Here is an example, using comma-delimited data:
-- contents of dept.csv:
10,ACCOUNTING,NEW YORK
20,RESEARCH,DALLAS
30,SALES,CHICAGO
40,OPERATIONS,BOSTON
--
-- create Oracle table to load data into:
scott@ORA92> create table dept2
  2  (deptno number,
  3   dname  varchar2(14),
  4   loc    varchar2(13))
  5  /

Table created.
--
-- contents of SQL*Loader control file dept_csv.ctl:
load data
infile 'dept.csv'
into table dept2
fields terminated by ','
trailing nullcols
(deptno, dname, loc)
--
-- load data:
-- (can be run from operating system 
--  or from sql*plus using host command, as below)
scott@ORA92> host sqlldr scott/tiger control=dept_csv.ctl log=dept.log
--
-- results:
scott@ORA92> select * from dept2
  2  /
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON

4 rows selected.

Here is another example, using a fixed format file:
-- contents of dept.sdf:
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON
--
-- create Oracle table to load data into:
scott@ORA92> create table dept2
  2  (deptno number,
  3   dname  varchar2(14),
  4   loc    varchar2(13))
  5  /

Table created.
--
-- contents of SQL*Loader control file dept_sdf.ctl:
load data
infile 'dept.sdf'
into table dept2
trailing nullcols
(deptno position (9:10), 
 dname  position (12:25), 
 loc    position (27:35))
--
-- load data:
-- (can be run from operating system 
--  or from sql*plus using host command, as below)
scott@ORA92> host sqlldr scott/tiger control=dept_sdf.ctl log=dept.log
--
-- results:
scott@ORA92> select * from dept2
  2  /
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON

4 rows selected.