Large tc dumps (tc -s {qdisc|class} sh dev ethX) done by Google BwE host
agent [1] are problematic at scale :
For each qdisc/class found in the dump, we currently lock the root qdisc
spinlock in order to get stats. Sampling stats every 5 seconds from
thousands of HTB classes is a challenge when the root qdisc spinlock is
under high pressure. Not only the dumps take time, they also slow
down the fast path (queue/dequeue packets) by 10 % to 20 % in some cases.
An audit of existing qdiscs showed that sch_fq_codel is the only qdisc
that might need the qdisc lock in fq_codel_dump_stats() and
fq_codel_dump_class_stats()
In v2 of this patch, I now use the Qdisc running seqcount to provide
consistent reads of packets/bytes counters, regardless of 32/64 bit arches.
I also changed rate estimators to use the same infrastructure
so that they no longer need to lock root qdisc lock.
[1]
http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43838.pdf
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: Kevin Athey <kda@google.com>
Cc: Xiaotian Pei <xiaotian@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 22e0f8b932 ("net: sched: make bstats per cpu and estimator RCU safe")
added the argument cpu_bstats to functions gen_new_estimator and
gen_replace_estimator and now the descriptions of these are missing for the
documentation. Adding them.
Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rate estimators are limited to 4 Mpps, which was fine years ago, but
too small with current hardware generation.
Lets use 2^5 scaling instead of 2^10 to get 128 Mpps new limit.
On 64bit arch, use an "unsigned long" for temp storage and remove limit.
(We do not expect 32bit arches to be able to reach this point)
Tested:
tc -s -d filter sh dev eth0 parent ffff:
filter protocol ip pref 1 u32
filter protocol ip pref 1 u32 fh 800: ht divisor 1
filter protocol ip pref 1 u32 fh 800::800 order 2048 key ht 800 bkt 0 flowid 1:15
match 07000000/ff000000 at 12
action order 1: gact action drop
random type none pass val 0
index 1 ref 1 bind 1 installed 166 sec
Action statistics:
Sent 39734251496 bytes 863788076 pkt (dropped 863788117, overlimits 0 requeues 0)
rate 4067Mbit 11053596pps backlog 0b 0p requeues 0
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to run qdisc's without locking statistics and estimators
need to be handled correctly.
To resolve bstats make the statistics per cpu. And because this is
only needed for qdiscs that are running without locks which is not
the case for most qdiscs in the near future only create percpu
stats when qdiscs set the TCQ_F_CPUSTATS flag.
Next because estimators use the bstats to calculate packets per
second and bytes per second the estimator code paths are updated
to use the per cpu statistics.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fix spelling typo found in DocBook/networking.xml.
It is because the neworking.xml is generated from comments
in the source, I have to fix typo in comments within the source.
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct gnet_stats_rate_est contains u32 fields, so the bytes per second
field can wrap at 34360Mbit.
Add a new gnet_stats_rate_est64 structure to get 64bit bps/pps fields,
and switch the kernel to use this structure natively.
This structure is dumped to user space as a new attribute :
TCA_STATS_RATE_EST64
Old tc command will now display the capped bps (to 34360Mbit), instead
of wrapped values, and updated tc command will display correct
information.
Old tc command output, after patch :
eric:~# tc -s -d qd sh dev lo
qdisc pfifo 8001: root refcnt 2 limit 1000p
Sent 80868245400 bytes 1978837 pkt (dropped 0, overlimits 0 requeues 0)
rate 34360Mbit 189696pps backlog 0b 0p requeues 0
This patch carefully reorganizes "struct Qdisc" layout to get optimal
performance on SMP.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove all #inclusions of asm/system.h preparatory to splitting and killing
it. Performed with the following command:
perl -p -i -e 's!^#\s*include\s*<asm/system[.]h>.*\n!!' `grep -Irl '^#\s*include\s*<asm/system[.]h>' *`
Signed-off-by: David Howells <dhowells@redhat.com>
The rcu callback __gen_kill_estimator() just calls a kfree(),
so we use kfree_rcu() instead of the call_rcu(__gen_kill_estimator).
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Now that est_tree_lock is acquired with BH protection, the other
call is unnecessary.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes a lockdep warning:
[ 516.287584] =========================================================
[ 516.288386] [ INFO: possible irq lock inversion dependency detected ]
[ 516.288386] 2.6.35b #7
[ 516.288386] ---------------------------------------------------------
[ 516.288386] swapper/0 just changed the state of lock:
[ 516.288386] (&qdisc_tx_lock){+.-...}, at: [<c12eacda>] est_timer+0x62/0x1b4
[ 516.288386] but this lock took another, SOFTIRQ-unsafe lock in the past:
[ 516.288386] (est_tree_lock){+.+...}
[ 516.288386]
[ 516.288386] and interrupts could create inverse lock ordering between them.
...
So, est_tree_lock needs BH protection because it's taken by
qdisc_tx_lock, which is used both in BH and process contexts.
(Full warning with this patch at netdev, 02 Sep 2010.)
Fixes commit: ae638c47dc
("pkt_sched: gen_estimator: add a new lock")
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_kill_estimator() API is incomplete or not well documented, since
caller should make sure an RCU grace period is respected before
freeing stats_lock.
This was partially addressed in commit 5d944c640b
(gen_estimator: deadlock fix), but same problem exist for all
gen_kill_estimator() users, if lock they use is not already RCU
protected.
A code review shows xt_RATEEST.c, act_api.c, act_police.c have this
problem. Other are ok because they use qdisc lock, already RCU
protected.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_kill_estimator() / gen_new_estimator() is not always called with
RTNL held.
net/netfilter/xt_RATEEST.c is one user of these API that do not hold
RTNL, so random corruptions can occur between "tc" and "iptables".
Add a new fine grained lock instead of trying to use RTNL in netfilter.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
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: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
In 5e140dfc1f "net: reorder struct Qdisc
for better SMP performance" the definition of struct gnet_stats_basic
changed incompatibly, as copies of this struct are shipped to
userland via netlink.
Restoring old behavior is not welcome, for performance reason.
Fix is to use a private structure for kernel, and
teach gnet_stats_copy_basic() to convert from kernel to user land,
using legacy structure (struct gnet_stats_basic)
Based on a report and initial patch from Michael Spang.
Reported-by: Michael Spang <mspang@csclub.uwaterloo.ca>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Right-shifts of signed integers are implementation-defined so unportable.
With feedback from: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_estimator can overflow bps (bytes per second) with Gb links, while
it was designed with a u32 API, with a theorical limit of 34360Mbit
(2^32 bytes)
Using 64 bit intermediate avbps/brate counters can allow us to reach
this theorical limit.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since all other gen_estimator functions use bstats and rate_est params
together, and searching for them is optimized now, let's use this also
in gen_estimator_active(). The return type of gen_estimator_active()
is changed to bool, and gen_find_node() parameters to const, btw.
In tcf_act_police_locate() a check for ACT_P_CREATED is added before
calling gen_estimator_active().
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Found that while trying average rate policing, it was possible to
request average rate policing without a rate estimator. This results
in no policing which is harmless but incorrect.
Since policing could be setup in two steps, need to check
in the kernel.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_kill_estimator() linear lists lookups are very slow, and e.g. while
deleting a large number of HTB classes soft lockups were reported. Here
is another try to fix this problem: this time internally, with rbtree,
so similarly to Jamal's hashing idea IIRC. (Looking for next hits could
be still optimized, but it's really fast as it is.)
Reported-by: Badalian Vyacheslav <slavon@bigtelecom.ru>
Reported-by: Denys Fedoryshchenko <denys@visp.net.lb>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit d4766692e7.
qdisc_destroy() now runs in RTNL fully again, so this
change is no longer needed.
Signed-off-by: David S. Miller <davem@davemloft.net>
gen_kill_estimator() required rtnl_lock() protection, but since it is
moved to an RCU callback __qdisc_destroy() let's use est_lock instead.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David 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: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
White spaces etc. are changed in gen_replace_estimator() to make it
similar to others in a file.
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can void divides (as seen with CONFIG_CC_OPTIMIZE_FOR_SIZE=y on x86)
changing ((HZ<<idx)/4) to ((HZ/4) << idx)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
remove asm/bitops.h includes
including asm/bitops directly may cause compile errors. don't include it
and include linux/bitops instead. next patch will deny including asm header
directly.
Cc: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-Fixes ABBA deadlock noted by Patrick McHardy <kaber@trash.net>:
> There is at least one ABBA deadlock, est_timer() does:
> read_lock(&est_lock)
> spin_lock(e->stats_lock) (which is dev->queue_lock)
>
> and qdisc_destroy calls htb_destroy under dev->queue_lock, which
> calls htb_destroy_class, then gen_kill_estimator and this
> write_locks est_lock.
To fix the ABBA deadlock the rate estimators are now kept on an rcu list.
-The est_lock changes the use from protecting the list to protecting
the update to the 'bstat' pointer in order to avoid NULL dereferencing.
-The 'interval' member of the gen_estimator structure removed as it is
not needed.
Signed-off-by: Ranko Zivojnovic <ranko@spidernet.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noticed by Jarek Poplawski <jarkao2@o2.pl>, the timer removal in
gen_kill_estimator races with the timer function rearming the timer.
Check whether the timer list is empty before rearming the timer
in the timer function to fix this.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Acked-by: Jarek Poplawski <jarkao2@o2.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Eric Dumazet