diff options
author | Michael Jones <michael_p_jones@apple.com> | 2021-10-14 15:53:40 +0300 |
---|---|---|
committer | Michael Jones <michael_p_jones@apple.com> | 2021-10-14 18:14:43 +0300 |
commit | a0f269f682dab848afc80cd322d04a0c4a815cae (patch) | |
tree | 0978b1888273fbaa2d14550bde484c5247fa89ff /intern/cycles/kernel/bvh/bvh.h | |
parent | 47caeb8c26686e24ea7e694f94fabee44f3d2dca (diff) |
Cycles: Kernel address space changes for MSL
This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation.
MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness.
The vast majority of deltas in this patch fall into one of two cases:
- Ensuring ccl_private is specified for thread-local pointer types
- Ensuring ccl_global is specified for device-wide pointer types
Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant.
In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture.
The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation.
Ref T92212
Reviewed By: brecht
Maniphest Tasks: T92212
Differential Revision: https://developer.blender.org/D12864
Diffstat (limited to 'intern/cycles/kernel/bvh/bvh.h')
-rw-r--r-- | intern/cycles/kernel/bvh/bvh.h | 36 |
1 files changed, 18 insertions, 18 deletions
diff --git a/intern/cycles/kernel/bvh/bvh.h b/intern/cycles/kernel/bvh/bvh.h index 0b44cc5db34..8f6dcd0adb9 100644 --- a/intern/cycles/kernel/bvh/bvh.h +++ b/intern/cycles/kernel/bvh/bvh.h @@ -139,7 +139,7 @@ CCL_NAMESPACE_BEGIN #endif /* __KERNEL_OPTIX__ */ -ccl_device_inline bool scene_intersect_valid(const Ray *ray) +ccl_device_inline bool scene_intersect_valid(ccl_private const Ray *ray) { /* NOTE: Due to some vectorization code non-finite origin point might * cause lots of false-positive intersections which will overflow traversal @@ -154,10 +154,10 @@ ccl_device_inline bool scene_intersect_valid(const Ray *ray) return isfinite_safe(ray->P.x) && isfinite_safe(ray->D.x) && len_squared(ray->D) != 0.0f; } -ccl_device_intersect bool scene_intersect(const KernelGlobals *kg, - const Ray *ray, +ccl_device_intersect bool scene_intersect(ccl_global const KernelGlobals *kg, + ccl_private const Ray *ray, const uint visibility, - Intersection *isect) + ccl_private Intersection *isect) { #ifdef __KERNEL_OPTIX__ uint p0 = 0; @@ -248,11 +248,11 @@ ccl_device_intersect bool scene_intersect(const KernelGlobals *kg, } #ifdef __BVH_LOCAL__ -ccl_device_intersect bool scene_intersect_local(const KernelGlobals *kg, - const Ray *ray, - LocalIntersection *local_isect, +ccl_device_intersect bool scene_intersect_local(ccl_global const KernelGlobals *kg, + ccl_private const Ray *ray, + ccl_private LocalIntersection *local_isect, int local_object, - uint *lcg_state, + ccl_private uint *lcg_state, int max_hits) { # ifdef __KERNEL_OPTIX__ @@ -360,12 +360,12 @@ ccl_device_intersect bool scene_intersect_local(const KernelGlobals *kg, #endif #ifdef __SHADOW_RECORD_ALL__ -ccl_device_intersect bool scene_intersect_shadow_all(const KernelGlobals *kg, - const Ray *ray, - Intersection *isect, +ccl_device_intersect bool scene_intersect_shadow_all(ccl_global const KernelGlobals *kg, + ccl_private const Ray *ray, + ccl_private Intersection *isect, uint visibility, uint max_hits, - uint *num_hits) + ccl_private uint *num_hits) { # ifdef __KERNEL_OPTIX__ uint p0 = ((uint64_t)isect) & 0xFFFFFFFF; @@ -445,9 +445,9 @@ ccl_device_intersect bool scene_intersect_shadow_all(const KernelGlobals *kg, #endif /* __SHADOW_RECORD_ALL__ */ #ifdef __VOLUME__ -ccl_device_intersect bool scene_intersect_volume(const KernelGlobals *kg, - const Ray *ray, - Intersection *isect, +ccl_device_intersect bool scene_intersect_volume(ccl_global const KernelGlobals *kg, + ccl_private const Ray *ray, + ccl_private Intersection *isect, const uint visibility) { # ifdef __KERNEL_OPTIX__ @@ -507,9 +507,9 @@ ccl_device_intersect bool scene_intersect_volume(const KernelGlobals *kg, #endif /* __VOLUME__ */ #ifdef __VOLUME_RECORD_ALL__ -ccl_device_intersect uint scene_intersect_volume_all(const KernelGlobals *kg, - const Ray *ray, - Intersection *isect, +ccl_device_intersect uint scene_intersect_volume_all(ccl_global const KernelGlobals *kg, + ccl_private const Ray *ray, + ccl_private Intersection *isect, const uint max_hits, const uint visibility) { |