On Tue, Dec 01, 2020 at 12:37:24PM +0100, Greg Kurz wrote:
> Igor recently suggested that instead of failing in spapr_drc_attach()
> at plug time we should rather check that the DRC is attachable at
> pre-plug time. This allows to error out before the hot-plugged device
> is even realized and to come up with simpler plug callbacks.
>
> sPAPR currently supports hotplug of PCI devices, PHBs, CPU cores,
> PC-DIMM/NVDIMM memory and TPM proxy devices. Some of these already
> do sanity checks at pre-plug that are sufficient to ensure the DRC
> are attachables. Some others don't even have a pre-plug handler.
>
> This series adds the missing pieces so that all failing conditions
> are caught at pre-plug time instead of plug time for all devices.
>
> v2: - hopefully less fragile way of setting compat mode for hot-plugged
> CPUs
Applied to ppc-for-6.0, thanks.
>
> Greg Kurz (4):
> spapr: Fix pre-2.10 dummy ICP hack
> spapr: Abort if ppc_set_compat() fails for hot-plugged CPUs
> spapr: Simplify error path of spapr_core_plug()
> spapr: spapr_drc_attach() cannot fail
>
> include/hw/ppc/spapr_drc.h | 8 ++++++-
> hw/ppc/spapr.c | 49 ++++++++++++++++++--------------------
> hw/ppc/spapr_drc.c | 8 ++-----
> hw/ppc/spapr_nvdimm.c | 2 +-
> hw/ppc/spapr_pci.c | 2 +-
> 5 files changed, 34 insertions(+), 35 deletions(-)
>
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson