tests : consolidate context state tests into test-llama-context (#30023)

* tests : consolidate the context state tests into test-llama-context

test-save-load-state, test-recurrent-state-rollback and
test-state-restore-fragmented check the same object from three places:
the KV cache and the recurrent state of a llama_context, saved, loaded,
copied, rolled back. Each had its own main, its own loop over the
models, its own copy of the token and the comparison helpers, and a run
of all three loaded the same models several times.

They are one binary now, and the case is the unit of the test:

  tests/llama-context/test-case.h   the interface a case implements
  tests/llama-context/cases-kv.cpp  the cases of the KV cache
  tests/llama-context/cases-rs.cpp  the cases of the recurrent state
  tests/llama-context/test-util.cpp tokens, generation, comparison
  tests/llama-context/test-runner.cpp the models of a case, on threads
  tests/llama-context/test-log.cpp  the log file of a model

A run is a table: a row per model, a column per case, a cell holds the
verdict of that pair. The models load on several threads (-j, --threads,
min(cores, 4) by default), the table is printed when it is complete and
the log of a model that failed is printed after it, in the order of the
rows. A case returns a verdict and the reason for it, and the runner is
the only one who reports - the test does not use tests/testing.h, a run
over 130 models is a matrix of verdicts and not a tree of subtests.

Staging no longer travels through the environment: llama_model_set_attn_rot
replaces the LLAMA_ATTN_ROT_DISABLE variable, which a model loaded in one
thread used to change for the models the other threads were loading, and
llama_memory_get_buffers is added for the cases that copy the buffers of
a memory. common_log_set_stdo keeps the console quiet while the table is
being built.

ggml_aligned_malloc allocates with posix_memalign when the thread
sanitizer is on. On macOS it allocates with the Mach VM, which the
sanitizer does not intercept, so it never sees a page range change hands
and reports the first write into recycled memory as a race - 30 of them
in this test, and every one of them false.

Assisted-by: pi:llama.cpp/Qwen3.8-Flash-Next

* cont : run single-threaded by default

* cont : try to build the tests on all platforms

* cont : trim down some body fat

* cont : fix

* cont : fix the fix
This commit is contained in:
Georgi Gerganov authored and GitHub committed 2026-10-13 16:21:55 +03:00
1 parent b476d9a762
commit a17e94d229
27 files changed
+2772 -1965

No files matched your search

+33
View File
@@ -1,5 +1,11 @@
#include "llama-memory.h"
#include "llama-ext.h"
#include "llama-io.h"
#include <set>
#include <vector>
llama_memory_status llama_memory_status_combine(llama_memory_status s0, llama_memory_status s1) {
bool has_update = false;
@@ -57,3 +63,30 @@ bool llama_memory_status_is_fail(llama_memory_status status) {
return false;
}
// collects the buffers of the tensors that a state write visits, without keeping the data
struct llama_memory_buffers_writer : public llama_io_write_i {
std::set<ggml_backend_buffer_t> buffers;
size_t size = 0;
void write(const void *, size_t n) override {
size += n;
}
void write_tensor(ggml_tensor * tensor, size_t, size_t n) override {
buffers.insert(tensor->buffer);
size += n;
}
size_t n_bytes() override {
return size;
}
};
std::vector<ggml_backend_buffer_t> llama_memory_get_buffers(const llama_memory_t mem) {
llama_memory_buffers_writer writer;
mem->state_write(writer);
return std::vector<ggml_backend_buffer_t>(writer.buffers.begin(), writer.buffers.end());
}