diff options
author | Jeroen Bakker <jbakker> | 2021-05-31 18:11:01 +0300 |
---|---|---|
committer | Jeroen Bakker <jeroen@blender.org> | 2021-05-31 18:11:25 +0300 |
commit | 44d2479dc36fe3409afc660eea369bea8a517987 (patch) | |
tree | 6f22a9983ca211c6759c352b8ff2ccd946e7dc23 /source/blender/draw/intern/draw_cache_inline.h | |
parent | aebeb85fe013a671f71f47d6f210c192598f408f (diff) |
Refactor: DRW Mesh Extractor: Join the extractors in a same loop
This patch replaces / redoes the entire MeshExtractors system.
Although they were useful and facilitated the addition of new buffers, they made it difficult to control the threads and added a lot of threading overhead.
Part of the problem was in traversing the same loop type in different threads. The concurrent access of the BMesh Elements slowed the reading.
This patch simplifies the use of threads by merging all the old callbacks from the extracts into a single series of iteration functions.
The type of extraction can be chosen using flags.
This optimized the process by around 34%.
Initial idea and implementation By @mano-wii.
Fine-tuning, cleanup by @atmind.
MASTER:
large_mesh_editing:
- rdata 9ms iter 50ms (frame 155ms)
- Average: 6.462874 FPS
PATCH:
large_mesh_editing:
- rdata 9ms iter 34ms (frame 136ms)
- Average: 7.379491 FPS
Differential Revision: https://developer.blender.org/D11425
Diffstat (limited to 'source/blender/draw/intern/draw_cache_inline.h')
-rw-r--r-- | source/blender/draw/intern/draw_cache_inline.h | 4 |
1 files changed, 0 insertions, 4 deletions
diff --git a/source/blender/draw/intern/draw_cache_inline.h b/source/blender/draw/intern/draw_cache_inline.h index bfc714e5d6a..6e537a3bffa 100644 --- a/source/blender/draw/intern/draw_cache_inline.h +++ b/source/blender/draw/intern/draw_cache_inline.h @@ -40,10 +40,6 @@ (flag |= DRW_ibo_requested(ibo) ? (value) : 0) #endif -/* Test and assign NULL if test fails */ -#define DRW_TEST_ASSIGN_VBO(v) (v = (DRW_vbo_requested(v) ? (v) : NULL)) -#define DRW_TEST_ASSIGN_IBO(v) (v = (DRW_ibo_requested(v) ? (v) : NULL)) - BLI_INLINE GPUBatch *DRW_batch_request(GPUBatch **batch) { /* XXX TODO(fclem): We are writing to batch cache here. Need to make this thread safe. */ |