diff --git a/src/slurmctld/job_mgr.c b/src/slurmctld/job_mgr.c index 9834c291ed863c143cbc0e05c68cb707a28dbd0e..c225916b871aee87c8b4ab4e4975bf0d0212e3ba 100644 --- a/src/slurmctld/job_mgr.c +++ b/src/slurmctld/job_mgr.c @@ -6429,18 +6429,18 @@ extern int job_limits_check(struct job_record **job_pptr, bool check_min_time) part_ptr = job_ptr->part_ptr; qos_ptr = job_ptr->qos_ptr; assoc_ptr = job_ptr->assoc_ptr; - if (!detail_ptr) { /* To prevent CLANG error */ + if (!detail_ptr) { fatal("job %u has NULL details_ptr", job_ptr->job_id); - return WAIT_NO_REASON; + return WAIT_NO_REASON; /* To prevent CLANG error */ } fail_reason = WAIT_NO_REASON; /* * Here we need to pretend we are just submitting the job so we can - * utilize the already existing function _part_access_check. If - * anything else is ever checked in that function this will most likely - * have to be updated. + * utilize the already existing function _part_access_check. If any + * additional fields in that function are ever checked, the fields set + * below will need to be modified. */ slurm_init_job_desc_msg(&job_desc); job_desc.reservation = job_ptr->resv_name; @@ -6448,7 +6448,8 @@ extern int job_limits_check(struct job_record **job_pptr, bool check_min_time) job_desc.alloc_node = job_ptr->alloc_node; job_desc.min_cpus = detail_ptr->min_cpus; #ifdef HAVE_BG - /* The node counts have been altered to reflect slurm nodes instead of + /* + * The node counts have been altered to reflect slurm nodes instead of * cnodes, so we need to figure out the cnode count * by using the cpu counts. The partitions have been altered as well * so we have to use the original values.