diff options
| author | Roland Scheidegger <[email protected]> | 2017-01-23 18:06:03 +0100 |
|---|---|---|
| committer | Roland Scheidegger <[email protected]> | 2017-01-24 19:12:46 +0100 |
| commit | f4df21ed9509af3d7bff3da4dc7f4b2949124b3f (patch) | |
| tree | 61ae65fce5785f025eb48f7726e297b71df3ba65 | |
| parent | 25208949d7293aa060a3416f8cf3cdb3ca1fbfdd (diff) | |
gallivm: don't try to use fast rcp for fdiv
The use of fast rcp instruction is disabled, and will always fall back
to use a division instead (1 / x). Hence, if we get a division opcode,
it doesn't make much sense trying to split that into rcp/mul.
Reviewed-by: Jose Fonseca <[email protected]>
| -rw-r--r-- | src/gallium/auxiliary/gallivm/lp_bld_arit.c | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/src/gallium/auxiliary/gallivm/lp_bld_arit.c b/src/gallium/auxiliary/gallivm/lp_bld_arit.c index 5553cb10dc2..04f86bef286 100644 --- a/src/gallium/auxiliary/gallivm/lp_bld_arit.c +++ b/src/gallium/auxiliary/gallivm/lp_bld_arit.c @@ -1372,7 +1372,9 @@ lp_build_div(struct lp_build_context *bld, return LLVMConstUDiv(a, b); } - if(((util_cpu_caps.has_sse && type.width == 32 && type.length == 4) || + /* fast rcp is disabled (just uses div), so makes no sense to try that */ + if(FALSE && + ((util_cpu_caps.has_sse && type.width == 32 && type.length == 4) || (util_cpu_caps.has_avx && type.width == 32 && type.length == 8)) && type.floating) return lp_build_mul(bld, a, lp_build_rcp(bld, b)); |