* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: Paul Moore @ 2013-04-08 21:09 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, netdev, mvadkert
In-Reply-To: <1365454501.3887.45.camel@edumazet-glaptop>
On Monday, April 08, 2013 01:55:01 PM Eric Dumazet wrote:
> On Mon, 2013-04-08 at 16:37 -0400, Paul Moore wrote:
> > The people who use this functionality almost never use upstream kernels,
> > they need to protection/certification/warm-fuzzies/etc. that come from a
> > distribution kernel and a support infrastructure. I didn't catch it
> > because I use a slightly different configuration that didn't expose this
> > bug; while I would like to run a full regression test every release I
> > simply don't have the time to do that myself.
> >
> > > This sounds like a very small issue to me, a revert is simply overkill.
> >
> > It all depends on your use case. To you, whom I assume doesn't use
> > SELinux, it is indeed a trivial issue. To someone who relies on SELinux
> > for its network access controls this is a pretty significant issue.
>
> Is the patch I sent addressing the problem or not ?
>
> Note that I do have : CONFIG_SECURITY=y
>
> So this patch basically adds the overhead back, and I'll have to use
> real hook later in net-next.
Please repost the patch to the LSM list, it needs to be discussed there.
> At least my patch clearly _shows_ the security requirement, instead of
> relying on a side effect of a previous sock_wmalloc()
I don't see it as a side effect, and as far as demonstration, I think the
SELinux network access controls in their entirety shows the security
requirement. If we want to make the security requirements even more explicit
in the networking stack, let's add a security blob to the sk_buff and allow
some proper LSM hooks.
> Again, it would be nice you understand the plan.
I have no idea what the above sentence is trying to say.
--
paul moore
security and virtualization @ redhat
^ permalink raw reply
* Re: unix STREAM/SEQPACKET sockets and SO_PASSCRED race
From: David Miller @ 2013-04-08 21:02 UTC (permalink / raw)
To: kay; +Cc: netdev
In-Reply-To: <CAPXgP106kvseQ_0ATmY38vPk1Sw+Y6+GBosJUcJ2YjfZKrHfzA@mail.gmail.com>
From: Kay Sievers <kay@vrfy.org>
Date: Mon, 8 Apr 2013 22:55:28 +0200
> If not, could we inherit/copy over the PASSCRED bit from the listen
> socket to the connection socket. Servers could set the flag there to
> receive it for the accepted connections?
We could start doing that yes, this is similar to what we did for
things like non-nonblocking flags on TCP sockets a long time ago.
^ permalink raw reply
* Re: [PATCH net-next] net: cdc_ncm: demote "unexpected notification" to debug level
From: David Miller @ 2013-04-08 20:58 UTC (permalink / raw)
To: bjorn-yOkvZcmFvRU
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-usb-u79uwXL29TY76Z2rM5mHXA,
alexey.orishko-0IS4wlFg1OjSUeElwK9/Pw
In-Reply-To: <1365445583-15124-1-git-send-email-bjorn-yOkvZcmFvRU@public.gmane.org>
From: Bjørn Mork <bjorn-yOkvZcmFvRU@public.gmane.org>
Date: Mon, 8 Apr 2013 20:26:23 +0200
> Receiving unhandled notifications is most certainly not an error
> and should not be logged as one. Knowing that the device sends
> notifications we don't handle is useful for developers, but there
> is very little a user can do about this. The message is therefore
> just annoying noise to most users with devices sending unhandled
> notifications like e.g. USB_CDC_NOTIFY_RESPONSE_AVAILABLE
>
> Cc: Alexey Orishko <alexey.orishko-0IS4wlFg1OjSUeElwK9/Pw@public.gmane.org>
> Signed-off-by: Bjørn Mork <bjorn-yOkvZcmFvRU@public.gmane.org>
Applied.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" 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] Documentation: cgroup: add documentation for net_cls cgroups.
From: David Miller @ 2013-04-08 20:58 UTC (permalink / raw)
To: ramirose-Re5JQEeQqe8AvxtiuMwx3w
Cc: cgroups-u79uwXL29TY76Z2rM5mHXA, tj-DgEjT+Ai2ygdnm+yROfE0A,
lizefan-hv44wF8Li93QT0dZR+AlfA, netdev-u79uwXL29TY76Z2rM5mHXA,
tgraf-G/eBtMaohhA
In-Reply-To: <1365429860-3184-1-git-send-email-ramirose-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
From: Rami Rosen <ramirose-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date: Mon, 8 Apr 2013 17:04:20 +0300
> This patch adds a new file, Documentation/cgroups/net_cls.txt, with info
> about net_cls cgroups, and updates the 00-INDEX accordingly.
>
> Signed-off-by: Rami Rosen <ramirose-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Applied.
^ permalink raw reply
* Re: [PATCH net-next] net: ipv6: add tokenized interface identifier support
From: David Miller @ 2013-04-08 20:57 UTC (permalink / raw)
To: dborkman; +Cc: netdev, hannes, yoshfuji, tgraf
In-Reply-To: <1365429690-17342-1-git-send-email-dborkman@redhat.com>
From: Daniel Borkmann <dborkman@redhat.com>
Date: Mon, 8 Apr 2013 16:01:30 +0200
> This patch adds support for IPv6 tokenized IIDs, that allow
> for administrators to assign well-known host-part addresses
> to nodes whilst still obtaining global network prefix from
> Router Advertisements. It is currently in draft status.
>
> The primary target for such support is server platforms
> where addresses are usually manually configured, rather
> than using DHCPv6 or SLAAC. By using tokenised identifiers,
> hosts can still determine their network prefix by use of
> SLAAC, but more readily be automatically renumbered should
> their network prefix change. [...]
>
> The disadvantage with static addresses is that they are
> likely to require manual editing should the network prefix
> in use change. If instead there were a method to only
> manually configure the static identifier part of the IPv6
> address, then the address could be automatically updated
> when a new prefix was introduced, as described in [RFC4192]
> for example. In such cases a DNS server might be
> configured with such a tokenised interface identifier of
> ::53, and SLAAC would use the token in constructing the
> interface address, using the advertised prefix. [...]
>
> http://tools.ietf.org/html/draft-chown-6man-tokenised-ipv6-identifiers-02
>
> The implementation is partially based on top of Mark K.
> Thompson's proof of concept. However, it uses the Netlink
> interface for configuration resp. data retrival, so that
> it can be easily extended in future. Successfully tested
> by myself.
>
> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
> Cc: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
> Cc: Thomas Graf <tgraf@suug.ch>
> Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Applied.
^ permalink raw reply
* Re: [net-next.git 0/5] stmmac code tidy-up (v2)
From: David Miller @ 2013-04-08 20:57 UTC (permalink / raw)
To: peppe.cavallaro; +Cc: netdev, eric.dumazet, bhutchings, shiraz.hashim
In-Reply-To: <1365423003-23959-1-git-send-email-peppe.cavallaro@st.com>
From: Giuseppe CAVALLARO <peppe.cavallaro@st.com>
Date: Mon, 8 Apr 2013 14:09:58 +0200
> These patches review and tidy-up the driver to respect the codying-style rules
> where there were some problems.
> Also now kernel-doc passes w/o any warnings.
> These review the stmmac.txt where some new fields have not been
> updated after adding new supports.
> There is a patch to review the private structure trying to reorganize all the fields
> in order to keep adjacent fields for cache effect (this patch is the V2 because
> already discussed in the mailing list). I have seen that there are not
> regressions when test on UP and SMP systems indeed.
> In these patches I have removed two old patches for memory barrier and GRO.
> We discussed about these in the mainling list and these had to be discarded.
Series applied, thanks.
^ permalink raw reply
* Re: [PATCH net-next RESEND] stmmac: modified pcs mode support for SGMII
From: David Miller @ 2013-04-08 20:57 UTC (permalink / raw)
To: bh74.an; +Cc: netdev, peppe.cavallaro, kgene.kim, cpgs
In-Reply-To: <015701ce340d$052a71d0$0f7f5570$%an@samsung.com>
From: Byungho An <bh74.an@samsung.com>
Date: Mon, 08 Apr 2013 12:56:16 +0900
>
> This patch modifies the pcs mode support for SGMII. Even though
> SGMII does auto-negotiation with phy, it needs stmmac_init_phy and
> stmmac_mdio_register function for initializing phy.
>
> Signed-off-by: Byungho An <bh74.an@samsung.com>
Applied.
^ permalink raw reply
* Re: [PATCH] net: of_mdio: scan mdiobus for PHYs without reg property
From: David Miller @ 2013-04-08 20:56 UTC (permalink / raw)
To: sebastian.hesselbarth-Re5JQEeQqe8AvxtiuMwx3w
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ,
linux-kernel-u79uwXL29TY76Z2rM5mHXA,
rob.herring-bsGFqQB8/DxBDgjK7y7TUQ,
florian-p3rKhJxN3npAfugRpC6u6w
In-Reply-To: <1365332988-9053-2-git-send-email-sebastian.hesselbarth-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
From: Sebastian Hesselbarth <sebastian.hesselbarth-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date: Sun, 7 Apr 2013 13:09:48 +0200
> Using DT for mdiobus and ethernet-phy requires to know the PHY address, which
> is hard to guess if you don't know it. This patch extends of_mdiobus_register
> to scan mdiobus for PHYs if reg property of the corresponding node is not set.
> This also allows to have phy nodes in SoC DT files where the reg property can
> be overwritten in the board file later. To encourage people to finally set the
> actual phy address, the mdiobus scan is noisier than required.
>
> Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Applied.
^ permalink raw reply
* Re: [PATCH] net: mvmdio: get and enable optional clock
From: David Miller @ 2013-04-08 20:56 UTC (permalink / raw)
To: sebastian.hesselbarth
Cc: florian, thomas.petazzoni, gregkh, netdev, linux-kernel
In-Reply-To: <1365332988-9053-1-git-send-email-sebastian.hesselbarth@gmail.com>
From: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Date: Sun, 7 Apr 2013 13:09:47 +0200
> Marvell mdio driver uses internal registers that can be clock gated on
> some SoCs. This patch just adds optional clock handling, to allow to pass
> and enable the corresponding clock.
>
> Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH 2/2] net: ks8851: Use dev_pm_ops
From: David Miller @ 2013-04-08 20:56 UTC (permalink / raw)
To: lars; +Cc: ben-linux, netdev, linux-kernel
In-Reply-To: <1365327184-22956-2-git-send-email-lars@metafoo.de>
From: Lars-Peter Clausen <lars@metafoo.de>
Date: Sun, 7 Apr 2013 11:33:04 +0200
> Use dev_pm_ops instead of the deprecated legacy suspend/resume callbacks.
>
> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Applied.
^ permalink raw reply
* Re: [PATCH 1/2] net: ks8851: Use module_spi_driver
From: David Miller @ 2013-04-08 20:56 UTC (permalink / raw)
To: lars; +Cc: ben-linux, netdev, linux-kernel
In-Reply-To: <1365327184-22956-1-git-send-email-lars@metafoo.de>
From: Lars-Peter Clausen <lars@metafoo.de>
Date: Sun, 7 Apr 2013 11:33:03 +0200
> By using module_spi_driver we can eliminate a few lines of boilerplate code.
>
> Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Applied.
^ permalink raw reply
* unix STREAM/SEQPACKET sockets and SO_PASSCRED race
From: Kay Sievers @ 2013-04-08 20:55 UTC (permalink / raw)
To: netdev
While checking the issues caused by:
"af_unix: dont send SCM_CREDENTIAL when dest socket is NULL"
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=14134f6584212d585b310ce95428014b653dfaf6
which was later reverted by:
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=25da0e3e9d3fb2b522bc2a598076735850310eb1
Seems, if messages are sent by the client during the window between
accept() and setsockopt(SO_PASSCRED) on the server, the kernel will
not attach the credentials to the message.
The original change opened the race window significantly, but it seems
the window existed even before the change, it's just hard to hit.
Is there a way to race-free set up a connection on the server side
which has scm credentials enabled? If scm credential handling should
only be needed to be requested on the server side, is there any way to
do that?
If not, could we inherit/copy over the PASSCRED bit from the listen
socket to the connection socket. Servers could set the flag there to
receive it for the accepted connections?
Thanks,
Kay
^ permalink raw reply
* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: Eric Dumazet @ 2013-04-08 20:55 UTC (permalink / raw)
To: Paul Moore; +Cc: David Miller, netdev, mvadkert
In-Reply-To: <3294227.D2rod7xgQB@sifl>
On Mon, 2013-04-08 at 16:37 -0400, Paul Moore wrote:
> The people who use this functionality almost never use upstream kernels, they
> need to protection/certification/warm-fuzzies/etc. that come from a
> distribution kernel and a support infrastructure. I didn't catch it because I
> use a slightly different configuration that didn't expose this bug; while I
> would like to run a full regression test every release I simply don't have the
> time to do that myself.
>
> > This sounds like a very small issue to me, a revert is simply overkill.
>
> It all depends on your use case. To you, whom I assume doesn't use SELinux,
> it is indeed a trivial issue. To someone who relies on SELinux for its
> network access controls this is a pretty significant issue.
>
Is the patch I sent addressing the problem or not ?
Note that I do have : CONFIG_SECURITY=y
So this patch basically adds the overhead back, and I'll have to use
real hook later in net-next.
At least my patch clearly _shows_ the security requirement, instead of
relying on a side effect of a previous sock_wmalloc()
Again, it would be nice you understand the plan.
^ permalink raw reply
* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: Paul Moore @ 2013-04-08 20:53 UTC (permalink / raw)
To: David Miller, eric.dumazet; +Cc: netdev, mvadkert
In-Reply-To: <20130408.164439.2282891058177074789.davem@davemloft.net>
On Monday, April 08, 2013 04:44:39 PM David Miller wrote:
> From: Paul Moore <pmoore@redhat.com>
> Date: Mon, 08 Apr 2013 16:37:22 -0400
>
> > On Monday, April 08, 2013 11:30:25 AM Eric Dumazet wrote:
> >> On Mon, 2013-04-08 at 11:21 -0700, Eric Dumazet wrote:
> >> > On Mon, 2013-04-08 at 14:12 -0400, Paul Moore wrote:
> >> > > It seems a bit fragile to me, perhaps even hacky, but in some ways I
> >> > > guess it isn't anymore fragile than relying on skb->sk - as this
> >> > > problem demonstrates. My other concern is that adding this hook
> >> > > *correctly* is likely to touch a lot of files and may be a bit much
> >> > > so late in the 3.9 cycle, Dave, what say you?>
> >> >
> >> > I don't get it, 90ba9b1986b5ac4b2d18 was in 3.6, why do you care of
> >> > 3.9 ?
> >> >
> >> > I am preparing a fix right now. Not a revert, thank you.
> >>
> >> Is the following patch not good enough ?
> >
> > I think it is somewhat telling that the hook you're proposing doesn't ever
> > make any calls into any of the individual LSMs, it only calls back into
> > the networking stack. In my mind, this makes it an abuse of the LSM
> > mechanism.
>
> Without LSMs the socket reference is spurious and pointless overhead,
> therefore the only acceptable fix one which only takes the socket
> reference when there are LSMs with a need.
Well that's frustrating, but if that is the game so be it.
Eric, your patch didn't cross-post to the LSM list, do you want to post it
there for review/merging? I would recommend also adding a pointer to this
thread for reference.
--
paul moore
security and virtualization @ redhat
^ permalink raw reply
* Re: [PATCH 1/2] crypto: add CMAC support to CryptoAPI
From: David Miller @ 2013-04-08 20:49 UTC (permalink / raw)
To: herbert; +Cc: steffen.klassert, jussi.kivilinna, linux-crypto, netdev, tstdenis
In-Reply-To: <20130408093340.GA18594@gondor.apana.org.au>
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 8 Apr 2013 17:33:40 +0800
> On Mon, Apr 08, 2013 at 10:24:16AM +0200, Steffen Klassert wrote:
>> On Mon, Apr 08, 2013 at 10:48:44AM +0300, Jussi Kivilinna wrote:
>> > Patch adds support for NIST recommended block cipher mode CMAC to CryptoAPI.
>> >
>> > This work is based on Tom St Denis' earlier patch,
>> > http://marc.info/?l=linux-crypto-vger&m=135877306305466&w=2
>> >
>> > Cc: Tom St Denis <tstdenis@elliptictech.com>
>> > Signed-off-by: Jussi Kivilinna <jussi.kivilinna@iki.fi>
>>
>> This patch does not apply clean to the ipsec-next tree
>> because of some crypto changes I don't have in ipsec-next.
>> The IPsec part should apply to the cryptodev tree,
>> so it's probaply the best if we route this patchset
>> through the cryptodev tree.
>>
>> Herbert,
>>
>> are you going to take these patches?
>
> Sure I can do that.
I'm fine with this:
Acked-by: David S. Miller <davem@davemloft.net>
^ permalink raw reply
* Re: [PATCH 2/2] bonding: fix bonding_masters race condition in bond unloading
From: David Miller @ 2013-04-08 20:45 UTC (permalink / raw)
To: nikolay; +Cc: netdev, andy, fubar, vfalico
In-Reply-To: <1365245678-4455-2-git-send-email-nikolay@redhat.com>
From: Nikolay Aleksandrov <nikolay@redhat.com>
Date: Sat, 6 Apr 2013 12:54:38 +0200
> While the bonding module is unloading, it is considered that after
> rtnl_link_unregister all bond devices are destroyed but since no
> synchronization mechanism exists, a new bond device can be created
> via bonding_masters before unregister_pernet_subsys which would
> lead to multiple problems (e.g. NULL pointer dereference, wrong RIP,
> list corruption).
>
> This patch fixes the issue by removing any bond devices left in the
> netns after bonding_masters is removed from sysfs.
>
> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Applied.
^ permalink raw reply
* Re: [PATCH 1/2] Revert "bonding: remove sysfs before removing devices"
From: David Miller @ 2013-04-08 20:45 UTC (permalink / raw)
To: nikolay; +Cc: netdev, andy, fubar, vfalico
In-Reply-To: <1365245678-4455-1-git-send-email-nikolay@redhat.com>
From: Nikolay Aleksandrov <nikolay@redhat.com>
Date: Sat, 6 Apr 2013 12:54:37 +0200
> This reverts commit 4de79c737b200492195ebc54a887075327e1ec1d.
>
> This patch introduces a new bug which causes access to freed memory.
> In bond_uninit: list_del(&bond->bond_list);
> bond_list is linked in bond_net's dev_list which is freed by
> unregister_pernet_subsys.
>
> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>
Applied.
^ permalink raw reply
* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: David Miller @ 2013-04-08 20:44 UTC (permalink / raw)
To: pmoore; +Cc: eric.dumazet, netdev, mvadkert
In-Reply-To: <3294227.D2rod7xgQB@sifl>
From: Paul Moore <pmoore@redhat.com>
Date: Mon, 08 Apr 2013 16:37:22 -0400
> On Monday, April 08, 2013 11:30:25 AM Eric Dumazet wrote:
>> On Mon, 2013-04-08 at 11:21 -0700, Eric Dumazet wrote:
>> > On Mon, 2013-04-08 at 14:12 -0400, Paul Moore wrote:
>> > > It seems a bit fragile to me, perhaps even hacky, but in some ways I
>> > > guess it isn't anymore fragile than relying on skb->sk - as this
>> > > problem demonstrates. My other concern is that adding this hook
>> > > *correctly* is likely to touch a lot of files and may be a bit much so
>> > > late in the 3.9 cycle, Dave, what say you?>
>> > I don't get it, 90ba9b1986b5ac4b2d18 was in 3.6, why do you care of
>> > 3.9 ?
>> >
>> > I am preparing a fix right now. Not a revert, thank you.
>>
>> Is the following patch not good enough ?
>
> I think it is somewhat telling that the hook you're proposing doesn't ever
> make any calls into any of the individual LSMs, it only calls back into the
> networking stack. In my mind, this makes it an abuse of the LSM mechanism.
Without LSMs the socket reference is spurious and pointless overhead,
therefore the only acceptable fix one which only takes the socket
reference when there are LSMs with a need.
^ permalink raw reply
* Re: pull request: wireless 2013-04-08
From: David Miller @ 2013-04-08 20:41 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, netdev, linux-kernel
In-Reply-To: <20130408192102.GB25564@tuxdriver.com>
From: "John W. Linville" <linville@tuxdriver.com>
Date: Mon, 8 Apr 2013 15:21:02 -0400
> Please consider this set of fixes for the 3.9 stream...
...
> git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless.git for-davem
Pulled, thanks John.
^ permalink raw reply
* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: Paul Moore @ 2013-04-08 20:37 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, netdev, mvadkert
In-Reply-To: <1365445825.3887.35.camel@edumazet-glaptop>
On Monday, April 08, 2013 11:30:25 AM Eric Dumazet wrote:
> On Mon, 2013-04-08 at 11:21 -0700, Eric Dumazet wrote:
> > On Mon, 2013-04-08 at 14:12 -0400, Paul Moore wrote:
> > > It seems a bit fragile to me, perhaps even hacky, but in some ways I
> > > guess it isn't anymore fragile than relying on skb->sk - as this
> > > problem demonstrates. My other concern is that adding this hook
> > > *correctly* is likely to touch a lot of files and may be a bit much so
> > > late in the 3.9 cycle, Dave, what say you?>
> > I don't get it, 90ba9b1986b5ac4b2d18 was in 3.6, why do you care of
> > 3.9 ?
> >
> > I am preparing a fix right now. Not a revert, thank you.
>
> Is the following patch not good enough ?
I think it is somewhat telling that the hook you're proposing doesn't ever
make any calls into any of the individual LSMs, it only calls back into the
networking stack. In my mind, this makes it an abuse of the LSM mechanism.
On Monday, April 08, 2013 11:34:11 AM Eric Dumazet wrote:
> On Mon, 2013-04-08 at 14:26 -0400, Paul Moore wrote:
> > I guess we'll have to wait and see then; the more I think about the new
> > hook you proposed the less enthused I am about it.
> >
> > I'm still curious to hear what Dave has to say on this.
>
> 90ba9b1986b5ac4b2 is 10 months old, and nobody complained until today ?
The people who use this functionality almost never use upstream kernels, they
need to protection/certification/warm-fuzzies/etc. that come from a
distribution kernel and a support infrastructure. I didn't catch it because I
use a slightly different configuration that didn't expose this bug; while I
would like to run a full regression test every release I simply don't have the
time to do that myself.
> This sounds like a very small issue to me, a revert is simply overkill.
It all depends on your use case. To you, whom I assume doesn't use SELinux,
it is indeed a trivial issue. To someone who relies on SELinux for its
network access controls this is a pretty significant issue.
--
paul moore
security and virtualization @ redhat
^ permalink raw reply
* Re: [PATCH 2/4] Move execute_set_action to lib/odp-util.c
From: Jesse Gross @ 2013-04-08 20:29 UTC (permalink / raw)
To: Simon Horman
Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org, netdev, Ravi K,
Isaku Yamahata
In-Reply-To: <1365403431-18102-3-git-send-email-horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org>
On Sun, Apr 7, 2013 at 11:43 PM, Simon Horman <horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org> wrote:
> Move execute_set_action from lib/dpif-netedev.c to lib/odp-util.c
>
> This is in preparation for using execute_set_action()
> in lib/odp-util.c to handle recirculation/
>
> Signed-off-by: Simon Horman <horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org>
>
> ---
>
> packet.c might be a better place for execute_set_action()
> but I'm unsure if accessing struct ovs_key_ethernet would
> lead to a layering violation.
I'd be tempted to just put this in it's own file. As you say, it
doesn't really fit in either of the two existing ones.
> diff --git a/lib/odp-util.c b/lib/odp-util.c
> index e18e109..ad5873c 100644
> --- a/lib/odp-util.c
> +++ b/lib/odp-util.c
> @@ -2420,3 +2420,79 @@ commit_odp_actions(const struct flow *flow, struct flow *base,
> commit_set_priority_action(flow, base, odp_actions);
> commit_set_skb_mark_action(flow, base, odp_actions);
> }
> +
> +static void
> +dp_netdev_set_dl(struct ofpbuf *packet, const struct ovs_key_ethernet *eth_key)
I think this function should be given a more generic name and possibly
moved to packet.c.
> +void
> +execute_set_action(struct ofpbuf *packet, const struct nlattr *a,
> + uint32_t *skb_mark)
> +{
> + enum ovs_key_attr type = nl_attr_type(a);
> + const struct ovs_key_ipv4 *ipv4_key;
> + const struct ovs_key_ipv6 *ipv6_key;
> + const struct ovs_key_tcp *tcp_key;
> + const struct ovs_key_udp *udp_key;
> +
> + switch (type) {
> + case OVS_KEY_ATTR_PRIORITY:
> + case OVS_KEY_ATTR_TUNNEL:
> + /* not implemented */
> + break;
Don't we need to carry this information along as well similar to skb->mark?
Also, is there a reason to not have the code for push/pop actions here as well?
^ permalink raw reply
* Re: how to test multipe unicast MAC address
From: John Fastabend @ 2013-04-08 20:28 UTC (permalink / raw)
To: Rayagond K; +Cc: Eric Dumazet, netdev
In-Reply-To: <CAJ3bTp4HnmZ9pvzEOUg5VAKiLYGxS+f6eqtb5HdXkmuUYdyZUw@mail.gmail.com>
On 04/08/2013 11:13 AM, Rayagond K wrote:
> On Mon, Apr 8, 2013 at 7:46 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
>> On Mon, 2013-04-08 at 12:38 +0530, Rayagond K wrote:
>>> Hi All,
>>>
>>> How to assign multiple unicast MAC address to single interface ?
>>> How to test multiple unicast MAC address filtering ?
>>>
>>> Thanks in advance.
>>>
>>
>> Standard way would be to use macvlan
>>
>> ip link add link eth0 name foo02 addr 44:44:44:44:00:02 type macvlan
>> ip link add link eth0 name foo03 addr 44:44:44:44:00:03 type macvlan
>
> I tried this, but this won't update the uc address and in netdev
> struct and won't call .ndo_set_multicast_list.
>
This should work macvlan calls uc_sync and mc_sync on the lowerdev. It
is working on the tip at least,
# bridge fdb show dev p3p2
01:00:5e:00:00:01 self permanent
33:33:00:00:00:01 self permanent
33:33:ff:69:9f:09 self permanent
# ip link add link p3p2 address 00:11:22:33:44:55 type macvlan
# ip link set dev macvlan0 up
# bridge fdb show dev p3p2
00:11:22:33:44:55 self permanent
01:00:5e:00:00:01 self permanent
33:33:00:00:00:01 self permanent
33:33:ff:69:9f:09 self permanent
33:33:ff:33:44:55 self permanent
> I am planning to use vlan through vconfig as below,
>
> #vconfig add eth0 100
> #ifconfig eth0.100 10.144.134.1 netmask 255.255.255.0 up
> #ifconfig eth0.100 hw ether 00:55:22:44:f7:80
>
> This will update uc in netdev and calls .ndo_set_multicast_list.
>
--
John Fastabend Intel Corporation
^ permalink raw reply
* Re: [patch 1/2] af_iucv: fix recvmsg by replacing skb_pull() function
From: Eric Dumazet @ 2013-04-08 20:16 UTC (permalink / raw)
To: frank.blaschka; +Cc: davem, netdev, linux-s390, Ursula Braun
In-Reply-To: <20130408082003.666955316@de.ibm.com>
On Mon, 2013-04-08 at 10:19 +0200, frank.blaschka@de.ibm.com wrote:
> plain text document attachment (601-af-iucv-skb-pull.diff)
> From: Ursula Braun <ursula.braun@de.ibm.com>
>
> When receiving data messages, the "BUG_ON(skb->len < skb->data_len)" in
> the skb_pull() function triggers a kernel panic.
>
> Replace the skb_pull logic by a per skb offset as advised by
> Eric Dumazet.
>
> Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
> Signed-off-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
> Reviewed-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
> ---
This is indeed a nicer patch ;)
Acked-by: Eric Dumazet <edumazet@google.com>
^ permalink raw reply
* Re: [PATCH] tcp: assign the sock correctly to an outgoing SYNACK packet
From: David Miller @ 2013-04-08 19:25 UTC (permalink / raw)
To: pmoore; +Cc: eric.dumazet, netdev, mvadkert
In-Reply-To: <3541094.5siDbVn1lC@sifl>
From: Paul Moore <pmoore@redhat.com>
Date: Mon, 08 Apr 2013 13:22:50 -0400
> Contrary to popular belief, my goal is to not destroy the scalability and/or
> performance of our network stack, I just want to make sure we have a quality
> network stack that is not only fast and scalable, but also preserves the
> security functionality that makes Linux attractive to a number of users.
Get the violin out.
> To that end, we could put a #ifdef in the middle of
> tcp_make_synack(), but that seems very ugly to me and I think sets a
> bad precedence for the network stack and kernel as a whole.
Not an ifdef, a run time state test.
^ permalink raw reply
* pull request: wireless 2013-04-08
From: John W. Linville @ 2013-04-08 19:21 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
[-- Attachment #1: Type: text/plain, Size: 31953 bytes --]
Dave,
Please consider this set of fixes for the 3.9 stream...
For the cfg80211 fix, Johannes says:
"I have another straggler for 3.9, adding locking forgotten in a previous
fix."
On top of that:
Bing Zhao provides an mwifiex fix to properly order a scan completion.
Franky Lin gives us a brcmfmac fix to fail at the firmware loading
stage if the nvram cannot be downloaded.
Gabor Juhos brings what at first looks like a rather big rt2x00 patch.
I think it is OK because it is really just reorganizing some code
within the rt2x00 driver in order to fix a build failure.
Hante Meuleman offers a trio of brcmfmac fixes related to running in
AP mode.
Robert Shade sends an ath9k fix to reenable interrupts even if a
channel change fails.
Tim Gardner gives us an rt2x00 fix to cut-down on some log SPAM.
Please let me know if there are problems!
Thanks,
John
---
The following changes since commit b50b72de2f2feed4adfbd8e18610a393b5a04cc7:
net: mvneta: enable features before registering the driver (2013-04-08 12:16:38 -0400)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless.git for-davem
for you to fetch changes up to 64e5751918b4d03015b185750a46ecd050e1ce77:
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless into for-davem (2013-04-08 14:26:57 -0400)
----------------------------------------------------------------
Artem Savkov (1):
cfg80211: sched_scan_mtx lock in cfg80211_conn_work()
Bing Zhao (1):
mwifiex: complete last internal scan
Franky Lin (1):
brcmfmac: do not proceed if fail to download nvram to dongle
Gabor Juhos (1):
rt2x00: rt2x00pci: fix build error on Ralink RT3x5x SoCs
Hante Meuleman (3):
brcmfmac: fix tkip mic tx/rx ap swap bug.
brcmfmac: fix stopping AP.
brcmfmac: fix returning cipher_suite for get_key operation.
John W. Linville (2):
Merge branch 'for-john' of git://git.kernel.org/.../jberg/mac80211
Merge branch 'master' of git://git.kernel.org/.../linville/wireless into for-davem
Robert Shade (1):
ath9k: Re-enable interrupts after a channel change failure
Tim Gardner (1):
rt2x00: rt2x00pci_regbusy_read() - only print register access failure once
drivers/net/wireless/ath/ath9k/main.c | 4 +
drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c | 6 +-
.../net/wireless/brcm80211/brcmfmac/wl_cfg80211.c | 46 +++--
drivers/net/wireless/mwifiex/scan.c | 11 +-
drivers/net/wireless/rt2x00/Kconfig | 7 +
drivers/net/wireless/rt2x00/Makefile | 1 +
drivers/net/wireless/rt2x00/rt2400pci.c | 1 +
drivers/net/wireless/rt2x00/rt2500pci.c | 1 +
drivers/net/wireless/rt2x00/rt2800pci.c | 1 +
drivers/net/wireless/rt2x00/rt2x00mmio.c | 216 +++++++++++++++++++++
drivers/net/wireless/rt2x00/rt2x00mmio.h | 119 ++++++++++++
drivers/net/wireless/rt2x00/rt2x00pci.c | 176 -----------------
drivers/net/wireless/rt2x00/rt2x00pci.h | 88 ---------
drivers/net/wireless/rt2x00/rt61pci.c | 1 +
net/wireless/sme.c | 2 +
15 files changed, 390 insertions(+), 290 deletions(-)
create mode 100644 drivers/net/wireless/rt2x00/rt2x00mmio.c
create mode 100644 drivers/net/wireless/rt2x00/rt2x00mmio.h
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index 6e66f9c..988372d 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -280,6 +280,10 @@ static int ath_reset_internal(struct ath_softc *sc, struct ath9k_channel *hchan)
if (r) {
ath_err(common,
"Unable to reset channel, reset status %d\n", r);
+
+ ath9k_hw_enable_interrupts(ah);
+ ath9k_queue_reset(sc, RESET_TYPE_BB_HANG);
+
goto out;
}
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c b/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
index 4469321..35fc68b 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/dhd_sdio.c
@@ -3317,15 +3317,15 @@ static int _brcmf_sdbrcm_download_firmware(struct brcmf_sdio *bus)
goto err;
}
- /* External image takes precedence if specified */
if (brcmf_sdbrcm_download_code_file(bus)) {
brcmf_err("dongle image file download failed\n");
goto err;
}
- /* External nvram takes precedence if specified */
- if (brcmf_sdbrcm_download_nvram(bus))
+ if (brcmf_sdbrcm_download_nvram(bus)) {
brcmf_err("dongle nvram file download failed\n");
+ goto err;
+ }
/* Take arm out of reset */
if (brcmf_sdbrcm_download_state(bus, false)) {
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
index 2af9c0f..ec46fff 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
@@ -1891,8 +1891,10 @@ static s32
brcmf_add_keyext(struct wiphy *wiphy, struct net_device *ndev,
u8 key_idx, const u8 *mac_addr, struct key_params *params)
{
+ struct brcmf_if *ifp = netdev_priv(ndev);
struct brcmf_wsec_key key;
s32 err = 0;
+ u8 keybuf[8];
memset(&key, 0, sizeof(key));
key.index = (u32) key_idx;
@@ -1916,8 +1918,9 @@ brcmf_add_keyext(struct wiphy *wiphy, struct net_device *ndev,
brcmf_dbg(CONN, "Setting the key index %d\n", key.index);
memcpy(key.data, params->key, key.len);
- if (params->cipher == WLAN_CIPHER_SUITE_TKIP) {
- u8 keybuf[8];
+ if ((ifp->vif->mode != WL_MODE_AP) &&
+ (params->cipher == WLAN_CIPHER_SUITE_TKIP)) {
+ brcmf_dbg(CONN, "Swapping RX/TX MIC key\n");
memcpy(keybuf, &key.data[24], sizeof(keybuf));
memcpy(&key.data[24], &key.data[16], sizeof(keybuf));
memcpy(&key.data[16], keybuf, sizeof(keybuf));
@@ -2013,7 +2016,7 @@ brcmf_cfg80211_add_key(struct wiphy *wiphy, struct net_device *ndev,
break;
case WLAN_CIPHER_SUITE_TKIP:
if (ifp->vif->mode != WL_MODE_AP) {
- brcmf_dbg(CONN, "Swapping key\n");
+ brcmf_dbg(CONN, "Swapping RX/TX MIC key\n");
memcpy(keybuf, &key.data[24], sizeof(keybuf));
memcpy(&key.data[24], &key.data[16], sizeof(keybuf));
memcpy(&key.data[16], keybuf, sizeof(keybuf));
@@ -2118,8 +2121,7 @@ brcmf_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
err = -EAGAIN;
goto done;
}
- switch (wsec & ~SES_OW_ENABLED) {
- case WEP_ENABLED:
+ if (wsec & WEP_ENABLED) {
sec = &profile->sec;
if (sec->cipher_pairwise & WLAN_CIPHER_SUITE_WEP40) {
params.cipher = WLAN_CIPHER_SUITE_WEP40;
@@ -2128,16 +2130,13 @@ brcmf_cfg80211_get_key(struct wiphy *wiphy, struct net_device *ndev,
params.cipher = WLAN_CIPHER_SUITE_WEP104;
brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_WEP104\n");
}
- break;
- case TKIP_ENABLED:
+ } else if (wsec & TKIP_ENABLED) {
params.cipher = WLAN_CIPHER_SUITE_TKIP;
brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_TKIP\n");
- break;
- case AES_ENABLED:
+ } else if (wsec & AES_ENABLED) {
params.cipher = WLAN_CIPHER_SUITE_AES_CMAC;
brcmf_dbg(CONN, "WLAN_CIPHER_SUITE_AES_CMAC\n");
- break;
- default:
+ } else {
brcmf_err("Invalid algo (0x%x)\n", wsec);
err = -EINVAL;
goto done;
@@ -3824,8 +3823,9 @@ exit:
static int brcmf_cfg80211_stop_ap(struct wiphy *wiphy, struct net_device *ndev)
{
struct brcmf_if *ifp = netdev_priv(ndev);
- s32 err = -EPERM;
+ s32 err;
struct brcmf_fil_bss_enable_le bss_enable;
+ struct brcmf_join_params join_params;
brcmf_dbg(TRACE, "Enter\n");
@@ -3833,16 +3833,21 @@ static int brcmf_cfg80211_stop_ap(struct wiphy *wiphy, struct net_device *ndev)
/* Due to most likely deauths outstanding we sleep */
/* first to make sure they get processed by fw. */
msleep(400);
- err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_AP, 0);
- if (err < 0) {
- brcmf_err("setting AP mode failed %d\n", err);
- goto exit;
- }
+
+ memset(&join_params, 0, sizeof(join_params));
+ err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SET_SSID,
+ &join_params, sizeof(join_params));
+ if (err < 0)
+ brcmf_err("SET SSID error (%d)\n", err);
err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_UP, 0);
- if (err < 0) {
+ if (err < 0)
brcmf_err("BRCMF_C_UP error %d\n", err);
- goto exit;
- }
+ err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_AP, 0);
+ if (err < 0)
+ brcmf_err("setting AP mode failed %d\n", err);
+ err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_INFRA, 0);
+ if (err < 0)
+ brcmf_err("setting INFRA mode failed %d\n", err);
} else {
bss_enable.bsscfg_idx = cpu_to_le32(ifp->bssidx);
bss_enable.enable = cpu_to_le32(0);
@@ -3855,7 +3860,6 @@ static int brcmf_cfg80211_stop_ap(struct wiphy *wiphy, struct net_device *ndev)
set_bit(BRCMF_VIF_STATUS_AP_CREATING, &ifp->vif->sme_state);
clear_bit(BRCMF_VIF_STATUS_AP_CREATED, &ifp->vif->sme_state);
-exit:
return err;
}
diff --git a/drivers/net/wireless/mwifiex/scan.c b/drivers/net/wireless/mwifiex/scan.c
index d215b4d..e7f6dea 100644
--- a/drivers/net/wireless/mwifiex/scan.c
+++ b/drivers/net/wireless/mwifiex/scan.c
@@ -1393,8 +1393,10 @@ int mwifiex_scan_networks(struct mwifiex_private *priv,
queue_work(adapter->workqueue, &adapter->main_work);
/* Perform internal scan synchronously */
- if (!priv->scan_request)
+ if (!priv->scan_request) {
+ dev_dbg(adapter->dev, "wait internal scan\n");
mwifiex_wait_queue_complete(adapter, cmd_node);
+ }
} else {
spin_unlock_irqrestore(&adapter->scan_pending_q_lock,
flags);
@@ -1793,7 +1795,12 @@ check_next_scan:
/* Need to indicate IOCTL complete */
if (adapter->curr_cmd->wait_q_enabled) {
adapter->cmd_wait_q.status = 0;
- mwifiex_complete_cmd(adapter, adapter->curr_cmd);
+ if (!priv->scan_request) {
+ dev_dbg(adapter->dev,
+ "complete internal scan\n");
+ mwifiex_complete_cmd(adapter,
+ adapter->curr_cmd);
+ }
}
if (priv->report_scan_result)
priv->report_scan_result = false;
diff --git a/drivers/net/wireless/rt2x00/Kconfig b/drivers/net/wireless/rt2x00/Kconfig
index 2bf4efa..76cd47e 100644
--- a/drivers/net/wireless/rt2x00/Kconfig
+++ b/drivers/net/wireless/rt2x00/Kconfig
@@ -20,6 +20,7 @@ if RT2X00
config RT2400PCI
tristate "Ralink rt2400 (PCI/PCMCIA) support"
depends on PCI
+ select RT2X00_LIB_MMIO
select RT2X00_LIB_PCI
select EEPROM_93CX6
---help---
@@ -31,6 +32,7 @@ config RT2400PCI
config RT2500PCI
tristate "Ralink rt2500 (PCI/PCMCIA) support"
depends on PCI
+ select RT2X00_LIB_MMIO
select RT2X00_LIB_PCI
select EEPROM_93CX6
---help---
@@ -43,6 +45,7 @@ config RT61PCI
tristate "Ralink rt2501/rt61 (PCI/PCMCIA) support"
depends on PCI
select RT2X00_LIB_PCI
+ select RT2X00_LIB_MMIO
select RT2X00_LIB_FIRMWARE
select RT2X00_LIB_CRYPTO
select CRC_ITU_T
@@ -57,6 +60,7 @@ config RT2800PCI
tristate "Ralink rt27xx/rt28xx/rt30xx (PCI/PCIe/PCMCIA) support"
depends on PCI || SOC_RT288X || SOC_RT305X
select RT2800_LIB
+ select RT2X00_LIB_MMIO
select RT2X00_LIB_PCI if PCI
select RT2X00_LIB_SOC if SOC_RT288X || SOC_RT305X
select RT2X00_LIB_FIRMWARE
@@ -185,6 +189,9 @@ endif
config RT2800_LIB
tristate
+config RT2X00_LIB_MMIO
+ tristate
+
config RT2X00_LIB_PCI
tristate
select RT2X00_LIB
diff --git a/drivers/net/wireless/rt2x00/Makefile b/drivers/net/wireless/rt2x00/Makefile
index 349d5b8..f069d8b 100644
--- a/drivers/net/wireless/rt2x00/Makefile
+++ b/drivers/net/wireless/rt2x00/Makefile
@@ -9,6 +9,7 @@ rt2x00lib-$(CONFIG_RT2X00_LIB_FIRMWARE) += rt2x00firmware.o
rt2x00lib-$(CONFIG_RT2X00_LIB_LEDS) += rt2x00leds.o
obj-$(CONFIG_RT2X00_LIB) += rt2x00lib.o
+obj-$(CONFIG_RT2X00_LIB_MMIO) += rt2x00mmio.o
obj-$(CONFIG_RT2X00_LIB_PCI) += rt2x00pci.o
obj-$(CONFIG_RT2X00_LIB_SOC) += rt2x00soc.o
obj-$(CONFIG_RT2X00_LIB_USB) += rt2x00usb.o
diff --git a/drivers/net/wireless/rt2x00/rt2400pci.c b/drivers/net/wireless/rt2x00/rt2400pci.c
index 221beaa..dcfb54e 100644
--- a/drivers/net/wireless/rt2x00/rt2400pci.c
+++ b/drivers/net/wireless/rt2x00/rt2400pci.c
@@ -34,6 +34,7 @@
#include <linux/slab.h>
#include "rt2x00.h"
+#include "rt2x00mmio.h"
#include "rt2x00pci.h"
#include "rt2400pci.h"
diff --git a/drivers/net/wireless/rt2x00/rt2500pci.c b/drivers/net/wireless/rt2x00/rt2500pci.c
index 39edc59..e1d2dc9 100644
--- a/drivers/net/wireless/rt2x00/rt2500pci.c
+++ b/drivers/net/wireless/rt2x00/rt2500pci.c
@@ -34,6 +34,7 @@
#include <linux/slab.h>
#include "rt2x00.h"
+#include "rt2x00mmio.h"
#include "rt2x00pci.h"
#include "rt2500pci.h"
diff --git a/drivers/net/wireless/rt2x00/rt2800pci.c b/drivers/net/wireless/rt2x00/rt2800pci.c
index ded73da..ba5a056 100644
--- a/drivers/net/wireless/rt2x00/rt2800pci.c
+++ b/drivers/net/wireless/rt2x00/rt2800pci.c
@@ -41,6 +41,7 @@
#include <linux/eeprom_93cx6.h>
#include "rt2x00.h"
+#include "rt2x00mmio.h"
#include "rt2x00pci.h"
#include "rt2x00soc.h"
#include "rt2800lib.h"
diff --git a/drivers/net/wireless/rt2x00/rt2x00mmio.c b/drivers/net/wireless/rt2x00/rt2x00mmio.c
new file mode 100644
index 0000000..d84a680
--- /dev/null
+++ b/drivers/net/wireless/rt2x00/rt2x00mmio.c
@@ -0,0 +1,216 @@
+/*
+ Copyright (C) 2004 - 2009 Ivo van Doorn <IvDoorn-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+ <http://rt2x00.serialmonkey.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.,
+ 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+/*
+ Module: rt2x00mmio
+ Abstract: rt2x00 generic mmio device routines.
+ */
+
+#include <linux/dma-mapping.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include "rt2x00.h"
+#include "rt2x00mmio.h"
+
+/*
+ * Register access.
+ */
+int rt2x00pci_regbusy_read(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ const struct rt2x00_field32 field,
+ u32 *reg)
+{
+ unsigned int i;
+
+ if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags))
+ return 0;
+
+ for (i = 0; i < REGISTER_BUSY_COUNT; i++) {
+ rt2x00pci_register_read(rt2x00dev, offset, reg);
+ if (!rt2x00_get_field32(*reg, field))
+ return 1;
+ udelay(REGISTER_BUSY_DELAY);
+ }
+
+ printk_once(KERN_ERR "%s() Indirect register access failed: "
+ "offset=0x%.08x, value=0x%.08x\n", __func__, offset, *reg);
+ *reg = ~0;
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(rt2x00pci_regbusy_read);
+
+bool rt2x00pci_rxdone(struct rt2x00_dev *rt2x00dev)
+{
+ struct data_queue *queue = rt2x00dev->rx;
+ struct queue_entry *entry;
+ struct queue_entry_priv_pci *entry_priv;
+ struct skb_frame_desc *skbdesc;
+ int max_rx = 16;
+
+ while (--max_rx) {
+ entry = rt2x00queue_get_entry(queue, Q_INDEX);
+ entry_priv = entry->priv_data;
+
+ if (rt2x00dev->ops->lib->get_entry_state(entry))
+ break;
+
+ /*
+ * Fill in desc fields of the skb descriptor
+ */
+ skbdesc = get_skb_frame_desc(entry->skb);
+ skbdesc->desc = entry_priv->desc;
+ skbdesc->desc_len = entry->queue->desc_size;
+
+ /*
+ * DMA is already done, notify rt2x00lib that
+ * it finished successfully.
+ */
+ rt2x00lib_dmastart(entry);
+ rt2x00lib_dmadone(entry);
+
+ /*
+ * Send the frame to rt2x00lib for further processing.
+ */
+ rt2x00lib_rxdone(entry, GFP_ATOMIC);
+ }
+
+ return !max_rx;
+}
+EXPORT_SYMBOL_GPL(rt2x00pci_rxdone);
+
+void rt2x00pci_flush_queue(struct data_queue *queue, bool drop)
+{
+ unsigned int i;
+
+ for (i = 0; !rt2x00queue_empty(queue) && i < 10; i++)
+ msleep(10);
+}
+EXPORT_SYMBOL_GPL(rt2x00pci_flush_queue);
+
+/*
+ * Device initialization handlers.
+ */
+static int rt2x00pci_alloc_queue_dma(struct rt2x00_dev *rt2x00dev,
+ struct data_queue *queue)
+{
+ struct queue_entry_priv_pci *entry_priv;
+ void *addr;
+ dma_addr_t dma;
+ unsigned int i;
+
+ /*
+ * Allocate DMA memory for descriptor and buffer.
+ */
+ addr = dma_alloc_coherent(rt2x00dev->dev,
+ queue->limit * queue->desc_size,
+ &dma, GFP_KERNEL);
+ if (!addr)
+ return -ENOMEM;
+
+ memset(addr, 0, queue->limit * queue->desc_size);
+
+ /*
+ * Initialize all queue entries to contain valid addresses.
+ */
+ for (i = 0; i < queue->limit; i++) {
+ entry_priv = queue->entries[i].priv_data;
+ entry_priv->desc = addr + i * queue->desc_size;
+ entry_priv->desc_dma = dma + i * queue->desc_size;
+ }
+
+ return 0;
+}
+
+static void rt2x00pci_free_queue_dma(struct rt2x00_dev *rt2x00dev,
+ struct data_queue *queue)
+{
+ struct queue_entry_priv_pci *entry_priv =
+ queue->entries[0].priv_data;
+
+ if (entry_priv->desc)
+ dma_free_coherent(rt2x00dev->dev,
+ queue->limit * queue->desc_size,
+ entry_priv->desc, entry_priv->desc_dma);
+ entry_priv->desc = NULL;
+}
+
+int rt2x00pci_initialize(struct rt2x00_dev *rt2x00dev)
+{
+ struct data_queue *queue;
+ int status;
+
+ /*
+ * Allocate DMA
+ */
+ queue_for_each(rt2x00dev, queue) {
+ status = rt2x00pci_alloc_queue_dma(rt2x00dev, queue);
+ if (status)
+ goto exit;
+ }
+
+ /*
+ * Register interrupt handler.
+ */
+ status = request_irq(rt2x00dev->irq,
+ rt2x00dev->ops->lib->irq_handler,
+ IRQF_SHARED, rt2x00dev->name, rt2x00dev);
+ if (status) {
+ ERROR(rt2x00dev, "IRQ %d allocation failed (error %d).\n",
+ rt2x00dev->irq, status);
+ goto exit;
+ }
+
+ return 0;
+
+exit:
+ queue_for_each(rt2x00dev, queue)
+ rt2x00pci_free_queue_dma(rt2x00dev, queue);
+
+ return status;
+}
+EXPORT_SYMBOL_GPL(rt2x00pci_initialize);
+
+void rt2x00pci_uninitialize(struct rt2x00_dev *rt2x00dev)
+{
+ struct data_queue *queue;
+
+ /*
+ * Free irq line.
+ */
+ free_irq(rt2x00dev->irq, rt2x00dev);
+
+ /*
+ * Free DMA
+ */
+ queue_for_each(rt2x00dev, queue)
+ rt2x00pci_free_queue_dma(rt2x00dev, queue);
+}
+EXPORT_SYMBOL_GPL(rt2x00pci_uninitialize);
+
+/*
+ * rt2x00mmio module information.
+ */
+MODULE_AUTHOR(DRV_PROJECT);
+MODULE_VERSION(DRV_VERSION);
+MODULE_DESCRIPTION("rt2x00 mmio library");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/wireless/rt2x00/rt2x00mmio.h b/drivers/net/wireless/rt2x00/rt2x00mmio.h
new file mode 100644
index 0000000..4ecaf60
--- /dev/null
+++ b/drivers/net/wireless/rt2x00/rt2x00mmio.h
@@ -0,0 +1,119 @@
+/*
+ Copyright (C) 2004 - 2009 Ivo van Doorn <IvDoorn-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+ <http://rt2x00.serialmonkey.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.,
+ 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+/*
+ Module: rt2x00mmio
+ Abstract: Data structures for the rt2x00mmio module.
+ */
+
+#ifndef RT2X00MMIO_H
+#define RT2X00MMIO_H
+
+#include <linux/io.h>
+
+/*
+ * Register access.
+ */
+static inline void rt2x00pci_register_read(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ u32 *value)
+{
+ *value = readl(rt2x00dev->csr.base + offset);
+}
+
+static inline void rt2x00pci_register_multiread(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ void *value, const u32 length)
+{
+ memcpy_fromio(value, rt2x00dev->csr.base + offset, length);
+}
+
+static inline void rt2x00pci_register_write(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ u32 value)
+{
+ writel(value, rt2x00dev->csr.base + offset);
+}
+
+static inline void rt2x00pci_register_multiwrite(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ const void *value,
+ const u32 length)
+{
+ __iowrite32_copy(rt2x00dev->csr.base + offset, value, length >> 2);
+}
+
+/**
+ * rt2x00pci_regbusy_read - Read from register with busy check
+ * @rt2x00dev: Device pointer, see &struct rt2x00_dev.
+ * @offset: Register offset
+ * @field: Field to check if register is busy
+ * @reg: Pointer to where register contents should be stored
+ *
+ * This function will read the given register, and checks if the
+ * register is busy. If it is, it will sleep for a couple of
+ * microseconds before reading the register again. If the register
+ * is not read after a certain timeout, this function will return
+ * FALSE.
+ */
+int rt2x00pci_regbusy_read(struct rt2x00_dev *rt2x00dev,
+ const unsigned int offset,
+ const struct rt2x00_field32 field,
+ u32 *reg);
+
+/**
+ * struct queue_entry_priv_pci: Per entry PCI specific information
+ *
+ * @desc: Pointer to device descriptor
+ * @desc_dma: DMA pointer to &desc.
+ * @data: Pointer to device's entry memory.
+ * @data_dma: DMA pointer to &data.
+ */
+struct queue_entry_priv_pci {
+ __le32 *desc;
+ dma_addr_t desc_dma;
+};
+
+/**
+ * rt2x00pci_rxdone - Handle RX done events
+ * @rt2x00dev: Device pointer, see &struct rt2x00_dev.
+ *
+ * Returns true if there are still rx frames pending and false if all
+ * pending rx frames were processed.
+ */
+bool rt2x00pci_rxdone(struct rt2x00_dev *rt2x00dev);
+
+/**
+ * rt2x00pci_flush_queue - Flush data queue
+ * @queue: Data queue to stop
+ * @drop: True to drop all pending frames.
+ *
+ * This will wait for a maximum of 100ms, waiting for the queues
+ * to become empty.
+ */
+void rt2x00pci_flush_queue(struct data_queue *queue, bool drop);
+
+/*
+ * Device initialization handlers.
+ */
+int rt2x00pci_initialize(struct rt2x00_dev *rt2x00dev);
+void rt2x00pci_uninitialize(struct rt2x00_dev *rt2x00dev);
+
+#endif /* RT2X00MMIO_H */
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.c b/drivers/net/wireless/rt2x00/rt2x00pci.c
index a0c8cae..e87865e 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.c
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.c
@@ -33,182 +33,6 @@
#include "rt2x00pci.h"
/*
- * Register access.
- */
-int rt2x00pci_regbusy_read(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- const struct rt2x00_field32 field,
- u32 *reg)
-{
- unsigned int i;
-
- if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags))
- return 0;
-
- for (i = 0; i < REGISTER_BUSY_COUNT; i++) {
- rt2x00pci_register_read(rt2x00dev, offset, reg);
- if (!rt2x00_get_field32(*reg, field))
- return 1;
- udelay(REGISTER_BUSY_DELAY);
- }
-
- ERROR(rt2x00dev, "Indirect register access failed: "
- "offset=0x%.08x, value=0x%.08x\n", offset, *reg);
- *reg = ~0;
-
- return 0;
-}
-EXPORT_SYMBOL_GPL(rt2x00pci_regbusy_read);
-
-bool rt2x00pci_rxdone(struct rt2x00_dev *rt2x00dev)
-{
- struct data_queue *queue = rt2x00dev->rx;
- struct queue_entry *entry;
- struct queue_entry_priv_pci *entry_priv;
- struct skb_frame_desc *skbdesc;
- int max_rx = 16;
-
- while (--max_rx) {
- entry = rt2x00queue_get_entry(queue, Q_INDEX);
- entry_priv = entry->priv_data;
-
- if (rt2x00dev->ops->lib->get_entry_state(entry))
- break;
-
- /*
- * Fill in desc fields of the skb descriptor
- */
- skbdesc = get_skb_frame_desc(entry->skb);
- skbdesc->desc = entry_priv->desc;
- skbdesc->desc_len = entry->queue->desc_size;
-
- /*
- * DMA is already done, notify rt2x00lib that
- * it finished successfully.
- */
- rt2x00lib_dmastart(entry);
- rt2x00lib_dmadone(entry);
-
- /*
- * Send the frame to rt2x00lib for further processing.
- */
- rt2x00lib_rxdone(entry, GFP_ATOMIC);
- }
-
- return !max_rx;
-}
-EXPORT_SYMBOL_GPL(rt2x00pci_rxdone);
-
-void rt2x00pci_flush_queue(struct data_queue *queue, bool drop)
-{
- unsigned int i;
-
- for (i = 0; !rt2x00queue_empty(queue) && i < 10; i++)
- msleep(10);
-}
-EXPORT_SYMBOL_GPL(rt2x00pci_flush_queue);
-
-/*
- * Device initialization handlers.
- */
-static int rt2x00pci_alloc_queue_dma(struct rt2x00_dev *rt2x00dev,
- struct data_queue *queue)
-{
- struct queue_entry_priv_pci *entry_priv;
- void *addr;
- dma_addr_t dma;
- unsigned int i;
-
- /*
- * Allocate DMA memory for descriptor and buffer.
- */
- addr = dma_alloc_coherent(rt2x00dev->dev,
- queue->limit * queue->desc_size,
- &dma, GFP_KERNEL);
- if (!addr)
- return -ENOMEM;
-
- memset(addr, 0, queue->limit * queue->desc_size);
-
- /*
- * Initialize all queue entries to contain valid addresses.
- */
- for (i = 0; i < queue->limit; i++) {
- entry_priv = queue->entries[i].priv_data;
- entry_priv->desc = addr + i * queue->desc_size;
- entry_priv->desc_dma = dma + i * queue->desc_size;
- }
-
- return 0;
-}
-
-static void rt2x00pci_free_queue_dma(struct rt2x00_dev *rt2x00dev,
- struct data_queue *queue)
-{
- struct queue_entry_priv_pci *entry_priv =
- queue->entries[0].priv_data;
-
- if (entry_priv->desc)
- dma_free_coherent(rt2x00dev->dev,
- queue->limit * queue->desc_size,
- entry_priv->desc, entry_priv->desc_dma);
- entry_priv->desc = NULL;
-}
-
-int rt2x00pci_initialize(struct rt2x00_dev *rt2x00dev)
-{
- struct data_queue *queue;
- int status;
-
- /*
- * Allocate DMA
- */
- queue_for_each(rt2x00dev, queue) {
- status = rt2x00pci_alloc_queue_dma(rt2x00dev, queue);
- if (status)
- goto exit;
- }
-
- /*
- * Register interrupt handler.
- */
- status = request_irq(rt2x00dev->irq,
- rt2x00dev->ops->lib->irq_handler,
- IRQF_SHARED, rt2x00dev->name, rt2x00dev);
- if (status) {
- ERROR(rt2x00dev, "IRQ %d allocation failed (error %d).\n",
- rt2x00dev->irq, status);
- goto exit;
- }
-
- return 0;
-
-exit:
- queue_for_each(rt2x00dev, queue)
- rt2x00pci_free_queue_dma(rt2x00dev, queue);
-
- return status;
-}
-EXPORT_SYMBOL_GPL(rt2x00pci_initialize);
-
-void rt2x00pci_uninitialize(struct rt2x00_dev *rt2x00dev)
-{
- struct data_queue *queue;
-
- /*
- * Free irq line.
- */
- free_irq(rt2x00dev->irq, rt2x00dev);
-
- /*
- * Free DMA
- */
- queue_for_each(rt2x00dev, queue)
- rt2x00pci_free_queue_dma(rt2x00dev, queue);
-}
-EXPORT_SYMBOL_GPL(rt2x00pci_uninitialize);
-
-/*
* PCI driver handlers.
*/
static void rt2x00pci_free_reg(struct rt2x00_dev *rt2x00dev)
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.h b/drivers/net/wireless/rt2x00/rt2x00pci.h
index e2c99f2..60d90b2 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.h
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.h
@@ -36,94 +36,6 @@
#define PCI_DEVICE_DATA(__ops) .driver_data = (kernel_ulong_t)(__ops)
/*
- * Register access.
- */
-static inline void rt2x00pci_register_read(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- u32 *value)
-{
- *value = readl(rt2x00dev->csr.base + offset);
-}
-
-static inline void rt2x00pci_register_multiread(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- void *value, const u32 length)
-{
- memcpy_fromio(value, rt2x00dev->csr.base + offset, length);
-}
-
-static inline void rt2x00pci_register_write(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- u32 value)
-{
- writel(value, rt2x00dev->csr.base + offset);
-}
-
-static inline void rt2x00pci_register_multiwrite(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- const void *value,
- const u32 length)
-{
- __iowrite32_copy(rt2x00dev->csr.base + offset, value, length >> 2);
-}
-
-/**
- * rt2x00pci_regbusy_read - Read from register with busy check
- * @rt2x00dev: Device pointer, see &struct rt2x00_dev.
- * @offset: Register offset
- * @field: Field to check if register is busy
- * @reg: Pointer to where register contents should be stored
- *
- * This function will read the given register, and checks if the
- * register is busy. If it is, it will sleep for a couple of
- * microseconds before reading the register again. If the register
- * is not read after a certain timeout, this function will return
- * FALSE.
- */
-int rt2x00pci_regbusy_read(struct rt2x00_dev *rt2x00dev,
- const unsigned int offset,
- const struct rt2x00_field32 field,
- u32 *reg);
-
-/**
- * struct queue_entry_priv_pci: Per entry PCI specific information
- *
- * @desc: Pointer to device descriptor
- * @desc_dma: DMA pointer to &desc.
- * @data: Pointer to device's entry memory.
- * @data_dma: DMA pointer to &data.
- */
-struct queue_entry_priv_pci {
- __le32 *desc;
- dma_addr_t desc_dma;
-};
-
-/**
- * rt2x00pci_rxdone - Handle RX done events
- * @rt2x00dev: Device pointer, see &struct rt2x00_dev.
- *
- * Returns true if there are still rx frames pending and false if all
- * pending rx frames were processed.
- */
-bool rt2x00pci_rxdone(struct rt2x00_dev *rt2x00dev);
-
-/**
- * rt2x00pci_flush_queue - Flush data queue
- * @queue: Data queue to stop
- * @drop: True to drop all pending frames.
- *
- * This will wait for a maximum of 100ms, waiting for the queues
- * to become empty.
- */
-void rt2x00pci_flush_queue(struct data_queue *queue, bool drop);
-
-/*
- * Device initialization handlers.
- */
-int rt2x00pci_initialize(struct rt2x00_dev *rt2x00dev);
-void rt2x00pci_uninitialize(struct rt2x00_dev *rt2x00dev);
-
-/*
* PCI driver handlers.
*/
int rt2x00pci_probe(struct pci_dev *pci_dev, const struct rt2x00_ops *ops);
diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c
index f95792c..9e3c8ff 100644
--- a/drivers/net/wireless/rt2x00/rt61pci.c
+++ b/drivers/net/wireless/rt2x00/rt61pci.c
@@ -35,6 +35,7 @@
#include <linux/eeprom_93cx6.h>
#include "rt2x00.h"
+#include "rt2x00mmio.h"
#include "rt2x00pci.h"
#include "rt61pci.h"
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index 09d994d..482c70e 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -224,6 +224,7 @@ void cfg80211_conn_work(struct work_struct *work)
rtnl_lock();
cfg80211_lock_rdev(rdev);
mutex_lock(&rdev->devlist_mtx);
+ mutex_lock(&rdev->sched_scan_mtx);
list_for_each_entry(wdev, &rdev->wdev_list, list) {
wdev_lock(wdev);
@@ -248,6 +249,7 @@ void cfg80211_conn_work(struct work_struct *work)
wdev_unlock(wdev);
}
+ mutex_unlock(&rdev->sched_scan_mtx);
mutex_unlock(&rdev->devlist_mtx);
cfg80211_unlock_rdev(rdev);
rtnl_unlock();
--
John W. Linville Someday the world will need a hero, and you
linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org might be all we have. Be ready.
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply related
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