lib: add Zstandard decompression support

Add the zstd library from Linux kernel (only decompression support).
There are minimal changes to build with U-Boot, otherwise the files are
identical to Linux commit dc35da16 from March 2018, the files had not
been touched since in kernel. Also SPDX lincese tags were added.
Signed-off-by: Marek Behún <marek.behun@nic.cz>

lib/Kconfig

@@ -374,6 +374,12 @@ config ZLIB
 	help
 	  This enables ZLIB compression lib.
 
+config ZSTD
+	bool "Enable Zstandard decompression support"
+	select XXHASH
+	help
+	  This enables Zstandard decompression library.
+
 config SPL_LZ4
 	bool "Enable LZ4 decompression support in SPL"
 	help
@@ -398,6 +404,12 @@ config SPL_ZLIB
 	help
 	  This enables compression lib for SPL boot.
 
+config SPL_ZSTD
+	bool "Enable Zstandard decompression support in SPL"
+	select XXHASH
+	help
+	  This enables Zstandard decompression library in the SPL.
+
 endmenu
 
 config ERRNO_STR

lib/Makefile

@@ -59,6 +59,7 @@ obj-$(CONFIG_SHA1) += sha1.o
 obj-$(CONFIG_SHA256) += sha256.o
 
 obj-$(CONFIG_$(SPL_)ZLIB) += zlib/
+obj-$(CONFIG_$(SPL_)ZSTD) += zstd/
 obj-$(CONFIG_$(SPL_)GZIP) += gunzip.o
 obj-$(CONFIG_$(SPL_)LZO) += lzo/
 obj-$(CONFIG_$(SPL_)LZ4) += lz4_wrapper.o

lib/zstd/Makefile

+obj-y += zstd_decompress.o
+
+zstd_decompress-y := huf_decompress.o decompress.o \
+		     entropy_common.o fse_decompress.o zstd_common.o

lib/zstd/bitstream.h

+/* SPDX-License-Identifier: (GPL-2.0 or BSD-2-Clause) */
+/*
+ * bitstream
+ * Part of FSE library
+ * header file (to include)
+ * Copyright (C) 2013-2016, Yann Collet.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ */
+#ifndef BITSTREAM_H_MODULE
+#define BITSTREAM_H_MODULE
+
+/*
+*  This API consists of small unitary functions, which must be inlined for best performance.
+*  Since link-time-optimization is not available for all compilers,
+*  these functions are defined into a .h to be included.
+*/
+
+/*-****************************************
+*  Dependencies
+******************************************/
+#include "error_private.h" /* error codes and messages */
+#include "mem.h"	   /* unaligned access routines */
+
+/*=========================================
+*  Target specific
+=========================================*/
+#define STREAM_ACCUMULATOR_MIN_32 25
+#define STREAM_ACCUMULATOR_MIN_64 57
+#define STREAM_ACCUMULATOR_MIN ((U32)(ZSTD_32bits() ? STREAM_ACCUMULATOR_MIN_32 : STREAM_ACCUMULATOR_MIN_64))
+
+/*-******************************************
+*  bitStream encoding API (write forward)
+********************************************/
+/* bitStream can mix input from multiple sources.
+*  A critical property of these streams is that they encode and decode in **reverse** direction.
+*  So the first bit sequence you add will be the last to be read, like a LIFO stack.
+*/
+typedef struct {
+	size_t bitContainer;
+	int bitPos;
+	char *startPtr;
+	char *ptr;
+	char *endPtr;
+} BIT_CStream_t;
+
+ZSTD_STATIC size_t BIT_initCStream(BIT_CStream_t *bitC, void *dstBuffer, size_t dstCapacity);
+ZSTD_STATIC void BIT_addBits(BIT_CStream_t *bitC, size_t value, unsigned nbBits);
+ZSTD_STATIC void BIT_flushBits(BIT_CStream_t *bitC);
+ZSTD_STATIC size_t BIT_closeCStream(BIT_CStream_t *bitC);
+
+/* Start with initCStream, providing the size of buffer to write into.
+*  bitStream will never write outside of this buffer.
+*  `dstCapacity` must be >= sizeof(bitD->bitContainer), otherwise @return will be an error code.
+*
+*  bits are first added to a local register.
+*  Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems.
+*  Writing data into memory is an explicit operation, performed by the flushBits function.
+*  Hence keep track how many bits are potentially stored into local register to avoid register overflow.
+*  After a flushBits, a maximum of 7 bits might still be stored into local register.
+*
+*  Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers.
+*
+*  Last operation is to close the bitStream.
+*  The function returns the final size of CStream in bytes.
+*  If data couldn't fit into `dstBuffer`, it will return a 0 ( == not storable)
+*/
+
+/*-********************************************
+*  bitStream decoding API (read backward)
+**********************************************/
+typedef struct {
+	size_t bitContainer;
+	unsigned bitsConsumed;
+	const char *ptr;
+	const char *start;
+} BIT_DStream_t;
+
+typedef enum {
+	BIT_DStream_unfinished = 0,
+	BIT_DStream_endOfBuffer = 1,
+	BIT_DStream_completed = 2,
+	BIT_DStream_overflow = 3
+} BIT_DStream_status; /* result of BIT_reloadDStream() */
+/* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */
+
+ZSTD_STATIC size_t BIT_initDStream(BIT_DStream_t *bitD, const void *srcBuffer, size_t srcSize);
+ZSTD_STATIC size_t BIT_readBits(BIT_DStream_t *bitD, unsigned nbBits);
+ZSTD_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t *bitD);
+ZSTD_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t *bitD);
+
+/* Start by invoking BIT_initDStream().
+*  A chunk of the bitStream is then stored into a local register.
+*  Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t).
+*  You can then retrieve bitFields stored into the local register, **in reverse order**.
+*  Local register is explicitly reloaded from memory by the BIT_reloadDStream() method.
+*  A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished.
+*  Otherwise, it can be less than that, so proceed accordingly.
+*  Checking if DStream has reached its end can be performed with BIT_endOfDStream().
+*/
+
+/*-****************************************
+*  unsafe API
+******************************************/
+ZSTD_STATIC void BIT_addBitsFast(BIT_CStream_t *bitC, size_t value, unsigned nbBits);
+/* faster, but works only if value is "clean", meaning all high bits above nbBits are 0 */
+
+ZSTD_STATIC void BIT_flushBitsFast(BIT_CStream_t *bitC);
+/* unsafe version; does not check buffer overflow */
+
+ZSTD_STATIC size_t BIT_readBitsFast(BIT_DStream_t *bitD, unsigned nbBits);
+/* faster, but works only if nbBits >= 1 */
+
+/*-**************************************************************
+*  Internal functions
+****************************************************************/
+ZSTD_STATIC unsigned BIT_highbit32(register U32 val) { return 31 - __builtin_clz(val); }
+
+/*=====    Local Constants   =====*/
+static const unsigned BIT_mask[] = {0,       1,       3,       7,	0xF,      0x1F,     0x3F,     0x7F,      0xFF,
+				    0x1FF,   0x3FF,   0x7FF,   0xFFF,    0x1FFF,   0x3FFF,   0x7FFF,   0xFFFF,    0x1FFFF,
+				    0x3FFFF, 0x7FFFF, 0xFFFFF, 0x1FFFFF, 0x3FFFFF, 0x7FFFFF, 0xFFFFFF, 0x1FFFFFF, 0x3FFFFFF}; /* up to 26 bits */
+
+/*-**************************************************************
+*  bitStream encoding
+****************************************************************/
+/*! BIT_initCStream() :
+ *  `dstCapacity` must be > sizeof(void*)
+ *  @return : 0 if success,
+			  otherwise an error code (can be tested using ERR_isError() ) */
+ZSTD_STATIC size_t BIT_initCStream(BIT_CStream_t *bitC, void *startPtr, size_t dstCapacity)
+{
+	bitC->bitContainer = 0;
+	bitC->bitPos = 0;
+	bitC->startPtr = (char *)startPtr;
+	bitC->ptr = bitC->startPtr;
+	bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->ptr);
+	if (dstCapacity <= sizeof(bitC->ptr))
+		return ERROR(dstSize_tooSmall);
+	return 0;
+}
+
+/*! BIT_addBits() :
+	can add up to 26 bits into `bitC`.
+	Does not check for register overflow ! */
+ZSTD_STATIC void BIT_addBits(BIT_CStream_t *bitC, size_t value, unsigned nbBits)
+{
+	bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos;
+	bitC->bitPos += nbBits;
+}
+
+/*! BIT_addBitsFast() :
+ *  works only if `value` is _clean_, meaning all high bits above nbBits are 0 */
+ZSTD_STATIC void BIT_addBitsFast(BIT_CStream_t *bitC, size_t value, unsigned nbBits)
+{
+	bitC->bitContainer |= value << bitC->bitPos;
+	bitC->bitPos += nbBits;
+}
+
+/*! BIT_flushBitsFast() :
+ *  unsafe version; does not check buffer overflow */
+ZSTD_STATIC void BIT_flushBitsFast(BIT_CStream_t *bitC)
+{
+	size_t const nbBytes = bitC->bitPos >> 3;
+	ZSTD_writeLEST(bitC->ptr, bitC->bitContainer);
+	bitC->ptr += nbBytes;
+	bitC->bitPos &= 7;
+	bitC->bitContainer >>= nbBytes * 8; /* if bitPos >= sizeof(bitContainer)*8 --> undefined behavior */
+}
+
+/*! BIT_flushBits() :
+ *  safe version; check for buffer overflow, and prevents it.
+ *  note : does not signal buffer overflow. This will be revealed later on using BIT_closeCStream() */
+ZSTD_STATIC void BIT_flushBits(BIT_CStream_t *bitC)
+{
+	size_t const nbBytes = bitC->bitPos >> 3;
+	ZSTD_writeLEST(bitC->ptr, bitC->bitContainer);
+	bitC->ptr += nbBytes;
+	if (bitC->ptr > bitC->endPtr)
+		bitC->ptr = bitC->endPtr;
+	bitC->bitPos &= 7;
+	bitC->bitContainer >>= nbBytes * 8; /* if bitPos >= sizeof(bitContainer)*8 --> undefined behavior */
+}
+
+/*! BIT_closeCStream() :
+ *  @return : size of CStream, in bytes,
+			  or 0 if it could not fit into dstBuffer */
+ZSTD_STATIC size_t BIT_closeCStream(BIT_CStream_t *bitC)
+{
+	BIT_addBitsFast(bitC, 1, 1); /* endMark */
+	BIT_flushBits(bitC);
+
+	if (bitC->ptr >= bitC->endPtr)
+		return 0; /* doesn't fit within authorized budget : cancel */
+
+	return (bitC->ptr - bitC->startPtr) + (bitC->bitPos > 0);
+}
+
+/*-********************************************************
+* bitStream decoding
+**********************************************************/
+/*! BIT_initDStream() :
+*   Initialize a BIT_DStream_t.
+*   `bitD` : a pointer to an already allocated BIT_DStream_t structure.
+*   `srcSize` must be the *exact* size of the bitStream, in bytes.
+*   @return : size of stream (== srcSize) or an errorCode if a problem is detected
+*/
+ZSTD_STATIC size_t BIT_initDStream(BIT_DStream_t *bitD, const void *srcBuffer, size_t srcSize)
+{
+	if (srcSize < 1) {
+		memset(bitD, 0, sizeof(*bitD));
+		return ERROR(srcSize_wrong);
+	}
+
+	if (srcSize >= sizeof(bitD->bitContainer)) { /* normal case */
+		bitD->start = (const char *)srcBuffer;
+		bitD->ptr = (const char *)srcBuffer + srcSize - sizeof(bitD->bitContainer);
+		bitD->bitContainer = ZSTD_readLEST(bitD->ptr);
+		{
+			BYTE const lastByte = ((const BYTE *)srcBuffer)[srcSize - 1];
+			bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */
+			if (lastByte == 0)
+				return ERROR(GENERIC); /* endMark not present */
+		}
+	} else {
+		bitD->start = (const char *)srcBuffer;
+		bitD->ptr = bitD->start;
+		bitD->bitContainer = *(const BYTE *)(bitD->start);
+		switch (srcSize) {
+		case 7: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[6]) << (sizeof(bitD->bitContainer) * 8 - 16);
+		case 6: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[5]) << (sizeof(bitD->bitContainer) * 8 - 24);
+		case 5: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[4]) << (sizeof(bitD->bitContainer) * 8 - 32);
+		case 4: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[3]) << 24;
+		case 3: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[2]) << 16;
+		case 2: bitD->bitContainer += (size_t)(((const BYTE *)(srcBuffer))[1]) << 8;
+		default:;
+		}
+		{
+			BYTE const lastByte = ((const BYTE *)srcBuffer)[srcSize - 1];
+			bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0;
+			if (lastByte == 0)
+				return ERROR(GENERIC); /* endMark not present */
+		}
+		bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize) * 8;
+	}
+
+	return srcSize;
+}
+
+ZSTD_STATIC size_t BIT_getUpperBits(size_t bitContainer, U32 const start) { return bitContainer >> start; }
+
+ZSTD_STATIC size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits) { return (bitContainer >> start) & BIT_mask[nbBits]; }
+
+ZSTD_STATIC size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits) { return bitContainer & BIT_mask[nbBits]; }
+
+/*! BIT_lookBits() :
+ *  Provides next n bits from local register.
+ *  local register is not modified.
+ *  On 32-bits, maxNbBits==24.
+ *  On 64-bits, maxNbBits==56.
+ *  @return : value extracted
+ */
+ZSTD_STATIC size_t BIT_lookBits(const BIT_DStream_t *bitD, U32 nbBits)
+{
+	U32 const bitMask = sizeof(bitD->bitContainer) * 8 - 1;
+	return ((bitD->bitContainer << (bitD->bitsConsumed & bitMask)) >> 1) >> ((bitMask - nbBits) & bitMask);
+}
+
+/*! BIT_lookBitsFast() :
+*   unsafe version; only works only if nbBits >= 1 */
+ZSTD_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t *bitD, U32 nbBits)
+{
+	U32 const bitMask = sizeof(bitD->bitContainer) * 8 - 1;
+	return (bitD->bitContainer << (bitD->bitsConsumed & bitMask)) >> (((bitMask + 1) - nbBits) & bitMask);
+}
+
+ZSTD_STATIC void BIT_skipBits(BIT_DStream_t *bitD, U32 nbBits) { bitD->bitsConsumed += nbBits; }
+
+/*! BIT_readBits() :
+ *  Read (consume) next n bits from local register and update.
+ *  Pay attention to not read more than nbBits contained into local register.
+ *  @return : extracted value.
+ */
+ZSTD_STATIC size_t BIT_readBits(BIT_DStream_t *bitD, U32 nbBits)
+{
+	size_t const value = BIT_lookBits(bitD, nbBits);
+	BIT_skipBits(bitD, nbBits);
+	return value;
+}
+
+/*! BIT_readBitsFast() :
+*   unsafe version; only works only if nbBits >= 1 */
+ZSTD_STATIC size_t BIT_readBitsFast(BIT_DStream_t *bitD, U32 nbBits)
+{
+	size_t const value = BIT_lookBitsFast(bitD, nbBits);
+	BIT_skipBits(bitD, nbBits);
+	return value;
+}
+
+/*! BIT_reloadDStream() :
+*   Refill `bitD` from buffer previously set in BIT_initDStream() .
+*   This function is safe, it guarantees it will not read beyond src buffer.
+*   @return : status of `BIT_DStream_t` internal register.
+			  if status == BIT_DStream_unfinished, internal register is filled with >= (sizeof(bitD->bitContainer)*8 - 7) bits */
+ZSTD_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t *bitD)
+{
+	if (bitD->bitsConsumed > (sizeof(bitD->bitContainer) * 8)) /* should not happen => corruption detected */
+		return BIT_DStream_overflow;
+
+	if (bitD->ptr >= bitD->start + sizeof(bitD->bitContainer)) {
+		bitD->ptr -= bitD->bitsConsumed >> 3;
+		bitD->bitsConsumed &= 7;
+		bitD->bitContainer = ZSTD_readLEST(bitD->ptr);
+		return BIT_DStream_unfinished;
+	}
+	if (bitD->ptr == bitD->start) {
+		if (bitD->bitsConsumed < sizeof(bitD->bitContainer) * 8)
+			return BIT_DStream_endOfBuffer;
+		return BIT_DStream_completed;
+	}
+	{
+		U32 nbBytes = bitD->bitsConsumed >> 3;
+		BIT_DStream_status result = BIT_DStream_unfinished;
+		if (bitD->ptr - nbBytes < bitD->start) {
+			nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */
+			result = BIT_DStream_endOfBuffer;
+		}
+		bitD->ptr -= nbBytes;
+		bitD->bitsConsumed -= nbBytes * 8;
+		bitD->bitContainer = ZSTD_readLEST(bitD->ptr); /* reminder : srcSize > sizeof(bitD) */
+		return result;
+	}
+}
+
+/*! BIT_endOfDStream() :
+*   @return Tells if DStream has exactly reached its end (all bits consumed).
+*/
+ZSTD_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t *DStream)
+{
+	return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer) * 8));
+}
+
+#endif /* BITSTREAM_H_MODULE */

lib/zstd/entropy_common.c

+// SPDX-License-Identifier: (GPL-2.0 or BSD-2-Clause)
+/*
+ * Common functions of New Generation Entropy library
+ * Copyright (C) 2016, Yann Collet.
+ *
+ * You can contact the author at :
+ * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ */
+
+/* *************************************
+*  Dependencies
+***************************************/
+#include "error_private.h" /* ERR_*, ERROR */
+#include "fse.h"
+#include "huf.h"
+#include "mem.h"
+
+/*===   Version   ===*/
+unsigned FSE_versionNumber(void) { return FSE_VERSION_NUMBER; }
+
+/*===   Error Management   ===*/
+unsigned FSE_isError(size_t code) { return ERR_isError(code); }
+
+unsigned HUF_isError(size_t code) { return ERR_isError(code); }
+
+/*-**************************************************************
+*  FSE NCount encoding-decoding
+****************************************************************/
+size_t FSE_readNCount(short *normalizedCounter, unsigned *maxSVPtr, unsigned *tableLogPtr, const void *headerBuffer, size_t hbSize)
+{
+	const BYTE *const istart = (const BYTE *)headerBuffer;
+	const BYTE *const iend = istart + hbSize;
+	const BYTE *ip = istart;
+	int nbBits;
+	int remaining;
+	int threshold;
+	U32 bitStream;
+	int bitCount;
+	unsigned charnum = 0;
+	int previous0 = 0;
+
+	if (hbSize < 4)
+		return ERROR(srcSize_wrong);
+	bitStream = ZSTD_readLE32(ip);
+	nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */
+	if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX)
+		return ERROR(tableLog_tooLarge);
+	bitStream >>= 4;
+	bitCount = 4;
+	*tableLogPtr = nbBits;
+	remaining = (1 << nbBits) + 1;
+	threshold = 1 << nbBits;
+	nbBits++;
+
+	while ((remaining > 1) & (charnum <= *maxSVPtr)) {
+		if (previous0) {
+			unsigned n0 = charnum;
+			while ((bitStream & 0xFFFF) == 0xFFFF) {
+				n0 += 24;
+				if (ip < iend - 5) {
+					ip += 2;
+					bitStream = ZSTD_readLE32(ip) >> bitCount;
+				} else {
+					bitStream >>= 16;
+					bitCount += 16;
+				}
+			}
+			while ((bitStream & 3) == 3) {
+				n0 += 3;
+				bitStream >>= 2;
+				bitCount += 2;
+			}
+			n0 += bitStream & 3;
+			bitCount += 2;
+			if (n0 > *maxSVPtr)
+				return ERROR(maxSymbolValue_tooSmall);
+			while (charnum < n0)
+				normalizedCounter[charnum++] = 0;
+			if ((ip <= iend - 7) || (ip + (bitCount >> 3) <= iend - 4)) {
+				ip += bitCount >> 3;
+				bitCount &= 7;
+				bitStream = ZSTD_readLE32(ip) >> bitCount;
+			} else {
+				bitStream >>= 2;
+			}
+		}
+		{
+			int const max = (2 * threshold - 1) - remaining;
+			int count;
+
+			if ((bitStream & (threshold - 1)) < (U32)max) {
+				count = bitStream & (threshold - 1);
+				bitCount += nbBits - 1;
+			} else {
+				count = bitStream & (2 * threshold - 1);
+				if (count >= threshold)
+					count -= max;
+				bitCount += nbBits;
+			}
+
+			count--;				 /* extra accuracy */
+			remaining -= count < 0 ? -count : count; /* -1 means +1 */
+			normalizedCounter[charnum++] = (short)count;
+			previous0 = !count;
+			while (remaining < threshold) {
+				nbBits--;
+				threshold >>= 1;
+			}
+
+			if ((ip <= iend - 7) || (ip + (bitCount >> 3) <= iend - 4)) {
+				ip += bitCount >> 3;
+				bitCount &= 7;
+			} else {
+				bitCount -= (int)(8 * (iend - 4 - ip));
+				ip = iend - 4;
+			}
+			bitStream = ZSTD_readLE32(ip) >> (bitCount & 31);
+		}
+	} /* while ((remaining>1) & (charnum<=*maxSVPtr)) */
+	if (remaining != 1)
+		return ERROR(corruption_detected);
+	if (bitCount > 32)
+		return ERROR(corruption_detected);
+	*maxSVPtr = charnum - 1;
+
+	ip += (bitCount + 7) >> 3;
+	return ip - istart;
+}
+
+/*! HUF_readStats() :
+	Read compact Huffman tree, saved by HUF_writeCTable().
+	`huffWeight` is destination buffer.
+	`rankStats` is assumed to be a table of at least HUF_TABLELOG_MAX U32.
+	@return : size read from `src` , or an error Code .
+	Note : Needed by HUF_readCTable() and HUF_readDTableX?() .
+*/
+size_t HUF_readStats_wksp(BYTE *huffWeight, size_t hwSize, U32 *rankStats, U32 *nbSymbolsPtr, U32 *tableLogPtr, const void *src, size_t srcSize, void *workspace, size_t workspaceSize)
+{
+	U32 weightTotal;
+	const BYTE *ip = (const BYTE *)src;
+	size_t iSize;
+	size_t oSize;
+
+	if (!srcSize)
+		return ERROR(srcSize_wrong);
+	iSize = ip[0];
+	/* memset(huffWeight, 0, hwSize);   */ /* is not necessary, even though some analyzer complain ... */
+
+	if (iSize >= 128) { /* special header */
+		oSize = iSize - 127;
+		iSize = ((oSize + 1) / 2);
+		if (iSize + 1 > srcSize)
+			return ERROR(srcSize_wrong);
+		if (oSize >= hwSize)
+			return ERROR(corruption_detected);
+		ip += 1;
+		{
+			U32 n;
+			for (n = 0; n < oSize; n += 2) {
+				huffWeight[n] = ip[n / 2] >> 4;
+				huffWeight[n + 1] = ip[n / 2] & 15;
+			}
+		}
+	} else {						 /* header compressed with FSE (normal case) */
+		if (iSize + 1 > srcSize)
+			return ERROR(srcSize_wrong);
+		oSize = FSE_decompress_wksp(huffWeight, hwSize - 1, ip + 1, iSize, 6, workspace, workspaceSize); /* max (hwSize-1) values decoded, as last one is implied */
+		if (FSE_isError(oSize))
+			return oSize;
+	}
+
+	/* collect weight stats */
+	memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32));
+	weightTotal = 0;
+	{
+		U32 n;
+		for (n = 0; n < oSize; n++) {
+			if (huffWeight[n] >= HUF_TABLELOG_MAX)
+				return ERROR(corruption_detected);
+			rankStats[huffWeight[n]]++;
+			weightTotal += (1 << huffWeight[n]) >> 1;
+		}
+	}
+	if (weightTotal == 0)
+		return ERROR(corruption_detected);
+
+	/* get last non-null symbol weight (implied, total must be 2^n) */
+	{
+		U32 const tableLog = BIT_highbit32(weightTotal) + 1;
+		if (tableLog > HUF_TABLELOG_MAX)
+			return ERROR(corruption_detected);
+		*tableLogPtr = tableLog;
+		/* determine last weight */
+		{
+			U32 const total = 1 << tableLog;
+			U32 const rest = total - weightTotal;
+			U32 const verif = 1 << BIT_highbit32(rest);
+			U32 const lastWeight = BIT_highbit32(rest) + 1;
+			if (verif != rest)
+				return ERROR(corruption_detected); /* last value must be a clean power of 2 */
+			huffWeight[oSize] = (BYTE)lastWeight;
+			rankStats[lastWeight]++;
+		}
+	}
+
+	/* check tree construction validity */
+	if ((rankStats[1] < 2) || (rankStats[1] & 1))
+		return ERROR(corruption_detected); /* by construction : at least 2 elts of rank 1, must be even */
+
+	/* results */
+	*nbSymbolsPtr = (U32)(oSize + 1);
+	return iSize + 1;
+}

lib/zstd/error_private.h

+/* SPDX-License-Identifier: (GPL-2.0 or BSD-3-Clause-Clear) */
+/**
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * All rights reserved.
+ */
+
+/* Note : this module is expected to remain private, do not expose it */
+
+#ifndef ERROR_H_MODULE
+#define ERROR_H_MODULE
+
+/* ****************************************
+*  Dependencies
+******************************************/
+#include <linux/types.h> /* size_t */
+#include <linux/zstd.h>  /* enum list */
+
+/* ****************************************
+*  Compiler-specific
+******************************************/
+#define ERR_STATIC static __attribute__((unused))
+
+/*-****************************************
+*  Customization (error_public.h)
+******************************************/
+typedef ZSTD_ErrorCode ERR_enum;
+#define PREFIX(name) ZSTD_error_##name
+
+/*-****************************************
+*  Error codes handling
+******************************************/
+#define ERROR(name) ((size_t)-PREFIX(name))
+
+ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); }
+
+ERR_STATIC ERR_enum ERR_getErrorCode(size_t code)
+{
+	if (!ERR_isError(code))
+		return (ERR_enum)0;
+	return (ERR_enum)(0 - code);
+}
+
+#endif /* ERROR_H_MODULE */

lib/zstd/mem.h

+/* SPDX-License-Identifier: (GPL-2.0 or BSD-3-Clause-Clear) */
+/**
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * All rights reserved.
+ */
+
+#ifndef MEM_H_MODULE
+#define MEM_H_MODULE
+
+/*-****************************************
+*  Dependencies
+******************************************/
+#include <asm/unaligned.h>
+#include <compiler.h>
+#include <linux/string.h> /* memcpy */
+#include <linux/types.h>  /* size_t, ptrdiff_t */
+
+/*-****************************************
+*  Compiler specifics
+******************************************/
+#define ZSTD_STATIC static __inline __attribute__((unused))
+
+/*-**************************************************************
+*  Basic Types
+*****************************************************************/
+typedef uint8_t BYTE;
+typedef uint16_t U16;
+typedef int16_t S16;
+typedef uint32_t U32;
+typedef int32_t S32;
+typedef uint64_t U64;
+typedef int64_t S64;
+typedef ptrdiff_t iPtrDiff;
+typedef uintptr_t uPtrDiff;
+
+/*-**************************************************************
+*  Memory I/O
+*****************************************************************/
+ZSTD_STATIC unsigned ZSTD_32bits(void) { return sizeof(size_t) == 4; }
+ZSTD_STATIC unsigned ZSTD_64bits(void) { return sizeof(size_t) == 8; }
+
+#if defined(__LITTLE_ENDIAN)
+#define ZSTD_LITTLE_ENDIAN 1
+#else
+#define ZSTD_LITTLE_ENDIAN 0
+#endif
+
+ZSTD_STATIC unsigned ZSTD_isLittleEndian(void) { return ZSTD_LITTLE_ENDIAN; }
+
+ZSTD_STATIC U16 ZSTD_read16(const void *memPtr) { return get_unaligned((const U16 *)memPtr); }
+
+ZSTD_STATIC U32 ZSTD_read32(const void *memPtr) { return get_unaligned((const U32 *)memPtr); }
+
+ZSTD_STATIC U64 ZSTD_read64(const void *memPtr) { return get_unaligned((const U64 *)memPtr); }
+
+ZSTD_STATIC size_t ZSTD_readST(const void *memPtr) { return get_unaligned((const size_t *)memPtr); }
+
+ZSTD_STATIC void ZSTD_write16(void *memPtr, U16 value) { put_unaligned(value, (U16 *)memPtr); }
+
+ZSTD_STATIC void ZSTD_write32(void *memPtr, U32 value) { put_unaligned(value, (U32 *)memPtr); }
+
+ZSTD_STATIC void ZSTD_write64(void *memPtr, U64 value) { put_unaligned(value, (U64 *)memPtr); }
+
+/*=== Little endian r/w ===*/
+
+ZSTD_STATIC U16 ZSTD_readLE16(const void *memPtr) { return get_unaligned_le16(memPtr); }
+
+ZSTD_STATIC void ZSTD_writeLE16(void *memPtr, U16 val) { put_unaligned_le16(val, memPtr); }
+
+ZSTD_STATIC U32 ZSTD_readLE24(const void *memPtr) { return ZSTD_readLE16(memPtr) + (((const BYTE *)memPtr)[2] << 16); }
+
+ZSTD_STATIC void ZSTD_writeLE24(void *memPtr, U32 val)
+{
+	ZSTD_writeLE16(memPtr, (U16)val);
+	((BYTE *)memPtr)[2] = (BYTE)(val >> 16);
+}
+
+ZSTD_STATIC U32 ZSTD_readLE32(const void *memPtr) { return get_unaligned_le32(memPtr); }
+
+ZSTD_STATIC void ZSTD_writeLE32(void *memPtr, U32 val32) { put_unaligned_le32(val32, memPtr); }
+
+ZSTD_STATIC U64 ZSTD_readLE64(const void *memPtr) { return get_unaligned_le64(memPtr); }
+
+ZSTD_STATIC void ZSTD_writeLE64(void *memPtr, U64 val64) { put_unaligned_le64(val64, memPtr); }
+
+ZSTD_STATIC size_t ZSTD_readLEST(const void *memPtr)
+{
+	if (ZSTD_32bits())
+		return (size_t)ZSTD_readLE32(memPtr);
+	else
+		return (size_t)ZSTD_readLE64(memPtr);
+}
+
+ZSTD_STATIC void ZSTD_writeLEST(void *memPtr, size_t val)
+{
+	if (ZSTD_32bits())
+		ZSTD_writeLE32(memPtr, (U32)val);
+	else
+		ZSTD_writeLE64(memPtr, (U64)val);
+}
+
+/*=== Big endian r/w ===*/
+
+ZSTD_STATIC U32 ZSTD_readBE32(const void *memPtr) { return get_unaligned_be32(memPtr); }
+
+ZSTD_STATIC void ZSTD_writeBE32(void *memPtr, U32 val32) { put_unaligned_be32(val32, memPtr); }
+
+ZSTD_STATIC U64 ZSTD_readBE64(const void *memPtr) { return get_unaligned_be64(memPtr); }
+
+ZSTD_STATIC void ZSTD_writeBE64(void *memPtr, U64 val64) { put_unaligned_be64(val64, memPtr); }
+
+ZSTD_STATIC size_t ZSTD_readBEST(const void *memPtr)
+{
+	if (ZSTD_32bits())
+		return (size_t)ZSTD_readBE32(memPtr);
+	else
+		return (size_t)ZSTD_readBE64(memPtr);
+}
+
+ZSTD_STATIC void ZSTD_writeBEST(void *memPtr, size_t val)
+{
+	if (ZSTD_32bits())
+		ZSTD_writeBE32(memPtr, (U32)val);
+	else
+		ZSTD_writeBE64(memPtr, (U64)val);
+}
+
+/* function safe only for comparisons */
+ZSTD_STATIC U32 ZSTD_readMINMATCH(const void *memPtr, U32 length)
+{
+	switch (length) {
+	default:
+	case 4: return ZSTD_read32(memPtr);
+	case 3:
+		if (ZSTD_isLittleEndian())
+			return ZSTD_read32(memPtr) << 8;
+		else
+			return ZSTD_read32(memPtr) >> 8;
+	}
+}
+
+#endif /* MEM_H_MODULE */