mbox series

[v3,0/1] Qemu crashes on VM migration after an handled memory error

Message ID 20230920235301.1622672-1-william.roche@oracle.com (mailing list archive)
Headers show
Series Qemu crashes on VM migration after an handled memory error | expand

Message

William Roche Sept. 20, 2023, 11:53 p.m. UTC 
From: William Roche <william.roche@oracle.com>

A Qemu VM can survive a memory error, as qemu can relay the error to the
VM kernel which could also deal with it -- poisoning/off-lining the impacted
page.
This situation creates a hole in the VM memory address space that the VM kernel
knows about (an unreadable page or set of pages).

But the migration of this VM (live migration through the network or
pseudo-migration with the creation of a state file) will crash Qemu when
it sequentially reads the memory address space and stumbles on the
existing hole.

In order to correct this problem, I suggest to treat the poisoned pages as if
they were zero-pages for the migration copy.
This fix also works with underlying large pages, taking into account the
RAMBlock segment "page-size".
This fix is scripts/checkpatch.pl clean.

v2:
  - adding compressed transfer handling of poisoned pages
 
Testing: I could verify that migration now works with a poisoned page
through standard and compressed migration with 4k and large (2M) pages.

v3:
  - Included the Reviewed-by and Tested-by information
  - added a TODO comment above control_save_page()
    mentioning Zhijian's feedback about migration failure.


William Roche (1):
  migration: skip poisoned memory pages on "ram saving" phase

 accel/kvm/kvm-all.c      | 14 ++++++++++++++
 accel/stubs/kvm-stub.c   |  5 +++++
 include/sysemu/kvm.h     | 10 ++++++++++
 migration/ram-compress.c |  3 ++-
 migration/ram.c          | 24 ++++++++++++++++++++++--
 migration/ram.h          |  2 ++
 6 files changed, 55 insertions(+), 3 deletions(-)