- 13 Jan, 2012 1 commit
-
-
Eric Dumazet authored
Adds an optional Random Early Detection on each SFQ flow queue. Traditional SFQ limits count of packets, while RED permits to also control number of bytes per flow, and adds ECN capability as well. 1) We dont handle the idle time management in this RED implementation, since each 'new flow' begins with a null qavg. We really want to address backlogged flows. 2) if headdrop is selected, we try to ecn mark first packet instead of currently enqueued packet. This gives faster feedback for tcp flows compared to traditional RED [ marking the last packet in queue ] Example of use : tc qdisc add dev $DEV parent 1:1 handle 10: est 1sec 4sec sfq \ limit 3000 headdrop flows 512 divisor 16384 \ redflowlimit 100000 min 8000 max 60000 probability 0.20 ecn qdisc sfq 10: parent 1:1 limit 3000p quantum 1514b depth 127 headdrop flows 512/16384 divisor 16384 ewma 6 min 8000b max 60000b probability 0.2 ecn prob_mark 0 prob_mark_head 4876 prob_drop 6131 forced_mark 0 forced_mark_head 0 forced_drop 0 Sent 1175211782 bytes 777537 pkt (dropped 6131, overlimits 11007 requeues 0) rate 99483Kbit 8219pps backlog 689392b 456p requeues 0 In this test, with 64 netperf TCP_STREAM sessions, 50% using ECN enabled flows, we can see number of packets CE marked is smaller than number of drops (for non ECN flows) If same test is run, without RED, we can check backlog is much bigger. qdisc sfq 10: parent 1:1 limit 3000p quantum 1514b depth 127 headdrop flows 512/16384 divisor 16384 Sent 1148683617 bytes 795006 pkt (dropped 0, overlimits 0 requeues 0) rate 98429Kbit 8521pps backlog 1221290b 841p requeues 0 Signed-off-by:
Eric Dumazet <eric.dumazet@gmail.com> CC: Stephen Hemminger <shemminger@vyatta.com> CC: Dave Taht <dave.taht@gmail.com> Tested-by:
Dave Taht <dave.taht@gmail.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- 12 Jan, 2012 7 commits
-
-
Manfred Rudigier authored
Similar problem as in 481a8199 ("can: fix NOHZ local_softirq_pending 08 warning"). This fix replaces netif_rx() with netif_rx_ni() which has to be used from process/softirq context. Signed-off-by:
Manfred Rudigier <manfred.rudigier@omicron.at> Signed-off-by:
-
Manfred Rudigier authored
When TX time stamping for PTP messages is enabled on a socket, a time stamp is returned on the socket error queue to the user space application after the frame was transmitted. The transmitted frame is also returned on the error queue so that an application knows to which frame the time stamp belongs. In the current implementation the TxFCB is immediately followed by the frame. Since the eTSEC inserts the TX time stamp 8 bytes after the TxFCB, parts of the frame have been overwritten and an invalid frame was returned on the socket error queue. This patch fixes the described problem by adding additional 16 padding bytes between the TxFCB and the frame for all messages sent from a time stamping enabled socket (other sockets are not affected). Signed-off-by:
-
Manfred Rudigier authored
When there is not enough headroom in the skb a private copy will be made. However, the private copy had no reference to the socket and consequently no time stamp could be queued on the socket error queue during the skb_tstamp_tx function. This patch fixes this issue by also stealing the sock reference from the original skb after making the private copy. Signed-off-by:
-
Timur Tabi authored
Introduce function mdiobus_alloc_size() as an alternative to mdiobus_alloc(). Most callers of mdiobus_alloc() also allocate a private data structure, and then manually point bus->priv to this object. mdiobus_alloc_size() combines the two operations into one, which simplifies memory management. The original mdiobus_alloc() now just calls mdiobus_alloc_size(0). Signed-off-by:
Timur Tabi <timur@freescale.com> Signed-off-by:
-
Glauber Costa authored
The logic of the current code is that whenever we destroy a cgroup that had its limit set (set meaning different than maximum), we should decrement the jump_label counter. Otherwise we assume it was never incremented. But what the code actually does is test for RES_USAGE instead of RES_LIMIT. Usage being different than maximum is likely to be true most of the time. The effect of this is that the key must become negative, and since the jump_label test says: !!atomic_read(&key->enabled); we'll have jump_labels still on when no one else is using this functionality. Signed-off-by:Glauber Costa <glommer@parallels.com> CC: David S. Miller <davem@davemloft.net> Signed-off-by:
-
Eric Dumazet authored
commit a9b3cd7f (rcu: convert uses of rcu_assign_pointer(x, NULL) to RCU_INIT_POINTER) did a lot of incorrect changes, since it did a complete conversion of rcu_assign_pointer(x, y) to RCU_INIT_POINTER(x, y). We miss needed barriers, even on x86, when y is not NULL. Signed-off-by:
-
-
- 11 Jan, 2012 26 commits
-
-
Pavel Emelyanov authored
Signed-off-by:
Pavel Emelyanov <xemul@parallels.com> Signed-off-by:
-
Pavel Emelyanov authored
Signed-off-by:
-
Maxim Uvarov authored
No need to lock soft irqs under bond_alb_xmit() which already has softirq disabled. Changes: 1. add non-bh/bh version to tlb_clear_slave() 2. represent BH and non BH hash table locks _lock_rx_hashtbl_bh/_unlock_rx_hashtbl_bh _lock_rx_hashtbl/_unlock_rx_hashtbl _lock_tx_hashtbl_bh/_unlock_tx_hashtbl_bh _lock_tx_hashtbl/_unlock_tx_hashtbl Signed-off-by:
Maxim Uvarov <maxim.uvarov@oracle.com> Signed-off-by:
Cong Wang <amwang@redhat.com> Signed-off-by:
-
Stanislaw Gruszka authored
Since: commit 816c04fe Author: Christian Lamparter <chunkeey@googlemail.com> Date: Sat Apr 30 15:24:30 2011 +0200 mac80211: consolidate MIC failure report handling is possible to that we dereference rx->key == NULL when driver set RX_FLAG_MMIC_STRIPPED and not RX_FLAG_IV_STRIPPED and we are in promiscuous mode. This happen with rt73usb and rt61pci at least. Before the commit we always check rx->key against NULL, so I assume fix should be done in mac80211 (also mic_fail path has similar check). References: https://bugzilla.redhat.com/show_bug.cgi?id=769766 http://rt2x00.serialmonkey.com/pipermail/users_rt2x00.serialmonkey.com/2012-January/004395.html Cc: stable@vger.kernel.org # 3.0+ Reported-by:
Stuart D Gathman <stuart@gathman.org> Reported-by:
Kai Wohlfahrt <kai.scorpio@gmail.com> Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Johannes Berg authored
My patch to validate station flags broke compatibility with the old station flags setting where all flags are always set at once since it always set the mask as all possible flags which ended up being rejected later in the station add/modify code. Fix by parsing only the current flags in the old flags attribute -- new applications and new flags should use (and will now require) the new flags attribute where the mask is given by the application. Reported-and-tested-by:
Thomas Hilber <ath9k-dev@toh.cx> Signed-off-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
Florian Fainelli <florian@openwrt.org> Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
Florian Fainelli authored
Signed-off-by:
-
- 10 Jan, 2012 6 commits
-
-
Jussi Kivilinna authored
In kernel v3.2 initialization sequence for Asix 88178 devices was changed so that hardware is reseted on every time interface is brought up (ifconfig up), instead just at USB probe time. This causes problem with setting custom MAC address to device as ax88178_reset causes reload of MAC address from EEPROM. This patch fixes the issue by rewriting MAC address at end of ax88178_reset. Signed-off-by:
Jussi Kivilinna <jussi.kivilinna@mbnet.fi> Acked-by:
Grant Grundler <grundler@chromium.org> Cc: Allan Chou <allan@asix.com.tw> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
Jussi Kivilinna authored
In kernel v3.2 initialization sequence for Asix 88772 devices was changed so that hardware is reseted on every time interface is brought up (ifconfig up), instead just at USB probe time. This causes problem with setting custom MAC address to device as ax88772_reset causes reload of MAC address from EEPROM. This patch fixes the issue by rewriting MAC address at end of ax88772_reset. Signed-off-by:
-
Fabio Estevam authored
Fix the following build warning: warning: (ISDN_PPP) selects SLHC which has unmet direct dependencies (NETDEVICES) Signed-off-by:
Fabio Estevam <fabio.estevam@freescale.com> Signed-off-by:
-
Stefan Roese authored
Otherwise the output looks like this: ... STMMAC - user ID: 0x10, Synopsys ID: 0x32 No HW DMA feature register supported Normal descriptors Remote wake-up capable Checksum Offload Engine supported No MAC Management Counters availableIP-Config: Complete: device=eth0, addr=192.168.20.42, mask=255.255.0.0, gw=192.168.1.254, ... Signed-off-by:Stefan Roese <sr@denx.de> Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com> Acked-by:
Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by:
-
Stefan Roese authored
Fix this error: CC drivers/net/ethernet/stmicro/stmmac/mmc_core.o drivers/net/ethernet/stmicro/stmmac/mmc_core.c: In function 'dwmac_mmc_ctrl': drivers/net/ethernet/stmicro/stmmac/mmc_core.c:143:2: error: implicit declaration of function 'pr_debug' [-Werror=implicit-function-declaration] Signed-off-by:
-
-
