[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v9 08/12] migration: Test new fd infrastructure
From: |
Dr. David Alan Gilbert |
Subject: |
Re: [Qemu-devel] [PATCH v9 08/12] migration: Test new fd infrastructure |
Date: |
Tue, 17 Oct 2017 12:11:07 +0100 |
User-agent: |
Mutt/1.9.1 (2017-09-22) |
* Juan Quintela (address@hidden) wrote:
> We just send the address through the alternate channels and test that it
> is ok.
>
> Signed-off-by: Juan Quintela <address@hidden>
I remember questions on this patch from last time as well; this is just
test isn't it, and all this gets changed in later patches. So I'm
not too sure of the point, especially since you could use
qio_channel_writev_all here and make the changes smaller.
Dave
> --
>
> Use qio_channel_*all functions
> ---
> migration/ram.c | 39 +++++++++++++++++++++++++++++++++++++++
> 1 file changed, 39 insertions(+)
>
> diff --git a/migration/ram.c b/migration/ram.c
> index 745da2971d..4c16d0775b 100644
> --- a/migration/ram.c
> +++ b/migration/ram.c
> @@ -492,8 +492,24 @@ static void *multifd_send_thread(void *opaque)
> break;
> }
> if (p->pages.num) {
> + Error *local_err = NULL;
> + size_t ret;
> + int i;
> + int num;
> +
> + num = p->pages.num;
> p->pages.num = 0;
> qemu_mutex_unlock(&p->mutex);
> +
> + for (i = 0; i < num; i++) {
> + ret = qio_channel_write_all(p->c,
> + (const char *)&p->pages.iov[i].iov_base,
> + sizeof(uint8_t *), &local_err);
> + if (ret != 0) {
> + terminate_multifd_send_threads(local_err);
> + return NULL;
> + }
> + }
> qemu_mutex_lock(&multifd_send_state->mutex);
> p->done = true;
> qemu_mutex_unlock(&multifd_send_state->mutex);
> @@ -675,6 +691,7 @@ int multifd_load_cleanup(Error **errp)
> static void *multifd_recv_thread(void *opaque)
> {
> MultiFDRecvParams *p = opaque;
> + uint8_t *recv_address;
>
> qemu_sem_post(&p->ready);
> while (true) {
> @@ -684,7 +701,29 @@ static void *multifd_recv_thread(void *opaque)
> break;
> }
> if (p->pages.num) {
> + Error *local_err = NULL;
> + size_t ret;
> + int i;
> + int num;
> +
> + num = p->pages.num;
> p->pages.num = 0;
> +
> + for (i = 0; i < num; i++) {
> + ret = qio_channel_read_all(p->c, (char *)&recv_address,
> + sizeof(uint8_t *), &local_err);
> + if (ret != 0) {
> + terminate_multifd_recv_threads(local_err);
> + return NULL;
> + }
> + if (recv_address != p->pages.iov[i].iov_base) {
> + error_setg(&local_err, "received %p and expecting %p
> (%d)",
> + recv_address, p->pages.iov[i].iov_base, i);
> + terminate_multifd_recv_threads(local_err);
> + return NULL;
> + }
> + }
> +
> p->done = true;
> qemu_mutex_unlock(&p->mutex);
> qemu_sem_post(&p->ready);
> --
> 2.13.5
>
--
Dr. David Alan Gilbert / address@hidden / Manchester, UK
- [Qemu-devel] [PATCH v9 04/12] migration: Start of multiple fd work, (continued)
- [Qemu-devel] [PATCH v9 04/12] migration: Start of multiple fd work, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 06/12] migration: Send the fd number which we are going to use for this page, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 07/12] migration: Create thread infrastructure for multifd recv side, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 08/12] migration: Test new fd infrastructure, Juan Quintela, 2017/10/04
- Re: [Qemu-devel] [PATCH v9 08/12] migration: Test new fd infrastructure,
Dr. David Alan Gilbert <=
- [Qemu-devel] [PATCH v9 05/12] migration: Create ram_multifd_page, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 09/12] migration: Rename initial_bytes, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 11/12] migration: Flush receive queue, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 10/12] migration: Transfer pages over new channels, Juan Quintela, 2017/10/04
- [Qemu-devel] [PATCH v9 12/12] migration: Add multifd test, Juan Quintela, 2017/10/04