Netdev List
 help / color / mirror / Atom feed
From: Liran Alon <liran.alon@oracle.com>
To: netanel@amazon.com, davem@davemloft.net, netdev@vger.kernel.org
Cc: saeedb@amazon.com, zorik@amazon.com, sameehj@amazon.com,
	igorch@amazon.com, akiyano@amazon.com, evgenys@amazon.com,
	gtzalik@amazon.com, ndagan@amazon.com, matua@amazon.com,
	galpress@amazon.com
Subject: [PATCH 0/2] net: AWS ENA: Fix memory barrier usage when using LLQ
Date: Thu,  2 Jan 2020 20:08:28 +0200	[thread overview]
Message-ID: <20200102180830.66676-1-liran.alon@oracle.com> (raw)

Hi,

This simple patch-series address 2 issues found during code-review of AWS ENA NIC driver
related to how it use memory barriers when NIC is running in "low-latency queue" mode (LLQ).

1st patch removes an unnecessary wmb().

2nd patch fix a bug of not flushing write-combined buffers holding LLQ transmit descriptors
and first packet bytes before writing new SQ tail to device. This bug is introduced because
of a weird behaviour of x86 AMD CPU that it doesn't flush write-combined buffers when CPU
writes to UC memory as x86 Intel CPU does. Which makes writeX() insufficient in order to
guarantee write-combined buffers are flushed.
This patch makes sure to just fix AWS ENA to handle this case properly, but a future patch-series
should be submitted to probably introduce a new flush_wc_writeX() macro that handles this case
properly for all CPU archs and vendors. For more info, see patch commit message itself.

Regards,
-Liran


             reply	other threads:[~2020-01-02 18:09 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-02 18:08 Liran Alon [this message]
2020-01-02 18:08 ` [PATCH 1/2] net: AWS ENA: Remove unncessary wmb() to flush bounce buffer Liran Alon
2020-01-17 15:54   ` Liran Alon
2020-01-02 18:08 ` [PATCH 2/2] net: AWS ENA: Flush WCBs before writing new SQ tail to doorbell Liran Alon
2020-01-03 18:46   ` Liran Alon
2020-01-04  4:55     ` Machulsky, Zorik
2020-01-05  9:53       ` Bshara, Saeed
2020-01-05 10:22         ` Liran Alon
2020-01-05 11:49           ` Bshara, Saeed

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=20200102180830.66676-1-liran.alon@oracle.com \
    --to=liran.alon@oracle.com \
    --cc=akiyano@amazon.com \
    --cc=davem@davemloft.net \
    --cc=evgenys@amazon.com \
    --cc=galpress@amazon.com \
    --cc=gtzalik@amazon.com \
    --cc=igorch@amazon.com \
    --cc=matua@amazon.com \
    --cc=ndagan@amazon.com \
    --cc=netanel@amazon.com \
    --cc=netdev@vger.kernel.org \
    --cc=saeedb@amazon.com \
    --cc=sameehj@amazon.com \
    --cc=zorik@amazon.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