Re: [PATCH] signalfd: don't dequeue the forced fatal signals

Re: [PATCH] signalfd: don't dequeue the forced fatal signals

[kernel test robot]

Hi Oleg,

kernel test robot noticed the following build warnings:

[auto build test WARNING on brauner-vfs/vfs.all]
[also build test WARNING on akpm-mm/mm-everything kees/for-next/pstore kees/for-next/kspp linus/master v7.0-rc7 next-20260406]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Oleg-Nesterov/signalfd-don-t-dequeue-the-forced-fatal-signals/20260407-131556
base:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git vfs.all
patch link:    https://lore.kernel.org/r/adKJMRkQJXEwHs-j%40redhat.com
patch subject: [PATCH] signalfd: don't dequeue the forced fatal signals
config: x86_64-randconfig-123-20260407 (https://download.01.org/0day-ci/archive/20260408/202604080450.mkKRp9Mk-lkp@intel.com/config)
compiler: gcc-13 (Debian 13.3.0-16) 13.3.0
sparse: v0.6.5-rc1
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260408/202604080450.mkKRp9Mk-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202604080450.mkKRp9Mk-lkp@intel.com/

sparse warnings: (new ones prefixed by >>)
>> fs/signalfd.c:53:40: sparse: sparse: incorrect type in initializer (different address spaces) @@     expected struct k_sigaction *k @@     got struct k_sigaction [noderef] __rcu * @@
   fs/signalfd.c:53:40: sparse:     expected struct k_sigaction *k
   fs/signalfd.c:53:40: sparse:     got struct k_sigaction [noderef] __rcu *
   fs/signalfd.c:69:33: sparse: sparse: incorrect type in argument 2 (different address spaces) @@     expected struct wait_queue_head [usertype] *wait_address @@     got struct wait_queue_head [noderef] __rcu * @@
   fs/signalfd.c:69:33: sparse:     expected struct wait_queue_head [usertype] *wait_address
   fs/signalfd.c:69:33: sparse:     got struct wait_queue_head [noderef] __rcu *
   fs/signalfd.c:71:31: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:71:31: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:71:31: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:76:33: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:76:33: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:76:33: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:110:32: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:128:33: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:131:33: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:135:33: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:151:32: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:155:38: sparse: sparse: cast removes address space '__user' of expression
   fs/signalfd.c:175:31: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:175:31: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:175:31: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:185:41: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:185:41: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:185:41: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:189:32: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct wait_queue_head *wq_head @@     got struct wait_queue_head [noderef] __rcu * @@
   fs/signalfd.c:189:32: sparse:     expected struct wait_queue_head *wq_head
   fs/signalfd.c:189:32: sparse:     got struct wait_queue_head [noderef] __rcu *
   fs/signalfd.c:199:41: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:199:41: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:199:41: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:201:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:201:39: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:201:39: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:204:33: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:204:33: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:204:33: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:206:35: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct wait_queue_head *wq_head @@     got struct wait_queue_head [noderef] __rcu * @@
   fs/signalfd.c:206:35: sparse:     expected struct wait_queue_head *wq_head
   fs/signalfd.c:206:35: sparse:     got struct wait_queue_head [noderef] __rcu *
   fs/signalfd.c:305:39: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:305:39: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:305:39: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:307:41: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct spinlock [usertype] *lock @@     got struct spinlock [noderef] __rcu * @@
   fs/signalfd.c:307:41: sparse:     expected struct spinlock [usertype] *lock
   fs/signalfd.c:307:41: sparse:     got struct spinlock [noderef] __rcu *
   fs/signalfd.c:309:17: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected struct wait_queue_head *wq_head @@     got struct wait_queue_head [noderef] __rcu * @@
   fs/signalfd.c:309:17: sparse:     expected struct wait_queue_head *wq_head
   fs/signalfd.c:309:17: sparse:     got struct wait_queue_head [noderef] __rcu *

vim +53 fs/signalfd.c

    50	
    51	static void mk_sigmask(struct signalfd_ctx *ctx, sigset_t *sigmask)
    52	{
  > 53		struct k_sigaction *k = current->sighand->action;
    54		int n;
    55	
    56		*sigmask = ctx->sigmask;
    57		for (n = 1; n <= _NSIG; ++n, ++k) {
    58			if (k->sa.sa_flags & SA_IMMUTABLE)
    59				sigaddset(sigmask, n);
    60		}
    61	}
    62	

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH] signalfd: don't dequeue the forced fatal signals

[Oleg Nesterov]

On 04/08, kernel test robot wrote:
>
> kernel test robot noticed the following build warnings:

...

> sparse warnings: (new ones prefixed by >>)
> >> fs/signalfd.c:53:40: sparse: sparse: incorrect type in initializer (different address spaces) @@     expected struct k_sigaction *k @@     got struct k_sigaction [noderef] __rcu * @@

...

> vim +53 fs/signalfd.c
>
>     50	
>     51	static void mk_sigmask(struct signalfd_ctx *ctx, sigset_t *sigmask)
>     52	{
>   > 53		struct k_sigaction *k = current->sighand->action;

I am going to ignore this new warning...

Yes, task_struct->sighand is __rcu. Not sure this annotation makes a lot of sense.

In any case. current->sighand is always stable. Plus task->sighand is stable under siglock.

We have a lot of (correct) non-rcu deferences of ->sighand.

I think that only lock_task_sighand() needs rcu_dereference(tsk->sighand).

Say, __exit_signal() does

	sighand = rcu_dereference_check(tsk->sighand,
					lockdep_tasklist_lock_is_held());

To me this just adds the unnecessary noise. I do not want to add another precedent.

Oleg.