Proper setting of read->eof in pipe code.

Setting read->eof to 0 seems to be just a typo.  It appeared in
nginx-0.0.1-2003-10-28-18:45:41 import (r164), while identical code in
ngx_recv.c introduced in the same import do actually set read->eof to 1.

Failure to set read->eof to 1 results in EOF not being generally detectable
from connection flags.  On the other hand, kqueue won't report any read
events on such a connection since we use EV_CLEAR.  This resulted in read
timeouts if such connection was cached and used for another request.

diff --git a/src/event/ngx_event_pipe.c b/src/event/ngx_event_pipe.c
index d01b204463fcf54f6a3319b99c4b1ed4fa3c9bc0..53d738f0a81eaa7de4a5c24b2b02aaefc331ead3 100644 (file)
--- a/src/event/ngx_event_pipe.c
+++ b/src/event/ngx_event_pipe.c
@@ -149,7 +149,7 @@ ngx_event_pipe_read_upstream(ngx_event_pipe_t *p)
                 && p->upstream->read->pending_eof)
             {
                 p->upstream->read->ready = 0;
-                p->upstream->read->eof = 0;
+                p->upstream->read->eof = 1;
                 p->upstream_eof = 1;
                 p->read = 1;