Here is a respin of patch.

I'll send a short patch to make SFQ more fair in presence of large
packets as well.

Thanks

[PATCH v3 net-next-2.6] net_sched: sch_sfq: better struct layouts

This patch shrinks sizeof(struct sfq_sched_data)
from 0x14f8 (or more if spinlocks are bigger) to 0x1180 bytes, and
reduce text size as well.

   text    data     bss     dec     hex filename
   4821     152       0    4973    136d old/net/sched/sch_sfq.o
   4627     136       0    4763    129b new/net/sched/sch_sfq.o

All data for a slot/flow is now grouped in a compact and cache friendly
structure, instead of being spreaded in many different points.

struct sfq_slot {
        struct sk_buff  *skblist_next;
        struct sk_buff  *skblist_prev;
        sfq_index       qlen; /* number of skbs in skblist */
        sfq_index       next; /* next slot in sfq chain */
        struct sfq_head dep; /* anchor in dep[] chains */
        unsigned short  hash; /* hash value (index in ht[]) */
        short           allot; /* credit for this slot */
};
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Cc: Jarek Poplawski <jarkao2@gmail.com>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When deploying SFQ/IFB here at work, I found the allot management was
pretty wrong in sfq, even changing allot from short to int...

We should init allot for each new flow, not using a previous value found
in slot.

Before patch, I saw bursts of several packets per flow, apparently
denying the default "quantum 1514" limit I had on my SFQ class.

class sfq 11:1 parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 7p requeues 0 
 allot 11546 

class sfq 11:46 parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 1p requeues 0 
 allot -23873 

class sfq 11:78 parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 5p requeues 0 
 allot 11393 

After patch, better fairness among each flow, allot limit being
respected, allot is positive :

class sfq 11:e parent 11: 
 (dropped 0, overlimits 0 requeues 86) 
 backlog 0b 3p requeues 86 
 allot 596 

class sfq 11:94 parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 3p requeues 0 
 allot 1468 

class sfq 11:a4 parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 4p requeues 0 
 allot 650 

class sfq 11:bb parent 11: 
 (dropped 0, overlimits 0 requeues 0) 
 backlog 0b 3p requeues 0 
 allot 596 
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We currently return for each active SFQ slot the number of packets in
queue. We can also give number of bytes accounted for these packets.

tc -s class show dev ifb0

Before patch :

class sfq 11:3d9 parent 11:
 (dropped 0, overlimits 0 requeues 0)
 backlog 0b 3p requeues 0
 allot 1266

After patch :

class sfq 11:3e4 parent 11:
 (dropped 0, overlimits 0 requeues 0)
 backlog 4380b 3p requeues 0
 allot 1212
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NChangli Gao <xiaosuo@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

sch_sfq as a classful qdisc needs the .leaf handler. Otherwise, there
is an oops possible in tc_modify_qdisc()/check_loop().

Fixes commit 7d2681a6Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is based on work originally done by Patric McHardy.
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

pkt_sched: sch_sfq: Add dummy unbind_tcf and put handles. Was: [PATCH] sfq: add dummy bind/unbind handles

Add dummy .unbind_tcf and .put qdisc class ops for easier verification.
(All other schedulers have it like this.)
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since there was added ->tcf_chain() method without ->bind_tcf() to
sch_sfq class options, there is oops when a filter is added with
the classid parameter.

Fixes commit 7d2681a6
netdev thread: null pointer at cls_api.c
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Reported-by: NFranchoze Eric <franchoze@yandex.ru>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The packet length should be checked before the packet data is dereferenced.
Signed-off-by: NChangli Gao <xiaosuo@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Sparse can help us find endianness bugs, but we need to make some
cleanups to be able to more easily spot real bugs.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

Some schedulers don't support creating, changing or deleting classes.
Make the respective callbacks optionally and consistently return
-EOPNOTSUPP for unsupported operations, instead of currently either
-EOPNOTSUPP, -ENOSYS or no error.

In case of sch_prio and sch_multiq, the removed operations additionally
checked for an invalid class. This is not necessary since the class
argument can only orginate from ->get() or in case of ->change is 0
for creation of new classes, in which case ->change() incorrectly
returned -ENOENT.

As a side-effect, this patch fixes a possible (root-only) NULL pointer
function call in sch_ingress, which didn't implement a so far mandatory
->delete() operation.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Define three accessors to get/set dst attached to a skb

struct dst_entry *skb_dst(const struct sk_buff *skb)

void skb_dst_set(struct sk_buff *skb, struct dst_entry *dst)

void skb_dst_drop(struct sk_buff *skb)
This one should replace occurrences of :
dst_release(skb->dst)
skb->dst = NULL;

Delete skb->dst field
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: NTheodore Ts'o <tytso@mit.edu>
Acked-by: NMark Fasheh <mfasheh@suse.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Cc: James Morris <jmorris@namei.org>
Acked-by: NCasey Schaufler <casey@schaufler-ca.com>
Acked-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some gcc versions warn that ret may be used uninitialized in
sfq_enqueue(). It's a false positive, so let's annotate this.
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After implementing qdisc->ops->peek() and changing sch_netem into
classless qdisc there are no more qdisc->ops->requeue() users. This
patch removes this method with its wrappers (qdisc_requeue()), and
also unused qdisc->requeue structure. There are a few minor fixes of
warnings (htb_enqueue()) and comments btw.

The idea to kill ->requeue() and a similar patch were first developed
by David S. Miller.
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

From: Patrick McHardy <kaber@trash.net>

Just as a demonstration how easy adding a peek operation to the
work-conserving qdiscs actually is. It doesn't need to keep or change
any internal state in many cases thanks to the guarantee that the
packet will either be dequeued or, if another packet arrives, the
upper qdisc will immediately ->peek again to reevaluate the state.

(This is only slightly modified Patrick's patch.)
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Patrick McHardy <kaber@trash.net> noticed that it would be nice to
handle NET_XMIT_BYPASS by NET_XMIT_SUCCESS with an internal qdisc flag
__NET_XMIT_BYPASS and to remove the mapping from dev_queue_xmit().

David Miller <davem@davemloft.net> spotted a serious bug in the first
version of this patch.
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Patrick McHardy <kaber@trash.net> noticed:
"The other problem that affects all qdiscs supporting actions is
TC_ACT_QUEUED/TC_ACT_STOLEN getting mapped to NET_XMIT_SUCCESS
even though the packet is not queued, corrupting upper qdiscs'
qlen counters."

and later explained:
"The reason why it translates it at all seems to be to not increase
the drops counter. Within a single qdisc this could be avoided by
other means easily, upper qdiscs would still increase the counter
when we return anything besides NET_XMIT_SUCCESS though.

This means we need a new NET_XMIT return value to indicate this to
the upper qdiscs. So I'd suggest to introduce NET_XMIT_STOLEN,
return that to upper qdiscs and translate it to NET_XMIT_SUCCESS
in dev_queue_xmit, similar to NET_XMIT_BYPASS."

David Miller <davem@davemloft.net> noticed:
"Maybe these NET_XMIT_* values being passed around should be a set of
bits. They could be composed of base meanings, combined with specific
attributes.

So you could say "NET_XMIT_DROP | __NET_XMIT_NO_DROP_COUNT"

The attributes get masked out by the top-level ->enqueue() caller,
such that the base meanings are the only thing that make their
way up into the stack. If it's only about communication within the
qdisc tree, let's simply code it that way."

This patch is trying to realize these ideas.
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit f867e6af.

Based upon discussions between Jarek and Patrick McHardy
this is field being set is more a config parameter than a
statistic.  And we should add a true statistic to provide
this information if we really want it.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Dump the "flows" number according to the number of active flows
instead of repeating the "limit".
Reported-by: NDenys Fedoryshchenko <denys@visp.net.lb>
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NJussi Kivilinna <jussi.kivilinna@mbnet.fi>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It can be obtained via the netdev_queue.  So create a helper routine,
qdisc_dev(), to make the transformations nicer looking.

Now, qdisc_alloc() now no longer needs a net_device pointer argument.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass double tcf_proto pointers to tcf_destroy_chain() to make it
clear the start of the filter list for more consistency.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Let's delete timer reliably in sfq_destroy().
Signed-off-by: NJarek Poplawski <jarkao2@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support for dumping statistics and make internal queues visible as
classes.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add support for external classifiers to allow using different flow
hash functions similar to ESFQ. When no classifier is attached the
built-in hash is used as before.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Convert packet schedulers to use the netlink API. Unfortunately a gradual
conversion is not possible without breaking compilation in the middle or
adding lots of casts, so this patch converts them all in one step. The
patch has been mostly generated automatically with some minor edits to
at least allow seperate conversion of classifiers and actions.
Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add whitespace around operators, and add a few blank lines to improve
readability.
Signed-off-by: NStephen Hemminger <stephen.hemminger@vyatta.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

SFQ doesn't need true random numbers, it is only using them to salt a
hash. Therefore it is better to use net_random() and avoid any
possible problems with depleting the entropy pool.
Signed-off-by: NStephen Hemminger <stephen.hemminger@vyatta.com>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The perturbation timer used for re-keying can be deferred, it doesn't
need to be deterministic.
Signed-off-by: NStephen Hemminger <stephen.hemminger@vyatta.com>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Qdisc_class_ops are const, and Qdisc_ops are mostly read.

Using "const" and "__read_mostly" qualifiers helps to reduce false
sharing.
Signed-off-by: NEric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Many-many code in the kernel initialized the timer->function
and  timer->data together with calling init_timer(timer). There
is already a helper for this. Use it for networking code.

The patch is HUGE, but makes the code 130 lines shorter
(98 insertions(+), 228 deletions(-)).
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Acked-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is followup to Patrick's patch. A little optimization to enqueue
routine allows to remove artificial limitation on queue length.

Plus, testing showed that hash function used by SFQ is too bad or even worse.
It does not even sweep the whole range of hash values.
Switched to Jenkins' hash.
Signed-off-by: NAlexey Kuznetsov <kaber@ms2.inr.ac.ru>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Acked-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NPatrick McHardy <kaber@trash.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

So that it is also an offset from skb->head, reduces its size from 8 to 4 bytes
on 64bit architectures, allowing us to combine the 4 bytes hole left by the
layer headers conversion, reducing struct sk_buff size to 256 bytes, i.e. 4
64byte cachelines, and since the sk_buff slab cache is SLAB_HWCACHE_ALIGN...
:-)

Many calculations that previously required that skb->{transport,network,
mac}_header be first converted to a pointer now can be done directly, being
meaningful as offsets or pointers.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now the skb->nh union has just one member, .raw, i.e. it is just like the
skb->mac union, strange, no? I'm just leaving it like that till the transport
layer is done with, when we'll rename skb->mac.raw to skb->mac_header (or
->mac_header_offset?), ditto for ->{h,nh}.
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>