Qlge is freeing the buffers before stopping the card DMA, and
this can cause some severe error, as a EEH event on PPC.
This patch just stop the card and then free the resources.
Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
close https://bugzilla.kernel.org/show_bug.cgi?id=16529
Before calling dev_forward_skb(), we should make sure skb head contains
at least an ethernet header, even if length included in upper layer said
so. Use pskb_may_pull() to make sure this ethernet header is present in
skb head.
Reported-by: Thomas Heil <heil@terminal-consulting.de>
Reported-by: Ian Campbell <Ian.Campbell@eu.citrix.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This issue come from ruby language community. Below test program
hang up when only run on Linux.
% uname -mrsv
Linux 2.6.26-2-486 #1 Sat Dec 26 08:37:39 UTC 2009 i686
% ruby -rsocket -ve '
BasicSocket.do_not_reverse_lookup = true
serv = TCPServer.open("127.0.0.1", 0)
s1 = TCPSocket.open("127.0.0.1", serv.addr[1])
s2 = serv.accept
s2.close
s1.write("a") rescue p $!
s1.write("a") rescue p $!
Thread.new {
s1.write("a")
}.join'
ruby 1.9.3dev (2010-07-06 trunk 28554) [i686-linux]
#<Errno::EPIPE: Broken pipe>
[Hang Here]
FreeBSD, Solaris, Mac doesn't. because Ruby's write() method call
select() internally. and tcp_poll has a bug.
SUS defined 'ready for writing' of select() as following.
| A descriptor shall be considered ready for writing when a call to an output
| function with O_NONBLOCK clear would not block, whether or not the function
| would transfer data successfully.
That said, EPIPE situation is clearly one of 'ready for writing'.
We don't have read-side issue because tcp_poll() already has read side
shutdown care.
| if (sk->sk_shutdown & RCV_SHUTDOWN)
| mask |= POLLIN | POLLRDNORM | POLLRDHUP;
So, Let's insert same logic in write side.
- reference url
http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31065http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/31068
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As discovered by Anton Blanchard, current code to autotune
tcp_death_row.sysctl_max_tw_buckets, sysctl_tcp_max_orphans and
sysctl_max_syn_backlog makes little sense.
The bigger a page is, the less tcp_max_orphans is : 4096 on a 512GB
machine in Anton's case.
(tcp_hashinfo.bhash_size * sizeof(struct inet_bind_hashbucket))
is much bigger if spinlock debugging is on. Its wrong to select bigger
limits in this case (where kernel structures are also bigger)
bhash_size max is 65536, and we get this value even for small machines.
A better ground is to use size of ehash table, this also makes code
shorter and more obvious.
Based on a patch from Anton, and another from David.
Reported-and-tested-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported by Anton Blanchard when we use
percpu_counter_read_positive() to make our orphan socket limit checks,
the check can be off by up to num_cpus_online() * batch (which is 32
by default) which on a 128 cpu machine can be as large as the default
orphan limit itself.
Fix this by doing the full expensive sum check if the optimized check
triggers.
Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Casting "pep->tx_desc_dma" to to a struct tx_desc pointer makes gcc
complain:
drivers/net/pxa168_eth.c:657: warning:
cast to pointer from integer of different size
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The phy_mii_ioctl() function changed recently. It now takes a struct
ifreq pointer directly.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A couple issues here:
* Some resources weren't released.
* If alloc_etherdev() failed it would have caused a NULL dereference
because "pep" would be null when we checked "if (pep->clk)".
* Also it's better to propagate the error codes from mdiobus_register()
instead of just returning -ENOMEM.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
"pep->pd" isn't checked consistently in this function. For example it's
dereferenced unconditionally on the next line after the end of the if
condition. This function is only called from pxa168_eth_probe() and
pep->pd is always non-NULL so I removed the check.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is possible that phylib will call adjust_link before returning
from {,of_}phy_connect(), which may cause the following [very rare,
though] oops upon reopening the device:
Unable to handle kernel paging request for data at address 0x0000024c
Oops: Kernel access of bad area, sig: 11 [#1]
PREEMPT SMP NR_CPUS=2 LTT NESTING LEVEL : 0
P1021 RDB
Modules linked in:
NIP: c0345dac LR: c0345dac CTR: c0345d84
TASK = dffab6b0[30] 'events/0' THREAD: c0d24000 CPU: 0
[...]
NIP [c0345dac] adjust_link+0x28/0x19c
LR [c0345dac] adjust_link+0x28/0x19c
Call Trace:
[c0d25f00] [000045e1] 0x45e1 (unreliable)
[c0d25f30] [c036c158] phy_state_machine+0x3ac/0x554
[...]
Here is why. Drivers store phydev in their private structures, e.g.
gianfar driver:
static int init_phy(struct net_device *dev)
{
...
priv->phydev = of_phy_connect(...);
...
}
So that adjust_link could retrieve it back:
static void adjust_link(struct net_device *dev)
{
...
struct phy_device *phydev = priv->phydev;
...
}
If the device has been opened before, then phydev->state is set to
PHY_HALTED (or undefined if the driver didn't call phy_stop()).
Now, phy_connect starts the PHY state machine before returning phydev to
the driver:
phy_start_machine(phydev, NULL);
if (phydev->irq > 0)
phy_start_interrupts(phydev);
return phydev;
The time between 'phy_start_machine()' and 'return phydev' is undefined.
The start machine routine delays execution for 1 second, which is enough
for most cases. But under heavy load, or if you're unlucky, it is quite
possible that PHY state machine will execute before phy_connect()
returns, and so adjust_link callback will try to dereference phydev,
which is not yet ready.
To fix the issue, simply initialize the PHY's state to PHY_READY during
phy_attach(). This will ensure that phylib won't call adjust_link before
phy_start().
Signed-off-by: Anton Vorontsov <avorontsov@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix this error on an s390 allyesconfig build:
linux-2.6/drivers/net/caif/caif_spi.c:98:
undefined reference to `dma_free_coherent'
Cc: Sjur Braendeland <sjur.brandeland@stericsson.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If netconsole is in use, there is a possibility for deadlock in 3c59x between
boomerang_interrupt and boomerang_start_xmit. Both routines take the vp->lock,
and if netconsole is in use, a pr_* call from the boomerang_interrupt routine
will result in the netconsole code attempting to trnasmit an skb, which can try
to take the same spin lock, resulting in deadlock.
The fix is pretty straightforward. This patch allocats a bit in the 3c59x
private structure to indicate that its handling an interrupt. If we get into
the transmit routine and that bit is set, we can be sure that we have recursed
and will deadlock if we continue, so instead we just return NETDEV_TX_BUSY, so
the stack requeues the skb to try again later.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Function qlcnic_intr has pointer to qlcnic_host_sds_ring
as second parameter not pointer to qlcnic_adapter.
Signed-off-by: Yinglin Luan <synmyth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Function netxen_intr has pointer to nx_host_sds_ring
as second parameter not pointer to netxen_adapter.
Signed-off-by: Yinglin Luan <synmyth@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
nf_bridge_alloc() always reset the skb->nf_bridge, so we should always
put the old one.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: Bart De Schuymer <bdschuym@pandora.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit f3c5c1bfd4
(netfilter: xtables: make ip_tables reentrant) forgot to
also compute the jumpstack size in the compat handlers.
Result is that "iptables -I INPUT -j userchain" turns into -j DROP.
Reported by Sebastian Roesner on #netfilter, closes
http://bugzilla.netfilter.org/show_bug.cgi?id=669.
Note: arptables change is compile-tested only.
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: David S. Miller <davem@davemloft.net>
__packed is only defined in kernel space, so we should use
__attribute__((packed)) for the code shared between kernel and user space.
Two __attribute() annotations are replaced with __attribute__() too.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
From: Bruce Allan <bruce.w.allan@intel.com>
The alternate MAC address feature is only supported by 80003ES2LAN and
82571 LOMs as well as a couple 82571 mezzanine cards. Checking for an
alternate MAC address on other parts can fail leading to the driver not
able to load. This patch limits the check for an alternate MAC address
to be done only for parts that support the feature.
This issue has been around since support for the feature was introduced
to the e1000e driver in 2.6.34.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Reported-by: Fabio Varesano <fax8@users.sourceforge.net>
Cc: stable@kernel.org
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On the e1000-devel mailing list, Nils Faerber reported latency issues with
the 82573 LOM on a ThinkPad X60. It was found to be caused by ASPM L1;
disabling it resolves the latency. The issue is present in kernels back
to 2.6.34 and possibly 2.6.33.
Reported-by: Nils Faerber <nils.faerber@kernelconcepts.de>
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Cc: stable@kernel.org
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Functions ll_temac_rx_irq and ll_temac_tx_irq
have pointer to net_device as second parameter not
pointer to temac_local.
Signed-off-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dont clear netdev->stats, it might give transient wrong values to
concurrent stat readers.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dont clear netdev->stats, it might give transient wrong values to
concurrent stat readers.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After skb is queued, its illegal to dereference it.
Cache skb->len into a temporary variable.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Spin lock rds_ring->lock is used in poll routine, so other users should
use spin_lock_bh(). While posting rx buffers from netxen_nic_attach,
rds_ring->lock is not required, so cleaning it instead of fixing it by
spin_lock_bh().
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When adding a new vlan, if the underlying interface has no carrier,
then the newly added vlan interface should also have no carrier.
At present, this is not true - the newly added vlan is added with
carrier up. Fix by checking state of real device.
Signed-off-by: Phil Oester <kernel@linuxace.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ehea: Fix synchronization between HW and SW send queue
When memory is added to / removed from a partition via the Memory DLPAR
mechanism, the eHEA driver has to do a couple of things to reflect the
memory change in its own IO address translation tables. This involves
stopping and restarting the HW queues.
During this operation, it is possible that HW and SW pointer into these
queues get out of sync. This results in a situation where packets that
are attached to a send queue are not transmitted immediately, but
delayed until further X packets have been put on the queue.
This patch detects such loss of synchronization, and resets the ehea
port when needed.
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Andre Detsch <adetsch@br.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update bnx2x version to 1.52.53-4
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
PHY locking is required between two ports for some external PHYs. Since
initialization was done in the common init function (called only on the
first port initialization) rather than in the port init function, there
was in fact no PHY locking between the ports.
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct rds_rdma_notify contains a 32 bits hole on 64bit arches,
make sure it is zeroed before copying it to user.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since
commit 1dacc76d00
Author: Johannes Berg <johannes@sipsolutions.net>
Date: Wed Jul 1 11:26:02 2009 +0000
net/compat/wext: send different messages to compat tasks
we had a race condition when setting and then
restoring frag_list. Eric attempted to fix it,
but the fix created even worse problems.
However, the original motivation I had when I
added the code that turned out to be racy is
no longer clear to me, since we only copy up
to skb->len to userspace, which doesn't include
the frag_list length. As a result, not doing
any frag_list clearing and restoring avoids
the race condition, while not introducing any
other problems.
Additionally, while preparing this patch I found
that since none of the remaining netlink code is
really aware of the frag_list, we need to use the
original skb's information for packet information
and credentials. This fixes, for example, the
group information received by compat tasks.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: stable@kernel.org [2.6.31+, for 2.6.35 revert 1235f504aa]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
"make headers_check" issued the following warning:
CHECK include/linux/netfilter (64 files)
usr/include/linux/netfilter/xt_ipvs.h:19: found __[us]{8,16,32,64} type without #include <linux/types.h>
Fix this by as suggested including linux/types.h.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
For 5000 series of devices, use long monitor timer to check
stuck tx queues.
This modification apply to all the 5000 series including 5300 and others.
Cc: stable@kernel.org [2.6.35]
Reported-by: drago01 <drago01@gmail.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Commit c96c31e499
"(drivers/net/wireless: Use wiphy_<level>)"
inadvertently changed some upper case words to
lower case. Restore the original case.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
>Xin Xiaohui wrote:
> I looked into the code dev_gro_receive(), found the code here:
> if the frags[0] is pulled to 0, then the page will be released,
> and memmove() frags left.
> Is that right? I'm not sure if memmove do right or not, but
> frags[0].size is never set after memove at least. what I think
> a simple way is not to do anything if we found frags[0].size == 0.
> The patch is as followed.
...
This version of the patch fixes the bug directly in memmove.
Reported-by: "Xin, Xiaohui" <xiaohui.xin@intel.com>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We leak at least 32bits of kernel memory to user land in tc dump,
because we dont init all fields (capab ?) of the dumped structure.
Use C99 initializers so that holes and non explicit fields are zeroed.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit 24b36f019 (netfilter: {ip,ip6,arp}_tables: dont block
bottom half more than necessary), lockdep can raise a warning
because we attempt to lock a spinlock with BH enabled, while
the same lock is usually locked by another cpu in a softirq context.
Disable again BH to avoid these lockdep warnings.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Diagnosed-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Applying the filter flags directly as done since
commit 3474ad635d
Author: Johannes Berg <johannes.berg@intel.com>
Date: Thu Apr 29 04:43:05 2010 -0700
iwlwifi: apply filter flags directly
broke 3945 under some unknown circumstances, as
reported by Alex.
Since I want to keep the direct application of
filter flags on iwlagn, duplicate the code into
both 3945 and agn and remove committing the
RXON that broke things from the 3945 version.
Cc: stable@kernel.org [2.6.35]
Reported-by: Alex Romosan <romosan@sycorax.lbl.gov>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
This reverts commit 1235f504aa.
It causes regressions worse than the problem it was trying
to fix. Eric will try to solve the problem another way.
Signed-off-by: David S. Miller <davem@davemloft.net>
sysctl output ipv6 gc_elasticity and min_adv_mss as values divided by
HZ. However, they are not in unit of jiffies, since ip6_rt_min_advmss
refers to packet size and ip6_rt_fc_elasticity is used as scaler as in
expire>>ip6_rt_gc_elasticity, so replace the jiffies conversion
handler will regular handler for them.
This has impact on scripts that are currently working assuming the
divide by HZ, will yield different results with this patch in place.
Signed-off-by: Min Zhang <mzhang@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As xfrm_compile_policy runs within a read_lock, we cannot use
GFP_KERNEL for memory allocations.
Reported-by: Luca Tettamanti <kronos.it@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
For 5000 and 6000g2b series of devices, use long monitor timer to check
stuck tx queues.
.6000g2b series device, it is WiFi/BT combo device, there are some cases,
tx queues are not move for a period of time because the WiFi/BT coex.
.5000 series device, it is being reported firmware got reload more
often than necessary, so extend the timer to avoid un-necessary reload.
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Change the name for monitor timer, also adding define for long monitor
timer; long monitor timer can be used for the type of devices require longer
time to determine the uCode is stuck on tx and needed reload.
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>
Breno Leitao
Eric Dumazet
KOSAKI Motohiro
David S. Miller
Dan Carpenter
Anton Vorontsov
Heiko Carstens
Neil Horman
Yinglin Luan
Changli Gao
Florian Westphal
Randy Dunlap
Bruce Allan
Michal Simek
Kuninori Morimoto
Amit Kumar Salecha
Phil Oester
Robert Jennings
Andre Detsch
Yaniv Rosner
Johannes Berg
Sam Ravnborg
Sachin Sanap
Wey-Yi Guy
Joe Perches
Jarek Poplawski
John W. Linville
Min Zhang
Herbert Xu