commit 561b7521370fc0f3c97b86aa71005baa9fea08ce
parent 8f6ac739c35c1d94adc504eb74a41b64d94a953f
Author: Sebastiano Tronto <sebastiano@tronto.net>
Date: Sun, 6 Oct 2024 08:49:30 +0200
Fix concurrency issues
So I did not have any misconception about Atomics, I just could
not apply it where I did because the logic doesn't allow for it.
Diffstat:
3 files changed, 73 insertions(+), 44 deletions(-)
diff --git a/src/nissy.c b/src/nissy.c
@@ -75,7 +75,7 @@ checkdata(const void *buf, const tableinfo_t *info)
getdistribution_cocsep(
(uint32_t *)((char *)buf + INFOSIZE), distr);
} else if (!strncmp(info->solver, "h48", 3)) {
- getdistribution_h48((_Atomic uint8_t *)buf + INFOSIZE, distr,
+ getdistribution_h48((uint8_t *)buf + INFOSIZE, distr,
info->h48h, info->bits);
} else {
LOG("checkdata: unknown solver %s\n", info->solver);
diff --git a/src/solvers/h48/gendata_h48.h b/src/solvers/h48/gendata_h48.h
@@ -1,43 +1,27 @@
-/*
-TODO: This loop over similar h48 coordinates can be improved by only
-transforming edges, but we need to compose transformations (i.e. conjugate
-_t by _ttrep).
-*/
-#define FOREACH_H48SIM(ARG_CUBE, ARG_COCSEPDATA, ARG_SELFSIM, ARG_ACTION) \
- int64_t VAR_COCSEP = coord_cocsep(ARG_CUBE); \
- uint8_t VAR_TTREP = TTREP(ARG_COCSEPDATA[VAR_COCSEP]); \
- uint8_t VAR_INVERSE_TTREP = inverse_trans(VAR_TTREP); \
- int64_t VAR_COCLASS = COCLASS(ARG_COCSEPDATA[VAR_COCSEP]); \
- cube_t VAR_REP = transform(ARG_CUBE, VAR_TTREP); \
- uint64_t VAR_S = ARG_SELFSIM[VAR_COCLASS]; \
- for (uint8_t VAR_T = 0; VAR_T < 48 && VAR_S; VAR_T++, VAR_S >>= 1) { \
- if (!(VAR_S & 1)) continue; \
- ARG_CUBE = transform(VAR_REP, VAR_T); \
- ARG_CUBE = transform(ARG_CUBE, VAR_INVERSE_TTREP); \
- ARG_ACTION \
- }
-
STATIC uint64_t gendata_h48short(gendata_h48short_arg_t *);
STATIC size_t gendata_h48(gendata_h48_arg_t *);
STATIC size_t gendata_h48h0k4(gendata_h48_arg_t *);
STATIC size_t gendata_h48k2(gendata_h48_arg_t *);
STATIC void * gendata_h48h0k4_runthread(void *);
+STATIC_INLINE void gendata_h48_mark_atomic(gendata_h48_mark_t *);
STATIC_INLINE void gendata_h48_mark(gendata_h48_mark_t *);
STATIC_INLINE bool gendata_h48k2_dfs_stop(cube_t, int8_t, h48k2_dfs_arg_t *);
STATIC size_t gendata_h48k2_realcoord(gendata_h48_arg_t *);
STATIC void gendata_h48k2_dfs(h48k2_dfs_arg_t *arg);
STATIC void * gendata_h48k2_runthread(void *);
STATIC tableinfo_t makeinfo_h48k2(gendata_h48_arg_t *);
-STATIC void getdistribution_h48(_Atomic const uint8_t *,
+STATIC void getdistribution_h48(const uint8_t *,
uint64_t [static INFO_DISTRIBUTION_LEN], uint8_t, uint8_t);
STATIC const uint32_t *get_cocsepdata_constptr(const void *);
STATIC const uint8_t *get_h48data_constptr(const void *);
+STATIC_INLINE uint8_t get_h48_pval(const uint8_t *, int64_t, uint8_t);
+STATIC_INLINE void set_h48_pval(uint8_t *, int64_t, uint8_t, uint8_t);
STATIC_INLINE uint8_t get_h48_pval_atomic(
_Atomic const uint8_t *, int64_t, uint8_t);
-STATIC_INLINE uint8_t get_h48_pval(const uint8_t *, int64_t, uint8_t);
-STATIC_INLINE void set_h48_pval(_Atomic uint8_t *, int64_t, uint8_t, uint8_t);
+STATIC_INLINE void set_h48_pval_atomic(
+ _Atomic uint8_t *, int64_t, uint8_t, uint8_t);
STATIC_INLINE uint8_t get_h48_bound(
cube_t, uint32_t, uint8_t, uint8_t, const uint8_t *);
@@ -166,7 +150,7 @@ gendata_h48h0k4(gendata_h48_arg_t *arg)
h48max = (int64_t)H48_COORDMAX(0);
sc = coord_h48(SOLVED_CUBE, arg->cocsepdata, 0);
- set_h48_pval(table, sc, 4, 0);
+ set_h48_pval_atomic(table, sc, 4, 0);
arg->info.distribution[0] = 1;
isize = h48max / THREADS;
@@ -235,7 +219,7 @@ gendata_h48h0k4_runthread(void *arg)
.k = 4,
.cocsepdata = bfsarg->cocsepdata,
.selfsim = bfsarg->selfsim,
- .table = bfsarg->table,
+ .table_atomic = bfsarg->table,
.table_mutex = bfsarg->table_mutex,
};
@@ -259,12 +243,12 @@ gendata_h48h0k4_runthread(void *arg)
if (c <= bfsarg->depth)
continue;
markarg.cube = moved;
- gendata_h48_mark(&markarg);
+ gendata_h48_mark_atomic(&markarg);
} else {
if (c >= bfsarg->depth)
continue;
markarg.cube = cube;
- gendata_h48_mark(&markarg);
+ gendata_h48_mark_atomic(&markarg);
break; /* Enough to find one, skip the rest */
}
}
@@ -326,7 +310,7 @@ gendata_h48k2(gendata_h48_arg_t *arg)
};
uint8_t t;
- _Atomic uint8_t *table;
+ uint8_t *table;
int64_t j;
uint64_t i, ii, inext, count;
h48map_t shortcubes;
@@ -338,7 +322,7 @@ gendata_h48k2(gendata_h48_arg_t *arg)
if (arg->buf == NULL)
goto gendata_h48k2_return_size;
- table = (_Atomic uint8_t *)arg->h48buf + INFOSIZE;
+ table = (uint8_t *)arg->h48buf + INFOSIZE;
if (arg->buf != NULL)
memset(table, 0xFF, H48_TABLESIZE(arg->h, arg->k));
@@ -389,7 +373,7 @@ gendata_h48k2(gendata_h48_arg_t *arg)
h48map_destroy(&shortcubes);
for (j = 0; j < H48_COORDMAX(arg->h); j++) {
- t = get_h48_pval_atomic(table, j, 2);
+ t = get_h48_pval(table, j, 2);
arg->info.distribution[t]++;
}
@@ -517,24 +501,42 @@ gendata_h48k2_dfs(h48k2_dfs_arg_t *arg)
}
STATIC_INLINE void
-gendata_h48_mark(gendata_h48_mark_t *arg)
+gendata_h48_mark_atomic(gendata_h48_mark_t *arg)
{
uint8_t oldval, newval;
int64_t coord, mutex;
FOREACH_H48SIM(arg->cube, arg->cocsepdata, arg->selfsim,
coord = coord_h48(arg->cube, arg->cocsepdata, arg->h);
- oldval = get_h48_pval_atomic(arg->table, coord, arg->k);
+ oldval = get_h48_pval_atomic(arg->table_atomic, coord, arg->k);
newval = (uint8_t)MAX(arg->depth, 0);
if (newval < oldval) {
mutex = H48_INDEX(coord, arg->k) % CHUNKS;
pthread_mutex_lock(arg->table_mutex[mutex]);
- set_h48_pval(arg->table, coord, arg->k, newval);
+ set_h48_pval_atomic(
+ arg->table_atomic, coord, arg->k, newval);
pthread_mutex_unlock(arg->table_mutex[mutex]);
}
)
}
+STATIC_INLINE void
+gendata_h48_mark(gendata_h48_mark_t *arg)
+{
+ uint8_t oldval, newval;
+ int64_t coord, mutex;
+
+ FOREACH_H48SIM(arg->cube, arg->cocsepdata, arg->selfsim,
+ coord = coord_h48(arg->cube, arg->cocsepdata, arg->h);
+ mutex = H48_INDEX(coord, arg->k) % CHUNKS;
+ pthread_mutex_lock(arg->table_mutex[mutex]);
+ oldval = get_h48_pval(arg->table, coord, arg->k);
+ newval = (uint8_t)MAX(arg->depth, 0);
+ set_h48_pval(arg->table, coord, arg->k, MIN(newval, oldval));
+ pthread_mutex_unlock(arg->table_mutex[mutex]);
+ )
+}
+
STATIC_INLINE bool
gendata_h48k2_dfs_stop(cube_t cube, int8_t depth, h48k2_dfs_arg_t *arg)
{
@@ -548,7 +550,7 @@ gendata_h48k2_dfs_stop(cube_t cube, int8_t depth, h48k2_dfs_arg_t *arg)
coord = coord_h48(cube, arg->cocsepdata, arg->h);
mutex = H48_INDEX(coord, arg->k) % CHUNKS;
pthread_mutex_lock(arg->table_mutex[mutex]);
- oldval = get_h48_pval_atomic(arg->table, coord, arg->k);
+ oldval = get_h48_pval(arg->table, coord, arg->k);
pthread_mutex_unlock(arg->table_mutex[mutex]);
return oldval <= depth;
} else {
@@ -603,7 +605,7 @@ makeinfo_h48k2(gendata_h48_arg_t *arg)
STATIC void
getdistribution_h48(
- _Atomic const uint8_t *table,
+ const uint8_t *table,
uint64_t distr[static INFO_DISTRIBUTION_LEN],
uint8_t h,
uint8_t k
@@ -615,7 +617,7 @@ getdistribution_h48(
h48max = H48_COORDMAX(h);
for (i = 0; i < h48max; i++) {
- val = get_h48_pval_atomic(table, i, k);
+ val = get_h48_pval(table, i, k);
distr[val]++;
}
}
@@ -633,19 +635,26 @@ get_h48data_constptr(const void *data)
}
STATIC_INLINE uint8_t
-get_h48_pval_atomic(_Atomic const uint8_t *table, int64_t i, uint8_t k)
+get_h48_pval(const uint8_t *table, int64_t i, uint8_t k)
{
return (table[H48_INDEX(i, k)] & H48_MASK(i, k)) >> H48_SHIFT(i, k);
}
STATIC_INLINE uint8_t
-get_h48_pval(const uint8_t *table, int64_t i, uint8_t k)
+get_h48_pval_atomic(_Atomic const uint8_t *table, int64_t i, uint8_t k)
{
return (table[H48_INDEX(i, k)] & H48_MASK(i, k)) >> H48_SHIFT(i, k);
}
STATIC_INLINE void
-set_h48_pval(_Atomic uint8_t *table, int64_t i, uint8_t k, uint8_t val)
+set_h48_pval(uint8_t *table, int64_t i, uint8_t k, uint8_t val)
+{
+ table[H48_INDEX(i, k)] = (table[H48_INDEX(i, k)] & (~H48_MASK(i, k)))
+ | (val << H48_SHIFT(i, k));
+}
+
+STATIC_INLINE void
+set_h48_pval_atomic(_Atomic uint8_t *table, int64_t i, uint8_t k, uint8_t val)
{
table[H48_INDEX(i, k)] = (table[H48_INDEX(i, k)] & (~H48_MASK(i, k)))
| (val << H48_SHIFT(i, k));
@@ -671,7 +680,7 @@ gendata_h48_derive(uint8_t h, const void *fulltable, void *buf)
size_t cocsepsize, h48size;
uint8_t val_full, val_derive;
const uint8_t *h48full;
- _Atomic uint8_t *h48derive;
+ uint8_t *h48derive;
int64_t i, j, h48max;
gendata_h48_arg_t arg;
tableinfo_t cocsepinfo, fulltableinfo;
@@ -712,7 +721,7 @@ gendata_h48_derive(uint8_t h, const void *fulltable, void *buf)
}
h48full = (const uint8_t *)fulltable + cocsepsize + INFOSIZE;
- h48derive = (_Atomic uint8_t *)arg.h48buf + INFOSIZE;
+ h48derive = (uint8_t *)arg.h48buf + INFOSIZE;
memset(h48derive, 0xFF, H48_TABLESIZE(h, arg.k));
memset(arg.info.distribution, 0,
INFO_DISTRIBUTION_LEN * sizeof(uint64_t));
@@ -723,7 +732,7 @@ gendata_h48_derive(uint8_t h, const void *fulltable, void *buf)
LOG("Processing %" PRId64 "th coordinate\n", i);
j = i >> (int64_t)(fulltableinfo.h48h - h);
val_full = get_h48_pval(h48full, i, arg.k);
- val_derive = get_h48_pval_atomic(h48derive, j, arg.k);
+ val_derive = get_h48_pval(h48derive, j, arg.k);
set_h48_pval(
h48derive, j, arg.k, MIN(val_full, val_derive));
}
diff --git a/src/solvers/h48/gendata_types_macros.h b/src/solvers/h48/gendata_types_macros.h
@@ -22,6 +22,25 @@
#define MAXLEN 20
#define CHUNKS COCSEP_CLASSES
+/*
+TODO: This loop over similar h48 coordinates can be improved by only
+transforming edges, but we need to compose transformations (i.e. conjugate
+_t by _ttrep).
+*/
+#define FOREACH_H48SIM(ARG_CUBE, ARG_COCSEPDATA, ARG_SELFSIM, ARG_ACTION) \
+ int64_t VAR_COCSEP = coord_cocsep(ARG_CUBE); \
+ uint8_t VAR_TTREP = TTREP(ARG_COCSEPDATA[VAR_COCSEP]); \
+ uint8_t VAR_INVERSE_TTREP = inverse_trans(VAR_TTREP); \
+ int64_t VAR_COCLASS = COCLASS(ARG_COCSEPDATA[VAR_COCSEP]); \
+ cube_t VAR_REP = transform(ARG_CUBE, VAR_TTREP); \
+ uint64_t VAR_S = ARG_SELFSIM[VAR_COCLASS]; \
+ for (uint8_t VAR_T = 0; VAR_T < 48 && VAR_S; VAR_T++, VAR_S >>= 1) { \
+ if (!(VAR_S & 1)) continue; \
+ ARG_CUBE = transform(VAR_REP, VAR_T); \
+ ARG_CUBE = transform(ARG_CUBE, VAR_INVERSE_TTREP); \
+ ARG_ACTION \
+ }
+
typedef struct {
cube_t cube;
uint8_t depth;
@@ -72,7 +91,7 @@ typedef struct {
uint8_t base;
uint8_t shortdepth;
uint32_t *cocsepdata;
- _Atomic uint8_t *table;
+ uint8_t *table;
uint64_t *selfsim;
cube_t *crep;
h48map_t *shortcubes;
@@ -89,6 +108,7 @@ typedef struct {
uint8_t k;
uint32_t *cocsepdata;
uint64_t *selfsim;
- _Atomic uint8_t *table;
+ uint8_t *table;
+ _Atomic uint8_t *table_atomic;
pthread_mutex_t **table_mutex;
} gendata_h48_mark_t;
