[PATCH 0/2] Add support for ZSTD-compressed kernel

[PATCH 0/2] Add support for ZSTD-compressed kernel

[Nick Terrell]

Hi,

This patch set adds support for a ZSTD-compressed kernel and ramdisk
images in the kernel boot process. It only integrates the support with
x86, though the first patch is generic to all architectures.

Zstandard requires slightly more memory during the kernel decompression
on x86 (192 KB vs 64 KB), and the memory usage is independent of the
window size.

Zstandard requires memory proprortional to the window size used during
compression for decompressing the ramdisk image, since streaming mode is
used. Newer versions of zstd (1.3.2+) list the window size of a file
with `zstd -lv <file>'. The absolute maximum amount of memory required
is just over 8 MB.

Thanks,
Nick Terrell

Nick Terrell (2):
  lib: Add support for ZSTD-compressed kernel
  x86: Add support for ZSTD-compressed kernel

 Documentation/x86/boot.txt        |   6 +-
 arch/x86/Kconfig                  |   1 +
 arch/x86/boot/compressed/Makefile |   5 +-
 arch/x86/boot/compressed/misc.c   |   4 +
 arch/x86/boot/header.S            |   8 +-
 arch/x86/include/asm/boot.h       |   6 +-
 include/linux/decompress/unzstd.h |  26 +++
 init/Kconfig                      |  14 +-
 lib/Kconfig                       |   4 +
 lib/Makefile                      |   1 +
 lib/decompress.c                  |   5 +
 lib/decompress_unzstd.c           | 341 ++++++++++++++++++++++++++++++++++++++
 lib/xxhash.c                      |  21 ++-
 lib/zstd/decompress.c             |   2 +
 lib/zstd/fse_decompress.c         |   9 +-
 scripts/Makefile.lib              |  15 ++
 usr/Kconfig                       |  22 +++
 17 files changed, 465 insertions(+), 25 deletions(-)
 create mode 100644 include/linux/decompress/unzstd.h
 create mode 100644 lib/decompress_unzstd.c

-- 
2.9.5

[PATCH 1/2] lib: Add support for ZSTD-compressed kernel

[Nick Terrell]

Add support for extracting ZSTD-compressed kernel images, as well as
ZSTD-compressed ramdisk images in the kernel boot process.

When neither `fill' nor `flush' are used, the decompression function
requires a constant amount of memory (192 KB is sufficient). When either
is used the decompression function requires memory proportional to the
window size used during compression, which is limited to 8 MB. The maximum
memory usage is just over 8 MB.

Signed-off-by: Nick Terrell <terrelln@fb.com>
---
 include/linux/decompress/unzstd.h |  26 +++
 init/Kconfig                      |  14 +-
 lib/Kconfig                       |   4 +
 lib/Makefile                      |   1 +
 lib/decompress.c                  |   5 +
 lib/decompress_unzstd.c           | 341 ++++++++++++++++++++++++++++++++++++++
 lib/xxhash.c                      |  21 ++-
 lib/zstd/decompress.c             |   2 +
 lib/zstd/fse_decompress.c         |   9 +-
 scripts/Makefile.lib              |  15 ++
 usr/Kconfig                       |  22 +++
 11 files changed, 442 insertions(+), 18 deletions(-)
 create mode 100644 include/linux/decompress/unzstd.h
 create mode 100644 lib/decompress_unzstd.c

diff --git a/include/linux/decompress/unzstd.h b/include/linux/decompress/unzstd.h
new file mode 100644
index 0000000..6f3022c
--- /dev/null
+++ b/include/linux/decompress/unzstd.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2017 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#ifndef LINUX_DECOMPRESS_UNZSTD_H
+#define LINUX_DECOMPRESS_UNZSTD_H
+
+int unzstd(unsigned char *inbuf, long len,
+	   long (*fill)(void*, unsigned long),
+	   long (*flush)(void*, unsigned long),
+	   unsigned char *output,
+	   long *pos,
+	   void (*error_fn)(char *x));
+#endif
diff --git a/init/Kconfig b/init/Kconfig
index 78cb246..3caa314 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -127,10 +127,13 @@ config HAVE_KERNEL_LZO
 config HAVE_KERNEL_LZ4
 	bool
 
+config HAVE_KERNEL_ZSTD
+	bool
+
 choice
 	prompt "Kernel compression mode"
 	default KERNEL_GZIP
-	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
+	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_ZSTD
 	help
 	  The linux kernel is a kind of self-extracting executable.
 	  Several compression algorithms are available, which differ
@@ -209,6 +212,15 @@ config KERNEL_LZ4
 	  is about 8% bigger than LZO. But the decompression speed is
 	  faster than LZO.
 
+config KERNEL_ZSTD
+	bool "ZSTD"
+	depends on HAVE_KERNEL_ZSTD
+	help
+	  ZSTD is a compression algorithm targeting intermediate compression
+	  with fast decompression speed. It will compress better than GZIP and
+	  decompress around the same speed as LZO, but slower than LZ4. You
+	  will need at least 192 KB RAM or more for booting.
+
 endchoice
 
 config DEFAULT_HOSTNAME
diff --git a/lib/Kconfig b/lib/Kconfig
index b1445b2..4d32bec 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -285,6 +285,10 @@ config DECOMPRESS_LZ4
 	select LZ4_DECOMPRESS
 	tristate
 
+config DECOMPRESS_ZSTD
+	select ZSTD_DECOMPRESS
+	tristate
+
 #
 # Generic allocator support is selected if needed
 #
diff --git a/lib/Makefile b/lib/Makefile
index dafa796..8afcdc3 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -128,6 +128,7 @@ lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
 lib-$(CONFIG_DECOMPRESS_XZ) += decompress_unxz.o
 lib-$(CONFIG_DECOMPRESS_LZO) += decompress_unlzo.o
 lib-$(CONFIG_DECOMPRESS_LZ4) += decompress_unlz4.o
+lib-$(CONFIG_DECOMPRESS_ZSTD) += decompress_unzstd.o
 
 obj-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
diff --git a/lib/decompress.c b/lib/decompress.c
index 62696df..b17bb07 100644
--- a/lib/decompress.c
+++ b/lib/decompress.c
@@ -12,6 +12,7 @@
 #include <linux/decompress/inflate.h>
 #include <linux/decompress/unlzo.h>
 #include <linux/decompress/unlz4.h>
+#include <linux/decompress/unzstd.h>
 
 #include <linux/types.h>
 #include <linux/string.h>
@@ -36,6 +37,9 @@
 #ifndef CONFIG_DECOMPRESS_LZ4
 # define unlz4 NULL
 #endif
+#ifndef CONFIG_DECOMPRESS_ZSTD
+# define unzstd NULL
+#endif
 
 struct compress_format {
 	unsigned char magic[2];
@@ -51,6 +55,7 @@ static const struct compress_format compressed_formats[] __initconst = {
 	{ {0xfd, 0x37}, "xz", unxz },
 	{ {0x89, 0x4c}, "lzo", unlzo },
 	{ {0x02, 0x21}, "lz4", unlz4 },
+	{ {0x28, 0xb5}, "zstd", unzstd },
 	{ {0, 0}, NULL, NULL }
 };
 
diff --git a/lib/decompress_unzstd.c b/lib/decompress_unzstd.c
new file mode 100644
index 0000000..84315dc
--- /dev/null
+++ b/lib/decompress_unzstd.c
@@ -0,0 +1,341 @@
+/*
+ * Copyright (C) 2017 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <https://www.gnu.org/licenses/>.
+ */
+
+/*
+ * Important notes about in-place decompression
+ *
+ * At least on x86, the kernel is decompressed in place: the compressed data
+ * is placed to the end of the output buffer, and the decompressor overwrites
+ * most of the compressed data. There must be enough safety margin to
+ * guarantee that the write position is always behind the read position.
+ *
+ * The safety margin for ZSTD with a 128 KB block size is calculated below.
+ * Note that the margin with ZSTD is bigger than with GZIP or XZ!
+ *
+ * The worst case for in-place decompression is that the beginning of
+ * the file is compressed extremely well, and the rest of the file is
+ * uncompressible. Thus, we must look for worst-case expansion when the
+ * compressor is encoding uncompressible data.
+ *
+ * The structure of the .zst file in case of a compresed kernel is as follows.
+ * Maximum sizes (as bytes) of the fields are in parenthesis.
+ *
+ *    Frame Header: (18)
+ *    Blocks: (N)
+ *    Checksum: (4)
+ *
+ * The frame header and checksum overhead is at most 22 bytes.
+ *
+ * ZSTD stores the data in blocks. Each block has a header whose size is
+ * a 3 bytes. After the block header, there is up to 128 KB of payload.
+ * The maximum uncompressed size of the payload is 128 KB. The minimum
+ * uncompressed size of the payload is never less than the payload size
+ * (excluding the block header).
+ *
+ * The assumption, that the uncompressed size of the payload is never
+ * smaller than the payload itself, is valid only when talking about
+ * the payload as a whole. It is possible that the payload has parts where
+ * the decompressor consumes more input than it produces output. Calculating
+ * the worst case for this would be tricky. Instead of trying to do that,
+ * let's simply make sure that the decompressor never overwrites any bytes
+ * of the payload which it is currently reading.
+ *
+ * Now we have enough information to calculate the safety margin. We need
+ *   - 22 bytes for the .zst file format headers;
+ *   - 3 bytes per every 128 KiB of uncompressed size (one block header per
+ *     block); and
+ *   - 128 KiB (biggest possible zstd block size) to make sure that the
+ *     decompressor never overwrites anything from the block it is currently
+ *     reading.
+ *
+ * We get the following formula:
+ *
+ *    safety_margin = 22 + uncompressed_size * 3 / 131072 + 131072
+ *                 <= 22 + (uncompressed_size >> 15) + 131072
+ */
+
+#ifdef STATIC
+	/* Preboot environments #include "path/to/decompress_unzstd.c".
+	 * All of the source files we depend on must be #included.
+	 * zstd's only source dependeny is xxhash, which has no source
+	 * dependencies.
+	 *
+	 * zstd and xxhash both avoid declaring themselves as modules
+	 * when PREBOOT is defined.
+	 */
+#	define PREBOOT
+#	include "xxhash.c"
+#	include "zstd/entropy_common.c"
+#	include "zstd/fse_decompress.c"
+#	include "zstd/huf_decompress.c"
+#	include "zstd/zstd_common.c"
+#	include "zstd/decompress.c"
+#endif
+
+#include <linux/decompress/mm.h>
+#include <linux/kernel.h>
+#include <linux/zstd.h>
+
+/* 8 MB maximum window size */
+#define ZSTD_WINDOWSIZE_MAX	(1 << 23)
+/* Size of the input and output buffers in multi-call mdoe */
+#define ZSTD_IOBUF_SIZE		4096
+
+static int INIT handle_zstd_error(size_t ret, void (*error)(char *x))
+{
+	const int err = ZSTD_getErrorCode(ret);
+
+	if (!ZSTD_isError(ret))
+		return 0;
+
+	switch (err) {
+	case ZSTD_error_memory_allocation:
+		error("ZSTD decompressor ran out of memory");
+		break;
+	case ZSTD_error_prefix_unknown:
+		error("Input is not in the ZSTD format (wrong magic bytes)");
+		break;
+	case ZSTD_error_dstSize_tooSmall:
+	case ZSTD_error_corruption_detected:
+	case ZSTD_error_checksum_wrong:
+		error("ZSTD-compressed data is corrupt");
+		break;
+	default:
+		error("ZSTD-compressed data is probably corrupt");
+		break;
+	}
+	return -1;
+}
+
+/* Handle the case where we have the entire input and output in one segment.
+ * We can allocate less memory (no circular buffer for the sliding window),
+ * and avoid some memcpy() calls.
+ */
+static int INIT decompress_single(const u8 *in_buf, long in_len, u8 *out_buf,
+				  long out_len, long *in_pos,
+				  void (*error)(char *x))
+{
+	const size_t wksp_size = ZSTD_DCtxWorkspaceBound();
+	void *wksp = large_malloc(wksp_size);
+	ZSTD_DCtx *dctx = ZSTD_initDCtx(wksp, wksp_size);
+	int err;
+	size_t ret;
+
+	if (dctx == NULL) {
+		error("Out of memory while allocating ZSTD_DCtx");
+		err = -1;
+		goto out;
+	}
+	/* Find out how large the frame actually is, there may be junk at
+	 * the end of the frame that ZSTD_decompressDCtx() can't handle.
+	 */
+	ret = ZSTD_findFrameCompressedSize(in_buf, in_len);
+	err = handle_zstd_error(ret, error);
+	if (err)
+		goto out;
+	in_len = (long)ret;
+
+	ret = ZSTD_decompressDCtx(dctx, out_buf, out_len, in_buf, in_len);
+	err = handle_zstd_error(ret, error);
+	if (err)
+		goto out;
+
+	if (in_pos != NULL)
+		*in_pos = in_len;
+
+	err = 0;
+out:
+	if (wksp != NULL)
+		large_free(wksp);
+	return err;
+}
+
+static int INIT __unzstd(unsigned char *in_buf, long in_len,
+			 long (*fill)(void*, unsigned long),
+			 long (*flush)(void*, unsigned long),
+			 unsigned char *out_buf, long out_len,
+			 long *in_pos,
+			 void (*error)(char *x))
+{
+	ZSTD_inBuffer in;
+	ZSTD_outBuffer out;
+	ZSTD_frameParams params;
+	void *in_allocated = NULL;
+	void *out_allocated = NULL;
+	void *wksp = NULL;
+	size_t wksp_size;
+	ZSTD_DStream *dstream;
+	int err;
+	size_t ret;
+
+	if (out_len == 0)
+		out_len = LONG_MAX; /* no limit */
+
+	if (fill == NULL && flush == NULL)
+		/* We can decompress faster and with less memory when we have a
+		 * single chunk.
+		 */
+		return decompress_single(in_buf, in_len, out_buf, out_len,
+					 in_pos, error);
+
+	/* If in_buf is not provided, we must be using fill(), so allocate
+	 * a large enough buffer. If it is provided, it must be at least
+	 * ZSTD_IOBUF_SIZE large.
+	 */
+	if (in_buf == NULL) {
+		in_allocated = malloc(ZSTD_IOBUF_SIZE);
+		if (in_allocated == NULL) {
+			error("Out of memory while allocating input buffer");
+			err = -1;
+			goto out;
+		}
+		in_buf = in_allocated;
+		in_len = 0;
+	}
+	/* Read the first chunk, since we need to decode the frame header */
+	if (fill != NULL)
+		in_len = fill(in_buf, ZSTD_IOBUF_SIZE);
+	if (in_len < 0) {
+		error("ZSTD-compressed data is truncated");
+		err = -1;
+		goto out;
+	}
+	/* Set the first non-empty input buffer */
+	in.src = in_buf;
+	in.pos = 0;
+	in.size = in_len;
+	/* Allocate the output buffer if we are using flush(). */
+	if (flush != NULL) {
+		out_allocated = malloc(ZSTD_IOBUF_SIZE);
+		if (out_allocated == NULL) {
+			error("Out of memory while allocating output buffer");
+			err = -1;
+			goto out;
+		}
+		out_buf = out_allocated;
+		out_len = ZSTD_IOBUF_SIZE;
+	}
+	/* Set the output buffer */
+	out.dst = out_buf;
+	out.pos = 0;
+	out.size = out_len;
+
+	/* We need to know the window size to allocate the ZSTD_DStream.
+	 * Since we are streaming, we need to allocate a buffer for the sliding
+	 * window. The window size varies from 1 KB to ZSTD_WINDOWSIZE_MAX
+	 * (8 MB), so it is important to use the actual value so as not to
+	 * waste memory when it is smaller.
+	 */
+	ret = ZSTD_getFrameParams(&params, in.src, in.size);
+	err = handle_zstd_error(ret, error);
+	if (err)
+		goto out;
+	if (ret != 0) {
+		error("ZSTD-compressed data has an incomplete frame header");
+		err = -1;
+		goto out;
+	}
+	if (params.windowSize > ZSTD_WINDOWSIZE_MAX) {
+		error("ZSTD-compressed data has too large a window size");
+		err = -1;
+		goto out;
+	}
+
+	/* Allocate the ZSTD_DStream now that we know how much memory is
+	 * required.
+	 */
+	wksp_size = ZSTD_DStreamWorkspaceBound(params.windowSize);
+	wksp = large_malloc(wksp_size);
+	dstream = ZSTD_initDStream(params.windowSize, wksp, wksp_size);
+	if (dstream == NULL) {
+		error("Out of memory while allocating ZSTD_DStream");
+		err = -1;
+		goto out;
+	}
+	/* Decompression loop:
+	 * Read more data if necessary (error if no more data can be read).
+	 * Call the decompression function, which returns 0 when finished.
+	 * Flush any data produced if using flush().
+	 */
+	if (in_pos != NULL)
+		*in_pos = 0;
+	do {
+		/* If we need to reload data, either we have fill() and can
+		 * try to get more data, or we don't and the input is truncated.
+		 */
+		if (in.pos == in.size) {
+			if (in_pos != NULL)
+				*in_pos += in.pos;
+			in_len = fill ? fill(in_buf, ZSTD_IOBUF_SIZE) : -1;
+			if (in_len < 0) {
+				error("ZSTD-compressed data is truncated");
+				err = -1;
+				goto out;
+			}
+			in.pos = 0;
+			in.size = in_len;
+		}
+		/* Returns zero when the frame is complete */
+		ret = ZSTD_decompressStream(dstream, &out, &in);
+		err = handle_zstd_error(ret, error);
+		if (err)
+			goto out;
+		/* Flush all of the data produced if using flush() */
+		if (flush != NULL && out.pos > 0) {
+			if (out.pos != flush(out.dst, out.pos)) {
+				error("Failed to flush()");
+				err = -1;
+				goto out;
+			}
+			out.pos = 0;
+		}
+	} while (ret != 0);
+
+	if (in_pos != NULL)
+		*in_pos += in.pos;
+
+	err = 0;
+out:
+	if (in_allocated != NULL)
+		free(in_allocated);
+	if (out_allocated != NULL)
+		free(out_allocated);
+	if (wksp != NULL)
+		large_free(wksp);
+	return err;
+}
+
+#ifndef PREBOOT
+STATIC int INIT unzstd(unsigned char *buf, long len,
+		       long (*fill)(void*, unsigned long),
+		       long (*flush)(void*, unsigned long),
+		       unsigned char *out_buf,
+		       long *pos,
+		       void (*error)(char *x))
+{
+	return __unzstd(buf, len, fill, flush, out_buf, 0, pos, error);
+}
+#else
+STATIC int INIT __decompress(unsigned char *buf, long len,
+			     long (*fill)(void*, unsigned long),
+			     long (*flush)(void*, unsigned long),
+			     unsigned char *out_buf, long out_len,
+			     long *pos,
+			     void (*error)(char *x))
+{
+	return __unzstd(buf, len, fill, flush, out_buf, out_len, pos, error);
+}
+#endif
diff --git a/lib/xxhash.c b/lib/xxhash.c
index aa61e2a..7f1d3cb 100644
--- a/lib/xxhash.c
+++ b/lib/xxhash.c
@@ -80,13 +80,11 @@ void xxh32_copy_state(struct xxh32_state *dst, const struct xxh32_state *src)
 {
 	memcpy(dst, src, sizeof(*dst));
 }
-EXPORT_SYMBOL(xxh32_copy_state);
 
 void xxh64_copy_state(struct xxh64_state *dst, const struct xxh64_state *src)
 {
 	memcpy(dst, src, sizeof(*dst));
 }
-EXPORT_SYMBOL(xxh64_copy_state);
 
 /*-***************************
  * Simple Hash Functions
@@ -151,7 +149,6 @@ uint32_t xxh32(const void *input, const size_t len, const uint32_t seed)
 
 	return h32;
 }
-EXPORT_SYMBOL(xxh32);
 
 static uint64_t xxh64_round(uint64_t acc, const uint64_t input)
 {
@@ -234,7 +231,6 @@ uint64_t xxh64(const void *input, const size_t len, const uint64_t seed)
 
 	return h64;
 }
-EXPORT_SYMBOL(xxh64);
 
 /*-**************************************************
  * Advanced Hash Functions
@@ -251,7 +247,6 @@ void xxh32_reset(struct xxh32_state *statePtr, const uint32_t seed)
 	state.v4 = seed - PRIME32_1;
 	memcpy(statePtr, &state, sizeof(state));
 }
-EXPORT_SYMBOL(xxh32_reset);
 
 void xxh64_reset(struct xxh64_state *statePtr, const uint64_t seed)
 {
@@ -265,7 +260,6 @@ void xxh64_reset(struct xxh64_state *statePtr, const uint64_t seed)
 	state.v4 = seed - PRIME64_1;
 	memcpy(statePtr, &state, sizeof(state));
 }
-EXPORT_SYMBOL(xxh64_reset);
 
 int xxh32_update(struct xxh32_state *state, const void *input, const size_t len)
 {
@@ -334,7 +328,6 @@ int xxh32_update(struct xxh32_state *state, const void *input, const size_t len)
 
 	return 0;
 }
-EXPORT_SYMBOL(xxh32_update);
 
 uint32_t xxh32_digest(const struct xxh32_state *state)
 {
@@ -372,7 +365,6 @@ uint32_t xxh32_digest(const struct xxh32_state *state)
 
 	return h32;
 }
-EXPORT_SYMBOL(xxh32_digest);
 
 int xxh64_update(struct xxh64_state *state, const void *input, const size_t len)
 {
@@ -439,7 +431,6 @@ int xxh64_update(struct xxh64_state *state, const void *input, const size_t len)
 
 	return 0;
 }
-EXPORT_SYMBOL(xxh64_update);
 
 uint64_t xxh64_digest(const struct xxh64_state *state)
 {
@@ -494,7 +485,19 @@ uint64_t xxh64_digest(const struct xxh64_state *state)
 
 	return h64;
 }
+
+#ifndef PREBOOT
+EXPORT_SYMBOL(xxh32_copy_state);
+EXPORT_SYMBOL(xxh64_copy_state);
+EXPORT_SYMBOL(xxh32);
+EXPORT_SYMBOL(xxh64);
+EXPORT_SYMBOL(xxh32_reset);
+EXPORT_SYMBOL(xxh64_reset);
+EXPORT_SYMBOL(xxh32_update);
+EXPORT_SYMBOL(xxh32_digest);
+EXPORT_SYMBOL(xxh64_update);
 EXPORT_SYMBOL(xxh64_digest);
 
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_DESCRIPTION("xxHash");
+#endif
diff --git a/lib/zstd/decompress.c b/lib/zstd/decompress.c
index b178467..b7b6599 100644
--- a/lib/zstd/decompress.c
+++ b/lib/zstd/decompress.c
@@ -2487,6 +2487,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream *zds, ZSTD_outBuffer *output, ZSTD_inB
 	}
 }
 
+#ifndef PREBOOT
 EXPORT_SYMBOL(ZSTD_DCtxWorkspaceBound);
 EXPORT_SYMBOL(ZSTD_initDCtx);
 EXPORT_SYMBOL(ZSTD_decompressDCtx);
@@ -2526,3 +2527,4 @@ EXPORT_SYMBOL(ZSTD_insertBlock);
 
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_DESCRIPTION("Zstd Decompressor");
+#endif
diff --git a/lib/zstd/fse_decompress.c b/lib/zstd/fse_decompress.c
index a84300e..0b35353 100644
--- a/lib/zstd/fse_decompress.c
+++ b/lib/zstd/fse_decompress.c
@@ -47,6 +47,7 @@
 ****************************************************************/
 #include "bitstream.h"
 #include "fse.h"
+#include "zstd_internal.h"
 #include <linux/compiler.h>
 #include <linux/kernel.h>
 #include <linux/string.h> /* memcpy, memset */
@@ -60,14 +61,6 @@
 		enum { FSE_static_assert = 1 / (int)(!!(c)) }; \
 	} /* use only *after* variable declarations */
 
-/* check and forward error code */
-#define CHECK_F(f)                  \
-	{                           \
-		size_t const e = f; \
-		if (FSE_isError(e)) \
-			return e;   \
-	}
-
 /* **************************************************************
 *  Templates
 ****************************************************************/
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 5e975fe..807c1fa 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -412,6 +412,21 @@ cmd_xzmisc = (cat $(filter-out FORCE,$^) | \
 	xz --check=crc32 --lzma2=dict=1MiB) > $@ || \
 	(rm -f $@ ; false)
 
+# ZSTD
+# ---------------------------------------------------------------------------
+# Appends the uncompressed size of the data using size_append. The .zst
+# format has the size information available at the beginning of the file too,
+# but it's in a more complex format and it's good to avoid changing the part
+# of the boot code that reads the uncompressed size.
+# Note that the bytes added by size_append will make the zstd tool think that
+# the file is corrupt. This is expected.
+
+quiet_cmd_zstd = ZSTD    $@
+cmd_zstd = (cat $(filter-out FORCE,$^) | \
+	zstd -19 && \
+        $(call size_append, $(filter-out FORCE,$^))) > $@ || \
+	(rm -f $@ ; false)
+
 # ASM offsets
 # ---------------------------------------------------------------------------
 
diff --git a/usr/Kconfig b/usr/Kconfig
index d53112fd..b63d8ee 100644
--- a/usr/Kconfig
+++ b/usr/Kconfig
@@ -105,6 +105,15 @@ config RD_LZ4
 	  Support loading of a LZ4 encoded initial ramdisk or cpio buffer
 	  If unsure, say N.
 
+config RD_ZSTD
+	bool "Support initial ramdisk/ramfs compressed using ZSTD"
+	default y
+	depends on BLK_DEV_INITRD
+	select DECOMPRESS_ZSTD
+	help
+	  Support loading of a ZSTD encoded initial ramdisk or cpio buffer.
+	  If unsure, say N.
+
 choice
 	prompt "Built-in initramfs compression mode"
 	depends on INITRAMFS_SOURCE!=""
@@ -213,6 +222,17 @@ config INITRAMFS_COMPRESSION_LZ4
 	  If you choose this, keep in mind that most distros don't provide lz4
 	  by default which could cause a build failure.
 
+config INITRAMFS_COMPRESSION_ZSTD
+	bool "ZSTD"
+	depends on RD_ZSTD
+	help
+	  ZSTD is a compression algorithm targeting intermediate compression
+	  with fast decompression speed. It will compress better than GZIP and
+	  decompress around the same speed as LZO, but slower than LZ4.
+
+	  If you choose this, keep in mind that you may need to install the zstd
+	  tool to be able to compress the initram.
+
 endchoice
 
 config INITRAMFS_COMPRESSION
@@ -225,10 +245,12 @@ config INITRAMFS_COMPRESSION
 	default ".xz"   if INITRAMFS_COMPRESSION_XZ
 	default ".lzo"  if INITRAMFS_COMPRESSION_LZO
 	default ".lz4"  if INITRAMFS_COMPRESSION_LZ4
+	default ".zst"  if INITRAMFS_COMPRESSION_ZSTD
 	default ".gz"   if RD_GZIP
 	default ".lz4"  if RD_LZ4
 	default ".lzo"  if RD_LZO
 	default ".xz"   if RD_XZ
 	default ".lzma" if RD_LZMA
 	default ".bz2"  if RD_BZIP2
+	default ".zst"  if RD_ZSTD
 	default ""
-- 
2.9.5

[PATCH 2/2] x86: Add support for ZSTD-compressed kernel

[Nick Terrell]

Integrates the ZSTD decompression code to the x86 pre-boot code.

Zstandard requires slightly more memory during the kernel decompression
on x86 (192 KB vs 64 KB), and the memory usage is independent of the
window size.

Zstandard requires memory proportional to the window size used during
compression for decompressing the ramdisk image, since streaming mode is
used. Newer versions of zstd (1.3.2+) list the window size of a file
with `zstd -lv <file>'. The absolute maximum amount of memory required
is just over 8 MB.

Signed-off-by: Nick Terrell <terrelln@fb.com>
---
 Documentation/x86/boot.txt        | 6 +++---
 arch/x86/Kconfig                  | 1 +
 arch/x86/boot/compressed/Makefile | 5 ++++-
 arch/x86/boot/compressed/misc.c   | 4 ++++
 arch/x86/boot/header.S            | 8 +++++++-
 arch/x86/include/asm/boot.h       | 6 ++++--
 6 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/Documentation/x86/boot.txt b/Documentation/x86/boot.txt
index 5e9b826..7f5d1f9 100644
--- a/Documentation/x86/boot.txt
+++ b/Documentation/x86/boot.txt
@@ -666,9 +666,9 @@ Protocol:	2.08+
   uncompressed data should be determined using the standard magic
   numbers.  The currently supported compression formats are gzip
   (magic numbers 1F 8B or 1F 9E), bzip2 (magic number 42 5A), LZMA
-  (magic number 5D 00), XZ (magic number FD 37), and LZ4 (magic number
-  02 21).  The uncompressed payload is currently always ELF (magic
-  number 7F 45 4C 46).
+  (magic number 5D 00), XZ (magic number FD 37), LZ4 (magic number
+  02 21) and ZSTD (magic number 28 B5). The uncompressed payload is
+  currently always ELF (magic number 7F 45 4C 46).
 
 Field name:	payload_length
 Type:		read
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 971feac..e34e3e5 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -150,6 +150,7 @@ config X86
 	select HAVE_KERNEL_LZMA
 	select HAVE_KERNEL_LZO
 	select HAVE_KERNEL_XZ
+	select HAVE_KERNEL_ZSTD
 	select HAVE_KPROBES
 	select HAVE_KPROBES_ON_FTRACE
 	select HAVE_KRETPROBES
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 8a95827..118a423 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -23,7 +23,7 @@ OBJECT_FILES_NON_STANDARD	:= y
 KCOV_INSTRUMENT		:= n
 
 targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma \
-	vmlinux.bin.xz vmlinux.bin.lzo vmlinux.bin.lz4
+	vmlinux.bin.xz vmlinux.bin.lzo vmlinux.bin.lz4 vmlinux.bin.zst
 
 KBUILD_CFLAGS := -m$(BITS) -D__KERNEL__ -O2
 KBUILD_CFLAGS += -fno-strict-aliasing $(call cc-option, -fPIE, -fPIC)
@@ -133,6 +133,8 @@ $(obj)/vmlinux.bin.lzo: $(vmlinux.bin.all-y) FORCE
 	$(call if_changed,lzo)
 $(obj)/vmlinux.bin.lz4: $(vmlinux.bin.all-y) FORCE
 	$(call if_changed,lz4)
+$(obj)/vmlinux.bin.zst: $(vmlinux.bin.all-y) FORCE
+	$(call if_changed,zstd)
 
 suffix-$(CONFIG_KERNEL_GZIP)	:= gz
 suffix-$(CONFIG_KERNEL_BZIP2)	:= bz2
@@ -140,6 +142,7 @@ suffix-$(CONFIG_KERNEL_LZMA)	:= lzma
 suffix-$(CONFIG_KERNEL_XZ)	:= xz
 suffix-$(CONFIG_KERNEL_LZO) 	:= lzo
 suffix-$(CONFIG_KERNEL_LZ4) 	:= lz4
+suffix-$(CONFIG_KERNEL_ZSTD)	:= zst
 
 quiet_cmd_mkpiggy = MKPIGGY $@
       cmd_mkpiggy = $(obj)/mkpiggy $< > $@ || ( rm -f $@ ; false )
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index c14217c..8140895 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -74,6 +74,10 @@ static int lines, cols;
 #ifdef CONFIG_KERNEL_LZ4
 #include "../../../../lib/decompress_unlz4.c"
 #endif
+
+#ifdef CONFIG_KERNEL_ZSTD
+#include "../../../../lib/decompress_unzstd.c"
+#endif
 /*
  * NOTE: When adding a new decompressor, please update the analysis in
  * ../header.S.
diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S
index 1bb08ec..11b70e1 100644
--- a/arch/x86/boot/header.S
+++ b/arch/x86/boot/header.S
@@ -526,8 +526,14 @@ pref_address:		.quad LOAD_PHYSICAL_ADDR	# preferred load addr
 # the size-dependent part now grows so fast.
 #
 # extra_bytes = (uncompressed_size >> 8) + 65536
+#
+# ZSTD compressed data grows by at most 3 bytes per 128K, and only has a 22
+# byte fixed overhead but has a maximum block size of 128K, so it needs a
+# larger margin.
+#
+# extra_bytes = (uncompressed_size >> 8) + 131072
 
-#define ZO_z_extra_bytes	((ZO_z_output_len >> 8) + 65536)
+#define ZO_z_extra_bytes	((ZO_z_output_len >> 8) + 131072)
 #if ZO_z_output_len > ZO_z_input_len
 # define ZO_z_extract_offset	(ZO_z_output_len + ZO_z_extra_bytes - \
 				 ZO_z_input_len)
diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index abd06b1..dcfdd4e 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -23,9 +23,11 @@
 # error "Invalid value for CONFIG_PHYSICAL_ALIGN"
 #endif
 
-#ifdef CONFIG_KERNEL_BZIP2
+#if defined(CONFIG_KERNEL_BZIP2)
 # define BOOT_HEAP_SIZE		0x400000
-#else /* !CONFIG_KERNEL_BZIP2 */
+#elif defined(CONFIG_KERNEL_ZSTD)
+# define BOOT_HEAP_SIZE		 0x30000
+#else
 # define BOOT_HEAP_SIZE		 0x10000
 #endif
 
-- 
2.9.5

Re: [PATCH 0/2] Add support for ZSTD-compressed kernel

[hpa]

On October 10, 2017 2:22:42 PM PDT, Nick Terrell <terrelln@fb.com> wrote:
>Hi,
>
>This patch set adds support for a ZSTD-compressed kernel and ramdisk
>images in the kernel boot process. It only integrates the support with
>x86, though the first patch is generic to all architectures.
>
>Zstandard requires slightly more memory during the kernel decompression
>on x86 (192 KB vs 64 KB), and the memory usage is independent of the
>window size.
>
>Zstandard requires memory proprortional to the window size used during
>compression for decompressing the ramdisk image, since streaming mode
>is
>used. Newer versions of zstd (1.3.2+) list the window size of a file
>with `zstd -lv <file>'. The absolute maximum amount of memory required
>is just over 8 MB.
>
>Thanks,
>Nick Terrell
>
>Nick Terrell (2):
>  lib: Add support for ZSTD-compressed kernel
>  x86: Add support for ZSTD-compressed kernel
>
> Documentation/x86/boot.txt        |   6 +-
> arch/x86/Kconfig                  |   1 +
> arch/x86/boot/compressed/Makefile |   5 +-
> arch/x86/boot/compressed/misc.c   |   4 +
> arch/x86/boot/header.S            |   8 +-
> arch/x86/include/asm/boot.h       |   6 +-
> include/linux/decompress/unzstd.h |  26 +++
> init/Kconfig                      |  14 +-
> lib/Kconfig                       |   4 +
> lib/Makefile                      |   1 +
> lib/decompress.c                  |   5 +
>lib/decompress_unzstd.c           | 341
>++++++++++++++++++++++++++++++++++++++
> lib/xxhash.c                      |  21 ++-
> lib/zstd/decompress.c             |   2 +
> lib/zstd/fse_decompress.c         |   9 +-
> scripts/Makefile.lib              |  15 ++
> usr/Kconfig                       |  22 +++
> 17 files changed, 465 insertions(+), 25 deletions(-)
> create mode 100644 include/linux/decompress/unzstd.h
> create mode 100644 lib/decompress_unzstd.c

And, pray tell, what are the actual results? What is the trade-off of kernel size versus decompression performance versus the other algorithms that we already support? Adding algorithms for their own sake is a bad thing not a good thing.
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Re: [PATCH 0/2] Add support for ZSTD-compressed kernel

[Nick Terrell]

On 10/10/17, 2:56 PM, "hpa@zytor.com" <hpa@zytor.com> wrote:
>On October 10, 2017 2:22:42 PM PDT, Nick Terrell <terrelln@fb.com> wrote:
>>Hi,
>>
>>This patch set adds support for a ZSTD-compressed kernel and ramdisk
>>images in the kernel boot process. It only integrates the support with
>>x86, though the first patch is generic to all architectures.
>>
>>Zstandard requires slightly more memory during the kernel decompression
>>on x86 (192 KB vs 64 KB), and the memory usage is independent of the
>>window size.
>>
>>Zstandard requires memory proprortional to the window size used during
>>compression for decompressing the ramdisk image, since streaming mode
>>is
>>used. Newer versions of zstd (1.3.2+) list the window size of a file
>>with `zstd -lv <file>'. The absolute maximum amount of memory required
>>is just over 8 MB.
>>
>>Thanks,
>>Nick Terrell
>>
>>Nick Terrell (2):
>>  lib: Add support for ZSTD-compressed kernel
>>  x86: Add support for ZSTD-compressed kernel
>>
>> Documentation/x86/boot.txt        |   6 +-
>> arch/x86/Kconfig                  |   1 +
>> arch/x86/boot/compressed/Makefile |   5 +-
>> arch/x86/boot/compressed/misc.c   |   4 +
>> arch/x86/boot/header.S            |   8 +-
>> arch/x86/include/asm/boot.h       |   6 +-
>> include/linux/decompress/unzstd.h |  26 +++
>> init/Kconfig                      |  14 +-
>> lib/Kconfig                       |   4 +
>> lib/Makefile                      |   1 +
>> lib/decompress.c                  |   5 +
>>lib/decompress_unzstd.c           | 341
>>++++++++++++++++++++++++++++++++++++++
>> lib/xxhash.c                      |  21 ++-
>> lib/zstd/decompress.c             |   2 +
>> lib/zstd/fse_decompress.c         |   9 +-
>> scripts/Makefile.lib              |  15 ++
>> usr/Kconfig                       |  22 +++
>> 17 files changed, 465 insertions(+), 25 deletions(-)
>> create mode 100644 include/linux/decompress/unzstd.h
>> create mode 100644 lib/decompress_unzstd.c
>
>And, pray tell, what are the actual results? What is the trade-off of kernel size versus decompression performance versus the other algorithms that we already support? Adding algorithms for their own sake is a bad thing not a good thing.

Sorry I neglected to include the benchmarks I've run so far. I've included
them below, and will add them to the next version's cover letter.

Comparing the command line tools on a kernel image that is 68970616 B large:

| Algorithm | Compression Ratio | Decompression MB/s |
|-----------|-------------------|--------------------|
| zstd      |              4.42 |              436.5 |
| gzip      |              3.72 |              134.1 |
| xz        |              4.83 |               53.1 |
| lz4       |              3.18 |             1682.2 |
| lzo       |              3.36 |              389.6 |
| bzip2     |              4.03 |               33.3 |

This isn't actually benchmarks of the kernel decompression, but I expect
the relative results to be the same. It compresses the second best and
decompresses the second fastest.

I know that this isn't a real benchmark of the kernel decompression. I
still need to figure out how to time the kernel decompression. If you have
any suggestions let me know. Otherwise, I'll get back to you when I've
figured out how to run the benchmark.

Best,
Nick Terrell

Re: [PATCH 0/2] Add support for ZSTD-compressed kernel

[Adam Borowski]

On Tue, Oct 10, 2017 at 10:40:13PM +0000, Nick Terrell wrote:
> On 10/10/17, 2:56 PM, "hpa@zytor.com" <hpa@zytor.com> wrote:
> >On October 10, 2017 2:22:42 PM PDT, Nick Terrell <terrelln@fb.com> wrote:
> >>This patch set adds support for a ZSTD-compressed kernel and ramdisk
> >>images in the kernel boot process. It only integrates the support with
> >>x86, though the first patch is generic to all architectures.
> >>
> >>Zstandard requires slightly more memory during the kernel decompression
> >>on x86 (192 KB vs 64 KB), and the memory usage is independent of the
> >>window size.
> >>
> >>Zstandard requires memory proprortional to the window size used during
> >>compression for decompressing the ramdisk image, since streaming mode
> >>is
> >>used. Newer versions of zstd (1.3.2+) list the window size of a file
> >>with `zstd -lv <file>'. The absolute maximum amount of memory required
> >>is just over 8 MB.

> > And, pray tell, what are the actual results?  What is the trade-off of
> > kernel size versus decompression performance versus the other algorithms
> > that we already support?  Adding algorithms for their own sake is a bad
> > thing not a good thing.
> 
> Sorry I neglected to include the benchmarks I've run so far. I've included
> them below, and will add them to the next version's cover letter.
> 
> Comparing the command line tools on a kernel image that is 68970616 B large:
> 
> | Algorithm | Compression Ratio | Decompression MB/s |
> |-----------|-------------------|--------------------|
> | zstd      |              4.42 |              436.5 |
> | gzip      |              3.72 |              134.1 |
> | xz        |              4.83 |               53.1 |
> | lz4       |              3.18 |             1682.2 |
> | lzo       |              3.36 |              389.6 |
> | bzip2     |              4.03 |               33.3 |

Perhaps it'd be a good idea to cull some of bad algorithms?  I don't know
the memory used by those, but envelope of the table you just shown suggests
using bzip2 and lzo is pointless.  So is gzip, but it's widespread as the
default for initramfs producers, thus it'd be unsafe to kill it.


Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀ We domesticated dogs 36000 years ago; together we chased
⣾⠁⢰⠒⠀⣿⡁ animals, hung out and licked or scratched our private parts.
⢿⡄⠘⠷⠚⠋⠀ Cats domesticated us 9500 years ago, and immediately we got
⠈⠳⣄⠀⠀⠀⠀ agriculture, towns then cities.     -- whitroth on /.

Re: [PATCH 0/2] Add support for ZSTD-compressed kernel

[Nick Terrell]

On 10/10/17, 5:08 PM, "Adam Borowski" <kilobyte@angband.pl> wrote:
> On Tue, Oct 10, 2017 at 10:40:13PM +0000, Nick Terrell wrote:
> > On 10/10/17, 2:56 PM, "hpa@zytor.com" <hpa@zytor.com> wrote:
> > >On October 10, 2017 2:22:42 PM PDT, Nick Terrell <terrelln@fb.com> wrote:
> > >>This patch set adds support for a ZSTD-compressed kernel and ramdisk
> > >>images in the kernel boot process. It only integrates the support with
> > >>x86, though the first patch is generic to all architectures.
> > >>
> > >>Zstandard requires slightly more memory during the kernel decompression
> > >>on x86 (192 KB vs 64 KB), and the memory usage is independent of the
> > >>window size.
> > >>
> > >>Zstandard requires memory proprortional to the window size used during
> > >>compression for decompressing the ramdisk image, since streaming mode
> > >>is
> > >>used. Newer versions of zstd (1.3.2+) list the window size of a file
> > >>with `zstd -lv <file>'. The absolute maximum amount of memory required
> > >>is just over 8 MB.

> > > And, pray tell, what are the actual results?  What is the trade-off of
> > > kernel size versus decompression performance versus the other algorithms
> > > that we already support?  Adding algorithms for their own sake is a bad
> > > thing not a good thing.
> >
> > Sorry I neglected to include the benchmarks I've run so far. I've included
> > them below, and will add them to the next version's cover letter.
> >
> > Comparing the command line tools on a kernel image that is 68970616 B large:
> >
> > | Algorithm | Compression Ratio | Decompression MB/s |
> > |-----------|-------------------|--------------------|
> > | zstd      |              4.42 |              436.5 |
> > | gzip      |              3.72 |              134.1 |
> > | xz        |              4.83 |               53.1 |
> > | lz4       |              3.18 |             1682.2 |
> > | lzo       |              3.36 |              389.6 |
> > | bzip2     |              4.03 |               33.3 |

> Perhaps it'd be a good idea to cull some of bad algorithms?  I don't know
> the memory used by those, but envelope of the table you just shown suggests
> using bzip2 and lzo is pointless.  So is gzip, but it's widespread as the
> default for initramfs producers, thus it'd be unsafe to kill it.

I'm not sure there is a great use case for bzip2. It requires more memory
than xz, compresses worse, and decompresses slower. lzo in the kernel might
decompress a bit faster than zstd (looking back at the BtrFS benchmarks, it
did). More importantly, it uses less memory than zstd. When decompressing
the kernel zstd only needs 192 KB, but for initramfs, it will need more.
Still, unless you really need 5% more compression, lz4 is probably a better
option than lzo for speed.

> > I know that this isn't a real benchmark of the kernel decompression. I
> > still need to figure out how to time the kernel decompression. If you have
> > any suggestions let me know. Otherwise, I'll get back to you when I've
> > figured out how to run the benchmark.

I've found a way to benchmark the kernel decompression time during boot
with QEMU. I add timestamps to every line of the output. I also had to
print 100 lines before the decompression starts to get consistent results.

I've found that zstd is decompressing 2x slower than it should. I narrowed
down the problem to ZSTD_wildcopy() and ZSTD_copy8() in
lib/zstd/zstd_internal.h. ZSTD_wildcopy() calls memcpy(src, dst, 8) in
a loop and doesn't handle the freestanding memcpy() well. Replacing it with
__builtin_mcmpy(src, dst, 8) doubles the speed.

I'm not an expert in freestanding gcc compilation, but I believe it is okay
to call __builtin_memcpy() in freestanding mode, and gcc will either
inline it, or add the right function call. The difference being that gcc
will be able to apply its memcpy() analysis. I also see that
arch/x86/boot/string.h defines memcpy() to __builtin_memcpy. Is it safe to
directly use __builtin_memcpy() in lib/zstd/zstd_internal.h?

If so, I'll submit a separate patch, and make sure to benchmark the
existing use cases (BtrFS and SquashFS).

Re: [PATCH 0/2] Add support for ZSTD-compressed kernel

[Adam Borowski]

On Wed, Oct 11, 2017 at 02:01:41AM +0000, Nick Terrell wrote:
> On 10/10/17, 5:08 PM, "Adam Borowski" <kilobyte@angband.pl> wrote:
> > On Tue, Oct 10, 2017 at 10:40:13PM +0000, Nick Terrell wrote:
> > > On 10/10/17, 2:56 PM, "hpa@zytor.com" <hpa@zytor.com> wrote:
> > > >On October 10, 2017 2:22:42 PM PDT, Nick Terrell <terrelln@fb.com> wrote:
> > > >>This patch set adds support for a ZSTD-compressed kernel and ramdisk
> > > >>images in the kernel boot process. It only integrates the support with
> > > >>x86, though the first patch is generic to all architectures.
> 
> > > Comparing the command line tools on a kernel image that is 68970616 B large:
> > >
> > > | Algorithm | Compression Ratio | Decompression MB/s |
> > > |-----------|-------------------|--------------------|
> > > | zstd      |              4.42 |              436.5 |
> > > | gzip      |              3.72 |              134.1 |
> > > | xz        |              4.83 |               53.1 |
> > > | lz4       |              3.18 |             1682.2 |
> > > | lzo       |              3.36 |              389.6 |
> > > | bzip2     |              4.03 |               33.3 |
> 
> > Perhaps it'd be a good idea to cull some of bad algorithms?  I don't know
> > the memory used by those, but envelope of the table you just shown suggests
> > using bzip2 and lzo is pointless.  So is gzip, but it's widespread as the
> > default for initramfs producers, thus it'd be unsafe to kill it.
> 
> I'm not sure there is a great use case for bzip2. It requires more memory
> than xz, compresses worse, and decompresses slower. lzo in the kernel might
> decompress a bit faster than zstd (looking back at the BtrFS benchmarks, it
> did). More importantly, it uses less memory than zstd. When decompressing
> the kernel zstd only needs 192 KB, but for initramfs, it will need more.
> Still, unless you really need 5% more compression, lz4 is probably a better
> option than lzo for speed.

The main reason I'm raising this question is, bzip2 has no other users in
the kernel, thus removing support for it would allow us to delete its code.

As for lzo, even if there are cases when it's a bit faster (overcoming the
9% lead zstd has in your userspace benchmark), it wouldn't be faster by
much -- certainly nowhere to make anyone want the massive compression loss.

Thus, let's enumerate use cases:
* a fast modern machine: zstd boots a second faster at the cost of 1.3MB
  image size; doesn't matter much either way.  No one wants weaker options.
* a weaker machine: zstd rapidly gains as boot times rise.
* a very slow machine with fast I/O that boots often: lz4.
* disk space at extreme premium: xz.

In no case I can think of other algorithms are the rational choice.
There's little gain in removing the rest, though: the code is used for other
purposes in the kernel, and doesn't get compiled in unless manually
selected.

Zstd looks like a good default, but it's nowhere near mature enough: this
patch is x86-only, tools which people may use to analyze compressed kernels
don't know about zstd yet, etc.

Thus, I'd propose the following plan:
* add zstd
* remove bzip2
* update recommendations (Kconfig text)
* in a year or two, consider making zstd the default

> > > I know that this isn't a real benchmark of the kernel decompression. I
> > > still need to figure out how to time the kernel decompression. If you have
> > > any suggestions let me know. Otherwise, I'll get back to you when I've
> > > figured out how to run the benchmark.
> 
> I've found a way to benchmark the kernel decompression time during boot
> with QEMU. I add timestamps to every line of the output. I also had to
> print 100 lines before the decompression starts to get consistent results.
> 
> I've found that zstd is decompressing 2x slower than it should. I narrowed
> down the problem to ZSTD_wildcopy() and ZSTD_copy8() in
> lib/zstd/zstd_internal.h. ZSTD_wildcopy() calls memcpy(src, dst, 8) in
> a loop and doesn't handle the freestanding memcpy() well. Replacing it with
> __builtin_mcmpy(src, dst, 8) doubles the speed.

... and by maturing I meant issues like this.

> I'm not an expert in freestanding gcc compilation, but I believe it is okay
> to call __builtin_memcpy() in freestanding mode, and gcc will either
> inline it, or add the right function call. The difference being that gcc
> will be able to apply its memcpy() analysis. I also see that
> arch/x86/boot/string.h defines memcpy() to __builtin_memcpy. Is it safe to
> directly use __builtin_memcpy() in lib/zstd/zstd_internal.h?

Try it, and see if it builds and fails to crash. :)


Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀ We domesticated dogs 36000 years ago; together we chased
⣾⠁⢰⠒⠀⣿⡁ animals, hung out and licked or scratched our private parts.
⢿⡄⠘⠷⠚⠋⠀ Cats domesticated us 9500 years ago, and immediately we got
⠈⠳⣄⠀⠀⠀⠀ agriculture, towns then cities.     -- whitroth on /.

[PATCH 1/2] lib: add support for ZSTD-compressed kernel

[Petr Malat]

Add support for extracting ZSTD-compressed kernel images, as well as
ZSTD-compressed initramfs.

Signed-off-by: Petr Malat <oss@malat.biz>
---
 include/linux/decompress/unzstd.h |  12 +++
 init/Kconfig                      |  15 ++-
 lib/Kconfig                       |   4 +
 lib/Makefile                      |   1 +
 lib/decompress.c                  |   5 +
 lib/decompress_unzstd.c           | 159 ++++++++++++++++++++++++++++++
 lib/zstd/decompress.c             |   2 +
 lib/zstd/fse_decompress.c         |   4 +-
 scripts/Makefile.lib              |   3 +
 usr/Kconfig                       |  24 +++++
 10 files changed, 227 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/decompress/unzstd.h
 create mode 100644 lib/decompress_unzstd.c

diff --git a/include/linux/decompress/unzstd.h b/include/linux/decompress/unzstd.h
new file mode 100644
index 000000000000..dd2c49d47456
--- /dev/null
+++ b/include/linux/decompress/unzstd.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef DECOMPRESS_UNZSTD_H
+#define DECOMPRESS_UNZSTD_H
+
+int unzstd(unsigned char *inbuf, long len,
+	long (*fill)(void*, unsigned long),
+	long (*flush)(void*, unsigned long),
+	unsigned char *output,
+	long *pos,
+	void (*error)(char *x));
+#endif
+
diff --git a/init/Kconfig b/init/Kconfig
index a34064a031a5..628eb3c290a2 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -172,13 +172,16 @@ config HAVE_KERNEL_LZO
 config HAVE_KERNEL_LZ4
 	bool
 
+config HAVE_KERNEL_ZSTD
+	bool
+
 config HAVE_KERNEL_UNCOMPRESSED
 	bool
 
 choice
 	prompt "Kernel compression mode"
 	default KERNEL_GZIP
-	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_UNCOMPRESSED
+	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_ZSTD || HAVE_KERNEL_UNCOMPRESSED
 	help
 	  The linux kernel is a kind of self-extracting executable.
 	  Several compression algorithms are available, which differ
@@ -257,6 +260,16 @@ config KERNEL_LZ4
 	  is about 8% bigger than LZO. But the decompression speed is
 	  faster than LZO.
 
+config KERNEL_ZSTD
+	bool "ZSTD"
+	depends on HAVE_KERNEL_ZSTD
+	help
+	  Its compression ratio is roughly 10% worst than xz, but the
+	  decompression is 10x faster. Currently, this is one of the optimal
+	  algorithms available in the kernel, as there isn't an algorithm,
+	  which would provide a better compression ratio and a shorter
+	  decompression time.
+
 config KERNEL_UNCOMPRESSED
 	bool "None"
 	depends on HAVE_KERNEL_UNCOMPRESSED
diff --git a/lib/Kconfig b/lib/Kconfig
index 6e790dc55c5b..df301bd888d7 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -329,6 +329,10 @@ config DECOMPRESS_LZ4
 	select LZ4_DECOMPRESS
 	tristate
 
+config DECOMPRESS_ZSTD
+	select ZSTD_DECOMPRESS
+	tristate
+
 #
 # Generic allocator support is selected if needed
 #
diff --git a/lib/Makefile b/lib/Makefile
index 93217d44237f..3ab9f4c31f8b 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -158,6 +158,7 @@ lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
 lib-$(CONFIG_DECOMPRESS_XZ) += decompress_unxz.o
 lib-$(CONFIG_DECOMPRESS_LZO) += decompress_unlzo.o
 lib-$(CONFIG_DECOMPRESS_LZ4) += decompress_unlz4.o
+lib-$(CONFIG_DECOMPRESS_ZSTD) += decompress_unzstd.o
 
 obj-$(CONFIG_TEXTSEARCH) += textsearch.o
 obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
diff --git a/lib/decompress.c b/lib/decompress.c
index 857ab1af1ef3..ab3fc90ffc64 100644
--- a/lib/decompress.c
+++ b/lib/decompress.c
@@ -13,6 +13,7 @@
 #include <linux/decompress/inflate.h>
 #include <linux/decompress/unlzo.h>
 #include <linux/decompress/unlz4.h>
+#include <linux/decompress/unzstd.h>
 
 #include <linux/types.h>
 #include <linux/string.h>
@@ -37,6 +38,9 @@
 #ifndef CONFIG_DECOMPRESS_LZ4
 # define unlz4 NULL
 #endif
+#ifndef CONFIG_DECOMPRESS_ZSTD
+# define unzstd NULL
+#endif
 
 struct compress_format {
 	unsigned char magic[2];
@@ -52,6 +56,7 @@ static const struct compress_format compressed_formats[] __initconst = {
 	{ {0xfd, 0x37}, "xz", unxz },
 	{ {0x89, 0x4c}, "lzo", unlzo },
 	{ {0x02, 0x21}, "lz4", unlz4 },
+	{ {0x28, 0xb5}, "zstd", unzstd },
 	{ {0, 0}, NULL, NULL }
 };
 
diff --git a/lib/decompress_unzstd.c b/lib/decompress_unzstd.c
new file mode 100644
index 000000000000..5af647a49885
--- /dev/null
+++ b/lib/decompress_unzstd.c
@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Wrapper for decompressing ZSTD-compressed kernel, initramfs, and initrd
+ * Based on decompress_unlz4.c
+ *
+ * Copyright (C) 2020, Petr Malat <oss@malat.biz>
+ */
+
+#ifdef STATIC
+#define PREBOOT
+#include "zstd/zstd_internal.h"
+#include "zstd/huf_decompress.c"
+#include "zstd/entropy_common.c"
+#include "zstd/fse_decompress.c"
+#include "zstd/zstd_common.c"
+#include "zstd/decompress.c"
+#include "xxhash.c"
+#else
+#include <linux/decompress/unzstd.h>
+#include <linux/zstd.h>
+#endif
+#include <linux/types.h>
+#include <linux/decompress/mm.h>
+#include <linux/compiler.h>
+
+STATIC inline int INIT unzstd(u8 *input, long in_len,
+				long (*fill)(void *, unsigned long),
+				long (*flush)(void *, unsigned long),
+				u8 *output, long *posp,
+				void (*error)(char *x))
+{
+	int ret = -1, ws = 1 << ZSTD_WINDOWLOG_MAX;
+	u8 *inp, *outp;
+	ZSTD_DStream *zstd;
+	void *workspace;
+	size_t workspace_size;
+	ZSTD_outBuffer out;
+	ZSTD_inBuffer in;
+	unsigned long out_len;
+	unsigned long pos;
+
+	if (output) {
+		out_len = ULONG_MAX; // Caller knows data will fit
+		outp = output;
+	} else if (!flush) {
+		error("NULL output pointer and no flush function provided");
+		goto exit_0;
+	} else {
+		out_len = ZSTD_DStreamOutSize();
+		outp = large_malloc(out_len);
+		if (!outp) {
+			error("Could not allocate output buffer");
+			goto exit_0;
+		}
+	}
+
+	if (input && fill) {
+		error("Both input pointer and fill function provided,");
+		goto exit_1;
+	} else if (input) {
+		ZSTD_frameParams p;
+
+		inp = input;
+		if (!ZSTD_getFrameParams(&p, input, in_len))
+			ws = p.windowSize;
+	} else if (!fill) {
+		error("NULL input pointer and missing fill function");
+		goto exit_1;
+	} else {
+		in_len = ZSTD_DStreamInSize();
+		inp = large_malloc(in_len);
+		if (!inp) {
+			error("Could not allocate input buffer");
+			goto exit_1;
+		}
+	}
+
+	workspace_size = ZSTD_DStreamWorkspaceBound(ws);
+	workspace = large_malloc(workspace_size);
+	if (!workspace) {
+		error("Could not allocate workspace");
+		goto exit_2;
+	}
+
+	zstd = ZSTD_initDStream(ws, workspace, workspace_size);
+	if (!zstd) {
+		error("Could not initialize ZSTD");
+		goto exit_3;
+	}
+
+	in.src = inp;
+	in.size = in_len;
+	in.pos = 0;
+	if (posp)
+		*posp = 0;
+
+	for (;;) {
+		if (fill) {
+			in.size = fill(inp, in_len);
+			if (in.size == 0)
+				break;
+		} else if (in.size == in.pos) {
+			break;
+		}
+init:		out.dst = outp;
+		out.size = out_len;
+		out.pos = 0;
+		pos = in.pos;
+
+		ret = ZSTD_decompressStream(zstd, &out, &in);
+		if (posp)
+			*posp += in.pos - pos;
+		if (ZSTD_isError(ret)) {
+			error("Decompression failed");
+			ret = -EIO;
+			goto exit_3;
+		}
+
+		if (flush && out.pos) {
+			if (flush(out.dst, out.pos) != out.pos) {
+				ret = -EIO;
+				goto exit_3;
+			}
+			goto init;
+		}
+
+		if (ret == 0) {
+			ret = ZSTD_resetDStream(zstd);
+			if (ZSTD_isError(ret)) {
+				ret = -EIO;
+				goto exit_3;
+			}
+		}
+		if (in.pos < in.size)
+			goto init_out;
+	}
+
+	ret = 0;
+
+exit_3:	large_free(workspace);
+exit_2:	if (!input)
+		large_free(inp);
+exit_1:	if (!output)
+		large_free(outp);
+exit_0:	return ret;
+}
+
+#ifdef PREBOOT
+STATIC int INIT __decompress(unsigned char *buf, long in_len,
+			      long (*fill)(void*, unsigned long),
+			      long (*flush)(void*, unsigned long),
+			      unsigned char *output, long out_len,
+			      long *posp,
+			      void (*error)(char *x)
+	)
+{
+	return unzstd(buf, in_len, fill, flush, output, posp, error);
+}
+#endif
diff --git a/lib/zstd/decompress.c b/lib/zstd/decompress.c
index 269ee9a796c1..6a5e1ce22719 100644
--- a/lib/zstd/decompress.c
+++ b/lib/zstd/decompress.c
@@ -42,9 +42,11 @@
 /*-*************************************
 *  Macros
 ***************************************/
+#ifndef PREBOOT
 #define ZSTD_isError ERR_isError /* for inlining */
 #define FSE_isError ERR_isError
 #define HUF_isError ERR_isError
+#endif
 
 /*_*******************************************************
 *  Memory operations
diff --git a/lib/zstd/fse_decompress.c b/lib/zstd/fse_decompress.c
index a84300e5a013..bd4e9c891d96 100644
--- a/lib/zstd/fse_decompress.c
+++ b/lib/zstd/fse_decompress.c
@@ -54,12 +54,13 @@
 /* **************************************************************
 *  Error Management
 ****************************************************************/
-#define FSE_isError ERR_isError
 #define FSE_STATIC_ASSERT(c)                                   \
 	{                                                      \
 		enum { FSE_static_assert = 1 / (int)(!!(c)) }; \
 	} /* use only *after* variable declarations */
 
+#ifndef PREBOOT
+#define FSE_isError ERR_isError
 /* check and forward error code */
 #define CHECK_F(f)                  \
 	{                           \
@@ -67,6 +68,7 @@
 		if (FSE_isError(e)) \
 			return e;   \
 	}
+#endif
 
 /* **************************************************************
 *  Templates
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 3fa32f83b2d7..1c2f2dc528dc 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -337,6 +337,9 @@ quiet_cmd_lz4 = LZ4     $@
       cmd_lz4 = { cat $(real-prereqs) | lz4c -l -c1 stdin stdout; \
                   $(size_append); } > $@
 
+quiet_cmd_zstd = ZSTD    $@
+      cmd_zstd = { cat $(real-prereqs) | zstd -19 --zstd=wlog=21; $(size_append); } > $@
+
 # U-Boot mkimage
 # ---------------------------------------------------------------------------
 
diff --git a/usr/Kconfig b/usr/Kconfig
index a6b68503d177..892eb15957db 100644
--- a/usr/Kconfig
+++ b/usr/Kconfig
@@ -106,6 +106,15 @@ config RD_LZ4
 	  Support loading of a LZ4 encoded initial ramdisk or cpio buffer
 	  If unsure, say N.
 
+config RD_ZSTD
+	bool "Support initial ramdisk/ramfs compressed using ZSTD"
+	default y
+	depends on BLK_DEV_INITRD
+	select DECOMPRESS_ZSTD
+	help
+	  Support loading of a ZSTD encoded initial ramdisk or cpio buffer
+	  If unsure, say N.
+
 choice
 	prompt "Built-in initramfs compression mode"
 	depends on INITRAMFS_SOURCE!=""
@@ -214,6 +223,19 @@ config INITRAMFS_COMPRESSION_LZ4
 	  If you choose this, keep in mind that most distros don't provide lz4
 	  by default which could cause a build failure.
 
+config INITRAMFS_COMPRESSION_ZSTD
+	bool "ZSTD"
+	depends on RD_ZSTD
+	help
+	  Its compression ratio is roughly 10% worst than xz, but the
+	  decompression is 10x faster. Currently, this is one of the optimal
+	  algorithms available in the kernel, as there isn't an algorithm,
+	  which would provide a better compression ratio and a shorter
+	  decompression time.
+
+	  If you choose this, keep in mind that you may need to install the zstd
+	  tool to be able to compress the initram.
+
 endchoice
 
 config INITRAMFS_COMPRESSION
@@ -226,10 +248,12 @@ config INITRAMFS_COMPRESSION
 	default ".xz"   if INITRAMFS_COMPRESSION_XZ
 	default ".lzo"  if INITRAMFS_COMPRESSION_LZO
 	default ".lz4"  if INITRAMFS_COMPRESSION_LZ4
+	default ".zst"  if INITRAMFS_COMPRESSION_ZSTD
 	default ".gz"   if RD_GZIP
 	default ".lz4"  if RD_LZ4
 	default ".lzo"  if RD_LZO
 	default ".xz"   if RD_XZ
 	default ".lzma" if RD_LZMA
 	default ".bz2"  if RD_BZIP2
+	default ".zst"  if RD_ZSTD
 	default ""
-- 
2.20.1

Re: [PATCH 1/2] lib: add support for ZSTD-compressed kernel

[Greg KH]

On Mon, Mar 16, 2020 at 02:50:24PM +0100, Petr Malat wrote:
> Add support for extracting ZSTD-compressed kernel images, as well as
> ZSTD-compressed initramfs.
> 
> Signed-off-by: Petr Malat <oss@malat.biz>

That says _what_ you did here, but not _why_ you did this, or why anyone
would even want this.

thanks,

greg k-h

Re: [PATCH 1/2] lib: add support for ZSTD-compressed kernel

[Petr Malat]

I have extended the description, thx.
  Petr

On Mon, Mar 16, 2020 at 03:07:26PM +0100, Greg KH wrote:
> On Mon, Mar 16, 2020 at 02:50:24PM +0100, Petr Malat wrote:
> > Add support for extracting ZSTD-compressed kernel images, as well as
> > ZSTD-compressed initramfs.
> > 
> > Signed-off-by: Petr Malat <oss@malat.biz>
> 
> That says _what_ you did here, but not _why_ you did this, or why anyone
> would even want this.
> 
> thanks,
> 
> greg k-h

Re: [PATCH 1/2] lib: add support for ZSTD-compressed kernel

[Nick Terrell]

> On Mar 16, 2020, at 6:50 AM, Petr Malat <oss@malat.biz> wrote:
> 
> Add support for extracting ZSTD-compressed kernel images, as well as
> ZSTD-compressed initramfs.

Hi Petr,

Thanks for putting up this patch!

In the last month I rebased, retested, and refactored my patches [0][1], because
we had a new use case for a zstd compressed initramfs. We have been using
the patches I put up for quite some time internally, so we know they work well.

I will re-submit my patches today, once I write all the commit summaries and a
new cover letter. Thanks for the kick in the butt to resubmit it.

Your patches are missing several things that I have included in my patches [0][1]:
* Invocations of EXPORT_SYMBOL() and MODULE_LICENSE() in the pre-boot
  either emitted warnings or didn’t work in the pre-boot environment, unless that
  has changed.
* The memcpy() inside of ZSTD_copy8() which is the core of zstd’s hot loop gets
  outlined in the x86 preboot environment unless you use __builtin_memcpy().
  This destroys decompression speed.
* unzstd() can use less memory when fill & flush are both NULL by calling
  ZSTD_decompressDCtx().
* ZO_z_extra_bytes needs to be bumped because zstd can overlap more than
  Other compressors. If it isn’t you could corrupt the kernel.

I think it would be better to go with my patches. They have already been tested
in production for both correctness and performance, and I have boot tested on
x86, arm, and aarch64. I will make my case fully when I resubmit my patches.

Thanks again for submitting this, because I have been dawdling!
Nick

[0] https://lore.kernel.org/patchwork/patch/839674/
[1] https://lore.kernel.org/patchwork/patch/839675/

> Signed-off-by: Petr Malat <oss@malat.biz>
> ---
> include/linux/decompress/unzstd.h |  12 +++
> init/Kconfig                      |  15 ++-
> lib/Kconfig                       |   4 +
> lib/Makefile                      |   1 +
> lib/decompress.c                  |   5 +
> lib/decompress_unzstd.c           | 159 ++++++++++++++++++++++++++++++
> lib/zstd/decompress.c             |   2 +
> lib/zstd/fse_decompress.c         |   4 +-
> scripts/Makefile.lib              |   3 +
> usr/Kconfig                       |  24 +++++
> 10 files changed, 227 insertions(+), 2 deletions(-)
> create mode 100644 include/linux/decompress/unzstd.h
> create mode 100644 lib/decompress_unzstd.c
> 
> diff --git a/include/linux/decompress/unzstd.h b/include/linux/decompress/unzstd.h
> new file mode 100644
> index 000000000000..dd2c49d47456
> --- /dev/null
> +++ b/include/linux/decompress/unzstd.h
> @@ -0,0 +1,12 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef DECOMPRESS_UNZSTD_H
> +#define DECOMPRESS_UNZSTD_H
> +
> +int unzstd(unsigned char *inbuf, long len,
> +	long (*fill)(void*, unsigned long),
> +	long (*flush)(void*, unsigned long),
> +	unsigned char *output,
> +	long *pos,
> +	void (*error)(char *x));
> +#endif
> +
> diff --git a/init/Kconfig b/init/Kconfig
> index a34064a031a5..628eb3c290a2 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -172,13 +172,16 @@ config HAVE_KERNEL_LZO
> config HAVE_KERNEL_LZ4
> 	bool
> 
> +config HAVE_KERNEL_ZSTD
> +	bool
> +
> config HAVE_KERNEL_UNCOMPRESSED
> 	bool
> 
> choice
> 	prompt "Kernel compression mode"
> 	default KERNEL_GZIP
> -	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_UNCOMPRESSED
> +	depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_ZSTD || HAVE_KERNEL_UNCOMPRESSED
> 	help
> 	  The linux kernel is a kind of self-extracting executable.
> 	  Several compression algorithms are available, which differ
> @@ -257,6 +260,16 @@ config KERNEL_LZ4
> 	  is about 8% bigger than LZO. But the decompression speed is
> 	  faster than LZO.
> 
> +config KERNEL_ZSTD
> +	bool "ZSTD"
> +	depends on HAVE_KERNEL_ZSTD
> +	help
> +	  Its compression ratio is roughly 10% worst than xz, but the
> +	  decompression is 10x faster. Currently, this is one of the optimal
> +	  algorithms available in the kernel, as there isn't an algorithm,
> +	  which would provide a better compression ratio and a shorter
> +	  decompression time.
> +
> config KERNEL_UNCOMPRESSED
> 	bool "None"
> 	depends on HAVE_KERNEL_UNCOMPRESSED
> diff --git a/lib/Kconfig b/lib/Kconfig
> index 6e790dc55c5b..df301bd888d7 100644
> --- a/lib/Kconfig
> +++ b/lib/Kconfig
> @@ -329,6 +329,10 @@ config DECOMPRESS_LZ4
> 	select LZ4_DECOMPRESS
> 	tristate
> 
> +config DECOMPRESS_ZSTD
> +	select ZSTD_DECOMPRESS
> +	tristate
> +
> #
> # Generic allocator support is selected if needed
> #
> diff --git a/lib/Makefile b/lib/Makefile
> index 93217d44237f..3ab9f4c31f8b 100644
> --- a/lib/Makefile
> +++ b/lib/Makefile
> @@ -158,6 +158,7 @@ lib-$(CONFIG_DECOMPRESS_LZMA) += decompress_unlzma.o
> lib-$(CONFIG_DECOMPRESS_XZ) += decompress_unxz.o
> lib-$(CONFIG_DECOMPRESS_LZO) += decompress_unlzo.o
> lib-$(CONFIG_DECOMPRESS_LZ4) += decompress_unlz4.o
> +lib-$(CONFIG_DECOMPRESS_ZSTD) += decompress_unzstd.o
> 
> obj-$(CONFIG_TEXTSEARCH) += textsearch.o
> obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o
> diff --git a/lib/decompress.c b/lib/decompress.c
> index 857ab1af1ef3..ab3fc90ffc64 100644
> --- a/lib/decompress.c
> +++ b/lib/decompress.c
> @@ -13,6 +13,7 @@
> #include <linux/decompress/inflate.h>
> #include <linux/decompress/unlzo.h>
> #include <linux/decompress/unlz4.h>
> +#include <linux/decompress/unzstd.h>
> 
> #include <linux/types.h>
> #include <linux/string.h>
> @@ -37,6 +38,9 @@
> #ifndef CONFIG_DECOMPRESS_LZ4
> # define unlz4 NULL
> #endif
> +#ifndef CONFIG_DECOMPRESS_ZSTD
> +# define unzstd NULL
> +#endif
> 
> struct compress_format {
> 	unsigned char magic[2];
> @@ -52,6 +56,7 @@ static const struct compress_format compressed_formats[] __initconst = {
> 	{ {0xfd, 0x37}, "xz", unxz },
> 	{ {0x89, 0x4c}, "lzo", unlzo },
> 	{ {0x02, 0x21}, "lz4", unlz4 },
> +	{ {0x28, 0xb5}, "zstd", unzstd },
> 	{ {0, 0}, NULL, NULL }
> };
> 
> diff --git a/lib/decompress_unzstd.c b/lib/decompress_unzstd.c
> new file mode 100644
> index 000000000000..5af647a49885
> --- /dev/null
> +++ b/lib/decompress_unzstd.c
> @@ -0,0 +1,159 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Wrapper for decompressing ZSTD-compressed kernel, initramfs, and initrd
> + * Based on decompress_unlz4.c
> + *
> + * Copyright (C) 2020, Petr Malat <oss@malat.biz>
> + */
> +
> +#ifdef STATIC
> +#define PREBOOT
> +#include "zstd/zstd_internal.h"
> +#include "zstd/huf_decompress.c"
> +#include "zstd/entropy_common.c"
> +#include "zstd/fse_decompress.c"
> +#include "zstd/zstd_common.c"
> +#include "zstd/decompress.c"
> +#include "xxhash.c"
> +#else
> +#include <linux/decompress/unzstd.h>
> +#include <linux/zstd.h>
> +#endif
> +#include <linux/types.h>
> +#include <linux/decompress/mm.h>
> +#include <linux/compiler.h>
> +
> +STATIC inline int INIT unzstd(u8 *input, long in_len,
> +				long (*fill)(void *, unsigned long),
> +				long (*flush)(void *, unsigned long),
> +				u8 *output, long *posp,
> +				void (*error)(char *x))
> +{
> +	int ret = -1, ws = 1 << ZSTD_WINDOWLOG_MAX;
> +	u8 *inp, *outp;
> +	ZSTD_DStream *zstd;
> +	void *workspace;
> +	size_t workspace_size;
> +	ZSTD_outBuffer out;
> +	ZSTD_inBuffer in;
> +	unsigned long out_len;
> +	unsigned long pos;
> +
> +	if (output) {
> +		out_len = ULONG_MAX; // Caller knows data will fit
> +		outp = output;
> +	} else if (!flush) {
> +		error("NULL output pointer and no flush function provided");
> +		goto exit_0;
> +	} else {
> +		out_len = ZSTD_DStreamOutSize();
> +		outp = large_malloc(out_len);
> +		if (!outp) {
> +			error("Could not allocate output buffer");
> +			goto exit_0;
> +		}
> +	}
> +
> +	if (input && fill) {
> +		error("Both input pointer and fill function provided,");
> +		goto exit_1;
> +	} else if (input) {
> +		ZSTD_frameParams p;
> +
> +		inp = input;
> +		if (!ZSTD_getFrameParams(&p, input, in_len))
> +			ws = p.windowSize;
> +	} else if (!fill) {
> +		error("NULL input pointer and missing fill function");
> +		goto exit_1;
> +	} else {
> +		in_len = ZSTD_DStreamInSize();
> +		inp = large_malloc(in_len);
> +		if (!inp) {
> +			error("Could not allocate input buffer");
> +			goto exit_1;
> +		}
> +	}
> +
> +	workspace_size = ZSTD_DStreamWorkspaceBound(ws);
> +	workspace = large_malloc(workspace_size);
> +	if (!workspace) {
> +		error("Could not allocate workspace");
> +		goto exit_2;
> +	}
> +
> +	zstd = ZSTD_initDStream(ws, workspace, workspace_size);
> +	if (!zstd) {
> +		error("Could not initialize ZSTD");
> +		goto exit_3;
> +	}
> +
> +	in.src = inp;
> +	in.size = in_len;
> +	in.pos = 0;
> +	if (posp)
> +		*posp = 0;
> +
> +	for (;;) {
> +		if (fill) {
> +			in.size = fill(inp, in_len);
> +			if (in.size == 0)
> +				break;
> +		} else if (in.size == in.pos) {
> +			break;
> +		}
> +init:		out.dst = outp;
> +		out.size = out_len;
> +		out.pos = 0;
> +		pos = in.pos;
> +
> +		ret = ZSTD_decompressStream(zstd, &out, &in);
> +		if (posp)
> +			*posp += in.pos - pos;
> +		if (ZSTD_isError(ret)) {
> +			error("Decompression failed");
> +			ret = -EIO;
> +			goto exit_3;
> +		}
> +
> +		if (flush && out.pos) {
> +			if (flush(out.dst, out.pos) != out.pos) {
> +				ret = -EIO;
> +				goto exit_3;
> +			}
> +			goto init;
> +		}
> +
> +		if (ret == 0) {
> +			ret = ZSTD_resetDStream(zstd);
> +			if (ZSTD_isError(ret)) {
> +				ret = -EIO;
> +				goto exit_3;
> +			}
> +		}
> +		if (in.pos < in.size)
> +			goto init_out;
> +	}
> +
> +	ret = 0;
> +
> +exit_3:	large_free(workspace);
> +exit_2:	if (!input)
> +		large_free(inp);
> +exit_1:	if (!output)
> +		large_free(outp);
> +exit_0:	return ret;
> +}
> +
> +#ifdef PREBOOT
> +STATIC int INIT __decompress(unsigned char *buf, long in_len,
> +			      long (*fill)(void*, unsigned long),
> +			      long (*flush)(void*, unsigned long),
> +			      unsigned char *output, long out_len,
> +			      long *posp,
> +			      void (*error)(char *x)
> +	)
> +{
> +	return unzstd(buf, in_len, fill, flush, output, posp, error);
> +}
> +#endif
> diff --git a/lib/zstd/decompress.c b/lib/zstd/decompress.c
> index 269ee9a796c1..6a5e1ce22719 100644
> --- a/lib/zstd/decompress.c
> +++ b/lib/zstd/decompress.c
> @@ -42,9 +42,11 @@
> /*-*************************************
> *  Macros
> ***************************************/
> +#ifndef PREBOOT
> #define ZSTD_isError ERR_isError /* for inlining */
> #define FSE_isError ERR_isError
> #define HUF_isError ERR_isError
> +#endif
> 
> /*_*******************************************************
> *  Memory operations
> diff --git a/lib/zstd/fse_decompress.c b/lib/zstd/fse_decompress.c
> index a84300e5a013..bd4e9c891d96 100644
> --- a/lib/zstd/fse_decompress.c
> +++ b/lib/zstd/fse_decompress.c
> @@ -54,12 +54,13 @@
> /* **************************************************************
> *  Error Management
> ****************************************************************/
> -#define FSE_isError ERR_isError
> #define FSE_STATIC_ASSERT(c)                                   \
> 	{                                                      \
> 		enum { FSE_static_assert = 1 / (int)(!!(c)) }; \
> 	} /* use only *after* variable declarations */
> 
> +#ifndef PREBOOT
> +#define FSE_isError ERR_isError
> /* check and forward error code */
> #define CHECK_F(f)                  \
> 	{                           \
> @@ -67,6 +68,7 @@
> 		if (FSE_isError(e)) \
> 			return e;   \
> 	}
> +#endif
> 
> /* **************************************************************
> *  Templates
> diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
> index 3fa32f83b2d7..1c2f2dc528dc 100644
> --- a/scripts/Makefile.lib
> +++ b/scripts/Makefile.lib
> @@ -337,6 +337,9 @@ quiet_cmd_lz4 = LZ4     $@
>       cmd_lz4 = { cat $(real-prereqs) | lz4c -l -c1 stdin stdout; \
>                   $(size_append); } > $@
> 
> +quiet_cmd_zstd = ZSTD    $@
> +      cmd_zstd = { cat $(real-prereqs) | zstd -19 --zstd=wlog=21; $(size_append); } > $@
> +
> # U-Boot mkimage
> # ---------------------------------------------------------------------------
> 
> diff --git a/usr/Kconfig b/usr/Kconfig
> index a6b68503d177..892eb15957db 100644
> --- a/usr/Kconfig
> +++ b/usr/Kconfig
> @@ -106,6 +106,15 @@ config RD_LZ4
> 	  Support loading of a LZ4 encoded initial ramdisk or cpio buffer
> 	  If unsure, say N.
> 
> +config RD_ZSTD
> +	bool "Support initial ramdisk/ramfs compressed using ZSTD"
> +	default y
> +	depends on BLK_DEV_INITRD
> +	select DECOMPRESS_ZSTD
> +	help
> +	  Support loading of a ZSTD encoded initial ramdisk or cpio buffer
> +	  If unsure, say N.
> +
> choice
> 	prompt "Built-in initramfs compression mode"
> 	depends on INITRAMFS_SOURCE!=""
> @@ -214,6 +223,19 @@ config INITRAMFS_COMPRESSION_LZ4
> 	  If you choose this, keep in mind that most distros don't provide lz4
> 	  by default which could cause a build failure.
> 
> +config INITRAMFS_COMPRESSION_ZSTD
> +	bool "ZSTD"
> +	depends on RD_ZSTD
> +	help
> +	  Its compression ratio is roughly 10% worst than xz, but the
> +	  decompression is 10x faster. Currently, this is one of the optimal
> +	  algorithms available in the kernel, as there isn't an algorithm,
> +	  which would provide a better compression ratio and a shorter
> +	  decompression time.
> +
> +	  If you choose this, keep in mind that you may need to install the zstd
> +	  tool to be able to compress the initram.
> +
> endchoice
> 
> config INITRAMFS_COMPRESSION
> @@ -226,10 +248,12 @@ config INITRAMFS_COMPRESSION
> 	default ".xz"   if INITRAMFS_COMPRESSION_XZ
> 	default ".lzo"  if INITRAMFS_COMPRESSION_LZO
> 	default ".lz4"  if INITRAMFS_COMPRESSION_LZ4
> +	default ".zst"  if INITRAMFS_COMPRESSION_ZSTD
> 	default ".gz"   if RD_GZIP
> 	default ".lz4"  if RD_LZ4
> 	default ".lzo"  if RD_LZO
> 	default ".xz"   if RD_XZ
> 	default ".lzma" if RD_LZMA
> 	default ".bz2"  if RD_BZIP2
> +	default ".zst"  if RD_ZSTD
> 	default ""
> -- 
> 2.20.1
>