qemu-kvm/kvm-qemu-thread-posix-cleanup-fix-document-QemuEvent.patch

147 lines
5.3 KiB
Diff
Raw Normal View History

2023-04-04 12:34:43 +00:00
From d46ca52c3f42add549bd3790a41d06594821334e Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 08:10:57 -0500
Subject: [PATCH 03/13] qemu-thread-posix: cleanup, fix, document QemuEvent
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 263: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2168472
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Commit: [3/10] 746070c4d78c7f0a9ac4456d9aee69475acb8964
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2168472
commit 9586a1329f5dce6c1d7f4de53cf0536644d7e593
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Thu Mar 2 11:19:52 2023 +0100
qemu-thread-posix: cleanup, fix, document QemuEvent
QemuEvent is currently broken on ARM due to missing memory barriers
after qatomic_*(). Apart from adding the memory barrier, a closer look
reveals some unpaired memory barriers too. Document more clearly what
is going on.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
util/qemu-thread-posix.c | 69 ++++++++++++++++++++++++++++------------
1 file changed, 49 insertions(+), 20 deletions(-)
diff --git a/util/qemu-thread-posix.c b/util/qemu-thread-posix.c
index e1225b63bd..dd3b6d4670 100644
--- a/util/qemu-thread-posix.c
+++ b/util/qemu-thread-posix.c
@@ -430,13 +430,21 @@ void qemu_event_destroy(QemuEvent *ev)
void qemu_event_set(QemuEvent *ev)
{
- /* qemu_event_set has release semantics, but because it *loads*
+ assert(ev->initialized);
+
+ /*
+ * Pairs with both qemu_event_reset() and qemu_event_wait().
+ *
+ * qemu_event_set has release semantics, but because it *loads*
* ev->value we need a full memory barrier here.
*/
- assert(ev->initialized);
smp_mb();
if (qatomic_read(&ev->value) != EV_SET) {
- if (qatomic_xchg(&ev->value, EV_SET) == EV_BUSY) {
+ int old = qatomic_xchg(&ev->value, EV_SET);
+
+ /* Pairs with memory barrier in kernel futex_wait system call. */
+ smp_mb__after_rmw();
+ if (old == EV_BUSY) {
/* There were waiters, wake them up. */
qemu_futex_wake(ev, INT_MAX);
}
@@ -445,18 +453,19 @@ void qemu_event_set(QemuEvent *ev)
void qemu_event_reset(QemuEvent *ev)
{
- unsigned value;
-
assert(ev->initialized);
- value = qatomic_read(&ev->value);
- smp_mb_acquire();
- if (value == EV_SET) {
- /*
- * If there was a concurrent reset (or even reset+wait),
- * do nothing. Otherwise change EV_SET->EV_FREE.
- */
- qatomic_or(&ev->value, EV_FREE);
- }
+
+ /*
+ * If there was a concurrent reset (or even reset+wait),
+ * do nothing. Otherwise change EV_SET->EV_FREE.
+ */
+ qatomic_or(&ev->value, EV_FREE);
+
+ /*
+ * Order reset before checking the condition in the caller.
+ * Pairs with the first memory barrier in qemu_event_set().
+ */
+ smp_mb__after_rmw();
}
void qemu_event_wait(QemuEvent *ev)
@@ -464,20 +473,40 @@ void qemu_event_wait(QemuEvent *ev)
unsigned value;
assert(ev->initialized);
- value = qatomic_read(&ev->value);
- smp_mb_acquire();
+
+ /*
+ * qemu_event_wait must synchronize with qemu_event_set even if it does
+ * not go down the slow path, so this load-acquire is needed that
+ * synchronizes with the first memory barrier in qemu_event_set().
+ *
+ * If we do go down the slow path, there is no requirement at all: we
+ * might miss a qemu_event_set() here but ultimately the memory barrier in
+ * qemu_futex_wait() will ensure the check is done correctly.
+ */
+ value = qatomic_load_acquire(&ev->value);
if (value != EV_SET) {
if (value == EV_FREE) {
/*
- * Leave the event reset and tell qemu_event_set that there
- * are waiters. No need to retry, because there cannot be
- * a concurrent busy->free transition. After the CAS, the
- * event will be either set or busy.
+ * Leave the event reset and tell qemu_event_set that there are
+ * waiters. No need to retry, because there cannot be a concurrent
+ * busy->free transition. After the CAS, the event will be either
+ * set or busy.
+ *
+ * This cmpxchg doesn't have particular ordering requirements if it
+ * succeeds (moving the store earlier can only cause qemu_event_set()
+ * to issue _more_ wakeups), the failing case needs acquire semantics
+ * like the load above.
*/
if (qatomic_cmpxchg(&ev->value, EV_FREE, EV_BUSY) == EV_SET) {
return;
}
}
+
+ /*
+ * This is the final check for a concurrent set, so it does need
+ * a smp_mb() pairing with the second barrier of qemu_event_set().
+ * The barrier is inside the FUTEX_WAIT system call.
+ */
qemu_futex_wait(ev, EV_BUSY);
}
}
--
2.37.3