Use correct shift factor for extracting the SGE DMA Ingress Padding
Boundary. Was accidentally using the register field's shift which was
close enough (4 instead of the propper value of 5) that it actually
sort of worked for various packet sizes ...
Signed-off-by: Casey Leedom <leedom@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove obsolete comment about the lack of a TX Timer Callback -- which
we now _do_ have ...
Signed-off-by: Casey Leedom <leedom@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When two systems using bonding devices in adaptive load
balancing (ALB) communicates with each other, an endless
ping-pong of ARP replies starts between these two systems.
What happens? In the ALB mode, bonding driver keeps track
of each client connected in a hash table, so it can do the
receive load balancing (RLB). This hash table is updated
when an ARP reply is received, then it scans for the client
entry, updates its MAC address and flag it to be announced
later. Therefore, two seconds later, the alb monitor runs
and send for each updated client entry two ARP replies
updating this specific client. The same process happens on
the receiving system, causing the endless ping-pong of arp
replies.
See more information including the relevant functions below:
System 1 System 2
bond0 bond0
ping <system2>
ARP request --------->
<--------- ARP reply
+->rlb_arp_recv <---------------------+ <--- loop begins
| rlb_update_entry_from_arp |
| client_info->ntt = 1; |
| bond_info->rx_ntt = 1; |
| |
| <communication succeed> |
| |
| bond_alb_monitor |
| rlb_update_rx_clients |
| rlb_update_client |
| arp_create(ARPOP_REPLY) |
| send ARP reply --------------> V
| send ARP reply -------------->
| rlb_arp_recv
| rlb_update_entry_from_arp
| client_info->ntt = 1;
| bond_info->rx_ntt = 1;
| < snipped, same as in system 1>
+------- <-------------- send ARP reply
<-------------- send ARP reply
Besides the unneeded networking traffic, this loop breaks
a cluster because a backup system can't take over the IP
address. There is always one system sending an ARP reply
poisoning the network.
This patch fixes the problem adding a check for the MAC
address before updating it. Thus, if the MAC address didn't
change, there is no need to update neither to announce it later.
Signed-off-by: Flavio Leitner <fleitner@redhat.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
add fast path for in-order fragments
As the fragments are sent in order in most of OSes, such as Windows, Darwin and
FreeBSD, it is likely the new fragments are at the end of the inet_frag_queue.
In the fast path, we check if the skb at the end of the inet_frag_queue is the
prev we expect.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
----
include/net/inet_frag.h | 1 +
net/ipv4/ip_fragment.c | 12 ++++++++++++
net/ipv6/reassembly.c | 11 +++++++++++
3 files changed, 24 insertions(+)
Signed-off-by: David S. Miller <davem@davemloft.net>
/proc/net/snmp and /proc/net/netstat expose SNMP counters.
Width of these counters is either 32 or 64 bits, depending on the size
of "unsigned long" in kernel.
This means user program parsing these files must already be prepared to
deal with 64bit values, regardless of user program being 32 or 64 bit.
This patch introduces 64bit snmp values for IPSTAT mib, where some
counters can wrap pretty fast if they are 32bit wide.
# netstat -s|egrep "InOctets|OutOctets"
InOctets: 244068329096
OutOctets: 244069348848
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ibm p7 architecure seems to reorder memory accesses more
aggressively than previous ppc64 architectures. This requires memory
barriers to ensure that rx/tx doorbells are pressed only after
memory to be DMAed is written.
Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The original code is off by one because we should start counting at
zero. So the size of the resource is end - start + 1. I switched it to
use resource_size() to do the calculation.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix abuse of the preincrement operator as detected when building with gcc
4.6.0:
CC [M] drivers/bluetooth/hci_bcsp.o
drivers/bluetooth/hci_bcsp.c: In function 'bcsp_prepare_pkt':
drivers/bluetooth/hci_bcsp.c:247:20: warning: operation on 'bcsp->msgq_txseq' may be undefined
Reported-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Gustavo F. Padovan <padovan@profusion.mobi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some controllers (KW, Dove) limits the TX IP/layer4 checksum offloading to a max size.
Signed-off-by: Saeed Bishara <saeed@marvell.com>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
act_nat: use stack variable
structure tc_nat isn't too big for stack, so we can put it in stack.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
----
net/sched/act_nat.c | 31 ++++++++++---------------------
1 file changed, 10 insertions(+), 21 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>
act_mirred: combine duplicate code
tcf_bstats is updated in any way, so we can do it earlier to reduce the size of
the code.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
----
net/sched/act_mirred.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow selection of minstrel_ht as default rate control algorithm. At
the moment minstrel_ht can only be requested by the driver code but
not selected as default in make menuconfig. Fix this by using
minstrel_ht when minstrel was selected as default and minstrel_ht
is available.
This change won't affect legacy devices as minstrel_ht falls back to
minstrel in that case.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
When issuing a reset, the TSF value is lost in the hardware because of
the 913x specific cold reset. As with some AR9280 cards, the TSF needs
to be preserved in software here.
Additionally, there's an issue that frequently prevents a successful
TSF write directly after the chip reset. In this case, repeating the
TSF write after the initval-writes usually works.
This patch detects failed TSF writes and recovers from them, taking
into account the delay caused by the initval writes.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Reported-by: Björn Smedman <bjorn.smedman@venatech.se>
Cc: stable@kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>
CC [M] drivers/net/wireless/rt2x00/rt2800lib.o
drivers/net/wireless/rt2x00/rt2800lib.c: In function 'rt2800_ampdu_action':
drivers/net/wireless/rt2x00/rt2800lib.c:2821: warning: unused variable 'rt2x00dev'
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Latest rt2870 legacy driver also sets BBP_CSR_CFG_BBP_RW_MODE to 1
when reading or writing the EEPROM. This means we can make the
BBP reading and writing completely equal on all platforms.
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Legacy driver indicates that BBP1_TX_ANTENNA must be set
to 0 for TXPATH values of 1 and 3. So the previous statement
that nothing should be done for TXPATH = 3, is false.
Furthermore, remove the false BBP3_RX_ANTENNA initialization
when TXPATH is 1 for PCI and SOC devices. This field will always
be overridden in the next switch statement, making this initialization
bogus. History of this line indicates it was there from the beginning,
and was once caught as typo. Instead of replacing the line with the
correct line, the correct line was added...
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
IEEE80211_HT_CAP_RX_STBC is a 2 bit flag, and should thus
never be set as normal flag. Instead we must read the number
of RX paths from the EEPROM and set the IEEE80211_HT_CAP_RX_STBC
with the correct value (using the same logic as the number of TX
streams).
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
When an interface is removed the according beacon entry should be reset.
The current approach to only clear the first word is not enough to stop
the device from sending out the beacon, hence resulting in beacons being
sent out for already removed interfaces.
Fix this by invalidating the entire TXWI in front of the beacon instead
of only the first word.
Also clear all beacons during startup in the same way.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Since the link tuning is based on average RSSI values taken from all received
frames it doesn't make sense to enable it in AP mode where every associated
station provides independent RSSI values. Furthermore the legacy drivers
don't enable link tuning in AP mode as well.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Fix rt61pci beacon updates in the same way as rt2800pci. rt61pci didn't
update the beacon template after each beacon interval, resulting in the
DTIM count being incorrect (if DTIM period > 1). Fix this by calling
rt2x00lib_beacondone after the current beacon was sent out.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
rt2800pci didn't update the beacon template after each beacon interval,
resulting in the DTIM count being incorrect (if DTIM period > 1). Fix this
by calling rt2x00lib_beacondone after the current beacon was sent out.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
MAC_BSSID_DW1_BSS_ID_MASK must be set to the mask 3, to
enable 8 BSSID's. The MAC_BSSID_DW1_BSS_BCN_NUM is initialized
to 7 to enable the 8 beacons.
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Tested-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Align with the latest versions of the Ralink legacy driver(s).
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Align with the way PCI devices are handled, even though it is not
strictly necessary.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Now that the {usb,pci} specific write_tx_data functions are no longer
present we can rename the write_tx_datadesc callback function back to
its old name.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Now that the write_tx_data functions are merged, also merge the relevant
parts of the txdone handling into common code, rather than {usb,pci}
specific code.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Now that rt2x00pci_write_tx_data and rt2x00usb_write_tx_data are similar
we can merge them in a single function in rt2x00queue.c.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
There is no need to fill the TX URB this early, and moving it to the
rt2x00usb_kick_tx_entry function allows us to merge the PCI and USB
variants of the write_tx_data function.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
We forgot to clear the SKBDESC_DESC_IN_SKB when the descriptor was removed
from the front of the skb.
Signed-off-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The update_bssid is set only when BSS_CHANGED_BSSID is used,
but the check if that field is true is done later in the function
but also only when BSS_CHANGED_BSSID is set. This makes the
variable useless, as it can never result in a negative check.
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Helmut Schaa <helmut.schaa@googlemail.com>
Acked-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
For the Master mode case, we initialized the BSSID as the MAC
address, but never wrote it into the registers. This causes
Hardware crypto to break in Master mode when receiving frames
which require the BSSID to be filled in.
This is safe for STA mode since the BSSID will be initialized
to 00:00:00:00:00 at this point, but will be set to the correct
value later when the device associates.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Acked-by: Gertjan van Wingerde <gwingerde@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In order to implement tx mpdu aggregation we only have to implement
the ampdu_action callback such that mac80211 allows negotiation of
blockack sessions.
The hardware will handle everything on its own as long as the ampdu
flag in the TXWI struct is set up correctly and we translate the tx
status correctly.
For now, refuse requests to start rx aggregation.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
MPC8313ECE says:
"If the controller receives a 1- or 2-byte frame (such as an illegal
runt packet or a packet with RX_ER asserted) before GRS is asserted
and does not receive any other frames, the controller may fail to set
GRSC even when the receive logic is completely idle. Any subsequent
receive frame that is larger than two bytes will reset the state so
the graceful stop can complete. A MAC receiver (Rx) reset will also
reset the state."
This patch implements the proposed workaround:
"If IEVENT[GRSC] is still not set after the timeout, read the eTSEC
register at offset 0xD1C. If bits 7-14 are the same as bits 23-30,
the eTSEC Rx is assumed to be idle and the Rx can be safely reset.
If the register fields are not equal, wait for another timeout
period and check again."
Signed-off-by: Anton Vorontsov <avorontsov@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
MPC8313ECE says:
"For TOE=1 huge or jumbo frames, the data required to generate the
checksum may exceed the 2500-byte threshold beyond which the controller
constrains itself to one memory fetch every 256 eTSEC system clocks.
This throttling threshold is supposed to trigger only when the
controller has sufficient data to keep transmit active for the duration
of the memory fetches. The state machine handling this threshold,
however, fails to take large TOE frames into account. As a result,
TOE=1 frames larger than 2500 bytes often see excess delays before start
of transmission."
This patch implements the workaround as suggested by the errata
document, i.e.:
"Limit TOE=1 frames to less than 2500 bytes to avoid excess delays due to
memory throttling.
When using packets larger than 2700 bytes, it is recommended to turn TOE
off."
To be sure, we limit the TOE frames to 2500 bytes, and do software
checksumming instead.
Signed-off-by: Anton Vorontsov <avorontsov@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
MPC8313ECE says:
"If MACCFG2[Huge Frame]=0 and the Ethernet controller receives frames
which are larger than MAXFRM, the controller truncates the frames to
length MAXFRM and marks RxBD[TR]=1 to indicate the error. The controller
also erroneously marks RxBD[TR]=1 if the received frame length is MAXFRM
or MAXFRM-1, even though those frames are not truncated.
No truncation or truncation error occurs if MACCFG2[Huge Frame]=1."
There are two options to workaround the issue:
"1. Set MACCFG2[Huge Frame]=1, so no truncation occurs for invalid large
frames. Software can determine if a frame is larger than MAXFRM by
reading RxBD[LG] or RxBD[Data Length].
2. Set MAXFRM to 1538 (0x602) instead of the default 1536 (0x600), so
normal-length frames are not marked as truncated. Software can examine
RxBD[Data Length] to determine if the frame was larger than MAXFRM-2."
This patch implements the first workaround option by setting HUGEFRAME
bit, and gfar_clean_rx_ring() already checks the RxBD[Data Length].
Signed-off-by: Anton Vorontsov <avorontsov@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Determine the size of the xfrm_mark struct, not of its pointer.
Signed-off-by: Andreas Steffen <andreas.steffen@strongswan.org>
Acked-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is only noticed by people that are not doing everything correct in
the first place.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
max_desync_factor can be configured per-interface, but nothing is
using the value.
Reported-by: Piotr Lewandowski <piotr.lewandowski@gmail.com>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since addresses are only revalidated every 2 minutes, the reported
valid_lft can underflow shortly before the address is deleted.
Clamp it to a minimum of 0, as for prefered_lft.
Reported-by: Piotr Lewandowski <piotr.lewandowski@gmail.com>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixed brace, static initialization, comment, whitespace and spacing
coding style issues.
Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This avoids scheduling in atomic context and also means that IRQs
will only be deferred for relatively short periods of time.
Previously discussed in:
http://article.gmane.org/gmane.linux.network/155024
Reported-by: Arne Nordmark <nordmark@mech.kth.se>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
In driver ixgbe, ixgbe_atr may cause crashes for non-ipv4 packets. Just
add a test to check skb->protocol. It may crash on short packets due
to ip_hdr() access.
Signed-off-by: Guillaume Gaudonville <guillaume.gaudonville@6wind.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Disabling the tx laser while receiving DMA requests
can hang the device. After this occurs the device
is in a bad state. The GPIO bit never clears when
PCI master access is disabled and a reboot is required
to get the device in a good state again.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch added to 2.6.34:
commit 5f6c018199
Author: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Wed Apr 14 16:04:23 2010 -0700
ixgbe: fix bug with vlan strip in promsic mode
among other things added a function called ixgbe_vlan_filter_enable.
This new function wants to access and set some rx_ring parameters, but
adapter->rx_ring has already been freed. This simply moves the free
until after the access and makes __ixgbe_shutdown look more like
ixgbe_remove.
Signed-off-by: Andy Gospodarek <andy@greyhouse.net>
Acked-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Emil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based on community feedback, EEE should be disabled by default until the
IEEE802.3az specification has been finalized.
Cc: bhutchings@solarflare.com
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Emil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Casey Leedom