* Re: [PATCH] kvm tools: Setup bridged network by a script
@ 2011-04-16 9:55 Ingo Molnar
2011-04-18 0:12 ` [PATCH] kvm tools: Add robust error handling for fork/waitpid() Amos Kong
0 siblings, 1 reply; 3+ messages in thread
From: Ingo Molnar @ 2011-04-16 9:55 UTC (permalink / raw)
To: Pekka Enberg; +Cc: KVM devel mailing list, Amos Kong
Hi, i noticed this small detail:
+ if (strcmp(params->script, "none")) {
+ pid = fork();
+ if (pid == 0) {
+ execl(params->script, params->script, net_device.tap_name, NULL);
+ _exit(1);
+ } else {
+ waitpid(pid, &status, 0);
+ if (WIFEXITED(status) && WEXITSTATUS(status) != 0) {
+ warning("Fail to setup tap by %s", params->script);
+ goto fail;
+ }
+ }
this path does not seem to have robust error handling: both fork() and
waitpid() can (and sometimes do) fail. If waitpid() fails then 'status' might
be ununitialized as well, IIRC.
Thanks,
Ingo
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH] kvm tools: Add robust error handling for fork/waitpid()
2011-04-16 9:55 [PATCH] kvm tools: Setup bridged network by a script Ingo Molnar
@ 2011-04-18 0:12 ` Amos Kong
2011-04-18 6:34 ` Ingo Molnar
0 siblings, 1 reply; 3+ messages in thread
From: Amos Kong @ 2011-04-18 0:12 UTC (permalink / raw)
To: kvm; +Cc: penberg, asias.hejun, mingo
> Hi, i noticed this small detail:
>
> + if (strcmp(params->script, "none")) {
> + pid = fork();
> + if (pid == 0) {
> + execl(params->script, params->script, net_device.tap_name, NULL);
> + _exit(1);
> + } else {
> + waitpid(pid, &status, 0);
> + if (WIFEXITED(status) && WEXITSTATUS(status) != 0) {
> + warning("Fail to setup tap by %s", params->script);
> + goto fail;
> + }
> + }
>
> this path does not seem to have robust error handling: both fork() and
> waitpid() can (and sometimes do) fail. If waitpid() fails then 'status' might
> be ununitialized as well, IIRC.
>
> Thanks,
>
> Ingo
Signed-off-by: Amos Kong <kongjianjun@gmail.com>
---
tools/kvm/virtio-net.c | 11 ++++++-----
1 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/tools/kvm/virtio-net.c b/tools/kvm/virtio-net.c
index f8d7276..430769a 100644
--- a/tools/kvm/virtio-net.c
+++ b/tools/kvm/virtio-net.c
@@ -311,13 +311,14 @@ static bool virtio_net__tap_init(const struct virtio_net_parameters *params)
if (pid == 0) {
execl(params->script, params->script, net_device.tap_name, NULL);
_exit(1);
- } else {
- waitpid(pid, &status, 0);
- if (WIFEXITED(status) && WEXITSTATUS(status) != 0) {
- warning("Fail to setup tap by %s", params->script);
- goto fail;
+ } else if (pid > 0) {
+ while(waitpid(pid, &status, 0) != pid) {
}
}
+ if (pid < 0 || (WIFEXITED(status) && WEXITSTATUS(status) != 0)) {
+ warning("Fail to setup tap by %s", params->script);
+ goto fail;
+ }
} else {
memset(&ifr, 0, sizeof(ifr));
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] kvm tools: Add robust error handling for fork/waitpid()
2011-04-18 0:12 ` [PATCH] kvm tools: Add robust error handling for fork/waitpid() Amos Kong
@ 2011-04-18 6:34 ` Ingo Molnar
0 siblings, 0 replies; 3+ messages in thread
From: Ingo Molnar @ 2011-04-18 6:34 UTC (permalink / raw)
To: Amos Kong; +Cc: kvm, penberg, asias.hejun
* Amos Kong <kongjianjun@gmail.com> wrote:
> + } else if (pid > 0) {
> + while(waitpid(pid, &status, 0) != pid) {
> }
> }
Doesn't that look like an infinite loop when waitpid() returns an error?
Thanks,
Ingo
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2011-04-18 6:34 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-04-16 9:55 [PATCH] kvm tools: Setup bridged network by a script Ingo Molnar
2011-04-18 0:12 ` [PATCH] kvm tools: Add robust error handling for fork/waitpid() Amos Kong
2011-04-18 6:34 ` Ingo Molnar
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox