| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <1599725880.10649.35.camel@mtkswgap22>
Date: Thu, 10 Sep 2020 16:18:00 +0800
From: Stanley Chu <stanley.chu@...iatek.com>
To: James Bottomley <James.Bottomley@...senPartnership.com>
CC: "Martin K. Petersen" <martin.petersen@...cle.com>,
Can Guo <cang@...eaurora.org>, <asutoshd@...eaurora.org>,
<nguyenb@...eaurora.org>, <hongwus@...eaurora.org>,
<ziqichen@...eaurora.org>, <rnayak@...eaurora.org>,
<linux-scsi@...r.kernel.org>, <kernel-team@...roid.com>,
<saravanak@...gle.com>, <salyzyn@...gle.com>,
Alim Akhtar <alim.akhtar@...sung.com>,
Avri Altman <avri.altman@....com>,
Matthias Brugger <matthias.bgg@...il.com>,
Bean Huo <beanhuo@...ron.com>,
Bart Van Assche <bvanassche@....org>,
open list <linux-kernel@...r.kernel.org>,
"moderated list:ARM/Mediatek SoC support"
<linux-arm-kernel@...ts.infradead.org>,
"moderated list:ARM/Mediatek SoC support"
<linux-mediatek@...ts.infradead.org>
Subject: Re: [PATCH v2 1/2] scsi: ufs: Abort tasks before clear them from
doorbell
Hi James,
On Wed, 2020-09-09 at 23:18 -0700, James Bottomley wrote:
> On Thu, 2020-09-10 at 10:48 +0800, Stanley Chu wrote:
> > Hi Martin, Can,
> >
> > On Wed, 2020-09-09 at 22:32 -0400, Martin K. Petersen wrote:
> > > Can and Stanley,
> > >
> > > > I can't reconcile this hunk:
> > >
> > > Please provide a resolution for these conflicting commits in fixes
> > > and
> > > queue:
> > >
> > > 307348f6ab14 scsi: ufs: Abort tasks before clearing them from
> > > doorbell
> > >
> > > b10178ee7fa8 scsi: ufs: Clean up completed request without
> > > interrupt
> > > notification
> > >
> >
> > Can's patch has considered my fix in the new flow.
> >
> > To be more clear, for the fixing case in my patch,
> > ufshcd_try_to_abort_task() will return 0 (err = 0) and finally the
> > target tag can be completed and cleared by
> > __ufshcd_transfer_req_compl()
> > in Can's new flow.
> >
> > Thus I think the resolution can just using the code in Can's patch.
> >
> > Can, please correct me if I was wrong.
>
> Well, that really doesn't make for an easy merge. The resolution I took
> is below.
>
> James
>
> ---
>
> commit 5399a4aa684d491c35a386effe385c06b41398fa
> Merge: 59958f7a956b 8c6572356646
> Author: James Bottomley <James.Bottomley@...senPartnership.com>
> Date: Wed Sep 9 23:12:52 2020 -0700
>
> Merge branch 'misc' into for-next
>
> Conflicts:
> drivers/scsi/ufs/ufshcd.c
> drivers/scsi/ufs/ufshcd.h
>
> diff --cc drivers/scsi/ufs/ufshcd.c
> index 34e1ab407b05,05716f62febe..49478c8a601f
> --- a/drivers/scsi/ufs/ufshcd.c
> +++ b/drivers/scsi/ufs/ufshcd.c
> @@@ -6574,84 -6739,22 +6736,25 @@@ static int ufshcd_abort(struct scsi_cmn
> }
> hba->req_abort_count++;
>
> - /* Skip task abort in case previous aborts failed and report failure */
> - if (lrbp->req_abort_skip) {
> - err = -EIO;
> - goto out;
> + if (!(reg & (1 << tag))) {
> + dev_err(hba->dev,
> + "%s: cmd was completed, but without a notifying intr, tag = %d",
> + __func__, tag);
> + goto cleanup;
> }
>
> - err = ufshcd_try_to_abort_task(hba, tag);
> - if (err)
> - goto out;
> -
> - spin_lock_irqsave(host->host_lock, flags);
> - __ufshcd_transfer_req_compl(hba, (1UL << tag));
> - spin_unlock_irqrestore(host->host_lock, flags);
> + /* Skip task abort in case previous aborts failed and report failure */
> - if (lrbp->req_abort_skip) {
> ++ if (lrbp->req_abort_skip)
> + err = -EIO;
> - goto out;
> - }
> -
> - for (poll_cnt = 100; poll_cnt; poll_cnt--) {
> - err = ufshcd_issue_tm_cmd(hba, lrbp->lun, lrbp->task_tag,
> - UFS_QUERY_TASK, &resp);
> - if (!err && resp == UPIU_TASK_MANAGEMENT_FUNC_SUCCEEDED) {
> - /* cmd pending in the device */
> - dev_err(hba->dev, "%s: cmd pending in the device. tag = %d\n",
> - __func__, tag);
> - break;
> - } else if (!err && resp == UPIU_TASK_MANAGEMENT_FUNC_COMPL) {
> - /*
> - * cmd not pending in the device, check if it is
> - * in transition.
> - */
> - dev_err(hba->dev, "%s: cmd at tag %d not pending in the device.\n",
> - __func__, tag);
> - reg = ufshcd_readl(hba, REG_UTP_TRANSFER_REQ_DOOR_BELL);
> - if (reg & (1 << tag)) {
> - /* sleep for max. 200us to stabilize */
> - usleep_range(100, 200);
> - continue;
> - }
> - /* command completed already */
> - dev_err(hba->dev, "%s: cmd at tag %d successfully cleared from DB.\n",
> - __func__, tag);
> - goto cleanup;
> - } else {
> - dev_err(hba->dev,
> - "%s: no response from device. tag = %d, err %d\n",
> - __func__, tag, err);
> - if (!err)
> - err = resp; /* service response error */
> - goto out;
> - }
> - }
> -
> - if (!poll_cnt) {
> - err = -EBUSY;
> - goto out;
> - }
> -
> - err = ufshcd_issue_tm_cmd(hba, lrbp->lun, lrbp->task_tag,
> - UFS_ABORT_TASK, &resp);
> - if (err || resp != UPIU_TASK_MANAGEMENT_FUNC_COMPL) {
> - if (!err) {
> - err = resp; /* service response error */
> - dev_err(hba->dev, "%s: issued. tag = %d, err %d\n",
> - __func__, tag, err);
> - }
> - goto out;
> - }
> -
> - err = ufshcd_clear_cmd(hba, tag);
> - if (err) {
> - dev_err(hba->dev, "%s: Failed clearing cmd at tag %d, err %d\n",
> - __func__, tag, err);
> - goto out;
> - }
> ++ else
> ++ err = ufshcd_try_to_abort_task(hba, tag);
>
> -out:
> + if (!err) {
> +cleanup:
Yeah, considering Bean Huo's patch "scsi: ufs: No need to send Abort
Task if the task in DB was cleared", "cleanup" label shall be added back
here.
So your resolution looks good to me.
Thanks so much : )
Stanley Chu
> - spin_lock_irqsave(host->host_lock, flags);
> - __ufshcd_transfer_req_compl(hba, (1UL << tag));
> - spin_unlock_irqrestore(host->host_lock, flags);
> -
> ++ spin_lock_irqsave(host->host_lock, flags);
> ++ __ufshcd_transfer_req_compl(hba, (1UL << tag));
> ++ spin_unlock_irqrestore(host->host_lock, flags);
> +out:
> - if (!err) {
> err = SUCCESS;
> } else {
> dev_err(hba->dev, "%s: failed with err %d\n", __func__, err);
> diff --cc drivers/scsi/ufs/ufshcd.h
> index b5b2761456fb,8011fdc89fb1..6663325ed8a0
> --- a/drivers/scsi/ufs/ufshcd.h
> +++ b/drivers/scsi/ufs/ufshcd.h
> @@@ -531,11 -531,10 +531,16 @@@ enum ufshcd_quirks
> */
> UFSHCD_QUIRK_BROKEN_OCS_FATAL_ERROR = 1 << 10,
>
> + /*
> + * This quirk needs to be enabled if the host controller has
> + * auto-hibernate capability but it doesn't work.
> + */
> + UFSHCD_QUIRK_BROKEN_AUTO_HIBERN8 = 1 << 11,
> ++
> + /*
> + * This quirk needs to disable manual flush for write booster
> + */
> - UFSHCI_QUIRK_SKIP_MANUAL_WB_FLUSH_CTRL = 1 << 11,
> ++ UFSHCI_QUIRK_SKIP_MANUAL_WB_FLUSH_CTRL = 1 << 12,
> };
>
> enum ufshcd_caps {
Powered by blists - more mailing lists