aboutsummaryrefslogtreecommitdiffstats
path: root/tun (follow)
Commit message (Collapse)AuthorAgeFilesLines
* tun: windows: mitigate infinite loop in Flush()Jason A. Donenfeld2019-05-311-29/+35
| | | | | It's possible that for whatever reason, we keep returning EOF, resulting in repeated close/open/write operations, except with empty packets.
* device, tun: rearrange code and fix device testsMatt Layher2019-05-291-93/+0
| | | | Signed-off-by: Matt Layher <mdlayher@gmail.com>
* wintun: remove extra /Jason A. Donenfeld2019-05-261-1/+1
|
* wintun: revise GetInterface()Simon Rozman2019-05-242-3/+6
| | | | | | | | | | | - Make foreign interface found error numeric to ease condition detection. - Update GetInterface() documentation. - Make tun.CreateTUN() quit when foreign interface found before attempting to create a Wintun interface with a duplicate name. Creation is futile. Signed-off-by: Simon Rozman <simon@rozman.si>
* wintun: never return nil, nilJason A. Donenfeld2019-05-231-1/+1
|
* setupapi: trim "Get" from gettersSimon Rozman2019-05-225-53/+53
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* wintun: don't try to flush interface, but rather deleteJason A. Donenfeld2019-05-172-25/+8
|
* wintun: make certain methods privateJason A. Donenfeld2019-05-172-33/+41
|
* wintun: change acronyms to uppercaseSimon Rozman2019-05-171-5/+5
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* global: regroup all importsJason A. Donenfeld2019-05-146-11/+16
|
* wintun: registry: fix nitsJason A. Donenfeld2019-05-111-17/+20
|
* wintun: registry: revise value readingSimon Rozman2019-05-112-67/+82
| | | | | | | | | | | | | | - 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>
* wintun: call HrRenameConnection in another threadJason A. Donenfeld2019-05-101-2/+3
|
* wintun: add LUID accessorJason A. Donenfeld2019-05-101-0/+7
|
* wintun: enumerate faster by using COMPATDRIVER instead of CLASSDRIVERJason A. Donenfeld2019-05-101-1/+1
|
* wintun: destroy devinfolist after usageJason A. Donenfeld2019-05-101-0/+1
|
* wintun: registry: replace REG_NOTIFY with NOTIFYSimon Rozman2019-05-103-13/+9
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* wintun: IpConfig is a MULTI_SZ, and fix errorsJason A. Donenfeld2019-05-102-49/+72
|
* wintun: poll for device keyJason A. Donenfeld2019-05-101-9/+10
| | | | It's actually pretty hard to guess where it is.
* wintun: fix scope of error objectJason A. Donenfeld2019-05-101-5/+6
|
* wintun: wait for interface registry key on device creationSimon Rozman2019-05-107-133/+534
| | | | | | By using RegNotifyChangeKeyValue(). Also disable dead gateway detection. Signed-off-by: Simon Rozman <simon@rozman.si>
* wintun: fix GUID leading zero paddingSimon Rozman2019-05-091-1/+1
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* setupapi: safer aliasing of slice typesJason A. Donenfeld2019-05-035-23/+40
|
* wintun: work around GetInterface staleness bugJason A. Donenfeld2019-05-032-18/+30
|
* wintun: don't retry when not creatingJason A. Donenfeld2019-05-022-9/+26
| | | | | | 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.
* wintun: try harder to open registry keyJason A. Donenfeld2019-04-291-2/+2
| | | | | This sucks. Can we please find a deterministic way of doing this instead?
* tun: freebsd: work around numerous kernel panics on shutdownJason A. Donenfeld2019-04-231-23/+82
| | | | | | | | | 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.
* setupapi: Fix struct size mismatchesSimon Rozman2019-04-194-8/+27
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* setupapi: actually fix padding by rounding up to sizeof(void*)Jason A. Donenfeld2019-04-191-1/+1
|
* tun: darwin: write routeSocket variable in helperJason A. Donenfeld2019-04-191-2/+4
| | | | Otherwise the race detector "complains".
* setupapi: Revise DrvInfoDetailData struct size calculationSimon Rozman2019-04-182-3/+7
| | | | | | | 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>
* windows: use proper constants from updated x/sysJason A. Donenfeld2019-04-133-16/+16
|
* tun: windows: Adopt new error codes returned by WintunSimon Rozman2019-04-111-3/+2
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: do not sleep after OPERATION_ABORTED on writeSimon Rozman2019-04-111-2/+1
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: do not sleep after OPERATION_ABORTEDJason A. Donenfeld2019-04-091-2/+1
|
* tun: windows: Retry R/W on ERROR_OPERATION_ABORTEDSimon Rozman2019-04-041-18/+44
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: Attempt to reopen handle on all errorsSimon Rozman2019-04-031-15/+9
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* wintun: rename device using undocumented API that netsh.exe usesJason A. Donenfeld2019-04-012-1/+41
|
* wintun: add more retry loopsJason A. Donenfeld2019-04-013-30/+56
|
* tun: windows: cancel ongoing reads on closing and delete after closeJason A. Donenfeld2019-03-261-4/+11
| | | | | This reverts commit 52ec440d7977fad966002c3710ed9df957943407 and adds some spice.
* wintun: query for NetCfgInstanceId several timesJason A. Donenfeld2019-03-221-5/+16
|
* tun: windows: delete interface before deleting file handlesJason A. Donenfeld2019-03-221-9/+4
|
* tun: windows: Make adapter rename asynchronousSimon Rozman2019-03-222-60/+13
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: Adapter devices renamed to WINTUN<LUID Index>Simon Rozman2019-03-224-52/+63
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: Increase unavailable adapter timeout to 30secSimon Rozman2019-03-221-1/+1
| | | | | | 5 seconds was too short when debugging. Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: Make writing persistent tooSimon Rozman2019-03-221-15/+17
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: Fix paused adapter testSimon Rozman2019-03-221-1/+1
| | | | Signed-off-by: Simon Rozman <simon@rozman.si>
* tun: windows: just open two file handlesJason A. Donenfeld2019-03-213-162/+83
|
* receive: implement flush semanticsJason A. Donenfeld2019-03-217-5/+28
|
* tun: windows: add dummy overlapped events backJason A. Donenfeld2019-03-211-2/+21
| | | | These seem basically wrong to me, but we get crashes without them.