qemu-kvm/kvm-async-update-documentation-of-the-memory-barriers.patch
Miroslav Rezanina de1852f087 * Sun Mar 12 2023 Miroslav Rezanina <mrezanin@redhat.com> - 7.2.0-12
- kvm-scsi-protect-req-aiocb-with-AioContext-lock.patch [bz#2155748]
- kvm-dma-helpers-prevent-dma_blk_cb-vs-dma_aio_cancel-rac.patch [bz#2155748]
- kvm-virtio-scsi-reset-SCSI-devices-from-main-loop-thread.patch [bz#2155748]
- kvm-qatomic-add-smp_mb__before-after_rmw.patch [bz#2175660]
- kvm-qemu-thread-posix-cleanup-fix-document-QemuEvent.patch [bz#2175660]
- kvm-qemu-thread-win32-cleanup-fix-document-QemuEvent.patch [bz#2175660]
- kvm-edu-add-smp_mb__after_rmw.patch [bz#2175660]
- kvm-aio-wait-switch-to-smp_mb__after_rmw.patch [bz#2175660]
- kvm-qemu-coroutine-lock-add-smp_mb__after_rmw.patch [bz#2175660]
- kvm-physmem-add-missing-memory-barrier.patch [bz#2175660]
- kvm-async-update-documentation-of-the-memory-barriers.patch [bz#2175660]
- kvm-async-clarify-usage-of-barriers-in-the-polling-case.patch [bz#2175660]
- Resolves: bz#2155748
  (qemu crash on void blk_drain(BlockBackend *): Assertion qemu_in_main_thread() failed)
- Resolves: bz#2175660
  (Guest hangs when starting or rebooting)
2023-03-12 23:37:54 -04:00

112 lines
4.4 KiB
Diff

From 29bcf843d796ffc2a0906dea947e4cdfe9f7ec60 Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 08:24:36 -0500
Subject: [PATCH 11/12] async: update documentation of the memory barriers
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 158: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2175660
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Commit: [8/9] 5ca20e4c8983e0bc1ecee66bead3472777abe4d1 (eesposit/qemu-kvm)
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2175660
commit 8dd48650b43dfde4ebea34191ac267e474bcc29e
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Mon Mar 6 10:15:06 2023 +0100
async: update documentation of the memory barriers
Ever since commit 8c6b0356b539 ("util/async: make bh_aio_poll() O(1)",
2020-02-22), synchronization between qemu_bh_schedule() and aio_bh_poll()
is happening when the bottom half is enqueued in the bh_list; not
when the flags are set. Update the documentation to match.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
util/async.c | 33 +++++++++++++++++++--------------
1 file changed, 19 insertions(+), 14 deletions(-)
diff --git a/util/async.c b/util/async.c
index 63434ddae4..37d3e6036d 100644
--- a/util/async.c
+++ b/util/async.c
@@ -73,14 +73,21 @@ static void aio_bh_enqueue(QEMUBH *bh, unsigned new_flags)
unsigned old_flags;
/*
- * The memory barrier implicit in qatomic_fetch_or makes sure that:
- * 1. idle & any writes needed by the callback are done before the
- * locations are read in the aio_bh_poll.
- * 2. ctx is loaded before the callback has a chance to execute and bh
- * could be freed.
+ * Synchronizes with atomic_fetch_and() in aio_bh_dequeue(), ensuring that
+ * insertion starts after BH_PENDING is set.
*/
old_flags = qatomic_fetch_or(&bh->flags, BH_PENDING | new_flags);
+
if (!(old_flags & BH_PENDING)) {
+ /*
+ * At this point the bottom half becomes visible to aio_bh_poll().
+ * This insertion thus synchronizes with QSLIST_MOVE_ATOMIC in
+ * aio_bh_poll(), ensuring that:
+ * 1. any writes needed by the callback are visible from the callback
+ * after aio_bh_dequeue() returns bh.
+ * 2. ctx is loaded before the callback has a chance to execute and bh
+ * could be freed.
+ */
QSLIST_INSERT_HEAD_ATOMIC(&ctx->bh_list, bh, next);
}
@@ -106,11 +113,8 @@ static QEMUBH *aio_bh_dequeue(BHList *head, unsigned *flags)
QSLIST_REMOVE_HEAD(head, next);
/*
- * The qatomic_and is paired with aio_bh_enqueue(). The implicit memory
- * barrier ensures that the callback sees all writes done by the scheduling
- * thread. It also ensures that the scheduling thread sees the cleared
- * flag before bh->cb has run, and thus will call aio_notify again if
- * necessary.
+ * Synchronizes with qatomic_fetch_or() in aio_bh_enqueue(), ensuring that
+ * the removal finishes before BH_PENDING is reset.
*/
*flags = qatomic_fetch_and(&bh->flags,
~(BH_PENDING | BH_SCHEDULED | BH_IDLE));
@@ -157,6 +161,7 @@ int aio_bh_poll(AioContext *ctx)
BHListSlice *s;
int ret = 0;
+ /* Synchronizes with QSLIST_INSERT_HEAD_ATOMIC in aio_bh_enqueue(). */
QSLIST_MOVE_ATOMIC(&slice.bh_list, &ctx->bh_list);
QSIMPLEQ_INSERT_TAIL(&ctx->bh_slice_list, &slice, next);
@@ -446,15 +451,15 @@ LuringState *aio_get_linux_io_uring(AioContext *ctx)
void aio_notify(AioContext *ctx)
{
/*
- * Write e.g. bh->flags before writing ctx->notified. Pairs with smp_mb in
- * aio_notify_accept.
+ * Write e.g. ctx->bh_list before writing ctx->notified. Pairs with
+ * smp_mb() in aio_notify_accept().
*/
smp_wmb();
qatomic_set(&ctx->notified, true);
/*
- * Write ctx->notified before reading ctx->notify_me. Pairs
- * with smp_mb in aio_ctx_prepare or aio_poll.
+ * Write ctx->notified (and also ctx->bh_list) before reading ctx->notify_me.
+ * Pairs with smp_mb() in aio_ctx_prepare or aio_poll.
*/
smp_mb();
if (qatomic_read(&ctx->notify_me)) {
--
2.39.1