lkml.org 
[lkml]   [2020]   [Oct]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH v1 07/10] bus: mhi: core: Move to SYS_ERROR regardless of RDDM capability
On 2020-10-09 09:32, Manivannan Sadhasivam wrote:
> On Fri, Sep 18, 2020 at 07:02:32PM -0700, Bhaumik Bhatt wrote:
>> In some cases, the entry of device to RDDM execution environment
>> can occur after a significant amount of time has elapsed after the
>> SYS_ERROR state change event has arrived. This can result in scenarios
>> where users of the MHI bus are unaware of the error state of the
>
> Who are all the users of MHI bus? Client drivers?
>
Both client and controller drivers. I will change it to that.
>> device. Hence, moving the MHI bus to a SYS_ERROR detected state will
>> prevent further client activity and wait for the RDDM entry.
>>
>> Signed-off-by: Bhaumik Bhatt <bbhatt@codeaurora.org>
>> ---
>> drivers/bus/mhi/core/main.c | 24 ++++++++++++++++--------
>> 1 file changed, 16 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/bus/mhi/core/main.c b/drivers/bus/mhi/core/main.c
>> index 2cff5dd..1c8e332 100644
>> --- a/drivers/bus/mhi/core/main.c
>> +++ b/drivers/bus/mhi/core/main.c
>> @@ -376,6 +376,7 @@ irqreturn_t mhi_intvec_threaded_handler(int
>> irq_number, void *priv)
>> enum mhi_state state = MHI_STATE_MAX;
>> enum mhi_pm_state pm_state = 0;
>> enum mhi_ee_type ee = 0;
>> + bool handle_rddm = false;
>>
>> write_lock_irq(&mhi_cntrl->pm_lock);
>> if (!MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) {
>> @@ -400,6 +401,17 @@ irqreturn_t mhi_intvec_threaded_handler(int
>> irq_number, void *priv)
>> /* If device supports RDDM don't bother processing SYS error */
>> if (mhi_cntrl->rddm_image) {
>> if (mhi_cntrl->ee == MHI_EE_RDDM && mhi_cntrl->ee != ee) {
>> + /* prevent clients from queueing any more packets */
>> + write_lock_irq(&mhi_cntrl->pm_lock);
>> + pm_state = mhi_tryset_pm_state(mhi_cntrl,
>> + MHI_PM_SYS_ERR_DETECT);
>
> The condition above already moves MHI to MHI_PM_SYS_ERR_DETECT if the
> state
> is MHI_STATE_SYS_ERR. Why are you doing it here again?
>
> Thanks,
> Mani
>
I added it there because any first move to RDDM required the MHI host to
be
inactive or in an "error" state.
However, upon further thought, I have made changes that negate this need
and
instead make the if (mhi_cntrl->rddm_image) check dependent on the
pm_state being
MHI_PM_SYS_ERR_DETECT.

Reason is: a first move RDDM comes after a SYS_ERROR in MHI state, since
PM state
will already by SYS_ERROR detect by then, no client drivers or
controllers will be able
to use the bus.
>> + if (pm_state == MHI_PM_SYS_ERR_DETECT)
>> + handle_rddm = true;
>> + write_unlock_irq(&mhi_cntrl->pm_lock);
>> + }
>> +
>> + if (handle_rddm) {
>> + dev_err(dev, "RDDM event occurred!\n");
>> mhi_cntrl->status_cb(mhi_cntrl, MHI_CB_EE_RDDM);
>> wake_up_all(&mhi_cntrl->state_event);
>> }
>> @@ -733,19 +745,15 @@ int mhi_process_ctrl_ev_ring(struct
>> mhi_controller *mhi_cntrl,
>> break;
>> case MHI_STATE_SYS_ERR:
>> {
>> - enum mhi_pm_state new_state;
>> -
>> - /* skip SYS_ERROR handling if RDDM supported */
>> - if (mhi_cntrl->ee == MHI_EE_RDDM ||
>> - mhi_cntrl->rddm_image)
>> - break;
>> + enum mhi_pm_state state = MHI_PM_STATE_MAX;
>>
>> dev_dbg(dev, "System error detected\n");
>> write_lock_irq(&mhi_cntrl->pm_lock);
>> - new_state = mhi_tryset_pm_state(mhi_cntrl,
>> + if (mhi_cntrl->ee != MHI_EE_RDDM)
>> + state = mhi_tryset_pm_state(mhi_cntrl,
>> MHI_PM_SYS_ERR_DETECT);
>> write_unlock_irq(&mhi_cntrl->pm_lock);
>> - if (new_state == MHI_PM_SYS_ERR_DETECT)
>> + if (state == MHI_PM_SYS_ERR_DETECT)
>> mhi_pm_sys_err_handler(mhi_cntrl);
>> break;
>> }
>> --
>> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
>> Forum,
>> a Linux Foundation Collaborative Project
>>

--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora
Forum,
a Linux Foundation Collaborative Project

\
 
 \ /
  Last update: 2020-10-14 03:07    [W:0.097 / U:2.516 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site