On 3/26/26 8:22 AM, David Marchand wrote: > On Mon, 23 Mar 2026 at 17:59, Kevin Traynor <[email protected]> wrote: >> >> On 3/23/26 10:52 AM, David Marchand wrote: >>> The close operation was never closing probed devices. >>> >>> Taking a step back, reevaluating the devargs makes no sense during the >>> close step, as a probed device must have passed the allow/block list >>> evaluation initially. >>> >>> Since the device contains a reference to the driver that probed it, >>> simply call this driver remove op. >>> >>> Fixes: 274fd921ff7f ("bus/fslmc: support close operation") >>> Cc: [email protected] >>> >>> Signed-off-by: David Marchand <[email protected]> >>> --- >>> drivers/bus/fslmc/fslmc_vfio.c | 21 ++++----------------- >>> 1 file changed, 4 insertions(+), 17 deletions(-) >>> >>> diff --git a/drivers/bus/fslmc/fslmc_vfio.c b/drivers/bus/fslmc/fslmc_vfio.c >>> index 550d4e0e8d..7daa18d850 100644 >>> --- a/drivers/bus/fslmc/fslmc_vfio.c >>> +++ b/drivers/bus/fslmc/fslmc_vfio.c >>> @@ -1393,7 +1393,7 @@ fslmc_close_iodevices(struct rte_dpaa2_device *dev, >>> { >>> struct rte_dpaa2_object *object = NULL; >>> struct rte_dpaa2_driver *drv; >>> - int ret, probe_all; >>> + int ret; >>> >>> switch (dev->dev_type) { >>> case DPAA2_IO: >>> @@ -1411,22 +1411,9 @@ fslmc_close_iodevices(struct rte_dpaa2_device *dev, >>> case DPAA2_ETH: >>> case DPAA2_CRYPTO: >>> case DPAA2_QDMA: >>> - probe_all = rte_fslmc_bus.bus.conf.scan_mode != >>> - RTE_BUS_SCAN_ALLOWLIST; >>> - TAILQ_FOREACH(drv, &rte_fslmc_bus.driver_list, next) { >>> - if (drv->drv_type != dev->dev_type) >>> - continue; >>> - if (rte_dev_is_probed(&dev->device)) >>> - continue; >>> - if (probe_all || >>> - (dev->device.devargs && >>> - dev->device.devargs->policy == >>> - RTE_DEV_ALLOWED)) { >>> - ret = drv->remove(dev); >>> - if (ret) >>> - DPAA2_BUS_ERR("Unable to remove"); >>> - } >>> - } >>> + drv = dev->driver; >>> + if (drv && drv->remove && drv->remove(dev)) >> >> Not new but dpaa2_qdma_remove doesn't return an error which seems like a >> bug, but it has it's own log and it's just a log here anyway, so not >> critical. >> >>> + DPAA2_BUS_ERR("Unable to remove"); > > Indeed, the dma/dpaa2 driver does not report errors. > And there is probably more to fix in the fslmc bus, as I see that the > bus .unplug() does not check the driver .remove() return value either. > > I did not dig deeper in this bus for now as I was focusing on the > devargs evaluation. > > Could you open a bugzilla for tracking this issue? >
Sure, I will do that > Thanks. > >
