* Re: [PATCH 2/4] Phonet: advise against enabling the pipe controller
From: David Miller @ 2010-10-08 21:17 UTC (permalink / raw)
To: remi; +Cc: netdev, remi.denis-courmont
In-Reply-To: <1286546523-3340-2-git-send-email-remi@remlab.net>
From: Rémi Denis-Courmont <remi@remlab.net>
Date: Fri, 8 Oct 2010 17:02:01 +0300
> From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
>
> 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>
This really needs to be designed such that it can co-exist with current
userland technology.
Applied.
^ permalink raw reply
* Re: [PATCH 1/4] Phonet: add to MAINTAINERS and add myself
From: David Miller @ 2010-10-08 21:17 UTC (permalink / raw)
To: remi; +Cc: netdev, remi.denis-courmont
In-Reply-To: <1286546523-3340-1-git-send-email-remi@remlab.net>
From: Rémi Denis-Courmont <remi@remlab.net>
Date: Fri, 8 Oct 2010 17:02:00 +0300
> From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
>
> Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Applied.
^ permalink raw reply
* Re: pull request: wireless-next-2.6 2010-10-08
From: David Miller @ 2010-10-08 21:01 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, netdev, linux-kernel
In-Reply-To: <20101008201008.GA2414@tuxdriver.com>
From: "John W. Linville" <linville@tuxdriver.com>
Date: Fri, 8 Oct 2010 16:10:09 -0400
> This offering is mostly the usual driver updates and such for ath9k,
> iwlwifi, rt2x00, wl1271, and a few others. Also featured are a number
> of mac80211, cfg80211 and nl80211 improvements from Johannes, as well
> as some mac80211 scanning improvements from Stanislaw. Johannes is
> also sneaking a genetlink feature through this tree, of which you
> are already aware. :-)
That sneaky Johannes.
Pulled, thanks John.
^ permalink raw reply
* Re: [PATCH NEXT 0/7]qlcnic: driver update
From: David Miller @ 2010-10-08 21:01 UTC (permalink / raw)
To: amit.salecha; +Cc: netdev, ameen.rahman, anirban.chakraborty
In-Reply-To: <1286531171-21173-1-git-send-email-amit.salecha@qlogic.com>
From: Amit Kumar Salecha <amit.salecha@qlogic.com>
Date: Fri, 8 Oct 2010 02:46:04 -0700
> Hi
> Series of 7 to support quiescent mode and fixes.
> Please include them in net-next.
All applied, thank you.
^ permalink raw reply
* tbf/htb qdisc limitations
From: Steven Brudenell @ 2010-10-08 20:58 UTC (permalink / raw)
To: netdev
hi folks,
i was disappointed recently to find that i can't set the "burst"
parameters very high on the tbf or htb qdiscs. the actual limit of the
burst parameters varies, according to the rate parameter. at the
relatively low rate i want to set, i want to have the burst parameter
be several gigabytes, but i'm actually limited to only a few
megabytes.
(motivation: a fully-automated way to stay inside the monthly transfer
limits imposed by many ISPs these days, without resorting to a
constant rate limit. for example, comcast limits its customers to
250GB/month, which is about 101KB/s; many cellular data plans in the
US limit to 5GB/month =~ 2KB/s).
i'll gladly code a patch, but i'd like the list's advice on whether
this is necessary, and a little bit about how to proceed:
1) what is the purpose of the "rate tables" used in these qdiscs --
why use them in favor of dividing bytes by time to compute a rate? i
assume the answer has something to do with restrictions on using
floating point math (maybe even integer division?) at different places
/ interruptability states in the kernel. maybe this is documented on
kernelnewbies somewhere but i couldn't find it.
2) is there an established procedure for versioning a netlink
interface? today the netlink interface for tbf and htb is horribly
implementation-coupled (the "burst" parameters need to be munged
according to the "rate" parameters and kernel tick rate). i think i
would need to change these interfaces in order to change the
accounting implementation in the corresponding qdisc. however, i
probably want to remain compatible with old userspace.
~steve
^ permalink raw reply
* Re: [PATCH] net/fec: carrier off initially to avoid root mount failure
From: David Miller @ 2010-10-08 20:50 UTC (permalink / raw)
To: oskar; +Cc: netdev, dan, bigeasy, hjk, gerg
In-Reply-To: <20101008203530.GA11926@www.tglx.de>
From: Oskar Schirmer <oskar@linutronix.de>
Date: Fri, 8 Oct 2010 22:35:30 +0200
> So when asking for the most correct place to fix this, I'ld propose
> phylib itself: Shouldn't the correct carrier state be set with
> phy_start_machine or phy_device_create when phy state is set to
> PHY_DOWN?
I agree, phylib is the best place to handle this.
The carrier defaults to "on" for newly created net devices
since that's what software devices and hardware ones which
lack a link want.
phy_device_create() currently lacks access to the network device
struct.
phy_connect_direct() and phy_attach_direct() might be good spots.
^ permalink raw reply
* Re: [PATCH] net/fec: carrier off initially to avoid root mount failure
From: Oskar Schirmer @ 2010-10-08 20:35 UTC (permalink / raw)
To: David Miller; +Cc: oskar, netdev, dan, bigeasy, hjk, gerg
In-Reply-To: <20101008.103112.189689736.davem@davemloft.net>
On Fri, Oct 08, 2010 at 10:31:12 -0700, David Miller wrote:
> From: Oskar Schirmer <oskar@linutronix.de>
> Date: Thu, 7 Oct 2010 14:30:30 +0200
>
> > 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>
>
> Maybe fs_enet_open() is a better place for this netif_carrier_off()
> call?
The patch is for fec, so I guess You propose fec_enet_open().
> Every open the driver probes the PHY and does phy_start().
netif_carrier handling is done in phylib properly
except with initialisation. So when asking for the most
correct place to fix this, I'ld propose phylib itself:
Shouldn't the correct carrier state be set with
phy_start_machine or phy_device_create when phy state is
set to PHY_DOWN?
Seen from this point of view, the patch I did propose is
a workaround, as much as doing it in the ndo_open, I guess.
^ permalink raw reply
* pull request: wireless-next-2.6 2010-10-08
From: John W. Linville @ 2010-10-08 20:10 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
Dave,
Another week, another huge pull request for -next...
This offering is mostly the usual driver updates and such for ath9k,
iwlwifi, rt2x00, wl1271, and a few others. Also featured are a number
of mac80211, cfg80211 and nl80211 improvements from Johannes, as well
as some mac80211 scanning improvements from Stanislaw. Johannes is
also sneaking a genetlink feature through this tree, of which you
are already aware. :-)
Please let me know if there are problems!
Thanks,
John
---
The following changes since commit dd53df265b1ee7a1fbbc76bb62c3bec2383bbd44:
bonding: add Speed/Duplex information to /proc/net/bonding/bond (2010-10-06 18:43:34 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next-2.6.git for-davem
Ben Greear (4):
wireless: Use first phyX name available when registering phy devices.
ath5k: Allow ath5k to support virtual STA and AP interfaces.
ath5k: Print rx/tx bytes in debugfs
ath9k: Report total tx/rx bytes and packets in debugfs.
Ben Hutchings (1):
ar9170: Remove MODULE_FIRMWARE references to two-stage firmware
Bill Jordan (3):
ath9k: add WDS interfaces to ath9k
cfg80211: patches to allow setting the WDS peer
mac80211: fix for WDS interfaces
Bruno Randolf (3):
mac80211: Add WME information element for IBSS
ath5k: Don't wake internal queues
nl80211/mac80211: Add retry and failed transmission count to station info
Christian Lamparter (3):
mac80211: fix rx monitor filter refcounters
carl9170: support firmware-based rx filter
p54usb: add five more USBIDs
Dan Carpenter (1):
ipw2200: check for allocation failures
Felix Fietkau (17):
ath9k_htc: remove use of common->ani.noise_floor
ath9k_hw: keep calibrated noise floor values per channel
ath9k: do not return default noise floor values via survey
ath9k: remove the noise floor value in the ani struct
nl80211: allow drivers to indicate whether the survey data channel is in use
ath9k: return survey data for all channels instead of just the current one
ath9k: fix channel flag / regd issues with multiple cards
ath9k_hw: remove some useless calibration data
ath9k_hw: fix regression in ANI listen time calculation
ath9k_hw: clean up calibration flags
ath9k_hw: merge codepaths that access the cycle counter registers
ath9k_hw: clean up register write buffering
ath9k_hw: add a helper function to check for the new ANI implementation
ath9k_hw: clean up ANI state handling
ath9k_hw: remove code duplication in phy error counter handling
ath9k_hw: remove function pointer abstraction for internal ANI ops
ath9k_hw: merge ath9k_hw_ani_monitor_old and ath9k_hw_ani_monitor_new
Gabor Juhos (1):
ath9k: get correct tx gain type in ath9k_hw_4k_get_eeprom
Helmut Schaa (18):
rt2x00: Don't overwrite beacon buffers in pairwise key setup
rt2x00: Split out parts of the rt2800_txdone function for easier reuse
rt2x00: rework tx status handling in rt2800pci
rt2x00: Fix SM PS check
rt2x00: Implement HT protection for rt2800
rt2x00: Don't initialize MM40 HT protection to RTS/CTS on PCI devices
rt2x00: Fix race between dma mapping and clearing rx entries in rt2800pci
rt2x00: Allow tx duplication for legacy rates in HT40 mode
rt2x00: Add register definition for busy time on secondary channel
rt2x00: add field definitions for the TBTT_SYNC_CFG register
rt2x00: Don't enable broad- and multicast buffering on USB devices
mac80211: distinct between max rates and the number of rates the hw can report
rt2x00: correctly set max_report_rates in rt61pci and rt2800
rt2x00: Enable rx aggregation in rt2800
rt2x00: Update comment about the AMPDU flag in the TXWI
rt2x00: Fix oops caused by error path in rt2x00lib_start
rt2x00: Improve cooperation between rt2800pci and minstrel
rt2x00: Work around hw aggregation oddity in rt2800
Ido Yariv (3):
wl1271: Handle large SPI transfers
wl1271: Support firmware RX packet aggregation
wl1271: Support firmware TX packet aggregation
Ivo van Doorn (2):
rt2x00: Add rt73usb device ID
rt2x00: Improve TX status entry validation
Jay Sternberg (1):
iwlwifi: define 100 devices
Johannes Berg (17):
cfg80211/mac80211: allow management frame TX in AP mode
cfg80211: remove spurious __KERNEL__ ifdef
cfg80211: fix locking
genetlink: introduce pre_doit/post_doit hooks
nl80211: use the new genetlink pre/post_doit hooks
nl80211: use generic check for netif_running
nl80211: reduce dumping boilerplate
mac80211: fix deadlock with multiple interfaces
nl80211: fix remain-on-channel documentation
mac80211: delete AddBA response timer
mac80211: avoid transmitting delBA to old AP
cfg80211/mac80211: allow per-station GTKs
nl80211: fix error in generic netif_running check
cfg80211: fix BSS double-unlinking
mac80211: fix sw scan locking
nl80211: use new genl helpers for WDS
cfg80211: constify WDS address
John W. Linville (9):
Merge branch 'wireless-next-2.6' of git://git.kernel.org/.../iwlwifi/iwlwifi-2.6
Merge branch 'wireless-next' of git://git.kernel.org/.../luca/wl12xx
Merge branch 'master' of git://git.kernel.org/.../linville/wireless-2.6
mac80211: avoid uninitialized var warning in ieee80211_scan_cancel
Revert "rtl8180: use NAPI for bottom-half processing"
Revert "mac80211: use netif_receive_skb in ieee80211_tx_status callpath"
Merge branch 'master' of git://git.kernel.org/.../linville/wireless-2.6
Merge branch 'wireless-next-2.6' of git://git.kernel.org/.../iwlwifi/iwlwifi-2.6
Merge branch 'master' of git://git.kernel.org/.../linville/wireless-next-2.6 into for-davem
Juuso Oikarinen (27):
wl1271: Implement CMD_SET_STA_STATE to indicate connection completion to FW
wl1271: Add trigger to net_device oper_state to change BT coex priority
wl1271: Fix scan race condition
wl1271: Change supported channel order for a more optimal scan
wl1271: Remove outdated FIXME's
wl1271: Move setting of wake-up conditions to ps.c from cmd.c
wl1271: Remove acx_fw_version function
wl1271: Release interrupt *after* releasing the driver context
wl1271: Remove function cmd_read_memory
wl1271: Enable/disable 11a support based on INI configuration
wl1271: Reduce rate used for last PSM entry attempt
wl1271: Enable 11a support always, prevent scanning for unsupporting chips
wl1271: Fix tid-configuration of TX frames
wl1271: Fix AC/TID default configuration
wl1271: Move scan complete invocation into work function
wl1271: Fix work cancelling when shutting down the driver
wl1271: Separate interface removal to another function
wl1271: Add hardware recovery mechanism
wl1271: Add handling for failing hardware scan command
wl1271: Optimize scan duration
wl1271: Increase connection reliability
wl1271: Add support for hardware GEM cipher
wl1271: Move work-init calls to hw allocation
wl1271: Indicate disconnection on hardware failure
wl1271: Add extended radio parameter initialization
mac80211: Add validity check for beacon_crc value
wl1271: Support FEM autodetection
Luciano Coelho (5):
wl1271: remove deprecated ACX definition
wl1271: remove useless 11a check when scanning
wl1271: fix forced passive scan
MAINTAINERS: Update git url for the wl1271 driver
wl1271: remove deprecated __attribute__ ((packed))
Luis R. Rodriguez (1):
mac80211: fix channel assumption for association done work
Mohammed Shafi Shajakhan (1):
ath9k : Fix for displaying the channel number
Ohad Ben-Cohen (1):
wl12xx: fix non-wl12xx build scenarios
Rafał Miłecki (3):
b43: N-PHY: simplify channel switching
b43: define B43_SHM_SH_CHAN_40MHZ
b43: N-PHY: don't duplicate setting channel in shared memory
Rajkumar Manoharan (1):
ath9k_hw: Fix hw reset failure with HTC driver
Senthil Balasubramanian (2):
ath9k: remove unnecessary power save flags.
ath9k: Introduce a wrapper for power save disable.
Shanyu Zhao (3):
iwlwifi: add iwl_nic_ops structure to iwl_ops
iwlagn: set CSR register for 6050g2 devices
iwlwifi: send DC calib config to runtime ucode
Stanislaw Gruszka (8):
iwlwifi: change WARN_ON to IWL_WARN in iwl_mac_add_interface
mac80211: perform scan cancel in hw reset work
mac80211: reduce number of __ieee80211_scan_completed calls
mac80211: keep lock when calling __ieee80211_scan_completed()
mac80211: assure we also cancel deferred scan request
mac80211: do not requeue scan work when not needed
mac80211: compete scan to cfg80211 if deferred scan fail to start
Revert "iwlwifi: do not perferm force reset while doing scan"
Wey-Yi Guy (9):
iwlagn: API v5 for 6050 series device
iwlwifi: schedule to deprecate software scan support
iwlagn: reduce redundant parameter definitions
iwlagn: rename iwl5000_tx_power_dbm_cmd
iwlwifi: remove uncorrect comments
iwlagn: rename iwl5000_tx_resp
iwlagn: define 130 series of WiFi/BT combo devices
iwlagn: fix module firmware name for 130 series device
iwlagn: add define for MODULE_FIRMWARE
ext Anand Gadiyar (1):
omap: zoom2/3: fix build caused by wl1271 support
Omnibus patch available here:
http://www.kernel.org/pub/linux/kernel/people/linville/wireless-next-2.6-2010-10-08.patch.bz2
--
John W. Linville Someday the world will need a hero, and you
linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org might be all we have. Be ready.
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: Jason Glasgow @ 2010-10-08 19:18 UTC (permalink / raw)
To: David Miller; +Cc: mjg59, ellyjones, netdev, msb, olofj
In-Reply-To: <20101008.120205.70182089.davem@davemloft.net>
David,
Yes, your concern is valid. A distribution would have trouble
supporting this out of the box unless they negotiated with Qualcomm.
I believe that many years ago we had a similar situation with WiFi
devices and firmware.
I think the right thing to do is to encourage vendors to allow
redistribution of their firmware by showing them that the firmware is
already freely available on other platforms.
-Jason
On Fri, Oct 8, 2010 at 12:02 PM, David Miller <davem@davemloft.net> wrote:
>
> From: Jason Glasgow <jglasgow@google.com>
> Date: Fri, 8 Oct 2010 11:59:19 -0700
>
> > Firmware is available in Windows packages. For example,
> >
> > DW5620 Mobile Broadband mini-card
> > http://support.dell.com/support/downloads/download.aspx?c=us&cs=2684&l=en&s=pub&releaseid=R275082&SystemID=Lat_13&servicetag=4PLG5L1&os=W732&osl=en&deviceid=21977&devlib=0&typecnt=0&vercnt=2&catid=-1&impid=-1&formatcnt=0&libid=20&typeid=-1&dateid=-1&formatid=-1&source=-1&fileid=406043
> >
> > <http://support.dell.com/support/downloads/download.aspx?c=us&cs=2684&l=en&s=pub&releaseid=R275082&SystemID=Lat_13&servicetag=4PLG5L1&os=W732&osl=en&deviceid=21977&devlib=0&typecnt=0&vercnt=2&catid=-1&impid=-1&formatcnt=0&libid=20&typeid=-1&dateid=-1&formatid=-1&source=-1&fileid=406043>The
> > problem is that the firmware is non-redistributable, not non-distributable.
>
> So address my concern, how could a distribution support this device
> out of the box?
>
> I don't think it's possible.
^ permalink raw reply
* Re: [PATCH] Tarpit target for the last stable (2.6.35.7): right version
From: Ben Hutchings @ 2010-10-08 19:10 UTC (permalink / raw)
To: Nicola; +Cc: netfilter-devel, netdev
In-Reply-To: <4CAF6A83.6090004@gmail.com>
On Fri, 2010-10-08 at 21:01 +0200, Nicola wrote:
> [Sorry for the inconvenience: I just noticed that Gmail has a bad
> formatting. I resend the patch hoping that now things are right.]
[...]
It's still word-wrapped. You may have to make the patch an attachment,
but see Documentation/email-clients.txt for configuration hints.
It is also not based on the right version; networking patches adding new
features should be based on David Miller's net-next-2.6 tree.
You should also read and follow Documentation/SubmittingPatches.
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: Matthew Garrett @ 2010-10-08 19:05 UTC (permalink / raw)
To: David Miller; +Cc: ellyjones, netdev, jglasgow, msb, olofj
In-Reply-To: <20101008.115050.28825457.davem@davemloft.net>
On Fri, Oct 08, 2010 at 11:50:50AM -0700, David Miller wrote:
> I don't see how any distribution can properly support this device
> if it cannot work without non-distributable firmware.
Individual vendors may negotiate redistribution agreements with
Qualcomm. We've carried drivers without redistributable firmware in the
past (b43 had none for some time), so I don't see that we have any clear
and established position on this.
--
Matthew Garrett | mjg59@srcf.ucam.org
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: David Miller @ 2010-10-08 19:02 UTC (permalink / raw)
To: jglasgow; +Cc: mjg59, ellyjones, netdev, msb, olofj
In-Reply-To: <AANLkTimbpPZt4FSeHDNcfuoYeKDYtV2YsLPxP-2ngELO@mail.gmail.com>
From: Jason Glasgow <jglasgow@google.com>
Date: Fri, 8 Oct 2010 11:59:19 -0700
> Firmware is available in Windows packages. For example,
>
> DW5620 Mobile Broadband mini-card
> http://support.dell.com/support/downloads/download.aspx?c=us&cs=2684&l=en&s=pub&releaseid=R275082&SystemID=Lat_13&servicetag=4PLG5L1&os=W732&osl=en&deviceid=21977&devlib=0&typecnt=0&vercnt=2&catid=-1&impid=-1&formatcnt=0&libid=20&typeid=-1&dateid=-1&formatid=-1&source=-1&fileid=406043
>
> <http://support.dell.com/support/downloads/download.aspx?c=us&cs=2684&l=en&s=pub&releaseid=R275082&SystemID=Lat_13&servicetag=4PLG5L1&os=W732&osl=en&deviceid=21977&devlib=0&typecnt=0&vercnt=2&catid=-1&impid=-1&formatcnt=0&libid=20&typeid=-1&dateid=-1&formatid=-1&source=-1&fileid=406043>The
> problem is that the firmware is non-redistributable, not non-distributable.
So address my concern, how could a distribution support this device
out of the box?
I don't think it's possible.
^ permalink raw reply
* [PATCH] Tarpit target for the last stable (2.6.35.7): right version
From: Nicola @ 2010-10-08 19:01 UTC (permalink / raw)
To: netfilter-devel, netdev
[Sorry for the inconvenience: I just noticed that Gmail has a bad
formatting. I resend the patch hoping that now things are right.]
Kernel module to capture and hold incoming TCP connections sending a
zero window packet to slow down the worm action.
More informations and newer version at:
http://npadovano.altervista.org/tarpit.html (or in the source code
comments)
Below you can find the patch and the relative userspace library
(compile it and put the .so output in /lib/xtables)
----tarpit-2.6.35.7.patch----
diff -rupN linux-2.6.35.7.orig/drivers/char/random.c
linux-2.6.35.7/drivers/char/random.c
--- linux-2.6.35.7.orig/drivers/char/random.c 2010-09-29
03:09:08.000000000 +0200
+++ linux-2.6.35.7/drivers/char/random.c 2010-10-08
02:51:35.000000000 +0200
@@ -1514,6 +1514,8 @@ __u32 secure_ip_id(__be32 daddr)
return half_md4_transform(hash, keyptr->secret);
}
+EXPORT_SYMBOL(secure_ip_id);
+
#ifdef CONFIG_INET
__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr,
@@ -1551,6 +1553,8 @@ __u32 secure_tcp_sequence_number(__be32
return seq;
}
+EXPORT_SYMBOL(secure_tcp_sequence_number);
+
/* Generate secure starting point for ephemeral IPV4 transport port
search */
u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport)
{
diff -rupN linux-2.6.35.7.orig/net/netfilter/Kconfig
linux-2.6.35.7/net/netfilter/Kconfig
--- linux-2.6.35.7.orig/net/netfilter/Kconfig 2010-09-29
03:09:08.000000000 +0200
+++ linux-2.6.35.7/net/netfilter/Kconfig 2010-10-08
12:51:07.000000000 +0200
@@ -548,6 +548,32 @@ config NETFILTER_XT_TARGET_SECMARK
To compile it as a module, choose M here. If unsure, say N.
+config NETFILTER_XT_TARGET_TARPIT
+ tristate '"TARPIT" target support'
+ depends on NETFILTER_XTABLES
+ default m if NETFILTER_ADVANCED=n
+ ---help---
+ Adds a TARPIT target to iptables, which captures and holds
+ incoming TCP connections using no local per-connection resources.
+ Connections are accepted, but immediately switched to the persist
+ state (0 byte window), in which the remote side stops sending data
+ and asks to continue every 60-240 seconds (with a probe packet).
+ Attempts to close the connection are ignored, forcing the remote
+ side to time out the connection in 12-24 minutes. Any TCP port that
+ you would normally DROP or REJECT can instead become a tarpit.
+ Example:
+
+ 1) iptables -A INPUT -p tcp --dport 1111 -j TARPIT
+ 2) iptables -A FORWARD -p tcp -s x.y.z.k --dport 1111 -j TARPIT
+
+ In the first example all the incoming TCP packets sent to the port
+ 1111 will be tarpitted. In the second one, we're tarpitting all the
+ forwarded packets sent to the host x.y.z.k on the port 1111.
+
+ You can find more informations and newer versions at:
+ <http://npadovano.altervista.org/tarpit.html>. Reporting bugs and
+ improvements are welcome.
+
config NETFILTER_XT_TARGET_TCPMSS
tristate '"TCPMSS" target support'
depends on (IPV6 || IPV6=n)
diff -rupN linux-2.6.35.7.orig/net/netfilter/Makefile
linux-2.6.35.7/net/netfilter/Makefile
--- linux-2.6.35.7.orig/net/netfilter/Makefile 2010-09-29
03:09:08.000000000 +0200
+++ linux-2.6.35.7/net/netfilter/Makefile 2010-10-08
02:53:44.000000000 +0200
@@ -56,6 +56,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE
obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o
obj-$(CONFIG_NETFILTER_XT_TARGET_RATEEST) += xt_RATEEST.o
obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o
+obj-$(CONFIG_NETFILTER_XT_TARGET_TARPIT) += xt_TARPIT.o
obj-$(CONFIG_NETFILTER_XT_TARGET_TPROXY) += xt_TPROXY.o
obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o
obj-$(CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP) += xt_TCPOPTSTRIP.o
diff -rupN linux-2.6.35.7.orig/net/netfilter/xt_TARPIT.c
linux-2.6.35.7/net/netfilter/xt_TARPIT.c
--- linux-2.6.35.7.orig/net/netfilter/xt_TARPIT.c 1970-01-01
01:00:00.000000000 +0100
+++ linux-2.6.35.7/net/netfilter/xt_TARPIT.c 2010-10-08
13:27:54.000000000 +0200
@@ -0,0 +1,304 @@
+/*
+ * Kernel module to capture and hold incoming TCP connections using
+ * no local per-connection resources.
+ *
+ * Based on ipt_REJECT.c and offering functionality similar to
+ * LaBrea <http://labrea.sourceforge.net/>.
+ *
+ * Original idea of:
+ * Aaron Hopkins, <tools@die.net>
+ *
+ * Version 2.6.35.7 kernel release, written by:
+ * Nicola Padovano, <nicola.padovano@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ *
+ * Goal:
+ * - Allow incoming TCP connections to be established.
+ * - Passing data should result in the connection being switched to the
+ * persist state (0 byte window), in which the remote side stops sending
+ * data and asks to continue every 60-240 seconds.
+ * - Prevent spoofing (don't allow the victim to connect with server worm)
+ * - Attempts to shut down the connection should be ignored completely, so
+ * the remote side ends up having to time it out.
+ *
+ * This means:
+ * - Reply to TCP SYN,!ACK,!RST,!FIN with SYN-ACK, window 5 bytes
+ * - Reply to TCP !SYN,!RST,!FIN with ACK, window 0 bytes, rate-limited
+ * - Reply to TCP SYN,ACK,!RST,!FIN with RST to prevent spoofing
+ * - No reply to TCP RST or FIN
+ *
+ * More informations and newer versions at:
+ * <http://npadovano.altervista.com/tarpit.html>
+ */
+
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <net/ip.h>
+#include <net/tcp.h>
+#include <net/icmp.h>
+#include <net/route.h>
+#include <linux/random.h>
+#include <linux/netfilter_ipv4/ip_tables.h>
+
+
+MODULE_DESCRIPTION("TARPIT iptables target, info at: "
+ "http://npadovano.altervista.org/tarpit.html");
+MODULE_AUTHOR("Nicola Padovano <nicola.padovano@gmail.com>");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("xt_TARPIT");
+
+/*
+ * send a packet (to L2 output processing function)
+ *
+ * Note: there is also rtable* argument: it's only
+ * used for debugging reason. Delete it if
+ * you're sure things work.
+ */
+static int ip_direct_send(struct sk_buff* skb)
+{
+ /* get skb dst_entry pointer */
+ struct dst_entry* dst = skb_dst(skb);
+
+ /* if we already sent a packet to our neighbour */
+ if (dst->hh != NULL)
+ return neigh_hh_output(dst->hh,skb);
+
+ /* if there isn't header cache, send our packet with simple output
function */
+ else if (dst->neighbour != NULL)
+ return dst->neighbour->output(skb);
+
+ /* no neighbour, no header cache: it's impossible send the packet */
+ if (net_ratelimit())
+ printk(KERN_DEBUG "TARPIT> ip_direct_send: no neighbour, no header
cache\n");
+
+ kfree_skb(skb);
+
+ return -EINVAL;
+}
+
+/*
+ * tarpit service routine: it accepts a pointer to
+ * sk_buff (the received packet) and its (input)
+ * cache entry.
+ * Note: the received packet it's called "old" in
+ * the comments below. The packet to send
+ * it's called "new".
+ */
+static void tarpit_tcp(const struct sk_buff *oskb,
+ struct rtable *ort)
+{
+ struct iphdr *oiph; /* old ip header pointer */
+ struct tcphdr *otcph; /* old tcp header pointer */
+ unsigned int oiphlen; /* old ip header len */
+ unsigned int otcplen; /* old tcp (payload + header) len */
+
+ struct sk_buff *nskb; /* new skb buffer */
+ struct iphdr *niph; /* new ip header pointer */
+ struct tcphdr *ntcph; /* new tcp header pointer */
+ struct rtable *nrt; /* new routing table */
+
+ struct flowi fl = {}; /* search key used for the cache
lookups */
+
+ u_int32_t tmp32; /* swapping variables*/
+ u_int16_t tmp16;
+
+ /* fill old structs */
+ oiph = ip_hdr(oskb); /* get old ip header pointer */
+ oiphlen = ip_hdrlen(oskb);
+ otcph = (void*)oiph + oiphlen; /* do not use
skb_transport_header! */
+ otcplen = oskb->len - oiphlen;
+
+ /* packets to drop */
+ if ( (oskb->len < oiphlen + sizeof(struct tcphdr))
|| /* truncate tcp header: 1st too fragmented packet...*/
+ (otcph->fin || otcph->rst)
|| /* RST or FIN packet*/
+ (!otcph->ack && !otcph->syn)
|| /* !ACK,!SYN packet */
+ (tcp_v4_check(otcplen, oiph->saddr,
oiph->daddr, /* bad tcp checksum e ip checksum */
+ csum_partial((char *)otcph, otcplen, 0))) ||
+ (oiph->frag_off & htons(IP_OFFSET))
|| /* fragment packets (after the first one) */
+ (!otcph->syn && otcph->ack && !xrlim_allow(&ort->u.dst,HZ))
|| /* rate-limit answer to SYN,ACK packets */
+ (ort == NULL)
|| /* check for an
input routing cache entry */
+ (oskb->pkt_type != PACKET_HOST
&& /* no replies to physical
multi/broadcast */
+ oskb->pkt_type != PACKET_OTHERHOST) ||
+ (ort->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
) /* check IP multi/broadcast */
+ return;
+
+
+ if (!(nskb = skb_copy(oskb, GFP_ATOMIC))) /* we start with a
yet-prepared buffer */
+ return; /* return if there's
an error in copy */
+
+ /* This packet will not be the same as the other: clear nf fields */
+ nf_conntrack_put(nskb->nfct);
+ nskb->nfct = NULL;
+#ifdef CONFIG_NETFILTER_DEBUG
+ nskb->nf_debug = 0;
+#endif
+
+ skb_trim(nskb, ip_hdrlen(nskb) + sizeof(struct tcphdr)); /* trim
skb size to IP LEN + TCP LEN*/
+ niph = (struct iphdr*)skb_network_header(nskb);
+ ntcph = (void*)niph + ip_hdrlen(nskb);
+
+ /* set new ip header */
+ tmp32 = niph->saddr;
+ niph->saddr = niph->daddr;
+ niph->daddr = tmp32;
+
+ niph->tot_len = htons(nskb->len);
+ niph->id = htons(secure_ip_id(niph->daddr));
+
+ niph->frag_off = htons(IP_DF); /* set don't frag flag */
+ niph->ttl = (u32)64;
+
+ niph->check = 0;
+ niph->check = ip_fast_csum((unsigned char*)niph ,niph->ihl);
+
+ /* set new tcp header */
+ tmp16 = ntcph->source;
+ ntcph->source = ntcph->dest;
+ ntcph->dest = tmp16;
+
+ ntcph->doff = sizeof(struct tcphdr)/4; /* 20bytes, doff is a
4bit variable */
+
+ ((u_int8_t*)ntcph)[13] = 0; /* reset all flag */
+ ntcph->urg_ptr = 0;
+
+ /* 1st goal: new tcp connection */
+ if (otcph->syn && !otcph->ack)
+ {
+ ntcph->syn = 1;
+ ntcph->ack = 1;
+ ntcph->seq = htonl(secure_tcp_sequence_number(niph->saddr,
+ niph->daddr,
+ ntcph->source,
+ ntcph->dest));
+ ntcph->ack_seq = htonl(ntohl(otcph->seq) + 1);
+ ntcph->window = htons(5);
+ }
+
+ /* 2nd goal: RST tcp: RST+ACK */
+ else if (otcph->syn && otcph->ack)
+ {
+ ntcph->syn = 0;
+ ntcph->ack = 1;
+ ntcph->rst = 1;
+
+ ntcph->seq = otcph->ack_seq;
+ ntcph->ack_seq = 0;
+ }
+
+ /* 3rd goal: 0 byte window */
+ else if (!otcph->syn && otcph->ack)
+ {
+ ntcph->syn = 0;
+ ntcph->ack = 1;
+ ntcph->window = 0;
+ ntcph->seq = otcph->ack_seq;
+ ntcph->ack_seq = otcph->seq; /* it's saying: "i'm not received
anything" */
+ }
+ else /* unacknowledged packet: drop it*/
+ return;
+
+ ntcph->check = 0;
+ ntcph->check = tcp_v4_check(sizeof(struct tcphdr),
+ niph->saddr,
+ niph->daddr,
+ csum_partial((char *)ntcph,
+ sizeof(struct tcphdr), 0));
+
+
+ /* create a search key to find packet route in cache*/
+ fl.nl_u.ip4_u.daddr = niph->daddr;
+ fl.nl_u.ip4_u.saddr = 0;
+ fl.nl_u.ip4_u.tos = RT_TOS(niph->tos) | RTO_CONN;
+ fl.oif = 0;
+
+ if (ip_route_output_key(&init_net, &nrt, &fl)) /* fill
nrt struct */
+ goto free_nskb; /* exit if
errors in filling*/
+
+ dst_release(skb_dst(nskb));
+ nskb->_skb_refdst = ((unsigned long)&(nrt->u.dst));
+
+ ip_direct_send(nskb); /* send
the packet */
+
+ return;
+
+ free_nskb:
+ kfree_skb(nskb);
+}
+
+
+/*
+ * target function, called everyone the rule is satisfied
+ * standard behaviour: NF_DROP
+ */
+static unsigned int xt_tarpit_target(struct sk_buff *skb,
+ const struct xt_target_param *par)
+{
+ struct rtable *rt = (void *)skb->_skb_refdst;
+ tarpit_tcp(skb,rt);
+ return NF_DROP;
+}
+
+/*
+ * xt_tarpit_check allows only:
+ * 1. raw table & PRE_ROUTING hook or
+ * 2. filter table & (LOCAL_IN or FORWARD) hook
+ * Note: for new kernels version: returns _false_
+ * if checking is OK, otherwise returns _true_
+ */
+static bool xt_tarpit_check(const struct xt_mtchk_param *par)
+{
+ if (!strcmp(par->table, "raw") &&
+ par->hook_mask == NF_INET_PRE_ROUTING)
+ return false;
+
+ if (strcmp(par->table, "filter"))
+ return true;
+
+ return (par->hook_mask & ~((1 << NF_INET_LOCAL_IN) |
+ (1 << NF_INET_FORWARD)));
+}
+
+
+static struct xt_target xt_tar_reg = {
+ .name = "TARPIT", /* target name */
+ .family = AF_INET, /* level 3 protocol */
+ .proto = IPPROTO_TCP, /* we recognize only tcp protocol */
+ .target = xt_tarpit_target, /* pointer to target function */
+ .checkentry = xt_tarpit_check, /* pointer to check-entry function */
+ .me = THIS_MODULE,
+};
+
+/*
+ * initing module function
+ */
+static int __init xt_tarpit_init(void)
+{
+ return xt_register_target(&xt_tar_reg);
+}
+
+/*
+ * delete module
+ */
+static void __exit xt_tarpit_exit(void)
+{
+ xt_unregister_target(&xt_tar_reg);
+}
+
+module_init(xt_tarpit_init);
+module_exit(xt_tarpit_exit);
----libxt_TARPIT.c----
#include <stdio.h>
#include <getopt.h>
#include <xtables.h>
static void tarpit_tg_help(void)
{
printf("TARPIT takes no options\n\n");
}
static int tarpit_tg_parse(int c, char **argv, int invert, unsigned int
*flags,
const void *entry, struct xt_entry_target
**target)
{
return 0;
}
static void tarpit_tg_check(unsigned int flags)
{
}
static struct xtables_target tarpit_tg_reg = {
.version = XTABLES_VERSION,
.name = "TARPIT",
.family = AF_INET,
.help = tarpit_tg_help,
.parse = tarpit_tg_parse,
.final_check = tarpit_tg_check,
};
static __attribute__((constructor)) void tarpit_tg_ldr(void)
{
xtables_register_target(&tarpit_tg_reg);
}
^ permalink raw reply
* Re: [PATCH 4/5] rds: stop including asm-generic/bitops/le.h directly
From: David Miller @ 2010-10-08 18:59 UTC (permalink / raw)
To: andy.grover
Cc: akinobu.mita, linux-kernel, linux-arch, arnd, akpm, rds-devel,
netdev
In-Reply-To: <4CAF693B.9030706@oracle.com>
From: Andy Grover <andy.grover@oracle.com>
Date: Fri, 08 Oct 2010 11:55:55 -0700
> On 10/08/2010 03:03 AM, Akinobu Mita wrote:
>> asm-generic/bitops/le.h is only intended to be included directly from
>> asm-generic/bitops/ext2-non-atomic.h or asm-generic/bitops/minix-le.h
>> which implements generic ext2 or minix bit operations.
>>
>> This stops including asm-generic/bitops/le.h directly and use ext2
>> non-atomic bit operations instead.
>>
>> An alternative approach is introducing little endian bit operations
>> in linux/bitops.h. But it needs to touch more files than this change
>> does.
>
>> - generic___set_le_bit(off, (void *)map->m_page_addrs[i]);
>> + ext2_set_bit(off, (void *)map->m_page_addrs[i]);
>
> I'd like a solution that doesn't have rds calling a function starting
> with ext2_. Seems wrong.
Agreed.
^ permalink raw reply
* Re: [PATCH 4/5] rds: stop including asm-generic/bitops/le.h directly
From: Andy Grover @ 2010-10-08 18:55 UTC (permalink / raw)
To: Akinobu Mita
Cc: linux-kernel, linux-arch, Arnd Bergmann, Andrew Morton, rds-devel,
David S. Miller, netdev
In-Reply-To: <1286532193-29814-4-git-send-email-akinobu.mita@gmail.com>
On 10/08/2010 03:03 AM, Akinobu Mita wrote:
> asm-generic/bitops/le.h is only intended to be included directly from
> asm-generic/bitops/ext2-non-atomic.h or asm-generic/bitops/minix-le.h
> which implements generic ext2 or minix bit operations.
>
> This stops including asm-generic/bitops/le.h directly and use ext2
> non-atomic bit operations instead.
>
> An alternative approach is introducing little endian bit operations
> in linux/bitops.h. But it needs to touch more files than this change does.
> - generic___set_le_bit(off, (void *)map->m_page_addrs[i]);
> + ext2_set_bit(off, (void *)map->m_page_addrs[i]);
I'd like a solution that doesn't have rds calling a function starting
with ext2_. Seems wrong.
Thanks -- Regards -- Andy
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: David Miller @ 2010-10-08 18:50 UTC (permalink / raw)
To: mjg59; +Cc: ellyjones, netdev, jglasgow, msb, olofj
In-Reply-To: <20101008181851.GA27756@srcf.ucam.org>
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Fri, 8 Oct 2010 19:18:51 +0100
> On Fri, Oct 08, 2010 at 11:14:06AM -0700, David Miller wrote:
>
>> I don't think we want a driver in the tree for a device for which the
>> firmware isn't even distributable.
>
> Greg didn't seem to think this was a problem when he merged the serial
> driver.
That's Greg's problem.
I don't see how any distribution can properly support this device
if it cannot work without non-distributable firmware.
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: Matthew Garrett @ 2010-10-08 18:18 UTC (permalink / raw)
To: David Miller; +Cc: ellyjones, netdev, jglasgow, msb, olofj
In-Reply-To: <20101008.111406.246542521.davem@davemloft.net>
On Fri, Oct 08, 2010 at 11:14:06AM -0700, David Miller wrote:
> I don't think we want a driver in the tree for a device for which the
> firmware isn't even distributable.
Greg didn't seem to think this was a problem when he merged the serial
driver.
--
Matthew Garrett | mjg59@srcf.ucam.org
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: David Miller @ 2010-10-08 18:14 UTC (permalink / raw)
To: mjg59; +Cc: ellyjones, netdev, jglasgow, msb, olofj
In-Reply-To: <20101008180310.GA27126@srcf.ucam.org>
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Fri, 8 Oct 2010 19:03:10 +0100
> On Fri, Oct 08, 2010 at 10:28:03AM -0700, David Miller wrote:
>
>> I've looked at the gobi_loader code and it's simpler than many
>> of the gigabit ethernet driver firmware loader sequences in
>> the tree already.
>>
>> Requiring udev et al. magic only makes this networking device
>> that much harder to use from an initrd.
>>
>> I understand how this might be a bit clumsy since we'd need to make a
>> dependency on the serial device since that is the mechanism by which
>> the firmware is uploaded, but really I'd like you to consider it
>> seriously.
>
> The device is capable of functioning in either UMTS or CDMA modes, and
> requires different firmware for each. Further, there's separate firmware
> images per carrier due to regulatory differences. To make things even
> more complicated, there's no consistency in the naming of the firmware
> (and we can't fix that, because the firmware is undistributable). The
> only way to do this in-kernel would seem to be to have a module
> parameter to pass a firmware name, which really doesn't seem elegant.
I don't think we want a driver in the tree for a device for which the
firmware isn't even distributable.
^ permalink raw reply
* Re: [PATCH v2] Add Qualcomm Gobi 2000 driver.
From: Matthew Garrett @ 2010-10-08 18:03 UTC (permalink / raw)
To: David Miller; +Cc: ellyjones, netdev, jglasgow, msb, olofj
In-Reply-To: <20101008.102803.104075511.davem@davemloft.net>
On Fri, Oct 08, 2010 at 10:28:03AM -0700, David Miller wrote:
> I've looked at the gobi_loader code and it's simpler than many
> of the gigabit ethernet driver firmware loader sequences in
> the tree already.
>
> Requiring udev et al. magic only makes this networking device
> that much harder to use from an initrd.
>
> I understand how this might be a bit clumsy since we'd need to make a
> dependency on the serial device since that is the mechanism by which
> the firmware is uploaded, but really I'd like you to consider it
> seriously.
The device is capable of functioning in either UMTS or CDMA modes, and
requires different firmware for each. Further, there's separate firmware
images per carrier due to regulatory differences. To make things even
more complicated, there's no consistency in the naming of the firmware
(and we can't fix that, because the firmware is undistributable). The
only way to do this in-kernel would seem to be to have a module
parameter to pass a firmware name, which really doesn't seem elegant.
--
Matthew Garrett | mjg59@srcf.ucam.org
^ permalink raw reply
* Re: [PATCH] remove leftover rcu_read_unlock calls from __mkroute_output
From: David Miller @ 2010-10-08 17:51 UTC (permalink / raw)
To: eric.dumazet; +Cc: dm, netdev
In-Reply-To: <1286513278.6536.470.camel@edumazet-laptop>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 08 Oct 2010 06:47:58 +0200
> Le jeudi 07 octobre 2010 à 17:48 -0700, Dimitris Michailidis a écrit :
>> 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>
Applied, thanks everyone.
^ permalink raw reply
* Re: [PATCH] net: clear heap allocation for ETHTOOL_GRXCLSRLALL
From: David Miller @ 2010-10-08 17:49 UTC (permalink / raw)
To: bhutchings
Cc: kees.cook, linux-kernel, jgarzik, jeffrey.t.kirsher,
peter.p.waskiewicz.jr, netdev
In-Reply-To: <1286483338.2271.34.camel@achroite.uk.solarflarecom.com>
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 07 Oct 2010 21:28:58 +0100
> On Thu, 2010-10-07 at 13:03 -0700, Kees Cook wrote:
>> 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.
>>
>> Cc: stable@kernel.org
>> Signed-off-by: Kees Cook <kees.cook@canonical.com>
>
> Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Applied and queued up for -stable, thanks.
^ permalink raw reply
* Re: BUG ? ipip unregister_netdevice_many()
From: Daniel Lezcano @ 2010-10-08 17:47 UTC (permalink / raw)
Cc: Eric W. Biederman, Hans Schillstrom, netdev@vger.kernel.org
In-Reply-To: <4CAF54E2.20204@free.fr>
On 10/08/2010 07:29 PM, Daniel Lezcano wrote:
> On 10/08/2010 06:58 PM, Eric W. Biederman wrote:
>> Daniel Lezcano<daniel.lezcano@free.fr> writes:
>>
>>> On 10/08/2010 05:53 PM, Daniel Lezcano wrote:
>>>> On 10/08/2010 02:28 PM, Hans Schillstrom wrote:
>>>>> Hi Eric,
>>>>> Any advice how to trace this down ?
>>>>> This rollback_registered_many() seems to have on the lists before...
>>>>> All IPv4 and IPv6 tunnels causes this crash, all you have to do is
>>>>> load the tunnel module(s)
>>>>> enter a new ns and exit from it.
>>>>>
>>>>> Have not tested any more devices than tunnels,
>>>>> I did an "ip link delete" on my macvlans before exiting the ns.
>>>> Ah ! I succeed to reproduce it.
>>>> It does not appear immediately in fact.
>>>>
>>>> I am trying to simplify the configuration but I am falling in the bug I
>>>> talked about in the previous email.
>>> Ok, so after investigating, we just need a macvlan and specify an
>>> ipv6 address
>>> for it (inside a new netns of course), and the loopback is not
>>> released. I
>>> compiled out the tunnels, so they are not related to this problem I
>>> think.
>>>
>>> That reduces the scope of investigation :)
>> This reproduces the unable to free nedevice problem? Or the bad
>> pointer in macvlan_close problem?
>
> The free netdevice problem.
>
> For the macvlan problem, you have to create 2 macvlans on the *same*
Just to clarify. The problem happens with a different macvlans on
different physical interfaces but it is harder to reproduce.
Sauf indication contraire ci-dessus:
Compagnie IBM France
Siège Social : Tour Descartes, 2, avenue Gambetta, La Défense 5, 92400
Courbevoie
RCS Nanterre 552 118 465
Forme Sociale : S.A.S.
Capital Social : 542.737.118 ?
SIREN/SIRET : 552 118 465 02430
^ permalink raw reply
* Re: [PATCH] net/tg3: simplify conditional
From: Matt Carlson @ 2010-10-08 17:43 UTC (permalink / raw)
To: Nicolas Kaiser
Cc: Matthew Carlson, Michael Chan, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
In-Reply-To: <20101008112927.0ca949cf@absol.kitzblitz>
On Fri, Oct 08, 2010 at 02:29:27AM -0700, Nicolas Kaiser wrote:
> Simplify: ((a && !b) || (!a && b)) => (a != b)
>
> Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
> ---
> drivers/net/tg3.c | 3 +--
> 1 files changed, 1 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
> index 16e1a95..714f0fb 100644
> --- a/drivers/net/tg3.c
> +++ b/drivers/net/tg3.c
> @@ -9967,8 +9967,7 @@ static int tg3_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam
>
> if (!(phydev->supported & SUPPORTED_Pause) ||
> (!(phydev->supported & SUPPORTED_Asym_Pause) &&
> - ((epause->rx_pause && !epause->tx_pause) ||
> - (!epause->rx_pause && epause->tx_pause))))
> + (epause->rx_pause != epause->tx_pause)))
> return -EINVAL;
>
> tp->link_config.flowctrl = 0;
Looks good to me.
^ permalink raw reply
* Re: pull request: wireless-2.6 2010-10-07
From: David Miller @ 2010-10-08 17:38 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, netdev, linux-kernel
In-Reply-To: <20101007185859.GD20950@tuxdriver.com>
From: "John W. Linville" <linville@tuxdriver.com>
Date: Thu, 7 Oct 2010 14:58:59 -0400
> Here are a few more more-or-less-one-liner fixes intended for 2.6.36.
> The one from Felix fixes a regression introduced in the 2.6.36 cycle.
> The one from Johannes fixes a crash reported by Ben Greear (as
> documented in the changelog). The one from me reverts and earlier patch
> from me that can result in stuff in dmesg from not calling
> netif_receive_skb in the proper context.
Pulled, thanks John.
^ permalink raw reply
* Re: [PATCH net-next-2.6] sfc: Don't try to set filters with search depths we know won't work
From: David Miller @ 2010-10-08 17:36 UTC (permalink / raw)
To: bhutchings; +Cc: netdev, linux-net-drivers
In-Reply-To: <1286473831.2271.26.camel@achroite.uk.solarflarecom.com>
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Thu, 07 Oct 2010 18:50:31 +0100
> 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>
Applied, thanks.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox