Skip to content
GitLab
Projects
Groups
Snippets
/
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
SWIFT
SWIFTsim
Commits
3bcdf6eb
Commit
3bcdf6eb
authored
Mar 14, 2017
by
James Willis
Browse files
Removed cache structure needed for auto-vectorisation.
parent
fd3d64a8
Changes
1
Hide whitespace changes
Inline
Side-by-side
src/cache.h
View file @
3bcdf6eb
...
...
@@ -39,28 +39,6 @@
* properties required for density/force calculations.*/
struct
cache
{
#ifdef DOPAIR1_AUTO_VEC
float
x
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* x position*/
float
y
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* y position*/
float
z
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* z position*/
float
m
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Mass */
float
vx
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* x velocity */
float
vy
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* y velocity */
float
vz
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* z velocity */
float
h
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Smoothing length */
/*Cached arrays to hold particle updates*/
float
rho
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Density */
float
rho_dh
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Density gradient */
float
wcount
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* No. of contributions*/
float
wcount_dh
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Mass */
float
div_v
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Velocity divergence */
float
curl_vx
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Velocity curl x */
float
curl_vy
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Velocity curl y */
float
curl_vz
[
MAX_NO_OF_PARTS
]
__attribute__
((
aligned
(
sizeof
(
float
)
*
VEC_SIZE
)));
/* Velocity curl z */
int
count
;
#else
/* Particle x position. */
float
*
restrict
x
__attribute__
((
aligned
(
CACHE_ALIGN
)));
...
...
@@ -91,13 +69,8 @@ struct cache {
/* Cache size. */
int
count
;
#endif
};
#ifdef DOPAIR1_AUTO_VEC
struct
cache
ci_cache
,
cj_cache
;
#endif
/* Secondary cache struct to hold a list of interactions between two
* particles.*/
struct
c2_cache
{
...
...
@@ -260,17 +233,6 @@ __attribute__((always_inline)) INLINE void cache_read_cell_sorted(
ci_cache
->
vx
[
i
]
=
ci
->
parts
[
idx
].
v
[
0
];
ci_cache
->
vy
[
i
]
=
ci
->
parts
[
idx
].
v
[
1
];
ci_cache
->
vz
[
i
]
=
ci
->
parts
[
idx
].
v
[
2
];
#ifdef DOPAIR1_AUTO_VEC
ci_cache
->
rho
[
i
]
=
0
.
0
f
;
ci_cache
->
rho_dh
[
i
]
=
0
.
0
f
;
ci_cache
->
wcount
[
i
]
=
0
.
0
f
;
ci_cache
->
wcount_dh
[
i
]
=
0
.
0
f
;
ci_cache
->
div_v
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vx
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vy
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vz
[
i
]
=
0
.
0
f
;
#endif
}
}
...
...
@@ -305,17 +267,6 @@ __attribute__((always_inline)) INLINE void cache_read_two_cells_sorted(
ci_cache
->
vx
[
i
]
=
ci
->
parts
[
idx
].
v
[
0
];
ci_cache
->
vy
[
i
]
=
ci
->
parts
[
idx
].
v
[
1
];
ci_cache
->
vz
[
i
]
=
ci
->
parts
[
idx
].
v
[
2
];
#ifdef DOPAIR1_AUTO_VEC
ci_cache
->
rho
[
i
]
=
0
.
0
f
;
ci_cache
->
rho_dh
[
i
]
=
0
.
0
f
;
ci_cache
->
wcount
[
i
]
=
0
.
0
f
;
ci_cache
->
wcount_dh
[
i
]
=
0
.
0
f
;
ci_cache
->
div_v
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vx
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vy
[
i
]
=
0
.
0
f
;
ci_cache
->
curl_vz
[
i
]
=
0
.
0
f
;
#endif
}
#if defined(WITH_VECTORIZATION) && defined(__ICC)
...
...
@@ -332,16 +283,6 @@ __attribute__((always_inline)) INLINE void cache_read_two_cells_sorted(
cj_cache
->
vx
[
i
]
=
cj
->
parts
[
idx
].
v
[
0
];
cj_cache
->
vy
[
i
]
=
cj
->
parts
[
idx
].
v
[
1
];
cj_cache
->
vz
[
i
]
=
cj
->
parts
[
idx
].
v
[
2
];
#ifdef DOPAIR1_AUTO_VEC
cj_cache
->
rho
[
i
]
=
0
.
0
f
;
cj_cache
->
rho_dh
[
i
]
=
0
.
0
f
;
cj_cache
->
wcount
[
i
]
=
0
.
0
f
;
cj_cache
->
wcount_dh
[
i
]
=
0
.
0
f
;
cj_cache
->
div_v
[
i
]
=
0
.
0
f
;
cj_cache
->
curl_vx
[
i
]
=
0
.
0
f
;
cj_cache
->
curl_vy
[
i
]
=
0
.
0
f
;
cj_cache
->
curl_vz
[
i
]
=
0
.
0
f
;
#endif
}
}
...
...
@@ -418,40 +359,6 @@ __attribute__((always_inline)) INLINE void cache_read_two_partial_cells_sorted(
}
}
__attribute__
((
always_inline
))
INLINE
static
void
cache_write_sorted_particles
(
const
struct
cache
*
const
ci_cache
,
const
struct
cache
*
const
cj_cache
,
const
struct
cell
*
const
ci
,
const
struct
cell
*
const
cj
,
const
struct
entry
*
restrict
sort_i
,
const
struct
entry
*
restrict
sort_j
)
{
#ifdef DOPAIR1_AUTO_VEC
struct
part
*
restrict
pi
,
*
restrict
pj
;
int
idx
=
0
;
for
(
int
i
=
0
;
i
<
ci
->
count
;
i
++
)
{
idx
=
sort_i
[
i
].
i
;
pi
=
&
ci
->
parts
[
idx
];
pi
->
rho
+=
ci_cache
->
rho
[
i
];
pi
->
density
.
rho_dh
+=
ci_cache
->
rho_dh
[
i
];
pi
->
density
.
wcount
+=
ci_cache
->
wcount
[
i
];
pi
->
density
.
wcount_dh
+=
ci_cache
->
wcount_dh
[
i
];
pi
->
density
.
div_v
+=
ci_cache
->
div_v
[
i
];
pi
->
density
.
rot_v
[
0
]
+=
ci_cache
->
curl_vx
[
i
];
pi
->
density
.
rot_v
[
1
]
+=
ci_cache
->
curl_vy
[
i
];
pi
->
density
.
rot_v
[
2
]
+=
ci_cache
->
curl_vz
[
i
];
}
for
(
int
i
=
0
;
i
<
cj
->
count
;
i
++
)
{
idx
=
sort_j
[
i
].
i
;
pj
=
&
cj
->
parts
[
idx
];
pj
->
rho
+=
cj_cache
->
rho
[
i
];
pj
->
density
.
rho_dh
+=
cj_cache
->
rho_dh
[
i
];
pj
->
density
.
wcount
+=
cj_cache
->
wcount
[
i
];
pj
->
density
.
wcount_dh
+=
cj_cache
->
wcount_dh
[
i
];
pj
->
density
.
div_v
+=
cj_cache
->
div_v
[
i
];
pj
->
density
.
rot_v
[
0
]
+=
cj_cache
->
curl_vx
[
i
];
pj
->
density
.
rot_v
[
1
]
+=
cj_cache
->
curl_vy
[
i
];
pj
->
density
.
rot_v
[
2
]
+=
cj_cache
->
curl_vz
[
i
];
}
#endif
}
/* @brief Clean the memory allocated by a #cache object.
*
* @param c The #cache to clean.
...
...
Write
Preview
Supports
Markdown
0%
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment