[LTP] [Fwd: [Announce] Linux Test Project]

[Aaron Laffin <alaffin@sgi.com>]

I've taken this patch posted by Andrea Arcangeli on lk and integrated it
into fcntl09.c and fcntl10.c.  I've discovered that fcntl09 and fcntl10
are _very_ similar.  09 uses F_SETLK and 10 uses F_SETLKW.  For purposes
of maintainability, its probably worthwhile to combine them.

Back to Andrea's patch, I like the changes and will check them in soon.

--aaron

Andrea Arcangeli wrote:
> 
> I've a little fix for the fcntl09 test that was wrongly return a positive.
> 
> --- ltp-20000804-alpha/tests/fcntl09.c.~1~      Fri Aug  4 22:48:23 2000
> +++ ltp-20000804-alpha/tests/fcntl09.c  Sat Aug 12 05:14:25 2000
> @@ -155,11 +155,14 @@
>       * check looping state if -c option given
>       ***************************************************************/
>      for (lc=0; TEST_LOOPING(lc); lc++) {
> +     int type;
> +     for (type = 0; type < 2; type ++) {
> 
>         /* reset Tst_count in case we are looping. */
>         Tst_count=0;
> 
> -       flocks.l_type = F_RDLCK | F_WRLCK;
> +       flocks.l_type = type ? F_RDLCK : F_WRLCK;
> +
>         /*
>          * Call fcntl(2) with F_SETLK argument on fname
>          */
> @@ -169,8 +172,8 @@
>         if ( TEST_RETURN == -1 ) {
>             TEST_ERROR_LOG(TEST_ERRNO);
>             tst_resm(TFAIL,
> -                    "fcntl(%s, F_SETLK, &flocks) flocks.l_type = F_RDLCK | F_WRLCK Failed, errno=%d : %s",
> -                    fname, TEST_ERRNO, strerror(TEST_ERRNO));
> +                    "fcntl(%s, F_SETLK, &flocks) flocks.l_type = %s Failed, errno=%d : %s",
> +                    fname, type ? "F_RDLCK" : "F_WRLCK", TEST_ERRNO, strerror(TEST_ERRNO));
>         } else {
> 
>             /***************************************************************
> @@ -179,8 +182,8 @@
>             if ( STD_FUNCTIONAL_TEST ) {
>                 /* No Verification test, yet... */
>                 tst_resm(TPASS,
> -                        "fcntl(%s, F_SETLK, &flocks) flocks.l_type = F_RDLCK | F_WRLCK returned %d",
> -                        fname, TEST_RETURN);
> +                        "fcntl(%s, F_SETLK, &flocks) flocks.l_type = %s returned %d",
> +                        fname, type ? "F_RDLCK" : "F_WRLCK" ,TEST_RETURN);
>             }
>         }
> 
> @@ -208,6 +211,7 @@
>                          fname, TEST_RETURN);
>             }
>         }
> +     }
> 
>      }  /* End for TEST_LOOPING */
> 
> 
> 
> You can set the l_type either RDLK or WRLK but not at the same time or
> flock_to_posix_lock will rightly fail here (will get the default path).
> 
>         switch (l->l_type) {
>         case F_RDLCK:
>         case F_WRLCK:
>         case F_UNLCK:
>                 fl->fl_type = l->l_type;
>                 break;
>         default:
>                 return (0);
>         }
> 
> (the WRITE lock just locks both writes and reads)
> 
> I also did an hack to test contention of the lock, but it's not clean
> code, if you want it let me know.
> 
> Andrea
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/

-- 
Aaron Laffin
Silicon Graphics, Inc. OS Test Development
Email: alaffin@sgi.com Voice: 651-683-5756
USA/MN/CRP/F5233/SSBU