little stuff
This commit is contained in:
parent
6145fc79e5
commit
845d572b87
2 changed files with 2 additions and 1 deletions
|
@ -19,7 +19,7 @@ void calc_superblock(const uint a_offset, const uint b_offset, const uint itid,
|
|||
if (i < num_blocks_per_row) {
|
||||
sccache[ix][0][itid] = FLOAT_TYPE((data_a[ib0 + i].scales[itid8] >> v_im*4) & 0xF); // lower 8 bytes
|
||||
sccache[ix][1][itid] = FLOAT_TYPE((data_a[ib0 + i].scales[itid8+8] >> v_im*4) & 0xF); // upper 8 bytes
|
||||
}
|
||||
}
|
||||
barrier();
|
||||
|
||||
if (i >= num_blocks_per_row)
|
||||
|
|
|
@ -6,6 +6,7 @@
|
|||
layout(local_size_x_id = 0, local_size_y = 1, local_size_z = 1) in;
|
||||
|
||||
shared FLOAT_TYPE sccache[BLOCK_SIZE/16][2][8];
|
||||
|
||||
FLOAT_TYPE temp[NUM_COLS][NUM_ROWS];
|
||||
|
||||
void calc_superblock(const uint a_offset, const uint b_offset, const uint ix, const uint itid8, const uint v_im, const uint v_im4, const uint v_in, const uint32_t hm_m[4], const uint q_offset, const uint y_offset, const uint s_shift, const uint i, const uint num_blocks_per_row, const uint first_row, const uint num_rows, const bool all_threads) {
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue