From: Matt Sealey <[email protected]>
Most compilers should be able to merge adjacent loads/stores of sizes which
are less than but effect a multiple of a machine word size (in effect a
memcpy() of a constant amount). However the semantics of the macro are that
it just does the copy, the pointer increment is in the code, hence we see
*a = *b
a += 8
b += 8
*a = *b
a += 8
b += 8
This introduces a dependency between the two groups of statements which
seems to defeat said compiler optimizers and generate some very strange
sequences of addition and subtraction of address offsets (i.e. it is
overcomplicated).
Since COPY8 is only ever used to copy amounts of 16 bytes (in pairs), just
define COPY16 as COPY8,COPY8. We leave the definition to preserve the need
to do unaligned accesses to machine-sized words per the original code
intent, we just don't use it in the code proper.
COPY16 then gives us code like:
*a = *b
*(a+8) = *(b+8)
a += 16
b += 16
This seems to allow compilers to generate much better code by using
base register writeback or simply positively incrementing offsets which
seems to positively affect performance. It is, at least, fewer instructions
to do the same job.
Signed-off-by: Matt Sealey <[email protected]>
---
lib/lzo/lzo1x_compress.c | 9 +++------
lib/lzo/lzo1x_decompress_safe.c | 18 ++++++------------
lib/lzo/lzodefs.h | 3 +++
3 files changed, 12 insertions(+), 18 deletions(-)
diff --git a/lib/lzo/lzo1x_compress.c b/lib/lzo/lzo1x_compress.c
index 236eb21167b5..82fb5571ce5e 100644
--- a/lib/lzo/lzo1x_compress.c
+++ b/lib/lzo/lzo1x_compress.c
@@ -60,8 +60,7 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
op += t;
} else if (t <= 16) {
*op++ = (t - 3);
- COPY8(op, ii);
- COPY8(op + 8, ii + 8);
+ COPY16(op, ii);
op += t;
} else {
if (t <= 18) {
@@ -76,8 +75,7 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
*op++ = tt;
}
do {
- COPY8(op, ii);
- COPY8(op + 8, ii + 8);
+ COPY16(op, ii);
op += 16;
ii += 16;
t -= 16;
@@ -255,8 +253,7 @@ int lzo1x_1_compress(const unsigned char *in, size_t in_len,
*op++ = tt;
}
if (t >= 16) do {
- COPY8(op, ii);
- COPY8(op + 8, ii + 8);
+ COPY16(op, ii);
op += 16;
ii += 16;
t -= 16;
diff --git a/lib/lzo/lzo1x_decompress_safe.c b/lib/lzo/lzo1x_decompress_safe.c
index a1c387f6afba..aa95d3066b7d 100644
--- a/lib/lzo/lzo1x_decompress_safe.c
+++ b/lib/lzo/lzo1x_decompress_safe.c
@@ -86,12 +86,9 @@ int lzo1x_decompress_safe(const unsigned char *in, size_t in_len,
const unsigned char *ie = ip + t;
unsigned char *oe = op + t;
do {
- COPY8(op, ip);
- op += 8;
- ip += 8;
- COPY8(op, ip);
- op += 8;
- ip += 8;
+ COPY16(op, ip);
+ op += 16;
+ ip += 16;
} while (ip < ie);
ip = ie;
op = oe;
@@ -187,12 +184,9 @@ int lzo1x_decompress_safe(const unsigned char *in, size_t in_len,
unsigned char *oe = op + t;
if (likely(HAVE_OP(t + 15))) {
do {
- COPY8(op, m_pos);
- op += 8;
- m_pos += 8;
- COPY8(op, m_pos);
- op += 8;
- m_pos += 8;
+ COPY16(op, m_pos);
+ op += 16;
+ m_pos += 16;
} while (op < oe);
op = oe;
if (HAVE_IP(6)) {
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index 4edefd2f540c..9f167eaabb28 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -23,6 +23,9 @@
COPY4(dst, src); COPY4((dst) + 4, (src) + 4)
#endif
+#define COPY16(dst, src) \
+ do { COPY8(dst, src); COPY8((dst) + 8, (src) + 8); } while (0)
+
#if defined(__BIG_ENDIAN) && defined(__LITTLE_ENDIAN)
#error "conflicting endian definitions"
#elif defined(__x86_64__)
--
2.16.4
When using zram, we frequently encounter long runs of zero bytes.
This adds a special case which identifies runs of zeros and encodes
them using run-length encoding.
This is faster for both compression and decompresion. For
high-entropy data which doesn't hit this case, impact is minimal.
Compression ratio is within a few percent in all cases.
This modifies the bitstream in a way which is backwards compatible
(i.e., we can decompress old bitstreams, but old versions of lzo
cannot decompress new bitstreams).
Signed-off-by: Dave Rodgman <[email protected]>
---
Documentation/lzo.txt | 35 ++++++++++++---
include/linux/lzo.h | 2 +-
lib/lzo/lzo1x_compress.c | 98 ++++++++++++++++++++++++++++++++++++-----
lib/lzo/lzo1x_decompress_safe.c | 75 +++++++++++++++++++++----------
lib/lzo/lzodefs.h | 12 ++++-
5 files changed, 180 insertions(+), 42 deletions(-)
diff --git a/Documentation/lzo.txt b/Documentation/lzo.txt
index 6fa6a93d0949..306c60344ca7 100644
--- a/Documentation/lzo.txt
+++ b/Documentation/lzo.txt
@@ -78,16 +78,30 @@ Description
is an implementation design choice independent on the algorithm or
encoding.
+Versions
+
+0: Original version
+1: LZO-RLE
+
+Version 1 of LZO implements an extension to encode runs of zeros using run
+length encoding. This improves speed for data with many zeros, which is a
+common case for zram. This modifies the bitstream in a backwards compatible way
+(v1 can correctly decompress v0 compressed data, but v0 cannot read v1 data).
+
Byte sequences
==============
First byte encoding::
- 0..17 : follow regular instruction encoding, see below. It is worth
- noting that codes 16 and 17 will represent a block copy from
- the dictionary which is empty, and that they will always be
+ 0..16 : follow regular instruction encoding, see below. It is worth
+ noting that code 16 will represent a block copy from the
+ dictionary which is empty, and that it will always be
invalid at this place.
+ 17 : bitstream version. If the first byte is 17, the next byte
+ gives the bitstream version. If the first byte is not 17,
+ the bitstream version is 0.
+
18..21 : copy 0..3 literals
state = (byte - 17) = 0..3 [ copy <state> literals ]
skip byte
@@ -140,6 +154,11 @@ Byte sequences
state = S (copy S literals after this block)
End of stream is reached if distance == 16384
+ In version 1, this instruction is also used to encode a run of zeros if
+ distance = 0xbfff, i.e. H = 1 and the D bits are all 1.
+ In this case, it is followed by a fourth byte, X.
+ run length = ((X << 3) | (0 0 0 0 0 L L L)) + 4.
+
0 0 1 L L L L L (32..63)
Copy of small block within 16kB distance (preferably less than 34B)
length = 2 + (L ?: 31 + (zero_bytes * 255) + non_zero_byte)
@@ -165,7 +184,9 @@ Authors
=======
This document was written by Willy Tarreau <[email protected]> on 2014/07/19 during an
- analysis of the decompression code available in Linux 3.16-rc5. The code is
- tricky, it is possible that this document contains mistakes or that a few
- corner cases were overlooked. In any case, please report any doubt, fix, or
- proposed updates to the author(s) so that the document can be updated.
+ analysis of the decompression code available in Linux 3.16-rc5, and updated
+ by Dave Rodgman <[email protected]> on 2018/10/30 to introduce run-length
+ encoding. The code is tricky, it is possible that this document contains
+ mistakes or that a few corner cases were overlooked. In any case, please
+ report any doubt, fix, or proposed updates to the author(s) so that the
+ document can be updated.
diff --git a/include/linux/lzo.h b/include/linux/lzo.h
index 2ae27cb89927..547a86c71e1b 100644
--- a/include/linux/lzo.h
+++ b/include/linux/lzo.h
@@ -18,7 +18,7 @@
#define LZO1X_1_MEM_COMPRESS (8192 * sizeof(unsigned short))
#define LZO1X_MEM_COMPRESS LZO1X_1_MEM_COMPRESS
-#define lzo1x_worst_compress(x) ((x) + ((x) / 16) + 64 + 3)
+#define lzo1x_worst_compress(x) ((x) + ((x) / 16) + 64 + 3 + 2)
/* This requires 'wrkmem' of size LZO1X_1_MEM_COMPRESS */
int lzo1x_1_compress(const unsigned char *src, size_t src_len,
diff --git a/lib/lzo/lzo1x_compress.c b/lib/lzo/lzo1x_compress.c
index 82fb5571ce5e..fa8d4ff38531 100644
--- a/lib/lzo/lzo1x_compress.c
+++ b/lib/lzo/lzo1x_compress.c
@@ -20,7 +20,7 @@
static noinline size_t
lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
unsigned char *out, size_t *out_len,
- size_t ti, void *wrkmem)
+ size_t ti, void *wrkmem, signed char *state_offset)
{
const unsigned char *ip;
unsigned char *op;
@@ -38,24 +38,82 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
const unsigned char *m_pos;
size_t t, m_len, m_off;
u32 dv;
+ u32 run_length = 0;
literal:
ip += 1 + ((ip - ii) >> 5);
next:
if (unlikely(ip >= ip_end))
break;
dv = get_unaligned_le32(ip);
- t = ((dv * 0x1824429d) >> (32 - D_BITS)) & D_MASK;
- m_pos = in + dict[t];
- dict[t] = (lzo_dict_t) (ip - in);
- if (unlikely(dv != get_unaligned_le32(m_pos)))
- goto literal;
+
+ if (dv == 0) {
+ const unsigned char *ir = ip + 4;
+ const unsigned char *limit = ip_end
+ < (ip + MAX_ZERO_RUN_LENGTH + 1)
+ ? ip_end : ip + MAX_ZERO_RUN_LENGTH + 1;
+#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && \
+ defined(LZO_FAST_64BIT_MEMORY_ACCESS)
+ u64 dv64;
+
+ for (; (ir + 32) <= limit; ir += 32) {
+ dv64 = get_unaligned((u64 *)ir);
+ dv64 |= get_unaligned((u64 *)ir + 1);
+ dv64 |= get_unaligned((u64 *)ir + 2);
+ dv64 |= get_unaligned((u64 *)ir + 3);
+ if (dv64)
+ break;
+ }
+ for (; (ir + 8) <= limit; ir += 8) {
+ dv64 = get_unaligned((u64 *)ir);
+ if (dv64) {
+# if defined(__LITTLE_ENDIAN)
+ ir += __builtin_ctzll(dv64) >> 3;
+# elif defined(__BIG_ENDIAN)
+ ir += __builtin_clzll(dv64) >> 3;
+# else
+# error "missing endian definition"
+# endif
+ break;
+ }
+ }
+#else
+ while ((ir < (const unsigned char *)
+ ALIGN((uintptr_t)ir, 4)) &&
+ (ir < limit) && (*ir == 0))
+ ir++;
+ for (; (ir + 4) <= limit; ir += 4) {
+ dv = *((u32 *)ir);
+ if (dv) {
+# if defined(__LITTLE_ENDIAN)
+ ir += __builtin_ctz(dv) >> 3;
+# elif defined(__BIG_ENDIAN)
+ ir += __builtin_clz(dv) >> 3;
+# else
+# error "missing endian definition"
+# endif
+ break;
+ }
+ }
+#endif
+ while (likely(ir < limit) && unlikely(*ir == 0))
+ ir++;
+ run_length = ir - ip;
+ if (run_length > MAX_ZERO_RUN_LENGTH)
+ run_length = MAX_ZERO_RUN_LENGTH;
+ } else {
+ t = ((dv * 0x1824429d) >> (32 - D_BITS)) & D_MASK;
+ m_pos = in + dict[t];
+ dict[t] = (lzo_dict_t) (ip - in);
+ if (unlikely(dv != get_unaligned_le32(m_pos)))
+ goto literal;
+ }
ii -= ti;
ti = 0;
t = ip - ii;
if (t != 0) {
if (t <= 3) {
- op[-2] |= t;
+ op[*state_offset] |= t;
COPY4(op, ii);
op += t;
} else if (t <= 16) {
@@ -86,6 +144,17 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
}
}
+ if (unlikely(run_length)) {
+ ip += run_length;
+ run_length -= MIN_ZERO_RUN_LENGTH;
+ put_unaligned_le32((run_length << 21) | 0xfffc18
+ | (run_length & 0x7), op);
+ op += 4;
+ run_length = 0;
+ *state_offset = -3;
+ goto finished_writing_instruction;
+ }
+
m_len = 4;
{
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && defined(LZO_USE_CTZ64)
@@ -168,7 +237,6 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
m_off = ip - m_pos;
ip += m_len;
- ii = ip;
if (m_len <= M2_MAX_LEN && m_off <= M2_MAX_OFFSET) {
m_off -= 1;
*op++ = (((m_len - 1) << 5) | ((m_off & 7) << 2));
@@ -205,6 +273,9 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
*op++ = (m_off << 2);
*op++ = (m_off >> 6);
}
+ *state_offset = -2;
+finished_writing_instruction:
+ ii = ip;
goto next;
}
*out_len = op - out;
@@ -219,6 +290,12 @@ int lzo1x_1_compress(const unsigned char *in, size_t in_len,
unsigned char *op = out;
size_t l = in_len;
size_t t = 0;
+ signed char state_offset = -2;
+
+ // LZO v0 will never write 17 as first byte,
+ // so this is used to version the bitstream
+ *op++ = 17;
+ *op++ = LZO_VERSION;
while (l > 20) {
size_t ll = l <= (M4_MAX_OFFSET + 1) ? l : (M4_MAX_OFFSET + 1);
@@ -227,7 +304,8 @@ int lzo1x_1_compress(const unsigned char *in, size_t in_len,
break;
BUILD_BUG_ON(D_SIZE * sizeof(lzo_dict_t) > LZO1X_1_MEM_COMPRESS);
memset(wrkmem, 0, D_SIZE * sizeof(lzo_dict_t));
- t = lzo1x_1_do_compress(ip, ll, op, out_len, t, wrkmem);
+ t = lzo1x_1_do_compress(ip, ll, op, out_len,
+ t, wrkmem, &state_offset);
ip += ll;
op += *out_len;
l -= ll;
@@ -240,7 +318,7 @@ int lzo1x_1_compress(const unsigned char *in, size_t in_len,
if (op == out && t <= 238) {
*op++ = (17 + t);
} else if (t <= 3) {
- op[-2] |= t;
+ op[state_offset] |= t;
} else if (t <= 18) {
*op++ = (t - 3);
} else {
diff --git a/lib/lzo/lzo1x_decompress_safe.c b/lib/lzo/lzo1x_decompress_safe.c
index aa95d3066b7d..b8f88d5ea3ff 100644
--- a/lib/lzo/lzo1x_decompress_safe.c
+++ b/lib/lzo/lzo1x_decompress_safe.c
@@ -46,11 +46,23 @@ int lzo1x_decompress_safe(const unsigned char *in, size_t in_len,
const unsigned char * const ip_end = in + in_len;
unsigned char * const op_end = out + *out_len;
+ unsigned char bitstream_version;
+
op = out;
ip = in;
if (unlikely(in_len < 3))
goto input_overrun;
+
+ if (likely(*ip == 17)) {
+ bitstream_version = ip[1];
+ ip += 2;
+ if (unlikely(in_len < 5))
+ goto input_overrun;
+ } else {
+ bitstream_version = 0;
+ }
+
if (*ip > 17) {
t = *ip++ - 17;
if (t < 4) {
@@ -151,32 +163,49 @@ int lzo1x_decompress_safe(const unsigned char *in, size_t in_len,
m_pos -= next >> 2;
next &= 3;
} else {
- m_pos = op;
- m_pos -= (t & 8) << 11;
- t = (t & 7) + (3 - 1);
- if (unlikely(t == 2)) {
- size_t offset;
- const unsigned char *ip_last = ip;
+ NEED_IP(2);
+ next = get_unaligned_le16(ip);
+ if (((next & 0xfffc) == 0xfffc) &&
+ ((t & 0xf8) == 0x18) &&
+ likely(bitstream_version)) {
+ NEED_IP(3);
+ t &= 7;
+ t |= ip[2] << 3;
+ t += MIN_ZERO_RUN_LENGTH;
+ NEED_OP(t);
+ memset(op, 0, t);
+ op += t;
+ next &= 3;
+ ip += 3;
+ goto match_next;
+ } else {
+ m_pos = op;
+ m_pos -= (t & 8) << 11;
+ t = (t & 7) + (3 - 1);
+ if (unlikely(t == 2)) {
+ size_t offset;
+ const unsigned char *ip_last = ip;
- while (unlikely(*ip == 0)) {
- ip++;
- NEED_IP(1);
- }
- offset = ip - ip_last;
- if (unlikely(offset > MAX_255_COUNT))
- return LZO_E_ERROR;
+ while (unlikely(*ip == 0)) {
+ ip++;
+ NEED_IP(1);
+ }
+ offset = ip - ip_last;
+ if (unlikely(offset > MAX_255_COUNT))
+ return LZO_E_ERROR;
- offset = (offset << 8) - offset;
- t += offset + 7 + *ip++;
- NEED_IP(2);
+ offset = (offset << 8) - offset;
+ t += offset + 7 + *ip++;
+ NEED_IP(2);
+ next = get_unaligned_le16(ip);
+ }
+ ip += 2;
+ m_pos -= next >> 2;
+ next &= 3;
+ if (m_pos == op)
+ goto eof_found;
+ m_pos -= 0x4000;
}
- next = get_unaligned_le16(ip);
- ip += 2;
- m_pos -= next >> 2;
- next &= 3;
- if (m_pos == op)
- goto eof_found;
- m_pos -= 0x4000;
}
TEST_LB(m_pos);
#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index 341d0f6095ab..436bb400ac87 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -13,6 +13,12 @@
*/
+/* Version
+ * 0: original lzo version
+ * 1: lzo with support for RLE
+ */
+#define LZO_VERSION 1
+
#define COPY4(dst, src) \
put_unaligned(get_unaligned((const u32 *)(src)), (u32 *)(dst))
#if defined(__x86_64__) || defined(__aarch64__)
@@ -31,6 +37,7 @@
#elif defined(__x86_64__) || defined(__aarch64__)
#define LZO_USE_CTZ64 1
#define LZO_USE_CTZ32 1
+#define LZO_FAST_64BIT_MEMORY_ACCESS
#elif defined(__i386__) || defined(__powerpc__)
#define LZO_USE_CTZ32 1
#elif defined(__arm__)
@@ -45,7 +52,7 @@
#define M1_MAX_OFFSET 0x0400
#define M2_MAX_OFFSET 0x0800
#define M3_MAX_OFFSET 0x4000
-#define M4_MAX_OFFSET 0xbfff
+#define M4_MAX_OFFSET 0xbffe
#define M1_MIN_LEN 2
#define M1_MAX_LEN 2
@@ -61,6 +68,9 @@
#define M3_MARKER 32
#define M4_MARKER 16
+#define MIN_ZERO_RUN_LENGTH 4
+#define MAX_ZERO_RUN_LENGTH (2047 + MIN_ZERO_RUN_LENGTH)
+
#define lzo_dict_t unsigned short
#define D_BITS 13
#define D_SIZE (1u << D_BITS)
--
2.16.4
To prevent any issues with persistent data, separate lzo-rle
from lzo so that it is treated as a separate algorithm, and
lzo is still available.
Use lzo-rle as the default algorithm for
zram.
Signed-off-by: Dave Rodgman <[email protected]>
---
Documentation/lzo.txt | 12 ++-
crypto/Makefile | 2 +-
crypto/lzo-rle.c | 171 ++++++++++++++++++++++++++++++++++++++++++
crypto/tcrypt.c | 4 +-
drivers/block/zram/zcomp.c | 1 +
drivers/block/zram/zram_drv.c | 2 +-
include/linux/lzo.h | 4 +
lib/lzo/lzo1x_compress.c | 42 ++++++++---
lib/lzo/lzodefs.h | 3 +-
9 files changed, 223 insertions(+), 18 deletions(-)
create mode 100644 crypto/lzo-rle.c
diff --git a/Documentation/lzo.txt b/Documentation/lzo.txt
index 306c60344ca7..f79934225d8d 100644
--- a/Documentation/lzo.txt
+++ b/Documentation/lzo.txt
@@ -88,6 +88,10 @@ length encoding. This improves speed for data with many zeros, which is a
common case for zram. This modifies the bitstream in a backwards compatible way
(v1 can correctly decompress v0 compressed data, but v0 cannot read v1 data).
+For maximum compatibility, both versions are available under different names
+(lzo and lzo-rle). Differences in the encoding are noted in this document with
+e.g.: version 1 only.
+
Byte sequences
==============
@@ -99,8 +103,8 @@ Byte sequences
invalid at this place.
17 : bitstream version. If the first byte is 17, the next byte
- gives the bitstream version. If the first byte is not 17,
- the bitstream version is 0.
+ gives the bitstream version (version 1 only). If the first byte
+ is not 17, the bitstream version is 0.
18..21 : copy 0..3 literals
state = (byte - 17) = 0..3 [ copy <state> literals ]
@@ -154,8 +158,8 @@ Byte sequences
state = S (copy S literals after this block)
End of stream is reached if distance == 16384
- In version 1, this instruction is also used to encode a run of zeros if
- distance = 0xbfff, i.e. H = 1 and the D bits are all 1.
+ In version 1 only, this instruction is also used to encode a run of
+ zeros if distance = 0xbfff, i.e. H = 1 and the D bits are all 1.
In this case, it is followed by a fourth byte, X.
run length = ((X << 3) | (0 0 0 0 0 L L L)) + 4.
diff --git a/crypto/Makefile b/crypto/Makefile
index 6d1d40eeb964..4b482fa11fee 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -125,7 +125,7 @@ obj-$(CONFIG_CRYPTO_CRC32C) += crc32c_generic.o
obj-$(CONFIG_CRYPTO_CRC32) += crc32_generic.o
obj-$(CONFIG_CRYPTO_CRCT10DIF) += crct10dif_common.o crct10dif_generic.o
obj-$(CONFIG_CRYPTO_AUTHENC) += authenc.o authencesn.o
-obj-$(CONFIG_CRYPTO_LZO) += lzo.o
+obj-$(CONFIG_CRYPTO_LZO) += lzo.o lzo-rle.o
obj-$(CONFIG_CRYPTO_LZ4) += lz4.o
obj-$(CONFIG_CRYPTO_LZ4HC) += lz4hc.o
obj-$(CONFIG_CRYPTO_842) += 842.o
diff --git a/crypto/lzo-rle.c b/crypto/lzo-rle.c
new file mode 100644
index 000000000000..d51baebbd8d7
--- /dev/null
+++ b/crypto/lzo-rle.c
@@ -0,0 +1,171 @@
+/*
+ * Cryptographic API.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 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.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/crypto.h>
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <linux/lzo.h>
+#include <crypto/internal/scompress.h>
+
+struct lzorle_ctx {
+ void *lzorle_comp_mem;
+};
+
+static void *lzorle_alloc_ctx(struct crypto_scomp *tfm)
+{
+ void *ctx;
+
+ ctx = kvmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL);
+ if (!ctx)
+ return ERR_PTR(-ENOMEM);
+
+ return ctx;
+}
+
+static int lzorle_init(struct crypto_tfm *tfm)
+{
+ struct lzorle_ctx *ctx = crypto_tfm_ctx(tfm);
+
+ ctx->lzorle_comp_mem = lzorle_alloc_ctx(NULL);
+ if (IS_ERR(ctx->lzorle_comp_mem))
+ return -ENOMEM;
+
+ return 0;
+}
+
+static void lzorle_free_ctx(struct crypto_scomp *tfm, void *ctx)
+{
+ kvfree(ctx);
+}
+
+static void lzorle_exit(struct crypto_tfm *tfm)
+{
+ struct lzorle_ctx *ctx = crypto_tfm_ctx(tfm);
+
+ lzorle_free_ctx(NULL, ctx->lzorle_comp_mem);
+}
+
+static int __lzorle_compress(const u8 *src, unsigned int slen,
+ u8 *dst, unsigned int *dlen, void *ctx)
+{
+ size_t tmp_len = *dlen; /* size_t(ulong) <-> uint on 64 bit */
+ int err;
+
+ err = lzorle1x_1_compress(src, slen, dst, &tmp_len, ctx);
+
+ if (err != LZO_E_OK)
+ return -EINVAL;
+
+ *dlen = tmp_len;
+ return 0;
+}
+
+static int lzorle_compress(struct crypto_tfm *tfm, const u8 *src,
+ unsigned int slen, u8 *dst, unsigned int *dlen)
+{
+ struct lzorle_ctx *ctx = crypto_tfm_ctx(tfm);
+
+ return __lzorle_compress(src, slen, dst, dlen, ctx->lzorle_comp_mem);
+}
+
+static int lzorle_scompress(struct crypto_scomp *tfm, const u8 *src,
+ unsigned int slen, u8 *dst, unsigned int *dlen,
+ void *ctx)
+{
+ return __lzorle_compress(src, slen, dst, dlen, ctx);
+}
+
+static int __lzorle_decompress(const u8 *src, unsigned int slen,
+ u8 *dst, unsigned int *dlen)
+{
+ int err;
+ size_t tmp_len = *dlen; /* size_t(ulong) <-> uint on 64 bit */
+
+ err = lzo1x_decompress_safe(src, slen, dst, &tmp_len);
+
+ if (err != LZO_E_OK)
+ return -EINVAL;
+
+ *dlen = tmp_len;
+ return 0;
+}
+
+static int lzorle_decompress(struct crypto_tfm *tfm, const u8 *src,
+ unsigned int slen, u8 *dst, unsigned int *dlen)
+{
+ return __lzorle_decompress(src, slen, dst, dlen);
+}
+
+static int lzorle_sdecompress(struct crypto_scomp *tfm, const u8 *src,
+ unsigned int slen, u8 *dst, unsigned int *dlen,
+ void *ctx)
+{
+ return __lzorle_decompress(src, slen, dst, dlen);
+}
+
+static struct crypto_alg alg = {
+ .cra_name = "lzo-rle",
+ .cra_flags = CRYPTO_ALG_TYPE_COMPRESS,
+ .cra_ctxsize = sizeof(struct lzorle_ctx),
+ .cra_module = THIS_MODULE,
+ .cra_init = lzorle_init,
+ .cra_exit = lzorle_exit,
+ .cra_u = { .compress = {
+ .coa_compress = lzorle_compress,
+ .coa_decompress = lzorle_decompress } }
+};
+
+static struct scomp_alg scomp = {
+ .alloc_ctx = lzorle_alloc_ctx,
+ .free_ctx = lzorle_free_ctx,
+ .compress = lzorle_scompress,
+ .decompress = lzorle_sdecompress,
+ .base = {
+ .cra_name = "lzo-rle",
+ .cra_driver_name = "lzo-rle-scomp",
+ .cra_module = THIS_MODULE,
+ }
+};
+
+static int __init lzorle_mod_init(void)
+{
+ int ret;
+
+ ret = crypto_register_alg(&alg);
+ if (ret)
+ return ret;
+
+ ret = crypto_register_scomp(&scomp);
+ if (ret) {
+ crypto_unregister_alg(&alg);
+ return ret;
+ }
+
+ return ret;
+}
+
+static void __exit lzorle_mod_fini(void)
+{
+ crypto_unregister_alg(&alg);
+ crypto_unregister_scomp(&scomp);
+}
+
+module_init(lzorle_mod_init);
+module_exit(lzorle_mod_fini);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("LZO-RLE Compression Algorithm");
+MODULE_ALIAS_CRYPTO("lzo-rle");
diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c
index bdde95e8d369..c154bc657e3a 100644
--- a/crypto/tcrypt.c
+++ b/crypto/tcrypt.c
@@ -76,8 +76,8 @@ static char *check[] = {
"cast6", "arc4", "michael_mic", "deflate", "crc32c", "tea", "xtea",
"khazad", "wp512", "wp384", "wp256", "tnepres", "xeta", "fcrypt",
"camellia", "seed", "salsa20", "rmd128", "rmd160", "rmd256", "rmd320",
- "lzo", "cts", "zlib", "sha3-224", "sha3-256", "sha3-384", "sha3-512",
- NULL
+ "lzo", "lzo-rle", "cts", "zlib", "sha3-224", "sha3-256", "sha3-384",
+ "sha3-512", NULL
};
static u32 block_sizes[] = { 16, 64, 256, 1024, 8192, 0 };
diff --git a/drivers/block/zram/zcomp.c b/drivers/block/zram/zcomp.c
index 4ed0a78fdc09..4d9a38890965 100644
--- a/drivers/block/zram/zcomp.c
+++ b/drivers/block/zram/zcomp.c
@@ -20,6 +20,7 @@
static const char * const backends[] = {
"lzo",
+ "lzo-rle",
#if IS_ENABLED(CONFIG_CRYPTO_LZ4)
"lz4",
#endif
diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
index a1d6b5597c17..bc127f21d1e9 100644
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@@ -41,7 +41,7 @@ static DEFINE_IDR(zram_index_idr);
static DEFINE_MUTEX(zram_index_mutex);
static int zram_major;
-static const char *default_compressor = "lzo";
+static const char *default_compressor = "lzo-rle";
/* Module params (documentation at end) */
static unsigned int num_devices = 1;
diff --git a/include/linux/lzo.h b/include/linux/lzo.h
index 547a86c71e1b..e95c7d1092b2 100644
--- a/include/linux/lzo.h
+++ b/include/linux/lzo.h
@@ -24,6 +24,10 @@
int lzo1x_1_compress(const unsigned char *src, size_t src_len,
unsigned char *dst, size_t *dst_len, void *wrkmem);
+/* This requires 'wrkmem' of size LZO1X_1_MEM_COMPRESS */
+int lzorle1x_1_compress(const unsigned char *src, size_t src_len,
+ unsigned char *dst, size_t *dst_len, void *wrkmem);
+
/* safe decompression with overrun testing */
int lzo1x_decompress_safe(const unsigned char *src, size_t src_len,
unsigned char *dst, size_t *dst_len);
diff --git a/lib/lzo/lzo1x_compress.c b/lib/lzo/lzo1x_compress.c
index fa8d4ff38531..1a1cd5e84391 100644
--- a/lib/lzo/lzo1x_compress.c
+++ b/lib/lzo/lzo1x_compress.c
@@ -20,7 +20,8 @@
static noinline size_t
lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
unsigned char *out, size_t *out_len,
- size_t ti, void *wrkmem, signed char *state_offset)
+ size_t ti, void *wrkmem, signed char *state_offset,
+ const unsigned char bitstream_version)
{
const unsigned char *ip;
unsigned char *op;
@@ -46,7 +47,7 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
break;
dv = get_unaligned_le32(ip);
- if (dv == 0) {
+ if (dv == 0 && bitstream_version) {
const unsigned char *ir = ip + 4;
const unsigned char *limit = ip_end
< (ip + MAX_ZERO_RUN_LENGTH + 1)
@@ -282,30 +283,36 @@ lzo1x_1_do_compress(const unsigned char *in, size_t in_len,
return in_end - (ii - ti);
}
-int lzo1x_1_compress(const unsigned char *in, size_t in_len,
+int lzogeneric1x_1_compress(const unsigned char *in, size_t in_len,
unsigned char *out, size_t *out_len,
- void *wrkmem)
+ void *wrkmem, const unsigned char bitstream_version)
{
const unsigned char *ip = in;
unsigned char *op = out;
size_t l = in_len;
size_t t = 0;
signed char state_offset = -2;
+ unsigned int m4_max_offset;
// LZO v0 will never write 17 as first byte,
// so this is used to version the bitstream
- *op++ = 17;
- *op++ = LZO_VERSION;
+ if (bitstream_version > 0) {
+ *op++ = 17;
+ *op++ = bitstream_version;
+ m4_max_offset = M4_MAX_OFFSET_V1;
+ } else {
+ m4_max_offset = M4_MAX_OFFSET_V0;
+ }
while (l > 20) {
- size_t ll = l <= (M4_MAX_OFFSET + 1) ? l : (M4_MAX_OFFSET + 1);
+ size_t ll = l <= (m4_max_offset + 1) ? l : (m4_max_offset + 1);
uintptr_t ll_end = (uintptr_t) ip + ll;
if ((ll_end + ((t + ll) >> 5)) <= ll_end)
break;
BUILD_BUG_ON(D_SIZE * sizeof(lzo_dict_t) > LZO1X_1_MEM_COMPRESS);
memset(wrkmem, 0, D_SIZE * sizeof(lzo_dict_t));
- t = lzo1x_1_do_compress(ip, ll, op, out_len,
- t, wrkmem, &state_offset);
+ t = lzo1x_1_do_compress(ip, ll, op, out_len, t, wrkmem,
+ &state_offset, bitstream_version);
ip += ll;
op += *out_len;
l -= ll;
@@ -348,7 +355,24 @@ int lzo1x_1_compress(const unsigned char *in, size_t in_len,
*out_len = op - out;
return LZO_E_OK;
}
+
+int lzo1x_1_compress(const unsigned char *in, size_t in_len,
+ unsigned char *out, size_t *out_len,
+ void *wrkmem)
+{
+ return lzogeneric1x_1_compress(in, in_len, out, out_len, wrkmem, 0);
+}
+
+int lzorle1x_1_compress(const unsigned char *in, size_t in_len,
+ unsigned char *out, size_t *out_len,
+ void *wrkmem)
+{
+ return lzogeneric1x_1_compress(in, in_len, out, out_len,
+ wrkmem, LZO_VERSION);
+}
+
EXPORT_SYMBOL_GPL(lzo1x_1_compress);
+EXPORT_SYMBOL_GPL(lzorle1x_1_compress);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("LZO1X-1 Compressor");
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index 436bb400ac87..60b5c63afcdb 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -52,7 +52,8 @@
#define M1_MAX_OFFSET 0x0400
#define M2_MAX_OFFSET 0x0800
#define M3_MAX_OFFSET 0x4000
-#define M4_MAX_OFFSET 0xbffe
+#define M4_MAX_OFFSET_V0 0xbfff
+#define M4_MAX_OFFSET_V1 0xbffe
#define M1_MIN_LEN 2
#define M1_MAX_LEN 2
--
2.16.4
From: Matt Sealey <[email protected]>
Enable faster 8-byte copies on arm64.
Signed-off-by: Dave Rodgman <[email protected]>
Signed-off-by: Matt Sealey <[email protected]>
---
lib/lzo/lzodefs.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index d167f8fc6795..341d0f6095ab 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -15,7 +15,7 @@
#define COPY4(dst, src) \
put_unaligned(get_unaligned((const u32 *)(src)), (u32 *)(dst))
-#if defined(__x86_64__)
+#if defined(__x86_64__) || defined(__aarch64__)
#define COPY8(dst, src) \
put_unaligned(get_unaligned((const u64 *)(src)), (u64 *)(dst))
#else
--
2.16.4
From: Matt Sealey <[email protected]>
ARMv6 Thumb state introduced an RBIT instruction which, combined with CLZ
as present in ARMv5, introduces an extremely fast path for counting
trailing zeroes.
Enable the use of the GCC builtin for this on ARMv6+ with
CONFIG_THUMB2_KERNEL to ensure we get the 'new' instruction usage.
We do not bother enabling LZO_USE_CTZ64 support for ARMv5 as the builtin
code path does the same thing as the LZO_USE_CTZ32 code, only with more
register pressure.
Signed-off-by: Matt Sealey <[email protected]>
---
lib/lzo/lzodefs.h | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index 9f167eaabb28..a9927bbce21f 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -33,9 +33,14 @@
#define LZO_USE_CTZ32 1
#elif defined(__i386__) || defined(__powerpc__)
#define LZO_USE_CTZ32 1
-#elif defined(__arm__) && (__LINUX_ARM_ARCH__ >= 5)
+#elif defined(__arm__)
+#if (__LINUX_ARM_ARCH__ >= 5)
#define LZO_USE_CTZ32 1
#endif
+#if (__LINUX_ARM_ARCH__ >= 6) && (CONFIG_THUMB2_KERNEL)
+#define LZO_USE_CTZ64 1
+#endif
+#endif
#define M1_MAX_OFFSET 0x0400
#define M2_MAX_OFFSET 0x0800
--
2.16.4
From: Matt Sealey <[email protected]>
LZO leaves some performance on the table by not realising that AArch64 can
optimize count-trailing-zeros bit operations.
Add __aarch64__ to the checked definitions alongside __x86_64__ to enable
the use of rbit/clz instructions on full 64-bit quantities.
Signed-off-by: Matt Sealey <[email protected]>
Signed-off-by: Dave Rodgman <[email protected]>
---
lib/lzo/lzodefs.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/lzo/lzodefs.h b/lib/lzo/lzodefs.h
index a9927bbce21f..d167f8fc6795 100644
--- a/lib/lzo/lzodefs.h
+++ b/lib/lzo/lzodefs.h
@@ -28,7 +28,7 @@
#if defined(__BIG_ENDIAN) && defined(__LITTLE_ENDIAN)
#error "conflicting endian definitions"
-#elif defined(__x86_64__)
+#elif defined(__x86_64__) || defined(__aarch64__)
#define LZO_USE_CTZ64 1
#define LZO_USE_CTZ32 1
#elif defined(__i386__) || defined(__powerpc__)
--
2.16.4
> #define LZO_USE_CTZ32 1
> #elif defined(__i386__) || defined(__powerpc__)
> #define LZO_USE_CTZ32 1
> -#elif defined(__arm__) && (__LINUX_ARM_ARCH__ >= 5)
> +#elif defined(__arm__)
> +#if (__LINUX_ARM_ARCH__ >= 5)
> #define LZO_USE_CTZ32 1
> #endif
> +#if (__LINUX_ARM_ARCH__ >= 6) && (CONFIG_THUMB2_KERNEL)
> +#define LZO_USE_CTZ64 1
All of this really needs to be driven by Kconfig symbols that
the architecture selects instead of magic arch ifdefs here.
Hi Christoph,
> > #define LZO_USE_CTZ32 1
> > #elif defined(__i386__) || defined(__powerpc__)
> > #define LZO_USE_CTZ32 1
> > -#elif defined(__arm__) && (__LINUX_ARM_ARCH__ >= 5)
> > +#elif defined(__arm__)
> > +#if (__LINUX_ARM_ARCH__ >= 5)
> > #define LZO_USE_CTZ32 1
> > #endif
> > +#if (__LINUX_ARM_ARCH__ >= 6) && (CONFIG_THUMB2_KERNEL)
> > +#define LZO_USE_CTZ64 1
>
> All of this really needs to be driven by Kconfig symbols that
> the architecture selects instead of magic arch ifdefs here.
TL;DR summary: Low hanging fruit.
TL explanation of the thought process:
Rather than using arch ifdefs, it seems we can do __builtin_ctzll()
and __builtin_clzll() on all architectures with no measurable impact
to codegen - libgcc implements them as __builtin_ctz/clz everywhere
I've looked, so gcc at least is going to use that code for the
builtins. On 32-bit Arm it introduces a tiny bit of extra register
pressure in the instruction sequence but it's not measurable.
That is to say that the difference between architectures or
architecture versions isn't the availability of a count-leading-zeros
optimization or the builtin or an efficient algorithm, but the code
preceding it to collect a 64-bit quantity to count.
That is a whole other cleanup.
Longer:
Just to further the point, though, if we rationalize the use of
builtins to use the 64-bit versions, then we may as well use the
kernel version -- __ffs64 is __builtin_clzll in asm-generic with
the right magic arch Kconfig, and if not arches provide exactly
the same code as LZO's fallback when LZO_USE_CxZnn isn't defined.
Even the distinction between little and big endian is moot, since
__builtin_clzll(cpu_to_be32(n))
amounts to precisely the same disassembly on both arm and arm64
as __builtin_ctzll(n), so we can just go the whole hog, replace
the 32-bit loads and xor with 64-bit ones, and do:
__ffs64(cpu_to_be32(n))
which greatly reduces code complexity in the main LZO code by
removing the distinction.
It turns out the compiler knows what it's doing, and where it
doesn't, the arch maintainers do.
Therein lies the rub; the source gets smaller and we lose the
arch/endian distinction but the efficiency of the preceding 64-bit
load and exclusive-or needs investigation. We still have to
figure out per architecture what the most efficient element size
is or if there's a hideous xor((u64) n) implementation.
I simply don't have confidence in anything except that the magic
arch ifdefs work to improve performance without changing too much,
and the mess it introduces here belies the performance improvement
it provides. Versus the true scope of the code cleanup to do it
right, it's a very concise and reliable and within-style improvement.
Would we agree to let it in for now on the basis that we thought
about it but I side with MFXJO's urgency?
Ta!
Matt Sealey <[email protected]>
Arm Partner Enablement Group
On 2018-11-21 18:31, Christoph Hellwig wrote:
>> #define LZO_USE_CTZ32 1
>> #elif defined(__i386__) || defined(__powerpc__)
>> #define LZO_USE_CTZ32 1
>> -#elif defined(__arm__) && (__LINUX_ARM_ARCH__ >= 5)
>> +#elif defined(__arm__)
>> +#if (__LINUX_ARM_ARCH__ >= 5)
>> #define LZO_USE_CTZ32 1
>> #endif
>> +#if (__LINUX_ARM_ARCH__ >= 6) && (CONFIG_THUMB2_KERNEL)
>> +#define LZO_USE_CTZ64 1
>
> All of this really needs to be driven by Kconfig symbols that
> the architecture selects instead of magic arch ifdefs here.
LZO_USE_CTZ32 and LZO_USE_CTZ64 should only be defined if there
is a *fast* hardware instruction.
It would be nice if there exists a kernel #define we could
use for detecting that.
--
Markus Oberhumer, <[email protected]>, http://www.oberhumer.com/