Linux PCI Non-Transparent Bridge framework and drivers
 help / color / mirror / Atom feed
From: Logan Gunthorpe <logang@deltatee.com>
To: Jon Mason <jdmason@kudzu.us>, Dave Jiang <dave.jiang@intel.com>,
	Allen Hubbe <Allen.Hubbe@emc.com>,
	John Kading <john.kading@gd-ms.com>,
	Sudip Mukherjee <sudipm.mukherjee@gmail.com>,
	Arnd Bergmann <arnd@arndb.de>
Cc: linux-ntb@googlegroups.com, linux-kernel@vger.kernel.org,
	Logan Gunthorpe <logang@deltatee.com>
Subject: [PATCH 0/3] ntb: Fixes and enhancements to ntb tools
Date: Fri,  3 Jun 2016 14:50:30 -0600	[thread overview]
Message-ID: <cover.1464986161.git.logang@deltatee.com> (raw)

Hi,

I've been working on developing an experimental NTB driver for some
custom hardware and I've found the need to make some minor enhancements
to the NTB layer.

1) I've modified ntb_perf to take an option similar to ntb_transport
which limits the memory window size. This was useful seeing the mws
I'm dealing with are much larger than the available coherent memory I
can allocate.

2) I've added code to ntb_perf and ntb_transport to check that there
are enough scratchpad registers. (As I was hit by a problem where my
hardware did not have enough for ntb_transport, and I would have liked
better information on the cause of the issue.)

3) Added support to debug and test memory windows to ntb_tool. A
coherent buffer is added when the link comes up, and then a debugfs
file for each mw and peer mw is added which allows reading and writing
the buffer. This was useful for me to debug window alignment issues I
was having.

I'm happy to make any revisions to these patches if anyone finds any
issues.

Thanks,

Logan


Logan Gunthorpe (3):
  ntb_perf: Allow limiting the size of the memory windows
  ntb_transport: Check the number of spads the hardware supports
  ntb_tool: Add memory window debug support

 drivers/ntb/ntb_transport.c |   9 +-
 drivers/ntb/test/ntb_perf.c |  16 ++-
 drivers/ntb/test/ntb_tool.c | 258 +++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 278 insertions(+), 5 deletions(-)

--
2.1.4

             reply	other threads:[~2016-06-03 20:53 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-03 20:50 Logan Gunthorpe [this message]
2016-06-03 20:50 ` [PATCH 1/3] ntb_perf: Allow limiting the size of the memory windows Logan Gunthorpe
2016-06-03 21:03   ` Jiang, Dave
2016-06-04 15:25     ` Jon Mason
2016-06-03 20:50 ` [PATCH 2/3] ntb_transport: Check the number of spads the hardware supports Logan Gunthorpe
2016-06-04 15:40   ` Jon Mason
2016-06-07 17:15     ` Logan Gunthorpe
2016-06-07 17:20       ` [PATCH v2 " Logan Gunthorpe
2016-06-07 17:24         ` Jiang, Dave
2016-06-09 14:51           ` Jon Mason
2016-06-03 20:50 ` [PATCH 3/3] ntb_tool: Add memory window debug support Logan Gunthorpe
2016-06-03 21:20   ` Allen Hubbe
2016-06-04 15:25     ` Jon Mason

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=cover.1464986161.git.logang@deltatee.com \
    --to=logang@deltatee.com \
    --cc=Allen.Hubbe@emc.com \
    --cc=arnd@arndb.de \
    --cc=dave.jiang@intel.com \
    --cc=jdmason@kudzu.us \
    --cc=john.kading@gd-ms.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-ntb@googlegroups.com \
    --cc=sudipm.mukherjee@gmail.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