LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Haiyang Zhang <haiyangz@microsoft.com>
To: Jason Wang <jasowang@redhat.com>
Cc: "davem@davemloft.net" <davem@davemloft.net>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	KY Srinivasan <kys@microsoft.com>,
	"olaf@aepfle.de" <olaf@aepfle.de>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"driverdev-devel@linuxdriverproject.org" 
	<driverdev-devel@linuxdriverproject.org>
Subject: RE: [PATCH net] hyperv: Fix the error processing in netvsc_send()
Date: Wed, 4 Feb 2015 22:26:51 +0000	[thread overview]
Message-ID: <BN1PR0301MB0770914291EB27A38F9C69EFCA3A0@BN1PR0301MB0770.namprd03.prod.outlook.com> (raw)
In-Reply-To: <1423034952.10558.3@smtp.corp.redhat.com>

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 2850 bytes --]



> -----Original Message-----
> From: Jason Wang [mailto:jasowang@redhat.com]
> Sent: Wednesday, February 4, 2015 2:29 AM
> > The EAGAIN error doesn't normally happen, because we set the hi water
> > mark
> > to stop send queue.
> 
> This is not true since only txq was stopped which means only network
> stack stop sending packets but not for control path e.g
> rndis_filter_send_request() or other callers who call
> vmbus_sendpacket() directly (e.g recv completion).
> 
> For control path, user may meet several errors when they want to change
> mac address under heavy load.
> 
> What's more serious is netvsc_send_recv_completion(), it can not even
> recover from more than 3 times of EAGAIN.
> 
> I must say mixing data packets with control packets with the same
> channel sounds really scary. Since control packets could be blocked or
> even dropped because of data packets already queued during heavy load,
> and you need to synchronize two paths carefully (e.g I didn't see any
> tx lock were held if rndis_filter_send_request() call netsc_send()
> which may stop or start a queue).

The RING_AVAIL_PERCENT_HIWATER is defined to be 20, so the data traffic
can only occupy 20% of the ring buffer before stopping the txq. So, this
mechanism ensures the control messages are not blocked by data traffic.

> >  If in really rare case, the ring buffer is full and there
> > is no outstanding sends, we can't stop queue here because there will
> > be no
> > send-completion msg to wake it up.
> 
> Confused, I believe only txq is stopped but we may still get completion
> interrupt in this case.

If there is no outstanding sends in this queue (queue_sends[q_idx]), we 
won't receive any more send-completion msg.

> 
> > And, the ring buffer is likely to be
> > occupied by other special msg, e.g. receive-completion msg (not a
> > normal case),
> > so we can't assume there are available slots.
> 
> Then why not checking hv_ringbuf_avail_percent() instead? And there's
> no need to check queue_sends since it does not count recv completion.

When ret == -EAGAIN, which means the ring is full, we don't need to check
hv_ringbuf_avail_percent().

> > We don't request retry from
> > the upper layer in this case to avoid possible busy retry.
> 
> Can't we just do this by stopping txq and depending on tx interrupt to
> wake it?

There is no tx interrupt. Do you mean rx interrupt for the send-completion?

In usual cases, when we hit the high water mark, the stopped queue depends on
the send-completion msg to wake up. But, not in some special cases.
As said above, we won't receive any more send-completion msg when there is 
no outstanding sends in this queue.

Thanks,
- Haiyang

ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥

  reply	other threads:[~2015-02-04 22:41 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-29 20:34 Haiyang Zhang
2015-01-30 10:25 ` Jason Wang
2015-01-30 15:05   ` Haiyang Zhang
     [not found]     ` <BN1PR0301MB0770FCDA58F3BC9E25382D95CA310@BN1PR0301MB0770.namprd03.prod.out look.com>
2015-02-02  6:49       ` Jason Wang
2015-02-03 15:46         ` Haiyang Zhang
     [not found]           ` <BN1PR0301MB077018D4A512E3AA9B8583E0CA3D0@BN1PR0301MB0770.namprd03.prod.out look.com>
2015-02-04  7:29             ` Jason Wang
2015-02-04 22:26               ` Haiyang Zhang [this message]
2015-02-01  1:32 ` David Miller

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=BN1PR0301MB0770914291EB27A38F9C69EFCA3A0@BN1PR0301MB0770.namprd03.prod.outlook.com \
    --to=haiyangz@microsoft.com \
    --cc=davem@davemloft.net \
    --cc=driverdev-devel@linuxdriverproject.org \
    --cc=jasowang@redhat.com \
    --cc=kys@microsoft.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=olaf@aepfle.de \
    --subject='RE: [PATCH net] hyperv: Fix the error processing in netvsc_send()' \
    /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

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).