Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ggml-vulkan: don't assert compute_ctx empty before perf timestamp

Browse Source 
Pinned async host uploads can record into compute_ctx before graph
compute, so the perf logger's opening timestamp must append after them
instead of asserting the context is expired. Without this, profiling
(GGML_VK_PERF_LOGGER) crashes whenever host-register pinning is on.

Assisted-by: Claude
This commit is contained in:
Lumpiasty
2026-07-26 21:58:43 +02:00
parent cfdc33ec83
commit 0abb8e7ece
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-vulkan/ggml-vulkan.cpp
+1 -1
View File
@@ -16685,7 +16685,7 @@ static ggml_status ggml_backend_vk_graph_compute(ggml_backend_t backend, ggml_cg
std::fill(ctx->query_nodes.begin(), ctx->query_nodes.end(), nullptr); std::fill(ctx->query_nodes.begin(), ctx->query_nodes.end(), nullptr);
std::fill(ctx->query_node_idx.begin(), ctx->query_node_idx.end(), 0); std::fill(ctx->query_node_idx.begin(), ctx->query_node_idx.end(), 0);
GGML_ASSERT(ctx->compute_ctx.expired()); // compute_ctx may hold async host uploads recorded before the graph; append the timestamp after them
compute_ctx = ggml_vk_get_compute_ctx(ctx); compute_ctx = ggml_vk_get_compute_ctx(ctx);
ctx->query_idx = 0; ctx->query_idx = 0;
compute_ctx->s->buffer->buf.writeTimestamp(vk::PipelineStageFlagBits::eAllCommands, ctx->query_pool, ctx->query_idx++); compute_ctx->s->buffer->buf.writeTimestamp(vk::PipelineStageFlagBits::eAllCommands, ctx->query_pool, ctx->query_idx++);