Jeff Garzik [Wed, 13 Jun 2007 02:27:30 +0000 (22:27 -0400)]
Merge branch 'libertas' of git://git./linux/kernel/git/linville/wireless-2.6 into upstream-fixes
Brice Goglin [Mon, 11 Jun 2007 18:27:07 +0000 (20:27 +0200)]
myri10ge: update driver version
Update myri10ge driver version to 1.3.1-1.248.
Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Brice Goglin [Mon, 11 Jun 2007 18:26:50 +0000 (20:26 +0200)]
myri10ge: report when the link partner is running in Myrinet mode
Since Myri-10G boards may also run in Myrinet mode instead of Ethernet,
add a message when we detect that the link partner is not running in the
right mode.
Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Brice Goglin [Mon, 11 Jun 2007 18:26:31 +0000 (20:26 +0200)]
myri10ge: limit the number of recoveries
Limit the number of recoveries from a NIC hw watchdog reset to 1 by default.
It enables detection of defective NICs immediately since these memory parity
errors are expected to happen very rarely (less than once per century*NIC).
Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Mithlesh Thukral [Mon, 11 Jun 2007 10:30:58 +0000 (03:30 -0700)]
NetXen: Fix link status messages
NetXen: Fix incorrect link status even with switch turned OFF.
NetXen driver failed to accurately indicate when a link is up or down.
This was encountered during failover testing, when the first port
indicated that the link was up even when the 10G switch it was assigned
to in the Bladecenter was turned off completely.
Signed-off by: Wen Xiong <wenxiong@us.ibm.com>
Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Jeff Garzik [Tue, 12 Jun 2007 22:52:31 +0000 (18:52 -0400)]
Revert "[netdrvr e100] experiment with doing RX in a similar manner to eepro100"
This reverts commit
d52df4a35af569071fda3f4eb08e47cc7023f094.
This patch attempted to fix e100 for non-cache coherent memory
architectures by using the cb style code that eepro100 had and using
the EL and s bits from the RFD list. Unfortunately the hardware
doesn't work exactly like this and therefore this patch actually
breaks e100. Reverting the change brings it back to the previously
known good state for 2.6.22. The pending rewrite in progress to this
code can then be safely merged later.
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Dan Williams [Tue, 29 May 2007 04:03:31 +0000 (00:03 -0400)]
[PATCH] libertas: reduce SSID and BSSID mixed-case abuse
Kill mixed case function names from scan.c/scan.h.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Tue, 29 May 2007 03:56:10 +0000 (23:56 -0400)]
[PATCH] libertas: remove WPA_SUPPLICANT structure
Unused.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Tue, 29 May 2007 03:54:55 +0000 (23:54 -0400)]
[PATCH] libertas: remove structure WLAN_802_11_SSID and libertas_escape_essid
Replace WLAN_802_11_SSID with direct 'ssid' and 'ssid_len' members
like ieee80211. In the process, remove private libertas_escape_essid
and depend on the ieee80211 implementation of escape_essid instead.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 03:51:12 +0000 (23:51 -0400)]
[PATCH] libertas: tweak association debug output
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:41:16 +0000 (23:41 -0400)]
[PATCH] libertas: fix big-endian associate command.
Byte-swapping length fields and then passing them to memcpy() considered
harmful.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:40:21 +0000 (23:40 -0400)]
[PATCH] libertas: don't byte-swap firmware version number. It's a byte array.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:39:34 +0000 (23:39 -0400)]
[PATCH] libertas: more endianness fixes, in tx.c this time
Now we finally get connectivity. For a while, before something else dies...
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:38:14 +0000 (23:38 -0400)]
[PATCH] libertas: More endianness fixes.
Now it at least manages to load the firmware.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:36:54 +0000 (23:36 -0400)]
[PATCH] libertas: first pass at fixing up endianness issues
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 03:23:43 +0000 (23:23 -0400)]
[PATCH] libertas: sparse fixes
Fix various issues reported by sparse.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
David Woodhouse [Sat, 26 May 2007 03:15:27 +0000 (23:15 -0400)]
[PATCH] libertas: fix character set in README
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Sat, 26 May 2007 03:12:19 +0000 (23:12 -0400)]
[PATCH] libertas: support for mesh autostart on firmware 5.220.11
Signed-off-by: Luis Carlos Cobo Rus <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Sat, 26 May 2007 03:08:34 +0000 (23:08 -0400)]
[PATCH] libertas: split wext for eth and msh
Separate wireless handlers of mshX and ethX. ethX remains as before.
For mshX, it has been disabled set/get essid, wap and set mode. Get mode always
returns "Repeater" and by now we use the Nickname to show if the mesh is active
("Mesh") or not (empty). The rest remains as before.
Signed-off-by: Luis Carlos Cobo Rus <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Sat, 26 May 2007 03:05:27 +0000 (23:05 -0400)]
[PATCH] libertas: make mac address configuration work with mesh interface too
Signed-off-by: Luis Carlos Cobo Rus <luiscarlos@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Sat, 26 May 2007 03:04:13 +0000 (23:04 -0400)]
[PATCH] libertas: updated readme file
Fix some wording and blinding table command options, clarify argument list for
fwt_add and fwt_list, simplify fwt_list_route
Signed-off-by: Luis Carlos Cobo Rus <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 03:01:24 +0000 (23:01 -0400)]
[PATCH] libertas: Make WPA work through supplicant handshake
Fix WPA so it works up through the supplicant 4-Way handshake process.
Doesn't successfully pass traffic yet; may be problems installing
the GTK to the firmware.
- RSN needs to be enabled before the association command is sent
- Use keys from the association request not the adapter structure
- cmd_act_mac_strict_protection_enable != IW_AUTH_DROP_UNENCRYPTED
- Fix network filtering logic in is_network_compatible() WPA helpers
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:58:55 +0000 (22:58 -0400)]
[PATCH] libertas: add more verbose debugging to libertas_cmd_80211_authenticate
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:54:50 +0000 (22:54 -0400)]
[PATCH] libertas: debug print spacing fixes in assoc.c
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:38:41 +0000 (22:38 -0400)]
[PATCH] libertas: send SIOCGIWSCAN event after partial scans too
Any time the driver gets new scan results, even from partial scans,
it should send the scan event to userspace.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:30:48 +0000 (22:30 -0400)]
[PATCH] libertas: honor specific channel requests during association
Previously if a fixed channel was specified along with an SSID,
the channel request would be ignored during the association
process. Instead, when searching for an adhoc or infrastructure
network to join, allow filtering results based on channel so
that the driver doesn't pick a BSS on a different channel than
requested.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:19:59 +0000 (22:19 -0400)]
[PATCH] libertas: fix default adhoc channel
Inadvertently removed on a previous commit; causes the first
adhoc start to fail if a channel has not been set or no
other association has been made.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 02:18:47 +0000 (22:18 -0400)]
[PATCH] libertas: fix deadlock SIOCGIWSCAN handler
Update signal quality before the locked scan result translation loop,
because calling libertas_prepare_and_send_command() with the
'waitforrsp' option grabs adapter->lock in the command return processing,
leading to the deadlock.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Sat, 26 May 2007 01:59:29 +0000 (21:59 -0400)]
[PATCH] libertas: correct error report paths for wlan_fwt_list_ioctl
Ensure the leave debug print gets triggered when necessary.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 21:33:28 +0000 (17:33 -0400)]
[PATCH] libertas: correctly balance locking in libertas_process_rx_command
adapter->lock should released after unlocking adapter->driver_lock to
balance the order in which they were locked at the top of the function.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 21:29:34 +0000 (17:29 -0400)]
[PATCH] libertas: fix debug enter/leave prints for libertas_execute_next_command
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 21:28:30 +0000 (17:28 -0400)]
[PATCH] libertas: use compare_ether_addr() rather than memcmp() where appropriate
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 21:27:31 +0000 (17:27 -0400)]
[PATCH] libertas: use MAC_FMT and MAC_ARG where appropriate
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 21:09:41 +0000 (17:09 -0400)]
[PATCH] libertas: make association paths consistent
The BSS to associate with (in either Infrastructure or IBSS join
operations) is now stored in _one_ place in the association request (the
bss member), not two places as before (pattemptedbss and
curbssparams->bssdescriptor).
Association requests are passed to the necessary association functions
to (a) give them access to the bss member and (b) ensure that
association/join/start setup uses settings from the request, not the
current adapter settings (which may not be valid for the requested
settings).
Because the 'bss' member of the association request is used now, the
command return functions from associate and adhoc join/start need access
to the in-progress association request to update curbssparams when
everything is done. The association worker moves the request from
pending to in-progress for the duration of the association attempt.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 20:46:33 +0000 (16:46 -0400)]
[PATCH] libertas: move channel changing into association framework
Handle channel changes through the deferred association framework
rather than directly. Fixes errors when setting channels along
with other parameters like mode and SSID.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Marcelo Tosatti [Fri, 25 May 2007 20:28:20 +0000 (16:28 -0400)]
[PATCH] libertas: fix oops on rmmod
Use list_for_each_entry_safe, to protect against list_del().
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Fri, 25 May 2007 20:26:47 +0000 (16:26 -0400)]
[PATCH] libertas: cleanup of fwt_list_route processing
Signed-off-by: Luis Carlos Cobo Rus <luisca@cozybit.com>
Acked-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 20:25:21 +0000 (16:25 -0400)]
[PATCH] libertas: fix 'keep previous scan' behavior
Do not clear the scan list except under specific conditions, such as
when (a) user-requested, or (b) joining/starting an adhoc network.
Furthermore, only clear entries which match the SSID or BSSID of the
request, not the whole scan list.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 20:15:56 +0000 (16:15 -0400)]
[PATCH] libertas: make scan result handling more flexible
- use a linked list for scan results
- age scan results
- pass bss_descriptors around instead of indexes into the scan table
- lock access to the scan results
- stop returning EAGAIN from SIOCGIWSCAN handler
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo [Fri, 25 May 2007 17:53:26 +0000 (13:53 -0400)]
[PATCH] libertas: updated mesh commands for 5.220.9.p11
Updated commands fwt_add and fwt_list, bt_list.
New commands: bt_get_invert, bt_set_invert, to invert the blinding table,
i.e., receive only frames from nodes listed in the BT.
This patch needs/is needed for firmware 5.220.9.p11.
Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo Rus [Fri, 25 May 2007 17:48:54 +0000 (13:48 -0400)]
[PATCH] libertas: version bump (321p0) and cmds update for new fw (5.220.10.p0)
This patch along with the previous commands update one, is necessary for
mesh and fwt ioctls to work properly with firmware version 5.220.10.p0
and later.
Signed-off-by: Luis Carlos Cobo Rus <luiscarlos@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 17:36:57 +0000 (13:36 -0400)]
[PATCH] libertas: don't tear down netdev in libertas_activate_card
libertas_activate_card() doesn't create the netdev, and shouldn't
free it on error. The caller of libertas_activate_card() is
responsible for cleaning up errors from libertas_add_card(),
not libertas_activate_card().
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 17:35:23 +0000 (13:35 -0400)]
[PATCH] libertas: correctly unregister mesh netdev on error
Subject says it all.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 17:16:38 +0000 (13:16 -0400)]
[PATCH] libertas: replace 'macaddress' with 'bssid'
Start to normalize bss_descriptor with ieee80211_network so we can
eventually replace bss_descriptor more easily.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Dan Williams [Fri, 25 May 2007 17:13:25 +0000 (13:13 -0400)]
[PATCH] libertas: call SET_NETDEV_DEV from common code
Move usage of SET_NETDEV_DEV into common code since it has nothing
to do with bus-specific devices. Also fixes a bug where the mesh
device was getting SET_NETDEV_DEV called after register_netdevice,
resulting in no 'device' link in /sys/class/net/mshX/.
Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo [Fri, 25 May 2007 17:11:13 +0000 (13:11 -0400)]
[PATCH] libertas: fixed kernel oops on module/card removal
Fixed kernel oops on module/card removal (using dongles)
Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo [Fri, 25 May 2007 17:10:18 +0000 (13:10 -0400)]
[PATCH] libertas: add URB debug info
Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Luis Carlos Cobo [Fri, 25 May 2007 17:08:33 +0000 (13:08 -0400)]
[PATCH] libertas: fixed incorrect assigment of fcs errors to frag errors
Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 17:05:16 +0000 (13:05 -0400)]
[PATCH] libertas: remove unused variables in wlan_dev_t
Actually, this patch removev wlan_dev_t totally and puts the used variables
of it directly into wlan_private. That reduces one level of indirection and
looks a little bit simpler. It's now "priv->card" and not
"priv->wlan_dev.card" and "priv->dev" instead of "priv->wlan_dev.netdev"
Changed two occurences of "((wlan_private *) dev->priv)->wlan_dev.netdev"
into "dev", because I didn't see the point in doing pointer-ping-pong.
The variables "ioport", "upld_rcv" and "upld_type" where unused. They have
been removed.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:49:10 +0000 (12:49 -0400)]
[PATCH] libertas: let DRV_NAME be overridable
For now, it's "libertas" by default, but that is overwritten in
if_usb.c/if_bootcmd.c and in if_cs.c.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:41:52 +0000 (12:41 -0400)]
[PATCH] libertas: fix RESET logic at unload time
Previously, we had a fixed array of 5 elements where we remembered all
initialized devices. This has been changed to use a "struct list_head"
organization, which is IMHO cleaner.
Also renamed usb_cardp to cardp, as in the reset of the code.
Renamed reset_device() to if_usb_reset_device() like many other functions.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:37:58 +0000 (12:37 -0400)]
[PATCH] libertas: split module into two (libertas.ko and usb8xxx.ko)
* add CONFIG_LIBERTAS to Kconfig
* remove global variable libertas_fw_name, the USB module might want to
use a different default FW name than the CF module, so libertas_fw_name
is now local to if_usb.c
* exported some symbols as GPL
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:18:36 +0000 (12:18 -0400)]
[PATCH] libertas: move contents of fw.h to decl.h
Also removes some useless "extern" declarations from function declaration.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:17:06 +0000 (12:17 -0400)]
[PATCH] libertas: indirect all hardware access via hw_XXXX functions
This functions makes all libertas_sbi_XXX functions static to the
if_usb.c file and renames them to if_usb_XXXX(). The get called from
other places of the source code via priv->hw_XXXX().
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Chris Ball [Fri, 25 May 2007 16:13:24 +0000 (12:13 -0400)]
[PATCH] libertas: wakeup both mesh and normal wakeup when getting out of scan
The previous patch wakes up the mesh device *instead* of the wlan device
when coming out of scan. We need to wake up both of them.
Signed-off-by: Chris Ball <cjb@laptop.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Javier Cardona [Fri, 25 May 2007 16:12:06 +0000 (12:12 -0400)]
[PATCH] libertas: added transmission failures to mesh statistics
Added transmission failures to mesh statistics.
Removed whitespace before newlines.
Signed-off-by: Javier Cardona <javier@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Marcelo Tosatti [Fri, 25 May 2007 16:09:13 +0000 (12:09 -0400)]
[PATCH] libertas: fix error handling of card initialization
Subject says it all.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Javier Cardona [Fri, 25 May 2007 16:06:56 +0000 (12:06 -0400)]
[PATCH] libertas: fixed transmission flow control on the mesh interface
This patch implements proper transmission flow control on mshX.
Signed-off-by: Javier Cardona <javier@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:04:31 +0000 (12:04 -0400)]
[PATCH] libertas: split wlan_add_card()
Split wlan_add_card() into a part that just setups kernel parameters and
into the function libertas_activate_card(), which will implizitly use
hardware functions by the started thread.
This allows us later to do something like this:
priv = libertas_add_card();
priv->hw_command_to_host = if_usb_command_to_host;
priv->hw_xxxx = if_usb_xxxx;
priv->hw_yyyy = if_usb_yyyy;
wlan_activate_card()
and of course the CF driver can set it's own functions.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 16:01:42 +0000 (12:01 -0400)]
[PATCH] libertas: move reset_device() code main.c to if_usb.c
The reset_device() logic is only needed for USB devices, not for CF
devices.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:58:22 +0000 (11:58 -0400)]
[PATCH] libertas: changed some occurences of kmalloc() + memset(&a,0,sz) to kzalloc()
The subject says it all.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:56:37 +0000 (11:56 -0400)]
[PATCH] libertas: fix SSID output
* a newline was missing
* changed %32s to '%s', no need to right justify the ESSID
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:55:19 +0000 (11:55 -0400)]
[PATCH] libertas: get rid of libertas_sbi_get_priv()
It's not really needed, because we can call wlan_remove_card() with
wlan_private* anyway.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:52:42 +0000 (11:52 -0400)]
[PATCH] libertas: change debug output of libertas_interrupt()
It used to be LBS_DEB_MAIN, now it's LBS_DEB_THREAD
Also fixed a missing ":" in lbs_deb_enter()
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:49:19 +0000 (11:49 -0400)]
[PATCH] libertas: single out mesh code
This patches adds the two functions wlan_add_mesh() and wlan_remove_mesh(),
which are responsible for the mshX interface. In a CF driver with a non-
mesh-aware firmware you can omit the calls to this functions.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:32:07 +0000 (11:32 -0400)]
[PATCH] libertas: tune debug code
* renamed module parameter back to libertas_debug
* change from bit shifts to constants, that way it's easier to look at the
source and specify the libertas_debug=0xXXXX module parameter
* moved module_param from fw.c to main.c, where it belongs better
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 15:27:16 +0000 (11:27 -0400)]
[PATCH] libertas: make debug configurable
The debug output of libertas was either not present or it was overwhelming.
This patch adds the possibility to specify a bitmask for the area of
interest. One should then only get the desired output.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 04:37:28 +0000 (00:37 -0400)]
[PATCH] libertas: exclude non-used code when PROC_DEBUG is not set
This reduces usb8xxx.ko by 951 bytes (text) and 256 bytes (data)
when PROC_DEBUG isn't defined.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Marcelo Tosatti [Fri, 25 May 2007 04:33:28 +0000 (00:33 -0400)]
[PATCH] libertas: fix scanning from associate path
The previous scan fix did not account for scan paths other than set_scan()
that need to do a full scan at once.
Add a "full_scan" parameter to wlan_scan_networks() to control such
behaviour.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 04:14:38 +0000 (00:14 -0400)]
[PATCH] libertas: make libertas_wlan_data_rates static
Move libertas_wlan_data_rates into wext.c and make it static. wext.c is the
only user of this array.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 04:11:58 +0000 (00:11 -0400)]
[PATCH] libertas: move vendor & product id's into if_usb.c
For me it looks cleaner, because it removes one level of indirection.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 04:09:54 +0000 (00:09 -0400)]
[PATCH] libertas: remove unused/superfluous definitions of DEV_NAME_LEN
DEV_NAME_LEN is already defined in defs.h and that is sufficient.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 04:07:38 +0000 (00:07 -0400)]
[PATCH] libertas: remove __FILE__ from debug output
Remove filename from debug output because it's way too long.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Marcelo Tosatti [Fri, 25 May 2007 04:04:11 +0000 (00:04 -0400)]
[PATCH] libertas: remove deprecated pm_register and associated code
Subject says it all.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 03:47:34 +0000 (23:47 -0400)]
[PATCH] libertas: fix removal of all debugfs files
rmmod did not remove /sys/kernel/debug/libertas_wireless/eth1/
subscribed_events/high_snr. After I fixed this, I noticed that
it also didn't remove /sys/kernel/debug/libertas_wireless/eth1
as well.
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 03:42:42 +0000 (23:42 -0400)]
[PATCH] libertas: a debug output was missing a newline
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Holger Schurig [Fri, 25 May 2007 03:41:15 +0000 (23:41 -0400)]
[PATCH] libertas: rename wlan_association_worker
Renames wlan_association_worker into libertas_association_worker
Signed-off-by: Holger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Marcelo Tosatti [Fri, 25 May 2007 03:37:28 +0000 (23:37 -0400)]
[PATCH] libertas: scan two channels per scan command
Scan two channels per each command on set_scan(), then bail out and let
get_scan() continue the scanning work up to the last channel.
This gives time to the firmware so it can go back to the association
channel and keep the connection alive.
Fixes http://dev.laptop.org/ticket/841
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Sam Ravnborg [Sat, 9 Jun 2007 22:50:51 +0000 (00:50 +0200)]
net: fix typo in drivers/net/usb/Kconfig
Replace invisible character with a space.
The diff looks like this on my terminal:
- <A0>Choose this option if you're using a host-to-host cable
- <A0>with one of these chips.
+ Choose this option if you're using a host-to-host cable
+ with one of these chips.
Reported by: Massimo Maiurana <maiurana@gmail.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Massimo Maiurana <maiurana@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Kim Phillips [Tue, 5 Jun 2007 10:46:47 +0000 (18:46 +0800)]
phylib: add RGMII-ID mode to the Marvell m88e1111 PHY to fix broken ucc_geth
Support for configuring RGMII-ID (RGMII with internal delay) mode on the
88e1111 and
88e1145. Ucc_geth on MPC8360EMDS(the main user of ucc_geth)
is broken after changed to use phylib. It is fixed by adding this
internal delay.
Also renamed 88e1111s ->
88e1111 (no references to an 88e1111s part were
found), and fixed some whitespace.
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Li Yang <leoli@freescale.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Thomas Klein [Wed, 6 Jun 2007 18:53:16 +0000 (20:53 +0200)]
ehea: Fixed possible kernel panic on VLAN packet recv
This patch fixes a possible kernel panic due to not checking the vlan group
when processing received VLAN packets and a malfunction in VLAN/hypervisor
registration.
Signed-off-by: Thomas Klein <tklein@de.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Mithlesh Thukral [Thu, 7 Jun 2007 11:36:36 +0000 (04:36 -0700)]
NetXen: Fix compile failure seen on PPC architecture
NetXen: Add NETXEN prefixes to macros to clean them up.
This is a cleanup patch which adds NETXEN prefix to some stand
alone macro names.
These posed compile errors when NetXen driver was backported to 2.6.9
on PPC architecture as macros like USER_START are defined in file
arch/ppc64/mm/hash_utils.c
Signed-off-by: Andy Gospodarek <andy@greyhouse.net>
Signed-off by: Wen Xiong <wenxiong@us.ibm.com>
Acked-off by: Mithlesh Thukral <mithlesh@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Mithlesh Thukral [Thu, 7 Jun 2007 11:33:02 +0000 (04:33 -0700)]
NetXen: Fix ping issue after reboot on Blades with 3.4.19 firmware
NetXen: Fix initialization and subsequent ping issue on 3.4.19 firmware
This patch fixes the ping problem seen X/PBlades after the adapter's
firmware was moved to 3.4.19. After configured interface up, ping
failed.
NetXen adapter couldn't accept ARP broadcast packet. Manual addition of
MAC address in the ARP table, made ping work.
NetXen adapter should finish initilization after system boot. But looks
NetXen adapter didn't initilization correctly after system boot up.
So have to re-load the firmware again in probe routine.
Also re-initilization netxen_config_0 and netxen_config_1 registers.
Signed-off by: Wen Xiong <wenxiong@us.ibm.com>
Signed-off by: Mithlesh Thukral <mithlesh@netxen.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Dave Jones [Wed, 6 Jun 2007 07:07:52 +0000 (03:07 -0400)]
typo in via-velocity.c
http://bugzilla.kernel.org/show_bug.cgi?id=8160
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Brian King [Fri, 8 Jun 2007 19:05:17 +0000 (14:05 -0500)]
ibmveth: Automatically enable larger rx buffer pools for larger mtu
Currently, ibmveth maintains several rx buffer pools, which can
be modified through sysfs. By default, pools are not allocated by
default such that jumbo frames cannot be supported without first
activating larger rx buffer pools. This results in failures when attempting
to change the mtu. This patch makes ibmveth automatically allocate
these larger buffer pools when the mtu is changed.
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Brian King [Fri, 8 Jun 2007 19:05:16 +0000 (14:05 -0500)]
ibmveth: Fix h_free_logical_lan error on pool resize
When attempting to activate additional rx buffer pools on an ibmveth interface that
was not yet up, the error below was seen. The patch fixes this by only closing
and opening the interface to activate the resize if the interface is already
opened.
(drivers/net/ibmveth.c:597 ua:
30000004) ERROR: h_free_logical_lan failed with
fffffffffffffffc, continuing with close
Unable to handle kernel paging request for data at address 0x00000ff8
Faulting instruction address: 0xd0000000002540e0
Oops: Kernel access of bad area, sig: 11 [#1]
SMP NR_CPUS=128 NUMA PSERIES LPAR
Modules linked in: ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle ipta
ble_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables i
p6table_filter ip6_tables x_tables ipv6 apparmor aamatch_pcre loop dm_mod ibmvet
h sg ibmvscsic sd_mod scsi_mod
NIP:
D0000000002540E0 LR:
D0000000002540D4 CTR:
80000000001AF404
REGS:
c00000001cd27870 TRAP: 0300 Not tainted (2.6.16.46-0.4-ppc64)
MSR:
8000000000009032 <EE,ME,IR,DR> CR:
24242422 XER:
00000007
DAR:
0000000000000FF8, DSISR:
0000000040000000
TASK =
c00000001ca7b4e0[1636] 'sh' THREAD:
c00000001cd24000 CPU: 0
GPR00:
D0000000002540D4 C00000001CD27AF0 D000000000265650 C00000001C936500
GPR04:
8000000000009032 FFFFFFFFFFFFFFFF 0000000000000007 000000000002C2EF
GPR08:
FFFFFFFFFFFFFFFF 0000000000000000 C000000000652A10 C000000000652AE0
GPR12:
0000000000004000 C0000000004A3300 00000000100A0000 0000000000000000
GPR16:
00000000100B8808 00000000100C0F60 0000000000000000 0000000010084878
GPR20:
0000000000000000 00000000100C0CB0 00000000100AF498 0000000000000002
GPR24:
00000000100BA488 C00000001C936760 D000000000258DD0 C00000001C936000
GPR28:
0000000000000000 C00000001C936500 D000000000265180 C00000001C936000
NIP [
D0000000002540E0] .ibmveth_close+0xc8/0xf4 [ibmveth]
LR [
D0000000002540D4] .ibmveth_close+0xbc/0xf4 [ibmveth]
Call Trace:
[
C00000001CD27AF0] [
D0000000002540D4] .ibmveth_close+0xbc/0xf4 [ibmveth] (unreliable)
[
C00000001CD27B80] [
D0000000002545FC] .veth_pool_store+0xd0/0x260 [ibmveth]
[
C00000001CD27C40] [
C00000000012E0E8] .sysfs_write_file+0x118/0x198
[
C00000001CD27CF0] [
C0000000000CDAF0] .vfs_write+0x130/0x218
[
C00000001CD27D90] [
C0000000000CE52C] .sys_write+0x4c/0x8c
[
C00000001CD27E30] [
C00000000000871C] syscall_exit+0x0/0x40
Instruction dump:
419affd8 2fa30000 419e0020 e93d0000 e89e8040 38a00255 e87e81b0 80c90018
48001531 e8410028 e93d00e0 7fa3eb78 <
e8090ff8>
f81d0430 4bfffdc9 38210090
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Linus Torvalds [Sat, 9 Jun 2007 02:44:16 +0000 (19:44 -0700)]
Merge branch 'upstream-linus' of git://git./linux/kernel/git/mfasheh/ocfs2
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2:
ocfs2: Fix invalid assertion during write on 64k pages
ocfs2: Fix masklog breakage
Linus Torvalds [Sat, 9 Jun 2007 01:15:49 +0000 (18:15 -0700)]
Merge branch 'master' of /linux/kernel/git/davem/net-2.6
* 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
[CIPSO]: Fix several unaligned kernel accesses in the CIPSO engine.
[NetLabel]: consolidate the struct socket/sock handling to just struct sock
[IPV4]: Do not remove idev when addresses are cleared
Linus Torvalds [Sat, 9 Jun 2007 01:15:23 +0000 (18:15 -0700)]
Merge branch 'master' of /linux/kernel/git/davem/sparc-2.6
* 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6:
[SPARC64]: Handle PCI bridges without 'ranges' property.
[SPARC64]: Include <linux/rwsem.h> instead of <asm/rwsem.h>.
Linus Torvalds [Sat, 9 Jun 2007 01:14:42 +0000 (18:14 -0700)]
Merge /pub/scm/linux/kernel/git/gregkh/usb-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6:
OHCI: Fix machine check in ohci_hub_status_data
USB: Fix up bogus bInterval values in endpoint descriptors
USB: cxacru: ignore error trying to start ADSL in atm_start
USB: cxacru: create sysfs attributes in atm_start instead of bind
USB: cxacru: add Documentation file
USB: UNUSUAL_DEV: Sync up some reported devices from Ubuntu
USB: usb gadgets avoid le{16,32}_to_cpup()
usblp: Don't let suspend to kill ->used
USB: set default y for CONFIG_USB_DEVICE_CLASS
Linus Torvalds [Sat, 9 Jun 2007 01:11:47 +0000 (18:11 -0700)]
Merge /pub/scm/linux/kernel/git/gregkh/driver-2.6
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6:
firmware: remove orphaned Email
kobject: use the proper printk level for kobject error
Driver core: kill unused code
Driver core: keep PHYSDEV for old struct class_device
update Documentation/driver-model/platform.txt
Robert P. J. Day [Fri, 8 Jun 2007 20:47:07 +0000 (13:47 -0700)]
Protect <linux/console_struct.h> from multiple inclusion
Prevent <linux/console_struct.h> from being included more than once,
otherwise you get a redefinition error if you happen to include
<linux/vt_kern.h> first.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andy Whitcroft [Fri, 8 Jun 2007 20:47:06 +0000 (13:47 -0700)]
update checkpatch.pl to version 0.04
This version brings a some new tests, and a host of changes to fix
false positives, of particular note:
- check for and report #if 0
- extend checking of line lengths and spacing for .pl, .sh etc
- extends the pointer type checks to multiple levels
- updates printk handling to track newlines
- adds a wrapped patch detector
- drops the leading component of the filenames
- extends switch indent handling to switch statmentes rooted in
the context
- adds foo * bar single pointer checks
This version of checkpatch.pl can be found at the following URL:
http://www.shadowen.org/~apw/public/checkpatch/checkpatch.pl-0.04
Full Changelog:
Andy Whitcroft (16):
allow checking line lengths and spacing on other source files
clean up that whitespace
sanitise the input line standardising the content of quotes
clean up pointer type * and space checks
fix up the sanitiser so it maintains the line length
apply the printk facility checks only to the first printk in a set
switch/case indent checks may anchor in the context
add a wrapped patch detector
put the #ifdef in C file checks on ice
asm volatile is acceptable
check for and report #if 0
drop the leading component of the filename as patches are -p1
use the original line when reporting operator errors
correct spelling of Joel's name
Version: 0.04
add support for struct foo * bar checks
Geert Uytterhoeven (1):
Fix checkpatch.pl name in usage template
Randy Dunlap (1):
checkpatch: produce fewer lines of output
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Joel Schopp <jschopp@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Randy Dunlap [Fri, 8 Jun 2007 20:47:04 +0000 (13:47 -0700)]
hexdump: more output formatting
Add a prefix string parameter. Callers are responsible for any string
length/alignment that they want to see in the output. I.e., callers should
pad strings to achieve alignment if they want that.
Add rowsize parameter. This is the number of raw data bytes to be printed
per line. Must be 16 or 32.
Add a groupsize parameter. This allows callers to dump values as 1-byte,
2-byte, 4-byte, or 8-byte numbers. Default is 1-byte numbers. If the
total length is not an even multiple of groupsize, 1-byte numbers are
printed.
Add an "ascii" output parameter. This causes ASCII data output following
the hex data output.
Clean up some doc examples.
Align the ASCII output on all lines that are produced by one call.
Add a new interface, print_hex_dump_bytes(), that is a shortcut to
print_hex_dump(), using default parameter values to print 16 bytes in
byte-size chunks of hex + ASCII output, using printk level KERN_DEBUG.
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Cox [Fri, 8 Jun 2007 20:47:03 +0000 (13:47 -0700)]
MAINTAINERS: corrections
The UFS entry was misformatted
The NEC V850 links are all broken
The Berkshire watchdog links are all broken
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Randy Dunlap [Fri, 8 Jun 2007 20:47:03 +0000 (13:47 -0700)]
checkpatch: produce fewer lines of output
Produce one less line of output per flagged incident.
Change this:
use tabs not spaces
PATCH: /home/rddunlap/arcmsr1200014.patch4:756:
FILE: b/drivers/scsi/arcmsr/arcmsr_hba.c:1843:
+ return PCI_ERS_RESULT_NEED_RESET;$
to this:
use tabs not spaces
#756: FILE: b/drivers/scsi/arcmsr/arcmsr_hba.c:1843:
+ return PCI_ERS_RESULT_NEED_RESET;$
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jeff Mahoney [Fri, 8 Jun 2007 20:47:02 +0000 (13:47 -0700)]
reiserfs: mailing list has moved
This patch changes MAINTAINERS to reflect the new location of the reiserfs
development mailing list. The old list forwards to the new one.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Morton [Fri, 8 Jun 2007 20:47:01 +0000 (13:47 -0700)]
x86_64: oops_begin() fix
We don't want to see this:
> BUG: using smp_processor_id() in preemptible [
00000001] code: bash/3857
> caller is oops_begin+0xb/0x6f
>
> Call Trace:
> [<
ffffffff8020ab4d>] show_trace+0x34/0x4f
> [<
ffffffff8020ab7a>] dump_stack+0x12/0x17
> [<
ffffffff8030d92d>] debug_smp_processor_id+0xad/0xbc
> [<
ffffffff8042388f>] oops_begin+0xb/0x6f
> [<
ffffffff8042520b>] do_page_fault+0x66a/0x7c0
> [<
ffffffff804234bd>] error_exit+0x0/0x84
>
coming out when the kernel is trying to oops.
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Bob Picco [Fri, 8 Jun 2007 20:47:00 +0000 (13:47 -0700)]
fix sysrq-m oops
We aren't sampling for holes in memory. Thus we encounter a section hole
with empty section map pointer for SPARSEMEM and OOPs for show_mem. This
issue has been seen in 2.6.21, current git and current mm. The patch below
is for mainline and mm. It was boot tested for SPARSEMEM, current VMEMMAP
of Andy's in mm ml and DISCONTIGMEM. A slightly different patch will be
posted to stable for 2.6.21.
Previous to commit
f0a5a58aa812b31fd9f197c4ba48245942364eae memory_present
was called for node_start_pfn to node_end_pfn. This would cover the
hole(s) with reserved pages and valid sections. Most SPARSEMEM supported
arches do a pfn_valid check in show_mem before computing the page structure
address.
This issue was brought to my attention on IRC by Arnaldo Carvalho de Melo.
Thanks to Arnaldo for testing.
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Chuck Ebbert <cebbert@redhat.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alexey Kuznetsov [Fri, 8 Jun 2007 20:47:00 +0000 (13:47 -0700)]
pi-futex: fix exit races and locking problems
1. New entries can be added to tsk->pi_state_list after task completed
exit_pi_state_list(). The result is memory leakage and deadlocks.
2. handle_mm_fault() is called under spinlock. The result is obvious.
3. results in self-inflicted deadlock inside glibc.
Sometimes futex_lock_pi returns -ESRCH, when it is not expected
and glibc enters to for(;;) sleep() to simulate deadlock. This problem
is quite obvious and I think the patch is right. Though it looks like
each "if" in futex_lock_pi() got some stupid special case "else if". :-)
4. sometimes futex_lock_pi() returns -EDEADLK,
when nobody has the lock. The reason is also obvious (see comment
in the patch), but correct fix is far beyond my comprehension.
I guess someone already saw this, the chunk:
if (rt_mutex_trylock(&q.pi_state->pi_mutex))
ret = 0;
is obviously from the same opera. But it does not work, because the
rtmutex is really taken at this point: wake_futex_pi() of previous
owner reassigned it to us. My fix works. But it looks very stupid.
I would think about removal of shift of ownership in wake_futex_pi()
and making all the work in context of process taking lock.
From: Thomas Gleixner <tglx@linutronix.de>
Fix 1) Avoid the tasklist lock variant of the exit race fix by adding
an additional state transition to the exit code.
This fixes also the issue, when a task with recursive segfaults
is not able to release the futexes.
Fix 2) Cleanup the lookup_pi_state() failure path and solve the -ESRCH
problem finally.
Fix 3) Solve the fixup_pi_state_owner() problem which needs to do the fixup
in the lock protected section by using the in_atomic userspace access
functions.
This removes also the ugly lock drop / unqueue inside of fixup_pi_state()
Fix 4) Fix a stale lock in the error path of futex_wake_pi()
Added some error checks for verification.
The -EDEADLK problem is solved by the rtmutex fixups.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>