From: Wei Wang <wei.w.wang@intel.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
akpm@linux-foundation.org, mawilcox@microsoft.com,
david@redhat.com, cornelia.huck@de.ibm.com,
mgorman@techsingularity.net, aarcange@redhat.com,
amit.shah@redhat.com, pbonzini@redhat.com, willy@infradead.org,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu@aliyun.com
Subject: [virtio-dev] Re: [PATCH v14 3/5] virtio-balloon: VIRTIO_BALLOON_F_SG
Date: Fri, 18 Aug 2017 15:39:27 +0800 [thread overview]
Message-ID: <599699AF.1090705@intel.com> (raw)
In-Reply-To: <20170818051451-mutt-send-email-mst@kernel.org>
On 08/18/2017 10:22 AM, Michael S. Tsirkin wrote:
> +static void send_balloon_page_sg(struct virtio_balloon *vb,
> + struct virtqueue *vq,
> + void *addr,
> + uint32_t size)
> +{
> + unsigned int len;
> + int ret;
> +
> + do {
> + ret = add_one_sg(vq, addr, size);
> + virtqueue_kick(vq);
> + wait_event(vb->acked, virtqueue_get_buf(vq, &len));
> + /*
> + * It is uncommon to see the vq is full, because the sg is sent
> + * one by one and the device is able to handle it in time. But
> + * if that happens, we go back to retry after an entry gets
> + * released.
> + */
> Why send one by one though? Why not batch some s/gs and wait for all
> of them to be completed? If memory if fragmented, waiting every time is
> worse than what we have now (VIRTIO_BALLOON_ARRAY_PFNS_MAX at a time).
>
OK, I'll do batching in some fashion.
Best,
Wei
---------------------------------------------------------------------
To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org
WARNING: multiple messages have this Message-ID (diff)
From: Wei Wang <wei.w.wang@intel.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
akpm@linux-foundation.org, mawilcox@microsoft.com,
david@redhat.com, cornelia.huck@de.ibm.com,
mgorman@techsingularity.net, aarcange@redhat.com,
amit.shah@redhat.com, pbonzini@redhat.com, willy@infradead.org,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu@aliyun.com
Subject: Re: [PATCH v14 3/5] virtio-balloon: VIRTIO_BALLOON_F_SG
Date: Fri, 18 Aug 2017 15:39:27 +0800 [thread overview]
Message-ID: <599699AF.1090705@intel.com> (raw)
In-Reply-To: <20170818051451-mutt-send-email-mst@kernel.org>
On 08/18/2017 10:22 AM, Michael S. Tsirkin wrote:
> +static void send_balloon_page_sg(struct virtio_balloon *vb,
> + struct virtqueue *vq,
> + void *addr,
> + uint32_t size)
> +{
> + unsigned int len;
> + int ret;
> +
> + do {
> + ret = add_one_sg(vq, addr, size);
> + virtqueue_kick(vq);
> + wait_event(vb->acked, virtqueue_get_buf(vq, &len));
> + /*
> + * It is uncommon to see the vq is full, because the sg is sent
> + * one by one and the device is able to handle it in time. But
> + * if that happens, we go back to retry after an entry gets
> + * released.
> + */
> Why send one by one though? Why not batch some s/gs and wait for all
> of them to be completed? If memory if fragmented, waiting every time is
> worse than what we have now (VIRTIO_BALLOON_ARRAY_PFNS_MAX at a time).
>
OK, I'll do batching in some fashion.
Best,
Wei
WARNING: multiple messages have this Message-ID (diff)
From: Wei Wang <wei.w.wang@intel.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
akpm@linux-foundation.org, mawilcox@microsoft.com,
david@redhat.com, cornelia.huck@de.ibm.com,
mgorman@techsingularity.net, aarcange@redhat.com,
amit.shah@redhat.com, pbonzini@redhat.com, willy@infradead.org,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu@aliyun.com
Subject: Re: [PATCH v14 3/5] virtio-balloon: VIRTIO_BALLOON_F_SG
Date: Fri, 18 Aug 2017 15:39:27 +0800 [thread overview]
Message-ID: <599699AF.1090705@intel.com> (raw)
In-Reply-To: <20170818051451-mutt-send-email-mst@kernel.org>
On 08/18/2017 10:22 AM, Michael S. Tsirkin wrote:
> +static void send_balloon_page_sg(struct virtio_balloon *vb,
> + struct virtqueue *vq,
> + void *addr,
> + uint32_t size)
> +{
> + unsigned int len;
> + int ret;
> +
> + do {
> + ret = add_one_sg(vq, addr, size);
> + virtqueue_kick(vq);
> + wait_event(vb->acked, virtqueue_get_buf(vq, &len));
> + /*
> + * It is uncommon to see the vq is full, because the sg is sent
> + * one by one and the device is able to handle it in time. But
> + * if that happens, we go back to retry after an entry gets
> + * released.
> + */
> Why send one by one though? Why not batch some s/gs and wait for all
> of them to be completed? If memory if fragmented, waiting every time is
> worse than what we have now (VIRTIO_BALLOON_ARRAY_PFNS_MAX at a time).
>
OK, I'll do batching in some fashion.
Best,
Wei
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Wei Wang <wei.w.wang@intel.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtio-dev@lists.oasis-open.org, linux-kernel@vger.kernel.org,
qemu-devel@nongnu.org, virtualization@lists.linux-foundation.org,
kvm@vger.kernel.org, linux-mm@kvack.org, mhocko@kernel.org,
akpm@linux-foundation.org, mawilcox@microsoft.com,
david@redhat.com, cornelia.huck@de.ibm.com,
mgorman@techsingularity.net, aarcange@redhat.com,
amit.shah@redhat.com, pbonzini@redhat.com, willy@infradead.org,
liliang.opensource@gmail.com, yang.zhang.wz@gmail.com,
quan.xu@aliyun.com
Subject: Re: [Qemu-devel] [PATCH v14 3/5] virtio-balloon: VIRTIO_BALLOON_F_SG
Date: Fri, 18 Aug 2017 15:39:27 +0800 [thread overview]
Message-ID: <599699AF.1090705@intel.com> (raw)
In-Reply-To: <20170818051451-mutt-send-email-mst@kernel.org>
On 08/18/2017 10:22 AM, Michael S. Tsirkin wrote:
> +static void send_balloon_page_sg(struct virtio_balloon *vb,
> + struct virtqueue *vq,
> + void *addr,
> + uint32_t size)
> +{
> + unsigned int len;
> + int ret;
> +
> + do {
> + ret = add_one_sg(vq, addr, size);
> + virtqueue_kick(vq);
> + wait_event(vb->acked, virtqueue_get_buf(vq, &len));
> + /*
> + * It is uncommon to see the vq is full, because the sg is sent
> + * one by one and the device is able to handle it in time. But
> + * if that happens, we go back to retry after an entry gets
> + * released.
> + */
> Why send one by one though? Why not batch some s/gs and wait for all
> of them to be completed? If memory if fragmented, waiting every time is
> worse than what we have now (VIRTIO_BALLOON_ARRAY_PFNS_MAX at a time).
>
OK, I'll do batching in some fashion.
Best,
Wei
next prev parent reply other threads:[~2017-08-18 7:37 UTC|newest]
Thread overview: 115+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-17 3:26 [virtio-dev] [PATCH v14 0/5] Virtio-balloon Enhancement Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` [virtio-dev] [PATCH v14 1/5] lib/xbitmap: Introduce xbitmap Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-19 20:30 ` kbuild test robot
2017-08-19 20:30 ` kbuild test robot
2017-08-19 20:30 ` [Qemu-devel] " kbuild test robot
2017-08-19 20:30 ` kbuild test robot
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` [PATCH v14 2/5] lib/xbitmap: add xb_find_next_bit() and xb_zero() Wei Wang
2017-08-17 3:26 ` [virtio-dev] " Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` [PATCH v14 3/5] virtio-balloon: VIRTIO_BALLOON_F_SG Wei Wang
2017-08-17 3:26 ` [virtio-dev] " Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-18 2:22 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 2:22 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 2:22 ` Michael S. Tsirkin
2017-08-18 2:22 ` Michael S. Tsirkin
2017-08-18 7:39 ` Wei Wang [this message]
2017-08-18 7:39 ` [Qemu-devel] " Wei Wang
2017-08-18 7:39 ` Wei Wang
2017-08-18 7:39 ` Wei Wang
2017-08-21 20:22 ` Michael S. Tsirkin
2017-08-21 20:22 ` [virtio-dev] " Michael S. Tsirkin
2017-08-21 20:22 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-21 20:22 ` Michael S. Tsirkin
2017-08-21 20:22 ` Michael S. Tsirkin
2017-08-18 7:39 ` Wei Wang
2017-08-18 2:22 ` Michael S. Tsirkin
2017-08-19 21:37 ` kbuild test robot
2017-08-19 21:37 ` [Qemu-devel] " kbuild test robot
2017-08-19 21:37 ` kbuild test robot
2017-08-19 21:37 ` kbuild test robot
2017-08-17 3:26 ` [virtio-dev] [PATCH v14 4/5] mm: support reporting free page blocks Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-18 13:46 ` Michal Hocko
2017-08-18 13:46 ` Michal Hocko
2017-08-18 13:46 ` [Qemu-devel] " Michal Hocko
2017-08-18 13:46 ` Michal Hocko
2017-08-21 6:12 ` [virtio-dev] " Wei Wang
2017-08-21 6:12 ` [Qemu-devel] " Wei Wang
2017-08-21 6:12 ` Wei Wang
2017-08-21 6:12 ` Wei Wang
2017-08-21 6:14 ` Michal Hocko
2017-08-21 6:14 ` [Qemu-devel] " Michal Hocko
2017-08-21 6:14 ` Michal Hocko
2017-08-21 6:14 ` Michal Hocko
2017-08-21 6:12 ` Wei Wang
2017-08-18 17:23 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 17:23 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 17:23 ` Michael S. Tsirkin
2017-08-18 17:23 ` Michael S. Tsirkin
2017-08-21 6:18 ` Michal Hocko
2017-08-21 6:18 ` [Qemu-devel] " Michal Hocko
2017-08-21 6:18 ` Michal Hocko
2017-08-21 6:18 ` Michal Hocko
2017-08-18 17:23 ` Michael S. Tsirkin
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` [PATCH v14 5/5] virtio-balloon: VIRTIO_BALLOON_F_FREE_PAGE_VQ Wei Wang
2017-08-17 3:26 ` [virtio-dev] " Wei Wang
2017-08-17 3:26 ` [Qemu-devel] " Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-17 3:26 ` Wei Wang
2017-08-18 2:13 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 2:13 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 2:13 ` Michael S. Tsirkin
2017-08-18 2:13 ` Michael S. Tsirkin
2017-08-18 8:41 ` [virtio-dev] " Wei Wang
2017-08-18 8:41 ` [Qemu-devel] " Wei Wang
2017-08-18 8:41 ` Wei Wang
2017-08-18 8:41 ` Wei Wang
2017-08-18 18:26 ` Michael S. Tsirkin
2017-08-18 18:26 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 18:26 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 18:26 ` Michael S. Tsirkin
2017-08-18 18:26 ` Michael S. Tsirkin
2017-08-21 5:21 ` [virtio-dev] " Wei Wang
2017-08-21 5:21 ` [Qemu-devel] " Wei Wang
2017-08-21 5:21 ` Wei Wang
2017-08-21 5:21 ` Wei Wang
2017-08-21 5:21 ` Wei Wang
2017-08-18 8:41 ` Wei Wang
2017-08-18 2:13 ` Michael S. Tsirkin
2017-08-18 2:28 ` Michael S. Tsirkin
2017-08-18 2:28 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 2:28 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 2:28 ` Michael S. Tsirkin
2017-08-18 2:28 ` Michael S. Tsirkin
2017-08-18 8:36 ` Wei Wang
2017-08-18 8:36 ` [virtio-dev] " Wei Wang
2017-08-18 8:36 ` [Qemu-devel] " Wei Wang
2017-08-18 8:36 ` Wei Wang
2017-08-18 8:36 ` Wei Wang
2017-08-18 18:10 ` Michael S. Tsirkin
2017-08-18 18:10 ` [virtio-dev] " Michael S. Tsirkin
2017-08-18 18:10 ` [Qemu-devel] " Michael S. Tsirkin
2017-08-18 18:10 ` Michael S. Tsirkin
2017-08-18 18:10 ` Michael S. Tsirkin
2017-08-21 5:28 ` [virtio-dev] " Wei Wang
2017-08-21 5:28 ` [Qemu-devel] " Wei Wang
2017-08-21 5:28 ` Wei Wang
2017-08-21 5:28 ` Wei Wang
2017-08-21 5:28 ` Wei Wang
2017-08-21 5:28 ` [virtio-dev] " Wei 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=599699AF.1090705@intel.com \
--to=wei.w.wang@intel.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=amit.shah@redhat.com \
--cc=cornelia.huck@de.ibm.com \
--cc=david@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=liliang.opensource@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mawilcox@microsoft.com \
--cc=mgorman@techsingularity.net \
--cc=mhocko@kernel.org \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=quan.xu@aliyun.com \
--cc=virtio-dev@lists.oasis-open.org \
--cc=virtualization@lists.linux-foundation.org \
--cc=willy@infradead.org \
--cc=yang.zhang.wz@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.