Commit 2ae7cdab authored by Paolo Bonzini's avatar Paolo Bonzini Committed by Ben Hutchings

srcu: Allow use of Classic SRCU from both process and interrupt context

commit 1123a604 upstream.

Linu Cherian reported a WARN in cleanup_srcu_struct() when shutting
down a guest running iperf on a VFIO assigned device.  This happens
because irqfd_wakeup() calls srcu_read_lock(&kvm->irq_srcu) in interrupt
context, while a worker thread does the same inside kvm_set_irq().  If the
interrupt happens while the worker thread is executing __srcu_read_lock(),
updates to the Classic SRCU ->lock_count[] field or the Tree SRCU
->srcu_lock_count[] field can be lost.

The docs say you are not supposed to call srcu_read_lock() and
srcu_read_unlock() from irq context, but KVM interrupt injection happens
from (host) interrupt context and it would be nice if SRCU supported the
use case.  KVM is using SRCU here not really for the "sleepable" part,
but rather due to its IPI-free fast detection of grace periods.  It is
therefore not desirable to switch back to RCU, which would effectively
revert commit 719d93cd ("kvm/irqchip: Speed up KVM_SET_GSI_ROUTING",
2014-01-16).

However, the docs are overly conservative.  You can have an SRCU instance
only has users in irq context, and you can mix process and irq context
as long as process context users disable interrupts.  In addition,
__srcu_read_unlock() actually uses this_cpu_dec() on both Tree SRCU and
Classic SRCU.  For those two implementations, only srcu_read_lock()
is unsafe.

When Classic SRCU's __srcu_read_unlock() was changed to use this_cpu_dec(),
in commit 5a41344a ("srcu: Simplify __srcu_read_unlock() via
this_cpu_dec()", 2012-11-29), __srcu_read_lock() did two increments.
Therefore it kept __this_cpu_inc(), with preempt_disable/enable in
the caller.  Tree SRCU however only does one increment, so on most
architectures it is more efficient for __srcu_read_lock() to use
this_cpu_inc(), and any performance differences appear to be down in
the noise.

Fixes: 719d93cd ("kvm/irqchip: Speed up KVM_SET_GSI_ROUTING")
Reported-by: default avatarLinu Cherian <linuc.decode@gmail.com>
Suggested-by: default avatarLinu Cherian <linuc.decode@gmail.com>
Cc: kvm@vger.kernel.org
Signed-off-by: default avatarPaolo Bonzini <pbonzini@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: default avatarPaul E. McKenney <paulmck@linux.vnet.ibm.com>
[bwh: Backported to 3.16: __srcu_read_lock() still updates two different
 counters.  So follow what  _this_cpu_generic_to_op() does and use
 raw_local_irq_{save,restore}() and raw_cpu_ptr().]
Signed-off-by: default avatarBen Hutchings <ben@decadent.org.uk>
parent b689af65
......@@ -219,9 +219,7 @@ static inline int srcu_read_lock(struct srcu_struct *sp) __acquires(sp)
{
int retval;
preempt_disable();
retval = __srcu_read_lock(sp);
preempt_enable();
rcu_lock_acquire(&(sp)->dep_map);
return retval;
}
......
......@@ -289,17 +289,20 @@ EXPORT_SYMBOL_GPL(cleanup_srcu_struct);
/*
* Counts the new reader in the appropriate per-CPU element of the
* srcu_struct. Must be called from process context.
* srcu_struct.
* Returns an index that must be passed to the matching srcu_read_unlock().
*/
int __srcu_read_lock(struct srcu_struct *sp)
{
unsigned long flags;
int idx;
idx = ACCESS_ONCE(sp->completed) & 0x1;
ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->c[idx]) += 1;
raw_local_irq_save(flags);
ACCESS_ONCE(raw_cpu_ptr(sp->per_cpu_ref)->c[idx]) += 1;
smp_mb(); /* B */ /* Avoid leaking the critical section. */
ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->seq[idx]) += 1;
ACCESS_ONCE(raw_cpu_ptr(sp->per_cpu_ref)->seq[idx]) += 1;
raw_local_irq_restore(flags);
return idx;
}
EXPORT_SYMBOL_GPL(__srcu_read_lock);
......@@ -308,7 +311,6 @@ EXPORT_SYMBOL_GPL(__srcu_read_lock);
* Removes the count for the old reader from the appropriate per-CPU
* element of the srcu_struct. Note that this may well be a different
* CPU than that which was incremented by the corresponding srcu_read_lock().
* Must be called from process context.
*/
void __srcu_read_unlock(struct srcu_struct *sp, int idx)
{
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment