From 0ec87f122721b13c747279703b963242bbc23998 Mon Sep 17 00:00:00 2001 From: Sergey Sharybin Date: Wed, 28 Sep 2016 17:05:33 +0200 Subject: Cycles: Cleanup, indentation --- intern/cycles/kernel/closure/bsdf_microfacet.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'intern/cycles/kernel/closure/bsdf_microfacet.h') diff --git a/intern/cycles/kernel/closure/bsdf_microfacet.h b/intern/cycles/kernel/closure/bsdf_microfacet.h index 9da73f66da0..7c36f05b6cc 100644 --- a/intern/cycles/kernel/closure/bsdf_microfacet.h +++ b/intern/cycles/kernel/closure/bsdf_microfacet.h @@ -267,7 +267,7 @@ ccl_device bool bsdf_microfacet_merge(const ShaderClosure *a, const ShaderClosur (isequal_float3(bsdf_a->T, bsdf_b->T)) && (bsdf_a->ior == bsdf_b->ior) && ((!bsdf_a->extra && !bsdf_b->extra) || - ((bsdf_a->extra && bsdf_b->extra) && + ((bsdf_a->extra && bsdf_b->extra) && (isequal_float3(bsdf_a->extra->color, bsdf_b->extra->color)))); } -- cgit v1.2.3 From a3abb020e37a072eb71fd30de9ab125d1c16623a Mon Sep 17 00:00:00 2001 From: Brecht Van Lommel Date: Sun, 2 Oct 2016 14:48:39 +0200 Subject: Fix Cycles CUDA performance on CUDA 8.0. Mostly this is making inlining match CUDA 7.5 in a few performance critical places. The end result is that performance is now better than before, possibly due to less register spilling or other CUDA 8.0 compiler improvements. On benchmarks scenes, there are 3% to 35% render time reductions. Stack memory usage is reduced a little too. Reviewed By: sergey Differential Revision: https://developer.blender.org/D2269 --- intern/cycles/kernel/closure/bsdf_microfacet.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'intern/cycles/kernel/closure/bsdf_microfacet.h') diff --git a/intern/cycles/kernel/closure/bsdf_microfacet.h b/intern/cycles/kernel/closure/bsdf_microfacet.h index 7c36f05b6cc..0a8d14a00c2 100644 --- a/intern/cycles/kernel/closure/bsdf_microfacet.h +++ b/intern/cycles/kernel/closure/bsdf_microfacet.h @@ -183,7 +183,7 @@ ccl_device_inline void microfacet_ggx_sample_slopes( *slope_y = S * z * safe_sqrtf(1.0f + (*slope_x)*(*slope_x)); } -ccl_device_inline float3 microfacet_sample_stretched( +ccl_device_forceinline float3 microfacet_sample_stretched( KernelGlobals *kg, const float3 omega_i, const float alpha_x, const float alpha_y, const float randu, const float randv, -- cgit v1.2.3