Alexander Duyck [Fri, 21 Nov 2008 05:05:08 +0000 (21:05 -0800)]
DCB: Add interface to query for the DCB capabilities of an device.
Adds to the netlink interface for Data Center Bridging (DCB), allowing
the DCB capabilities supported by a device to be queried.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Alexander Duyck [Fri, 21 Nov 2008 04:52:10 +0000 (20:52 -0800)]
ixgbe: this patch adds support for DCB to the kernel and ixgbe driver
This adds support for Data Center Bridging (DCB) features in the ixgbe
driver and adds an rtnetlink interface for configuring DCB to the
kernel. The DCB feature support included are Priority Grouping (PG) -
which allows bandwidth guarantees to be allocated to groups to traffic
based on the 802.1q priority, and Priority Based Flow Control (PFC) -
which introduces a new MAC control PAUSE frame which works at
granularity of the 802.1p priority instead of the link (IEEE 802.3x).
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Convert to network device ops. Needed to change to directly call
the init routine since two sides share same ops. In the process
found by inspection a device ref count leak if register_netdevice failed.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch moves neigh_setup and hard_start_xmit into the network device ops
structure. For bisection, fix all the previously converted drivers as well.
Bonding driver took the biggest hit on this.
Added a prefetch of the hard_start_xmit in the fast path to try and reduce
any impact this would have.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
USB: ACE1001 patch for cp2101.c
USB: usbmon: fix read(2)
USB: gadget rndis: send notifications
USB: gadget rndis: stop windows self-immolation
USB: storage: update unusual_devs entries for Nokia 5300 and 5310
USB: storage: updates unusual_devs entry for the Nokia 6300
usb: musb: fix bug in musb_schedule
USB: fix SB700 usb subsystem hang bug
pv_cpu_ops.getreg(_IA64_REG_IP) returned constant.
But the returned ip valued should be the one in the caller, not of the callee.
This patch fixes that.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp> Signed-off-by: Tony Luck <tony.luck@intel.com>
Hidetoshi Seto [Mon, 17 Nov 2008 01:18:08 +0000 (10:18 +0900)]
[IA64] use mprintk instead of printk, in ia64_mca_modify_original_stack
Using printk from MCA/INIT context is unsafe since it can cause deadlock.
The ia64_mca_modify_original_stack is called from both of mca handler and
init handler, so it should use mprintk instead of printk.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
Tony Luck [Thu, 20 Nov 2008 21:27:12 +0000 (13:27 -0800)]
[IA64] Rationalize kernel mode alignment checking
Itanium processors can handle some misaligned data accesses. They
also provide a mode where all such accesses are forced to trap. The
kernel was schizophrenic about use of this mode:
* Base kernel code ran in permissive mode where the only traps
generated were from those cases that the h/w could not handle.
* Interrupt, syscall and trap code ran in strict mode where all
unaligned accesses caused traps to the 0x5a00 unaligned reference
vector.
Use strict alignment checking throughout the kernel, but make
sure that we continue to let user mode use more relaxed mode
as the default.
Matthew Wilcox [Thu, 20 Nov 2008 21:09:33 +0000 (14:09 -0700)]
x86: Fix interrupt leak due to migration
When we migrate an interrupt from one CPU to another, we set the
move_in_progress flag and clean up the vectors later once they're not
being used. If you're unlucky and call destroy_irq() before the vectors
become un-used, the move_in_progress flag is never cleared, which causes
the interrupt to become unusable.
This was discovered by Jesse Brandeburg for whom it manifested as an
MSI-X device refusing to use MSI-X mode when the driver was unloaded
and reloaded repeatedly.
Signed-off-by: Matthew Wilcox <willy@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Trond Myklebust [Thu, 20 Nov 2008 21:06:21 +0000 (16:06 -0500)]
SUNRPC: Fix a performance regression in the RPC authentication code
Fix a regression reported by Max Kellermann whereby kernel profiling
showed that his clients were spending 45% of their time in
rpcauth_lookup_credcache.
It turns out that although his processes had identical uid/gid/groups,
generic_match() was failing to detect this, because the task->group_info
pointers were not shared. This again lead to the creation of a huge number
of identical credentials at the RPC layer.
The regression is fixed by comparing the contents of task->group_info
if the actual pointers are not identical.
* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
[CIFS] Do not attempt to close invalidated file handles
[CIFS] fix check for dead tcon in smb_init
Linus Torvalds [Thu, 20 Nov 2008 21:13:48 +0000 (13:13 -0800)]
Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
MIPS: csrc-r4k: Fix declaration depending on the wrong CONFIG_ symbol.
MIPS: csrc-r4k: Fix spelling mistake.
MIPS: RB532: Provide functions for gpio configuration
MIPS: IP22: Make indy_sc_ops variable static
MIPS: RB532: GPIO register offsets are relative to GPIOBASE
MIPS: Malta: Fix include paths in malta-amon.c
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (23 commits)
net: fix tiny output corruption of /proc/net/snmp6
atl2: don't request irq on resume if netif running
ipv6: use seq_release_private for ip6mr.c /proc entries
pkt_sched: fix missing check for packet overrun in qdisc_dump_stab()
smc911x: Fix printf format typo in smc911x driver.
asix: Fix asix-based cards connecting to 10/100Mbs LAN.
mv643xx_eth: fix recycle check bound
mv643xx_eth: fix the order of mdiobus_{unregister, free}() calls
sh: sh_eth: Update to change of mii_bus
TPROXY: supply a struct flowi->flags argument in inet_sk_rebuild_header()
TPROXY: fill struct flowi->flags in udp_sendmsg()
net: ipg.c fix bracing on endian swapping
phylib: Fix auto-negotiation restart avoidance
net: jme.c rxdesc.flags is __le16, other missing endian swaps
phylib: fix phy name example in documentation
net: Do not fire linkwatch events until the device is registered.
phonet: fix compilation with gcc-3.4
ixgbe: fix compilation with gcc-3.4
pktgen: fix multiple queue warning
net: fix ip_mr_init() error path
...
Linus Torvalds [Thu, 20 Nov 2008 21:09:32 +0000 (13:09 -0800)]
Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: uaccess_64: fix return value in __copy_from_user()
x86: quirk for reboot stalls on a Dell Optiplex 330
Helge Deller [Thu, 20 Nov 2008 09:54:09 +0000 (10:54 +0100)]
parisc: fix bug in compat_arch_ptrace
Commit 81e192d6ce303b6792aa38ff35f41a1a7357f23a ("parisc: convert to
generic compat_sys_ptrace") introduced a bug which segfaults the parisc
64bit kernel when stracing 32bit applications:
The problem is that compat_arch_ptrace() enters with an addr value of
type compat_ulong_t and calls translate_usr_offset() to translate the
address offset into a struct pt_regs offset like this:
addr = translate_usr_offset(addr)
this means that any return value of translate_usr_offset() is stored
back as compat_ulong_t type into the addr variable.
But since translate_usr_offset() returns -1 for invalid offsets, addr
can now get the value 0xffffffff which then fails the next return-value
sanity check and thus the kernel tries to access invalid memory:
if (addr < 0)
break;
Fix this bug by modifying translate_usr_offset() to take and return
values of type compat_ulong_t, and by returning the value
"sizeof(struct pt_regs)" as an error indicator.
Additionally change the sanity check to check for return values
for >= sizeof(struct pt_regs).
Steve French [Thu, 20 Nov 2008 20:00:44 +0000 (20:00 +0000)]
[CIFS] Do not attempt to close invalidated file handles
If a connection with open file handles has gone down
and come back up and reconnected without reopening
the file handle yet, do not attempt to send an SMB close
request for this handle in cifs_close. We were
checking for the connection being invalid in cifs_close
but since the connection may have been reconnected
we also need to check whether the file handle
was marked invalid (otherwise we could close the
wrong file handle by accident).
Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
Phil Sutter [Sat, 1 Nov 2008 14:13:21 +0000 (15:13 +0100)]
MIPS: RB532: Provide functions for gpio configuration
As gpiolib doesn't support pin multiplexing, it provides no way to
access the GPIOFUNC register. Also there is no support for setting
interrupt status and level. These functions provide access to them and
are needed by the CompactFlash driver.
Signed-off-by: Phil Sutter <n0-1@freewrt.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Florian Fainelli [Fri, 31 Oct 2008 13:24:29 +0000 (14:24 +0100)]
MIPS: RB532: GPIO register offsets are relative to GPIOBASE
This patch fixes the wrong use of GPIO register offsets
in devices.c. To avoid further problems, use gpio_get_value
to return the NAND status instead of our own expanded code.
Also define the zero offset of the alternate function register to allow
consistent access.
Wang Chen [Thu, 20 Nov 2008 12:25:06 +0000 (04:25 -0800)]
netdevice ni65: Convert directly reference of netdev->priv
This patch is to kill directly reference of netdev->priv too.
Because the private data needs special memory: lower 16MB DMA.
alloc_etherdev() can not do this work.
So we can't use netdev->priv to point to netdev's private data.
Use netdev->ml_priv instead.
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
LAN9500 supports tx checksum offload, which slightly decreases cpu
utilisation. The benefit isn't very large because we still require
the skb to be linearized, but it does save a few cycles.
This patch adds support for it, and enables it by default.
Signed-off-by: Steve Glendinning <steve.glendinning@smsc.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Alan Jenkins [Thu, 20 Nov 2008 12:18:25 +0000 (04:18 -0800)]
atl2: don't request irq on resume if netif running
If the device is suspended with the cable disconnected, then
resumed with the cable connected, dev->open is called before
resume. During resume, we request an IRQ, but the IRQ was
already assigned during dev->open, resulting in the warning
shown below.
Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk> Signed-off-by: Jay Cliburn <jcliburn@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Hutchings [Thu, 20 Nov 2008 12:17:42 +0000 (04:17 -0800)]
sfc: Fix dependency for SFC_MTD
Randy Dunlap found that SFC_MTD was selected when sfc was built-in and
the MTD core was a module. Don't allow that combination.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Benjamin Thery [Thu, 20 Nov 2008 12:16:12 +0000 (04:16 -0800)]
ipv6: use seq_release_private for ip6mr.c /proc entries
In ip6mr.c, /proc entries /proc/net/ip6_mr_cache and /proc/net/ip6_mr_vif
are opened with seq_open_private(), thus seq_release_private() should be
used to release them.
Should fix a small memory leak.
Signed-off-by: Benjamin Thery <benjamin.thery@bull.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Wed, 19 Nov 2008 08:03:09 +0000 (08:03 +0000)]
pkt_sched: remove unnecessary xchg() in packet classifiers
The use of xchg() hasn't been necessary since 2.2.something when proper
locking was added to packet schedulers. In the case of classifiers they
mostly weren't even necessary before that since they're mainly used
to assign a NULL pointer to the filter root in the ->destroy path;
the root is destroyed immediately after that.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Patrick McHardy [Thu, 20 Nov 2008 12:10:00 +0000 (04:10 -0800)]
pkt_sched: add DRR scheduler
Add classful DRR scheduler as a more flexible replacement for SFQ.
The main difference to the algorithm described in "Efficient Fair Queueing
using Deficit Round Robin" is that this implementation doesn't drop packets
from the longest queue on overrun because its classful and limits are
handled by each individual child qdisc.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Wang Chen [Thu, 20 Nov 2008 12:05:56 +0000 (04:05 -0800)]
netdevice lance: Convert directly reference of netdev->priv
This patch is to kill directly reference of netdev->priv too.
Because the private data should be allocated in DMA area, alloc_etherdev()
can't satisfy this needs.
Use netdev->ml_priv to point to lance_private.
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, if multiple unicast addresses are programmed into a
mv643xx_eth interface, the core networking will resort to enabling
promiscuous mode on the interface, as mv643xx_eth does not implement
->set_rx_mode().
This patch switches mv643xx_eth over from ->set_multicast_list()
to ->set_rx_mode(), and implements support for secondary unicast
addresses. The hardware can handle multiple unicast addresses as
long as their first 11 nibbles are the same (i.e. are of the form
xx:xx:xx:xx:xx:xy where the x part is the same for all addresses), so
if that is the case, we use that mode. If it's not the case, we enable
unicast promiscuous mode in the hardware, which is slightly better than
enabling promiscuous mode for multicasts as well, which is what would
happen before.
While we are at it, change the programming sequence so that we
don't clear all filter bits first, so we don't lose all incoming
packets while the filter is being reprogrammed.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Since txq_alloc_desc_index() is a very simple function, and since
descriptor ring index handling for transmit reclaim, receive
processing and receive refill is already handled inline as well,
inline txq_alloc_desc_index() into its two call sites.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
By giving the per-port 'struct mv643xx_eth_private' its own
'void __iomem *base' pointer that points to the per-port register
area, we can get rid of both the double indirection and the << 10
that is done for every per-port register access -- this patch does
that.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes sparse warnings:
net/ipv4/ip_sockglue.c:146:15: warning: incorrect type in assignment (different base types)
net/ipv4/ip_sockglue.c:146:15: expected restricted __be16 [assigned] [usertype] sin_port
net/ipv4/ip_sockglue.c:146:15: got unsigned short [unsigned] [short] [usertype] <noident>
net/ipv4/ip_sockglue.c:130:6: warning: symbol 'ip_cmsg_recv_dstaddr' was not declared. Should it be static?
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
asix: Fix asix-based cards connecting to 10/100Mbs LAN.
Add AX_MEDIUM_ENCK also when speed = 10/100Mbps. This allows my belkin f5d5055 to work with my 100Mbps switch and with an old 10Mbps ISA card.
Without this patch, the card is recognized and the interface is brought
up fine, but no packets actually flow through the interface.
Signed-off-by: Pantelis Koukousoulas <pktoss@gmail.com> Acked-by: David Hollis <dhollis@davehollis.com> Signed-off-by: David S. Miller <davem@davemloft.net>
When mv643xx_eth allocates skbuffs, it adds
'dma_get_cache_alignment() - 1' to the length it needs, so that it can
align the skb's ->data pointer to a cache boundary. When checking
whether a transmitted skbuff can be reused as a receive buffer, these
bytes needs to be included into the minimum bound for the recycle check.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Frysinger [Mon, 17 Nov 2008 21:23:40 +0000 (21:23 +0000)]
netdev/smc91x: unify Blackfin code a bit and use proper read/write functions
Signed-off-by: Mike Frysinger <vapier.adi@gmail.com> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Balazs Scheidler [Thu, 20 Nov 2008 09:08:06 +0000 (01:08 -0800)]
TPROXY: supply a struct flowi->flags argument in inet_sk_rebuild_header()
inet_sk_rebuild_header() does a new route lookup if the dst_entry
associated with a socket becomes stale. However inet_sk_rebuild_header()
didn't use struct flowi->flags, causing the route lookup to
fail for foreign-bound IP_TRANSPARENT sockets, causing an error
state to be set for the sockets in question.
Signed-off-by: Balazs Scheidler <bazsi@balabit.hu> Signed-off-by: David S. Miller <davem@davemloft.net>
Balazs Scheidler [Thu, 20 Nov 2008 09:07:24 +0000 (01:07 -0800)]
TPROXY: fill struct flowi->flags in udp_sendmsg()
udp_sendmsg() didn't fill struct flowi->flags, which means that
the route lookup would fail for non-local IPs even if the
IP_TRANSPARENT sockopt was set.
This prevents sendto() to work properly for UDP sockets, whereas
bind(foreign-ip) + connect() + send() worked fine.
Signed-off-by: Balazs Scheidler <bazsi@balabit.hu> Signed-off-by: David S. Miller <davem@davemloft.net>
Wang Chen [Thu, 20 Nov 2008 09:02:05 +0000 (01:02 -0800)]
netdevice hamradio: Convert directly reference of netdev->priv
Since all the memory, which pointed by netdev->priv, are allocated in
advance instead of by alloc_netdev(). Use netdev->ml_priv to point to
those memory.
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
filter: add SKF_AD_NLATTR_NEST to look for nested attributes
SKF_AD_NLATTR allows us to find the first matching attribute in a
stream of netlink attributes from one offset to the end of the
netlink message. This is not suitable to look for a specific
matching inside a set of nested attributes.
For example, in ctnetlink messages, if we look for the CTA_V6_SRC
attribute in a message that talks about an IPv4 connection,
SKF_AD_NLATTR returns the offset of CTA_STATUS which has the same
value of CTA_V6_SRC but outside the nest. To differenciate
CTA_STATUS and CTA_V6_SRC, we would have to make assumptions on the
size of the attribute and the usual offset, resulting in horrible
BSF code.
This patch adds SKF_AD_NLATTR_NEST, which is a variant of
SKF_AD_NLATTR, that looks for an attribute inside the limits of
a nested attributes, but not further.
This patch validates that we have enough room to look for the
nested attributes - based on a suggestion from Patrick McHardy.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Alexander Duyck [Thu, 20 Nov 2008 08:48:10 +0000 (00:48 -0800)]
igb: Fix tx/rx_ring_count parameters for igb on suspend/resume/ring resize
When suspending the device the ring structure is freed which causes it to
loose track of the count. To resolve this we need to move the ring count
outside of the ring structure and store it in the adapter struct.
In addition to resolving the suspend/resume issue this patch also addresses
issues seen in the event of memory allocation errors causing uneven ring
sizes on multiple queues.
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>
Alexander Duyck [Thu, 20 Nov 2008 08:47:34 +0000 (00:47 -0800)]
igb: simplify swap in clean_rx_irq if using packet split
This update replaces the xchg calls that were added with a pair of
assignments as there is no need for the xchg calls and they were found to
cause issues on some architectures.
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>
Alexander Duyck [Thu, 20 Nov 2008 08:46:57 +0000 (00:46 -0800)]
igb: update name to reflect new hardware
This patch adds the 82576 device to the description for igb in Kconfig.
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>
Convert to new network device ops interface. Slight additional complexity
here because the second port does not allow netpoll and therefore has
different virtual function table.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Convert to net_device_ops table.
Note: for some operations move error checking into generic networking
layer (rather than looking at pointers in bonding).
A couple of gratituous style cleanups to get rid of extra {}
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Code Review / linux-2.6.git / log