Skip to content

GitLab

Switch branch/tag
  1. 02 Mar, 2016 1 commit
  3. 26 Feb, 2016 1 commit
  5. 25 Feb, 2016 3 commits
  7. 23 Feb, 2016 2 commits
    • Lorenzo Colitti's avatar
      Rename the netd service from "android.net.INetd" to "netd". · c2c7b755
      Lorenzo Colitti authored 
      Bug: 27239233
Change-Id: I34a5554604e6d8b8a50838c6911c28503cbf4540
      c2c7b755
    • Lorenzo Colitti's avatar
      Initial netd binder interface. · e4d626ea
      Lorenzo Colitti authored 
      In this change:

1. AIDL files for a new, android.net.INetd service, and
   corresponding implementation using generated code. For now the
   interface is just a prototype: it only has one trivial method.
2. Permission checking code to check for CONNECTIVITY_INTERNAL.
3. Add a Big Netd Lock and provide a wrapper that makes it easy
   to ensure that it is taken by every CommandListener command.

Bug: 27239233
Change-Id: I448d0ac233edd0e351a7fe7f13901fb6871683a2
      e4d626ea
  9. 18 Feb, 2016 1 commit
  11. 17 Feb, 2016 2 commits
  13. 16 Feb, 2016 2 commits
    • Erik Kline's avatar
      Switch to ifc_add_address() for setcfg · d7761bf0
      Erik Kline authored 
      am: 397fbc05

* commit '397fbc05':
  Switch to ifc_add_address() for setcfg
      d7761bf0
    • Erik Kline's avatar
      Switch to ifc_add_address() for setcfg · 5e69c7f6
      Erik Kline authored 
      This avoids the issue where, for example, 10.0.0.1/8 is added first,
then 10.0.0.1/8 is deleted, and 10.0.0.1/24 is added.  Netlink changes
observed at the wrong time would make it seem as though IPv4 had
been lost when in fact everthing was fine.

Bug: 26991160

(cherry picked from commit 397fbc05)

Change-Id: Ia5a5fcdd01e0950ac099c2054d74f0c52aa20b9c
      5e69c7f6
  15. 15 Feb, 2016 1 commit
    • Lorenzo Colitti's avatar
      Support killing sockets using SOCK_DESTROY. · 8464e1ed
      Lorenzo Colitti authored 
      This gives netd the ability to close sockets on a particular
source IP address using SOCK_DESTROY. It does not yet enable
this behaviour.

The microbenchmark is able to close 500 IPv6 sockets in ~30ms on
my angler. Specifically:

- Scanning 500 socketpairs: ~5ms
- Scanning 500 socketpairs and killing one half of each: ~30ms
- Scanning 500 socketpairs and killing both halves of each: ~40ms

This is about ~2.5x-3.5x slower than SIOCKILLADDR:
 - For 500 sockets, it's 9.5ms vs. 22.9ms.
 - For 4000 sockets, it's ~40ms vs ~135ms.

A large part of that is due to sending RST packets, which
SIOCKILLADDR does not do. If the kernel is modified so that
SOCK_DESTROY does not send RSTs, the time taken to kill 4000
sockets goes down to ~70ms

Batching the destroy operations does not help much. It saves
5-10%, but it complicates error handling.

Bug: 26976388
Change-Id: I2e1ac30af5dbcdb98dbb7c6e4d4d67c55b9fd00f
      8464e1ed
  17. 10 Feb, 2016 1 commit
    • Erik Kline's avatar
      Switch to ifc_add_address() for setcfg · 397fbc05
      Erik Kline authored 
      This avoids the issue where, for example, 10.0.0.1/8 is added first,
then 10.0.0.1/8 is deleted, and 10.0.0.1/24 is added.  Netlink changes
observed at the wrong time would make it seem as though IPv4 had
been lost when in fact everthing was fine.

Bug: 26991160
Change-Id: If0dad2993f818686ef8a99618e779813e9c7af90
      397fbc05
  19. 05 Jan, 2016 2 commits
  21. 24 Dec, 2015 1 commit
  23. 17 Dec, 2015 1 commit
  25. 05 Dec, 2015 1 commit
  27. 04 Dec, 2015 3 commits
  29. 25 Nov, 2015 1 commit
  31. 20 Nov, 2015 2 commits
  33. 19 Nov, 2015 4 commits
  35. 18 Nov, 2015 1 commit
    • Sreeram Ramachandran's avatar
      Don't use framework permission strings for netd permissions. · 5c8c42e9
      Sreeram Ramachandran authored 
      These framework permission strings were being used as arbitrary labels
that mapped to netd permissions that have completely different meaning.
This leads to confusion, so use different strings.

This is being cherry picked from lmp-mr1-dev to lmp-dev to fix failures
when creating restricted networks due to prior back-port a120442.

Bug: 21900139
Bug: 18194858
Change-Id: Ib3ec377ab26ce904d3d4678f04edec6cb1260517
(cherry picked from commit 584dbee5)
      5c8c42e9
  37. 13 Nov, 2015 3 commits
    • Erik Kline's avatar
      Tell dnsmasq not to start wildcard sockets · f1dfabed
      Erik Kline authored 
      Bug: 9580643
Change-Id: I8c95c15970d6ed72dc44cec42c4eb78e18ed46ba
      f1dfabed
    • Erik Kline's avatar
      Pass both IPv4 and IPv6 nameservers to dnsmasq · f90a8b92
      Erik Kline authored 
      Additionally, store and return a list of strings for
DNS forwarders, since that's what is most useful. These
strings are guaranteed to be valid IP string literals by
virtue of validation on input.

Bug: 9580643
Change-Id: I4fbaf19835055b12e38d585c0bfba0d2b18b9717
      f90a8b92
    • Erik Kline's avatar
      Change dnsmasq input command argument separator · 13fa01fa
      Erik Kline authored 
      Heretofore netd issued commands to dnsmasq over a stdin channel
using ":" as an argument separator. This make it very complicated
to issue commands that involve IPv6 addresses.

This changes the separator from ":" to "|".

Additionally:

    - Pass DHCP range string literals through to startTethering(),
      still performing string literal to IPv4 address validation.

Bug: 9580643
Change-Id: I210daf311a4c3d84dae3ed054b92c9fa399a791f
      13fa01fa
  39. 05 Nov, 2015 3 commits
  41. 02 Nov, 2015 4 commits