linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bjorn Helgaas <helgaas@kernel.org>
To: Wesley Sheng <wesley.sheng@microchip.com>
Cc: kurt.schwemmer@microsemi.com, logang@deltatee.com,
	linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
	wesleyshenggit@sina.com
Subject: Re: [PATCH 5/5] switchtec: MRPC DMA mode implementation
Date: Thu, 13 Dec 2018 09:17:56 -0600	[thread overview]
Message-ID: <20181213151756.GD4701@google.com> (raw)
In-Reply-To: <1544433144-7563-6-git-send-email-wesley.sheng@microchip.com>

On Mon, Dec 10, 2018 at 05:12:24PM +0800, Wesley Sheng wrote:
> MRPC normal mode requires the host to read the MRPC command status and
> output data from BAR. This results in high latency responses from the
> Memory Read TLP and potential Completion Timeout (CTO).
> 
> MRPC DMA mode implementation includes:
> Macro definitions for registers and data structures corresponding to
> MRPC DMA mode.
> 
> Add module parameter use_dma_mrpc to select between MRPC DMA mode and
> MRPC normal mode.
> 
> Add MRPC mode functionality to:
> * Retrieve MRPC DMA mode version
> * Allocate DMA buffer, ISR registration, and enable DMA function during
>   initialization
> * Check MRPC execution status and collect execution results from DMA buffer
> * Release DMA buffer and disable DMA function when unloading module
> 
> MRPC DMA mode is a new feature of firmware and the driver will fall back
> to MRPC normal mode if there is no support in the legacy firmware.
> 
> Include <linux/io-64-nonatomic-lo-hi.h> so that readq/writeq is replaced
> by two readl/writel on systems that do not support it.
> 
> Signed-off-by: Wesley Sheng <wesley.sheng@microchip.com>
> Reviewed-by: Logan Gunthorpe <logang@deltatee.com>

>  static void init_pff(struct switchtec_dev *stdev)
> @@ -1294,6 +1367,19 @@ static int switchtec_init_pci(struct switchtec_dev *stdev,
>  
>  	pci_set_drvdata(pdev, stdev);
>  
> +	if (!use_dma_mrpc)
> +		return 0;
> +
> +	if (!(ioread32(&stdev->mmio_mrpc->dma_ver) ? true : false))
> +		return 0;

This is ... harder to decode than necessary.  It's obvious that the
intent is to return if the adapter firmware doesn't support DMA, but I
lost interest before I could verify that it works as intended.

Bjorn

  parent reply	other threads:[~2018-12-13 15:18 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-10  9:12 [PATCH 0/5] Switchtec MRPC DMA mode support Wesley Sheng
2018-12-10  9:12 ` [PATCH 1/5] switchtec: Remove immediate status check after submit a MRPC command Wesley Sheng
2018-12-10  9:12 ` [PATCH 2/5] switchtec: Set DMA coherent mask in Switchtec driver Wesley Sheng
2018-12-10  9:12 ` [PATCH 3/5] switchtec: A temporary variable should be used for the flags of switchtec_ioctl_event_ctl Wesley Sheng
2018-12-12 22:43   ` Bjorn Helgaas
2018-12-12 22:52     ` Logan Gunthorpe
2018-12-10  9:12 ` [PATCH 4/5] switchtec: Improve MRPC efficiency by leveraging write combining Wesley Sheng
2018-12-10  9:12 ` [PATCH 5/5] switchtec: MRPC DMA mode implementation Wesley Sheng
2018-12-12 22:52   ` Bjorn Helgaas
2018-12-12 22:58     ` Logan Gunthorpe
2018-12-13 15:16       ` Bjorn Helgaas
2018-12-13 15:17   ` Bjorn Helgaas [this message]
2018-12-13 16:46     ` Logan Gunthorpe
2018-12-13 15:20 ` [PATCH 0/5] Switchtec MRPC DMA mode support Bjorn Helgaas
  -- strict thread matches above, loose matches on Subject: below --
2018-11-15  9:43 Wesley Sheng
2018-11-15  9:44 ` [PATCH 5/5] switchtec: MRPC DMA mode implementation Wesley Sheng

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=20181213151756.GD4701@google.com \
    --to=helgaas@kernel.org \
    --cc=kurt.schwemmer@microsemi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=logang@deltatee.com \
    --cc=wesley.sheng@microchip.com \
    --cc=wesleyshenggit@sina.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).