Dan Williams [Wed, 11 Feb 2009 18:26:06 +0000 (13:26 -0500)]
ipw2x00: age scan results on resume
Scanned BSS entries are timestamped with jiffies, which doesn't
increment across suspend and hibernate. On resume, every BSS in the
scan list looks like it was scanned within the last 10 seconds,
irregardless of how long the machine was actually asleep. Age scan
results on resume with the time spent during sleep so userspace has a
clue how old they really are.
Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 18 Feb 2009 18:32:08 +0000 (19:32 +0100)]
mac80211/cfg80211: move iwrange handler to cfg80211
The previous patch made cfg80211 generally aware of the signal
type a given hardware will give, so now it can implement
SIOCGIWRANGE itself, removing more wext stuff from mac80211.
Might need to be a little more parametrized once we have
more hardware using cfg80211 and new hardware capabilities.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 18 Feb 2009 17:45:06 +0000 (18:45 +0100)]
cfg80211: clean up signal type
It wasn't a good idea to make the signal type a per-BSS option,
although then it is closer to the actual value. Move it to be
a per-wiphy setting, update mac80211 to match.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 18 Feb 2009 09:48:07 +0000 (10:48 +0100)]
nl80211: remove admin requirement from station get
There's no particular reason to not let untrusted users see
this information -- it's just the stations we're talking to,
packet counters for them and possibly some mesh things.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Some APs don't start BA negotiation with the client before it is done
with the key handshake in WPA/RSN. With those APs, key handshake times
out if EAPOL frames are sent after addba request. So defer the BA negotiation
until we are done with tx/rx of all EAPOL frames.
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 18 Feb 2009 17:27:22 +0000 (18:27 +0100)]
cfg80211/mac80211: fill qual.qual value/adjust max_qual.qual
Due to various bugs in the software stack we end up having
to fill qual.qual; level should be used, but wpa_supplicant
doesn't properly ignore qual.qual, NM should use qual.level
regardless of that because qual.qual is 0 but doesn't handle
IW_QUAL_DBM right now.
So fill qual.qual with the qual.level value clamped to
-110..-40 dBm or just the regular 'unspecified' signal level.
This requires a mac80211 change to properly announce the
max_qual.qual and avg_qual.qual values.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Wed, 11 Feb 2009 22:14:43 +0000 (17:14 -0500)]
cfg80211: age scan results on resume
Scanned BSS entries are timestamped with jiffies, which doesn't
increment across suspend and hibernate. On resume, every BSS in the
scan list looks like it was scanned within the last 10 seconds,
irregardless of how long the machine was actually asleep. Age scan
results on resume with the time spent during sleep so userspace has a
clue how old they really are.
Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ivo van Doorn [Tue, 17 Feb 2009 13:04:29 +0000 (14:04 +0100)]
rt2x00: Fix RF offset
The word_base is in bytes instead of word index number,
this means that when using it, it should be transformed into
a word index first.
Otherwise RF register reading will fail through debugfs since
we would start reading 4 words starting with word 4 (which is the last
valid word for RF).
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Jouni Malinen [Tue, 17 Feb 2009 11:24:57 +0000 (13:24 +0200)]
nl80211: Provide access to STA TX/RX packet counters
The TX/RX packet counters are needed to fill in RADIUS Accounting
attributes Acct-Output-Packets and Acct-Input-Packets. We already
collect the needed information, but only the TX/RX bytes were
previously exposed through nl80211. Allow applications to fetch the
packet counters, too, to provide more complete support for accounting.
Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Jouni Malinen [Mon, 16 Feb 2009 17:39:13 +0000 (19:39 +0200)]
nl80211: Optional IEs into scan request
This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications
to specify a set of information element(s) to be added into Probe
Request frames with NL80211_ATTR_IE. This provides support for the
MLME-SCAN.request primitive parameter VendorSpecificInfo and can be
used, e.g., to implement WPS scanning.
Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Sun, 15 Feb 2009 17:06:11 +0000 (12:06 -0500)]
ath5k: move beacon processing to a tasklet
We currently send beacons directly from the interrupt routine. This
can hold up interrupt processing in beaconing modes and makes the
ISR somewhat more complex. Move it to a tasklet like rx and tx.
Changes-licensed-under: 3-Clause-BSD
Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Sun, 15 Feb 2009 17:06:10 +0000 (12:06 -0500)]
ath5k: use spin_lock_irqsave for beacon lock
ath5k_reset can be called from process context, which in turn can
call ath5k_beacon_config which takes the sc->block spinlock. Since
it can also be taken in hard irq context, use spin_lock_irqsave
everywhere. This fixes a potential deadlock in adhoc mode.
Changes-licensed-under: 3-Clause-BSD
Cc: stable@kernel.org Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Sun, 15 Feb 2009 17:06:09 +0000 (12:06 -0500)]
ath5k: don't mask off interrupt bits
Since interrupts are already masked by the hardware, there's no need to
discard interrupt bits in the ISR itself. Also, in ath5k_beacon_config
we mask off a couple of bits without locking, so doing this mask in
software can lead to unhandled beacon timer and beacon miss interrupts.
Changes-licensed-under: 3-Clause-BSD
Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Ivo van Doorn [Sun, 15 Feb 2009 16:42:48 +0000 (17:42 +0100)]
rt2x00: Remove check for rf word 0
The only way rf_write() can be called with word 0 is
when the user sends the wrong word index through debugfs.
However the values which are send through debugfs are
validated using the RF_BASE and RF_SIZE macro values,
the most logical solution is to increase RF_BASE with 4
and decrease RF_SIZE with 4 (RF_SIZE has always been
1 word too big)
Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Randy Dunlap [Sat, 14 Feb 2009 00:39:35 +0000 (16:39 -0800)]
wireless: fix for CONFIG_NL80211=n
Add empty function for case of CONFIG_NL80211=n:
net/wireless/scan.c:35: error: implicit declaration of function 'nl80211_send_scan_aborted'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Reinette Chatre [Fri, 13 Feb 2009 19:51:19 +0000 (11:51 -0800)]
iwlwifi: Fix and rework Kconfig file
Fixes:
- iwlwifi is an optional driver and should thus not default to 'y'.
- 3945 now depends on IWLCORE.
Rework:
- There is not a case when IWLCORE should not be selected. At the same time
the driver does not use IWLWIFI or IWLCORE. We can just merge the usage of
these two. With IWLWIFI being the driver name we proceed to use just it and
replace instances of IWLCORE with it. The module name does not change
and is still iwlcore.
- Both IWLAGN and IWL3945 are selecting FW_LOADER, we can thus just move
this up to one select when IWLWIFI is selected.
- IWL5000 now supports Intel Wireless Wifi 100, 6000, and 6050 series.
- Now that 3945 depends on IWLWIFI we can also indicate its dependency on
MAC80211_LEDS and LEDS_CLASS at this level.
- IWLAGN_LEDS is not used by driver - remove it.
- IWLAGN_SPECTRUM_MEASUREMENT actually depends on IWLWIFI as it forms part
of iwlcore module. Move this config up in Kconfig to reflect that and also
change name to IWLWIFI_SPECTRUM_MEASUREMENT.
- CONFIG_IWLWIFI_RFKILL is used by iwlagn as well as iwl3945, add text to
description that indicates this.
- CONFIG_IWL3945_RFKILL does not exist - remove usage from driver.
- Add "iwlagn" to end of description of IWLAGN to help people understand
what iwlagn means in rest of Kconfig text.
- Add "iwl3945" to end of description of IWL3945 to help people understand
what iwlagn means in rest of Kconfig text.
- Change IWLWIFI_DEBUGFS description to indicate that only iwlagn supports
it (for now).
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com> Reported-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Reinette Chatre [Fri, 13 Feb 2009 19:51:18 +0000 (11:51 -0800)]
iwlwifi: use singlethread workqueue
Use one workqueue instead of one per CPU.
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
cc: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Jiri Slaby [Thu, 12 Feb 2009 21:29:40 +0000 (22:29 +0100)]
ipw2200, fix ipw io functions
- some of them are defined as follows:
#define ipw_write32 expr1; expr2
and are called from loops or ifs without a compound statement, so
they are broken. Fix it by putting them into do {} while (0) for
writes and ({ }) for reads.
- also unify and cleanup them while at it -- convert them from
macros to inline functions, so that we get some basic typechecking
Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Acked-by: Zhu Yi <yi.zhu@intel.com> Cc: James Ketrenos <jketreno@linux.intel.com> Cc: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:58 +0000 (13:38 -0500)]
ath9k: move common regulatory code out of if() branches
Both branches of the regulatory check in ath_attach() set up a
custom regulatory domain and apply radar and world flags, so extract
those into a single path.
While at it, fix a couple of spelling errors and an unnecessary extra
pointer traversal.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:57 +0000 (13:38 -0500)]
ath9k: remove prototype for ath9k_regd_get_current_country
ath9k_regd_get_current_country() doesn't exist.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:56 +0000 (13:38 -0500)]
ath9k: remove ath9k_regd_get_rd()
The function ath9k_regd_get_rd() is unused.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:55 +0000 (13:38 -0500)]
ath9k: convert isWwrSKU macro into C code
Write isWwrSKU as an inline function and nix the camel-case to make
the routine slightly clearer. Change its argument to take the regd
value directly so it can eventually be used by ath5k as well.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:54 +0000 (13:38 -0500)]
ath9k: save a few calls to ath9k_regd_get_eepromRD
Since we already have a stack variable to track the eeprom regd,
we can grab it up front it and save three calls to fetch it again.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bob Copeland [Thu, 12 Feb 2009 18:38:53 +0000 (13:38 -0500)]
ath9k: remove write-only current_rd_inuse
The current_rd_inuse regulatory value is assigned but not used anywhere.
Signed-off-by: Bob Copeland <me@bobcopeland.com> Acked-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Helmut Schaa [Thu, 12 Feb 2009 17:51:03 +0000 (18:51 +0100)]
iwlagn: clean up error path in iwl_pci_probe
This avoids triggering a BUG_ON in pci_disable_msi in the error path.
Furthermore remove the first call to pci_disable_device as it is already
called at out_pci_disable_device.
Both issues were introduced in the patch "iwlagn: fix hw-rfkill while
the interface is down".
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com> Acked-by: Reinette Chatre <reinette.chatre@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Sujith [Thu, 12 Feb 2009 06:08:37 +0000 (11:38 +0530)]
mac80211: Extend the rate control API with an update callback
The AP can switch dynamically between 20/40 Mhz channel width,
in which case we switch the local operating channel, but the
rate control algorithm is not notified. This patch adds a new callback
to indicate such changes to the RC algorithm.
Currently, HT channel width change is notified, but this callback
can be used to indicate any new requirements that might come up later on.
Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Sujith [Thu, 12 Feb 2009 04:36:45 +0000 (10:06 +0530)]
ath9k: Enable Fractional N mode
This patch enables Fractional N mode for all channel
if the EEPROM says so, and also fixes the INI only
when the device is not a 2 GHz only capable device.
Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Sun, 15 Feb 2009 11:44:28 +0000 (12:44 +0100)]
mac80211: split IBSS/managed code
This patch splits out the ibss code and data from managed (station) mode.
The reason to do this is to better separate the state machines, and have
the code be contained better so it gets easier to determine what exactly
a given change will affect, that in turn makes it easier to understand.
This is quite some churn, especially because I split sdata->u.sta into
sdata->u.mgd and sdata->u.ibss, but I think it's easier to maintain that
way. I've also shuffled around some code -- null function sending is only
applicable to managed interfaces so put that into that file, some other
functions are needed from various places so put them into util, and also
rearranged the prototypes in ieee80211_i.h accordingly.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 11 Feb 2009 23:51:53 +0000 (00:51 +0100)]
mac80211: fix aggregation for hardware with ampdu queues
Hardware with AMPDU queues currently has broken aggregation.
This patch fixes it by making all A-MPDUs go over the regular AC queues,
but keeping track of the hardware queues in mac80211. As a first rough
version, it actually stops the AC queue for extended periods of time,
which can be removed by adding buffering internal to mac80211, but is
currently not a huge problem because people rarely use multiple TIDs
that are in the same AC (and iwlwifi currently doesn't operate as AP).
This is a short-term fix, my current medium-term plan, which I hope to
execute soon as well, but am not sure can finish before .30, looks like
this:
1) rework the internal queuing layer in mac80211 that we use for
fragments if the driver stopped queue in the middle of a fragmented
frame to be able to queue more frames at once (rather than just a
single frame with its fragments)
2) instead of stopping the entire AC queue, queue up the frames in a
per-station/per-TID queue during aggregation session initiation,
when the session has come up take all those frames and put them
onto the queue from 1)
3) push the ampdu queue layer abstraction this patch introduces in
mac80211 into the driver, and remove the virtual queue stuff from
mac80211 again
This plan will probably also affect ath9k in that mac80211 queues the
frames instead of passing them down, even when there are no ampdu queues.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Johannes Berg [Wed, 11 Feb 2009 19:27:30 +0000 (20:27 +0100)]
mac80211: disallow moving netns
mac80211 currently assumes init_net for all interfaces,
so really will not cope well with network namespaces,
at least at this time.
To change this, we would have keep track of the netns
in addition to the ifindex, which is not something I
want to think about right now.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
mac80211: Make sure non-HT connection when IEEE80211_STA_TKIP_WEP_USED is set
It is possible that some broken AP might send HT IEs in it's
assoc response even though the STA has not sent them in assoc req
when WEP/TKIP is used as pairwise cipher suite. Also it is important
to check this bit before enabling ht mode in beacon receive path.
Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
Bing Zhao [Thu, 5 Feb 2009 06:22:39 +0000 (22:22 -0800)]
libertas: fix power save issue in libertas_sdio module
The problem: "iwconfig ethX power on" returns error
The cause: "ps_supported" flag was never set for SD8385/8686
The fix: check firmware capabilities returned by GET_HW_SPEC command.
Set "ps_supported" to 1 if FW_CAPINFO_PS bit is on. This fix applies
to SDIO interface only.
Signed-off-by: Bing Zhao <bzhao@marvell.com> Acked-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
This patch adds two new device ids to the asix driver.
One comes directly from the asix driver on their web site, the other was
reported by Armani Liao as needed for the MSI X320 to get the driver to
work properly for it.
Reported-by: Armani Liao <aliao@novell.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Ron Mercer [Mon, 23 Feb 2009 10:42:17 +0000 (10:42 +0000)]
qlge: Use one path to (re)fill rx buffers.
Currently there are two paths for filling rx buffer queues. One is
used during initialization and the other during runtime. This patch
removes ql_alloc_sbq_buffers() and ql_alloc_lbq_buffers() and replaces
them with a call to the runtime functions ql_update_lbq() and
ql_update_sbq().
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ron Mercer [Mon, 23 Feb 2009 10:42:16 +0000 (10:42 +0000)]
qlge: Optimize rx buffer refill process.
RX Buffers are refilled in chunks of 16 at a time before notifying the
hardware with a register write. This can cause several writes to take
place in a given napi poll call. This change causes the write to take place
only once at the end of the call.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of taking/giving the hw semaphore repeatedly when iterating over
several frame to queue route settings, we have the caller hold it until
all are done.
This reduces PCI bus chatter and possible waits.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ron Mercer [Mon, 23 Feb 2009 10:42:14 +0000 (10:42 +0000)]
qlge: Increase MAC addr hw sem granularity.
Instead of taking/giving the semaphore repeatedly when iterating over
several adderesses, we have the caller hold it until all are done. This
reduces PCI bus chatter and possible waits.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ron Mercer [Mon, 23 Feb 2009 10:42:13 +0000 (10:42 +0000)]
qlge: Clean up mac address and frame route settings.
Setting MAC addresses and routing frames to various queues will need to
be done in response to firmware events as well as during initialization.
This change encapsulates the facilities into a single call that can
later me made from other places.
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch changes the return value of nlmsg_notify() as follows:
If NETLINK_BROADCAST_ERROR is set by any of the listeners and
an error in the delivery happened, return the broadcast error;
else if there are no listeners apart from the socket that
requested a change with the echo flag, return the result of the
unicast notification. Thus, with this patch, the unicast
notification is handled in the same way of a broadcast listener
that has set the NETLINK_BROADCAST_ERROR socket flag.
This patch is useful in case that the caller of nlmsg_notify()
wants to know the result of the delivery of a netlink notification
(including the broadcast delivery) and take any action in case
that the delivery failed. For example, ctnetlink can drop packets
if the event delivery failed to provide reliable logging and
state-synchronization at the cost of dropping packets.
This patch also modifies the rtnetlink code to ignore the return
value of rtnl_notify() in all callers. The function rtnl_notify()
(before this patch) returned the error of the unicast notification
which makes rtnl_set_sk_err() reports errors to all listeners. This
is not of any help since the origin of the change (the socket that
requested the echoing) notices the ENOBUFS error if the notification
fails and should resync itself.
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>
mv643xx_eth: set sane default receive coalescing timeout
A receive coalescing timeout of 250 usec appears to strike a good
balance between allowing enough received frames to be aggregated for
LRO to do its job and not allowing the connection to stall due to
delaying ACKs to the remote end for too long.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
mv643xx_eth: move a couple of init actions from ->open() to port probe
Move the netif_carrier_off() call in ->open() to port probe, so that
ethtool doesn't report the link as being up before we have up'd the
interface.
Move initialisation of the rx/tx coalescing timers from ->open() to
port probe, so that we don't reset the coalescing timers every time
the interface is up'd.
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jesse Brandeburg [Wed, 25 Feb 2009 00:37:31 +0000 (16:37 -0800)]
ixbge: fix bug when using large pages and jumbo frames
it was pointed out on the list that ixgbe was failing when using 64kB pages
and large 16kB MTU.
since with a 64kB PAGE_SIZE MAX_SKB_FRAGS = 3, the way the driver was
configuring page usage was assuming 2kB is half a page, and was only
ever dmaing that much data to a half page.
(16kB - header size) / 2048 = 7 or 8 pages, which would far exceed 3
adjust the driver to account for these large pages, the hardware can
support DMA to up to 16kB for each descriptor.
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>
Shannon Nelson [Wed, 25 Feb 2009 00:36:38 +0000 (16:36 -0800)]
ixgbe: Move ring features into an enum, allowing easier future maintenance
From: Shannon Nelson <shannon.nelson@intel.com>
The ring_feature member of ixgbe_adapter is statically allocated based on
the supported features of the device. When a new feature is added, we need
to manually update the static allocation. This patch makes the feature
list an enum, eliminating the need for multiple updates to the code when
adding a new feature.
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>
Josef Drexler [Tue, 24 Feb 2009 13:53:12 +0000 (14:53 +0100)]
netfilter: xt_recent: fix proc-file addition/removal of IPv4 addresses
Fix regression introduded by commit 079aa88 (netfilter: xt_recent: IPv6 support):
From http://bugzilla.kernel.org/show_bug.cgi?id=12753:
Problem Description:
An uninitialized buffer causes IPv4 addresses added manually (via the +IP
command to the proc interface) to never match any packets. Similarly, the -IP
command fails to remove IPv4 addresses.
Details:
In the function recent_entry_lookup, the xt_recent module does comparisons of
the entire nf_inet_addr union value, both for IPv4 and IPv6 addresses. For
addresses initialized from actual packets the remaining 12 bytes not occupied
by the IPv4 are zeroed so this works correctly. However when setting the
nf_inet_addr addr variable in the recent_mt_proc_write function, only the IPv4
bytes are initialized and the remaining 12 bytes contain garbage.
Hence addresses added in this way never match any packets, unless these
uninitialized 12 bytes happened to be zero by coincidence. Similarly, addresses
cannot consistently be removed using the proc interface due to mismatch of the
garbage bytes (although it will sometimes work to remove an address that was
added manually).
Reading the /proc/net/xt_recent/ entries hides this problem because this only
uses the first 4 bytes when displaying IPv4 addresses.
Doc: Refer to ip-sysctl.txt for strict vs. loose rp_filter mode
The IP_ADVANCED_ROUTER Kconfig describes the rp_filter
proc option. Recent changes added a loose mode.
Instead of documenting this change too places, refer to
the document describing it:
Documentation/networking/ip-sysctl.txt
I'm considering moving the rp_filter description away
from the Kconfig file into ip-sysctl.txt.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk> Signed-off-by: David S. Miller <davem@davemloft.net>
Dhananjay Phadke [Tue, 24 Feb 2009 11:42:59 +0000 (03:42 -0800)]
netxen: fix physical port mapping
The PCI function to physical port mapping is valid only for
old firmware. New firmware (4.0.0+) abstracts this.
So driver should never try to access phy using invalid
mapping. The behavior is unpredictable when PCI functions
4-7 are enabled on the same NIC.
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
net: amend the fix for SO_BSDCOMPAT gsopt infoleak
netns: build fix for net_alloc_generic
proc: proc_get_inode should de_put when inode already initialized
de_get is called before every proc_get_inode, but corresponding de_put is
called only when dropping last reference to an inode. This might cause
something like
remove_proc_entry: /proc/stats busy, count=14496
to be printed to the syslog.
The fix is to call de_put in case of an already initialized inode in
proc_get_inode.
Signed-off-by: Krzysztof Sachanowicz <analyzer1@gmail.com> Tested-by: Marcin Pilipczuk <marcin.pilipczuk@gmail.com> Acked-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jesse Barnes [Mon, 23 Feb 2009 23:41:09 +0000 (15:41 -0800)]
i915: suspend/resume interrupt state
In the KMS case, enter/leavevt won't fix up the interrupt handler for
us, so we need to do it at suspend/resume time. Make sure we don't fail
the resume if the chip is hung either.
Eugene Teo [Mon, 23 Feb 2009 23:38:41 +0000 (15:38 -0800)]
net: amend the fix for SO_BSDCOMPAT gsopt infoleak
The fix for CVE-2009-0676 (upstream commit df0bca04) is incomplete. Note
that the same problem of leaking kernel memory will reappear if someone
on some architecture uses struct timeval with some internal padding (for
example tv_sec 64-bit and tv_usec 32-bit) --- then, you are going to
leak the padded bytes to userspace.
Signed-off-by: Eugene Teo <eugeneteo@kernel.sg> Reported-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
netns: fix double free at netns creation
veth : add the set_mac_address capability
sunlance: Beyond ARRAY_SIZE of ib->btx_ring
sungem: another error printed one too early
ISDN: fix sc/shmem printk format warning
SMSC: timeout reaches -1
smsc9420: handle magic field of ethtool_eeprom
sundance: missing parentheses?
smsc9420: fix another postfixed timeout
wimax/i2400m: driver loads firmware v1.4 instead of v1.3
vlan: Update skb->mac_header in __vlan_put_tag().
cxgb3: Add support for PCI ID 0x35.
tcp: remove obsoleted comment about different passes
TG3: &&/|| confusion
ATM: misplaced parentheses?
net/mv643xx: don't disable the mib timer too early and lock properly
net/mv643xx: use GFP_ATOMIC while atomic
atl1c: Atheros L1C Gigabit Ethernet driver
net: Kill skb_truesize_check(), it only catches false-positives.
net: forcedeth: Fix wake-on-lan regression
Larry Finger [Tue, 17 Feb 2009 20:31:12 +0000 (14:31 -0600)]
rtl8187: New USB ID's for RTL8187L
Add new USB ID codes. These come from two postings on forums and
mailing lists, and four are derived from the .inf that accompanies
the latest Realtek Windows driver for the RTL8187L.
Thanks to Viktor Ilijašić <viktor.ilijasic@gmail.com> and Xose Vazquez
Perez <xose.vazquez@gmail.com> for reporting these new ID's.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>