rpms/zlib
7
0
Fork 0
Code Issues Pull Requests Releases Activity
a4a3718ffd
zlib/zlib-1.2.11-IBM-Z-hw-accelrated-deflate-fix.patch

517 lines
19 KiB
Diff
Raw Normal View History

RHEL 9.0.0 Alpha bootstrap The content of this branch was automatically imported from Fedora ELN with the following as its source: https://src.fedoraproject.org/rpms/zlib#c440ac9c208c74536e4b43d0fbb7cc8339a73509
2020-11-18 18:03:13 +00:00
From 608b71008c16ce6fbf2305145c5ffb69cd88ef59 Mon Sep 17 00:00:00 2001
From: Ondrej Dubaj <odubaj@redhat.com>
Date: Fri, 7 Aug 2020 07:12:50 +0200
Subject: [PATCH] Fix for Z hardware-accelerated deflate for s390x
---
configure | 7 +
contrib/s390/dfltcc.c | 244 +++++++++++++++++++++-------------
contrib/s390/dfltcc_deflate.h | 10 +-
deflate.c | 21 +--
4 files changed, 177 insertions(+), 105 deletions(-)
diff --git a/configure b/configure
index 66caece..bfe4386 100755
--- a/configure
+++ b/configure
@@ -114,6 +114,7 @@ case "$1" in
echo ' configure [--const] [--zprefix] [--prefix=PREFIX] [--eprefix=EXPREFIX]' | tee -a configure.log
echo ' [--static] [--64] [--libdir=LIBDIR] [--sharedlibdir=LIBDIR]' | tee -a configure.log
echo ' [--includedir=INCLUDEDIR] [--archs="-arch i386 -arch x86_64"]' | tee -a configure.log
+ echo ' [--dfltcc]' | tee -a configure.log
exit 0 ;;
-p*=* | --prefix=*) prefix=`echo $1 | sed 's/.*=//'`; shift ;;
-e*=* | --eprefix=*) exec_prefix=`echo $1 | sed 's/.*=//'`; shift ;;
@@ -137,6 +138,12 @@ case "$1" in
-c* | --const) zconst=1; shift ;;
-w* | --warn) warn=1; shift ;;
-d* | --debug) debug=1; shift ;;
+ --dfltcc)
+ CFLAGS="$CFLAGS -DDFLTCC"
+ OBJC="$OBJC dfltcc.o"
+ PIC_OBJC="$PIC_OBJC dfltcc.lo"
+ shift
+ ;;
*)
echo "unknown option: $1" | tee -a configure.log
echo "$0 --help for help" | tee -a configure.log
diff --git a/contrib/s390/dfltcc.c b/contrib/s390/dfltcc.c
index d187796..d88a0d6 100644
--- a/contrib/s390/dfltcc.c
+++ b/contrib/s390/dfltcc.c
@@ -2,12 +2,13 @@
/*
Use the following commands to build zlib with DFLTCC support:
- $ CFLAGS=-DDFLTCC ./configure
- $ make OBJA=dfltcc.o PIC_OBJA=dfltcc.lo
+ $ ./configure --dfltcc
+ $ make
*/
#define _GNU_SOURCE
#include <ctype.h>
+#include <errno.h>
#include <inttypes.h>
#include <stddef.h>
#include <stdio.h>
@@ -230,31 +231,28 @@ struct dfltcc_state {
/*
Compress.
*/
-local inline int dfltcc_are_params_ok(int level,
- uInt window_bits,
- int strategy,
- uLong level_mask);
-local inline int dfltcc_are_params_ok(level, window_bits, strategy, level_mask)
+local inline int dfltcc_can_deflate_with_params(z_streamp strm,
+ int level,
+ uInt window_bits,
+ int strategy);
+local inline int dfltcc_can_deflate_with_params(strm,
+ level,
+ window_bits,
+ strategy)
+ z_streamp strm;
int level;
uInt window_bits;
int strategy;
- uLong level_mask;
-{
- return (level_mask & (1 << level)) != 0 &&
- (window_bits == HB_BITS) &&
- (strategy == Z_FIXED || strategy == Z_DEFAULT_STRATEGY);
-}
-
-
-int ZLIB_INTERNAL dfltcc_can_deflate(strm)
- z_streamp strm;
{
deflate_state FAR *state = (deflate_state FAR *)strm->state;
struct dfltcc_state FAR *dfltcc_state = GET_DFLTCC_STATE(state);
/* Unsupported compression settings */
- if (!dfltcc_are_params_ok(state->level, state->w_bits, state->strategy,
- dfltcc_state->level_mask))
+ if ((dfltcc_state->level_mask & (1 << level)) == 0)
+ return 0;
+ if (window_bits != HB_BITS)
+ return 0;
+ if (strategy != Z_FIXED && strategy != Z_DEFAULT_STRATEGY)
return 0;
/* Unsupported hardware */
@@ -266,6 +264,17 @@ int ZLIB_INTERNAL dfltcc_can_deflate(strm)
return 1;
}
+int ZLIB_INTERNAL dfltcc_can_deflate(strm)
+ z_streamp strm;
+{
+ deflate_state FAR *state = (deflate_state FAR *)strm->state;
+
+ return dfltcc_can_deflate_with_params(strm,
+ state->level,
+ state->w_bits,
+ state->strategy);
+}
+
local void dfltcc_gdht OF((z_streamp strm));
local void dfltcc_gdht(strm)
z_streamp strm;
@@ -349,22 +358,24 @@ again:
soft_bcc = 0;
no_flush = flush == Z_NO_FLUSH;
- /* Trailing empty block. Switch to software, except when Continuation Flag
- * is set, which means that DFLTCC has buffered some output in the
- * parameter block and needs to be called again in order to flush it.
+ /* No input data. Return, except when Continuation Flag is set, which means
+ * that DFLTCC has buffered some output in the parameter block and needs to
+ * be called again in order to flush it.
*/
- if (flush == Z_FINISH && strm->avail_in == 0 && !param->cf) {
- if (param->bcf) {
- /* A block is still open, and the hardware does not support closing
- * blocks without adding data. Thus, close it manually.
- */
+ if (strm->avail_in == 0 && !param->cf) {
+ /* A block is still open, and the hardware does not support closing
+ * blocks without adding data. Thus, close it manually.
+ */
+ if (!no_flush && param->bcf) {
send_eobs(strm, param);
param->bcf = 0;
}
- return 0;
- }
-
- if (strm->avail_in == 0 && !param->cf) {
+ /* Let one of deflate_* functions write a trailing empty block. */
+ if (flush == Z_FINISH)
+ return 0;
+ /* Clear history. */
+ if (flush == Z_FULL_FLUSH)
+ param->hl = 0;
*result = need_more;
return 1;
}
@@ -418,7 +429,7 @@ again:
param->cvt = state->wrap == 2 ? CVT_CRC32 : CVT_ADLER32;
if (!no_flush)
/* We need to close a block. Always do this in software - when there is
- * no input data, the hardware will not nohor BCC. */
+ * no input data, the hardware will not honor BCC. */
soft_bcc = 1;
if (flush == Z_FINISH && !param->bcf)
/* We are about to open a BFINAL block, set Block Header Final bit
@@ -433,8 +444,8 @@ again:
param->sbb = (unsigned int)state->bi_valid;
if (param->sbb > 0)
*strm->next_out = (Bytef)state->bi_buf;
- if (param->hl)
- param->nt = 0; /* Honor history */
+ /* Honor history and check value */
+ param->nt = 0;
param->cv = state->wrap == 2 ? ZSWAP32(strm->adler) : strm->adler;
/* When opening a block, choose a Huffman-Table Type */
@@ -642,27 +653,86 @@ int ZLIB_INTERNAL dfltcc_inflate_disable(strm)
return 0;
}
-/*
- Memory management.
- DFLTCC requires parameter blocks and window to be aligned. zlib allows
- users to specify their own allocation functions, so using e.g.
- `posix_memalign' is not an option. Thus, we overallocate and take the
- aligned portion of the buffer.
-*/
+local int env_dfltcc_disabled;
+local int env_source_date_epoch;
+local unsigned long env_level_mask;
+local unsigned long env_block_size;
+local unsigned long env_block_threshold;
+local unsigned long env_dht_threshold;
+local unsigned long env_ribm;
+local uint64_t cpu_facilities[(DFLTCC_FACILITY / 64) + 1];
+local struct dfltcc_qaf_param cpu_af __attribute__((aligned(8)));
+
local inline int is_dfltcc_enabled OF((void));
local inline int is_dfltcc_enabled(void)
+{
+ if (env_dfltcc_disabled)
+ /* User has explicitly disabled DFLTCC. */
+ return 0;
+
+ return is_bit_set((const char *)cpu_facilities, DFLTCC_FACILITY);
+}
+
+local unsigned long xstrtoul OF((const char *s, unsigned long _default));
+local unsigned long xstrtoul(s, _default)
+ const char *s;
+ unsigned long _default;
+{
+ char *endptr;
+ unsigned long result;
+
+ if (!(s && *s))
+ return _default;
+ errno = 0;
+ result = strtoul(s, &endptr, 0);
+ return (errno || *endptr) ? _default : result;
+}
+
+__attribute__((constructor)) local void init_globals OF((void));
+__attribute__((constructor)) local void init_globals(void)
{
const char *env;
- uint64_t facilities[(DFLTCC_FACILITY / 64) + 1];
register char r0 __asm__("r0");
env = secure_getenv("DFLTCC");
- if (env && !strcmp(env, "0"))
- /* User has explicitly disabled DFLTCC. */
- return 0;
+
+
+ env_dfltcc_disabled = env && !strcmp(env, "0");
+
+ env = secure_getenv("SOURCE_DATE_EPOCH");
+ env_source_date_epoch = !!env;
+
+#ifndef DFLTCC_LEVEL_MASK
+#define DFLTCC_LEVEL_MASK 0x2
+#endif
+ env_level_mask = xstrtoul(secure_getenv("DFLTCC_LEVEL_MASK"),
+ DFLTCC_LEVEL_MASK);
+
+#ifndef DFLTCC_BLOCK_SIZE
+#define DFLTCC_BLOCK_SIZE 1048576
+#endif
+ env_block_size = xstrtoul(secure_getenv("DFLTCC_BLOCK_SIZE"),
+ DFLTCC_BLOCK_SIZE);
- memset(facilities, 0, sizeof(facilities));
- r0 = sizeof(facilities) / sizeof(facilities[0]) - 1;
+#ifndef DFLTCC_FIRST_FHT_BLOCK_SIZE
+#define DFLTCC_FIRST_FHT_BLOCK_SIZE 4096
+#endif
+ env_block_threshold = xstrtoul(secure_getenv("DFLTCC_FIRST_FHT_BLOCK_SIZE"),
+ DFLTCC_FIRST_FHT_BLOCK_SIZE);
+
+#ifndef DFLTCC_DHT_MIN_SAMPLE_SIZE
+#define DFLTCC_DHT_MIN_SAMPLE_SIZE 4096
+#endif
+ env_dht_threshold = xstrtoul(secure_getenv("DFLTCC_DHT_MIN_SAMPLE_SIZE"),
+ DFLTCC_DHT_MIN_SAMPLE_SIZE);
+
+#ifndef DFLTCC_RIBM
+#define DFLTCC_RIBM 0
+#endif
+ env_ribm = xstrtoul(secure_getenv("DFLTCC_RIBM"), DFLTCC_RIBM);
+
+ memset(cpu_facilities, 0, sizeof(cpu_facilities));
+ r0 = sizeof(cpu_facilities) / sizeof(cpu_facilities[0]) - 1;
/* STFLE is supported since z9-109 and only in z/Architecture mode. When
* compiling with -m31, gcc defaults to ESA mode, however, since the kernel
* is 64-bit, it's always z/Architecture mode at runtime.
@@ -671,31 +741,35 @@ local inline int is_dfltcc_enabled(void)
".machinemode zarch\n"
"stfle %[facilities]\n"
".machinemode pop\n"
- : [facilities] "=Q" (facilities)
+ : [facilities] "=Q" (cpu_facilities)
, [r0] "+r" (r0)
:
: "cc");
- return is_bit_set((const char *)facilities, DFLTCC_FACILITY);
+ /* Initialize available functions */
+ if (is_dfltcc_enabled())
+ dfltcc(DFLTCC_QAF, &cpu_af, NULL, NULL, NULL, NULL, NULL);
+ else
+ memset(&cpu_af, 0, sizeof(cpu_af));
}
+/*
+ Memory management.
+
+ DFLTCC requires parameter blocks and window to be aligned. zlib allows
+ users to specify their own allocation functions, so using e.g.
+ `posix_memalign' is not an option. Thus, we overallocate and take the
+ aligned portion of the buffer.
+*/
void ZLIB_INTERNAL dfltcc_reset(strm, size)
z_streamp strm;
uInt size;
{
struct dfltcc_state *dfltcc_state =
(struct dfltcc_state *)((char FAR *)strm->state + ALIGN_UP(size, 8));
- struct dfltcc_qaf_param *param =
- (struct dfltcc_qaf_param *)&dfltcc_state->param;
- const char *s;
- /* Initialize available functions */
- if (is_dfltcc_enabled()) {
- dfltcc(DFLTCC_QAF, param, NULL, NULL, NULL, NULL, NULL);
- memmove(&dfltcc_state->af, param, sizeof(dfltcc_state->af));
- } else
- memset(&dfltcc_state->af, 0, sizeof(dfltcc_state->af));
+ memcpy(&dfltcc_state->af, &cpu_af, sizeof(dfltcc_state->af));
- if (secure_getenv("SOURCE_DATE_EPOCH"))
+ if (env_source_date_epoch)
/* User needs reproducible results, but the output of DFLTCC_CMPR
* depends on buffers' page offsets.
*/
@@ -706,36 +780,11 @@ void ZLIB_INTERNAL dfltcc_reset(strm, size)
dfltcc_state->param.nt = 1;
/* Initialize tuning parameters */
-#ifndef DFLTCC_LEVEL_MASK
-#define DFLTCC_LEVEL_MASK 0x2
-#endif
- s = secure_getenv("DFLTCC_LEVEL_MASK");
- dfltcc_state->level_mask = (s && *s) ? strtoul(s, NULL, 0) :
- DFLTCC_LEVEL_MASK;
-#ifndef DFLTCC_BLOCK_SIZE
-#define DFLTCC_BLOCK_SIZE 1048576
-#endif
- s = secure_getenv("DFLTCC_BLOCK_SIZE");
- dfltcc_state->block_size = (s && *s) ? strtoul(s, NULL, 0) :
- DFLTCC_BLOCK_SIZE;
-#ifndef DFLTCC_FIRST_FHT_BLOCK_SIZE
-#define DFLTCC_FIRST_FHT_BLOCK_SIZE 4096
-#endif
- s = secure_getenv("DFLTCC_FIRST_FHT_BLOCK_SIZE");
- dfltcc_state->block_threshold = (s && *s) ? strtoul(s, NULL, 0) :
- DFLTCC_FIRST_FHT_BLOCK_SIZE;
-#ifndef DFLTCC_DHT_MIN_SAMPLE_SIZE
-#define DFLTCC_DHT_MIN_SAMPLE_SIZE 4096
-#endif
- s = secure_getenv("DFLTCC_DHT_MIN_SAMPLE_SIZE");
- dfltcc_state->dht_threshold = (s && *s) ? strtoul(s, NULL, 0) :
- DFLTCC_DHT_MIN_SAMPLE_SIZE;
-#ifndef DFLTCC_RIBM
-#define DFLTCC_RIBM 0
-#endif
- s = secure_getenv("DFLTCC_RIBM");
- dfltcc_state->param.ribm = (s && *s) ? strtoul(s, NULL, 0) :
- DFLTCC_RIBM;
+ dfltcc_state->level_mask = env_level_mask;
+ dfltcc_state->block_size = env_block_size;
+ dfltcc_state->block_threshold = env_block_threshold;
+ dfltcc_state->dht_threshold = env_dht_threshold;
+ dfltcc_state->param.ribm = env_ribm;
}
voidpf ZLIB_INTERNAL dfltcc_alloc_state(strm, items, size)
@@ -787,22 +836,26 @@ void ZLIB_INTERNAL dfltcc_free_window(strm, w)
/*
Switching between hardware and software compression.
+
DFLTCC does not support all zlib settings, e.g. generation of non-compressed
blocks or alternative window sizes. When such settings are applied on the
fly with deflateParams, we need to convert between hardware and software
window formats.
*/
-int ZLIB_INTERNAL dfltcc_deflate_params(strm, level, strategy)
+int ZLIB_INTERNAL dfltcc_deflate_params(strm, level, strategy, flush)
z_streamp strm;
int level;
int strategy;
+ int *flush;
{
deflate_state FAR *state = (deflate_state FAR *)strm->state;
struct dfltcc_state FAR *dfltcc_state = GET_DFLTCC_STATE(state);
struct dfltcc_param_v0 FAR *param = &dfltcc_state->param;
int could_deflate = dfltcc_can_deflate(strm);
- int can_deflate = dfltcc_are_params_ok(level, state->w_bits, strategy,
- dfltcc_state->level_mask);
+ int can_deflate = dfltcc_can_deflate_with_params(strm,
+ level,
+ state->w_bits,
+ strategy);
if (can_deflate == could_deflate)
/* We continue to work in the same mode - no changes needed */
@@ -812,8 +865,11 @@ int ZLIB_INTERNAL dfltcc_deflate_params(strm, level, strategy)
/* DFLTCC was not used yet - no changes needed */
return Z_OK;
- /* Switching between hardware and software is not implemented */
- return Z_STREAM_ERROR;
+ /* For now, do not convert between window formats - simply get rid of the
+ * old data instead.
+ */
+ *flush = Z_FULL_FLUSH;
+ return Z_OK;
}
/*
diff --git a/contrib/s390/dfltcc_deflate.h b/contrib/s390/dfltcc_deflate.h
index a129a91..de36784 100644
--- a/contrib/s390/dfltcc_deflate.h
+++ b/contrib/s390/dfltcc_deflate.h
@@ -9,7 +9,8 @@ int ZLIB_INTERNAL dfltcc_deflate OF((z_streamp strm,
block_state *result));
int ZLIB_INTERNAL dfltcc_deflate_params OF((z_streamp strm,
int level,
- int strategy));
+ int strategy,
+ int *flush));
int ZLIB_INTERNAL dfltcc_deflate_set_dictionary OF((z_streamp strm,
const Bytef *dictionary,
uInt dict_length));
@@ -29,11 +30,14 @@ int ZLIB_INTERNAL dfltcc_deflate_get_dictionary OF((z_streamp strm,
} while (0)
#define DEFLATE_RESET_KEEP_HOOK(strm) \
dfltcc_reset((strm), sizeof(deflate_state))
-#define DEFLATE_PARAMS_HOOK(strm, level, strategy) \
+#define DEFLATE_PARAMS_HOOK(strm, level, strategy, hook_flush) \
do { \
int err; \
\
- err = dfltcc_deflate_params((strm), (level), (strategy)); \
+ err = dfltcc_deflate_params((strm), \
+ (level), \
+ (strategy), \
+ (hook_flush)); \
if (err == Z_STREAM_ERROR) \
return err; \
} while (0)
diff --git a/deflate.c b/deflate.c
index b17a7dd..a80bd3e 100644
--- a/deflate.c
+++ b/deflate.c
@@ -74,7 +74,7 @@ const char deflate_copyright[] =
#define DEFLATE_SET_DICTIONARY_HOOK(strm, dict, dict_len) do {} while (0)
#define DEFLATE_GET_DICTIONARY_HOOK(strm, dict, dict_len) do {} while (0)
#define DEFLATE_RESET_KEEP_HOOK(strm) do {} while (0)
-#define DEFLATE_PARAMS_HOOK(strm, level, strategy) do {} while (0)
+#define DEFLATE_PARAMS_HOOK(strm, level, strategy, hook_flush) do {} while (0)
#define DEFLATE_BOUND_ADJUST_COMPLEN(strm, complen, sourceLen) do {} while (0)
#define DEFLATE_NEED_CONSERVATIVE_BOUND(strm) 0
#define DEFLATE_HOOK(strm, flush, bstate) 0
@@ -589,6 +589,7 @@ int ZEXPORT deflateParams(strm, level, strategy)
{
deflate_state *s;
compress_func func;
+ int hook_flush = Z_NO_FLUSH;
if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
s = strm->state;
@@ -601,13 +602,14 @@ int ZEXPORT deflateParams(strm, level, strategy)
if (level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED) {
return Z_STREAM_ERROR;
}
- DEFLATE_PARAMS_HOOK(strm, level, strategy);
+ DEFLATE_PARAMS_HOOK(strm, level, strategy, &hook_flush);
func = configuration_table[s->level].func;
- if ((strategy != s->strategy || func != configuration_table[level].func) &&
- s->high_water) {
+ if ((strategy != s->strategy || func != configuration_table[level].func ||
+ hook_flush != Z_NO_FLUSH) && s->high_water) {
/* Flush the last buffer: */
- int err = deflate(strm, Z_BLOCK);
+ int err = deflate(strm, RANK(hook_flush) > RANK(Z_BLOCK) ?
+ hook_flush : Z_BLOCK);
if (err == Z_STREAM_ERROR)
return err;
if (strm->avail_out == 0)
@@ -1065,7 +1067,6 @@ int ZEXPORT deflate (strm, flush)
}
if (flush != Z_FINISH) return Z_OK;
- if (s->wrap <= 0) return Z_STREAM_END;
/* Write the trailer */
#ifdef GZIP
@@ -1081,7 +1082,7 @@ int ZEXPORT deflate (strm, flush)
}
else
#endif
- {
+ if (s->wrap == 1) {
putShortMSB(s, (uInt)(strm->adler >> 16));
putShortMSB(s, (uInt)(strm->adler & 0xffff));
}
@@ -1090,7 +1091,11 @@ int ZEXPORT deflate (strm, flush)
* to flush the rest.
*/
if (s->wrap > 0) s->wrap = -s->wrap; /* write the trailer only once! */
- return s->pending != 0 ? Z_OK : Z_STREAM_END;
+ if (s->pending == 0) {
+ Assert(s->bi_valid == 0, "bi_buf not flushed");
+ return Z_STREAM_END;
+ }
+ return Z_OK;
}
/* ========================================================================= */
--
2.26.0
Reference in New Issue Copy Permalink