This patch adds [,resourcecontrol=deny] to `-sandbox on' option. It
blacklists all process affinity and scheduler priority system calls to
avoid any bigger of the process.
Signed-off-by: Eduardo Otubo <ot...@redhat.com>
---
include/sysemu/seccomp.h | 1 +
qemu-options.hx | 9 ++++++---
qemu-seccomp.c | 28 ++++++++++++++++++++++++++++
vl.c | 11 +++++++++++
4 files changed, 46 insertions(+), 3 deletions(-)
diff --git a/include/sysemu/seccomp.h b/include/sysemu/seccomp.h
index f1614d6514..c7003dd197 100644
--- a/include/sysemu/seccomp.h
+++ b/include/sysemu/seccomp.h
@@ -18,6 +18,7 @@
#define OBSOLETE 0x0001
#define PRIVILEGED 0x0010
#define SPAWN 0x0100
+#define RESOURCECTL 0x1000
#include <seccomp.h>
diff --git a/qemu-options.hx b/qemu-options.hx
index 3d612f0fd1..7192236dee 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -4004,13 +4004,14 @@ Old param mode (ARM only).
ETEXI
DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
- "-sandbox
on[,obsolete=allow][,elevateprivileges=allow|deny|children][,spawn=deny]
Enable seccomp mode 2 system call filter (default 'off').\n" \
+ "-sandbox
on[,obsolete=allow][,elevateprivileges=allow|deny|children][,spawn=deny][,resourcecontrol=deny]
Enable seccomp mode 2 system call filter (default 'off').\n" \
" obsolete: Allow obsolete system calls\n"
" elevateprivileges: allows or denies Qemu process to
elevate its privileges by blacklisting all set*uid|gid system calls. 'children'
will deny set*uid|gid system calls for main Qemu process but will allow forks
and execves to run unprivileged\n"
- " spawn: avoids Qemu to spawn new threads or processes by
blacklisting *fork and execve\n",
+ " spawn: avoids Qemu to spawn new threads or processes by
blacklisting *fork and execve\n"
+ " resourcecontrol: disable process affinity and schedular
priority\n",
QEMU_ARCH_ALL)
STEXI
-@item -sandbox
@var{arg}[,obsolete=@var{string}][,elevateprivileges=@var{string}][,spawn=@var{string}]
+@item -sandbox
@var{arg}[,obsolete=@var{string}][,elevateprivileges=@var{string}][,spawn=@var{string}][,resourcecontrol=@var{string}]
@findex -sandbox
Enable Seccomp mode 2 system call filter. 'on' will enable syscall filtering
and 'off' will
disable it. The default is 'off'.
@@ -4021,6 +4022,8 @@ Enable Obsolete system calls
Disable set*uid|gid systema calls
@item spawn=@var{string}
Disable *fork and execve
+@item resourcecontrol=@var{string}
+Disable process affinity and schedular priority
@end table
ETEXI
diff --git a/qemu-seccomp.c b/qemu-seccomp.c
index 22a093ca1b..95c8e31d1a 100644
--- a/qemu-seccomp.c
+++ b/qemu-seccomp.c
@@ -31,6 +31,19 @@ struct QemuSeccompSyscall {
uint8_t priority;
};
+static const struct QemuSeccompSyscall resourcecontrol_syscalls[] = {
+ { SCMP_SYS(getpriority), 255 },
+ { SCMP_SYS(setpriority), 255 },
+ { SCMP_SYS(sched_setparam), 255 },
+ { SCMP_SYS(sched_getparam), 255 },
+ { SCMP_SYS(sched_setscheduler), 255 },
+ { SCMP_SYS(sched_getscheduler), 255 },
+ { SCMP_SYS(sched_setaffinity), 255 },
+ { SCMP_SYS(sched_getaffinity), 255 },
+ { SCMP_SYS(sched_get_priority_max), 255 },
+ { SCMP_SYS(sched_get_priority_min), 255 },
+};
+
static const struct QemuSeccompSyscall spawn_syscalls[] = {
{ SCMP_SYS(fork), 255 },
{ SCMP_SYS(vfork), 255 },
@@ -158,6 +171,21 @@ int seccomp_start(uint8_t seccomp_opts)
}
}
+ if (seccomp_opts & RESOURCECTL) {
+ for (i = 0; i < ARRAY_SIZE(resourcecontrol_syscalls); i++) {
+ rc = seccomp_rule_add(ctx, SCMP_ACT_KILL,
+ resourcecontrol_syscalls[i].num, 0);
+ if (rc < 0) {
+ goto seccomp_return;
+ }
+ rc = seccomp_syscall_priority(ctx, resourcecontrol_syscalls[i].num,
+
resourcecontrol_syscalls[i].priority);
+ if (rc < 0) {
+ goto seccomp_return;
+ }
+ }
+ }
+
rc = seccomp_load(ctx);
seccomp_return:
diff --git a/vl.c b/vl.c
index ae3fa8b139..d9d4b025b9 100644
--- a/vl.c
+++ b/vl.c
@@ -284,6 +284,10 @@ static QemuOptsList qemu_sandbox_opts = {
.name = "spawn",
.type = QEMU_OPT_STRING,
},
+ {
+ .name = "resourcecontrol",
+ .type = QEMU_OPT_STRING,
+ },
{ /* end of list */ }
},
};
@@ -1076,6 +1080,13 @@ static int parse_sandbox(void *opaque, QemuOpts *opts,
Error **errp)
}
}
+ value = qemu_opt_get(opts, "resourcecontrol");
+ if (value) {
+ if (strcmp(value, "deny") == 0) {
+ seccomp_opts |= RESOURCECTL;
+ }
+ }
+
if (seccomp_start(seccomp_opts) < 0) {
error_report("failed to install seccomp syscall filter "
"in the kernel");
--
2.13.3
