diff options
| author | Rob Clark <[email protected]> | 2019-09-25 10:10:39 -0700 |
|---|---|---|
| committer | Rob Clark <[email protected]> | 2019-10-18 15:08:54 -0700 |
| commit | 0568761f8e7b9249ec8cffbd7826c24e38bf16d6 (patch) | |
| tree | 174362a432ebe93bf918d9461264a3ffbfec80cd /src/compiler/nir/nir_builder.h | |
| parent | bc2ccdc45ab6de74d72412244984b4551ece855c (diff) | |
nir: Add a new ALU nir_op_imul24
Some hardware can do 24b multiply in a single instruction, but not 32b.
However in most cases 24b is sufficient for address/offset calculation.
Signed-off-by: Rob Clark <[email protected]>
Acked-by: Jason Ekstrand <[email protected]>
Reviewed-by: Kristian H. Kristensen <[email protected]>
Reviewed-by: Eduardo Lima Mitev <[email protected]>
Diffstat (limited to 'src/compiler/nir/nir_builder.h')
0 files changed, 0 insertions, 0 deletions