| Message ID | 1623300218-9454-8-git-send-email-cang@codeaurora.org (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Complementary changes for error handling | expand |
On 10/06/21 7:43 am, Can Guo wrote: > UFS error handling now is doing more than just re-probing, but also sending > scsi cmds, e.g., for clearing UACs, and recovering runtime PM error, which > may change runtime status of scsi devices. To protect system suspend/resume > from being disturbed by error handling, move the host_sem from wl pm ops > to ufshcd_suspend_prepare() and ufshcd_resume_complete(). Have you checked whether error handling might actually be needed after ufshcd_suspend_prepare()? Wouldn't this complexity go away if we just did recovery directly in __ufshcd_wl_suspend() and __ufshcd_wl_resume()? > > Signed-off-by: Can Guo <cang@codeaurora.org> > --- > drivers/scsi/ufs/ufshcd.c | 8 +++----- > drivers/scsi/ufs/ufshcd.h | 2 +- > 2 files changed, 4 insertions(+), 6 deletions(-) > > diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c > index c418a19..861942b 100644 > --- a/drivers/scsi/ufs/ufshcd.c > +++ b/drivers/scsi/ufs/ufshcd.c > @@ -9060,16 +9060,13 @@ static int ufshcd_wl_suspend(struct device *dev) > ktime_t start = ktime_get(); > > hba = shost_priv(sdev->host); > - down(&hba->host_sem); > > if (pm_runtime_suspended(dev)) > goto out; > > ret = __ufshcd_wl_suspend(hba, UFS_SYSTEM_PM); > - if (ret) { > + if (ret) > dev_err(&sdev->sdev_gendev, "%s failed: %d\n", __func__, ret); > - up(&hba->host_sem); > - } > > out: > if (!ret) > @@ -9102,7 +9099,6 @@ static int ufshcd_wl_resume(struct device *dev) > hba->curr_dev_pwr_mode, hba->uic_link_state); > if (!ret) > hba->is_wl_sys_suspended = false; > - up(&hba->host_sem); > return ret; > } > #endif > @@ -9665,6 +9661,7 @@ void ufshcd_resume_complete(struct device *dev) > ufshcd_rpmb_rpm_put(hba); > hba->rpmb_complete_put = false; > } > + up(&hba->host_sem); > } > EXPORT_SYMBOL_GPL(ufshcd_resume_complete); > > @@ -9691,6 +9688,7 @@ int ufshcd_suspend_prepare(struct device *dev) > ufshcd_rpmb_rpm_get_sync(hba); > hba->rpmb_complete_put = true; > } > + down(&hba->host_sem); > return 0; > } > EXPORT_SYMBOL_GPL(ufshcd_suspend_prepare); > diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h > index eaebb4e..47da47c 100644 > --- a/drivers/scsi/ufs/ufshcd.h > +++ b/drivers/scsi/ufs/ufshcd.h > @@ -693,7 +693,7 @@ struct ufs_hba_monitor { > * @ee_ctrl_mask: Exception event control mask > * @is_powered: flag to check if HBA is powered > * @shutting_down: flag to check if shutdown has been invoked > - * @host_sem: semaphore used to serialize concurrent contexts > + * @host_sem: semaphore used to avoid concurrency of contexts > * @eh_wq: Workqueue that eh_work works on > * @eh_work: Worker to handle UFS errors that require s/w attention > * @eeh_work: Worker to handle exception events >
Hi Adrian, On 2021-06-10 21:32, Adrian Hunter wrote: > On 10/06/21 7:43 am, Can Guo wrote: >> UFS error handling now is doing more than just re-probing, but also >> sending >> scsi cmds, e.g., for clearing UACs, and recovering runtime PM error, >> which >> may change runtime status of scsi devices. To protect system >> suspend/resume >> from being disturbed by error handling, move the host_sem from wl pm >> ops >> to ufshcd_suspend_prepare() and ufshcd_resume_complete(). > > Have you checked whether error handling might actually be needed after > ufshcd_suspend_prepare()? I intend to make it this (simple) way - if error handling is invoked during system suspend/resume, it should just wait until system resume is finished. suspend/resume does not count on error handling, if suspend/resume run into errors, they just fail and bail. > > Wouldn't this complexity go away if we just did recovery > directly in __ufshcd_wl_suspend() and __ufshcd_wl_resume()? > Please kindly check my reply in patch #5. Thanks, Can Guo. >> >> Signed-off-by: Can Guo <cang@codeaurora.org> >> --- >> drivers/scsi/ufs/ufshcd.c | 8 +++----- >> drivers/scsi/ufs/ufshcd.h | 2 +- >> 2 files changed, 4 insertions(+), 6 deletions(-) >> >> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c >> index c418a19..861942b 100644 >> --- a/drivers/scsi/ufs/ufshcd.c >> +++ b/drivers/scsi/ufs/ufshcd.c >> @@ -9060,16 +9060,13 @@ static int ufshcd_wl_suspend(struct device >> *dev) >> ktime_t start = ktime_get(); >> >> hba = shost_priv(sdev->host); >> - down(&hba->host_sem); >> >> if (pm_runtime_suspended(dev)) >> goto out; >> >> ret = __ufshcd_wl_suspend(hba, UFS_SYSTEM_PM); >> - if (ret) { >> + if (ret) >> dev_err(&sdev->sdev_gendev, "%s failed: %d\n", __func__, ret); >> - up(&hba->host_sem); >> - } >> >> out: >> if (!ret) >> @@ -9102,7 +9099,6 @@ static int ufshcd_wl_resume(struct device *dev) >> hba->curr_dev_pwr_mode, hba->uic_link_state); >> if (!ret) >> hba->is_wl_sys_suspended = false; >> - up(&hba->host_sem); >> return ret; >> } >> #endif >> @@ -9665,6 +9661,7 @@ void ufshcd_resume_complete(struct device *dev) >> ufshcd_rpmb_rpm_put(hba); >> hba->rpmb_complete_put = false; >> } >> + up(&hba->host_sem); >> } >> EXPORT_SYMBOL_GPL(ufshcd_resume_complete); >> >> @@ -9691,6 +9688,7 @@ int ufshcd_suspend_prepare(struct device *dev) >> ufshcd_rpmb_rpm_get_sync(hba); >> hba->rpmb_complete_put = true; >> } >> + down(&hba->host_sem); >> return 0; >> } >> EXPORT_SYMBOL_GPL(ufshcd_suspend_prepare); >> diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h >> index eaebb4e..47da47c 100644 >> --- a/drivers/scsi/ufs/ufshcd.h >> +++ b/drivers/scsi/ufs/ufshcd.h >> @@ -693,7 +693,7 @@ struct ufs_hba_monitor { >> * @ee_ctrl_mask: Exception event control mask >> * @is_powered: flag to check if HBA is powered >> * @shutting_down: flag to check if shutdown has been invoked >> - * @host_sem: semaphore used to serialize concurrent contexts >> + * @host_sem: semaphore used to avoid concurrency of contexts >> * @eh_wq: Workqueue that eh_work works on >> * @eh_work: Worker to handle UFS errors that require s/w attention >> * @eeh_work: Worker to handle exception events >>
On 6/9/21 9:43 PM, Can Guo wrote: > UFS error handling now is doing more than just re-probing, but also sending > scsi cmds, e.g., for clearing UACs, and recovering runtime PM error, which > may change runtime status of scsi devices. To protect system suspend/resume > from being disturbed by error handling, move the host_sem from wl pm ops > to ufshcd_suspend_prepare() and ufshcd_resume_complete(). If lock_system_sleep() and unlock_system_sleep() would be used in the error handler, would that allow to remove host_sem? Thanks, Bart.
On 2021-06-12 05:00, Bart Van Assche wrote: > On 6/9/21 9:43 PM, Can Guo wrote: >> UFS error handling now is doing more than just re-probing, but also >> sending >> scsi cmds, e.g., for clearing UACs, and recovering runtime PM error, >> which >> may change runtime status of scsi devices. To protect system >> suspend/resume >> from being disturbed by error handling, move the host_sem from wl pm >> ops >> to ufshcd_suspend_prepare() and ufshcd_resume_complete(). > > If lock_system_sleep() and unlock_system_sleep() would be used in the > error handler, would that allow to remove host_sem? Please kindly check my reply in patch #5. Thanks, Can Guo. > > Thanks, > > Bart.
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index c418a19..861942b 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -9060,16 +9060,13 @@ static int ufshcd_wl_suspend(struct device *dev) ktime_t start = ktime_get(); hba = shost_priv(sdev->host); - down(&hba->host_sem); if (pm_runtime_suspended(dev)) goto out; ret = __ufshcd_wl_suspend(hba, UFS_SYSTEM_PM); - if (ret) { + if (ret) dev_err(&sdev->sdev_gendev, "%s failed: %d\n", __func__, ret); - up(&hba->host_sem); - } out: if (!ret) @@ -9102,7 +9099,6 @@ static int ufshcd_wl_resume(struct device *dev) hba->curr_dev_pwr_mode, hba->uic_link_state); if (!ret) hba->is_wl_sys_suspended = false; - up(&hba->host_sem); return ret; } #endif @@ -9665,6 +9661,7 @@ void ufshcd_resume_complete(struct device *dev) ufshcd_rpmb_rpm_put(hba); hba->rpmb_complete_put = false; } + up(&hba->host_sem); } EXPORT_SYMBOL_GPL(ufshcd_resume_complete); @@ -9691,6 +9688,7 @@ int ufshcd_suspend_prepare(struct device *dev) ufshcd_rpmb_rpm_get_sync(hba); hba->rpmb_complete_put = true; } + down(&hba->host_sem); return 0; } EXPORT_SYMBOL_GPL(ufshcd_suspend_prepare); diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h index eaebb4e..47da47c 100644 --- a/drivers/scsi/ufs/ufshcd.h +++ b/drivers/scsi/ufs/ufshcd.h @@ -693,7 +693,7 @@ struct ufs_hba_monitor { * @ee_ctrl_mask: Exception event control mask * @is_powered: flag to check if HBA is powered * @shutting_down: flag to check if shutdown has been invoked - * @host_sem: semaphore used to serialize concurrent contexts + * @host_sem: semaphore used to avoid concurrency of contexts * @eh_wq: Workqueue that eh_work works on * @eh_work: Worker to handle UFS errors that require s/w attention * @eeh_work: Worker to handle exception events
UFS error handling now is doing more than just re-probing, but also sending scsi cmds, e.g., for clearing UACs, and recovering runtime PM error, which may change runtime status of scsi devices. To protect system suspend/resume from being disturbed by error handling, move the host_sem from wl pm ops to ufshcd_suspend_prepare() and ufshcd_resume_complete(). Signed-off-by: Can Guo <cang@codeaurora.org> --- drivers/scsi/ufs/ufshcd.c | 8 +++----- drivers/scsi/ufs/ufshcd.h | 2 +- 2 files changed, 4 insertions(+), 6 deletions(-)