From: Fam Zheng <famz@redhat.com>
To: Jason Wang <jasowang@redhat.com>
Cc: Peter Maydell <peter.maydell@linaro.org>,
Peter Crosthwaite <peter.crosthwaite@xilinx.com>,
Rob Herring <robh@kernel.org>,
qemu-devel@nongnu.org, Michael Walle <michael@walle.cc>,
Gerd Hoffmann <kraxel@redhat.com>,
stefanha@redhat.com,
"Edgar E. Iglesias" <edgar.iglesias@gmail.com>
Subject: Re: [Qemu-devel] [PATCH for-2.4 06/12] etsec: Flush queue when rx buffer is consumed
Date: Tue, 14 Jul 2015 17:48:40 +0800 [thread overview]
Message-ID: <20150714094840.GD27873@ad.nay.redhat.com> (raw)
In-Reply-To: <55A4D775.40006@redhat.com>
On Tue, 07/14 17:33, Jason Wang wrote:
>
>
> On 07/14/2015 03:53 PM, Fam Zheng wrote:
> > The BH will be scheduled when etsec->rx_buffer_len is becoming 0, which
> > is the condition of queuing.
> >
> > Signed-off-by: Fam Zheng <famz@redhat.com>
>
> I suggest to squash this into patch 05/12 to unbreak bisection.
05/12 didn't change the logic, it just moved the semantics from returning 0 in
.can_receive() to .receive(). So I think it's OK to split to make reviewing
easier.
>
> And can we do this without a bh? Otherwise, we may need to stop and
> restart the bh during vm stop and start?
A bh doesn't hurt when vm stop and restart (we get superfluous flush),
otherwise the call stack could go very deap with a long queue, something like:
etsec_receive
etsec_rx_ring_write
etsec_walk_rx_ring
etsec_flush_rx_queue
qemu_flush_queued_packets
...
etsec_receive
etsec_rx_ring_write
etsec_walk_rx_ring
etsec_flush_rx_queue
qemu_flush_queued_packets
/* loop */
...
>
> > ---
> > hw/net/fsl_etsec/etsec.c | 9 +++++++++
> > hw/net/fsl_etsec/etsec.h | 2 ++
> > hw/net/fsl_etsec/rings.c | 1 +
> > 3 files changed, 12 insertions(+)
> >
> > diff --git a/hw/net/fsl_etsec/etsec.c b/hw/net/fsl_etsec/etsec.c
> > index f5170ae..fcde9b4 100644
> > --- a/hw/net/fsl_etsec/etsec.c
> > +++ b/hw/net/fsl_etsec/etsec.c
> > @@ -366,6 +366,13 @@ static NetClientInfo net_etsec_info = {
> > .link_status_changed = etsec_set_link_status,
> > };
> >
> > +static void etsec_flush_rx_queue(void *opaque)
> > +{
> > + eTSEC *etsec = opaque;
> > +
> > + qemu_flush_queued_packets(qemu_get_queue(etsec->nic));
> > +}
> > +
> > static void etsec_realize(DeviceState *dev, Error **errp)
> > {
> > eTSEC *etsec = ETSEC_COMMON(dev);
> > @@ -378,6 +385,8 @@ static void etsec_realize(DeviceState *dev, Error **errp)
> > etsec->bh = qemu_bh_new(etsec_timer_hit, etsec);
> > etsec->ptimer = ptimer_init(etsec->bh);
> > ptimer_set_freq(etsec->ptimer, 100);
> > +
> > + etsec->flush_bh = qemu_bh_new(etsec_flush_rx_queue, etsec);
> > }
> >
> > static void etsec_instance_init(Object *obj)
> > diff --git a/hw/net/fsl_etsec/etsec.h b/hw/net/fsl_etsec/etsec.h
> > index fc41773..05bb7f8 100644
> > --- a/hw/net/fsl_etsec/etsec.h
> > +++ b/hw/net/fsl_etsec/etsec.h
> > @@ -144,6 +144,8 @@ typedef struct eTSEC {
> > QEMUBH *bh;
> > struct ptimer_state *ptimer;
> >
> > + QEMUBH *flush_bh;
> > +
> > } eTSEC;
> >
> > #define TYPE_ETSEC_COMMON "eTSEC"
> > diff --git a/hw/net/fsl_etsec/rings.c b/hw/net/fsl_etsec/rings.c
> > index a11280b..7aae93e 100644
> > --- a/hw/net/fsl_etsec/rings.c
> > +++ b/hw/net/fsl_etsec/rings.c
> > @@ -646,6 +646,7 @@ void etsec_walk_rx_ring(eTSEC *etsec, int ring_nbr)
> > } else {
> > etsec->rx_buffer_len = 0;
> > etsec->rx_buffer = NULL;
> > + qemu_bh_schedule(etsec->flush_bh);
> > }
> >
> > RING_DEBUG("eTSEC End of ring_write: remaining_data:%zu\n", remaining_data);
>
next prev parent reply other threads:[~2015-07-14 9:48 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-14 7:53 [Qemu-devel] [PATCH for-2.4 00/12] hw/net: Fix .can_receive() for NICs Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 01/12] xgmac: Drop packets with eth_can_rx is false Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 02/12] pcnet: Drop pcnet_can_receive Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 03/12] eepro100: Drop nic_can_receive Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 04/12] usbnet: Drop usbnet_can_receive Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 05/12] etsec: Move etsec_can_receive into etsec_receive Fam Zheng
2015-07-14 9:30 ` Jason Wang
2015-07-14 9:49 ` Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 06/12] etsec: Flush queue when rx buffer is consumed Fam Zheng
2015-07-14 9:33 ` Jason Wang
2015-07-14 9:48 ` Fam Zheng [this message]
2015-07-15 5:10 ` Jason Wang
2015-07-15 6:01 ` Fam Zheng
2015-07-15 7:38 ` Jason Wang
2015-07-15 8:04 ` Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 07/12] mcf_fec: Drop mcf_fec_can_receive Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 08/12] milkymist-minimac2: Flush queued packets when link comes up Fam Zheng
2015-07-14 11:02 ` Michael Walle
2015-07-14 11:07 ` Fam Zheng
2015-07-15 7:50 ` Michael Walle
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 09/12] mipsnet: Flush queued packets when receiving is enabled Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 10/12] stellaris_enet: Flush queued packets when read done Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 11/12] dp8393x: Flush packets when link comes up Fam Zheng
2015-07-14 7:53 ` [Qemu-devel] [PATCH for-2.4 12/12] axienet: Flush queued packets when rx is done Fam Zheng
2015-07-14 8:34 ` [Qemu-devel] [PATCH for-2.4 00/12] hw/net: Fix .can_receive() for NICs Wen Congyang
2015-07-15 8:50 ` Stefan Hajnoczi
2015-07-14 14:40 ` Stefan Hajnoczi
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=20150714094840.GD27873@ad.nay.redhat.com \
--to=famz@redhat.com \
--cc=edgar.iglesias@gmail.com \
--cc=jasowang@redhat.com \
--cc=kraxel@redhat.com \
--cc=michael@walle.cc \
--cc=peter.crosthwaite@xilinx.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=robh@kernel.org \
--cc=stefanha@redhat.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).