All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sinan Kaya <okaya@codeaurora.org>
To: linux-scsi@vger.kernel.org, timur@codeaurora.org, sulrich@codeaurora.org
Cc: linux-arm-msm@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	Sinan Kaya <okaya@codeaurora.org>,
	"James E.J. Bottomley" <jejb@linux.vnet.ibm.com>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	Varun Prakash <varun@chelsio.com>,
	Peter Senna Tschudin <peter.senna@collabora.com>,
	Romain Perier <romain.perier@collabora.com>,
	linux-kernel@vger.kernel.org
Subject: [PATCH v4 7/7] scsi: csiostor: Eliminate duplicate barriers on weakly-ordered archs
Date: Mon, 19 Mar 2018 22:50:07 -0400	[thread overview]
Message-ID: <1521514207-10695-8-git-send-email-okaya@codeaurora.org> (raw)
In-Reply-To: <1521514207-10695-1-git-send-email-okaya@codeaurora.org>

Code includes barrier() followed by writel(). writel() already has a
barrier
on some architectures like arm64.

This ends up CPU observing two barriers back to back before executing the
register write.

Create a new wrapper function with relaxed write operator. Use the new
wrapper when a write is following a barrier().

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
---
 drivers/scsi/csiostor/csio_hw.h | 4 ++++
 drivers/scsi/csiostor/csio_wr.c | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/csiostor/csio_hw.h b/drivers/scsi/csiostor/csio_hw.h
index 30f5f52..9fd8b00 100644
--- a/drivers/scsi/csiostor/csio_hw.h
+++ b/drivers/scsi/csiostor/csio_hw.h
@@ -512,6 +512,10 @@ struct csio_hw {
 						csio_reg((_h)->regstart, (_r)))
 #define	csio_wr_reg16(_h, _v, _r)	writew((_v), \
 						csio_reg((_h)->regstart, (_r)))
+
+#define	csio_wr_reg32_relaxed(_h, _v, _r) \
+	writel_relaxed((_v), csio_reg((_h)->regstart, (_r)))
+
 #define	csio_wr_reg32(_h, _v, _r)	writel((_v), \
 						csio_reg((_h)->regstart, (_r)))
 #define	csio_wr_reg64(_h, _v, _r)	writeq((_v), \
diff --git a/drivers/scsi/csiostor/csio_wr.c b/drivers/scsi/csiostor/csio_wr.c
index c0a1778..db26222 100644
--- a/drivers/scsi/csiostor/csio_wr.c
+++ b/drivers/scsi/csiostor/csio_wr.c
@@ -984,7 +984,7 @@ csio_wr_issue(struct csio_hw *hw, int qidx, bool prio)
 
 	wmb();
 	/* Ring SGE Doorbell writing q->pidx into it */
-	csio_wr_reg32(hw, DBPRIO_V(prio) | QID_V(q->un.eq.physeqid) |
+	csio_wr_reg32_relaxed(hw, DBPRIO_V(prio) | QID_V(q->un.eq.physeqid) |
 			  PIDX_T5_V(q->inc_idx) | DBTYPE_F,
 			  MYPF_REG(SGE_PF_KDOORBELL_A));
 	q->inc_idx = 0;
-- 
2.7.4

WARNING: multiple messages have this Message-ID (diff)
From: okaya@codeaurora.org (Sinan Kaya)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v4 7/7] scsi: csiostor: Eliminate duplicate barriers on weakly-ordered archs
Date: Mon, 19 Mar 2018 22:50:07 -0400	[thread overview]
Message-ID: <1521514207-10695-8-git-send-email-okaya@codeaurora.org> (raw)
In-Reply-To: <1521514207-10695-1-git-send-email-okaya@codeaurora.org>

Code includes barrier() followed by writel(). writel() already has a
barrier
on some architectures like arm64.

This ends up CPU observing two barriers back to back before executing the
register write.

Create a new wrapper function with relaxed write operator. Use the new
wrapper when a write is following a barrier().

Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
---
 drivers/scsi/csiostor/csio_hw.h | 4 ++++
 drivers/scsi/csiostor/csio_wr.c | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/csiostor/csio_hw.h b/drivers/scsi/csiostor/csio_hw.h
index 30f5f52..9fd8b00 100644
--- a/drivers/scsi/csiostor/csio_hw.h
+++ b/drivers/scsi/csiostor/csio_hw.h
@@ -512,6 +512,10 @@ struct csio_hw {
 						csio_reg((_h)->regstart, (_r)))
 #define	csio_wr_reg16(_h, _v, _r)	writew((_v), \
 						csio_reg((_h)->regstart, (_r)))
+
+#define	csio_wr_reg32_relaxed(_h, _v, _r) \
+	writel_relaxed((_v), csio_reg((_h)->regstart, (_r)))
+
 #define	csio_wr_reg32(_h, _v, _r)	writel((_v), \
 						csio_reg((_h)->regstart, (_r)))
 #define	csio_wr_reg64(_h, _v, _r)	writeq((_v), \
diff --git a/drivers/scsi/csiostor/csio_wr.c b/drivers/scsi/csiostor/csio_wr.c
index c0a1778..db26222 100644
--- a/drivers/scsi/csiostor/csio_wr.c
+++ b/drivers/scsi/csiostor/csio_wr.c
@@ -984,7 +984,7 @@ csio_wr_issue(struct csio_hw *hw, int qidx, bool prio)
 
 	wmb();
 	/* Ring SGE Doorbell writing q->pidx into it */
-	csio_wr_reg32(hw, DBPRIO_V(prio) | QID_V(q->un.eq.physeqid) |
+	csio_wr_reg32_relaxed(hw, DBPRIO_V(prio) | QID_V(q->un.eq.physeqid) |
 			  PIDX_T5_V(q->inc_idx) | DBTYPE_F,
 			  MYPF_REG(SGE_PF_KDOORBELL_A));
 	q->inc_idx = 0;
-- 
2.7.4

  parent reply	other threads:[~2018-03-20  2:50 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-20  2:50 [PATCH v4 0/7] scsi: Eliminate duplicate barriers on weakly-ordered archs Sinan Kaya
2018-03-20  2:50 ` Sinan Kaya
2018-03-20  2:50 ` [PATCH v4 1/7] scsi: hpsa: " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-03-20 14:01   ` Don Brace
2018-03-20 14:01     ` Don Brace
2018-03-20 16:51   ` Laurence Oberman
2018-03-20 16:51     ` Laurence Oberman
2018-03-20  2:50 ` [PATCH v4 2/7] scsi: megaraid: eliminate " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-03-20  2:50 ` [PATCH v4 3/7] scsi: dpt_i2o: " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-03-20 10:23   ` Christoph Hellwig
2018-03-20 10:23     ` Christoph Hellwig
2018-03-20 12:13     ` okaya
2018-03-20 12:13       ` okaya at codeaurora.org
2018-03-20  2:50 ` [PATCH v4 4/7] scsi: lpfc: Eliminate " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-03-20  2:50 ` [PATCH v4 5/7] scsi: ipr: " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-05-29 20:45   ` Brian King
2018-05-29 20:45     ` Brian King
2018-03-20  2:50 ` [PATCH v4 6/7] scsi: bnx2i: " Sinan Kaya
2018-03-20  2:50   ` Sinan Kaya
2018-03-20  2:50 ` Sinan Kaya [this message]
2018-03-20  2:50   ` [PATCH v4 7/7] scsi: csiostor: " Sinan Kaya

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=1521514207-10695-8-git-send-email-okaya@codeaurora.org \
    --to=okaya@codeaurora.org \
    --cc=jejb@linux.vnet.ibm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=peter.senna@collabora.com \
    --cc=romain.perier@collabora.com \
    --cc=sulrich@codeaurora.org \
    --cc=timur@codeaurora.org \
    --cc=varun@chelsio.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.