myxce/src/scheduler/testsuite/test_handlers.c
Rongsong Shen e819a50600
feat(scheduler-core): implement scheduler-side admin request handler (task 9.4)
Implements the scheduler's first request-reply opcode (OPCode_schedAdmin).
sched_handle_admin_request() (handler.c) decodes SchedAdminRequest and
dispatches by action: recordStatus answers synchronously (pure read);
recordStart/recordStop are deferred onto the existing pend_updates queue
(XCE_SCHED_OP_ADMIN_RECORD) since sched_record_start()'s snapshot call
requires the session_active==0 precondition only guaranteed on the driver
thread, not the async request-handling thread. driver.c's
xce_sched_driver_translate_pending_events() performs the deferred action
and sends the reply via a new xcecli_reply_schedadmin() (client.c/client.h),
correlated by seq copied before the original request is freed.

Adversarial review found a genuine cross-thread race: sched_record_status()
read driver-owned recording state (including a pointer the driver thread
frees) from the vtask thread with no lock -- the codebase's first such
race, missed by ASan (doesn't detect races). Fixed with a dedicated
record_lock scoped strictly to in-memory field access, never spanning file
I/O; sched_record_status() now strdup()s under the lock instead of
returning a borrowed live pointer. Verified with a real two-thread
lock-contention test (independently confirmed non-vacuous by removing the
lock and observing the test fail) plus a deterministic use-after-free
reproduction.

26/26 scheduler suite, clang-format clean.
2026-08-05 14:11:43 +00:00

214 lines
7.4 KiB
C

/*
* test_handlers.c - white-box tests for the scheduler notify/dataset enqueue
* handlers (bug-765 G-6, tasks 9.2 + 9.3). Pulls the real (static) handlers in
* handler.c into this TU; enq_pend_update is local to handler.c, so we drive a
* real sched->pend_updates dataq and assert the correct op was enqueued (or
* skipped / -EINVAL) per each handler's contract.
*/
#define UNIT_TEST 1
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unity/unity.h>
#include "dataq.h"
#include "mylock.h"
#include "scheduler.h"
/* Pull the real handlers under test into this TU. handler.c is self-contained
* for the enqueue path: enq_pend_update and sched_job_state_change are both
* static within it; the dispatchers call lib helpers (linked). The
* scheduler-internal externs handler.c references but this file does not
* exercise are stubbed below: sched_init_config (config-init request path)
* and sched_get_job (job-array-single-wal-record task 5.3's per-element
* dedup check, XceData_data_jobArray — not tested here, see
* test_job_array_expand.c for the real kvmgr-backed sched_get_job). */
int sched_init_config(xce_scheduler_t* s, xce_conf4sched_t* c)
{
(void)s;
(void)c;
return 0;
}
/* sched_get_job relocated to core.h (scheduler-core spec, task 4.1) -- now
* operates on xce_sched_core_t*, not xce_scheduler_t*. */
xce_sched_job_t* sched_get_job(xce_sched_core_t* sched, uint64_t jobid)
{
(void)sched;
(void)jobid;
return NULL;
}
/* sched_record_status (recording.c, scheduler-core spec, task 9.4) --
* handler.c's OPCode_schedAdmin/recordStatus path references it; this file
* never exercises OPCode_schedAdmin (its own scope is notify/dataset
* enqueue), so a link-satisfying stub is enough, same rationale as
* sched_init_config above. */
int sched_record_status(xce_scheduler_t* sched, sched_admin_reply_t* reply)
{
(void)sched;
(void)reply;
return -EINVAL;
}
#include "../src/handler.c"
void setUp(void) { return; }
void tearDown(void)
{
printf("\n-----------------------------------------------------\n");
return;
}
static void fixture_init(xce_scheduler_t* sched)
{
memset(sched, 0, sizeof(*sched));
mylock_init(&(sched->core.lock));
sched->core.pend_updates[0] = (dataq_t*)calloc(1, sizeof(dataq_t));
dataq_init(sched->core.pend_updates[0], NULL);
sched->core.pend_updates[1] = (dataq_t*)calloc(1, sizeof(dataq_t));
dataq_init(sched->core.pend_updates[1], NULL);
}
/* Dequeue one pending request (NULL if the queue is empty) and return its op
* via *op_out; frees the request shell (payloads are test-owned/leaked under
* the suite's leaks-off convention). Returns 1 if an item was present. */
static int pop_op(xce_scheduler_t* sched, xce_sched_op_t* op_out)
{
data_item_t* it = dataq_deq(sched->core.pend_updates[0], NULL);
xce_sched_update_request_t* req;
if (it == NULL) {
return 0;
}
req = ITPTR_TO_UPDATE_REQ(it);
if (op_out != NULL) {
*op_out = req->op;
}
free(req);
return 1;
}
/* ── 9.3: notify single-item handlers (NULL -> -EINVAL, valid -> enqueue) ── */
static void test_handle_update_job_enqueues_job_fields(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
fixture_init(&sched);
TEST_ASSERT_EQUAL_INT(-EINVAL, sched_handle_sched_update_job(&sched, NULL));
TEST_ASSERT_EQUAL_INT(0, pop_op(&sched, NULL)); /* NULL enqueued nothing */
xce_job_t* job = (xce_job_t*)calloc(1, sizeof(*job));
job->jobid = 42;
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_update_job(&sched, job));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
/* 5.6: notify job update takes the in-place UPDATE_JOB_FIELDS path, NOT
* the dataset bulk-create UPDATE_JOB path. */
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_UPDATE_JOB_FIELDS, op);
free(job);
}
static void test_handle_update_queue_enqueues_queue(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
fixture_init(&sched);
TEST_ASSERT_EQUAL_INT(-EINVAL,
sched_handle_sched_update_queue(&sched, NULL));
xce_queue_t* q = (xce_queue_t*)calloc(1, sizeof(*q));
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_update_queue(&sched, q));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_UPDATE_QUEUE, op);
free(q);
}
static void test_handle_job_rusage_enqueues_rusage(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
fixture_init(&sched);
TEST_ASSERT_EQUAL_INT(-EINVAL, sched_handle_sched_job_rusage(&sched, NULL));
xce_job_rusages_t* ru = (xce_job_rusages_t*)calloc(1, sizeof(*ru));
ru->jobid = 7;
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_job_rusage(&sched, ru));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_JOB_RUSAGE, op);
free(ru);
}
static void test_handle_update_resvals_skips_empty_enqueues_nonempty(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
xce_resval_list_t empty = { .count = 0, .vals = NULL };
xce_resval_list_t full = { .count = 1, .vals = NULL };
fixture_init(&sched);
/* NULL or count==0 -> skip (return 0, nothing enqueued) */
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_update_resvals(&sched, NULL));
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_update_resvals(&sched, &empty));
TEST_ASSERT_EQUAL_INT(0, pop_op(&sched, NULL));
/* count>0 -> enqueue UPDATE_RESVALS */
TEST_ASSERT_EQUAL_INT(0, sched_handle_sched_update_resvals(&sched, &full));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_UPDATE_RESVALS, op);
}
/* ── 9.2: dataset handlers (skip on NULL/empty, enqueue on valid) ── */
static void test_handle_resources_skips_empty_enqueues_nonempty(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
xce_resource_t* one[1];
fixture_init(&sched);
TEST_ASSERT_EQUAL_INT(0, sched_handle_resources(&sched, 0, NULL));
TEST_ASSERT_EQUAL_INT(0, pop_op(&sched, NULL));
one[0] = (xce_resource_t*)calloc(1, sizeof(xce_resource_t));
TEST_ASSERT_EQUAL_INT(0, sched_handle_resources(&sched, 1, one));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_UPDATE_RESOURCES, op);
free(one[0]);
}
static void test_handle_resvals_full_skips_empty_enqueues_nonempty(void)
{
xce_scheduler_t sched;
xce_sched_op_t op = (xce_sched_op_t)-1;
xce_resval_t* one[1];
fixture_init(&sched);
TEST_ASSERT_EQUAL_INT(0, sched_handle_resvals(&sched, 0, NULL));
TEST_ASSERT_EQUAL_INT(0, pop_op(&sched, NULL));
one[0] = (xce_resval_t*)calloc(1, sizeof(xce_resval_t));
TEST_ASSERT_EQUAL_INT(0, sched_handle_resvals(&sched, 1, one));
TEST_ASSERT_EQUAL_INT(1, pop_op(&sched, &op));
TEST_ASSERT_EQUAL_INT(XCE_SCHED_OP_UPDATE_RESVALS_FULL, op);
free(one[0]);
}
int main(int argc, char* argv[])
{
(void)argc;
(void)argv;
UNITY_BEGIN();
RUN_TEST(test_handle_update_job_enqueues_job_fields, 0);
RUN_TEST(test_handle_update_queue_enqueues_queue, 0);
RUN_TEST(test_handle_job_rusage_enqueues_rusage, 0);
RUN_TEST(test_handle_update_resvals_skips_empty_enqueues_nonempty, 0);
RUN_TEST(test_handle_resources_skips_empty_enqueues_nonempty, 0);
RUN_TEST(test_handle_resvals_full_skips_empty_enqueues_nonempty, 0);
return UNITY_END();
}