public inbox for pve-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Stefan Reiter <s.reiter@proxmox.com>
To: pve-devel@lists.proxmox.com
Subject: Re: [pve-devel] [PATCH pve-qemu] update patches with some pbs-state migration cleanups
Date: Tue, 10 Nov 2020 17:18:43 +0100	[thread overview]
Message-ID: <78161bd3-5af5-5b2b-2280-a4177394639c@proxmox.com> (raw)
In-Reply-To: <20201110161615.23702-1-s.reiter@proxmox.com>

Eh, if it wasn't clear, this is the 'other thing' from the previous 
patch [0], so it applies on top.

[0] https://lists.proxmox.com/pipermail/pve-devel/2020-November/045884.html

On 11/10/20 5:16 PM, Stefan Reiter wrote:
> ...and literal cleanup, as in, call save_cleanup after success or error.
> 
> Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
> ---
> 
> I found it more confusing to set opaque to a value and then never use it, so I
> put it back to NULL and documented it.
> 
> Behaviour tested to be the same.
> 
>   ...igrate-dirty-bitmap-state-via-savevm.patch | 33 ++++++++++++-------
>   1 file changed, 21 insertions(+), 12 deletions(-)
> 
> diff --git a/debian/patches/pve/0055-PVE-Migrate-dirty-bitmap-state-via-savevm.patch b/debian/patches/pve/0055-PVE-Migrate-dirty-bitmap-state-via-savevm.patch
> index 77c0c76..a39e3ca 100644
> --- a/debian/patches/pve/0055-PVE-Migrate-dirty-bitmap-state-via-savevm.patch
> +++ b/debian/patches/pve/0055-PVE-Migrate-dirty-bitmap-state-via-savevm.patch
> @@ -13,13 +13,13 @@ safe migration is possible and makes sense.
>   
>   Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
>   ---
> - include/migration/misc.h |  3 ++
> - migration/Makefile.objs  |  1 +
> - migration/pbs-state.c    | 97 ++++++++++++++++++++++++++++++++++++++++
> - pve-backup.c             |  1 +
> - qapi/block-core.json     |  6 +++
> - softmmu/vl.c             |  1 +
> - 6 files changed, 109 insertions(+)
> + include/migration/misc.h |   3 ++
> + migration/Makefile.objs  |   1 +
> + migration/pbs-state.c    | 106 +++++++++++++++++++++++++++++++++++++++
> + pve-backup.c             |   1 +
> + qapi/block-core.json     |   6 +++
> + softmmu/vl.c             |   1 +
> + 6 files changed, 118 insertions(+)
>    create mode 100644 migration/pbs-state.c
>   
>   diff --git a/include/migration/misc.h b/include/migration/misc.h
> @@ -48,10 +48,10 @@ index 0fc619e380..20b3792599 100644
>    common-obj-$(CONFIG_RDMA) += rdma.o
>   diff --git a/migration/pbs-state.c b/migration/pbs-state.c
>   new file mode 100644
> -index 0000000000..c711498c3e
> +index 0000000000..29f2b3860d
>   --- /dev/null
>   +++ b/migration/pbs-state.c
> -@@ -0,0 +1,97 @@
> +@@ -0,0 +1,106 @@
>   +/*
>   + * PBS (dirty-bitmap) state migration
>   + */
> @@ -67,6 +67,7 @@ index 0000000000..c711498c3e
>   +    bool active;
>   +} PBSState;
>   +
> ++/* state is accessed via this static variable directly, 'opaque' is NULL */
>   +static PBSState pbs_state;
>   +
>   +static void pbs_state_save_pending(QEMUFile *f, void *opaque,
> @@ -115,8 +116,9 @@ index 0000000000..c711498c3e
>   +
>   +static bool pbs_state_is_active(void *opaque)
>   +{
> -+    // we need to be return active once, else .save_setup is never called, but,
> -+    // if we'd just would return true, we'd freeze the migration for ~ 5 - 10s
> ++    /* we need to return active exactly once, else .save_setup is never called,
> ++     * but if we'd just return true the migration doesn't make progress since
> ++     * it'd be waiting for us */
>   +    return pbs_state.active;
>   +}
>   +
> @@ -133,6 +135,12 @@ index 0000000000..c711498c3e
>   +    return false;
>   +}
>   +
> ++static void pbs_state_save_cleanup(void *opaque)
> ++{
> ++    /* reset active after migration succeeds or fails */
> ++    pbs_state.active = false;
> ++}
> ++
>   +static SaveVMHandlers savevm_pbs_state_handlers = {
>   +    .save_setup = pbs_state_save_setup,
>   +    .has_postcopy = pbs_state_has_postcopy,
> @@ -140,6 +148,7 @@ index 0000000000..c711498c3e
>   +    .is_active_iterate = pbs_state_is_active_iterate,
>   +    .load_state = pbs_state_load,
>   +    .is_active = pbs_state_is_active,
> ++    .save_cleanup = pbs_state_save_cleanup,
>   +};
>   +
>   +void pbs_state_mig_init(void)
> @@ -147,7 +156,7 @@ index 0000000000..c711498c3e
>   +    pbs_state.active = true;
>   +    register_savevm_live("pbs-state", 0, 1,
>   +                         &savevm_pbs_state_handlers,
> -+                         &pbs_state);
> ++                         NULL);
>   +}
>   diff --git a/pve-backup.c b/pve-backup.c
>   index c7cde0fb0e..f65f1dda26 100644
> 




  reply	other threads:[~2020-11-10 16:18 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-10 16:16 Stefan Reiter
2020-11-10 16:18 ` Stefan Reiter [this message]
2020-11-24 16:32 ` Fabian Grünbichler
2020-11-25 11:03   ` [pve-devel] applied: " Fabian Grünbichler

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=78161bd3-5af5-5b2b-2280-a4177394639c@proxmox.com \
    --to=s.reiter@proxmox.com \
    --cc=pve-devel@lists.proxmox.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal