Make the bsinc table layout more efficient

The old layout separated filters, scale deltas, phase deltas, and scale phase
deltas into separate segments that each contained a numbers of scale and phase
entries, Since processing a sample needed a filter and one of each delta entry
relating to a particular scale and phase, the memory needed would be spread
across the whole table. And since subsequent samples would use a different
phase, it would jump around the table a whole lot as well.

The new layout packs the data in a way more consistent with its use. The
filters, scale deltas, phase deltas, and scale phase deltas are interleaved,
such that for a particular scale and phase, the filter and delta entries used
are contiguous. And the phase entries for a particular scale are kept together,
so the ~500 to ~1000 samples processed per source update stay within the same
3KB to 6KB area of the 70+KB table, which is much more cache friendly.

1 files changed, 4 insertions, 4 deletions

diff --git a/Alc/ALu.c b/Alc/ALu.c
index 0bf9fc11..d559649a 100644
--- a/Alc/ALu.c
+++ b/Alc/ALu.c
@@ -267,10 +267,10 @@ ALboolean BsincPrepare(const ALuint increment, BsincState *state)
      */
     for(pi = 0;pi < BSINC_PHASE_COUNT;pi++)
     {
-        state->coeffs[pi].filter  = &bsinc.Tab[bsinc.to[0][si] + bsinc.tm[0][si]*pi];
-        state->coeffs[pi].scDelta = &bsinc.Tab[bsinc.to[1][si] + bsinc.tm[1][si]*pi];
-        state->coeffs[pi].phDelta = &bsinc.Tab[bsinc.to[2][si] + bsinc.tm[0][si]*pi];
-        state->coeffs[pi].spDelta = &bsinc.Tab[bsinc.to[3][si] + bsinc.tm[1][si]*pi];
+        state->coeffs[pi].filter  = &bsinc.Tab[bsinc.filterOffset[si] + bsinc.m[si]*(pi*4 + 0)];
+        state->coeffs[pi].scDelta = &bsinc.Tab[bsinc.filterOffset[si] + bsinc.m[si]*(pi*4 + 1)];
+        state->coeffs[pi].phDelta = &bsinc.Tab[bsinc.filterOffset[si] + bsinc.m[si]*(pi*4 + 2)];
+        state->coeffs[pi].spDelta = &bsinc.Tab[bsinc.filterOffset[si] + bsinc.m[si]*(pi*4 + 3)];
     }
     return uncut;
 }