This fixes a performance regression introduced by
7fbb9d8415 (xen-netback: release pending
index before pushing Tx responses)
Moving the notify outside of the spin locks means it can be delayed a
long time (if the dealloc thread is descheduled or there is an
interrupt or softirq).
Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Zoltan Kiss <zoltan.kiss@linaro.org>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 814d488c61 ("tcp: fix the timid additive increase on stretch
ACKs") fixed a bug where tcp_cong_avoid_ai() would either credit a
connection with an increase of snd_cwnd_cnt, or increase snd_cwnd, but
not both, resulting in cwnd increasing by 1 packet on at most every
alternate invocation of tcp_cong_avoid_ai().
Although the commit correctly implemented the CUBIC algorithm, which
can increase cwnd by as much as 1 packet per 1 packet ACKed (2x per
RTT), in practice that could be too aggressive: in tests on network
paths with small buffers, YouTube server retransmission rates nearly
doubled.
This commit restores CUBIC to a maximum cwnd growth rate of 1 packet
per 2 packets ACKed (1.5x per RTT). In YouTube tests this restored
retransmit rates to low levels.
Testing: This patch has been tested in datacenter netperf transfers
and live youtube.com and google.com servers.
Fixes: 9cd981dcf1 ("tcp: fix stretch ACK bugs in CUBIC")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The recent change to tcp_cong_avoid_ai() to handle stretch ACKs
introduced a bug where snd_cwnd_cnt could accumulate a very large
value while w was large, and then if w was reduced snd_cwnd could be
incremented by a large delta, leading to a large burst and high packet
loss. This was tickled when CUBIC's bictcp_update() sets "ca->cnt =
100 * cwnd".
This bug crept in while preparing the upstream version of
814d488c61.
Testing: This patch has been tested in datacenter netperf transfers
and live youtube.com and google.com servers.
Fixes: 814d488c61 ("tcp: fix the timid additive increase on stretch ACKs")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If rtnl_newlink() fails on it's call to dev_change_net_namespace(), we
have to make use of the ->dellink() method, if present, just like we
do when rtnl_configure_link() fails.
Fixes: 317f4810e4 ("rtnl: allow to create device with IFLA_LINK_NETNSID set")
Signed-off-by: David S. Miller <davem@davemloft.net>
iwlwifi:
* fix ROC removal - avoids a firmware crash
* fix throughput regression on iwldvm devices
* fix panic in BT Coex
* fixes in rate control
* fixes in scan
b43:
* fix support for 5 GHz only BCM43228 model
rtlwifi:
* improve handling of IPv6 packets
brcmfmac:
* perform bound checking on vendor command buffer
Signed-off-by: David S. Miller <davem@davemloft.net>
The current driver support receive VLAN CTAG HW acceleration feature
(NETIF_F_HW_VLAN_CTAG_RX) through software simulation. There calls the
api .skb_copy_to_linear_data_offset() to skip the VLAN tag, but there
have overlap between the two memory data point range. The patch just fix
the issue.
V2:
Michael Grzeschik suggest to use memmove() instead of skb_copy_to_linear_data_offset().
Reported-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
Fixes: 1b7bde6d65 ("net: fec: implement rx_copybreak to improve rx performance")
Signed-off-by: Fugang Duan <B38611@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The interrupt is enabled before napi_complete(). A network timeout
occurs if the interrupt handler is called before napi_complete().
Fix the bug by enabling the interrupt after napi_complete().
Signed-off-by: Yongbae Park <yongbae2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The interrupt is enabled before napi_complete(). A network timeout
occurs if the interrupt handler is called before napi_complete().
Fix the bug by enabling the interrupt after napi_complete().
Signed-off-by: Yongbae Park <yongbae2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We dynamically allocate divisor+1 entries for ->ht[] in tc_u_hnode:
ht = kzalloc(sizeof(*ht) + divisor*sizeof(void *), GFP_KERNEL);
So ->ht is supposed to be the last field of this struct, however
this is broken, since an rcu head is appended after it.
Fixes: 1ce87720d4 ("net: sched: make cls_u32 lockless")
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull kvm/s390 bugfixes from Marcelo Tosatti.
* git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: s390: non-LPAR case obsolete during facilities mask init
KVM: s390: include guest facilities in kvm facility test
KVM: s390: fix in memory copy of facility lists
KVM: s390/cpacf: Fix kernel bug under z/VM
KVM: s390/cpacf: Enable key wrapping by default
Pull s390 fixes from Martin Schwidefsky:
"One performance optimization for page_clear and a couple of bug fixes"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390/mm: fix incorrect ASCE after crst_table_downgrade
s390/ftrace: fix crashes when switching tracers / add notrace to cpu_relax()
s390/pci: unify pci_iomap symbol exports
s390/pci: fix [un]map_resources sequence
s390: let the compiler do page clearing
s390/pci: fix possible information leak in mmio syscall
s390/dcss: array index 'i' is used before limits check.
s390/scm_block: fix off by one during cluster reservation
s390/jump label: improve and fix sanity check
s390/jump label: add missing jump_label_apply_nops() call
Pull seq-buf/ftrace fixes from Steven Rostedt:
"This includes fixes for seq_buf_bprintf() truncation issue. It also
contains fixes to ftrace when /proc/sys/kernel/ftrace_enabled and
function tracing are started. Doing the following causes some issues:
# echo 0 > /proc/sys/kernel/ftrace_enabled
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
# echo 1 > /proc/sys/kernel/ftrace_enabled
# echo nop > /sys/kernel/debug/tracing/current_tracer
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
As well as with function tracing too. Pratyush Anand first reported
this issue to me and supplied a patch. When I tested this on my x86
test box, it caused thousands of backtraces and warnings to appear in
dmesg, which also caused a denial of service (a warning for every
function that was listed). I applied Pratyush's patch but it did not
fix the issue for me. I looked into it and found a slight problem
with trampoline accounting. I fixed it and sent Pratyush a patch, but
he said that it did not fix the issue for him.
I later learned tha Pratyush was using an ARM64 server, and when I
tested on my ARM board, I was able to reproduce the same issue as
Pratyush. After applying his patch, it fixed the problem. The above
test uncovered two different bugs, one in x86 and one in ARM and
ARM64. As this looked like it would affect PowerPC, I tested it on my
PPC64 box. It too broke, but neither the patch that fixed ARM or x86
fixed this box (the changes were all in generic code!). The above
test, uncovered two more bugs that affected PowerPC. Again, the
changes were only done to generic code. It's the way the arch code
expected things to be done that was different between the archs. Some
where more sensitive than others.
The rest of this series fixes the PPC bugs as well"
* tag 'trace-fixes-v4.0-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Fix ftrace enable ordering of sysctl ftrace_enabled
ftrace: Fix en(dis)able graph caller when en(dis)abling record via sysctl
ftrace: Clear REGS_EN and TRAMP_EN flags on disabling record via sysctl
seq_buf: Fix seq_buf_bprintf() truncation
seq_buf: Fix seq_buf_vprintf() truncation
Pull networking fixes from David Miller:
1) nft_compat accidently truncates ethernet protocol to 8-bits, from
Arturo Borrero.
2) Memory leak in ip_vs_proc_conn(), from Julian Anastasov.
3) Don't allow the space required for nftables rules to exceed the
maximum value representable in the dlen field. From Patrick
McHardy.
4) bcm63xx_enet can accidently leave interrupts permanently disabled
due to errors in the NAPI polling exit logic. Fix from Nicolas
Schichan.
5) Fix OOPSes triggerable by the ping protocol module, due to missing
address family validations etc. From Lorenzo Colitti.
6) Don't use RCU locking in sleepable context in team driver, from Jiri
Pirko.
7) xen-netback miscalculates statistic offset pointers when reporting
the stats to userspace. From David Vrabel.
8) Fix a leak of up to 256 pages per VIF destroy in xen-netaback, also
from David Vrabel.
9) ip_check_defrag() cannot assume that skb_network_offset(),
particularly when it is used by the AF_PACKET fanout defrag code.
From Alexander Drozdov.
10) gianfar driver doesn't query OF node names properly when trying to
determine the number of hw queues available. Fix it to explicitly
check for OF nodes named queue-group. From Tobias Waldekranz.
11) MID field in macb driver should be 12 bits, not 16. From Punnaiah
Choudary Kalluri.
12) Fix unintentional regression in traceroute due to timestamp socket
option changes. Empty ICMP payloads should be allowed in
non-timestamp cases. From Willem de Bruijn.
13) When devices are unregistered, we have to get rid of AF_PACKET
multicast list entries that point to it via ifindex. Fix from
Francesco Ruggeri.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (38 commits)
tipc: fix bug in link failover handling
net: delete stale packet_mclist entries
net: macb: constify macb configuration data
MAINTAINERS: add Marc Kleine-Budde as co maintainer for CAN networking layer
MAINTAINERS: linux-can moved to github
can: kvaser_usb: Read all messages in a bulk-in URB buffer
can: kvaser_usb: Avoid double free on URB submission failures
can: peak_usb: fix missing ctrlmode_ init for every dev
can: add missing initialisations in CAN related skbuffs
ip: fix error queue empty skb handling
bgmac: Clean warning messages
tcp: align tcp_xmit_size_goal() on tcp_tso_autosize()
net: fec: fix unbalanced clk disable on driver unbind
net: macb: Correct the MID field length value
net: gianfar: correctly determine the number of queue groups
ipv4: ip_check_defrag should not assume that skb_network_offset is zero
net: bcmgenet: properly disable password matching
net: eth: xgene: fix booting with devicetree
bnx2x: Force fundamental reset for EEH recovery
xen-netback: refactor xenvif_handle_frag_list()
...
Pull regulator fixes from Mark Brown:
"A couple of driver specific fixes plus a fix for a regression in the
core where the updates to use sysfs group registration were overly
enthusiastic in eliding properties and removed some that had been
previously present"
* tag 'regulator-v4.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: Fix regression due to NULL constraints check
regulator: rk808: Set the enable time for LDOs
regulator: da9210: Mask all interrupt sources to deassert interrupt line
Pull spi fixes from Mark Brown:
"A collection of driver specific fixes to which the usual comments
about them being important if you see them mostly apply (except for
the comment fix). The pl022 one is particularly nasty for anyone
affected by it"
* tag 'spi-v4.0-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: pl022: Fix race in giveback() leading to driver lock-up
spi: dw-mid: avoid potential NULL dereference
spi: img-spfi: Verify max spfi transfer length
spi: fix a typo in comment.
spi: atmel: Fix interrupt setup for PDC transfers
spi: dw: revisit FIFO size detection again
spi: dw-pci: correct number of chip selects
drivers: spi: ti-qspi: wait for busy bit clear before data write/read
Pull tpm fixes from James Morris:
"fixes for the TPM driver"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security:
tpm: fix call order in tpm-chip.c
tpm/ibmvtpm: Additional LE support for tpm_ibmvtpm_send
Pull fbdev fixes from Tomi Valkeinen:
- Fix regression in with omapdss when using i2c displays
- Fix possible null deref in fbmon
- Check kalloc return value in AMBA CLCD
* tag 'fbdev-fixes-4.0' of git://git.kernel.org/pub/scm/linux/kernel/git/tomba/linux:
OMAPDSS: fix regression with display sysfs files
video: fbdev: fix possible null dereference
video: ARM CLCD: Add missing error check for devm_kzalloc
Pull cgroup fixes from Tejun Heo:
"The cgroup iteration update two years ago and the recent cpuset
restructuring introduced regressions in subset of cpuset
configurations. Three patches to fix them.
All are marked for -stable"
* 'for-4.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cpuset: Fix cpuset sched_relax_domain_level
cpuset: fix a warning when clearing configured masks in old hierarchy
cpuset: initialize effective masks when clone_children is enabled
Pull libata fixlet from Tejun Heo:
"Speed limiting fix for sata_fsl"
* 'for-4.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
sata-fsl: Apply link speed limits
Pull workqueue fix from Tejun Heo:
"One fix patch for a subtle livelock condition which can happen on
PREEMPT_NONE kernels involving two racing cancel_work calls. Whoever
comes in the second has to wait for the previous one to finish. This
was implemented by making the later one block for the same condition
that the former would be (work item completion) and then loop and
retest; unfortunately, depending on the wake up order, the later one
could lock out the former one to finish by busy looping on the cpu.
This is fixed by implementing explicit wait mechanism. Work item
might not belong anywhere at this point and there's remote possibility
of thundering herd problem. I originally tried to use bit_waitqueue
but it didn't work for static work items on modules. It's currently
using single wait queue with filtering wake up function and exclusive
wakeup. If this ever becomes a problem, which is not very likely, we
can try to figure out a way to piggy back on bit_waitqueue"
* 'for-4.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: fix hang involving racing cancel[_delayed]_work_sync()'s for PREEMPT_NONE
David Vrabel
Alexey Kodanev
Neal Cardwell
chas williams - CONTRACTOR
David S. Miller
Oliver Hartkopp
Hariprasad Shenai
Nimrod Andy
Yongbae Park
WANG Cong
Linus Torvalds