| 44692de7aeb1dad7b87874dc41508b60be24eb82 |
|
16-Nov-2015 |
Lorenzo Colitti <lorenzo@google.com> |
Don't enable MDNS on non-multicast, non-broadcast interfaces.
This is unlikely to work, and in the case of cell interfaces, it
may be harmful (e.g., use metered bandwidth, or compete for
resources with carrier traffic such as VoLTE).
Bug: 25629907
Bug: 25669326
Change-Id: I2622182ce1a3bb1c605de8afce61cf5ae1f4cbfb
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| 0b6b4e32c36ce143bfd095119b4939e4cf826021 |
|
10-Nov-2014 |
Joshua Melcon <jmelcon@google.com> |
Sync google3/third_party/mDNSResponder with this version.
This includes updating mDNSResponder to a more recent version.
Additionally, we introduce a make target that is suitable for
GmsCore to use. Soon GmsCore will embed a version of mDNS
locally for its 'Local App Connections API'.
Change-Id: I95c1e39beb45c06256b1c7959b34af24e827135a
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| 54bbef089d469dba7032c9cefa703647a0e917f1 |
|
25-Mar-2014 |
Dave Platt <dplatt@google.com> |
Close FILE* handles after consuming contents
Two more file-descriptor leaks. Code was using fopen() to get a
handle, but was never calling fclose() once the file's contents had
been read and processed.
Bug: 13490553
Change-Id: I8c7dbf9d913d6363a14bb10bb5087836181eb1c5
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| f48f8a2e3417f93bc1f45930dbf9c4ea7523523d |
|
19-Mar-2014 |
Dave Platt <dplatt@google.com> |
Fix mDNS socket leak during network configuration changes
Several rather important close() calls were embedded within
assert() macros. Not a good idea, as the macros don't compile
into anything at all in non-debug builds, and the close() calls
disappear entirely. This could eventually lead to failure of
the mDNS daemon when its file-descriptor table fills up with
orphans.
Bug: 13490553
Change-Id: I5978c055e8475d8f5be7e0242b80593b45ed71a1
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| 2fd068f98fa215f2681ddbe89fd7404d53218be3 |
|
20-Feb-2014 |
Christopher Lane <lanechr@google.com> |
Always use SO_REUSEADDR, never SO_REUSEPORT
We need to build mdnsd for 3.4.x and 3.10 while only using one set of kernel
headers, so for 3.4.x SO_REUSEPORT is set in the headers but isn't actually
supported in the kernel. Since this is UDP, SO_REUSEPORT doesn't get much over
SO_REUSEADDR, so just always use SO_REUSEADDR.
Bug: 12995869
Change-Id: Idce363c3bb3b53f7f4bd3694a2d4f6350a981e0c
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| afc86e3216439a9437ea54905853275111eac4e2 |
|
06-Apr-2012 |
Robert Greenwalt <rgreenwalt@google.com> |
Reduce MDNS logging
If it's compiled with no debugging turned on we should not log all
the info and warning and stuff.
Change-Id: Ifacfeca81d24a8a542e3ce2eb0684643eb35e166
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| dd52342e088715929196886494c5055f9186d5c3 |
|
29-Mar-2012 |
Robert Greenwalt <rgreenwalt@google.com> |
Some fixes for Android
- Lets us not start as root and control who we switch to - switching
to "nobody" prevents us from opening sockets!
- Ratches down the debug output - packet debuggins is very verbose
- Removes some random debugging I'd added
Change-Id: I6d1acf9653dbf0d2c46f7e6081672f26982ffad4
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|
| 47e4cebad7397422144bb03a21f3f7682c062c4a |
|
27-Mar-2012 |
Robert Greenwalt <rgreenwalt@google.com> |
Initial code submission for mDnsSd
AKA known as Bonjour.
Change-Id: I7ddf2238deaebe1bdd4c5da83b5ec9038ebec5c6
/external/mdnsresponder/mDNSPosix/mDNSPosix.c
|