snow/wireguard-go
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity
634 commits 2 branches 0 tags 1.3 MiB
Commit graph

634 commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Jason A. Donenfeld a6dbe4f475 wintun: don't try to flush interface, but rather delete 2019-05-17 16:06:02 +02:00
Jason A. Donenfeld c718f3940d device: fail to give bind if it doesn't exist 2019-05-17 15:35:20 +02:00
Jason A. Donenfeld 95c70b8032 wintun: make certain methods private 2019-05-17 15:01:08 +02:00
Jason A. Donenfeld 583ebe99f1 version: bump snapshot 2019-05-17 10:28:04 +02:00
Jason A. Donenfeld a6dd282600 makefile: do not show warning on non-linux 2019-05-17 10:27:51 +02:00
Simon Rozman 7d5f5bcc0d wintun: change acronyms to uppercase 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-05-17 10:22:34 +02:00
Jason A. Donenfeld 3bf41b06ae global: regroup all imports 2019-05-14 09:09:52 +02:00
Jason A. Donenfeld 3147f00089 wintun: registry: fix nits 2019-05-11 17:25:48 +02:00
Simon Rozman 6c1b66802f wintun: registry: revise value reading 
- Make getStringValueRetry() reusable for reading any value type. This
  merges code from GetIntegerValueWait().
- expandString() >> toString() and extend to support REG_MULTI_SZ
  (to return first value of REG_MULTI_SZ). Furthermore, doing our own
  UTF-16 to UTF-8 conversion works around a bug in windows/registry's
  GetStringValue() non-zero terminated string handling.
- Provide toInteger() analogous to toString()
- GetStringValueWait() tolerates and reads REG_MULTI_SZ too now. It
  returns REG_MULTI_SZ[0], making GetFirstStringValueWait() redundant.

Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-05-11 17:14:37 +02:00
Jason A. Donenfeld 5669ed326f wintun: call HrRenameConnection in another thread 2019-05-10 21:31:37 +02:00
Jason A. Donenfeld 2d847a38a2 wintun: add LUID accessor 2019-05-10 21:30:23 +02:00
Jason A. Donenfeld 7a8553aef0 wintun: enumerate faster by using COMPATDRIVER instead of CLASSDRIVER 2019-05-10 20:30:59 +02:00
Jason A. Donenfeld a6045ac042 wintun: destroy devinfolist after usage 2019-05-10 20:19:11 +02:00
Simon Rozman 1c92b48415 wintun: registry: replace REG_NOTIFY with NOTIFY 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-05-10 18:09:20 +02:00
Jason A. Donenfeld c267965bf8 wintun: IpConfig is a MULTI_SZ, and fix errors 2019-05-10 18:06:49 +02:00
Jason A. Donenfeld 1bf1dadf15 wintun: poll for device key 
It's actually pretty hard to guess where it is.
 2019-05-10 17:34:03 +02:00
Jason A. Donenfeld f9dcfccbb7 wintun: fix scope of error object 2019-05-10 16:59:24 +02:00
Simon Rozman 7e962a9932 wintun: wait for interface registry key on device creation 
By using RegNotifyChangeKeyValue(). Also disable dead gateway detection.

Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-05-10 16:43:58 +02:00
Jason A. Donenfeld 586112b5d7 conn: remove scope when sanity checking IP address format 2019-05-09 15:42:35 +02:00
Simon Rozman dcb8f1aa6b wintun: fix GUID leading zero padding 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-05-09 12:16:21 +02:00
Jason A. Donenfeld b16b0e4cf7 mod: update deps 2019-05-03 09:37:29 +02:00
Jason A. Donenfeld 81ca08f1b3 setupapi: safer aliasing of slice types 2019-05-03 09:34:00 +02:00
Jason A. Donenfeld 2e988467c2 wintun: work around GetInterface staleness bug 2019-05-03 00:42:36 +02:00
Jason A. Donenfeld 46dbf54040 wintun: don't retry when not creating 
The only time we're trying to counteract the race condition is when
we're creating a driver. When we're simply looking up all drivers, it
doesn't make sense to retry.
 2019-05-02 23:53:15 +02:00
Jason A. Donenfeld 247e14693a wintun: try harder to open registry key 
This sucks. Can we please find a deterministic way of doing this
instead?
 2019-04-29 14:00:49 +02:00
Jason A. Donenfeld 3945a299ff go.mod: use vendored winio 2019-04-29 08:09:38 +02:00
Jason A. Donenfeld bb42ec7d18 tun: freebsd: work around numerous kernel panics on shutdown 
There are numerous race conditions. But even this will crash it:

while true; do ifconfig tun0 create; ifconfig tun0 destroy; done

It seems like LLv6 is related, which we're not using anyway, so
explicitly disable it on the interface.
 2019-04-23 18:00:23 +09:00
Simon Rozman f1dc167901 setupapi: Fix struct size mismatches 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-19 10:08:11 +02:00
Jason A. Donenfeld c7a26dfef3 setupapi: actually fix padding by rounding up to sizeof(void*) 2019-04-19 10:19:00 +09:00
Jason A. Donenfeld d024393335 tun: darwin: write routeSocket variable in helper 
Otherwise the race detector "complains".
 2019-04-19 07:53:19 +09:00
Jason A. Donenfeld d9078fe772 main: revise warnings 2019-04-19 07:48:09 +09:00
Jason A. Donenfeld d3dd991e4e device: send: check packet length before freeing element 2019-04-18 23:23:03 +09:00
Simon Rozman 5811447b38 setupapi: Revise DrvInfoDetailData struct size calculation 
Go adds trailing padding to DrvInfoDetailData struct in GOARCH=386 which
confuses SetupAPI expecting exactly sizeof(SP_DRVINFO_DETAIL_DATA).

Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-18 10:39:22 +02:00
Jason A. Donenfeld e0a8c22aa6 windows: use proper constants from updated x/sys 2019-04-13 02:02:02 +02:00
Jason A. Donenfeld 0b77bf78cd conn: linux: RTA_MARK has moved to x/sys 2019-04-13 02:01:20 +02:00
Simon Rozman ef5f3ad80a tun: windows: Adopt new error codes returned by Wintun 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-11 19:38:11 +02:00
Simon Rozman a291fdd746 tun: windows: do not sleep after OPERATION_ABORTED on write 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-11 19:37:04 +02:00
Jason A. Donenfeld d50e390904 main_windows: use proper version constant 2019-04-09 10:45:40 +02:00
Jason A. Donenfeld 18fa270472 version: put version in right place 2019-04-09 10:39:48 +02:00
Jason A. Donenfeld f156a53ff4 version: bump snapshot 2019-04-09 07:37:22 +02:00
Jason A. Donenfeld e680008700 tun: windows: do not sleep after OPERATION_ABORTED 2019-04-09 07:36:03 +02:00
Simon Rozman 767c86f8cb tun: windows: Retry R/W on ERROR_OPERATION_ABORTED 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-04 09:20:18 +02:00
Simon Rozman 421c1f9143 tun: windows: Attempt to reopen handle on all errors 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-04-03 05:41:38 +02:00
Jason A. Donenfeld ac25702eaf wintun: rename device using undocumented API that netsh.exe uses 2019-04-01 12:04:44 +02:00
Jason A. Donenfeld 92f8474832 wintun: add more retry loops 2019-04-01 09:07:43 +02:00
Jason A. Donenfeld 2e0ed4614a tun: windows: cancel ongoing reads on closing and delete after close 
This reverts commit 52ec440d79 and adds
some spice.
 2019-03-26 16:14:32 +01:00
Jason A. Donenfeld 2fa80c0cb7 wintun: query for NetCfgInstanceId several times 2019-03-22 16:48:40 -06:00
Jason A. Donenfeld 52ec440d79 tun: windows: delete interface before deleting file handles 2019-03-22 16:45:58 -06:00
Simon Rozman 2faf2dcf90 tun: windows: Make adapter rename asynchronous 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-03-22 16:36:30 +01:00
Simon Rozman 41c30a7279 tun: windows: Adapter devices renamed to WINTUN<LUID Index> 
Signed-off-by: Simon Rozman <simon@rozman.si>
 2019-03-22 15:29:14 +01:00