Count the GPU time of copies into VRAM mirrors, so GPU-bound frames are no longer reported as CPU-bound.

A mirror prologue is closed after the part it was made for but runs first on the GPU; its timestamps were read in the wrong order, so its copies counted as idle time with work queued. ts_submitted now queues the reads in GPU order, and the perf line shows the copies on their own. From a tester's fixes: in Kokiri Forest at 4K on PCIe 3.0, 7.1 of 20.4 ms.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
chrg6
2026-11-08 15:33:48 -05:00
parent 3731155efc
commit 8c4c2f809c
3 changed files with 32 additions and 15 deletions

View File

@@ -60,6 +60,7 @@ typedef struct {
int have_idle; // calibrated timestamps: the idle split below is filled
uint64_t idle_ns, starved_ns, queued_ns; // GPU idle between timed parts; of it, before the part was submitted; submit to start
uint64_t idle_gaps, max_gap_ns; // gaps over 0.5 ms; the longest
uint64_t mirror_ns; // of gpu_ns, the copies into VRAM mirrors (vk_sync.c ts_submitted)
} PerfGpu;
extern void (*perf_gpu_stats)(PerfGpu *out);
// The performance overlay (perfmon.c): mode 0 off, 1 fps, 2 full. While it is on, perf_present hands every frame to