OSDN Git Service

android-x86/system-netd.git
9 years agoam bc9da4e6: am 6b6f22fa: Enable tcp mss clamp for Tethering
Gordon Gao [Wed, 29 Oct 2014 07:00:19 +0000 (07:00 +0000)]
am bc9da4e6: am 6b6f22fa: Enable tcp mss clamp for Tethering

* commit 'bc9da4e633165700bbcc6ba33e09f100acb88d3f':

9 years agoam 10f9a166: am 4c95a125: Support manipulating throw routes.
Lorenzo Colitti [Wed, 29 Oct 2014 07:00:18 +0000 (07:00 +0000)]
am 10f9a166: am 4c95a125: Support manipulating throw routes.

* commit '10f9a166f6bfc625a07f67cf3bcc3d96b170a5de':

9 years agoam 388cd671: (-s ours) Merge "Fix MDnsSdListener::Monitor::Monitor()"
Robert Greenwalt [Thu, 9 Oct 2014 22:10:38 +0000 (22:10 +0000)]
am 388cd671: (-s ours) Merge "Fix MDnsSdListener::Monitor::Monitor()"

* commit '388cd671ff5237aabcd728794af49d33e18fda8b':
  Fix MDnsSdListener::Monitor::Monitor()

9 years agoMerge "Fix MDnsSdListener::Monitor::Monitor()"
Robert Greenwalt [Thu, 9 Oct 2014 19:13:59 +0000 (19:13 +0000)]
Merge "Fix MDnsSdListener::Monitor::Monitor()"

9 years agoFix MDnsSdListener::Monitor::Monitor()
Kévin Petit [Thu, 9 Oct 2014 14:20:56 +0000 (15:20 +0100)]
Fix MDnsSdListener::Monitor::Monitor()

* Initialisation of mLiveCount

The value of mLiveCount was being used without being initialised, and
this was causing the CTS class android.net.wifi.cts.NsdManagerTest
method testAndroidTestCaseSetupProperly test to fail when the calls
to calloc in the MDnsSdListener::Monitor::rescan() method returned a
value of NULL due to the large value of mPollSize, which had been set
to "mLiveCount + 5". The mPollFds and mPollRefs pointers were then used
without checking that they were not NULL, resulting in a segmentation
violation when the NULL pointer value was dereferenced. Initialising
the value of mLiveCount to zero in MDnsSdListener::Monitor::Monitor()
fixes this problem, and the test then passes.

* Also initilize mPollFds, mPollRefs and mPollSize

Change-Id: I1a8f7ee0f3ed38f46854e9ba6d5c432c9412b7c3
Signed-off-by: Marcus Oakland <marcus.oakland@arm.com>
Signed-off-by: Kévin Petit <kevin.petit@arm.com>
9 years agoam 3d48bc23: am a1067c8d: Improve error return values on network selection.
Lorenzo Colitti [Fri, 3 Oct 2014 06:12:47 +0000 (06:12 +0000)]
am 3d48bc23: am a1067c8d: Improve error return values on network selection.

* commit '3d48bc23465becc6992756d4160550bf01799ffc':
  Improve error return values on network selection.

9 years agoam 1c57d2b7: am e8164ddc: Fix MSS clamping.
Lorenzo Colitti [Fri, 3 Oct 2014 06:12:47 +0000 (06:12 +0000)]
am 1c57d2b7: am e8164ddc: Fix MSS clamping.

* commit '1c57d2b7141ba81a2a4e991267082a7258c09015':
  Fix MSS clamping.

9 years agoam 6fa06b78: softap: Ensure entropy file creation
Dmitry Shmidt [Fri, 3 Oct 2014 06:12:47 +0000 (06:12 +0000)]
am 6fa06b78: softap: Ensure entropy file creation

* commit '6fa06b78e31f9e99c4e61f2c4c8241823b781f89':
  softap: Ensure entropy file creation

9 years agoam a1067c8d: Improve error return values on network selection.
Lorenzo Colitti [Fri, 3 Oct 2014 05:55:27 +0000 (05:55 +0000)]
am a1067c8d: Improve error return values on network selection.

* commit 'a1067c8d2b2165f1058a3a8216bed4efacfa1c80':
  Improve error return values on network selection.

9 years agoam e8164ddc: Fix MSS clamping.
Lorenzo Colitti [Fri, 3 Oct 2014 05:54:55 +0000 (05:54 +0000)]
am e8164ddc: Fix MSS clamping.

* commit 'e8164ddc8204b626c1144a0a504754bf6622c6fd':
  Fix MSS clamping.

9 years agoImprove error return values on network selection.
Lorenzo Colitti [Thu, 2 Oct 2014 13:47:41 +0000 (22:47 +0900)]
Improve error return values on network selection.

It's very confusing to see EPERM when opening or connecting a
socket when the problem is not security-related. In the (common)
case where an app cannot select a network because it does not
exist, return ENONET ("Machine is not on network") instead.

Also, return EREMOTEIO for when we can't figure out who the user
is, and use EPERM for VPN denials and EACCES for permission
bits.

Bug: 17702933
Change-Id: Ia680c485e0ea1efad1ad374231d994e9bfd4cd5a

9 years agoFix MSS clamping.
Lorenzo Colitti [Thu, 2 Oct 2014 11:46:23 +0000 (20:46 +0900)]
Fix MSS clamping.

http://ag/553410 added an iptables chain and rule to do MSS
clamping for tethered clients, but did not add the chain to any
other chains, so the rule had no effect.

Fix this by adding the chain to the proper forwarding chains.
Also rename some of the new variables and constants so they are
more consistent with the previous code.

Bug: 17552732
Bug: 17727533
Change-Id: I9fcae31de5c0283d7d9f1dac989de84f77c5e53c

9 years agosoftap: Ensure entropy file creation
Dmitry Shmidt [Fri, 5 Sep 2014 23:39:29 +0000 (16:39 -0700)]
softap: Ensure entropy file creation

Bug: 17408011

Change-Id: I558270f1ce2603f305d38ad8604379ea533f0a47
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
9 years agoam 6b6f22fa: Enable tcp mss clamp for Tethering
Gordon Gao [Fri, 26 Sep 2014 23:57:03 +0000 (23:57 +0000)]
am 6b6f22fa: Enable tcp mss clamp for Tethering

* commit '6b6f22fac4c33fcd349aaf6970bbdc191db752a3':
  Enable tcp mss clamp for Tethering

9 years agoam 6b6f22fa: Enable tcp mss clamp for Tethering
Gordon Gao [Fri, 26 Sep 2014 23:44:04 +0000 (23:44 +0000)]
am 6b6f22fa: Enable tcp mss clamp for Tethering

* commit '6b6f22fac4c33fcd349aaf6970bbdc191db752a3':
  Enable tcp mss clamp for Tethering

9 years agoEnable tcp mss clamp for Tethering
Gordon Gao [Thu, 18 Sep 2014 18:50:09 +0000 (11:50 -0700)]
Enable tcp mss clamp for Tethering

With this change, IPv4 tethering clients (such as PC) with default
MTU size 1500 will be configured again in the phone to use MTU
matching the carrier network, then no IP fragemention in the phone
and have higher data throughput for tethering.

bug:17552732
Change-Id: Ia306b1b8454da1b0e2d9c320e770aeff172d5fc4

9 years agoam 4c95a125: Support manipulating throw routes.
Lorenzo Colitti [Mon, 22 Sep 2014 20:02:01 +0000 (20:02 +0000)]
am 4c95a125: Support manipulating throw routes.

* commit '4c95a125e0930c112555437589f7620575482095':
  Support manipulating throw routes.

9 years agoam 4c95a125: Support manipulating throw routes.
Lorenzo Colitti [Mon, 22 Sep 2014 17:25:28 +0000 (17:25 +0000)]
am 4c95a125: Support manipulating throw routes.

* commit '4c95a125e0930c112555437589f7620575482095':
  Support manipulating throw routes.

9 years agoSupport manipulating throw routes.
Lorenzo Colitti [Thu, 18 Sep 2014 07:01:50 +0000 (16:01 +0900)]
Support manipulating throw routes.

We already supported unreachable routes. Throw routes are
necessary so we can exempt the VPN endpoint from being routed
through the VPN in legacy VPN modes that do not pass traffic
through a tun or ppp interface but just directly appply IPsec
transformations to outgoing packets.

Bug: 17462989
Change-Id: I8635472ca3e96ec2866af2de48e6260ab2da13fb

9 years agoam 100457df: (-s ours) Merge "Remove makefile cruft."
Dan Albert [Fri, 12 Sep 2014 20:01:53 +0000 (20:01 +0000)]
am 100457df: (-s ours) Merge "Remove makefile cruft."

* commit '100457df47b8a453eb94df3cf2e87f6248e783d1':
  Remove makefile cruft.

9 years agoMerge "Remove makefile cruft."
Dan Albert [Fri, 12 Sep 2014 18:07:44 +0000 (18:07 +0000)]
Merge "Remove makefile cruft."

9 years agoRemove makefile cruft.
Dan Albert [Fri, 12 Sep 2014 17:47:14 +0000 (10:47 -0700)]
Remove makefile cruft.

Don't manually include stlport, and don't use private bionic headers.

Change-Id: Ibe91ed010b8558fba4ce08fdcf789cbed006f6ef

9 years agoam 6a834abb: Merge "Reset connections whenever an address goes away." into lmp-dev
Sreeram Ramachandran [Wed, 10 Sep 2014 00:51:21 +0000 (00:51 +0000)]
am 6a834abb: Merge "Reset connections whenever an address goes away." into lmp-dev

* commit '6a834abb5acca22e47b70ee88e55f394f909559b':
  Reset connections whenever an address goes away.

9 years agoMerge "Reset connections whenever an address goes away." into lmp-dev
Sreeram Ramachandran [Tue, 9 Sep 2014 23:24:52 +0000 (23:24 +0000)]
Merge "Reset connections whenever an address goes away." into lmp-dev

9 years agoReset connections whenever an address goes away.
Sreeram Ramachandran [Mon, 8 Sep 2014 23:03:18 +0000 (16:03 -0700)]
Reset connections whenever an address goes away.

Bug: 15414453
Change-Id: Ibf1f052e12b88838a328148e25817a96f10d50f5

9 years agoam bbdde990: Reserve NetIds 1..50 for OEM use.
Sreeram Ramachandran [Mon, 8 Sep 2014 23:50:49 +0000 (23:50 +0000)]
am bbdde990: Reserve NetIds 1..50 for OEM use.

* commit 'bbdde9909b7b4fd31c5857156ceb00049bf4992d':
  Reserve NetIds 1..50 for OEM use.

9 years agoReserve NetIds 1..50 for OEM use.
Sreeram Ramachandran [Fri, 5 Sep 2014 23:05:03 +0000 (16:05 -0700)]
Reserve NetIds 1..50 for OEM use.

Also reserve 51..100 for our use.

Bug: 17303534
Change-Id: Id082368b493dc3c5b5c479e51d273458d80ec6aa

9 years agoam 928294e4: Merge "softap: Add wowlan_triggers=any parameter" into lmp-dev
Dmitry Shmidt [Thu, 4 Sep 2014 17:49:28 +0000 (17:49 +0000)]
am 928294e4: Merge "softap: Add wowlan_triggers=any parameter" into lmp-dev

* commit '928294e4e778338c2a120c96373a779de8027ec6':
  softap: Add wowlan_triggers=any parameter

9 years agoMerge "softap: Add wowlan_triggers=any parameter" into lmp-dev
Dmitry Shmidt [Thu, 4 Sep 2014 17:23:26 +0000 (17:23 +0000)]
Merge "softap: Add wowlan_triggers=any parameter" into lmp-dev

9 years agosoftap: Add wowlan_triggers=any parameter
Dmitry Shmidt [Wed, 3 Sep 2014 22:19:36 +0000 (15:19 -0700)]
softap: Add wowlan_triggers=any parameter

New kernels in wiphy_suspend() will call cfg80211_leave_all()
that will eventually end up in cfg80211_stop_ap() unless
wowlan_triggers were set.

Bug: 17269024

Change-Id: I609d78f5bb573b4c6a9dae25a3b2a345499f6853
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
9 years agoam 667c4771: Add a socket mark parameter to tethering DNS commands.
Lorenzo Colitti [Wed, 3 Sep 2014 05:49:15 +0000 (05:49 +0000)]
am 667c4771: Add a socket mark parameter to tethering DNS commands.

* commit '667c477133318e4779819d34364194c8e5eaf19c':
  Add a socket mark parameter to tethering DNS commands.

9 years agoAdd a socket mark parameter to tethering DNS commands.
Lorenzo Colitti [Tue, 26 Aug 2014 21:13:07 +0000 (14:13 -0700)]
Add a socket mark parameter to tethering DNS commands.

This allows the framework to pass down to dnsmasq the socket mark
to use for DNS queries.

Bug: 16357676
Change-Id: I8933b6de198a92c2aaf0291931ace8966ddba275

10 years agoam 99286fe1: Make destroying networks more robust.
Lorenzo Colitti [Fri, 15 Aug 2014 11:49:37 +0000 (11:49 +0000)]
am 99286fe1: Make destroying networks more robust.

* commit '99286fe1ef6fc325c28dd10b651b5adedd549495':
  Make destroying networks more robust.

10 years agoMake destroying networks more robust.
Lorenzo Colitti [Tue, 12 Aug 2014 06:08:00 +0000 (15:08 +0900)]
Make destroying networks more robust.

1. Retry route flushes if they fail.
2. Make destroyNetwork ignore (but return) errors.

Bug: 16944962
Change-Id: I26301613437d7cc373ff64955fd44d716e9982b9

10 years agoam 6a46f338: Revert DNS cache flush command deletion from "Cleanup: Delete dead code."
Paul Jensen [Thu, 7 Aug 2014 02:42:07 +0000 (02:42 +0000)]
am 6a46f338: Revert DNS cache flush command deletion from "Cleanup: Delete dead code."

* commit '6a46f3384b3f48e6ef77a83a5e058a47a2939a0d':
  Revert DNS cache flush command deletion from "Cleanup: Delete dead code."

10 years agoRevert DNS cache flush command deletion from "Cleanup: Delete dead code."
Paul Jensen [Wed, 6 Aug 2014 18:42:27 +0000 (18:42 +0000)]
Revert DNS cache flush command deletion from "Cleanup: Delete dead code."

This reverts part of commit 1604e18615f824adb2d54c36320069ba6fcb4796.
The DNS cache flushing code needs to be called when routes change.

Change-Id: I20d39eabfa088deeeeeff0c3292b1910fd948b0d
bug:16549455

10 years agoam 10190d06: Merge "Fix initialization of Monitor fields." into lmp-dev
Narayan Kamath [Mon, 4 Aug 2014 16:00:40 +0000 (16:00 +0000)]
am 10190d06: Merge "Fix initialization of Monitor fields." into lmp-dev

* commit '10190d06d273577d85c42a947e4e0ab804411733':
  Fix initialization of Monitor fields.

10 years agoam 47f8597b: Add logging of calloc failures.
Robert Greenwalt [Mon, 4 Aug 2014 16:00:39 +0000 (16:00 +0000)]
am 47f8597b: Add logging of calloc failures.

* commit '47f8597b42b849d22c6cc128f27bba4a7692d371':
  Add logging of calloc failures.

10 years agoMerge "Fix initialization of Monitor fields." into lmp-dev
Narayan Kamath [Mon, 4 Aug 2014 15:54:00 +0000 (15:54 +0000)]
Merge "Fix initialization of Monitor fields." into lmp-dev

10 years agoFix initialization of Monitor fields.
Narayan Kamath [Mon, 4 Aug 2014 15:20:47 +0000 (16:20 +0100)]
Fix initialization of Monitor fields.

mLiveCount was never initialized. Also moves
initialization of mPollCount to the constructor, where
it belongs. Also, fields are now initialized in the order
in which they are declared.

bug: 16734745

Change-Id: I9cc668a6fc5649d3ea550bc3a47c0eef210cad07

10 years agoAdd logging of calloc failures.
Robert Greenwalt [Fri, 1 Aug 2014 23:19:02 +0000 (16:19 -0700)]
Add logging of calloc failures.

Re-adding after build-break and revert

bug:16734745
Change-Id: I26f9ff8c43836fd35778db61915ce37fd2025ba7

10 years agoam 515a5909: Merge "Revert "Add logging of calloc failures."" into lmp-dev
Ben Murdoch [Sat, 2 Aug 2014 12:38:48 +0000 (12:38 +0000)]
am 515a5909: Merge "Revert "Add logging of calloc failures."" into lmp-dev

* commit '515a59097e8b6d87023b824554e82b98aca8edf5':
  Revert "Add logging of calloc failures."

10 years agoam 3bc81979: Merge "Add logging of calloc failures." into lmp-dev
Robert Greenwalt [Fri, 1 Aug 2014 23:52:51 +0000 (23:52 +0000)]
am 3bc81979: Merge "Add logging of calloc failures." into lmp-dev

* commit '3bc819792bc617427512b0fe7b60c7ce2653094e':
  Add logging of calloc failures.

10 years agoMerge "Revert "Add logging of calloc failures."" into lmp-dev
Ben Murdoch [Sat, 2 Aug 2014 09:47:35 +0000 (09:47 +0000)]
Merge "Revert "Add logging of calloc failures."" into lmp-dev

10 years agoRevert "Add logging of calloc failures."
Ben Murdoch [Sat, 2 Aug 2014 09:43:22 +0000 (09:43 +0000)]
Revert "Add logging of calloc failures."

This reverts commit 96dd5bfec1693fee29ed0c728cf6ab4f0f0ec6e9 as it breaks the
64b build.

Bug: 16734745
Change-Id: Icf83b35ca0f99c2b5d6d919cfc8c2b02ee980e0b

10 years agoAdd logging of calloc failures.
Robert Greenwalt [Fri, 1 Aug 2014 23:19:02 +0000 (16:19 -0700)]
Add logging of calloc failures.

bug:16734745
Change-Id: I30437983e93fc2420e291e3d2e6ffbe4d8a47af4

10 years agoam 6d7e6235: Remove default routing rule & table before we forget table number.
Paul Jensen [Fri, 1 Aug 2014 15:47:49 +0000 (15:47 +0000)]
am 6d7e6235: Remove default routing rule & table before we forget table number.

* commit '6d7e6235b560be2ead9889c4035184573ab1b70a':
  Remove default routing rule & table before we forget table number.

10 years agoRemove default routing rule & table before we forget table number.
Paul Jensen [Fri, 1 Aug 2014 14:54:03 +0000 (10:54 -0400)]
Remove default routing rule & table before we forget table number.

We clear the interface index cache when we remove an interface from a
network, so we must only do this after we remove the default rule so
we still know the table number (calculated from interface index).

bug:16728065
Change-Id: I538673dcba49820a9e21f60407754fae30de02cf

10 years agoMerge "Add logging of calloc failures." into lmp-dev
Robert Greenwalt [Fri, 1 Aug 2014 23:46:24 +0000 (23:46 +0000)]
Merge "Add logging of calloc failures." into lmp-dev

10 years agoam 738c93ee: Minor improvements to NetworkController error reporting.
Lorenzo Colitti [Thu, 31 Jul 2014 13:27:36 +0000 (13:27 +0000)]
am 738c93ee: Minor improvements to NetworkController error reporting.

* commit '738c93ee075354ffafb3a8ceef76e9aa711f057b':
  Minor improvements to NetworkController error reporting.

10 years agoMinor improvements to NetworkController error reporting.
Lorenzo Colitti [Wed, 30 Jul 2014 08:46:08 +0000 (17:46 +0900)]
Minor improvements to NetworkController error reporting.

Currently, when trying to perform an operation on netId that
does not exist, we return EINVAL. This can be confusing
because lots of things can return EINVAL for many different
reasons.

Instead, change this to ENONET ("Machine is not on network"),
which was what Sreeram originally implemented before we changed
it to EINVAL. I also considered ENOENT ("No such file or
directory"), but on reflection that seems less appropriate, and
it's used by other things as well. ENONET appears to have no
uses in our tree. It is also clearly separate from the ESRCH we
return for nonexistent rules.

While I'm at it, also disambiguate some of the
if (foo || bar ) { return EFOO; } to return two different error
codes.

Bug: 16667349
Change-Id: Ief2d720a58679aa746f5ba273d545487d0825e52

10 years agoam db74dba7: Stop copying directly-connected routes to the main table, #2.
Lorenzo Colitti [Tue, 29 Jul 2014 13:15:23 +0000 (13:15 +0000)]
am db74dba7: Stop copying directly-connected routes to the main table, #2.

* commit 'db74dba7ccfe9e9504e0acd440a23fed96682842':
  Stop copying directly-connected routes to the main table, #2.

10 years agoam a2c23052: Revert "Stop copying directly-connected routes to the main table."
Lorenzo Colitti [Tue, 29 Jul 2014 13:15:23 +0000 (13:15 +0000)]
am a2c23052: Revert "Stop copying directly-connected routes to the main table."

* commit 'a2c230520be5fdafce8bbc0b6ee52262f981f75a':
  Revert "Stop copying directly-connected routes to the main table."

10 years agoam 2f5ea0e9: Stop copying directly-connected routes to the main table.
Lorenzo Colitti [Tue, 29 Jul 2014 13:15:22 +0000 (13:15 +0000)]
am 2f5ea0e9: Stop copying directly-connected routes to the main table.

* commit '2f5ea0e99e9a436cd43901b1772b77a410a62f8d':
  Stop copying directly-connected routes to the main table.

10 years agoam 2bff72e0: Add "scope link" when adding routes without a nexthop.
Sreeram Ramachandran [Tue, 29 Jul 2014 13:15:21 +0000 (13:15 +0000)]
am 2bff72e0: Add "scope link" when adding routes without a nexthop.

* commit '2bff72e0cf091f4d71af0c81cfc74e8ebd8c6644':
  Add "scope link" when adding routes without a nexthop.

10 years agoStop copying directly-connected routes to the main table, #2.
Lorenzo Colitti [Tue, 29 Jul 2014 09:26:21 +0000 (18:26 +0900)]
Stop copying directly-connected routes to the main table, #2.

For a long time we have thought that copying directly-connected
routes to the main table was necessary to add gatewayed routes
to other routing tables. However, this is not necessary when the
directly-connected routes are properly created with "scope link"
as we do in http://ag/513100 .

Delete the copying code, but don't delete the rule that looks up
the main table or the code that dumps it. The main table is used
for things like cell networking, because the RIL emulates cell
networks, which are actually point-to-point, as directly
connected broadcast subnets (e.g., a /30 or a /27) with a fake
default gateway.  The directly-connected route that covers the
fake default gateway is implicitly created by adding the IP
address, but it's in the main table, so we can't add the default
route without looking up the main table.

Change-Id: I93bd4764ac75fdcc98fa4206c601524100d53fc3

10 years agoRevert "Stop copying directly-connected routes to the main table."
Lorenzo Colitti [Tue, 29 Jul 2014 09:25:44 +0000 (09:25 +0000)]
Revert "Stop copying directly-connected routes to the main table."

This reverts commit 2f5ea0e99e9a436cd43901b1772b77a410a62f8d.

Change-Id: I1fe1df0249714cb650a34fae56476236ac0108e3

10 years agoStop copying directly-connected routes to the main table.
Lorenzo Colitti [Tue, 29 Jul 2014 06:46:56 +0000 (15:46 +0900)]
Stop copying directly-connected routes to the main table.

For a long time we have thought that copying directly-connected
routes to the main table was necessary to add gatewayed routes
to other routing tables. However, this is not necessary when the
directly-connected routes are properly created with "scope link"
as we do in http://ag/513100 .

Delete the copying code, but keep dumping the main table in
bugreports, so we can see if third-party code such as RIL
daemons is putting anything in it.

Change-Id: Iddd531daaf9881ffd82f0a4b4f6cc857ce8788fd

10 years agoAdd "scope link" when adding routes without a nexthop.
Sreeram Ramachandran [Fri, 18 Jul 2014 20:03:47 +0000 (13:03 -0700)]
Add "scope link" when adding routes without a nexthop.

This is consistent with what /sbin/ip does, and it makes
recursive lookups on secondary tables work even when the main
table is empty.

This was originally proposed by Sreeram as http://ag/506456 to
fix a VPN issue, but then abandoned because it did not fix that
particular problem.

Bug: 16628572
Change-Id: I85753389c683ae2127b21af722463a35a33b60eb

10 years agoam 060d855a: Merge "Prohibit address families by default unless a VPN explicitly...
Sreeram Ramachandran [Mon, 28 Jul 2014 19:03:46 +0000 (19:03 +0000)]
am 060d855a: Merge "Prohibit address families by default unless a VPN explicitly allows them." into lmp-dev

* commit '060d855a00cb94e8903fd1551c9bf040f42b57d1':
  Prohibit address families by default unless a VPN explicitly allows them.

10 years agoam 507e7c5b: Merge "Fix fwmark handling for bypassable VPNs and DNS." into lmp-dev
Sreeram Ramachandran [Mon, 28 Jul 2014 19:03:46 +0000 (19:03 +0000)]
am 507e7c5b: Merge "Fix fwmark handling for bypassable VPNs and DNS." into lmp-dev

* commit '507e7c5bbfcd1d68f67477866497adb1748c5fd5':
  Fix fwmark handling for bypassable VPNs and DNS.

10 years agoam 48e19b03: Implement the fallthrough rule to support split tunnel VPNs.
Sreeram Ramachandran [Mon, 28 Jul 2014 14:29:59 +0000 (14:29 +0000)]
am 48e19b03: Implement the fallthrough rule to support split tunnel VPNs.

* commit '48e19b037e7e20674048ef76bf31ce65c741347c':
  Implement the fallthrough rule to support split tunnel VPNs.

10 years agoMerge "Prohibit address families by default unless a VPN explicitly allows them....
Sreeram Ramachandran [Mon, 28 Jul 2014 18:53:50 +0000 (18:53 +0000)]
Merge "Prohibit address families by default unless a VPN explicitly allows them." into lmp-dev

10 years agoProhibit address families by default unless a VPN explicitly allows them.
Sreeram Ramachandran [Sun, 27 Jul 2014 01:43:25 +0000 (18:43 -0700)]
Prohibit address families by default unless a VPN explicitly allows them.

Bug: 15972465
Change-Id: I3278d94536fefacc86390c1ba4231680f7be8589

10 years agoImplement the fallthrough rule to support split tunnel VPNs.
Sreeram Ramachandran [Wed, 23 Jul 2014 05:23:20 +0000 (22:23 -0700)]
Implement the fallthrough rule to support split tunnel VPNs.

Change-Id: Ibc48caedb5954c6b12bfa553d978bab56c4b09aa

10 years agoam 95684ba1: Implement support for bypassable VPNs.
Sreeram Ramachandran [Fri, 25 Jul 2014 23:08:55 +0000 (23:08 +0000)]
am 95684ba1: Implement support for bypassable VPNs.

* commit '95684ba176a9fe5ea59207d7202e47fa12bbfdbe':
  Implement support for bypassable VPNs.

10 years agoImplement support for bypassable VPNs.
Sreeram Ramachandran [Wed, 23 Jul 2014 20:27:31 +0000 (13:27 -0700)]
Implement support for bypassable VPNs.

Bypassable VPNs grab all traffic by default (just like secure VPNs), but:
+ They allow all apps to choose other networks using the multinetwork APIs.
  If these other networks are insecure ("untrusted"), they will enforce that the
  app holds the necessary permissions, such as CHANGE_NETWORK_STATE.
+ They support consistent routing. If an app has an existing connection over
  some other network when the bypassable VPN comes up, it's not interrupted.

Bug: 15347374
Change-Id: Iaee9c6f6fa8103215738570d2b65d3fcf10343f3

10 years agoMerge "Fix fwmark handling for bypassable VPNs and DNS." into lmp-dev
Sreeram Ramachandran [Mon, 28 Jul 2014 18:52:15 +0000 (18:52 +0000)]
Merge "Fix fwmark handling for bypassable VPNs and DNS." into lmp-dev

10 years agoFix fwmark handling for bypassable VPNs and DNS.
Sreeram Ramachandran [Fri, 25 Jul 2014 02:04:32 +0000 (19:04 -0700)]
Fix fwmark handling for bypassable VPNs and DNS.

This is a significant change to the way fwmarks are handled for two purposes:

1. Bypassable VPN.

   This was introduced in http://ag/510058 and had an issue that if there's a
   default network, it would always be used in connect(), so the bypassable VPN
   wouldn't get any traffic. This CL fixes that issue by using the bypassable
   VPN's NetId in connect(). See the comments in the code for more details.

2. DNS.

   The previous DNS code (specifically, getNetworkForUser()) had two problems:

   + Even if a user asks for a NetId they have permission for, we'd always use
     the user's VPN if they were subject to one. So, for example, a system IMS
     app that brings up the mobile network in the presence of a VPN would still
     have its DNS queries sent over the VPN, instead of mobile as desired.

   + Any user could perform DNS over any valid network, even one they didn't
     have permissions for, as long as they weren't subject to a VPN. So, for
     example, an app could use the DNS servers of a different profile's VPN.

   This CL fixes those problems. See getNetworkForDns() for more details.

The two pieces above are inter-related. Previously, we never set the explicit
bit from the DNS code. But we need to do that if the user asks for a network
explicitly, for two reasons:

o So that the DNS query is really restricted to that network and doesn't
  fallthrough to the default network.

o So that the heuristic described in ON_CONNECT works in all cases. I.e., if the
  DNS proxy's connect() request comes in with the explicit bit NOT set, we know
  that the NetId can only be either the default network or a VPN.

This CL is not intended to be robust against race conditions. In general, very
little of the netd code is resilient. A separate effort needs to be undertaken
to carefully audit all the code and logic to guard against things like:

* A VPN being established between calls to getNetworkForDns() and connect().
* State changes between multiple calls to NetworkController from clients such as
  FwmarkServer and DnsProxyListener.
* Routing rules / iptables rules being set up in a less-than-ideal order.
* ... etc.

Bug: 15347374
Change-Id: I5baad9168c4f4f3ef4129e07234b4bf24b0d8ba2

10 years agoam 111bec20: Implement the rule to forward traffic to local networks via VPNs.
Sreeram Ramachandran [Wed, 23 Jul 2014 19:36:49 +0000 (19:36 +0000)]
am 111bec20: Implement the rule to forward traffic to local networks via VPNs.

* commit '111bec203e82bdc9fb2c27df7c232465dffeee5f':
  Implement the rule to forward traffic to local networks via VPNs.

10 years agoImplement the rule to forward traffic to local networks via VPNs.
Sreeram Ramachandran [Wed, 23 Jul 2014 01:51:06 +0000 (18:51 -0700)]
Implement the rule to forward traffic to local networks via VPNs.

Change-Id: I4bffb2ce854a6fa7d4c0d35b97f123d91c6a84d6

10 years agoam fa9f4dcd: Cosmetic: Move some functions around to match rule priority order.
Sreeram Ramachandran [Wed, 23 Jul 2014 01:45:50 +0000 (01:45 +0000)]
am fa9f4dcd: Cosmetic: Move some functions around to match rule priority order.

* commit 'fa9f4dcd79dcfd24af276263159c5315abb06df6':
  Cosmetic: Move some functions around to match rule priority order.

10 years agoCosmetic: Move some functions around to match rule priority order.
Sreeram Ramachandran [Wed, 23 Jul 2014 01:16:44 +0000 (18:16 -0700)]
Cosmetic: Move some functions around to match rule priority order.

Change-Id: I81af639b66ac66272e0fae2d53119de1bfa12e69

10 years agoam 182a26ff: Remove the oif=iface clause when tethering.
Sreeram Ramachandran [Wed, 23 Jul 2014 00:23:05 +0000 (00:23 +0000)]
am 182a26ff: Remove the oif=iface clause when tethering.

* commit '182a26ff95515179b1a7e62a806f90279cdcd9b3':
  Remove the oif=iface clause when tethering.

10 years agoresolved conflicts for merge of 1604e186 to lmp-dev-plus-aosp
Sreeram Ramachandran [Wed, 23 Jul 2014 00:08:02 +0000 (17:08 -0700)]
resolved conflicts for merge of 1604e186 to lmp-dev-plus-aosp

Change-Id: Ia37af0e92adc8c119ce170e072f4cd3899927bf6

10 years agoRemove the oif=iface clause when tethering.
Sreeram Ramachandran [Tue, 22 Jul 2014 23:44:37 +0000 (16:44 -0700)]
Remove the oif=iface clause when tethering.

The rule is meant to be "iif=inputIface => lookup <table_for_outputIface>", not
"iif=foo oif=bar => lookup <table_for_bar>" (the latter would never match
because packets coming in on the input interface will not have any output
interface binding).

Bug: 16242255
Change-Id: I98a2a8ab90765aee833134297b58d4f7a212036f

10 years agoCleanup: Delete dead code.
Sreeram Ramachandran [Sun, 20 Jul 2014 06:22:33 +0000 (23:22 -0700)]
Cleanup: Delete dead code.

Bug: 15413389
Change-Id: I315468832ef18ffc84174e54774ab63b86d284dc

10 years agoam 8b3b91c6: Try to undo iptables rules if a command fails.
Sreeram Ramachandran [Tue, 22 Jul 2014 21:53:46 +0000 (21:53 +0000)]
am 8b3b91c6: Try to undo iptables rules if a command fails.

* commit '8b3b91c6cad577e2928a29073fc962c57ef75af5':
  Try to undo iptables rules if a command fails.

10 years agoTry to undo iptables rules if a command fails.
Sreeram Ramachandran [Tue, 22 Jul 2014 19:40:36 +0000 (12:40 -0700)]
Try to undo iptables rules if a command fails.

As per review comments at:
https://googleplex-android-review.git.corp.google.com/#/c/504240/5/server/NatController.cpp@194

Bug: 15413694
Bug: 15413741
Change-Id: I99e2f83792fa81498447e4c03f5f8be1d2c43b7d

10 years agoam b717e74d: Flush routing rules and add an "unreachable" rule on Init().
Sreeram Ramachandran [Tue, 22 Jul 2014 18:54:06 +0000 (18:54 +0000)]
am b717e74d: Flush routing rules and add an "unreachable" rule on Init().

* commit 'b717e74df3571951890cf36ee8bd559501d7fdc4':
  Flush routing rules and add an "unreachable" rule on Init().

10 years agoam 87475a14: Fix WiFi-Direct and Tethering.
Sreeram Ramachandran [Tue, 22 Jul 2014 18:54:05 +0000 (18:54 +0000)]
am 87475a14: Fix WiFi-Direct and Tethering.

* commit '87475a1471373b72ffc9f81f17dfd7884723fa86':
  Fix WiFi-Direct and Tethering.

10 years agoFlush routing rules and add an "unreachable" rule on Init().
Sreeram Ramachandran [Sat, 19 Jul 2014 07:22:15 +0000 (00:22 -0700)]
Flush routing rules and add an "unreachable" rule on Init().

Without the flush, rules accumulate duplicates when netd is restarted due to a
runtime restart. Nothing functionally wrong with having duplicates; it just
makes the output of "ip rule" look as though something went wrong in the system.

Time to add the unreachable rule, to suss out issues with corner cases. With the
flush, there's no more a naked "from main" rule that we need to protect by
adding the unreachable rule. But it's a good idea to add the unreachable rule
anyway, in case somebody comes along and adds a rule below it later.

Change-Id: I975b2221868b7f5366bd7cf60937a82fb4b75913

10 years agoFix WiFi-Direct and Tethering.
Sreeram Ramachandran [Tue, 15 Jul 2014 23:20:28 +0000 (16:20 -0700)]
Fix WiFi-Direct and Tethering.

A LocalNetwork object now always exists in the NetworkController, with a fixed
NetId that's guaranteed not to collide with NetIds created by the framework.

When routes are added on an interface tracked by the LocalNetwork, they are
added to a fixed "local_network" table.

When NAT is enabled, we add a special "iif -> oif" tethering rule.

Bug: 15413694
Bug: 15413741

Change-Id: I36effc438d5ac193a77174493bf196cb68a5b97a

10 years agoam f5d949ef: Merge "Don\'t choke if IPv6 rules fail."
Sreeram Ramachandran [Sat, 19 Jul 2014 11:49:24 +0000 (11:49 +0000)]
am f5d949ef: Merge "Don\'t choke if IPv6 rules fail."

* commit 'f5d949ef0991737af9daa7ba702cc2ec638e435b':
  Don't choke if IPv6 rules fail.

10 years agoMerge "Don't choke if IPv6 rules fail."
Sreeram Ramachandran [Sat, 19 Jul 2014 07:56:18 +0000 (07:56 +0000)]
Merge "Don't choke if IPv6 rules fail."

10 years agoDon't choke if IPv6 rules fail.
Sreeram Ramachandran [Thu, 17 Jul 2014 20:27:41 +0000 (13:27 -0700)]
Don't choke if IPv6 rules fail.

On networks with a low MTU, these IPv6 commands may fail because v6 requires a
minimum MTU of 1280. Ignore the failures, since these networks are clearly meant
to be IPv4 only.

Bug: 15466895
Change-Id: Ie8f24f6e7a316633fb60661a07a3e715b6ab6add

10 years agoDon't reset a VPN's NetId in the connect() shim.
Sreeram Ramachandran [Sat, 12 Jul 2014 00:06:12 +0000 (17:06 -0700)]
Don't reset a VPN's NetId in the connect() shim.

Change-Id: I0cc6c0e221a40c9100c8f4c0c5e761fce3f9b0ae

10 years agoIntroduce a SELECT_FOR_USER fwmarkd command.
Sreeram Ramachandran [Fri, 11 Jul 2014 23:27:02 +0000 (16:27 -0700)]
Introduce a SELECT_FOR_USER fwmarkd command.

This lets SYSTEM apps mark a socket with the network that would apply to an
arbitrary UID. I.e., either the VPN that applies to that user, or the default
network, if there's no such VPN.

This command will be used by system apps that proxy stuff for a user, so that
they can route the same way that user would have. Examples of such system apps
are the DnsProxyListener, MediaServer and DownloadManager.

The "explicit" bit is NOT set, so that if the user's VPN is a split tunnel, the
route lookup will fall-through to the default network as desired.

The "protect" bit is set, so that the socket bypasses any VPN applicable to the
system app itself. Note that even if the uid being proxied for is also subject
to the same VPN, this still works because the relevant rule doesn't care about
the protected bit (see modifyVpnSystemPermissionRule() in RouteController.cpp).

Change-Id: I4d501e5214b127f4ae9eaeb7befb1751cd102308

10 years agoCosmetic: Remove unnecessary constructor.
Sreeram Ramachandran [Fri, 11 Jul 2014 23:01:33 +0000 (16:01 -0700)]
Cosmetic: Remove unnecessary constructor.

Change-Id: Ib1de6e8988c4728e962da8607d0ecd98d6121e0c

10 years agoMerge "Add default routing tables to rt_tables."
Sreeram Ramachandran [Fri, 11 Jul 2014 22:29:49 +0000 (22:29 +0000)]
Merge "Add default routing tables to rt_tables."

10 years agoAdd default routing tables to rt_tables.
Sreeram Ramachandran [Fri, 11 Jul 2014 18:45:14 +0000 (11:45 -0700)]
Add default routing tables to rt_tables.

This is so that these tables also get dumped by bugreport / dumpstate.

Change-Id: Ia5ed8a23911d7b34ba964934dec42849f4a73824

10 years agoFix tethering in the case of a regular upstream connection.
Sreeram Ramachandran [Fri, 11 Jul 2014 16:10:20 +0000 (09:10 -0700)]
Fix tethering in the case of a regular upstream connection.

Fixes tethering via Ethernet, Bluetooth and WiFi (hotspot).

Tethering when the upstream has a DUN-specific APN is likely still broken
(untested).

For now, assign a fixed NetId (a hack) until we can change the framework to
create a valid NetworkAgent and all that jazz.

Bug: 15968336
Bug: 14988803
Change-Id: Idcf4d492d9329a9c87913e27be6dd835a792bea2

10 years agoCosmetic: Move constants into file scope.
Sreeram Ramachandran [Wed, 9 Jul 2014 14:39:30 +0000 (07:39 -0700)]
Cosmetic: Move constants into file scope.

Change-Id: Ic1398a5867693bb8f1e588e727869ec6a4574432

10 years agoAdd symbolic table names for routing table numbers.
Sreeram Ramachandran [Tue, 8 Jul 2014 05:11:37 +0000 (22:11 -0700)]
Add symbolic table names for routing table numbers.

This makes the output from "ip rule" much more readable.

Companion changes are in AOSP.

Change-Id: I69deb1a64d5d6647470823405bf0cc55b24b22de

10 years agoAdd full support for UIDs in VPNs.
Sreeram Ramachandran [Sun, 6 Jul 2014 00:15:14 +0000 (17:15 -0700)]
Add full support for UIDs in VPNs.

Major:
+ Implement the functions mentioned in http://go/android-multinetwork-routing
  correctly, including handling accept(), connect(), setNetworkForSocket()
  and protect() and supporting functions like canUserSelectNetwork().
+ Eliminate the old code path of getting/setting UID ranges through
  SecondaryTableController (which is currently unused) and mUidMap.

Minor:
+ Rename some methods/variables for clarity and consistency.
+ Moved some methods in .cpp files to match declaration order in the .h files.

Bug: 15409918
Change-Id: Ic6ce3646c58cf645db0d9a53cbeefdd7ffafff93

10 years agoUpdate routing rules.
Sreeram Ramachandran [Thu, 3 Jul 2014 19:20:48 +0000 (12:20 -0700)]
Update routing rules.

As per the latest changes to: http://go/android-multinetwork-routing

Functional changes:
+ Add explicit=NO to the implicit network rules, though it's a no-op.
+ Remove most of the UID=0 (kernel access) rules since they are no longer
  needed, except in one case to allow access to a VPN.
+ Add the explicit, protect and permissions bits to the incoming packet mark.
+ VPNs now don't need an implicit network rule.
+ Modifying network permissions now modifies the incoming packet mark as well.

Cosmetic changes:
+ Renamed the legacy tables to match their permissions (SYSTEM and NETWORK).
+ Renamed most functions and methods for clarity and consistency.
+ Renamed and adjusted some ule priorities.
+ Move most rule modifications into their own functions, to prevent brittle
  reliance on the previous state of the fwmark/mask variables.

Change-Id: I958a7e158ee918d5254de606fcfa55fe23327438

10 years agoFix permissions handling.
Sreeram Ramachandran [Sat, 5 Jul 2014 19:31:05 +0000 (12:31 -0700)]
Fix permissions handling.

+ Rename the permissions as per: http://go/android-multinetwork-routing
+ Make the SYSTEM permission explicitly include NETWORK.
+ Grant the SYSTEM permission to system UIDs by default, but allow the framework
  to override them if necessary.
+ Move the "string to permission" parsing to CommandListener.cpp, thus allowing
  us to get rid of Permission.cpp.
+ There's no need to support multiple permissions string arguments, so tighten
  that up.

Change-Id: I73d51b5e2f44a97e6d5ab5943ff198cebfbcc0c4

10 years agoOnly add permissions in accept(); do not try to enforce anything.
Sreeram Ramachandran [Sat, 5 Jul 2014 18:00:55 +0000 (11:00 -0700)]
Only add permissions in accept(); do not try to enforce anything.

Previously, we were enforcing that the user has access to the network over which
the socket is accepted. This has two problems:

1. We weren't handling the loopback interface ('lo') correctly. It's not part of
   any network, so the NetId in the fwmark is NETID_UNSET. In
   NetworkController::isUserPermittedOnNetwork(), we would fail to find a valid
   network, and so we would return false.

2. We have decided that in fact, we don't want to enforce this even for other
   interfaces, due to the SYN-ACK problem. See the comments and discussion at:
   http://go/android-multinetwork-routing

Bug: 16079376
Change-Id: I3f07f67d06dd7b48bfdfa5616ee22e098e31f8af

10 years agoFix regressions in flushRoutes().
Sreeram Ramachandran [Thu, 3 Jul 2014 01:06:34 +0000 (18:06 -0700)]
Fix regressions in flushRoutes().

http://ag/486276 left out a comma, causing the command to be "routeflush" (which
is invalid), instead of the correct "route flush".

http://ag/495068 changed the order, causing the interface to be removed from the
interfaceToIndex map before we are done using it in modifyPerNetworkRules().

Change-Id: I65af0b5763b13e47b48e7e2b81d243dc19e1a03b

10 years agoMinor style change.
Sreeram Ramachandran [Wed, 2 Jul 2014 21:49:33 +0000 (14:49 -0700)]
Minor style change.

Change-Id: Ib1f89bff55506cf3881047dd876adbf97c789423