kernel/linux.git/include/linux/rculist.h, branch linux-5.8.y

Merge branch 'proc-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace

2020-06-04T20:54:34+00:00

Pull proc updates from Eric Biederman: "This has four sets of changes: - modernize proc to support multiple private instances - ensure we see the exit of each process tid exactly - remove has_group_leader_pid - use pids not tasks in posix-cpu-timers lookup Alexey updated proc so each mount of proc uses a new superblock. This allows people to actually use mount options with proc with no fear of messing up another mount of proc. Given the kernel's internal mounts of proc for things like uml this was a real problem, and resulted in Android's hidepid mount options being ignored and introducing security issues. The rest of the changes are small cleanups and fixes that came out of my work to allow this change to proc. In essence it is swapping the pids in de_thread during exec which removes a special case the code had to handle. Then updating the code to stop handling that special case" * 'proc-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: proc: proc_pid_ns takes super_block as an argument remove the no longer needed pid_alive() check in __task_pid_nr_ns() posix-cpu-timers: Replace __get_task_for_clock with pid_for_clock posix-cpu-timers: Replace cpu_timer_pid_type with clock_pid_type posix-cpu-timers: Extend rcu_read_lock removing task_struct references signal: Remove has_group_leader_pid exec: Remove BUG_ON(has_group_leader_pid) posix-cpu-timer: Unify the now redundant code in lookup_task posix-cpu-timer: Tidy up group_leader logic in lookup_task proc: Ensure we see the exit of each process tid exactly once rculist: Add hlists_swap_heads_rcu proc: Use PIDTYPE_TGID in next_tgid Use proc_pid_ns() to get pid_namespace from the proc superblock proc: use named enums for better readability proc: use human-readable values for hidepid docs: proc: add documentation for "hidepid=4" and "subset=pid" options and new mount behavior proc: add option to mount only a pids subset proc: instantiate only pids that we can ptrace on 'hidepid=4' mount option proc: allow to mount many instances of proc in one pid namespace proc: rename struct proc_fs_info to proc_fs_opts

rculist: Add hlists_swap_heads_rcu

2020-04-28T21:08:54+00:00

Using the struct pid to refer to two tasks in de_thread was a clever idea and ultimately too clever, as it has lead to proc_flush_task being called inconsistently. To support rectifying this add hlists_swap_heads_rcu. An hlist primitive that just swaps the hlist heads of two lists. This is exactly what is needed for exchanging the pids of two tasks. Only consideration of correctness of the code has been given, as the caller is expected to be a slowpath. Link: https://lore.kernel.org/lkml/87mu6vajnq.fsf_-_@x220.int.ebiederm.org/ Acked-by: Linus Torvalds Acked-by: Oleg Nesterov Signed-off-by: "Eric W. Biederman"

Revert "rculist: Describe variadic macro argument in a Sphinx-compatible way"

2020-04-27T18:01:16+00:00

This reverts commit f452ee096d95482892b101bde4fd037fa025d3cc. The workaround became unnecessary with commit 43756e347f21 ("scripts/kernel-doc: Add support for named variable macro arguments"). Signed-off-by: Jonathan Neuschäfer Signed-off-by: Paul E. McKenney

rculist: Add brackets around cond argument in __list_check_rcu macro

2020-02-20T23:58:23+00:00

Passing a complex lockdep condition to __list_check_rcu results in false positive lockdep splat due to incorrect expression evaluation. For example, a lockdep check condition `cond1 || cond2` is evaluated as `!cond1 || cond2 && !rcu_read_lock_any_held()` which, according to operator precedence, evaluates to `!cond1 || (cond2 && !rcu_read_lock_any_held())`. This would result in a lockdep splat when cond1 is false and cond2 is true which is logically incorrect. Signed-off-by: Amol Grover Acked-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney

rculist.h: Add list_tail_rcu()

2020-01-10T22:00:58+00:00

This patch adds the macro list_tail_rcu() and documents it. Signed-off-by: Madhuparna Bhowmik [ paulmck: Reword a bit. ] Signed-off-by: Paul E. McKenney

list: Add hlist_unhashed_lockless()

2019-12-09T20:36:58+00:00

We would like to use hlist_unhashed() from timer_pending(), which runs without protection of a lock. Note that other callers might also want to use this variant. Instead of forcing a READ_ONCE() for all hlist_unhashed() callers, add a new helper with an explicit _lockless suffix in the name to better document what is going on. Also add various WRITE_ONCE() in __hlist_del(), hlist_add_head() and hlist_add_before()/hlist_add_behind() to pair with the READ_ONCE(). Signed-off-by: Eric Dumazet Cc: Thomas Gleixner [ paulmck: Also add WRITE_ONCE() to rculist.h. ] Signed-off-by: Paul E. McKenney

rculist: Describe variadic macro argument in a Sphinx-compatible way

2019-12-09T20:36:57+00:00

Without this patch, Sphinx shows "variable arguments" as the description of the cond argument, rather than the intended description, and prints the following warnings: ./include/linux/rculist.h:374: warning: Excess function parameter 'cond' description in 'list_for_each_entry_rcu' ./include/linux/rculist.h:651: warning: Excess function parameter 'cond' description in 'hlist_for_each_entry_rcu' Signed-off-by: Jonathan Neuschäfer Acked-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney

rcu: Add support for consolidated-RCU reader checking

2019-08-09T18:00:35+00:00

This commit adds RCU-reader checks to list_for_each_entry_rcu() and hlist_for_each_entry_rcu(). These checks are optional, and are indicated by a lockdep expression passed to a new optional argument to these two macros. If this optional lockdep expression is omitted, these two macros act as before, checking for an RCU read-side critical section. Signed-off-by: Joel Fernandes (Google) [ paulmck: Update to eliminate return within macro and update comment. ] Signed-off-by: Paul E. McKenney

treewide: Rename rcu_dereference_raw_notrace() to _check()

2019-08-01T21:16:21+00:00

The rcu_dereference_raw_notrace() API name is confusing. It is equivalent to rcu_dereference_raw() except that it also does sparse pointer checking. There are only a few users of rcu_dereference_raw_notrace(). This patches renames all of them to be rcu_dereference_raw_check() with the "_check()" indicating sparse checking. Signed-off-by: Joel Fernandes (Google) [ paulmck: Fix checkpatch warnings about parentheses. ] Signed-off-by: Paul E. McKenney

rcu: Clean up flavor-related definitions and comments in rculist.h

2018-08-30T23:03:31+00:00

Signed-off-by: Paul E. McKenney