Le mardi 12 octobre 2010 à 00:02 +0200, Eric Dumazet a écrit :
> Here is the followup patch.
>
> Thanks !
>
Oops, this was an old version, the up2date ones also took care of "used"
field.
I guess its time for a sleep, sorry again.
[PATCH net-next V2] neigh: reorder struct neighbour fields
(refcnt) and (ha_lock, ha, used, dev, output, ops, primary_key) should
be placed on a separate cache lines.
refcnt can be often written, while other fields are mostly read.
This gave me good result on stress test :
before:
real 0m45.570s
user 0m15.525s
sys 9m56.669s
After:
real 0m41.841s
user 0m15.261s
sys 8m45.949s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
struct dst_ops tracks number of allocated dst in an atomic_t field,
subject to high cache line contention in stress workload.
Switch to a percpu_counter, to reduce number of time we need to dirty a
central location. Place it on a separate cache line to avoid dirtying
read only fields.
Stress test :
(Sending 160.000.000 UDP frames,
IP route cache disabled, dual E5540 @2.53GHz,
32bit kernel, FIB_TRIE, SLUB/NUMA)
Before:
real 0m51.179s
user 0m15.329s
sys 10m15.942s
After:
real 0m45.570s
user 0m15.525s
sys 9m56.669s
With a small reordering of struct neighbour fields, subject of a
following patch, (to separate refcnt from other read mostly fields)
real 0m41.841s
user 0m15.261s
sys 8m45.949s
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a seqlock in struct neighbour to protect neigh->ha[], and avoid
dirtying neighbour in stress situation (many different flows / dsts)
Dirtying takes place because of read_lock(&n->lock) and n->used writes.
Switching to a seqlock, and writing n->used only on jiffies changes
permits less dirtying.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several other ethtool functions leave heap uncleared (potentially) by
drivers. Some interfaces appear safe (eeprom, etc), in that the sizes
are well controlled. In some situations (e.g. unchecked error conditions),
the heap will remain unchanged in areas before copying back to userspace.
Note that these are less of an issue since these all require CAP_NET_ADMIN.
Cc: stable@kernel.org
Signed-off-by: Kees Cook <kees.cook@canonical.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stanse found that pch_gbe_xmit_frame uses skb after it is freed. Fix
that.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Masayuki Ohtake <masa-korg@dsn.okisemi.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stanse found that i2400m_rx frees skb, but still uses skb->len even
though it has skb_len defined. So use skb_len properly in the code.
And also define it unsinged int rather than size_t to solve
compilation warnings.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Cc: linux-wimax@intel.com
Acked-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stanse found that ia_init_one locks a spinlock and inside of that it
calls ia_start which calls:
* request_irq
* tx_init which does kmalloc(GFP_KERNEL)
Both of them can thus sleep and result in a deadlock. I don't see a
reason to have a per-device spinlock there which is used only there
and inited right before the lock location. So remove it completely.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stanse found that mpc_push frees skb and then it dereferences it. It
is a typo, new_skb should be dereferenced there.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stanse found we do in console_show:
kfree_skb(skb);
return skb->len;
which is not good. Fix that by remembering the len and use it in the
function instead.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Chas Williams <chas@cmf.nrl.navy.mil>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a new dst is used to send a frame, neigh_resolve_output() tries to
associate an struct hh_cache to this dst, calling neigh_hh_init() with
the neigh rwlock write locked.
Most of the time, hh_cache is already known and linked into neighbour,
so we find it and increment its refcount.
This patch changes the logic so that we call neigh_hh_init() with
neighbour lock read locked only, so that fast path can be run in
parallel by concurrent cpus.
This brings part of the speedup we got with commit c7d4426a98
(introduce DST_NOCACHE flag) for non cached dsts, even for cached ones,
removing one of the contention point that routers hit on multiqueue
enabled machines.
Further improvements would need to use a seqlock instead of an rwlock to
protect neigh->ha[], to not dirty neigh too often and remove two atomic
ops.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
with hardware slow in negotiation, the system did freeze
while trying to mount root on nfs at boot time.
the link state has not been initialised so network stack
tried to start transmission right away. this caused instant
retries, as the driver solely stated business upon link down,
rendering the system unusable.
notify carrier off initially to prevent transmission until
phylib will report link up.
Signed-off-by: Oskar Schirmer <oskar@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
While parsing the GetValuebyClass command frame, we could potentially write
passed the skb->data pointer.
Cc: stable@kernel.org
Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
Test whether index exceeds fw->size before reading the element
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Samuel Ortiz <samuel@sortiz.org>
sundance get_stats() should not be run concurrently, add a lock to avoid
potential losses.
Note: Remove unused rx_lock field
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simplify: ((a && b) || (!a && !b)) => (a == b)
Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Acked-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use DMA API as PCI equivalents will be deprecated. This change also
allow to allocate with GFP_KERNEL where possible.
Tested-by: Neal Becker <ndbecker2@gmail.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We have fedora bug report where driver fail to initialize after
suspend/resume because of memory allocation errors:
https://bugzilla.redhat.com/show_bug.cgi?id=629158
To fix use GFP_KERNEL allocation where possible.
Tested-by: Neal Becker <ndbecker2@gmail.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The rx->count reference is used to track reference counts to the
number of rx-queue kobjects created for the device. This patch
eliminates initialization of the counter in netif_alloc_rx_queues
and instead increments the counter each time a kobject is created.
This is now symmetric with the decrement that is done when an object is
released.
Signed-off-by: Tom Herbert <therbert@google.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are a bunch of issues that need to be fixed, including:
- GFP_KERNEL allocations from atomic context
(and GFP_ATOMIC in process context),
- abuse of the setsockopt() call convention,
- unprotected/unlocked static variables...
IMHO, we will need to alter the userspace ABI when we fix it. So mark
the configuration option as EXPERIMENTAL for the time being (or should
it be BROKEN instead?).
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current code works like this:
int garbage, status;
socklen_t len = sizeof(status);
/* enable pipe */
setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &garbage, sizeof(garbage));
/* disable pipe */
setsockopt(fd, SOL_PNPIPE, PNPIPE_DISABLE, &garbage, sizeof(garbage));
/* get status */
getsockopt(fd, SOL_PNPIPE, PNPIPE_INQ, &status, &len);
...which does not follow the usual socket option pattern. This patch
merges all three "options" into a single gettable&settable option,
before Linux 2.6.37 gets out:
int status;
socklen_t len = sizeof(status);
/* enable pipe */
status = 1;
setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, sizeof(status));
/* disable pipe */
status = 0;
setsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, sizeof(status));
/* get status */
getsockopt(fd, SOL_PNPIPE, PNPIPE_ENABLE, &status, &len);
This also fixes the error code from EFAULT to ENOTCONN.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Cc: Kumar Sanghvi <kumar.sanghvi@stericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As it currently is, the new code path is not compatible with existing
Nokia modems. This would break existing userspace for Nokia modem, such
as the existing oFono ISI driver.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch just rename all P3 #define to P3P.
Signed-off-by: Sritej Velaga <sritej.velaga@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
o Allow promiscous mode setting for VF's depending upon the configuration.
Signed-off-by: Rajesh Borundia <rajesh.borundia@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove "Flex-10" from board description.
Signed-off-by: Sritej Velaga <sritej.velaga@qlogic.com>
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
LRO was not getting enable after interface down/up.
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Put device in quiescent mode during internal loopback test.
Before running test, set state to NEED_QUISCENT. After getting
ack from all function, change state to QUISCENT and perform test.
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently fw recovery usage global workqueue.
As same workqueue used by kernel for ethtool and etc., supporting
quiescent mode is not possible, without driver private workqueue.
Signed-off-by: Amit Kumar Salecha <amit.salecha@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit "fib: RCU conversion of fib_lookup()" removed rcu_read_lock() from
__mkroute_output but left a couple of calls to rcu_read_unlock() in there.
This causes lockdep to complain that the rcu_read_unlock() call in
__ip_route_output_key causes a lock inbalance and quickly crashes the
kernel. The below fixes this for me.
Signed-off-by: Dimitris Michailidis <dm@chelsio.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Calling ETHTOOL_GRXCLSRLALL with a large rule_cnt will allocate kernel
heap without clearing it. For the one driver (niu) that implements it,
it will leave the unused portion of heap unchanged and copy the full
contents back to userspace.
Signed-off-by: Kees Cook <kees.cook@canonical.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The filter engine will time-out and ignore filters beyond
200-something hops. We also need to avoid infinite loops in
efx_filter_search() when the table is full.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Synchronise the comment with the preceding implementation change.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
setup.phone and setup.eazmsn are 32 character buffers.
rcvmsg.msg_data.byte_array is a 48 character buffer.
sc_adapter[card]->channel[rcvmsg.phy_link_no - 1].dn is 50 chars.
The rcvmsg struct comes from the memcpy_fromio() in receivemessage().
I guess that means it's data off the wire. I'm not very familiar with
this code but I don't see any reason to assume these strings are NULL
terminated.
Also it's weird that "dn" in a 50 character buffer but we only seem to
use 32 characters. In drivers/isdn/sc/scioc.h, "dn" is only a 49
character buffer. So potentially there is still an issue there.
The important thing for now is to prevent the memory corruption.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The following commit removed DISABLE_REGWRITE_BUFFER ops. The unnecessary
REGWRITE_BUFFER_FLUSH was not removed properly which is causing failure on
hw reset.
Author: Felix Fietkau <nbd@openwrt.org>
Date: Tue Oct 5 12:03:42 2010 +0200
ath9k_hw: clean up register write buffering
Signed-off-by: Rajkumar Manoharan <rmanoharan@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
There's no need for the WDS peer address
to not be const, so make it const.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bill Jordan's patch to allow setting the WDS
peer crossed with my patch removing all the
boilerplate code in nl80211, and consequently
he didn't make use of it yet. Fix that.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
The recent scan overhaul broke locking
because now we can jump to code that
attempts to unlock, while we don't have
the mutex held. Fix this by holding the
mutex around all the relevant code.
Reported-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
In the NVS/INI-parameters for the driver/firmware there is the FEM
(front end module) selection option, which may configure the hardware FEM
via autodetection or manual selection. So far, there has been support for
manual selection only.
Add support for FEM autodetection.
Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
