On 24/02/2015 18:38, John Snow wrote:
The day we all feared is here, and I am proposing we allow the
migration of the AHCI device. The series that precedes this
which fixes AHCI migration has improved the stability of the
device and as this test series proves, is stable.
I am justifying this checkin based on a series of ping-pong
migration tests I ran under heavy load (using google's stressapptest)
and saw over 300 successful migrations without a single failure.
This series does a few things:
(1) Add migration facilities to libqos
(2) Enable AHCI and ICH9 migration
(3) Add a series of migration tests to ahci-test
This patch has several dependencies:
(1) stefanha/block
(2) [PATCH v4 00/17] ide: rerror/werror migration fixes for IDE/ISA and AHCI
(3) [PATCH 0/8] ahci: add more IO tests
(4) [PATCH 0/6] ahci: rerror/werror=stop resume tests
1, 3 and 4 are ahci-test framework dependencies, but 2 actually allows the
AHCI migration to become stable.
Does rerror/werror=stop work even for queued DMA reads/writes (I didn't
try doing that, but I never even checked whether it just worked)?
Paolo