commonlib: Add commonlib/bsd

This patch creates a new commonlib/bsd subdirectory with a similar
purpose to the existing commonlib, with the difference that all files
under this subdirectory shall be licensed under the BSD-3-Clause license
(or compatible permissive license). The goal is to allow more code to be
shared with libpayload in the future.

Initially, I'm going to move a few files there that have already been
BSD-licensed in the existing commonlib. I am also exracting most
contents of the often-needed <commonlib/helpers.h> as long as they have
either been written by me (and are hereby relicensed) or have an
existing equivalent in BSD-licensed libpayload code. I am also
relicensing <commonlib/compression.h> (written by me) and
<commonlib/compiler.h> (same stuff exists in libpayload).

Finally, I am extracting the cb_err error code definitions from
<types.h> into a new BSD-licensed header so that future commonlib/bsd
code can build upon a common set of error values. I am making the
assumption here that the enum constants and the half-sentence fragments
of documentation next to them by themselves do not meet the threshold of
copyrightability.

Change-Id: I316cea70930f131e8e93d4218542ddb5ae4b63a2
Signed-off-by: Julius Werner <jwerner@chromium.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/38420
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>

8 files changed, 890 insertions, 0 deletions

diff --git a/src/commonlib/bsd/include/commonlib/bsd/cb_err.h b/src/commonlib/bsd/include/commonlib/bsd/cb_err.h
new file mode 100644
index 0000000000..ab419a7709
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/cb_err.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-or-later */
+
+#ifndef _COMMONLIB_BSD_CB_ERR_H_
+#define _COMMONLIB_BSD_CB_ERR_H_
+
+#include <stdint.h>
+
+/**
+ * coreboot error codes
+ *
+ * Common error definitions that can be used for any function. All error values
+ * should be negative -- when useful, positive values can also be used to denote
+ * success. Allocate a new group or errors every 100 values.
+ */
+enum cb_err {
+	CB_SUCCESS = 0,		/**< Call completed successfully */
+	CB_ERR = -1,		/**< Generic error code */
+	CB_ERR_ARG = -2,	/**< Invalid argument */
+
+	/* NVRAM/CMOS errors */
+	CB_CMOS_OTABLE_DISABLED = -100,		/**< Option table disabled */
+	CB_CMOS_LAYOUT_NOT_FOUND = -101,	/**< Layout file not found */
+	CB_CMOS_OPTION_NOT_FOUND = -102,	/**< Option string not found */
+	CB_CMOS_ACCESS_ERROR = -103,		/**< CMOS access error */
+	CB_CMOS_CHECKSUM_INVALID = -104,	/**< CMOS checksum is invalid */
+
+	/* Keyboard test failures */
+	CB_KBD_CONTROLLER_FAILURE = -200,
+	CB_KBD_INTERFACE_FAILURE = -201,
+
+	/* I2C controller failures */
+	CB_I2C_NO_DEVICE	= -300,	/**< Device is not responding */
+	CB_I2C_BUSY		= -301,	/**< Device tells it's busy */
+	CB_I2C_PROTOCOL_ERROR	= -302,	/**< Data lost or spurious slave
+					     device response, try again? */
+	CB_I2C_TIMEOUT		= -303, /**< Transmission timed out */
+};
+
+/* Don't typedef the enum directly, so the size is unambiguous for serialization. */
+typedef int32_t cb_err_t;
+
+#endif	/* _COMMONLIB_BSD_CB_ERR_H_ */
diff --git a/src/commonlib/bsd/include/commonlib/bsd/cbfs_serialized.h b/src/commonlib/bsd/include/commonlib/bsd/cbfs_serialized.h
new file mode 100644
index 0000000000..d2fc6267ad
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/cbfs_serialized.h
@@ -0,0 +1,190 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#ifndef _CBFS_SERIALIZED_H_
+#define _CBFS_SERIALIZED_H_
+
+#include <stdint.h>
+
+/** These are standard values for the known compression
+    algorithms that coreboot knows about for stages and
+    payloads.  Of course, other CBFS users can use whatever
+    values they want, as long as they understand them. */
+
+#define CBFS_COMPRESS_NONE  0
+#define CBFS_COMPRESS_LZMA  1
+#define CBFS_COMPRESS_LZ4   2
+
+/** These are standard component types for well known
+    components (i.e - those that coreboot needs to consume.
+    Users are welcome to use any other value for their
+    components */
+
+#define CBFS_TYPE_DELETED    0x00000000
+#define CBFS_TYPE_DELETED2   0xffffffff
+#define CBFS_TYPE_STAGE      0x10
+#define CBFS_TYPE_SELF       0x20
+#define CBFS_TYPE_FIT        0x21
+#define CBFS_TYPE_OPTIONROM  0x30
+#define CBFS_TYPE_BOOTSPLASH 0x40
+#define CBFS_TYPE_RAW        0x50
+#define CBFS_TYPE_VSA        0x51
+#define CBFS_TYPE_MBI        0x52
+#define CBFS_TYPE_MICROCODE  0x53
+#define CBFS_TYPE_FSP        0x60
+#define CBFS_TYPE_MRC        0x61
+#define CBFS_TYPE_MMA        0x62
+#define CBFS_TYPE_EFI        0x63
+#define CBFS_TYPE_STRUCT     0x70
+#define CBFS_COMPONENT_CMOS_DEFAULT 0xaa
+#define CBFS_TYPE_SPD          0xab
+#define CBFS_TYPE_MRC_CACHE    0xac
+#define CBFS_COMPONENT_CMOS_LAYOUT 0x01aa
+
+#define CBFS_HEADER_MAGIC  0x4F524243
+#define CBFS_HEADER_VERSION1 0x31313131
+#define CBFS_HEADER_VERSION2 0x31313132
+#define CBFS_HEADER_VERSION  CBFS_HEADER_VERSION2
+
+/* this is the master cbfs header - it must be located somewhere available
+ * to bootblock (to load romstage). The last 4 bytes in the image contain its
+ * relative offset from the end of the image (as a 32-bit signed integer). */
+
+struct cbfs_header {
+	uint32_t magic;
+	uint32_t version;
+	uint32_t romsize;
+	uint32_t bootblocksize;
+	uint32_t align; /* fixed to 64 bytes */
+	uint32_t offset;
+	uint32_t architecture;
+	uint32_t pad[1];
+} __packed;
+
+/* this used to be flexible, but wasn't ever set to something different. */
+#define CBFS_ALIGNMENT 64
+
+/* "Unknown" refers to CBFS headers version 1,
+ * before the architecture was defined (i.e., x86 only).
+ */
+#define CBFS_ARCHITECTURE_UNKNOWN  0xFFFFFFFF
+#define CBFS_ARCHITECTURE_X86      0x00000001
+#define CBFS_ARCHITECTURE_ARM      0x00000010
+
+/** This is a component header - every entry in the CBFS
+    will have this header.
+
+    This is how the component is arranged in the ROM:
+
+    --------------   <- 0
+    component header
+    --------------   <- sizeof(struct component)
+    component name
+    --------------   <- offset
+    data
+    ...
+    --------------   <- offset + len
+*/
+
+#define CBFS_FILE_MAGIC "LARCHIVE"
+
+struct cbfs_file {
+	char magic[8];
+	uint32_t len;
+	uint32_t type;
+	uint32_t attributes_offset;
+	uint32_t offset;
+} __packed;
+
+/* The common fields of extended cbfs file attributes.
+   Attributes are expected to start with tag/len, then append their
+   specific fields. */
+struct cbfs_file_attribute {
+	uint32_t tag;
+	/* len covers the whole structure, incl. tag and len */
+	uint32_t len;
+	uint8_t data[0];
+} __packed;
+
+/* Depending on how the header was initialized, it may be backed with 0x00 or
+ * 0xff. Support both. */
+#define CBFS_FILE_ATTR_TAG_UNUSED 0
+#define CBFS_FILE_ATTR_TAG_UNUSED2 0xffffffff
+#define CBFS_FILE_ATTR_TAG_COMPRESSION 0x42435a4c
+#define CBFS_FILE_ATTR_TAG_HASH 0x68736148
+#define CBFS_FILE_ATTR_TAG_POSITION 0x42435350  /* PSCB */
+#define CBFS_FILE_ATTR_TAG_ALIGNMENT 0x42434c41 /* ALCB */
+#define CBFS_FILE_ATTR_TAG_IBB 0x32494242 /* Initial BootBlock */
+
+struct cbfs_file_attr_compression {
+	uint32_t tag;
+	uint32_t len;
+	/* whole file compression format. 0 if no compression. */
+	uint32_t compression;
+	uint32_t decompressed_size;
+} __packed;
+
+struct cbfs_file_attr_hash {
+	uint32_t tag;
+	uint32_t len;
+	uint32_t hash_type;
+	/* hash_data is len - sizeof(struct) bytes */
+	uint8_t  hash_data[];
+} __packed;
+
+struct cbfs_file_attr_position {
+	uint32_t tag;
+	uint32_t len;
+	uint32_t position;
+} __packed;
+
+struct cbfs_file_attr_align {
+	uint32_t tag;
+	uint32_t len;
+	uint32_t alignment;
+} __packed;
+
+
+/*** Component sub-headers ***/
+
+/* Following are component sub-headers for the "standard"
+   component types */
+
+/** This is the sub-header for stage components.  Stages are
+    loaded by coreboot during the normal boot process */
+
+struct cbfs_stage {
+	uint32_t compression;  /** Compression type */
+	uint64_t entry;  /** entry point */
+	uint64_t load;   /** Where to load in memory */
+	uint32_t len;          /** length of data to load */
+	uint32_t memlen;	   /** total length of object in memory */
+} __packed;
+
+/** this is the sub-header for payload components.  Payloads
+    are loaded by coreboot at the end of the boot process */
+
+struct cbfs_payload_segment {
+	uint32_t type;
+	uint32_t compression;
+	uint32_t offset;
+	uint64_t load_addr;
+	uint32_t len;
+	uint32_t mem_len;
+} __packed;
+
+struct cbfs_payload {
+	struct cbfs_payload_segment segments;
+};
+
+#define PAYLOAD_SEGMENT_CODE   0x434F4445
+#define PAYLOAD_SEGMENT_DATA   0x44415441
+#define PAYLOAD_SEGMENT_BSS    0x42535320
+#define PAYLOAD_SEGMENT_PARAMS 0x50415241
+#define PAYLOAD_SEGMENT_ENTRY  0x454E5452
+
+struct cbfs_optionrom {
+	uint32_t compression;
+	uint32_t len;
+} __packed;
+
+#endif /* _CBFS_SERIALIZED_H_ */
diff --git a/src/commonlib/bsd/include/commonlib/bsd/compiler.h b/src/commonlib/bsd/include/commonlib/bsd/compiler.h
new file mode 100644
index 0000000000..ee2ff88d10
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/compiler.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#ifndef _COMMONLIB_BSD_COMPILER_H_
+#define _COMMONLIB_BSD_COMPILER_H_
+
+#ifndef __packed
+#if defined(__WIN32) || defined(__WIN64)
+#define __packed __attribute__((gcc_struct, packed))
+#else
+#define __packed __attribute__((packed))
+#endif
+#endif
+
+#ifndef __aligned
+#define __aligned(x) __attribute__((aligned(x)))
+#endif
+
+#ifndef __always_unused
+#define __always_unused __attribute__((unused))
+#endif
+
+#ifndef __must_check
+#define __must_check __attribute__((warn_unused_result))
+#endif
+
+#ifndef __weak
+#define __weak __attribute__((weak))
+#endif
+
+#ifndef __noreturn
+#define __noreturn __attribute__((noreturn))
+#endif
+
+#ifndef __always_inline
+#define __always_inline inline __attribute__((always_inline))
+#endif
+
+/* This evaluates to the type of the first expression, unless that is constant
+   in which case it evalutates to the type of the second. This is useful when
+   assigning macro parameters to temporary variables, because that would
+   normally circumvent the special loosened type promotion rules for integer
+   literals. By using this macro, the promotion can happen at the time the
+   literal is assigned to the temporary variable. If the literal doesn't fit in
+   the chosen type, -Werror=overflow will catch it, so this should be safe. */
+#define __TYPEOF_UNLESS_CONST(expr, fallback_expr) __typeof__( \
+	__builtin_choose_expr(__builtin_constant_p(expr), fallback_expr, expr))
+
+/* This creates a unique local variable name for use in macros. */
+#define __TMPNAME_3(i) __tmpname_##i
+#define __TMPNAME_2(i) __TMPNAME_3(i)
+#define __TMPNAME __TMPNAME_2(__COUNTER__)
+
+#endif
diff --git a/src/commonlib/bsd/include/commonlib/bsd/compression.h b/src/commonlib/bsd/include/commonlib/bsd/compression.h
new file mode 100644
index 0000000000..873e7e4e15
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/compression.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#ifndef _COMMONLIB_COMPRESSION_H_
+#define _COMMONLIB_COMPRESSION_H_
+
+#include <stddef.h>
+
+/* Decompresses an LZ4F image (multiple LZ4 blocks with frame header) from src
+ * to dst, ensuring that it doesn't read more than srcn bytes and doesn't write
+ * more than dstn. Buffer sizes must stay below 2GB. Can decompress files loaded
+ * to the end of a buffer in-place, as long as buffer is larger than the final
+ * output size. (Usually just a few bytes, but may be up to (8 + dstn/255) in
+ * worst case. Will reliably return an error if buffer was too small.)
+ * Returns amount of decompressed bytes, or 0 on error.
+ */
+size_t ulz4fn(const void *src, size_t srcn, void *dst, size_t dstn);
+
+/* Same as ulz4fn() but does not perform any bounds checks. */
+size_t ulz4f(const void *src, void *dst);
+
+#endif	/* _COMMONLIB_COMPRESSION_H_ */
diff --git a/src/commonlib/bsd/include/commonlib/bsd/fmap_serialized.h b/src/commonlib/bsd/include/commonlib/bsd/fmap_serialized.h
new file mode 100644
index 0000000000..3d328c45b8
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/fmap_serialized.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#ifndef FLASHMAP_SERIALIZED_H__
+#define FLASHMAP_SERIALIZED_H__
+
+#include <stdint.h>
+
+#define FMAP_SIGNATURE		"__FMAP__"
+#define FMAP_VER_MAJOR		1	/* this header's FMAP minor version */
+#define FMAP_VER_MINOR		1	/* this header's FMAP minor version */
+#define FMAP_STRLEN		32	/* maximum length for strings, */
+					/* including null-terminator */
+
+enum fmap_flags {
+	FMAP_AREA_STATIC	= 1 << 0,
+	FMAP_AREA_COMPRESSED	= 1 << 1,
+	FMAP_AREA_RO		= 1 << 2,
+	FMAP_AREA_PRESERVE	= 1 << 3,
+};
+
+/* Mapping of volatile and static regions in firmware binary */
+struct fmap_area {
+	uint32_t offset;                /* offset relative to base */
+	uint32_t size;                  /* size in bytes */
+	uint8_t  name[FMAP_STRLEN];     /* descriptive name */
+	uint16_t flags;                 /* flags for this area */
+}  __packed;
+
+struct fmap {
+	uint8_t  signature[8];		/* "__FMAP__" (0x5F5F464D41505F5F) */
+	uint8_t  ver_major;		/* major version */
+	uint8_t  ver_minor;		/* minor version */
+	uint64_t base;			/* address of the firmware binary */
+	uint32_t size;			/* size of firmware binary in bytes */
+	uint8_t  name[FMAP_STRLEN];	/* name of this firmware binary */
+	uint16_t nareas;		/* number of areas described by
+					   fmap_areas[] below */
+	struct fmap_area areas[];
+} __packed;
+
+#endif	/* FLASHMAP_SERIALIZED_H__ */
diff --git a/src/commonlib/bsd/include/commonlib/bsd/helpers.h b/src/commonlib/bsd/include/commonlib/bsd/helpers.h
new file mode 100644
index 0000000000..a305df0cd5
--- /dev/null
+++ b/src/commonlib/bsd/include/commonlib/bsd/helpers.h
@@ -0,0 +1,89 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#ifndef COMMONLIB_BSD_HELPERS_H
+#define COMMONLIB_BSD_HELPERS_H
+
+#ifndef __ASSEMBLER__
+#include <commonlib/bsd/compiler.h>
+#include <stddef.h>
+#endif
+
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0]))
+#endif
+
+#define ALIGN(x, a)             __ALIGN_MASK(x, (__typeof__(x))(a)-1UL)
+#define __ALIGN_MASK(x, mask)   (((x)+(mask))&~(mask))
+#define ALIGN_UP(x, a)          ALIGN((x), (a))
+#define ALIGN_DOWN(x, a)        ((x) & ~((__typeof__(x))(a)-1UL))
+#define IS_ALIGNED(x, a)        (((x) & ((__typeof__(x))(a)-1UL)) == 0)
+
+/* Double-evaluation unsafe min/max, for bitfields and outside of functions */
+#define __CMP_UNSAFE(a, b, op) ((a) op (b) ? (a) : (b))
+#define MIN_UNSAFE(a, b) __CMP_UNSAFE(a, b, <)
+#define MAX_UNSAFE(a, b) __CMP_UNSAFE(a, b, >)
+
+#define __CMP_SAFE(a, b, op, var_a, var_b) ({ \
+	__TYPEOF_UNLESS_CONST(a, b) var_a = (a); \
+	__TYPEOF_UNLESS_CONST(b, a) var_b = (b); \
+	var_a op var_b ? var_a : var_b; \
+})
+
+#define __CMP(a, b, op) __builtin_choose_expr( \
+	__builtin_constant_p(a) && __builtin_constant_p(b), \
+	__CMP_UNSAFE(a, b, op), __CMP_SAFE(a, b, op, __TMPNAME, __TMPNAME))
+
+#ifndef MIN
+#define MIN(a, b) __CMP(a, b, <)
+#endif
+#ifndef MAX
+#define MAX(a, b) __CMP(a, b, >)
+#endif
+
+#ifndef ABS
+#define ABS(a) ({ \
+	__typeof__(a) _abs_local_a = (a); \
+	(_abs_local_a < 0) ? (-_abs_local_a) : _abs_local_a; \
+})
+#endif
+
+#define IS_POWER_OF_2(x) ({ \
+	__typeof__(x) _power_local_x = (x); \
+	(_power_local_x & (_power_local_x - 1)) == 0; \
+})
+
+#define DIV_ROUND_UP(x, y) ({ \
+	__typeof__(x) _div_local_x = (x); \
+	__typeof__(y) _div_local_y = (y); \
+	(_div_local_x + _div_local_y - 1) / _div_local_y; \
+})
+
+#define SWAP(a, b) do { \
+	__typeof__(&(a)) _swap_local_a = &(a); \
+	__typeof__(&(b)) _swap_local_b = &(b); \
+	__typeof__(a) _swap_local_tmp = *_swap_local_a; \
+	*_swap_local_a = *_swap_local_b; \
+	*_swap_local_b = _swap_local_tmp; \
+} while (0)
+
+/* Standard units. */
+#define KiB (1<<10)
+#define MiB (1<<20)
+#define GiB (1<<30)
+
+#define KHz (1000)
+#define MHz (1000 * KHz)
+#define GHz (1000 * MHz)
+
+#ifndef offsetof
+#define offsetof(TYPE, MEMBER) __builtin_offsetof(TYPE, MEMBER)
+#endif
+
+#define check_member(structure, member, offset) _Static_assert( \
+	offsetof(struct structure, member) == offset, \
+	"`struct " #structure "` offset for `" #member "` is not " #offset)
+
+/* Calculate size of structure member. */
+#define member_size(type, member)	(sizeof(((type *)0)->member))
+
+#endif /* COMMONLIB_BSD_HELPERS_H */
diff --git a/src/commonlib/bsd/lz4.c.inc b/src/commonlib/bsd/lz4.c.inc
new file mode 100644
index 0000000000..b3be4e5b44
--- /dev/null
+++ b/src/commonlib/bsd/lz4.c.inc
@@ -0,0 +1,280 @@
+/*
+   LZ4 - Fast LZ compression algorithm
+   Copyright (C) 2011-2015, Yann Collet.
+
+   BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+   Redistribution and use in source and binary forms, with or without
+   modification, are permitted provided that the following conditions are
+   met:
+
+       * Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+       * Redistributions in binary form must reproduce the above
+   copyright notice, this list of conditions and the following disclaimer
+   in the documentation and/or other materials provided with the
+   distribution.
+
+   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+   You can contact the author at :
+   - LZ4 source repository : https://github.com/Cyan4973/lz4
+   - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
+*/
+
+
+/**************************************
+*  Reading and writing into memory
+**************************************/
+
+/* customized variant of memcpy, which can overwrite up to 7 bytes beyond dstEnd */
+static void LZ4_wildCopy(void* dstPtr, const void* srcPtr, void* dstEnd)
+{
+    BYTE* d = (BYTE*)dstPtr;
+    const BYTE* s = (const BYTE*)srcPtr;
+    BYTE* const e = (BYTE*)dstEnd;
+
+#if 0
+    const size_t l2 = 8 - (((size_t)d) & (sizeof(void*)-1));
+    LZ4_copy8(d,s); if (d>e-9) return;
+    d+=l2; s+=l2;
+#endif /* join to align */
+
+    do { LZ4_copy8(d,s); d+=8; s+=8; } while (d<e);
+}
+
+
+/**************************************
+*  Common Constants
+**************************************/
+#define MINMATCH 4
+
+#define WILDCOPYLENGTH 8
+#define LASTLITERALS 5
+#define MFLIMIT (WILDCOPYLENGTH+MINMATCH)
+static const int LZ4_minLength = (MFLIMIT+1);
+
+#define KB *(1 <<10)
+#define MB *(1 <<20)
+#define GB *(1U<<30)
+
+#define MAXD_LOG 16
+#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
+
+#define ML_BITS  4
+#define ML_MASK  ((1U<<ML_BITS)-1)
+#define RUN_BITS (8-ML_BITS)
+#define RUN_MASK ((1U<<RUN_BITS)-1)
+
+
+/**************************************
+*  Local Structures and types
+**************************************/
+typedef enum { noDict = 0, withPrefix64k, usingExtDict } dict_directive;
+typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
+typedef enum { full = 0, partial = 1 } earlyEnd_directive;
+
+
+
+/*******************************
+*  Decompression functions
+*******************************/
+/*
+ * This generic decompression function cover all use cases.
+ * It shall be instantiated several times, using different sets of directives
+ * Note that it is essential this generic function is really inlined,
+ * in order to remove useless branches during compilation optimization.
+ */
+FORCE_INLINE int LZ4_decompress_generic(
+                 const char* const source,
+                 char* const dest,
+                 int inputSize,
+                 int outputSize,         /* If endOnInput==endOnInputSize, this value is the max size of Output Buffer. */
+
+                 int endOnInput,         /* endOnOutputSize, endOnInputSize */
+                 int partialDecoding,    /* full, partial */
+                 int targetOutputSize,   /* only used if partialDecoding==partial */
+                 int dict,               /* noDict, withPrefix64k, usingExtDict */
+                 const BYTE* const lowPrefix,  /* == dest if dict == noDict */
+                 const BYTE* const dictStart,  /* only if dict==usingExtDict */
+                 const size_t dictSize         /* note : = 0 if noDict */
+                 )
+{
+    /* Local Variables */
+    const BYTE* ip = (const BYTE*) source;
+    const BYTE* const iend = ip + inputSize;
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const oend = op + outputSize;
+    BYTE* cpy;
+    BYTE* oexit = op + targetOutputSize;
+    const BYTE* const lowLimit = lowPrefix - dictSize;
+
+    const BYTE* const dictEnd = (const BYTE*)dictStart + dictSize;
+    const unsigned dec32table[] = {4, 1, 2, 1, 4, 4, 4, 4};
+    const int dec64table[] = {0, 0, 0, -1, 0, 1, 2, 3};
+
+    const int safeDecode = (endOnInput==endOnInputSize);
+    const int checkOffset = ((safeDecode) && (dictSize < (int)(64 KB)));
+    const int inPlaceDecode = ((ip >= op) && (ip < oend));
+
+
+    /* Special cases */
+    if ((partialDecoding) && (oexit> oend-MFLIMIT)) oexit = oend-MFLIMIT;                         /* targetOutputSize too high => decode everything */
+    if ((endOnInput) && (unlikely(outputSize==0))) return ((inputSize==1) && (*ip==0)) ? 0 : -1;  /* Empty output buffer */
+    if ((!endOnInput) && (unlikely(outputSize==0))) return (*ip==0?1:-1);
+
+
+    /* Main Loop */
+    while (1)
+    {
+        unsigned token;
+        size_t length;
+        const BYTE* match;
+        size_t offset;
+
+        if (unlikely((inPlaceDecode) && (op + WILDCOPYLENGTH > ip))) goto _output_error;   /* output stream ran over input stream */
+
+        /* get literal length */
+        token = *ip++;
+        if ((length=(token>>ML_BITS)) == RUN_MASK)
+        {
+            unsigned s;
+            do
+            {
+                s = *ip++;
+                length += s;
+            }
+            while ( likely(endOnInput ? ip<iend-RUN_MASK : 1) && (s==255) );
+            if ((safeDecode) && unlikely((size_t)(op+length)<(size_t)(op))) goto _output_error;   /* overflow detection */
+            if ((safeDecode) && unlikely((size_t)(ip+length)<(size_t)(ip))) goto _output_error;   /* overflow detection */
+        }
+
+        /* copy literals */
+        cpy = op+length;
+        if (((endOnInput) && ((cpy>(partialDecoding?oexit:oend-MFLIMIT)) || (ip+length>iend-(2+1+LASTLITERALS))) )
+            || ((!endOnInput) && (cpy>oend-WILDCOPYLENGTH)))
+        {
+            if (partialDecoding)
+            {
+                if (cpy > oend) goto _output_error;                           /* Error : write attempt beyond end of output buffer */
+                if ((endOnInput) && (ip+length > iend)) goto _output_error;   /* Error : read attempt beyond end of input buffer */
+            }
+            else
+            {
+                if ((!endOnInput) && (cpy != oend)) goto _output_error;       /* Error : block decoding must stop exactly there */
+                if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error;   /* Error : input must be consumed */
+            }
+            memmove(op, ip, length);
+            ip += length;
+            op += length;
+            break;     /* Necessarily EOF, due to parsing restrictions */
+        }
+        LZ4_wildCopy(op, ip, cpy);
+        ip += length; op = cpy;
+
+        /* get offset */
+        offset = LZ4_readLE16(ip); ip+=2;
+        match = op - offset;
+        if ((checkOffset) && (unlikely(match < lowLimit))) goto _output_error;   /* Error : offset outside buffers */
+
+        /* get matchlength */
+        length = token & ML_MASK;
+        if (length == ML_MASK)
+        {
+            unsigned s;
+            do
+            {
+                if ((endOnInput) && (ip > iend-LASTLITERALS)) goto _output_error;
+                s = *ip++;
+                length += s;
+            } while (s==255);
+            if ((safeDecode) && unlikely((size_t)(op+length)<(size_t)op)) goto _output_error;   /* overflow detection */
+        }
+        length += MINMATCH;
+
+        /* check external dictionary */
+        if ((dict==usingExtDict) && (match < lowPrefix))
+        {
+            if (unlikely(op+length > oend-LASTLITERALS)) goto _output_error;   /* doesn't respect parsing restriction */
+
+            if (length <= (size_t)(lowPrefix-match))
+            {
+                /* match can be copied as a single segment from external dictionary */
+                match = dictEnd - (lowPrefix-match);
+                memmove(op, match, length); op += length;
+            }
+            else
+            {
+                /* match encompass external dictionary and current block */
+                size_t copySize = (size_t)(lowPrefix-match);
+                memcpy(op, dictEnd - copySize, copySize);
+                op += copySize;
+                copySize = length - copySize;
+                if (copySize > (size_t)(op-lowPrefix))   /* overlap copy */
+                {
+                    BYTE* const endOfMatch = op + copySize;
+                    const BYTE* copyFrom = lowPrefix;
+                    while (op < endOfMatch) *op++ = *copyFrom++;
+                }
+                else
+                {
+                    memcpy(op, lowPrefix, copySize);
+                    op += copySize;
+                }
+            }
+            continue;
+        }
+
+        /* copy match within block */
+        cpy = op + length;
+        if (unlikely(offset<8))
+        {
+            const int dec64 = dec64table[offset];
+            op[0] = match[0];
+            op[1] = match[1];
+            op[2] = match[2];
+            op[3] = match[3];
+            match += dec32table[offset];
+            memcpy(op+4, match, 4);
+            match -= dec64;
+        } else { LZ4_copy8(op, match); match+=8; }
+        op += 8;
+
+        if (unlikely(cpy>oend-12))
+        {
+            BYTE* const oCopyLimit = oend-(WILDCOPYLENGTH-1);
+            if (cpy > oend-LASTLITERALS) goto _output_error;    /* Error : last LASTLITERALS bytes must be literals (uncompressed) */
+            if (op < oCopyLimit)
+            {
+                LZ4_wildCopy(op, match, oCopyLimit);
+                match += oCopyLimit - op;
+                op = oCopyLimit;
+            }
+            while (op<cpy) *op++ = *match++;
+        }
+        else
+            LZ4_wildCopy(op, match, cpy);
+        op=cpy;   /* correction */
+    }
+
+    /* end of decoding */
+    if (endOnInput)
+       return (int) (((char*)op)-dest);     /* Nb of output bytes decoded */
+    else
+       return (int) (((const char*)ip)-source);   /* Nb of input bytes read */
+
+    /* Overflow error detected */
+_output_error:
+    return (int) (-(((const char*)ip)-source))-1;
+}
diff --git a/src/commonlib/bsd/lz4_wrapper.c b/src/commonlib/bsd/lz4_wrapper.c
new file mode 100644
index 0000000000..2367afceaf
--- /dev/null
+++ b/src/commonlib/bsd/lz4_wrapper.c
@@ -0,0 +1,174 @@
+/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0-only */
+
+#include <commonlib/bsd/compression.h>
+#include <commonlib/bsd/helpers.h>
+#include <endian.h>
+#include <stdint.h>
+#include <string.h>
+
+/* LZ4 comes with its own supposedly portable memory access functions, but they
+ * seem to be very inefficient in practice (at least on ARM64). Since coreboot
+ * knows about endinaness and allows some basic assumptions (such as unaligned
+ * access support), we can easily write the ones we need ourselves. */
+static uint16_t LZ4_readLE16(const void *src)
+{
+	return le16toh(*(const uint16_t *)src);
+}
+static void LZ4_copy8(void *dst, const void *src)
+{
+/* ARM32 needs to be a special snowflake to prevent GCC from coalescing the
+ * access into LDRD/STRD (which don't support unaligned accesses). */
+#ifdef __arm__	/* ARMv < 6 doesn't support unaligned accesses at all. */
+	#if defined(__COREBOOT_ARM_ARCH__) && __COREBOOT_ARM_ARCH__ < 6
+		int i;
+		for (i = 0; i < 8; i++)
+			((uint8_t *)dst)[i] = ((uint8_t *)src)[i];
+	#else
+		uint32_t x0, x1;
+		__asm__ ("ldr %[x0], [%[src]]"
+			: [x0]"=r"(x0)
+			: [src]"r"(src), "m"(*(const uint32_t *)src));
+		__asm__ ("ldr %[x1], [%[src], #4]"
+			: [x1]"=r"(x1)
+			: [src]"r"(src), "m"(*(const uint32_t *)(src + 4)));
+		__asm__ ("str %[x0], [%[dst]]"
+			: "=m"(*(uint32_t *)dst)
+			: [x0]"r"(x0), [dst]"r"(dst));
+		__asm__ ("str %[x1], [%[dst], #4]"
+			: "=m"(*(uint32_t *)(dst + 4))
+			: [x1]"r"(x1), [dst]"r"(dst));
+	#endif
+#elif defined(__riscv)
+	/* RISC-V implementations may trap on any unaligned access. */
+	int i;
+	for (i = 0; i < 8; i++)
+		((uint8_t *)dst)[i] = ((uint8_t *)src)[i];
+#else
+	*(uint64_t *)dst = *(const uint64_t *)src;
+#endif
+}
+
+typedef  uint8_t BYTE;
+typedef uint16_t U16;
+typedef uint32_t U32;
+typedef  int32_t S32;
+typedef uint64_t U64;
+
+#define FORCE_INLINE static __always_inline
+#define likely(expr) __builtin_expect((expr) != 0, 1)
+#define unlikely(expr) __builtin_expect((expr) != 0, 0)
+
+/* Unaltered (just removed unrelated code) from github.com/Cyan4973/lz4/dev. */
+#include "lz4.c.inc"	/* #include for inlining, do not link! */
+
+#define LZ4F_MAGICNUMBER 0x184D2204
+
+struct lz4_frame_header {
+	uint32_t magic;
+	union {
+		uint8_t flags;
+		struct {
+			uint8_t reserved0		: 2;
+			uint8_t has_content_checksum	: 1;
+			uint8_t has_content_size	: 1;
+			uint8_t has_block_checksum	: 1;
+			uint8_t independent_blocks	: 1;
+			uint8_t version			: 2;
+		};
+	};
+	union {
+		uint8_t block_descriptor;
+		struct {
+			uint8_t reserved1		: 4;
+			uint8_t max_block_size		: 3;
+			uint8_t reserved2		: 1;
+		};
+	};
+	/* + uint64_t content_size iff has_content_size is set */
+	/* + uint8_t header_checksum */
+} __packed;
+
+struct lz4_block_header {
+	union {
+		uint32_t raw;
+		struct {
+			uint32_t size		: 31;
+			uint32_t not_compressed	: 1;
+		};
+	};
+	/* + size bytes of data */
+	/* + uint32_t block_checksum iff has_block_checksum is set */
+} __packed;
+
+size_t ulz4fn(const void *src, size_t srcn, void *dst, size_t dstn)
+{
+	const void *in = src;
+	void *out = dst;
+	size_t out_size = 0;
+	int has_block_checksum;
+
+	{ /* With in-place decompression the header may become invalid later. */
+		const struct lz4_frame_header *h = in;
+
+		if (srcn < sizeof(*h) + sizeof(uint64_t) + sizeof(uint8_t))
+			return 0;	/* input overrun */
+
+		/* We assume there's always only a single, standard frame. */
+		if (le32toh(h->magic) != LZ4F_MAGICNUMBER || h->version != 1)
+			return 0;	/* unknown format */
+		if (h->reserved0 || h->reserved1 || h->reserved2)
+			return 0;	/* reserved must be zero */
+		if (!h->independent_blocks)
+			return 0;	/* we don't support block dependency */
+		has_block_checksum = h->has_block_checksum;
+
+		in += sizeof(*h);
+		if (h->has_content_size)
+			in += sizeof(uint64_t);
+		in += sizeof(uint8_t);
+	}
+
+	while (1) {
+		struct lz4_block_header b = {
+			{ .raw = le32toh(*(const uint32_t *)in) }
+		};
+		in += sizeof(struct lz4_block_header);
+
+		if ((size_t)(in - src) + b.size > srcn)
+			break;			/* input overrun */
+
+		if (!b.size) {
+			out_size = out - dst;
+			break;			/* decompression successful */
+		}
+
+		if (b.not_compressed) {
+			size_t size = MIN((uintptr_t)b.size, (uintptr_t)dst
+				+ dstn - (uintptr_t)out);
+			memcpy(out, in, size);
+			if (size < b.size)
+				break;		/* output overrun */
+			out += size;
+		} else {
+			/* constant folding essential, do not touch params! */
+			int ret = LZ4_decompress_generic(in, out, b.size,
+					dst + dstn - out, endOnInputSize,
+					full, 0, noDict, out, NULL, 0);
+			if (ret < 0)
+				break;		/* decompression error */
+			out += ret;
+		}
+
+		in += b.size;
+		if (has_block_checksum)
+			in += sizeof(uint32_t);
+	}
+
+	return out_size;
+}
+
+size_t ulz4f(const void *src, void *dst)
+{
+	/* LZ4 uses signed size parameters, so can't just use ((u32)-1) here. */
+	return ulz4fn(src, 1*GiB, dst, 1*GiB);
+}