netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Steve Wise" <swise@opengridcomputing.com>
To: "'Sinan Kaya'" <okaya@codeaurora.org>, <netdev@vger.kernel.org>,
	<timur@codeaurora.org>, <sulrich@codeaurora.org>
Cc: <linux-arm-msm@vger.kernel.org>,
	<linux-arm-kernel@lists.infradead.org>,
	"'Steve Wise'" <swise@chelsio.com>,
	"'Doug Ledford'" <dledford@redhat.com>,
	"'Jason Gunthorpe'" <jgg@ziepe.ca>, <linux-rdma@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>,
	"'Michael Werner'" <werner@chelsio.com>,
	"'Casey Leedom'" <leedom@chelsio.com>
Subject: RE: [PATCH v3 18/18] infiniband: cxgb4: Eliminate duplicate barriers on weakly-ordered archs
Date: Fri, 16 Mar 2018 16:05:10 -0500	[thread overview]
Message-ID: <003601d3bd6a$783d6970$68b83c50$@opengridcomputing.com> (raw)
In-Reply-To: <1521216991-28706-19-git-send-email-okaya@codeaurora.org>

> Code includes wmb() 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.
> 
> Since code already has an explicit barrier call, changing writel() to
> writel_relaxed().
> 
> Signed-off-by: Sinan Kaya <okaya@codeaurora.org>

NAK - This isn't correct for PowerPC.  For PowerPC, writeX_relaxed() is just
writeX().  

I was just looking at this with Chelsio developers, and they said the
writeX() should be replaced with __raw_writeX(), not writeX_relaxed(), to
get rid of the extra barrier for all architectures.

Also, t4.h:pio_copy() needs to  use __raw_writeq() to enable the write
combining fastpath for ARM and PowerPC.  The code as it stands doesn't
achieve any write combining on PowerPC at least.  

And the writel()s at the end of the ring functions (the non bar2 udb path)
needs a mmiowb() afterwards if you're going to use __raw_writeX() there.
However that path is only used for very old hardware (T4), so I wouldn't
worry about them. 

Steve.


> ---
>  drivers/infiniband/hw/cxgb4/t4.h | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/infiniband/hw/cxgb4/t4.h
> b/drivers/infiniband/hw/cxgb4/t4.h
> index 8369c7c..7a48c9e 100644
> --- a/drivers/infiniband/hw/cxgb4/t4.h
> +++ b/drivers/infiniband/hw/cxgb4/t4.h
> @@ -477,15 +477,15 @@ static inline void t4_ring_sq_db(struct t4_wq *wq,
> u16 inc, union t4_wr *wqe)
>  				 (u64 *)wqe);
>  		} else {
>  			pr_debug("DB wq->sq.pidx = %d\n", wq->sq.pidx);
> -			writel(PIDX_T5_V(inc) | QID_V(wq->sq.bar2_qid),
> -			       wq->sq.bar2_va + SGE_UDB_KDOORBELL);
> +			writel_relaxed(PIDX_T5_V(inc) | QID_V(wq-
> >sq.bar2_qid),
> +				       wq->sq.bar2_va +
> SGE_UDB_KDOORBELL);
>  		}
> 
>  		/* Flush user doorbell area writes. */
>  		wmb();
>  		return;
>  	}
> -	writel(QID_V(wq->sq.qid) | PIDX_V(inc), wq->db);
> +	writel_relaxed(QID_V(wq->sq.qid) | PIDX_V(inc), wq->db);
>  }
> 
>  static inline void t4_ring_rq_db(struct t4_wq *wq, u16 inc,
> @@ -502,15 +502,15 @@ static inline void t4_ring_rq_db(struct t4_wq *wq,
> u16 inc,
>  				 (void *)wqe);
>  		} else {
>  			pr_debug("DB wq->rq.pidx = %d\n", wq->rq.pidx);
> -			writel(PIDX_T5_V(inc) | QID_V(wq->rq.bar2_qid),
> -			       wq->rq.bar2_va + SGE_UDB_KDOORBELL);
> +			writel_relaxed(PIDX_T5_V(inc) | QID_V(wq-
> >rq.bar2_qid),
> +				       wq->rq.bar2_va +
> SGE_UDB_KDOORBELL);
>  		}
> 
>  		/* Flush user doorbell area writes. */
>  		wmb();
>  		return;
>  	}
> -	writel(QID_V(wq->rq.qid) | PIDX_V(inc), wq->db);
> +	writel_relaxed(QID_V(wq->rq.qid) | PIDX_V(inc), wq->db);
>  }
> 
>  static inline int t4_wq_in_error(struct t4_wq *wq)
> --
> 2.7.4

  reply	other threads:[~2018-03-16 21:05 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-16 16:16 [PATCH v3 00/18] Eliminate duplicate barriers on weakly-ordered archs Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 01/18] i40e/i40evf: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 02/18] ixgbe: eliminate " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 03/18] igbvf: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 04/18] igb: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 05/18] ixgbevf: keep writel() closer to wmb() Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 06/18] ixgbevf: eliminate duplicate barriers on weakly-ordered archs Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 07/18] drivers: net: cxgb: Eliminate " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 08/18] scsi: hpsa: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 09/18] fm10k: " Sinan Kaya
2018-03-16 16:30   ` [Intel-wired-lan] " Alexander Duyck
2018-03-16 16:33     ` Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 10/18] net: qla3xxx: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 11/18] qlcnic: " Sinan Kaya
2018-03-19 20:10   ` Chopra, Manish
2018-03-16 16:16 ` [PATCH v3 12/18] bnx2x: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 13/18] net: cxgb4/cxgb4vf: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 14/18] net: cxgb3: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 15/18] RDMA/bnxt_re: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 16/18] IB/mlx4: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 17/18] RDMA/i40iw: " Sinan Kaya
2018-03-16 16:16 ` [PATCH v3 18/18] infiniband: cxgb4: " Sinan Kaya
2018-03-16 21:05   ` Steve Wise [this message]
2018-03-16 21:46     ` Sinan Kaya
2018-03-16 23:05       ` Steve Wise
2018-03-17  3:40         ` Sinan Kaya
2018-03-17  4:03           ` Sinan Kaya
2018-03-17  4:25             ` Sinan Kaya
2018-03-17  4:30               ` Timur Tabi
2018-03-17 13:23               ` Steve Wise
2018-03-17 13:27               ` David Miller
2018-03-17 15:05               ` Jason Gunthorpe
2018-03-17 18:30                 ` Sinan Kaya
2018-03-19  1:48                   ` Jason Gunthorpe
2018-03-16 22:13     ` Jason Gunthorpe
2018-03-16 23:04       ` Steve Wise
2018-03-17  4:08         ` Timur Tabi

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='003601d3bd6a$783d6970$68b83c50$@opengridcomputing.com' \
    --to=swise@opengridcomputing.com \
    --cc=dledford@redhat.com \
    --cc=jgg@ziepe.ca \
    --cc=leedom@chelsio.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=okaya@codeaurora.org \
    --cc=sulrich@codeaurora.org \
    --cc=swise@chelsio.com \
    --cc=timur@codeaurora.org \
    --cc=werner@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).