fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-23 15:30:14 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

radeonsi: use better sample locations for 4x MSAA

Browse source 
Discovered by luck. Verified with the piglit MSAA accuracy test.
It also shows that the worst case EQAA 16s4f results in very good 4x MSAA
in the worst case.

Nine might not like these positions, but they are prettier to the eye and
GL doesn't care.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
This commit is contained in:
Marek Olšák 2018-05-01 22:45:24 -04:00
parent 8d8b71ccfa
commit 01fd543c82
1 changed files with 10 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

21
src/gallium/drivers/radeonsi/si_state_msaa.c
View file

@ -92,11 +92,6 @@ static const uint32_t sample_locs_2x =
FILL_SREG(-4,-4, 4, 4, 0, 0, 0, 0); /* S2 & S3 fields are not used by 2x MSAA */ FILL_SREG(-4,-4, 4, 4, 0, 0, 0, 0); /* S2 & S3 fields are not used by 2x MSAA */
static const uint64_t centroid_priority_2x = 0x1010101010101010ull; static const uint64_t centroid_priority_2x = 0x1010101010101010ull;
/* 4x MSAA */
static const uint32_t sample_locs_4x =
FILL_SREG(-2,-6, 2, 6, -6, 2, 6,-2);
static const uint64_t centroid_priority_4x = 0x3210321032103210ull;
/* 8x MSAA */ /* 8x MSAA */
static const uint32_t sample_locs_8x[] = { static const uint32_t sample_locs_8x[] = {
FILL_SREG(-3,-5, 5, 1, -5, 5, 7,-7), FILL_SREG(-3,-5, 5, 1, -5, 5, 7,-7),
@ -106,13 +101,17 @@ static const uint32_t sample_locs_8x[] = {
}; };
static const uint64_t centroid_priority_8x = 0x3542017635420176ull; static const uint64_t centroid_priority_8x = 0x3542017635420176ull;
/* 16x MSAA */ /* 4x and 16x MSAA
static const uint32_t sample_locs_16x[] = { * (the first 4 locations happen to be optimal for 4x MSAA, better than
* the standard DX 4x locations)
*/
static const uint32_t sample_locs_4x_16x[] = {
FILL_SREG(-5,-2, 5, 3, -2, 6, 3,-5), FILL_SREG(-5,-2, 5, 3, -2, 6, 3,-5),
FILL_SREG(-7,-8, 1, 1, -6, 4, 7,-4), FILL_SREG(-7,-8, 1, 1, -6, 4, 7,-4),
FILL_SREG(-1,-3, 6, 7, -3, 2, 0,-7), FILL_SREG(-1,-3, 6, 7, -3, 2, 0,-7),
FILL_SREG(-4,-6, 2, 5, -8, 0, 4,-1), FILL_SREG(-4,-6, 2, 5, -8, 0, 4,-1),
}; };
static const uint64_t centroid_priority_4x = 0x2310231023102310ull;
static const uint64_t centroid_priority_16x = 0x497ec6b231d0fa85ull; static const uint64_t centroid_priority_16x = 0x497ec6b231d0fa85ull;
static void si_get_sample_position(struct pipe_context *ctx, unsigned sample_count, static void si_get_sample_position(struct pipe_context *ctx, unsigned sample_count,
@ -129,13 +128,13 @@ static void si_get_sample_position(struct pipe_context *ctx, unsigned sample_cou
sample_locs = &sample_locs_2x; sample_locs = &sample_locs_2x;
break; break;
case 4: case 4:
sample_locs = &sample_locs_4x; sample_locs = sample_locs_4x_16x;
break; break;
case 8: case 8:
sample_locs = sample_locs_8x; sample_locs = sample_locs_8x;
break; break;
case 16: case 16:
sample_locs = sample_locs_16x; sample_locs = sample_locs_4x_16x;
break; break;
} }
@ -183,13 +182,13 @@ void si_emit_sample_locations(struct radeon_winsys_cs *cs, int nr_samples)
si_emit_max_4_sample_locs(cs, centroid_priority_2x, sample_locs_2x); si_emit_max_4_sample_locs(cs, centroid_priority_2x, sample_locs_2x);
break; break;
case 4: case 4:
si_emit_max_4_sample_locs(cs, centroid_priority_4x, sample_locs_4x); si_emit_max_4_sample_locs(cs, centroid_priority_4x, sample_locs_4x_16x[0]);
break; break;
case 8: case 8:
si_emit_max_16_sample_locs(cs, centroid_priority_8x, sample_locs_8x, 8); si_emit_max_16_sample_locs(cs, centroid_priority_8x, sample_locs_8x, 8);
break; break;
case 16: case 16:
si_emit_max_16_sample_locs(cs, centroid_priority_16x, sample_locs_16x, 16); si_emit_max_16_sample_locs(cs, centroid_priority_16x, sample_locs_4x_16x, 16);
break; break;
} }
} }