Ask for a reused compute request before its shader module is made, and report the reuse with the prebuild.

A compute warm request the same as one built before still made its shader module, with the SPIR-V fixes, pool rebind
and marks, before the reuse table answered; the table is now asked first, all of its key being known by then. The
table is cleared under its lock, the count of reused requests is in the prebuild's and the preparation's report
instead of a line at every autosave, and NX2NAVI_COMPUTE_WARM_REUSE is a known setting, so nx2navi.ini can set it
without an "unknown key" line.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
chrg6
2026-11-12 22:54:02 -05:00
parent 9e23aec717
commit f3856c8dca
2 changed files with 12 additions and 11 deletions

View File

@@ -54,7 +54,7 @@ static const char *const known[] = {
"HDR_BLOWOUT", "HDR_CONTRAST", "HDR_EXPOSURE", "HDR_FLARE", "HDR_GAMMA", "HDR_GRADE", "HDR_HIGHLIGHT_SAT",
"HDR_HIGHLIGHTS", "HDR_HUE", "HDR_HUE_CLIP", "HDR_MATCH_SLOPE", "HDR_MIDGRAY", "HDR_SATURATION", "HDR_SHADOWS",
"HDR_TONEMAP", "HDR_TONE_SLOPE", "HIGHFPS_FIX", "INPUT", "INPUT_RECORD", "INPUT_SCRIPT", "LANG", "LIB_PREBUILD",
"BIG_BUILD_FREE_MB", "BIG_BUILD_MAX_MB", "BUILD_FREE_MB", "BUILD_RESERVE_MB", "LIB_PREBUILD_BIG", "LIB_PREBUILD_BIG_BG", "LIB_PREBUILD_KEEP", "LIB_PREBUILD_THREADS", "LIB_WARM", "LINE_WIDTH", "POOL_SHOWN_FIRST", "POOL_DATA_FIRST", "TEX_REUSE_NAMED", "TEX_OVERWRITTEN_USED", "PIPE_LINK_MS", "FRAME_HOLD_MS", "PACE_SMOOTH", "OQ_BATCH", "CLEAR_SMALL_CPU", "GATE_WAIT_MS", "LOCALE", "LOD_GUARD", "LSTICK_DEADZONE", "MEMORY", "MEM_CENSUS", "MEM_NEED_MB", "MENU_LANG", "MENU_SOUND_VOLUME", "MENU_TEST",
"BIG_BUILD_FREE_MB", "BIG_BUILD_MAX_MB", "BUILD_FREE_MB", "BUILD_RESERVE_MB", "LIB_PREBUILD_BIG", "LIB_PREBUILD_BIG_BG", "COMPUTE_WARM_REUSE", "LIB_PREBUILD_KEEP", "LIB_PREBUILD_THREADS", "LIB_WARM", "LINE_WIDTH", "POOL_SHOWN_FIRST", "POOL_DATA_FIRST", "TEX_REUSE_NAMED", "TEX_OVERWRITTEN_USED", "PIPE_LINK_MS", "FRAME_HOLD_MS", "PACE_SMOOTH", "OQ_BATCH", "CLEAR_SMALL_CPU", "GATE_WAIT_MS", "LOCALE", "LOD_GUARD", "LSTICK_DEADZONE", "MEMORY", "MEM_CENSUS", "MEM_NEED_MB", "MENU_LANG", "MENU_SOUND_VOLUME", "MENU_TEST",
"MENU_CURSOR", "MENU_CURSOR_LOG", "MENU_TEST_CLICK", "MENU_TEST_DIR", "MENU_TEST_SET", "MINIMAL_LAYOUT", "MICROPHONE", "MIRROR_GAP", "MODS_DIR", "MODS_DISABLED", "MODS_ORDER", "MODS_REBUILD", "MONITOR", "MOTION", "MOUSE_MODE", "MOUSE_DIRECT_PATCH", "MOUSE_SENS", "MOUSE_TRACE", "MOVIE_HOST", "MOVIE_PACE", "MOVIE_CLOCK", "TEST_HITCH",
"MOTION_GUARD_DEGREES", "MOTION_POLICY_TRACE", "SELECTIVE_MOTION", "SELECTIVE_MOTION_RAW", // motion_owner.c, vk_fg_pose.inc
"SKIN_MOTION", // vk_skin_motion.inc

View File

@@ -6310,10 +6310,12 @@ static int compute_warm_remember(const ComputeWarmKey *key, int completed) {
}
static void compute_warm_clear(void) {
plat_mutex_lock(&compute_warm_lock);
for (size_t i = 0; i < COMPUTE_WARM_BUCKETS; i++) {
for (ComputeWarmEntry *e = compute_warm[i], *n; e; e = n) n = e->next, free(e);
compute_warm[i] = NULL;
}
plat_mutex_unlock(&compute_warm_lock);
}
// A compute pipeline for cs in layout l; VK_NULL_HANDLE on failure. No lock needed (P.cache is driver-synchronised).
@@ -6331,12 +6333,10 @@ static VkPipeline compute_create(const GfxShader *cs, const Layout *l, int cache
if (over) code = over, words = ow;
uint32_t *fog = spirv_fog_keep_froxels(code, words);
if (fog) free(over), over = fog, code = fog;
VkShaderModule m = make_module(code, words, 0);
if (!m) return free(over), VK_NULL_HANDLE;
VkPipelineShaderStageRequiredSubgroupSizeCreateInfo rqc;
VkComputePipelineCreateInfo ci = {.sType = VK_STRUCTURE_TYPE_COMPUTE_PIPELINE_CREATE_INFO,
.stage = {.sType = VK_STRUCTURE_TYPE_PIPELINE_SHADER_STAGE_CREATE_INFO,
.stage = VK_SHADER_STAGE_COMPUTE_BIT, .module = m, .pName = "main"},
.stage = VK_SHADER_STAGE_COMPUTE_BIT, .pName = "main"},
.layout = l->layout, .flags = noopt_flag()};
if (cached_only) ci.flags |= VK_PIPELINE_CREATE_FAIL_ON_PIPELINE_COMPILE_REQUIRED_BIT;
pin_subgroup32(&ci.stage, &rqc, code, words);
@@ -6353,14 +6353,16 @@ static VkPipeline compute_create(const GfxShader *cs, const Layout *l, int cache
rt_sha256(code, (size_t)words * 4, warm_key.code);
warm_key.layout = l->layout, warm_key.words = words;
warm_key.flags = ci.flags, warm_key.stage_flags = ci.stage.flags;
if (compute_warm_remember(&warm_key, 0)) {
if (compute_warm_remember(&warm_key, 0)) { // before the module: a reused request makes none
*already_warm = 1;
__atomic_fetch_add(&compute_warm_reused, 1, __ATOMIC_RELAXED);
vkDestroyShaderModule(P.device, m, NULL);
free(over);
return VK_NULL_HANDLE;
}
}
VkShaderModule m = make_module(code, words, 0);
if (!m) return free(over), VK_NULL_HANDLE;
ci.stage.module = m;
VkPipeline pipe = VK_NULL_HANDLE;
VkPipelineCreationFeedbackCreateInfo fci;
VkPipelineCreationFeedback fb;
@@ -7435,7 +7437,8 @@ static void lb_report(const char *how) {
"%llu vertex/fragment skipped (graphics pipeline library off), %llu skipped (the driver crashed on them before, "
"shader-prepare-skip.txt), %llu large compute shaders left to their first dispatch (NX2NAVI_LIB_PREBUILD_BIG), %zu "
"queued for the background builder (NX2NAVI_LIB_PREBUILD_BIG_BG), %llu large compute shaders not built because "
"no run has dispatched them; process working set %.0f MB", how, (unsigned long long)LB.done, LB.n,
"no run has dispatched them; %llu compute requests the same as one built before, not built again "
"(NX2NAVI_COMPUTE_WARM_REUSE); process working set %.0f MB", how, (unsigned long long)LB.done, LB.n,
now_ms() - LB.t0, (unsigned long long)LB.built[0], LB.built[0] ? LB.ms[0] / LB.built[0] : 0.0,
(unsigned long long)LB.have[0], (unsigned long long)LB.built[1], LB.built[1] ? LB.ms[1] / LB.built[1] : 0.0,
(unsigned long long)LB.have[1], (unsigned long long)LB.built[2], LB.built[2] ? LB.ms[2] / LB.built[2] : 0.0,
@@ -7444,7 +7447,8 @@ static void lb_report(const char *how) {
(unsigned long long)LB.warmed[2], LB.warmed[2] ? LB.warm_ms[2] / LB.warmed[2] : 0.0,
(unsigned long long)(LB.fail[0] + LB.fail[1] + LB.fail[2]),
(unsigned long long)LB.noshader, (unsigned long long)LB.nolib, (unsigned long long)LB.skipped,
(unsigned long long)LB.deferred, BG.n, (unsigned long long)LB.big_unused, (double)mem_proc);
(unsigned long long)LB.deferred, BG.n, (unsigned long long)LB.big_unused,
(unsigned long long)__atomic_load_n(&compute_warm_reused, __ATOMIC_RELAXED), (double)mem_proc);
plat_mutex_unlock_shared(&LB.stat);
}
@@ -8615,9 +8619,6 @@ static int cache_save(uint32_t wait_ms, int exiting) {
// Returns cache_save's answer.
static int save_with(uint32_t wait_ms, int exiting) {
if (P.ready) keys_flush(), pb_flush(), cu_flush(), cf_log();
uint64_t reused = __atomic_load_n(&compute_warm_reused, __ATOMIC_RELAXED);
if (reused) rt_log("vk_pipeline: %llu duplicate compute warm requests reused a successful identical request",
(unsigned long long)reused);
if (P.ready && P.gpl) {
rt_log("vk_pipeline: gpl: %llu fast links (%llu by workers, %llu of those by the library workers for skipped draws; "
"%llu of pipelines with a feedback loop; avg %.2f ms, "