2021-06-23 07:40:45

by Can Guo

[permalink] [raw]
Subject: [PATCH v4 03/10] scsi: ufs: Update the return value of supplier pm ops

rpm_get_suppliers() is returning an error only if the error is negative.
However, ufshcd_wl_resume() may return a positive error code, e.g., when
hibern8 or SSU cmd fails. Make the positive return value a negative error
code so that consumers are aware of any resume failure from their supplier.
Make the same change to ufshcd_wl_suspend() just to keep symmetry.

Signed-off-by: Can Guo <[email protected]>
---
drivers/scsi/ufs/ufshcd.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index abe5f2d..ee70522 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -8922,7 +8922,7 @@ static int __ufshcd_wl_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
ufshcd_release(hba);
}
hba->wlu_pm_op_in_progress = false;
- return ret;
+ return ret <= 0 ? ret : -EINVAL;
}

static int __ufshcd_wl_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
@@ -9009,7 +9009,7 @@ static int __ufshcd_wl_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
hba->clk_gating.is_suspended = false;
ufshcd_release(hba);
hba->wlu_pm_op_in_progress = false;
- return ret;
+ return ret <= 0 ? ret : -EINVAL;
}

static int ufshcd_wl_runtime_suspend(struct device *dev)
--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.


2021-06-23 21:11:05

by Bart Van Assche

[permalink] [raw]
Subject: Re: [PATCH v4 03/10] scsi: ufs: Update the return value of supplier pm ops

On 6/23/21 12:35 AM, Can Guo wrote:
> rpm_get_suppliers() is returning an error only if the error is negative.
> However, ufshcd_wl_resume() may return a positive error code, e.g., when
> hibern8 or SSU cmd fails. Make the positive return value a negative error
> code so that consumers are aware of any resume failure from their supplier.
> Make the same change to ufshcd_wl_suspend() just to keep symmetry.
>
> Signed-off-by: Can Guo <[email protected]>
> ---
> drivers/scsi/ufs/ufshcd.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
> index abe5f2d..ee70522 100644
> --- a/drivers/scsi/ufs/ufshcd.c
> +++ b/drivers/scsi/ufs/ufshcd.c
> @@ -8922,7 +8922,7 @@ static int __ufshcd_wl_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
> ufshcd_release(hba);
> }
> hba->wlu_pm_op_in_progress = false;
> - return ret;
> + return ret <= 0 ? ret : -EINVAL;
> }
>
> static int __ufshcd_wl_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
> @@ -9009,7 +9009,7 @@ static int __ufshcd_wl_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
> hba->clk_gating.is_suspended = false;
> ufshcd_release(hba);
> hba->wlu_pm_op_in_progress = false;
> - return ret;
> + return ret <= 0 ? ret : -EINVAL;
> }

I think the above patch shows that indicating failure by either
returning a positive or a negative value is a booby trap. Please modify
ufshcd_send_request_sense() and ufshcd_set_dev_pwr_mode() such that
these return a value that is either zero or negative. Are there any
other functions than that need to be modified?

Thanks,

Bart.

2021-06-24 02:14:42

by Can Guo

[permalink] [raw]
Subject: Re: [PATCH v4 03/10] scsi: ufs: Update the return value of supplier pm ops

On 2021-06-24 05:08, Bart Van Assche wrote:
> On 6/23/21 12:35 AM, Can Guo wrote:
>> rpm_get_suppliers() is returning an error only if the error is
>> negative.
>> However, ufshcd_wl_resume() may return a positive error code, e.g.,
>> when
>> hibern8 or SSU cmd fails. Make the positive return value a negative
>> error
>> code so that consumers are aware of any resume failure from their
>> supplier.
>> Make the same change to ufshcd_wl_suspend() just to keep symmetry.
>>
>> Signed-off-by: Can Guo <[email protected]>
>> ---
>> drivers/scsi/ufs/ufshcd.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
>> index abe5f2d..ee70522 100644
>> --- a/drivers/scsi/ufs/ufshcd.c
>> +++ b/drivers/scsi/ufs/ufshcd.c
>> @@ -8922,7 +8922,7 @@ static int __ufshcd_wl_suspend(struct ufs_hba
>> *hba, enum ufs_pm_op pm_op)
>> ufshcd_release(hba);
>> }
>> hba->wlu_pm_op_in_progress = false;
>> - return ret;
>> + return ret <= 0 ? ret : -EINVAL;
>> }
>>
>> static int __ufshcd_wl_resume(struct ufs_hba *hba, enum ufs_pm_op
>> pm_op)
>> @@ -9009,7 +9009,7 @@ static int __ufshcd_wl_resume(struct ufs_hba
>> *hba, enum ufs_pm_op pm_op)
>> hba->clk_gating.is_suspended = false;
>> ufshcd_release(hba);
>> hba->wlu_pm_op_in_progress = false;
>> - return ret;
>> + return ret <= 0 ? ret : -EINVAL;
>> }
>
> I think the above patch shows that indicating failure by either
> returning a positive or a negative value is a booby trap. Please modify
> ufshcd_send_request_sense() and ufshcd_set_dev_pwr_mode() such that
> these return a value that is either zero or negative. Are there any
> other functions than that need to be modified?

Yes, there are more of them - both enter/exit_hibern8() and
reset_and_restore()
can return positive values. I will modify all of them in next ver.

Thanks,

Can Guo.

>
> Thanks,
>
> Bart.