- 14 Nov, 2009 19 commits
-
-
Frank Blaschka authored
State transition to DEV_STATE_STOPPED indicates all outstanding I/O has finished. Add wait queue to wait for this state. Signed-off-by:
Frank Blaschka <frank.blaschka@de.ibm.com> Signed-off-by:
Ursula Braun <ursula.braun@de.ibm.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Ursula Braun authored
If iucv_work_queue is not empty during kernel freeze, a kernel panic occurs. This suspend-patch adds flushing of the work queue for pending connection requests and severing of remaining pending connections. Signed-off-by:
-
Eric Dumazet authored
While investigating for network latencies, I found inet_getid() was a contention point for some workloads, as inet_peer_idlock is shared by all inet_getid() users regardless of peers. One way to fix this is to make ip_id_count an atomic_t instead of __u16, and use atomic_add_return(). In order to keep sizeof(struct inet_peer) = 64 on 64bit arches tcp_ts_stamp is also converted to __u32 instead of "unsigned long". Signed-off-by:
Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by:
-
Eric Dumazet authored
When handling large number of netdevices, inet6_dump_addr() is very slow because it has O(N^2) complexity. Instead of scanning one single list, we can use the NETDEV_HASHENTRIES sub lists of the dev_index hash table, and RCU lookups. Signed-off-by:
-
Lucian Adrian Grijincu authored
The first "node" is supposed to be the cursor used in the for_each. The second "node" is ment literally and should not be macro expanded: it's the name of the hlist_node field from the inet_bind_bucket. This currently works because when inet_bind_bucket_for_each is called it's argument is still "node". Signed-off-by:
Lucian Adrian Grijincu <lgrijincu@ixiacom.com> Signed-off-by:
-
Eric Dumazet authored
Stephen Hemminger a écrit : > On Thu, 12 Nov 2009 15:11:36 +0100 > Eric Dumazet <eric.dumazet@gmail.com> wrote: > >> When handling large number of netdevices, inet_dump_ifaddr() >> is very slow because it has O(N^2) complexity. >> >> Instead of scanning one single list, we can use the NETDEV_HASHENTRIES >> sub lists of the dev_index hash table, and RCU lookups. >> >> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> > > You might be able to make RCU critical section smaller by moving > it into loop. > Indeed. But we dump at most one skb (<= 8192 bytes ?), so rcu_read_lock holding time is small, unless we meet many netdevices without addresses. I wonder if its really common... Thanks [PATCH net-next-2.6] ipv4: speedup inet_dump_ifaddr() When handling large number of netdevices, inet_dump_ifaddr() is very slow because it has O(N2) complexity. Instead of scanning one single list, we can use the NETDEV_HASHENTRIES sub lists of the dev_index hash table, and RCU lookups. Signed-off-by:
Stephen Hemminger <shemminger@vyatta.com> Signed-off-by:
-
PJ Waskiewicz authored
This patch pairs similar-numbered Rx and Tx queues onto a single MSI-X vector. For example, Tx queue 0 and Rx queue 0's interrupt with be ethX-RxTx-0. This allows for more efficient cleanup, since fewer interrupts will be firing during device operation. It also helps with a cleaner CPU affinity for IRQ affinity. Signed-off-by:
Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by:
-
Nelson, Shannon authored
Signed-off-by:
Shannon Nelson <shannon.nelson@intel.com> Signed-off-by:
-
Alexander Duyck authored
This patch makes it so that we only recycle pages when they are from the local NUMA node. Non-local pages are freed and replaced with locally allocated pages. Signed-off-by:
Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by:
-
Alexander Duyck authored
We were previously only checking the first tx ring to see if it had any packets in it when the link when down. However we should be checking all of the rings so this patch makes it so that all of the rings are now being checked. Signed-off-by:
-
Alexander Duyck authored
This patch removes unused variables total_tx_bytes, total_tx_packets, total_rx_bytes, and total_rx_packets from the adapter struct. Signed-off-by:
-
Alexander Duyck authored
This patch does some refactoring work that I felt was needed after reviewing the changes recently submitted relating to the replacement of net_stats with netdev->stats. This patch essentially creates two different collections of stats. The first handles the adapter specific states and is stored in gstring_stats, and the second is for netdev specific stats and is stored in gstring_net_stats. Signed-off-by:
-
Alexander Duyck authored
This patch makes it so that nics such as 82576 and newer can support more hardware queues when there are more than 4 cpus by combining a tx/rx queue pair onto one interrupt so that 8 queue pairs can be supported and thus allow for more queues. Signed-off-by:
-
Alexander Duyck authored
Current code is quite large and making igb_probe difficult to read. Signed-off-by:
-
Alexander Duyck authored
Change the type for the ring size values to u16 and use min/max_t instead of min/max. Signed-off-by:
-
Wan ZongShun authored
To fix the bug of checking on signed return value using unsigned statement. Thanks Roel Kluin for digging out it. Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
Wan ZongShun <mcuos.com@gmail.com> Signed-off-by:
-
Eric Dumazet authored
We need to use next_det_device_rcu() in RCU protected section. We also can avoid in_dev_get()/in_dev_put() overhead (code size mainly) in rcu_read_lock() sections. Signed-off-by:
-
Eric Dumazet authored
No longer need read_lock(&dev_base_lock), use RCU instead. We also can avoid taking references on inet6_dev structs. Signed-off-by:
-
William Allen Simpson authored
Define two symbols needed in both kernel and user space. Remove old (somewhat incorrect) kernel variant that wasn't used in most cases. Default should apply to both RMSS and SMSS (RFC2581). Replace numeric constants with defined symbols. Stand-alone patch, originally developed for TCPCT. Signed-off-by: William.Allen.Simpson@gmail.com Acked-by:
-
- 13 Nov, 2009 3 commits
-
-
Patrick McHardy authored
Both vlan and macvlan devices usually don't use a qdisc and immediately queue packets to the underlying device. Propagate transmission state of the underlying device to the upper layers so they can react on congestion and/or inform the sending process. Signed-off-by:
Patrick McHardy <kaber@trash.net> Signed-off-by:
-
Patrick McHardy authored
Currently the ->ndo_hard_start_xmit() callbacks are only permitted to return one of the NETDEV_TX codes. This prevents any kind of error propagation for virtual devices, like queue congestion of the underlying device in case of layered devices, or unreachability in case of tunnels. This patches changes the NET_XMIT codes to avoid clashes with the NETDEV_TX codes and changes the two callers of dev_hard_start_xmit() to expect either errno codes, NET_XMIT codes or NETDEV_TX codes as return value. In case of qdisc_restart(), all non NETDEV_TX codes are mapped to NETDEV_TX_OK since no error propagation is possible when using qdiscs. In case of dev_queue_xmit(), the error is propagated upwards. Signed-off-by:
-
Joe Perches authored
Untested, no hardware Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
- 12 Nov, 2009 18 commits
-
-
Arnd Bergmann authored
We have two implementations of the compat_ioctl handling for ATM, the one that we have had for ages in fs/compat_ioctl.c and the one added to net/atm/ioctl.c by David Woodhouse. Unfortunately, both versions are incomplete, and in practice we use a very confusing combination of the two. For ioctl numbers that have the same identifier on 32 and 64 bit systems, we go directly through the compat_ioctl socket operation, for those that differ, we do a conversion in fs/compat_ioctl.c. This patch moves both variants into the vcc_compat_ioctl() function, while preserving the current behaviour. It also kills off the COMPATIBLE_IOCTL definitions that we never use here. Doing it this way is clearly not a good solution, but I hope it is a step into the right direction, so that someone is able to clean up this mess for real. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: David Woodhouse <dwmw2@infradead.org> Signed-off-by:
-
Arnd Bergmann authored
Handling for SIOCSHWTSTAMP is broken on architectures with a split user/kernel address space like s390, because it passes a real user pointer while using set_fs(KERNEL_DS). A similar problem might arise the next time somebody adds code to dev_ifsioc. Split up dev_ifsioc into three separate functions for SIOCSHWTSTAMP, SIOC*IFMAP and all other numbers so we can get rid of set_fs in all potentially affected cases. Signed-off-by:
-
Ben Dooks authored
Add support for Wake on LAN (WOL) reception and waking the device up from this signal via the ethtool interface. Currently we are only supporting the magic-packet variant of wakeup. WOL is enabled by specifying a second interrupt resource to the driver which indicates where the interrupt for the WOL is being signalled. This then enables the necessary ethtool calls to leave the device in a state to receive WOL frames when going into suspend. Signed-off-by:
Ben Dooks <ben@simtec.co.uk> Signed-off-by:
Simtec Linux Team <linux@simtec.co.uk> Signed-off-by:
-
Ali Gholami Rudi authored
The values of r_idx and rx_ring are not used after the last time they are set in ixgbe_msix_clean_rx(), so they can be removed. Signed-off-by:
Ali Gholami Rudi <ali@rudi.ir> Acked-by:
-
stephen hemminger authored
There is no reason for this lock to be reader/writer since the reader only has lock held for a very brief period. The overhead of read_lock is more expensive than spinlock. Compile tested only, I am not a decnet user. Signed-off-by:
-
stephen hemminger authored
Add missing locking in the case of auto binding to the default device. The address list might change while this code is looking at the list. Compile tested only, I am not a decnet user. Signed-off-by:
-
stephen hemminger authored
The full_name_hash function does not produce well distributed values in the lower bits, so most code uses hash_32() to fold it. This is really a bug introduced when name hashing was added, back in 2.5 when I added name hashing. hash_32 is all that is needed since full_name_hash returns unsigned int which is only 32 bits on 64 bit platforms. Also, there is no point in using hash_32 on ifindex, because the is naturally sequential and usually well distributed. Signed-off-by:
-
Ron Mercer authored
Signed-off-by:
Ron Mercer <ron.mercer@qlogic.com> Signed-off-by:
-
Ron Mercer authored
Change it to match qlge_xxx convention. Signed-off-by:
-
Ron Mercer authored
Force asic to known state at open(). Signed-off-by:
-
Ron Mercer authored
We do not need to change the frame routing to direct all frames to the management fifo during suspend. This is now done by the firmware. Signed-off-by:
-
David S. Miller authored
Noticed by Thomas GLeixner. Signed-off-by: -
Anton Vorontsov authored
Before calling gfar_clean_tx_ring() the driver grabs an irqsave spinlock, and then tries to recycle skbs. But since skb_recycle_check() returns 0 with IRQs disabled, we'll never recycle any skbs. It appears that gfar_clean_tx_ring() and gfar_start_xmit() are mostly idependent and can work in parallel, except when they modify num_txbdfree. So we can drop the lock from most sections and thus fix the skb recycling. Signed-off-by:
Anton Vorontsov <avorontsov@ru.mvista.com> Acked-by:
Kumar Gala <galak@kernel.crashing.org> Signed-off-by:
-
Anton Vorontsov authored
gfar_error() can arrive at the middle of gfar_start_xmit() processing, and so it can trigger transfers of BDs that we don't yet expect to be transmitted. Fix this by locking the tx queues in gfar_error(). Signed-off-by:
-
Anton Vorontsov authored
We obviously want to write a modified 'temp' value back to the register, not the saved IRQ flags. Signed-off-by:
-
Anton Vorontsov authored
commit fba4ed03 ("gianfar: Add Multiple Queue Support") introduced the following build failure: CC gianfar.o gianfar.c: In function 'gfar_restore': gianfar.c:1249: error: request for member 'napi' in something not a structure or union This patch fixes the issue. Signed-off-by:
-
Anton Vorontsov authored
It is OK to poll with disabled IRQs, so remove the warning. Signed-off-by:
-
Anton Vorontsov authored
NAPI drivers try to recycle SKBs in their polling routine, but we generally don't know the context in which the polling will be called, and the skb recycling itself may require IRQs to be enabled. This patch adds irqs_disabled() test to the skb_recycle_check() routine, so that we'll not let the drivers hit the skb recycling path with IRQs disabled. As a side effect, this patch actually disables skb recycling for some [broken] drivers. E.g. gianfar driver grabs an irqsave spinlock during TX ring processing, and then tries to recycle an skb, and that caused the following badness: nf_conntrack version 0.5.0 (1008 buckets, 4032 max) ------------[ cut here ]------------ Badness at kernel/softirq.c:143 NIP: c003e3c4 LR: c423a528 CTR: c003e344 ... NIP [c003e3c4] local_bh_enable+0x80/0xc4 LR [c423a528] destroy_conntrack+0xd4/0x13c [nf_conntrack] Call Trace: [c15d1b60] [c003e32c] local_bh_disable+0x1c/0x34 (unreliable) [c15d1b70] [c423a528] destroy_conntrack+0xd4/0x13c [nf_conntrack] [c15d1b80] [c02c6370] nf_conntrack_destroy+0x3c/0x70 Signed-off-by:
-
