%%edi,%%ecx,) \n\t" // write back updated value - "movq %%mm7, %%mm1 \n\t" // now mm1 will be used as - // Raw(x-bpp) - // now do Paeth for 2nd set of bytes (3-5) - "psrlq _ShiftBpp, %%mm2 \n\t" // load b=Prior(x) step 2 - "punpcklbw %%mm0, %%mm1 \n\t" // unpack High bytes of a - "pxor %%mm7, %%mm7 \n\t" - "punpcklbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "psubw %%mm3, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = - // pav + pbv = pbv + pav - "movq %%mm5, %%mm6 \n\t" - "paddw %%mm4, %%mm6 \n\t" - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm5, %%mm0 \n\t" // create mask pbv bytes < 0 - "pcmpgtw %%mm4, %%mm7 \n\t" // create mask pav bytes < 0 - "pand %%mm5, %%mm0 \n\t" // only pbv bytes < 0 in mm0 - "pand %%mm4, %%mm7 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm0, %%mm5 \n\t" - "psubw %%mm7, %%mm4 \n\t" - "psubw %%mm0, %%mm5 \n\t" - "psubw %%mm7, %%mm4 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "pxor %%mm1, %%mm1 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "packuswb %%mm1, %%mm7 \n\t" - "movq %%mm2, %%mm3 \n\t" // load c=Prior(x-bpp) step 1 - "pand _ActiveMask, %%mm7 \n\t" - "punpckhbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - "psllq _ShiftBpp, %%mm7 \n\t" // shift bytes to 2nd group of - // 3 bytes - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "paddb (%%edi,%%ecx,), %%mm7 \n\t" // add Paeth predictor with Raw(x) - "psllq _ShiftBpp, %%mm3 \n\t" // load c=Prior(x-bpp) step 2 - "movq %%mm7, (%%edi,%%ecx,) \n\t" // write back updated value - "movq %%mm7, %%mm1 \n\t" - "punpckhbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - "psllq _ShiftBpp, %%mm1 \n\t" // shift bytes - // now mm1 will be used as Raw(x-bpp) - // now do Paeth for 3rd, and final, set of bytes (6-7) - "pxor %%mm7, %%mm7 \n\t" - "punpckhbw %%mm0, %%mm1 \n\t" // unpack High bytes of a - "psubw %%mm3, %%mm4 \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "paddw %%mm5, %%mm6 \n\t" - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm1, %%mm1 \n\t" - "packuswb %%mm7, %%mm1 \n\t" - // step ecx to next set of 8 bytes and repeat loop til done - "addl $8, %%ecx \n\t" - "pand _ActiveMaskEnd, %%mm1 \n\t" - "paddb -8(%%edi,%%ecx,), %%mm1 \n\t" // add Paeth predictor with - // Raw(x) - - "cmpl _MMXLength, %%ecx \n\t" - "pxor %%mm0, %%mm0 \n\t" // pxor does not affect flags - "movq %%mm1, -8(%%edi,%%ecx,) \n\t" // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - // mm3 ready to be used as Prior(x-bpp) next loop - "jb paeth_3lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 3 bpp - - case 6: - //case 7: // GRR BOGUS - //case 5: // GRR BOGUS - { - _ActiveMask.use = 0x00000000ffffffffLL; - _ActiveMask2.use = 0xffffffff00000000LL; - _ShiftBpp.use = bpp << 3; // == bpp * 8 - _ShiftRem.use = 64 - _ShiftBpp.use; - - __asm__ __volatile__ ( - "movl _dif, %%ecx \n\t" -// preload "movl row, %%edi \n\t" -// preload "movl prev_row, %%esi \n\t" - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" - "pxor %%mm0, %%mm0 \n\t" - - "paeth_6lp: \n\t" - // must shift to position Raw(x-bpp) data - "psrlq _ShiftRem, %%mm1 \n\t" - // do first set of 4 bytes - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // read c=Prior(x-bpp) bytes - "punpcklbw %%mm0, %%mm1 \n\t" // unpack Low bytes of a - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "punpcklbw %%mm0, %%mm2 \n\t" // unpack Low bytes of b - // must shift to position Prior(x-bpp) data - "psrlq _ShiftRem, %%mm3 \n\t" - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "punpcklbw %%mm0, %%mm3 \n\t" // unpack Low bytes of c - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "packuswb %%mm1, %%mm7 \n\t" - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // load c=Prior(x-bpp) - "pand _ActiveMask, %%mm7 \n\t" - "psrlq _ShiftRem, %%mm3 \n\t" - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) step 1 - "paddb (%%edi,%%ecx,), %%mm7 \n\t" // add Paeth predictor and Raw(x) - "movq %%mm2, %%mm6 \n\t" - "movq %%mm7, (%%edi,%%ecx,) \n\t" // write back updated value - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" - "psllq _ShiftBpp, %%mm6 \n\t" - "movq %%mm7, %%mm5 \n\t" - "psrlq _ShiftRem, %%mm1 \n\t" - "por %%mm6, %%mm3 \n\t" - "psllq _ShiftBpp, %%mm5 \n\t" - "punpckhbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - "por %%mm5, %%mm1 \n\t" - // do second set of 4 bytes - "punpckhbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - "punpckhbw %%mm0, %%mm1 \n\t" // unpack High bytes of a - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "pxor %%mm1, %%mm1 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - // step ecx to next set of 8 bytes and repeat loop til done - "addl $8, %%ecx \n\t" - "packuswb %%mm7, %%mm1 \n\t" - "paddb -8(%%edi,%%ecx,), %%mm1 \n\t" // add Paeth predictor with Raw(x) - "cmpl _MMXLength, %%ecx \n\t" - "movq %%mm1, -8(%%edi,%%ecx,) \n\t" // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - "jb paeth_6lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 6 bpp - - case 4: - { - _ActiveMask.use = 0x00000000ffffffffLL; - - __asm__ __volatile__ ( - "movl _dif, %%ecx \n\t" -// preload "movl row, %%edi \n\t" -// preload "movl prev_row, %%esi \n\t" - "pxor %%mm0, %%mm0 \n\t" - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" // only time should need to read - // a=Raw(x-bpp) bytes - "paeth_4lp: \n\t" - // do first set of 4 bytes - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // read c=Prior(x-bpp) bytes - "punpckhbw %%mm0, %%mm1 \n\t" // unpack Low bytes of a - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "punpcklbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "punpckhbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "packuswb %%mm1, %%mm7 \n\t" - "movq (%%esi,%%ecx,), %%mm3 \n\t" // load c=Prior(x-bpp) - "pand _ActiveMask, %%mm7 \n\t" - "movq %%mm3, %%mm2 \n\t" // load b=Prior(x) step 1 - "paddb (%%edi,%%ecx,), %%mm7 \n\t" // add Paeth predictor with Raw(x) - "punpcklbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - "movq %%mm7, (%%edi,%%ecx,) \n\t" // write back updated value - "movq %%mm7, %%mm1 \n\t" // now mm1 will be used as Raw(x-bpp) - // do second set of 4 bytes - "punpckhbw %%mm0, %%mm2 \n\t" // unpack Low bytes of b - "punpcklbw %%mm0, %%mm1 \n\t" // unpack Low bytes of a - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "pxor %%mm1, %%mm1 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - // step ecx to next set of 8 bytes and repeat loop til done - "addl $8, %%ecx \n\t" - "packuswb %%mm7, %%mm1 \n\t" - "paddb -8(%%edi,%%ecx,), %%mm1 \n\t" // add predictor with Raw(x) - "cmpl _MMXLength, %%ecx \n\t" - "movq %%mm1, -8(%%edi,%%ecx,) \n\t" // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - "jb paeth_4lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 4 bpp - - case 8: // bpp == 8 - { - _ActiveMask.use = 0x00000000ffffffffLL; - - __asm__ __volatile__ ( - "movl _dif, %%ecx \n\t" -// preload "movl row, %%edi \n\t" -// preload "movl prev_row, %%esi \n\t" - "pxor %%mm0, %%mm0 \n\t" - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" // only time should need to read - // a=Raw(x-bpp) bytes - "paeth_8lp: \n\t" - // do first set of 4 bytes - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // read c=Prior(x-bpp) bytes - "punpcklbw %%mm0, %%mm1 \n\t" // unpack Low bytes of a - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "punpcklbw %%mm0, %%mm2 \n\t" // unpack Low bytes of b - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - "punpcklbw %%mm0, %%mm3 \n\t" // unpack Low bytes of c - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "packuswb %%mm1, %%mm7 \n\t" - "movq -8(%%esi,%%ecx,), %%mm3 \n\t" // read c=Prior(x-bpp) bytes - "pand _ActiveMask, %%mm7 \n\t" - "movq (%%esi,%%ecx,), %%mm2 \n\t" // load b=Prior(x) - "paddb (%%edi,%%ecx,), %%mm7 \n\t" // add Paeth predictor with Raw(x) - "punpckhbw %%mm0, %%mm3 \n\t" // unpack High bytes of c - "movq %%mm7, (%%edi,%%ecx,) \n\t" // write back updated value - "movq -8(%%edi,%%ecx,), %%mm1 \n\t" // read a=Raw(x-bpp) bytes - - // do second set of 4 bytes - "punpckhbw %%mm0, %%mm2 \n\t" // unpack High bytes of b - "punpckhbw %%mm0, %%mm1 \n\t" // unpack High bytes of a - // pav = p - a = (a + b - c) - a = b - c - "movq %%mm2, %%mm4 \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movq %%mm1, %%mm5 \n\t" - "psubw %%mm3, %%mm4 \n\t" - "pxor %%mm7, %%mm7 \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "movq %%mm4, %%mm6 \n\t" - "psubw %%mm3, %%mm5 \n\t" - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - "pcmpgtw %%mm4, %%mm0 \n\t" // create mask pav bytes < 0 - "paddw %%mm5, %%mm6 \n\t" - "pand %%mm4, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "pcmpgtw %%mm5, %%mm7 \n\t" // create mask pbv bytes < 0 - "psubw %%mm0, %%mm4 \n\t" - "pand %%mm5, %%mm7 \n\t" // only pbv bytes < 0 in mm0 - "psubw %%mm0, %%mm4 \n\t" - "psubw %%mm7, %%mm5 \n\t" - "pxor %%mm0, %%mm0 \n\t" - "pcmpgtw %%mm6, %%mm0 \n\t" // create mask pcv bytes < 0 - "pand %%mm6, %%mm0 \n\t" // only pav bytes < 0 in mm7 - "psubw %%mm7, %%mm5 \n\t" - "psubw %%mm0, %%mm6 \n\t" - // test pa <= pb - "movq %%mm4, %%mm7 \n\t" - "psubw %%mm0, %%mm6 \n\t" - "pcmpgtw %%mm5, %%mm7 \n\t" // pa > pb? - "movq %%mm7, %%mm0 \n\t" - // use mm7 mask to merge pa & pb - "pand %%mm7, %%mm5 \n\t" - // use mm0 mask copy to merge a & b - "pand %%mm0, %%mm2 \n\t" - "pandn %%mm4, %%mm7 \n\t" - "pandn %%mm1, %%mm0 \n\t" - "paddw %%mm5, %%mm7 \n\t" - "paddw %%mm2, %%mm0 \n\t" - // test ((pa <= pb)? pa:pb) <= pc - "pcmpgtw %%mm6, %%mm7 \n\t" // pab > pc? - "pxor %%mm1, %%mm1 \n\t" - "pand %%mm7, %%mm3 \n\t" - "pandn %%mm0, %%mm7 \n\t" - "pxor %%mm1, %%mm1 \n\t" - "paddw %%mm3, %%mm7 \n\t" - "pxor %%mm0, %%mm0 \n\t" - // step ecx to next set of 8 bytes and repeat loop til done - "addl $8, %%ecx \n\t" - "packuswb %%mm7, %%mm1 \n\t" - "paddb -8(%%edi,%%ecx,), %%mm1 \n\t" // add Paeth predictor with Raw(x) - "cmpl _MMXLength, %%ecx \n\t" - "movq %%mm1, -8(%%edi,%%ecx,) \n\t" // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - "jb paeth_8lp \n\t" - - : "=S" (dummy_value_S), // output regs (dummy) - "=D" (dummy_value_D) - - : "0" (prev_row), // esi // input regs - "1" (row) // edi - - : "%ecx" // clobber list -#if 0 /* %mm0, ..., %mm7 not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; // end 8 bpp - - case 1: // bpp = 1 - case 2: // bpp = 2 - default: // bpp > 8 - { - __asm__ __volatile__ ( -#ifdef __PIC__ - "pushl %%ebx \n\t" // save Global Offset Table index -#endif - "movl _dif, %%ebx \n\t" - "cmpl _FullLength, %%ebx \n\t" - "jnb paeth_dend \n\t" - -// preload "movl row, %%edi \n\t" -// preload "movl prev_row, %%esi \n\t" - // do Paeth decode for remaining bytes - "movl %%ebx, %%edx \n\t" -// preload "subl bpp, %%edx \n\t" // (bpp is preloaded into ecx) - "subl %%ecx, %%edx \n\t" // edx = ebx - bpp - "xorl %%ecx, %%ecx \n\t" // zero ecx before using cl & cx - - "paeth_dlp: \n\t" - "xorl %%eax, %%eax \n\t" - // pav = p - a = (a + b - c) - a = b - c - "movb (%%esi,%%ebx,), %%al \n\t" // load Prior(x) into al - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, _patemp \n\t" // Save pav for later use - "xorl %%eax, %%eax \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movb (%%edi,%%edx,), %%al \n\t" // load Raw(x-bpp) into al - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, %%ecx \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "addl _patemp, %%eax \n\t" // pcv = pav + pbv - // pc = abs(pcv) - "testl $0x80000000, %%eax \n\t" - "jz paeth_dpca \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_dpca: \n\t" - "movl %%eax, _pctemp \n\t" // save pc for later use - // pb = abs(pbv) - "testl $0x80000000, %%ecx \n\t" - "jz paeth_dpba \n\t" - "negl %%ecx \n\t" // reverse sign of neg values - - "paeth_dpba: \n\t" - "movl %%ecx, _pbtemp \n\t" // save pb for later use - // pa = abs(pav) - "movl _patemp, %%eax \n\t" - "testl $0x80000000, %%eax \n\t" - "jz paeth_dpaa \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_dpaa: \n\t" - "movl %%eax, _patemp \n\t" // save pa for later use - // test if pa <= pb - "cmpl %%ecx, %%eax \n\t" - "jna paeth_dabb \n\t" - // pa > pb; now test if pb <= pc - "cmpl _pctemp, %%ecx \n\t" - "jna paeth_dbbc \n\t" - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_dpaeth \n\t" - - "paeth_dbbc: \n\t" - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - "movb (%%esi,%%ebx,), %%cl \n\t" // load Prior(x) into cl - "jmp paeth_dpaeth \n\t" - - "paeth_dabb: \n\t" - // pa <= pb; now test if pa <= pc - "cmpl _pctemp, %%eax \n\t" - "jna paeth_dabc \n\t" - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_dpaeth \n\t" - - "paeth_dabc: \n\t" - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - "movb (%%edi,%%edx,), %%cl \n\t" // load Raw(x-bpp) into cl - - "paeth_dpaeth: \n\t" - "incl %%ebx \n\t" - "incl %%edx \n\t" - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - "addb %%cl, -1(%%edi,%%ebx,) \n\t" - "cmpl _FullLength, %%ebx \n\t" - "jb paeth_dlp \n\t" - - "paeth_dend: \n\t" -#ifdef __PIC__ - "popl %%ebx \n\t" // index to Global Offset Table -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list -#ifndef __PIC__ - , "%ebx" -#endif - ); - } - return; // No need to go further with this one - - } // end switch (bpp) - - __asm__ __volatile__ ( - // MMX acceleration complete; now do clean-up - // check if any remaining bytes left to decode -#ifdef __PIC__ - "pushl %%ebx \n\t" // save index to Global Offset Table -#endif - "movl _MMXLength, %%ebx \n\t" - "cmpl _FullLength, %%ebx \n\t" - "jnb paeth_end \n\t" -//pre "movl row, %%edi \n\t" -//pre "movl prev_row, %%esi \n\t" - // do Paeth decode for remaining bytes - "movl %%ebx, %%edx \n\t" -//pre "subl bpp, %%edx \n\t" // (bpp is preloaded into ecx) - "subl %%ecx, %%edx \n\t" // edx = ebx - bpp - "xorl %%ecx, %%ecx \n\t" // zero ecx before using cl & cx below - - "paeth_lp2: \n\t" - "xorl %%eax, %%eax \n\t" - // pav = p - a = (a + b - c) - a = b - c - "movb (%%esi,%%ebx,), %%al \n\t" // load Prior(x) into al - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, _patemp \n\t" // Save pav for later use - "xorl %%eax, %%eax \n\t" - // pbv = p - b = (a + b - c) - b = a - c - "movb (%%edi,%%edx,), %%al \n\t" // load Raw(x-bpp) into al - "subl %%ecx, %%eax \n\t" // subtract Prior(x-bpp) - "movl %%eax, %%ecx \n\t" - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - "addl _patemp, %%eax \n\t" // pcv = pav + pbv - // pc = abs(pcv) - "testl $0x80000000, %%eax \n\t" - "jz paeth_pca2 \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_pca2: \n\t" - "movl %%eax, _pctemp \n\t" // save pc for later use - // pb = abs(pbv) - "testl $0x80000000, %%ecx \n\t" - "jz paeth_pba2 \n\t" - "negl %%ecx \n\t" // reverse sign of neg values - - "paeth_pba2: \n\t" - "movl %%ecx, _pbtemp \n\t" // save pb for later use - // pa = abs(pav) - "movl _patemp, %%eax \n\t" - "testl $0x80000000, %%eax \n\t" - "jz paeth_paa2 \n\t" - "negl %%eax \n\t" // reverse sign of neg values - - "paeth_paa2: \n\t" - "movl %%eax, _patemp \n\t" // save pa for later use - // test if pa <= pb - "cmpl %%ecx, %%eax \n\t" - "jna paeth_abb2 \n\t" - // pa > pb; now test if pb <= pc - "cmpl _pctemp, %%ecx \n\t" - "jna paeth_bbc2 \n\t" - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_paeth2 \n\t" - - "paeth_bbc2: \n\t" - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - "movb (%%esi,%%ebx,), %%cl \n\t" // load Prior(x) into cl - "jmp paeth_paeth2 \n\t" - - "paeth_abb2: \n\t" - // pa <= pb; now test if pa <= pc - "cmpl _pctemp, %%eax \n\t" - "jna paeth_abc2 \n\t" - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - "movb (%%esi,%%edx,), %%cl \n\t" // load Prior(x-bpp) into cl - "jmp paeth_paeth2 \n\t" - - "paeth_abc2: \n\t" - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - "movb (%%edi,%%edx,), %%cl \n\t" // load Raw(x-bpp) into cl - - "paeth_paeth2: \n\t" - "incl %%ebx \n\t" - "incl %%edx \n\t" - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - "addb %%cl, -1(%%edi,%%ebx,) \n\t" - "cmpl _FullLength, %%ebx \n\t" - "jb paeth_lp2 \n\t" - - "paeth_end: \n\t" - "EMMS \n\t" // end MMX; prep for poss. FP instrs. -#ifdef __PIC__ - "popl %%ebx \n\t" // restore index to Global Offset Table -#endif - - : "=c" (dummy_value_c), // output regs (dummy) - "=S" (dummy_value_S), - "=D" (dummy_value_D) - - : "0" (bpp), // ecx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%edx" // clobber list (no input regs!) -#ifndef __PIC__ - , "%ebx" -#endif - ); - -} /* end png_read_filter_row_mmx_paeth() */ -#endif - - - - -#ifdef PNG_THREAD_UNSAFE_OK -//===========================================================================// -// // -// P N G _ R E A D _ F I L T E R _ R O W _ M M X _ S U B // -// // -//===========================================================================// - -// Optimized code for PNG Sub filter decoder - -static void /* PRIVATE */ -png_read_filter_row_mmx_sub(png_row_infop row_info, png_bytep row) -{ - int bpp; - int dummy_value_a; - int dummy_value_D; - - bpp = (row_info->pixel_depth + 7) >> 3; // calc number of bytes per pixel - _FullLength = row_info->rowbytes - bpp; // number of bytes to filter - - __asm__ __volatile__ ( -//pre "movl row, %%edi \n\t" - "movl %%edi, %%esi \n\t" // lp = row -//pre "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp -//irr "xorl %%eax, %%eax \n\t" - // get # of bytes to alignment - "movl %%edi, _dif \n\t" // take start of row - "addl $0xf, _dif \n\t" // add 7 + 8 to incr past - // alignment boundary - "xorl %%ecx, %%ecx \n\t" - "andl $0xfffffff8, _dif \n\t" // mask to alignment boundary - "subl %%edi, _dif \n\t" // subtract from start ==> value - "jz sub_go \n\t" // ecx at alignment - - "sub_lp1: \n\t" // fix alignment - "movb (%%esi,%%ecx,), %%al \n\t" - "addb %%al, (%%edi,%%ecx,) \n\t" - "incl %%ecx \n\t" - "cmpl _dif, %%ecx \n\t" - "jb sub_lp1 \n\t" - - "sub_go: \n\t" - "movl _FullLength, %%eax \n\t" - "movl %%eax, %%edx \n\t" - "subl %%ecx, %%edx \n\t" // subtract alignment fix - "andl $0x00000007, %%edx \n\t" // calc bytes over mult of 8 - "subl %%edx, %%eax \n\t" // drop over bytes from length - "movl %%eax, _MMXLength \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%esi", "%ecx", "%edx" // clobber list - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - - // now do the math for the rest of the row - switch (bpp) - { - case 3: - { - _ActiveMask.use = 0x0000ffffff000000LL; - _ShiftBpp.use = 24; // == 3 * 8 - _ShiftRem.use = 40; // == 64 - 24 - - __asm__ __volatile__ ( -// preload "movl row, %%edi \n\t" - "movq _ActiveMask, %%mm7 \n\t" // load _ActiveMask for 2nd - // active byte group - "movl %%edi, %%esi \n\t" // lp = row -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - "movq %%mm7, %%mm6 \n\t" - "movl _dif, %%edx \n\t" - "psllq _ShiftBpp, %%mm6 \n\t" // move mask in mm6 to cover - // 3rd active byte group - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%edx,), %%mm1 \n\t" - - "sub_3lp: \n\t" // shift data for adding first - "psrlq _ShiftRem, %%mm1 \n\t" // bpp bytes (no need for mask; - // shift clears inactive bytes) - // add 1st active group - "movq (%%edi,%%edx,), %%mm0 \n\t" - "paddb %%mm1, %%mm0 \n\t" - - // add 2nd active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "pand %%mm7, %%mm1 \n\t" // mask to use 2nd active group - "paddb %%mm1, %%mm0 \n\t" - - // add 3rd active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "pand %%mm6, %%mm1 \n\t" // mask to use 3rd active group - "addl $8, %%edx \n\t" - "paddb %%mm1, %%mm0 \n\t" - - "cmpl _MMXLength, %%edx \n\t" - "movq %%mm0, -8(%%edi,%%edx,) \n\t" // write updated Raws to array - "movq %%mm0, %%mm1 \n\t" // prep 1st add at top of loop - "jb sub_3lp \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm6", "%mm7" -#endif - ); - } - break; - - case 1: - { - __asm__ __volatile__ ( - "movl _dif, %%edx \n\t" -// preload "movl row, %%edi \n\t" - "cmpl _FullLength, %%edx \n\t" - "jnb sub_1end \n\t" - "movl %%edi, %%esi \n\t" // lp = row - "xorl %%eax, %%eax \n\t" -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - - "sub_1lp: \n\t" - "movb (%%esi,%%edx,), %%al \n\t" - "addb %%al, (%%edi,%%edx,) \n\t" - "incl %%edx \n\t" - "cmpl _FullLength, %%edx \n\t" - "jb sub_1lp \n\t" - - "sub_1end: \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list - ); - } - return; - - case 6: - case 4: - //case 7: // GRR BOGUS - //case 5: // GRR BOGUS - { - _ShiftBpp.use = bpp << 3; - _ShiftRem.use = 64 - _ShiftBpp.use; - - __asm__ __volatile__ ( -// preload "movl row, %%edi \n\t" - "movl _dif, %%edx \n\t" - "movl %%edi, %%esi \n\t" // lp = row -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%edx,), %%mm1 \n\t" - - "sub_4lp: \n\t" // shift data for adding first - "psrlq _ShiftRem, %%mm1 \n\t" // bpp bytes (no need for mask; - // shift clears inactive bytes) - "movq (%%edi,%%edx,), %%mm0 \n\t" - "paddb %%mm1, %%mm0 \n\t" - - // add 2nd active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "addl $8, %%edx \n\t" - "paddb %%mm1, %%mm0 \n\t" - - "cmpl _MMXLength, %%edx \n\t" - "movq %%mm0, -8(%%edi,%%edx,) \n\t" - "movq %%mm0, %%mm1 \n\t" // prep 1st add at top of loop - "jb sub_4lp \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1" -#endif - ); - } - break; - - case 2: - { - _ActiveMask.use = 0x00000000ffff0000LL; - _ShiftBpp.use = 16; // == 2 * 8 - _ShiftRem.use = 48; // == 64 - 16 - - __asm__ __volatile__ ( - "movq _ActiveMask, %%mm7 \n\t" // load _ActiveMask for 2nd - // active byte group - "movl _dif, %%edx \n\t" - "movq %%mm7, %%mm6 \n\t" -// preload "movl row, %%edi \n\t" - "psllq _ShiftBpp, %%mm6 \n\t" // move mask in mm6 to cover - // 3rd active byte group - "movl %%edi, %%esi \n\t" // lp = row - "movq %%mm6, %%mm5 \n\t" -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - "psllq _ShiftBpp, %%mm5 \n\t" // move mask in mm5 to cover - // 4th active byte group - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%edx,), %%mm1 \n\t" - - "sub_2lp: \n\t" // shift data for adding first - "psrlq _ShiftRem, %%mm1 \n\t" // bpp bytes (no need for mask; - // shift clears inactive bytes) - // add 1st active group - "movq (%%edi,%%edx,), %%mm0 \n\t" - "paddb %%mm1, %%mm0 \n\t" - - // add 2nd active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "pand %%mm7, %%mm1 \n\t" // mask to use 2nd active group - "paddb %%mm1, %%mm0 \n\t" - - // add 3rd active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "pand %%mm6, %%mm1 \n\t" // mask to use 3rd active group - "paddb %%mm1, %%mm0 \n\t" - - // add 4th active group - "movq %%mm0, %%mm1 \n\t" // mov updated Raws to mm1 - "psllq _ShiftBpp, %%mm1 \n\t" // shift data to pos. correctly - "pand %%mm5, %%mm1 \n\t" // mask to use 4th active group - "addl $8, %%edx \n\t" - "paddb %%mm1, %%mm0 \n\t" - "cmpl _MMXLength, %%edx \n\t" - "movq %%mm0, -8(%%edi,%%edx,) \n\t" // write updated Raws to array - "movq %%mm0, %%mm1 \n\t" // prep 1st add at top of loop - "jb sub_2lp \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; - - case 8: - { - __asm__ __volatile__ ( -// preload "movl row, %%edi \n\t" - "movl _dif, %%edx \n\t" - "movl %%edi, %%esi \n\t" // lp = row -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - "movl _MMXLength, %%ecx \n\t" - - // prime the pump: load the first Raw(x-bpp) data set - "movq -8(%%edi,%%edx,), %%mm7 \n\t" - "andl $0x0000003f, %%ecx \n\t" // calc bytes over mult of 64 - - "sub_8lp: \n\t" - "movq (%%edi,%%edx,), %%mm0 \n\t" // load Sub(x) for 1st 8 bytes - "paddb %%mm7, %%mm0 \n\t" - "movq 8(%%edi,%%edx,), %%mm1 \n\t" // load Sub(x) for 2nd 8 bytes - "movq %%mm0, (%%edi,%%edx,) \n\t" // write Raw(x) for 1st 8 bytes - - // Now mm0 will be used as Raw(x-bpp) for the 2nd group of 8 bytes. - // This will be repeated for each group of 8 bytes with the 8th - // group being used as the Raw(x-bpp) for the 1st group of the - // next loop. - - "paddb %%mm0, %%mm1 \n\t" - "movq 16(%%edi,%%edx,), %%mm2 \n\t" // load Sub(x) for 3rd 8 bytes - "movq %%mm1, 8(%%edi,%%edx,) \n\t" // write Raw(x) for 2nd 8 bytes - "paddb %%mm1, %%mm2 \n\t" - "movq 24(%%edi,%%edx,), %%mm3 \n\t" // load Sub(x) for 4th 8 bytes - "movq %%mm2, 16(%%edi,%%edx,) \n\t" // write Raw(x) for 3rd 8 bytes - "paddb %%mm2, %%mm3 \n\t" - "movq 32(%%edi,%%edx,), %%mm4 \n\t" // load Sub(x) for 5th 8 bytes - "movq %%mm3, 24(%%edi,%%edx,) \n\t" // write Raw(x) for 4th 8 bytes - "paddb %%mm3, %%mm4 \n\t" - "movq 40(%%edi,%%edx,), %%mm5 \n\t" // load Sub(x) for 6th 8 bytes - "movq %%mm4, 32(%%edi,%%edx,) \n\t" // write Raw(x) for 5th 8 bytes - "paddb %%mm4, %%mm5 \n\t" - "movq 48(%%edi,%%edx,), %%mm6 \n\t" // load Sub(x) for 7th 8 bytes - "movq %%mm5, 40(%%edi,%%edx,) \n\t" // write Raw(x) for 6th 8 bytes - "paddb %%mm5, %%mm6 \n\t" - "movq 56(%%edi,%%edx,), %%mm7 \n\t" // load Sub(x) for 8th 8 bytes - "movq %%mm6, 48(%%edi,%%edx,) \n\t" // write Raw(x) for 7th 8 bytes - "addl $64, %%edx \n\t" - "paddb %%mm6, %%mm7 \n\t" - "cmpl %%ecx, %%edx \n\t" - "movq %%mm7, -8(%%edi,%%edx,) \n\t" // write Raw(x) for 8th 8 bytes - "jb sub_8lp \n\t" - - "cmpl _MMXLength, %%edx \n\t" - "jnb sub_8lt8 \n\t" - - "sub_8lpA: \n\t" - "movq (%%edi,%%edx,), %%mm0 \n\t" - "addl $8, %%edx \n\t" - "paddb %%mm7, %%mm0 \n\t" - "cmpl _MMXLength, %%edx \n\t" - "movq %%mm0, -8(%%edi,%%edx,) \n\t" // -8 to offset early addl edx - "movq %%mm0, %%mm7 \n\t" // move calculated Raw(x) data - // to mm1 to be new Raw(x-bpp) - // for next loop - "jb sub_8lpA \n\t" - - "sub_8lt8: \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%ecx", "%edx", "%esi" // clobber list -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3", "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - } - break; - - default: // bpp greater than 8 bytes GRR BOGUS - { - __asm__ __volatile__ ( - "movl _dif, %%edx \n\t" -// preload "movl row, %%edi \n\t" - "movl %%edi, %%esi \n\t" // lp = row -// preload "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - - "sub_Alp: \n\t" - "movq (%%edi,%%edx,), %%mm0 \n\t" - "movq (%%esi,%%edx,), %%mm1 \n\t" - "addl $8, %%edx \n\t" - "paddb %%mm1, %%mm0 \n\t" - "cmpl _MMXLength, %%edx \n\t" - "movq %%mm0, -8(%%edi,%%edx,) \n\t" // mov does not affect flags; - // -8 to offset addl edx - "jb sub_Alp \n\t" - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1" -#endif - ); - } - break; - - } // end switch (bpp) - - __asm__ __volatile__ ( - "movl _MMXLength, %%edx \n\t" -//pre "movl row, %%edi \n\t" - "cmpl _FullLength, %%edx \n\t" - "jnb sub_end \n\t" - - "movl %%edi, %%esi \n\t" // lp = row -//pre "movl bpp, %%eax \n\t" - "addl %%eax, %%edi \n\t" // rp = row + bpp - "xorl %%eax, %%eax \n\t" - - "sub_lp2: \n\t" - "movb (%%esi,%%edx,), %%al \n\t" - "addb %%al, (%%edi,%%edx,) \n\t" - "incl %%edx \n\t" - "cmpl _FullLength, %%edx \n\t" - "jb sub_lp2 \n\t" - - "sub_end: \n\t" - "EMMS \n\t" // end MMX instructions - - : "=a" (dummy_value_a), // 0 // output regs (dummy) - "=D" (dummy_value_D) // 1 - - : "0" (bpp), // eax // input regs - "1" (row) // edi - - : "%edx", "%esi" // clobber list - ); - -} // end of png_read_filter_row_mmx_sub() -#endif - - - - -//===========================================================================// -// // -// P N G _ R E A D _ F I L T E R _ R O W _ M M X _ U P // -// // -//===========================================================================// - -// Optimized code for PNG Up filter decoder - -static void /* PRIVATE */ -png_read_filter_row_mmx_up(png_row_infop row_info, png_bytep row, - png_bytep prev_row) -{ - png_uint_32 len; - int dummy_value_d; // fix 'forbidden register 3 (dx) was spilled' error - int dummy_value_S; - int dummy_value_D; - - len = row_info->rowbytes; // number of bytes to filter - - __asm__ __volatile__ ( -//pre "movl row, %%edi \n\t" - // get # of bytes to alignment -#ifdef __PIC__ - "pushl %%ebx \n\t" -#endif - "movl %%edi, %%ecx \n\t" - "xorl %%ebx, %%ebx \n\t" - "addl $0x7, %%ecx \n\t" - "xorl %%eax, %%eax \n\t" - "andl $0xfffffff8, %%ecx \n\t" -//pre "movl prev_row, %%esi \n\t" - "subl %%edi, %%ecx \n\t" - "jz up_go \n\t" - - "up_lp1: \n\t" // fix alignment - "movb (%%edi,%%ebx,), %%al \n\t" - "addb (%%esi,%%ebx,), %%al \n\t" - "incl %%ebx \n\t" - "cmpl %%ecx, %%ebx \n\t" - "movb %%al, -1(%%edi,%%ebx,) \n\t" // mov does not affect flags; -1 to - "jb up_lp1 \n\t" // offset incl ebx - - "up_go: \n\t" -//pre "movl len, %%edx \n\t" - "movl %%edx, %%ecx \n\t" - "subl %%ebx, %%edx \n\t" // subtract alignment fix - "andl $0x0000003f, %%edx \n\t" // calc bytes over mult of 64 - "subl %%edx, %%ecx \n\t" // drop over bytes from length - - // unrolled loop - use all MMX registers and interleave to reduce - // number of branch instructions (loops) and reduce partial stalls - "up_loop: \n\t" - "movq (%%esi,%%ebx,), %%mm1 \n\t" - "movq (%%edi,%%ebx,), %%mm0 \n\t" - "movq 8(%%esi,%%ebx,), %%mm3 \n\t" - "paddb %%mm1, %%mm0 \n\t" - "movq 8(%%edi,%%ebx,), %%mm2 \n\t" - "movq %%mm0, (%%edi,%%ebx,) \n\t" - "paddb %%mm3, %%mm2 \n\t" - "movq 16(%%esi,%%ebx,), %%mm5 \n\t" - "movq %%mm2, 8(%%edi,%%ebx,) \n\t" - "movq 16(%%edi,%%ebx,), %%mm4 \n\t" - "movq 24(%%esi,%%ebx,), %%mm7 \n\t" - "paddb %%mm5, %%mm4 \n\t" - "movq 24(%%edi,%%ebx,), %%mm6 \n\t" - "movq %%mm4, 16(%%edi,%%ebx,) \n\t" - "paddb %%mm7, %%mm6 \n\t" - "movq 32(%%esi,%%ebx,), %%mm1 \n\t" - "movq %%mm6, 24(%%edi,%%ebx,) \n\t" - "movq 32(%%edi,%%ebx,), %%mm0 \n\t" - "movq 40(%%esi,%%ebx,), %%mm3 \n\t" - "paddb %%mm1, %%mm0 \n\t" - "movq 40(%%edi,%%ebx,), %%mm2 \n\t" - "movq %%mm0, 32(%%edi,%%ebx,) \n\t" - "paddb %%mm3, %%mm2 \n\t" - "movq 48(%%esi,%%ebx,), %%mm5 \n\t" - "movq %%mm2, 40(%%edi,%%ebx,) \n\t" - "movq 48(%%edi,%%ebx,), %%mm4 \n\t" - "movq 56(%%esi,%%ebx,), %%mm7 \n\t" - "paddb %%mm5, %%mm4 \n\t" - "movq 56(%%edi,%%ebx,), %%mm6 \n\t" - "movq %%mm4, 48(%%edi,%%ebx,) \n\t" - "addl $64, %%ebx \n\t" - "paddb %%mm7, %%mm6 \n\t" - "cmpl %%ecx, %%ebx \n\t" - "movq %%mm6, -8(%%edi,%%ebx,) \n\t" // (+56)movq does not affect flags; - "jb up_loop \n\t" // -8 to offset addl ebx - - "cmpl $0, %%edx \n\t" // test for bytes over mult of 64 - "jz up_end \n\t" - - "cmpl $8, %%edx \n\t" // test for less than 8 bytes - "jb up_lt8 \n\t" // [added by lcreeve at netins.net] - - "addl %%edx, %%ecx \n\t" - "andl $0x00000007, %%edx \n\t" // calc bytes over mult of 8 - "subl %%edx, %%ecx \n\t" // drop over bytes from length - "jz up_lt8 \n\t" - - "up_lpA: \n\t" // use MMX regs to update 8 bytes sim. - "movq (%%esi,%%ebx,), %%mm1 \n\t" - "movq (%%edi,%%ebx,), %%mm0 \n\t" - "addl $8, %%ebx \n\t" - "paddb %%mm1, %%mm0 \n\t" - "cmpl %%ecx, %%ebx \n\t" - "movq %%mm0, -8(%%edi,%%ebx,) \n\t" // movq does not affect flags; -8 to - "jb up_lpA \n\t" // offset add ebx - "cmpl $0, %%edx \n\t" // test for bytes over mult of 8 - "jz up_end \n\t" - - "up_lt8: \n\t" - "xorl %%eax, %%eax \n\t" - "addl %%edx, %%ecx \n\t" // move over byte count into counter - - "up_lp2: \n\t" // use x86 regs for remaining bytes - "movb (%%edi,%%ebx,), %%al \n\t" - "addb (%%esi,%%ebx,), %%al \n\t" - "incl %%ebx \n\t" - "cmpl %%ecx, %%ebx \n\t" - "movb %%al, -1(%%edi,%%ebx,) \n\t" // mov does not affect flags; -1 to - "jb up_lp2 \n\t" // offset inc ebx - - "up_end: \n\t" - "EMMS \n\t" // conversion of filtered row complete -#ifdef __PIC__ - "popl %%ebx \n\t" -#endif - - : "=d" (dummy_value_d), // 0 // output regs (dummy) - "=S" (dummy_value_S), // 1 - "=D" (dummy_value_D) // 2 - - : "0" (len), // edx // input regs - "1" (prev_row), // esi - "2" (row) // edi - - : "%eax", "%ecx" // clobber list (no input regs!) -#ifndef __PIC__ - , "%ebx" -#endif - -#if 0 /* MMX regs (%mm0, etc.) not supported by gcc 2.7.2.3 or egcs 1.1 */ - , "%mm0", "%mm1", "%mm2", "%mm3" - , "%mm4", "%mm5", "%mm6", "%mm7" -#endif - ); - -} // end of png_read_filter_row_mmx_up() - -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - - - - -/*===========================================================================*/ -/* */ -/* P N G _ R E A D _ F I L T E R _ R O W */ -/* */ -/*===========================================================================*/ - - -/* Optimized png_read_filter_row routines */ - -void /* PRIVATE */ -png_read_filter_row(png_structp png_ptr, png_row_infop row_info, png_bytep - row, png_bytep prev_row, int filter) -{ -#ifdef PNG_DEBUG - char filnm[10]; -#endif - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) -/* GRR: these are superseded by png_ptr->asm_flags: */ -#define UseMMX_sub 1 // GRR: converted 20000730 -#define UseMMX_up 1 // GRR: converted 20000729 -#define UseMMX_avg 1 // GRR: converted 20000828 (+ 16-bit bugfix 20000916) -#define UseMMX_paeth 1 // GRR: converted 20000828 - - if (_mmx_supported == 2) { - /* this should have happened in png_init_mmx_flags() already */ -#if !defined(PNG_1_0_X) - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - -#ifdef PNG_DEBUG - png_debug(1, "in png_read_filter_row (pnggccrd.c)\n"); - switch (filter) - { - case 0: sprintf(filnm, "none"); - break; - case 1: sprintf(filnm, "sub-%s", -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_SUB)? "MMX" : -#endif -#endif -"x86"); - break; - case 2: sprintf(filnm, "up-%s", -#ifdef PNG_ASSEMBLER_CODE_SUPPORTED -#if !defined(PNG_1_0_X) - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_UP)? "MMX" : -#endif -#endif - "x86"); - break; - case 3: sprintf(filnm, "avg-%s", -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_AVG)? "MMX" : -#endif -#endif - "x86"); - break; - case 4: sprintf(filnm, "Paeth-%s", -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_PAETH)? "MMX": -#endif -#endif -"x86"); - break; - default: sprintf(filnm, "unknw"); - break; - } - png_debug2(0, "row_number=%5ld, %5s, ", png_ptr->row_number, filnm); - png_debug1(0, "row=0x%08lx, ", (unsigned long)row); - png_debug2(0, "pixdepth=%2d, bytes=%d, ", (int)row_info->pixel_depth, - (int)((row_info->pixel_depth + 7) >> 3)); - png_debug1(0,"rowbytes=%8ld\n", row_info->rowbytes); -#endif /* PNG_DEBUG */ - - switch (filter) - { - case PNG_FILTER_VALUE_NONE: - break; - - case PNG_FILTER_VALUE_SUB: -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_SUB) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (_mmx_supported) -#endif - { - png_read_filter_row_mmx_sub(row_info, row); - } - else -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - png_uint_32 i; - png_uint_32 istop = row_info->rowbytes; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_bytep rp = row + bpp; - png_bytep lp = row; - - for (i = bpp; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*lp++)) & 0xff); - rp++; - } - } /* end !UseMMX_sub */ - break; - - case PNG_FILTER_VALUE_UP: -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_UP) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (_mmx_supported) -#endif - { - png_read_filter_row_mmx_up(row_info, row, prev_row); - } - else -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - png_uint_32 i; - png_uint_32 istop = row_info->rowbytes; - png_bytep rp = row; - png_bytep pp = prev_row; - - for (i = 0; i < istop; ++i) - { - *rp = (png_byte)(((int)(*rp) + (int)(*pp++)) & 0xff); - rp++; - } - } /* end !UseMMX_up */ - break; - - case PNG_FILTER_VALUE_AVG: -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_AVG) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (_mmx_supported) -#endif - { - png_read_filter_row_mmx_avg(row_info, row, prev_row); - } - else -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - png_uint_32 i; - png_bytep rp = row; - png_bytep pp = prev_row; - png_bytep lp = row; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_uint_32 istop = row_info->rowbytes - bpp; - - for (i = 0; i < bpp; i++) - { - *rp = (png_byte)(((int)(*rp) + - ((int)(*pp++) >> 1)) & 0xff); - rp++; - } - - for (i = 0; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + - ((int)(*pp++ + *lp++) >> 1)) & 0xff); - rp++; - } - } /* end !UseMMX_avg */ - break; - - case PNG_FILTER_VALUE_PAETH: -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_THREAD_UNSAFE_OK) -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_PAETH) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (_mmx_supported) -#endif - { - png_read_filter_row_mmx_paeth(row_info, row, prev_row); - } - else -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - { - png_uint_32 i; - png_bytep rp = row; - png_bytep pp = prev_row; - png_bytep lp = row; - png_bytep cp = prev_row; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_uint_32 istop = row_info->rowbytes - bpp; - - for (i = 0; i < bpp; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*pp++)) & 0xff); - rp++; - } - - for (i = 0; i < istop; i++) /* use leftover rp,pp */ - { - int a, b, c, pa, pb, pc, p; - - a = *lp++; - b = *pp++; - c = *cp++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - /* - if (pa <= pb && pa <= pc) - p = a; - else if (pb <= pc) - p = b; - else - p = c; - */ - - p = (pa <= pb && pa <= pc) ? a : (pb <= pc) ? b : c; - - *rp = (png_byte)(((int)(*rp) + p) & 0xff); - rp++; - } - } /* end !UseMMX_paeth */ - break; - - default: - png_warning(png_ptr, "Ignoring bad row-filter type"); - *row=0; - break; - } -} - -#endif /* PNG_HAVE_ASSEMBLER_READ_FILTER_ROW */ - - -/*===========================================================================*/ -/* */ -/* P N G _ M M X _ S U P P O R T */ -/* */ -/*===========================================================================*/ - -/* GRR NOTES: (1) the following code assumes 386 or better (pushfl/popfl) - * (2) all instructions compile with gcc 2.7.2.3 and later - * (3) the function is moved down here to prevent gcc from - * inlining it in multiple places and then barfing be- - * cause the ".NOT_SUPPORTED" label is multiply defined - * [is there a way to signal that a *single* function should - * not be inlined? is there a way to modify the label for - * each inlined instance, e.g., by appending _1, _2, etc.? - * maybe if don't use leading "." in label name? (nope...sigh)] - */ +static int _mmx_supported = 2; // 0: no MMX; 1: MMX supported; 2: not tested int PNGAPI -png_mmx_support(void) +png_dummy_mmx_support(void) __attribute__((noinline)); + +int PNGAPI +png_dummy_mmx_support(void) { -#if defined(PNG_MMX_CODE_SUPPORTED) - int result; + int result; +#if defined(PNG_MMX_CODE_SUPPORTED) // superfluous, but what the heck __asm__ __volatile__ ( +#if defined(__x86_64__) + "pushq %%rbx \n\t" // rbx gets clobbered by CPUID instruction + "pushq %%rcx \n\t" // so does rcx... + "pushq %%rdx \n\t" // ...and rdx (but rcx & rdx safe on Linux) + "pushfq \n\t" // save Eflag to stack + "popq %%rax \n\t" // get Eflag from stack into rax + "movq %%rax, %%rcx \n\t" // make another copy of Eflag in rcx + "xorl $0x200000, %%eax \n\t" // toggle ID bit in Eflag (i.e., bit 21) + "pushq %%rax \n\t" // save modified Eflag back to stack + "popfq \n\t" // restore modified value to Eflag reg + "pushfq \n\t" // save Eflag to stack + "popq %%rax \n\t" // get Eflag from stack + "pushq %%rcx \n\t" // save original Eflag to stack + "popfq \n\t" // restore original Eflag +#else "pushl %%ebx \n\t" // ebx gets clobbered by CPUID instruction "pushl %%ecx \n\t" // so does ecx... "pushl %%edx \n\t" // ...and edx (but ecx & edx safe on Linux) -// ".byte 0x66 \n\t" // convert 16-bit pushf to 32-bit pushfd -// "pushf \n\t" // 16-bit pushf "pushfl \n\t" // save Eflag to stack "popl %%eax \n\t" // get Eflag from stack into eax "movl %%eax, %%ecx \n\t" // make another copy of Eflag in ecx "xorl $0x200000, %%eax \n\t" // toggle ID bit in Eflag (i.e., bit 21) "pushl %%eax \n\t" // save modified Eflag back to stack -// ".byte 0x66 \n\t" // convert 16-bit popf to 32-bit popfd -// "popf \n\t" // 16-bit popf "popfl \n\t" // restore modified value to Eflag reg "pushfl \n\t" // save Eflag to stack "popl %%eax \n\t" // get Eflag from stack "pushl %%ecx \n\t" // save original Eflag to stack "popfl \n\t" // restore original Eflag +#endif "xorl %%ecx, %%eax \n\t" // compare new Eflag with original Eflag "jz 0f \n\t" // if same, CPUID instr. is not supported @@ -5392,9 +71,15 @@ png_mmx_support(void) "0: \n\t" // .NOT_SUPPORTED: target label for jump instructions "movl $0, %%eax \n\t" // set return value to 0 "1: \n\t" // .RETURN: target label for jump instructions +#if defined(__x86_64__) + "popq %%rdx \n\t" // restore rdx + "popq %%rcx \n\t" // restore rcx + "popq %%rbx \n\t" // restore rbx +#else "popl %%edx \n\t" // restore edx "popl %%ecx \n\t" // restore ecx "popl %%ebx \n\t" // restore ebx +#endif // "ret \n\t" // DONE: no MMX support // (fall through to standard C "ret") @@ -5415,6 +100,4 @@ png_mmx_support(void) return _mmx_supported; } - - -#endif /* PNG_USE_PNGGCCRD */ +#endif diff --git a/pngget.c b/pngget.c index df7658587..a0e90bb6a 100644 --- a/pngget.c +++ b/pngget.c @@ -1,9 +1,9 @@ /* pngget.c - retrieval of values from info struct * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.15 January 5, 2007 * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2007 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -512,8 +512,11 @@ png_get_sPLT(png_structp png_ptr, png_infop info_ptr, png_sPLT_tpp spalettes) { if (png_ptr != NULL && info_ptr != NULL && spalettes != NULL) + { *spalettes = info_ptr->splt_palettes; - return ((png_uint_32)info_ptr->splt_palettes_num); + return ((png_uint_32)info_ptr->splt_palettes_num); + } + return (0); } #endif @@ -800,8 +803,11 @@ png_get_unknown_chunks(png_structp png_ptr, png_infop info_ptr, png_unknown_chunkpp unknowns) { if (png_ptr != NULL && info_ptr != NULL && unknowns != NULL) + { *unknowns = info_ptr->unknown_chunks; - return ((png_uint_32)info_ptr->unknown_chunks_num); + return ((png_uint_32)info_ptr->unknown_chunks_num); + } + return (0); } #endif @@ -829,96 +835,53 @@ png_get_compression_buffer_size(png_structp png_ptr) } #endif -#ifndef PNG_1_0_X #ifdef PNG_ASSEMBLER_CODE_SUPPORTED +#ifndef PNG_1_0_X /* this function was added to libpng 1.2.0 and should exist by default */ png_uint_32 PNGAPI png_get_asm_flags (png_structp png_ptr) { - return (png_uint_32)(png_ptr? png_ptr->asm_flags : 0L); + /* obsolete, to be removed from libpng-1.4.0 */ + return (png_ptr? 0L: 0L); } /* this function was added to libpng 1.2.0 and should exist by default */ png_uint_32 PNGAPI png_get_asm_flagmask (int flag_select) { - png_uint_32 settable_asm_flags = 0; - - if (flag_select & PNG_SELECT_READ) - settable_asm_flags |= - PNG_ASM_FLAG_MMX_READ_COMBINE_ROW | - PNG_ASM_FLAG_MMX_READ_INTERLACE | - PNG_ASM_FLAG_MMX_READ_FILTER_SUB | - PNG_ASM_FLAG_MMX_READ_FILTER_UP | - PNG_ASM_FLAG_MMX_READ_FILTER_AVG | - PNG_ASM_FLAG_MMX_READ_FILTER_PAETH ; - /* no non-MMX flags yet */ - -#if 0 - /* GRR: no write-flags yet, either, but someday... */ - if (flag_select & PNG_SELECT_WRITE) - settable_asm_flags |= - PNG_ASM_FLAG_MMX_WRITE_ [whatever] ; -#endif /* 0 */ - - return settable_asm_flags; /* _theoretically_ settable capabilities only */ + /* obsolete, to be removed from libpng-1.4.0 */ + flag_select=flag_select; + return 0L; } -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED */ - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) /* GRR: could add this: && defined(PNG_MMX_CODE_SUPPORTED) */ /* this function was added to libpng 1.2.0 */ png_uint_32 PNGAPI png_get_mmx_flagmask (int flag_select, int *compilerID) { - png_uint_32 settable_mmx_flags = 0; - - if (flag_select & PNG_SELECT_READ) - settable_mmx_flags |= - PNG_ASM_FLAG_MMX_READ_COMBINE_ROW | - PNG_ASM_FLAG_MMX_READ_INTERLACE | - PNG_ASM_FLAG_MMX_READ_FILTER_SUB | - PNG_ASM_FLAG_MMX_READ_FILTER_UP | - PNG_ASM_FLAG_MMX_READ_FILTER_AVG | - PNG_ASM_FLAG_MMX_READ_FILTER_PAETH ; -#if 0 - /* GRR: no MMX write support yet, but someday... */ - if (flag_select & PNG_SELECT_WRITE) - settable_mmx_flags |= - PNG_ASM_FLAG_MMX_WRITE_ [whatever] ; -#endif /* 0 */ - - if (compilerID != NULL) { -#ifdef PNG_USE_PNGVCRD - *compilerID = 1; /* MSVC */ -#else -#ifdef PNG_USE_PNGGCCRD - *compilerID = 2; /* gcc/gas */ -#else - *compilerID = -1; /* unknown (i.e., no asm/MMX code compiled) */ -#endif -#endif - } - - return settable_mmx_flags; /* _theoretically_ settable capabilities only */ + /* obsolete, to be removed from libpng-1.4.0 */ + flag_select=flag_select; + *compilerID = -1; /* unknown (i.e., no asm/MMX code compiled) */ + return 0L; } /* this function was added to libpng 1.2.0 */ png_byte PNGAPI png_get_mmx_bitdepth_threshold (png_structp png_ptr) { - return (png_byte)(png_ptr? png_ptr->mmx_bitdepth_threshold : 0); + /* obsolete, to be removed from libpng-1.4.0 */ + return (png_ptr? 0: 0); } /* this function was added to libpng 1.2.0 */ png_uint_32 PNGAPI png_get_mmx_rowbytes_threshold (png_structp png_ptr) { - return (png_uint_32)(png_ptr? png_ptr->mmx_rowbytes_threshold : 0L); + /* obsolete, to be removed from libpng-1.4.0 */ + return (png_ptr? 0L: 0L); } -#endif /* ?PNG_ASSEMBLER_CODE_SUPPORTED */ #endif /* ?PNG_1_0_X */ +#endif /* ?PNG_ASSEMBLER_CODE_SUPPORTED */ #ifdef PNG_SET_USER_LIMITS_SUPPORTED /* these functions were added to libpng 1.2.6 */ @@ -933,5 +896,6 @@ png_get_user_height_max (png_structp png_ptr) return (png_ptr? png_ptr->user_height_max : 0); } #endif /* ?PNG_SET_USER_LIMITS_SUPPORTED */ + #endif /* PNG_READ_SUPPORTED || PNG_WRITE_SUPPORTED */ diff --git a/pngmem.c b/pngmem.c index cfe3e0fde..13cc60c03 100644 --- a/pngmem.c +++ b/pngmem.c @@ -1,9 +1,9 @@ /* pngmem.c - stub functions for memory allocation * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -137,6 +137,9 @@ png_malloc_default(png_structp png_ptr, png_uint_32 size) png_voidp ret; #endif /* PNG_USER_MEM_SUPPORTED */ + if (png_ptr == NULL || size == 0) + return (NULL); + #ifdef PNG_MAX_MALLOC_64K if (size > (png_uint_32)65536L) { @@ -290,6 +293,8 @@ png_free_default(png_structp png_ptr, png_voidp ptr) { #endif /* PNG_USER_MEM_SUPPORTED */ + if(png_ptr == NULL || ptr == NULL) return; + if (png_ptr->offset_table != NULL) { int i; @@ -451,7 +456,7 @@ png_malloc_default(png_structp png_ptr, png_uint_32 size) if (size > (png_uint_32)65536L) { #ifndef PNG_USER_MEM_SUPPORTED - if(png_ptr->flags&PNG_FLAG_MALLOC_NULL_MEM_OK) == 0) + if ((png_ptr->flags&PNG_FLAG_MALLOC_NULL_MEM_OK) == 0) png_error(png_ptr, "Cannot Allocate > 64K"); else #endif @@ -536,8 +541,10 @@ png_voidp PNGAPI png_malloc_warn(png_structp png_ptr, png_uint_32 size) { png_voidp ptr; - png_uint_32 save_flags=png_ptr->flags; + png_uint_32 save_flags; + if(png_ptr == NULL) return (NULL); + save_flags=png_ptr->flags; png_ptr->flags|=PNG_FLAG_MALLOC_NULL_MEM_OK; ptr = (png_voidp)png_malloc((png_structp)png_ptr, size); png_ptr->flags=save_flags; @@ -580,9 +587,11 @@ void PNGAPI png_set_mem_fn(png_structp png_ptr, png_voidp mem_ptr, png_malloc_ptr malloc_fn, png_free_ptr free_fn) { + if(png_ptr != NULL) { png_ptr->mem_ptr = mem_ptr; png_ptr->malloc_fn = malloc_fn; png_ptr->free_fn = free_fn; + } } /* This function returns a pointer to the mem_ptr associated with the user @@ -592,6 +601,7 @@ png_set_mem_fn(png_structp png_ptr, png_voidp mem_ptr, png_malloc_ptr png_voidp PNGAPI png_get_mem_ptr(png_structp png_ptr) { + if(png_ptr == NULL) return (NULL); return ((png_voidp)png_ptr->mem_ptr); } #endif /* PNG_USER_MEM_SUPPORTED */ diff --git a/pngpread.c b/pngpread.c index f585239d3..aa7151c4b 100644 --- a/pngpread.c +++ b/pngpread.c @@ -1,9 +1,9 @@ /* pngpread.c - read a png file in push mode * - * Last changed in libpng 1.2.8 - December 3, 2004 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2004 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -28,6 +28,7 @@ void PNGAPI png_process_data(png_structp png_ptr, png_infop info_ptr, png_bytep buffer, png_size_t buffer_size) { + if(png_ptr == NULL || info_ptr == NULL) return; png_push_restore_buffer(png_ptr, buffer, buffer_size); while (png_ptr->buffer_size) @@ -42,6 +43,7 @@ png_process_data(png_structp png_ptr, png_infop info_ptr, void /* PRIVATE */ png_process_some_data(png_structp png_ptr, png_infop info_ptr) { + if(png_ptr == NULL) return; switch (png_ptr->process_mode) { case PNG_READ_SIG_MODE: @@ -135,60 +137,60 @@ void /* PRIVATE */ png_push_read_chunk(png_structp png_ptr, png_infop info_ptr) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IHDR; - PNG_IDAT; - PNG_IEND; - PNG_PLTE; + PNG_CONST PNG_IHDR; + PNG_CONST PNG_IDAT; + PNG_CONST PNG_IEND; + PNG_CONST PNG_PLTE; #if defined(PNG_READ_bKGD_SUPPORTED) - PNG_bKGD; + PNG_CONST PNG_bKGD; #endif #if defined(PNG_READ_cHRM_SUPPORTED) - PNG_cHRM; + PNG_CONST PNG_cHRM; #endif #if defined(PNG_READ_gAMA_SUPPORTED) - PNG_gAMA; + PNG_CONST PNG_gAMA; #endif #if defined(PNG_READ_hIST_SUPPORTED) - PNG_hIST; + PNG_CONST PNG_hIST; #endif #if defined(PNG_READ_iCCP_SUPPORTED) - PNG_iCCP; + PNG_CONST PNG_iCCP; #endif #if defined(PNG_READ_iTXt_SUPPORTED) - PNG_iTXt; + PNG_CONST PNG_iTXt; #endif #if defined(PNG_READ_oFFs_SUPPORTED) - PNG_oFFs; + PNG_CONST PNG_oFFs; #endif #if defined(PNG_READ_pCAL_SUPPORTED) - PNG_pCAL; + PNG_CONST PNG_pCAL; #endif #if defined(PNG_READ_pHYs_SUPPORTED) - PNG_pHYs; + PNG_CONST PNG_pHYs; #endif #if defined(PNG_READ_sBIT_SUPPORTED) - PNG_sBIT; + PNG_CONST PNG_sBIT; #endif #if defined(PNG_READ_sCAL_SUPPORTED) - PNG_sCAL; + PNG_CONST PNG_sCAL; #endif #if defined(PNG_READ_sRGB_SUPPORTED) - PNG_sRGB; + PNG_CONST PNG_sRGB; #endif #if defined(PNG_READ_sPLT_SUPPORTED) - PNG_sPLT; + PNG_CONST PNG_sPLT; #endif #if defined(PNG_READ_tEXt_SUPPORTED) - PNG_tEXt; + PNG_CONST PNG_tEXt; #endif #if defined(PNG_READ_tIME_SUPPORTED) - PNG_tIME; + PNG_CONST PNG_tIME; #endif #if defined(PNG_READ_tRNS_SUPPORTED) - PNG_tRNS; + PNG_CONST PNG_tRNS; #endif #if defined(PNG_READ_zTXt_SUPPORTED) - PNG_zTXt; + PNG_CONST PNG_zTXt; #endif #endif /* PNG_USE_LOCAL_ARRAYS */ /* First we make sure we have enough data for the 4 byte chunk name @@ -214,7 +216,7 @@ png_push_read_chunk(png_structp png_ptr, png_infop info_ptr) png_ptr->mode |= PNG_HAVE_CHUNK_HEADER; } - if (!png_memcmp(png_ptr->chunk_name, (png_bytep)png_IDAT, 4)) + if (!png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) if(png_ptr->mode & PNG_AFTER_IDAT) png_ptr->mode |= PNG_HAVE_CHUNK_AFTER_IDAT; @@ -222,6 +224,8 @@ png_push_read_chunk(png_structp png_ptr, png_infop info_ptr) { if (png_ptr->push_length + 4 > png_ptr->buffer_size) { + if (png_ptr->push_length != 13) + png_error(png_ptr, "Invalid IHDR length"); png_push_save_buffer(png_ptr); return; } @@ -271,7 +275,7 @@ png_push_read_chunk(png_structp png_ptr, png_infop info_ptr) } png_handle_PLTE(png_ptr, info_ptr, png_ptr->push_length); } - else if (!png_memcmp(png_ptr->chunk_name, (png_bytep)png_IDAT, 4)) + else if (!png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) { /* If we reach an IDAT chunk, this means we have read all of the * header chunks, and we can start reading the image (or if this @@ -561,6 +565,7 @@ png_push_fill_buffer(png_structp png_ptr, png_bytep buffer, png_size_t length) { png_bytep ptr; + if(png_ptr == NULL) return; ptr = buffer; if (png_ptr->save_buffer_size) { @@ -657,7 +662,7 @@ void /* PRIVATE */ png_push_read_IDAT(png_structp png_ptr) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IDAT; + PNG_CONST PNG_IDAT; #endif if (!(png_ptr->mode & PNG_HAVE_CHUNK_HEADER)) { @@ -675,7 +680,7 @@ png_push_read_IDAT(png_structp png_ptr) png_crc_read(png_ptr, png_ptr->chunk_name, 4); png_ptr->mode |= PNG_HAVE_CHUNK_HEADER; - if (png_memcmp(png_ptr->chunk_name, (png_bytep)png_IDAT, 4)) + if (png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) { png_ptr->process_mode = PNG_READ_CHUNK_MODE; if (!(png_ptr->flags & PNG_FLAG_ZLIB_FINISHED)) @@ -987,25 +992,20 @@ png_read_push_finish_row(png_structp png_ptr) /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* start of interlace block */ - const int FARDATA png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; + PNG_CONST int FARDATA png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; /* offset to next interlace block */ - const int FARDATA png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; + PNG_CONST int FARDATA png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; /* start of interlace block in the y direction */ - const int FARDATA png_pass_ystart[] = {0, 0, 4, 0, 2, 0, 1}; + PNG_CONST int FARDATA png_pass_ystart[] = {0, 0, 4, 0, 2, 0, 1}; /* offset to next interlace block in the y direction */ - const int FARDATA png_pass_yinc[] = {8, 8, 8, 4, 4, 2, 2}; - - /* Width of interlace block. This is not currently used - if you need - * it, uncomment it here and in png.h - const int FARDATA png_pass_width[] = {8, 4, 4, 2, 2, 1, 1}; - */ + PNG_CONST int FARDATA png_pass_yinc[] = {8, 8, 8, 4, 4, 2, 2}; /* Height of interlace block. This is not currently used - if you need * it, uncomment it here and in png.h - const int FARDATA png_pass_height[] = {8, 8, 4, 4, 2, 2, 1}; + PNG_CONST int FARDATA png_pass_height[] = {8, 8, 4, 4, 2, 2, 1}; */ #endif @@ -1020,32 +1020,37 @@ png_read_push_finish_row(png_structp png_ptr) png_ptr->rowbytes + 1); do { - png_ptr->pass++; - if ((png_ptr->pass == 1 && png_ptr->width < 5) || - (png_ptr->pass == 3 && png_ptr->width < 3) || - (png_ptr->pass == 5 && png_ptr->width < 2)) - png_ptr->pass++; + int pass; + pass = png_ptr->pass; + pass++; + if ((pass == 1 && png_ptr->width < 5) || + (pass == 3 && png_ptr->width < 3) || + (pass == 5 && png_ptr->width < 2)) + pass++; - if (png_ptr->pass > 7) - png_ptr->pass--; - if (png_ptr->pass >= 7) - break; + if (pass > 7) + pass--; + png_ptr->pass = (png_byte) pass; + if (pass < 7) + { + png_ptr->iwidth = (png_ptr->width + + png_pass_inc[pass] - 1 - + png_pass_start[pass]) / + png_pass_inc[pass]; - png_ptr->iwidth = (png_ptr->width + - png_pass_inc[png_ptr->pass] - 1 - - png_pass_start[png_ptr->pass]) / - png_pass_inc[png_ptr->pass]; + png_ptr->irowbytes = PNG_ROWBYTES(png_ptr->pixel_depth, + png_ptr->iwidth) + 1; - png_ptr->irowbytes = PNG_ROWBYTES(png_ptr->pixel_depth, - png_ptr->iwidth) + 1; + if (png_ptr->transformations & PNG_INTERLACE) + break; - if (png_ptr->transformations & PNG_INTERLACE) - break; - - png_ptr->num_rows = (png_ptr->height + - png_pass_yinc[png_ptr->pass] - 1 - - png_pass_ystart[png_ptr->pass]) / - png_pass_yinc[png_ptr->pass]; + png_ptr->num_rows = (png_ptr->height + + png_pass_yinc[pass] - 1 - + png_pass_ystart[pass]) / + png_pass_yinc[pass]; + } + else + break; } while (png_ptr->iwidth == 0 || png_ptr->num_rows == 0); } @@ -1059,8 +1064,7 @@ png_push_handle_tEXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 if (!(png_ptr->mode & PNG_HAVE_IHDR) || (png_ptr->mode & PNG_HAVE_IEND)) { png_error(png_ptr, "Out of place tEXt"); - /* to quiet some compiler warnings */ - if(info_ptr == NULL) return; + info_ptr = info_ptr; /* to quiet some compiler warnings */ } #ifdef PNG_MAX_MALLOC_64K @@ -1123,7 +1127,7 @@ png_push_read_tEXt(png_structp png_ptr, png_infop info_ptr) for (text = key; *text; text++) /* empty loop */ ; - if (text != key + png_ptr->current_text_size) + if (text < key + png_ptr->current_text_size) text++; text_ptr = (png_textp)png_malloc(png_ptr, @@ -1156,8 +1160,7 @@ png_push_handle_zTXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 if (!(png_ptr->mode & PNG_HAVE_IHDR) || (png_ptr->mode & PNG_HAVE_IEND)) { png_error(png_ptr, "Out of place zTXt"); - /* to quiet some compiler warnings */ - if(info_ptr == NULL) return; + info_ptr = info_ptr; /* to quiet some compiler warnings */ } #ifdef PNG_MAX_MALLOC_64K @@ -1219,7 +1222,7 @@ png_push_read_zTXt(png_structp png_ptr, png_infop info_ptr) /* empty loop */ ; /* zTXt can't have zero text */ - if (text == key + png_ptr->current_text_size) + if (text >= key + png_ptr->current_text_size) { png_ptr->current_text = NULL; png_free(png_ptr, key); @@ -1349,8 +1352,7 @@ png_push_handle_iTXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 if (!(png_ptr->mode & PNG_HAVE_IHDR) || (png_ptr->mode & PNG_HAVE_IEND)) { png_error(png_ptr, "Out of place iTXt"); - /* to quiet some compiler warnings */ - if(info_ptr == NULL) return; + info_ptr = info_ptr; /* to quiet some compiler warnings */ } #ifdef PNG_MAX_MALLOC_64K @@ -1417,7 +1419,7 @@ png_push_read_iTXt(png_structp png_ptr, png_infop info_ptr) for (lang = key; *lang; lang++) /* empty loop */ ; - if (lang != key + png_ptr->current_text_size) + if (lang < key + png_ptr->current_text_size - 3) lang++; comp_flag = *lang++; @@ -1427,10 +1429,14 @@ png_push_read_iTXt(png_structp png_ptr, png_infop info_ptr) /* empty loop */ ; lang_key++; /* skip NUL separator */ - for (text = lang_key; *text; text++) - /* empty loop */ ; + text=lang_key; + if (lang_key < key + png_ptr->current_text_size - 1) + { + for (; *text; text++) + /* empty loop */ ; + } - if (text != key + png_ptr->current_text_size) + if (text < key + png_ptr->current_text_size) text++; text_ptr = (png_textp)png_malloc(png_ptr, @@ -1468,55 +1474,66 @@ png_push_handle_unknown(png_structp png_ptr, png_infop info_ptr, png_uint_32 if (!(png_ptr->chunk_name[0] & 0x20)) { #if defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) - if(png_handle_as_unknown(png_ptr, png_ptr->chunk_name) != - PNG_HANDLE_CHUNK_ALWAYS + if(png_handle_as_unknown(png_ptr, png_ptr->chunk_name) != + PNG_HANDLE_CHUNK_ALWAYS #if defined(PNG_READ_USER_CHUNKS_SUPPORTED) - && png_ptr->read_user_chunk_fn == NULL + && png_ptr->read_user_chunk_fn == NULL #endif - ) + ) #endif - png_chunk_error(png_ptr, "unknown critical chunk"); + png_chunk_error(png_ptr, "unknown critical chunk"); - /* to quiet compiler warnings about unused info_ptr */ - if (info_ptr == NULL) - return; + info_ptr = info_ptr; /* to quiet some compiler warnings */ } #if defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) if (png_ptr->flags & PNG_FLAG_KEEP_UNKNOWN_CHUNKS) { - png_unknown_chunk chunk; - #ifdef PNG_MAX_MALLOC_64K - if (length > (png_uint_32)65535L) - { - png_warning(png_ptr, "unknown chunk too large to fit in memory"); - skip = length - (png_uint_32)65535L; - length = (png_uint_32)65535L; - } + if (length > (png_uint_32)65535L) + { + png_warning(png_ptr, "unknown chunk too large to fit in memory"); + skip = length - (png_uint_32)65535L; + length = (png_uint_32)65535L; + } #endif + png_memcpy((png_charp)png_ptr->unknown_chunk.name, + (png_charp)png_ptr->chunk_name, + png_sizeof(png_ptr->unknown_chunk.name)); + png_ptr->unknown_chunk.name[png_sizeof(png_ptr->unknown_chunk.name)-1]='\0'; - png_strcpy((png_charp)chunk.name, (png_charp)png_ptr->chunk_name); - chunk.data = (png_bytep)png_malloc(png_ptr, length); - png_crc_read(png_ptr, chunk.data, length); - chunk.size = length; + png_ptr->unknown_chunk.size = (png_size_t)length; + if (length == 0) + png_ptr->unknown_chunk.data = NULL; + else + { + png_ptr->unknown_chunk.data = (png_bytep)png_malloc(png_ptr, length); + png_crc_read(png_ptr, (png_bytep)png_ptr->unknown_chunk.data, length); + } #if defined(PNG_READ_USER_CHUNKS_SUPPORTED) - if(png_ptr->read_user_chunk_fn != NULL) - { - /* callback to user unknown chunk handler */ - if ((*(png_ptr->read_user_chunk_fn)) (png_ptr, &chunk) <= 0) - { - if (!(png_ptr->chunk_name[0] & 0x20)) - if(png_handle_as_unknown(png_ptr, png_ptr->chunk_name) != - PNG_HANDLE_CHUNK_ALWAYS) - png_chunk_error(png_ptr, "unknown critical chunk"); - } - png_set_unknown_chunks(png_ptr, info_ptr, &chunk, 1); - } - else + if(png_ptr->read_user_chunk_fn != NULL) + { + /* callback to user unknown chunk handler */ + int ret; + ret = (*(png_ptr->read_user_chunk_fn)) + (png_ptr, &png_ptr->unknown_chunk); + if (ret < 0) + png_chunk_error(png_ptr, "error in user chunk"); + if (ret == 0) + { + if (!(png_ptr->chunk_name[0] & 0x20)) + if(png_handle_as_unknown(png_ptr, png_ptr->chunk_name) != + PNG_HANDLE_CHUNK_ALWAYS) + png_chunk_error(png_ptr, "unknown critical chunk"); + png_set_unknown_chunks(png_ptr, info_ptr, + &png_ptr->unknown_chunk, 1); + } + } + else #endif - png_set_unknown_chunks(png_ptr, info_ptr, &chunk, 1); - png_free(png_ptr, chunk.data); + png_set_unknown_chunks(png_ptr, info_ptr, &png_ptr->unknown_chunk, 1); + png_free(png_ptr, png_ptr->unknown_chunk.data); + png_ptr->unknown_chunk.data = NULL; } else #endif @@ -1551,9 +1568,10 @@ png_progressive_combine_row (png_structp png_ptr, png_bytep old_row, png_bytep new_row) { #ifdef PNG_USE_LOCAL_ARRAYS - const int FARDATA png_pass_dsp_mask[7] = + PNG_CONST int FARDATA png_pass_dsp_mask[7] = {0xff, 0x0f, 0xff, 0x33, 0xff, 0x55, 0xff}; #endif + if(png_ptr == NULL) return; if (new_row != NULL) /* new_row must == png_ptr->row_buf here. */ png_combine_row(png_ptr, old_row, png_pass_dsp_mask[png_ptr->pass]); } @@ -1563,6 +1581,7 @@ png_set_progressive_read_fn(png_structp png_ptr, png_voidp progressive_ptr, png_progressive_info_ptr info_fn, png_progressive_row_ptr row_fn, png_progressive_end_ptr end_fn) { + if(png_ptr == NULL) return; png_ptr->info_fn = info_fn; png_ptr->row_fn = row_fn; png_ptr->end_fn = end_fn; @@ -1573,6 +1592,7 @@ png_set_progressive_read_fn(png_structp png_ptr, png_voidp progressive_ptr, png_voidp PNGAPI png_get_progressive_ptr(png_structp png_ptr) { + if(png_ptr == NULL) return (NULL); return png_ptr->io_ptr; } #endif /* PNG_PROGRESSIVE_READ_SUPPORTED */ diff --git a/pngread.c b/pngread.c index 9dd83bc7b..bd8bcd98a 100644 --- a/pngread.c +++ b/pngread.c @@ -1,9 +1,9 @@ /* pngread.c - read a PNG file * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.25 [February 18, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -55,12 +55,6 @@ png_create_read_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, if (png_ptr == NULL) return (NULL); -#if !defined(PNG_1_0_X) -#ifdef PNG_ASSEMBLER_CODE_SUPPORTED - png_init_mmx_flags(png_ptr); /* 1.2.0 addition */ -#endif -#endif /* PNG_1_0_X */ - /* added at libpng-1.2.6 */ #ifdef PNG_SET_USER_LIMITS_SUPPORTED png_ptr->user_width_max=PNG_USER_WIDTH_MAX; @@ -95,12 +89,18 @@ png_create_read_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, png_set_error_fn(png_ptr, error_ptr, error_fn, warn_fn); - i=0; - do + if(user_png_ver) { - if(user_png_ver[i] != png_libpng_ver[i]) + i=0; + do + { + if(user_png_ver[i] != png_libpng_ver[i]) + png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; + } while (png_libpng_ver[i++]); + } + else png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; - } while (png_libpng_ver[i++]); + if (png_ptr->flags & PNG_FLAG_LIBRARY_MISMATCH) { @@ -117,11 +117,13 @@ png_create_read_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, char msg[80]; if (user_png_ver) { - sprintf(msg, "Application was compiled with png.h from libpng-%.20s", + png_snprintf(msg, 80, + "Application was compiled with png.h from libpng-%.20s", user_png_ver); png_warning(png_ptr, msg); } - sprintf(msg, "Application is running with png.c from libpng-%.20s", + png_snprintf(msg, 80, + "Application is running with png.c from libpng-%.20s", png_libpng_ver); png_warning(png_ptr, msg); #endif @@ -188,6 +190,7 @@ png_read_init_2(png_structp png_ptr, png_const_charp user_png_ver, png_size_t png_struct_size, png_size_t png_info_size) { /* We only come here via pre-1.0.12-compiled applications */ + if(png_ptr == NULL) return; #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) if(png_sizeof(png_struct) > png_struct_size || png_sizeof(png_info) > png_info_size) @@ -196,11 +199,13 @@ png_read_init_2(png_structp png_ptr, png_const_charp user_png_ver, png_ptr->warning_fn=NULL; if (user_png_ver) { - sprintf(msg, "Application was compiled with png.h from libpng-%.20s", + png_snprintf(msg, 80, + "Application was compiled with png.h from libpng-%.20s", user_png_ver); png_warning(png_ptr, msg); } - sprintf(msg, "Application is running with png.c from libpng-%.20s", + png_snprintf(msg, 80, + "Application is running with png.c from libpng-%.20s", png_libpng_ver); png_warning(png_ptr, msg); } @@ -239,6 +244,8 @@ png_read_init_3(png_structpp ptr_ptr, png_const_charp user_png_ver, png_structp png_ptr=*ptr_ptr; + if(png_ptr == NULL) return; + do { if(user_png_ver[i] != png_libpng_ver[i]) @@ -317,6 +324,7 @@ png_read_init_3(png_structpp ptr_ptr, png_const_charp user_png_ver, void PNGAPI png_read_info(png_structp png_ptr, png_infop info_ptr) { + if(png_ptr == NULL || info_ptr == NULL) return; png_debug(1, "in png_read_info\n"); /* If we haven't checked all of the PNG signature bytes, do so now. */ if (png_ptr->sig_bytes < 8) @@ -342,60 +350,60 @@ png_read_info(png_structp png_ptr, png_infop info_ptr) for(;;) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IHDR; - PNG_IDAT; - PNG_IEND; - PNG_PLTE; + PNG_CONST PNG_IHDR; + PNG_CONST PNG_IDAT; + PNG_CONST PNG_IEND; + PNG_CONST PNG_PLTE; #if defined(PNG_READ_bKGD_SUPPORTED) - PNG_bKGD; + PNG_CONST PNG_bKGD; #endif #if defined(PNG_READ_cHRM_SUPPORTED) - PNG_cHRM; + PNG_CONST PNG_cHRM; #endif #if defined(PNG_READ_gAMA_SUPPORTED) - PNG_gAMA; + PNG_CONST PNG_gAMA; #endif #if defined(PNG_READ_hIST_SUPPORTED) - PNG_hIST; + PNG_CONST PNG_hIST; #endif #if defined(PNG_READ_iCCP_SUPPORTED) - PNG_iCCP; + PNG_CONST PNG_iCCP; #endif #if defined(PNG_READ_iTXt_SUPPORTED) - PNG_iTXt; + PNG_CONST PNG_iTXt; #endif #if defined(PNG_READ_oFFs_SUPPORTED) - PNG_oFFs; + PNG_CONST PNG_oFFs; #endif #if defined(PNG_READ_pCAL_SUPPORTED) - PNG_pCAL; + PNG_CONST PNG_pCAL; #endif #if defined(PNG_READ_pHYs_SUPPORTED) - PNG_pHYs; + PNG_CONST PNG_pHYs; #endif #if defined(PNG_READ_sBIT_SUPPORTED) - PNG_sBIT; + PNG_CONST PNG_sBIT; #endif #if defined(PNG_READ_sCAL_SUPPORTED) - PNG_sCAL; + PNG_CONST PNG_sCAL; #endif #if defined(PNG_READ_sPLT_SUPPORTED) - PNG_sPLT; + PNG_CONST PNG_sPLT; #endif #if defined(PNG_READ_sRGB_SUPPORTED) - PNG_sRGB; + PNG_CONST PNG_sRGB; #endif #if defined(PNG_READ_tEXt_SUPPORTED) - PNG_tEXt; + PNG_CONST PNG_tEXt; #endif #if defined(PNG_READ_tIME_SUPPORTED) - PNG_tIME; + PNG_CONST PNG_tIME; #endif #if defined(PNG_READ_tRNS_SUPPORTED) - PNG_tRNS; + PNG_CONST PNG_tRNS; #endif #if defined(PNG_READ_zTXt_SUPPORTED) - PNG_zTXt; + PNG_CONST PNG_zTXt; #endif #endif /* PNG_USE_LOCAL_ARRAYS */ png_byte chunk_length[4]; @@ -413,7 +421,7 @@ png_read_info(png_structp png_ptr, png_infop info_ptr) /* This should be a binary subdivision search or a hash for * matching the chunk name rather than a linear search. */ - if (!png_memcmp(png_ptr->chunk_name, (png_bytep)png_IDAT, 4)) + if (!png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) if(png_ptr->mode & PNG_AFTER_IDAT) png_ptr->mode |= PNG_HAVE_CHUNK_AFTER_IDAT; @@ -533,6 +541,7 @@ void PNGAPI png_read_update_info(png_structp png_ptr, png_infop info_ptr) { png_debug(1, "in png_read_update_info\n"); + if(png_ptr == NULL) return; if (!(png_ptr->flags & PNG_FLAG_ROW_INIT)) png_read_start_row(png_ptr); else @@ -551,6 +560,7 @@ void PNGAPI png_start_read_image(png_structp png_ptr) { png_debug(1, "in png_start_read_image\n"); + if(png_ptr == NULL) return; if (!(png_ptr->flags & PNG_FLAG_ROW_INIT)) png_read_start_row(png_ptr); } @@ -561,11 +571,13 @@ void PNGAPI png_read_row(png_structp png_ptr, png_bytep row, png_bytep dsp_row) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IDAT; - const int png_pass_dsp_mask[7] = {0xff, 0x0f, 0xff, 0x33, 0xff, 0x55, 0xff}; - const int png_pass_mask[7] = {0x80, 0x08, 0x88, 0x22, 0xaa, 0x55, 0xff}; + PNG_CONST PNG_IDAT; + PNG_CONST int png_pass_dsp_mask[7] = {0xff, 0x0f, 0xff, 0x33, 0xff, 0x55, + 0xff}; + PNG_CONST int png_pass_mask[7] = {0x80, 0x08, 0x88, 0x22, 0xaa, 0x55, 0xff}; #endif int ret; + if(png_ptr == NULL) return; png_debug2(1, "in png_read_row (row %lu, pass %d)\n", png_ptr->row_number, png_ptr->pass); if (!(png_ptr->flags & PNG_FLAG_ROW_INIT)) @@ -824,6 +836,7 @@ png_read_rows(png_structp png_ptr, png_bytepp row, png_bytepp dp; png_debug(1, "in png_read_rows\n"); + if(png_ptr == NULL) return; rp = row; dp = display_row; if (rp != NULL && dp != NULL) @@ -872,6 +885,7 @@ png_read_image(png_structp png_ptr, png_bytepp image) png_bytepp rp; png_debug(1, "in png_read_image\n"); + if(png_ptr == NULL) return; #ifdef PNG_READ_INTERLACING_SUPPORTED pass = png_set_interlace_handling(png_ptr); @@ -910,65 +924,66 @@ png_read_end(png_structp png_ptr, png_infop info_ptr) png_uint_32 length; png_debug(1, "in png_read_end\n"); + if(png_ptr == NULL) return; png_crc_finish(png_ptr, 0); /* Finish off CRC from last IDAT chunk */ do { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IHDR; - PNG_IDAT; - PNG_IEND; - PNG_PLTE; + PNG_CONST PNG_IHDR; + PNG_CONST PNG_IDAT; + PNG_CONST PNG_IEND; + PNG_CONST PNG_PLTE; #if defined(PNG_READ_bKGD_SUPPORTED) - PNG_bKGD; + PNG_CONST PNG_bKGD; #endif #if defined(PNG_READ_cHRM_SUPPORTED) - PNG_cHRM; + PNG_CONST PNG_cHRM; #endif #if defined(PNG_READ_gAMA_SUPPORTED) - PNG_gAMA; + PNG_CONST PNG_gAMA; #endif #if defined(PNG_READ_hIST_SUPPORTED) - PNG_hIST; + PNG_CONST PNG_hIST; #endif #if defined(PNG_READ_iCCP_SUPPORTED) - PNG_iCCP; + PNG_CONST PNG_iCCP; #endif #if defined(PNG_READ_iTXt_SUPPORTED) - PNG_iTXt; + PNG_CONST PNG_iTXt; #endif #if defined(PNG_READ_oFFs_SUPPORTED) - PNG_oFFs; + PNG_CONST PNG_oFFs; #endif #if defined(PNG_READ_pCAL_SUPPORTED) - PNG_pCAL; + PNG_CONST PNG_pCAL; #endif #if defined(PNG_READ_pHYs_SUPPORTED) - PNG_pHYs; + PNG_CONST PNG_pHYs; #endif #if defined(PNG_READ_sBIT_SUPPORTED) - PNG_sBIT; + PNG_CONST PNG_sBIT; #endif #if defined(PNG_READ_sCAL_SUPPORTED) - PNG_sCAL; + PNG_CONST PNG_sCAL; #endif #if defined(PNG_READ_sPLT_SUPPORTED) - PNG_sPLT; + PNG_CONST PNG_sPLT; #endif #if defined(PNG_READ_sRGB_SUPPORTED) - PNG_sRGB; + PNG_CONST PNG_sRGB; #endif #if defined(PNG_READ_tEXt_SUPPORTED) - PNG_tEXt; + PNG_CONST PNG_tEXt; #endif #if defined(PNG_READ_tIME_SUPPORTED) - PNG_tIME; + PNG_CONST PNG_tIME; #endif #if defined(PNG_READ_tRNS_SUPPORTED) - PNG_tRNS; + PNG_CONST PNG_tRNS; #endif #if defined(PNG_READ_zTXt_SUPPORTED) - PNG_zTXt; + PNG_CONST PNG_zTXt; #endif #endif /* PNG_USE_LOCAL_ARRAYS */ @@ -1090,13 +1105,22 @@ png_destroy_read_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr, png_structp png_ptr = NULL; png_infop info_ptr = NULL, end_info_ptr = NULL; #ifdef PNG_USER_MEM_SUPPORTED - png_free_ptr free_fn; - png_voidp mem_ptr; + png_free_ptr free_fn = NULL; + png_voidp mem_ptr = NULL; #endif png_debug(1, "in png_destroy_read_struct\n"); if (png_ptr_ptr != NULL) + { png_ptr = *png_ptr_ptr; + } + if (png_ptr == NULL) + return; + +#ifdef PNG_USER_MEM_SUPPORTED + free_fn = png_ptr->free_fn; + mem_ptr = png_ptr->mem_ptr; +#endif if (info_ptr_ptr != NULL) info_ptr = *info_ptr_ptr; @@ -1104,11 +1128,6 @@ png_destroy_read_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr, if (end_info_ptr_ptr != NULL) end_info_ptr = *end_info_ptr_ptr; -#ifdef PNG_USER_MEM_SUPPORTED - free_fn = png_ptr->free_fn; - mem_ptr = png_ptr->mem_ptr; -#endif - png_read_destroy(png_ptr, info_ptr, end_info_ptr); if (info_ptr != NULL) @@ -1140,16 +1159,13 @@ png_destroy_read_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr, *end_info_ptr_ptr = NULL; } - if (png_ptr != NULL) - { #ifdef PNG_USER_MEM_SUPPORTED - png_destroy_struct_2((png_voidp)png_ptr, (png_free_ptr)free_fn, - (png_voidp)mem_ptr); + png_destroy_struct_2((png_voidp)png_ptr, (png_free_ptr)free_fn, + (png_voidp)mem_ptr); #else - png_destroy_struct((png_voidp)png_ptr); + png_destroy_struct((png_voidp)png_ptr); #endif - *png_ptr_ptr = NULL; - } + *png_ptr_ptr = NULL; } /* free all memory used by the read (old method) */ @@ -1300,6 +1316,7 @@ png_read_destroy(png_structp png_ptr, png_infop info_ptr, png_infop end_info_ptr void PNGAPI png_set_read_status_fn(png_structp png_ptr, png_read_status_ptr read_row_fn) { + if(png_ptr == NULL) return; png_ptr->read_row_fn = read_row_fn; } @@ -1313,6 +1330,7 @@ png_read_png(png_structp png_ptr, png_infop info_ptr, { int row; + if(png_ptr == NULL) return; #if defined(PNG_READ_INVERT_ALPHA_SUPPORTED) /* invert the alpha channel from opacity to transparency */ @@ -1452,8 +1470,8 @@ png_read_png(png_structp png_ptr, png_infop info_ptr, /* read rest of file, and get additional chunks in info_ptr - REQUIRED */ png_read_end(png_ptr, info_ptr); - if(transforms == 0 || params == NULL) - /* quiet compiler warnings */ return; + transforms = transforms; /* quiet compiler warnings */ + params = params; } #endif /* PNG_INFO_IMAGE_SUPPORTED */ diff --git a/pngrio.c b/pngrio.c index ce05cada7..7d2522f1f 100644 --- a/pngrio.c +++ b/pngrio.c @@ -1,7 +1,7 @@ /* pngrio.c - functions for data input * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.13 November 13, 2006 * For conditions of distribution and use, see copyright notice in png.h * Copyright (c) 1998-2006 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) @@ -46,6 +46,7 @@ png_default_read_data(png_structp png_ptr, png_bytep data, png_size_t length) { png_size_t check; + if(png_ptr == NULL) return; /* fread() returns 0 on error, so it is OK to store this in a png_size_t * instead of an int, which is what fread() actually returns. */ @@ -76,6 +77,7 @@ png_default_read_data(png_structp png_ptr, png_bytep data, png_size_t length) png_byte *n_data; png_FILE_p io_ptr; + if(png_ptr == NULL) return; /* Check if data really is near. If so, use usual code. */ n_data = (png_byte *)CVT_PTR_NOCHECK(data); io_ptr = (png_FILE_p)CVT_PTR(png_ptr->io_ptr); @@ -136,6 +138,7 @@ void PNGAPI png_set_read_fn(png_structp png_ptr, png_voidp io_ptr, png_rw_ptr read_data_fn) { + if(png_ptr == NULL) return; png_ptr->io_ptr = io_ptr; #if !defined(PNG_NO_STDIO) diff --git a/pngrtran.c b/pngrtran.c index 8b580807c..873b22cf3 100644 --- a/pngrtran.c +++ b/pngrtran.c @@ -1,9 +1,9 @@ /* pngrtran.c - transforms the data in a row for PNG readers * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -24,6 +24,7 @@ png_set_crc_action(png_structp png_ptr, int crit_action, int ancil_action) { png_debug(1, "in png_set_crc_action\n"); /* Tell libpng how we react to CRC errors in critical chunks */ + if(png_ptr == NULL) return; switch (crit_action) { case PNG_CRC_NO_CHANGE: /* leave setting as is */ @@ -80,6 +81,7 @@ png_set_background(png_structp png_ptr, int need_expand, double background_gamma) { png_debug(1, "in png_set_background\n"); + if(png_ptr == NULL) return; if (background_gamma_code == PNG_BACKGROUND_GAMMA_UNKNOWN) { png_warning(png_ptr, "Application must supply a known background gamma"); @@ -92,17 +94,6 @@ png_set_background(png_structp png_ptr, png_ptr->background_gamma = (float)background_gamma; png_ptr->background_gamma_type = (png_byte)(background_gamma_code); png_ptr->transformations |= (need_expand ? PNG_BACKGROUND_EXPAND : 0); - - /* Note: if need_expand is set and color_type is either RGB or RGB_ALPHA - * (in which case need_expand is superfluous anyway), the background color - * might actually be gray yet not be flagged as such. This is not a problem - * for the current code, which uses PNG_BACKGROUND_IS_GRAY only to - * decide when to do the png_do_gray_to_rgb() transformation. - */ - if ((need_expand && !(png_ptr->color_type & PNG_COLOR_MASK_COLOR)) || - (!need_expand && background_color->red == background_color->green && - background_color->red == background_color->blue)) - png_ptr->mode |= PNG_BACKGROUND_IS_GRAY; } #endif @@ -112,6 +103,7 @@ void PNGAPI png_set_strip_16(png_structp png_ptr) { png_debug(1, "in png_set_strip_16\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_16_TO_8; } #endif @@ -121,6 +113,7 @@ void PNGAPI png_set_strip_alpha(png_structp png_ptr) { png_debug(1, "in png_set_strip_alpha\n"); + if(png_ptr == NULL) return; png_ptr->flags |= PNG_FLAG_STRIP_ALPHA; } #endif @@ -150,6 +143,7 @@ png_set_dither(png_structp png_ptr, png_colorp palette, int full_dither) { png_debug(1, "in png_set_dither\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_DITHER; if (!full_dither) @@ -533,6 +527,7 @@ void PNGAPI png_set_gamma(png_structp png_ptr, double scrn_gamma, double file_gamma) { png_debug(1, "in png_set_gamma\n"); + if(png_ptr == NULL) return; if ((fabs(scrn_gamma * file_gamma - 1.0) > PNG_GAMMA_THRESHOLD) || (png_ptr->color_type & PNG_COLOR_MASK_ALPHA) || (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE)) @@ -551,7 +546,9 @@ void PNGAPI png_set_expand(png_structp png_ptr) { png_debug(1, "in png_set_expand\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); + png_ptr->flags &= ~PNG_FLAG_ROW_INIT; } /* GRR 19990627: the following three functions currently are identical @@ -576,7 +573,9 @@ void PNGAPI png_set_palette_to_rgb(png_structp png_ptr) { png_debug(1, "in png_set_palette_to_rgb\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); + png_ptr->flags &= ~PNG_FLAG_ROW_INIT; } #if !defined(PNG_1_0_X) @@ -585,7 +584,9 @@ void PNGAPI png_set_expand_gray_1_2_4_to_8(png_structp png_ptr) { png_debug(1, "in png_set_expand_gray_1_2_4_to_8\n"); - png_ptr->transformations |= PNG_EXPAND_tRNS; + if(png_ptr == NULL) return; + png_ptr->transformations |= PNG_EXPAND; + png_ptr->flags &= ~PNG_FLAG_ROW_INIT; } #endif @@ -596,6 +597,7 @@ void PNGAPI png_set_gray_1_2_4_to_8(png_structp png_ptr) { png_debug(1, "in png_set_gray_1_2_4_to_8\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); } #endif @@ -605,8 +607,9 @@ png_set_gray_1_2_4_to_8(png_structp png_ptr) void PNGAPI png_set_tRNS_to_alpha(png_structp png_ptr) { - png_debug(1, "in png_set_expand\n"); + png_debug(1, "in png_set_tRNS_to_alpha\n"); png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); + png_ptr->flags &= ~PNG_FLAG_ROW_INIT; } #endif /* defined(PNG_READ_EXPAND_SUPPORTED) */ @@ -616,6 +619,7 @@ png_set_gray_to_rgb(png_structp png_ptr) { png_debug(1, "in png_set_gray_to_rgb\n"); png_ptr->transformations |= PNG_GRAY_TO_RGB; + png_ptr->flags &= ~PNG_FLAG_ROW_INIT; } #endif @@ -631,6 +635,7 @@ png_set_rgb_to_gray(png_structp png_ptr, int error_action, double red, { int red_fixed = (int)((float)red*100000.0 + 0.5); int green_fixed = (int)((float)green*100000.0 + 0.5); + if(png_ptr == NULL) return; png_set_rgb_to_gray_fixed(png_ptr, error_action, red_fixed, green_fixed); } #endif @@ -640,6 +645,7 @@ png_set_rgb_to_gray_fixed(png_structp png_ptr, int error_action, png_fixed_point red, png_fixed_point green) { png_debug(1, "in png_set_rgb_to_gray\n"); + if(png_ptr == NULL) return; switch(error_action) { case 1: png_ptr->transformations |= PNG_RGB_TO_GRAY; @@ -690,6 +696,7 @@ png_set_read_user_transform_fn(png_structp png_ptr, png_user_transform_ptr read_user_transform_fn) { png_debug(1, "in png_set_read_user_transform_fn\n"); + if(png_ptr == NULL) return; #if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) png_ptr->transformations |= PNG_USER_TRANSFORM; png_ptr->read_user_transform_fn = read_user_transform_fn; @@ -719,6 +726,32 @@ png_init_read_transformations(png_structp png_ptr) #endif #if defined(PNG_READ_EXPAND_SUPPORTED) && defined(PNG_READ_BACKGROUND_SUPPORTED) + +#if defined(PNG_READ_GRAY_TO_RGB_SUPPORTED) + /* Detect gray background and attempt to enable optimization + * for gray --> RGB case */ + /* Note: if PNG_BACKGROUND_EXPAND is set and color_type is either RGB or + * RGB_ALPHA (in which case need_expand is superfluous anyway), the + * background color might actually be gray yet not be flagged as such. + * This is not a problem for the current code, which uses + * PNG_BACKGROUND_IS_GRAY only to decide when to do the + * png_do_gray_to_rgb() transformation. + */ + if ((png_ptr->transformations & PNG_BACKGROUND_EXPAND) && + !(color_type & PNG_COLOR_MASK_COLOR)) + { + png_ptr->mode |= PNG_BACKGROUND_IS_GRAY; + } else if ((png_ptr->transformations & PNG_BACKGROUND) && + !(png_ptr->transformations & PNG_BACKGROUND_EXPAND) && + (png_ptr->transformations & PNG_GRAY_TO_RGB) && + png_ptr->background.red == png_ptr->background.green && + png_ptr->background.red == png_ptr->background.blue) + { + png_ptr->mode |= PNG_BACKGROUND_IS_GRAY; + png_ptr->background.gray = png_ptr->background.red; + } +#endif + if ((png_ptr->transformations & PNG_BACKGROUND_EXPAND) && (png_ptr->transformations & PNG_EXPAND)) { @@ -814,7 +847,7 @@ png_init_read_transformations(png_structp png_ptr) k=1; /* partial transparency is present */ } if (k == 0) - png_ptr->transformations &= (~PNG_GAMMA); + png_ptr->transformations &= ~PNG_GAMMA; } if ((png_ptr->transformations & (PNG_GAMMA | PNG_RGB_TO_GRAY)) && @@ -921,6 +954,14 @@ png_init_read_transformations(png_structp png_ptr) palette[i].blue = png_ptr->gamma_table[palette[i].blue]; } } + /* Prevent the transformations being done again, and make sure + * that the now spurious alpha channel is stripped - the code + * has just reduced background composition and gamma correction + * to a simple alpha channel strip. + */ + png_ptr->transformations &= ~PNG_BACKGROUND; + png_ptr->transformations &= ~PNG_GAMMA; + png_ptr->transformations |= PNG_STRIP_ALPHA; } /* if (png_ptr->background_gamma_type!=PNG_BACKGROUND_GAMMA_UNKNOWN) */ else @@ -995,6 +1036,9 @@ png_init_read_transformations(png_structp png_ptr) palette[i].green = png_ptr->gamma_table[palette[i].green]; palette[i].blue = png_ptr->gamma_table[palette[i].blue]; } + + /* Done the gamma correction. */ + png_ptr->transformations &= ~PNG_GAMMA; } } #if defined(PNG_READ_BACKGROUND_SUPPORTED) @@ -1032,6 +1076,10 @@ png_init_read_transformations(png_structp png_ptr) png_ptr->trans[i], back.blue); } } + + /* Handled alpha, still need to strip the channel. */ + png_ptr->transformations &= ~PNG_BACKGROUND; + png_ptr->transformations |= PNG_STRIP_ALPHA; } #endif /* PNG_READ_BACKGROUND_SUPPORTED */ @@ -1080,7 +1128,8 @@ png_read_transform_info(png_structp png_ptr, png_infop info_ptr) { if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) { - if (png_ptr->num_trans && (png_ptr->transformations & PNG_EXPAND_tRNS)) + if (png_ptr->num_trans && + (png_ptr->transformations & PNG_EXPAND_tRNS)) info_ptr->color_type = PNG_COLOR_TYPE_RGB_ALPHA; else info_ptr->color_type = PNG_COLOR_TYPE_RGB; @@ -1093,8 +1142,10 @@ png_read_transform_info(png_structp png_ptr, png_infop info_ptr) { if (png_ptr->transformations & PNG_EXPAND_tRNS) info_ptr->color_type |= PNG_COLOR_MASK_ALPHA; +#if 0 /* Removed from libpng-1.2.27 */ else info_ptr->color_type |= PNG_COLOR_MASK_COLOR; +#endif } if (info_ptr->bit_depth < 8) info_ptr->bit_depth = 8; @@ -1216,19 +1267,29 @@ void /* PRIVATE */ png_do_read_transformations(png_structp png_ptr) { png_debug(1, "in png_do_read_transformations\n"); -#if !defined(PNG_USELESS_TESTS_SUPPORTED) if (png_ptr->row_buf == NULL) { #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) char msg[50]; - sprintf(msg, "NULL row buffer for row %ld, pass %d", png_ptr->row_number, + png_snprintf2(msg, 50, + "NULL row buffer for row %ld, pass %d", png_ptr->row_number, png_ptr->pass); png_error(png_ptr, msg); #else png_error(png_ptr, "NULL row buffer"); #endif } +#ifdef PNG_WARN_UNINITIALIZED_ROW + if (!(png_ptr->flags & PNG_FLAG_ROW_INIT)) + /* Application has failed to call either png_read_start_image() + * or png_read_update_info() after setting transforms that expand + * pixels. This check added to libpng-1.2.19 */ +#if (PNG_WARN_UNINITIALIZED_ROW==1) + png_error(png_ptr, "Uninitialized row"); +#else + png_warning(png_ptr, "Uninitialized row"); +#endif #endif #if defined(PNG_READ_EXPAND_SUPPORTED) @@ -1241,7 +1302,8 @@ png_do_read_transformations(png_structp png_ptr) } else { - if (png_ptr->num_trans && (png_ptr->transformations & PNG_EXPAND_tRNS)) + if (png_ptr->num_trans && + (png_ptr->transformations & PNG_EXPAND_tRNS)) png_do_expand(&(png_ptr->row_info), png_ptr->row_buf + 1, &(png_ptr->trans_values)); else @@ -1265,9 +1327,11 @@ png_do_read_transformations(png_structp png_ptr) if(rgb_error) { png_ptr->rgb_to_gray_status=1; - if(png_ptr->transformations == PNG_RGB_TO_GRAY_WARN) + if((png_ptr->transformations & PNG_RGB_TO_GRAY) == + PNG_RGB_TO_GRAY_WARN) png_warning(png_ptr, "png_do_rgb_to_gray found nongray pixel"); - if(png_ptr->transformations == PNG_RGB_TO_GRAY_ERR) + if((png_ptr->transformations & PNG_RGB_TO_GRAY) == + PNG_RGB_TO_GRAY_ERR) png_error(png_ptr, "png_do_rgb_to_gray found nongray pixel"); } } @@ -3671,7 +3735,7 @@ png_do_expand(png_row_infop row_info, png_bytep row, { case 1: { - gray = (png_uint_16)(gray*0xff); + gray = (png_uint_16)((gray&0x01)*0xff); sp = row + (png_size_t)((row_width - 1) >> 3); dp = row + (png_size_t)row_width - 1; shift = 7 - (int)((row_width + 7) & 0x07); @@ -3695,7 +3759,7 @@ png_do_expand(png_row_infop row_info, png_bytep row, } case 2: { - gray = (png_uint_16)(gray*0x55); + gray = (png_uint_16)((gray&0x03)*0x55); sp = row + (png_size_t)((row_width - 1) >> 2); dp = row + (png_size_t)row_width - 1; shift = (int)((3 - ((row_width + 3) & 0x03)) << 1); @@ -3718,7 +3782,7 @@ png_do_expand(png_row_infop row_info, png_bytep row, } case 4: { - gray = (png_uint_16)(gray*0x11); + gray = (png_uint_16)((gray&0x0f)*0x11); sp = row + (png_size_t)((row_width - 1) >> 1); dp = row + (png_size_t)row_width - 1; shift = (int)((1 - ((row_width + 1) & 0x01)) << 2); @@ -3748,6 +3812,7 @@ png_do_expand(png_row_infop row_info, png_bytep row, { if (row_info->bit_depth == 8) { + gray = gray & 0xff; sp = row + (png_size_t)row_width - 1; dp = row + (png_size_t)(row_width << 1) - 1; for (i = 0; i < row_width; i++) @@ -3761,12 +3826,13 @@ png_do_expand(png_row_infop row_info, png_bytep row, } else if (row_info->bit_depth == 16) { + png_byte gray_high = (gray >> 8) & 0xff; + png_byte gray_low = gray & 0xff; sp = row + row_info->rowbytes - 1; dp = row + (row_info->rowbytes << 1) - 1; for (i = 0; i < row_width; i++) { - if (((png_uint_16)*(sp) | - ((png_uint_16)*(sp - 1) << 8)) == gray) + if (*(sp-1) == gray_high && *(sp) == gray_low) { *dp-- = 0; *dp-- = 0; @@ -3791,13 +3857,14 @@ png_do_expand(png_row_infop row_info, png_bytep row, { if (row_info->bit_depth == 8) { + png_byte red = trans_value->red & 0xff; + png_byte green = trans_value->green & 0xff; + png_byte blue = trans_value->blue & 0xff; sp = row + (png_size_t)row_info->rowbytes - 1; dp = row + (png_size_t)(row_width << 2) - 1; for (i = 0; i < row_width; i++) { - if (*(sp - 2) == trans_value->red && - *(sp - 1) == trans_value->green && - *(sp - 0) == trans_value->blue) + if (*(sp - 2) == red && *(sp - 1) == green && *(sp) == blue) *dp-- = 0; else *dp-- = 0xff; @@ -3808,16 +3875,22 @@ png_do_expand(png_row_infop row_info, png_bytep row, } else if (row_info->bit_depth == 16) { + png_byte red_high = (trans_value->red >> 8) & 0xff; + png_byte green_high = (trans_value->green >> 8) & 0xff; + png_byte blue_high = (trans_value->blue >> 8) & 0xff; + png_byte red_low = trans_value->red & 0xff; + png_byte green_low = trans_value->green & 0xff; + png_byte blue_low = trans_value->blue & 0xff; sp = row + row_info->rowbytes - 1; dp = row + (png_size_t)(row_width << 3) - 1; for (i = 0; i < row_width; i++) { - if ((((png_uint_16)*(sp - 4) | - ((png_uint_16)*(sp - 5) << 8)) == trans_value->red) && - (((png_uint_16)*(sp - 2) | - ((png_uint_16)*(sp - 3) << 8)) == trans_value->green) && - (((png_uint_16)*(sp - 0) | - ((png_uint_16)*(sp - 1) << 8)) == trans_value->blue)) + if (*(sp - 5) == red_high && + *(sp - 4) == red_low && + *(sp - 3) == green_high && + *(sp - 2) == green_low && + *(sp - 1) == blue_high && + *(sp ) == blue_low) { *dp-- = 0; *dp-- = 0; @@ -3937,8 +4010,8 @@ png_do_dither(png_row_infop row_info, png_bytep row, #ifdef PNG_FLOATING_POINT_SUPPORTED #if defined(PNG_READ_GAMMA_SUPPORTED) -static int png_gamma_shift[] = - {0x10, 0x21, 0x42, 0x84, 0x110, 0x248, 0x550, 0xff0}; +static PNG_CONST int png_gamma_shift[] = + {0x10, 0x21, 0x42, 0x84, 0x110, 0x248, 0x550, 0xff0, 0x00}; /* We build the 8- or 16-bit gamma tables here. Note that for 16-bit * tables, we don't make a full table if we are reducing to 8-bit in diff --git a/pngrutil.c b/pngrutil.c index 2bde4b416..531cb0570 100644 --- a/pngrutil.c +++ b/pngrutil.c @@ -1,9 +1,9 @@ /* pngrutil.c - utilities to read a PNG file * - * Last changed in libpng 1.2.11 June 4, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -16,27 +16,33 @@ #if defined(PNG_READ_SUPPORTED) -#if defined(_WIN32_WCE) +#if defined(_WIN32_WCE) && (_WIN32_WCE<0x500) +# define WIN32_WCE_OLD +#endif + +#ifdef PNG_FLOATING_POINT_SUPPORTED +# if defined(WIN32_WCE_OLD) /* strtod() function is not supported on WindowsCE */ -# ifdef PNG_FLOATING_POINT_SUPPORTED -__inline double strtod(const char *nptr, char **endptr) +__inline double png_strtod(png_structp png_ptr, PNG_CONST char *nptr, char **endptr) { double result = 0; int len; wchar_t *str, *end; len = MultiByteToWideChar(CP_ACP, 0, nptr, -1, NULL, 0); - str = (wchar_t *)malloc(len * sizeof(wchar_t)); + str = (wchar_t *)png_malloc(png_ptr, len * sizeof(wchar_t)); if ( NULL != str ) { MultiByteToWideChar(CP_ACP, 0, nptr, -1, str, len); result = wcstod(str, &end); len = WideCharToMultiByte(CP_ACP, 0, end, -1, NULL, 0, NULL, NULL); *endptr = (char *)nptr + (png_strlen(nptr) - len + 1); - free(str); + png_free(png_ptr, str); } return result; } +# else +# define png_strtod(p,a,b) strtod(a,b) # endif #endif @@ -90,6 +96,7 @@ png_get_uint_16(png_bytep buf) void /* PRIVATE */ png_crc_read(png_structp png_ptr, png_bytep buf, png_size_t length) { + if(png_ptr == NULL) return; png_read_data(png_ptr, buf, length); png_calculate_crc(png_ptr, buf, length); } @@ -178,7 +185,7 @@ png_decompress_chunk(png_structp png_ptr, int comp_type, png_charp chunkdata, png_size_t chunklength, png_size_t prefix_size, png_size_t *newlength) { - static char msg[] = "Error decoding compressed text"; + static PNG_CONST char msg[] = "Error decoding compressed text"; png_charp text; png_size_t text_size; @@ -223,7 +230,7 @@ png_decompress_chunk(png_structp png_ptr, int comp_type, text_size = (png_size_t)(chunklength - (text - chunkdata) - 1); text_size = png_sizeof(msg) > text_size ? text_size : png_sizeof(msg); - png_memcpy(text + prefix_size, msg, text_size + 1); + png_memcpy(text + prefix_size, msg, text_size); break; } if (!png_ptr->zstream.avail_out || ret == Z_STREAM_END) @@ -276,16 +283,19 @@ png_decompress_chunk(png_structp png_ptr, int comp_type, if (ret != Z_STREAM_END) { #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) - char umsg[50]; + char umsg[52]; if (ret == Z_BUF_ERROR) - sprintf(umsg,"Buffer error in compressed datastream in %s chunk", + png_snprintf(umsg, 52, + "Buffer error in compressed datastream in %s chunk", png_ptr->chunk_name); else if (ret == Z_DATA_ERROR) - sprintf(umsg,"Data error in compressed datastream in %s chunk", + png_snprintf(umsg, 52, + "Data error in compressed datastream in %s chunk", png_ptr->chunk_name); else - sprintf(umsg,"Incomplete compressed datastream in %s chunk", + png_snprintf(umsg, 52, + "Incomplete compressed datastream in %s chunk", png_ptr->chunk_name); png_warning(png_ptr, umsg); #else @@ -318,7 +328,8 @@ png_decompress_chunk(png_structp png_ptr, int comp_type, #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) char umsg[50]; - sprintf(umsg, "Unknown zTXt compression type %d", comp_type); + png_snprintf(umsg, 50, + "Unknown zTXt compression type %d", comp_type); png_warning(png_ptr, umsg); #else png_warning(png_ptr, "Unknown zTXt compression type"); @@ -560,8 +571,7 @@ png_handle_IEND(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) } png_crc_finish(png_ptr, length); - if (&info_ptr == NULL) /* quiet compiler warnings about unused info_ptr */ - return; + info_ptr =info_ptr; /* quiet compiler warnings about unused info_ptr */ } #if defined(PNG_READ_gAMA_SUPPORTED) @@ -1034,7 +1044,7 @@ png_handle_iCCP(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) /* there should be at least one zero (the compression type byte) following the separator, and we should be on it */ - if ( profile >= chunkdata + slength) + if ( profile >= chunkdata + slength - 1) { png_free(png_ptr, chunkdata); png_warning(png_ptr, "Malformed iCCP chunk"); @@ -1138,7 +1148,7 @@ png_handle_sPLT(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) ++entry_start; /* a sample depth should follow the separator, and we should be on it */ - if (entry_start > chunkdata + slength) + if (entry_start > chunkdata + slength - 2) { png_free(png_ptr, chunkdata); png_warning(png_ptr, "malformed sPLT chunk"); @@ -1311,7 +1321,10 @@ png_handle_tRNS(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) } if (png_crc_finish(png_ptr, 0)) + { + png_ptr->num_trans = 0; return; + } png_set_tRNS(png_ptr, info_ptr, readbuf, png_ptr->num_trans, &(png_ptr->trans_values)); @@ -1374,7 +1387,7 @@ png_handle_bKGD(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) { png_ptr->background.index = buf[0]; - if(info_ptr->num_palette) + if (info_ptr && info_ptr->num_palette) { if(buf[0] > info_ptr->num_palette) { @@ -1654,7 +1667,7 @@ png_handle_pCAL(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) buf++; /* Skip the null string terminator from previous parameter. */ png_debug1(3, "Reading pCAL parameter %d\n", i); - for (params[i] = buf; *buf != 0x00 && buf <= endptr; buf++) + for (params[i] = buf; buf <= endptr && *buf != 0x00; buf++) /* Empty loop to move past each parameter string */ ; /* Make sure we haven't run out of data yet */ @@ -1730,7 +1743,7 @@ png_handle_sCAL(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) ep = buffer + 1; /* skip unit byte */ #ifdef PNG_FLOATING_POINT_SUPPORTED - width = strtod(ep, &vp); + width = png_strtod(png_ptr, ep, &vp); if (*vp) { png_warning(png_ptr, "malformed width string in sCAL chunk"); @@ -1752,8 +1765,19 @@ png_handle_sCAL(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) /* empty loop */ ; ep++; + if (buffer + slength < ep) + { + png_warning(png_ptr, "Truncated sCAL chunk"); +#if defined(PNG_FIXED_POINT_SUPPORTED) && \ + !defined(PNG_FLOATING_POINT_SUPPORTED) + png_free(png_ptr, swidth); +#endif + png_free(png_ptr, buffer); + return; + } + #ifdef PNG_FLOATING_POINT_SUPPORTED - height = strtod(ep, &vp); + height = png_strtod(png_ptr, ep, &vp); if (*vp) { png_warning(png_ptr, "malformed height string in sCAL chunk"); @@ -1762,7 +1786,7 @@ png_handle_sCAL(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) #else #ifdef PNG_FIXED_POINT_SUPPORTED sheight = (png_charp)png_malloc_warn(png_ptr, png_strlen(ep) + 1); - if (swidth == NULL) + if (sheight == NULL) { png_warning(png_ptr, "Out of memory while processing sCAL chunk height"); return; @@ -1975,10 +1999,11 @@ png_handle_zTXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) /* empty loop */ ; /* zTXt must have some text after the chunkdataword */ - if (text == chunkdata + slength) + if (text >= chunkdata + slength - 2) { - comp_type = PNG_TEXT_COMPRESSION_NONE; - png_warning(png_ptr, "Zero length zTXt chunk"); + png_warning(png_ptr, "Truncated zTXt chunk"); + png_free(png_ptr, chunkdata); + return; } else { @@ -2078,10 +2103,11 @@ png_handle_iTXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) translated keyword (possibly empty), and possibly some text after the keyword */ - if (lang >= chunkdata + slength) + if (lang >= chunkdata + slength - 3) { - comp_flag = PNG_TEXT_COMPRESSION_NONE; - png_warning(png_ptr, "Zero length iTXt chunk"); + png_warning(png_ptr, "Truncated iTXt chunk"); + png_free(png_ptr, chunkdata); + return; } else { @@ -2093,9 +2119,22 @@ png_handle_iTXt(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) /* empty loop */ ; lang_key++; /* skip NUL separator */ + if (lang_key >= chunkdata + slength) + { + png_warning(png_ptr, "Truncated iTXt chunk"); + png_free(png_ptr, chunkdata); + return; + } + for (text = lang_key; *text; text++) /* empty loop */ ; text++; /* skip NUL separator */ + if (text >= chunkdata + slength) + { + png_warning(png_ptr, "Malformed iTXt chunk"); + png_free(png_ptr, chunkdata); + return; + } prefix_len = text - chunkdata; @@ -2145,7 +2184,7 @@ png_handle_unknown(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) if (png_ptr->mode & PNG_HAVE_IDAT) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IDAT; + PNG_CONST PNG_IDAT; #endif if (png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) /* not an IDAT */ png_ptr->mode |= PNG_AFTER_IDAT; @@ -2170,8 +2209,6 @@ png_handle_unknown(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) if ((png_ptr->flags & PNG_FLAG_KEEP_UNKNOWN_CHUNKS) || (png_ptr->read_user_chunk_fn != NULL)) { - png_unknown_chunk chunk; - #ifdef PNG_MAX_MALLOC_64K if (length > (png_uint_32)65535L) { @@ -2180,30 +2217,42 @@ png_handle_unknown(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) length = (png_uint_32)65535L; } #endif - png_strcpy((png_charp)chunk.name, (png_charp)png_ptr->chunk_name); - chunk.data = (png_bytep)png_malloc(png_ptr, length); - chunk.size = (png_size_t)length; - png_crc_read(png_ptr, (png_bytep)chunk.data, length); + png_memcpy((png_charp)png_ptr->unknown_chunk.name, + (png_charp)png_ptr->chunk_name, + png_sizeof(png_ptr->unknown_chunk.name)); + png_ptr->unknown_chunk.name[png_sizeof(png_ptr->unknown_chunk.name)-1] = '\0'; + png_ptr->unknown_chunk.size = (png_size_t)length; + if (length == 0) + png_ptr->unknown_chunk.data = NULL; + else + { + png_ptr->unknown_chunk.data = (png_bytep)png_malloc(png_ptr, length); + png_crc_read(png_ptr, (png_bytep)png_ptr->unknown_chunk.data, length); + } #if defined(PNG_READ_USER_CHUNKS_SUPPORTED) if(png_ptr->read_user_chunk_fn != NULL) { /* callback to user unknown chunk handler */ - if ((*(png_ptr->read_user_chunk_fn)) (png_ptr, &chunk) <= 0) + int ret; + ret = (*(png_ptr->read_user_chunk_fn)) + (png_ptr, &png_ptr->unknown_chunk); + if (ret < 0) + png_chunk_error(png_ptr, "error in user chunk"); + if (ret == 0) { if (!(png_ptr->chunk_name[0] & 0x20)) if(png_handle_as_unknown(png_ptr, png_ptr->chunk_name) != PNG_HANDLE_CHUNK_ALWAYS) - { - png_free(png_ptr, chunk.data); png_chunk_error(png_ptr, "unknown critical chunk"); - } - png_set_unknown_chunks(png_ptr, info_ptr, &chunk, 1); + png_set_unknown_chunks(png_ptr, info_ptr, + &png_ptr->unknown_chunk, 1); } } else #endif - png_set_unknown_chunks(png_ptr, info_ptr, &chunk, 1); - png_free(png_ptr, chunk.data); + png_set_unknown_chunks(png_ptr, info_ptr, &png_ptr->unknown_chunk, 1); + png_free(png_ptr, png_ptr->unknown_chunk.data); + png_ptr->unknown_chunk.data = NULL; } else #endif @@ -2212,8 +2261,7 @@ png_handle_unknown(png_structp png_ptr, png_infop info_ptr, png_uint_32 length) png_crc_finish(png_ptr, skip); #if !defined(PNG_READ_USER_CHUNKS_SUPPORTED) - if (&info_ptr == NULL) /* quiet compiler warnings about unused info_ptr */ - return; + info_ptr = info_ptr; /* quiet compiler warnings about unused info_ptr */ #endif } @@ -2246,7 +2294,7 @@ png_check_chunk_name(png_structp png_ptr, png_bytep chunk_name) a zero indicates the pixel is to be skipped. This is in addition to any alpha or transparency value associated with the pixel. If you want all pixels to be combined, pass 0xff (255) in mask. */ -#ifndef PNG_HAVE_ASSEMBLER_COMBINE_ROW + void /* PRIVATE */ png_combine_row(png_structp png_ptr, png_bytep row, int mask) { @@ -2447,10 +2495,8 @@ png_combine_row(png_structp png_ptr, png_bytep row, int mask) } } } -#endif /* !PNG_HAVE_ASSEMBLER_COMBINE_ROW */ #ifdef PNG_READ_INTERLACING_SUPPORTED -#ifndef PNG_HAVE_ASSEMBLER_READ_INTERLACE /* else in pngvcrd.c, pnggccrd.c */ /* OLD pre-1.0.9 interface: void png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass, png_uint_32 transformations) @@ -2465,10 +2511,10 @@ png_do_read_interlace(png_structp png_ptr) #ifdef PNG_USE_LOCAL_ARRAYS /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* offset to next interlace block */ - const int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; + PNG_CONST int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; #endif - png_debug(1,"in png_do_read_interlace (stock C version)\n"); + png_debug(1,"in png_do_read_interlace\n"); if (row != NULL && row_info != NULL) { png_uint_32 final_width; @@ -2672,14 +2718,11 @@ png_do_read_interlace(png_structp png_ptr) row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth,final_width); } #if !defined(PNG_READ_PACKSWAP_SUPPORTED) - if (&transformations == NULL) /* silence compiler warning */ - return; + transformations = transformations; /* silence compiler warning */ #endif } -#endif /* !PNG_HAVE_ASSEMBLER_READ_INTERLACE */ #endif /* PNG_READ_INTERLACING_SUPPORTED */ -#ifndef PNG_HAVE_ASSEMBLER_READ_FILTER_ROW void /* PRIVATE */ png_read_filter_row(png_structp png_ptr, png_row_infop row_info, png_bytep row, png_bytep prev_row, int filter) @@ -2802,25 +2845,26 @@ png_read_filter_row(png_structp png_ptr, png_row_infop row_info, png_bytep row, break; } } -#endif /* !PNG_HAVE_ASSEMBLER_READ_FILTER_ROW */ void /* PRIVATE */ png_read_finish_row(png_structp png_ptr) { #ifdef PNG_USE_LOCAL_ARRAYS +#ifdef PNG_READ_INTERLACING_SUPPORTED /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* start of interlace block */ - const int png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; + PNG_CONST int png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; /* offset to next interlace block */ - const int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; + PNG_CONST int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; /* start of interlace block in the y direction */ - const int png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; + PNG_CONST int png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; /* offset to next interlace block in the y direction */ - const int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; + PNG_CONST int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; +#endif /* PNG_READ_INTERLACING_SUPPORTED */ #endif png_debug(1, "in png_read_finish_row\n"); @@ -2828,10 +2872,12 @@ png_read_finish_row(png_structp png_ptr) if (png_ptr->row_number < png_ptr->num_rows) return; +#ifdef PNG_READ_INTERLACING_SUPPORTED if (png_ptr->interlaced) { png_ptr->row_number = 0; - png_memset_check(png_ptr, png_ptr->prev_row, 0, png_ptr->rowbytes + 1); + png_memset_check(png_ptr, png_ptr->prev_row, 0, + png_ptr->rowbytes + 1); do { png_ptr->pass++; @@ -2861,11 +2907,12 @@ png_read_finish_row(png_structp png_ptr) if (png_ptr->pass < 7) return; } +#endif /* PNG_READ_INTERLACING_SUPPORTED */ if (!(png_ptr->flags & PNG_FLAG_ZLIB_FINISHED)) { #ifdef PNG_USE_LOCAL_ARRAYS - PNG_IDAT; + PNG_CONST PNG_IDAT; #endif char extra; int ret; @@ -2886,7 +2933,7 @@ png_read_finish_row(png_structp png_ptr) png_ptr->idat_size = png_get_uint_31(png_ptr, chunk_length); png_reset_crc(png_ptr); png_crc_read(png_ptr, png_ptr->chunk_name, 4); - if (png_memcmp(png_ptr->chunk_name, (png_bytep)png_IDAT, 4)) + if (png_memcmp(png_ptr->chunk_name, png_IDAT, 4)) png_error(png_ptr, "Not enough image data"); } @@ -2935,19 +2982,21 @@ void /* PRIVATE */ png_read_start_row(png_structp png_ptr) { #ifdef PNG_USE_LOCAL_ARRAYS +#ifdef PNG_READ_INTERLACING_SUPPORTED /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ /* start of interlace block */ - const int png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; + PNG_CONST int png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; /* offset to next interlace block */ - const int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; + PNG_CONST int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; /* start of interlace block in the y direction */ - const int png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; + PNG_CONST int png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; /* offset to next interlace block in the y direction */ - const int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; + PNG_CONST int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; +#endif #endif int max_pixel_depth; @@ -2956,6 +3005,7 @@ png_read_start_row(png_structp png_ptr) png_debug(1, "in png_read_start_row\n"); png_ptr->zstream.avail_in = 0; png_init_read_transformations(png_ptr); +#ifdef PNG_READ_INTERLACING_SUPPORTED if (png_ptr->interlaced) { if (!(png_ptr->transformations & PNG_INTERLACE)) @@ -2976,6 +3026,7 @@ png_read_start_row(png_structp png_ptr) png_error(png_ptr, "Rowbytes overflow in png_read_start_row"); } else +#endif /* PNG_READ_INTERLACING_SUPPORTED */ { png_ptr->num_rows = png_ptr->height; png_ptr->iwidth = png_ptr->width; @@ -3094,20 +3145,29 @@ defined(PNG_USER_TRANSFORM_PTR_SUPPORTED) if (row_bytes > (png_uint_32)65536L) png_error(png_ptr, "This image requires a row greater than 64KB"); #endif - png_ptr->big_row_buf = (png_bytep)png_malloc(png_ptr, row_bytes+64); - png_ptr->row_buf = png_ptr->big_row_buf+32; -#if defined(PNG_DEBUG) && defined(PNG_USE_PNGGCCRD) - png_ptr->row_buf_size = row_bytes; -#endif + + if(row_bytes + 64 > png_ptr->old_big_row_buf_size) + { + png_free(png_ptr,png_ptr->big_row_buf); + png_ptr->big_row_buf = (png_bytep)png_malloc(png_ptr, row_bytes+64); + png_ptr->row_buf = png_ptr->big_row_buf+32; + png_ptr->old_big_row_buf_size = row_bytes+64; + } #ifdef PNG_MAX_MALLOC_64K if ((png_uint_32)png_ptr->rowbytes + 1 > (png_uint_32)65536L) png_error(png_ptr, "This image requires a row greater than 64KB"); #endif - if ((png_uint_32)png_ptr->rowbytes > PNG_SIZE_MAX - 1) + if ((png_uint_32)png_ptr->rowbytes > (png_uint_32)(PNG_SIZE_MAX - 1)) png_error(png_ptr, "Row has too many bytes to allocate in memory."); - png_ptr->prev_row = (png_bytep)png_malloc(png_ptr, (png_uint_32)( - png_ptr->rowbytes + 1)); + + if(png_ptr->rowbytes+1 > png_ptr->old_prev_row_size) + { + png_free(png_ptr,png_ptr->prev_row); + png_ptr->prev_row = (png_bytep)png_malloc(png_ptr, (png_uint_32)( + png_ptr->rowbytes + 1)); + png_ptr->old_prev_row_size = png_ptr->rowbytes+1; + } png_memset_check(png_ptr, png_ptr->prev_row, 0, png_ptr->rowbytes + 1); diff --git a/pngset.c b/pngset.c index 9e499c9bc..8b25ca504 100644 --- a/pngset.c +++ b/pngset.c @@ -1,9 +1,9 @@ /* pngset.c - storage of image information into info struct * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -41,7 +41,13 @@ png_set_cHRM(png_structp png_ptr, png_infop info_ptr, png_debug1(1, "in %s storage function\n", "cHRM"); if (png_ptr == NULL || info_ptr == NULL) return; - + if (!(white_x || white_y || red_x || red_y || green_x || green_y || + blue_x || blue_y)) + { + png_warning(png_ptr, + "Ignoring attempt to set all-zero chromaticity values"); + return; + } if (white_x < 0.0 || white_y < 0.0 || red_x < 0.0 || red_y < 0.0 || green_x < 0.0 || green_y < 0.0 || @@ -93,6 +99,13 @@ png_set_cHRM_fixed(png_structp png_ptr, png_infop info_ptr, if (png_ptr == NULL || info_ptr == NULL) return; + if (!(white_x || white_y || red_x || red_y || green_x || green_y || + blue_x || blue_y)) + { + png_warning(png_ptr, + "Ignoring attempt to set all-zero chromaticity values"); + return; + } if (white_x < 0 || white_y < 0 || red_x < 0 || red_y < 0 || green_x < 0 || green_y < 0 || @@ -102,25 +115,14 @@ png_set_cHRM_fixed(png_structp png_ptr, png_infop info_ptr, "Ignoring attempt to set negative chromaticity value"); return; } -#ifdef PNG_FLOATING_POINT_SUPPORTED - if (white_x > (double) PNG_UINT_31_MAX || - white_y > (double) PNG_UINT_31_MAX || - red_x > (double) PNG_UINT_31_MAX || - red_y > (double) PNG_UINT_31_MAX || - green_x > (double) PNG_UINT_31_MAX || - green_y > (double) PNG_UINT_31_MAX || - blue_x > (double) PNG_UINT_31_MAX || - blue_y > (double) PNG_UINT_31_MAX) -#else - if (white_x > (png_fixed_point) PNG_UINT_31_MAX/100000L || - white_y > (png_fixed_point) PNG_UINT_31_MAX/100000L || - red_x > (png_fixed_point) PNG_UINT_31_MAX/100000L || - red_y > (png_fixed_point) PNG_UINT_31_MAX/100000L || - green_x > (png_fixed_point) PNG_UINT_31_MAX/100000L || - green_y > (png_fixed_point) PNG_UINT_31_MAX/100000L || - blue_x > (png_fixed_point) PNG_UINT_31_MAX/100000L || - blue_y > (png_fixed_point) PNG_UINT_31_MAX/100000L) -#endif + if (white_x > (png_fixed_point) PNG_UINT_31_MAX || + white_y > (png_fixed_point) PNG_UINT_31_MAX || + red_x > (png_fixed_point) PNG_UINT_31_MAX || + red_y > (png_fixed_point) PNG_UINT_31_MAX || + green_x > (png_fixed_point) PNG_UINT_31_MAX || + green_y > (png_fixed_point) PNG_UINT_31_MAX || + blue_x > (png_fixed_point) PNG_UINT_31_MAX || + blue_y > (png_fixed_point) PNG_UINT_31_MAX ) { png_warning(png_ptr, "Ignoring attempt to set chromaticity value exceeding 21474.83"); @@ -222,7 +224,7 @@ png_set_hIST(png_structp png_ptr, png_infop info_ptr, png_uint_16p hist) png_debug1(1, "in %s storage function\n", "hIST"); if (png_ptr == NULL || info_ptr == NULL) return; - if (info_ptr->num_palette <= 0 || info_ptr->num_palette + if (info_ptr->num_palette == 0 || info_ptr->num_palette > PNG_MAX_PALETTE_LENGTH) { png_warning(png_ptr, @@ -353,7 +355,7 @@ png_set_IHDR(png_structp png_ptr, png_infop info_ptr, info_ptr->pixel_depth = (png_byte)(info_ptr->channels * info_ptr->bit_depth); /* check for potential overflow */ - if ( width > (PNG_UINT_32_MAX + if (width > (PNG_UINT_32_MAX >> 3) /* 8-byte RGBA pixels */ - 64 /* bigrowbuf hack */ - 1 /* filter byte */ @@ -484,7 +486,9 @@ png_set_sCAL_s(png_structp png_ptr, png_infop info_ptr, info_ptr->scal_s_width = (png_charp)png_malloc_warn(png_ptr, length); if (info_ptr->scal_s_width == NULL) { - png_warning(png_ptr, "Memory allocation failed while processing sCAL."); + png_warning(png_ptr, + "Memory allocation failed while processing sCAL."); + return; } png_memcpy(info_ptr->scal_s_width, swidth, (png_size_t)length); @@ -494,10 +498,11 @@ png_set_sCAL_s(png_structp png_ptr, png_infop info_ptr, if (info_ptr->scal_s_height == NULL) { png_free (png_ptr, info_ptr->scal_s_width); - png_warning(png_ptr, "Memory allocation failed while processing sCAL."); + png_warning(png_ptr, + "Memory allocation failed while processing sCAL."); + return; } png_memcpy(info_ptr->scal_s_height, sheight, (png_size_t)length); - info_ptr->valid |= PNG_INFO_sCAL; #ifdef PNG_FREE_ME_SUPPORTED info_ptr->free_me |= PNG_FREE_SCAL; @@ -677,18 +682,20 @@ png_set_iCCP(png_structp png_ptr, png_infop info_ptr, { png_charp new_iccp_name; png_charp new_iccp_profile; + png_uint_32 length; png_debug1(1, "in %s storage function\n", "iCCP"); if (png_ptr == NULL || info_ptr == NULL || name == NULL || profile == NULL) return; - new_iccp_name = (png_charp)png_malloc_warn(png_ptr, png_strlen(name)+1); + length = png_strlen(name)+1; + new_iccp_name = (png_charp)png_malloc_warn(png_ptr, length); if (new_iccp_name == NULL) { png_warning(png_ptr, "Insufficient memory to process iCCP chunk."); return; } - png_strcpy(new_iccp_name, name); + png_memcpy(new_iccp_name, name, length); new_iccp_profile = (png_charp)png_malloc_warn(png_ptr, proflen); if (new_iccp_profile == NULL) { @@ -877,7 +884,6 @@ png_set_text_2(png_structp png_ptr, png_infop info_ptr, png_textp text_ptr, textp->itxt_length = 0; #endif } - info_ptr->text[info_ptr->num_text]= *textp; info_ptr->num_text++; png_debug1(3, "transferred text chunk %d\n", info_ptr->num_text); } @@ -908,6 +914,8 @@ png_set_tRNS(png_structp png_ptr, png_infop info_ptr, if (png_ptr == NULL || info_ptr == NULL) return; + png_free_data(png_ptr, info_ptr, PNG_FREE_TRNS, 0); + if (trans != NULL) { /* @@ -915,30 +923,41 @@ png_set_tRNS(png_structp png_ptr, png_infop info_ptr, * we do it for backward compatibility with the way the png_handle_tRNS * function used to do the allocation. */ -#ifdef PNG_FREE_ME_SUPPORTED - png_free_data(png_ptr, info_ptr, PNG_FREE_TRNS, 0); -#endif + /* Changed from num_trans to PNG_MAX_PALETTE_LENGTH in version 1.2.1 */ png_ptr->trans = info_ptr->trans = (png_bytep)png_malloc(png_ptr, (png_uint_32)PNG_MAX_PALETTE_LENGTH); - if (num_trans <= PNG_MAX_PALETTE_LENGTH) + if (num_trans > 0 && num_trans <= PNG_MAX_PALETTE_LENGTH) png_memcpy(info_ptr->trans, trans, (png_size_t)num_trans); -#ifdef PNG_FREE_ME_SUPPORTED - info_ptr->free_me |= PNG_FREE_TRNS; -#else - png_ptr->flags |= PNG_FLAG_FREE_TRNS; -#endif } if (trans_values != NULL) { + int sample_max = (1 << info_ptr->bit_depth); + if ((info_ptr->color_type == PNG_COLOR_TYPE_GRAY && + (int)trans_values->gray > sample_max) || + (info_ptr->color_type == PNG_COLOR_TYPE_RGB && + ((int)trans_values->red > sample_max || + (int)trans_values->green > sample_max || + (int)trans_values->blue > sample_max))) + png_warning(png_ptr, + "tRNS chunk has out-of-range samples for bit_depth"); png_memcpy(&(info_ptr->trans_values), trans_values, png_sizeof(png_color_16)); if (num_trans == 0) num_trans = 1; } + info_ptr->num_trans = (png_uint_16)num_trans; - info_ptr->valid |= PNG_INFO_tRNS; + if (num_trans != 0) + { + info_ptr->valid |= PNG_INFO_tRNS; +#ifdef PNG_FREE_ME_SUPPORTED + info_ptr->free_me |= PNG_FREE_TRNS; +#else + png_ptr->flags |= PNG_FLAG_FREE_TRNS; +#endif + } } #endif @@ -970,16 +989,29 @@ png_set_sPLT(png_structp png_ptr, { png_sPLT_tp to = np + info_ptr->splt_palettes_num + i; png_sPLT_tp from = entries + i; + png_uint_32 length; - to->name = (png_charp)png_malloc(png_ptr, - png_strlen(from->name) + 1); - /* TODO: use png_malloc_warn */ - png_strcpy(to->name, from->name); - to->entries = (png_sPLT_entryp)png_malloc(png_ptr, - from->nentries * png_sizeof(png_sPLT_t)); - /* TODO: use png_malloc_warn */ + length = png_strlen(from->name) + 1; + to->name = (png_charp)png_malloc_warn(png_ptr, length); + if (to->name == NULL) + { + png_warning(png_ptr, + "Out of memory while processing sPLT chunk"); + continue; + } + png_memcpy(to->name, from->name, length); + to->entries = (png_sPLT_entryp)png_malloc_warn(png_ptr, + from->nentries * png_sizeof(png_sPLT_entry)); + if (to->entries == NULL) + { + png_warning(png_ptr, + "Out of memory while processing sPLT chunk"); + png_free(png_ptr,to->name); + to->name = NULL; + continue; + } png_memcpy(to->entries, from->entries, - from->nentries * png_sizeof(png_sPLT_t)); + from->nentries * png_sizeof(png_sPLT_entry)); to->nentries = from->nentries; to->depth = from->depth; } @@ -1009,7 +1041,8 @@ png_set_unknown_chunks(png_structp png_ptr, png_sizeof(png_unknown_chunk)); if (np == NULL) { - png_warning(png_ptr, "Out of memory while processing unknown chunk."); + png_warning(png_ptr, + "Out of memory while processing unknown chunk."); return; } @@ -1020,23 +1053,31 @@ png_set_unknown_chunks(png_structp png_ptr, for (i = 0; i < num_unknowns; i++) { - png_unknown_chunkp to = np + info_ptr->unknown_chunks_num + i; - png_unknown_chunkp from = unknowns + i; + png_unknown_chunkp to = np + info_ptr->unknown_chunks_num + i; + png_unknown_chunkp from = unknowns + i; - png_strncpy((png_charp)to->name, (png_charp)from->name, 5); - to->data = (png_bytep)png_malloc_warn(png_ptr, from->size); - if (to->data == NULL) - { - png_warning(png_ptr, "Out of memory processing unknown chunk."); - } - else - { - png_memcpy(to->data, from->data, from->size); - to->size = from->size; + png_memcpy((png_charp)to->name, + (png_charp)from->name, + png_sizeof(from->name)); + to->name[png_sizeof(to->name)-1] = '\0'; + to->size = from->size; + /* note our location in the read or write sequence */ + to->location = (png_byte)(png_ptr->mode & 0xff); - /* note our location in the read or write sequence */ - to->location = (png_byte)(png_ptr->mode & 0xff); - } + if (from->size == 0) + to->data=NULL; + else + { + to->data = (png_bytep)png_malloc_warn(png_ptr, from->size); + if (to->data == NULL) + { + png_warning(png_ptr, + "Out of memory while processing unknown chunk."); + to->size=0; + } + else + png_memcpy(to->data, from->data, from->size); + } } info_ptr->unknown_chunks = np; @@ -1067,7 +1108,7 @@ png_permit_empty_plte (png_structp png_ptr, int empty_plte_permitted) if (png_ptr == NULL) return; png_ptr->mng_features_permitted = (png_byte) - ((png_ptr->mng_features_permitted & (~(PNG_FLAG_MNG_EMPTY_PLTE))) | + ((png_ptr->mng_features_permitted & (~PNG_FLAG_MNG_EMPTY_PLTE)) | ((empty_plte_permitted & PNG_FLAG_MNG_EMPTY_PLTE))); } #endif @@ -1168,8 +1209,7 @@ png_set_compression_buffer_size(png_structp png_ptr, png_uint_32 size) { if (png_ptr == NULL) return; - if(png_ptr->zbuf) - png_free(png_ptr, png_ptr->zbuf); + png_free(png_ptr, png_ptr->zbuf); png_ptr->zbuf_size = (png_size_t)size; png_ptr->zbuf = (png_bytep)png_malloc(png_ptr, size); png_ptr->zstream.next_out = png_ptr->zbuf; @@ -1181,68 +1221,30 @@ void PNGAPI png_set_invalid(png_structp png_ptr, png_infop info_ptr, int mask) { if (png_ptr && info_ptr) - info_ptr->valid &= ~(mask); + info_ptr->valid &= ~mask; } #ifndef PNG_1_0_X #ifdef PNG_ASSEMBLER_CODE_SUPPORTED -/* this function was added to libpng 1.2.0 and should always exist by default */ +/* function was added to libpng 1.2.0 and should always exist by default */ void PNGAPI png_set_asm_flags (png_structp png_ptr, png_uint_32 asm_flags) { - png_uint_32 settable_asm_flags; - png_uint_32 settable_mmx_flags; - - if (png_ptr == NULL) - return; - - settable_mmx_flags = -#ifdef PNG_HAVE_ASSEMBLER_COMBINE_ROW - PNG_ASM_FLAG_MMX_READ_COMBINE_ROW | -#endif -#ifdef PNG_HAVE_ASSEMBLER_READ_INTERLACE - PNG_ASM_FLAG_MMX_READ_INTERLACE | -#endif -#ifdef PNG_HAVE_ASSEMBLER_READ_FILTER_ROW - PNG_ASM_FLAG_MMX_READ_FILTER_SUB | - PNG_ASM_FLAG_MMX_READ_FILTER_UP | - PNG_ASM_FLAG_MMX_READ_FILTER_AVG | - PNG_ASM_FLAG_MMX_READ_FILTER_PAETH | -#endif - 0; - - /* could be some non-MMX ones in the future, but not currently: */ - settable_asm_flags = settable_mmx_flags; - - if (!(png_ptr->asm_flags & PNG_ASM_FLAG_MMX_SUPPORT_COMPILED) || - !(png_ptr->asm_flags & PNG_ASM_FLAG_MMX_SUPPORT_IN_CPU)) - { - /* clear all MMX flags if MMX isn't supported */ - settable_asm_flags &= ~settable_mmx_flags; - png_ptr->asm_flags &= ~settable_mmx_flags; - } - - /* we're replacing the settable bits with those passed in by the user, - * so first zero them out of the master copy, then logical-OR in the - * allowed subset that was requested */ - - png_ptr->asm_flags &= ~settable_asm_flags; /* zero them */ - png_ptr->asm_flags |= (asm_flags & settable_asm_flags); /* set them */ +/* Obsolete as of libpng-1.2.20 and will be removed from libpng-1.4.0 */ + if (png_ptr != NULL) + png_ptr->asm_flags = 0; } -#endif /* ?PNG_ASSEMBLER_CODE_SUPPORTED */ -#ifdef PNG_ASSEMBLER_CODE_SUPPORTED /* this function was added to libpng 1.2.0 */ void PNGAPI png_set_mmx_thresholds (png_structp png_ptr, png_byte mmx_bitdepth_threshold, png_uint_32 mmx_rowbytes_threshold) { +/* Obsolete as of libpng-1.2.20 and will be removed from libpng-1.4.0 */ if (png_ptr == NULL) return; - png_ptr->mmx_bitdepth_threshold = mmx_bitdepth_threshold; - png_ptr->mmx_rowbytes_threshold = mmx_rowbytes_threshold; } #endif /* ?PNG_ASSEMBLER_CODE_SUPPORTED */ @@ -1256,6 +1258,7 @@ png_set_user_limits (png_structp png_ptr, png_uint_32 user_width_max, * rejected by png_set_IHDR(). To accept any PNG datastream * regardless of dimensions, set both limits to 0x7ffffffL. */ + if(png_ptr == NULL) return; png_ptr->user_width_max = user_width_max; png_ptr->user_height_max = user_height_max; } diff --git a/pngtest.c b/pngtest.c index 00afcd734..60980c275 100644 --- a/pngtest.c +++ b/pngtest.c @@ -1,9 +1,9 @@ /* pngtest.c - a simple test program to test libpng * - * Last changed in libpng 1.2.6 - August 15, 2004 + * Last changed in libpng 1.2.27 - [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2004 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -81,21 +81,10 @@ static float t_start, t_stop, t_decode, t_encode, t_misc; #include #endif -/* Define png_jmpbuf() in case we are using a pre-1.0.6 version of libpng */ -#ifndef png_jmpbuf -# define png_jmpbuf(png_ptr) png_ptr->jmpbuf -#endif - -#ifdef PNGTEST_TIMING -static float t_start, t_stop, t_decode, t_encode, t_misc; -#if !defined(PNG_tIME_SUPPORTED) -#include -#endif -#endif - #if defined(PNG_TIME_RFC1123_SUPPORTED) +#define PNG_tIME_STRING_LENGTH 30 static int tIME_chunk_present=0; -static char tIME_string[30] = "no tIME chunk present in file"; +static char tIME_string[PNG_tIME_STRING_LENGTH] = "no tIME chunk present in file"; #endif static int verbose = 0; @@ -115,6 +104,16 @@ static int status_pass=1; static int status_dots_requested=0; static int status_dots=1; +/* In case a system header (e.g., on AIX) defined jmpbuf */ +#ifdef jmpbuf +# undef jmpbuf +#endif + +/* Define png_jmpbuf() in case we are using a pre-1.0.6 version of libpng */ +#ifndef png_jmpbuf +# define png_jmpbuf(png_ptr) png_ptr->jmpbuf +#endif + void #ifdef PNG_1_0_X PNGAPI @@ -436,8 +435,9 @@ pngtest_write_data(png_structp png_ptr, png_bytep data, png_size_t length) png_error(png_ptr, "Write Error"); } } - #endif /* USE_FAR_KEYWORD */ +#endif /* PNG_NO_STDIO */ +/* END of code to validate stdio-free compilation */ /* This function is called when there is a warning, but the library thinks * it can continue anyway. Replacement functions don't have to do anything @@ -465,8 +465,6 @@ pngtest_error(png_structp png_ptr, png_const_charp message) /* We can return because png_error calls the default handler, which is * actually OK in this case. */ } -#endif /* PNG_NO_STDIO */ -/* END of code to validate stdio-free compilation */ /* START of code to validate memory allocation and deallocation */ #if defined(PNG_USER_MEM_SUPPORTED) && PNG_DEBUG @@ -536,7 +534,8 @@ png_debug_malloc(png_structp png_ptr, png_uint_32 size) /* Make sure the caller isn't assuming zeroed memory. */ png_memset(pinfo->pointer, 0xdd, pinfo->size); if(verbose) - printf("png_malloc %lu bytes at %x\n",size,pinfo->pointer); + printf("png_malloc %lu bytes at %x\n",(unsigned long)size, + pinfo->pointer); return (png_voidp)(pinfo->pointer); } } @@ -659,10 +658,8 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) read_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING, png_voidp_NULL, png_error_ptr_NULL, png_error_ptr_NULL); #endif -#if defined(PNG_NO_STDIO) png_set_error_fn(read_ptr, (png_voidp)inname, pngtest_error, pngtest_warning); -#endif #ifdef PNG_WRITE_SUPPORTED #if defined(PNG_USER_MEM_SUPPORTED) && PNG_DEBUG write_ptr = png_create_write_struct_2(PNG_LIBPNG_VER_STRING, png_voidp_NULL, @@ -672,10 +669,8 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) write_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING, png_voidp_NULL, png_error_ptr_NULL, png_error_ptr_NULL); #endif -#if defined(PNG_NO_STDIO) png_set_error_fn(write_ptr, (png_voidp)inname, pngtest_error, pngtest_warning); -#endif #endif png_debug(0, "Allocating read_info, write_info and end_info structures\n"); read_info_ptr = png_create_info_struct(read_ptr); @@ -694,8 +689,7 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) #endif { fprintf(STDERR, "%s -> %s: libpng read error\n", inname, outname); - if (row_buf) - png_free(read_ptr, row_buf); + png_free(read_ptr, row_buf); png_destroy_read_struct(&read_ptr, &read_info_ptr, &end_info_ptr); #ifdef PNG_WRITE_SUPPORTED png_destroy_info_struct(write_ptr, &write_end_info_ptr); @@ -1008,10 +1002,13 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) { png_set_tIME(write_ptr, write_info_ptr, mod_time); #if defined(PNG_TIME_RFC1123_SUPPORTED) - /* we have to use png_strcpy instead of "=" because the string + /* we have to use png_memcpy instead of "=" because the string pointed to by png_convert_to_rfc1123() gets free'ed before we use it */ - png_strcpy(tIME_string,png_convert_to_rfc1123(read_ptr, mod_time)); + png_memcpy(tIME_string, + png_convert_to_rfc1123(read_ptr, mod_time), + png_sizeof(tIME_string)); + tIME_string[png_sizeof(tIME_string)-1] = '\0'; tIME_chunk_present++; #endif /* PNG_TIME_RFC1123_SUPPORTED */ } @@ -1026,8 +1023,16 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) if (png_get_tRNS(read_ptr, read_info_ptr, &trans, &num_trans, &trans_values)) { - png_set_tRNS(write_ptr, write_info_ptr, trans, num_trans, - trans_values); + int sample_max = (1 << read_info_ptr->bit_depth); + /* libpng doesn't reject a tRNS chunk with out-of-range samples */ + if (!((read_info_ptr->color_type == PNG_COLOR_TYPE_GRAY && + (int)trans_values->gray > sample_max) || + (read_info_ptr->color_type == PNG_COLOR_TYPE_RGB && + ((int)trans_values->red > sample_max || + (int)trans_values->green > sample_max || + (int)trans_values->blue > sample_max)))) + png_set_tRNS(write_ptr, write_info_ptr, trans, num_trans, + trans_values); } } #endif @@ -1148,10 +1153,13 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) { png_set_tIME(write_ptr, write_end_info_ptr, mod_time); #if defined(PNG_TIME_RFC1123_SUPPORTED) - /* we have to use png_strcpy instead of "=" because the string + /* we have to use png_memcpy instead of "=" because the string pointed to by png_convert_to_rfc1123() gets free'ed before we use it */ - png_strcpy(tIME_string,png_convert_to_rfc1123(read_ptr, mod_time)); + png_memcpy(tIME_string, + png_convert_to_rfc1123(read_ptr, mod_time), + png_sizeof(tIME_string)); + tIME_string[png_sizeof(tIME_string)-1] = '\0'; tIME_chunk_present++; #endif /* PNG_TIME_RFC1123_SUPPORTED */ } @@ -1188,7 +1196,7 @@ test_one_file(PNG_CONST char *inname, PNG_CONST char *outname) iwidth = png_get_image_width(write_ptr, write_info_ptr); iheight = png_get_image_height(write_ptr, write_info_ptr); fprintf(STDERR, "Image width = %lu, height = %lu\n", - iwidth, iheight); + (unsigned long)iwidth, (unsigned long)iheight); } #endif @@ -1312,7 +1320,8 @@ main(int argc, char *argv[]) fprintf(STDERR, " with zlib version %s\n", ZLIB_VERSION); fprintf(STDERR,"%s",png_get_copyright(NULL)); /* Show the version of libpng used in building the library */ - fprintf(STDERR," library (%lu):%s", png_access_version_number(), + fprintf(STDERR," library (%lu):%s", + (unsigned long)png_access_version_number(), png_get_header_version(NULL)); /* Show the version of libpng used in building the application */ fprintf(STDERR," pngtest (%lu):%s", (unsigned long)PNG_LIBPNG_VER, @@ -1400,7 +1409,8 @@ main(int argc, char *argv[]) if (kerror == 0) { #if defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) - fprintf(STDERR, "\n PASS (%lu zero samples)\n",zero_samples); + fprintf(STDERR, "\n PASS (%lu zero samples)\n", + (unsigned long)zero_samples); #else fprintf(STDERR, " PASS\n"); #endif @@ -1408,7 +1418,7 @@ main(int argc, char *argv[]) for (k=0; k<256; k++) if(filters_used[k]) fprintf(STDERR, " Filter %d was used %lu times\n", - k,filters_used[k]); + k,(unsigned long)filters_used[k]); #endif #if defined(PNG_TIME_RFC1123_SUPPORTED) if(tIME_chunk_present != 0) @@ -1433,7 +1443,7 @@ main(int argc, char *argv[]) current_allocation); while (pinfo != NULL) { - fprintf(STDERR, " %lu bytes at %x\n", pinfo->size, + fprintf(STDERR, " %lu bytes at %x\n", (unsigned long)pinfo->size, (unsigned int) pinfo->pointer); pinfo = pinfo->next; } @@ -1473,7 +1483,8 @@ main(int argc, char *argv[]) int k; #endif #if defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) - fprintf(STDERR, "\n PASS (%lu zero samples)\n",zero_samples); + fprintf(STDERR, "\n PASS (%lu zero samples)\n", + (unsigned long)zero_samples); #else fprintf(STDERR, " PASS\n"); #endif @@ -1481,7 +1492,7 @@ main(int argc, char *argv[]) for (k=0; k<256; k++) if(filters_used[k]) fprintf(STDERR, " Filter %d was used %lu times\n", - k,filters_used[k]); + k,(unsigned long)filters_used[k]); #endif #if defined(PNG_TIME_RFC1123_SUPPORTED) if(tIME_chunk_present != 0) @@ -1509,7 +1520,7 @@ main(int argc, char *argv[]) while (pinfo != NULL) { fprintf(STDERR," %lu bytes at %x\n", - pinfo->size, (unsigned int)pinfo->pointer); + (unsigned long)pinfo->size, (unsigned int)pinfo->pointer); pinfo = pinfo->next; } } @@ -1549,4 +1560,4 @@ main(int argc, char *argv[]) } /* Generate a compiler error if there is an old png.h in the search path. */ -typedef version_1_2_11beta4 your_png_h_is_not_version_1_2_11beta4; +typedef version_1_2_29 your_png_h_is_not_version_1_2_29; diff --git a/pngtrans.c b/pngtrans.c index 9c4d67cd0..164009502 100644 --- a/pngtrans.c +++ b/pngtrans.c @@ -1,9 +1,9 @@ /* pngtrans.c - transforms the data in a row (used by both readers and writers) * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.17 May 15, 2007 * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2007 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -18,6 +18,7 @@ void PNGAPI png_set_bgr(png_structp png_ptr) { png_debug(1, "in png_set_bgr\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_BGR; } #endif @@ -28,6 +29,7 @@ void PNGAPI png_set_swap(png_structp png_ptr) { png_debug(1, "in png_set_swap\n"); + if(png_ptr == NULL) return; if (png_ptr->bit_depth == 16) png_ptr->transformations |= PNG_SWAP_BYTES; } @@ -39,6 +41,7 @@ void PNGAPI png_set_packing(png_structp png_ptr) { png_debug(1, "in png_set_packing\n"); + if(png_ptr == NULL) return; if (png_ptr->bit_depth < 8) { png_ptr->transformations |= PNG_PACK; @@ -53,6 +56,7 @@ void PNGAPI png_set_packswap(png_structp png_ptr) { png_debug(1, "in png_set_packswap\n"); + if(png_ptr == NULL) return; if (png_ptr->bit_depth < 8) png_ptr->transformations |= PNG_PACKSWAP; } @@ -63,6 +67,7 @@ void PNGAPI png_set_shift(png_structp png_ptr, png_color_8p true_bits) { png_debug(1, "in png_set_shift\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_SHIFT; png_ptr->shift = *true_bits; } @@ -74,7 +79,7 @@ int PNGAPI png_set_interlace_handling(png_structp png_ptr) { png_debug(1, "in png_set_interlace handling\n"); - if (png_ptr->interlaced) + if (png_ptr && png_ptr->interlaced) { png_ptr->transformations |= PNG_INTERLACE; return (7); @@ -94,6 +99,7 @@ void PNGAPI png_set_filler(png_structp png_ptr, png_uint_32 filler, int filler_loc) { png_debug(1, "in png_set_filler\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_FILLER; png_ptr->filler = (png_byte)filler; if (filler_loc == PNG_FILLER_AFTER) @@ -126,6 +132,7 @@ void PNGAPI png_set_add_alpha(png_structp png_ptr, png_uint_32 filler, int filler_loc) { png_debug(1, "in png_set_add_alpha\n"); + if(png_ptr == NULL) return; png_set_filler(png_ptr, filler, filler_loc); png_ptr->transformations |= PNG_ADD_ALPHA; } @@ -139,6 +146,7 @@ void PNGAPI png_set_swap_alpha(png_structp png_ptr) { png_debug(1, "in png_set_swap_alpha\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_SWAP_ALPHA; } #endif @@ -149,6 +157,7 @@ void PNGAPI png_set_invert_alpha(png_structp png_ptr) { png_debug(1, "in png_set_invert_alpha\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_INVERT_ALPHA; } #endif @@ -158,6 +167,7 @@ void PNGAPI png_set_invert_mono(png_structp png_ptr) { png_debug(1, "in png_set_invert_mono\n"); + if(png_ptr == NULL) return; png_ptr->transformations |= PNG_INVERT_MONO; } @@ -621,6 +631,7 @@ png_set_user_transform_info(png_structp png_ptr, png_voidp user_transform_ptr, int user_transform_depth, int user_transform_channels) { png_debug(1, "in png_set_user_transform_info\n"); + if(png_ptr == NULL) return; #if defined(PNG_USER_TRANSFORM_PTR_SUPPORTED) png_ptr->user_transform_ptr = user_transform_ptr; png_ptr->user_transform_depth = (png_byte)user_transform_depth; @@ -642,10 +653,9 @@ png_voidp PNGAPI png_get_user_transform_ptr(png_structp png_ptr) { #if defined(PNG_USER_TRANSFORM_PTR_SUPPORTED) + if (png_ptr == NULL) return (NULL); return ((png_voidp)png_ptr->user_transform_ptr); #else - if(png_ptr) - return (NULL); return (NULL); #endif } diff --git a/pngvcrd.c b/pngvcrd.c index d1c6405e5..ce4233efe 100644 --- a/pngvcrd.c +++ b/pngvcrd.c @@ -1,3904 +1 @@ - -/* pngvcrd.c - mixed C/assembler version of utilities to read a PNG file - * - * For Intel x86 CPU and Microsoft Visual C++ compiler - * - * Last changed in libpng 1.2.6 - August 15, 2004 - * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2004 Glenn Randers-Pehrson - * Copyright (c) 1998, Intel Corporation - * - * Contributed by Nirav Chhatrapati, Intel Corporation, 1998 - * Interface to libpng contributed by Gilles Vollant, 1999 - * - * - * In png_do_read_interlace() in libpng versions 1.0.3a through 1.0.4d, - * a sign error in the post-MMX cleanup code for each pixel_depth resulted - * in bad pixels at the beginning of some rows of some images, and also - * (due to out-of-range memory reads and writes) caused heap corruption - * when compiled with MSVC 6.0. The error was fixed in version 1.0.4e. - * - * [png_read_filter_row_mmx_avg() bpp == 2 bugfix, GRR 20000916] - * - * [runtime MMX configuration, GRR 20010102] - * - */ - -#define PNG_INTERNAL -#include "png.h" - -#if defined(PNG_ASSEMBLER_CODE_SUPPORTED) && defined(PNG_USE_PNGVCRD) - -static int mmx_supported=2; - - -int PNGAPI -png_mmx_support(void) -{ - int mmx_supported_local = 0; - _asm { - push ebx //CPUID will trash these - push ecx - push edx - - pushfd //Save Eflag to stack - pop eax //Get Eflag from stack into eax - mov ecx, eax //Make another copy of Eflag in ecx - xor eax, 0x200000 //Toggle ID bit in Eflag [i.e. bit(21)] - push eax //Save modified Eflag back to stack - - popfd //Restored modified value back to Eflag reg - pushfd //Save Eflag to stack - pop eax //Get Eflag from stack - push ecx // save original Eflag to stack - popfd // restore original Eflag - xor eax, ecx //Compare the new Eflag with the original Eflag - jz NOT_SUPPORTED //If the same, CPUID instruction is not supported, - //skip following instructions and jump to - //NOT_SUPPORTED label - - xor eax, eax //Set eax to zero - - _asm _emit 0x0f //CPUID instruction (two bytes opcode) - _asm _emit 0xa2 - - cmp eax, 1 //make sure eax return non-zero value - jl NOT_SUPPORTED //If eax is zero, mmx not supported - - xor eax, eax //set eax to zero - inc eax //Now increment eax to 1. This instruction is - //faster than the instruction "mov eax, 1" - - _asm _emit 0x0f //CPUID instruction - _asm _emit 0xa2 - - and edx, 0x00800000 //mask out all bits but mmx bit(24) - cmp edx, 0 // 0 = mmx not supported - jz NOT_SUPPORTED // non-zero = Yes, mmx IS supported - - mov mmx_supported_local, 1 //set return value to 1 - -NOT_SUPPORTED: - mov eax, mmx_supported_local //move return value to eax - pop edx //CPUID trashed these - pop ecx - pop ebx - } - - //mmx_supported_local=0; // test code for force don't support MMX - //printf("MMX : %u (1=MMX supported)\n",mmx_supported_local); - - mmx_supported = mmx_supported_local; - return mmx_supported_local; -} - -/* Combines the row recently read in with the previous row. - This routine takes care of alpha and transparency if requested. - This routine also handles the two methods of progressive display - of interlaced images, depending on the mask value. - The mask value describes which pixels are to be combined with - the row. The pattern always repeats every 8 pixels, so just 8 - bits are needed. A one indicates the pixel is to be combined; a - zero indicates the pixel is to be skipped. This is in addition - to any alpha or transparency value associated with the pixel. If - you want all pixels to be combined, pass 0xff (255) in mask. */ - -/* Use this routine for x86 platform - uses faster MMX routine if machine - supports MMX */ - -void /* PRIVATE */ -png_combine_row(png_structp png_ptr, png_bytep row, int mask) -{ -#ifdef PNG_USE_LOCAL_ARRAYS - const int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; -#endif - - png_debug(1,"in png_combine_row_asm\n"); - - if (mmx_supported == 2) { -#if !defined(PNG_1_0_X) - /* this should have happened in png_init_mmx_flags() already */ - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } - - if (mask == 0xff) - { - png_memcpy(row, png_ptr->row_buf + 1, - (png_size_t)PNG_ROWBYTES(png_ptr->row_info.pixel_depth, - png_ptr->width)); - } - /* GRR: add "else if (mask == 0)" case? - * or does png_combine_row() not even get called in that case? */ - else - { - switch (png_ptr->row_info.pixel_depth) - { - case 1: - { - png_bytep sp; - png_bytep dp; - int s_inc, s_start, s_end; - int m; - int shift; - png_uint_32 i; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 7; - s_inc = 1; - } - else -#endif - { - s_start = 7; - s_end = 0; - s_inc = -1; - } - - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - int value; - - value = (*sp >> shift) & 0x1; - *dp &= (png_byte)((0x7f7f >> (7 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 2: - { - png_bytep sp; - png_bytep dp; - int s_start, s_end, s_inc; - int m; - int shift; - png_uint_32 i; - int value; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 6; - s_inc = 2; - } - else -#endif - { - s_start = 6; - s_end = 0; - s_inc = -2; - } - - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - value = (*sp >> shift) & 0x3; - *dp &= (png_byte)((0x3f3f >> (6 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 4: - { - png_bytep sp; - png_bytep dp; - int s_start, s_end, s_inc; - int m; - int shift; - png_uint_32 i; - int value; - - sp = png_ptr->row_buf + 1; - dp = row; - m = 0x80; -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (png_ptr->transformations & PNG_PACKSWAP) - { - s_start = 0; - s_end = 4; - s_inc = 4; - } - else -#endif - { - s_start = 4; - s_end = 0; - s_inc = -4; - } - shift = s_start; - - for (i = 0; i < png_ptr->width; i++) - { - if (m & mask) - { - value = (*sp >> shift) & 0xf; - *dp &= (png_byte)((0xf0f >> (4 - shift)) & 0xff); - *dp |= (png_byte)(value << shift); - } - - if (shift == s_end) - { - shift = s_start; - sp++; - dp++; - } - else - shift += s_inc; - if (m == 1) - m = 0x80; - else - m >>= 1; - } - break; - } - - case 8: - { - png_bytep srcptr; - png_bytep dstptr; - png_uint_32 len; - int m; - int diff, unmask; - - __int64 mask0=0x0102040810204080; - -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - srcptr = png_ptr->row_buf + 1; - dstptr = row; - m = 0x80; - unmask = ~mask; - len = png_ptr->width &~7; //reduce to multiple of 8 - diff = png_ptr->width & 7; //amount lost - - _asm - { - movd mm7, unmask //load bit pattern - psubb mm6,mm6 //zero mm6 - punpcklbw mm7,mm7 - punpcklwd mm7,mm7 - punpckldq mm7,mm7 //fill register with 8 masks - - movq mm0,mask0 - - pand mm0,mm7 //nonzero if keep byte - pcmpeqb mm0,mm6 //zeros->1s, v versa - - mov ecx,len //load length of line (pixels) - mov esi,srcptr //load source - mov ebx,dstptr //load dest - cmp ecx,0 //lcr - je mainloop8end - -mainloop8: - movq mm4,[esi] - pand mm4,mm0 - movq mm6,mm0 - pandn mm6,[ebx] - por mm4,mm6 - movq [ebx],mm4 - - add esi,8 //inc by 8 bytes processed - add ebx,8 - sub ecx,8 //dec by 8 pixels processed - - ja mainloop8 -mainloop8end: - - mov ecx,diff - cmp ecx,0 - jz end8 - - mov edx,mask - sal edx,24 //make low byte the high byte - -secondloop8: - sal edx,1 //move high bit to CF - jnc skip8 //if CF = 0 - mov al,[esi] - mov [ebx],al -skip8: - inc esi - inc ebx - - dec ecx - jnz secondloop8 -end8: - emms - } - } - else /* mmx not supported - use modified C routine */ - { - register unsigned int incr1, initial_val, final_val; - png_size_t pixel_bytes; - png_uint_32 i; - register int disp = png_pass_inc[png_ptr->pass]; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - srcptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dstptr = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dstptr, srcptr, pixel_bytes); - srcptr += incr1; - dstptr += incr1; - } - } /* end of else */ - - break; - } // end 8 bpp - - case 16: - { - png_bytep srcptr; - png_bytep dstptr; - png_uint_32 len; - int unmask, diff; - __int64 mask1=0x0101020204040808, - mask0=0x1010202040408080; - -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - srcptr = png_ptr->row_buf + 1; - dstptr = row; - - unmask = ~mask; - len = (png_ptr->width)&~7; - diff = (png_ptr->width)&7; - _asm - { - movd mm7, unmask //load bit pattern - psubb mm6,mm6 //zero mm6 - punpcklbw mm7,mm7 - punpcklwd mm7,mm7 - punpckldq mm7,mm7 //fill register with 8 masks - - movq mm0,mask0 - movq mm1,mask1 - - pand mm0,mm7 - pand mm1,mm7 - - pcmpeqb mm0,mm6 - pcmpeqb mm1,mm6 - - mov ecx,len //load length of line - mov esi,srcptr //load source - mov ebx,dstptr //load dest - cmp ecx,0 //lcr - jz mainloop16end - -mainloop16: - movq mm4,[esi] - pand mm4,mm0 - movq mm6,mm0 - movq mm7,[ebx] - pandn mm6,mm7 - por mm4,mm6 - movq [ebx],mm4 - - movq mm5,[esi+8] - pand mm5,mm1 - movq mm7,mm1 - movq mm6,[ebx+8] - pandn mm7,mm6 - por mm5,mm7 - movq [ebx+8],mm5 - - add esi,16 //inc by 16 bytes processed - add ebx,16 - sub ecx,8 //dec by 8 pixels processed - - ja mainloop16 - -mainloop16end: - mov ecx,diff - cmp ecx,0 - jz end16 - - mov edx,mask - sal edx,24 //make low byte the high byte -secondloop16: - sal edx,1 //move high bit to CF - jnc skip16 //if CF = 0 - mov ax,[esi] - mov [ebx],ax -skip16: - add esi,2 - add ebx,2 - - dec ecx - jnz secondloop16 -end16: - emms - } - } - else /* mmx not supported - use modified C routine */ - { - register unsigned int incr1, initial_val, final_val; - png_size_t pixel_bytes; - png_uint_32 i; - register int disp = png_pass_inc[png_ptr->pass]; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - srcptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dstptr = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dstptr, srcptr, pixel_bytes); - srcptr += incr1; - dstptr += incr1; - } - } /* end of else */ - - break; - } // end 16 bpp - - case 24: - { - png_bytep srcptr; - png_bytep dstptr; - png_uint_32 len; - int unmask, diff; - - __int64 mask2=0x0101010202020404, //24bpp - mask1=0x0408080810101020, - mask0=0x2020404040808080; - - srcptr = png_ptr->row_buf + 1; - dstptr = row; - - unmask = ~mask; - len = (png_ptr->width)&~7; - diff = (png_ptr->width)&7; - -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - _asm - { - movd mm7, unmask //load bit pattern - psubb mm6,mm6 //zero mm6 - punpcklbw mm7,mm7 - punpcklwd mm7,mm7 - punpckldq mm7,mm7 //fill register with 8 masks - - movq mm0,mask0 - movq mm1,mask1 - movq mm2,mask2 - - pand mm0,mm7 - pand mm1,mm7 - pand mm2,mm7 - - pcmpeqb mm0,mm6 - pcmpeqb mm1,mm6 - pcmpeqb mm2,mm6 - - mov ecx,len //load length of line - mov esi,srcptr //load source - mov ebx,dstptr //load dest - cmp ecx,0 - jz mainloop24end - -mainloop24: - movq mm4,[esi] - pand mm4,mm0 - movq mm6,mm0 - movq mm7,[ebx] - pandn mm6,mm7 - por mm4,mm6 - movq [ebx],mm4 - - - movq mm5,[esi+8] - pand mm5,mm1 - movq mm7,mm1 - movq mm6,[ebx+8] - pandn mm7,mm6 - por mm5,mm7 - movq [ebx+8],mm5 - - movq mm6,[esi+16] - pand mm6,mm2 - movq mm4,mm2 - movq mm7,[ebx+16] - pandn mm4,mm7 - por mm6,mm4 - movq [ebx+16],mm6 - - add esi,24 //inc by 24 bytes processed - add ebx,24 - sub ecx,8 //dec by 8 pixels processed - - ja mainloop24 - -mainloop24end: - mov ecx,diff - cmp ecx,0 - jz end24 - - mov edx,mask - sal edx,24 //make low byte the high byte -secondloop24: - sal edx,1 //move high bit to CF - jnc skip24 //if CF = 0 - mov ax,[esi] - mov [ebx],ax - xor eax,eax - mov al,[esi+2] - mov [ebx+2],al -skip24: - add esi,3 - add ebx,3 - - dec ecx - jnz secondloop24 - -end24: - emms - } - } - else /* mmx not supported - use modified C routine */ - { - register unsigned int incr1, initial_val, final_val; - png_size_t pixel_bytes; - png_uint_32 i; - register int disp = png_pass_inc[png_ptr->pass]; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - srcptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dstptr = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dstptr, srcptr, pixel_bytes); - srcptr += incr1; - dstptr += incr1; - } - } /* end of else */ - - break; - } // end 24 bpp - - case 32: - { - png_bytep srcptr; - png_bytep dstptr; - png_uint_32 len; - int unmask, diff; - - __int64 mask3=0x0101010102020202, //32bpp - mask2=0x0404040408080808, - mask1=0x1010101020202020, - mask0=0x4040404080808080; - - srcptr = png_ptr->row_buf + 1; - dstptr = row; - - unmask = ~mask; - len = (png_ptr->width)&~7; - diff = (png_ptr->width)&7; - -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - _asm - { - movd mm7, unmask //load bit pattern - psubb mm6,mm6 //zero mm6 - punpcklbw mm7,mm7 - punpcklwd mm7,mm7 - punpckldq mm7,mm7 //fill register with 8 masks - - movq mm0,mask0 - movq mm1,mask1 - movq mm2,mask2 - movq mm3,mask3 - - pand mm0,mm7 - pand mm1,mm7 - pand mm2,mm7 - pand mm3,mm7 - - pcmpeqb mm0,mm6 - pcmpeqb mm1,mm6 - pcmpeqb mm2,mm6 - pcmpeqb mm3,mm6 - - mov ecx,len //load length of line - mov esi,srcptr //load source - mov ebx,dstptr //load dest - - cmp ecx,0 //lcr - jz mainloop32end - -mainloop32: - movq mm4,[esi] - pand mm4,mm0 - movq mm6,mm0 - movq mm7,[ebx] - pandn mm6,mm7 - por mm4,mm6 - movq [ebx],mm4 - - movq mm5,[esi+8] - pand mm5,mm1 - movq mm7,mm1 - movq mm6,[ebx+8] - pandn mm7,mm6 - por mm5,mm7 - movq [ebx+8],mm5 - - movq mm6,[esi+16] - pand mm6,mm2 - movq mm4,mm2 - movq mm7,[ebx+16] - pandn mm4,mm7 - por mm6,mm4 - movq [ebx+16],mm6 - - movq mm7,[esi+24] - pand mm7,mm3 - movq mm5,mm3 - movq mm4,[ebx+24] - pandn mm5,mm4 - por mm7,mm5 - movq [ebx+24],mm7 - - add esi,32 //inc by 32 bytes processed - add ebx,32 - sub ecx,8 //dec by 8 pixels processed - - ja mainloop32 - -mainloop32end: - mov ecx,diff - cmp ecx,0 - jz end32 - - mov edx,mask - sal edx,24 //make low byte the high byte -secondloop32: - sal edx,1 //move high bit to CF - jnc skip32 //if CF = 0 - mov eax,[esi] - mov [ebx],eax -skip32: - add esi,4 - add ebx,4 - - dec ecx - jnz secondloop32 - -end32: - emms - } - } - else /* mmx _not supported - Use modified C routine */ - { - register unsigned int incr1, initial_val, final_val; - png_size_t pixel_bytes; - png_uint_32 i; - register int disp = png_pass_inc[png_ptr->pass]; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - srcptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dstptr = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dstptr, srcptr, pixel_bytes); - srcptr += incr1; - dstptr += incr1; - } - } /* end of else */ - - break; - } // end 32 bpp - - case 48: - { - png_bytep srcptr; - png_bytep dstptr; - png_uint_32 len; - int unmask, diff; - - __int64 mask5=0x0101010101010202, - mask4=0x0202020204040404, - mask3=0x0404080808080808, - mask2=0x1010101010102020, - mask1=0x2020202040404040, - mask0=0x4040808080808080; - -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_COMBINE_ROW) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - srcptr = png_ptr->row_buf + 1; - dstptr = row; - - unmask = ~mask; - len = (png_ptr->width)&~7; - diff = (png_ptr->width)&7; - _asm - { - movd mm7, unmask //load bit pattern - psubb mm6,mm6 //zero mm6 - punpcklbw mm7,mm7 - punpcklwd mm7,mm7 - punpckldq mm7,mm7 //fill register with 8 masks - - movq mm0,mask0 - movq mm1,mask1 - movq mm2,mask2 - movq mm3,mask3 - movq mm4,mask4 - movq mm5,mask5 - - pand mm0,mm7 - pand mm1,mm7 - pand mm2,mm7 - pand mm3,mm7 - pand mm4,mm7 - pand mm5,mm7 - - pcmpeqb mm0,mm6 - pcmpeqb mm1,mm6 - pcmpeqb mm2,mm6 - pcmpeqb mm3,mm6 - pcmpeqb mm4,mm6 - pcmpeqb mm5,mm6 - - mov ecx,len //load length of line - mov esi,srcptr //load source - mov ebx,dstptr //load dest - - cmp ecx,0 - jz mainloop48end - -mainloop48: - movq mm7,[esi] - pand mm7,mm0 - movq mm6,mm0 - pandn mm6,[ebx] - por mm7,mm6 - movq [ebx],mm7 - - movq mm6,[esi+8] - pand mm6,mm1 - movq mm7,mm1 - pandn mm7,[ebx+8] - por mm6,mm7 - movq [ebx+8],mm6 - - movq mm6,[esi+16] - pand mm6,mm2 - movq mm7,mm2 - pandn mm7,[ebx+16] - por mm6,mm7 - movq [ebx+16],mm6 - - movq mm7,[esi+24] - pand mm7,mm3 - movq mm6,mm3 - pandn mm6,[ebx+24] - por mm7,mm6 - movq [ebx+24],mm7 - - movq mm6,[esi+32] - pand mm6,mm4 - movq mm7,mm4 - pandn mm7,[ebx+32] - por mm6,mm7 - movq [ebx+32],mm6 - - movq mm7,[esi+40] - pand mm7,mm5 - movq mm6,mm5 - pandn mm6,[ebx+40] - por mm7,mm6 - movq [ebx+40],mm7 - - add esi,48 //inc by 32 bytes processed - add ebx,48 - sub ecx,8 //dec by 8 pixels processed - - ja mainloop48 -mainloop48end: - - mov ecx,diff - cmp ecx,0 - jz end48 - - mov edx,mask - sal edx,24 //make low byte the high byte - -secondloop48: - sal edx,1 //move high bit to CF - jnc skip48 //if CF = 0 - mov eax,[esi] - mov [ebx],eax -skip48: - add esi,4 - add ebx,4 - - dec ecx - jnz secondloop48 - -end48: - emms - } - } - else /* mmx _not supported - Use modified C routine */ - { - register unsigned int incr1, initial_val, final_val; - png_size_t pixel_bytes; - png_uint_32 i; - register int disp = png_pass_inc[png_ptr->pass]; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - srcptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dstptr = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dstptr, srcptr, pixel_bytes); - srcptr += incr1; - dstptr += incr1; - } - } /* end of else */ - - break; - } // end 48 bpp - - default: - { - png_bytep sptr; - png_bytep dp; - png_size_t pixel_bytes; - int offset_table[7] = {0, 4, 0, 2, 0, 1, 0}; - unsigned int i; - register int disp = png_pass_inc[png_ptr->pass]; // get the offset - register unsigned int incr1, initial_val, final_val; - - pixel_bytes = (png_ptr->row_info.pixel_depth >> 3); - sptr = png_ptr->row_buf + 1 + offset_table[png_ptr->pass]* - pixel_bytes; - dp = row + offset_table[png_ptr->pass]*pixel_bytes; - initial_val = offset_table[png_ptr->pass]*pixel_bytes; - final_val = png_ptr->width*pixel_bytes; - incr1 = (disp)*pixel_bytes; - for (i = initial_val; i < final_val; i += incr1) - { - png_memcpy(dp, sptr, pixel_bytes); - sptr += incr1; - dp += incr1; - } - break; - } - } /* end switch (png_ptr->row_info.pixel_depth) */ - } /* end if (non-trivial mask) */ - -} /* end png_combine_row() */ - - -#if defined(PNG_READ_INTERLACING_SUPPORTED) - -void /* PRIVATE */ -png_do_read_interlace(png_structp png_ptr) -{ - png_row_infop row_info = &(png_ptr->row_info); - png_bytep row = png_ptr->row_buf + 1; - int pass = png_ptr->pass; - png_uint_32 transformations = png_ptr->transformations; -#ifdef PNG_USE_LOCAL_ARRAYS - const int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; -#endif - - png_debug(1,"in png_do_read_interlace\n"); - - if (mmx_supported == 2) { -#if !defined(PNG_1_0_X) - /* this should have happened in png_init_mmx_flags() already */ - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } - - if (row != NULL && row_info != NULL) - { - png_uint_32 final_width; - - final_width = row_info->width * png_pass_inc[pass]; - - switch (row_info->pixel_depth) - { - case 1: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_byte v; - png_uint_32 i; - int j; - - sp = row + (png_size_t)((row_info->width - 1) >> 3); - dp = row + (png_size_t)((final_width - 1) >> 3); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (int)((row_info->width + 7) & 7); - dshift = (int)((final_width + 7) & 7); - s_start = 7; - s_end = 0; - s_inc = -1; - } - else -#endif - { - sshift = 7 - (int)((row_info->width + 7) & 7); - dshift = 7 - (int)((final_width + 7) & 7); - s_start = 0; - s_end = 7; - s_inc = 1; - } - - for (i = row_info->width; i; i--) - { - v = (png_byte)((*sp >> sshift) & 0x1); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0x7f7f >> (7 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - case 2: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_uint_32 i; - - sp = row + (png_size_t)((row_info->width - 1) >> 2); - dp = row + (png_size_t)((final_width - 1) >> 2); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (png_size_t)(((row_info->width + 3) & 3) << 1); - dshift = (png_size_t)(((final_width + 3) & 3) << 1); - s_start = 6; - s_end = 0; - s_inc = -2; - } - else -#endif - { - sshift = (png_size_t)((3 - ((row_info->width + 3) & 3)) << 1); - dshift = (png_size_t)((3 - ((final_width + 3) & 3)) << 1); - s_start = 0; - s_end = 6; - s_inc = 2; - } - - for (i = row_info->width; i; i--) - { - png_byte v; - int j; - - v = (png_byte)((*sp >> sshift) & 0x3); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0x3f3f >> (6 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - case 4: - { - png_bytep sp, dp; - int sshift, dshift; - int s_start, s_end, s_inc; - png_uint_32 i; - - sp = row + (png_size_t)((row_info->width - 1) >> 1); - dp = row + (png_size_t)((final_width - 1) >> 1); -#if defined(PNG_READ_PACKSWAP_SUPPORTED) - if (transformations & PNG_PACKSWAP) - { - sshift = (png_size_t)(((row_info->width + 1) & 1) << 2); - dshift = (png_size_t)(((final_width + 1) & 1) << 2); - s_start = 4; - s_end = 0; - s_inc = -4; - } - else -#endif - { - sshift = (png_size_t)((1 - ((row_info->width + 1) & 1)) << 2); - dshift = (png_size_t)((1 - ((final_width + 1) & 1)) << 2); - s_start = 0; - s_end = 4; - s_inc = 4; - } - - for (i = row_info->width; i; i--) - { - png_byte v; - int j; - - v = (png_byte)((*sp >> sshift) & 0xf); - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp &= (png_byte)((0xf0f >> (4 - dshift)) & 0xff); - *dp |= (png_byte)(v << dshift); - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - else - dshift += s_inc; - } - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - else - sshift += s_inc; - } - break; - } - - default: // This is the place where the routine is modified - { - __int64 const4 = 0x0000000000FFFFFF; - // __int64 const5 = 0x000000FFFFFF0000; // unused... - __int64 const6 = 0x00000000000000FF; - png_bytep sptr, dp; - png_uint_32 i; - png_size_t pixel_bytes; - int width = row_info->width; - - pixel_bytes = (row_info->pixel_depth >> 3); - - sptr = row + (width - 1) * pixel_bytes; - dp = row + (final_width - 1) * pixel_bytes; - // New code by Nirav Chhatrapati - Intel Corporation - // sign fix by GRR - // NOTE: there is NO MMX code for 48-bit and 64-bit images - - // use MMX routine if machine supports it -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_INTERLACE) - /* && mmx_supported */ ) -#else - if (mmx_supported) -#endif - { - if (pixel_bytes == 3) - { - if (((pass == 0) || (pass == 1)) && width) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width - sub edi, 21 // (png_pass_inc[pass] - 1)*pixel_bytes -loop_pass0: - movd mm0, [esi] ; X X X X X v2 v1 v0 - pand mm0, const4 ; 0 0 0 0 0 v2 v1 v0 - movq mm1, mm0 ; 0 0 0 0 0 v2 v1 v0 - psllq mm0, 16 ; 0 0 0 v2 v1 v0 0 0 - movq mm2, mm0 ; 0 0 0 v2 v1 v0 0 0 - psllq mm0, 24 ; v2 v1 v0 0 0 0 0 0 - psrlq mm1, 8 ; 0 0 0 0 0 0 v2 v1 - por mm0, mm2 ; v2 v1 v0 v2 v1 v0 0 0 - por mm0, mm1 ; v2 v1 v0 v2 v1 v0 v2 v1 - movq mm3, mm0 ; v2 v1 v0 v2 v1 v0 v2 v1 - psllq mm0, 16 ; v0 v2 v1 v0 v2 v1 0 0 - movq mm4, mm3 ; v2 v1 v0 v2 v1 v0 v2 v1 - punpckhdq mm3, mm0 ; v0 v2 v1 v0 v2 v1 v0 v2 - movq [edi+16] , mm4 - psrlq mm0, 32 ; 0 0 0 0 v0 v2 v1 v0 - movq [edi+8] , mm3 - punpckldq mm0, mm4 ; v1 v0 v2 v1 v0 v2 v1 v0 - sub esi, 3 - movq [edi], mm0 - sub edi, 24 - //sub esi, 3 - dec ecx - jnz loop_pass0 - EMMS - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width - sub edi, 9 // (png_pass_inc[pass] - 1)*pixel_bytes -loop_pass2: - movd mm0, [esi] ; X X X X X v2 v1 v0 - pand mm0, const4 ; 0 0 0 0 0 v2 v1 v0 - movq mm1, mm0 ; 0 0 0 0 0 v2 v1 v0 - psllq mm0, 16 ; 0 0 0 v2 v1 v0 0 0 - movq mm2, mm0 ; 0 0 0 v2 v1 v0 0 0 - psllq mm0, 24 ; v2 v1 v0 0 0 0 0 0 - psrlq mm1, 8 ; 0 0 0 0 0 0 v2 v1 - por mm0, mm2 ; v2 v1 v0 v2 v1 v0 0 0 - por mm0, mm1 ; v2 v1 v0 v2 v1 v0 v2 v1 - movq [edi+4], mm0 ; move to memory - psrlq mm0, 16 ; 0 0 v2 v1 v0 v2 v1 v0 - movd [edi], mm0 ; move to memory - sub esi, 3 - sub edi, 12 - dec ecx - jnz loop_pass2 - EMMS - } - } - else if (width) /* && ((pass == 4) || (pass == 5)) */ - { - int width_mmx = ((width >> 1) << 1) - 8; - if (width_mmx < 0) - width_mmx = 0; - width -= width_mmx; // 8 or 9 pix, 24 or 27 bytes - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 3 - sub edi, 9 -loop_pass4: - movq mm0, [esi] ; X X v2 v1 v0 v5 v4 v3 - movq mm7, mm0 ; X X v2 v1 v0 v5 v4 v3 - movq mm6, mm0 ; X X v2 v1 v0 v5 v4 v3 - psllq mm0, 24 ; v1 v0 v5 v4 v3 0 0 0 - pand mm7, const4 ; 0 0 0 0 0 v5 v4 v3 - psrlq mm6, 24 ; 0 0 0 X X v2 v1 v0 - por mm0, mm7 ; v1 v0 v5 v4 v3 v5 v4 v3 - movq mm5, mm6 ; 0 0 0 X X v2 v1 v0 - psllq mm6, 8 ; 0 0 X X v2 v1 v0 0 - movq [edi], mm0 ; move quad to memory - psrlq mm5, 16 ; 0 0 0 0 0 X X v2 - pand mm5, const6 ; 0 0 0 0 0 0 0 v2 - por mm6, mm5 ; 0 0 X X v2 v1 v0 v2 - movd [edi+8], mm6 ; move double to memory - sub esi, 6 - sub edi, 12 - sub ecx, 2 - jnz loop_pass4 - EMMS - } - } - - sptr -= width_mmx*3; - dp -= width_mmx*6; - for (i = width; i; i--) - { - png_byte v[8]; - int j; - - png_memcpy(v, sptr, 3); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 3); - dp -= 3; - } - sptr -= 3; - } - } - } /* end of pixel_bytes == 3 */ - - else if (pixel_bytes == 1) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 2) << 2); - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub edi, 31 - sub esi, 3 -loop1_pass0: - movd mm0, [esi] ; X X X X v0 v1 v2 v3 - movq mm1, mm0 ; X X X X v0 v1 v2 v3 - punpcklbw mm0, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3 - movq mm2, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3 - punpcklwd mm0, mm0 ; v2 v2 v2 v2 v3 v3 v3 v3 - movq mm3, mm0 ; v2 v2 v2 v2 v3 v3 v3 v3 - punpckldq mm0, mm0 ; v3 v3 v3 v3 v3 v3 v3 v3 - punpckhdq mm3, mm3 ; v2 v2 v2 v2 v2 v2 v2 v2 - movq [edi], mm0 ; move to memory v3 - punpckhwd mm2, mm2 ; v0 v0 v0 v0 v1 v1 v1 v1 - movq [edi+8], mm3 ; move to memory v2 - movq mm4, mm2 ; v0 v0 v0 v0 v1 v1 v1 v1 - punpckldq mm2, mm2 ; v1 v1 v1 v1 v1 v1 v1 v1 - punpckhdq mm4, mm4 ; v0 v0 v0 v0 v0 v0 v0 v0 - movq [edi+16], mm2 ; move to memory v1 - movq [edi+24], mm4 ; move to memory v0 - sub esi, 4 - sub edi, 32 - sub ecx, 4 - jnz loop1_pass0 - EMMS - } - } - - sptr -= width_mmx; - dp -= width_mmx*8; - for (i = width; i; i--) - { - int j; - - /* I simplified this part in version 1.0.4e - * here and in several other instances where - * pixel_bytes == 1 -- GR-P - * - * Original code: - * - * png_byte v[8]; - * png_memcpy(v, sptr, pixel_bytes); - * for (j = 0; j < png_pass_inc[pass]; j++) - * { - * png_memcpy(dp, v, pixel_bytes); - * dp -= pixel_bytes; - * } - * sptr -= pixel_bytes; - * - * Replacement code is in the next three lines: - */ - - for (j = 0; j < png_pass_inc[pass]; j++) - *dp-- = *sptr; - sptr--; - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 2) << 2); - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub edi, 15 - sub esi, 3 -loop1_pass2: - movd mm0, [esi] ; X X X X v0 v1 v2 v3 - punpcklbw mm0, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3 - movq mm1, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3 - punpcklwd mm0, mm0 ; v2 v2 v2 v2 v3 v3 v3 v3 - punpckhwd mm1, mm1 ; v0 v0 v0 v0 v1 v1 v1 v1 - movq [edi], mm0 ; move to memory v2 and v3 - sub esi, 4 - movq [edi+8], mm1 ; move to memory v1 and v0 - sub edi, 16 - sub ecx, 4 - jnz loop1_pass2 - EMMS - } - } - - sptr -= width_mmx; - dp -= width_mmx*4; - for (i = width; i; i--) - { - int j; - - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - sptr --; - } - } - else if (width) /* && ((pass == 4) || (pass == 5))) */ - { - int width_mmx = ((width >> 3) << 3); - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub edi, 15 - sub esi, 7 -loop1_pass4: - movq mm0, [esi] ; v0 v1 v2 v3 v4 v5 v6 v7 - movq mm1, mm0 ; v0 v1 v2 v3 v4 v5 v6 v7 - punpcklbw mm0, mm0 ; v4 v4 v5 v5 v6 v6 v7 v7 - //movq mm1, mm0 ; v0 v0 v1 v1 v2 v2 v3 v3 - punpckhbw mm1, mm1 ;v0 v0 v1 v1 v2 v2 v3 v3 - movq [edi+8], mm1 ; move to memory v0 v1 v2 and v3 - sub esi, 8 - movq [edi], mm0 ; move to memory v4 v5 v6 and v7 - //sub esi, 4 - sub edi, 16 - sub ecx, 8 - jnz loop1_pass4 - EMMS - } - } - - sptr -= width_mmx; - dp -= width_mmx*2; - for (i = width; i; i--) - { - int j; - - for (j = 0; j < png_pass_inc[pass]; j++) - { - *dp-- = *sptr; - } - sptr --; - } - } - } /* end of pixel_bytes == 1 */ - - else if (pixel_bytes == 2) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 1) << 1); - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 2 - sub edi, 30 -loop2_pass0: - movd mm0, [esi] ; X X X X v1 v0 v3 v2 - punpcklwd mm0, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2 - movq mm1, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2 - punpckldq mm0, mm0 ; v3 v2 v3 v2 v3 v2 v3 v2 - punpckhdq mm1, mm1 ; v1 v0 v1 v0 v1 v0 v1 v0 - movq [edi], mm0 - movq [edi + 8], mm0 - movq [edi + 16], mm1 - movq [edi + 24], mm1 - sub esi, 4 - sub edi, 32 - sub ecx, 2 - jnz loop2_pass0 - EMMS - } - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*16 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 2 - sub edi, 14 -loop2_pass2: - movd mm0, [esi] ; X X X X v1 v0 v3 v2 - punpcklwd mm0, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2 - movq mm1, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2 - punpckldq mm0, mm0 ; v3 v2 v3 v2 v3 v2 v3 v2 - punpckhdq mm1, mm1 ; v1 v0 v1 v0 v1 v0 v1 v0 - movq [edi], mm0 - sub esi, 4 - movq [edi + 8], mm1 - //sub esi, 4 - sub edi, 16 - sub ecx, 2 - jnz loop2_pass2 - EMMS - } - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*8 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - else if (width) // pass == 4 or 5 - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 2 - sub edi, 6 -loop2_pass4: - movd mm0, [esi] ; X X X X v1 v0 v3 v2 - punpcklwd mm0, mm0 ; v1 v0 v1 v0 v3 v2 v3 v2 - sub esi, 4 - movq [edi], mm0 - sub edi, 8 - sub ecx, 2 - jnz loop2_pass4 - EMMS - } - } - - sptr -= (width_mmx*2 - 2); // sign fixed - dp -= (width_mmx*4 - 2); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 2; - png_memcpy(v, sptr, 2); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 2; - png_memcpy(dp, v, 2); - } - } - } - } /* end of pixel_bytes == 2 */ - - else if (pixel_bytes == 4) - { - if (((pass == 0) || (pass == 1)) && width) - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 4 - sub edi, 60 -loop4_pass0: - movq mm0, [esi] ; v3 v2 v1 v0 v7 v6 v5 v4 - movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4 - punpckldq mm0, mm0 ; v7 v6 v5 v4 v7 v6 v5 v4 - punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0 - movq [edi], mm0 - movq [edi + 8], mm0 - movq [edi + 16], mm0 - movq [edi + 24], mm0 - movq [edi+32], mm1 - movq [edi + 40], mm1 - movq [edi+ 48], mm1 - sub esi, 8 - movq [edi + 56], mm1 - sub edi, 64 - sub ecx, 2 - jnz loop4_pass0 - EMMS - } - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*32 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - else if (((pass == 2) || (pass == 3)) && width) - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 4 - sub edi, 28 -loop4_pass2: - movq mm0, [esi] ; v3 v2 v1 v0 v7 v6 v5 v4 - movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4 - punpckldq mm0, mm0 ; v7 v6 v5 v4 v7 v6 v5 v4 - punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0 - movq [edi], mm0 - movq [edi + 8], mm0 - movq [edi+16], mm1 - movq [edi + 24], mm1 - sub esi, 8 - sub edi, 32 - sub ecx, 2 - jnz loop4_pass2 - EMMS - } - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*16 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - else if (width) // pass == 4 or 5 - { - int width_mmx = ((width >> 1) << 1) ; - width -= width_mmx; - if (width_mmx) - { - _asm - { - mov esi, sptr - mov edi, dp - mov ecx, width_mmx - sub esi, 4 - sub edi, 12 -loop4_pass4: - movq mm0, [esi] ; v3 v2 v1 v0 v7 v6 v5 v4 - movq mm1, mm0 ; v3 v2 v1 v0 v7 v6 v5 v4 - punpckldq mm0, mm0 ; v7 v6 v5 v4 v7 v6 v5 v4 - punpckhdq mm1, mm1 ; v3 v2 v1 v0 v3 v2 v1 v0 - movq [edi], mm0 - sub esi, 8 - movq [edi + 8], mm1 - sub edi, 16 - sub ecx, 2 - jnz loop4_pass4 - EMMS - } - } - - sptr -= (width_mmx*4 - 4); // sign fixed - dp -= (width_mmx*8 - 4); // sign fixed - for (i = width; i; i--) - { - png_byte v[8]; - int j; - sptr -= 4; - png_memcpy(v, sptr, 4); - for (j = 0; j < png_pass_inc[pass]; j++) - { - dp -= 4; - png_memcpy(dp, v, 4); - } - } - } - - } /* end of pixel_bytes == 4 */ - - else if (pixel_bytes == 6) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, 6); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, 6); - dp -= 6; - } - sptr -= 6; - } - } /* end of pixel_bytes == 6 */ - - else - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr-= pixel_bytes; - } - } - } /* end of mmx_supported */ - - else /* MMX not supported: use modified C code - takes advantage - * of inlining of memcpy for a constant */ - { - if (pixel_bytes == 1) - { - for (i = width; i; i--) - { - int j; - for (j = 0; j < png_pass_inc[pass]; j++) - *dp-- = *sptr; - sptr--; - } - } - else if (pixel_bytes == 3) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - else if (pixel_bytes == 2) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - else if (pixel_bytes == 4) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - else if (pixel_bytes == 6) - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - else - { - for (i = width; i; i--) - { - png_byte v[8]; - int j; - png_memcpy(v, sptr, pixel_bytes); - for (j = 0; j < png_pass_inc[pass]; j++) - { - png_memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - sptr -= pixel_bytes; - } - } - - } /* end of MMX not supported */ - break; - } - } /* end switch (row_info->pixel_depth) */ - - row_info->width = final_width; - - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth,final_width); - } - -} - -#endif /* PNG_READ_INTERLACING_SUPPORTED */ - - -// These variables are utilized in the functions below. They are declared -// globally here to ensure alignment on 8-byte boundaries. - -union uAll { - __int64 use; - double align; -} LBCarryMask = {0x0101010101010101}, - HBClearMask = {0x7f7f7f7f7f7f7f7f}, - ActiveMask, ActiveMask2, ActiveMaskEnd, ShiftBpp, ShiftRem; - - -// Optimized code for PNG Average filter decoder -void /* PRIVATE */ -png_read_filter_row_mmx_avg(png_row_infop row_info, png_bytep row - , png_bytep prev_row) -{ - int bpp; - png_uint_32 FullLength; - png_uint_32 MMXLength; - //png_uint_32 len; - int diff; - - bpp = (row_info->pixel_depth + 7) >> 3; // Get # bytes per pixel - FullLength = row_info->rowbytes; // # of bytes to filter - _asm { - // Init address pointers and offset - mov edi, row // edi ==> Avg(x) - xor ebx, ebx // ebx ==> x - mov edx, edi - mov esi, prev_row // esi ==> Prior(x) - sub edx, bpp // edx ==> Raw(x-bpp) - - xor eax, eax - // Compute the Raw value for the first bpp bytes - // Raw(x) = Avg(x) + (Prior(x)/2) -davgrlp: - mov al, [esi + ebx] // Load al with Prior(x) - inc ebx - shr al, 1 // divide by 2 - add al, [edi+ebx-1] // Add Avg(x); -1 to offset inc ebx - cmp ebx, bpp - mov [edi+ebx-1], al // Write back Raw(x); - // mov does not affect flags; -1 to offset inc ebx - jb davgrlp - // get # of bytes to alignment - mov diff, edi // take start of row - add diff, ebx // add bpp - add diff, 0xf // add 7 + 8 to incr past alignment boundary - and diff, 0xfffffff8 // mask to alignment boundary - sub diff, edi // subtract from start ==> value ebx at alignment - jz davggo - // fix alignment - // Compute the Raw value for the bytes upto the alignment boundary - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - xor ecx, ecx -davglp1: - xor eax, eax - mov cl, [esi + ebx] // load cl with Prior(x) - mov al, [edx + ebx] // load al with Raw(x-bpp) - add ax, cx - inc ebx - shr ax, 1 // divide by 2 - add al, [edi+ebx-1] // Add Avg(x); -1 to offset inc ebx - cmp ebx, diff // Check if at alignment boundary - mov [edi+ebx-1], al // Write back Raw(x); - // mov does not affect flags; -1 to offset inc ebx - jb davglp1 // Repeat until at alignment boundary -davggo: - mov eax, FullLength - mov ecx, eax - sub eax, ebx // subtract alignment fix - and eax, 0x00000007 // calc bytes over mult of 8 - sub ecx, eax // drop over bytes from original length - mov MMXLength, ecx - } // end _asm block - // Now do the math for the rest of the row - switch ( bpp ) - { - case 3: - { - ActiveMask.use = 0x0000000000ffffff; - ShiftBpp.use = 24; // == 3 * 8 - ShiftRem.use = 40; // == 64 - 24 - _asm { - // Re-init address pointers and offset - movq mm7, ActiveMask - mov ebx, diff // ebx ==> x = offset to alignment boundary - movq mm5, LBCarryMask - mov edi, row // edi ==> Avg(x) - movq mm4, HBClearMask - mov esi, prev_row // esi ==> Prior(x) - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm2, [edi + ebx - 8] // Load previous aligned 8 bytes - // (we correct position in loop below) -davg3lp: - movq mm0, [edi + ebx] // Load mm0 with Avg(x) - // Add (Prev_row/2) to Average - movq mm3, mm5 - psrlq mm2, ShiftRem // Correct position Raw(x-bpp) data - movq mm1, [esi + ebx] // Load mm1 with Prior(x) - movq mm6, mm7 - pand mm3, mm1 // get lsb for each prev_row byte - psrlq mm1, 1 // divide prev_row bytes by 2 - pand mm1, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte - // Add 1st active group (Raw(x-bpp)/2) to Average with LBCarry - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 1 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active - // byte - // Add 2nd active group (Raw(x-bpp)/2) to Average with LBCarry - psllq mm6, ShiftBpp // shift the mm6 mask to cover bytes 3-5 - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active - // byte - - // Add 3rd active group (Raw(x-bpp)/2) to Average with LBCarry - psllq mm6, ShiftBpp // shift the mm6 mask to cover the last two - // bytes - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - // Data only needs to be shifted once here to - // get the correct x-bpp offset. - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - add ebx, 8 - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active - // byte - - // Now ready to write back to memory - movq [edi + ebx - 8], mm0 - // Move updated Raw(x) to use as Raw(x-bpp) for next loop - cmp ebx, MMXLength - movq mm2, mm0 // mov updated Raw(x) to mm2 - jb davg3lp - } // end _asm block - } - break; - - case 6: - case 4: - case 7: - case 5: - { - ActiveMask.use = 0xffffffffffffffff; // use shift below to clear - // appropriate inactive bytes - ShiftBpp.use = bpp << 3; - ShiftRem.use = 64 - ShiftBpp.use; - _asm { - movq mm4, HBClearMask - // Re-init address pointers and offset - mov ebx, diff // ebx ==> x = offset to alignment boundary - // Load ActiveMask and clear all bytes except for 1st active group - movq mm7, ActiveMask - mov edi, row // edi ==> Avg(x) - psrlq mm7, ShiftRem - mov esi, prev_row // esi ==> Prior(x) - movq mm6, mm7 - movq mm5, LBCarryMask - psllq mm6, ShiftBpp // Create mask for 2nd active group - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm2, [edi + ebx - 8] // Load previous aligned 8 bytes - // (we correct position in loop below) -davg4lp: - movq mm0, [edi + ebx] - psrlq mm2, ShiftRem // shift data to position correctly - movq mm1, [esi + ebx] - // Add (Prev_row/2) to Average - movq mm3, mm5 - pand mm3, mm1 // get lsb for each prev_row byte - psrlq mm1, 1 // divide prev_row bytes by 2 - pand mm1, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte - // Add 1st active group (Raw(x-bpp)/2) to Average with LBCarry - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm7 // Leave only Active Group 1 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active - // byte - // Add 2nd active group (Raw(x-bpp)/2) to Average with LBCarry - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - add ebx, 8 - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active - // byte - cmp ebx, MMXLength - // Now ready to write back to memory - movq [edi + ebx - 8], mm0 - // Prep Raw(x-bpp) for next loop - movq mm2, mm0 // mov updated Raws to mm2 - jb davg4lp - } // end _asm block - } - break; - case 2: - { - ActiveMask.use = 0x000000000000ffff; - ShiftBpp.use = 16; // == 2 * 8 [BUGFIX] - ShiftRem.use = 48; // == 64 - 16 [BUGFIX] - _asm { - // Load ActiveMask - movq mm7, ActiveMask - // Re-init address pointers and offset - mov ebx, diff // ebx ==> x = offset to alignment boundary - movq mm5, LBCarryMask - mov edi, row // edi ==> Avg(x) - movq mm4, HBClearMask - mov esi, prev_row // esi ==> Prior(x) - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm2, [edi + ebx - 8] // Load previous aligned 8 bytes - // (we correct position in loop below) -davg2lp: - movq mm0, [edi + ebx] - psrlq mm2, ShiftRem // shift data to position correctly [BUGFIX] - movq mm1, [esi + ebx] - // Add (Prev_row/2) to Average - movq mm3, mm5 - pand mm3, mm1 // get lsb for each prev_row byte - psrlq mm1, 1 // divide prev_row bytes by 2 - pand mm1, mm4 // clear invalid bit 7 of each byte - movq mm6, mm7 - paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte - // Add 1st active group (Raw(x-bpp)/2) to Average with LBCarry - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 1 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active byte - // Add 2nd active group (Raw(x-bpp)/2) to Average with LBCarry - psllq mm6, ShiftBpp // shift the mm6 mask to cover bytes 2 & 3 - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active byte - - // Add rdd active group (Raw(x-bpp)/2) to Average with LBCarry - psllq mm6, ShiftBpp // shift the mm6 mask to cover bytes 4 & 5 - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - // Data only needs to be shifted once here to - // get the correct x-bpp offset. - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active byte - - // Add 4th active group (Raw(x-bpp)/2) to Average with LBCarry - psllq mm6, ShiftBpp // shift the mm6 mask to cover bytes 6 & 7 - movq mm2, mm0 // mov updated Raws to mm2 - psllq mm2, ShiftBpp // shift data to position correctly - // Data only needs to be shifted once here to - // get the correct x-bpp offset. - add ebx, 8 - movq mm1, mm3 // now use mm1 for getting LBCarrys - pand mm1, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 (Only valid for active group) - psrlq mm2, 1 // divide raw bytes by 2 - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm2, mm1 // add LBCarrys to (Raw(x-bpp)/2) for each byte - pand mm2, mm6 // Leave only Active Group 2 bytes to add to Avg - paddb mm0, mm2 // add (Raw/2) + LBCarrys to Avg for each Active byte - - cmp ebx, MMXLength - // Now ready to write back to memory - movq [edi + ebx - 8], mm0 - // Prep Raw(x-bpp) for next loop - movq mm2, mm0 // mov updated Raws to mm2 - jb davg2lp - } // end _asm block - } - break; - - case 1: // bpp == 1 - { - _asm { - // Re-init address pointers and offset - mov ebx, diff // ebx ==> x = offset to alignment boundary - mov edi, row // edi ==> Avg(x) - cmp ebx, FullLength // Test if offset at end of array - jnb davg1end - // Do Paeth decode for remaining bytes - mov esi, prev_row // esi ==> Prior(x) - mov edx, edi - xor ecx, ecx // zero ecx before using cl & cx in loop below - sub edx, bpp // edx ==> Raw(x-bpp) -davg1lp: - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - xor eax, eax - mov cl, [esi + ebx] // load cl with Prior(x) - mov al, [edx + ebx] // load al with Raw(x-bpp) - add ax, cx - inc ebx - shr ax, 1 // divide by 2 - add al, [edi+ebx-1] // Add Avg(x); -1 to offset inc ebx - cmp ebx, FullLength // Check if at end of array - mov [edi+ebx-1], al // Write back Raw(x); - // mov does not affect flags; -1 to offset inc ebx - jb davg1lp -davg1end: - } // end _asm block - } - return; - - case 8: // bpp == 8 - { - _asm { - // Re-init address pointers and offset - mov ebx, diff // ebx ==> x = offset to alignment boundary - movq mm5, LBCarryMask - mov edi, row // edi ==> Avg(x) - movq mm4, HBClearMask - mov esi, prev_row // esi ==> Prior(x) - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm2, [edi + ebx - 8] // Load previous aligned 8 bytes - // (NO NEED to correct position in loop below) -davg8lp: - movq mm0, [edi + ebx] - movq mm3, mm5 - movq mm1, [esi + ebx] - add ebx, 8 - pand mm3, mm1 // get lsb for each prev_row byte - psrlq mm1, 1 // divide prev_row bytes by 2 - pand mm3, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 - psrlq mm2, 1 // divide raw bytes by 2 - pand mm1, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm3 // add LBCarrys to Avg for each byte - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte - paddb mm0, mm2 // add (Raw/2) to Avg for each byte - cmp ebx, MMXLength - movq [edi + ebx - 8], mm0 - movq mm2, mm0 // reuse as Raw(x-bpp) - jb davg8lp - } // end _asm block - } - break; - default: // bpp greater than 8 - { - _asm { - movq mm5, LBCarryMask - // Re-init address pointers and offset - mov ebx, diff // ebx ==> x = offset to alignment boundary - mov edi, row // edi ==> Avg(x) - movq mm4, HBClearMask - mov edx, edi - mov esi, prev_row // esi ==> Prior(x) - sub edx, bpp // edx ==> Raw(x-bpp) -davgAlp: - movq mm0, [edi + ebx] - movq mm3, mm5 - movq mm1, [esi + ebx] - pand mm3, mm1 // get lsb for each prev_row byte - movq mm2, [edx + ebx] - psrlq mm1, 1 // divide prev_row bytes by 2 - pand mm3, mm2 // get LBCarrys for each byte where both - // lsb's were == 1 - psrlq mm2, 1 // divide raw bytes by 2 - pand mm1, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm3 // add LBCarrys to Avg for each byte - pand mm2, mm4 // clear invalid bit 7 of each byte - paddb mm0, mm1 // add (Prev_row/2) to Avg for each byte - add ebx, 8 - paddb mm0, mm2 // add (Raw/2) to Avg for each byte - cmp ebx, MMXLength - movq [edi + ebx - 8], mm0 - jb davgAlp - } // end _asm block - } - break; - } // end switch ( bpp ) - - _asm { - // MMX acceleration complete now do clean-up - // Check if any remaining bytes left to decode - mov ebx, MMXLength // ebx ==> x = offset bytes remaining after MMX - mov edi, row // edi ==> Avg(x) - cmp ebx, FullLength // Test if offset at end of array - jnb davgend - // Do Paeth decode for remaining bytes - mov esi, prev_row // esi ==> Prior(x) - mov edx, edi - xor ecx, ecx // zero ecx before using cl & cx in loop below - sub edx, bpp // edx ==> Raw(x-bpp) -davglp2: - // Raw(x) = Avg(x) + ((Raw(x-bpp) + Prior(x))/2) - xor eax, eax - mov cl, [esi + ebx] // load cl with Prior(x) - mov al, [edx + ebx] // load al with Raw(x-bpp) - add ax, cx - inc ebx - shr ax, 1 // divide by 2 - add al, [edi+ebx-1] // Add Avg(x); -1 to offset inc ebx - cmp ebx, FullLength // Check if at end of array - mov [edi+ebx-1], al // Write back Raw(x); - // mov does not affect flags; -1 to offset inc ebx - jb davglp2 -davgend: - emms // End MMX instructions; prep for possible FP instrs. - } // end _asm block -} - -// Optimized code for PNG Paeth filter decoder -void /* PRIVATE */ -png_read_filter_row_mmx_paeth(png_row_infop row_info, png_bytep row, - png_bytep prev_row) -{ - png_uint_32 FullLength; - png_uint_32 MMXLength; - //png_uint_32 len; - int bpp; - int diff; - //int ptemp; - int patemp, pbtemp, pctemp; - - bpp = (row_info->pixel_depth + 7) >> 3; // Get # bytes per pixel - FullLength = row_info->rowbytes; // # of bytes to filter - _asm - { - xor ebx, ebx // ebx ==> x offset - mov edi, row - xor edx, edx // edx ==> x-bpp offset - mov esi, prev_row - xor eax, eax - - // Compute the Raw value for the first bpp bytes - // Note: the formula works out to be always - // Paeth(x) = Raw(x) + Prior(x) where x < bpp -dpthrlp: - mov al, [edi + ebx] - add al, [esi + ebx] - inc ebx - cmp ebx, bpp - mov [edi + ebx - 1], al - jb dpthrlp - // get # of bytes to alignment - mov diff, edi // take start of row - add diff, ebx // add bpp - xor ecx, ecx - add diff, 0xf // add 7 + 8 to incr past alignment boundary - and diff, 0xfffffff8 // mask to alignment boundary - sub diff, edi // subtract from start ==> value ebx at alignment - jz dpthgo - // fix alignment -dpthlp1: - xor eax, eax - // pav = p - a = (a + b - c) - a = b - c - mov al, [esi + ebx] // load Prior(x) into al - mov cl, [esi + edx] // load Prior(x-bpp) into cl - sub eax, ecx // subtract Prior(x-bpp) - mov patemp, eax // Save pav for later use - xor eax, eax - // pbv = p - b = (a + b - c) - b = a - c - mov al, [edi + edx] // load Raw(x-bpp) into al - sub eax, ecx // subtract Prior(x-bpp) - mov ecx, eax - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - add eax, patemp // pcv = pav + pbv - // pc = abs(pcv) - test eax, 0x80000000 - jz dpthpca - neg eax // reverse sign of neg values -dpthpca: - mov pctemp, eax // save pc for later use - // pb = abs(pbv) - test ecx, 0x80000000 - jz dpthpba - neg ecx // reverse sign of neg values -dpthpba: - mov pbtemp, ecx // save pb for later use - // pa = abs(pav) - mov eax, patemp - test eax, 0x80000000 - jz dpthpaa - neg eax // reverse sign of neg values -dpthpaa: - mov patemp, eax // save pa for later use - // test if pa <= pb - cmp eax, ecx - jna dpthabb - // pa > pb; now test if pb <= pc - cmp ecx, pctemp - jna dpthbbc - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthpaeth -dpthbbc: - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - mov cl, [esi + ebx] // load Prior(x) into cl - jmp dpthpaeth -dpthabb: - // pa <= pb; now test if pa <= pc - cmp eax, pctemp - jna dpthabc - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthpaeth -dpthabc: - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - mov cl, [edi + edx] // load Raw(x-bpp) into cl -dpthpaeth: - inc ebx - inc edx - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - add [edi + ebx - 1], cl - cmp ebx, diff - jb dpthlp1 -dpthgo: - mov ecx, FullLength - mov eax, ecx - sub eax, ebx // subtract alignment fix - and eax, 0x00000007 // calc bytes over mult of 8 - sub ecx, eax // drop over bytes from original length - mov MMXLength, ecx - } // end _asm block - // Now do the math for the rest of the row - switch ( bpp ) - { - case 3: - { - ActiveMask.use = 0x0000000000ffffff; - ActiveMaskEnd.use = 0xffff000000000000; - ShiftBpp.use = 24; // == bpp(3) * 8 - ShiftRem.use = 40; // == 64 - 24 - _asm - { - mov ebx, diff - mov edi, row - mov esi, prev_row - pxor mm0, mm0 - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] -dpth3lp: - psrlq mm1, ShiftRem // shift last 3 bytes to 1st 3 bytes - movq mm2, [esi + ebx] // load b=Prior(x) - punpcklbw mm1, mm0 // Unpack High bytes of a - movq mm3, [esi+ebx-8] // Prep c=Prior(x-bpp) bytes - punpcklbw mm2, mm0 // Unpack High bytes of b - psrlq mm3, ShiftRem // shift last 3 bytes to 1st 3 bytes - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - punpcklbw mm3, mm0 // Unpack High bytes of c - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - paddw mm7, mm3 - pxor mm0, mm0 - packuswb mm7, mm1 - movq mm3, [esi + ebx] // load c=Prior(x-bpp) - pand mm7, ActiveMask - movq mm2, mm3 // load b=Prior(x) step 1 - paddb mm7, [edi + ebx] // add Paeth predictor with Raw(x) - punpcklbw mm3, mm0 // Unpack High bytes of c - movq [edi + ebx], mm7 // write back updated value - movq mm1, mm7 // Now mm1 will be used as Raw(x-bpp) - // Now do Paeth for 2nd set of bytes (3-5) - psrlq mm2, ShiftBpp // load b=Prior(x) step 2 - punpcklbw mm1, mm0 // Unpack High bytes of a - pxor mm7, mm7 - punpcklbw mm2, mm0 // Unpack High bytes of b - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - psubw mm5, mm3 - psubw mm4, mm3 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = - // pav + pbv = pbv + pav - movq mm6, mm5 - paddw mm6, mm4 - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm5 // Create mask pbv bytes < 0 - pcmpgtw mm7, mm4 // Create mask pav bytes < 0 - pand mm0, mm5 // Only pbv bytes < 0 in mm0 - pand mm7, mm4 // Only pav bytes < 0 in mm7 - psubw mm5, mm0 - psubw mm4, mm7 - psubw mm5, mm0 - psubw mm4, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - movq mm2, [esi + ebx] // load b=Prior(x) - pand mm3, mm7 - pandn mm7, mm0 - pxor mm1, mm1 - paddw mm7, mm3 - pxor mm0, mm0 - packuswb mm7, mm1 - movq mm3, mm2 // load c=Prior(x-bpp) step 1 - pand mm7, ActiveMask - punpckhbw mm2, mm0 // Unpack High bytes of b - psllq mm7, ShiftBpp // Shift bytes to 2nd group of 3 bytes - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - paddb mm7, [edi + ebx] // add Paeth predictor with Raw(x) - psllq mm3, ShiftBpp // load c=Prior(x-bpp) step 2 - movq [edi + ebx], mm7 // write back updated value - movq mm1, mm7 - punpckhbw mm3, mm0 // Unpack High bytes of c - psllq mm1, ShiftBpp // Shift bytes - // Now mm1 will be used as Raw(x-bpp) - // Now do Paeth for 3rd, and final, set of bytes (6-7) - pxor mm7, mm7 - punpckhbw mm1, mm0 // Unpack High bytes of a - psubw mm4, mm3 - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - pxor mm0, mm0 - paddw mm6, mm5 - - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - pandn mm0, mm1 - pandn mm7, mm4 - paddw mm0, mm2 - paddw mm7, mm5 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pand mm3, mm7 - pandn mm7, mm0 - paddw mm7, mm3 - pxor mm1, mm1 - packuswb mm1, mm7 - // Step ebx to next set of 8 bytes and repeat loop til done - add ebx, 8 - pand mm1, ActiveMaskEnd - paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x) - - cmp ebx, MMXLength - pxor mm0, mm0 // pxor does not affect flags - movq [edi + ebx - 8], mm1 // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - // mm3 ready to be used as Prior(x-bpp) next loop - jb dpth3lp - } // end _asm block - } - break; - - case 6: - case 7: - case 5: - { - ActiveMask.use = 0x00000000ffffffff; - ActiveMask2.use = 0xffffffff00000000; - ShiftBpp.use = bpp << 3; // == bpp * 8 - ShiftRem.use = 64 - ShiftBpp.use; - _asm - { - mov ebx, diff - mov edi, row - mov esi, prev_row - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] - pxor mm0, mm0 -dpth6lp: - // Must shift to position Raw(x-bpp) data - psrlq mm1, ShiftRem - // Do first set of 4 bytes - movq mm3, [esi+ebx-8] // read c=Prior(x-bpp) bytes - punpcklbw mm1, mm0 // Unpack Low bytes of a - movq mm2, [esi + ebx] // load b=Prior(x) - punpcklbw mm2, mm0 // Unpack Low bytes of b - // Must shift to position Prior(x-bpp) data - psrlq mm3, ShiftRem - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - punpcklbw mm3, mm0 // Unpack Low bytes of c - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - paddw mm7, mm3 - pxor mm0, mm0 - packuswb mm7, mm1 - movq mm3, [esi + ebx - 8] // load c=Prior(x-bpp) - pand mm7, ActiveMask - psrlq mm3, ShiftRem - movq mm2, [esi + ebx] // load b=Prior(x) step 1 - paddb mm7, [edi + ebx] // add Paeth predictor with Raw(x) - movq mm6, mm2 - movq [edi + ebx], mm7 // write back updated value - movq mm1, [edi+ebx-8] - psllq mm6, ShiftBpp - movq mm5, mm7 - psrlq mm1, ShiftRem - por mm3, mm6 - psllq mm5, ShiftBpp - punpckhbw mm3, mm0 // Unpack High bytes of c - por mm1, mm5 - // Do second set of 4 bytes - punpckhbw mm2, mm0 // Unpack High bytes of b - punpckhbw mm1, mm0 // Unpack High bytes of a - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - pxor mm1, mm1 - paddw mm7, mm3 - pxor mm0, mm0 - // Step ex to next set of 8 bytes and repeat loop til done - add ebx, 8 - packuswb mm1, mm7 - paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x) - cmp ebx, MMXLength - movq [edi + ebx - 8], mm1 // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - jb dpth6lp - } // end _asm block - } - break; - - case 4: - { - ActiveMask.use = 0x00000000ffffffff; - _asm { - mov ebx, diff - mov edi, row - mov esi, prev_row - pxor mm0, mm0 - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] // Only time should need to read - // a=Raw(x-bpp) bytes -dpth4lp: - // Do first set of 4 bytes - movq mm3, [esi+ebx-8] // read c=Prior(x-bpp) bytes - punpckhbw mm1, mm0 // Unpack Low bytes of a - movq mm2, [esi + ebx] // load b=Prior(x) - punpcklbw mm2, mm0 // Unpack High bytes of b - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - punpckhbw mm3, mm0 // Unpack High bytes of c - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - paddw mm7, mm3 - pxor mm0, mm0 - packuswb mm7, mm1 - movq mm3, [esi + ebx] // load c=Prior(x-bpp) - pand mm7, ActiveMask - movq mm2, mm3 // load b=Prior(x) step 1 - paddb mm7, [edi + ebx] // add Paeth predictor with Raw(x) - punpcklbw mm3, mm0 // Unpack High bytes of c - movq [edi + ebx], mm7 // write back updated value - movq mm1, mm7 // Now mm1 will be used as Raw(x-bpp) - // Do second set of 4 bytes - punpckhbw mm2, mm0 // Unpack Low bytes of b - punpcklbw mm1, mm0 // Unpack Low bytes of a - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - pxor mm1, mm1 - paddw mm7, mm3 - pxor mm0, mm0 - // Step ex to next set of 8 bytes and repeat loop til done - add ebx, 8 - packuswb mm1, mm7 - paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x) - cmp ebx, MMXLength - movq [edi + ebx - 8], mm1 // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - jb dpth4lp - } // end _asm block - } - break; - case 8: // bpp == 8 - { - ActiveMask.use = 0x00000000ffffffff; - _asm { - mov ebx, diff - mov edi, row - mov esi, prev_row - pxor mm0, mm0 - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] // Only time should need to read - // a=Raw(x-bpp) bytes -dpth8lp: - // Do first set of 4 bytes - movq mm3, [esi+ebx-8] // read c=Prior(x-bpp) bytes - punpcklbw mm1, mm0 // Unpack Low bytes of a - movq mm2, [esi + ebx] // load b=Prior(x) - punpcklbw mm2, mm0 // Unpack Low bytes of b - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - punpcklbw mm3, mm0 // Unpack Low bytes of c - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - paddw mm7, mm3 - pxor mm0, mm0 - packuswb mm7, mm1 - movq mm3, [esi+ebx-8] // read c=Prior(x-bpp) bytes - pand mm7, ActiveMask - movq mm2, [esi + ebx] // load b=Prior(x) - paddb mm7, [edi + ebx] // add Paeth predictor with Raw(x) - punpckhbw mm3, mm0 // Unpack High bytes of c - movq [edi + ebx], mm7 // write back updated value - movq mm1, [edi+ebx-8] // read a=Raw(x-bpp) bytes - - // Do second set of 4 bytes - punpckhbw mm2, mm0 // Unpack High bytes of b - punpckhbw mm1, mm0 // Unpack High bytes of a - // pav = p - a = (a + b - c) - a = b - c - movq mm4, mm2 - // pbv = p - b = (a + b - c) - b = a - c - movq mm5, mm1 - psubw mm4, mm3 - pxor mm7, mm7 - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - movq mm6, mm4 - psubw mm5, mm3 - // pa = abs(p-a) = abs(pav) - // pb = abs(p-b) = abs(pbv) - // pc = abs(p-c) = abs(pcv) - pcmpgtw mm0, mm4 // Create mask pav bytes < 0 - paddw mm6, mm5 - pand mm0, mm4 // Only pav bytes < 0 in mm7 - pcmpgtw mm7, mm5 // Create mask pbv bytes < 0 - psubw mm4, mm0 - pand mm7, mm5 // Only pbv bytes < 0 in mm0 - psubw mm4, mm0 - psubw mm5, mm7 - pxor mm0, mm0 - pcmpgtw mm0, mm6 // Create mask pcv bytes < 0 - pand mm0, mm6 // Only pav bytes < 0 in mm7 - psubw mm5, mm7 - psubw mm6, mm0 - // test pa <= pb - movq mm7, mm4 - psubw mm6, mm0 - pcmpgtw mm7, mm5 // pa > pb? - movq mm0, mm7 - // use mm7 mask to merge pa & pb - pand mm5, mm7 - // use mm0 mask copy to merge a & b - pand mm2, mm0 - pandn mm7, mm4 - pandn mm0, mm1 - paddw mm7, mm5 - paddw mm0, mm2 - // test ((pa <= pb)? pa:pb) <= pc - pcmpgtw mm7, mm6 // pab > pc? - pxor mm1, mm1 - pand mm3, mm7 - pandn mm7, mm0 - pxor mm1, mm1 - paddw mm7, mm3 - pxor mm0, mm0 - // Step ex to next set of 8 bytes and repeat loop til done - add ebx, 8 - packuswb mm1, mm7 - paddb mm1, [edi + ebx - 8] // add Paeth predictor with Raw(x) - cmp ebx, MMXLength - movq [edi + ebx - 8], mm1 // write back updated value - // mm1 will be used as Raw(x-bpp) next loop - jb dpth8lp - } // end _asm block - } - break; - - case 1: // bpp = 1 - case 2: // bpp = 2 - default: // bpp > 8 - { - _asm { - mov ebx, diff - cmp ebx, FullLength - jnb dpthdend - mov edi, row - mov esi, prev_row - // Do Paeth decode for remaining bytes - mov edx, ebx - xor ecx, ecx // zero ecx before using cl & cx in loop below - sub edx, bpp // Set edx = ebx - bpp -dpthdlp: - xor eax, eax - // pav = p - a = (a + b - c) - a = b - c - mov al, [esi + ebx] // load Prior(x) into al - mov cl, [esi + edx] // load Prior(x-bpp) into cl - sub eax, ecx // subtract Prior(x-bpp) - mov patemp, eax // Save pav for later use - xor eax, eax - // pbv = p - b = (a + b - c) - b = a - c - mov al, [edi + edx] // load Raw(x-bpp) into al - sub eax, ecx // subtract Prior(x-bpp) - mov ecx, eax - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - add eax, patemp // pcv = pav + pbv - // pc = abs(pcv) - test eax, 0x80000000 - jz dpthdpca - neg eax // reverse sign of neg values -dpthdpca: - mov pctemp, eax // save pc for later use - // pb = abs(pbv) - test ecx, 0x80000000 - jz dpthdpba - neg ecx // reverse sign of neg values -dpthdpba: - mov pbtemp, ecx // save pb for later use - // pa = abs(pav) - mov eax, patemp - test eax, 0x80000000 - jz dpthdpaa - neg eax // reverse sign of neg values -dpthdpaa: - mov patemp, eax // save pa for later use - // test if pa <= pb - cmp eax, ecx - jna dpthdabb - // pa > pb; now test if pb <= pc - cmp ecx, pctemp - jna dpthdbbc - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthdpaeth -dpthdbbc: - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - mov cl, [esi + ebx] // load Prior(x) into cl - jmp dpthdpaeth -dpthdabb: - // pa <= pb; now test if pa <= pc - cmp eax, pctemp - jna dpthdabc - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthdpaeth -dpthdabc: - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - mov cl, [edi + edx] // load Raw(x-bpp) into cl -dpthdpaeth: - inc ebx - inc edx - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - add [edi + ebx - 1], cl - cmp ebx, FullLength - jb dpthdlp -dpthdend: - } // end _asm block - } - return; // No need to go further with this one - } // end switch ( bpp ) - _asm - { - // MMX acceleration complete now do clean-up - // Check if any remaining bytes left to decode - mov ebx, MMXLength - cmp ebx, FullLength - jnb dpthend - mov edi, row - mov esi, prev_row - // Do Paeth decode for remaining bytes - mov edx, ebx - xor ecx, ecx // zero ecx before using cl & cx in loop below - sub edx, bpp // Set edx = ebx - bpp -dpthlp2: - xor eax, eax - // pav = p - a = (a + b - c) - a = b - c - mov al, [esi + ebx] // load Prior(x) into al - mov cl, [esi + edx] // load Prior(x-bpp) into cl - sub eax, ecx // subtract Prior(x-bpp) - mov patemp, eax // Save pav for later use - xor eax, eax - // pbv = p - b = (a + b - c) - b = a - c - mov al, [edi + edx] // load Raw(x-bpp) into al - sub eax, ecx // subtract Prior(x-bpp) - mov ecx, eax - // pcv = p - c = (a + b - c) -c = (a - c) + (b - c) = pav + pbv - add eax, patemp // pcv = pav + pbv - // pc = abs(pcv) - test eax, 0x80000000 - jz dpthpca2 - neg eax // reverse sign of neg values -dpthpca2: - mov pctemp, eax // save pc for later use - // pb = abs(pbv) - test ecx, 0x80000000 - jz dpthpba2 - neg ecx // reverse sign of neg values -dpthpba2: - mov pbtemp, ecx // save pb for later use - // pa = abs(pav) - mov eax, patemp - test eax, 0x80000000 - jz dpthpaa2 - neg eax // reverse sign of neg values -dpthpaa2: - mov patemp, eax // save pa for later use - // test if pa <= pb - cmp eax, ecx - jna dpthabb2 - // pa > pb; now test if pb <= pc - cmp ecx, pctemp - jna dpthbbc2 - // pb > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthpaeth2 -dpthbbc2: - // pb <= pc; Raw(x) = Paeth(x) + Prior(x) - mov cl, [esi + ebx] // load Prior(x) into cl - jmp dpthpaeth2 -dpthabb2: - // pa <= pb; now test if pa <= pc - cmp eax, pctemp - jna dpthabc2 - // pa > pc; Raw(x) = Paeth(x) + Prior(x-bpp) - mov cl, [esi + edx] // load Prior(x-bpp) into cl - jmp dpthpaeth2 -dpthabc2: - // pa <= pc; Raw(x) = Paeth(x) + Raw(x-bpp) - mov cl, [edi + edx] // load Raw(x-bpp) into cl -dpthpaeth2: - inc ebx - inc edx - // Raw(x) = (Paeth(x) + Paeth_Predictor( a, b, c )) mod 256 - add [edi + ebx - 1], cl - cmp ebx, FullLength - jb dpthlp2 -dpthend: - emms // End MMX instructions; prep for possible FP instrs. - } // end _asm block -} - -// Optimized code for PNG Sub filter decoder -void /* PRIVATE */ -png_read_filter_row_mmx_sub(png_row_infop row_info, png_bytep row) -{ - //int test; - int bpp; - png_uint_32 FullLength; - png_uint_32 MMXLength; - int diff; - - bpp = (row_info->pixel_depth + 7) >> 3; // Get # bytes per pixel - FullLength = row_info->rowbytes - bpp; // # of bytes to filter - _asm { - mov edi, row - mov esi, edi // lp = row - add edi, bpp // rp = row + bpp - xor eax, eax - // get # of bytes to alignment - mov diff, edi // take start of row - add diff, 0xf // add 7 + 8 to incr past - // alignment boundary - xor ebx, ebx - and diff, 0xfffffff8 // mask to alignment boundary - sub diff, edi // subtract from start ==> value - // ebx at alignment - jz dsubgo - // fix alignment -dsublp1: - mov al, [esi+ebx] - add [edi+ebx], al - inc ebx - cmp ebx, diff - jb dsublp1 -dsubgo: - mov ecx, FullLength - mov edx, ecx - sub edx, ebx // subtract alignment fix - and edx, 0x00000007 // calc bytes over mult of 8 - sub ecx, edx // drop over bytes from length - mov MMXLength, ecx - } // end _asm block - - // Now do the math for the rest of the row - switch ( bpp ) - { - case 3: - { - ActiveMask.use = 0x0000ffffff000000; - ShiftBpp.use = 24; // == 3 * 8 - ShiftRem.use = 40; // == 64 - 24 - _asm { - mov edi, row - movq mm7, ActiveMask // Load ActiveMask for 2nd active byte group - mov esi, edi // lp = row - add edi, bpp // rp = row + bpp - movq mm6, mm7 - mov ebx, diff - psllq mm6, ShiftBpp // Move mask in mm6 to cover 3rd active - // byte group - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] -dsub3lp: - psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes - // no need for mask; shift clears inactive bytes - // Add 1st active group - movq mm0, [edi+ebx] - paddb mm0, mm1 - // Add 2nd active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - pand mm1, mm7 // mask to use only 2nd active group - paddb mm0, mm1 - // Add 3rd active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - pand mm1, mm6 // mask to use only 3rd active group - add ebx, 8 - paddb mm0, mm1 - cmp ebx, MMXLength - movq [edi+ebx-8], mm0 // Write updated Raws back to array - // Prep for doing 1st add at top of loop - movq mm1, mm0 - jb dsub3lp - } // end _asm block - } - break; - - case 1: - { - // Placed here just in case this is a duplicate of the - // non-MMX code for the SUB filter in png_read_filter_row below - // - // png_bytep rp; - // png_bytep lp; - // png_uint_32 i; - // bpp = (row_info->pixel_depth + 7) >> 3; - // for (i = (png_uint_32)bpp, rp = row + bpp, lp = row; - // i < row_info->rowbytes; i++, rp++, lp++) - // { - // *rp = (png_byte)(((int)(*rp) + (int)(*lp)) & 0xff); - // } - _asm { - mov ebx, diff - mov edi, row - cmp ebx, FullLength - jnb dsub1end - mov esi, edi // lp = row - xor eax, eax - add edi, bpp // rp = row + bpp -dsub1lp: - mov al, [esi+ebx] - add [edi+ebx], al - inc ebx - cmp ebx, FullLength - jb dsub1lp -dsub1end: - } // end _asm block - } - return; - - case 6: - case 7: - case 4: - case 5: - { - ShiftBpp.use = bpp << 3; - ShiftRem.use = 64 - ShiftBpp.use; - _asm { - mov edi, row - mov ebx, diff - mov esi, edi // lp = row - add edi, bpp // rp = row + bpp - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] -dsub4lp: - psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes - // no need for mask; shift clears inactive bytes - movq mm0, [edi+ebx] - paddb mm0, mm1 - // Add 2nd active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - // there is no need for any mask - // since shift clears inactive bits/bytes - add ebx, 8 - paddb mm0, mm1 - cmp ebx, MMXLength - movq [edi+ebx-8], mm0 - movq mm1, mm0 // Prep for doing 1st add at top of loop - jb dsub4lp - } // end _asm block - } - break; - - case 2: - { - ActiveMask.use = 0x00000000ffff0000; - ShiftBpp.use = 16; // == 2 * 8 - ShiftRem.use = 48; // == 64 - 16 - _asm { - movq mm7, ActiveMask // Load ActiveMask for 2nd active byte group - mov ebx, diff - movq mm6, mm7 - mov edi, row - psllq mm6, ShiftBpp // Move mask in mm6 to cover 3rd active - // byte group - mov esi, edi // lp = row - movq mm5, mm6 - add edi, bpp // rp = row + bpp - psllq mm5, ShiftBpp // Move mask in mm5 to cover 4th active - // byte group - // PRIME the pump (load the first Raw(x-bpp) data set - movq mm1, [edi+ebx-8] -dsub2lp: - // Add 1st active group - psrlq mm1, ShiftRem // Shift data for adding 1st bpp bytes - // no need for mask; shift clears inactive - // bytes - movq mm0, [edi+ebx] - paddb mm0, mm1 - // Add 2nd active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - pand mm1, mm7 // mask to use only 2nd active group - paddb mm0, mm1 - // Add 3rd active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - pand mm1, mm6 // mask to use only 3rd active group - paddb mm0, mm1 - // Add 4th active group - movq mm1, mm0 // mov updated Raws to mm1 - psllq mm1, ShiftBpp // shift data to position correctly - pand mm1, mm5 // mask to use only 4th active group - add ebx, 8 - paddb mm0, mm1 - cmp ebx, MMXLength - movq [edi+ebx-8], mm0 // Write updated Raws back to array - movq mm1, mm0 // Prep for doing 1st add at top of loop - jb dsub2lp - } // end _asm block - } - break; - case 8: - { - _asm { - mov edi, row - mov ebx, diff - mov esi, edi // lp = row - add edi, bpp // rp = row + bpp - mov ecx, MMXLength - movq mm7, [edi+ebx-8] // PRIME the pump (load the first - // Raw(x-bpp) data set - and ecx, 0x0000003f // calc bytes over mult of 64 -dsub8lp: - movq mm0, [edi+ebx] // Load Sub(x) for 1st 8 bytes - paddb mm0, mm7 - movq mm1, [edi+ebx+8] // Load Sub(x) for 2nd 8 bytes - movq [edi+ebx], mm0 // Write Raw(x) for 1st 8 bytes - // Now mm0 will be used as Raw(x-bpp) for - // the 2nd group of 8 bytes. This will be - // repeated for each group of 8 bytes with - // the 8th group being used as the Raw(x-bpp) - // for the 1st group of the next loop. - paddb mm1, mm0 - movq mm2, [edi+ebx+16] // Load Sub(x) for 3rd 8 bytes - movq [edi+ebx+8], mm1 // Write Raw(x) for 2nd 8 bytes - paddb mm2, mm1 - movq mm3, [edi+ebx+24] // Load Sub(x) for 4th 8 bytes - movq [edi+ebx+16], mm2 // Write Raw(x) for 3rd 8 bytes - paddb mm3, mm2 - movq mm4, [edi+ebx+32] // Load Sub(x) for 5th 8 bytes - movq [edi+ebx+24], mm3 // Write Raw(x) for 4th 8 bytes - paddb mm4, mm3 - movq mm5, [edi+ebx+40] // Load Sub(x) for 6th 8 bytes - movq [edi+ebx+32], mm4 // Write Raw(x) for 5th 8 bytes - paddb mm5, mm4 - movq mm6, [edi+ebx+48] // Load Sub(x) for 7th 8 bytes - movq [edi+ebx+40], mm5 // Write Raw(x) for 6th 8 bytes - paddb mm6, mm5 - movq mm7, [edi+ebx+56] // Load Sub(x) for 8th 8 bytes - movq [edi+ebx+48], mm6 // Write Raw(x) for 7th 8 bytes - add ebx, 64 - paddb mm7, mm6 - cmp ebx, ecx - movq [edi+ebx-8], mm7 // Write Raw(x) for 8th 8 bytes - jb dsub8lp - cmp ebx, MMXLength - jnb dsub8lt8 -dsub8lpA: - movq mm0, [edi+ebx] - add ebx, 8 - paddb mm0, mm7 - cmp ebx, MMXLength - movq [edi+ebx-8], mm0 // use -8 to offset early add to ebx - movq mm7, mm0 // Move calculated Raw(x) data to mm1 to - // be the new Raw(x-bpp) for the next loop - jb dsub8lpA -dsub8lt8: - } // end _asm block - } - break; - - default: // bpp greater than 8 bytes - { - _asm { - mov ebx, diff - mov edi, row - mov esi, edi // lp = row - add edi, bpp // rp = row + bpp -dsubAlp: - movq mm0, [edi+ebx] - movq mm1, [esi+ebx] - add ebx, 8 - paddb mm0, mm1 - cmp ebx, MMXLength - movq [edi+ebx-8], mm0 // mov does not affect flags; -8 to offset - // add ebx - jb dsubAlp - } // end _asm block - } - break; - - } // end switch ( bpp ) - - _asm { - mov ebx, MMXLength - mov edi, row - cmp ebx, FullLength - jnb dsubend - mov esi, edi // lp = row - xor eax, eax - add edi, bpp // rp = row + bpp -dsublp2: - mov al, [esi+ebx] - add [edi+ebx], al - inc ebx - cmp ebx, FullLength - jb dsublp2 -dsubend: - emms // End MMX instructions; prep for possible FP instrs. - } // end _asm block -} - -// Optimized code for PNG Up filter decoder -void /* PRIVATE */ -png_read_filter_row_mmx_up(png_row_infop row_info, png_bytep row, - png_bytep prev_row) -{ - png_uint_32 len; - len = row_info->rowbytes; // # of bytes to filter - _asm { - mov edi, row - // get # of bytes to alignment - mov ecx, edi - xor ebx, ebx - add ecx, 0x7 - xor eax, eax - and ecx, 0xfffffff8 - mov esi, prev_row - sub ecx, edi - jz dupgo - // fix alignment -duplp1: - mov al, [edi+ebx] - add al, [esi+ebx] - inc ebx - cmp ebx, ecx - mov [edi + ebx-1], al // mov does not affect flags; -1 to offset inc ebx - jb duplp1 -dupgo: - mov ecx, len - mov edx, ecx - sub edx, ebx // subtract alignment fix - and edx, 0x0000003f // calc bytes over mult of 64 - sub ecx, edx // drop over bytes from length - // Unrolled loop - use all MMX registers and interleave to reduce - // number of branch instructions (loops) and reduce partial stalls -duploop: - movq mm1, [esi+ebx] - movq mm0, [edi+ebx] - movq mm3, [esi+ebx+8] - paddb mm0, mm1 - movq mm2, [edi+ebx+8] - movq [edi+ebx], mm0 - paddb mm2, mm3 - movq mm5, [esi+ebx+16] - movq [edi+ebx+8], mm2 - movq mm4, [edi+ebx+16] - movq mm7, [esi+ebx+24] - paddb mm4, mm5 - movq mm6, [edi+ebx+24] - movq [edi+ebx+16], mm4 - paddb mm6, mm7 - movq mm1, [esi+ebx+32] - movq [edi+ebx+24], mm6 - movq mm0, [edi+ebx+32] - movq mm3, [esi+ebx+40] - paddb mm0, mm1 - movq mm2, [edi+ebx+40] - movq [edi+ebx+32], mm0 - paddb mm2, mm3 - movq mm5, [esi+ebx+48] - movq [edi+ebx+40], mm2 - movq mm4, [edi+ebx+48] - movq mm7, [esi+ebx+56] - paddb mm4, mm5 - movq mm6, [edi+ebx+56] - movq [edi+ebx+48], mm4 - add ebx, 64 - paddb mm6, mm7 - cmp ebx, ecx - movq [edi+ebx-8], mm6 // (+56)movq does not affect flags; - // -8 to offset add ebx - jb duploop - - cmp edx, 0 // Test for bytes over mult of 64 - jz dupend - - - // 2 lines added by lcreeve at netins.net - // (mail 11 Jul 98 in png-implement list) - cmp edx, 8 //test for less than 8 bytes - jb duplt8 - - - add ecx, edx - and edx, 0x00000007 // calc bytes over mult of 8 - sub ecx, edx // drop over bytes from length - jz duplt8 - // Loop using MMX registers mm0 & mm1 to update 8 bytes simultaneously -duplpA: - movq mm1, [esi+ebx] - movq mm0, [edi+ebx] - add ebx, 8 - paddb mm0, mm1 - cmp ebx, ecx - movq [edi+ebx-8], mm0 // movq does not affect flags; -8 to offset add ebx - jb duplpA - cmp edx, 0 // Test for bytes over mult of 8 - jz dupend -duplt8: - xor eax, eax - add ecx, edx // move over byte count into counter - // Loop using x86 registers to update remaining bytes -duplp2: - mov al, [edi + ebx] - add al, [esi + ebx] - inc ebx - cmp ebx, ecx - mov [edi + ebx-1], al // mov does not affect flags; -1 to offset inc ebx - jb duplp2 -dupend: - // Conversion of filtered row completed - emms // End MMX instructions; prep for possible FP instrs. - } // end _asm block -} - - -// Optimized png_read_filter_row routines -void /* PRIVATE */ -png_read_filter_row(png_structp png_ptr, png_row_infop row_info, png_bytep - row, png_bytep prev_row, int filter) -{ -#ifdef PNG_DEBUG - char filnm[10]; -#endif - - if (mmx_supported == 2) { -#if !defined(PNG_1_0_X) - /* this should have happened in png_init_mmx_flags() already */ - png_warning(png_ptr, "asm_flags may not have been initialized"); -#endif - png_mmx_support(); - } - -#ifdef PNG_DEBUG - png_debug(1, "in png_read_filter_row\n"); - switch (filter) - { - case 0: sprintf(filnm, "none"); - break; -#if !defined(PNG_1_0_X) - case 1: sprintf(filnm, "sub-%s", - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_SUB)? "MMX" : "x86"); - break; - case 2: sprintf(filnm, "up-%s", - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_UP)? "MMX" : "x86"); - break; - case 3: sprintf(filnm, "avg-%s", - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_AVG)? "MMX" : "x86"); - break; - case 4: sprintf(filnm, "Paeth-%s", - (png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_PAETH)? "MMX":"x86"); - break; -#else - case 1: sprintf(filnm, "sub"); - break; - case 2: sprintf(filnm, "up"); - break; - case 3: sprintf(filnm, "avg"); - break; - case 4: sprintf(filnm, "Paeth"); - break; -#endif - default: sprintf(filnm, "unknw"); - break; - } - png_debug2(0,"row=%5d, %s, ", png_ptr->row_number, filnm); - png_debug2(0, "pd=%2d, b=%d, ", (int)row_info->pixel_depth, - (int)((row_info->pixel_depth + 7) >> 3)); - png_debug1(0,"len=%8d, ", row_info->rowbytes); -#endif /* PNG_DEBUG */ - - switch (filter) - { - case PNG_FILTER_VALUE_NONE: - break; - - case PNG_FILTER_VALUE_SUB: - { -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_SUB) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (mmx_supported) -#endif - { - png_read_filter_row_mmx_sub(row_info, row); - } - else - { - png_uint_32 i; - png_uint_32 istop = row_info->rowbytes; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_bytep rp = row + bpp; - png_bytep lp = row; - - for (i = bpp; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*lp++)) & 0xff); - rp++; - } - } - break; - } - - case PNG_FILTER_VALUE_UP: - { -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_UP) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (mmx_supported) -#endif - { - png_read_filter_row_mmx_up(row_info, row, prev_row); - } - else - { - png_uint_32 i; - png_uint_32 istop = row_info->rowbytes; - png_bytep rp = row; - png_bytep pp = prev_row; - - for (i = 0; i < istop; ++i) - { - *rp = (png_byte)(((int)(*rp) + (int)(*pp++)) & 0xff); - rp++; - } - } - break; - } - - case PNG_FILTER_VALUE_AVG: - { -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_AVG) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (mmx_supported) -#endif - { - png_read_filter_row_mmx_avg(row_info, row, prev_row); - } - else - { - png_uint_32 i; - png_bytep rp = row; - png_bytep pp = prev_row; - png_bytep lp = row; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_uint_32 istop = row_info->rowbytes - bpp; - - for (i = 0; i < bpp; i++) - { - *rp = (png_byte)(((int)(*rp) + - ((int)(*pp++) >> 1)) & 0xff); - rp++; - } - - for (i = 0; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + - ((int)(*pp++ + *lp++) >> 1)) & 0xff); - rp++; - } - } - break; - } - - case PNG_FILTER_VALUE_PAETH: - { -#if !defined(PNG_1_0_X) - if ((png_ptr->asm_flags & PNG_ASM_FLAG_MMX_READ_FILTER_PAETH) && - (row_info->pixel_depth >= png_ptr->mmx_bitdepth_threshold) && - (row_info->rowbytes >= png_ptr->mmx_rowbytes_threshold)) -#else - if (mmx_supported) -#endif - { - png_read_filter_row_mmx_paeth(row_info, row, prev_row); - } - else - { - png_uint_32 i; - png_bytep rp = row; - png_bytep pp = prev_row; - png_bytep lp = row; - png_bytep cp = prev_row; - png_uint_32 bpp = (row_info->pixel_depth + 7) >> 3; - png_uint_32 istop=row_info->rowbytes - bpp; - - for (i = 0; i < bpp; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*pp++)) & 0xff); - rp++; - } - - for (i = 0; i < istop; i++) // use leftover rp,pp - { - int a, b, c, pa, pb, pc, p; - - a = *lp++; - b = *pp++; - c = *cp++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - /* - if (pa <= pb && pa <= pc) - p = a; - else if (pb <= pc) - p = b; - else - p = c; - */ - - p = (pa <= pb && pa <=pc) ? a : (pb <= pc) ? b : c; - - *rp = (png_byte)(((int)(*rp) + p) & 0xff); - rp++; - } - } - break; - } - - default: - png_warning(png_ptr, "Ignoring bad row filter type"); - *row=0; - break; - } -} - -#endif /* PNG_ASSEMBLER_CODE_SUPPORTED && PNG_USE_PNGVCRD */ +/* pnggvrd.c was removed from libpng-1.2.20. */ diff --git a/pngwio.c b/pngwio.c index a9c1dc53d..371a4fad6 100644 --- a/pngwio.c +++ b/pngwio.c @@ -1,9 +1,9 @@ /* pngwio.c - functions for data output * - * Last changed in libpng 1.2.3 - May 21, 2002 + * Last changed in libpng 1.2.13 November 13, 2006 * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2002 Glenn Randers-Pehrson + * Copyright (c) 1998-2006 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) * @@ -45,6 +45,7 @@ png_default_write_data(png_structp png_ptr, png_bytep data, png_size_t length) { png_uint_32 check; + if(png_ptr == NULL) return; #if defined(_WIN32_WCE) if ( !WriteFile((HANDLE)(png_ptr->io_ptr), data, length, &check, NULL) ) check = 0; @@ -70,6 +71,7 @@ png_default_write_data(png_structp png_ptr, png_bytep data, png_size_t length) png_byte *near_data; /* Needs to be "png_byte *" instead of "png_bytep" */ png_FILE_p io_ptr; + if(png_ptr == NULL) return; /* Check if data really is near. If so, use usual code. */ near_data = (png_byte *)CVT_PTR_NOCHECK(data); io_ptr = (png_FILE_p)CVT_PTR(png_ptr->io_ptr); @@ -131,6 +133,9 @@ png_default_flush(png_structp png_ptr) { #if !defined(_WIN32_WCE) png_FILE_p io_ptr; +#endif + if(png_ptr == NULL) return; +#if !defined(_WIN32_WCE) io_ptr = (png_FILE_p)CVT_PTR((png_ptr->io_ptr)); if (io_ptr != NULL) fflush(io_ptr); @@ -165,6 +170,7 @@ void PNGAPI png_set_write_fn(png_structp png_ptr, png_voidp io_ptr, png_rw_ptr write_data_fn, png_flush_ptr output_flush_fn) { + if(png_ptr == NULL) return; png_ptr->io_ptr = io_ptr; #if !defined(PNG_NO_STDIO) diff --git a/pngwrite.c b/pngwrite.c index 95984f6e5..7d02ad7dd 100644 --- a/pngwrite.c +++ b/pngwrite.c @@ -1,9 +1,9 @@ /* pngwrite.c - general routines to write a PNG file * - * Last changed in libpng 1.2.9 April 14, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -112,6 +112,8 @@ png_write_info_before_PLTE(png_structp png_ptr, png_infop info_ptr) ((up->name[3] & 0x20) || keep == PNG_HANDLE_CHUNK_ALWAYS || (png_ptr->flags & PNG_FLAG_KEEP_UNSAFE_CHUNKS))) { + if (up->size == 0) + png_warning(png_ptr, "Writing zero-length unknown chunk"); png_write_chunk(png_ptr, up->name, up->data, up->size); } } @@ -391,11 +393,6 @@ png_write_end(png_structp png_ptr, png_infop info_ptr) /* write end of PNG file */ png_write_IEND(png_ptr); -#if 0 -/* This flush, added in libpng-1.0.8, causes some applications to crash - because they do not set png_ptr->output_flush_fn */ - png_flush(png_ptr); -#endif } #if defined(PNG_WRITE_tIME_SUPPORTED) @@ -459,12 +456,6 @@ png_create_write_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, if (png_ptr == NULL) return (NULL); -#if !defined(PNG_1_0_X) -#ifdef PNG_ASSEMBLER_CODE_SUPPORTED - png_init_mmx_flags(png_ptr); /* 1.2.0 addition */ -#endif -#endif /* PNG_1_0_X */ - /* added at libpng-1.2.6 */ #ifdef PNG_SET_USER_LIMITS_SUPPORTED png_ptr->user_width_max=PNG_USER_WIDTH_MAX; @@ -493,12 +484,15 @@ png_create_write_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, #endif /* PNG_USER_MEM_SUPPORTED */ png_set_error_fn(png_ptr, error_ptr, error_fn, warn_fn); - i=0; - do + if(user_png_ver) { - if(user_png_ver[i] != png_libpng_ver[i]) - png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; - } while (png_libpng_ver[i++]); + i=0; + do + { + if(user_png_ver[i] != png_libpng_ver[i]) + png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; + } while (png_libpng_ver[i++]); + } if (png_ptr->flags & PNG_FLAG_LIBRARY_MISMATCH) { @@ -515,11 +509,13 @@ png_create_write_struct_2(png_const_charp user_png_ver, png_voidp error_ptr, char msg[80]; if (user_png_ver) { - sprintf(msg, "Application was compiled with png.h from libpng-%.20s", + png_snprintf(msg, 80, + "Application was compiled with png.h from libpng-%.20s", user_png_ver); png_warning(png_ptr, msg); } - sprintf(msg, "Application is running with png.c from libpng-%.20s", + png_snprintf(msg, 80, + "Application is running with png.c from libpng-%.20s", png_libpng_ver); png_warning(png_ptr, msg); #endif @@ -576,6 +572,7 @@ png_write_init_2(png_structp png_ptr, png_const_charp user_png_ver, png_size_t png_struct_size, png_size_t png_info_size) { /* We only come here via pre-1.0.12-compiled applications */ + if(png_ptr == NULL) return; #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) if(png_sizeof(png_struct) > png_struct_size || png_sizeof(png_info) > png_info_size) @@ -584,11 +581,13 @@ png_write_init_2(png_structp png_ptr, png_const_charp user_png_ver, png_ptr->warning_fn=NULL; if (user_png_ver) { - sprintf(msg, "Application was compiled with png.h from libpng-%.20s", + png_snprintf(msg, 80, + "Application was compiled with png.h from libpng-%.20s", user_png_ver); png_warning(png_ptr, msg); } - sprintf(msg, "Application is running with png.c from libpng-%.20s", + png_snprintf(msg, 80, + "Application is running with png.c from libpng-%.20s", png_libpng_ver); png_warning(png_ptr, msg); } @@ -668,12 +667,6 @@ png_write_init_3(png_structpp ptr_ptr, png_const_charp user_png_ver, png_ptr->user_height_max=PNG_USER_HEIGHT_MAX; #endif -#if !defined(PNG_1_0_X) -#ifdef PNG_ASSEMBLER_CODE_SUPPORTED - png_init_mmx_flags(png_ptr); /* 1.2.0 addition */ -#endif -#endif /* PNG_1_0_X */ - #ifdef PNG_SETJMP_SUPPORTED /* restore jump buffer */ png_memcpy(png_ptr->jmpbuf, tmp_jmp, png_sizeof (jmp_buf)); @@ -1012,21 +1005,32 @@ png_destroy_write_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr) #endif } +#ifdef PNG_USER_MEM_SUPPORTED + if (png_ptr != NULL) + { + free_fn = png_ptr->free_fn; + mem_ptr = png_ptr->mem_ptr; + } +#endif + if (info_ptr_ptr != NULL) info_ptr = *info_ptr_ptr; if (info_ptr != NULL) { - png_free_data(png_ptr, info_ptr, PNG_FREE_ALL, -1); + if (png_ptr != NULL) + { + png_free_data(png_ptr, info_ptr, PNG_FREE_ALL, -1); #if defined(PNG_UNKNOWN_CHUNKS_SUPPORTED) - if (png_ptr->num_chunk_list) - { - png_free(png_ptr, png_ptr->chunk_list); - png_ptr->chunk_list=NULL; - png_ptr->num_chunk_list=0; - } + if (png_ptr->num_chunk_list) + { + png_free(png_ptr, png_ptr->chunk_list); + png_ptr->chunk_list=NULL; + png_ptr->num_chunk_list=0; + } #endif + } #ifdef PNG_USER_MEM_SUPPORTED png_destroy_struct_2((png_voidp)info_ptr, (png_free_ptr)free_fn, @@ -1072,11 +1076,13 @@ png_write_destroy(png_structp png_ptr) /* free our memory. png_free checks NULL for us. */ png_free(png_ptr, png_ptr->zbuf); png_free(png_ptr, png_ptr->row_buf); +#ifndef PNG_NO_WRITE_FILTERING png_free(png_ptr, png_ptr->prev_row); png_free(png_ptr, png_ptr->sub_row); png_free(png_ptr, png_ptr->up_row); png_free(png_ptr, png_ptr->avg_row); png_free(png_ptr, png_ptr->paeth_row); +#endif #if defined(PNG_TIME_RFC1123_SUPPORTED) png_free(png_ptr, png_ptr->time_buffer); @@ -1132,15 +1138,26 @@ png_set_filter(png_structp png_ptr, int method, int filters) { switch (filters & (PNG_ALL_FILTERS | 0x07)) { +#ifndef PNG_NO_WRITE_FILTER case 5: case 6: case 7: png_warning(png_ptr, "Unknown row filter for method 0"); - case PNG_FILTER_VALUE_NONE: png_ptr->do_filter=PNG_FILTER_NONE; break; - case PNG_FILTER_VALUE_SUB: png_ptr->do_filter=PNG_FILTER_SUB; break; - case PNG_FILTER_VALUE_UP: png_ptr->do_filter=PNG_FILTER_UP; break; - case PNG_FILTER_VALUE_AVG: png_ptr->do_filter=PNG_FILTER_AVG; break; - case PNG_FILTER_VALUE_PAETH: png_ptr->do_filter=PNG_FILTER_PAETH;break; +#endif /* PNG_NO_WRITE_FILTER */ + case PNG_FILTER_VALUE_NONE: + png_ptr->do_filter=PNG_FILTER_NONE; break; +#ifndef PNG_NO_WRITE_FILTER + case PNG_FILTER_VALUE_SUB: + png_ptr->do_filter=PNG_FILTER_SUB; break; + case PNG_FILTER_VALUE_UP: + png_ptr->do_filter=PNG_FILTER_UP; break; + case PNG_FILTER_VALUE_AVG: + png_ptr->do_filter=PNG_FILTER_AVG; break; + case PNG_FILTER_VALUE_PAETH: + png_ptr->do_filter=PNG_FILTER_PAETH; break; default: png_ptr->do_filter = (png_byte)filters; break; +#else + default: png_warning(png_ptr, "Unknown row filter for method 0"); +#endif /* PNG_NO_WRITE_FILTER */ } /* If we have allocated the row_buf, this means we have already started @@ -1154,6 +1171,7 @@ png_set_filter(png_structp png_ptr, int method, int filters) */ if (png_ptr->row_buf != NULL) { +#ifndef PNG_NO_WRITE_FILTER if ((png_ptr->do_filter & PNG_FILTER_SUB) && png_ptr->sub_row == NULL) { png_ptr->sub_row = (png_bytep)png_malloc(png_ptr, @@ -1208,6 +1226,7 @@ png_set_filter(png_structp png_ptr, int method, int filters) } if (png_ptr->do_filter == PNG_NO_FILTERS) +#endif /* PNG_NO_WRITE_FILTER */ png_ptr->do_filter = PNG_FILTER_NONE; } } @@ -1506,8 +1525,8 @@ png_write_png(png_structp png_ptr, png_infop info_ptr, /* It is REQUIRED to call this to finish writing the rest of the file */ png_write_end(png_ptr, info_ptr); - if(transforms == 0 || params == NULL) - /* quiet compiler warnings */ return; + transforms = transforms; /* quiet compiler warnings */ + params = params; } #endif #endif /* PNG_WRITE_SUPPORTED */ diff --git a/pngwutil.c b/pngwutil.c index 4a9c52b93..0774080b4 100644 --- a/pngwutil.c +++ b/pngwutil.c @@ -1,9 +1,9 @@ /* pngwutil.c - utilities to write a PNG file * - * Last changed in libpng 1.2.11 June 4, 2006 + * Last changed in libpng 1.2.27 [April 29, 2008] * For conditions of distribution and use, see copyright notice in png.h - * Copyright (c) 1998-2006 Glenn Randers-Pehrson + * Copyright (c) 1998-2008 Glenn Randers-Pehrson * (Version 0.96 Copyright (c) 1996, 1997 Andreas Dilger) * (Version 0.88 Copyright (c) 1995, 1996 Guy Eric Schalnat, Group 42, Inc.) */ @@ -62,6 +62,7 @@ void PNGAPI png_write_chunk(png_structp png_ptr, png_bytep chunk_name, png_bytep data, png_size_t length) { + if(png_ptr == NULL) return; png_write_chunk_start(png_ptr, chunk_name, (png_uint_32)length); png_write_chunk_data(png_ptr, data, length); png_write_chunk_end(png_ptr); @@ -77,6 +78,7 @@ png_write_chunk_start(png_structp png_ptr, png_bytep chunk_name, { png_byte buf[4]; png_debug2(0, "Writing %s chunk (%lu bytes)\n", chunk_name, length); + if(png_ptr == NULL) return; /* write the length */ png_save_uint_32(buf, length); @@ -98,6 +100,7 @@ void PNGAPI png_write_chunk_data(png_structp png_ptr, png_bytep data, png_size_t length) { /* write the data, and run the CRC over it */ + if(png_ptr == NULL) return; if (data != NULL && length > 0) { png_calculate_crc(png_ptr, data, length); @@ -111,6 +114,8 @@ png_write_chunk_end(png_structp png_ptr) { png_byte buf[4]; + if(png_ptr == NULL) return; + /* write the crc */ png_save_uint_32(buf, png_ptr->crc); @@ -177,7 +182,7 @@ png_text_compress(png_structp png_ptr, { #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) char msg[50]; - sprintf(msg, "Unknown compression type %d", compression); + png_snprintf(msg, 50, "Unknown compression type %d", compression); png_warning(png_ptr, msg); #else png_warning(png_ptr, "Unknown compression type"); @@ -375,6 +380,8 @@ png_write_IHDR(png_structp png_ptr, png_uint_32 width, png_uint_32 height, #ifdef PNG_USE_LOCAL_ARRAYS PNG_IHDR; #endif + int ret; + png_byte buf[13]; /* buffer to store the IHDR info */ png_debug(1, "in png_write_IHDR\n"); @@ -489,7 +496,7 @@ png_write_IHDR(png_structp png_ptr, png_uint_32 width, png_uint_32 height, buf[12] = (png_byte)interlace_type; /* write the chunk */ - png_write_chunk(png_ptr, (png_bytep)png_IHDR, buf, (png_size_t)13); + png_write_chunk(png_ptr, png_IHDR, buf, (png_size_t)13); /* initialize zlib with PNG info */ png_ptr->zstream.zalloc = png_zalloc; @@ -518,9 +525,19 @@ png_write_IHDR(png_structp png_ptr, png_uint_32 width, png_uint_32 height, png_ptr->zlib_window_bits = 15; if (!(png_ptr->flags & PNG_FLAG_ZLIB_CUSTOM_METHOD)) png_ptr->zlib_method = 8; - deflateInit2(&png_ptr->zstream, png_ptr->zlib_level, - png_ptr->zlib_method, png_ptr->zlib_window_bits, - png_ptr->zlib_mem_level, png_ptr->zlib_strategy); + ret = deflateInit2(&png_ptr->zstream, png_ptr->zlib_level, + png_ptr->zlib_method, png_ptr->zlib_window_bits, + png_ptr->zlib_mem_level, png_ptr->zlib_strategy); + if (ret != Z_OK) + { + if (ret == Z_VERSION_ERROR) png_error(png_ptr, + "zlib failed to initialize compressor -- version error"); + if (ret == Z_STREAM_ERROR) png_error(png_ptr, + "zlib failed to initialize compressor -- stream error"); + if (ret == Z_MEM_ERROR) png_error(png_ptr, + "zlib failed to initialize compressor -- mem error"); + png_error(png_ptr, "zlib failed to initialize compressor"); + } png_ptr->zstream.next_out = png_ptr->zbuf; png_ptr->zstream.avail_out = (uInt)png_ptr->zbuf_size; /* libpng is not interested in zstream.data_type */ @@ -572,7 +589,7 @@ png_write_PLTE(png_structp png_ptr, png_colorp palette, png_uint_32 num_pal) png_ptr->num_palette = (png_uint_16)num_pal; png_debug1(3, "num_palette = %d\n", png_ptr->num_palette); - png_write_chunk_start(png_ptr, (png_bytep)png_PLTE, num_pal * 3); + png_write_chunk_start(png_ptr, png_PLTE, num_pal * 3); #ifndef PNG_NO_POINTER_INDEXING for (i = 0, pal_ptr = palette; i < num_pal; i++, pal_ptr++) { @@ -644,7 +661,7 @@ png_write_IDAT(png_structp png_ptr, png_bytep data, png_size_t length) "Invalid zlib compression method or flags in IDAT"); } - png_write_chunk(png_ptr, (png_bytep)png_IDAT, data, length); + png_write_chunk(png_ptr, png_IDAT, data, length); png_ptr->mode |= PNG_HAVE_IDAT; } @@ -656,7 +673,7 @@ png_write_IEND(png_structp png_ptr) PNG_IEND; #endif png_debug(1, "in png_write_IEND\n"); - png_write_chunk(png_ptr, (png_bytep)png_IEND, png_bytep_NULL, + png_write_chunk(png_ptr, png_IEND, png_bytep_NULL, (png_size_t)0); png_ptr->mode |= PNG_HAVE_IEND; } @@ -677,7 +694,7 @@ png_write_gAMA(png_structp png_ptr, double file_gamma) /* file_gamma is saved in 1/100,000ths */ igamma = (png_uint_32)(file_gamma * 100000.0 + 0.5); png_save_uint_32(buf, igamma); - png_write_chunk(png_ptr, (png_bytep)png_gAMA, buf, (png_size_t)4); + png_write_chunk(png_ptr, png_gAMA, buf, (png_size_t)4); } #endif #ifdef PNG_FIXED_POINT_SUPPORTED @@ -692,7 +709,7 @@ png_write_gAMA_fixed(png_structp png_ptr, png_fixed_point file_gamma) png_debug(1, "in png_write_gAMA\n"); /* file_gamma is saved in 1/100,000ths */ png_save_uint_32(buf, (png_uint_32)file_gamma); - png_write_chunk(png_ptr, (png_bytep)png_gAMA, buf, (png_size_t)4); + png_write_chunk(png_ptr, png_gAMA, buf, (png_size_t)4); } #endif #endif @@ -712,7 +729,7 @@ png_write_sRGB(png_structp png_ptr, int srgb_intent) png_warning(png_ptr, "Invalid sRGB rendering intent specified"); buf[0]=(png_byte)srgb_intent; - png_write_chunk(png_ptr, (png_bytep)png_sRGB, buf, (png_size_t)1); + png_write_chunk(png_ptr, png_sRGB, buf, (png_size_t)1); } #endif @@ -728,6 +745,7 @@ png_write_iCCP(png_structp png_ptr, png_charp name, int compression_type, png_size_t name_len; png_charp new_name; compression_state comp; + int embedded_profile_len = 0; png_debug(1, "in png_write_iCCP\n"); @@ -750,12 +768,33 @@ png_write_iCCP(png_structp png_ptr, png_charp name, int compression_type, if (profile == NULL) profile_len = 0; + if (profile_len > 3) + embedded_profile_len = + ((*( (png_bytep)profile ))<<24) | + ((*( (png_bytep)profile+1))<<16) | + ((*( (png_bytep)profile+2))<< 8) | + ((*( (png_bytep)profile+3)) ); + + if (profile_len < embedded_profile_len) + { + png_warning(png_ptr, + "Embedded profile length too large in iCCP chunk"); + return; + } + + if (profile_len > embedded_profile_len) + { + png_warning(png_ptr, + "Truncating profile to actual length in iCCP chunk"); + profile_len = embedded_profile_len; + } + if (profile_len) profile_len = png_text_compress(png_ptr, profile, (png_size_t)profile_len, PNG_COMPRESSION_TYPE_BASE, &comp); /* make sure we include the NULL after the name and the compression type */ - png_write_chunk_start(png_ptr, (png_bytep)png_iCCP, + png_write_chunk_start(png_ptr, png_iCCP, (png_uint_32)name_len+profile_len+2); new_name[name_len+1]=0x00; png_write_chunk_data(png_ptr, (png_bytep)new_name, name_len + 2); @@ -795,7 +834,7 @@ png_write_sPLT(png_structp png_ptr, png_sPLT_tp spalette) } /* make sure we include the NULL after the name */ - png_write_chunk_start(png_ptr, (png_bytep)png_sPLT, + png_write_chunk_start(png_ptr, png_sPLT, (png_uint_32)(name_len + 2 + palette_size)); png_write_chunk_data(png_ptr, (png_bytep)new_name, name_len + 1); png_write_chunk_data(png_ptr, (png_bytep)&spalette->depth, 1); @@ -903,7 +942,7 @@ png_write_sBIT(png_structp png_ptr, png_color_8p sbit, int color_type) buf[size++] = sbit->alpha; } - png_write_chunk(png_ptr, (png_bytep)png_sBIT, buf, size); + png_write_chunk(png_ptr, png_sBIT, buf, size); } #endif @@ -967,7 +1006,7 @@ png_write_cHRM(png_structp png_ptr, double white_x, double white_y, itemp = (png_uint_32)(blue_y * 100000.0 + 0.5); png_save_uint_32(buf + 28, itemp); - png_write_chunk(png_ptr, (png_bytep)png_cHRM, buf, (png_size_t)32); + png_write_chunk(png_ptr, png_cHRM, buf, (png_size_t)32); } #endif #ifdef PNG_FIXED_POINT_SUPPORTED @@ -1019,7 +1058,7 @@ png_write_cHRM_fixed(png_structp png_ptr, png_fixed_point white_x, png_save_uint_32(buf + 24, (png_uint_32)blue_x); png_save_uint_32(buf + 28, (png_uint_32)blue_y); - png_write_chunk(png_ptr, (png_bytep)png_cHRM, buf, (png_size_t)32); + png_write_chunk(png_ptr, png_cHRM, buf, (png_size_t)32); } #endif #endif @@ -1044,7 +1083,7 @@ png_write_tRNS(png_structp png_ptr, png_bytep trans, png_color_16p tran, return; } /* write the chunk out as it is */ - png_write_chunk(png_ptr, (png_bytep)png_tRNS, trans, (png_size_t)num_trans); + png_write_chunk(png_ptr, png_tRNS, trans, (png_size_t)num_trans); } else if (color_type == PNG_COLOR_TYPE_GRAY) { @@ -1056,7 +1095,7 @@ png_write_tRNS(png_structp png_ptr, png_bytep trans, png_color_16p tran, return; } png_save_uint_16(buf, tran->gray); - png_write_chunk(png_ptr, (png_bytep)png_tRNS, buf, (png_size_t)2); + png_write_chunk(png_ptr, png_tRNS, buf, (png_size_t)2); } else if (color_type == PNG_COLOR_TYPE_RGB) { @@ -1070,7 +1109,7 @@ png_write_tRNS(png_structp png_ptr, png_bytep trans, png_color_16p tran, "Ignoring attempt to write 16-bit tRNS chunk when bit_depth is 8"); return; } - png_write_chunk(png_ptr, (png_bytep)png_tRNS, buf, (png_size_t)6); + png_write_chunk(png_ptr, png_tRNS, buf, (png_size_t)6); } else { @@ -1103,7 +1142,7 @@ png_write_bKGD(png_structp png_ptr, png_color_16p back, int color_type) return; } buf[0] = back->index; - png_write_chunk(png_ptr, (png_bytep)png_bKGD, buf, (png_size_t)1); + png_write_chunk(png_ptr, png_bKGD, buf, (png_size_t)1); } else if (color_type & PNG_COLOR_MASK_COLOR) { @@ -1116,7 +1155,7 @@ png_write_bKGD(png_structp png_ptr, png_color_16p back, int color_type) "Ignoring attempt to write 16-bit bKGD chunk when bit_depth is 8"); return; } - png_write_chunk(png_ptr, (png_bytep)png_bKGD, buf, (png_size_t)6); + png_write_chunk(png_ptr, png_bKGD, buf, (png_size_t)6); } else { @@ -1127,7 +1166,7 @@ png_write_bKGD(png_structp png_ptr, png_color_16p back, int color_type) return; } png_save_uint_16(buf, back->gray); - png_write_chunk(png_ptr, (png_bytep)png_bKGD, buf, (png_size_t)2); + png_write_chunk(png_ptr, png_bKGD, buf, (png_size_t)2); } } #endif @@ -1152,7 +1191,7 @@ png_write_hIST(png_structp png_ptr, png_uint_16p hist, int num_hist) return; } - png_write_chunk_start(png_ptr, (png_bytep)png_hIST, (png_uint_32)(num_hist * 2)); + png_write_chunk_start(png_ptr, png_hIST, (png_uint_32)(num_hist * 2)); for (i = 0; i < num_hist; i++) { png_save_uint_16(buf, hist[i]); @@ -1203,12 +1242,14 @@ png_check_keyword(png_structp png_ptr, png_charp key, png_charpp new_key) /* Replace non-printing characters with a blank and print a warning */ for (kp = key, dp = *new_key; *kp != '\0'; kp++, dp++) { - if (*kp < 0x20 || (*kp > 0x7E && (png_byte)*kp < 0xA1)) + if ((png_byte)*kp < 0x20 || + ((png_byte)*kp > 0x7E && (png_byte)*kp < 0xA1)) { #if !defined(PNG_NO_STDIO) && !defined(_WIN32_WCE) char msg[40]; - sprintf(msg, "invalid keyword character 0x%02X", *kp); + png_snprintf(msg, 40, + "invalid keyword character 0x%02X", (png_byte)*kp); png_warning(png_ptr, msg); #else png_warning(png_ptr, "invalid character in keyword"); @@ -1316,7 +1357,7 @@ png_write_tEXt(png_structp png_ptr, png_charp key, png_charp text, text_len = png_strlen(text); /* make sure we include the 0 after the key */ - png_write_chunk_start(png_ptr, (png_bytep)png_tEXt, (png_uint_32)key_len+text_len+1); + png_write_chunk_start(png_ptr, png_tEXt, (png_uint_32)key_len+text_len+1); /* * We leave it to the application to meet PNG-1.0 requirements on the * contents of the text. PNG-1.0 through PNG-1.2 discourage the use of @@ -1369,17 +1410,17 @@ png_write_zTXt(png_structp png_ptr, png_charp key, png_charp text, text_len = png_strlen(text); - png_free(png_ptr, new_key); - /* compute the compressed data; do it now for the length */ text_len = png_text_compress(png_ptr, text, text_len, compression, &comp); /* write start of chunk */ - png_write_chunk_start(png_ptr, (png_bytep)png_zTXt, (png_uint_32) + png_write_chunk_start(png_ptr, png_zTXt, (png_uint_32) (key_len+text_len+2)); /* write key */ - png_write_chunk_data(png_ptr, (png_bytep)key, key_len + 1); + png_write_chunk_data(png_ptr, (png_bytep)new_key, key_len + 1); + png_free(png_ptr, new_key); + buf[0] = (png_byte)compression; /* write compression */ png_write_chunk_data(png_ptr, (png_bytep)buf, (png_size_t)1); @@ -1442,7 +1483,7 @@ png_write_iTXt(png_structp png_ptr, int compression, png_charp key, /* make sure we include the compression flag, the compression byte, * and the NULs after the key, lang, and lang_key parts */ - png_write_chunk_start(png_ptr, (png_bytep)png_iTXt, + png_write_chunk_start(png_ptr, png_iTXt, (png_uint_32)( 5 /* comp byte, comp flag, terminators for key, lang and lang_key */ + key_len @@ -1475,8 +1516,7 @@ png_write_iTXt(png_structp png_ptr, int compression, png_charp key, png_write_chunk_end(png_ptr); png_free(png_ptr, new_key); - if (new_lang) - png_free(png_ptr, new_lang); + png_free(png_ptr, new_lang); } #endif @@ -1499,10 +1539,9 @@ png_write_oFFs(png_structp png_ptr, png_int_32 x_offset, png_int_32 y_offset, png_save_int_32(buf + 4, y_offset); buf[8] = (png_byte)unit_type; - png_write_chunk(png_ptr, (png_bytep)png_oFFs, buf, (png_size_t)9); + png_write_chunk(png_ptr, png_oFFs, buf, (png_size_t)9); } #endif - #if defined(PNG_WRITE_pCAL_SUPPORTED) /* write the pCAL chunk (described in the PNG extensions document) */ void /* PRIVATE */ @@ -1541,7 +1580,7 @@ png_write_pCAL(png_structp png_ptr, png_charp purpose, png_int_32 X0, } png_debug1(3, "pCAL total length = %d\n", (int)total_len); - png_write_chunk_start(png_ptr, (png_bytep)png_pCAL, (png_uint_32)total_len); + png_write_chunk_start(png_ptr, png_pCAL, (png_uint_32)total_len); png_write_chunk_data(png_ptr, (png_bytep)new_purpose, purpose_len); png_save_int_32(buf, X0); png_save_int_32(buf + 4, X1); @@ -1594,14 +1633,14 @@ png_write_sCAL(png_structp png_ptr, int unit, double width, double height) total_len += wc_len; } #else - sprintf(buf + 1, "%12.12e", width); + png_snprintf(buf + 1, 63, "%12.12e", width); total_len = 1 + png_strlen(buf + 1) + 1; - sprintf(buf + total_len, "%12.12e", height); + png_snprintf(buf + total_len, 64-total_len, "%12.12e", height); total_len += png_strlen(buf + total_len); #endif png_debug1(3, "sCAL total length = %u\n", (unsigned int)total_len); - png_write_chunk(png_ptr, (png_bytep)png_sCAL, (png_bytep)buf, total_len); + png_write_chunk(png_ptr, png_sCAL, (png_bytep)buf, total_len); } #else #ifdef PNG_FIXED_POINT_SUPPORTED @@ -1631,7 +1670,7 @@ png_write_sCAL_s(png_structp png_ptr, int unit, png_charp width, png_memcpy(buf + wlen + 2, height, hlen); /* do NOT append the '\0' here */ png_debug1(3, "sCAL total length = %u\n", (unsigned int)total_len); - png_write_chunk(png_ptr, (png_bytep)png_sCAL, buf, total_len); + png_write_chunk(png_ptr, png_sCAL, buf, total_len); } #endif #endif @@ -1657,7 +1696,7 @@ png_write_pHYs(png_structp png_ptr, png_uint_32 x_pixels_per_unit, png_save_uint_32(buf + 4, y_pixels_per_unit); buf[8] = (png_byte)unit_type; - png_write_chunk(png_ptr, (png_bytep)png_pHYs, buf, (png_size_t)9); + png_write_chunk(png_ptr, png_pHYs, buf, (png_size_t)9); } #endif @@ -1689,7 +1728,7 @@ png_write_tIME(png_structp png_ptr, png_timep mod_time) buf[5] = mod_time->minute; buf[6] = mod_time->second; - png_write_chunk(png_ptr, (png_bytep)png_tIME, buf, (png_size_t)7); + png_write_chunk(png_ptr, png_tIME, buf, (png_size_t)7); } #endif @@ -1697,6 +1736,7 @@ png_write_tIME(png_structp png_ptr, png_timep mod_time) void /* PRIVATE */ png_write_start_row(png_structp png_ptr) { +#ifdef PNG_WRITE_INTERLACING_SUPPORTED #ifdef PNG_USE_LOCAL_ARRAYS /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ @@ -1711,6 +1751,7 @@ png_write_start_row(png_structp png_ptr) /* offset to next interlace block in the y direction */ int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; +#endif #endif png_size_t buf_size; @@ -1723,6 +1764,7 @@ png_write_start_row(png_structp png_ptr) png_ptr->row_buf = (png_bytep)png_malloc(png_ptr, (png_uint_32)buf_size); png_ptr->row_buf[0] = PNG_FILTER_VALUE_NONE; +#ifndef PNG_NO_WRITE_FILTERING /* set up filtering buffer, if using this filter */ if (png_ptr->do_filter & PNG_FILTER_SUB) { @@ -1740,7 +1782,7 @@ png_write_start_row(png_structp png_ptr) if (png_ptr->do_filter & PNG_FILTER_UP) { - png_ptr->up_row = (png_bytep )png_malloc(png_ptr, + png_ptr->up_row = (png_bytep)png_malloc(png_ptr, (png_ptr->rowbytes + 1)); png_ptr->up_row[0] = PNG_FILTER_VALUE_UP; } @@ -1754,10 +1796,11 @@ png_write_start_row(png_structp png_ptr) if (png_ptr->do_filter & PNG_FILTER_PAETH) { - png_ptr->paeth_row = (png_bytep )png_malloc(png_ptr, + png_ptr->paeth_row = (png_bytep)png_malloc(png_ptr, (png_ptr->rowbytes + 1)); png_ptr->paeth_row[0] = PNG_FILTER_VALUE_PAETH; } +#endif /* PNG_NO_WRITE_FILTERING */ } #ifdef PNG_WRITE_INTERLACING_SUPPORTED @@ -1791,6 +1834,7 @@ png_write_start_row(png_structp png_ptr) void /* PRIVATE */ png_write_finish_row(png_structp png_ptr) { +#ifdef PNG_WRITE_INTERLACING_SUPPORTED #ifdef PNG_USE_LOCAL_ARRAYS /* arrays to facilitate easy interlacing - use pass (0 - 6) as index */ @@ -1805,6 +1849,7 @@ png_write_finish_row(png_structp png_ptr) /* offset to next interlace block in the y direction */ int png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; +#endif #endif int ret; @@ -2079,7 +2124,9 @@ png_do_write_interlace(png_row_infop row_info, png_bytep row, int pass) void /* PRIVATE */ png_write_find_filter(png_structp png_ptr, png_row_infop row_info) { - png_bytep prev_row, best_row, row_buf; + png_bytep best_row; +#ifndef PNG_NO_WRITE_FILTER + png_bytep prev_row, row_buf; png_uint_32 mins, bpp; png_byte filter_to_do = png_ptr->do_filter; png_uint_32 row_bytes = row_info->rowbytes; @@ -2092,7 +2139,10 @@ png_write_find_filter(png_structp png_ptr, png_row_infop row_info) bpp = (row_info->pixel_depth + 7) >> 3; prev_row = png_ptr->prev_row; - best_row = row_buf = png_ptr->row_buf; +#endif + best_row = png_ptr->row_buf; +#ifndef PNG_NO_WRITE_FILTER + row_buf = best_row; mins = PNG_MAXSUM; /* The prediction method we use is to find which method provides the @@ -2667,11 +2717,12 @@ png_write_find_filter(png_structp png_ptr, png_row_infop row_info) best_row = png_ptr->paeth_row; } } - +#endif /* PNG_NO_WRITE_FILTER */ /* Do the actual writing of the filtered row data from the chosen filter. */ png_write_filtered_row(png_ptr, best_row); +#ifndef PNG_NO_WRITE_FILTER #if defined(PNG_WRITE_WEIGHTED_FILTER_SUPPORTED) /* Save the type of filter we picked this time for future calculations */ if (png_ptr->num_prev_filters > 0) @@ -2684,6 +2735,7 @@ png_write_find_filter(png_structp png_ptr, png_row_infop row_info) png_ptr->prev_filters[j] = best_row[0]; } #endif +#endif /* PNG_NO_WRITE_FILTER */ } diff --git a/trees.c b/trees.c index 395e4e168..94f28bbf6 100644 --- a/trees.c +++ b/trees.c @@ -1,5 +1,6 @@ /* trees.c -- output deflated data using Huffman coding - * Copyright (C) 1995-2005 Jean-loup Gailly + * Copyright (C) 1995-2006 Jean-loup Gailly + * detect_data_type() function provided freely by Cosmin Truta, 2006 * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -152,7 +153,7 @@ local void send_all_trees OF((deflate_state *s, int lcodes, int dcodes, int blcodes)); local void compress_block OF((deflate_state *s, ct_data *ltree, ct_data *dtree)); -local void set_data_type OF((deflate_state *s)); +local int detect_data_type OF((deflate_state *s)); local unsigned bi_reverse OF((unsigned value, int length)); local void bi_windup OF((deflate_state *s)); local void bi_flush OF((deflate_state *s)); @@ -250,11 +251,13 @@ local void tr_static_init() if (static_init_done) return; /* For some embedded targets, global variables are not initialized: */ +#ifdef NO_INIT_GLOBAL_POINTERS static_l_desc.static_tree = static_ltree; static_l_desc.extra_bits = extra_lbits; static_d_desc.static_tree = static_dtree; static_d_desc.extra_bits = extra_dbits; static_bl_desc.extra_bits = extra_blbits; +#endif /* Initialize the mapping length (0..255) -> length code (0..28) */ length = 0; @@ -931,8 +934,8 @@ void _tr_flush_block(s, buf, stored_len, eof) if (s->level > 0) { /* Check if the file is binary or text */ - if (stored_len > 0 && s->strm->data_type == Z_UNKNOWN) - set_data_type(s); + if (s->strm->data_type == Z_UNKNOWN) + s->strm->data_type = detect_data_type(s); /* Construct the literal and distance trees */ build_tree(s, (tree_desc *)(&(s->l_desc))); @@ -1118,24 +1121,45 @@ local void compress_block(s, ltree, dtree) } /* =========================================================================== - * Set the data type to BINARY or TEXT, using a crude approximation: - * set it to Z_TEXT if all symbols are either printable characters (33 to 255) - * or white spaces (9 to 13, or 32); or set it to Z_BINARY otherwise. + * Check if the data type is TEXT or BINARY, using the following algorithm: + * - TEXT if the two conditions below are satisfied: + * a) There are no non-portable control characters belonging to the + * "black list" (0..6, 14..25, 28..31). + * b) There is at least one printable character belonging to the + * "white list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255). + * - BINARY otherwise. + * - The following partially-portable control characters form a + * "gray list" that is ignored in this detection algorithm: + * (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}). * IN assertion: the fields Freq of dyn_ltree are set. */ -local void set_data_type(s) +local int detect_data_type(s) deflate_state *s; { + /* black_mask is the bit mask of black-listed bytes + * set bits 0..6, 14..25, and 28..31 + * 0xf3ffc07f = binary 11110011111111111100000001111111 + */ + unsigned long black_mask = 0xf3ffc07fUL; int n; - for (n = 0; n < 9; n++) + /* Check for non-textual ("black-listed") bytes. */ + for (n = 0; n <= 31; n++, black_mask >>= 1) + if ((black_mask & 1) && (s->dyn_ltree[n].Freq != 0)) + return Z_BINARY; + + /* Check for textual ("white-listed") bytes. */ + if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0 + || s->dyn_ltree[13].Freq != 0) + return Z_TEXT; + for (n = 32; n < LITERALS; n++) if (s->dyn_ltree[n].Freq != 0) - break; - if (n == 9) - for (n = 14; n < 32; n++) - if (s->dyn_ltree[n].Freq != 0) - break; - s->strm->data_type = (n == 32) ? Z_TEXT : Z_BINARY; + return Z_TEXT; + + /* There are no "black-listed" or "white-listed" bytes: + * this stream either is empty or has tolerated ("gray-listed") bytes only. + */ + return Z_BINARY; } /* =========================================================================== diff --git a/zconf.h b/zconf.h index 03a9431c8..0a9f5a4cd 100644 --- a/zconf.h +++ b/zconf.h @@ -1,5 +1,5 @@ /* zconf.h -- configuration of the zlib compression library - * Copyright (C) 1995-2005 Jean-loup Gailly. + * Copyright (C) 1995-2006 Jean-loup Gailly. * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -11,52 +11,115 @@ /* * If you *really* need a unique prefix for all types and library functions, * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it. + * Even better than compiling with -DZ_PREFIX would be to use configure to set + * this permanently in zconf.h using "./configure --zprefix". */ -#ifdef Z_PREFIX -# define deflateInit_ z_deflateInit_ -# define deflate z_deflate -# define deflateEnd z_deflateEnd -# define inflateInit_ z_inflateInit_ -# define inflate z_inflate -# define inflateEnd z_inflateEnd -# define deflateInit2_ z_deflateInit2_ -# define deflateSetDictionary z_deflateSetDictionary -# define deflateCopy z_deflateCopy -# define deflateReset z_deflateReset -# define deflateParams z_deflateParams -# define deflateBound z_deflateBound -# define deflatePrime z_deflatePrime -# define inflateInit2_ z_inflateInit2_ -# define inflateSetDictionary z_inflateSetDictionary -# define inflateSync z_inflateSync -# define inflateSyncPoint z_inflateSyncPoint -# define inflateCopy z_inflateCopy -# define inflateReset z_inflateReset -# define inflateBack z_inflateBack -# define inflateBackEnd z_inflateBackEnd +#ifdef Z_PREFIX /* may be set to #if 1 by ./configure */ + +/* all linked symbols */ +# define _dist_code z__dist_code +# define _length_code z__length_code +# define _tr_align z__tr_align +# define _tr_flush_block z__tr_flush_block +# define _tr_init z__tr_init +# define _tr_stored_block z__tr_stored_block +# define _tr_tally z__tr_tally +# define adler32 z_adler32 +# define adler32_combine z_adler32_combine # define compress z_compress # define compress2 z_compress2 # define compressBound z_compressBound -# define uncompress z_uncompress -# define adler32 z_adler32 # define crc32 z_crc32 +# define crc32_combine z_crc32_combine +# define deflate z_deflate +# define deflateBound z_deflateBound +# define deflateCopy z_deflateCopy +# define deflateEnd z_deflateEnd +# define deflateInit2_ z_deflateInit2_ +# define deflateInit_ z_deflateInit_ +# define deflateParams z_deflateParams +# define deflatePrime z_deflatePrime +# define deflateReset z_deflateReset +# define deflateSetDictionary z_deflateSetDictionary +# define deflateSetHeader z_deflateSetHeader +# define deflateTune z_deflateTune +# define deflate_copyright z_deflate_copyright # define get_crc_table z_get_crc_table +# define gzclearerr z_gzclearerr +# define gzclose z_gzclose +# define gzdirect z_gzdirect +# define gzdopen z_gzdopen +# define gzeof z_gzeof +# define gzerror z_gzerror +# define gzflush z_gzflush +# define gzgetc z_gzgetc +# define gzgets z_gzgets +# define gzopen z_gzopen +# define gzprintf z_gzprintf +# define gzputc z_gzputc +# define gzputs z_gzputs +# define gzread z_gzread +# define gzrewind z_gzrewind +# define gzseek z_gzseek +# define gzsetparams z_gzsetparams +# define gztell z_gztell +# define gzungetc z_gzungetc +# define gzwrite z_gzwrite +# define inflate z_inflate +# define inflateBack z_inflateBack +# define inflateBackEnd z_inflateBackEnd +# define inflateBackInit_ z_inflateBackInit_ +# define inflateCopy z_inflateCopy +# define inflateEnd z_inflateEnd +# define inflateGetHeader z_inflateGetHeader +# define inflateInit2_ z_inflateInit2_ +# define inflateInit_ z_inflateInit_ +# define inflatePrime z_inflatePrime +# define inflateReset z_inflateReset +# define inflateSetDictionary z_inflateSetDictionary +# define inflateSync z_inflateSync +# define inflateSyncPoint z_inflateSyncPoint +# define inflate_copyright z_inflate_copyright +# define inflate_fast z_inflate_fast +# define inflate_table z_inflate_table +# define uncompress z_uncompress # define zError z_zError +# define z_errmsg z_z_errmsg +# define zcalloc z_zcalloc +# define zcfree z_zcfree +# define zlibCompileFlags z_zlibCompileFlags +# define zlibVersion z_zlibVersion -# define alloc_func z_alloc_func -# define free_func z_free_func -# define in_func z_in_func -# define out_func z_out_func +/* all zlib typedefs in zlib.h and zconf.h */ # define Byte z_Byte -# define uInt z_uInt -# define uLong z_uLong # define Bytef z_Bytef +# define alloc_func z_alloc_func # define charf z_charf +# define free_func z_free_func +# define gzFile z_gzFile +# define gz_header z_gz_header +# define gz_headerp z_gz_headerp +# define in_func z_in_func # define intf z_intf +# define out_func z_out_func +# define uInt z_uInt # define uIntf z_uIntf +# define uLong z_uLong # define uLongf z_uLongf -# define voidpf z_voidpf # define voidp z_voidp +# define voidp z_voidp +# define voidpc z_voidpc +# define voidpc z_voidpc +# define voidpf z_voidpf +# define voidpf z_voidpf +# define z_stream z_z_stream +# define z_streamp z_z_streamp + +/* all zlib structs in zlib.h and zconf.h */ +# define gz_header_s z_gz_header_s +# define internal_state z_internal_state +# define z_stream_s z_z_stream_s + #endif #if defined(__MSDOS__) && !defined(MSDOS) @@ -243,6 +306,10 @@ # endif #endif +#ifdef HAVE_VISIBILITY_PRAGMA +# define ZEXTERN __attribute__((visibility ("default"))) extern +#endif + #ifndef ZEXTERN # define ZEXTERN extern #endif @@ -284,14 +351,8 @@ typedef uLong FAR uLongf; typedef Byte *voidp; #endif -#if 0 /* HAVE_UNISTD_H -- this line is updated by ./configure */ -# include /* for off_t */ -# include /* for SEEK_* and off_t */ -# ifdef VMS -# include /* for off_t */ -# endif -# define z_off_t off_t -#endif +#include "zlibdefs.h" /* created by configure */ + #ifndef SEEK_SET # define SEEK_SET 0 /* Seek from beginning of file. */ # define SEEK_CUR 1 /* Seek from current position. */ @@ -307,9 +368,6 @@ typedef uLong FAR uLongf; #if defined(__MVS__) # define NO_vsnprintf -# ifdef FAR -# undef FAR -# endif #endif /* MVS linker does not support external names larger than 8 bytes */ diff --git a/zlib.h b/zlib.h index 022817927..500b7d6c1 100644 --- a/zlib.h +++ b/zlib.h @@ -1,7 +1,7 @@ /* zlib.h -- interface of the 'zlib' general purpose compression library - version 1.2.3, July 18th, 2005 + version 1.2.3.3, October 2nd, 2006 - Copyright (C) 1995-2005 Jean-loup Gailly and Mark Adler + Copyright (C) 1995-2006 Jean-loup Gailly and Mark Adler This software is provided 'as-is', without any express or implied warranty. In no event will the authors be held liable for any damages @@ -37,8 +37,11 @@ extern "C" { #endif -#define ZLIB_VERSION "1.2.3" -#define ZLIB_VERNUM 0x1230 +#define ZLIB_VERSION "1.2.3.3" +#define ZLIB_VERNUM 0x1233 +#define ZLIB_VER_MAJOR 1 +#define ZLIB_VER_MINOR 2 +#define ZLIB_VER_REVISION 3 /* The 'zlib' compression library provides in-memory compression and @@ -157,6 +160,8 @@ typedef gz_header FAR *gz_headerp; a single step). */ +#define INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR + /* constants */ #define Z_NO_FLUSH 0 @@ -353,10 +358,14 @@ ZEXTERN int ZEXPORT inflateInit OF((z_streamp strm)); inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough memory, Z_VERSION_ERROR if the zlib library version is incompatible with the - version assumed by the caller. msg is set to null if there is no error - message. inflateInit does not perform any decompression apart from reading - the zlib header if present: this will be done by inflate(). (So next_in and - avail_in may be modified, but next_out and avail_out are unchanged.) + version assumed by the caller, or Z_STREAM_ERROR if the parameters are + invalid, such as a null pointer to the structure. msg is set to null if + there is no error message. inflateInit does not perform any decompression + apart from possibly reading the zlib header if present: actual decompression + will be done by inflate(). (So next_in and avail_in may be modified, but + next_out and avail_out are unused and unchanged.) The current + implementation of inflateInit() does not process any header information -- + that is deferred until inflate() is called. */ @@ -645,9 +654,10 @@ ZEXTERN uLong ZEXPORT deflateBound OF((z_streamp strm, uLong sourceLen)); /* deflateBound() returns an upper bound on the compressed size after - deflation of sourceLen bytes. It must be called after deflateInit() - or deflateInit2(). This would be used to allocate an output buffer - for deflation in a single pass, and so would be called before deflate(). + deflation of sourceLen bytes. It must be called after deflateInit() or + deflateInit2(), and after deflateSetHeader(), if used. This would be used + to allocate an output buffer for deflation in a single pass, and so would be + called before deflate(). */ ZEXTERN int ZEXPORT deflatePrime OF((z_streamp strm, @@ -726,11 +736,15 @@ ZEXTERN int ZEXPORT inflateInit2 OF((z_streamp strm, a crc32 instead of an adler32. inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_STREAM_ERROR if a parameter is invalid (such as a null strm). msg - is set to null if there is no error message. inflateInit2 does not perform - any decompression apart from reading the zlib header if present: this will - be done by inflate(). (So next_in and avail_in may be modified, but next_out - and avail_out are unchanged.) + memory, Z_VERSION_ERROR if the zlib library version is incompatible with the + version assumed by the caller, or Z_STREAM_ERROR if the parameters are + invalid, such as a null pointer to the structure. msg is set to null if + there is no error message. inflateInit2 does not perform any decompression + apart from possibly reading the zlib header if present: actual decompression + will be done by inflate(). (So next_in and avail_in may be modified, but + next_out and avail_out are unused and unchanged.) The current + implementation of inflateInit2() does not process any header information -- + that is deferred until inflate() is called. */ ZEXTERN int ZEXPORT inflateSetDictionary OF((z_streamp strm, @@ -1065,8 +1079,9 @@ ZEXTERN int ZEXPORT uncompress OF((Bytef *dest, uLongf *destLen, typedef voidp gzFile; -ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode)); /* +ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode)); + Opens a gzip (.gz) file for reading or writing. The mode parameter is as in fopen ("rb" or "wb") but can also include a compression level ("wb9") or a strategy: 'f' for filtered data as in "wb6f", 'h' for @@ -1180,9 +1195,10 @@ ZEXTERN int ZEXPORT gzflush OF((gzFile file, int flush)); degrade compression. */ +/* ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile file, z_off_t offset, int whence)); -/* + Sets the starting position for the next gzread or gzwrite on the given compressed file. The offset represents a number of bytes in the uncompressed data stream. The whence parameter is defined as in lseek(2); @@ -1205,8 +1221,9 @@ ZEXTERN int ZEXPORT gzrewind OF((gzFile file)); gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET) */ -ZEXTERN z_off_t ZEXPORT gztell OF((gzFile file)); /* +ZEXTERN z_off_t ZEXPORT gztell OF((gzFile file)); + Returns the starting position for the next gzread or gzwrite on the given compressed file. This position represents a number of bytes in the uncompressed data stream. @@ -1230,7 +1247,8 @@ ZEXTERN int ZEXPORT gzclose OF((gzFile file)); /* Flushes all pending output if necessary, closes the compressed file and deallocates all the (de)compression state. The return value is the zlib - error number (see function gzerror below). + error number. Note that once file is close, you cannot call gzerror with + file, since its structures have been deallocated. */ ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum)); @@ -1240,6 +1258,9 @@ ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum)); error occurred in the file system and not in the compression library, errnum is set to Z_ERRNO and the application may consult errno to get the exact error code. + + The application must not modify the returned string and future calls to + this function may invalidate the returned string. */ ZEXTERN void ZEXPORT gzclearerr OF((gzFile file)); @@ -1273,9 +1294,10 @@ ZEXTERN uLong ZEXPORT adler32 OF((uLong adler, const Bytef *buf, uInt len)); if (adler != original_adler) error(); */ +/* ZEXTERN uLong ZEXPORT adler32_combine OF((uLong adler1, uLong adler2, z_off_t len2)); -/* + Combine two Adler-32 checksums into one. For two sequences of bytes, seq1 and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for each, adler1 and adler2. adler32_combine() returns the Adler-32 checksum of @@ -1298,9 +1320,9 @@ ZEXTERN uLong ZEXPORT crc32 OF((uLong crc, const Bytef *buf, uInt len)); if (crc != original_crc) error(); */ +/* ZEXTERN uLong ZEXPORT crc32_combine OF((uLong crc1, uLong crc2, z_off_t len2)); -/* Combine two CRC-32 check values into one. For two sequences of bytes, seq1 and seq2 with lengths len1 and len2, CRC-32 check values were calculated for each, crc1 and crc2. crc32_combine() returns the CRC-32 @@ -1339,16 +1361,38 @@ ZEXTERN int ZEXPORT inflateBackInit_ OF((z_streamp strm, int windowBits, inflateInit2_((strm), (windowBits), ZLIB_VERSION, sizeof(z_stream)) #define inflateBackInit(strm, windowBits, window) \ inflateBackInit_((strm), (windowBits), (window), \ - ZLIB_VERSION, sizeof(z_stream)) + ZLIB_VERSION, sizeof(z_stream)) +#ifdef _LARGEFILE64_SOURCE + ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *)); + ZEXTERN off64_t ZEXPORT gzseek64 OF((gzFile, off64_t, int)); + ZEXTERN off64_t ZEXPORT gztell64 OF((gzFile)); + ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, off64_t)); + ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, off64_t)); +#endif + +#if _FILE_OFFSET_BITS == 64 +# define gzopen gzopen64 +# define gzseek gzseek64 +# define gztell gztell64 +# define adler32_combine adler32_combine64 +# define crc32_combine crc32_combine64 +#else + ZEXTERN gzFile ZEXPORT gzopen OF((const char *, const char *)); + ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile, z_off_t, int)); + ZEXTERN z_off_t ZEXPORT gztell OF((gzFile)); + ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t)); + ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t)); +#endif #if !defined(ZUTIL_H) && !defined(NO_DUMMY_DECL) struct internal_state {int dummy;}; /* hack for buggy compilers */ #endif ZEXTERN const char * ZEXPORT zError OF((int)); -ZEXTERN int ZEXPORT inflateSyncPoint OF((z_streamp z)); +ZEXTERN int ZEXPORT inflateSyncPoint OF((z_streamp)); ZEXTERN const uLongf * ZEXPORT get_crc_table OF((void)); +ZEXTERN int ZEXPORT inflateUndermine OF((z_streamp, int)); #ifdef __cplusplus } diff --git a/zlibdefs.h b/zlibdefs.h new file mode 100644 index 000000000..80d6e8171 --- /dev/null +++ b/zlibdefs.h @@ -0,0 +1,13 @@ +/* zlibdefs.h -- compile-time definitions for the zlib compression library + * Copyright (C) 1995-2006 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include /* for off_t */ +#include /* for SEEK_* and off_t */ +#ifdef VMS +# include /* for off_t */ +#endif +#ifndef z_off_t +# define z_off_t off_t +#endif diff --git a/zutil.h b/zutil.h index b7d5eff81..32e9c890d 100644 --- a/zutil.h +++ b/zutil.h @@ -1,5 +1,5 @@ /* zutil.h -- internal interface and configuration of the compression library - * Copyright (C) 1995-2005 Jean-loup Gailly. + * Copyright (C) 1995-2006 Jean-loup Gailly. * For conditions of distribution and use, see copyright notice in zlib.h */ @@ -23,7 +23,7 @@ # include # include #endif -#ifdef NO_ERRNO_H +#if defined(NO_ERRNO_H) || defined(_WIN32_WCE) # ifdef _WIN32_WCE /* The Microsoft C Run-Time Library for Windows CE doesn't have * errno. We define it as a global variable to simplify porting. @@ -34,9 +34,7 @@ # endif extern int errno; #else -# ifndef _WIN32_WCE -# include -# endif +# include #endif #ifndef local @@ -151,7 +149,7 @@ extern const char * const z_errmsg[10]; /* indexed by 2-zlib_error */ # define fdopen(fd,mode) NULL /* No fdopen() */ #endif -#if (defined(_MSC_VER) && (_MSC_VER > 600)) +#if (defined(_MSC_VER) && (_MSC_VER > 600)) && !defined __INTERIX # if defined(_WIN32_WCE) # define fdopen(fd,mode) NULL /* No fdopen() */ # ifndef _PTRDIFF_T_DEFINED @@ -161,6 +159,12 @@ extern const char * const z_errmsg[10]; /* indexed by 2-zlib_error */ # else # define fdopen(fd,type) _fdopen(fd,type) # endif +#endif + +#if defined(__BORLANDC__) +#pragma warn -8004 +#pragma warn -8008 +#pragma warn -8066 #endif /* common defaults */ @@ -171,6 +175,12 @@ extern const char * const z_errmsg[10]; /* indexed by 2-zlib_error */ #ifndef F_OPEN # define F_OPEN(name, mode) fopen((name), (mode)) +#endif + +#ifdef _LARGEFILE64_SOURCE +# define F_OPEN64(name, mode) fopen64((name), (mode)) +#else +# define F_OPEN64(name, mode) fopen((name), (mode)) #endif /* functions */