mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Avri Altman <avri.altman@wdc.com>
To: "James E . J . Bottomley" <jejb@linux.vnet.ibm.com>,
	"Martin K . Petersen" <martin.petersen@oracle.com>,
	linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: gregkh@linuxfoundation.org, Bart Van Assche <bvanassche@acm.org>,
	yongmyung lee <ymhungry.lee@samsung.com>,
	Daejun Park <daejun7.park@samsung.com>,
	alim.akhtar@samsung.com, asutoshd@codeaurora.org,
	Zang Leigang <zangleigang@hisilicon.com>,
	Avi Shchislowski <avi.shchislowski@wdc.com>,
	Bean Huo <beanhuo@micron.com>,
	cang@codeaurora.org, stanley.chu@mediatek.com,
	Avri Altman <avri.altman@wdc.com>
Subject: [PATCH v2 4/9] scsi: ufshpb: Make eviction depends on region's reads
Date: Tue,  2 Feb 2021 10:30:02 +0200	[thread overview]
Message-ID: <20210202083007.104050-5-avri.altman@wdc.com> (raw)
In-Reply-To: <20210202083007.104050-1-avri.altman@wdc.com>

In host mode, eviction is considered an extreme measure.
verify that the entering region has enough reads, and the exiting
region has much less reads.

Signed-off-by: Avri Altman <avri.altman@wdc.com>
---
 drivers/scsi/ufs/ufshpb.c | 42 ++++++++++++++++++++++++++++++++++++---
 1 file changed, 39 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/ufs/ufshpb.c b/drivers/scsi/ufs/ufshpb.c
index de4866d42df0..bae7dca105da 100644
--- a/drivers/scsi/ufs/ufshpb.c
+++ b/drivers/scsi/ufs/ufshpb.c
@@ -18,6 +18,7 @@
 
 #define WORK_PENDING 0
 #define ACTIVATION_THRSHLD 4 /* 4 IOs */
+#define EVICTION_THRSHLD (ACTIVATION_THRSHLD << 6) /* 256 IOs */
 
 /* memory management */
 static struct kmem_cache *ufshpb_mctx_cache;
@@ -644,6 +645,13 @@ static struct ufshpb_region *ufshpb_victim_lru_info(struct ufshpb_lu *hpb)
 		if (ufshpb_check_srgns_issue_state(hpb, rgn))
 			continue;
 
+		/*
+		 * in host control mode, verify that the exiting region
+		 * has less reads
+		 */
+		if (hpb->is_hcm && rgn->reads > (EVICTION_THRSHLD >> 1))
+			continue;
+
 		victim_rgn = rgn;
 		break;
 	}
@@ -799,7 +807,7 @@ static int ufshpb_issue_map_req(struct ufshpb_lu *hpb,
 
 static int ufshpb_add_region(struct ufshpb_lu *hpb, struct ufshpb_region *rgn)
 {
-	struct ufshpb_region *victim_rgn;
+	struct ufshpb_region *victim_rgn = NULL;
 	struct victim_select_info *lru_info = &hpb->lru_info;
 	unsigned long flags;
 	int ret = 0;
@@ -827,6 +835,16 @@ static int ufshpb_add_region(struct ufshpb_lu *hpb, struct ufshpb_region *rgn)
 			 * because the device could detect this region
 			 * by not issuing HPB_READ
 			 */
+
+			/*
+			 * in host control mode, verify that the entering
+			 * region has enough reads
+			 */
+			if (hpb->is_hcm && rgn->reads < EVICTION_THRSHLD) {
+				ret = -EACCES;
+				goto out;
+			}
+
 			victim_rgn = ufshpb_victim_lru_info(hpb);
 			if (!victim_rgn) {
 				dev_warn(&hpb->sdev_ufs_lu->sdev_dev,
@@ -1034,8 +1052,13 @@ static void ufshpb_run_active_subregion_list(struct ufshpb_lu *hpb)
 
 		rgn = hpb->rgn_tbl + srgn->rgn_idx;
 		ret = ufshpb_add_region(hpb, rgn);
-		if (ret)
-			goto active_failed;
+		if (ret) {
+			if (ret == -EACCES) {
+				spin_lock_irqsave(&hpb->rsp_list_lock, flags);
+				continue;
+			}
+			goto add_region_failed;
+		}
 
 		ret = ufshpb_issue_map_req(hpb, rgn, srgn);
 		if (ret) {
@@ -1049,9 +1072,22 @@ static void ufshpb_run_active_subregion_list(struct ufshpb_lu *hpb)
 	spin_unlock_irqrestore(&hpb->rsp_list_lock, flags);
 	return;
 
+add_region_failed:
+	if (hpb->is_hcm) {
+		/*
+		 * In host control mode, it is common that eviction trials
+		 * fail, either because the entering region didn't have enough
+		 * reads, or a poor-performing exiting region wasn't found.
+		 * No need to re-insert those regions to the "to-be-activated"
+		 * list.
+		 */
+		return;
+	}
+
 active_failed:
 	dev_err(&hpb->sdev_ufs_lu->sdev_dev, "failed to activate region %d - %d, will retry\n",
 		   rgn->rgn_idx, srgn->srgn_idx);
+
 	spin_lock_irqsave(&hpb->rsp_list_lock, flags);
 	ufshpb_add_active_list(hpb, rgn, srgn);
 	spin_unlock_irqrestore(&hpb->rsp_list_lock, flags);
-- 
2.25.1


  parent reply	other threads:[~2021-02-02  8:32 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-02  8:29 [PATCH v2 0/9] Add Host control mode to HPB Avri Altman
2021-02-02  8:29 ` [PATCH v2 1/9] scsi: ufshpb: Cache HPB Control mode on init Avri Altman
2021-02-02  8:30 ` [PATCH v2 2/9] scsi: ufshpb: Add host control mode support to rsp_upiu Avri Altman
2021-02-02 11:13   ` Greg KH
2021-02-02 11:24     ` Avri Altman
2021-02-02 11:51       ` Greg KH
2021-02-02 11:54         ` Avri Altman
2021-02-02  8:30 ` [PATCH v2 3/9] scsi: ufshpb: Add region's reads counter Avri Altman
2021-02-02 11:15   ` Greg KH
2021-02-02 11:23     ` Avri Altman
2021-02-02 11:51       ` Greg KH
2021-02-02 11:16   ` Greg KH
2021-02-02 11:22     ` Avri Altman
2021-02-02  8:30 ` Avri Altman [this message]
2021-02-02  8:30 ` [PATCH v2 5/9] scsi: ufshpb: Region inactivation in host mode Avri Altman
2021-02-02  8:30 ` [PATCH v2 6/9] scsi: ufshpb: Add hpb dev reset response Avri Altman
2021-02-08  6:34   ` Can Guo
2021-02-08  7:50     ` Avri Altman
2021-02-02  8:30 ` [PATCH v2 7/9] scsi: ufshpb: Add "Cold" regions timer Avri Altman
2021-02-02  8:30 ` [PATCH v2 8/9] scsi: ufshpb: Add support for host control mode Avri Altman
2021-02-02  8:30 ` [PATCH v2 9/9] scsi: ufshpb: Make host mode parameters configurable Avri Altman
2021-02-02 11:17   ` Greg KH
2021-02-02 11:21     ` Avri Altman
     [not found]     ` <CGME20210202112157epcas2p3b04677e1380e0fa2c3f38217dfcba8bf@epcms2p8>
2021-02-04  0:48       ` Daejun Park
2021-02-02 11:18   ` Greg KH
2021-02-02 11:20     ` Avri Altman
2021-02-05 11:36 ` [PATCH v2 0/9] Add Host control mode to HPB Bean Huo
2021-02-05 11:43   ` Bean Huo
2021-02-07 14:07     ` Avri Altman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210202083007.104050-5-avri.altman@wdc.com \
    --to=avri.altman@wdc.com \
    --cc=alim.akhtar@samsung.com \
    --cc=asutoshd@codeaurora.org \
    --cc=avi.shchislowski@wdc.com \
    --cc=beanhuo@micron.com \
    --cc=bvanassche@acm.org \
    --cc=cang@codeaurora.org \
    --cc=daejun7.park@samsung.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jejb@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=stanley.chu@mediatek.com \
    --cc=ymhungry.lee@samsung.com \
    --cc=zangleigang@hisilicon.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®