From: Spike Curtis <spike@coder.com>
To: wireguard@lists.zx2c4.com
Subject: wireguard-go - peer goroutines can SIGSEGV on device.Close()
Date: Thu, 21 Jul 2022 08:38:03 -0700 [thread overview]
Message-ID: <CAAi99JAdRNepDqE-+P1g7TnBmPa5AU8DBdTXbee8ZbGxZ70deQ@mail.gmail.com> (raw)
I have encountered what I think is a bug in wireguard-go which causes
a SIGSEGV like:
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x20 pc=0x1006ce478]
goroutine 150 [running]:
golang.zx2c4.com/wireguard/tun/netstack.(*netTun).Write(0x140004501e0,
{0x14000192000, 0x58, 0x100000000000000?}, 0x3?)
/Users/spike/go/pkg/mod/github.com/coder/wireguard-go/tun/netstack@v0.0.0-20220614153727-d82b4ba8619f/tun.go:192
+0x108
golang.zx2c4.com/wireguard/device.(*Peer).RoutineSequentialReceiver(0x14000558000)
/Users/spike/go/pkg/mod/golang.zx2c4.com/wireguard@v0.0.0-20220703234212-c31a7b1ab478/device/receive.go:476
+0x4a0
created by golang.zx2c4.com/wireguard/device.(*Peer).Start
/Users/spike/go/pkg/mod/golang.zx2c4.com/wireguard@v0.0.0-20220703234212-c31a7b1ab478/device/peer.go:199
+0x2c8
I believe this happens when I call `Close()` on the Device with this
Peer. Closing the device triggers gVisor to Attach a nil
NetworkDispatcher to the tun, which the Peer goroutine attempts to
dereference via Write. This is a race condition that depends on
whether any packets are queued up at the time: if the queues are
empty, we are fine.
I'm able to work around this issue by calling RemoveAllPeers() prior
to Close() on the device, but presumably it wasn't the intention to
make Close() on its own dangerous in this way.
I'm happy to offer a patch if you're interested.
--
Spike Curtis
Principal Engineer
coder.com
reply other threads:[~2022-07-26 13:12 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAAi99JAdRNepDqE-+P1g7TnBmPa5AU8DBdTXbee8ZbGxZ70deQ@mail.gmail.com \
--to=spike@coder.com \
--cc=wireguard@lists.zx2c4.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).