Virtuozzo QEMU development (svt-core@virtuozzo.com)
 help / color / mirror / Atom feed
From: no-reply@sashiko.virtuozzo.com
To: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>,
	 svt-core@virtuozzo.com
Subject: Re: [PATCH QEMU HCI-8.1 v6 04/34] migration/cpr-exec: require a file or exec migration URI
Date: Mon, 2 Nov 2026 10:42:42 +0000	[thread overview]
Message-ID: <010701a19bdab025-d1d88833-f0d3-43f2-9959-9be6cf8819c9-000000@eu-central-1.amazonses.com> (raw)
In-Reply-To: <20261102-vfio_v1-v6-4-9bc7cabc850d@virtuozzo.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] migration: cpr-exec URI check also rejects incoming socket/fd/rdma URIs, undocumented
--

commit 77db773c747012b4f8980124448a08cb0f5cd2ed
Author: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>

migration/cpr-exec: require a file or exec migration URI

This commit makes migration_channels_and_transport_compatible() refuse
cpr-exec mode unless the migration URI is a file or exec type. It also
updates CPR.rst to say that other URI types are refused.

> diff --git a/docs/devel/migration/CPR.rst b/docs/devel/migration/CPR.rst
> --- a/docs/devel/migration/CPR.rst
> +++ b/docs/devel/migration/CPR.rst

[ ... ]

> @@ -365,10 +365,10 @@ The VM must be started with the ``-machine aux-ram-share=on`` option.
>  Outgoing:
>    * Set the migration mode parameter to ``cpr-exec``.
>    * Set the ``cpr-exec-command`` parameter.
> -  * Issue the ``migrate`` command.  It is recommended that the URI be
> -    a ``file`` type, but one can use other types such as ``exec``,
> -    provided the command captures all the data from the outgoing side,
> -    and provides all the data to the incoming side.
> +  * Issue the ``migrate`` command.  The URI must be a ``file`` type,
> +    or an ``exec`` type, provided the command captures all the data
> +    from the outgoing side, and provides all the data to the incoming
> +    side.  Other types are refused.
>  
>  Incoming:
>    * You do not need to explicitly start new QEMU.  It is started as

[Severity: Low]
The restriction is documented only under Outgoing, but the check is also
applied on the incoming side.  Should the Incoming section mention it as
well?

> diff --git a/migration/migration.c b/migration/migration.c
> --- a/migration/migration.c
> +++ b/migration/migration.c

[ ... ]

> @@ -258,6 +258,14 @@ migration_channels_and_transport_compatible(MigrationAddress *addr,
>          return false;
>      }
>  
> +    if (migrate_mode() == MIG_MODE_CPR_EXEC &&
> +        addr->transport != MIGRATION_ADDRESS_TYPE_FILE &&
> +        addr->transport != MIGRATION_ADDRESS_TYPE_EXEC) {
> +        error_setg(errp,
> +                   "cpr-exec mode requires a file or exec migration URI");
> +        return false;
> +    }
> +
>      return true;
>  }

[Severity: Low]
Is it intended that this check also applies to incoming migration?  The
commit message, the CPR.rst text and the error string describe only the
outgoing side.

migration_channels_and_transport_compatible() is shared with the incoming
path:

qemu_start_incoming_migration()
  migration_transport_compatible()
    migration_channels_and_transport_compatible()

qmp_migrate_incoming() reaches the same helper.

migrate_mode() returns cpr_get_incoming_mode() first, so in the exec'd
QEMU, after cpr_state_load() has set the incoming mode, it returns
cpr-exec:

MigMode migrate_mode(void)
{
    MigMode mode = cpr_get_incoming_mode();

    if (mode == MIG_MODE_NONE) {
        mode = migrate_get_current()->parameters.mode;
    }
    ...
}

When the mode is cpr-exec, any incoming URI that is not file or exec is
now refused.  That includes tcp, unix, vsock, rdma and fd:.  An fd: URI
always parses as MIGRATION_ADDRESS_TYPE_SOCKET, even when the descriptor
refers to a regular file.

A plain destination with the mode parameter set to cpr-exec is affected as
well.

Could the commit message and documentation describe the incoming-side
behavior?  Alternatively, should the check be limited to the outgoing path?

-- 
Sashiko AI review · https://sashiko.vzint.dev/#/patchset/20261102-vfio_v1-v6-0-9bc7cabc850d@virtuozzo.com?part=4

  reply	other threads:[~2026-11-02 10:42 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-02 10:36 [PATCH QEMU HCI-8.1 v6 00/34] vfio: support qemu-update with passed through GPU's and vGPU's #VSTOR-103860 Andrey Drobyshev
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 01/34] VZ: migration/cpr-exec: fix Error leak in preserve_ram() Andrey Drobyshev
2026-11-02 10:39   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 02/34] migration/cpr-exec: restore RAM write access if exec fails Andrey Drobyshev
2026-11-02 10:40   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 03/34] vfio/region: create BAR dma-bufs with O_CLOEXEC Andrey Drobyshev
2026-11-02 10:40   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 04/34] migration/cpr-exec: require a file or exec migration URI Andrey Drobyshev
2026-11-02 10:42   ` no-reply [this message]
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 05/34] migration: close the CPR state if the outgoing channel fails to open Andrey Drobyshev
2026-11-02 10:42   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 06/34] migration/cpr-exec: close the persisted state if the " Andrey Drobyshev
2026-11-02 10:44   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 07/34] migration: don't refuse a blocker over a migration of another mode Andrey Drobyshev
2026-11-02 10:44   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 08/34] migration: Plug memory leaks after migrate_set_error() Andrey Drobyshev
2026-11-02 10:44   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 09/34] hw/vfio: Fix liveness check in vfio_connect_kvm_msi_virq() Andrey Drobyshev
2026-11-02 10:45   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 10/34] vfio/pci: Reject invalid MSI-X Table and PBA BIR values Andrey Drobyshev
2026-11-02 10:45   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 11/34] vfio/pci: Fix information leak in vfio_rom_read() Andrey Drobyshev
2026-11-02 10:45   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 12/34] vfio/pci: Initialize rom_read_failed in vfio_pci_load_rom() Andrey Drobyshev
2026-11-02 10:46   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 13/34] vfio/pci: Grow buffer in vfio_pci_host_match() Andrey Drobyshev
2026-11-02 10:46   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 14/34] q35: Fix migration of SMRAM state Andrey Drobyshev
2026-11-02 10:48   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 15/34] migration: ensure APIC is loaded prior to VFIO PCI devices Andrey Drobyshev
2026-11-02 10:47   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 16/34] migration: Fix regression of passing error_fatal into vmstate_load_state() Andrey Drobyshev
2026-11-02 10:51   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 17/34] migration: vmstate_save_state_v: fix double error_setg Andrey Drobyshev
2026-11-02 10:51   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 18/34] migration: yank: Move register instance earlier Andrey Drobyshev
2026-11-02 10:55   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 19/34] migration: Free cpr-transfer MigrationAddress along with gsource Andrey Drobyshev
2026-11-02 10:53   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 20/34] migration: Move CPR HUP watch to cpr-transfer.c Andrey Drobyshev
2026-11-02 10:54   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 21/34] migration: Fix crash on second migration when cancel early Andrey Drobyshev
2026-11-02 10:57   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 22/34] kvm: tolerate ioctls after kvm_close() Andrey Drobyshev
2026-11-02 11:01   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 23/34] kvm: fix coalesced MMIO flush racing with kvm_close() Andrey Drobyshev
2026-11-02 11:00   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 24/34] migration/cpr: add cpr_is_in_progress() Andrey Drobyshev
2026-11-02 11:01   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 25/34] vfio/migration: don't migrate device state during CPR Andrey Drobyshev
2026-11-02 11:07   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 26/34] vfio/migration: don't assume the device is running on CPR incoming Andrey Drobyshev
2026-11-02 11:06   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 27/34] migration/ram: don't start dirty logging in CPR modes Andrey Drobyshev
2026-11-02 11:07   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 28/34] VZ: vfio/cpr: don't close KVM on cpr-exec Andrey Drobyshev
2026-11-02 11:09   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 29/34] VZ: migration/cpr-exec: abort cpr-exec if memory can't be preserved Andrey Drobyshev
2026-11-02 11:09   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 30/34] VZ: migration/cpr-exec: take the vhost-net devices back if exec fails Andrey Drobyshev
2026-11-02 11:11   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 31/34] VZ: migration/cpr-exec: don't exec if a vhost-net device can't be released Andrey Drobyshev
2026-11-02 11:10   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 32/34] VZ: vfio/region: preserve BAR mappings across cpr-exec #VSTOR-103860 Andrey Drobyshev
2026-11-02 11:16   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 33/34] VZ: migration/cpr-exec: allow cpr-exec with VFIO devices #VSTOR-103860 Andrey Drobyshev
2026-11-02 11:12   ` no-reply
2026-11-02 10:36 ` [PATCH QEMU HCI-8.1 v6 34/34] VZ: vfio/migration: don't wait for stop-copy-only device state #VSTOR-141821 Andrey Drobyshev
2026-11-02 11:15   ` no-reply
  -- strict thread matches above, loose matches on Subject: below --
2026-10-30 14:45 [PATCH QEMU HCI-8.1 v6 00/34] vfio: support qemu-update with passed through GPU's and vGPU's #VSTOR-103860 Andrey Drobyshev
2026-10-30 14:45 ` [PATCH QEMU HCI-8.1 v6 04/34] migration/cpr-exec: require a file or exec migration URI Andrey Drobyshev
2026-10-30 14:50   ` no-reply

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=010701a19bdab025-d1d88833-f0d3-43f2-9959-9be6cf8819c9-000000@eu-central-1.amazonses.com \
    --to=no-reply@sashiko.virtuozzo.com \
    --cc=andrey.drobyshev@virtuozzo.com \
    --cc=svt-core@virtuozzo.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