[tip:locking/urgent 3/5] kernel/futex.c:2016:18: warning: variable 'vpid' set but not used

From: kernel test robot
Date: Thu Sep 02 2021 - 18:06:51 EST

Next message:
Previous message:
Messages sorted

tree: wget # If you fix the issue, Reported-by: kernel test
All warnings (new ones
kernel/futex.c: In >> kernel/futex.c:2016:18: 2016 | int ret, vpid;
| ^~~~

vim +/vpid +2016 kernel/futex.c

52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart bab5bc9e8576388 Darren Hart 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner bab5bc9e8576388 Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart 52400ba946759af Darren Hart 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 6c23cbbd5056b15 Randy Dunlap 7b4ff1adb57ad96 Mauro 7b4ff1adb57ad96 Mauro 7b4ff1adb57ad96 Mauro 52400ba946759af Darren Hart 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 3ef240eaff36b81 Thomas Gleixner 52400ba946759af Darren Hart bab5bc9e8576388 Darren Hart 52400ba946759af Darren Hart 866293ee5422758 Thomas Gleixner 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart ab51fbab39d864f Davidlohr Bueso ab51fbab39d864f Davidlohr Bueso ab51fbab39d864f Davidlohr Bueso bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart 93d0955e6cf562d Ingo Molnar bab5bc9e8576388 Darren Hart bab5bc9e8576388 Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart dc7109aaa233d83 Thomas Gleixner dc7109aaa233d83 Thomas Gleixner dc7109aaa233d83 Thomas Gleixner dc7109aaa233d83 Thomas Gleixner dc7109aaa233d83 Thomas Gleixner a974b54036f79dd Colin Ian King dc7109aaa233d83 Thomas Gleixner 84bc4af59081ee9 Darren Hart 84bc4af59081ee9 Darren Hart 84bc4af59081ee9 Darren Hart 84bc4af59081ee9 Darren Hart 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner ^1da177e4c3f415 Linus Torvalds 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 52400ba946759af Darren Hart 866293ee5422758 Thomas Gleixner bab5bc9e8576388 Darren Hart 3ef240eaff36b81 Thomas Gleixner 866293ee5422758 Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner 4f07ec0d76f242d Thomas Gleixner beda2c7ea2c15ed Darren Hart 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 07d91ef510fb16a Thomas Gleixner 866293ee5422758 Thomas Gleixner 52400ba946759af Darren Hart 52400ba946759af Darren Hart 52400ba946759af Darren Hart
:::::: The code at line :::::: 866293ee54227584ffcb4a42
:::::: TO: Thomas Gleixner :::::: CC: Thomas Gleixner
---
0-DAY CI Kernel Test Service, Description: application/gzip Andrzej Hajda: "Re: [PATCH v3 00/16] eDP: Support probing eDP panels dynamically instead of hardcoding" Vladimir Oltean: "Re: [RFC PATCH net-next 0/3] Make the PHY library stop being so greedy when binding the generic PHY driver" by: [ date ] [ thread ] [ subject ] [ author ] href="https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git">https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git locking/urgent
ad5bfa806d8ab473c
c6f7293c28254a554 [3/5] futex: Prevent inconsistent state and exit race
(attached as .config)
(GCC) 11.2.0
href="https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross">https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
href="https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=4f07ec0d76f242d4ca0f0c0c6f7293c28254a554">https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=4f07ec0d76f242d4ca0f0c0c6f7293c28254a554
rel="nofollow" href="https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git">https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git
tip locking/urgent
f242d4ca0f0c0c6f7293c28254a554
.config to linux build tree
COMPILER=gcc-11.2.0 make.cross ARCH=riscv
kindly add following tag as appropriate
robot <lkp@xxxxxxxxx>
prefixed by >>):
function 'futex_proxy_trylock_atomic':
warning: variable 'vpid' set but not used [-Wunused-but-set-variable]
2009-04-03 1981
2009-04-03 1982 /**
2009-04-03 1983 * futex_proxy_trylock_atomic() - Attempt an atomic lock for the top waiter
2009-04-03 1984 * @pifutex: the user address of the to futex
2009-04-03 1985 * @hb1: the from futex hash bucket, must be locked by the caller
2009-04-03 1986 * @hb2: the to futex hash bucket, must be locked by the caller
2009-04-03 1987 * @key1: the from futex key
2009-04-03 1988 * @key2: the to futex key
2009-04-07 1989 * @ps: address to store the pi_state pointer
2019-11-06 1990 * @exiting: Pointer to store the task pointer of the owner task
2019-11-06 1991 * which is in the middle of exiting
2009-04-07 1992 * @set_waiters: force setting the FUTEX_WAITERS bit (1) or not (0)
2009-04-03 1993 *
2009-04-03 1994 * Try and get the lock on behalf of the top waiter if we can do it atomically.
2009-04-07 1995 * Wake the top waiter if we succeed. If the caller specified set_waiters,
2009-04-07 1996 * then direct futex_lock_pi_atomic() to force setting the FUTEX_WAITERS bit.
2009-04-07 1997 * hb1 and hb2 must be held by the caller.
2009-04-03 1998 *
2019-11-06 1999 * @exiting is only set when the return value is -EBUSY. If so, this holds
2019-11-06 2000 * a refcount on the exiting task on return and the caller needs to drop it
2019-11-06 2001 * after waiting for the exit to complete.
2019-11-06 2002 *
2013-03-05 2003 * Return:
Carvalho Chehab 2017-05-11 2004 * - 0 - failed to acquire the lock atomically;
Carvalho Chehab 2017-05-11 2005 * - >0 - acquired the lock, return value is vpid of the top_waiter
Carvalho Chehab 2017-05-11 2006 * - <0 - error
2009-04-03 2007 */
2019-11-06 2008 static int
2019-11-06 2009 futex_proxy_trylock_atomic(u32 __user *pifutex, struct futex_hash_bucket *hb1,
2019-11-06 2010 struct futex_hash_bucket *hb2, union futex_key *key1,
2019-11-06 2011 union futex_key *key2, struct futex_pi_state **ps,
2019-11-06 2012 struct task_struct **exiting, int set_waiters)
2009-04-03 2013 {
2009-04-07 2014 struct futex_q *top_waiter = NULL;
2009-04-03 2015 u32 curval;
2014-05-12 @2016 int ret, vpid;
2009-04-03 2017
2009-04-03 2018 if (get_futex_value_locked(&curval, pifutex))
2009-04-03 2019 return -EFAULT;
2009-04-03 2020
2015-06-29 2021 if (unlikely(should_fail_futex(true)))
2015-06-29 2022 return -EFAULT;
2015-06-29 2023
2009-04-07 2024 /*
2009-04-07 2025 * Find the top_waiter and determine if there are additional waiters.
2009-04-07 2026 * If the caller intends to requeue more than 1 waiter to pifutex,
2009-04-07 2027 * force futex_lock_pi_atomic() to set the FUTEX_WAITERS bit now,
2009-04-07 2028 * as we have means to handle the possible fault. If not, don't set
2021-05-12 2029 * the bit unnecessarily as it will force the subsequent unlock to enter
2009-04-07 2030 * the kernel.
2009-04-07 2031 */
2009-04-03 2032 top_waiter = futex_top_waiter(hb1, key1);
2009-04-03 2033
2009-04-03 2034 /* There are no waiters, nothing for us to do. */
2009-04-03 2035 if (!top_waiter)
2009-04-03 2036 return 0;
2009-04-03 2037
2021-08-15 2038 /*
2021-08-15 2039 * Ensure that this is a waiter sitting in futex_wait_requeue_pi()
2021-08-15 2040 * and waiting on the 'waitqueue' futex which is always !PI.
2021-08-15 2041 */
2021-08-15 2042 if (!top_waiter->rt_waiter || top_waiter->pi_state)
2021-08-18 2043 return -EINVAL;
2021-08-15 2044
2009-08-13 2045 /* Ensure we requeue to the expected futex. */
2009-08-13 2046 if (!match_futex(top_waiter->requeue_pi_key, key2))
2009-08-13 2047 return -EINVAL;
2009-08-13 2048
2021-08-15 2049 /* Ensure that this does not race against an early wakeup */
2021-08-15 2050 if (!futex_requeue_pi_prepare(top_waiter, NULL))
2021-08-15 2051 return -EAGAIN;
2021-08-15 2052
2005-04-16 2053 /*
2021-09-02 2054 * Try to take the lock for top_waiter and set the FUTEX_WAITERS bit
2021-09-02 2055 * in the contended case or if @set_waiters is true.
2021-09-02 2056 *
2021-09-02 2057 * In the contended case PI state is attached to the lock owner. If
2021-09-02 2058 * the user space lock can be acquired then PI state is attached to
2021-09-02 2059 * the new owner (@top_waiter->task) when @set_waiters is true.
2009-04-03 2060 */
2014-05-12 2061 vpid = task_pid_vnr(top_waiter->task);
2009-04-07 2062 ret = futex_lock_pi_atomic(pifutex, hb2, key2, ps, top_waiter->task,
2019-11-06 2063 exiting, set_waiters);
2014-05-12 2064 if (ret == 1) {
2021-09-02 2065 /*
2021-09-02 2066 * Lock was acquired in user space and PI state was
2021-09-02 2067 * attached to @top_waiter->task. That means state is fully
2021-09-02 2068 * consistent and the waiter can return to user space
2021-09-02 2069 * immediately after the wakeup.
2021-09-02 2070 */
2009-08-09 2071 requeue_pi_wake_futex(top_waiter, key2, hb2);
2021-08-15 2072 } else if (ret < 0) {
2021-08-15 2073 /* Rewind top_waiter::requeue_state */
2021-08-15 2074 futex_requeue_pi_complete(top_waiter, ret);
2021-08-15 2075 } else {
2021-08-15 2076 /*
2021-08-15 2077 * futex_lock_pi_atomic() did not acquire the user space
2021-08-15 2078 * futex, but managed to establish the proxy lock and pi
2021-08-15 2079 * state. top_waiter::requeue_state cannot be fixed up here
2021-08-15 2080 * because the waiter is not enqueued on the rtmutex
2021-08-15 2081 * yet. This is handled at the callsite depending on the
2021-08-15 2082 * result of rt_mutex_start_proxy_lock() which is
2021-08-15 2083 * guaranteed to be reached with this function returning 0.
2021-08-15 2084 */
2014-05-12 2085 }
2009-04-03 2086 return ret;
2009-04-03 2087 }
2009-04-03 2088
2016 was first introduced by commit
f69c1f365974ba7f futex: Add another early deadlock detection check
<tglx@xxxxxxxxxxxxx>
<tglx@xxxxxxxxxxxxx>
Intel Corporation
//lists.01.org/hyperkitty/list/kbuild-all@xxxxxxxxxxxx">https://lists.01.org/hyperkitty/list/kbuild-all@xxxxxxxxxxxx
>.config.gz

Next message: Andrzej Hajda: "Re: [PATCH v3 00/16] eDP: Support probing eDP panels dynamically instead of hardcoding"
Previous message: Vladimir Oltean: "Re: [RFC PATCH net-next 0/3] Make the PHY library stop being so greedy when binding the generic PHY driver"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]