AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

gallivm: Improve lp_build_rcp_refine.

Browse Source 
Use the alternative more accurate expression from
https://en.wikipedia.org/wiki/Division_algorithm#Newton%E2%80%93Raphson_division

v2: Use lp_build_fmuladd as suggested by Roland

Tested by enabling this code path, and running lp_test_arit.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
This commit is contained in:
Jose Fonseca
2019-05-31 17:10:40 +01:00
parent 0ec8a292fb
commit 3573412981
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/gallium/auxiliary/gallivm/lp_bld_arit.c
+6 -6
View File
@@ -2707,11 +2707,11 @@ lp_build_sqrt(struct lp_build_context *bld,
/**
* Do one Newton-Raphson step to improve reciprocate precision:
*
* x_{i+1} = x_i * (2 - a * x_i)
* x_{i+1} = x_i + x_i * (1 - a * x_i)
*
* XXX: Unfortunately this won't give IEEE-754 conformant results for 0 or
* +/-Inf, giving NaN instead. Certain applications rely on this behavior,
* such as Google Earth, which does RCP(RSQRT(0.0) when drawing the Earth's
* such as Google Earth, which does RCP(RSQRT(0.0)) when drawing the Earth's
* halo. It would be necessary to clamp the argument to prevent this.
*
* See also:
@@ -2724,12 +2724,12 @@ lp_build_rcp_refine(struct lp_build_context *bld,
LLVMValueRef rcp_a)
{
LLVMBuilderRef builder = bld->gallivm->builder;
LLVMValueRef two = lp_build_const_vec(bld->gallivm, bld->type, 2.0);
LLVMValueRef neg_a;
LLVMValueRef res;
res = LLVMBuildFMul(builder, a, rcp_a, "");
res = LLVMBuildFSub(builder, two, res, "");
res = LLVMBuildFMul(builder, rcp_a, res, "");
neg_a = LLVMBuildFNeg(builder, a, "");
res = lp_build_fmuladd(builder, neg_a, rcp_a, bld->one);
res = lp_build_fmuladd(builder, res, rcp_a, rcp_a);
return res;
}