Re: [PATCH v2] ufs: core: fix deadlock when rtc update

On Mon, 2024-07-15 at 14:38 +0800, peter.wang@mediatek.com wrote:

@@ -8171,7 +8171,10 @@ static void ufshcd_update_rtc(struct ufs_hba *hba)          */         val = ts64.tv_sec - hba->dev_info.rtc_time_baseline;   -       ufshcd_rpm_get_sync(hba); +       /* Skip update RTC if RPM state is not RPM_ACTIVE */ +       if (ufshcd_rpm_get_if_active(hba) <= 0) +               return;

err = ufshcd_query_attr(hba, UPIU_QUERY_OPCODE_WRITE_ATTR, QUERY_ATTR_IDN_SECONDS_PASSED,                                 0, 0, &val);         ufshcd_rpm_put_sync(hba);

My suggestion would be to not return here and just skip the update, but reschedule it for the next time that doesn't affect the suspend/resume flow you're worried about.