BUG/MEDIUM: mux-h2: handle remaining read0 cases

Commit 3d4631fec ("BUG/MEDIUM: mux-h2: fix read0 handling on partial
frames") tried to address an issue introduced in commit aade4edc1 where
read0 wasn't properly handled in the middle of a frame. But the fix was
incomplete for two reasons:

  - first, it would set H2_CF_RCVD_SHUT in h2_recv() after detecting
    a read0 but the condition was guarded by h2_recv_allowed() which
    explicitly excludes read0 ;

  - second, h2_process would only call h2_process_demux() when there
    were still data in the buffer, but closing after a short pause to
    leave a buffer empty wouldn't be caught in this case.

This patch fixes this by properly taking care of the received shutdown
and by also waking up h2_process_demux() on an empty buffer if the demux
is not blocked.

Given the patches above were tagged for backporting to 2.0, this one
should be as well.
This commit is contained in:
Willy Tarreau 2021-02-05 11:41:46 +01:00
parent c1ddcafdf9
commit f09612289f

View File

@ -3634,11 +3634,11 @@ static int h2_recv(struct h2c *h2c)
ret = max ? conn->xprt->rcv_buf(conn, conn->xprt_ctx, buf, max, 0) : 0;
if (max && !ret && h2_recv_allowed(h2c)) {
if (max && !ret) {
if (conn_xprt_read0_pending(h2c->conn)) {
TRACE_DATA("received read0", H2_EV_H2C_RECV, h2c->conn);
h2c->flags |= H2_CF_RCVD_SHUT;
} else {
} else if (h2_recv_allowed(h2c)) {
TRACE_DATA("failed to receive data, subscribing", H2_EV_H2C_RECV, h2c->conn);
conn->xprt->subscribe(conn, conn->xprt_ctx, SUB_RETRY_RECV, &h2c->wait_event);
}
@ -3843,7 +3843,8 @@ static int h2_process(struct h2c *h2c)
TRACE_ENTER(H2_EV_H2C_WAKE, conn);
if (b_data(&h2c->dbuf) && !(h2c->flags & H2_CF_DEM_BLOCK_ANY)) {
if (!(h2c->flags & H2_CF_DEM_BLOCK_ANY) &&
(b_data(&h2c->dbuf) || (h2c->flags & H2_CF_RCVD_SHUT))) {
h2_process_demux(h2c);
if (h2c->st0 >= H2_CS_ERROR || conn->flags & CO_FL_ERROR)