[PATCH] net/passt: don't close socketpair fds after ownership transfer

[zhaoguohan@kylinos.cn]

From: GuoHan Zhao <zhaoguohan@kylinos.cn>

qemu_chr_add_client() and g_subprocess_launcher_take_fd() take
ownership of sv[0] and sv[1], respectively.

Closing them in the caller after the handoff can result in the same
fd being closed twice when the owning cleanup path runs.

Avoid this by tracking ownership and only closing descriptors that
are still owned by the caller.

Signed-off-by: GuoHan Zhao <zhaoguohan@kylinos.cn>
---
 net/passt.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/net/passt.c b/net/passt.c
index 4ff94ee509d4..930fe44cce4f 100644
--- a/net/passt.c
+++ b/net/passt.c
@@ -331,10 +331,8 @@ static int net_passt_stream_start(NetPasstState *s, Error **errp)
     /* start passt */
     if (net_passt_start_daemon(s, sv[1], errp) == -1) {
         close(sv[0]);
-        close(sv[1]);
         return -1;
     }
-    close(sv[1]);
 
     return 0;
 }
@@ -458,6 +456,7 @@ static int net_passt_vhost_user_init(NetPasstState *s, Error **errp)
         error_setg(errp, "Failed to make socket chardev");
         goto err;
     }
+    sv[0] = -1;
 
     s->vhost_user = g_new0(struct VhostUserState, 1);
     if (!qemu_chr_fe_init(&s->vhost_chr, chr, errp) ||
@@ -467,8 +466,10 @@ static int net_passt_vhost_user_init(NetPasstState *s, Error **errp)
 
     /* start passt */
     if (net_passt_start_daemon(s, sv[1], errp) == -1) {
+        sv[1] = -1;
         goto err;
     }
+    sv[1] = -1;
 
     do {
         if (qemu_chr_fe_wait_connected(&s->vhost_chr, errp) < 0) {
@@ -482,11 +483,14 @@ static int net_passt_vhost_user_init(NetPasstState *s, Error **errp)
 
     qemu_set_info_str(&s->data.nc, "vhost-user,connected to pid %d", s->pid);
 
-    close(sv[1]);
     return 0;
 err:
-    close(sv[0]);
-    close(sv[1]);
+    if (sv[0] >= 0) {
+        close(sv[0]);
+    }
+    if (sv[1] >= 0) {
+        close(sv[1]);
+    }
 
     return -1;
 }
-- 
2.43.0