qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jason Wang <jasowang@redhat.com>
To: qemu-devel@nongnu.org, peter.maydell@linaro.org
Cc: "Eugenio Pérez" <eperezma@redhat.com>,
	"Jason Wang" <jasowang@redhat.com>
Subject: [PULL 04/24] vdpa: Avoid compiler to squash reads to used idx
Date: Tue, 19 Jul 2022 21:16:17 +0800	[thread overview]
Message-ID: <20220719131637.46131-5-jasowang@redhat.com> (raw)
In-Reply-To: <20220719131637.46131-1-jasowang@redhat.com>

From: Eugenio Pérez <eperezma@redhat.com>

In the next patch we will allow busypolling of this value. The compiler
have a running path where shadow_used_idx, last_used_idx, and vring used
idx are not modified within the same thread busypolling.

This was not an issue before since we always cleared device event
notifier before checking it, and that could act as memory barrier.
However, the busypoll needs something similar to kernel READ_ONCE.

Let's add it here, separated from the polling.

Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 hw/virtio/vhost-shadow-virtqueue.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
index e2184a4..560d07a 100644
--- a/hw/virtio/vhost-shadow-virtqueue.c
+++ b/hw/virtio/vhost-shadow-virtqueue.c
@@ -327,11 +327,12 @@ static void vhost_handle_guest_kick_notifier(EventNotifier *n)
 
 static bool vhost_svq_more_used(VhostShadowVirtqueue *svq)
 {
+    uint16_t *used_idx = &svq->vring.used->idx;
     if (svq->last_used_idx != svq->shadow_used_idx) {
         return true;
     }
 
-    svq->shadow_used_idx = cpu_to_le16(svq->vring.used->idx);
+    svq->shadow_used_idx = cpu_to_le16(*(volatile uint16_t *)used_idx);
 
     return svq->last_used_idx != svq->shadow_used_idx;
 }
-- 
2.7.4



  parent reply	other threads:[~2022-07-19 13:25 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-19 13:16 [PULL 00/24] Net Patches Jason Wang
2022-07-19 13:16 ` [PULL 01/24] vhost: move descriptor translation to vhost_svq_vring_write_descs Jason Wang
2022-07-19 13:16 ` [PULL 02/24] virtio-net: Expose MAC_TABLE_ENTRIES Jason Wang
2022-07-19 13:16 ` [PULL 03/24] virtio-net: Expose ctrl virtqueue logic Jason Wang
2022-07-19 13:16 ` Jason Wang [this message]
2022-07-19 13:16 ` [PULL 05/24] vhost: Reorder vhost_svq_kick Jason Wang
2022-07-19 13:16 ` [PULL 06/24] vhost: Move vhost_svq_kick call to vhost_svq_add Jason Wang
2022-07-19 13:16 ` [PULL 07/24] vhost: Check for queue full at vhost_svq_add Jason Wang
2022-07-19 13:16 ` [PULL 08/24] vhost: Decouple vhost_svq_add from VirtQueueElement Jason Wang
2022-07-19 13:16 ` [PULL 09/24] vhost: Add SVQDescState Jason Wang
2022-07-19 13:16 ` [PULL 10/24] vhost: Track number of descs in SVQDescState Jason Wang
2022-07-19 13:16 ` [PULL 11/24] vhost: add vhost_svq_push_elem Jason Wang
2022-07-19 13:16 ` [PULL 12/24] vhost: Expose vhost_svq_add Jason Wang
2022-07-19 13:16 ` [PULL 13/24] vhost: add vhost_svq_poll Jason Wang
2022-07-19 13:16 ` [PULL 14/24] vhost: Add svq avail_handler callback Jason Wang
2022-07-19 13:16 ` [PULL 15/24] vdpa: Export vhost_vdpa_dma_map and unmap calls Jason Wang
2022-07-19 13:16 ` [PULL 16/24] vdpa: manual forward CVQ buffers Jason Wang
2022-07-19 13:16 ` [PULL 17/24] vdpa: Buffer CVQ support on shadow virtqueue Jason Wang
2022-07-19 13:16 ` [PULL 18/24] vdpa: Extract get features part from vhost_vdpa_get_max_queue_pairs Jason Wang
2022-07-19 13:16 ` [PULL 19/24] vdpa: Add device migration blocker Jason Wang
2022-07-19 13:16 ` [PULL 20/24] vdpa: Add x-svq to NetdevVhostVDPAOptions Jason Wang
2022-07-19 13:16 ` [PULL 21/24] softmmu/runstate.c: add RunStateTransition support form COLO to PRELAUNCH Jason Wang
2022-07-19 13:16 ` [PULL 22/24] net/colo: Fix a "double free" crash to clear the conn_list Jason Wang
2022-07-19 13:16 ` [PULL 23/24] net/colo.c: No need to track conn_list for filter-rewriter Jason Wang
2022-07-19 13:16 ` [PULL 24/24] net/colo.c: fix segmentation fault when packet is not parsed correctly Jason Wang
2022-07-19 16:40 ` [PULL 00/24] Net Patches Peter Maydell
2022-07-20  3:40   ` Jason Wang
2022-07-20  6:02     ` Eugenio Perez Martin
2022-07-20  6:06       ` Jason Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220719131637.46131-5-jasowang@redhat.com \
    --to=jasowang@redhat.com \
    --cc=eperezma@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).