If segmentation offload is enabled by the host, we currently allocate
maximum sized packet buffers and pass them to the host. This uses up
20 ring entries, allowing us to supply only 20 packet buffers to the
host with a 256 entry ring. This is a huge overhead when receiving
small packets, and is most keenly felt when receiving MTU sized
packets from off-host.
The VIRTIO_NET_F_MRG_RXBUF feature flag is set by hosts which support
using receive buffers which are smaller than the maximum packet size.
In order to transfer large packets to the guest, the host merges
together multiple receive buffers to form a larger logical buffer.
The number of merged buffers is returned to the guest via a field in
the virtio_net_hdr.
Make use of this support by supplying single page receive buffers to
the host. On receive, we extract the virtio_net_hdr, copy 128 bytes of
the payload to the skb's linear data buffer and adjust the fragment
offset to point to the remaining data. This ensures proper alignment
and allows us to not use any paged data for small packets. If the
payload occupies multiple pages, we simply append those pages as
fragments and free the associated skbs.
This scheme allows us to be efficient in our use of ring entries
while still supporting large packets. Benchmarking using netperf from
an external machine to a guest over a 10Gb/s network shows a 100%
improvement from ~1Gb/s to ~2Gb/s. With a local host->guest benchmark
with GSO disabled on the host side, throughput was seen to increase
from 700Mb/s to 1.7Gb/s.
Based on a patch from Herbert Xu.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (use netdev_priv)
Signed-off-by: David S. Miller <davem@davemloft.net>
Seems like an oversight that we have set-tx-csum and set-sg hooked
up, but not set-tso.
Also leads to the strange situation that if you e.g. disable tx-csum,
then tso doesn't get disabled.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Each time we re-fill the recv queue with buffers, we allocate
one too many skbs and free it again when adding fails. We should
recycle the pages allocated in this case.
A previous version of this patch made trim_pages() trim trailing
unused pages from skbs with some paged data, but this actually
caused a barely measurable slowdown.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (use netdev_priv)
Signed-off-by: David S. Miller <davem@davemloft.net>
Technically, patch changes format for modules, but I think nobody cares.
-86dd :ipv6:ipv6_rcv+0x0
+86dd ipv6_rcv+0x0/0x400 [ipv6]
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As found in the past (commit f1dd9c379c
[NET]: Fix tbench regression in 2.6.25-rc1), it is really
important that struct dst_entry refcount is aligned on a cache line.
We cannot use __atribute((aligned)), so manually pad the structure
for 32 and 64 bit arches.
for 32bit : offsetof(truct dst_entry, __refcnt) is 0x80
for 64bit : offsetof(truct dst_entry, __refcnt) is 0xc0
As it is not possible to guess at compile time cache line size,
we use a generic value of 64 bytes, that satisfies many current arches.
(Using 128 bytes alignment on 64bit arches would waste 64 bytes)
Add a BUILD_BUG_ON to catch future updates to "struct dst_entry" dont
break this alignment.
"tbench 8" is 4.4 % faster on a dual quad core (HP BL460c G1), Intel E5450 @3.00GHz
(2350 MB/s instead of 2250 MB/s)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adds Documentation/RCU/rculist_nulls.txt file to describe how 'nulls'
end-of-list can help in some RCU algos.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>
RCU was added to UDP lookups, using a fast infrastructure :
- sockets kmem_cache use SLAB_DESTROY_BY_RCU and dont pay the
price of call_rcu() at freeing time.
- hlist_nulls permits to use few memory barriers.
This patch uses same infrastructure for TCP/DCCP established
and timewait sockets.
Thanks to SLAB_DESTROY_BY_RCU, no slowdown for applications
using short lived TCP connections. A followup patch, converting
rwlocks to spinlocks will even speedup this case.
__inet_lookup_established() is pretty fast now we dont have to
dirty a contended cache line (read_lock/read_unlock)
Only established and timewait hashtable are converted to RCU
(bind table and listen table are still using traditional locking)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a straightforward patch, using hlist_nulls infrastructure.
RCUification already done on UDP two weeks ago.
Using hlist_nulls permits us to avoid some memory barriers, both
at lookup time and delete time.
Patch is large because it adds new macros to include/net/sock.h.
These macros will be used by TCP & DCCP in next patch.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hlist uses NULL value to finish a chain.
hlist_nulls variant use the low order bit set to 1 to signal an end-of-list marker.
This allows to store many different end markers, so that some RCU lockless
algos (used in TCP/UDP stack for example) can save some memory barriers in
fast paths.
Two new files are added :
include/linux/list_nulls.h
- mimics hlist part of include/linux/list.h, derived to hlist_nulls variant
include/linux/rculist_nulls.h
- mimics hlist part of include/linux/rculist.h, derived to hlist_nulls variant
Only four helpers are declared for the moment :
hlist_nulls_del_init_rcu(), hlist_nulls_del_rcu(),
hlist_nulls_add_head_rcu() and hlist_nulls_for_each_entry_rcu()
prefetches() were removed, since an end of list is not anymore NULL value.
prefetches() could trigger useless (and possibly dangerous) memory transactions.
Example of use (extracted from __udp4_lib_lookup())
struct sock *sk, *result;
struct hlist_nulls_node *node;
unsigned short hnum = ntohs(dport);
unsigned int hash = udp_hashfn(net, hnum);
struct udp_hslot *hslot = &udptable->hash[hash];
int score, badness;
rcu_read_lock();
begin:
result = NULL;
badness = -1;
sk_nulls_for_each_rcu(sk, node, &hslot->head) {
score = compute_score(sk, net, saddr, hnum, sport,
daddr, dport, dif);
if (score > badness) {
result = sk;
badness = score;
}
}
/*
* if the nulls value we got at the end of this lookup is
* not the expected one, we must restart lookup.
* We probably met an item that was moved to another chain.
*/
if (get_nulls_value(node) != hash)
goto begin;
if (result) {
if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
result = NULL;
else if (unlikely(compute_score(result, net, saddr, hnum, sport,
daddr, dport, dif) < badness)) {
sock_put(result);
goto begin;
}
}
rcu_read_unlock();
return result;
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>
In case UDP traffic is redirected to a local UDP socket,
the originally addressed destination address/port
cannot be recovered with the in-kernel tproxy.
This patch adds an IP_RECVORIGDSTADDR sockopt that enables
a IP_ORIGDSTADDR ancillary message in recvmsg(). This
ancillary message contains the original destination address/port
of the packet being received.
Signed-off-by: Balazs Scheidler <bazsi@balabit.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Greear wrote:
> I have 500 mac-vlans on a system talking to 500 other
> mac-vlans. My problem is that the arp-table gets extremely
> huge because every time an arp-request comes in on all mac-vlans,
> a stale arp entry is added for each mac-vlan. I have filtering
> turned on, but that doesn't help because the neigh_event_ns call
> below will cause a stale neighbor entry to be created regardless
> of whether a replay will be sent or not.
> Maybe the neigh_event code should be below the checks for dont_send,
> and only create check neigh_event_ns if we are !dont_send?
The attached patch makes it work much better for me. The patch
will cause the code to NOT create a stale neighbor entry if we
are not going to respond to the ARP request. The old code
*would* create a stale entry even if we are not going to respond.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use a similar approach to the SMB session sharing. Add a list of tcons
attached to each SMB session. Move the refcount to non-atomic. Protect
all of the above with the cifs_tcp_ses_lock. Add functions to
properly find and put references to the tcons.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
This change enables ECC correction for the packet buffer on all 82571
silicon.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some users reported that they have machines with BMCs enabled that cannot
receive IPMI traffic after e1000e is loaded.
http://marc.info/?l=e1000-devel&m=121909039127414&w=2http://marc.info/?l=e1000-devel&m=121365543823387&w=2
This fixes the issue if they load with the new parameter = 0 by disabling
crc stripping, but leaves the performance feature on for most users.
Based on work done by Hong Zhang.
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the driver fails to initialize the first time due to the failure in the
phy_id check the kernel triggers a warn_on on the second try to load the
driver because the driver did not free the msi/x resources in the first
load because of the previous failure in phy_id check.
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
make mdio-gpio work with non OpenFirmware gpio implementation.
Aditional changes to mdio-gpio:
- use gpio_request() and gpio_free()
- place irq[] array in struct mdio_gpio_info
- add module description, author and license
- add note about compiling this driver as module
- rename mdc and mdio function (were ugly names)
- change MII to MDIO in bus name
- add __init __exit to module (un)loading functions
- probe fails if no phys added to the bus
- kzalloc bitbang with sizeof(*bitbang)
Changes since v3:
- keep bus naming "%x" to be compatible with existing drivers.
Changes since v2:
- more #ifdefs reduction
- platform driver will be registered on OF platforms also
- unified platform and OF bus_id to phy%i
Changes since v1:
- removed NO_IRQ
- reduced #idefs
Laurent, please test this driver under OF.
Signed-off-by: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix an unitialized return value when compiling on parisc (with CONFIG_UNEVICTABLE_LRU=y):
mm/mlock.c: In function `__mlock_vma_pages_range':
mm/mlock.c:165: warning: `ret' might be used uninitialized in this function
Signed-off-by: Helge Deller <deller@gmx.de>
[ It isn't ever really used uninitialized, since no caller should ever
call this function with an empty range. But the compiler is correct
that from a local analysis standpoint that is impossible to see, and
fixing the warning is appropriate. ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Bug #11989: Suspend failure on NForce4-based boards due to chanes in
stop_machine
We should not access active.fnret outside the lock; in theory the next
stop_machine could overwrite it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Tested-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Reported by Stephen Rothwell:
drivers/net/dm9000.c:1450: error: expected ')' before ';' token
drivers/net/dm9000.c:1455: error: expected ';' before '}' token
Signed-off-by: David S. Miller <davem@davemloft.net>
The vendor BSP used for the WM8350 development provided an I2C driver
which incorrectly returned zero on succesful sends rather than the
number of transmitted bytes, an error which was then propagated into the
WM8350 I2C accessors.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
Commit 0794469da3: ("ACPI: struct device -
replace bus_id with dev_name(), dev_set_name()") introduced a bug by
testing 'dev_name(ldev)' instead of 'ldev->bus' for NULL when printing
out the bus information.
So if ldev->bus was NULL, we'd oops.
Reported-and-tested-by: Bruno Prémont <bonbons@linux-vserver.org>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
PHYID returns 0xffff and not 0xffffffff when not found and in some
case(at91sam9263) 0x0. Maybe this patch could be useful.
Signed-off-by: Giulio Benetti <giulio.benetti@micronovasrl.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The e100 driver triggers BUG_ON(buf->direction != dir)
by doing pci_map_single(..., PCI_DMA_BIDIRECTIONAL)
and pci_dma_sync_single_for_device(..., PCI_DMA_TODEVICE).
Changing the DMA direction, especially with dmabounce will result
in unexpected behaviour.
Reported-by: Anders Grafstrom <grfstrm@users.sourceforge.net>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use dev_printk() instead of printk() to give a little more context
and use consistent format.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Removed debug print statements and improved conditionals around informational statements.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since dev->power.should_wakeup bit is used by the PCI core to
decide whether the device should wake up the system from sleep
states, set/unset this bit whenever WOL is enabled/disabled using
igb_set_wol(). Accordingly, use device_can_wakeup() for checking
if wake-up is supported by the device.
Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since dev->power.should_wakeup bit is used by the PCI core to
decide whether the device should wake up the system from sleep
states, set/unset this bit whenever WOL is enabled/disabled using
e1000_set_wol(). Accordingly, use device_can_wakeup() for checking
if wake-up is supported by the device.
Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since dev->power.should_wakeup bit is used by the PCI core to
decide whether the device should wake up the system from sleep
states, set/unset this bit whenever WOL is enabled/disabled using
e1000_set_wol(). Accordingly, use device_can_wakeup() for checking
if wake-up is supported by the device.
Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Impact: fix es7000 build
CC arch/x86/kernel/es7000_32.o
arch/x86/kernel/es7000_32.c: In function find_unisys_acpi_oem_table:
arch/x86/kernel/es7000_32.c:255: error: implicit declaration of function acpi_get_table_with_size
arch/x86/kernel/es7000_32.c:261: error: implicit declaration of function early_acpi_os_unmap_memory
arch/x86/kernel/es7000_32.c: In function unmap_unisys_acpi_oem_table:
arch/x86/kernel/es7000_32.c:277: error: implicit declaration of function __acpi_unmap_table
make[1]: *** [arch/x86/kernel/es7000_32.o] Error 1
we applied one patch out of order...
| commit a73aaedd95
| Author: Yinghai Lu <yhlu.kernel@gmail.com>
| Date: Sun Sep 14 02:33:14 2008 -0700
|
| x86: check dsdt before find oem table for es7000, v2
|
| v2: use __acpi_unmap_table()
that patch need:
x86: use early_ioremap in __acpi_map_table
x86: always explicitly map acpi memory
acpi: remove final __acpi_map_table mapping before setting acpi_gbl_permanent_mmap
acpi/x86: introduce __apci_map_table, v4
submitted to the ACPI tree but not upstream yet.
fix it until those patches applied, need to revert this one
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make pegasus driver not allocate a workqueue until the driver
is bound to some device, which will need that workqueue if
the device is brought up. This conserves resources when the
driver is linked but there's no pegasus device connected.
Also shrink the runtime footprint a smidgeon by moving some
init-only code into its proper section, and move an obnoxious
(frequent and meaningless) message to be debug-only.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
netif_carrier_off() is sufficient to stop Tx into the driver. Stopping the Tx
queues is redundant and unnecessary. By the same token, netif_carrier_on()
will be sufficient to re-enable Tx, so waking the queues is unnecessary.
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: David S. Miller <davem@davemloft.net>
Impact: make output of available_filter_functions complete
phenomenon:
The first value of dyn_ftrace_total_info is not equal with
`cat available_filter_functions | wc -l`, but they should be equal.
root cause:
When printing functions with seq_printf in t_show, if the read buffer
is just overflowed by current function record, then this function
won't be printed to user space through read buffer, it will
just be dropped. So we can't see this function printing.
So, every time the last function to fill the read buffer, if overflowed,
will be dropped.
This also applies to set_ftrace_filter if set_ftrace_filter has
more bytes than read buffer.
fix:
Through checking return value of seq_printf, if less than 0, we know
this function doesn't be printed. Then we decrease position to force
this function to be printed next time, in next read buffer.
Another little fix is to show correct allocating pages count.
Signed-off-by: walimis <walimisdev@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix a problem where ds_request() returned an error without releasing the
ds lock.
Reported-by: Stephane Eranian <eranian@gmail.com>
Signed-off-by: Markus Metzger <markus.t.metzger@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid:
HID: don't grab devices with no input
HID: fix radio-mr800 hidquirks
HID: fix kworld fm700 radio hidquirks
HID: fix start/stop cycle in usbhid driver
HID: use single threaded work queue for hid_compat
HID: map macbook keys for "Expose" and "Dashboard"
HID: support for new unibody macbooks
HID: fix locking in hidraw_open()
This patch will add support for the Marvell 88E1118 PHY which supports gigabit ethernet among other things.
Signed-off-by: Ron Madrid <ron_madrid@sbcglobal.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before the change the driver reported the same pause parameters
for all the ports, even only one of them was modified.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
Inotify watch removals suck violently.
To kick the watch out we need (in this order) inode->inotify_mutex and
ih->mutex. That's fine if we have a hold on inode; however, for all
other cases we need to make damn sure we don't race with umount. We can
*NOT* just grab a reference to a watch - inotify_unmount_inodes() will
happily sail past it and we'll end with reference to inode potentially
outliving its superblock.
Ideally we just want to grab an active reference to superblock if we
can; that will make sure we won't go into inotify_umount_inodes() until
we are done. Cleanup is just deactivate_super().
However, that leaves a messy case - what if we *are* racing with
umount() and active references to superblock can't be acquired anymore?
We can bump ->s_count, grab ->s_umount, which will almost certainly wait
until the superblock is shut down and the watch in question is pining
for fjords. That's fine, but there is a problem - we might have hit the
window between ->s_active getting to 0 / ->s_count - below S_BIAS (i.e.
the moment when superblock is past the point of no return and is heading
for shutdown) and the moment when deactivate_super() acquires
->s_umount.
We could just do drop_super() yield() and retry, but that's rather
antisocial and this stuff is luser-triggerable. OTOH, having grabbed
->s_umount and having found that we'd got there first (i.e. that
->s_root is non-NULL) we know that we won't race with
inotify_umount_inodes().
So we could grab a reference to watch and do the rest as above, just
with drop_super() instead of deactivate_super(), right? Wrong. We had
to drop ih->mutex before we could grab ->s_umount. So the watch
could've been gone already.
That still can be dealt with - we need to save watch->wd, do idr_find()
and compare its result with our pointer. If they match, we either have
the damn thing still alive or we'd lost not one but two races at once,
the watch had been killed and a new one got created with the same ->wd
at the same address. That couldn't have happened in inotify_destroy(),
but inotify_rm_wd() could run into that. Still, "new one got created"
is not a problem - we have every right to kill it or leave it alone,
whatever's more convenient.
So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
"grab it and kill it" check. If it's been our original watch, we are
fine, if it's a newcomer - nevermind, just pretend that we'd won the
race and kill the fscker anyway; we are safe since we know that its
superblock won't be going away.
And yes, this is far beyond mere "not very pretty"; so's the entire
concept of inotify to start with.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Greg KH <greg@kroah.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The file(s) below do not use LINUX_VERSION_CODE nor KERNEL_VERSION.
drivers/hwmon/lis3lv02d.c
This patch removes the said #include <version.h>.
Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'sh/for-2.6.28' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
serial: sh-sci: Reorder the SCxTDR write after the TDxE clear.
sh: __copy_user function can corrupt the stack in case of exception
sh: Fixed the TMU0 reload value on resume
sh: Don't factor in PAGE_OFFSET for valid_phys_addr_range() check.
sh: early printk port type fix
i2c: fix i2c-sh_mobile rx underrun
sh: Provide a sane valid_phys_addr_range() to prevent TLB reset with PMB.
usb: r8a66597-hcd: fix wrong data access in SuperH on-chip USB
fix sci type for SH7723
serial: sh-sci: fix cannot work SH7723 SCIFA
sh: Handle fixmap TLB eviction more coherently.
