On Wed, Nov 25, 2020 at 12:23:16PM +0100, Loic Poulain wrote:
> This patch fixes the hierarchical structure of MHI devices. Indeed,
> MHI client devices are directly 'enumerated' from the mhi controller
> and therefore must be direct descendants/children of their mhi
> controller device, in accordance with the Linux Device Model.
>
> Today both MHI clients and controller devices are at the same level,
> this patch ensures that MHI controller is parent of its client devices.
>
> The hierarc /* Only destroy virtual devices thats attached to bus */
if (mhi_dev->dev_type == MHI_DEVICE_CONTROLLER)
return 0;hy is especially important for power management (safe
> suspend/resume order). It is also useful for userspace to determine
> relationship between MHI client devices and controllers.
>
> Signed-off-by: Loic Poulain <loic.poulain@xxxxxxxxxx>
> Reviewed-by: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
> ---
> v2: fix commit message
> v3: reword commit message
> v4: fix device destroy
>
> drivers/bus/mhi/core/init.c | 10 +++++++++-
> drivers/bus/mhi/core/pm.c | 4 ++--
> 2 files changed, 11 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
> index 436221c..c7a7354 100644
> --- a/drivers/bus/mhi/core/init.c
> +++ b/drivers/bus/mhi/core/init.c
> @@ -1137,7 +1137,15 @@ struct mhi_device *mhi_alloc_device(struct mhi_controller *mhi_cntrl)
> device_initialize(dev);
> dev->bus = &mhi_bus_type;
> dev->release = mhi_release_device;
> - dev->parent = mhi_cntrl->cntrl_dev;
> +
> + if (mhi_cntrl->mhi_dev) {
> + /* for MHI client devices, parent is the MHI controller device */
> + dev->parent = &mhi_cntrl->mhi_dev->dev;
> + } else {
> + /* for MHI controller device, parent is the bus device (e.g. pci device) */
> + dev->parent = mhi_cntrl->cntrl_dev;
> + }
> +
> mhi_dev->mhi_cntrl = mhi_cntrl;
> mhi_dev->dev_wake = 0;
>
> diff --git a/drivers/bus/mhi/core/pm.c b/drivers/bus/mhi/core/pm.c
> index a671f58..681960c 100644
> --- a/drivers/bus/mhi/core/pm.c
> +++ b/drivers/bus/mhi/core/pm.c
> @@ -504,7 +504,7 @@ static void mhi_pm_disable_transition(struct mhi_controller *mhi_cntrl)
> wake_up_all(&mhi_cntrl->state_event);
>
> dev_dbg(dev, "Reset all active channels and remove MHI devices\n");
> - device_for_each_child(mhi_cntrl->cntrl_dev, NULL, mhi_destroy_device);
> + device_for_each_child(&mhi_cntrl->mhi_dev->dev, NULL, mhi_destroy_device);
>
> mutex_lock(&mhi_cntrl->pm_mutex);
>
> @@ -637,7 +637,7 @@ static void mhi_pm_sys_error_transition(struct mhi_controller *mhi_cntrl)
> wake_up_all(&mhi_cntrl->state_event);
>
> dev_dbg(dev, "Reset all active channels and remove MHI devices\n");
> - device_for_each_child(mhi_cntrl->cntrl_dev, NULL, mhi_destroy_device);
> + device_for_each_child(&mhi_cntrl->mhi_dev->dev, NULL, mhi_destroy_device);
Hmm. Now I don't think we need below check in mhi_destroy_device():
/* Only destroy virtual devices thats attached to bus */
if (mhi_dev->dev_type == MHI_DEVICE_CONTROLLER)
return 0;
But let's keep it there for sanity.
Have you looked into mhi_debugfs_devices_show()? I guess you need to call
device_for_each_child() for both "mhi_cntrl->cntrl_dev" and
"mhi_cntrl->mhi_dev->dev".
Thanks,
Mani
>
> mutex_lock(&mhi_cntrl->pm_mutex);
>
> --
> 2.7.4
>
