Re: [PATCH 2/5] spapr.c: check unplug_request flag in spapr_memory_unplug_request()

[Greg Kurz]

On Fri, 26 Feb 2021 13:32:58 -0300
Daniel Henrique Barboza <danielhb413@gmail.com> wrote:

> Now that we're asserting the first DRC LMB earlier, use it to query if
> the DRC is already pending unplug and, in this case, issue the same
> error we already do.
> 
> The previous check was introduced in commit 2a129767ebb1 and it works,
> but it's easier to check the unplug_requested  flag instead of looking
> for the existence of the sPAPRDIMMState. It's also compliant with what
> is already done in other unplug_request functions for other devices.
> 
> Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
> ---

Reviewed-by: Greg Kurz <groug@kaod.org>

>  hw/ppc/spapr.c | 8 +-------
>  1 file changed, 1 insertion(+), 7 deletions(-)
> 
> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> index 74e046b522..149dc2113f 100644
> --- a/hw/ppc/spapr.c
> +++ b/hw/ppc/spapr.c
> @@ -3681,13 +3681,7 @@ static void spapr_memory_unplug_request(HotplugHandler 
> *hotplug_dev,
>                                  addr_start / SPAPR_MEMORY_BLOCK_SIZE);
>      g_assert(drc_start);
>  
> -    /*
> -     * An existing pending dimm state for this DIMM means that there is an
> -     * unplug operation in progress, waiting for the spapr_lmb_release
> -     * callback to complete the job (BQL can't cover that far). In this case,
> -     * bail out to avoid detaching DRCs that were already released.
> -     */
> -    if (spapr_pending_dimm_unplugs_find(spapr, dimm)) {
> +    if (spapr_drc_unplug_requested(drc_start)) {
>          error_setg(errp, "Memory unplug already in progress for device %s",
>                     dev->id);
>          return;