diff options
| author | Michael Pratt <mpratt@google.com> | 2019-06-05 13:59:01 -0700 |
|---|---|---|
| committer | Shentubot <shentubot@google.com> | 2019-06-05 14:00:13 -0700 |
| commit | d3ed9baac0dc967eaf6d3e3f986cafe60604121a (patch) | |
| tree | 47121539775297207ba205b60b136c9093d5d393 /pkg/sentry/kernel/task_exec.go | |
| parent | cecb71dc37a77d8e4e88cdfada92a37a72c67602 (diff) | |
Implement dumpability tracking and checks
We don't actually support core dumps, but some applications want to
get/set dumpability, which still has an effect in procfs.
Lack of support for set-uid binaries or fs creds simplifies things a
bit.
As-is, processes started via CreateProcess (i.e., init and sentryctl
exec) have normal dumpability. I'm a bit torn on whether sentryctl exec
tasks should be dumpable, but at least since they have no parent normal
UID/GID checks should protect them.
PiperOrigin-RevId: 251712714
Diffstat (limited to 'pkg/sentry/kernel/task_exec.go')
| -rw-r--r-- | pkg/sentry/kernel/task_exec.go | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/pkg/sentry/kernel/task_exec.go b/pkg/sentry/kernel/task_exec.go index 5d1425d5c..35d5cb90c 100644 --- a/pkg/sentry/kernel/task_exec.go +++ b/pkg/sentry/kernel/task_exec.go @@ -68,6 +68,7 @@ import ( "gvisor.googlesource.com/gvisor/pkg/abi/linux" "gvisor.googlesource.com/gvisor/pkg/sentry/arch" "gvisor.googlesource.com/gvisor/pkg/sentry/fs" + "gvisor.googlesource.com/gvisor/pkg/sentry/mm" "gvisor.googlesource.com/gvisor/pkg/syserror" ) @@ -198,6 +199,12 @@ func (r *runSyscallAfterExecStop) execute(t *Task) taskRunState { return flags.CloseOnExec }) + // NOTE(b/30815691): We currently do not implement privileged + // executables (set-user/group-ID bits and file capabilities). This + // allows us to unconditionally enable user dumpability on the new mm. + // See fs/exec.c:setup_new_exec. + r.tc.MemoryManager.SetDumpability(mm.UserDumpable) + // Switch to the new process. t.MemoryManager().Deactivate() t.mu.Lock() |