[LTP] [PATCH] ioctl: take use of TST_RETRY_FN* macro
Li Wang
liwang@redhat.com
Wed Sep 9 11:10:03 CEST 2020
On Wed, Sep 9, 2020 at 4:44 PM Jan Stancek <jstancek@redhat.com> wrote:
> > + value ? (time_delay = 30) : (time_delay = 1);
> > +
> > + ret = TST_RETRY_FN_EXP_BACKOFF(access(sys_loop_partpath, F_OK),
> > TST_RETVAL_EQ0, time_delay);
>
> Shouldn't we invert also the check when value == 0? At the moment
> there's TST_RETVAL_EQ0 for all cases, but we expect path to exist
> only for value == 1, correct?
>
Obviously yes, and I have thought a while to decrease the
waiting time for the nonzero return, but didn't realize there is also
a macro named TST_RETVAL_NOTNULL(x) can be used:).
Thanks for the suggestion, I think your method is better.
> diff --git a/testcases/kernel/syscalls/ioctl/ioctl09.c
> b/testcases/kernel/syscalls/ioctl/ioctl09.c
> index 151618df41db..dff31d58a054 100644
> --- a/testcases/kernel/syscalls/ioctl/ioctl09.c
> +++ b/testcases/kernel/syscalls/ioctl/ioctl09.c
> @@ -28,6 +28,9 @@ static void change_partition(const char *const cmd[])
> tst_brk(TBROK, "parted return %i", ret);
> }
>
> +#define RETVAL_CHECK(x) \
> + ({ value ? TST_RETVAL_EQ0(x) : TST_RETVAL_NOTNULL(x); })
> +
> static void check_partition(int part_num, bool value)
> {
> int ret;
> @@ -36,7 +39,7 @@ static void check_partition(int part_num, bool value)
> dev_num, dev_num, part_num);
> sprintf(loop_partpath, "%sp%d", dev_path, part_num);
>
> - ret = access(sys_loop_partpath, F_OK);
> + ret = TST_RETRY_FN_EXP_BACKOFF(access(sys_loop_partpath, F_OK),
> RETVAL_CHECK, 30);
> if (ret == 0)
> tst_res(value ? TPASS : TFAIL, "access %s succeeds",
> sys_loop_partpath);
> @@ -44,7 +47,7 @@ static void check_partition(int part_num, bool value)
> tst_res(value ? TFAIL : TPASS, "access %s fails",
> sys_loop_partpath);
>
> - ret = access(loop_partpath, F_OK);
> + ret = TST_RETRY_FN_EXP_BACKOFF(access(loop_partpath, F_OK),
> RETVAL_CHECK, 30);
> if (ret == 0)
> tst_res(value ? TPASS : TFAIL, "access %s succeeds",
> loop_partpath);
>
>
--
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20200909/aed24b16/attachment.htm>
More information about the ltp
mailing list