From: Rusty Russell <rusty-8n+1lVoiYb80n/F98K4Iww@public•gmane.org>
To: David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public•gmane.org>
Cc: herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public•gmane.org,
mb-fseUSCV1ubazQB+pC5nmwQ@public•gmane.org,
johannes-cdvu00un1VgdHxzADdlk8Q@public•gmane.org,
linux-wireless-u79uwXL29TY76Z2rM5mHXA@public•gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public•gmane.org,
ron.rindjunsky-ral2JQCrhuEAvxtiuMwx3w@public•gmane.org,
tomasw-Re5JQEeQqe8AvxtiuMwx3w@public•gmane.org,
ivdoorn-Re5JQEeQqe8AvxtiuMwx3w@public•gmane.org,
peter.p.waskiewicz.jr-ral2JQCrhuEAvxtiuMwx3w@public•gmane.org
Subject: Re: [PATCH] mac80211: rewrite fragmentation code
Date: Fri, 16 May 2008 20:32:48 +1000 [thread overview]
Message-ID: <200805162032.48469.rusty@rustcorp.com.au> (raw)
In-Reply-To: <20080515.215823.28841530.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
On Friday 16 May 2008 14:58:23 David Miller wrote:
> From: Rusty Russell <rusty-8n+1lVoiYb80n/F98K4Iww@public•gmane.org>
> Date: Fri, 16 May 2008 12:01:48 +1000
>
> > Dave, please allow me to ask a heretical question. Returning
> > TX_BUSY has some appeal for virtio_net: is it fundamentally a flawed
> > idea, or simply a matter of coding?
>
> Allowing TX_BUSY adds a special case to the caller which we'd
> like to remove at some point.
>
> > Currently we have no virtio interface to ask how many descriptors are
> > left; it's not clear that it's a fair question to ask, since for Xen it's
> > depends on the actual buffers we're trying to put in the descirptors.
>
> Two things:
>
> 1) You can always make sure that you have enough space for a
> TSO frame, with arbitrary page boundaries and thus buffer
> chopping.
>
> It can even be estimated, and if violated by some corner case
> you can punt and drop.
Yes, this is what we'd have to do. Wasting room in the ring feels wrong
though.
> 2) You can queue inside of the driver one packet when you hit
> the limits unexpectedly, netif_stop_queue(), and return
> success. Spit this packet out right before waking the
> queue again.
I put a patch in to do exactly that at Herbert's prompting, for 2.6.26, but
it's buggy in (at least) two ways. I have a fix for this, which adds a new
tasklet to xmit the packet. There's still some subtle race, however, since
I'm still seeing a stuck packet. I'll have to revert to TX_BUSY for 2.6.26
if I can't find it (unlikely).
And I haven't measured what it does to performance (should be OK, but still).
> Really, there are no hard reasons to ever return TX_BUSY,
> it's always a bug.
But it's *simple*, and seems like a common thing to want. Why not change
everything to use TX_BUSY and rip out the guestimate/buffering hacks?
> In fact, I want to move things more and more towards the driver
> queueing TX packets internally instead of the networking mid-layer.
>
> That will ahve benefits for things like TX multiqueue, we won't
> need any locking at all, nor have any knowledge about multiple
> queues at all, if the driver takes care of providing the buffer
> between what the kernel gives it and what the device can handle
> at the moment.
That would be great: then I could shove the packet back on the queue myself
and not have to ask you about it. It's adding a *second* queue inside the
driver which feels terribly ugly...
Cheers,
Rusty.
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public•gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2008-05-16 10:32 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-30 12:40 [RFC/RFT 0/4] mac80211 QoS-related enhancements Johannes Berg
2008-04-30 12:40 ` [RFC/RFT 1/4] mac80211: use rate index in TX control Johannes Berg
2008-04-30 12:40 ` [RFC/RFT 2/4] GSO: generalize for mac80211 Johannes Berg
[not found] ` <20080430130049.359549000-cdvu00un1VgdHxzADdlk8Q@public.gmane.org>
2008-05-06 16:12 ` Johannes Berg
2008-04-30 12:40 ` [RFC/RFT 3/4] mac80211: use GSO for fragmentation Johannes Berg
2008-05-07 7:10 ` Herbert Xu
2008-05-07 8:50 ` Johannes Berg
2008-05-07 9:00 ` Herbert Xu
[not found] ` <20080507090040.GA25186-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
2008-05-07 11:22 ` [PATCH] mac80211: rewrite fragmentation code Johannes Berg
[not found] ` <1210159339.5642.13.camel-YfaajirXv214zXjbi5bjpg@public.gmane.org>
2008-05-07 11:41 ` Herbert Xu
2008-05-07 11:52 ` Johannes Berg
[not found] ` <1210161133.5642.19.camel-YfaajirXv214zXjbi5bjpg@public.gmane.org>
2008-05-07 13:05 ` Herbert Xu
[not found] ` <20080507130548.GA26977-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
2008-05-07 13:48 ` Michael Buesch
2008-05-08 3:22 ` Herbert Xu
[not found] ` <20080508032208.GA401-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
2008-05-08 3:26 ` David Miller
[not found] ` <20080507.202606.242037993.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2008-05-08 9:00 ` Johannes Berg
2008-05-16 2:01 ` Rusty Russell
2008-05-16 3:28 ` Herbert Xu
2008-05-16 4:58 ` David Miller
[not found] ` <20080515.215823.28841530.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2008-05-16 10:32 ` Rusty Russell [this message]
[not found] ` <200805162032.48469.rusty-8n+1lVoiYb80n/F98K4Iww@public.gmane.org>
2008-05-16 10:38 ` Johannes Berg
2008-05-16 12:15 ` Herbert Xu
2008-05-16 19:40 ` David Miller
[not found] ` <20080516.124039.253626477.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2008-05-19 3:08 ` Rusty Russell
2008-05-19 7:03 ` David Miller
2008-05-08 13:00 ` Michael Buesch
[not found] ` <200805081500.00682.mb-fseUSCV1ubazQB+pC5nmwQ@public.gmane.org>
2008-05-08 13:08 ` Herbert Xu
2008-05-08 13:13 ` Michael Buesch
2008-05-08 13:15 ` Michael Buesch
[not found] ` <200805081513.56521.mb-fseUSCV1ubazQB+pC5nmwQ@public.gmane.org>
2008-05-08 13:32 ` Herbert Xu
2008-05-07 19:19 ` Johannes Berg
2008-04-30 12:40 ` [RFC/RFT 4/4] mac80211: use multi-queue master netdevice Johannes Berg
[not found] ` <20080430130051.397094000-cdvu00un1VgdHxzADdlk8Q@public.gmane.org>
2008-04-30 14:37 ` Ivo van Doorn
[not found] ` <200804301637.35170.IvDoorn-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2008-04-30 14:45 ` Johannes Berg
[not found] ` <1209566743.18659.30.camel-YfaajirXv214zXjbi5bjpg@public.gmane.org>
2008-04-30 15:00 ` Johannes Berg
[not found] ` <1209567609.18659.33.camel-YfaajirXv214zXjbi5bjpg@public.gmane.org>
2008-04-30 15:34 ` Ivo van Doorn
2008-04-30 15:38 ` Johannes Berg
2008-05-01 8:21 ` Ivo van Doorn
[not found] ` <200805011021.04435.IvDoorn-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2008-05-01 8:54 ` Johannes Berg
2008-04-30 19:39 ` Waskiewicz Jr, Peter P
[not found] ` <D5C1322C3E673F459512FB59E0DDC32904FE144F-O6kdQIuPh0Q64kNsxIetb7fspsVTdybXVpNB7YpNyf8@public.gmane.org>
2008-04-30 20:07 ` Johannes Berg
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=200805162032.48469.rusty@rustcorp.com.au \
--to=rusty-8n+1lvoiyb80n/f98k4iww@public$(echo .)gmane.org \
--cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public$(echo .)gmane.org \
--cc=herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public$(echo .)gmane.org \
--cc=ivdoorn-Re5JQEeQqe8AvxtiuMwx3w@public$(echo .)gmane.org \
--cc=johannes-cdvu00un1VgdHxzADdlk8Q@public$(echo .)gmane.org \
--cc=linux-wireless-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=mb-fseUSCV1ubazQB+pC5nmwQ@public$(echo .)gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public$(echo .)gmane.org \
--cc=peter.p.waskiewicz.jr-ral2JQCrhuEAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=ron.rindjunsky-ral2JQCrhuEAvxtiuMwx3w@public$(echo .)gmane.org \
--cc=tomasw-Re5JQEeQqe8AvxtiuMwx3w@public$(echo .)gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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