[LTP] [PATCH 1/2] syscalls/nice05: new test for nice()

zhaogongyi zhaogongyi@huawei.com
Wed Oct 26 10:54:39 CEST 2022


Hi,

> 
> Hello,
> 
> Zhao Gongyi via ltp <ltp@lists.linux.it> writes:
> 
> > Verify that user of root can decrease the nice value of the process
> > successfully by passing a lower increment value (< min. applicable
> > limits) to nice() system call.
> >
> > Signed-off-by: Zhao Gongyi <zhaogongyi@huawei.com>
> > ---
> >  runtest/syscalls                          |  1 +
> >  testcases/kernel/syscalls/nice/.gitignore |  1 +
> >  testcases/kernel/syscalls/nice/nice05.c   | 54
> +++++++++++++++++++++++
> >  3 files changed, 56 insertions(+)
> >  create mode 100644 testcases/kernel/syscalls/nice/nice05.c
> >
> > diff --git a/runtest/syscalls b/runtest/syscalls index
> > bd74373a4..7db399375 100644
> > --- a/runtest/syscalls
> > +++ b/runtest/syscalls
> > @@ -902,6 +902,7 @@ nice01 nice01
> >  nice02 nice02
> >  nice03 nice03
> >  nice04 nice04
> > +nice05 nice05
> 
> 05 has already been taken. Also this test is very similar to nice01. Maybe it
> could be added to that?

It seems better that move to nice01, thanks!

> 
> >
> >  open01 open01
> >  open01A symlink01 -T open01
> > diff --git a/testcases/kernel/syscalls/nice/.gitignore
> > b/testcases/kernel/syscalls/nice/.gitignore
> > index 9d7a1bb43..58d64779e 100644
> > --- a/testcases/kernel/syscalls/nice/.gitignore
> > +++ b/testcases/kernel/syscalls/nice/.gitignore
> > @@ -2,3 +2,4 @@
> >  /nice02
> >  /nice03
> >  /nice04
> > +/nice05
> > diff --git a/testcases/kernel/syscalls/nice/nice05.c
> > b/testcases/kernel/syscalls/nice/nice05.c
> > new file mode 100644
> > index 000000000..85f10fadf
> > --- /dev/null
> > +++ b/testcases/kernel/syscalls/nice/nice05.c
> > @@ -0,0 +1,54 @@
> > +// SPDX-License-Identifier: GPL-2.0-or-later
> > +/*
> > + * Copyright(c) 2022 Huawei Technologies Co., Ltd
> > + * Author: Zhao Gongyi <zhaogongyi@huawei.com>  */
> > +
> > +/*\
> > + * [Description]
> > + *
> > + * Verify that user of root can decrease the nice value of
> > + * the process successfully by passing a lower increment
> > + * value (< min. applicable limits) to nice() system call.
> > + */
> > +#include <unistd.h>
> > +#include "tst_test.h"
> > +
> > +#define	NICEINC -50
> > +#define MIN_PRIO  -20
> > +#define DEFAULT_PRIO 0
> > +
> > +static void verify_nice(void)
> > +{
> > +	int new_nice;
> > +
> > +	TEST(nice(NICEINC));
> 
> Why not use one of the TST_EXP* macros?

The errno seems difficult to check in TST_EXP*, TST_EXP_PASS have no checking of errno.

> 
> > +	if (TST_RET == -1) {
> > +		tst_res(TFAIL | TTERRNO, "nice(%d) returned -1", NICEINC);
> > +		return;
> > +	}
> > +
> > +	if (TST_ERR) {
> > +		tst_res(TFAIL | TTERRNO, "nice(%d) failed", NICEINC);
> > +		return;
> > +	}
> > +
> > +	new_nice = SAFE_GETPRIORITY(PRIO_PROCESS, 0);
> > +
> > +	if (new_nice != MIN_PRIO) {
> > +		tst_res(TFAIL,
> > +			"Process priority %i, expected %i", new_nice, MIN_PRIO);
> > +		return;
> > +	}
> > +
> > +	tst_res(TPASS, "nice(%d) passed", NICEINC);
> > +
> > +	TEST(nice(DEFAULT_PRIO));
> > +	if (TST_ERR)
> > +		tst_brk(TBROK | TTERRNO, "nice(%d) failed",
> > DEFAULT_PRIO);
> 
> Again, could use TST_EXP* macro. Also just checking TST_ERR is not strictly
> correct.

Similialy, TST_EXP_PASS has no checking of errno also, and it seems that check the return value is enough since the return value is not -1.

Regards,
Gongyi


More information about the ltp mailing list