[lkml]   [2019]   [Dec]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[PATCH] smp: Allow smp_call_function_single_async() to insert locked csd
Previously we will raise an warning if we want to insert a csd object
which is with the LOCK flag set, and if it happens we'll also wait for
the lock to be released. However, this operation does not match
perfectly with how the function is named - the name with "_async"
suffix hints that this function should not block, while we will.

This patch changed this behavior by simply return -EBUSY instead of
waiting, at the meantime we allow this operation to happen without
warning the user to change this into a feature when the caller wants
to "insert a csd object, if it's there, just wait for that one".

This is pretty safe because in flush_smp_call_function_queue() for
async csd objects (where csd->flags&SYNC is zero) we'll first do the
unlock then we call the csd->func(). So if we see the csd->flags&LOCK
is true in smp_call_function_single_async(), then it's guaranteed that
csd->func() will be called after this smp_call_function_single_async()
returns -EBUSY.

Update the comment of the function too to refect this.

CC: Marcelo Tosatti <>
CC: Thomas Gleixner <>
CC: Nadav Amit <>
CC: Josh Poimboeuf <>
CC: Greg Kroah-Hartman <>
CC: Peter Zijlstra <>
Signed-off-by: Peter Xu <>

The story starts from a test where we've encountered the WARN_ON() on
a customized kernel and the csd_wait() took merely forever to
complete (so we've got a WARN_ON plusing a hang host). The current
solution (which is downstream-only for now) is that from the caller's
side we use a boolean to store whether the csd is executed, we do:

if (atomic_cmpxchg(&in_progress, 0, 1))

While at the end of csd->func() we clear the bit. However imho that's
mostly what csd->flags&LOCK is doing. So I'm thinking maybe it would
worth it to raise this patch for upstream too so that it might help
other users of smp_call_function_single_async() when they need the
same semantic (and, I do think we shouldn't wait in _async()s...)

Signed-off-by: Peter Xu <>
kernel/smp.c | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/kernel/smp.c b/kernel/smp.c
index 7dbcb402c2fc..dd31e8228218 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -329,6 +329,11 @@ EXPORT_SYMBOL(smp_call_function_single);
* (ie: embedded in an object) and is responsible for synchronizing it
* such that the IPIs performed on the @csd are strictly serialized.
+ * If the function is called with one csd which has not yet been
+ * processed by previous call to smp_call_function_single_async(), the
+ * function will return immediately with -EBUSY showing that the csd
+ * object is still in progress.
+ *
* NOTE: Be careful, there is unfortunately no current debugging facility to
* validate the correctness of this serialization.
@@ -338,14 +343,17 @@ int smp_call_function_single_async(int cpu, call_single_data_t *csd)


- /* We could deadlock if we have to wait here with interrupts disabled! */
- if (WARN_ON_ONCE(csd->flags & CSD_FLAG_LOCK))
- csd_lock_wait(csd);
+ if (csd->flags & CSD_FLAG_LOCK) {
+ err = -EBUSY;
+ goto out;
+ }

csd->flags = CSD_FLAG_LOCK;

err = generic_exec_single(cpu, csd, csd->func, csd->info);

return err;
 \ /
  Last update: 2019-12-04 21:48    [W:0.055 / U:0.644 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site