Davide Caratti writes:
> hi Toke,
>
> thanks for answering.
>
> On Fri, 2020-06-26 at 14:52 +0200, Toke Høiland-Jørgensen wrote:
>> Davide Caratti writes:
>
> [...]
>
>> >
>> > > I guess I can trying going through them all and fig
Jose Blanquicet writes:
> Hi everyone,
>
> We have an embedded system with limited CPU resources that acts as a
> gateway to provide Internet access from LTE to a private USB-NCM
> network (And also to a Wi-Fi private network but we will work on it
> later). Our problem is that the bandwidth on
):
sch_cake: fix IP protocol handling in the presence of VLAN tags
sch_cake: don't try to reallocate or unshare skb unconditionally
Kevin Darbyshire-Bryant (1):
sch_cake: add RFC 8622 LE PHB support to CAKE diffserv handling
Toke Høiland-Jørgensen (2):
sch_cake: don't call
efore
changing DSCP bits")
Signed-off-by: Ilya Ponetayev
[ fix a few style issues, reflow commit message ]
Signed-off-by: Toke Høiland-Jørgensen
---
net/sched/sch_cake.c | 41 ++---
1 file changed, 30 insertions(+), 11 deletions(-)
diff --git a/net/sched/sc
From: Toke Høiland-Jørgensen
As a further optimisation of the diffserv parsing codepath, we can skip it
entirely if CAKE is neither configured to use diffserv-based
classification, nor to zero out the diffserv bits.
Fixes: c87b4ecdbe8d ("sch_cake: Make sure we can write the IP header b
mit message ]
Signed-off-by: Toke Høiland-Jørgensen
---
net/sched/sch_cake.c | 52 +++---
1 file changed, 49 insertions(+), 3 deletions(-)
diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index 60f8ae578819..0f594d88a957 100644
--- a/net
orities from the default order as
found in diffserv8, in case anyone is wondering why it looks a bit odd.
Signed-off-by: Kevin Darbyshire-Bryant
[ reword commit message slightly ]
Signed-off-by: Toke Høiland-Jørgensen
---
net/sched/sch_cake.c |8
1 file changed, 4 insertions(+),
From: Toke Høiland-Jørgensen
I spotted a few nits when comparing the in-tree version of sch_cake with
the out-of-tree one: A redundant error variable declaration shadowing an
outer declaration, and an indentation alignment issue. Fix both of these to
minimise the delta.
Signed-off-by: Toke
Davide Caratti writes:
> hello Toke,
>
> thanks for answering!
>
> On Fri, 2020-07-03 at 14:05 +0200, Toke Høiland-Jørgensen wrote:
>> while (proto == htons(ETH_P_8021Q) || proto == htons(ETH_P_8021AD)) {
>
> maybe this line be shortened, since if_vlan.h has [1]:
>
Daniel Borkmann writes:
> On 7/6/20 2:29 PM, Toke Høiland-Jørgensen wrote:
>> Toshiaki pointed out that we now have two very similar functions to extract
>> the L3 protocol number in the presence of VLAN tags. And Daniel pointed out
>> that the unbounded parsing l
Pete Heist writes:
> I'm using Cake on an EdgeRouter-X at a site that does around 50-
> 100GB/day, and we're quite happy with it so far. Egress and ingress
> stats are below after ~1TB download and ~100GB upload.
>
> I use "memlimit 8M", and noticed that it's reporting 8389696 bytes used
> on
sed
hashing mode. Add a check for this before using the skb->hash value as
flow_hash.
Fixes: b0c19ed6088a ("sch_cake: Take advantage of skb->hash where appropriate")
Reported-by: Pete Heist
Tested-by: Pete Heist
Signed-off-by: Toke Høiland-Jørgensen
---
net/sched/sch_cake.c | 2 +-
Christophe Leroy writes:
> Le 28/03/2022 à 17:43, Toke Høiland-Jørgensen a écrit :
>> Christophe Leroy writes:
>>
>>> French Administration has written a guideline that defines additional
>>> DSCP values for use in its networks.
>>
>> Huh, that
Matt Johnston writes:
> The CS1 priority (index 0x08) was changed from 0 to 1 when LE (index
> 0x01) was added. This looks unintentional, it doesn't match the
> docs and CS1 shouldn't be the same tin as AF1x
Hmm, Kevin, any comments?
-Toke
___
Cake
--- Begin Message ---
Hey everyone
The dual-mode fairness patch has been in the github repo for a few weeks
now, and no one has complained. If no one continues to complain, I'll
submit it upstream tomorrow along with Kevin's fwmark patch.
So if anyone else wants to test, now would be a good time
--- Begin Message ---
Pete Heist writes:
> Thanks for the nudge... I just tested this on the original one-armed
> router setup I encountered it on (3.16.7, routing from eth0 to
> eth0.3300). The patch makes a big improvement over the original (which
> was around 10/80):
>
> IP 1, 1 up: 47.0 Mbit
--- Begin Message ---
Hi everyone
The fwmark tin selection override feature has been accepted into
upstream and will be part of Linux 5.1. I just pushed the associated
userspace changed to the tc-adv repo, so it should now be possible to
test the feature "out of the box", so to speak.
So please
--- Begin Message ---
Dave Taht writes:
> the SCE draft is now submitted to IETF! I'm too tired to figure out
> what timezone conversion the UTC deadline is in PDT right now, and I
> guess the chairs need to move it into the tsvwg working group instead
> of individual submissions... and I have
--- Begin Message ---
Alban writes:
> Hi everybody,
>
> I am setting a new router with a non trivial setup and I really like to
> get some recommendations on how to best use cake. First of all the
> router is using VLAN on top of 2 bonded gigabit Ethernet interface:
>
>
--- Begin Message ---
Dave Taht writes:
> I am happy to see quic in last call. there are a ton of interoperble
> implementations now.
And in related news, this RFC of pacing from userspace was posted on
netdev yesterday:
Luca Muscariello writes:
> Is there a reason why in cake the quantum cannot be configured to a
> different value like in fq_codel?
I think this was mostly to be as no-knob as possible; so the quantum is
auto-scaled with the tin bandwidths, instead of being configurable.
Jonathan can probably
Jim Geo writes:
>>
>> > On 28 Jul, 2020, at 12:41 am, Jim Geo wrote:
>> >
>> > Thank you for all the efforts you have done to make internet usable.
>> >
>> > I currently use htb & fq_codel in my low speed ADSL 6Mbps downlink/1 Mbps
>> > uplink. I use fwmark to control both uplink and downlink
Dean Scarff writes:
> On Mon, 02 Nov 2020 13:37:00 +0100, Toke wrote:
>> Dean Scarff writes:
>>
>>> Hi,
>>>
>>> I've been happily running the out-of-tree sch_cake on my Raspberry
>>> Pi
>>> since 2015. However, I recently upgraded my kernel (to 5.4.72 from
>>> Raspbian's
Dean Scarff writes:
> Hi,
>
> I've been happily running the out-of-tree sch_cake on my Raspberry Pi
> since 2015. However, I recently upgraded my kernel (to 5.4.72 from
> Raspbian's raspberrypi-kernel 1.20201022-1), which comes with the
> sch_cake in mainline. Now, when running:
>
>
Dean Scarff writes:
> On Tue, 03 Nov 2020 12:00:55 +0100, Toke Høiland-Jørgensen wrote:
>> Dean Scarff writes:
>>
>>> On Mon, 02 Nov 2020 13:37:00 +0100, Toke wrote:
Dean Scarff writes:
> Hi,
>
> I've been happily running the out-of-tree sch_cake on my
>
Sebastian Moeller writes:
>> On Jun 24, 2021, at 15:59, Toke Høiland-Jørgensen wrote:
>>
>> Pete Heist writes:
>>
>>> I'm using Cake on an EdgeRouter-X at a site that does around 50-
>>> 100GB/day, and we're quite happy with it so far. Egress
Dave Taht writes:
> Currently centos (and I assume redhat) is at 4.18. Cake went into 4.19
> so I assume the next major
> redhat/centos releases will have it.
Note that the RHEL kernel version number is basically a complete
fabrication; it's the version that the kernel was forked from,
Pete Heist writes:
> I've always wanted a way to customize Cake's host and flow isolation in
> a way that would be usable e.g. for small ISPs, and this is what I came
> up with:
>
> https://github.com/heistp/cake-custom-isolation
>
> ipsets are used to set the skb priority or mark, then tc-flow
Pete Heist writes:
> On Sun, 2021-06-06 at 21:59 +0200, Toke Høiland-Jørgensen wrote:
>> Pete Heist writes:
>>
>> > I've always wanted a way to customize Cake's host and flow isolation
>> > in
>> > a way that would be usable e.g. for small
Dave Taht writes:
> One thing somewhat related to this was finally expanding the space
> available for the tc and iptables functionality for
> things like hashing and actions etc from 16 bits to 32. That is
> something of a fork lift upgrade, but... 64k queues is not
> enough in some cases, nor
"Holland, Jake via Bloat" writes:
> Hi David,
>
> That’s an interesting point, and I think you’re right that packet
> arrival is poorly modeled as a Poisson process, because in practice
> packet transmissions are very rarely unrelated to other packet
> transmissions.
>
> But now you’ve got me
Dave Taht writes:
> wow, that is (predictably) miserable, even with cake. The only
> solution that is going to
> work is to somehow actively monitor your link quality and adjust cake
> to suit. Or we can start trying to use kathie's passive ping tools.
We have a PhD student working on a
Taraldsen Erik writes:
> Disclamer: I'm working on the Fixed Wireless products for Telenor
> (Zyxel NR7101 outdoor wall mounted unit). Not the Mobile Broadband
> products. We are working with Zyxel and Qualcom to try and implement
> an upstream queue which adapts to available radio resources. To
Nils Andreas Svee writes:
> On 2/25/21 11:30 AM, Toke Høiland-Jørgensen wrote:
>
>> Ah, wireguard doesn't have XDP support, so that's likely not going to
>> work; and if you run it on the physical interface, even if you didn't
>> get driver errors, the tool would just s
Taraldsen Erik writes:
> This is getting LTE/5G spesific. Not sure if it belongs on the list.
> Let us know if we are generating noise.
I for one think it's fascinating - carry on! :)
-Toke
___
Cake mailing list
Cake@lists.bufferbloat.net
Dave Taht writes:
> https://github.com/dtaht/tc-adv/issues/27
>
> I don't have any of the hardware supported by this firmware lying
> around handy... can anyone test?
I believe the original reported already tested and confirmed the fix,
actually. I have a patch pending for upstream iproute2
On 24 February 2021 23:49:48 CET, Nils Andreas Svee wrote:
>I'll look into pping. Admittedly I'm quite ignorant about BPF, so I'll
>likely blunder about for a bit, but hey, got it to compile - *and* run,
>but I didn't get any output other than the messages from clean_map.
>Dunno if I did
"Nils Andreas Svee" writes:
> I ran it on my router though, which has a decent amount of TCP flows running
> at any given time.
> It's all going over a wg tunnel though, that might be wonky for all I
> know.
Ah, wireguard doesn't have XDP support, so that's likely not going to
work; and if you
John Sager writes:
> You will need to specify the hosts explicitly, unless you can live with them
> all sharing one bandwidth class. In that case if you have more than one
> using bandwidth they would share the bandwidth in that class equally. I
> assume from your original post that you want
John Sager writes:
> PS does the cake list allow attachments? It will be a small zip file.
It'll likely end up in the moderation queue (don't recall the size
limit), but I can manually approve it from there...
-Toke
___
Cake mailing list
Peter Lepeska writes:
> Not sure this is the right forum but I am using Cake on my OpenWRT router
> so I thought I would cross-post this question (
> https://forum.openwrt.org/t/use-sqm-to-configure-video-quality/88984) here.
CAKE doesn't have any facilities to do per-device shaping like you're
Peter Lepeska writes:
> A user on the OpenWrt forum suggested hashlimit rules supported by
> iptables. How does that idea sound to you?
That will result in a cliff-edge policer (i.e., as soon as a device goes
over its limits it will see every packet get dropped). This doesn't
interact too well
Pete Heist writes:
> As for triple-isolate, I added a test Jon suggested:
>
> "I would recommend a W configuration of flows
>
> one of the three hosts sends to or from both of the two hosts, while
> the other two send to or from only one each
>
> this should result in the two hosts getting equal
Pete Heist writes:
> One more tip, reverting this commit seems to fix it:
>
> https://github.com/torvalds/linux/commit/b0c19ed6088ab41dd2a727b60594b7297c15d6ce
Ah, I think I see what the problem is; could you please try the patch
below?
-Toke
diff --git a/net/sched/sch_cake.c
Jonathan Morton writes:
> On Wed, 4 Aug 2021 at 14:14, Toke Høiland-Jørgensen via Cake
> wrote:
>>
>> Pete Heist writes:
>>
>> > One more tip, reverting this commit seems to fix it:
>> >
>> > https://github.com/torvalds/linux/commit/b0c19ed60
Dave Taht writes:
> my more specific question was gro. On gro assembly is the dscp/ecn
> header examined?
Yes, and only packets with the same value get aggregated:
https://elixir.bootlin.com/linux/latest/source/net/ipv6/ip6_offload.c#L263
-Toke
___
Dave Taht writes:
> weirdly enough, my gmail account has not received anything from netdev
> since oct 11.
You're not alone in that:
https://lore.kernel.org/netdev/20211014112718.6aed7...@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com/T/#t
> yes, i think fq_codel will be better, and even the
Dave Taht writes:
> On Thu, Oct 14, 2021 at 2:44 PM Toke Høiland-Jørgensen wrote:
>>
>> Dave Taht writes:
>>
>> > weirdly enough, my gmail account has not received anything from netdev
>> > since oct 11.
>>
>> You're not alone in that:
>
Dave Taht writes:
> I have sometimes thought that allowing the deficit to go negative
> (either one quantum or one mtu) might
> result in slightly better service times, or tcp, or aqm behavior,
> given tcp's requirement for two back to back packets
> to release an ack.
Why? Isn't that just
Dave Taht writes:
> An open question for me:
>
> What happens when a GSO packet is marked? Do all the packets get the
> marking, or just the first?
When segmenting a GSO packet, the IP header is copied to all segments.
The code has grown quite complex over the years, but it's easier to see
in
Sebastian Moeller writes:
> Mmmh, I guess our approach at traffic shaping does not scale well at those
> speeds. Maybe this could be fixed with larger batching?
>
> I think it might be worth trying to switch to simple.qos/fq_codel and
> set a somewhat larger burst/quantum duration in
>> And either way, CAKE is still going to be limited by being single core
>> only, and fixing that requires some serious surgery that I seem to
>> recall looking into and giving up at some point :(
>
> That is sad, and pretty much rules out that I could make some
> progress in that
Sebastian Moeller writes:
> Hi Dave,
>
> On 23 November 2021 08:32:06 CET, Dave Taht wrote:
>>The context of my question is basically this:
>>
>>Is cake baked? Is it done?
>
> How about per MAC address fairness (useful for ISPs and to treat
> IPv4/6 equally)?
>
> How about configurable number
Dave Taht writes:
> On Tue, Nov 23, 2021 at 2:39 AM Toke Høiland-Jørgensen wrote:
>>
>> Sebastian Moeller writes:
>>
>> > Hi Dave,
>> >
>> > On 23 November 2021 08:32:06 CET, Dave Taht wrote:
>> >>The context of my ques
Dave Taht writes:
> https://github.com/rchac/LibreQoS
Cool! Would be an excellent use case for the BPF-based pping (to avoid
the overhead of libpcap)...
-Toke
___
Cake mailing list
Cake@lists.bufferbloat.net
Jim Geo writes:
> Hello,
>
> Was it the function of ingress argument in cake?
> Checking online man pages, I couldn't find an explanation.
> Does it make sense to use it on an ifb device? (sending ingress
> traffic through it)
Yes, that's exactly what it's meant for. The option changes cake's
Christophe Leroy writes:
> French Administration has written a guideline that defines additional
> DSCP values for use in its networks.
Huh, that's interesting!
> Add new CAKE diffserv tables to take those new values into account
> and add CONFIG_NET_SCH_CAKE_DGSIC to select those tables
eter is also checked before
> calling change function in .init(). Therefore, it's no need to check the
> input parameter opt in change function.
>
> Signed-off-by: Zhengchao Shao
for sch_cake:
Acked-by: Toke Høiland-Jørgensen
__
Jakub Kicinski writes:
> On Thu, 01 Sep 2022 00:13:24 +0200 Toke Høiland-Jørgensen wrote:
>> Ah, crossed streams (just sent v2[0]).
>
> Sorry about that, traveling knocked out my sense of time and I kept
> thinking it's Thursday, and the discussion happened yesterday :S
Ha
the original skb still exists, when it really has been freed. Fix
this by adding the __NET_XMIT_STOLEN flag to the return value in this case.
Fixes: 0c850344d388 ("sch_cake: Conditionally split GSO segments")
Signed-off-by: Toke Høiland-Jørgensen
---
net/sched/sch_cake.c | 4 +++-
1 fi
Eric Dumazet writes:
> On Wed, Aug 31, 2022 at 2:25 AM Toke Høiland-Jørgensen wrote:
>>
>> When the GSO splitting feature of sch_cake is enabled, GSO superpackets
>> will be broken up and the resulting segments enqueued in place of the
>> original skb. In this c
patchwork-bot+netdev...@kernel.org writes:
> Hello:
>
> This patch was applied to netdev/net.git (master)
> by Jakub Kicinski :
>
> On Wed, 31 Aug 2022 11:21:03 +0200 you wrote:
>> When the GSO splitting feature of sch_cake is enabled, GSO superpackets
>> will be broken up and the resulting
I figured y'all would find this an uplifting story - I certainly did!
https://forum.openwrt.org/t/nftables-and-qos-in-2021/112013/530
-Toke
___
Cake mailing list
Cake@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
_syscall_64+0x35/0x80
> entry_SYSCALL_64_after_hwframe+0x46/0xb0
> RIP: 0033:0x7f89e5122d04
>
>
> Fixes: 046f6fd5daef ("sched: Add Common Applications Kept Enhanced (cake)
> qdisc")
> Signed-off-by: Zhengchao Shao
Acked-by: Toke Høiland-Jørgensen
___
Cake mailing list
Cake@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
Zhengchao Shao writes:
> use tc_qdisc_stats_dump() in qdisc.
>
> Signed-off-by: Zhengchao Shao
> Reviewed-by: Victor Nogueira
> Tested-by: Victor Nogueira
For sch_cake:
Acked-by: Toke Høiland-Jørgensen
___
Cake ma
t. So we can reconstruct
> them.
>
> Signed-off-by: Zhengchao Shao
Acked-by: Toke Høiland-Jørgensen
___
Cake mailing list
Cake@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
Zhengchao Shao writes:
> The walk implementation of most qdisc class modules is basically the
> same. That is, the values of count and skip are checked first. If
> count is greater than or equal to skip, the registered fn function is
> executed. Otherwise, increase the value of count. So we can
Jonathan Morton via Cake writes:
>> On 23 Feb, 2023, at 3:35 am, Dave Taht via Cake
>> wrote:
>>
>> does this break cake?
>>
>> https://lore.kernel.org/netdev/de811bf3-e2d8-f727-72bc-c8a754a9d...@tessares.net/T/
>
> It looks like they've included patches *to* Cake to handle anticipated
>
Dave Taht via Cake writes:
> In looking how the code has morphed since I last looked at it, I found
> myself staring at this bit...
>
> skb_flow_dissect_flow_keys(skb, ,
>FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL);
>
> // so we have delved deeply into the
Dave Taht via Cake writes:
> I have been meaning to make this patch use UINT_MAX and submit to the
> kernel for a long time. Anyone want to give it a spin?
There's tooling to ease submission these days:
https://people.kernel.org/monsieuricon/sending-a-kernel-patch-with-b4-part-1
-Toke
"David P. Reed via Bloat" writes:
> (One such nightmare can be seen in LKML... Search for
> dpr...@deepplum.com patch emails. I tried hard, was worn down, then
> gave up, since I found a way to avoid the bug, in virtualization code
> on x86, and gave up on getting it fixed after a year. Life is
Dave Taht via Cake writes:
> It has been years since I looked at cake's code.
>
> Does anyone remember why we do not ack-filter a gso-split?
Because a GSO packet cannot be a pure ACK, so it wouldn't be filtered
anyway...
-Toke
___
Cake mailing list
Michal Koutný writes:
> On Wed, Dec 06, 2023 at 02:28:57PM -0800, Stephen Hemminger
> wrote:
>> It is not clear to me what this patchset is trying to fix.
>> Autoloading happens now, but it does depend on the name not alias.
>
> There are some more details in the thread of v1 [1] [2].
> Does
Dave Taht writes:
> On Sun, Feb 18, 2024 at 8:37 AM Toke Høiland-Jørgensen wrote:
>>
>> Dave Taht via Cake writes:
>>
>> > It has been years since I looked at cake's code.
>> >
>> > Does anyone remember why we do not ack-filter a gso-split?
>
Kuan-Wei Chiu writes:
> Improve the max-heap construction process by reducing unnecessary
> heapify operations. Specifically, adjust the starting condition from
> n / 2 to n / 2 - 1 in the loop that iterates over all non-leaf
> elements.
Please add an explanation for why this change is correct,
if conditions).
>
> Signed-off-by: Kuan-Wei Chiu
Acked-by: Toke Høiland-Jørgensen
___
Cake mailing list
Cake@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
Kuan-Wei Chiu writes:
> On Sun, Apr 07, 2024 at 06:10:04PM +0200, Toke Høiland-Jørgensen wrote:
>> Kuan-Wei Chiu writes:
>>
>> > Improve the max-heap construction process by reducing unnecessary
>> > heapify operations. Specifically, adjust the starting condi
Simon Horman writes:
> + Toke Høiland-Jørgensen
> cake@lists.bufferbloat.net
Thanks!
> On Mon, Apr 15, 2024 at 01:20:42PM +, Eric Dumazet wrote:
>> Instead of relying on RTNL, cake_dump() can use READ_ONCE()
>> annotations, paired with WRITE_ONCE() ones in cake_ch
501 - 578 of 578 matches
Mail list logo